Animals for food, animals for tools: fauna as a source of raw material at Abri Cellier, Dordogne, and the Grotte du Renne, Arcy-sur-Cure

Clare Tolmie

The point during the Middle Paleolithic: reflection of a specific status in Neanderthal hunting equipment and toolkits La pointe au Paléolithique moyen : reflet d’un statut particulier dans l’équipement de chasse et l’outillage des Néandertaliens During the course of the Middle Paleolithic, Neanderthals produced flakes, blades and points according to wide-ranging operative schemas. Previous research carried out in the north of France has shown the unique aspect of the laminar phenomenon (Révillion, 1994; Locht et al., 2002; Locht et al., 2010), but studies focusing on point production remained rare until recently. However, new studies have shown that the point was not just an anecdotal element of the Neanderthal toolkit, as was generally considered for a long time, but that it was associated with a specific status for these populations (Knecht, 1997; Soressi & Locht, 2010; Goval, 2012; Hérisson, 2012). Although it is universally accepted that Neanderthal Man was capable of hunting large herbivores for food supplies, hunting techniques and tools are often largely unexplained. To date, it is difficult to accurately identify the weapons used for hunting, although hafting and impact marks on points show that this tool must have been part of the hunting equipment (Boeda et al., 1996; Plisson & Beyries, 1998). Other use wear analyses indicate that the point was handheld, like a knife (with the pointed end serving as an apex). The status of the point thus remains enigmatic as the quantity and production methods of these tools vary hugely from site to site. In the north of France, points are produced and/or brought into the production, kill or processing sites by Man, but this phenomenon appears to be limited to certain specific geographic zones. The chronological scope of this communication is the Middle Paleolithic, from the earliest to the most recent phase, encompassing human occupations from the Saalian to the early Weichselian (from 300 000 BP to 70 000 BP). This chronological choice stems from the absence of points in the assemblages in the north of France after isotopic stage 4. This communication focuses on the point as an element of the toolkit of these populations, but also on the production, the function, the mobility and the representation of this object, or, in other words, on its status.

Did Neanderthal produce a bone industry? The recent discovery of a large bone tool assemblage at the Neanderthal site of Chagyrskaya (Altai, Siberia, Russia) and the increasing discoveries of isolated finds of bone tools in various Mousterian sites across Eurasia stimulate the debate. Assuming that the isolate finds may be the tip of the iceberg and that the Siberian occurrence did not result from a local adaptation of easternmost Neanderthals, we looked for evidence of a similar industry in the Western side of their spread area. We assessed the bone tool potential of the Quina bone-bed level currently under excavation at chez Pinaud site (Jonzac, Charente-Maritime, France) and found as many bone tools as flint ones: not only the well-known retouchers but also beveled tools, retouched artifacts and a smooth-ended rib. Their diversity opens a window on a range of activities not expected in a butchering site and not documented by the flint tools, all involved in the carcass processing...

The vast majority of tools recovered from Palaeolithic sites are made of stone varieties. Only rarely do non-lithic implements come to light, let alone tools produced on marine mollusc shell. Interestingly, a good number of shell implements made on Callista chione and Glycymeris sp. valves have been reported from 13 Middle Palaeolithic (Mousterian) sites in southern peninsular Europe. Of these, more than 300 specimens display evidence of deliberate edge retouch. They are all considered products of Neanderthals and date from ~110 ka BP to perhaps ~50 ka BP. In this paper, we review the evidence for Mousterian shell tool production in Italy and Greece—the only two countries in which such tools have been securely identified—and present experimental results obtained in the effort to understand the production process and typo-functional role(s) of the artefacts. We examine the general provisioning pattern of raw materials, as well as the typological, species-related and chronological data pertinent to the production of shell tools by Neanderthals. The data suggest that the Mousterian shell scrapers are a response to poor availability of lithic raw material in the areas of occurrence, and may be best described as an extension of chipped stone technologies to specific types of marine shell, their form defined by an existing mental template. As such, they constitute evidence for refined adaptation strategies and advanced provisioning of resources amongst Neanderthals, and may lend further support to the idea that these hominids displayed a degree of complex behaviour.

University of Iowa Masthead Logo Iowa Research Online Theses and Dissertations Spring 2013 Animals for food, animals for tools: fauna as a source of raw material at Abri Cellier, Dordogne, and the Grotte du Renne, Arcy-sur-Cure Clare Tolmie University of Iowa Copyright 2013 Clare Tolmie This dissertation is available at Iowa Research Online: https://ir.uiowa.edu/etd/2647 Recommended Citation Tolmie, Clare. "Animals for food, animals for tools: fauna as a source of raw material at Abri Cellier, Dordogne, and the Grotte du Renne, Arcy-sur-Cure." PhD (Doctor of Philosophy) thesis, University of Iowa, 2013. https://doi.org/10.17077/etd.2w6i1ycf Follow this and additional works at: https://ir.uiowa.edu/etd Part of the Anthropology Commons ANIMALS FOR FOOD, ANIMALS FOR TOOLS: FAUNA AS A SOURCE OF RAW MATERIAL AT ABRI CELLIER, DORDOGNE, AND THE GROTTE DU RENNE, ARCY-SUR-CURE by Clare Tolmie An Abstract Of a thesis submitted in partial fulfillment of the requirements for the Doctor of Philosophy degree in Anthropology in the Graduate College of The University of Iowa May 2013 Thesis Supervisor: Professor James G. Enloe 1 ABSTRACT The adoption of bone tool technology in the Early Upper Palaeolithic of Europe by Neanderthals and anatomically modern humans has been the focus of considerable debate. In particular this debate has focused on the origins of the technology and the possible implications for the extinction of Neanderthals. This dissertation examines the context of element selection for use as raw material to produce bone tools, related to prey species in the Châtelperronian of the Grotte du Renne, Arcy-sur Cure and the Aurignacian of Abri Cellier, Dordogne. Current research indicates that there was little difference in the subsistence organization of Neanderthals and modern humans. As a more nuanced view of Neanderthal behavior emerges from recent studies, it is becoming apparent that differences between the two hominins are a matter of degree rather than absolute difference. The faunal analysis of the two assemblages in this dissertation found that both Neanderthals and modern humans were pursuing a foraging strategy to obtain prime age herbivores for food. Locally available taxa were taken. Carcasses were processed for meat, marrow and fat. Both assemblages show a preference for non-marrow bearing long bones or long bone shaft fragments to make tools. The raw material was chosen with reference to the mechanical properties of the bones, which exhibit elasticity necessary for use as awls or hide scrapers. Raw material was a by-product of the larger subsistence strategy. There is a difference in the use of antler. This is not used by Neanderthals. In the Aurignacian, it appears that the amount of antler represented by the points and tools at Abri Cellier could be obtained as part of a general foraging strategy. The appearance of bone tools in the Early Upper Palaeolithic has been argued as evidence for ‘modern’ behavior. It might be more profitable to view the adoption of this new technology as a response by two different but related populations to particular 2 ecological problems. It could be argued that the archaeological visibility of bone tools reflects an increasing investment in the production of more effective clothing by both Neanderthals and modern humans. Abstract Approved: ____________________________________ Thesis Supervisor ____________________________________ Title and Department ____________________________________ Date ANIMALS FOR FOOD, ANIMALS FOR TOOLS: FAUNA AS A SOURCE OF RAW MATERIAL AT ABRI CELLIER, DORDOGNE, AND THE GROTTE DU RENNE, ARCY-SUR-CURE by Clare Tolmie A thesis submitted in partial fulfillment of the requirements for the Doctor of Philosophy degree in Anthropology in the Graduate College of The University of Iowa May 2013 Thesis Supervisor: Professor James G. Enloe Copyright by CLARE TOLMIE 2013 All Rights Reserved Graduate College The University of Iowa Iowa City, Iowa CERTIFICATE OF APPROVAL _______________________ PH.D. THESIS _______________ This is to certify that the Ph.D. thesis of Clare Tolmie has been approved by the Examining Committee for the thesis requirement for the Doctor of Philosophy degree in Anthropology at the May 2013 graduation. Thesis Committee: ___________________________________ James G. Enloe, Thesis Supervisor ___________________________________ E. Arthur Bettis III ___________________________________ Russell L. Ciochon ___________________________________ Robert G. Franciscus ___________________________________ Matthew E. Hill To David and all my family ii ACKNOWLEDGMENTS This project could not have been undertaken without access to two significant archaeological collections. I would like to thank Dr. Bill Green, director of the Logan Museum for permission to use the Abri Cellier faunal material as part of my dissertation research. I would also like to thank Nicolette Meister, Collections Manager, Dan Bartlett and all the Logan staff for their kindness and help during my visits to the museum. The other data set used in my research is the faunal material from Level Xc of the Grotte du Renne, Arcy-sur-Cure. Madame Francine David, of the Laboratoire d’Ethnologie Préhistorique, CNRS generously offered this material for my use. Working with Francine is a wonderful and enriching experience. I would also like to thank Pierre Bodu, Maurice Hardy, Francoise Audouze, Michele Julien, Olivier Bignon, and Nejma Goutas for their interest in my project and help during my stay in France. I also thank Francine, Mme Francine Morillot and her family, and Remi David and his family for providing me with accommodation. I thank Jean-Jacques Cley-Merle for allowing me to examine the Abri Cellier fauna and tools held the Musée National de La Préhistoire, Les Eyzies. I have also appreciated the advice of Dr. Randall White, Dr. Anne Grauer, Kathy Ehrhardt, Luc Doyon, Alex Woods, and my committee. My research was supported by funding from the Stanley Foundation, the Center for Global and Regional Environmental Research; a T. Anne Cleary Graduate Travel fellowship and a University of Iowa Summer Fellowship. This financial support is greatly appreciated. Finally I would like to thank all my friends in Iowa and elsewhere, including Rochelle Lurie, Catherine Bird and all my colleagues at Midwest Archaeological Research Services, Inc. I thank Rochelle in particular for encouraging me to return to academia to obtain my PhD. Of course, my family’s support and encouragement have iii been very important to me throughout my research project. My one regret is that my father is not here to see what his encouragement of my interest in history and archaeology has produced. This dissertation is dedicated to all my family, but especially to my husband, David McCallum. I cannot find the words to express what his love and support have meant to me. iv ABSTRACT The adoption of bone tool technology in the Early Upper Palaeolithic of Europe by Neanderthals and anatomically modern humans has been the focus of considerable debate. In particular this debate has focused on the origins of the technology and the possible implications for the extinction of Neanderthals. This dissertation examines the context of element selection for use as raw material to produce bone tools, related to prey species in the Châtelperronian of the Grotte du Renne, Arcy-sur Cure and the Aurignacian of Abri Cellier, Dordogne. Current research indicates that there was little difference in the subsistence organization of Neanderthals and modern humans. As a more nuanced view of Neanderthal behavior emerges from recent studies, it is becoming apparent that differences between the two hominins are a matter of degree rather than absolute difference. The faunal analysis of the two assemblages in this dissertation found that both Neanderthals and modern humans were pursuing a foraging strategy to obtain prime age herbivores for food. Locally available taxa were taken. Carcasses were processed for meat, marrow and fat. Both assemblages show a preference for non-marrow bearing long bones or long bone shaft fragments to make tools. The raw material was chosen with reference to the mechanical properties of the bones, which exhibit elasticity necessary for use as awls or hide scrapers. Raw material was a by-product of the larger subsistence strategy. There is a difference in the use of antler. This is not used by Neanderthals. In the Aurignacian, it appears that the amount of antler represented by the points and tools at Abri Cellier could be obtained as part of a general foraging strategy. The appearance of bone tools in the Early Upper Palaeolithic has been argued as evidence for ‘modern’ behavior. It might be more profitable to view the adoption of this new technology as a response by two different but related populations to particular v ecological problems. It could be argued that the archaeological visibility of bone tools reflects an increasing investment in the production of more effective clothing by both Neanderthals and modern humans. vi TABLE OF CONTENTS LIST OF TABLES ............................................................................................................. xi LIST OF FIGURES ......................................................................................................... xiv INTRODUCTION ...............................................................................................................1 Organization. ....................................................................................................2 CHAPTER 1: WHAT DO PALAEOANTHROPOLOGISTS MEAN WHEN THEY SAY ACCULTURATION? ..................................................................5 Introduction.......................................................................................................5 Archaeological cultures and typologies ............................................................6 Transitions, time, and technological innovation ...............................................8 The issue(s) of acculturation ...........................................................................10 Acculturation, transculturation or ethnogenesis .............................................13 Independent innovations .................................................................................15 Conclusion ......................................................................................................18 CHAPTER 2: LITHICS AND HUNTING-SIMILARITIES AND DIFFERENCES .......21 Introduction.....................................................................................................21 Lithic technology ............................................................................................23 Symbolic behavior ..........................................................................................25 Subsistence .....................................................................................................26 Landscape use .................................................................................................31 Conclusion ......................................................................................................35 CHAPTER 3: PROXIES FOR THE PALAEOLITHIC: HUNTER –GATHERER STUDIES ........................................................................................................37 Introduction.....................................................................................................37 Subsistence organization: models and reality .................................................38 Learning to hunt and forage – a lifetime learning experience ........................45 Clothing: the other time consuming by-product of hunting ...........................46 Conclusion ......................................................................................................51 CHAPTER 4: NEANDERTHAL LIFE HISTORIES AND IMPLICATIONS FOR SUBSISTENCE ..............................................................................................53 Introduction.....................................................................................................53 Neanderthal ontogeny .....................................................................................53 Nourishing a demanding brain........................................................................56 Provisioning and group organization ..............................................................59 Was life really nasty, brutish and short ..........................................................61 The Neanderthal who came in from the cold..................................................64 Conclusion ......................................................................................................66 vii CHAPTER 5: NEANDERTHALS, MODERNS AND BONE TOOL USE: THE RESEARCH PROJECT..................................................................................68 Introduction.....................................................................................................68 Description of the project ...............................................................................69 Research hypotheses and testable models ......................................................75 Testing Null Hypothesis 1 ................................................................76 Testing Null Hypothesis 2 ................................................................76 Testing Null Hypothesis 3 ................................................................77 Conclusion ......................................................................................................77 CHAPTER 6: TAPHONOMIC ISSUES, A REVIEW ......................................................79 Introduction.....................................................................................................79 Taphonomy and ethnoarchaeology .................................................................83 Non-human agents of accumulaiton ...............................................................87 Post-depositional taphonomic factors .............................................................89 Conclusion ......................................................................................................96 CHAPTER 7: THE GROTTE DU RENNE, ARCY-SUR-CURE: PREVIOUS RESEARCH AND CURRENT CONTROVERSY........................................98 Introduction.....................................................................................................98 Previous Research ...........................................................................................98 Description of the site ...................................................................................101 Environmental context ..................................................................................104 Cultural material from the Grotte du Renne .................................................106 A controversial site for a controversial period .............................................108 The issue of disturbance ...............................................................................114 Direct evidence for Neanderthal occupation of Level Xc ............................116 CHAPTER 8: FAUNAL ANALYSIS OF LEVEL XC OF THE GROTTE DU RENNE, ARCY-SUR-CURE .......................................................................118 Introduction...................................................................................................118 Taxa present in Level Xc ..............................................................................118 NISP and MNI ..............................................................................................122 Unidentified mammal bone and esquilles ......................................133 Taphonomy ...................................................................................................136 General condition of the assemblage .............................................137 Density values ................................................................................140 Damage by animal gnawing ...........................................................145 Staining...........................................................................................150 Burning ...........................................................................................150 Summary of taphonomy .................................................................152 Herbivores.....................................................................................................152 Reindeer (Rangifer tarandus).........................................................152 Horse (Equus caballus) ..................................................................165 Bovidae...........................................................................................172 Red deer (Cervus elaphus) .............................................................174 Megafauna ......................................................................................176 Hare (Lepus sp.) .............................................................................176 Carnivores .....................................................................................................177 viii Cave bear (Ursus speleaus) ............................................................177 Hyena (Crocuta spelaeus) ..............................................................181 Wolf (Canis lupus) .........................................................................184 Felidae ............................................................................................185 Neanderthal subsistence and behavior ..........................................................185 Evidence for fat processing ............................................................189 Discard patterns ..............................................................................191 Summary of subsistence activities .................................................196 Conclusion ....................................................................................................197 CHAPTER 9: ABRI CELLIER: POLITICS AND PREHISTORY ................................199 Introduction...................................................................................................199 Site location in a regional context ................................................................200 All politics is personal, even in archaeology ................................................202 Excavations at Abri Cellier ...........................................................................205 Previous research on the Cellier collection ..................................................209 Fauna: the orphan child of the Palaeolithic ..................................................210 Conclusion ....................................................................................................212 CHAPTER 10: FAUNAL ANALYSIS OF ABRI CELLIER .........................................213 Introduction...................................................................................................213 Taxa present at Abri Cellier ..........................................................................214 NISP and MNI ..............................................................................................218 Unidentifiable bone ........................................................................232 Taphonomy ...................................................................................................234 General condition of the assemblage .............................................235 Density and survivorship ................................................................237 Carnivore gnawing .........................................................................242 Staining...........................................................................................245 Burning ...........................................................................................245 Summary of taphonomy .................................................................246 Herbivores.....................................................................................................247 Reindeer (Rangifer tarandus).........................................................247 Horse (Equus caballus) ..................................................................255 Bovids.............................................................................................263 Red deer (Cervus elaphus) .............................................................269 Saiga (Saiga tataricus) ...................................................................275 Cervidae .........................................................................................276 Other herbivores .............................................................................276 Summary for herbivores .................................................................276 Non-herbivores .............................................................................................277 Wolf (Canis lupus) .........................................................................277 Fox ..................................................................................................277 Bear (Ursus speleaus) ....................................................................278 Wild boar (Sus scrofa)....................................................................278 Bird (Aves)......................................................................................278 Fish (Pisces) ...................................................................................278 Summary for non-herbivores .........................................................278 Prey selection ................................................................................................279 Tool blanks ...................................................................................................282 Conclusion ....................................................................................................283 ix CHAPTER 11: DISCUSSION: A BONE TO PICK, OR SCRAPE WITH ....................284 Introduction...................................................................................................284 Osseous material culture studies: a brief history ..........................................285 Bone formation and structure .......................................................................286 Antler and ivory formation ...........................................................................288 Bone tool manufacture and use: archaeological, ethnographic and experimental data ..........................................................................................290 Bone tool manufacture in the Upper Palaeolithic and the industries at Arcy-sur-Cure and Abri Cellier ....................................................................293 Tool use and manufacture at the Grotte du Renne .......................................297 Tool manufacture at Abri Cellier ..................................................................301 Antler supplies – logistical behavior or simple collection? ..........................306 Conclusion ....................................................................................................309 CHAPTER 12: CONCLUSIONS AND SUGGESTIONS FOR FURTHER RESEARCH .................................................................................................311 Introduction...................................................................................................311 Testing the null hypotheses ..........................................................................313 Conclusion and further research ...................................................................317 APPENDIX: CUTMARK LOCATIONS ON ELEMENTS FROM LEVEL RXC, GROTTE DU RENNE, AND ABRI CELLIER ...........................................320 REFERENCES CITED ....................................................................................................352 x LIST OF TABLES Table 7.1: Radiocarbon dates for the Grotte du Renne, for levels dating from the Mousterian (XII) through the Gravettian (V) periods. ...........................................110 8.1: Minimum Number of Individuals and Number of Identified Specimens identified to genus and/or species ...........................................................................122 8.2: Summary showing calculations of Minimum Number of Individuals for taxa in Level Xc. .................................................................................................................123 8.3. Summary of NISP for reindeer, horse, red deer, bovids and hare from Level Xc. ...........................................................................................................................126 8.4: Summary of NISP for cave bear, wolf, hyena felid and mammoth from Level Xc. ...........................................................................................................................128 8.5: Percentages of NISP by element for reindeer, horse, bovids, red deer and hare in Level Xc..............................................................................................................130 8.6: Percentage of NISP by element for cave bear, hyena, wolf, felid and mammoth in Level Xc..............................................................................................................132 8.7: Summary of unidentified mammal bone fragments and esquilles from Level Xc. ...........................................................................................................................134 8.8: Table showing MNE, survivorship and density for reindeer. ...................................141 8 9: Table showing MNE, survivorship and density for horse. .......................................143 8.10: Summary of the Minimum Number of Elements and Minimal Animal Units for reindeer in Level Xc. .........................................................................................154 8.11: Summary table of reindeer appendicular skeleton bone fragments, lengths in millimeters. .............................................................................................................157 8.12: Summary table of reindeer axial element fragments, lengths in millimeters. ........158 8.13: Table showing the proportions of dry, fresh and undetermined breaks by element for reindeer in Level Xc. ...........................................................................160 8.14: Summary table Minimum Number of Elements and Minimum Animal Units for horse in Level Xc. .............................................................................................166 8.15: Summary table of horse appendicular skeleton bone fragments, lengths in millimeters. .............................................................................................................168 8.16: Percentage of dry, fresh and undetermined breaks by element for horse. ..............170 8.17: Table showing Minimum Number of Elements, survivorship and Minimum Number of Animal Units for bovids in Level Xc. ..................................................173 xi 8.18: MNE, survivorship and NISP for red deer from Level Xc. ....................................175 8.19: Table showing the NME, % survival of elements and NISP for adult bears in level Xc. ..................................................................................................................178 8.20: Table showing percentage of dry, fresh and undetermined breaks by element for cave bear in Level Xc. .......................................................................................179 8.21 Table showing the Minimum Number of Elements for hyena in Level Xc.............183 10.1: Total Number of Identified Specimens by count from all collections of fauna from Abri Cellier. ...................................................................................................215 10. 2: Total Number of Identified Specimens for fauna from Abri Cellier curated at the Logan Museum. ................................................................................................216 10.3: Total Minimum Number of Individuals from all collections from Abri Cellier. ...219 10.4: Total MNI for fauna from Abri Cellier held at the Logan Museum. ......................221 10.5: Herbivores at Abri Cellier, NISP counts and percentages, excluding unidentified cervidae and hare. ...............................................................................225 10 6: Summary table of NISP per element by level and taxon for herbivores from the Abri Cellier fauna held at Beloit College. ........................................................226 10.7: Summary table of NISP per element for non-herbivores from Abri Cellier fauna held at Beloit College....................................................................................229 10.8: Table showing the total amount of unidentified bone by category from Abri Cellier......................................................................................................................232 10.9: Summary of expected, observed and survival rate of reindeer elements in the Upper Level and their representation as Minimum Animal Units..........................248 10.10: Summary of expected, observed and survival rate of reindeer elements in the Lower Level and their representation as Minimum Animal Units. ........................250 10.11: Table showing reindeer appendicular bone fragment lengths, both levels (in millimeters). ............................................................................................................252 10.12: Table showing proportions of dry, fresh and undetermined breaks for elements for reindeer in all levels of Abri Cellier. .................................................253 10.13: Summary table of expected, observed and survival rate of horse elements in the Upper Level and their representation as Minimum Animal Units....................256 10.14: Summary of expected, observed and survival rate of horse elements in the Lower Level and their representation as Minimum Animal Units. ........................258 10.15: Horse appendicular element fragment lengths, both levels, in millimeters. .........260 10.16: Proportions of dry, fresh and undetermined breaks by element for horse at Abri Cellier. ............................................................................................................262 xii 10.17: Observed and expected elements for bovids from the Upper Level of Abri Cellier, MAU and MGUI. .......................................................................................264 10.18: Observed and expected elements for bovids from the Lower Level of Abri Cellier, MAU and MGUI. .......................................................................................266 10.19: Lengths of bone fragments for bovids from Abri Cellier, in millimeters. ............268 10.20: Observed and expected elements for red deer, MAU and MGUI for the Upper Level of Abri Cellier. ...................................................................................270 10.21: Observed and expected elements for red deer, MAU and MGUI for the Lower Level of Abri Cellier. ..................................................................................272 10:22: Bone fragment lengths for red deer elements from Abri Cellier, all levels. .........274 11.1: Bone and antler tools from four Early Aurignacian sites in southwest and southern France. ......................................................................................................295 11.2: Percentage of NISP for herbivores at four Early Aurignacian sites in southwest and southern France. ..............................................................................296 11.3: Awls and sources of tool supports from the Châtelperronian and Aurignacian levels at the Grotte du Renne. .................................................................................298 11.4: Sources of tool supports from the Upper and Lower Levels of the Aurignacian occupation at Abri Cellier, excluding antler. .....................................305 xiii LIST OF FIGURES Figure 5.1. Map showing the locations of the Grotte du Renne and Abri Cellier.........................70 7.1: Map showing the location of the Grotte du Renne. ....................................................99 7.2. Sketch showing the prehistoric caves at Arcy-sur-Cure and previous excavations..............................................................................................................100 7.3: Sketch of Level Xc of the Grotte du Renne, showing the location of the hut area, ash areas and the talus or porche. ...................................................................102 8.1: NISP by count and percentage for Level Xc. ...........................................................124 8.2: MNI by count and percentage of total for Level Xc. ................................................124 8.3: Graph showing the proportion of dry, fresh and undetermined breaks on bone fragments larger than 2.5cm in size. .......................................................................135 8.4: Chart showing the proportion of weathering present in the Level Xc faunal assemblage. .............................................................................................................138 8.5: Chart showing the percentage of chemical weathering on bone fragments from Level Xc. .................................................................................................................139 8.6: Bivariate plot of the MNE for reindeer by density value for Level Xc of the Grotte du Renne. .....................................................................................................140 8.7: Bivariate plot of MNE of horse against density value for Level Xc. .......................145 8.8: Proportions of damage by carnivores to bones in the Level Xc assemblage. ...........147 8.9: Chart showing proportions of damage to bones within the Level Xc assemblage by different agents. ..............................................................................147 8.10: Chart showing proportion of bones by taxon with evidence for gnawing. .............149 8.11: Chart showing proportions of reindeer bones with evidence of gnawing. .............149 8.12: Percentage of burnt and unburnt bone at the Grotte du Renne level Xc for bone fragments and esquilles. .................................................................................151 8.13: Graph showing the number and percentage of MNE per element for reindeer in Level Xc..............................................................................................................153 8.14: Graph showing number and percentage of MAU per element for reindeer in Level Xc. .................................................................................................................156 8.15: Graph showing the counts of total MNE and total NISP per element. ...................157 xiv 8.16: Appendicular elements of reindeer showing the median, mode and longest and shortest lengths in millimeters. ........................................................................158 8.17: Axial elements of reindeer, showing the mean, median, mode and longest and shortest lengths in millimeters. ...............................................................................159 8.18: Chart showing the proportion of dry, fresh and undetermined breaks by element for reindeer in Level Xc. ...........................................................................159 8.19: Graph showing counts of mandibular molar wear for reindeer in level Xc. ..........162 8.20: Graph showing counts of maxillary molar wear for reindeer in Level Xc. ............163 8.21: Graph showing the total MNE and total NISP per element for horse in Level Xc. ...........................................................................................................................168 8.22: Graph showing lengths of long bone fragments for horse in Level Xc, in millimeters. .............................................................................................................169 8.23: Chart showing the proportions of dry, fresh and undetermined breaks by element for horse in Level Xc.................................................................................170 8.24. Percentage of dry, fresh and undetermined breaks by element for cave bear in Level Xc. .................................................................................................................180 8. 25: Photograph showing fused hyena unciform and third carpal with associated bone growth. ...........................................................................................................181 8.26: Element selection strategy by Neanderthals for reindeer in Level Xc. ..................186 8.27: Element selection strategy by Neanderthals for horse in Level Xc. .......................187 8.28: Map of Level Xc showing the percentage of all identified bones by grid square. .....................................................................................................................192 8.29: Map of level Xc showing the percentage of unidentified bone fragments by grid square...............................................................................................................193 8.30: Distribution map of unburnt bone splinters (esquilles) by percentage. ..................194 8. 31: Distribution map of burnt bone splinters (esquilles)s by percentage.....................195 9.1: Map showing the location of Abri Cellier, Commune de Tursac, Dordogne. ..........199 10.1: Graph showing count and percentage of NISP by taxon for the Upper Level of Abri Cellier, held at the Logan Museum. ...........................................................217 10.2: Graph showing count and percentage of NISP by taxon for the Lower Level of Abri Cellier, held at the Logan Museum. ...........................................................217 10.3: Graph showing proportions of NISP by taxon for the Abri Cellier faunal assemblage held at the Logan Museum. .................................................................218 xv 10.4: Graph showing count and percentage of MNI by taxon for the Upper Level of Abri Cellier, held at the Logan Museum. ...............................................................220 10.5: Graph showing count and percentage of MNI by taxon for the Lower Level of Abri Cellier, held at the Logan Museum. ...............................................................222 10.6: Graph showing proportions of NMI by taxon for the Abri Cellier faunal assemblage held at the Logan Museum. .................................................................222 10.7: Graph showing intra-taxon variation between levels at Abri Cellier, where MNI is greater than 1, excluding cervidae. .............................................................223 10.8: Graph showing the proportion of dry, fresh, recent and undetermined breaks on long bone fragments in the Abri Cellier assemblage (all levels). ......................233 10.9: Graph showing the proportion of weathering for the Upper and Lower Level of Abri Cellier. ........................................................................................................235 10.10: Graph showing the proportions of vermiculated elements in the Upper and Lower Levels. .........................................................................................................236 10.11: Graph showing proportions of drawer wear for the Upper and Lower Levels. ....237 10.12: Bivariate plot of MNE of reindeer against density value for the Upper Level. ....238 10.13: Bivariate plot of MNE of reindeer against density value for the Lower Level ....238 10.14: Bivariate plot of MNE of horse against density value for the Upper Level. ........239 10.15: Bivariate plot of MNE of horse against density value for the Lower Level. .......240 10.16: Bivariate plot of MNE of bovids against density value for the Upper Level. ......240 10.17: Bivariate plot of MNE for red deer against density value for the Upper Level. ......................................................................................................................241 10.18: Graph showing the number of elements damaged by gnawing per taxon for the Upper Level. .....................................................................................................243 10.19: Graph showing the number of elements damaged by gnawing per taxon for the Lower Level. .....................................................................................................244 10.20: Graph showing carnivore damage patterns on elements in the Upper Level. ......244 10.21: Graph showing carnivore damage patterns on elements in the Lower Level .......245 10.22: Graph showing lengths of reindeer long bones, in millimeters, all levels. ...........252 10.23: Graph showing the proportion of dry, fresh and undetermined breaks by element for reindeer at Abri Cellier. .......................................................................254 10.24: Graph showing the long bone length for horse elements from both levels of Abri Cellier. ............................................................................................................261 xvi 10.25: Graph showing the proportion of dry, fresh and undetermined breaks by element for horse at Abri Cellier. ...........................................................................261 10.26: Graph showing the lengths of bovid appendicular elements in the assemblage, in millimeters......................................................................................268 10.27: Graph showing the mean, median, longest and shortest length for red deer bone. ........................................................................................................................274 10.28: Bivariate plot of MAU and MGUI values for reindeer from the Upper Level of Abri Cellier. ........................................................................................................279 10. 29: Bivariate plot of MAU and MGUI values for reindeer from the Lower Level of Abri Cellier. ........................................................................................................280 10 30: Graph showing wear stages for reindeer mandibular molars from both levels of Abri Cellier. ........................................................................................................281 11.1: Drawing showing the tool fragments and areas of polish, and their location on the left proximal tibia shaft. ....................................................................................299 11.2: Sketch of three scrapers made on unidentified mammal bone fragments 61.63.A6; 63.C9; A5. Actual size. ..........................................................................300 11.3: Detail of shaped horn core from the Lower Level of Abri Cellier. ........................303 11.4: Worked wolf ulna, from the lower level of Abri Cellier,showing rounded distal end and dry break. .........................................................................................304 A.1: Level Xc, Grotte du Renne. Cutmark locations on reindeer skull and vertebrae..................................................................................................................320 A.2: Level Xc, Grotte du Renne. Cutmark locations and impact marks on reindeer right humeri.............................................................................................................321 A.3: Level Xc, Grotte du Renne. Cutmark locations and impact marks on reindeer left humeri. ..............................................................................................................322 A.4: Level Xc, Grotte du Renne. Cutmark locations and impact marks on reindeer right radii and ulnae. ...............................................................................................323 A.5: Level Xc, Grotte du Renne. Cutmark locations and impact marks on reindeer left radii and ulnae. .................................................................................................324 A.6: Level Xc, Grotte du Renne. Cutmark locations and impact marks on reindeer carpals (right) and right metacarpals (left). ............................................................325 A.7: Level Xc, Grotte du Renne. Cutmark locations and impact marks on reindeer indetrminate metacarpals. .......................................................................................326 A.8: Level Xc, Grotte du Renne. Cutmark locations on reindeer right femora (left) and indeterminate femora (right). ...........................................................................327 xvii A.9: Level Xc, Grotte du Renne. Cutmark locations and impact marks on reindeer right tibia. ................................................................................................................328 A.10: Level Xc, Grotte du Renne. Cutmark locations and impact marks on reindeer left tibia. ..................................................................................................................329 A.11: Level Xc, Grotte du Renne. Cutmark locations and impact marks on reindeer tarsals. .....................................................................................................................330 A.12: Level Xc, Grotte du Renne. Cutmark locations on reindeer right metatarsals. .....331 A.13: Level Xc, Grotte du Renne. Cutmark locations on reindeer left metatarsals. .......332 A.14: Level Xc, Grotte du Renne. Cutmark and impact locations on reindeer indeterminate metatarsals. ......................................................................................333 A.15: Level Xc, Grotte du Renne. Cutmark locations and impact fractures on reindeer phalanges. .................................................................................................334 A.16: Level Xc, Grotte du Renne. Cutmark locations on horse right humerus. ..............335 A.17: Level Xc, Grotte du Renne. Cutmark locations on horse indeterminate radius. ...336 A.18: Level Xc, Grotte du Renne. Impact locations on horse right tibia. .......................337 A.19: Level Xc, Grotte du Renne. Impact locations on horse left tibia. ..........................338 A.20: Level Xc, Grotte du Renne. Cutmark and impact locations on horse indeterminate metapodials. .....................................................................................339 A.21: Level Xc, Grotte du Renne. Cutmarks and impact locations on bear indeterminate humeri. .............................................................................................340 A.22: Level Xc, Grotte du Renne. Cutmark and impact locations on bear indeterminate femora. .............................................................................................341 A.23: Level Xc, Grotte du Renne. Cutmark and impact locations on bear indeterminate tibia. .................................................................................................342 A.24: Level Xc, Grotte du Renne. Cutmark locations on bear phalanges. ......................343 A.25: Level Xc, Grotte du Renne. Cutmark and impact locations on hyena radius (right) and possible cutmarks on hyena fibula (left). ..............................................344 A.26: Level Xc, Grotte du Renne. Cutmark and impact locations on hyena phalanges and tarsals. .............................................................................................345 A.27: Level Xc, Grotte du Renne. Cutmark location on felid third phalange. ................346 A.28: Abri Cellier. Cutmark locations on reindeer atlas. ................................................347 A.29: Abri Cellier. Cutmark locations on reindeer humeri. ............................................348 A.30: Abri Cellier. Cutmark locations on reindeer right radius and metacarpal. ............349 xviii A.31: Abri Cellier. Cutmark locations on reindeer right femur and tibia ........................350 A.32: Abri Cellier. Cutmark locations on reindeer right metatarsal. ...............................351 xix 1 INTRODUCTION The Early Upper Palaeolithic in western Europe saw the extinction of the Neanderthals and their replacement by modern humans. Many of the explanations for what was probably an extremely complex process focus on particular aspects of Neanderthal behavior. This is in contrast to behavioral patterns of their modern counterparts which are perceived as more adaptive. The interpretations of Early Upper Palaeolithic culture speak to how modern palaeoanthropologists interpret the limited record of past human behavior that survives from this period. Lithics are abundant, but the amount of fauna varies in relation to local preservational biases. No perishable materials, such as wood or hide, survive. Instead we have to infer behaviors from what remains in the archaeological record. One area of debate has been the use and manufacture of bone tools by Neanderthals, a behavior argued by some to be the result of acculturation from modern humans and by others to be the result of independent innovation. This thesis examines the evidence for tool use, not through the examination of manufacturing techniques, but through an examination of the acquisition of the basic raw material (bone) as part of the overall subsistence strategy. The goal of this thesis is to examine if there are any substantial differences between Neanderthals in the Châtelperronian and modern humans in the Aurignacian in the selection of particular bones for the manufacture of tools; or if there are any difference in carcass transportation choices that might reflect the need for particular skeletal elements for tool manufacture. The appearance of bone tools in the Châtelperronian and Aurignacian in Europe represents evidence for the regular manufacture of items made from fragile materials such as hide, intestine or plant materials. This further implies the manufacture of containers for storage or transportation, the manufacture of portable shelter in the form of clothing, or the construction of tents or wind breaks. The degree to which either hominin 2 population invested in the use of these tools, and by inference the manufacture of containers large and small may have implications for the success of one population, and the retreat and extinction of the other. Alternatively, if there is no significant difference in the manufacture and use of bone tools, it may be that there was little difference in the use of clothing, containers and shelter by the two groups and we, as palaeoanthropologists, cannot make any definitive statements about the adaptiveness of one particular tool kit. In this thesis I examine the faunal assemblages from two sites: Level Xc of the Grotte du Renne, Arcy-sur-Cure, Yonne, France and the Aurignacian I and II occupations of Abri Cellier, Dordogne, France. The Level Xc assemblage is the fauna generated during the earliest Châtelperronian occupation at the Grotte du Renne. This level also produced a large assemblage of bone tools, primarily awls, that has been the focus of considerable debate in the literature. Abri Cellier contained two separate levels of occupation, a Lower Level (Aurignacian I) and Upper Level (Aurignacian II) that both produced a rich assemblage of bone and antler tools. The tool assemblages from both sites have been described and published in the literature. To date, there has been no examination of how the raw material was acquired. The examination of this question will proceed as follows: Organization The first chapter will discuss the issue of acculturation or innovation associated with the evidence of bone tool use by Neanderthals and modern humans in the early Upper Palaeolithic. In this chapter, the arguments for acculturation and independent innovation on the part of Neanderthals will be discussed and examined with reference to the archaeological record. The issue of what is meant by acculturation and the possible mechanisms of transmission of knowledge between populations will also be addressed. The mechanisms that underlie the processes by which acculturation occurs is an aspect of the debate that is not considered in the palaeoanthropological literature to any great 3 degree. This debate on the ability of Neanderthals to innovate (a presumed ‘modern’ behavior) will be related to the larger pattern of subsistence behaviors in the second chapter. The second chapter discusses the changing view of Neanderthals within the evolutionary history of modern humans, and how the views of Neanderthals derived from this debate and early interpretation of the skeletal data has continued to color our interpretation of one of our closest extinct relatives. This section of the thesis will also examine similarities and differences in lithic technology, symbolic behavior, subsistence, and landscape use. The greatest differences are in symbolic behavior, where Neanderthal symbolic actions (apart from burials) have to be inferred from other data sources, in contrast to the evidence of artwork, personal adornment and long distance social networks found in the Aurignacian. This leads to how we reconstruct past subsistence behaviors and life histories. The third chapter examines the use of modern ethnoarchaeological data that serves as models or proxies for Palaeolithic hunter-gatherer social and subsistence behavior. The allocation of labor, decision-making and food sharing by modern huntergatherers is discussed. While modern societies talk of a gendered division of labor, it is clear from the ethnographic literature that members of hunter-gatherer or pastoral societies see each member as an integral part of the subsistence system. Different actors may operate in different spheres, but this is in a complementary manner, not a compartmentalized system. Particular attention is paid to how children and young adults learn to hunt and forage, and the role that children play in fulfilling their subsistence requirements. Another aspect of subsistence behavior, the acquisition of raw material and the manufacture of clothing, will also be examined in this chapter. The following chapter will discuss how the data from the archaeological record, the data from modern ethnographic studies and Neanderthal lifeways can be combined to understand how Neanderthals met the nutritional requirements of offspring and evidence for group 4 provisioning. This will be related to evidence of the acquisition of effective foraging and hunting skills, the size of social groups and local ranges and territories, and the evidence for interactions with larger social networks. It appears that these differences, particularly social networks, may be a matter of degree rather than absolute differences. To understand one aspect of this difference, this thesis will examine how raw material for bone tools was obtained. Chapter five describes the research question, examines other studies of Upper Palaeolithic worked bone and presents the research hypotheses and testable models and the project data sets. Following a general discussion of faunal analysis with reference to taphonomic issues in chapter six, chapters seven through ten describe the history of excavation at the two sites. This includes the data sets, and present the results of the analysis in terms of subsistence behavior and the raw material chosen at each site for bone tools. Following the presentation of the datasets, chapter eleven provides an overview of bone-tool studies in archaeology and overview of bone tool use and selection in the ethnographic and archaeological record, with particular reference to the Early Upper Palaeolithic in France. The final chapter presents conclusions with reference to the research hypotheses and suggests some avenues for further research. 5 CHAPTER 1: WHAT DO PALAEOANTHROPOLOGISTS MEAN WHEN THEY SAY ACCULTURATION? Introduction The transition from the Middle to Upper Palaeolithic in Western Europe is marked by changes in lithic technology and the appearance of bone and antler tools in the European archaeological record (Mellars 1996). The degree of contact or cultural exchange between the two human species in Europe (Neanderthals and modern humans) remains a matter of controversy. The role of material culture as an expression of group identity, the integration of new technology, evidence for or against cultural exchange and creation of new cultures and cultural forms plays a significant role in this debate (Tolmie, in press). The interpretations of Early Upper Palaeolithic culture also speak to how modern anthropologists create and interpret both ancient cultures and our own, in terms of how and what we perceive to be typical “human” behavior. What is significantly lacking is the theoretical consideration of culture and ethnogenesis by archaeologists and palaeoanthropologists who study this fascinating time period. The debate for and against acculturation corresponds to a larger debate regarding the distinctiveness or uniqueness of modern human behavior. Discussions of Middle Palaeolithic and Upper Palaeolithic behavior have been framed in oppositional terms related to the two human species (Finlayson 2004; Mellars 1996; Stringer and Gamble 1993; Trinkaus and Shipman 1993a). This in turn relates to our perceptions of what behavior represents “modern”, how we view ourselves and our own use of culture. Lack of artwork and distinctive regional variation has been argued to show non-modern behavior by Neanderthals, although there is little evidence for such behavior by contemporary modern humans (Chase and Dibble 1987; Gamble 1999; Henshilwood and Marean 2003; White and Knecht 1992). The interpretation of Mousterian culture as less complex than later Upper Palaeolithic cultures reflects an unfounded assumption that 6 complex behavior is mirrored in complex material culture and is informed by unverified assumptions about Neanderthals capacity for “modern” or “human” behavior (Speth 2004). Of particular significance are lithic manufacturing systems that focus on the production of blades to make tools, and the introduction of worked bone (osseous) technology. The Aurignacian, like other early Upper Palaeolithic lithic industries, shows a shift in core reduction techniques to a volumetric approach that produces blades of a consistent length and width (Shea 2007). However the association of blade dominated technologies with modern humans is uncritical and does not examine the socio-economic context of stone tool production (Bar Yosef and Kuhn 1999; Chase 2007; Clark 2002; Gamble 2007). In the preceding Middle Palaeolithic, both Neanderthals and modern (or near-modern) humans shared the same lithic technology in Europe and Western Asia: the Mousterian (Marks 1988; Shea 1998). The standardization of tools over a wide geographical and temporal range is interpreted as a conservative, unchanging culture. If stone tools served as important markers of social group identity (Farizy 1990a) the similarity of toolkits over a wide area could reflect a sparse population signaling membership of a larger community. Expression of similarity through the use of lithic technology may have been more important (and adaptive) than expression of difference. Archaeological cultures and typologies The analytical methods used to define Middle Palaeolithic Mousterian assemblages and Upper Palaeolithic assemblages are derived from methodologies that were developed in the caves and rockshelters of southwestern France (Bisson 2000). These are based largely on a geological-style classification system designed to create a chronological framework of archaeological assemblages referred to as facies, an approach that emphasizes the role of type fossils in defining cultural assemblages (Teyssandier 2007). The system works well for Lower Palaeolithic and Upper 7 Palaeolithic assemblages. In contrast, Middle Palaeolithic Mousterian assemblages lack type fossils but are organized into five traditions. These have been interpreted as expressions of ethnicity (Bordes 1972), different toolkits reflecting different subsistence activities (Binford and Binford 1966) or the end products of tool use and tool repair (Dibble 1987). Is the Mousterian the cultural monolith envisioned by archaeologists? Bisson has discussed this at length (Bisson 2000). Given that typologies are cultural artifacts, how do they help us understand past prehistoric behavior? While the five traditions approach creates a convenient shorthand, the Bordesian system does not allow for incorporation of a wide range of lithic material, and masks possible diversity. Nor does it factor in consideration of tool retouch (c.f. Dibble 1987), use wear, site function or local ecology (Bisson 2000). Lithic analysts have noted regional patterns; analysts argue that these distinct regional traditions correlate with interactions between band members that occupy particular areas (Burke 2006; White 2006; White and Pettitt 2011). Regional variation in Mousterian assemblages may correspond to expression of band membership by both Neanderthals and modern humans within a larger cultural tradition (Shea 2007). The use of typological schemes and type fossils throughout Europe has homogenized the material culture of the Palaeolithic and obscured regional variation in lithic assemblages that may reflect local cultural expression. Examination of reduction sequences is more productive in understanding how the tools were produced and used in ways that reflect cultural or social organization. Lithic reduction sequences in the Levant associated with the two species show distinctly different practices of core reduction, although the finished tools are similar in morphology (Shea 2007). The typological approach therefore results in equifinality where different processes result in similar morphological shapes, obscuring differences in tool production that may reflect culturally appropriate behavior. 8 Transitions, time, and technological innovation Traditionally, the archaeological record has been subdivided into different temporal and cultural subsets with apparently abrupt boundaries. In the Pleistocene, this is a product of the long-term focus on deeply stratified cave sites to determine chronologies and the reliance of fossiles directeurs as chronological markers. Realistically, archaeologists have to rely on some form of consistent markers to determine change over time, be they single distinctive tool-types, such as the hand axe or split based bone point, or a suite of traits, such as the Mousterian MTA. What remains problematic is how we view the apparent speed of change. As Gamble (2007) has noted, the analogies of recent history (the Industrial and Agricultural Revolutions or political revolutions) may not be suitable for explaining changes in the extreme past. How much of the “Human Revolution” as posited by Mellars, and also by Klein, is a product of sudden behavioral shifts, or an apparent reorganization of the neocortex? The concept of a ‘revolution’ in modern human behavior has been questioned by archaeologists (Kaufman 2002; McBrearty 2000). Does identifying the “earliest” example of any given technology really show a shift in behavior, or should we, as archaeologists, examine why some technologies become widespread and some appear and disappear throughout the Pleistocene, such as blade tool technology (Bar Yosef and Kuhn 1999). In Europe, the ‘transitional’ industries including the Châtelperronian, Uluzzian, Szeletian and Bachokirian, were originally interpreted as precursors of the Aurignacian based on the presence of tools made on blades, not flakes. As the Aurignacian was thought to be produced by modern humans, these earlier industries were also thought to be produced by modern humans. The discovery of Neanderthal remains in Châtelperronian contexts called these assumptions into question (Bailey and Hublin 2008; Lévèque et al. 1993). Interpretation of the Châtelperronian and central European and Italian facies thought to be produced by Neanderthals focuses on the evidence for acculturation or 9 indigenous development (Harrold 2002; Hublin 2000; Kuhn and Bietti 2000; Mellars 1999; de Quiros et al. 2001). Proponents of acculturation argue that technological innovation and a limited amount of symbolic behavior result from culture contact via diffusion or acculturation through direct contact (Davies 2007; Harrold and Otte 2001; Mellars and Gravina 2008). There is little or no consideration of the mechanisms of cultural transmission (Chase 2007; Eerkens and Lipo 2007), or whether acculturation, sensu strictu, is even an appropriate description for the mechanisms that underlie these developments. Other palaeoanthropologists argue that these new lithic and osseous technologies are an indigenous development (d’Errico et al. 1998; Zilhão and d’Errico 1999; Zilhão et al. 2006; Zilhão et al. 2008a; Zilhão et al. 2008b), because there is no evidence for interaction or contact between the two species. The underlying implication for this hypothesis is that these new cultures are an expression of cultural development related to changes in local socio-economic conditions. This hypothesis does not explicitly consider how material culture is embedded in social production of identity. In Upper Palaeolithic Europe, the appearance of personal adornment and art is viewed as evidence for evolution of social systems and symbolic expression associated with modern humans, implying less developed social structures and notions of self and group for Neanderthals (Gamble 1999; Mithen 2005). This conveniently ignores the absence of art and adornment in other assemblages associated with early modern humans (Speth 2004) and the operation of some material culture as an expression of identity and group membership that is not limited to personal adornment (Wiessner 1983; Wobst 1977). The debate over the origin of the Châtelperronian and other transitional cultures is embedded in a larger discussion of how culture and technology produced the extinction of the Neanderthals, represented by the spread of the Aurignacian. Acculturation proponents argue that Neanderthals were out-competed by technologically advanced colonists and that change in technology indicates ‘imitation’ of modern human behavior. 10 In contrast, proponents of indigenous innovation argue that both groups were equally well-adapted in terms of technology and other factors resulted in the extinction of Neanderthals. The environment was equally hostile to modern humans, and the early Aurignacian reflects modern human responses to similar socio-environmental stress. Ultimately the cause of Neanderthal extinction is probably far more complex that currently recognized (Stringer 2008). Nor is the archaeological record clear that modern humans, who are presumed to have produced the Aurignacian, were any more successful in colonizing an increasingly hostile environment (Bradtmöller et al. 2012). The issue(s) of acculturation The proponents of acculturation do not address the mechanisms by which acculturation occurs. The Aurignacian is envisioned as a single new technocomplex that has aspects that contemporary Neanderthal societies chose to (or had to) emulate through adoption of new lithic and osseous technologies in order to compete with modern or nearmodern humans. It is interesting that the most distinctive artifact (in archaeological terms), the antler point, is not produced in non-Aurignacian technologies, nor is antler a common raw material. As Cusik (1998:135) has noted, archaeologists tend to confuse changes in behavior with changes in identity, with changes in material culture over time equated with acculturation (Friemand 2009). Archaeologists tend to be apocalyptic when considering culture change, with a focus on rupture and change, as evidenced in the use of the term “revolution” (Gamble 2007:43; Kuhn 2012). There is also a tendency to view acculturation as a directional process, based on constant contact (Dohrenwend and Smith 1962). In the acculturation model proposed and maintained by Mellars, there is also the implication of cultural and technological superiority (i.e. dominance) on the part of modern humans, an implication that could almost be termed colonial or dominant in its approach. 11 Modern acculturation studies and identification of acculturation in the archaeological record suggest differences in social relations where subject groups react to the dominant group. Individuals and communities are actors in processes of integration, assimilation, separation and marginalization in relation to other socio-cultural networks that may be larger or more dominant (Berry 2003; Padilla 1980). While acculturation can lead to culture change, it is a reactive process, whereby the dominant culture can be rejected by increasing traditionalism (an explanation for the ‘Ebro frontier’ that no-one had used, to my knowledge). The degree or intensity of contact is a major factor in how acculturation operates between dominant and minority cultures (Berry 2003:23). The classic formulation of acculturation is “those phenomena which result when groups of individuals having different cultures come into continuous contact” (Dohrenwend and Smith 1962:30) (my emphasis). The process is also influenced by the contact situation (constant, sporadic, infrequent, imposed, or resisted), the size of the two populations and the conditions of contact which can range from extreme dominance to parity, where either culture can exclude the other. The response to acculturation can result in assimilation, where many changes occur on an individual or group level; integration, where valued features of a culture are retained and other features selectively adopted; or separation, where the fewest changes occur in the minority culture. If acculturation is occurring in the Châtelperronian, the best fit for the archaeological data would be a form of integration, where new technologies or forms of communication are adopted into existing Neanderthal culture. There is less change in lithic technology (the Châtelperronian clearly has roots in Mousterian lithic technology), and new hunting armatures (antler points) are not adopted. New technology to manufacture hide or vegetal containers, clothing or shelter does appear, as do new forms of personal ornament. As a corollary, this new technology implies that the groups also adopted new forms of manufacture, or significantly increased or improved existing forms of container or shelter production. 12 Assimilation is a model that may not apply in egalitarian systems, such as those inferred for the Châtelperronian and Aurignacian (Shortman and Urban 1998). Under the Aurignacian acculturation hypothesis, new technology implies regular contact and is unidirectional, although between groups of similar complexity the interaction would be bilateral and therefore bicultural (Schuyler 1998). Ethnoarchaeological studies among modern human demonstrate that interaction does not result in wholesale replacement of material culture, even where groups of different social or technological complexity are in regular contact. Adoption of new technology can vary greatly between neighboring groups. For example, while central Californian groups such as the Pomo and Miwok adopted metal awls for basketry in the nineteenth century, the Western Mono retained bone awls, despite the wide availability of metal. An informant noted that they preferred the sound that bone made when pushed into a basket coil (Meighan 1953). Modern hunter-gatherers operate within larger economic frameworks and ultimately the larger world system. They remain clear in their identities as parts of particular groups based on shared language, ideology, ritual and traditions (Silberbauer 1996). The incorporation of new materials and technology reflects culturally appropriate behavior, not an inferred desire to emulate neighboring cultures. An analogous situation of new populations, population displacement and population replacement occurs in the protohistoric period in North America after 1492. The archaeological record of Native American sites in the Midwestern United States demonstrate that indigenous groups maintained or created identity during periods of social upheaval as populations were displaced westwards; economic systems were altered as the Midwest became incorporated into the fur trade; political change occurred as European political systems and alliances were established, and European diseases impacted indigenous populations. Material culture shows the incorporation of trade items and the adaptation of indigenous material culture to new economic patterns (Ehrhardt, 2009; Griffitts 2009), where new material culture was adopted for new activities (Rogers 13 1993). At initial contact, the adoption of new technologies was mediated by existing social structures and institutions (Wilson and Rogers 1993). Later, acculturation was a coercive process enforced on indigenous groups by the colonial powers, even though indigenous groups succeeded in maintaining distinct cultural identities and traditions. The transmission of traditional values and culture to children through the family group appears an important factor (Bruner 1956). These studies all reflect the behaviors of modern humans who share the same biological, cognitive and social capacities. Debate continues regarding the extent to which Neanderthals shared these capacities. The proponents of acculturation tacitly imply that even if Neanderthals did not have the full suite of modern capacities, they were similar enough in behavior to be able to select and emulate aspects of modern human culture within their own social systems. Discussions of acculturation in the literature reviewed for the Châtelperronian and Aurignacian do not address any transfer mechanisms, nor do the authors specifically state how they define culture – a single “magic ingredient” such as symbolic behavior, or as a continuum of behaviors (Byrne 2004:341). As human culture is extremely complex, it can be described in terms of social learning and conformity, as a sign of cognitive complexity, a means of transmitting knowledge, and of course as a physical product. The chaîne opératoire is a result of the interaction of culturally appropriate gestures and the physical world. The arrival of new technology from external sources implies the development or adoption of associated manufacturing processes. Acculturation, transculturation or ethnogenesis What direct evidence do the so-called “transitional” cultures of the Early Upper Palaeolithic show for acculturation, transculturation or ethnogenesis? For acculturation to occur, some form of regular contact is required as well as the integration of culturally appropriate new behavior. In the more recent past, contact between cultures is expressed 14 in the archaeological record by the appearance of artifacts that are derived from an exogenous source (frequently trade items). Evidence for contemporaneous occupation of a geographic region by both cultures as proxies for species, or Châtelperronian material in good Aurignacian context or vice-versa would be direct evidence for contact. To date, there is no unequivocal evidence for such contact in Western Europe. Mitochondrial DNA studies have demonstrated that a few modern humans and Neanderthals came into extremely close contact, as modern Eurasians retains a small percentage of the Neanderthal genome within their DNA. This contact likely occurred in western Asia during the Middle Palaeolithic, prior to any expansion by modern humans into Europe (Green et al. 2010). Unfortunately, this contact also occurred prior to the development of the Aurignacian and transitional Neanderthal cultures, therefore it is not possible to infer any patterns of cultural as opposed to genetic exchanges. The archaeological data do not demonstrate any clear evidence for direct contact between the two species. In Eastern and Central Europe, all “transitional” material underlies the earliest Aurignacian material (Hoffecker et al. 2008). In France, three rockshelters (Grotte des Fées, Roc de Combe and Le Piage) are argued to contain interstratified Aurignacian and Châtelperronian layers, implying contemporaneity and, therefore, contact although the evidence is highly equivocal (Bordes 2003; Mellars and Gravina 2007, 2008; Zilhão et al. 2006, 2008a and b). Is interstratification direct evidence for contemporaneity, if it is even present? Cave sediments are the product of long term depositional processes. Interstratification (if it exists) could simply reflect expansion of one group into empty space abandoned by the other, as has been argued for Neanderthal and modern human occupations in the Near East. Direct evidence for contemporaneous occupation is lacking. Radiocarbon dating is at the extreme limit of viability and fluctuations in atmospheric carbon produce large 2-sigma error ranges which make any statements about contemporary occupations problematic (Blockley et al. 2008; Pettit and Pike 2001; Roebroeks 2008). As Kuhn (2012) had noted, there are also issues of scale in 15 considering the Middle to Upper Palaeolithic transition. The appearance of the Aurignacian is at a very large scale that reflects centuries of change, yet archaeological explanations (migration, diffusion or social upheaval) are local and short-term. When considered at the local level, archaeological sites, and cultures (as defined by archaeologists) are the results of “countless decisions made by individuals…following agendas that might have been quite divergent, and that certainly had little to do with creating and maintaining what we perceive as the Mousterian, the Magdalenian or the Aurignacian” (Kuhn 2012:3). Change and continuity vary according to the aspect of the archaeological record examined. For example, continuity occurs across the Upper: Middle Palaeolithic transition in faunal exploitation. In terms of geographic contemporaneity, occupation of the same continent by small populations separated by significant topographic barriers to movement does not imply a degree of contact that would result in acculturation. While some archaeologists have argued for a “Bow Wave” model, where information gained by a group in direct contact is then transmitted through a wide network (Tostevin 2007), there is no archaeological evidence of any cultural material moving at the same time. As will be discussed in Chapter 2, there is little long distance (over 100 km) movement of lithic raw material in Middle Palaeolithic and some transitional cultures. While genetic evidence clearly shows some contact between Neanderthals and modern humans, the small percentage of shared genetic material does not indicate any long-term or large-scale evidence for direct contact. Independent innovations Given that typological studies obscure the potential variation within the archaeological record for this period, and ignore the problem of equifinality, a more productive approach would be to examine how the bone tools are made, used and discarded (the chaîne opératoire). If Aurignacian and Châtelperronian osseous tools have 16 similar chaînes opératoire, this would imply a transfer of knowledge. Examination of bone awls from Châtelperronian and Aurignacian levels at the Grotte du Renne, Arcysur-Cure, indicate different methods of production (d’Errico et al. 2003). The tools are from clearly defined stratigraphic levels separated by a sterile layer, although some argue for post depositional reworking of sediments (Bar Yosef 2007; Higham et al. 2010). Châtelperronian manufacturing techniques are sophisticated and indicate a complex, well established bone technology. Some awls are also decorated, indicating expression of social value. Aurignacian awls demonstrate a different manufacturing technique, using grinding to shape the tool instead of shaving the tool to a point. The use sequence is also different, indicating a separate technological tradition. Based on the evidence for awl production at the Grotte du Renne, it appears that bone working in the Châtelperronian is an autochthonous development, used for practical purposes and also used to express cultural information. This independent development mirrors the apparent independent development of blade manufacturing technology in Châtelperronian, Uluzzian and Seletzian assemblages. All three are derived from the final local Middle Palaeolithic Mousterian tradition (Churchill and Smith 2000; Kozlowski 2007). While the Aurignacian is the product of modern humans, recent analyses have suggested that it is a lithic tradition that developed in eastern or central Europe and then spread both east and west (Belfer Cohen and Goring-Morris 2007; Davies 2007; Kozlowski 2007; Svoboda 2007). The Aurignacian appears to be response to stresses on the modern human population in Europe that occurs at approximately the same time as new technologies emerge in the indigenous Neanderthal population. Interestingly, the appearance of split-based points post-date changes in lithic production. So who, if anyone, is influencing whom? The archaeological data support the argument for independent adoption of worked bone technology by both Neanderthals and modern humans. These two groups of humans developed new technologies as a response to socio-climatic conditions. If there is no 17 evidence for acculturation, how can we explain the emergence of worked bone technology in various areas of Europe at the transition from the Middle to Upper Palaeolithic? The technological innovations reflect a change in subsistence strategy as a response to the onset of the last major glacial episode. It may also it express increasingly strong regional identities that begin to develop in the Mousterian which reinforced social relations among group members. Climatic reconstructions show a retreat of woodlands into more temperate areas and an associated spread of open grasslands (Finlayson 2004). The increasing instability and unpredictability of resources could have resulted in elaboration of material culture that reflected shifts in social organization to reinforce group ties and connections with other bands, and/or intensification of subsistence behavior. The appearance of a variety of new lithic traditions, combined with innovations in osseous tool and ornament production, suggest a response to complex external pressures: demographic, economic, social or environmental. If two populations shared a similar lithic tradition and technology (and implied social and cultural organization) were exposed to similar socio-environmental pressures in different regions of Europe, is it surprising that they responded in a similar manner independently? Wiessner (1983) has demonstrated that formal variation in material culture transmits information about personal and social identity, reflecting group membership. Individuals arrive at a number of styles independently; therefore diffusion of style within the archaeological record may not depend on the degree of contact between different regions (Friemand 2009). The adoption of bone working in the Châtelperronian by Neanderthals reflects a combination of social, economic and environmental factors. As the major big game predators of Europe, they would have been familiar with the mechanical properties of bone from butchering carcasses of large animals. The use of bone tools may indicate increasing efficiency in the production of clothing, shelter or containers using hides and/or cordage, which may be a response to less predictable resource availability 18 resulting from a less stable environment. At the same time, the production of this technology, particularly items interpreted as decorative, indicate an increased need to signal social relations and group membership. In the Middle Palaeolithic, neither Neanderthal nor modern human populations show any strong evidence for symbolic behavior. As both populations experienced stresses related to periods of environmental flux, the use of decorative material and cultural expression encoded into tools would mediate tensions and assist in establishing intra-specific social networks in new alignments. Conclusion The archaeological record does not support the hypothesis that the development of bone tool working in Europe is the product of acculturation of Neanderthals by modern humans. While it is assumed to occur, there is no theoretical consideration of why it should occur. Other changes in material culture associated with the development of the Châtelperronian are autochthonous - derived from the preceding Middle Palaeolithic lithic production sequences. In the early Upper Palaeolithic there was clearly a change in technological and social organization among both species of hominins in Eurasia. While proponents of acculturation have argued that modern humans introduced new boneworking technologies that were adopted by the indigenous populations, this ignores the usual imbalance in power relationships that occur with acculturation, where the less powerful group adopts part or all of the dominant culture. There is no evidence that culture contact (also vital to acculturation) even occurred, although there is evidence to suggest that the two groups did interact during the Middle Palaeolithic in western Asia. By the Early Upper Palaeolithic, data from the archaeological record suggest that both species were using technology to create new forms of cultural expression, most likely in the face of ecological change and greater climatic instability. 19 I believe that it would be more fruitful to examine the development of the regional cultures of the Early Upper Palaeolithic, including the Early Aurignacian, as expressions of regional identity, possibly ethnogenesis that reflects an increasing need to express cultural identity and group membership. Indeed, the creation of a group identity may have operated to restrict mating networks to particular regions, as the shared Mousterian culture became fragmented. Instead of seeking to explain the disappearance of the Neanderthals in relation to the presence of another human species, we would do better to examine how Neanderthals interacted with each other and their environment, and how they adopted new tools and technology into their existing cultural practices. The causes of Neanderthal extinction were probably far more complex than we understand at present. By examining Neanderthal behavior in its own terms, without arguing for or against adaptive fitness or technological superiority, we will come to a better understanding of their subsistence and social practice. In this thesis I examine one aspect of Neanderthal cultural practice, namely the selection and use of skeletal elements for bone tools. This new technology emerges among the last Neanderthal populations of Europe, and is most vividly expressed in the Châtelperronian levels of the Grotte du Renne at Arcy-sur-Cure. The lowest level, Level Xc, has produced a relatively rich assemblage of bone awls and thin bone “pins” in addition to items of personal adornment in the form of pendants. The question arises as to how or if Neanderthals had to modify their carcass transportation and processing practices to obtain suitable supports for bone tools. Alternatively, the occupants of Level Xc may simply have used items transported into the site for the primary purpose of butchery and the consumption of fat and meat. The same questions can be asked of the modern humans associated with the Aurignacian culture at the site of Abri Cellier. If both hominins are pragmatically selecting from the bones available through hunting, it could be argued that there is little difference in the underlying provisioning of supplies for bone tools. Therefore, the presence or absence of bone tools in the Early Upper Palaeolithic 20 does not point to any major behavioral differences in the methods of raw material acquisition, and we cannot make any inferences regarding the ‘modernity’ or otherwise of Neanderthals. In the next chapter we will consider the evidence for similarities and differences in general in the behavior of the two groups of hominins in terms of lithic manufacture, symbolic behavior and, particularly, the exploitation of animals for subsistence purposes. The interpretation of these behaviors has been, and is still used, to support arguments for and against behavioral modernity and the ability or otherwise to innovate. This ability to react to changing ecological circumstances is frequently cited in discussion of the ‘fate of the Neanderthals.’ 21 CHAPTER 2: LITHICS AND HUNTING-SIMILARITIES AND DIFFERENCES Introduction The debate about acculturation and adoption of new technology is intertwined with the debate concerning the ‘fate of the Neanderthals’ in western Europe, which remains problematic. It is generally agreed that Neanderthals were replaced by anatomically modern humans who migrated westwards across Europe (Howell 1999). The factors underlying replacement remain unresolved and speak to basic assumptions that we, as anthropologists, make about modern human behavior. At present, debate continues on the differences and similarities between Neanderthals and anatomically modern humans in intellectual capacity and social organization, inferred from lithic technology, symbolic behavior and subsistence practices. In the early 21st century it is becoming apparent that the reconstruction of Neanderthal behavior shows a species capable of a broad range of flexible behaviors. It is also becoming apparent that the full picture of Neanderthal behavior requires a broad range of macroscopic and microscopic studies of lithics, fauna, sediments and physical remains. Further, palaeoanthropologists are still seeking a single “prime mover” as the cause of Neanderthal extinction and this remains framed in an almost nineteenth-century paradigm of competition with a betteradapted adversary, namely our modern human ancestors. The role of Homo neanderthalensis in human evolution has been a matter of debate since the recognition of the taxon (Eisley 1957; Mellars 2000a; Trinkaus and Shipman 1993b). Initial analyses of the skeletal material and lithics depicted the species as intermediate between apes and humans, with limited intellectual and technological capabilities (Boule 1911-1913; Hrdlička 1929; Morant 1927; Weidenreich 1943). Perceptions of the “humanity” of Neanderthals changed as new data became available. Boule’s analysis was a product, in part, of his perception of human evolution as non- 22 linear. In this paradigm, Neanderthals represented an extinct hominin that was not ancestral to modern humans, and Boule magnified differences and downplayed similarities in skeletal morphology (Trinkaus and Shipman 1993). One influential paper argued that Boule had misinterpreted pathological alterations to reconstruct Neanderthal posture (Straus and Cave 1957); but Boule had recognized and corrected for these pathologies (Trinkaus and Shipman 1993). Reanalysis indicated that Neanderthals had similar post-crania to modern humans, albeit considerably more muscular and robust. Neanderthal post-crania also apparently showed more evidence for skeletal stress, interpreted as a result of hunting behavior (Berger and Trinkaus 1995). Skeletal differences were interpreted as a response to selective pressures that result in similar anatomical or kinesiological movement when compared to recent modern humans (Trinkaus 1983, 1987; Trinkaus, et al. 1991). More recent studies which compare Neanderthal physiology with early modern humans show little or no difference in post-cranial robusticity (Estabrook 2009; Trinkaus 2012) indicating little difference in hunting behaviors or mobility patterns. The most direct evidence for Neanderthals as a separate hominin lineage is derived from genetic studies. Genetic evidence has demonstrated that anatomically modern humans evolved in Africa and then spread to other continents (Stoneking and Cann 1989). These were used to support morphological evidence for lack of contact between Neanderthal and early modern humans (Howell 1999; Pearson 2000; Stringer and Andrews 1988; Weaver and Roseman 2005). DNA extracted from Neanderthal fossils has produced sequences that differ from both early modern and recent modern humans (Green, et al. 2006; Krings, et al. 1997; Serre, et al. 2004), indicating a separate evolutionary pathway for some period of time. Studies indicate that local environmental pressures resulted in similar evolutionary variation, such as selection for fairer skin (i.e. convergent evolution), although operating on a different segment of DNA than modern populations (Lalueza-Fox, et al. 2007). 23 The most recent genetic studies of modern populations in Europe and Asia indicate that although Neanderthals were not ancestral to modern humans, the two hominin species shared a relatively recent common ancestor. Neanderthals and modern humans were genetically close enough to interbreed, resulting in a small percentage of Neanderthal DNA surviving in modern Eurasian populations (Currat and Excoffier 2011; Green, et al. 2010). This interaction probably occurred in western Asia. The evidence for interbreeding indicates that the two populations at that time and place were similar enough in terms of behavior and appearance to consider each other suitable mates. Studies of the material culture of Neanderthals and their contemporary modern counterparts in the Middle Palaeolithic and early Upper Palaeolithic show similarities in lithic tool production and subsistence strategies. Lithic technology Neanderthals and contemporary modern human groups had similar lithic technologies in the Middle Palaeolithic. Mousterian lithic assemblages occur throughout Europe and Western Asia. Mousterian sites in the Near East cannot be ascribed to Neanderthals or anatomically modern humans based on technology alone: both groups utilized similar stone tool working technology and toolkits producing both flake tools and Levallois industries (Marks 1988; Shea 1998). The standardization of the tools over a wide geographical and temporal range has been interpreted as reflecting a conservative, unchanging culture. The lack of distinctive regional variation (and artwork) has been argued to show a lack of symbolic behavior (Chase and Dibble 1987; Clarke and Lindley 1991; Gamble 1999; Henshilwood and Marean 2003; White 1995). If stone tools served as such important markers of social group identity (Farizy 1990a), the similarity of toolkits over a wide area could reflect a sparse population signaling membership of a larger community (cf. Rowley Conwy 2001; Torrence 2001). Expressions of similarity 24 through the use of lithic technology may have been more important (and adaptive) than expression of difference. Early Upper Palaeolithic industries in Europe are characterized by the production of blades to make tools, and debate continues regarding autochthonous and allochthonous development, diffusion and acculturation coincident with the migration of anatomically modern humans into Europe. The earliest Upper Palaeolithic lithic culture in France is the Châtelperronian. This industry was assumed to be a precursor of the Aurignacian and produced by anatomically modern humans, until the discovery at St. Césaire in 1979 of a Neanderthal burial in direct association with a Châtelperronian assemblage (Lévèque, et al. 1993). Many studies of material culture remain influenced by an uncritical evolutionary framework that equates variation in lithic technology with evolutionary development; particularly the association of blade dominated technologies with modern humans, rather than examining the context of tool production (Bar Yosef and Kuhn 1999; Clark 2002; Cosgrove and Pike-Tay 2004; Kuhn 2011). Interpretation of the Châtelperronian (and central European and Italian facies also thought to be produced by Neanderthals) now focuses on the evidence for or against acculturation, both in the lithic and worked bone technologies (de Quiros, et al. 2001; Harrold 1988, 2002; Hublin 2000; Kuhn and Stiner 2000; Mellars 2000b; White and Knecht 1992). Some analysts argue for an indigenous development of blade technologies and bone tools (d’Errico, et al. 1998; Zilhão and d'Errico 1999; Zilhao, et al. 2008 a and b; Zilhão, et al. 2006), whereas others argue that the limited amount of personal adornment may result from culture contact via diffusion or acculturation (Harrold 1988; Harrold and Otte 2001; Mellars and Gravina 2007, 2008), albeit with little discussion of the mechanisms of cultural transmission (cf. Eerkens and Lipo 2007). 25 Symbolic behavior Inferred symbolic behavior by Neanderthals shows some similarities with anatomically modern humans. The presence of burials suggests symbolic behavior similar to modern humans with reference to the dead, but grave goods are absent. Although the presence of Neanderthal burials has been questioned (Gargett 1989); palaeoanthropologists generally agree that evidence for Neanderthal burial is present in Europe and southwest Asia (Gamble 1999; Mellars 1996; Riel-Salvatore and Clark 2001). Neanderthal burials are infrequent, and very simple, and some may in fact represent fortuitous survival of corpses rather than deliberate burial (e.g. Sandgathe, et al. 2011). Less apparent is any symbolic behavior related to personal adornment or expression of group membership (Chase and Dibble 1987). The considerable amount of ochre, including possible crayons, found in Mousterian sites has been used as evidence for body decorations, as seen in many recent reconstructions of Neanderthals. Direct evidence is rare: ochre covered shells from Cueva Antón and Cueva de los Aviones, Murcia, southeastern Spain (Zilhao et al. 2010) and, recently, the identification of bird elements that provided feathers at the site of Fumane in northern Italy (Peresani et al. 2011) and in the Mousterian at the Grotte du Bison,Yonne, France (Cecile Moirer pers. comm.). Raptor claws were also collected at Pech de l’Azé 1 (Rendu 2010:1807). Additional studies show selection of diurnal raptors, and a preference for corvidae feathers (Finlayson, et al. 2012; Morin and Laroulandie 2012). Archaeological evidence for the use of bone and ivory as raw material for tools and adornment first appears in the Châtelperronian (d’Errico, et al. 1998; David and Poulain 1990; Julien, et al. 2002). Antler is not used for tools until the Aurignacian (Morin 2004; Tartar, et al. 2006; White 1998). 26 Subsistence Exploitation of animals for subsistence shows a similar suite of behaviors by both groups of hominins. Early Upper Palaeolithic faunal exploitation was based on an encounter strategy with locally available herd animals (Bar-Oz, et al. 2004; Enloe 1993; Morin 2004; Pike-Tay 1993; Simek and Snyder 1988; Steele 2002). In general, Late Middle Palaeolithic and Early Upper Palaeolithic hunting strategies focused on the most abundant seasonally and locally available species: for example bison and goats in the Caucasus (Bar Oz et al. 2004), and horse and reindeer in southwest France (Delpeche 1993; Morin 2004; Peterkin 2001). It should be noted that examination of faunal assemblages in terms of subsistence behavior(s) did not form a significant part of Palaeolithic research until the 1980s (Mellars 1996:202). Earlier studies focused on the proportions of large animals present to determine local vegetation and environmental conditions. Biostratigraphic assemblages, chiefly of large mammals, were used with lithic fossiles directeurs, palynological, and geomorphological data to create a relative chronology by linking fauna, lithics, and geomorphology to European glacial sequences. These studies focused solely on the proportions of fauna present, with little or no discussion of subsistence practices or butchery patterns (e.g. Delpeche 1983; Gaudelli and Laville 1990). These studies also assumed (usually an unstated assumption) that the fauna present were in direct proportion to the abundance of particular species in the region. More recent studies utilize carnivore den assemblages and radiometric dating to examine environmental change or to examine species distribution across regions over time (Boyle 1990; Raynal and Gaudelli 1990). The use of microfauna to reconstruct local environment had also provided a more nuanced explication of local and regional biogeography and climate (Marquet 1993). The changes in approach to environmental reconstruction reflect the development of zooarchaeology as a discipline. As Boyle notes, until the mid-1970s faunal analysis was less important than lithic analysis in French Palaeolithic studies (Boyle 1990:16, 21). 27 Debate over subsistence behavior initially concerned zooarchaeological evidence for hunting or scavenging (Marean and Assefa 1999), with the overt assumption that the latter process was less efficient or adaptive, with the implicit assumption that it was more “primitive” or less “human”. For example, zooarchaeological analysis of fauna from Combe Grenal and Grotte Vaufrey led Binford to argue that the deposit was produced by Neanderthal scavenging behavior (Binford 1988). Chase, in a detailed analysis of the Combe Grenal material, concluded the opposite, that the deposits were accumulated as a result of hunting (Chase 1986). Assumptions about mortality curves, body parts selected and transported and the processing of carcasses result in differing interpretations about the same assemblage (e.g. Marean and Kim 1998 and comments). Zooarchaeologists have to rely on experimental, ethnographic and ethnoarchaeological data to create models of animal exploitation. Extrapolation of these data to different climates and extinct hominins results in debate but not necessarily resolution of many issues. Modern zooarchaeological analyses have examined dietary strategies (bulk or gourmet), provisioning and transportation strategies (carcass butchery, degree of processing, etc.), inferences for site function (hunting camp, butchery site or base camp), inferences for movement within an environment, season of occupation, and evidence for social organization such as food sharing. Research by Stiner on faunal material from caves in the Latium region of Italy demonstrated that Mousterian hominids utilized a variety of subsistence strategies, both scavenging and hunting bovids and cervids (Stiner 1991a, b, and c, 1993a and b, 1994, 2002 a and b). Studies of faunal assemblages created by Neanderthals and anatomically modern humans in Western Europe show a focus on the exploitation of prime age individuals that developed circa 50 000 KBP (Stiner 2002a:20) and continued into the Upper Palaeolithic (Kuhn and Stiner 2001, Morin 2004). Archaeologists and zooarchaeologists now generally agree that subsistence patterns in the late Mousterian and early Upper Palaeolithic are similar and focus on large herd animals (Chase 1986; David and Farizy 28 1994; Mellars 1996; Patou 1989). Bones of large, medium and small game are found in association with Mousterian and Châtelperronian levels of caves and open air sites (David and Farizy 1994; Enloe 1993, 2001; Pike-Tay 1993). All late Mousterian and Châtelperronian sites contain a variety of taxa, usually large herd animals, as do Aurignacian sites (Boyle 1990, 2000; Grayson 2003; Mouton and Joffroy 1958). A few Mousterian sites are dominated by a single species, for example bovids at Mauran (Farizy et al. 1994). Châtelperronian sites show a similar pattern with cervids dominant at St. Césaire (Morin 2004). Specialized hunting, focusing on the interception of a particular prey species appears in the late Upper Palaeolithic (Altuna 1989; Enloe 1997; Enloe and David 1995; Steele 2002; Stiner 1993b, 2002a), coincident with increasing production of a broad variety of bone projectile points, possibly related to changes in hunting behavior (Stettler 2000). Direct evidence for diet and prey species is derived from isotopic studies. Such studies have shown that Neanderthals in northern Europe were highly carnivorous, more so than later Upper Palaeolithic populations. Recent studies of the St Césaire and one of the Spy Neanderthals found that both individuals had isotopic signatures a diet based largely on megafuana in the form of mammoth and woolly rhinoceros (Balter and Simon 2006; Bocherens 2001, 2011). Others, such as the Neander Tal type specimen, ate reindeer (Richards and Schmitz 2008). Generally, both Neanderthal and the earliest modern humans in western Europe had similar meat based diets, both based on terrestrial resources (Bocherens and Drucker 2004; Drucker and Bocherens 2004; Richards et al. 2008). While Neanderthals in northern Europe clearly were meat eaters dependent on large game, southern Neanderthals ate a wider variety of mammals and mollusks (Hockett and Haws 2005). The exploitation of marine mollusks and other marine resources has been cited as an adaptive stratagem by modern humans, or even associated with ‘modernity’ but the same sessile resources are exploited at Gorham’s Cave by Neanderthals. Coastal sites are largely on the Atlantic littoral of Iberia (a result of post- 29 glacial sea level changes) and the absence of dense shell middens in the Mediterranean is more likely the product of the narrow inter-tidal zone which limits the population of sessile mollusks (Bailey and Flemming 2008). Many Upper Palaeolithic coastal sites show a continued focus on land resources, even when located directly on the coast (for example the site of Üçağizli Cave, Turkey), although the amount of marine resources does increase in the Early Upper Palaeolithic (Stiner et al. 2013). Much research has focused on the large amount of animal protein present in Neanderthal diets, particularly northern Neanderthals where isotopic studies show a similar reliance on meat to that of modern high-latitude hunter-gatherers. This data-set may be skewed by reliance on data reported from relatively short periods of ethnographic fieldwork that focused on male hunting behavior. More recent ethnobotanical studies have noted that as much as 70% of modern diets are derived from plant sources (greens, nuts, tubers, bark, fruit), indicating the importance of this gathered food item (Hardy 2010). This would indicate that plant foods played a role in Neanderthal subsistence that needs to be considered when attempting to reconstruct past behavior. Studies of dental calculus show the consumption of starchy plants and grass seeds by Neanderthals from Shanidar and Spy (Henry, et al. 2010). Microwear studies suggest that the amount of plant food consumed related to the local environments. Neanderthals living in wooded environments consumed more plant foods that those in grasslands (El Zaatari, et al. 2011). Plant foods are also necessary as a source of micronutrients and vital compounds. Temperate Europe is rich in plant foods (over 90 species of potential edible plants) which would be readily available to Neanderthals practicing a flexible subsistence strategy. Tooth abrasions indicate consumption of roots and tubers with wear patterns on teeth from St. Césaire falling within the range of mixed-diet hunter-gatherers (Hardy 2010). Clearly hunting was the primary source of energy, but plant foods also contributed to the Neanderthal diet. All studies indicate that subsistence was based on locally available resources, which varied by latitude, elevation and climate in glacial and temperate 30 conditions. Social herbivores were the primary source of protein and fat, supplemented by plant resources. Stiner (1994) has argued persuasively that the development of hunting prime age individuals through an encounter strategy represents occupation of an empty predator niche in Europe and western Asia during the Middle Palaeolithic. Isotopic analysis of Neanderthals indicates that they were among the top predators, if not the top predator on the food chain. Other carnivores were present – all however were cursorial hunters while Neanderthals would have relied on a form of encounter or ambush to take prey. A different hunting strategy would reduce competition and ambush/encounter would enable selection of prime age individuals rather than the slower and weaker animals taken by cursorial hunters. Temporal separation further reduced competition – hyenas are mostly nocturnal while humans are diurnal (Dusseldorp 2010). Humans therefore avoided competition with cursorial hunters (canids and hyenas) and gained access to high value subsistence items in the form of fat and meat protein. The arrival of modern humans in Western Europe resulted in two species of predators occupying the same environmental niche. Expansion of modern humans would eventually lead to competition for the same resources. The debate remains as to how anatomically modern humans expanded into the same niche as Neanderthals and (presumably) out-competed a population that was welladapted to its niche and environment. The sometimes acrimonious debate (e.g. Mellars and Gravina 2008, Zilhao et al. 2008), focuses on lithic production, subsistence practices and inferred behavioral capacity. However this assumes that there were two populations competing for the same resources. The late entry and slow expansion of modern humans into Europe (in comparison to Asia and Australia) could indicate that they did not successfully out-compete their Neanderthal neighbors but rather moved into areas vacated by a population that was in decline for other reasons. Recent radiocarbon dates from Iberia suggest that this may be the case, with no late survival of Neanderthals, although this can only be supported by further radiocarbon dating and a reevaluation of 31 the chronological evidence (Wood, et al. 2013). This method of migration has been proposed for the Caucasus, and the absence of any inter-stratification or evidence of exchange between the two populations in Western Europe further supports such a proposition. Landscape use The Pleistocene landscape of Europe during the late Middle Palaeolithic differed greatly from the modern landscape. During this period a large, rich lowland environment was present in what is now the English Channel and North Sea (White 2006). Recent advances in underwater mapping technology have mapped out the ancient braided river system that drained south west into the Atlantic Ocean. The surviving settlement system therefore reflects adaptations to upland environments, while the patterns of occupation in the now inundated lowlands are unknown. Sea-level variation in Mediterranean and Red Sea has also removed an area of lowlands from our current understanding of landscape use (Bailey and Flemming 2008). Differences in landscape use (mobility, wayfinding and site selection) have been noted between Neanderthals and modern humans in some areas of Eurasia but not in others. Mobility patterns are based on the presence/absence of lithic raw materials from local and non-local sources, and social networks in the Upper Palaeolithic are also inferred from the presence of marine shells at inland sites a considerable distance from the modern coastline. Archaeologists’ understanding of landscape use is anchored on the archaeological site as a focus of activities although determination of site function in the Pleistocene can be problematic. Burke (2006) has argued that site function may not be a major factor in site location, in contrast to many site taxonomies (base camp, extraction camp, butchery site etc.) and settlement systems developed in the anthropological literature. In the Crimea, Neanderthals followed a highly mobile prey, and mapped onto fixed resources 32 (such as lithics). Sites and resource distributions were mapped onto a highly legible landscape. The location of the Grotte du Renne is in a highly legible landscape, in a limestone cliff on the last meander of the Cure, a highly visible and recognizable way point. Similarly, Abri Cellier is located within a distinctive landform. While Neanderthal home ranges were not large, there is ample evidence of planning depth and forward planning and of flexible subsistence strategies (Burke 2012). What is also apparent is the absence of Neanderthals from the broad grasslands of Central Europe. Finlayson ascribes this to a subsistence strategy based on open temperate woodland or ecotones but Burke also notes the importance of way markers and landmarks in Neanderthals mapping practices, which tend to be absent in broad areas of grassland (Burke 2006). Early Aurignacian sites in the Middle Danube follow the river (Svoboda 2006) and Gravettian sites are located on low spurs in valleys (Goutas, pers. comm.) which suggests that landmarks remained important in navigating the open Central European Plain during the earliest movements of modern humans into the area. This is also an area of large-scale loess deposition during the last pleni-glacial which has likely resulted in the deep burial of Mousterian open-air sites. Unless survey protocols require deep testing, open air Mousterian sites in river valleys will remain unrecorded. Subsistence and mobility patterns vary among Neanderthal populations over time and space. In the Southern Caucasus both Neanderthals and later human populations followed the same mobility pattern, moving from lower to higher elevations as they followed the Capra caucasia herds over their seasonal migrations. While the subsistence strategies were similar, the lithics indicate the use of local materials by Neanderthals, suggesting small, local territories occupied by small groups (Adler, et al. 2006). In contrast, Early Upper Palaeolithic groups in the Caucasus and elsewhere had access to obsidian from sources over 100km distant (Adler, et al. 2008). A similar pattern of small territories with a high degree of planning and exploitation of local resources is evident at Abri Romani (Level M) where the majority of lithics are derived from sources at least 33 10km from the site and arrive on site as finished or partially finished reserve material (Fernández-Laso, et al. 2011). Again, the occupants of Level M had a flexible strategy that exploited three nearby ecosystems and were highly mobile within the home range of the site. Subsistence strategies underlie variation in mobility patterns in other regions. In Franco-Cantabria there is a shift in subsistence strategies to a more logistical pattern in the Early Upper Palaeolithic (Cosgrove and Pike-Tay 2004) although these authors argue strongly that lithic technology and complexity of toolkit should not be used as proxies for a particular hunting behavior, noting that logistical collectors can operate very effectively with apparently ‘simple’ armatures and lithic operational sequences. In their study, prey ethology is an important factor in subsistence organization. Other studies in other areas indicate that lithic technology and prey ethology may not predict site organization and subsistence. In the southern Levant, Neanderthals appear to have followed a radiating or logistical residential strategy in contrast to the circular or foraging residential strategy of earlier and later modern humans groups in the region (Lieberman and Shea 1994). These strategies appear to be largely independent of environmental factors, but reflect differing subsistence choices within a patchy environment. In an overview of Mousterian site occupation patterns in Eurasia, PatouMathis argues for a more logistical approach where prey is seasonal or the preferred prey species are relatively large (Patou-Mathis 2000), but individual sites reflect a broad range of subsistence patterns and choices, along a continuum from foraging or opportunistic strategies to planned or logistical patterns of resource acquisition. For example, at Peche de l’Azé 1 (Southwest France) subsistence strategies varied between interception and encounter based on season of occupation and prey availability, but the site functioned as a base camp regardless of the subsistence strategy (Rendu 2010). In southeastern Europe both Neanderthals and modern humans had similar flexible land-use strategies that correlate with the environmental variation. Lithic analysis 34 demonstrated greater mobility among both groups in colder climatic episodes (RielSalvatore, et al. 2008). In the far northwestern region of Neanderthal occupation (aka southern Britain) the archaeological evidence points to short term occupations that reflect a high degree of logistical planning, provisioning, cooperation, knowledge of the local landscape and prey behavior from bases in the lowlands of Doggerland (White and Pettitt 2011:77). In the Middle Vézère Valley of the Dordogne, recent GIS-based studies have demonstrated a shift in site location from higher to lower elevations from the Middle Palaeolithic to the Upper Palaeolithic. Mousterian sites are generally at higher elevations, located to exploit ecotones and are also located to exploit a large viewshed. This is interpreted to indicate an opportunistic or foraging strategy, where a larger viewshed permitted easier encounters with prey species (Sisk 2011), in contrast to the Early Upper Palaeolithic sites near fords where an intercept strategy is inferred (Sisk 2011; White 1985). Châtelperronian contexts in the Middle Vézère are generally collocated with Aurignacian deposits, and are relatively few. Where they do occur, the viewshed is generally larger than the average for Aurignacian and Gravettian occupations. This could indicate a more logistical approach to subsistence procurement along a continuum of subsistence strategies. Recent data from Western Europe indicates movement of a small amount of raw materials over considerable distances in the Middle Palaeolithic. At an open air site (Champ Grand) at the southern limit of the Paris Basin and northern limit of the Massif Central in the Loire Valley approximately 1% (n=568) of the formal tools were found to come from sources up to 250 km from the site, indicating contact between groups over a wide distance (Slimak and Giraud 2007). The preponderance of evidence indicates that Neanderthals had some forms of long distance contact with other groups, including occasional aggregation, but these contacts were not as substantial as the interaction networks that develop during the Upper Palaeolithic. Gamble argues that Neanderthal 35 material culture does not reflect enchainement in large social networks like those of the Upper Palaeolithic (Gamble 2011:159). These less robust indications of interaction with other populations in the larger region speak to the degree to which mobility indicates the depth of social networks and related access to information that form an important part of modern hunter-gatherer behavior (Whallon 2006). These networks are formed by the unique bisexual philopatry found in modern hunter-gatherer populations (Hill et al. 2011). The combination of philopatry and frequent visiting between nearby related groups facilitates information exchange and the rapid spread of new technology. Such networks also facilitate the retention of cultural and technical knowledge whereas innovation or even basic skills may be lost when networks are small or interactions are infrequent (Hill et al.2011: 1288). The lower proportion of exotic materials indicates less investment in long-distance communication networks by Neanderthals which could have severely inhibited the spread of new ideas. The absence of well-established pathways of interaction and information exchange may relate to the absence of symbolic behavior and the need to communicate beyond a regional territory on a regular basis. Conclusion In summary, the archaeological record shows little difference between Neanderthals and anatomically modern humans in terms of subsistence behavior and lithic tool manufacture. Evidence for symbolic behavior and long-distance information sharing do indicate differences between the taxa. While it seems that archaeologists are willing to accept that similar subsistence patterns demonstrate similar adaptive responses to similar subsistence and provisioning problems by different hominins, there is less agreement in the interpretation of symbolic behavior associated with Neanderthals. Different methodologies compete to confirm or deny the presence of symbolic behavior by Neanderthals, especially as expressed by items of personal adornment. Another area of difference for which methodologies may be successfully developed is the use of bone and 36 antler for tools. The appearance of items of bone artifacts but the lack of bone points in the Châtelperronian has caused debate as to the nature of contact between the groups, the ability of Neanderthals to create new technologies, and the purpose of personal adornment (d’Errico et al. 1998; White 1998). The organization of labor and technological resources among Neanderthal populations is significant for our understanding of their lifeways in general. One important aspect, fundamental to success in many ways, is the subsistence organization of the group. Before examining Neanderthal life histories and their significance for subsistence behavior, including the acquisition of bone as a raw material, it would be useful to examine modern hunter-gatherer social and subsistence organization to better understand how people learn to hunt and forage, and how labor is allocated across the group in terms of mobility, age cohort, gender and experience. This will be examined in the following chapter. 37 CHAPTER 3: PROXIES FOR THE PALAEOLITHIC: HUNTER – GATHERER STUDIES Introduction This research project examines the organization of labor and technological resources among Neanderthal and modern human societies during the Early Upper Palaeolithic in Western Europe. This is important for understanding their ecological roles and, perhaps, their relative evolutionary success. Human culture is not innate, but a learned behavior that is particular to specific social groups within the constraints of biological growth and human development. No modern human cultures live in environmental conditions that mirror the early Late Glacial habitat of the two hominins. Ethnoarchaeological research and hunter-gatherer studies provide data on the organization of labor and technological resources within the ecological constraints of different environments. From these data we can examine how individuals in modern hunter-gatherer societies acquire competence in a variety of subsistence tasks and how cooperative behavior within the group, at a variety of levels, aids in the successful rearing of offspring and maintains adequate access to nutritional and other resources within the group environment. Modern hunter-gatherers have often stood as proxies for Palaeolithic huntergatherers. Early encounters between European explorers and hunter-gatherer groups influenced the development of archaeological thought, providing antiquarians with explanatory models for the stone tools found in European fields, caves and gravel pits (Daniel 1976; Trigger 1989). Nineteenth century anthropologists and archaeologists used hunter-gatherer material culture as a proxy for evolutionary development, placing huntergatherer cultures on a lower rung of cultural and intellectual development than agriculturalists or that acme of evolution, the nineteenth century rich white male. With the abandonment of early racist attitudes, hunter-gatherer societies have been the subject 38 of a broad range of anthropological research. In the field of archaeology, hunter-gatherers have been of particular interest to Palaeolithic archaeologists, who have used field studies, particularly ethnoarchaeology, to examine modern material culture and to derive models and testable hypotheses of past behavior through these studies. While earlier studies focused primarily on hunting, more recent research has examined huntergatherers within their ecological and social contexts. More attention has also been paid to the development of subsistence skills and how hunter-gatherer lifeways are transmitted. Recent studies examine how hunting and gathering behaviors are learned by children (Bird and Bliege Bird 2001, 2005; Blurton Jones et al. 1994; Hawkes et al. 1995; Hrdy 2005; Konner 2005). Studies have also examined how and when hunters and gatherers are regarded as proficient in their subsistence activities, and how long they retain their proficiency (e.g. Bock 2005). Other research focusses on how the environment influences life histories and development of subsistence skills (e.g. Ellis, et al. 2009). Perhaps most importantly, recent hunter-gatherer research has focused more on how different subsistence behaviors (including hunting, gathering, meat processing and hide processing) are part of an integrated system in which different actors play important roles at different stages of food acquisition, carcass processing and food preparation (Gurven and Kaplan 2006; Hames and Draper 2004; Hawkes et al. 1997; Hurtado et al. 1992). Subsistence organization: models and reality All modern hunter gatherer societies have labor structured to a certain degree by gender and age. Early ethnoarchaeological studies focused on male behavior (primarily hunting) as a means of modeling the evolution of human behavior (Fedigan 1986; Kaplan and Hill 1985; Lee and DeVore 1969; Marchant 1991; Washburn and Lancaster 1968). Critiques of this androcentric approach and the realization that women contributed much of the predictable food supply, resulted in the broadening of subsistence studies to 39 include women’s roles in provisioning and food sharing (e.g. Linton 1971; Tanner 1981; Wylie 1991). More recently, the recognition that children participate in, and have their own, provisioning activities has added further nuance to the understanding of modern hunter-gatherer subsistence. It is clear that a focus on the actions of a single gender, sex or age group negates the importance of the complementary nature of members of that group in obtaining adequate subsistence (Gurven and Hill 2009; Hawkes, et al. 2001; McCreedy 1994). Labor practices of modern hunter gatherers are a product of their particular ecology, technology and culturally appropriate behavior. These labor practices strongly influence the decision-making process for each group. Anthropological models for hunter-gatherer decision-making are frequently based on optimal foraging models (diet breadth, patch choice and marginal value studies) which have been justly critiqued for the lack of attention to the cultural context of food procurement (Bettinger 1991; Jochim 1988b; Shott 1991, 2004; Winterhalder and Smith 1981). These models focus on calorific returns (protein, meat, etc.) and rarely consider the importance of prey animals as sources of raw materials for clothing. This may, in part, be a result of the sartorial habits of many surviving hunter-gatherer groups. All these societies have contacts with, and participate in, the larger, global economy and therefore have access to modern fabrics and clothing. Surviving Arctic cultures retain some of their extremely efficient winter clothing, but even in these regions, hide procurement is geared towards the modern global economy (Hatt and Taylor 1969; Oakes 1991). Decisions regarding labor allocation are the product of knowledge of local resources, seasonality, subsistence requirements and the estimation of probable success. Ethnoarchaeologists have examined decision-making in terms of resource exploitation and transportation costs relative to perceived nutritional or energetic benefits, often informed by optimal foraging models (which rarely consider post-transportation costs of hide or skin processing). Much research in the 1980s focused on male hunting and decision-making, with little examination of the role of women’s role in the gathering of 40 resources. More recent research has highlighted women’s and children’s roles in acquiring protein, fat and carbohydrates (Bird, et al. 2009; Bliege Bird and Bird 2008; Noss and Hewlett 2001). The degree to which women can participate in hunting of large and small game is determined by the local environment and culturally-specific child care practices. The main restriction on women hunting and traveling widely is the care of unweaned children. This period of restriction, in the 20-30 year age range, coincides with the period that male hunters are learning the skills necessary to track and hunt larger game (Gurven, et al. 2006). As a consequence older women, who could hunt, lack the necessary acquired knowledge to hunt as effectively as their male counterparts of the same age. Fertile women are constrained by the need to breastfeed infants, but child care and provisioning in hunter-gatherer societies beyond that point are flexible and relate to resource availability and the local environment. The sexual or gendered division of labor in food procurement, food processing, lithic and bone tool manufacture and even the production of art are/were stereotypical ideas that in themselves were the product of Western culture and therefore seldom questioned (Conkey and Gero 1991; Gero 1985, 1991; Owen 1999; Spector 1991; Waguespack 2005). Western anthropologists tend to categorize labor into separate spheres of work, which produces an oppositional or dichotomous relationship that obscures the integration of male and female labor to provision each other and offspring (Owen 2005:13). This apparent separation of male and female labor in anthropological and ethnoarchaeological studies creates a false barrier in a behavior system that is regarded as cooperative by the members of that system (Bodenhorn 1990; Jarvenpa and Brumbach 2009). It is better to consider female and male roles as complementary and the product of biology, ecology and social structure. In this context, the roles of individuals are enmeshed in a larger system predicated on the reduction of risk and the provision of adequate calories and nutrients within the immediate family and larger social network. 41 Some authors argue that large game hunting would never have been a viable subsistence strategy without sharing (e.g. Gurven and Hill 2009:52). Food sharing of large game redistributes protein and fat across a social network, a practice vital in maintaining access to resources in economies where storage is not practiced (Gurven and Hill 2009; Kent 1996). Sharing also minimizes the risk inherent in big game hunting, where return rates per hunter based on animal weight can be high, but the success rate is low (Hawkes, et al. 2001). Decisions made about where and when to hunt are therefore based both on the presence and absence of game, and also the need to maintain sharing networks. Decisions on when to hunt and at what point to abandon pursuit have mostly been examined through the lens of foraging theory and maximization of returns. These are linked to available technology and also to the ethology and distribution of prey species. Tropical, subtropical and Boreal forest groups practice an encounter strategy which intercepts dispersed animal resources with immediate consumption. This is largely caused by the ethology of the game species, which tend to live in small groups or as solitary individuals and also by the ecology of the game, which is dispersed across a patchy landscape with little or no seasonal movements or aggregation. Other huntergatherer groups, generally in higher latitudes, intercept predictable game species and practice delayed return in the form of storage (Binford 1978, 1981, 2001). In all societies, men may hunt and transport meat to camp, but women ultimately process the carcass and share out the meat. Women are known to be small game hunters and even ‘traditional’ studies of women’s foraging for plant foods have frequent references to the taking of small game by women (Owen 2005). Recent studies have emphasized the active role of women as hunters in Australia, Africa and North America. For example, gender differences in labor are subtle among the Martu. Women hunt smaller game that often has to be dug out, while men hunt more mobile bigger game (Bird, et al. 2009; Bliege Bird and Bird 2008). 42 In this case, the goal is to optimize household production. In Central Africa, Agta women participate in net hunts with and without men, and division of labor is fluid and dependent on gender and age. Data suggest that childless or childfree women hunt more frequently than women with young children (McCreedy 1994; Noss and Hewlett 2001). In boreal North America, Chipewyan Dene women hunt alongside their male partners, although less that formerly, as socio-political reorganization has resulted in women being more closely tied to permanent settlements by mandatory schooling requirements for children. Interestingly, this development has resulted in a shift in hunting from a foraging to a logistical pattern, further evidence that hunting behavior is a product of social, economic and ecological factors (Brumbach and Jarvenpa 1997). Women hunters are documented (usually anecdotally) among the Aché, Cree, !Kung and Inuit. As Owen (2005) has noted, it would be extremely maladaptive not to teach daughters how to hunt in situations where groups are heavily dependent on animal resources. Women’s foraging, while incorporating exploitation of small or sessile game, focuses largely on plant foods in temperate and tropical environments. Vegetal remains are far less visible in the archaeological record, therefore direct evidence of female labor are well documented in the ethnographic record, but largely invisible in the archaeological record. (Although advances in archaeological techniques such as phytolith studies and starch grain analysis are providing additional data in this area). While meat is a valuable source of protein and fat from marrow, plant foods supply carbohydrates and other nutrients that are also vital in human development and nutrition. Foraging for plants, like hunting, is a social behavior and foraged food is also shared, generally among the immediate family rather that within the larger social unit (Hames and Draper 2004; Hawkes, et al. 1997; Hurtado, et al. 1992). As with meat sharing this behavior is classed as altruistic, but here it improves the chance of child-survival and therefore reproductive fitness of the non-parent (often a grandmother). While meat sharing reinforces larger social bonds and reciprocity within a broader network of more distant relatives or non- 43 kin; enhancing social relationships, sharing of vegetal foods reinforces ties within the immediate family. Foraging behavior is also related to mobility and age. Among the Hadza, postmenopausal women forage more than mothers with children, and nursing mothers forage the least (Hawkes, et al. 1997). Among the Pumé, women fill each other’s baskets, and older women carry more and contribute more than younger women. Older women also travel more frequently to distant but desired food patches (Hilton and Greaves 2008; Sear and Mace 2008). Women in hunter-gatherer societies follow a pattern of cooperative breeding by working together as a group to enhance the survivorship of offspring (Burkart, et al. 2009; Hrdy 2005). It is clear that child care also affects female foraging patterns and local ecology is a strong predictor of foraging behavior (Blurton Jones, et al. 1996; Hawkes, et al. 1995; Hurtado, et al. 1992). In open savannah with few predators and good landmarks, navigation and movement is easy, therefore Hiwi and Hadza children can roam relatively unsupervised. Similarly in India, Chenchu children foraged in groups separate from their parents when considered old enough to leave the village, and younger children foraged on the village periphery (von Fürer-Haimendorf 1943). The major predators in this area were largely nocturnal and therefore not feared by both adults and children. In contrast, the Aché in South America regard the forest as unsafe and unhealthy for unsupervised children. Kalahari San groups do not let children forage at all in the topographically featureless landscape. In those circumstances children, who lack knowledge of wayfinding, could easily get lost. The long distances to resources and lack of water and shade make the cost of children’s foraging extremely high and an impediment to their care-givers in these environments. Cultural norms stress the dangers to children of getting lost or killed away from camp (Blurton Jones, et al. 1994; Blurton Jones, et al. 1996). The degree to which children can forage or process their own food therefore varies by environment and as a function of a child’s physiology and acquired skills. 44 There has been a tendency among anthropologists and ethnoarchaeologists to assume that children are passive consumers of foodstuffs (Bird and Bliege Bird 2001:461; Konner 2005). This is a product of Western researchers’ assumptions regarding food gathering and processing, and also, I would argue, a product of our own experience in Western culture where children are mainly consumers of food, rather than collectors or processors of foodstuffs. In hunter-gatherer culture children participate in food gathering and/or food processing. While !Kung children do not forage, they are responsible for cracking nuts for themselves and their younger siblings, enabling adults to devote more time tofood collecting and less time to food processing (Blurton Jones, et al. 1994; Hawkes, et al. 1995). The more mobile Hadza children collect up to half of their daily calorific allowance by age 10, thereby reducing the costs of child-provisioning, especially for their mothers and grandmothers (Blurton Jones, et al. 1994; Blurton Jones, et al. 1996; Konner 2005), and Chenchu children were able to collect and process much of their own food (von Fürer-Haimendorf 1943). This independent food collecting buttresses the older children against the diversion of food resources by their mothers to their younger siblings. Children forage in groups or with adults near their camps, and collect roots, tubers, baobab fruits, berries and nesting birds according to season and camp location. In western Australia children forage for grubs, fruits, roots and a species of lizard. When children forage with women, the women choose the location, but women and children forage at different locations and for different resources. When children forage by themselves the nature of the resource near the residence predicates children’s own foraging decisions (Bird and Bliege Bird 2001; Bird, et al. 2009; Bliege Bird and Bird 2008). In coastal zones, for example among the Meriam, children collect shellfish and their collection and processing strategies differ from those of adults. Meriam children do not cover as large an area and in consequence, have fewer encounters with higher ranked food, and therefore take a broader range of available shellfish. It is clear that children seek sessile resources, or small animals that are easily captured. They are also less likely 45 to process food in the field but return to camp with their gathered foods (Bird and Bliege Bird 2001). Children are recorded as taking small game, fishing and participating in net hunting (Noss and Hewlett 2001). From existing ethnographic data it is clear that children collect sessile resources and rarely participate in activities that place them at risk of injury or death, particularly large game hunting. The size, strength and stamina of children make it more likely that they will focus on food stuffs that are easy to gather and easy to process (Bock 2005). Foraging makes children less reliant on parental provisioning, although adolescents and even young adults are not the most efficient hunters or foragers. Learning to hunt and forage – a lifetime learning experience Foraging and hunting are learnt behaviors in modern hunter-gatherer societies. Individuals have to learn what is culturally appropriate subsistence behavior, beyond the basics of which plants and animals are non-toxic to humans. Humans have to acquire the motor processes to manufacture tools and to procure and prepare subsistence items (MacDonald 2007). Brain development and human growth patterns indicate that humans are capable of learning (by direct instruction, play or observation), at a relatively young age, but their physical development precludes active participation in tasks requiring more strength until body size increases (Bock 2005; von Fürer-Haimendorf 1943). Ceremonies are performed to mark a boy’s first adult hunt or kill, and a girl’s first adult foraging trip in some societies, a formal mark of a change in subsistence roles and social expectations (Fink 2004). There is cross-cultural variation in the age at which children begin to participate in hunting trips, but there appears to be an intensification around the age of 12. This coincides with an adolescent growth spurt in Inuit, !Kung and Australian Aboriginal groups. Strategies to capture large game are learnt relatively late, with large game kills not occurring until late adolescence (MacDonald 2007). Hunters of medium and larger game are at their most effective relatively late in life (Gurven, et al. 2006). 46 Hunting skill is less a factor of physical strength than the result of observational skills which peak between age 30 and 40 (Walker 2002). Non-meat foods also require proficiency in collection and processing (Gurven and Kaplan 2006). Stamina and strength are required to collect and process some plant foods, either because of the distance between patchy resources, or the nature of the resource itself. In general, individuals become more proficient in low skilled tasks most rapidly and return rates become size dependent as a task is mastered. Again, older individuals tend to supply more sessile and plant resources to the overall diet, but lack the stamina to hunt. Sharing within the family group or within the band buffers novices and the elderly alike and ensures adequate nutrition. It is vital in the context of the relatively long post-weaning growth period of human children. Clothing: the other time consuming by-product of hunting The majority of the ethnographic literature focuses on the acquisition of calories through the consumption of animals. But animals also provide the means to conserve calories as a source of hides for clothing or shelters, such as windbreaks or tents. There are few areas of the world where modern humans can survive without some form of clothing or shelter. As the last surviving species of a genus that originated in sub-Saharan Africa, modern humans retain an ability to adapt physiologically to heat rather than cold. To survive in temperate and arctic environments some form of shelter is vital. Various groups do exhibit some physiological adaptations to cold environments (for example Inuit or Tierra del Fuegans) but these are not enough to enable survival without the ultimate in extrasomatic adaptation – a shelter to sleep and eat in, and some form of clothing to retain heat around the body. We, as anthropologists, should also consider the practical need to protect our delicate skins – humans lack the protection of a thick hide or thick fur to prevent injury or irritation that can result from accidental contact with plants and insects. 47 The use of clothing and shelter varies by culture in temperate societies, and even within similar environments. For example, in Tierra del Fuego differences between northern and southern hunter-gatherer groups’ use of animals for shelter reflected differences in subsistence practices and local ecology despite similar mean annual temperatures; although there was a difference in annual precipitation of 8” between north (drier) and south (wetter) (Lothrop 1928). The northern Selk’nam were terrestrial huntergatherers who exploited guanaco hides for clothing, shelters, bags and thongs. Guanaco hides provided warm, waterproof cloaks and boots. Men’s cloaks were made of 2-3 guanaco hides and women’s of two. Children were rarely clothed, even in winter. The men’s cloaks were wrapped around the body and held in place by the hand and dropped to shoot arrows during hunts. Women’s cloaks were closed with thongs to keep their hands free to perform daily subsistence activities. The highly mobile groups (at least in the late nineteenth century) used simple three-sided shelters on their campsites, which were easy to transport and provided shelter from the wind and rain. The southern Yaghan focused on maritime resources in an area dominated by temperate rain forest. In contrast to the body-enveloping cloaks of the Selk’nam, the Yaghan wore shorter sealskin cloaks and (according to Lothrop) less effective shoes. However, their winter campsites contained semi-subterranean huts with a fire pit located at a lower level than the interior benches, providing a very effective means of staying warm, and their summer wigwams were effective waterproof shelters covered with leaves and bark (Lothrop 1928:127). Both shelters reflect a higher degree of investment in the construction of structures for protection from the environment than that seen in the northern Selk’nam. Differences in the use of shelter and clothing types in similar climatic conditions indicate that there is no one solution to the problem of retaining body heat in a temperate to cool climate. A more mobile group might rely more on immediate shelter provided by clothing and a less mobile group, or group that operates from a long-term base camp, might invest more energy in larger shelters. Seasonality would be an 48 important factor in the use of clothing – for example, modern ethnographies indicate that fur was regarded as too warm for summer use by the Tungu, Chucki, and Koryak, among others (Hatt and Taylor 1969:8) and different outer garments were used in summer and winter by men and women in the Taiga (Brandišauskas 2010). Another important factor is the degree of movement necessary in daily subsistence activities. Terrestrial hunters may require more effective clothing while stalking or ambushing game, whereas a large cloak would be extremely cumbersome while maneuvering a canoe, and a major danger during a capsize. Clearly a variety of responses to temperate climates were possible in the Palaeolithic of Europe, relative to the local environment. What clothing did provide was the means to create a new adaptive response to climate change. Instead of retreating into warmer or more stable refugia, hominin groups in OIS3 could remain in more northern areas of temperate Europe than before, therefore enlarging the niche available for occupation. Clearly, while food acquisition is extremely important in hunting decisions, the use of animals for clothing is another important factor in the processing of carcasses. The toolkit required to process hides utilizes bone tools, such as modified scapulae or longbones for hide scrapers, as well as bone awls and needles. While bone tools are also used for other process, such as bark removal, the almost universal requirements for shelter suggest that these tools should correlate with the more systematic use of hides for shelter. Ethnographic data on hide working indicates that hides can be treated and prepared using relatively simple tools, but the process can be time consuming where hide clothing is vital for survival. “Skin preparation is the never-ending work of the arctic women” (Hatt and Taylor 1969:20). This statement can be extended to other groups in the sub-arctic and temperate zones. Hide processing is undertaken to preserve the skin and, depending on intended use, to create a supple waterproof or warm enclosure for a human. Subcutaneous fat and flesh must be removed without tearing the skin. The range of tools used for this purpose 49 is broad and includes scapulae (Griffitts 2007; Hoffman 1980; Oakes 1991), long bones or metapodials (Beyries 2008; Gilmore 2005; Steinbring 1966). Long bones also serve as handles for stone scrapers. While certain elements from particular species might be preferred, bones that are readily available are used. For example, on the Little Black River Reserve, informants stated that a bear ulna made the best deflesher, but as bear were no longer available, moose metatarsals were used (Steinbring 1966:579). If a smooth hide is required, hair must be removed, either by relying on natural decay, soaking, abrasion or shaving with a bone or stone tool. Drying is an important part of hide preparation, which occurs before processing in European groups and after initial processing in North American Arctic cultures. The hide must be stretched out to avoid shrinkage. All skins will then require softening (or mechanical preparation) by pulling and stretching over a log or rope or through rubbing with a stone or pumice or by pounding (David et al. 1998: 124). This breaks down and realigns the stiff fibers in the hide and makes it supple for use (Hatt and Taylor 1969). An additional stage in the process is tanning, introducing fat into the hide and processing to make it more durable and rot resistant. A very common tanning agent is brains, cited by ethnographers in the Arctic, Boreal and sub boreal groups. This material is applied to the skin, left for a period of time and then removed by scraping. The skin is then dried and softened. Another preserving agent is woodsmoke, which also waterproofs the hide and alters the color. Hide curing and tanning is a time consuming and labor intensive process. Ewers (1945) lists the stages necessary for hide preparation by traditional Blackfeet women. i) the hide is pegged out and defleshed ii) sun dried “for several days” iii) scraped to even thickness, then reversed and hair removed if required. A rib bone beamer may be used for this. 50 The skin is now rawhide – used for containers, ropes, and moccasin soles. If clothing is required the hide is tanned: iv) soft tanning: brain fat and liver is applied by hand and then spread and rubbed in v) sun drying vi) saturating with warm water and wrapping into a bundle vii) stretching viii) final softening by rubbing to remove excess material and sawing through a loop of rawhide. ix) smoking to keep clothing flexible after wetting (this is optional). Therefore prior to cutting and sewing, hide processing requires several days and heavy manual labor. Sewing did not require needles: all clothes, bags etc. were sewn using awls to make holes in the skin. Stenton notes that needles are only necessary for waterproof seams (Stenton 1991). In Siberia, among the Orochem-Evenki, a similar pattern of hide processing is undertaken (Brandišauskas 2010). Hide is defleshed and dried, then depilated by scraping or shaving, and greased using animal fat to soften and waterproof it. It is then smoked and softened by hand. At this stage it is suitable for gear products but not clothing, which requires further processing through tanning. Hide is softened by the application of rotten moose liver, left for a few days and then oiled and left for 1 day (red deer, roe deer, reindeer) or 3 days (moose). The hide is then stretched, cleaned, smoked then scraped and softened by hand. For half a moose hide the entire process requires six days of liver softening and smoking and another 8.5 hours of processing activities (ibid: 113). Then additional time is needed to sew and decorate clothing and footwear. 51 There are little data in the ethnoarchaeological record regarding the collection of raw materials for hide working – both the hides and the bones. As noted above, requirements for clothing and shelter are related to the ecology of particular huntergatherer groups, and their access to the global economy. In the Arctic, among the Sami and other herders, reindeer hide usage varies by the age of the prey, with newborn calf hides (soft but not very durable) used for infants clothing; reindeer calf hides used for women’s garments and older calf hides used for men’s outer clothing (Delaporte, et al. 1980; Hatt and Taylor 1969). Among the Eastern Inuit the thickness and condition of the pelt is of extreme importance in the construction of inner and outer garments that permitted both easy movement, control of body temperature and the ability remain warm whilst watching seal blow holes on the Arctic ice (Stenton 1991). Clothing also varies by season, as would be expected, with lighter clothing or fewer layers necessary in the warmer months. Conclusion Hunter-gatherer societies form part of an interrelated world system of social organization. Within hunter-gatherer societies, labor is divided based on a number of factors related to physical strength, mobility, knowledge, skill sets and ecological factors. This results in an gendered division of labor, that may not be as dichotomous to the members of a given society as it is to a Western ethnographer who was raised in a system that has binary structures such as domestic: public or male: female tasks. Animal acquisition and processing is a continuum from the stalking or interception by more mobile members of the group (usually male) through initial processing to later processing and sharing at camp by less mobile members of the group (usually female). Collection of plant foods follows a similar pattern among the more and less mobile members of the group that do not hunt (usually women and children). Mobile individuals travel further or 52 collect more plant foods and other sessile resources. These are shared with less mobile individuals, usually on a more restricted basis within the immediate family group. Food sharing serves to buffer nutritional inequalities among the group and, very importantly, allows the time necessary for children and young adults to acquire the skills needed to provision their own families. Full competence is late (mid-20s for gathering and mid-30s for hunting) which has implications for the social structure of Neanderthals and their anatomically modern neighbors, which will be discussed further below. Ethnographic data also provides models of how clothing and shelter is used in a variety of ways to protect humans in temperate and cold environments. Whilst we should always be wary of taking any one model to explain the archaeological record, given that different subsistence and ecological factors weigh into clothing and shelter systems used by any group, it is clear that the processing of animal skins into material suitable for storage containers or containers for human limbs requires considerable investment in time and energy. The material culture associated with such processing behavior may be very simple, even if the results can be quite complex in terms of construction and use. This is another skill set that has to be acquired by members of the community and is vital to survival of the culture in the long term. For this research project the question arises – was there selection of prey for hides, or was the prey available used for containers? Ethnographic studies suggest that among hunter-gatherers and some pastoralists the main object is meat acquisition from herd species, but there is also hunting of fur-bearing carnivores that are not part of the diet. If these animals occur in Neanderthal contexts, with evidence of hide removal, it could be argued that clothing is not only an integral part of Neanderthal subsistence, but that there was also selection for particular hides (color or pattern) that were used for decorative and (therefore) arguably symbolic purposes. 53 CHAPTER 4: NEANDERTHAL LIFE HISTORIES AND IMPLICATIONS FOR SUBSISTENCE Introduction In this chapter I will utilize data from the archaeological record and combine this with information on modern hunter-gatherer lifeways discussed in Chapter 3 to model Neanderthal life histories. I will then consider if these life histories would have any significant impact on inferred Neanderthal subsistence behavior. As we have seen, Neanderthal social organization is rarely discussed in the literature. Neanderthals are assumed to live in small family groups within defined territories. Large-scale, wellestablished social networks are lacking, although there is evidence for some long distance transportation of exotic lithics in France and southern Italy. The absence of unequivocal symbolic behavior is also argued as evidence for a lack of social networks. Yet evidence is emerging for the use of bird feathers for adornment in the Mousterian, and for the selection of fur-bearing species, presumably for the decorative nature of the pelts. The degree to which Neanderthals wore clothes is also debated, although it is now clear that some form of protection was necessary for survival in temperate and cool zones of Europe. Another way to approach Neanderthal social organization and lifeways is to examine the growth and development of Neanderthals and discuss how and when our closest extinct relatives would become proficient food producers. By examining evidence for development and growth, and combining this data with modern ethnographic studies, we should be able to determine how Neanderthals developed their subsistence skills and how dependent they would remain on the group as a whole. Neanderthal ontogeny All studies of Neanderthal ontogeny are hampered by the absence of a living population or a large skeletal sample. Many analyses of tooth eruption sequences, postcranial growth and fusion rates are also structured in terms of the evolutionary 54 significance of the differences between the extinct Neanderthals and successful modern humans. There are differences between the two species, but it is neither clear if this is a result of genetic drift, nor is it easy to confirm that these differences are anything but neutral in evolutionary terms. This is especially true in discussion of skeletal growth, which is held as a proxy for brain development and learning capability. We now know that modern human brain growth is near complete by the age of 7 or 8, but neural development continues after full physical maturity is reached, and probably after the average age of first reproduction (Robson and Wood 2008). As we lack Neanderthal brains to study, we can only use modern hunter-gatherers as proxies. While past paradigmatic bias has focused on how Neanderthals are different (and therefore less welladapted or unable to compete with modern humans) we should perhaps examine evidence, based on growth patterns in modern and fossil populations to determine if Neanderthal life histories indicates similar or different patterns to their modern human contemporaries. Studies of primate foetal development, growth, reproductive age and senescence indicate that Neanderthals had a modern human pattern of life history (Robson and Wood 2008:417) and did not mature at a faster rate based on estimated brain size at birth and post-natal brain growth rates (Ponce de Leon, et al. 2008). Brain growth is more rapid in Neanderthals (although this may just be related to the overall larger brain) and the ontogeny of facial development appears to be different, again probably a result of the particular morphology of the Neanderthal face. It is not clear if Neanderthal dental development is more rapid that modern humans, based on comparisons of dental eruptions sequences, enamel formation and comparison with post cranial development (Guatelli-Steinberg 2009). Post cranial studies are rarer and have their own issues. Studies using tibial development tend to underestimate height, and those using femora overestimate adult height. Nevertheless, the growth curve for both modern humans and Neanderthals to the age of 60 months is the same shape, but Neanderthal growth is slower after the age of 15-20 months, indicating that growth rate differences are 55 established in the post-gestational and post exclusive lactation period (Martin-Gonzalez, et al. 2012). Dental hypoplasia in Neanderthal children between 2.3 and 2.8 years old strongly indicates weaning, although some authors argue for later weaning, between 4 and 5 years, suggesting longer birth spacing and slower population growth (Pettitt 2000). Slower overall growth rates in Neanderthal children may be linked to higher metabolic loading either as a response to the local climate or to the building of greater muscle mass present in adults. The data on growth rates indicates strong parental investment in child provisioning. The grown rate curve indicates that young Neanderthals remained dependent on adult provisioning for a long period of time, probably as long as that of modern humans. The demands on parents and ethnographic models indicate that alloparenting would be a vital component of successful child rearing in the Mousterian of Europe. The role of older siblings in provisioning and teaching foraging behavior should not be ignored (Pettitt 2000:359). Modern hunter-gatherer ethnographies indicate a considerable variation in the amount of direct provisioning undertaken by children, which is determined by the presence or absence of large predators, the presence of resources that children can access easily (such a fruits, nuts and other sessile resources) and the ease of navigation within a landscape. Both Abri Cellier and the Grotte du Renne are located in landscapes that contain a varied topography and a wide range of prominent landmarks. The sites also provide access to a range of ecotones from a river valley to an upland plateau. This relatively rich environment would likely contain a broad range of seasonally available sessile resources. In these locations it would therefore be possible for sub adults to acquire some of their calorific intake directly. Native European edible wild plants and wild tubers were available throughout the Middle Palaeolithic in Europe, and tubers in particular would have provided valuable carbohydrates in a high protein diet (Hardy 2010). 56 But before our little Neanderthals go running off into the woods, we need to consider the presence of other top level carnivores. Middle Pleistocene Europe was home to cave lions, wolves, and cave hyenas that were all obligate carnivores. (I exclude cave bears because isotopic data suggest that these were mostly omnivorous or frugivorous). Cave lion predation focused on medium to large herbivores, and it is unclear if these predators were cooperative hunters or hunted as individuals or in small groups (Bocherens, et al. 2011). Given the broad range of environments occupied by modern lions it seems likely that predation behavior varied with environment. These predators would be a threat to both adult and infant humans. Wolves, if hunting cooperatively, would also be a predation threat to unguarded individuals and individual wolves might be capable of taking a smaller child. Cave hyenas were probably nocturnal and were likely obligate scavengers, and therefore posed no threat to juveniles or infants. While the presence of predators might have limited the degree to which children foraged unsupervised, it seems probable that children could have foraged alongside adults, or close to adult groups on a daily basis. Early Upper Palaeolithic Neanderthals were not defenseless – they carried spears and were quite capable of throwing rocks – which might encourage local predators to seek less obnoxious prey. It is possible that Neanderthal children foraged near or with other adults, once they had developed the stamina necessary to participate in foraging trips. In this way, they could supplement parental provisioning and acquire valuable calories and nutrients necessary to support a compact but muscular body and a large brain. Nourishing a demanding brain In comparison with African apes, humans have large brains, slow growth, longer lifespans, a higher quality diet, and different foraging behavior, which reflect the energetics of human life and the constraining relationships in terms of energy: balancing growth, maintenance, activity and reproduction (Aiello and Wells 2002; Aiello and 57 Wheeler 1995). A large body, small gut and reduced teeth are present in Homo ergaster, indicating consumption of a higher quality and more digestible diet which would include carbohydrates as well as fat and protein from meat. If this is a pattern established in early Homo populations, it can be argued that Neanderthals and modern humans shared a common ancestor in which late maturity, a long childhood and dependence on a high quality diet were established. Large brain size is strongly associated with allomaternal care in both primates and social carnivores in the form of direct aid, such as provisioning, or indirect aid that includes carrying children, child care, or protection (Isler and van Schaik 2012). The cooperative breeding hypothesis posits that energy subsidies allow for a higher degree of encephalization without a reduction in reproductive rates. Humans are cooperative breeders with secondary altriciality in the ability to thermoregulate and achieve a high metabolic rate, but with very immature brains at birth. A human brain at birth is 28% adult size in contrast with brain size at birth size of 40% adult size for chimpanzees. Therefore human infants are more helpless, but maternal provisioning and alloparenting, particularly of weaned offspring, enables the brain to grow and develop during childhood. Some authors argue that this pattern of cooperative care and communal living was established in archaic Homo sapiens as early as1.77 MBP, in populations ancestral to both Neanderthals and modern humans (Bribiesčas, et al. 2012; Isler and van Schaik 2012). The relative lack of sexual dimorphism in humans and ancestral Homo (in contrast to the great apes) suggests a relatively high amount of paternal, as well as maternal, investment in offspring. While research has focused on the role of grandmothers in the reproductive success of their daughters’ offspring, paternal provisioning can also have a significant impact on child growth and development. Ethnographic data shows that modern hunter-gatherer men spend more time in proximity to, or directly caring for, young offspring (5-20% of waking hours depending on social organization, ecology and subsistence practices) than pastoralists or horticulturalists (Bribiescas, et al. 2012:428). 58 Another major developmental issue is the development of skill competence in Neanderthals and modern humans. As discussed above, ethnographic data shows that modern humans develop maximum skill competence in foraging efficiency in their mid20s (women) and in hunting efficiency after the age of 30 (men). This is after the average age at first reproduction, which is 19 years. Food sharing (alloparenting and meat redistribution) therefore plays a vital role in maintaining child growth and survivorship. Neanderthal children clearly remained dependent on the larger group for food and protection for some time after weaning. Food sharing and alloparenting indicate that there were mechanisms in place to control for group membership and, perhaps, fission and fusion of family and band sized groups depending on the ecology and resources available. It seems likely that such a behaviorally flexible species would have a variety of forms of group organization, dependent on local and regional social and economic factors. For other mammals, including primates, skill competence is achieved prior to full physical maturity and age at first reproduction. Late age of first reproduction is commonly linked to a high intensity niche, such as complex foraging, which is related to the need to learn how to acquire energy from the environment and to mitigate risk, Among mammals, co-operative hunting species reach skill competence significantly later than independently foraging species, and the age of skill competence is associated with the level of gregariousness, obligate sharing and slow growth development (Schuppli, et al. 2012). Modern humans combine all three factors, and the large brains of Neanderthals and modern humans imply an increased adult lifespan. Intergenerational transfers of food would be vital for the development of the human lifespan where competency occurs later than first reproduction. This complex niche may have been established with Homo erectus, again predating the evolution of both Neanderthals and modern humans. 59 Provisioning and group organization Another indication of cooperative hunting and food sharing with less mobile individuals by Neanderthals is the size and choice of prey species in the later Mousterian. Neanderthals were taking prime age large herbivore: bovids weighing between 600 and 1100 kg, red deer (100-340 kg), horse (200-350 kg), and reindeer (80-220 kg) (Hofmann 1989). The estimated mass of Neanderthals ranges from 54-65 kg. Studies of African carnivores (Owen-Smith and Mills 2008; Radloff and du Toit 2004) show a relationship between predator and prey size. Only cooperative hunters such as lions and spotted hyenas took large animals (over 1000 kg in mass) and took far more herbivores in the 100-900 kg range. Non-cooperative hunters were far less likely to take prey three or four times their body mass. If this applies to humans (and isotopic studies put Neanderthals in the top predator category) the prey choices indicate cooperative hunting to successfully acquire large animal carcasses. Transportation of all or part of the carcass from the kill site to an occupation site indicates some form of provisioning of less mobile members of the group. The consistent taking of large game also indicates that enough hunters were achieving skill competence and transferring this knowledge to younger members of the group. Based on modern hunter-gatherer ethnography this indicates that group members were surviving into their thirties. There are less data on gathered foods, but cementum studies indicate the use of plants in Neanderthals’ diet (Power et al. 2012). The presence of carbonized macrobotanical remains in hearths implies the use of containers to transport items to a campsite for processing. Palaeoanthropologists also need to consider just how Neanderthals and other hominins were transporting tools and raw materials, given that we never explicitly state that any containers were used; despite many use-wear studies of stone tools that indicate dry hide working, which would indicate production of clothing or containers for transportation. Again, if ethnographic analogies can be applied to Neanderthals, this suggests that some provisioning and sharing, perhaps at the intra 60 family level rather than the intra group level operated. Further, we could argue for evidence of alloparenting and it would be interesting to attempt to model the ages at which Neanderthal children may have begun to forage for themselves based on latitude and local topography. Unfortunately time and space does not allow for such a consideration here, but given Neanderthal occupation of areas with clear landmarks and varied environment, it seems likely that Neanderthal children could have foraged for plant foods and small game alongside adult foragers and thereby supplied calories necessary for growth while learning skills necessary to raise their future offspring. Other evidence for group provisioning is supplied by the presence in the fossil record of individuals incapacitated by age, disease or injury, most notably the Old Man of La Chapelle and Shanidar I. The survival of injured individuals or disabled individuals indicates social care within the group. However, other burials (Shanidar 3, and St Césaire 1) show evidence of intra-human violence, the latter dying from a thrown spear but the former surviving a blow to the skull (Churchill, et al. 2009; Zollikofer, et al. 2002). Such injuries point to incidences of stress and competition within the group and possibly between groups. Comparative studies of post-cranial skeletal morphology and physiological stress indicate that Neanderthals are most similar to modern hunter-gatherer groups that are intensive foragers in small territories. Both male and female specimens show a pattern of activities similar to broad spectrum foragers, and in fact, may have foraged more intensively than contemporary modern humans (Pearson, et al. 2008:151). Injury patterns also suggest that male and female Neanderthals had very similar activity patterns, although the general frequency of ante-mortem trauma is not high and generally reflects minor injuries rather than major incapacity (Estabrook 2009). Interestingly female Neanderthals have a higher number of fractures in the fibulae, suggesting more frequent falls, which implies a great deal of mobility over uneven ground, consistent with a high degree of foraging. 61 The provisioning of individuals and the implied social relationships leads to the question of socialization. The modern human brain modulates emotion and this appears to be a primary human characteristic that allows complex behaviors and interaction. These are necessary in negotiating group living which, while it has many benefits, also has considerable costs in terms of stress and conflict resolution or avoidance (Gamble 2011; Gowlett, et al. 2012). The theory of the social brain for Neanderthal brain size indicates that the average distributed group size was around 150, and that Neanderthals communicated using speech and metaphor (Gamble 2011:161). However this average distributed group is argued to be the limits of cognitive loading, i.e. person-to-person interaction for Neanderthals, rather than a mating network within a fission-fusion society. Neanderthals therefore had to negotiate the dissolution and amalgamation of larger and smaller groups, presumably related to seasonal variation in resources in a manner similar to modern hunter-gatherer societies. What remains different is the degree to which material culture was entrained in the Neanderthal social network, where exotic raw materials are less common in the lithic assemblage, and the use of personal adornment more ephemeral, and the degree to which the fission:fusion organization of Neanderthal bands enhanced or impeded mating networks outside the immediate region. The genetic evidence for interbreeding between Neanderthals and modern humans might suggest that Neanderthal behavior was not so dissimilar from modern behavior. As a result, individuals from either population were found to be acceptable mates. Was life really nasty, brutish and short Analyses of Neanderthal life and physical interactions with the environment followed the dominant paradigm of “primitive” behavior that relied far more on physical strength and less on the assumed “advanced” behaviors of modern humans who relied on culture and technology (Estabrook 2009). “For many, anatomically modern humans arrive like the rising sun” (Gowlett, et al. 2012:694). Early studies focused on the 62 apparent and real pathologies present in specimens with the result that trauma appeared ubiquitous in Neanderthal populations. This assumption appeared to be confirmed by a study that found trauma on Neanderthal skeletons to be similar to that of rodeo riders, suggesting direct contact with prey (Berger and Trinkaus 1995). However this sample was not compared with trauma on contemporary modern remains. Later research found that there was no significant difference between the two datasets in terms of trauma. Neanderthals, despite their greater musculature, were throwing or thrusting heavy spears and had the same injury patterns as atlatl-wielding Late Upper Palaeolithic populations. Neanderthal hunting behaviors were not responsible for the relatively moderate trauma visible on the fossil remains, which should be examined in terms of each individual, as the injuries probably relate to a variety of causes (Trinkaus 2012). Studies of injuries in Neanderthals show a cumulative acquisition of relatively minor trauma over a lifetime, with the oldest individuals showing the most injuries (Estabrook 2009:226). In terms of demographics, burial data indicate that most Neanderthals in the fossil record lived to maturity (at least 35 years) and there is no difference between Neanderthals and Pleistocene modern population mortality patterns in terms of age at death (Trinkaus 2011). This again shows that Neanderthals had the time to develop and pass on the knowledge needed to both forage and hunt effectively. Neanderthal trauma shows no statistical difference with modern foragers, where most injuries are the result of accidental falls or occupational stresses (Estabrook 2009). For example, the higher frequencies of injuries to fibulae in female Neanderthals might relate to the amount of walking (and potential for falls) required to gather food during foraging trips. These injuries suggest that Neanderthal females spent considerable time moving across the landscape. Physiological studies have been undertaken to determine the degree to which Neanderthals used projectile technology but the data is equivocal. Studies of humeri morphology suggest that only Late Upper Palaeolithic individuals used consistent spear 63 throwing motions to a degree that bone structure was affected, while both Neanderthals and early Upper Palaeolithic hunters were more reliant on close range hunting or spear thrusting to kill prey (Rhodes and Churchill 2009; Schmitt 2003). Adoption of the atlatl appears to occur later in the Upper Palaeolithic, with atlatls becoming common in the Gravettian. Physiological studies therefore indicate that there was little difference in the nature of the hunting equipment and strategies of both Neanderthals and early Upper Palaeolithic hunters. Both appear to have employed a variety of strategies and armatures to acquire animal protein. Recent evidence indicates that Neanderthals utilized different forms of armatures to complete different hunting tasks, utilizing both thrown spears and thrusting spears or pikes. In Northern Iberia, lithic analysis indicates a flexible hunting behavior utilizing two types of hunting tools – a “conventional” small point and a large and thick point. This indicates two different hunting strategies, with the larger point used like a pike and the smaller point used on a thrown spear (Lazuén 2012:2308). But does the focus on hunting behavior in the physiology of the Neanderthal arm obscure other repetitive behaviors? This interpretation of arm morphology as a product of hunting behavior has been called into question through the study of muscle activity, which suggests the morphology of the humerus seen in Neanderthals is the product of scraping activities rather than spear thrusting (Shaw, et al. 2012). Given the repetitive and intensive movements required to process hides described in the ethnographic record, this interpretation of skeletal morphology may be as valid as, or more valid than, the focus on hunting as an explanation for physiological traits. Hide working is another acquired skill that requires time to master. The question then becomes, at what point was it necessary to transform worked hides into shelters and containers that required additional tools to create effective energy retention in the form of clothing, or food and tools in the form of sacks or bags. Further, when would this behavior become archaeologically visible? 64 The Neanderthal who came in from the cold Neanderthal physiology reflects selection for cold adaptations that operated on both Neanderthals and modern human populations in Northern Europe, although genetic drift was probably also a factor in Neanderthal cranial and post-cranial features (Weaver 2009; Weaver and Steudel-Numbers 2005). Despite a physiological response to the cooler climate, ambient temperature estimates for interglacial periods indicate that some form of clothing, footwear and blankets would be required in the northern part of the Neanderthals’ range (Sorensen 2001). Bedding is also known from late Mousterian Cantabria (Cabanes, et al. 2010) in association with hearths, where bone was probably used as fuel. A similar pattern may be present at other sites where empty spaces occur between hearths (Cabanes, et al. 2010:2955; Vallverdu, et al. 2010). Hearths, particularly large hearths, also indicate group investment, given that a continually burning fire might require between 50 and 100 kg of wood per day (Gowlett, et al. 2012:705). Here we see evidence of foraging behavior to supply energy or conserve energy not by direct consumption, but by the use of shelters, insulation and combustion. Energy can be regarded as a form of liquid capital, which buffers an individual or group against external fluctuations (Wells 2012:469) through storage of adipose tissue; through behaviors such as cooperative breeding; or delayed return in the form of storage in containers. Over the human lifespan, these forms of energetic capital are intertwined: neonates and babies store large quantities of adipose tissue, and cooperative breeding lessens the energetic cost to the mother. But this behavior increases the energetic investment of other individuals. How to retain this energetic capital effectively in nontropical environments? Energy is lost via radiation, conduction and convection. Reconstruction of past environments and temperatures focusses on ambient temperature and calculated windchill. To date, I have seen no discussion of atmospheric humidity. While not denying the impact of windchill on the need for clothing, personal experience of a maritime 65 climate underscores the impact that higher humidity in winter conditions can have on heat loss and general comfort. In Western Europe this can be as severe a problem as windchill (in terms of the danger of exposure) and would encourage the use of some form of clothing in what appear (on paper) to be milder temperate environments. Clearly the need for clothing will vary by latitude and local climatic conditions. Recent research argues that in the Mousterian Neanderthals would require at least 70% of the body to be covered in northern territories, in addition to head coverings, hand protection and foot coverings (Wales 2012:789). This study also concludes that ineffective clothing was not a prime mover in the demise of the Neanderthals. Wales argues for relatively untailored clothing based on the absence of awls, but his analysis included Châtelperronian sites within the larger Mousterian assemblage. Further, he does not discuss the Arcy osseous assemblage which clearly indicates hideworking by Neanderthals. His database is also sadly lacking in temperate hunter-gatherer groups, a factor of modern hunter-gatherer population distribution. The inclusion of Plains and Midwestern horticultural groups would have been most informative in understanding how clothing is used by mid-latitude hunter-horticulturalists. The appearance of formal tools associated with the manufacture of containers from hide and other organic materials occurs in the Châtelperronian during a period of climatic instability. The Grotte du Renne is located near the northern boundary of Châtelperronian sites and is relatively late in date. The occupation occurred during a cold phase, and the site represents a series of winter occupations (Francine David, pers. comm.). Under these circumstances, the presence of awls associated with the construction of shelters and better constructed clothing should not be surprising. Neanderthals had a long history of hideworking, as indicated by lithic use-wear analysis (Texier, et al. 1996) and were capable of flexible behavior. To modify hides into shaped clothing would not be beyond their capacity, and could have been a simple response to local ecological conditions. 66 Conclusion Current data on Neanderthal ontogeny, physiology and energetics, combined with ethnographic data on modern hunter-gatherer lifeways and models of Homo erectus behavior indicate that Neanderthal life histories were probably very similar to those of modern hunter-gatherers. Neanderthal children grew at similar but slower rates than modern children, and it appears that they matured physically at a similar rate. Growing children absorb a lot of energy, and the involvement of both parents, older siblings and older relatives or other group members, would be important in successful child rearing. Given maturation rates and the time required to become an effective forager or hunter, ethnographic models also indicate that Neanderthals’ first reproduction occurred prior to their most effective period of foraging/hunting, as with modern humans, and that adult Neanderthals lived long enough to become extremely effective large game hunters, who could communicate this knowledge to their offspring and other younger group members. Based on modern ethnographic analogies, some biological constraints were placed on Neanderthal women’s mobility during their child-bearing years, resulting in a gendered or biological division of labor, with cooperative provisioning by subgroups sorted by gender/age/mobility focusing on large game procurement, and more sessile resources (small game, vegetable resources). This operated within a fission-fusion huntergatherer society where smaller groups formed weak associations with larger social groupings. Exotic raw materials occur in low frequencies at Mousterian and other Neanderthal sites which indicate some larger networks in place, but not to the degree expressed in Aurignacian sites. Clearly differences between modern humans and Neanderthals were a matter of degree rather than absolute differences. The question remains: were these differences enough to result in the disappearance of one species? Did modern humans out-compete Neanderthals, or are palaeoanthropologists fishing for trout in milk? In other words, apparent correlation does not imply causation. 67 Neanderthals were effective hunters and gatherers who show little difference in their subsistence behaviors from contemporary modern humans. Like modern humans, Neanderthals required some form of external covering to survive in Europe, even during temperate periods. The need for clothing, particularly in the later Châtelperronian would result in the development of tools to manufacture adequate protection from an unstable and cooling environment in the northern ranges of their territory. The retention of the Mousterian in Iberia is probably a reflection of less pressure to adapt to a changing environment. At the same time as Neanderthals were facing the problem of an unstable ecological niche, modern humans faced the same problems. Based on the ongoing reevaluation of Neanderthal lifeways and capacity for flexible behavior in the recent literature and conference proceedings, one has to ask why it would be surprising that the two human species solved similar problems in similar ways? When examining the use of animals as a source of both raw materials and protein, this apparent parallel evolution of subsistence and technical behaviors should be visible in the archaeological record. In the next four chapters I will present the two datasets that form the basis for my faunal analysis of the sites at Abri Cellier and the Grotte du Renne, Level Xc, and the results of these analyses. I will then discuss the evidence for the selection of tool supports and place this in the larger context of tool manufacture in the Châtelperronian and Aurignacian. 68 CHAPTER 5: NEANDERTHALS, MODERNS AND BONE TOOL USE: THE RESEARCH PROJECT Introduction In previous chapters I have explored the debate surrounding the evidence for innovation or adoption of bone tool use by Neanderthals and the implications for their contact or lack of contact with modern human populations in early Upper Palaeolithic Europe. Bone tool use implies the working of fragile materials such as hide or plant fibers, and the acquisition and processing of hides is an important part of subsistence strategies among modern Boreal and Arctic indigenous hunter-gatherers or pastoralists. Neanderthals and modern humans evolved in parallel in Europe and Africa and shared a suite of behaviors derived from their common ancestor. Differences between the two species are nuanced and reflect a difference in the intensity of certain social behaviors. There is little difference in subsistence behavior or lithic manufacturing. There may be differences in symbolic behavior, but, to date, this alone cannot fully explain the demise of the Neanderthals. Many explanations regarding the disappearance of Neanderthals from Europe between 40,000 and 30,000 BP remain colored by previous perceptions of Neanderthals’ lack of ‘humanity’ or an inability to innovate or react rapidly to new socioecological conditions. Most explanations also seek a single primary factor to explain the extinction of Neanderthals, which may be inadequate for a process that occurred over a period of approximately 10,000 years. All explanations assume some form of behavioral, technological or intellectual superiority on the part of modern humans. As we have seen, one highly contested area is the use of bone tools by Neanderthals in the Early Upper Palaeolithic in France, represented by the Châtelperronian. Debate has focused on the degree to which the appearance of bone tools in the Châtelperronian is the result of acculturation, in other words contact between Neanderthal and modern human societies 69 in Europe. Some argue for contact where Neanderthals adopted new technologies (e.g. Mellars 2000b). Others argue for independent innovation (e.g. Zilhao, et al. 2006). There is little discussion of the mechanisms for acculturation (should it have occurred), nor is there any discussion of why bone tools become part of the subsistence practices of either group. Clearly this new technology was developed as a solution to a suite of problems that faced both late Neanderthal populations and the incoming modern human populations in Europe. Ethnographic evidence and use-wear data show that bone tools are largely associated with the manufacture of clothing and other containers from organic materials (animal hides or plants). This indicates an expansion of these societies into a new ecological niche, or an adaptation to a new set of ecological circumstances (likewise a new niche). Description of the project This research project examines two datasets to determine what, if any, differences there were in the choices of supports (blanks) of bone tools and what the use of these tools represents in terms of subsistence practices and social organization. One data set is derived from the Aurignacian site of Abri Cellier. The second data set is the fauna from the lowest Châtelperronian level of the Grotte du Renne, Arcy-sur-Cure. Figure 5.1 shows the location of the two sites. The research will focus on the selection of prey animals and elements from the prey species to determine if there are differences in how Neanderthals and modern humans selected elements for use as bone tools. This will not include the use of antler to make projectile points, a practice that emerges after the initial colonization of Europe by modern humans. This thesis also seeks to integrate zooarchaeology with the study of worked bone, by examining if any particular species or elements were selected consistently as sources of raw material, or if bone tools were manufactured in a more ad hoc manner on any available support. The absence of close examination of bone shafts by faunal analysts for evidence of ad hoc tools also reduces 70 our understanding of how bone served in the manufacturing process. Studies of faunal remains for Early Upper Palaeolithic sites in Europe generally fall within two categories: first subsistence (including reconstruction of, and interaction with, the local environment); and, second, the use of bone and antler for tools and personal adornment. There rarely any integration of these areas of study: animals are Figure 5.1. Map showing the locations of the Grotte du Renne and Abri Cellier. Source: Google Earth 71 eaten in a warmer or cooler environment, or people used antler or bone tools while wearing personal ornaments made from teeth or bone (e.g. David and Poulain 1990; Julien, et al. 2002). When animals are considered as sources of raw material, it is often in terms of acquisition of shed or unshed antler (White 1983). In studies of bone and antler tools there is attention to form, but not to function, of tools other than projectile points (e.g. White & Knecht 1992). This thesis argues that reconstructing Early Upper Palaeolithic subsistence should combine both direct evidence for animal exploitation and consumption and the degree to which fauna provided raw material for the production of other artifacts, particularly tools for container manufacture or the production of rawhide or leather for clothing, shelter or containers. It is becoming apparent that the study of Neanderthal behavior requires close attention to the nuances of the archaeological record. For example, closer examination of faunal remains indicates exploitation of particular bird species for feathers for decorative and/or symbolic purposes. The identification of such behavior in Neanderthals requires an additional level of interpretation of the data, and this may also be the case for use of bone tools to manufacture containers. Extending faunal analysis to examine behaviors that conserve energy through the use of animals as a source for tools and raw materials will enhance our knowledge of early Upper Palaeolithic subsistence and scheduling behavior. This research project also has the potential to provide insights into similarities differences in subsistence and technological behaviors between both between Neanderthals and anatomically modern humans, and between different ecological zones in early Upper Palaeolithic France, using fauna from the lowest Châtelperronian level (level Xc) of the Grotte du Renne, Arcy-surCure, Yonne, France and Aurignacian I and Aurignacian II faunas from the site of Abri Cellier, Dordogne, France. This thesis will therefore focus on Early Upper Palaeolithic faunal exploitation both for direct subsistence and on fauna as a source of raw materials for tool 72 manufacture; through analysis of these faunal assemblages. Subsistence is the “direct link between humans and their environments” (Enloe 1993:103). The extraction of energy from the environment is basic to human survival. Fulfilling this task involves movement within the landscape, social organization, hunting and processing technology and other methods of food procurement. Ethnographic and ethnoarchaeological data on modern and/or recent hunter-gatherer societies provide models on how members of a group operate to meet the nutritional requirements for their immediate families and members of the larger social group. The provisioning of offspring for a considerable period of time after weaning requires alloparenting. Children also have to acquire skills to enable them to become productive adult members of the society. Modern humans are unusual in the late acquisition of competency in provisioning, which post-dates the average age of first reproduction. The late attainment of efficient hunting skills is particularly notable, which has implications for the transfer of knowledge to subsequent generations. Adult hunters would have to survive into their late 30s to be able to pass on their full suite of skills to younger individuals. At the same time, they would be vital in provisioning less mobile group members, such as children or nursing mothers. The survival of offspring therefore rests not simply on the parents but some form of alloparenting, by fathers, grandmothers and other members of the residential group. Ethnographic data also serves as a source of information on the production of containers. Here I use the term in the broadest sense (c.f. Gamble 2007) to include shelters and clothing as well as items used to transport food or equipment. Research on published data has made it apparent that production of basketry using botanical raw materials does not require any more specialized equipment than a sharp thumbnail and an ability to weave (Liapunova and Miklukho 1996). In contrast, rawhide and leather containers require the use of tools to process hides into material suitable for working, and tools to manufacture containers, usually bone tools to prevent tearing of the fragile hides or skins. Hideworking is part of an overall subsistence strategy, and can be time 73 consuming dependent on the end product desired. The integration of hideworking into the overall subsistence pattern will be examined in terms of time and energy required to process animals skins and also the necessity for the majority of human societies to conserve energy through external coverings. Hide production could therefore be seen as a means of conserving the energy acquired through hunting. By examining fauna as a source for tools, we as archaeologists can extend our examination of subsistence to include evidence for manufacturing of cordage, containers and infer improved storage (delayed consumption) or improved transportation of materials (Soffer 2004; Soffer, et al. 2000 a and b). One of the major differences between the Aurignacian and earlier or contemporary Neanderthal cultures is the amount and degree of bone and antler working (Marean and Assefa 1999; Peterkin 2001). Speth (2004) has discussed how analyses of bone working and symbolic behavior perpetuate the false assumption that performance alone indicates capacity to produce without assessing what impetus, cultural, ecological or otherwise, may encourage the adoption or rejection of new technology. Given the production of bone tools in the Châtelperronian, Neanderthals clearly had the capacity to work bone but the nature of the performance may differ from the production of bone tools in the Aurignacian. Even within the Aurignacian, production of bone and antler tools appears to have been differentiated between armature (antler) and bone tools (manufacturing) (Tartar, et al. 2006) Studies of Aurignacian and later worked bone focus on hunting technology (e.g.,Knecht 1991; Knecht 1993) or symbolic behavior (e.g.White and Breitborde 1992). With the exception of Soffer (2004), little research has focused on other uses of bone tools beyond the method of manufacture and possible use (chaîne opératoire) (e.g. d’Errico, et al. 2003, Julien, et al. 2002; Liolios 2003, 2006; Tartar 2009, 2012; Tartar, et al. 2006). The focus is more on hunting and less on hides. The continuing privileging of hunting technology obscures the capacity that non-projectile bone technology has to 74 provide data on other subsistence behaviors. The addition of a new ‘toolkit’ implies a new range of behaviors that, while not necessarily a major departure from previous subsistence behavior, could provide some adaptive advantage that, in the long term and combined with other behavioral changes, favored modern human population growth at the expense of Neanderthal groups. Many studies of Early Upper Palaeolithic worked bone have focused on the apparent symbolic aspect of personal adornment and art in the context of social organization and expression of group membership. Methods of production, the chaîne opératoire, have been reconstructed but there has been little discussion of possible function of regularly shaped and incised bone and antler tools, with the exception of projectile points (Julien, et al. 2002; Knecht 1991; Stettler 2000; White 2002). This is surprising given the investment required in terms of time and lithic tools to produce worked antler and bone tools (Griffitts 2006, 2007; Guthrie 1983; Knecht 1991, 1993), not to mention the investment in clothing or container manufacture that these manufacturing tools represent. Ethnography, ethnoarchaeology and experimental studies indicate that bone tool manufacture and technology require specialist knowledge of the properties of bone and antler and manufacturing techniques that utilize these properties (Currey 1990; Currey 1979; MacGregor 1985; MacGregor and Currey 1983). The use of bone tools implies access to a broader range of technology and behavior than currently considered in the Upper Palaeolithic. Soffer has demonstrated that bone implements from the German Aurignacian show evidence of wear damage analogous with use as a batten in weaving tools from ethnographic collections (Soffer 2004:410). Ethnographic studies indicate that bone tools are preferred to work fragile materials. Clothing was essential to occupation of temperate Middle Palaeolithic Europe, according to recent publications (Gilligan 2007, 2008, 2010; Wales 2012). It is possible that the appearance of bone tools in the 75 Châtelperronian reflects an increasing reliance on more efficient clothing and shelter as a means of adaptation to an unstable and cooling environment. Aurignacian and Châtelperronian lithic assemblages are rich in burins, scrapers and notched tools, all used to make other tools. The adoption of bone and antler as raw material for tool making, also to produce tools to make other tools, argues for an extension of the chaîne opératoire to examine not simply how tools are made but how raw materials are amassed and employed to create a suite of tools, and what those tools are used for. I believe that integration of faunal material into the toolkit argues for change in subsistence and technological organization that may have resulted in a longterm adaptive advantage. Jochim has argued that “perhaps the most promising approaches [to the Neanderthal/modern transition] are those that focus on changes in the behavior of organization” (Jochim 1988a:275). The addition of bone and antler to the suite of raw materials represents a change in organizational behavior that could, in part, explain the long-term success of anatomically modern humans at the expense of Neanderthals. The increase in these tools implies shifts in the organization of subsistence to allow for time to produce these new artifacts. There may also be implications for a sexual or otherwise gendered division of labor based on mobility and ability to process hides and manufacture containers. Research hypotheses and testable models This research project will analyze the faunal remains from the Aurignacian site of Abri Cellier, Dordogne, France and level Xc of the Grotte du Renne, Arcy-sur-Cure, Yonne, France to differences and similarities in subsistence behavior between two groups of hominins: Neanderthals (represented by the Châtelperronian) and modern humans (represented by the Aurignacian) in terms of exploitation of animals for direct subsistence (food) and as a source of raw materials for bone and antler working technology. While little difference is expected in terms of basic subsistence behaviors, there may be 76 differences in the manufacture and degree of use of bone tools that have implications for broader subsistence strategies. This research proposal will test the three null hypotheses stated below: Testing Null Hypothesis 1 Null Hypothesis 1: The faunal remains at Abri Cellier and Level Xc, Grotte du Renne are solely the product of hominin behavior. The faunal assemblages at the two sites are assumed to be the product of human hunting behavior. This will be tested by examining the taphonomy of the assemblage to confirm hominin agency as the prime depositional factor. The analysis will also consider the evidence of human and non-human transportation and consumption and assess the impact of post-depositional natural and cultural practices (including recovery techniques) on the final composition of the assemblage. The presence of large amounts of carnivore damage on bones and a lack of butchery marks will indicate significant input by other predators. This will invalidate Null Hypothesis 1. Testing Null Hypothesis 2 Null Hypothesis 2: There is no difference in subsistence behavior in terms of exploitation of animals for food between the Châtelperronian and Aurignacian. Both cultures followed the same subsistence practices over time and space. Any apparent differences will be explained by environmental change, not differences in social organization. If there is little evidence for non-hominin accumulation of the faunal material, it is assumed that both levels of Abri Cellier are the result of low intensity foraging for resources within the local environment. The same species, same age cohorts and the same elements will occur in both levels at the site. The two levels will be compared with the fauna from Level Xc of the Grotte du Renne to examine if any differences are simply the result of responses of the same subsistence strategy in time or in space. If differences 77 occur that cannot be ascribed to the local ecological setting Null Hypothesis 2 will be invalidated. Testing Null Hypothesis 3 Null Hypothesis 3: There is no difference in the selection and use of bone for tools between the Châtelperronian and Aurignacian. Selection for raw materials will be the same and both cultures will use tools for a similar suite of manufacturing and subsistence behaviors. The faunal assemblage will be examined for evidence of bone working debris, selection of particular elements for tool making and the types of tools present. Usewear analysis will be examined to assess similarities and differences in production sequences, tool type and degree of bone tool use between the upper and lower layers. These data will then be compared to data from the Grotte du Renne and other Early Upper Palaeolithic sites to place bone tool manufacture and use at Abri Cellier and the Grotte du Renne within a larger temporal and regional context. If bone tools suggesting use of containers, and delayed consumption of resources as an adaptive advantage, only occur in the Aurignacian Null Hypothesis 3 will be invalidated. It is possible that Null Hypothesis 3 will be partially invalidated – there may be use of bone tools the Châtelperronian but to a lesser degree, or vice-versa. Conclusion In summary I would expect little, if any, difference in faunal exploitation between Abri Cellier and Level Xc of the Grotte du Renne. If differences do occur, these must be examined in the context of variation in the game animals exploited based on changes in the larger animal community over time and the degree of intensification or other evidence for changes in hunting organization in comparison with other faunal assemblages. Further, at what stage do these differences (if any) become significant in explaining 78 cultural behavior that may or may not contribute to the long-term survival of modern humans but not Neanderthals? Based on the archaeological record, I expect that there will be significant differences between Aurignacian and Châtelperronian bone tool manufacturing techniques and the use of bone and antler tools that reflect more intense use of this technology by modern humans. Previous research suggests a difference in degree of bone use, and the introduction of antler points in the Aurignacian, with no evidence for antler working in the Châtelperronian. However, the question remains as to how significant these differences are. Do they reflect capacity to produce new items or simply different solutions or responses to similar subsistence or social requirements? The following chapters will consider general issues involved in taphonomic analysis, describe the two sites, and present the results of the faunal analyses. The selection of skeletal parts for bone tool blanks will then be assessed. Finally, the concluding chapter will assess how many of the null hypotheses listed above are validated on invalidated. And so we leave hunter-gatherers past and present and travel to the banks of the River Cure, in northern Burgundy and the Grotte du Renne, pausing to consider taphonomic factors that impact faunal assemblages. 79 CHAPTER 6: TAPHONOMIC ISSUES, A REVIEW Introduction The following four chapters will describe the history of excavations at the Grotte du Renne and Abri Cellier and the results of the faunal analysis. It is necessary to consider the variety of processes that produced the different assemblages, and to keep these processes in mind when considering the excavation methods and the underlying goals of the excavators. As Gifford (1981:424) has noted, taphonomic processes are complex, and any faunal analysis will be a rather complex exercise. The development of the study of site formation processes is intertwined with the development of archaeology as a discipline, from its beginnings as a science in the nineteenth century. These processes, now subsumed under taphonomy for faunal remains, seek to understand how biological material is incorporated into the archaeological record from the time of death of the individual, through processing for meat, fat and raw materials, and consumption and initial discard, followed by post-depositional processes that result in the final burial and subsequent recovery by archaeologists. Faunal remains associated with stone tools and/or hominin fossils were uncritically assumed to be the product of hominin behavior in the late nineteenth and much of the twentieth century. More recently, awareness of the complexity of the taphonomy of faunal assemblages and the problem of equifinality, where different agencies can produce similar assemblages, have resulted in the re-evaluation of faunal deposits associated with stone tools and/or hominin fossils (Gifford 1981; Grayson 1986; Lyman 1994). The theoretical emphasis on culture history and trait lists in both Europe and North America archaeological practice in the nineteenth and much of the twentieth centuries reduced interest in animal remains. This was despite Lartet’s attempt to use literal type-fossils as chrono-stratigraphic markers, dividing the Upper Palaeolithic into 80 Cave Bear, Woolly Mammoth and Rhinoceros, Reindeer and Aurochs/Bison periods (Reitz and Wing 1999:16). While there was little interest expressed in taphonomy or site formation processes in the zooarchaeological literature of the early twentieth century, some concern was expressed regarding the need to recover all faunal material, not just elements that the archaeologist felt could be identified, and for fauna to be analyzed by an expert (Olsen 1964). However, even in the 1960s unmodified faunal remains were often discarded and many descriptions unquantified (Reitz and Wing 1999:20). These are general observations, but the continued emphasis on lithic technology and culture history hindered the integration of faunal remains into Palaeolithic studies. Animal remains in association with stone tools were interpreted as the product of human or hominin behavior Pliocene and Pleistocene with no consideration for other taphonomic factors (for further discussion see Binford 1981:8-12). Major developments in taphonomy in the early twentieth century occurred within palaeontology. Weigelt defined biostratinomy (factors from death to final burial) and diagenesis (changes after final burial) and Efremov defined taphonomy at the study of the transition of organisms from the biosphere to the lithosphere (Efremov 1940:85, Weigelt 1989). As with archaeology, palaeontology moved from a period of description and classification to a research focus on reconstruction of past communities and population dynamics. Actualistic studies examined differences between living communities and death assemblages of vertebrates and invertebrates, considered transport and attrition, skeletal disarticulation and sedimentology. The inclusion of taphonomy in analysis of fossil assemblages led to a new and better understanding of the fossil record, and of population dynamics and ecological reconstruction. However, Lyman (1994) has argued that the paleontological tradition of taphonomy has a tendency to focus on bias in the fossil record and how to detect and compensate for such biases (Olsen 1980). Recent 81 taphonomic and palaeoanthropological studies have moved away from considerations of “bias” to a more holistic approach to the processes of accumulation, preservation and deletion within the faunal record (Lyman 1994). Significant impetus for taphonomic research in zooarchaeological method and theory came from East and South Africa in the late 1960s, following the discovery of Plio-Pleistocene sites containing stone tools in apparent association with animal bones. Research on hominin localities and re-evaluation of their depositional history coincided with the development of the New Archaeology in North America and its emphasis on Middle Range Theory and understanding of processes that created the archaeological record. At the same time in Europe, a “loss of innocence” was described by David L. Clarke (Trigger 1989:358), recognizing that a body of theory was necessary to link human behavior to archaeological remains. (Behrensmeyer and Boaz 1980; Brain 1980, 2004; Gifford 1981; Haynes 2004; Lyman 1987, 1994). Actualistic studies of bone loss or survival, carcass disarticulation, bone transport agencies and ethnoarchaeological approaches to the taphonomy of human sites, plus theoretical considerations of fluvial transportation, functional anatomy and bone chemistry were applied to the larger question of East African hominin palaeoecology. In sum, these studies and similar studies encouraged a more nuanced approach to sites with bone accumulations. The role of other carnivores in the production of cave faunas and better identification of scavenged or actively predated fauna sparked a reevaluation of the role of hominins as active collectors or passive scavengers of large animal carcasses. Hunting was no longer the default interpretation for bone assemblages in the Palaeolithic record, particularly the Lower Palaeolithic. Middle Palaeolithic assemblages were also reassessed to determine evidence for hunting or scavenging behavior (Chase 1986; Stiner 1994). The broader impact of this research on North American and European archaeologists resulted in the development of 82 new analytical techniques and an increasing awareness of the complexity of the Pliocene and Pleistocene archaeological record. The growth of interest in taphonomy and environmental reconstruction paralleled developments in archaeological method and theory and, therefore, zooarchaeology. In the 1960s archaeology moved away from classification and towards explanation. The development of Processual Archaeology in North America, and similar changes in methodology in Europe resulted in a new approach to the faunal record. If culture was a means of adapting to the local environment, faunal and floral remains held important information on how humans interacted with, and extracted energy from, their surroundings. Screening and flotation were adopted as standard practices to maximize recovery of data, reflecting these new research paradigms and increasing awareness of the finite nature of the archaeological record. The perceived need for Middle Range Theory as an explanatory tool in interpreting the archaeological record produced a new focus on humans as active creators of the archaeological record. Studies of site formation processes now included examinations of hunting strategies (including linear programming or optimal foraging), butchering and transport decisions, carcass processing, discard and bone destruction; but less attention was paid to the impact of recovery techniques on the final assemblage (Binford 1981; Schiffer 1987; Trigger 1989). The early 1980s saw a fluorescence of publications on taphonomy and faunal analysis in zooarchaeology. Two extremely influential publications appeared in 1981: Binford’s study of bone modification by human and non-human agents, (Bones: Ancient Men and Modern Myths) and Brain’s study of bone accumulators in South African Caves (The Hunters of the Hunted). Both volumes cover similar topics: agents of skeletal disarticulation, patterns of bone modification and destruction by humans and carnivores, and patterns of accumulation that can be ascribed to human/hominin and non-human agency. Binford (1981) was 83 extremely influential in his use of theory, ethnographic data and actualistic studies and remains a major reference point in taphonomy. Later studies have developed from his initial research, including further work on bone density, carnivore ravaging, butchery patterns, evidence for hunting or scavenging from element distribution, and transportation decisions by humans. The growth of multi-disciplinary or interdisciplinary studies included a greater emphasis on the ecological and environmental context of a given site or cultural system. Zooarchaeology moved from simple cataloguing of elements and species within an assemblage to generating profound insights into the nature of a given archaeological assemblage and the processes that operated to produce the excavated material. This is not to say that the new approaches and use of ethnographic data and animal studies resulted in consistent interpretation of faunal remains. Even after thirty years of taphonomic research, archaeologists can still disagree on the interpretation of faunal assemblages, for example ascribing evidence for primary or secondary scavenging or hunting to the same data set through the use of different analytical criteria (Bunn 1993; Domínguez and Pickering 2003; Lupo 2002; Lyman 1987; Marean and Spencer 1991; Monahan 1998). Taphonomy and ethnoarchaeology Site formation processes in zooarchaeology have focused on identification the processes that resulted in bone accumulation through the development of methodologies to identify human and non-human agency; factors resulting in assemblage attrition; and assessing how the chemical, structural and nutritional contents of bones interact to have a major influence on bone survivorship. While ethnoarchaeologists and zooarchaeologists have conducted fieldwork, field studies and controlled experiments, palaeontologists have been addressing many of the same issues, with particular focus on the formation and representative nature of the fossil record, and how to resolve issues of time resolution. 84 Zooarchaeologists in general have been more concerned with the formation processes that created the assemblage than post-depositional diagenetic processes studied by palaeontologists. Given that zooarchaeological assemblages are frequently the product of human behavior, ethnoarchaeology has been used to examine how contemporary hunter gatherer groups produce faunal assemblages. Binford’s studies of the Nunamuit and Navajo considered the economic anatomy of sheep and goat in butchery practices and established the Meat Utility Index (MUI), refined to include marrow and fat to produce the General Utility Index (GUI) for skeletal elements (Binford 1978; Binford and Bartram 1977; Metcalfe and Jones 1988). As meat values, marrow values and bone grease values are not constant for each element, butchery practices and consumption practices can be examined by using the indices as a predictive model, or as an explanatory model for patterns observed ethnographically and archaeologically. Unfortunately there can by an analytical tendency to focus on elements or parts of elements, obscuring the fact that most carcasses are not butchered into single skeletal elements, but transported as packages. The indices serve to examine carcass dispersal in terms of maximizing access to meat, fat or both, and therefore generate clearer understanding of operational decisions made by the butcher and/or consumer (Binford 1978:19). Binford’s determination of utility, related to the prey species age, sex and condition has proved a powerful and useful tool in analyzing faunal assemblages and butchery practices. Binford’s work also highlighted the complexity of decision-making processes related to provisioning of family, seasonality and abundance of game (1978:44). Ethnoarchaeology also demonstrated that different site categories (kill site, camp site, base camp) can be identified based on the elements of bone assemblage present. Butchery and transportation decisions are situational - dependent on prey species and method of transportation. Butchery practices are also dependent on whether the kill was 85 fresh or cached - with different patterns of disarticulation related to fresh or frozen carcasses. This, in turn, relates to the differences between immediate and delayed consumption utilized in a logistical collector strategy. A further product of Binford’s ethnoarchaeological work was an exhaustive consideration of modes of bone modification (Binford 1981), where he considered modification for food processing but not modifications for tool production (1981:88). Binford’s ethnoarchaeological data were derived from northern latitude huntergatherers and mid-latitude pastoralists who followed a predominantly logistical subsistence pattern. Processing and storage of meat, fat and marrow for immediate and delayed consumption, plus provisioning of dogs, resulted in a complex sequence of carcass disarticulation, bone breakage and attrition. The high protein diet of high-latitude hunter-gatherers and pastoralists requires a high proportion of animal fat and/or carbohydrate in the diet for optimum nutrition (Speth 2012; Speth and Spielman 1983). Marrow is easy to obtain and process and is highly valued for both taste and feel in the mouth, and also contains high proportions of unsaturated fatty acids as well as the oleic acid studied by Binford (Morin 2007). In contrast, white fat (also referred to as bone grease, although it exists around major organs such as the kidneys) is not a preferred comestible, as it is hard in the mouth and less palatable (Morin 2007:79). However white fat has superior properties of conservation and this is important in decisions regarding storage and consumption and must be factored into the time and energy in terms of fuel and labor required to render down bone grease (Binford 1978:159; Morin 2007). Ethnoarchaeological work in low-latitude environments has examined meat acquisition and consumption among foraging groups who practice an encounter strategy. Here food storage is rare or not practiced and food sharing is an important mechanism to maintain group nutrition and social ties. Meat utility indices were not good predictors of bone transportation for large carcasses among the Hadza, because the meat was removed 86 from the bones at the kill site. A similar pattern was observed in Kalahari hunting and butchery practices. Decisions regarding transportation were related to the size of the carcass, size of the butchery party and the distance to camp; smaller carcasses were transported whole (O'Connell, et al. 1988). While transportation was variable and context-dependent (as proposed by Binford), three sites could be identified for the Hazda based on the faunal remains: a single use kill/butchery site containing low value elements, ambush sites (re-used) with a more varied assemblage created over time, and base camps, which contained the most variable faunal assemblage. Similar discard and transportation processes were observed among Kalahari groups, with lower limbs of animals discarded at butchery sites (Kent 1993, 1996; O'Connell, et al. 1988). Bone attrition at base camps was a result in part of further processing: boiling greasy bones to extract fat, or boiling to cook joints resulted in further fragmentation, while roasting caused little further destruction. As a corollary, boiled bone was less likely to be damaged by scavenging canids as it was grease depleted (Kent 1993). Other ethnoarchaeological studies have examined the distribution of cutmarks and tooth marks on animal bones to determine the presence of humans and other predators (Lupo 2002) and the distribution of animal bones at camp sites (Bunn 1993; Kent 1993; Yellen 1977). While Bunn and others considered transport and destruction of bones, Yellen and Kent examined the role of sharing, cooking and scavenging by carnivores on the faunal assemblage. Carnivore ravaging appeared minimal at the camp, although dogs did disperse bones (Bunn 1993, Kent 1993). Other studies of Kalahari camp sites and kill sites have raised the issue of how marrow processing may impact interpretation of the archaeological record as it reduces the number of identifiable long bones considerably (Bartram and Marean 1999). Ethnoarchaeology has clearly provided a considerable body of data in terms of bone accumulation by humans, bone butchery and disarticulation patterns, the 87 complexities of human behavior in terms of decision making, carcass part transportation and redistribution within a camp site or other living area. None of the studies explicitly consider the removal of bone from the archaeological record through tool manufacture, use and discard. Non-human agents of accumulation While humans are important agents in bone assemblage formation, non-human actors are also important and their presence leads to the consideration of equifinality, especially when considering transportation of bones and bone destruction because meat and fat are highly valued by carnivores. Animals add and subtract material from both cultural and non-cultural faunal assemblage. Carnivores in particular create bone accumulations through transportation to dens or consumption sites, and delete bones from the record by destruction during consumption, usually described with the vivid term “carnivore ravaging”. The majority of studies have examined pack animals - wolves, dogs, and hyenas. There has been a particular focus on hyenas, indicating the continuing debate on the role of carnivores in generating or impacting East and South African faunal assemblages. These studies also inform the interpretation of Pleistocene sites in Eurasia that were occupied by carnivores and/or hominins. Early studies were undertaken by Brain with his examination of carnivore damage to bones in Kuisib River villages and of bone assemblages in hyena and leopard dens (Brain 1980, 1981); while Binford (1981) examined bones gnawed by dogs in Nunamuit villages and wolf kills and published detailed description of the damage patterns. Haynes (1983) examined difference in damage patterns and element selection among a range of carnivores. Other studies of bone destruction were undertaken through observation of bone consumption by captive animals, although the interpretation of bone consumption behavior by captive, well-fed animals is questionable (e.g. Marean and Spencer 1991). Actualistic studies of predator ethology are more valuable as they provide data and 88 models of predator interaction with prey, and responses by the study taxon to competition from other predators, resource availability and prey behavior. In North America, an actualistic study of wolf and coyote kills found the variation in bone destruction was influenced by seasonal variation in prey species and competition between predators (Burgess 1999). In the Amboseli, bone survivorship decreased following a shift in predator populations (Faith and Behrensmeyer 2006). A far greater degree of damage and destruction was noted, related to the increased hyena population in the park. Carnivores are frequently responsible for the disarticulation and deletion of faunal elements, and the amount of disarticulation and damage varies by the amount of meat available, season and the size and ethology of local predators (Hill 1980; Pasda 2005). An extremely direct approach to understanding bone consumption is the examination of digested material in scats or coprolites (Schmitt & Juell 1994) or raptor pellets, which can be used to reconstruct local prey populations, identify predators and obtain information on the local environment (Andrews 1980). Recent papers by Reed (2005), McGraw (2006) and Marín et al. (2009) have examined the agency of large owls and vultures, which both produce distinctive bone damage patterns and bone assemblages, in the production of faunal assemblages. The potential presence of denning animals and roosting raptors must be considered when examining any cave fauna (Bocheniski 2005, Larounlandie 2005). The majority of animal behavior studies consider the role of carnivores, but noncarnivores also have an impact on faunal remains. Large rodents such as porcupines can create impressive assemblages of dry, non-greasy bones in their dens and exhibit preferences for particular sized bone for ease of handling (Brain 1980). The taxon present is important, as rats and mice will remove greasy bone for consumption (Klippell & Synstylen 2007). The presence of rodent-gnawed bone will indicate secondary deposition 89 of the material from primary death or consumption context. Consideration of different bone generating or accumulating agents continue, with palaeontologists and zooarchaeologist now considering the role of carnivorous reptiles in site formation processes at Plio-Pleistocene lakeshore deposits in East Africa (Stephanie Drumheller, pers. comm.). Geological processes also result in bone accumulation and dispersal and have to be considered by zooarchaeologists (e.g. Chase, et al. 1994; Enloe 2006) Studies have examined how bones become incorporated into fluvial systems and redeposited in fluvial sediments to create models of disarticulation and transport of bones (Behrensmeyer 1982, Hanson 1980). The elements present, their shape, size, orientation and evidence for rolling can all be used to assess a faunal assemblage for evidence of fluvial transportation, rather than cultural accumulation. In cave sediments, faunal analysts must consider the original structure of a cave to determine the point of entry of bones. Animal bones frequently accumulate in caves through sink-holes, where animals may enter the cave, or carcasses or carcass fragments wash into the cave. Bones also accumulate through natural mortality during hibernation or denning. The location and condition of bones, taxa present, their surrounding sediment and evidence for damage through gnawing, licking or rolling can provide important indications of the different collection agents that produce the final archaeological or palaeontological faunal assemblage (Brain 1981; Lord, et al. 2007). Post-depositional taphonomic factors While bone accumulation is a complex process and any faunal assemblage reflects the behavior of a number of actors (predators, scavengers, natural processes) much bone survival is mediated by bone density and the presence or absence of fat and grease in the bones (Cleghorn and Marean 2004). Differential preservation resulting from bone structure and bone density can produce similar assemblages from different 90 taphonomic processes – another cause of equifinality. How much destruction is the result of bone structure and bone chemistry, and how much is the product of human or animal behavior operating on bone structure and bone density? Ethnographic studies noted consistent patterns of consumption and damage by hunter gatherers and in carnivore gnawed assemblages derived from human provisioning, allowing the creation of models of destruction based on the presence of fat, marrow and flesh and the testing of hypotheses based on these models to indicate site formation processes and site use (Binford and Bartram 1977). Initial attempts to calculate bone density by water displacement were provocative (Binford and Bartram 1977). Later studies utilized photon densitometry (Lyman 1984, 1993) and CT scanning (Lam 2003; Lam, et al. 1999; Symmons 2005) to examine interand intra-species variation in bone density. This expanding body of data permits zooarchaeologists and taphonomists to calculate bone density patterns for faunal assemblages and compare observed bone survivorship with expected survivorship. Analysts can then consider how and if density-mediated attrition is operating on the assemblage, and create models to examine possible causes of any such attritional processes, while considering the range of intra-species variation. The maturity of the individual is an important factor in bone density, and bone density measurements are not consistent between juvenile and adults of the same taxon (Symmons 2005). Fetal and juvenile remains deteriorate rapidly in exposed conditions (Behrensmeyer and Boaz 1980) and, because they are less dense, are easily destroyed by carnivore gnawing. The mortality curves of a faunal assemblage, or an individual taxon, are powerful tools to generate models of human and non-human hunting and scavenging, seasonality, and the underlying processes that created the deposit. Age of death is estimated by examination of epiphyseal fusion and also calculated from tooth eruption sequences and/or tooth wear patterns or cementum ring studies. 91 Tooth eruption sequences can only be used for sub-adult animals (Klein and Cruz-Aribe 1984; Klein et al. 1983; Niven and Hill 1995; Pike-Tay 1995; Pike-Tay, et al. 2000; Stallibrass 1982) and wear rates vary by diet and individual (Enloe 1997; Enloe and Turner 2006). Age assessment using tooth eruption sequences and wear patterns requires teeth in anatomical relationship. Cementum studies can be performed on single teeth, and can also be used to age adult teeth with greater potential accuracy than tooth wear patterns, although these can also be prone to error (Lubinski and O'Brien 2001; Pike-Tay 1995). Seasonality indications from tooth cementum or the presence of juveniles of particular birth cohorts can assist in interpretation of hunting patterns (intercept or encounter), butchery practices and bone transport selection based on age, sex and inferred fat quality; and, of course, the season of occupation of the site by humans. Estimates of age of death generate mortality profiles: U-shaped, catastrophic, or prime age. U-shaped curves are dominated by juvenile and older individuals, which can indicate normal mortality patterns or hunting by non-human predators as malnutrition, disease and carnivores all impact the most vulnerable members of a population (Stiner 1990, 1991a). The taphonomy of the deposit, presence of distinctive damage and bone survivorship can indicate the operation of attritional mortality or carnivore behavior. Catastrophic mortality curves represent all age cohorts and reflect the actual population. These are the product of sudden mortality events, frequently natural disasters such as flooding, fire or (in extreme cases) volcanic eruption (Lyman 1989). While catastrophic death assemblages are often the result of natural events, archaeological sites that represent the results of game drives may have a similar pattern (Haynes 2004). Prime age mortality curves are dominated by prime age individuals and this mortality pattern is indicative of human hunting behavior as no cursorial carnivores take prime age individuals (Stiner 1990, 1991a, 1994). Formerly expressed in bar graphs by age cohort, mortality patterns now are expressed as proportions of juveniles, prime age and old in 92 triangular graphs, which are more explanatory and easier to compare between assemblages (Steele and Weaver 2002). Mortality profiles and elements present in a faunal assemblage are used to indicate hunting or scavenging, and variables and relationships can be demonstrated graphically and tested statistically. Determination of element value is derived from the MGUI. Assemblages dominated by crania and low utility elements are interpreted as results of scavenging (e.g. Stiner 1994). The presence of high value elements at the site indicates early access to prey via hunting. Examination of such patterns relies on the preservation of identifiable bone, usually the proximal and distal ends of long bones. Some authors, particularly Marean, (Marean and Frey 1997; Marean and Kim 1998; Marean and Spencer 1991) have argued that inclusion of bone shafts through the refitting of diaphysis fragments will affect the element count and interpretation of results. However few archaeologists are in able to refit all long-bone fragments (as advocated by Marean). In older collections like the Cellier material, such material is rare or absent due to selection of potentially identifiable bones. In modern collections such as Level Xc of the Grotte du Renne, the sheer volume and small size of the fragments makes refitting too onerous for the return on investment. While long-bones may be underrepresented in a sample, examination of bone fragments for landmarks will increase the number of identifiable fragments, and no archaeozoologist would neglect examination of bone fragments for data on toothmarks, cutmarks, and other evidence for the taphonomic processes that operated on the assemblage (Stiner 1998, 2002b) While Marean has argued strongly that density mediated attrition and the fragmentation of long bones has resulted in false patterns of observed bone frequencies and that density mediated attrition obscures hominin and human food transportation patterns, Stiner (2002 and elsewhere) has riposted that zooarchaeology is a multi-step process that begins “…with questions about the agencies of bone collection, modification and destruction. Later…analysts may take 93 on questions about human behavior…” (Stiner 2002:979). Stiner further argues, I feel correctly, that bones rarely arrive on an archaeological site as single elements, but are transported or discarded as part of a package of valued or less-valued nutritional and economic components. By combining information on bone density and the larger picture of carcass part transportation, taphonomic processes in terms of bone destruction from processing, carnivore ravaging or diagenetic factors may become apparent. Bone density clearly affects bone survivorship, but density studies do not consider post-depositional factors that operate during diagenesis. All density studies have been conducted on fresh bone. There is no consideration of chemical changes that occur, such as destruction of collagen, leaching of carbonates or phosphates and external factors such as soil pH, crushing, sediment reworking and post recovery destruction through desiccation. On-going actualistic experiments in England (the Overton and Wareham Down project) have shown that soil chemical composition has a significant impact on bone survivorship within a single site and results in alterations in bone chemical structure (Crowther 2002). There is little discussion in the more recent literature regarding bone weathering or freeze-thaw cycles as they impact bone density (c.f. Behrensmeyer 1978; Conard, et al. 2008; Pasda 2005). The majority of zooarchaeological site formation studies focus on how portions of a carcass are or are not incorporated into the archaeological record. Bone density, selection for meat and differential destruction to obtain bone grease and marrow are interrelated as a function of the internal structure of bone. Breakage of bone for marrow will often result in the removal of less dense bone at an articular surface to remove the marrow contents. Bone grease occurs in cancellous material which is found in less dense portions of bone. Bone is therefore subject to destruction both as a result of innate structural patterns and is also more likely to be destroyed by carnivores seeking valued fats or humans breaking and boiling bones to extract grease. Another source of 94 destruction of fatty bones is the use of bones as a fuel source in the Upper Palaeolithic, although there is little discussion of why bone, an incredibly pungent heat source, is burnt in such quantities (Costamagno, et al. 2005; Goldberg, et al. 2010). The use of bone for fuel will therefore distort the faunal assemblages and element representation by removing greasier bones from the archaeological record through burning rather than processing for subsistence. There has been less theoretical consideration of recovery methods and analytical techniques, with the exception of the debate over the inclusion of long-bone fragments (Blumenschine, et al. 1996; Steele and Weaver 2002). Clearly recovery techniques and the area within the site that is excavated are taphonomic processes and will affect sample size and composition. While Lyman and Ames (2007) attempted to quantify the point at which enough of an archaeological site would be excavated to provide a representative faunal sample, this assumes a uniform use of space and uniform discard across the site. While random deposition might be assumed in a palaeontological assemblage, human behavior is patterned and non-random. By selecting areas within a site for excavation, archaeologists restrict the amount of data available for study. Site function, site reoccupation and activities conducted within the site result in non-random patterned archaeofaunas. The issue of dilution of behavior patterns (through the presence of both animal and repeated human occupations) should be considered where sites are attractive to both humans and carnivores (Mondini 2005). Failure to analyze the entire faunal assemblage, or a research focus on particular elements will result in different interpretations of site function and subsistence practices as noted above for East African Plio-Pleistocene sites and discussed by Wismer for Palaeoindian bison (2009). Research continues on the interaction of multiple factors in the creation and survival of faunal assemblages in the archaeological record. A review of the more recent (21st century) literature shows that zooarchaeologists and others continue to build on past 95 knowledge to create an increasingly nuanced picture of taphonomic processes. Debate continues on how to interpret patterns of element survival. Recent research augments our knowledge of animal ethology, and research has examined patterning of tooth mark frequencies. These are strongly correlated with bone length and bone density (Faith, et al. 2007). Variation in toothmark frequencies can be distinguished between species of bone collectors (Kuhn, et al. 2009). Limitations placed on data sets (discussed above) have major implications for tooth-mark frequencies and interpretation, therefore all material within an assemblage should be examined for evidence of carnivore behavior, and possible identification of the agent of damage. Recent experimental studies of caged hyena behavior have also examined evidence patterns of bone deletion and damage related to pack social organization (Faith, et al. 2007). Methodological issues under recent discussion relate to the recording and interpretation of cutmarks through ethnographic study and multivariate analysis to understand the variables important in toothmark and cutmark distribution, cutmark survival and documentation, and the impact of weathering on cutmark survival (Dominguez and Yravedra 2009; Phoca Costamentatou 2005). These taphonomic studies largely focus on the destruction of bone as part of the process of obtaining calories in the form of proteins and lipids. There has been little discussion of the deletion of bone from the faunal assemblage through tool manufacture, or the criteria for selecting bone for tool uses. Archaeologists have described chaînes opératoires for the production of bone tools (Averbouh 2000a and b; Averbouh and Provenzano 1998; Campana 1989; d’Errico, et al. 2003; David 2007; Julien, et al. 2002; Tartar 2009, 2012; Tartar, et al. 2006), and there are replication experiments that describe the manufacture and use of tools (e.g. Griffitts 2006; Le Moine 1991). Bone tool manufacture is another taphonomic factor that deletes elements from the faunal assemblage. Châtelperronian awl manufacturers utilized 96 herbivore and carnivore limb elements, including large, dense epiphyses; while Aurignacian awl manufacture utilized reindeer metapodials at some sites, but long bone fragments at others (d’Errico, et al. 2003; Tartar 2006). Bone tool manufacturing techniques may show distinct differences too, in terms of the methods used to shape tools and the degree of use and use, as indicated Early Upper Palaeolithic awl production (d’Errico, et al. 2003). Deletion of bone elements in cultures where bone tool manufacture is common should reflect choice of supports for tool manufacture. Conclusion In conclusion, the major theoretical issues in taphonomy have focused on the interaction of predator behavior and bone structure in explaining and interpreting the observed zooarchaeological record. There has been far less focus on the deletion of elements from the record as a result of tool manufacture, post depositional diagenetic processes, or collector bias, which can impact the faunal assemblage. The latter factor becomes extremely important when examining older, curated collections such as Abri Cellier. Few items have been deleted from the Level Xc assemblage. Some elements were destroyed during early radiocarbon dating assays, but in general the assemblage remains largely intact. The faunal material from Abri Cellier and Level Xc of the Grotte du Renne have both undergone a variety of taphonomic processes: transportation of fauna to the site, processing, and post depositional diagenetic alterations. In addition the Abri Cellier assemblage has been affected by selective recovery of elements and transportation to Beloit College. Further attritional processes were sale to other museums and drawer damage that occurred in the museum collection (White and Knecht 1992; Woods 2011). The absence of small bone fragments clearly reflects excavation practices and collection decisions of the Beloit excavators, who focused on identifieable elements and bone tools (Tolmie 2009). 97 The taphonomy of each assemblage will be discussed in the respective chapters that describe the results of the analysis. Both assemblages will be assessed for transportation decisions in relation to appropriate utility indices and fat content of bones. Attritional factors including carnivore damage, natural diagenetic processes and destruction of elements via bone processing will also be examined for each assemblage. The next four chapters will examine the history of excavation at the two sites, and the faunal analysis. 98 CHAPTER 7: THE GROTTE DU RENNE, ARCY-SUR-CURE: PREVIOUS RESEARCH AND CURRENT CONTROVERSY Introduction The Grotte du Renne (Yonne, Burgundy) is one of series of solution caves formed in a small limestone massif of Jurassic Era Rauracian corals known as the Massif d’Arcy, located between the granite of the Morvan and the south-east edge of the Paris Basin, approximately 20km from the edge of the Morvan plateau and 200 km southwest of Paris (David, et al. 2001, Girard 1980, Leroi Gourhan & Leroi Gourhan 1964). These caves are located on the last meander of the River Cure, a location that would form a natural way point or landmark for Neanderthals and modern humans (Figures7.1 and 7.2). The site is located in a limestone cliff, facing south towards the river, with a gently rolling upland plateau to the north. Previous Research Evidence for human occupation has been found in the Grand Grotte, the Grottes de Lagopède, Cheval, Hyène, Trilobite, Ours, Renne, Bison, Loup, Lion and the Grotte des Fées. Of these, only the Grand Grotte and the Grotte des Fées were known prior to the nineteenth century. (In a rather charming tradition began by Abbé Parat in the nineteenth century, new caves are named for the first fossil recovered). The caves of Arcy-sur-Cure have been the subject of archaeological investigations since the nineteenth century. Stone tools and animal bones were first discovered in the Grotte des Fées in 1853 and excavations in 1858 identified a level of cave bear occupation, followed by a level of stone tools and animal bones and capped by a level containing pottery. A human mandible was recovered, supposedly from the lower level, and is thought to be that of a Neanderthal. 99 Figure 7.1: Map showing the location of the Grotte du Renne. Source: David, et al 2009. 100 The Grotte du Trilobite was discovered in 1886 and the first scientific excavations were undertaken by the Abbe Parat in the 1890s. In 1894 he recovered an unusual lithic industry from the Grotte des Ours which contained tools similar to that of the Mousterian, mixed with more evolved items and he described “que cet outillage est l’oeuvre «d’un people primitive inaugarant une nouvelle voie dans l’industrie»” [that these tools are the work of “a primitive people initiating a new path for the industry”] possibly as the result of contact between Mousterian and Upper Palaeolithic groups (Baffier & Girard 1998: 14). This material was later identified by Abbé Breuil as Châtelperronian. Parat ceased excavations at Arcy in 1905. Figure 7.2. Sketch showing the prehistoric caves at Arcy-sur-Cure and previous excavations. 101 No further scientific work was undertaken until the discovery of engravings in the Grotte du Cheval in 1946 by local speleologists attracted the attention of André LeroiGourhan. He and his students excavated the grottes de l’Hyène and Loup (already partially explored by Parat) and the Grottes du Bison, Renne and the Lagopède rock shelter. Excavations ended in 1964, when Leroi-Gourhan began excavations at the openair Magdalenian site of Pincevent. Archaeological research resumed following the discovery of cave paintings in the Grande Grotte in 1990. Excavations within the cave were undertaken by a team led by Dominique Baffier and Michel Girard and research on the parietal art continues to the present. Excavations at the Grotte du Bison resumed in 1995 under the direction of Francine David. This research program also remains active. The Grotte du Bison is linked to the Grotte du Renne, and one research goal is to better understand the relationship of the occupation sequences of the two neighboring caves. Description of the site The Grotte du Renne was discovered by Pierre Poulain in 1949. The site contains the collapsed cave (the porche) which has been excavated, and the Gallerie Shoepflin, a long narrow gallery which extends north of the cave are, which also contains archaeological material (Schmider 2002). The collapsed cave area measures 6.5 meters by 9.0 meters and is aligned north-south. Two episodes of roof-fall are present: Level III which seals the archaeological deposits, and Level XIII, above the first Mousterian occupation (Girard 1980). Excavations were undertaken at the Grotte du Renne from 1949 to 1963 by a team lead by Andre Leroi-Gourhan. Further excavations were undertaken in 1995 when the witness column was excavated to verify the sedimentary descriptions (David et al 2001). The excavations procedure used at the Grotte du Renne is known in French as décapage – the removal of sediments over a large area to reveal a living floor or 102 occupation area. This type of excavation was pioneered by Leroi-Gourhan, whose prime research interest was in the social organization of Palaeolithic societies, rather than the type fossil chronologies created primarily through cave excavations. As he himself noted, when discussing Neanderthal research, it was to be regretted that the strong focus on chronology had resulted in prehistorians ignoring or losing the opportunity to “relever les innombrables détails qui auraient permis d’enricher notre connaissance sur les activités intellectuels et sociales des homes de cette époque” (Leroi-Gourhan 1964: 142) [recover innumerable details which would allow us to enrich our knowledge about the intellectual and social behavior of humans of this period]. As Movius (1969:122) noted, this style of excavation allowed archaeologists to go beyond “preoccupations that are limited to chronology and …to attempt an attack on the problem froma truly ethnological or cultural angle.” Figure 7.3: Sketch of Level Xc of the Grotte du Renne, showing the location of the hut area, ash areas and the talus or porche. 103 Excavations of the Châtelperronian levels revealed an area at the rear of the cave which contained postholes and mammoth tusks in association with concentrations of ash or hearths, which were interpreted as huts or shelters. By piece-plotting artifacts and exposing whole surfaces, the goal was to attempt to understand the organization of space and infer social behavior. In addition, the majority of excavated sediment was water screened and all material was curated, including small, unidentifiable bone fragments. All archaeological material was curated, including bone fragments, another practice unusual for that time period. Leroi-Gourhan’s excavations revealed 15 different levels (French: couches) which were defined by differences in sediments. Eleven of these contained archaeological material. The excavations did not reach bedrock, but halted at the river alluviums, which were archaeologically sterile and underlay the cave sediments (Girard 1980, Movius 1969). These alluvial sediments may date to the Riss-Wurm interglacial (Leroi-Gourhan & Leroi-Gourhan 1964). The lowest level (XV) was archaeologically sterile and contained alluvial sands. Levels XIV through XI contained different Mousterian facies (Typical, Transitional and Denticulate). Sedimentology shows a discontinuity between levels XI (the latest Mousterian) and X, the first Châtelperronian occupation. Levels X, IX and VIII contained Châtelperronian materials, overlain by Aurignacian (VII), Gravettian (VI, V) and Solutrean (IV) occupations. The upper three levels were archaeologically sterile (David, et al. 2001:207). The Châtelperronian occupations occur within accumulations of limestone plaquettes spalled from the roof and walls of the cave. In the 1998 study described in David et al. (2001) only the lowest portion of level Xc and all of level X were present in the witness column. Level X is divided into three parts – Xa, Xb and Xc. Layer Xa is 22 cm thick, and subdivided into two sub-levels separated by 5-7 cm of relatively sterile sediment. At the base of level Xa, large slabs and blocks separate it from level Xb. As 104 with layer Xa, level Xb contains two occupation layers, separated by a thin sterile layer. Layer Xb is 28cm thick. This is the thickest and richest Châtelperronian level in the site. In contrast, level Xc is thin (5cm thick) in a blackened sandy-clay matrix with relatively few plaquettes, unlike the upper sections of level X, where plaquettes are common. Châtelperronian material, particularly burnt flint, is common. This level overlies the uppermost Mousterian deposits, which occur in a bed of yellow clayey sands. The sediments of level Xc include alluvial clays. The source is probably from seasonal flooding of the Cure and the fine sediments indicate a low energy flood environment. David et al. argue that level Xc is contemporary with the later part of the Cottes interstadial, when the climate was returning to more glacial conditions. The Châtelperronian occupation at Arcy-sur-Cure therefore begins during a milder phase of the early Last Glacial, and continues through increasingly cold conditions and varying levels of humidity (David, et al. 2001; Farizy 1990b). Environmental context Environmental reconstruction based on pollen data (Leroi-Gourhan and LeroiGourhan 1964) indicates that the Châtelperronian occupation at the Grotte du Renne first occurred as the climate was cooling rapidly at the end of an interstadial (Leroi-Gourhan and Leroi-Gourhann1964:3). The Mousterian occupations at Arcy-sur-Cure occurred during climatic oscillations associated with the onset of the Würm glaciation. At Arcy, the warmest of these occupations, described as “legerement temperé” Leroi-Gourhan and Leroi-Gourhan 1964:3) coincided with the last Mousterian occupation, Level XI, of the Grotte du Renne. This was followed by rapid cooling of the climate and the mean annual temperatures had diminished considerably by the time of the first Châtelperronian occupation in level Xc, and continued to decline (Leroi-Gourhan and Leroi-Gourhan 1964:11). Thermophilous pollens decline rapidly during the early part of the Level X sequence with steppe vegetation on the plateau with pockets of pine and juniper, and 105 alder and hazelnut along the river valley. The climate became increasingly cold, with the coldest period corresponding to Level VIII, the last Châtelperronian occupation – where pine, alder and birch pollen are present. Artemesia pollen disappear and ericeaes are also absent. Sedimentological studies also characterize the climate during Xc occupation as cold and dry (Girard, et al. 1990). Again, level VIII represents the coldest period (Movius 1969). The Châtelperronian occupation ceases prior to the Arcy interstadial, a period of rapid rewarming associated with the Aurignacian level in Level VII. Thermophile tree pollen appears in this level. This is again followed by rapid cooling and extreme cold in the Late Glacial Maximum, capped by collapse of the remaining cave roof. The Châtelperronian occupation at the Grotte du Renne coincided with a cooling, dry environment. Pollen and sedimentological studies indicate that the climate cooled rapidly during the first Châtelperronian occupation in Level X, and continued a cooling trend, which peaked in Level VIII, followed by the rapid warming associated with the Arcy interstadial. Open habitats predominated, with trees present in sheltered areas and probably along the river floodplain. Preliminary analysis of the faunal remains from layer Xc confirm a cooler, open environment. While microfauna are rare, large fauna are dominated by reindeer, with smaller numbers of horse, elk and bovids (David and Poulain 1990). Humans occupying the site would need to adapt to the increasingly cold climate, and it appears that the Neanderthals at the Grotte du Renne had expanded their niche through the use of shelter and probably clothing. The Châtelperronian of the Grotte du Renne probably post-dates the Châtelperronian at St. Césaire (David, et al. 2001). Pollen data from the latter site are not easy to interpret as three separate samples from within the cave produced a highly variable amount of arboreal pollen. The pollen analysis shows the final Mousterian phase at the site to coincide with a warmer environment, with broadleaf tree pollen present. The overlying Châtelperronian samples varied in the proportions of arboreal pollen, but were dominated by pine and other cold adapted species. Mesophilous pollen returned in the 106 Proto-Aurignacian (Leroyer and Leroi-Gourhan 1993). Again, the climate associated with the Châtelperronian indicates a cooling to cold environment, followed by a warming trend that post-dates the Châtelperronian occupation. The Grand Pile pollen core shows a rapid cooling after the Hengelo oscillation around 40,000±800 rcbp. A minor warmer oscillation occurs around 34,100±290rcbp, followed by the coldest phase for the stadial followed by rapid warming around 30,820±210 rcbp associated with the Denekamp interstadial (Mangerud 1991).Within this 10,000 year time period, there are a number of second and third order (Dansgaard/Oescher) climate phases. The Châtelperronian occupation at the Grotte du Renne appears to coincide with a sudden cold phase, a minor warming trend, followed by cooling and then rapid warming (van Andel 2003:33). Cultural material from the Grotte du Renne The Châtelperronian occupations cover an area approximately 80m2 in a situation that is not quite a rock shelter and not quite in the open air, but sheltered under the cliff face (Farizy 1990). The lithic industry is rich: over 35,000 items, including 5000 tools. This is far richer than the preceding Mousterian assemblage, and the proportion of raw material is also different. Chert dominated the Mousterian assemblages (80%) but the situation is reversed in the Châtelperronian with flint dominant (70%). This coincides with changes in the chaîne opératoire of tool production for flint tools (Bodu 1990; Gouedo 1990), whilst older technological practices continue for the manufacture of items in chert. This dichotomy led to Leroi-Gourhan describing the Châtelperronian at Arcy as “Mousterian déguisé en Paléolithique supérieur” (Farizy 1990a:286). [the Mousterian disguised as the Upper Palaeolithic]. As noted above, there is a hiatus in occupation at the Grotte du Renne between the Mousterian and Châtelperronian. As well as changes in the patterns of lithic production, there is a change in spatial organization. There is little evidence of management of space 107 in the Mousterian levels at the various caves and rockshelters at Arcy. Zones of activity can be identified (c.f. Enloe & Lanoë 2012) but there is little evidence of any maintenance of space or removal or refuse. In contrast, the Châtelperronian of the Grotte du Renne has evidence of deliberate construction of windbreaks or cabins that contained hearths (Farizy 1990a; Movuis 1969). Some hearths show evidence of cleaning and reuse. This management of the occupation area has resulted in some mixing of earlier material through the digging of postholes and the leveling of sediments. The most significant change from the Mousterian is the appearance of a variety of bone tools, and also the appearance of personal ornaments. At least 120 bone or ivory tools have been recovered from levels X and XI (Baffier and Julien 1996). My recent faunal analysis has increased this number as additional tool fragments were identified within the faunal collection. At least 44 bone tools are now known from Level Xc. These include awls and thinner bone items referred to as “pins” by the analysts (d”Errico, et al. 2003). Ornaments from Level X comprise 16 incised or perforated mammal teeth. Three pendants were recovered from Level IX and six from Level VIII. Ivory rings were recovered from Beds X (n=2) and VIII (n=1) (Baffier and Julien 1990). While differences occur in the lithic and organic tool industries between the Châtelperronian and the Mousterian, the faunal remains are similar to those of the Mousterian, with a focus on the hunting of large herd animals. The major difference is the absence of a significant amount of occupation by carnivores in the Châtelperronian levels. Both hyena and cave bear are well documented in the Mousterian levels of the Grotte du Renne and the Grotte du Bison. There is a major change in the use of the site by carnivores in the Châtelperronian. Carnivore remains are few, and the amount of gnawed or digested bone is negligible. While cave bear occasionally hibernated at the site, there is little evidence for use of the shelter as a den by cave hyenas. This pattern is replicated in the Grotte du Bison, which shared an entrance with the Grotte du Renne at this time period (Enloe in press; Enloe and Lanoë 2012). Why this occurs is in unknown. 108 One explanation could reflect changes in season of use of the site by Neanderthals, which excluded the hyenas. Another possibility is that major shifts in the local hyena population occurred, reducing the density of occupation of the area by the scavengers. A controversial site for a controversial period The debate surrounding the interpretation of the spatial patterning and artefacts recovered from the Grotte du Renne reflect the debate over the cognitive capacities of Neanderthals and the underlying cause of their disappearance. The Grotte du Renne has produced the richest assemblage of worked bone and bone and ivory ornaments known in the Châtelperronian (Pelegrin & Soressi 2007). Palaeoanthropologists are divided as to the reason for this rich assemblage. While one school of thought argues that the tools and ornaments are the produced by Neanderthals, others argue that the osseous industry is the product of post-depositional taphonomic factors, mainly mixing. A third line of argument has argued that the “modern” material found in the Châtelperronian levels reflects contact and exchange between the two hominin species. We will examine the arguments for postdepositional mixing, exchange and in-situ production or use. Arguments for the mixing of sediments and post-depositional disturbance are were first made in White’s 2001 paper which described the osseous industry (White 2001). Another argument for disturbance is based on or methods of production (BarYosef 2006; White 2001). Radiocarbon dating has also been used to support this argument, particularly in a recent study using ultra-filtration to obtain new dates, which shows considerable variation in the ages of the tools examined, especially level Xc (Higham, et al. 2010). Although the principal author argued that the samples had been adequately treated to remove all possible contaminants, a more recent series of dates (Hublin et al. 2012) produced more coherent dates with smaller standard deviations, suggesting that the problems with the Oxford dates may be related to inadequate preparation or insufficient cleaning. Ultrafiltration is a new technique and it is interesting 109 that all new dates derived from this method are found to be earlier that dates established by AMS. The dates for the Grotte du Renne material also have substantial standard deviations when compared to the AMS dates. Statistically these dates may not be particularly early, given the wide 2-sigma variation. Until another laboratory replicates the ultrafiltration process and the dates, these early dates should be regarded with some skepticism (Pettitt and Pike 2001). AMS dates for Level VIII are in agreement with dates from contemporary levels in the Grotte du Bison. However both Level IX and Level X radiocarbon dates are very discordant (David, et al. 2001) and do not align well with sedimentological and environmental data (Table 7.1). The Oxford dating program found a large number of outliers, and the principal authors argued that this indicated a significant amount of reworking of the sediments in Level X (Higham, et al. 2011). This follows the arguments of White and others that the sediments containing the Châtelperronian are disturbed. Some of the issues with dating may reflect the choice of materials to be dated. In recent studies the focus has been on the worked bone items. The most recent studies found that unworked bone or bones with cutmarks were more consistent in their dates that the worked material (Hublin, et al. 2012). The latter produced skewed dates, while faunal remains with butchery marks or no cutmarks produced coherent dates. Reasons for this are unclear, but might relate to greater exposure and weathering resulting in loss of collagen or contamination for the bone tools. The Châtelperronian occupation of the Grotte du Renne, in broad terms, appears to begin between 39,000 and 40,000 BP and lasts until approximately 35-36,000BP, based on the most recent dates (Hublin, et al. 2012). Level d'Errico et al 1998 Higham et al 2011 Hublin et al 2012 V Ly-2161 20,150 ±500 OxA-21567* 23,070 ±210 OxA-21568* 23,180 ±210 OxA-X-2279-12 34,850 ±600 VI VII GrN-1717 30,800 ±250 OxA-21682 35,000 ±650 EVA-79 29,930 ±208 Ly-2162 31,800 ±1240 OxA-21569* 36,500 ±1300 EVA-81 33,850 ±311 OxA-21570* 34,600 ±800 EVA-92 31,610 ±131 OxA-21571* 34,050 ±750 EVA-93 33,010 ±182 OxA-21572* 34,600 ±750 EVA-95 34,810 ±210 OxA-X-2279-14 35,450 ±7503 EVA-52 35,980 ±432 VIII Ly-2163 33,000 ±1400 110 Table 7.1: Radiocarbon dates for the Grotte du Renne, for levels dating from the Mousterian (XII) through the Gravettian (V) periods. Level d'Errico et al 1998 Higham et al 2011 Hublin et al 2012 GrN-1742 33,860 ±250 OxA-21683 40,000 ±1200 EVA-53 36,230 ±435 GrN-1736 33,500 ±400 OxA-21573* 36,800 ±1000 EVA-54 35,380 ±390 EVA-55 36,630 ±452 EVA-56 37,710 ±533 IX OxA-21574* 38,800 ±1300 EVA-44 39,280 ±351 IXa OxA-21575* 32,100 ±550 EVA-46 39,930 ±361 IXa EVA-47 39,750 ±360 IXa EVA-33 40,970 ±424 IXb EVA-34 40,520 ±389 IXb EVA-35 39,240 ±341 IXb EVA-36 39,450 ±340 IXb EVA-37 37,740 ±307 IXb Table 7.1 continued. 111 Level d'Errico et al 1998 Higham et al 2011 Hublin et al 2012 X GrN-4251 25,500 ±380 OxA-21565 37,900 ±900 Xa EVA-38 36,540 ±248 Xa GrN4216 24,500 ±4216 OxA-21557 38,100 ±1300 Xa EVA-40 37,510 ±275 Xa OxA-21576* 40,800 ±1700 Xa EVA-41 38,730 ±333 Xa OxA-X-2222-21* 23,120 ±190 Xa EVA-42 38,070 ±311 Xa OxA-21577 34,650 ±800 Xa EVA-43 39,020 ±352 Xa OxA-X-2226-7 38,500 ±1300 Xb EVA-23 36,840 ±335 Xb1 OxA-21590 21,150 ±160 Xb1 EVA-24 38,400 ±317 Xb1 OxA-21591* 34,750 ±750 Xb1c EVA-25 36,210 ±250 Xb1 OxA-21592* 36,200 ±1100 Xb2 EVA-26 39,390 ±334 Xb1 OxA-21593 35,300 ±900 Xb2 EVA-27 40,230 ±395 Xb1 OxA-X-2226-12 41,500 ±1900 Xb2 EVA-28 40,930 ±393 Xb1 OxA-X-2226-13* 39,000 ±1400 Xc EVA-29 35,500 ±216 Xb2 Table 7.1: continued. 112 Level d'Errico et al 1998 Higham et al 2011 Hublin et al 2012 OxA-X-2279-18 40,600 ±1300 Xb/c EVA-30 37,980 ±284 Xb2 OxA-X-2279-44* 48,700 ±3600 Xb EVA-31 39,290 ±334 Xb2 OxA-X-2279-45* 40,900 ±1300 Xb EVA-32 36,820 ±257 Xb2 OxA-X-2279-46* 38,700 ±1000 Xb/c EVA-48 39,070 ±332 Xb2 EVA-49 40,830 ±778 Xb2 EVA-51 39,960 ±702 Xb2 EVA-77 42,120 ±805 EVA-83 41,980 ±821 EVA-84 43,270 ±929 EVA-85 40,900 ±719 XI Ly-2165 37,500 ±1600 XII 37,000 ±1000 OxA-21595* 38,200 ±1200 113 Table 7.1: concluded. OxA-21594* 114 The site is also one of the most northerly Châtelperronian sites. Given a cooling climate and a northerly location, there may have been more impetus to develop and use osseous tools to produce shelter and effective clothing, than at sites further south where less efficient shelters were adequate for protection. Further, Neanderthals in this region had time to fully develop a more formal bone tool technology. The issue of disturbance The presence or absence of disturbed soil or post depositional reworking is a major issue in the study of the Châtelperronian in Western Europe and is an argument used particularly by researchers from Bordeaux to negate any evidence for interstratification or for the presence or “modern” tools in Châtelperronian levels (Bordes 2003; Zilhao, et al. 2008; Zilhão, et al. 2006). In other words, it is argued that the Châtelperronian may be a construct of post-depositional processes. This argument cannot be validated at the Grotte du Renne. The richest Châtelperronian level is Level X, which produced 38 awls. Only 9 awls are known from the Aurignacian level (Level VII) and the upper Châtelperronian levels (IX and VIII) produced 5 awls apiece (d’Errico et al 2001). If Aurignacian material were moving down through the sediments it would be more likely to occur in Level VIII and distributed in manner that reflected the main activity area of the Aurignacian occupation. This is not the case (d’Errico, et al. 2001:254; Caron, et al. 2011). Further, the sediments containing Level VIII are 80 centimeters thick, with yellow clay and eboulis (Movius 1969; F. David pers. comm.) in contrast to the purplish Aurignacian levels. Large scale reworking of sediments would likely be reflected in the sediment structure. A recent statistical study assessing the evidence for mixing of the deposits at the Grotte du Renne, with particular reference to the recent dating program by Higham found no evidence for large scale or small scale displacement of artifacts (Caron, et al. 2011). Examination of the lithics found that 100% of Levallois flakes were recovered 115 from Mousterian Level XI; 99% of Châtelperronian points and scrapers were from Levels X, IX and VIII; and 100% of Dufour bladelets and blanks were from Level VII. A very small amount of mixing did occur, (for example, a Châtelperronian point in Level VII, four probably Aurignacian ivory fragments in Level VIII) but this was not statistically significant and might be the result of excavator error, or post excavation mixing. How to explain the radiocarbon dating anomaly? Caron et al. (2011) argue that the most parsimonious explanation for the outlier dates is poor collagen preservation or the retention of very small amounts of contaminants. They further note that while material treated with consolidants was rejected for radiocarbon dating at other sites in the project, 84% of the material used in the Grotte du Renne study had been treated in some form (Carron et al 2011: 5). Clearly the absolute dates for the Châtelperronian at the Grotte du Renne do show some inconsistencies, but even in Higham’s data, which are argued to show disturbance, two-thirds of the dates were consistent and in stratigraphic sequence. The large standard deviations for a number of the outliers also overlap with the “good” dates. Given that a radiocarbon date is a statistical probability as much as a measure of the amount of datable organic material, I would argue that the anomalous radiocarbon dates cannot be used to prove the existence of significant post-depositional modifications to the sediments. Some authors have taken a technological approach to argue for mixing of the sediments. White (2001) argues that the presence of teeth pierced by drilling (a technique common in the Aurignacian) alongside pendants prepared for suspension by graving around the root of a tooth (rainage) is the result of taphonomic processes that have resulted in the migration of Aurignacian material downwards into the Châtelperronian levels. This despite the fact the he himself notes that there is only one tooth “…pierced in an asymmetrical and idiosyncratic fashion” plus “a few” ornaments (White 2001:44). This is in contrast to the relatively rich Châtelperronian assemblage. One could reverse 116 White’s argument and argue that the richer assemblage provided the material that had been redeposited in the Aurignacian levels. A similar argument is used to explain the presence of Neanderthal teeth in association with the Châtelperronian. Bar Yosef (2006) is one of a number of authors who have suggested that this material was introduced into Level X through the digging of postholes associated with the cabins of wind breaks, which introduced Mousterian material into the level. All the debates on the validity of the Châtelperronian at Arcy-sur-Cure focus on the Grotte du Renne, largely because of the large amount of worked bone in the lowest Châtelperronian levels. It should be noted that the Châtelperronian also occurs at the Grotte des Ours (identified by Parat) and that post-depositional disturbance cannot be argued for Quinçay, where the unpublished Châtelperronian osseous assemblage is sealed by a layer of large limestone slabs and where the Aurignacian does not occur. Direct evidence for Neanderthal occupation of Level Xc Human fossils from the Châtelperronian levels of the Grotte du Renne consist of 29 teeth and a juvenile temporal bone (Bailey and Hublin 2008; Hublin, et al. 1996). The temporal bone is from Level Xb, as are the majority of the teeth. All traits in the adult teeth (n=15) fall within the Neanderthal pattern. No teeth (adult or juvenile) fall within the modern human cluster (Bailey and Hublin 2008). The authors further argue that the distribution of teeth of two individual (an infant and a juvenile) show good integrity and little post-depositional disturbance. The Châtelperronian is therefore associated with Neanderthal fossils at Arcy-sur-Cure. Why would palaeoanthropologists persist in arguing that the osseous tools at the Grotte du Renne should be regarded as anomalous? The site was occupied during a period of climatic deterioration when clothing and shelter would be increasingly important. The construction of huts or windbreaks and the investment of time in 117 preparing and maintaining the site suggest a longer term occupation or maintenance and reuse of the site. This would result in a larger and more representative sample of artifacts, manuports and evidence for personal decoration. I would also argue that palaeanthropologists have fallen into the trap of assuming that one site is representative of an entire cultural entity, without considering how local conditions or variations in behaviors by local groups could result in differences in material culture. The Grotte du Renne provides a large dataset that enables faunal analysis, not simply of a sample of the fauna, but of the entire faunal remains present at the site. The site was excavated using techniques that were innovative for the period and designed to answer questions about hominin social and economic behaviors. While there are some discrepancies in the dataset, the product of over 50 years of post-excavation storage, the persistence of the notion of large scale disturbance cannot be proven or even adequately demonstrated. The original argument for disturbance (White) is based on an interpretation of site notes by a researcher who was not part of the excavation team, not refitting studies, or other actualistic analyses, which would have more validity. Lithic refits are rare between the Châtelperronian levels (Hublin, et al. 2012), and largely occur in the talus, where the separate sediments peter out and slope towards the Cure. The collection is ideal for examination of the interaction of prey choice and bone tool supports, given the recovery methods used and the curation of all material. A full zooarchaeologocial analysis of the material will enable a better understanding of how the assemblage was accumulated (whether humans or carnivores were responsible); if humans are the main agent of accumulation, it will be possible to examine prey selection patterns, carcass transportation choices, and the material available for use as raw material for bone tools. This will be the focus of the next chapter. 118 CHAPTER 8: FAUNAL ANALYSIS OF LEVEL XC OF THE GROTTE DU RENNE, ARCY-SUR-CURE Introduction A wide variety of species occur in the faunal assemblage of Level Xc. In the first part of this chapter the faunal assemblage will be described and quantified. The processes of bone accumulation and bone attrition will then be examined to determine if the assemblage is the product of natural taphonomic processes (n-transforms), or cultural taphonomic processes (c-transforms). The presence of carnivores in the assemblage, particularly known bone accumulators such as hyenas, makes it necessary to how much of the assemblage is the product of carnivore denning behavior. No meaningful statements about hominin subsistence can be made without careful taphonomic analysis of the assemblage to determine the role of hominin and carnivores in creating the assemblage. It will also be necessary to consider the role of both groups as agents of destruction within the assemblage. Taphonomic analysis indicates that the prime agents of bone accumulation and destruction were Neanderthals rather than carnivores. The final section of the chapter will examine prey selection choices and transportation practices and spatial patterning of faunal remains within Level Xc to better understand the behavior of Neanderthals at the site. Taxa present in Level Xc The following animals were present in the fauna: reindeer (Rangifer tarandus), horse (Equus caballus), red deer (Cervus elaphus), bison or auroch (Bovidae), mammoth (Mammuthus primigenius), hare (Lepus), cave hyena (Crocuta spelaeus), cave bear (Ursus spelaeus), a large cat (Felidae) and wolf (Lupus sp.). Bird, microfauna and fish bones present within the fauna were not part of this study, but had been sent to other analysts for examination prior to the beginning of this thesis research. A single possible bird long bone fragment was identified in the faunal assemblage under study. This bone 119 was passed to the researcher responsible for the avifauna and will not form part of this analysis. The range of animals present at an archaeological site is a reflection of the subsistence strategies practiced by a particular population. Changes in the numbers of species present are generally interpreted as responses to changes in the natural or cultural environment. One must ask if using calculations based on the Number of Identified Specimens (NISP) is appropriate. As Schmitt and Lupo note “…prehistoric hunters were probably occupied with animal size…” (1995: 497) and not the number of bones in a carcass. Taxonomic diversity uses relative abundance calculations to compare diets between sites, between cultures or examine change over time and/or space. The most common measures of abundance are NISP and the Minimum Number of Individuals (MNI). Both have problems when used to address taxonomic diversity within an assemblage. Klein and Cruz-Uribe (1984) note that NISP will over-represent animals that are brought into the site as entire packages, and is also very sensitive to the amount of bone fragmentation present, or the presence of loose teeth. They argue that NISP is not suitable for abundance calculations (1984:25), yet it is used because MNI calculations give values that are too low to be of use in statistical calculations. Another reason NISP is used is convenience: it usually published and therefore available for use in comparisons. Other data, apart from MNI, may not be provided and it is tedious and time consuming to return to the original data and make the necessary calculations. In terms of MNI, Klein and Cruz-Uribe note that different analysts use different methods of calculation, and that a site MNI will be affected by the units of analysis (1984:28), a point enlarged by Grayson (1984:29 et seq.). The main problem with both MNI and NISP is that while they measure abundance of bones, they both ignore specific skeletal parts and therefore ignore contrasts in patterns between sites or samples (Klein & Cruz-Uribe 1984:30) Measures based on NMI, meat-weight or the general utility index 120 (GUI), would also be problematic, but would relate more closely to decisions made in the past by people taking and processing animals as packages of meat and fur, not pieces of bone. Zooarchaeologists use taxonomic frequency to recreate both the local environment and to examine the diet of site occupants. As part of diet reconstruction and site interpretation, zooarchaeologists use skeletal part frequencies to interpret meat procurement strategies. The results of such analyses can then serve to interpret broader issues of subsistence and social organization. Measures of dietary utility (meat, marrow and grease, usually incorporated into a general utility index or MGUI) are calculated for bones or bone parts, the Minimum Animal Units (MAU) or Minimum Number of Elements (MNE). The resulting plots of parts present against utility are used to infer transport decisions and diet. However, the relationship between element frequency, transportation decisions and diet is not simple, and is mediated by many factors before and after deposition of the bones at a site. Binford (1978) examined meat, marrow and grease utility in sheep and caribou to create indices by element for all three resources. Different body parts were taken at different season, based on the physical condition of the animals, and butchery patterns related to species of animal. Ethnoarchaeology found that there was “no single episode in which selection of anatomical parts was unambiguously conducted with respect to considerations of meat yield only” (Binford 1978:23). Marrow and bone grease were also extremely important, and the value of marrow and grease and processing time for both were factors in bone transportation. Transport decisions are complex and embedded in what is regarded as culturally appropriate (1978:40). In fact, transportation decisions appear to be the result of a complex of economic, social and situational contexts, evidenced by the informants’ responses to Binford when he asked about their “favorite part” of an animal - a concept that had no meaning to the Numamiut. 121 Ultimately Binford made two generalizations: when game is scarce there will be a maximization of food regardless, but when game is abundant labor considerations will converge with utility indices (1978:44). While these studies generated a testable model, they did not consider taphonomic issues related to bone survivorship after discard Lyman (1985) argued that taphonomy mediates information regarding subsistence choices left in the archaeological record. Lyman also addressed potential problems in the choice of element used to create the assemblage indices - NISP, MNI MAU and MNE. (1985:223) and argues for use of “butchery units”. However, butchery units are culturally determined and vary by season and carcass. Lyman’s calculations found the marrow indices were strongly correlated with density (not surprising as it occurs in long bones, metapodials and phalanges), while grease was probably correlated inversely (occurring in cancellous tissues, particularly the axial skeleton), and meat had a weak inverse correlation with density. Clearly both butchery, processing for marrow and grease and post-depositional factors will impact a faunal assemblage. Lyman raised awareness of these issues, but did not suggest any methods for assessing how much of an assemblage represents of human transportation versus post-depositional destruction. While Lyman examined the impact of taphonomy on a faunal assemblage, Speth and Spielmann (1983) examined the importance of nutrition as a taphonomic factor. They critiqued the focus on protein in faunal analysis, noting that fat, particularly carbohydrates and essential fatty acids provided by marrow, are extremely important to prevent starvation in times when most meat is lean - high in protein but low in fat. This has important implications for bone transport and frequencies in terms of seasonality, particularly in high latitudes where prey becomes lean in winter and early spring. They examine strategies used by foragers to avoid protein starvation including use of other fat-rich species when the main prey animals are in poor condition. The broadening of species diversity at some sites may reflect seasonality - if fat rich animals such as beaver, waterfowl, bear and fish are present it suggests maximization of 122 procurement of essential fats. Skeletal part frequency should therefore be considered in relation to species diversity at a site. By focusing only on larger animals, or the most common taxa, zooarchaeologists may obscure some dietary choices that represent responses to seasonality or nutritional needs. NISP and MNI The most common taxa by NISP were herbivores (n=1178), with only 211 items identified as carnivores (Tables 8.1, 8.2 and Figures 8.1 and 8.2). Genera MNI NISP %MNI %NISP Bovidae 2 22 5.26 1.58 Red deer 2 8 5.26 0.58 Hare 2 6 5.26 0.43 Horse 4 206 10.53 14.83 Mammoth 1 61 2.63 4.39 Reindeer 14 875 36.84 62.99 Bear 7 134 18.42 9.65 Felidae 1 1 2.63 0.07 Hyena 3 64 7.89 4.61 Wolf 2 12 5.26 0.86 Total 38 1389 100 100 Table 8.1: Minimum Number of Individuals and Number of Identified Specimens identified to genus and/or species 123 Taxon Total Count Side Element/Landmark Bird 1 1 U Long bone Bison 1 1 L Humerus cf Bison sp Bovid 1 1 L Femur/Nutrative foramen cf Bos sp Bear 7 2 L Tarsal/naviculocuboid 5 R Deciduous canine Felid 1 1 U Third phalange Hare 2 2 L Maxillae/Tooth row Horse 4 3 L 3rd mandibular molar 1 L 3rd mandibular molar, Comment unerupted Hyena 3 2 R Deciduous canine 1 R Radius/Styloid process fully fused Mammoth 1 1 u Third phalange Red deer 2 1 L Radius unfused Mandibular and maxillary Worn teeth 1 teeth Reindeer Wolf 14 2 12 L Tibia/Nutrative foramen 2 R Tibia/Medial malleolus unfused 1 U Metapodial unfused 1 U Second phalange/fused Table 8.2: Summary showing calculations of Minimum Number of Individuals for taxa in Level Xc. . . 124 Figure 8.1: NISP by count and percentage for Level Xc. Figure 8.2: MNI by count and percentage of total for Level Xc. 125 When minimum numbers of individuals were calculated, herbivores remained dominant, (n=25), with fewer carnivores (n=13). The most common herbivore is reindeer (NISP=875, MNI=14) and the most common carnivore is cave bear (NISP=134, MNI=7). Bone preservation of the majority of identifiable bone was good to excellent, with little weathering or leaching. In contrast to the unidentified bone, discussed below, the majority of the identified bone was broken during carcass processing. A variety of bone preservation techniques were utilized over the course of excavation. Conservation materials included paraffin wax, and, later, chemical conservation agents. No additional conservation was undertaken at this time, with the exception of gluing of refits. Bones were separated by taxa, then classified by element. Identifications were made using published resources (Barone 1976; Gilbert 1990; Pales and Garcia 1981a and b; Pales and Lambert 1971a,and b) and comparative material housed at the Maison Rene Ginouves (University of Paris 10, Nanterre). This collection was created by Andre Leroi Gourhan and continued by Mme Francine David. The NISP per taxon and element is summarized in the following tables. Table 8.3 presents the NISP counts by element for reindeer, horse, red deer, bovids, and hare. Table 8.4 contains the NISP counts by element for bear, hyena, wolf, felids and mammoth. Tables 8.6 and 8.7 present the percentage of total NISP by taxon. Table 8.6 contains the percentages for the herbivores, and table 8.7 presents the proportions of carnivores. 126 Reindeer Horse Red deer Bovids Lepus Antler 11 0 0 0 0 Cranuim 11 1 0 0 0 Mandible 14 0 0 0 0 Maxilla 0 0 0 0 2 Atlas 0 0 0 0 0 Axis 0 0 0 0 0 Vertebrae 10 0 0 0 0 Rib 20 0 0 0 0 Scapula 24 0 0 0 0 Humerus 53 9 0 1 0 Radius 32 3 1 0 0 Ulna 10 1 0 0 0 Radius/ulna 24 0 0 0 0 Carpals 19 1 0 1 0 Metacarpals 41 0 2 0 1 Inominate 3 0 0 0 0 Femur 67 0 0 1 0 Patella 2 0 0 0 0 Tibia 118 15 2 1 1 Fibula 0 0 0 0 0 Astragalus 5 0 0 1 0 Table 8.3. Summary of NISP for reindeer, horse, red deer, bovids and hare from Level Xc. 127 Reindeer Horse Red deer Bovids Lepus Calcaneus 5 0 0 0 0 Tarsals 8 0 0 0 0 Metatarsals 113 0 0 0 0 Sesamoids 16 2 1 1 0 First Phalange 35 1 0 0 0 Second Palange 34 0 0 0 1 Third Phalange 15 0 0 0 0 Metapodials 0 9 0 0 0 Residuals 29 3 0 0 0 Teeth 0 0 0 0 0 Incisors 50 38 0 1 0 Canines 0 2 0 0 0 Max teeth 42 69 1 5 2 Mand teeth 56 40 1 10 0 Unid teeth 6 12 0 0 0 875 206 8 22 7 Total Table 8.3: concluded. 128 Bear Hyena Wolf Felid Mammoth Antler 0 0 0 0 0 Cranium 0 0 0 0 0 Mandible 0 0 0 0 0 Maxilla 0 0 0 0 0 Atlas 0 0 0 0 0 Axis 0 0 0 0 0 Vertebrae 0 0 0 0 0 Rib 0 0 0 0 0 Scapula 0 0 0 0 0 Humerus 5 0 1 0 0 Radius 0 2 0 0 0 Ulna 12 0 1 0 0 Radius/ulna 0 0 0 0 0 Carpals 1 4 1 0 0 Metacarpals 1 3 0 0 0 Inominate 0 0 0 0 0 Femur 17 2 0 0 0 Patella 0 0 0 0 0 Tibia 8 1 0 0 0 Fibula 1 0 0 0 Astragalus 0 1 1 0 0 Calcaneus 0 2 0 0 0 Table 8.4: Summary of NISP for cave bear, wolf, hyena felid and mammoth from Level Xc. 129 Bear Hyena Wolf Felid Mammoth Tarsals 4 4 1 0 0 Metatarsals 6 5 0 0 0 Sesamoids 2 2 2 0 0 First Phalange 7 11 0 0 0 Second Palange 10 10 1 0 0 Third Phalange 10 6 1 1 1 Metapodials 0 7 0 0 0 Residuals 0 0 0 0 0 Teeth 0 0 0 0 0 Incisors 16 0 0 0 57 Canines 27 4 1 0 0 Max teeth 5 0 0 0 0 Mand teeth 2 0 2 0 1 Unid teeth 0 0 0 0 2 135 64 12 1 61 Total Table 8.4: concluded. 130 Reindeer Horse Bovid Red deer Lepus Antler 1.26 0.00 0.00 0.00 0.00 Cranium 1.26 0.49 0.00 0.00 0.00 Mandible 1.60 0.00 0.00 0.00 0.00 Maxilla 0.00 0.00 0.00 0.00 28.57 Atlas 0.00 0.00 0.00 0.00 0.00 Axis 0.00 0.00 0.00 0.00 0.00 Vertebrae 1.14 0.00 0.00 0.00 0.00 Rib 2.29 0.00 0.00 0.00 0.00 Scapula 2.74 0.00 0.00 0.00 0.00 Humerus 6.06 4.37 4.55 0.00 0.00 Radius 3.66 1.46 0.00 12.50 0.00 Ulna 1.14 0.49 0.00 0.00 0.00 Radius/ulna 2.74 0.00 0.00 0.00 0.00 Carpals 2.17 0.49 4.55 0.00 0.00 Metacarpals 4.69 0.00 0.00 25.00 28.57 Inominate 0.34 0.00 0.00 0.00 0.00 Femur 7.66 0.00 4.55 0.00 0.00 Patella 0.23 0.00 0.00 0.00 0.00 Tibia 13.49 7.28 4.55 25.00 14.29 Fibula 0.00 0.00 0.00 0.00 0.00 Table 8.5: Percentages of NISP by element for reindeer, horse, bovids, red deer and hare in Level Xc. 131 Reindeer Horse Bovid Red deer Lepus Astragalus 0.80 0.00 4.55 0.00 0.00 Calcaneus 0.57 0.00 0.00 0.00 0.00 Tarsals 0.91 0.00 0.00 0.00 0.00 Metatarsals 12.91 0.00 0.00 0.00 0.00 Sesamoids 1.83 0.97 4.55 12.50 0.00 First Phalange 4.00 0.49 0.00 0.00 0.00 Second Palange 3.89 0.00 0.00 0.00 14.29 Third Phalage 1.71 0.00 0.00 0.00 0.00 Metapodials 0.00 4.37 0.00 0.00 0.00 Residuals 3.31 1.46 0.00 0.00 0.00 Incisors 5.71 18.45 4.55 0.00 0.00 Canines 0.00 0.97 0.00 0.00 0.00 Max teeth 4.80 33.50 22.73 12.50 28.57 Mand teeth 6.40 19.42 45.45 12.50 0.00 Unid teeth 0.69 5.83 0.00 0.00 0.00 100.00 100.00 100.00 100.00 100.00 Total Table 8.5: concluded. 132 Bear Hyena Wolf Felid Mammoth Antler 0.00 0.00 0.00 0.00 0.00 Cranium 0.00 0.00 0.00 0.00 0.00 Mandible 0.00 0.00 0.00 0.00 0.00 Maxilla 0.00 0.00 0.00 0.00 0.00 Atlas 0.00 0.00 0.00 0.00 0.00 Axis 0.00 0.00 0.00 0.00 0.00 Vertebrae 0.00 0.00 0.00 0.00 0.00 Rib 0.00 0.00 0.00 0.00 0.00 Scapula 0.00 0.00 0.00 0.00 0.00 Humerus 3.73 0.00 8.33 0.00 0.00 Radius 8.96 3.13 0.00 0.00 0.00 Ulna 0.00 0.00 8.33 0.00 0.00 Radius/ulna 0.00 0.00 0.00 0.00 0.00 Carpals 0.75 6.25 8.33 0.00 0.00 Metacarpals 0.75 6.25 0.00 0.00 0.00 Inominate 0.00 0.00 0.00 0.00 0.00 Femur 12.69 1.56 0.00 0.00 0.00 Patella 0.00 0.00 0.00 0.00 0.00 Tibia 5.97 1.56 0.00 0.00 0.00 Fibula 0.75 0.00 0.00 0.00 0.00 Table 8.6: Percentage of NISP by element for cave bear, hyena, wolf, felid and mammoth in Level Xc. 133 Bear Hyena Wolf Felid Mammoth Astragalus 0.00 1.56 8.33 0.00 0.00 Calcaneus 0.00 3.13 0.00 0.00 0.00 Tarsals 2.99 6.25 8.33 0.00 0.00 Metatarsals 4.48 7.81 0.00 0.00 0.00 Sesamoids 1.49 3.13 16.67 0.00 0.00 First Phalange 5.22 17.19 0.00 0.00 0.00 Second Palange 7.46 15.63 8.33 0.00 0.00 Third Phalange 7.46 9.38 8.33 100.00 100.00 Metapodials 0.00 10.94 0.00 0.00 0.00 Residuals 0.00 0.00 0.00 0.00 0.00 Incisors 11.94 0.00 0.00 0.00 0.00 Canines 20.15 6.25 8.33 0.00 0.00 Max teeth 3.73 0.00 0.00 0.00 0.00 Mand teeth 1.49 0.00 16.67 0.00 0.00 Unid teeth 0.00 0.00 0.00 0.00 0.00 100.00 100.00 100.00 100.00 100.00 Total Table 8.6. concluded. Unidentified mammal bone and esquilles Bone fragments which could not be assigned to particular genus or species were classed by size and thickness in comparison to identified bones and comparative material. (Table 8.7). Bone was categorized as cancellous, flat or long. Cancellous bone refers to fragments of mostly cancellous tissue; largely fragments of proximal or distal epiphyses that could not be assigned to a taxon. Flat bone indicates two cortical surfaces with thin 134 internal tissue, such as rib, crania or scapulae. Long bone refers to fragments of diaphysis that could not be assigned to a particular taxon. Bone scraps are fragments of bone larger than 2.5 cm in size that could not be classified. Esquilles, bone fragments less than 2.5cm in size, are tabulated here in Table 8.7, but discussed in the next section. Cancellous Flat bone Small Epiphysis Long Tooth Nutrative Bone bone bone foramen Total scrap 0 0 1 18 0 0 0 19 2 16 2 845 0 2 0 867 12 15 2 588 0 0 1 618 Megafauna 0 0 0 16 0 0 0 16 Unknown 9 53 1 330 5 0 260 424 Esquilles 0 0 0 0 0 0 17,422 17,433 Total 23 84 6 1797 5 2 17469 mammal Medium mammal Large mammal 19,366 Table 8.7: Summary of unidentified mammal bone fragments and esquilles from Level Xc. Large mammal refers to animals the size of bear, bovids, horse and red deer; medium mammal to reindeer or wolf and small mammals to fox and hare. The apparent absence of megafauna (mammoth or woolly rhinoceros) is surprising, given the welldocumented presence of mammoth tusks at the site. However, preservation of megafaunal post crania at Arcy-sur-Cure is very poor. In the Grotte du Bison (the cave 135 connected to the Grotte du Renne), post-cranial bones are extremely rare. Those that are uncovered are extremely friable and almost impossible to conserve, resembling little more than a pile of crumbs or small pine needles. A similar situation was present in the Grotte du Renne, according to Mme Francine David. Given the excellent preservation of other animal bones at Arcy, the poor preservation of megafaunal elements is surprising. I suggest that it reflects a different collection strategy. Large, medium and small mammals present on the site largely reflect the transportation of fresh carcasses or carcass portions for processing by hominins or carnivores. In contrast the megafauna may indicate with the collection of older, dry bones, probably by Neanderthals, that subsequently deteriorated in the moist cave sediments. Figure 8.3: Graph showing the proportion of dry, fresh and undetermined breaks on bone fragments larger than 2.5cm in size. As can be seen from Table 8.7, the majority of bone fragments larger than esquilles are derived from long bones (n=1797 or 92%), and the size class reflects the 136 limited range of species identified, coming largely from mammals ranging in size from bovids to reindeer. A majority (55%) of these have only dry breaks (Figure 8.3). Fortyone percent of the total unidentified mammal bones have at least one fresh break, with 4% with indeterminate breaks. This indicates a high proportion of post depositional breakage following discard for the majority of unidentified mammal bone fragments. This might be the product of trampling or, a common problem in caves, breakage through rock fall. Three tools were identified among the unidentified mammal bones. All were made on the diaphysis of large mammals and all appear to have been used as scrapers with retouch along one or more edges (61.63.A6; 63.C9; A5). These items are currently under study by Michelle Julien, CNRS as part of a larger report on the Châtelperronian levels of the Grotte du Renne. They will be discussed further in Chapter 11. Esquilles are categorized as any unidentifiable bone fragment under 2.5 cm in size. These were collected by meter square, and are mostly the product of water screening. Esquilles were categorized as burnt or unburnt and tabulated by meter square. Of the 17,422 items examined, 679 were potentially identifiable, and largely unburnt. The remaining 16,725 items were classified as unburnt (56%) and burnt (44%). This will be discussed further in the taphonomy section of this chapter. Taphonomy Taphonomy is defined as the process by which organic materials are incorporated into the lithosphere. No substantive faunal analysis can be undertaken without a thorough understanding of the agents of accumulation, modification, deposition and post depositional practices (including excavation) (Lyman 1994). The collection was examined for evidence for humans or carnivores as the primary agents of accumulation by documenting the presence, location and proportion of cut marks and tooth marks and considering evidence for selection of prime age individuals via ambush hunting as 137 opposed to an attritional curve associated with cursorial hunters such as wolves (Binford 1978, 1981; Blurton Jones, et al. 1996; Bunn 1993; Grayson 1991; Kent 1993; Lord, et al. 2007; Lyman 2005; Stiner 1990). Depositional and post depositional processes such as density-mediated attrition (Lam 2003; Lam, et al. 1999; Lyman 1985, 1993), butchery and associated marrow and grease extraction (Binford 1978, 1981, 1984; Bunn 1993; Lyman 2005; Monahan 1998; O'Connell, et al. 1988, 1996; Otárola-Castillo 2010; Perkins and Daly 1968; Speth and Spielman 1983); weathering and other natural agents of destruction (Behrensmeyer and Boaz 1980; Cutler, et al. 1999; Haynes 1988; Hill 1980); carnivore ravaging (Bartram and Marean 1999; Cleghorn and Marean 2004; Faith and Behrensmeyer 2006; Haynes 1983; Lord, et al. 2007; Lyman 1993; Marean and Spencer 1991), and geological processes were considered with reference to reconstruction of the life assemblage. Issues of equifinality (Lyman 1993) and time averaging (Lyman 2003) were also considered. An important aspect of the study is the impact of excavation and collection methodology (Gifford 1981). Recovery processes were examined to understand what faunal remains merited collection or discard. Excavation techniques at the Grotte du Renne maximized recovery of material. Vertical and horizontal controls were in place, in the form of a grid system, and items were piece-plotted on graph paper or vertical photographs. All sediments were waterscreened and all faunal and lithic items recovered and retained. Some items have been deleted from the record as a result of processing for radiocarbon dating, but it is clear that the assemblage from the site has been minimally impacted by excavation procedures and post-excavation analyses. General condition of the assemblage The faunal assemblage from level Xc of the Grotte du Renne was in good condition (Figure 8.4). Only 10 items showed some damage from post-excavation storage (drawer wear or other recent damage to the surface or edges of the bone through 138 abrasion). Weathering was minimal and present on 961 bone fragments which represents 41% of items larger than 2.5 cm in size. Weathering categories were derived from Behrensmeyer (Behrensmeyer 1978). The majority were relatively lightly weathered. 61% showed some longditudinal cracks or light cracking (Level 2) and 22% showed light surface flaking with deeper cracks (Level 3) across less than 50% of the surface area of the bone. Figure 8.4: Chart showing the proportion of weathering present in the Level Xc faunal assemblage. Not all weathering could be ascribed to the loss of grease or drying or spalling of the bones. Chemical weathering was observed on 169 of the weathered bones (Figure 8.5). These are included in the general weathering data above. Chemical weathering intensity was defined by adapting Behrensmeyer’s categories. Light weathering, in the form of occasional pitting, was defined as Level 2 chemical weathering. More concentrated pitting across less than 50% of the bone surface was defined as Level 3 139 weathering. Level 4 pitting indicated an impact on more than 50% of the bone surface. Level 5 pitting was extremely heavy, resulting in little bone surface remaining and Level 6 pitting indicated only bone fibers remained. Figure 8.5: Chart showing the percentage of chemical weathering on bone fragments from Level Xc. Weathering was identified as pitting or leaching produced by acids in surrounding soil. Chemical weathering is distinct from pitting and damage associated with digestive acid etching. Digested bones are generally smooth and waxy in texture, with heavily rounded edges, whereas the chemically weathered bones are dry, with clear breaks and retain their entire cortical thickness. This chemical weathering is probably related to the presence of acid introduced through plant roots, or slightly acidic groundwater. The pitting is therefore the result of post-depositional taphonomic processes. 140 Density values Zooarchaeologists have recognized that assemblage must be controlled for density mediated attrition before any meaningful statements regarding carnivore or human agency (or other accumulation agents) can be made. Density data for Rangifer and Equus were derived from Lam et al. (1999). An Excel table was generated plotting numbers of landmarks by herbivore taxon against measurement sites as defined by Lyman (1984), and utilized by Lam and his colleagues in their later study (Figures 8.6 and 8.7). Red deer and bovids in Level Xc densities were not calculated because only one measurement site/landmark was present for each taxon. Further, only single elements were identified per taxon making any statements about density problematic. Table 8.8 and 8.9 list the elements and density values for horse and reindeer. Figure 8.6: Bivariate plot of the MNE for reindeer by density value for Level Xc of the Grotte du Renne. 141 MNE % Survival Density Antler/horn core 0 0 0 Cranium 7 58.33 0 Mandible 5 20.83 1.05 Atlas 0 0 0.47 Axis 0 0 0.42 vt Cervical 3 3.57 0.42 vt Thoracic 5 3.47 0.53 vt Lumbar 1 1.19 0.49 Rib 5 3.47 0.49 Scapula 4 16.67 0.66 Pelvis 3 25 0.65 P Humerus 10 41.67 0.44 D Humerus 5 20.83 1.08 P Radius 8 33.33 1.04 D Radius 4 16.67 1 P Ulna 4 16.67 0.68 D Ulna 0 0 0 Carpals 19 13.19 0 P Metacarpal 4 16.67 1.03 D Metacarpal 2 8.33 0.6 P Femur 16 66.67 0.52 Table 8.8: Table showing MNE, survivorship and density for reindeer. 142 MNE % Survival Density D Femur 10 41.67 0.61 P Tibia 22 91.67 0.35 D Tibia 13 54.17 1.02 Patella 2 8.33 0 P Fibula 0 0 0 D Fibula 0 0 0 Calcaneus 5 20.83 0.73 Astragalus 7 29.167 0.68 Tarsals 8 8.333 0 P Metarsal 5 20.83 1.08 D Metatarsal 6 25 0.59 P Metapodial 4 8.333 1.08 D Metapodial 4 8.33 0.59 1st Phalange 22 22.92 0.92 2nd Phalange 23 23.96 0.72 3rd Phalange 13 13.54 0.48 Table 8.8: concluded. 143 MNE % Survival Density Cranium 3 75 Mandible 3 37.5 96 Atlas 0 0 0.54 vt Cervical 0 0 0.4 vt Thoracic 0 0 0.49 vt Lumbar 0 0 0.43 Rib 0 0 0.36 Scapula 0 0 0.66 Pelvis 0 0 0.98 P Humerus 1 6.25 0.28 P Humerus 1 6.25 0.28 D Humerus 2 25 1.05 P Radius 0 0 1.04 D Radius 0 0 1 P Ulna 1 12.5 0.65 D Ulna 0 0 Carpals 1 2.08 P Metacarpal 0 0 1.03 D Metacarpal 0 0 0.6 P Femur 0 0 0.35 D Femur 0 0 0.99 Table 8 9: Table showing MNE, survivorship and density for horse. 144 MNE % Survival Density D Tibia 4 50 0.105 Patella 0 0 0 P Fibula 0 0 0 D Fibula 0 0 0 Calcaneum 0 0 0.55 Astragalus 0 0 0.67 Tarsals 0 0 P Metarsal 0 0 1.07 D Metatarsal 0 0 0.71 1st Phalange 0 0 1.02 2nd Phalange 0 0 0.62 3rd Phalange 1 6.25 0.57 Table 8.9: concluded. Bivariate plots for horse and reindeer show no correlation between density values and the number of elements. A strong positive or negative relationship would be indicated by a clear increase or decrease in the number of elements by density. To confirm this apparent lack of relationship statistical analyses were performed on the relationship between density and element parts present. Spearman’s rho was calculated for both taxa and no significant correlation was found between bone density values and survivorship. For reindeer, Spearman’s rho had a correlation coefficient of 0.160, p= 0.109. For horse, the relationship between density and number of elements had a correlation coefficient of 0.039, p= 0.686. The faunal assemblage is therefore not the 145 product of density based attrition. Other factors are operating most strongly on the assemblage. Figure 8.7: Bivariate plot of MNE of horse against density value for Level Xc. Damage by animal gnawing Carnivores are major actors in both bone accumulation and bone deletion from the archaeological record (Binford 1981; Brain 1980, 1981; Cleghorn & Marean 2007; Kent. 1993). Denning carnivores will transport meat to feed cubs or nursing mates, and hyenas in particular are well known for this behavior. Hyenas are also extremely well adapted for bone consumption, which is reflected by the presence of digested bone and bone rich coprolites in the archaeological record. Hyenas are capable of converting identifiable bone into unidentifiable long bone fragments, but other carnivores tend to remove the fat rich articular ends of long bones. Carnivore gnawing is distinctive, resulting in crenellated and rounded edges of bone fragments. Tooth punctures and gouges can also be distinguished on the surface of bones. 146 Rodents are also known to be bone accumulators and to gnaw bone. In contrast to carnivores, many rodents prefer to gnaw dry bone (Klippel and Synstelien 2007). This behavior is associated with the control of incisor growth, and rodent gnawing is usually represented by distinctive double channels from the rodent’s upper and lower incisors. The presence of rodent gnawing indicates use of bones that have lost the majority of their grease or fat. Possible agents of accumulation or destruction at Arcy-sur-Cure include cave hyenas, cave bears, other large and small canids, and at least one large felid. Cave hyenas are known to have been present at Arcy-sur-Cure during the Middle and Early Upper Palaeolithic. At the Grotte du Bison (which adjoins the Grotte du Renne), hyenas played a major role in the formation of the archaeological record during the Mousterian occupation. But a shift in behavior occurs in the Châtelperronian and later Upper Palaeolithic occupations. Spatial analysis indicates that hyena occupations become ephemeral or virtually cease and the caves were utilized predominantly by hominins (Enloe in press; Enloe and Lanoë 2012). Cave bears also utilized the caves for hibernation dens, but their role as bone accumulators is less clear. The same applies to the wolves and felid recorded at Arcy. These occur in very low numbers at the site and it seems that they were not major actors as bone accumulators or destroyers. The evidence indicates that animals played a minor role in bone accumulation and destruction. Only 67 bones showed evidence for gnawing by carnivores or rodents. . Crenellation along the edges of long bone fragments was the most common form of damage, on 49% on the elements, with 40% showing some evidence for channeling and only 11% with puncture marks (Figure 8.8). Of these, 64% showed damage by large or medium carnivores, 21% indicated damage by small carnivores and 15% were damaged by rodents (Figure 8.9). 147 Figure 8.8: Proportions of damage by carnivores to bones in the Level Xc assemblage. Figure 8.9: Chart showing proportions of damage to bones within the Level Xc assemblage by different agents. 148 There was a distinct difference in the size of crenellation indentations between the larger and small carnivores. Given the presence of bear cubs at the site, evidenced by milk teeth lost prior to the cubs leaving the maternal hibernation den, I suggest that the small carnivore damage is the result of teething behavior by cave bear cubs. Gnawing is present on reindeer, bovid, horse and bear bone and on bones of large, medium and small mammal bones. Figure 8.10 below shows the percentages of each taxon. The majority of the damage is on long bones, and primarily long bone shafts or diaphysis fragments. Despite representing 49% of the total gnawed bone, only 31 reindeer elements showed evidence carnivore or rodent damage (Figure 8.11). Gnawing is present on the edges of broken bones and some distal ends. Rodent gnawing is present on five specimens. Seven specimens show evidence of gnawing by a small-toothed carnivore. Four elements exhibit both cut marks and carnivore gnawing, indicating that these elements were butchered prior to damage from carnivores. The amount of carnivore and rodent damage is low, only 3.6% of the total NISP for reindeer. This suggests that Neanderthals were the prime accumulators of the reindeer bone assemblage. No evidence for any carnivore or rodent gnawing is present on the horse specimens. Channeling is present on the proximal shaft of a bovid tibia and at the distal and proximal ends of one red deer metacarpal. Both are consistent with carnivore gnawing. However, the presence of butchery marks on other limb bones of these taxon suggest that the gnawing of the long bones is the product of scavenging by carnivores of bones accumulated by Neanderthals. 149 Figure 8.10: Chart showing proportion of bones by taxon with evidence for gnawing. Figure 8.11: Chart showing proportions of reindeer bones with evidence of gnawing. 150 In addition to the evidence of damage by carnivores through gnawing, four bone fragments show evidence of digestive corrosion. One bone has rounded edges consisted with damage from licking and saliva and three bone fragments have heavy pitting and rounding consistent with passage through the digestive system of a hyena. The low level of carnivore damage within the faunal assemblage of Level Xc indicates that carnivores did not play a major role in the accumulation of the assemblage, nor were they major actors in the destruction of bones within the assemblage. Gnawing on bones represents occasional occupation of the site as a den by cave bears, or scavenging by hyenas or other large carnivores. This pattern is consistent with the finding for the Châtelperronian levels of the adjoining Grotte du Bison. Staining Three reindeer tibia fragments, a hyena meteacarpal , a bear first phalange and an unidentified large mammal bone fragment showed evidence of ochre staining. Ochre covered between 25 and 100% of the bone surface. This staining is likely the by-product of other activities that utilized red ocher at the site. All other staining identified on the bones was modern and associated with the early conservation practice of immersing bone fragments in liquid paraffin. In some cases, this resulted in darkening or blackening of the bones. Another conservation agent resulted in some bones acquiring an uneven rather sparkly (for want of a better term) surface coating, probably the result of super-saturation by a more recent chemical consolidant. Burning Only twenty five bones show evidence for damage by combustion in the Renne Level Xc assemblage. In contrast, 42% of the esquilles (small bone fragments) have been burnt (Figure 8.12). This discrepancy in the data suggests that some process at the site is deleting bones from the assemblage, but in such a way that the bones are completely 151 destroyed or reduced to small, unidentifiable fragments. One possible cause of the low number of burnt bone fragments is the use of bone for radiocarbon dating. Figure 8.12: Percentage of burnt and unburnt bone at the Grotte du Renne level Xc for bone fragments and esquilles. The first radiocarbon assays for Level X were undertaken 1962 on burnt bone which is lower in carbon content than charcoal, so at least a kilogram was needed for a date (David, et al 2001: 226). This was taken from the entire stratigraphic layer and may have considerably reduced the amount of burnt bone represented in the Level Xc faunal assemblage. Another question is how many bones the burnt esquilles represent? When compared with unburnt esquilles, burnt esquilles are generally smaller in size. This may be a by-product of damage to the internal structure of burnt bone, resulting in greater fragmentation. What is interesting is the difference in spatial patterning of the burnt and unburnt esquilles. This will be discussed in the last section of the chapter. 152 Summary of taphonomy The faunal assemblage from Level Xc of the Grotte du Renne is largely the product of hominin behavior. Neanderthals were the primary bone accumulators at the site, and probably the major agents of bone damage and destruction. There is no significant correlation between bone density and deletion of elements from the assemblage, as would be expected with density mediated attrition. Weathering damage to the bones is light, with the majority showing minor damage from gradual drying, or chemical weathering that is likely the product of root action or the slightly acidic groundwater. There is little evidence that carnivores were major agents of bone accumulation or destructions. The proportion of bones with evidence of gnawing or damage from digestion is low. Occasional scavenging of bone occurred, but there is no indication of any major input into the record by carnivores. Hominins may have deleted bone from the record through burning, although the proportion of burnt bone fragments is low. However the relatively high proportion of burnt esquilles suggest that bone was burnt at the site, either as part of processing activities, or through use as fuel. Now that the role of Neanderthals as primary bone accumulators has been established (at least to my satisfaction); the faunal assemblage will be described by taxon. The final section of the chapter will examine the evidence for Neanderthal subsistence behavior in terms of prey selection, site maintenance and selection of supports for bone tools. Herbivores Reindeer (Rangifer tarandus) Reindeer is the most common taxon by NISP and by MNI. Fourteen individuals were identified, 12 adults, identified by the nutritive foramen on the proximal left tibia and 2 juveniles, indicated by the presence of unfused phalanges and limb bones. A total of 875 specimens were identified as reindeer (148 teeth and 727 bone/bone fragments). 153 The reindeer assemblage is highly fragmented. Taphonomic analysis indicates that factors other than density are in operation. The assemblage is not density mediated and carnivore damage is minimal. The prime actors in the formation of the assemblage, and its fragmentation are Neanderthals. The highest numbers of elements by MNE for reindeer are first and second phalanges, proximal tibia and carpals (Table 8.10, Figures 8.13). The high number of phalanges and carpals are a product of the lack of damage to these elements. The MAU data indicate that all elements were transported to the site. Axial elements are underrepresented, with the exception of the cranium (Figure 8.14). Scapulae, vertebrae, ribs and the pelvic basin form less than 10% of the total %MAU. Figure 8.13: Graph showing the number and percentage of MNE per element for reindeer in Level Xc. 154 MNE Expected % Survival MAU %MAU MGUI Antler 0 24 0 0 0 1.02 Cranium 7 12 58.33 7 63.64 17.47 Mandible 5 24 20.83 2.5 22.73 30.26 Atlas 0 12 0 0 0 9.79 Axis 0 12 0 0 0 9.79 vt Cervical 3 84 3.57 0.6 5.45 35.71 vt Thoracic 5 144 3.47 0.4 3.64 45.53 vt Lumbar 1 84 1.19 1.7 15.45 32.05 Rib 5 144 3.47 0.2 1.82 49.77 Scapula 4 24 16.67 2 18.18 43.47 Pelvis 3 12 25 1.5 13.64 47.89 P Humerus 10 24 41.67 5 45.45 43.47 D Humerus 5 24 20.83 2.5 22.73 36.52 P Radius 8 24 33.33 4 36.36 26.64 D Radius 4 24 16.67 2 18.18 33.23 P Ulna 4 24 16.67 2 18.18 D Ulna 0 24 0 0 0 Carpals 19 144 13.19 1.6 14.545 15.53 P Metacarp 4 24 16.67 1 9.09 12.18 Table 8.10: Summary of the Minimum Number of Elements and Minimal Animal Units for reindeer in Level Xc. 155 MNE Expected % Survival MAU %MAU D Metacar 2 24 8.33 0.5 4.54 10.5 P Femur 16 24 66.67 8 72.73 100 D Femur 10 24 41.67 5 45.45 100 P Tibia 22 24 91.67 11 100 64.73 D Tibia 13 24 54.167 6.5 59.09 47.09 Patella 2 24 8.33 1 0 P Fibula 0 24 0 0 0 D Fibula 0 24 0 0 0 Calcaneum 5 24 20.83 2.5 22.73 31.66 Astragalus 7 24 29.17 3.5 31.82 31.66 Tarsals 8 96 8.33 1.33 12.09 31.66 P Metarsal 5 24 20.83 2.5 22.73 29.93 D Metatars 6 24 25 3 27.27 23.93 P Metapod 4 48 8.33 1 9.09 D Metapod 4 48 8.33 1 9.09 Ph 1 22 96 22.92 5.5 50 13.72 Ph 2 23 96 23.96 5.75 52.27 13.72 Ph 3 13 96 13.54 3.25 29.54 13.72 Table 8.10: concluded. MGUI 156 Figure 8.14: Graph showing number and percentage of MAU per element for reindeer in Level Xc. The collection is highly fragmented. The majority of the identified specimens are shaft fragments. Very few have landmarks that permitted calculation of MNE. The fragmentation of metatarsals, tibiae, metacarpals, femora, humeri and radii is striking when the NISP and MNI are compared (Figure 8.15). These bones all have relatively high marrow indices or grease indices (Binford 1978). Clearly the Neanderthals in Level Xc were investing energy in maximizing the returns for these elements. Only seven percent of the assemblage comprises intact elements. Unbroken bones are all small, dense bones: 14 sesamoids, 18 carpals, 7 tarsals 3 astragalae, 1 calcaneus and 10 phalanges. The remaining 93% percent of the bone assemblage comprises broken bones. All reindeer bone fragments are relatively small – the average size ranges from 3.81 cm to 4.8 cm for long bones; and 23.8 cm and 48.4 cm for axial elements (Tables 8.11, and 13; Figures 8.16 and 8.17). 157 Figure 8.15: Graph showing the counts of total MNE and total NISP per element. Humerus Tibia Metacarp. Metatars. Femur Radius Ulna Mean 38.46 46.78 43.13 41.71 42.13 38.21 43.40 Median 38.00 46.25 41.00 39.80 38.00 37.15 39.40 Mode 28.10 39.90 34.20 28.20 40.70 30.20 49.90 Longest 59.60 93.70 76.40 77.80 96.80 77.90 85.90 shortest 10.20 15.50 23.30 11.40 14.00 12.20 18.50 Table 8.11: Summary table of reindeer appendicular skeleton bone fragments, lengths in millimeters. 158 Figure 8.16: Appendicular elements of reindeer showing the median, mode and longest and shortest lengths in millimeters. Antler Crania Mandible Scapula Vertebra Rib Inominate mean 40.72 35.65 34.10 48.37 23.76 35.88 45.60 median 30.50 27.80 32.40 39.10 29.85 36.75 49.10 mode none none none none none none none longest 119.90 60.70 46.60 145.40 38.50 47.30 51.20 shortest 17.20 23.30 25.60 17.40 2.80 19.20 36.50 Table 8.12: Summary table of reindeer axial element fragments, lengths in millimeters. 159 Figure 8.17: Axial elements of reindeer, showing the mean, median, mode and longest and shortest lengths in millimeters. Figure 8.18: Chart showing the proportion of dry, fresh and undetermined breaks by element for reindeer in Level Xc. 160 %dry %fresh % indeterminate 27.27 18.18 45.45 Cranium 0 0 100 Mandible 64.29 0 21.43 Vertebrae 60 0 0 Rib 95 5 0 Scapula 88 0 0 Humerus 22.69 64.15 9.43 Radius 44.64 48.21 3.57 20 30 30 Carpals 5.26 0 0 Metacarpals 4.89 80.49 0 Inominate 100 75.61 12.19 Femur 2.98 0 0 Patella 0 74.63 10.45 4.24 72.03 15.25 Astragalus 0 0 14.29 Calcaneus 0 0 40 Tarsals 12.5 0 0 Metatarsals 14.16 76.99 3.54 Sesamoids 12.5 0 0 Antler Ulna Tibia Table 8.13: Table showing the proportions of dry, fresh and undetermined breaks by element for reindeer in Level Xc. 161 %dry %fresh % indeterminate First Phalange 14.29 60 5.71 Second Palange 2.95 38.24 35.29 Third Phalange 13.34 20 20 Residuals 17.249 3.45 10.34 Table 8.13: concluded. Axial elements are all broken after discard. All axial elements exhibit dry or undetermined breaks, the latter lacking the curvature of fresh breaks but absent a completely jagged edge consistent with dry breaks (Table 8.13). The majority of appendicular elements exhibit fresh breaks that occurred shortly after death (spiral or round fractures). This indicates deliberate breakage of appendicular elements to obtain marrow, and possibly processing for bone grease, as suggested by the high levels of fragmentation for marrow and grease rich elements. The absence of fresh breaks on the proximal mandible, vertebrae and all but one rib fragment (Figure 8.18) is consistent with ethnographic data that demonstrate that these bones contain yellow bone grease which is not valued as a comestible item. The distal mandible does contain marrow, which may explain the lack of intact mandibular tooth rows. Within the appendicular elements, the radius and second and third phalanges exhibit higher proportions of dry breaks than the other elements. The relatively low proportion of fresh breaks on the radius is surprising, because this is an element relatively rich in marrow. Mortality patterns Fusion rates indicate that at least two sub-adult reindeer were part of the faunal assemblage. No intact maxillary or mandibular tooth rows survived, therefore it was not 162 possible to calculate the age of any individuals. Tooth wear patterns suggest that Neanderthals were focusing on prime age individuals. Figure 8.19 shows the wear stages on mandibular molars, using Grant’s (1982) wear stages and Figure 8.20 shows the wear on maxillary molars, using Kilberger and Enloe’s (2005) wear stages. Figure 8.19: Graph showing counts of mandibular molar wear for reindeer in level Xc. Wear patterns on teeth indicate the general age of an animal, but wear rate is relative and impacted by the amount of coarse material consumed. However, it can give a relative indication of age at death. Wear patterns on the mandibular molars were assessed using tooth wear patterns for ungulates illustrated by Grant (1982). Wear patterns on maxillary molars were assessed using data derived from Kilberger & Enloe (2005). 163 Figure 8.20: Graph showing counts of maxillary molar wear for reindeer in Level Xc. Wear patterns and the absence of deciduous teeth indicate a focus on prime age individuals, where teeth show light to moderately heavy wear ,with at least one older individual also present. This pattern is consistent with the hunting strategies demonstrated for Neanderthals and modern humans in the later Mousterian and early Upper Palaeolithic. Butchery Cutmarks are present on reindeer crania, vertebrae, humeri, radii, tibia, carpals and tarsals, metacarpals, metatarsals and phalanges (Appendix, Figures A.1- A.20). Cut marks on the crania occur on the frontal and indicate hide removal around the antlers. Similarly, cutmarks on the first and second phalanges could relate to skinning, although the majority of these occur on the ventral mesial surface suggesting that tendon removal may also be a factor. Butchery on all other elements, with the exception of the carpals and tarsals, relates to meat or tendon removal. Cutmarks on humerus and tibia are largely placed around muscle and tendon attachments. The most interesting pattern for cutmarks 164 is on the metacarpals and metatarsals. These are not meat-bearing bones, but the majority of cutmarks occur on the anterior and lateral shafts, not at the proximal or distal articular surfaces. The location of the cutmarks is clearly not for meat removal, given the low meat values for metapodials (Binford 1978, 1981). Cutmarks are present on the posterior surface of carpals, and the lateral and medial facets of tarsals consistent with disarticulation of the fore and hind-limbs, rather than meat or tendon removal. As noted above, long bones show clear evidence for deliberate breakage. The majority of long bones have fresh breaks, in contrast to the dry breaks on axial bones. Long bones were broken by direct percussion to gain access to marrow. Percussion marks show distinct patterns of strikes above or below the proximal and/or distal epiphyses on the humerus, metacarpals, metatarsals, tibia and radiocubitus. Percussion marks also occur at the midshaft on the radius, tibia and phalanges. Impact marks on the various specimens shows a clustering of impacts in particular locales, suggesting that these breakage patterns are part of an enculturated system of carcass processing, rather than random fracturing patterns. This pattern is also consistent with the use of stone tools or hammerstones to break the bone to access marrow. Modern hunter gatherers such as the Inuit remove reindeer marrow as single unit, by removing the epiphyses and then extracting the marrow from the proximal or distal end of the bone shaft. But these processors have the use of metal knives or blades which remove the epiphyses very efficiently, a technological option not available in the Upper Palaeolithic. Tools Two lissoirs (defleshers) were identified in the reindeer faunal assemblage. Both were on the proximal end of the left tibia shaft. Other tools identified during the analysis included a possible ad hoc tool formed on the diaphysis of a humerus (61.A6(168)). This had been formed by flaking one end to a point to serve as an awl. Another possible tool from an identifiable element was a 165 fragment of an ulna exhibiting heavy polish (B6, no number). These items were given to Mme Julien for further study and no sketches were made. Summary for Reindeer In summary, the reindeer assemblage indicates that reindeer carcasses were transported intact to the site and processed in situ. Butchery patterns indicate that meat, tendons and hide were removed from the carcasses. Bone breakage patterns attest to heavy processing for marrow and bone grease, which will be discussed further in the last part of this chapter. Reindeer also provided supports for tools. Two tools were made on the proximal ends of left tibia shafts, another possible tool on an ulna, and one ad-hoc tools on a humerus fragment. Horse (Equus caballus) Horse was the second most common taxon by MNI and NISP. Four individuals were identified – three adults and a sub-adult, based on three erupted third mandibular molars and an unerupted third mandibular molar in the assemblage. A total of 206 elements was identified (161 teeth or teeth fragments and 45 bone fragments). The most common element (apart from teeth) was the tibia (n= 15). No axial elements were present. Only humerus, tibia, radius, ulna, sesamoids, carpals, metapodials, residual metapodials and a first phalange were identified within the sample, plus a femur diaphysis which was tentatively assigned to the horse category but is not included in the present analysis. Table 8.14 shows the MNE, survivorship and % MAU against MGUI. 166 MNE Expected % Survivorship MAU %MAU MGUI Cranium 3 4 75 3 100 17.9 Mandible 3 8 37.5 2 66.67 7.4 Atlas 0 4 0 0 0 7.8 vt Cervical 0 28 0 0 0 45.2 vt Thoracic 0 48 0 0 0 100 vt Lumbar 0 28 0 0 0 22.4 Rib 0 48 0 0 0 Scapula 0 8 0 0 0 15 Pelvis 1 4 0 0 0 53 P Humerus 1 16 6.25 0.5 16.67 15 D Humerus 2 8 25 1 33.33 14.1 P Radius 1 8 0 0 0 8.7 D Radius 0 8 0 0 0 6 P Ulna 1 8 12.5 0.5 16.67 D Ulna 0 8 0 0 0 Carpals 1 48 2.08 0.14 4.76 3.1 P Metacarp. 0 8 0 0 0 1.6 D Metacarp. 0 8 0 0 0 0.7 P Femur 0 8 0 0 0 45.4 D Femur 0 8 0 0 0 45.4 P Tibia 0 8 0 0 0 25.3 D Tibia 4 8 50 2 66.67 15.2 Table 8.14: Summary table Minimum Number of Elements and Minimum Animal Units for horse in Level Xc. 167 MNE Expected % Survivorship MAU %MAU MGUI Patella 0 8 0 0 0 P Fibula 0 8 0 0 0 D Fibula 0 8 0 0 0 Calcaneum 0 8 0 0 0 7.6 Astragalu 0 8 0 0 0 7.6 Tarsals 0 32 0 0 0 7.6 P Metatars. 0 8 0 0 0 3.8 D Metatars. 0 8 0 0 0 1.8 Metapodial 1 16 6.25 1 33.33 n/a Ph 1 0 16 0 0 0 0.9 Ph 2 0 16 0 0 0 0.9 Ph 3 1 16 6.25 0.25 8.33 0.9 Table 8.14. concluded. As noted above, the horse assemblage is dominated by teeth, which form 78% of the NISP. No intact tooth rows survive indicated destruction through processing of the horse crania, hence the NME of three for horse mandibles, calculated by the presence of mandibular third molars, in contrast to a NISP of zero (Figure 8.21). All post-cranial elements are fractured. This suggests heavy processing of the horse assemblage by Neanderthals (Table 8.15, Figure 8.22). All bones were fragmented. The longest element measured 17.2cm and the shortest 2.2cm, with an average length of 6.14cm. Table 8.15 below gives the mean, median and greatest and least lengths for long bones where more than one specimen was present. 168 Figure 8.21: Graph showing the total MNE and total NISP per element for horse in Level Xc. Humerus Radius Tibia Metapodial Residual metapodials mean 56.38 69.04 74.31 67.56 49.13 median 53.8 70.7 67.2 68.8 37.9 mode none 0 0 0 0 longest 87.1 81.7 172.1 92.1 79.6 shortest 38.6 54.7 24.3 36.2 29.9 Table 8.15: Summary table of horse appendicular skeleton bone fragments, lengths in millimeters. 169 Figure 8.22: Graph showing lengths of long bone fragments for horse in Level Xc, in millimeters. The ulna, a sesamoid and the majority of residual metapodials have only dry breaks. The majority of all other appendicular elements have at least one break that is fresh or recently post-mortem, indicative of processing at the site. This is surprising given the low quantities of marrow present in horse long bones, and given the investment of energy required to fracture the bones. All breaks on tibia and metapodials are fresh. The metapodials are relatively high in marrow quantity and marrow is relatively easy to extract. This suggests processing for marrow of these elements. The breakage pattern on the humerus, radius and tibia are harder to interpret. These are dense bones with relatively little marrow in small marrow cavities. However, these bones do contain white bone grease, and the damage to the bones may reflect breakage to obtain grease. This will be discussed further in the final part of the chapter. 170 %dry % fresh % indeterminate Cranium 0 0 100 Humerus 22.22 77.78 0 0 100 0 Ulna 100 0 0 Tibia 6.67 66.67 6.677 0 100 0 66.67 33.33 0 Sesamoids 50 0 0 Ph 1 100 0 0 Radius Metapodial Residual Table 8.16: Percentage of dry, fresh and undetermined breaks by element for horse. Figure 8.23: Chart showing the proportions of dry, fresh and undetermined breaks by element for horse in Level Xc. 171 Butchery patterns Evidence for carcass processing is weaker for horse than for reindeer. The differences in physiology may account for the absence of cut marks or elements of the skeleton at the site. This could be a result of the greater thickness of flesh ton the upper limb bones of horse carcasses. Cut marks might not be deep enough to impact the bone. Cutmarks are found on the radius, humerus and metapodials. On the humerus, two cutmarks were identified on the deltoid tuberosity. Cutmarks were also present on the ventral midshaft of the radius and on the upper and lower shafts of metapodials (Appendix, Figures A.16-A.20). Impact marks indicate breakage for marrow. Impact fractures are present on the ventral humerus above and below the medullary cavity, at various points on the midsection of the tibia, both ventral and dorsal; and all along the lateral side of the metapodials. Unlike reindeer, which were transported intact, it appears that only certain elements of horse carcasses were processed at the shelter. Crania (indicated by the presence of teeth) and limbs were processed on the site, but there is no evidence for transportation of axial elements such as vertebrae or inominates. This would suggest that horses were killed and preliminary carcass processing is occurring away from the Grotte du Renne. The meat from the richest (and heaviest) portions of the carcass was removed at or near the kill site and for transportation to the Grotte du Renne. Only elements that contained nutrients that require further processing (and their riders) were transported for further processing, namely marrow and fat extraction. Mortality patterns It was not possible to establish any mortality curve for horse at the site. Horse tooth eruption patterns preclude any use of tooth-wear to determine an approximate age curve, and the low number of teeth and elements would make any such determination highly questionable. One third molar was unerupted but well-formed. The third molar 172 erupts between three and four years in modern horses which suggests a sub adult under four years of age is in the assemblage. Summary for horse It seems likely that Neanderthals at the Grotte du Renne targeted prime age individuals in a similar manner to reindeer, but perhaps further from the cave resulting in differential transportation of the carcass parts. Horse longbones were processed for fat and marrow at the site, as indicated by the breakage patterns. As with reindeer,r it appears that Neanderthals were maximizing their returns in terms of fat and protein from the horse carcasses. Bovidae The third most common herbivores by element were bovids with a NISP of 22 and an MNI of 2. The MNI reflects the presence of a bison based on the morphology of the carpal, and a second individual with a femoral morphology indicative of an auroch. Teeth were the most common element of bovids present at the site (an NISP of 16). The presence of mandibular and maxillary teeth indicate that crania and mandibles were also processed at the site, hence the NME of 1 for crania and 2 for mandibles (Figure 8.17). Post crania was represented by single specimens of the humerus, radius, tibia, femur, carpals, astragalus and sesamoids. The longest bone fragment measures 8.97 cm and the shortest 2.47 cm, with an average of 4.41 cm. Butchery The transportation and carcass processing strategy indicates initial processing away from the site, with processing for marrow and meat at the Grotte du Renne. This is the same strategy as that utilized for processing horse carcasses. Crania and long bones were brought to the site. Breaks on the long bones are all fresh, indicating processing for marrow and/or grease at the site. Cutmarks are present above the teres tuberosity on the 173 MNE Left Right Unid Expected % Survivorship MAU Horn core 0 0 0 4 0.00 0 Cranium 0 1 0 2 50.00 0 Mandible 1 1 0 4 50.00 0 Maxilla 1 0 0 4 25.00 0 Atlas 0 0 0 2 0.00 0 Axis 0 0 0 2 0.00 0 Vertebrae 0 0 0 52 0.00 0 Rib 0 0 0 48 0.00 0 Scapula 0 0 0 4 0.00 0 Humerus 1 0 0 4 25.00 1 Radius 0 0 0 4 0.00 0 Ulna 0 0 0 4 0.00 0 Carpals 1 0 0 24 4.17 1 Metacarpals 0 0 0 4 0.00 0 Pelvis 0 0 0 2 0.00 0 Femur 1 0 0 4 25.00 1 Patella 0 0 0 4 0.00 0 Tibia 1 0 0 4 25.00 1 Fibula 0 1 0 4 25.00 0 Astragalus 0 0 0 4 0.00 1 Calcaneus 0 0 0 4 0.00 0 Table 8.17: Table showing Minimum Number of Elements, survivorship and Minimum Number of Animal Units for bovids in Level Xc. 174 MNE Left % Survivorship Right Unid Expected MAU Tarsals 0 0 0 16 0.00 0 Metatarsals 0 0 0 4 0.00 0 Ph 1 0 0 0 16 0.00 0 Ph 2 0 0 0 16 0.00 0 Ph 3 0 0 0 16 0.00 0 Metapodials 0 0 0 32 0.00 0 Table 8.17: concluded. humerus, and faint cutmarks were evident on at the base of the tubercle on the femur. A single impact fracture was noted on the distal portion of the midshaft of the humerus. Butchery and bone fracture patterns are consistent with the removal of meat or tendons and the acquisition of marrow and bone grease. Red deer (Cervus elaphus) Red deer had MNI of 2, based on an unfused radius, and the presence of fully erupted and worn maxillary and mandibular teeth. The NISP totaled 8. Elements present were largely from the appendicular skeleton: a radius, two metacarpals a tarsal, and two tibia. Teeth present indicated that elements of the mandible and maxilla had been present and therefore NME for the crania and mandible were based on the presence of these teeth. Bone fragments ranged in length from 10.5cm to 2.2cm with an average size of 72.8cm. All breaks on long bones were fresh, indicating processing soon after death. No traces of cutmarks were found on the bones, but one metacarpal exhibited fracture marks along one longitudinal edge, consistent with splitting for marrow. 175 MNE Expected % Survivorship MAU Antler 0 4 0 0 Cranium 1 2 50 0 Mandible 1 4 25 0 Maxilla 0 4 0 0 Atlas 0 2 0 0 Axis 0 2 0 0 Vertebrae 0 52 0 0 Rib 0 48 0 0 Scapula 0 4 0 0 Humerus 0 4 0 0 Radius 1 4 25 1 Ulna 0 4 0 0 Radius/ulna 0 4 0 0 Carpals 0 24 0 0 Metacarpals 1 8 12.5 2 Pelvis 0 2 0 0 Femur 0 4 0 0 Patella 0 4 0 0 Tibia 2 4 50 2 Fibula 0 4 0 0 Astragalus 0 4 0 0 Calcaneus 0 4 0 0 Table 8.18: MNE, survivorship and NISP for red deer from Level Xc. 176 MNE Expected % survivorship MAU Tarsals 0 16 0 0 Metatarsals 0 8 0 0 Ph 1 0 16 0 0 Ph 2 0 16 0 0 Ph 3 0 16 0 0 Table 8.18: concluded. The transportation of crania, as indicated by the teeth, and marrow rich appendicular elements indicates that red deer were processed in a similar manner to the other larger herbivores (horse and bovidae). Megafauna Megafauna are represented by a third phalanx of a mammoth and fragments of mammoth ivory. The mammoth ivory is likely associated with the use of mammoth tusks as part of the structures identified in level X. No statements can be made about the presence of the phalanx other than to recognize its presence at the site. No butchery marks or other traces of human or carnivore damage are present. Hare (Lepus sp.) The only small mammal represented in the collection is hare (NISP=6, MNI=2). Two left maxillary tooth rows indicate the presence of two individuals and two maxillary teeth are also present. Post crania comprise single specimens of a left tibia, a metacarpal and a phalange The few elements present were found in the midden areas, suggesting that this formed a small component of the diet. No butchery or other indications of processing were present. 177 Carnivores Four taxa of carnivore were present in Layer Xc: cave bear, cave hyena, wolf and a large cat. At Arcy-sur-Cure, both cave bear and cave hyena have a long history of using the caves for dens, particularly in the Mousterian. The faunal remains of carnivores in the Châtelperronian levels shows minor use of the site as dens, and the majority of the remains present strongly suggest human agency in bringing the material to the site. Cave bear (Ursus speleaus) This was the most common carnivore present (NISP=133, MNI=7) (Table 8.19). At least 5 individuals are cubs, represented by deciduous teeth. Two adults are also present, represented by permanent teeth and fully fused limb post crania The presence of cubs demonstrates occasional use of the site as a cave-bear den. These cubs left the cave in the spring, taking their new adult teeth with them. No sub-adult post crania are present in the bear assemblage. The adults in the cave do not appear to have died during hibernation. Butchery patterns indicate that these elements are present as a result of Neanderthal behavior. Fresh breaks occur largely on long bones with marrow cavities, indicating marrow extraction (Table 8.20, Figure 8.24). Ethnographic data shows that bear fat was valued as fuel for lighting and for oil in cooking (Pastoureau 2007). The processing of cave bear elements again indicates that Neanderthals in Level Xc were maximizing their returns in terms of oil and fat Butchery Traces of butchery are evident on the tibia, the femur, and ulna (Appendix, Figures A.21- A.24). There are at least 12 first and second phalanges with single or multiple cutmarks on both the dorsal and ventral sides of the shaft. In contrast, no cutmarks were identified on third phalanges. 178 MNE Left Right Unid Expected % Survival MAU Cranium 0 0 1 2 50.00 1 Mandible 1 1 0 4 50.00 1 Maxilla 0 0 0 4 0.00 0 Atlas 0 0 0 2 0.00 0 Axis 0 0 0 2 0.00 0 Vertebrae 0 0 0 52 0.00 0 Rib 0 0 0 48 0.00 0 Scapula 0 0 0 4 0.00 0 Humerus 2 1 0 4 75.00 1 Radius 0 0 0 4 0.00 0 Ulna 1 1 0 4 50.00 1 Carpals 0 1 0 28 3.57 1 Metacarpals 1 0 0 20 5.00 1 Inominate 0 0 0 2 0.00 0 Femur 2 1 0 4 75.00 1 Patella 0 0 0 4 0.00 0 Tibia 2 1 0 4 75.00 2 Fibula 1 0 0 4 25.00 1 Astragalus 0 0 0 4 0.00 0 Calcaneus 0 0 0 4 0.00 0 Tarsals 3 1 0 28 14.29 1 Table 8.19: Table showing the NME, % survival of elements and NISP for adult bears in level Xc. 179 MNE Left Right Unid expected % Survival MAU Metatarsals 3 3 0 20 30.00 1 Ph 1 0 0 2 40 5.00 1 Ph 2 0 7 0 40 17.50 1 Ph 3 7 2 1 40 25.00 1 Table 8.19: concluded. % dry % fresh % undetermined 20 80 0 8.33 75 0 Carpal 0 100 0 Femur 0 82.35 11.76 Tibia 0 62.5 0 Fibula 0 0 100 Tarsal 0 0 50 Ph2 0 0 10 Humerus Ulna Table 8.20: Table showing percentage of dry, fresh and undetermined breaks by element for cave bear in Level Xc. 180 Figure 8.24. Percentage of dry, fresh and undetermined breaks by element for cave bear in Level Xc. The cutmarks on the phalanges are in an intriguing pattern, which suggests that the third phalanges were either retained with the hide, or that the third phalanges were removed for other purposes. A possible explanation is a desire to use the bear claws for decorative purposes, either as part of a robe, or as personal ornaments. Modern ethnography has many examples of use of bear claws as pendants or as parts of necklaces, for example the Mesquaki bear-claw necklaces on display in the University of Iowa Museum of Natural History in Iowa Hall. Given that bear teeth are known to have been worn by Neanderthals at the Grotte du Renne, we can speculate (but no more than speculate) that bear claws were also reserved for symbolic expression. Percussion impacts were found on the lateral midshaft of the humerus, and on the midshaft of the femur, suggesting breakage to access the marrow cavity, as discussed above. It appears that by the Châtelperronian period, cave bears formed part of the susbsistence strategy at the Grotte du Renne – the adult bears appear to have been processed for meat and marrow, and skinned for their hides. As to the role of the bears in 181 personal adornment, it is likely that hides were taken for clothing, and that the teeth and possibly claws were used for personal adornment. Hyena (Crocuta spelaeus) Hyenas are represented by both adults and infants. (MNI=3, NISP=64). Of the three individuals present, two are represented by upper right deciduous canine teeth and partially fused long bones (Table 8.21). Only a single adult is present. The adult hyena had an unciform fused with the third carpal. There is a considerable amount of additional bone around this joint, which suggests either disease or a healed injury, or possibly some form of arthritis (Figure 8.25). Figure 8. 25: Photograph showing fused hyena unciform and third carpal with associated bone growth. 182 Only teeth and a femur, a fibula and carpals, tarsals, and elements of the feet were identified. All breaks are dry or indeterminate with the exception of four fresh breaks: two on metacarpals and two on radii. Some researchers argue for hyenas as bone accumulators in the Châtelperronian levels at the Grotte du Renne (e.g. Higham, quoted in Nature October 2012). In Level Xc, the low MNI, low NISP and the relative absence of digested bone does not support this argument. In fact, it seems that there is a strong human influence in the accumulation of hyena bones, given the amount of butchery present, and fresh breaks on some elements. Butchery Faint cutmarks are present on the fibula. There are two impact cones on the radius, and two cutmarks on the distal epiphyses of the same element. Cutmarks are also evident on the calcaneus, carpals and the midshaft of a metacarpal that also exhibits black and red ochre staining on the posterior surface (Appendix, Figures A.25- A.26). The location of the cutmarks on the carpals and tarsals suggests removal of hide rather than disarticulation to obtain meat. The acquisition of hides from hyenas is also suggested by the absence of other post-crania. The elements present at the site may be “riders” transported along with a hide that required further processing. Hyena pelts are attractive, with striped or spotted markings. I am not aware of any modern society that consumes hyena meat. Hyenas, even when killed by other carnivores, tend not to be eaten. In contrast to the cave bears (known from isotopic studies to be omnivorous) cave hyenas were obligate scavengers. It seems likely that, like many terrestrial carnivores and scavengers, their flesh was not palatable. 183 MNE Left Right Indet Total Cranium 0 0 0 0 Mandible 0 0 0 0 Maxilla 0 0 0 0 Atlas 0 0 0 0 Axis 0 0 0 0 Vertebrae 0 0 0 0 Rib 0 0 0 0 Scapula 0 0 0 0 Humerus 0 0 0 0 Radius 0 2 0 2 Ulna 0 0 0 0 Radius/ulna 0 0 0 0 Carpals 0 4 0 4 Metacarpals 1 3 0 4 Inominate 0 0 0 0 Femur 0 1 0 1 Tibia 0 1 0 1 Fibula 0 0 0 0 Table 8.21 Table showing the Minimum Number of Elements for hyena in Level Xc. 184 MNE Left Right Indet Total Astragalus 1 0 0 1 Calcaneus 2 0 0 2 Tarsals 3 1 0 4 Metatarsals 3 2 0 5 Sesamoids 0 0 2 2 First Phalange 3 5 3 11 Second Palange 3 4 3 10 Third Phalange 0 0 6 6 Metapodials 0 0 7 7 Table 8.21: concluded. Wolf (Canis lupus) Two wolves, an adult and a sub-adult, are present based on the amount of fusion present on a phalange and metapodial. The majority of the faunal remains (NISP =12) are from the lower limbs: two sesamoids and single specimens of carpals, metapodials, and phalanges, plus the trochlear notch of an ulna. Single teeth from the maxilla and mandible are also present. The dominance of lower limb, non-meat-bearing elements suggests transportation of hides or pelts to the site. The absence of teeth and larger limb bones suggests that these remains are not the result of natural deaths in the cave, but that the feet bones of wolves were imported by other agents, most probably Neanderthals. However, no butchery marks are present which would confirm hominins as the agent of accumulation. 185 Felidae A large felid is represented by a third phalange with a cutmark on the distal end (Appendix, Figure A.27). This hints at hide removal, but no further inferences can be made. Neanderthal subsistence and behavior Taphonomic analysis of the Level Xc assemblage determined that the processes that resulted in the faunal assemblage were not the product of density-mediated attrition or carnivore ravaging. The major accumulators of the assemblage were Neanderthals who practiced three different transportation strategies related to the fauna present at the site. First, only the lower appendicular elements of fur bearing carnivores are present, indicating transportation of pelts rather than carcasses. Second, large animals such as horse, red deer and bovids were not transported in their entirety to the site. Only elements low in meat value but high in marrow or bone grease were transported to the Grotte du Renne for additional processing. The same pattern applies to the adult cave bear at the site, which were also exploited for pelts, meat and fat. The third category was the medium sized herbivore, reindeer. Reindeer carcasses were transported to the site and then butchered in or near the Grotte du Renne. Element selection and transportation of preferred carcass parts and associated riders can strongly influence the archaeological faunal assemblage. The Modified General Utility Index (Binford 1978:74) indicates the general meat and fat values of carcass elements. A high MGUI indicates a high value element, and conversely a low MGUI an element that is unlikely to be transported from a kill site for its nutritional value in terms of protein and fat. The MGUI and the similar Food Utility Index (FUI) can indicate if an archaeological site is a kill site, a butchery site or a home base by calculation of the values of elements discarded on site. However the MGUI may obscure other carcass processing activities, particularly the rendering of bones for marrow and 186 bone grease (the hard white fat stored in cancellous bone). MGUI values for reindeer were calculated based on the MNI of 12 adult individuals, excluding the two sub-adults because the original MGUI values were calculated from an adult and it is not clear that sub-adults would have the same MGUI by element given their ongoing growth and possible depletion of resources. In addition, sub-adult skeletal elements may be less likely to survive as a factor of bone density rather than any processing by hominins on the site. The MGUI index is a calculation of the value of an item for both its meat and fat content. The relationship expressed here may be skewed by a focus on selection or deletion in of items of the faunal assemblage for fat processing. In consequence, it cannot be determined if a gourmet or generalist strategy was employed in the transportation of large herbivores. There are too few Minimum Animal Units (MAU) for bovids or red Figure 8.26: Element selection strategy by Neanderthals for reindeer in Level Xc. 187 deer for these data to be meaningful, and horse elements present are largely of low General Utility Value. The transportation and consumption strategy can be calculated for reindeer. The plot of %MAU against the MGUI for reindeer does not show a maximizing of quantity (a bulk strategy) or quality (a gourmet strategy) but rather an unbiased or generalist strategy (Figure 8.26) (Binford 1978:81). The transportation of horse elements was examined with reference to the food utility index (FUI) calculated by Outram and Rowley-Conwy (1998:845). Bivariate plotting of food utility against elements present does not appear to show a negative or reverse utility curve between the general utility of an element and the presence on the site (Figure 8.27). Survival rates also do not show a strong relationship between food utility and survival as both very low and very high valued items are absent from the archaeological assemblage. Horse crania and mandibles present were calculated by the presence of the second premolar or third molar. . Figure 8.27: Element selection strategy by Neanderthals for horse in Level Xc. 188 The plot of food utility against % MAU for horse confirms the evidence for transportation of parts of the carcass for processing, namely the crania (as evidenced by the high number of teeth) plus the humerus and tibia. The differential transportation could indicate that horse carcasses were obtained at a greater distance from the site, and initial processing and filleting of meat took place at the kill site. The elements transported to the site have low general utility measures. As these elements are relatively low in marrow, with the exception of the mandible, it appears that these elements may have been selected for fat processing. One confounding factor in Binford’s (1978 and 1981) discussion and other discussions of butchery patterns for subsistence purposes is the realization that the “schlepp effect” is real and has real meaning for the interpretation of zooarchaeological collections (Perkins and Daly 1968). At the Grotte du Renne, the main site function is that of a habitation, not a hunting camp or butchery stand. As a result, transportation decisions regarding carcass or carcass parts probably related to the distance to the site, the number of hunters, and the size and condition of the prey. It is clear that reindeer were processed as whole carcasses to the site, or processed near the site but larger animals (bovids, horse, red deer and bear) are only represented by limb bones and crania. This suggests differential transportation based on the size of the carcass or the distance to the site. Absence of meat-rich elements suggest that that meat from the axial skeleton of larger herbivores was removed from the bones and transported to the site, with the cranium (ready packaged brains and tongue) and the limb bones removed from the carcass for processing at the site. Never the less, these larger herbivores and one carnivore seem to represent a less important part of the diet. Unlike the herbivores (all prime age) and possibly the cave bear, hyenas were taken for their pelts. The adult hyena appears to be an older animal that was hunted for its pelt, with the hide removed from the carcass and tarsals and carpals and the feet discarded at the site. The presence of only lower limb bones and feet of the wolf also 189 suggest acquisition of hides – again with the lower limbs removed, but the absence of cutmarks means that this is only indicated, not demonstrated. Evidence for fat processing Ethnographic and physiological data have shown that fat is extremely important in any meat-based diet (Speth and Spielman 1983). Fat is obtained from liquid fat (marrow) and hard or white fat which is found around certain organs (for example the kidneys) and in the cancellous tissue within the skeleton. The absence of large fragments of proximal and distal long bones, and the absence of much of the axial skeleton may be the result of fat processing by Neanderthals. The presence or absence of kidney fat cannot be computed, but examination of evidence for processing of bones for marrow and grease is possible. Binford (Binford 1978:158) has described how bone grease is produced by the Numamuit. Articular ends of long bones were pounded into fragments and then boiled to render the grease. This is skimmed and stored and the pulverized bone discarded. Prior to the introduction of metal cooking pots, bone grease manufacture was an extremely intensive procedure requiring hot rocks to boil the grease in wooden buckets. This resulted in an extremely distinctive signature of heaps of fire cracked or fire altered rock, plus piles of bone fragments or bone chips (Binford 1978:159). Archaeologists assume that stone boiling is necessary to render bone grease from bone fragments based on ethnographic examples of stone boiling. I am not aware of any large concentrations of fire altered rocks in Level Xc, although there are a large number of burnt or heat altered lithics, but it is possible to process bone grease without the use of hot rocks. There are a number of ethnographic and experimental studies that document the cooking of food and heating of water in hide or bark containers over direct heat, if there is sufficient liquid to keep the container moist (Speth 2012:28). The absence of large amounts of fire cracked rock should not, therefore, 190 lead us to assume that grease was not rendered by Neanderthals at the site. The high level of bone breakage suggests intense processing of all marrow and grease bearing bones. According to Speth, the Inuit consume high amounts of fat, not lean meat, in their diet: protein forms approximately 25% of their total dietary intake (Speth 2012). Isotopic analysis indicates high levels of meat consumption by Neanderthals which , according to the Inuit data, indicates that a high level of fat intake would also be required to optimize protein absorption. In his study of the Numanuit, Binford found that bones containing white grease from appendicular elements were rendered. Scapulae, carpals and phalanges generally were not processed (probably because they were generally low in grease). Mandibles, ribs and vertebrae were not processed because they contained undesirable yellow fat. Binford calculated that the bones with the highest white grease values were the distal femur, the proximal humerus and the proximal tibia of reindeer. A similar pattern is seen in bison, where grease weight is highest in the proximal humerus, the distal femur, the proximal tibia, proximal radioulna and proximal femur (Morin 2007:77). No intact proximal or distal ends survive for large long bones in the Level Xc faunal assemblage. The high degree of fragmentation and dominance of shaft fragments in the assemblage suggests deliberate destruction by hominin occupants of the site. The bone breakage rate is high, particularly for herbivores. No intact long bones survive for reindeer, horse, bovids or red deer. Surviving complete bones are carpals, tarsals, sesamoids and phalanges. As seen above, the majority of the breaks on herbivore and omnivore bones are fresh, while the majority of breaks on carnivore bones are dry, indicating deliberate breakage. It appears that Neanderthals were optimizing their calorific returns by extracting marrow and grease from the carcasses and carcass parts brought to the site. 191 Discard patterns Neanderthals processed and discarded animal bone in Level Xc. The spatial organization may provide data on how space on the site was organized. One or two shelters are present at the north end (rear) of the cave, which are interpreted as living spaces. The processing and discard of the components of the faunal assemblage may give some indications as to where Neanderthals performed subsistence activities at the site. Discard patterns may also shed some light on how Neanderthals maintained their space. Both the identified bone and the unidentified elements show the same spatial patterning (Figures 8.28 and 8.29). A dense area of discarded bone is located in square A13, in association with a dense ash lens. This appears to be between the two possible structures shown by the post holes. In contrast to this concentration of ash, the two smaller ash concentrations, within the shelter, appear to be relatively free of bone fragments. This indicates that the space within the shelters was kept relatively clean and free of bone debris. A second area of bone disposal is located towards the front of the shelter, along the east side of the cave. A possible interpretation of this concentration of remains is an area of discard associated with processing of carcasses and bones that occurred outside the shelter area, as defined by the postholes. This is a talus area, and bones may have been dropped on the edge of the slope, or pushed away from the more level area outside the shelters. The latter area served as a processing area. The distribution of esquilles does not follow the same pattern as that of the larger bone fragments (Figures 8.30 and 8.31). There is no dense concentration associated with the bone disposal area and large ash lens in A13. Dense concentrations of both burnt and unburnt esquilles occur in Y12 and B12, associated with the nearby small ash/hearth areas. This may relate to hearth cleaning activities and general clearing of the areas within the shelters, that removed larger bone fragments but left smaller items trampled into the soil or fallen between rocks on the living surface. 192 Figure 8.28: Map of Level Xc showing the percentage of all identified bones by grid square. 193 Figure 8.29: Map of level Xc showing the percentage of unidentified bone fragments by grid square. 194 Figure 8.30: Distribution map of unburnt bone splinters (esquilles) by percentage. 195 Figure 8. 31: Distribution map of burnt bone splinters (esquilles)s by percentage. 196 A second large concentration of burnt esquilles in B7 may indicate the presence of another hearth nearby. This was destroyed by a test pit excavated by M. Poulain (F. David, pers. comm.). The distribution of esquilles in the front portion of the Grotte du Renne follows the pattern of the larger elements and again indicates systematic disposal of debris from nearby processing activities. Discard patterns confirm the primary role of hominins as the agents of accumulation. In the neighboring Grotte du Bison, cave hyenas were found to prefer the rear of the cave as a locus for denning activities, as were cave bear (Enloe & Lanoë 2012). The open area at the front of the cave is also unlikely to have been favored as a hibernation location by cave bears. In contrast the open. well-lit and south-facing talus area would have been a favored location for processing or other activities undertaken by hominin occupants of the site. Summary of subsistence activities Processing activities at the site included skinning and butchering of reindeer carcasses; processing of carcass parts of horse, red deer, bovids and bear; and hide processing. The latter is indicated by the presence of two broken defleshers made on reindeer tibia. Other indications of hide processing at the site are the awls found in Level Xc. Clearly the occupants of the site had to organize their subsistence to incorporate time for hide processing, marrow and fat processing and carcass rendering. The cleaning of the habitation areas and disposal of processing debris outside the shelters suggests that Neanderthals occupied the site for a sufficient period of time that it was necessary to keep the living areas clear. The dominance of reindeer is consistent with the cool, arid environment. Horse and at least one bison are present, indicating open grassland near the site. Red deer are generally known as a woodland species, but also occupy open parkland and moorland in the northern parts of temperate Europe. The faunal assemblage reflects an encounter 197 strategy to acquire protein, in the form of a variety of herbivores from the local environment. Conclusion In conclusion, the faunal assemblage from level Xc of the Grotte du Renne is the product of Neanderthal subsistence behavior. Analysis of the remains indicates that a generalist strategy was employed to hunt large and medium sized herbivores. Horse, bovid and red deer carcasses appear to have been butchered at or near the kill site and portions of the carcass, mostly the appendicular skeleton and head, were transported to the habitation site for additional processing. In contrast, reindeer carcasses appear to have been transported as a single package to the site and then processed for meat and fat. The location of surviving cutmarks indicates removal of meat and tendons for cordage. The high level of bone destruction and the absence of white grease bearing elements suggest that the carcasses were processed for bone grease as well as marrow. This would be consistent with the high fat requirement of a protein heavy diet, which northern Neanderthal populations are known to have followed. There is evidence of minor use of the site by carnivores 1, with cave bear deciduous teeth indicating use of the site as a cave bear hibernation den. However, adult bears appear to have been transported to the site as prey. Fresh breaks on long bones suggest marrow processing. Cut marks on the second phalanges indicate removal of the hide, while retaining the third phalanges and claws, most likely for display purposes. Cutmarks on hyena elements also suggest exploitation of this carnivore for fur, and the presence of the lower limbones of a wolf may also indicate hide acquisition. Cutmarks on reindeer crania and lower limbs also indicate removal of the hide for hide processing. The 1 David and Poulain (1990) also report fox (Vulpes vulpes) in Level Xc. This represents single individual and the materaial was not available for analysis during the current study. 198 faunal evidence suggests that Level Xc of the Grotte du Renne was a locus of meat, fat and hide processing. At least five bone tools were recovered during the analysis. These include three small scrapers made on long bone fragments, and two broken defleshers made on reindeer tibia. Two other elements (a horse humerus and an ulna fragment) may also show evidence of working. In Chapter 11 we will consider the use of bone tools at Arcysur-Cure and elsewhere. Did the animals that provided the hides also provide the tools, or were other elements specifically selected and curated for use? Before examining this question, we will consider the evidence from Abri Cellier for Aurignacian subsistence in the following two chapters. 199 CHAPTER 9: ABRI CELLIER: POLITICS AND PREHISTORY Introduction Abri Cellier is a large rock shelter situated on the north side of the Vézère River in the Dordogne region of France, located on a tongue of limestone west of the Combede-Vergne, a long narrow valley that provides access to upland areas, and east of a ravine (the Vallon de Plazac). Figure 9.1: Map showing the location of Abri Cellier, Commune de Tursac, Dordogne. 200 The site of Le Moustier lies 300 meters east, across the Combe-de-Vergne (IGN 1:24000 map). Abri Cellier overlooks the valley of the Vézère and across the Combe de Vergne. The site is therefore well-positioned to monitor the movement of game animals north and south (to and from uplands to the north), and east and west along the Vézère. A series of excavations in the early twentieth century revealed a series of occupations dating from the Mousterian to the Aurignacian, with a possible ephemeral occupation in the Gravettian. The use of Abri Cellier as a lookout continued during the historic period. Although not on the same scale as La Madelaine, the site was used as a troglodyte dwelling during the early medieval period, a refuge from Norse and other raiders who rowed up or down the river valley. The south facing aspect of the site provides both maximum insolation and shelter from northerly winds. Site location in a regional context Although Abri Cellier was also occupied during the Aurignacian, the site location reflects a pattern common in the Mousterian of the Middle Vézère valley. White (1985) argued that Upper Palaeolithic sites in the valley focused on fords and other river crossings and that the organization of settlements reflected a logistical subsistence pattern based on the interception of herd animals as they crossed the rivers or as they came to watering places. This interpretation presupposes large scale seasonal movements of animals. Rather like the problem of the Inuit in the Magdalenian, Palaeolithic researchers should be wary of assuming large scale migration in all regions of Europe (to use the Inuit analogy, we would see North Slope Herds in the Perigord). Some populations are more likely to migrate from low elevations in the winter to higher ground in the summer, within the same region or territory. Indictors of seasonality that might be of use in understanding reindeer migration would include the presence/absence of antler on male and female adults; tooth eruption sequences of juveniles and subadults; or isotopic analysis of trace element that can indicate movement across different landforms. Isotopic 201 analysis of bovid and reindeer teeth from the late Middle Palaeolithic site of Jonzac determined that the bison remained in the locality year-round while reindeer were present in the winter (Britton, et al. 2011). Reindeer migration is not completely predictable. Variation in movement occurs in relation to population density and resource stress. Despite these caveats, Mousterian sites would be well placed to exploit seasonal movement of migratory animals and more local movements of non-migratory animals between different elevations. White’s study assumed that the landscape of the valley remained relatively unchanged since the end of the Pleistocene. Apparently the long history of post-Pleistocene agriculture and related soil erosion had little impact on the structure of the valley floor. A more skeptical view would suggest that more alluviation and/or downcutting and reworking of the valley floor and stream terraces have buried Mousterian flood plain occupations. Despite the presence of Aurignacian deposits, the site of Abri Cellier does not fit the Upper Palaeolithic settlement pattern posited by White. It is located at a distance from any significant modern stream, and high above the current flood plain. The site location fits better with Mousterian occupation patterns in the region, which are interpreted as an expression of a foraging or encounter strategy within more heterogeneous upland environments. A recent re-analysis of early Upper Palaeolithic settlement patterns in the region shows that Early Upper Palaeolithic sites do not share the same settlement system as later Upper Palaeolithic sites in the middle Vézère valley (Sisk 2011). Early Upper Palaeolithic sites are located at lower elevations than Mousterian sites, and closer to rivers, but all retain good viewsheds and are able to exploit a variety of environments. This suggests a less logistical strategy than that posited by White for the Upper Palaeolithic, and indicates a flexible subsistence strategy that could have included both foraging for locally available prey animals as part of an encounter-based strategy and the interception of predictable prey species, as part of a more logistical strategy. Later Upper Palaeolithic sites are closer to river crossings, and this is argued to show a logistical 202 pattern focused on predictable migratory herds. However this only appears in the Solutrean and Magdalenian, indicating a shift in subsistence practices from the Early to the Later Palaeolithic (Sisk 2011). Interestingly, in his study, the 13 Châtelperronian occupations in the region all occur under Aurignacian strata, suggesting similarities in site choice and subsistence strategies in the transitional period. The location of Abri Cellier supports evidence for a Mousterian component to the site. This is not present in the material excavated by the Beloit College expedition, but a Mousterian component was reported by Peyrony (1946). The reuse of the site during the Aurignacian suggests that at some periods during the Early Upper Palaeolithic the site remained attractive as a base for subsistence activities. All politics is personal, even in archaeology The excavations at Abri Cellier occurred during a period when American museums were in the process of acquiring Palaeolithic materials for teaching or display purposes by a mixture of purchase, exchange and excavation (Straus 2002). In the late nineteenth and early twentieth century there was relatively little control over the excavation or sale of archaeological material in France. By this time there was a wellestablished market in Palaeolithic and later artifacts in Europe. American museums became particularly active after World War I, when impoverished French landowners or archaeologists were particularly ready to sell material. While the Smithsonian relied on exchange and excavation to obtain material, other museums with private funding were able to purchase items or entire collections from French excavators, researchers or landowners prior to the onset of the Depression. During this period White (2002) estimates that over 150,000 French artifacts were shipped to North America. It should be noted that other museums and collectors in Europe were also profiting from the economic hardships that encourages sales of antiquities. Sale of prehistoric items predated the arrival of the Americans. The French themselves were not loath to decorate their gardens 203 with menhirs and dolmens from Brittainy or stalactites from newly discovered caves (Hurel 2007). Late nineteenth century German interest in the Palaeolithic resulted in intensive research and collecting. In the Dordogne this occurred under the auspices of Otto Hauser (a Swiss national), who became the bête noire of the French archaeological establishment. His rupture with Peyrony (known as the Affaire Hauser), purchase or lease of many major sites, anti-clericalism, association with de Mortillet and work with German museums had consequences for later foreign archaeologists in the Perigord. Peyrony’s antipathy to Hauser benefitted American researchers such as McCurdy who were encouraged to excavate sites to keep Hauser out. After 1914 this situation changed somewhat. Peyrony was alarmed at the loss of the French patrimoine to foreigners, but there was no legal regulation to protect sites. After 1913 it was possible to give some protection to sites by classification as a ‘monument historique’ but the site and artifacts remained the private property of the landowner. The only option of the French authorities was to lease as many sites as possible and, in the 1920s, classify many sites as historic monuments. All foreigners needed (until legislation passed in 1941) to excavate was a good relationship with one of the archaeological authorities and legal access, via lease or purchase, to a particular site (White 2002). Furthermore, many of the French excavators were amateurs who lacked access to government funding. Sale of duplicate artifacts enabled them to fund their research and excavations. The Logan Museum of Anthropology at Beloit College became embroiled in the aftermath of the Hauser Affair, and became involved directly in Peyrony’s attempt to preserve Palaeolithic sites through the monument historique classification process (White 2002). Alonzo Pond, the Logan Museum representative, purchased a large collection of artifacts from Jean Leyssales, Hauser’s local collaborator and excavator, unaware that this material was contested by Peyrony who identified part of the collection as belonging to Hauser and therefore the property of the Musée National de Préhistoire. 204 The Logan Museum of Anthropology became interested in the excavating at Abri Cellier after excavating at Rocher de la Peine, again arousing Peyrony’s suspicions because he suspected that the landlord of the site was a member of “a secret ring of men, who are getting all the valuable material they can…and sending it for a big price to Germany” (Letter from Collie to Pond, quoted in White 2002). Pond identified Abri Cellier as a potential field school locale based on the collection dug out by Fernand Merlan. The Museum negotiated a lease of the site from the owner, Mme. Cellier, which included the rights of ownership of all material excavated. The museum also purchased the Merlan Collection. The path to excavation did not go smoothly. Denis Peyrony objected strongly to the proposed excavation. In an effort to prevent excavation he had Abri Cellier (also known at that time as La Ruth) classified as a National Monument, and delayed issuance of the excavation permit. The matter was finally resolved in favor of Beloit College following the intervention of the President of Beloit College, US President Calvin Coolidge, the American Ambassador to France and the French Minister for Cultural Affairs. Peyrony’s reasons for withholding the permit seem to be related in part to the presence of Fernand Merlan, who had worked for Otto Hauser. Control of the material and the excavation were also strong concerns for Peyrony, who dominated archaeology in the Dordogne in the inter-war period. It has also been suggested that his active collaboration with the Peabody Museum, who were also inaugurating a long term field school in the Dordogne, may have influenced him to discourage other such field programs. Finally Peyrony and other archaeologists were somewhat concerned at the impact of American purchasing power on the access to important sites and collections. Also, they were aware that they were also involved in the commerce of artifacts with Americans, albeit discretely (White 2002). While Beloit College ran into difficulties at Abri Cellier, McCurdy was able to excavate at the Abri des Merveilles at Sergeac from 1924 to 1931 and export the majority 205 of artifacts without difficulty. Peyrony later prevented McCurdy from excavating at another site. Abbé Breuil worked closely with Henry Field, a curator of the Field Museum (and a nephew of Marshall Field), traveling with him in France and Spain and facilitating contacts with owners of major collections. And H-M Ami, a Canadian archaeologist and close friend of Peyrony excavated for several years at Combe Capelle Bas (White 2002). While Peyrony was clearly concerned about the loss of material to foreigners, White notes that a certain double standard was in play, probably the result of Hauser’s direct competition with Peyrony (plus his personality) and his association with German museums in the late nineteenth century, a time of considerable anti-German sentiment in France. The Americans had the funds to excavate and, while wishing to collect material, would cooperate with Peyrony and other French archaeologists. Alonzo Pond and George Collie appear to have inadvertently become tainted by association with individuals linked to Hauser, and this added to their difficulties in obtaining permission to excavate at Abri Cellier. The cessation of excavations at Abri Cellier after the 1927 season likely resulted from lack of funding following the Stock Market Crash of 1929; the continued opposition of Peyrony; and the availability of other supposedly early Upper Palaeolithic (now known to be Epipalaeolithic in date) material in North Africa (W. Green, pers. comm.). Excavations at Abri Cellier The history of excavation at the site is derived from Knecht (1991) and White and Knecht (1992). Peyrony excavated a sondage in 1909, on the talus of the rock shelter, midway between the cliff face and a quarry on the southern margin and recovered Mousterian lithics approximately 3 meters below the surface and Aurignacian lithics near the surface. Using data from later testing by Ami (in 1930) and Lucas (circa 1909), Peyrony suggested that the Mousterian level extended along a rocky terrace running for 120 meters between the two valleys (Peyrony 1946). 206 The 1927 Logan Museum excavations (directed by George Collie and Paul Nesbitt) had the goal of recovering material and particularly skeletal material associated with “Early Man”, primarily for educational purposes. No human remains were recovered, but the excavations identified two major Aurignacian cultural levels as shown in Nesbitt’s thesis (Nesbitt 1928) and “perhaps a more restricted Upper Perigordian unit in the area excavated by Beloit College” (White and Knecht 1992:54, emphasis authors’), printed in Collie’s volume (Collie 1928). The Aurignacian overlies bedrock and the Mousterian deposits found on the talus did not extend into this part of the shelter. The Beloit excavations lasted for two months, expanding the north-south trench excavated by Merlan. All bone and lithic artifacts were plotted according to level and distance from the abri wall, and plotted on a graph of the section. Artifacts and fauna are catalogued by artifact type or taxonomic element, and simply noted as coming from the Upper or Lower Level. Unworked blades, nuclei and debitage were not collected but placed in two piles, one for each deposit. Sediment was screened for the first week only, but abandoned for reasons of time and economy (Nesbitt 1928:32). Fauna were dried in the sun and treated with gomme arabique and shellac. This is the sum total of the information regarding the excavations at the site. Collie’s diary (on file at Beloit College) contains sparse notes of the day’s work, or occasional trips to dig at other sites. All other documentation related to the excavation has been lost. From photographs on file at Beloit College it is clear that the excavators did not screen any material, and it is clear from the surviving collection that there was a strong selective bias towards formal stone tools, identifiable fauna and worked osseous items. The Aurignacian levels are designated in the collection as “Lower Aurignacian” and “Upper Aurignacian”. The “Lower Aurignacian” (herein referred to as the Lower Level) deposit occurred above bedrock in the rear of the shelter and extended at least 40 feet south of the rock shelter wall. The lower layer of the Aurignacian deposit ranged in 207 thickness from 4 inches (10cm) to 2 feet (61cm) and was a palimpsest of a series of occupations, including two hearths towards the front of the cave. Collie noted that there were 7 or 8 thin strata within the level (Collie 1928:66). All stratigraphic columns show hiatus in occupation overtopped by an “Upper Aurignacian” layer (Upper Level) that ended abruptly 30 feet south of the rear shelter wall. The upper Aurignacian layer was thinner than the earlier occupation - between 4 and 13 inches (10-33cm) thick, thickest under the abri and attenuating towards the talus and contained a single hearth. Formal tools are typologically consistent, assigning the Lower Level to Aurignacian I, and the Upper Level to Aurignacian II (Woods 2011:38). An unpublished section at the Logan Museum, edited by Collie, shows a third stratum, a Gravettian deposit very limited in range, extending no more than 10 feet from the shelter wall above the Upper Aurignacian level. Nesbit (1928) specifically states that there were no Upper Perigordian or Gravettian tools in the collection. However, the modern collection does contain a single Gravette point (from the Merlan Collection, which was not part of Nesbitt’s study) and nine tools labeled as Audi blades (White 1985; Woods 2011:153), all accessioned within the Upper Aurignacian level (either as Upper Aurignacian or part of the Merlan Collection). Thus it is possible that an ephemeral Gravettian/Upper Perigordian occupation was present, but excavated by Collie prior to Nesbitt’s arrival. Given the very low proportion of Gravettian/Upper Perigoridan tools (10 of a total of approximately 11,123 flint or chert tools or 0.09% of the total lithic assemblage), and the absence of any diagnostic worked bone material (c.f. White and Knecht 1992), it seems highly unlikely that any faunal material from the Gravettian/Upper Perigordian occupation was collected. Initially, the two Aurignacian layers were excavated as single levels; however the discovery of bone points (and a visit by Peyrony) resulted in a change in excavation procedures. The Lower Level was subsequently excavated in three sub-levels and the Upper Level was divided into two sub-levels. Bone and flint artifacts, were numbered 208 according to level and distance from the abri wall (Nesbitt 1928:31). Unfortunately this numbering system was not retained during cataloguing at the Logan Museum. These spatial data were utilized by Nesbitt in his thesis, but the notes are not present at the Logan today. They were probably retained by Nesbitt for his thesis, and remained with him when he left the Logan Museum. In contrast to the formal stone and bone tools, unworked bone was not excavated by general level, because “the scarcity of faunal remains and the fact that many bones overlapped in the layers of each deposit” (Nesbitt 1928:32 [my emphasis]). The faunal material therefore represents a sequence of occupations within the abri. Artifact recovery was limited. All lithics recovered were formal tools, and the criteria for bone collection is uncertain and will be addressed further below, in the analysis chapter. As a result, lithic manufacturing debris, microfauna, small beads and smaller bone fragments are absent. We should not assume that the collection is totally biased. Recent re-excavation of spoil heaps at Le Castanet and comparison with material in the National Museum at Le Eyzies has found that while much non-diagnostic material was discarded, the tools in the museum collections are proportional to diagnostic material still in the backdirt (R White, pers. comm.)(Tartar 2009). Material recovered from the excavations was returned to Beloit College, with the exception of “une bonne serie des pieces de chaque niveau” and parietal engravings, which were deposited at the Musée Préhistorique des Eyzies, in accordance with the excavation permit. The parietal engravings include the famous vulvas found on collapsed roof or wall debris. In a 2009 visit to the museum collection, no record of any fauna or worked bone from the Beloit Excavations was found at the museum. All material curated was from Peyrony’s earlier test excavations on the talus. After arrival at Beloit, some material was sold or exchanged with other museums, but majority of the collection remains at the Logan Museum. No further archaeological excavations have occurred at 209 the site. The back dirt remains, and it would be fruitful to undertake excavations of this material to obtain a fuller picture of the material culture present at the site. Previous research on the Cellier collection The collection under study results from excavations undertaken in 1927 by the Logan Museum, Beloit College, Wisconsin, now housed at Beloit College. Apart from a brief description in a Master’s Thesis (Nesbitt 1928) and studies of the worked bone and antler tools (White and Knecht 1992), no formal studies have been undertaken on the faunal material. Nesbitt provided a cursory faunal analysis based on the teeth recovered (Nesbitt 1928:62-63) but his primary focus was on the lithic tools from the site. Paul Nesbitt analyzed the formal stone tools for his Master’s Thesis at the Department of Anthropology, University of Chicago. Nesbitt noted that the most common tool class was the grattoir, followed by burins, side scrapers, and spokeshaves. The presence of grattoirs, burins and perçoirs (all “tools to make tools”) suggests that focus of production at the site was processing of non-lithic material - hides, bone or wood. Nesbitt’s primary research interest was the spatial patterning of the formal lithic tools, an early attempt to examine the organization of space within a rock shelter. This spatial analysis focused solely on the lithic material from the site. There was no discussion of the distribution of faunal elements across either level. Nesbitt also briefly examined the worked bone at Abri Cellier, noting that the majority of worked bone came from the Lower Level, with sparser material in the Upper Level (perhaps not surprising, given the difference in thickness of the deposits). Worked bone from the site was studied in greater detail by White and Knecht (1992) and Knecht (1992). Two of the teeth in the lower layer are circumincised for suspension, a method of working associated with the Châtelperronian (White and Knecht 1992:62). Nesbitt noted the presence of “Châtelperron blades” in the lower Aurignacian layer (Nesbitt 1928). 210 However, recent analysis by Alex Wood of the blades from Abri Cellier did not identify any typical Châtelperronian lithic material (Woods 2011). The lithic material from Cellier was used by Alexander Woods to examine raw material selection and knapping practices in the Aurignacian. Raw material at the site is generally of high quality. Woods identified locally available black Senonian flint, which was used as a support for most tools; plus blades of Bergerac flint (40 km southwest); chalcedony (23 km southwest); quartz (probably local); jasper from a variety of sources 30-35 km from the site; and a few pieces on Turonian flint, 50 km to the southwest. Exotic raw material was more common in the Lower Level (15.73%) than the Upper (8.46%), consistent with patterns observed by other analysts. Tool frequencies also showed a pattern similar to other Earlier Aurignacian sites, with a greater proportion of marginally retouched blades and endscrapers in the Lower Level, and a higher proportion of burins in the Upper Level (Woods 2011:67). The lithic analysis indicates a shift in behaviors at the site between the two occupations, including a reduction in the use of tools made on non-local raw materials. It may infer a shift in mobility patterns, with what appears to be a reduction in mobility. This may be related to shifts in subsistence related to changes in the environment. We will return to this issue in the analysis of the fauna. Fauna: the orphan child of the Palaeolithic The lack of interest in the faunal material from Abri Cellier over the past 85 years is symptomatic of a larger issue (albeit undiscussed) in Palaeolithic archaeology. Fauna just wasn’t interesting in terms of research questions asked of the data until the advent of Processualism in the 1960s. In the nineteenth and first half of the twentieth century, French Palaeolithic research was strongly influenced by a geological approach that focused on the construction of culture sequences. Caves and rock shelters were excavated to recover stratigraphic sequences of stone tools (fossiles directeurs) that replaced type fossils as chronological markers. The fauna was examined by archaeologists in a cursory 211 manner to provide information on the local or regional climate. It was assumed that the animal remains in the cave were largely the products of hunting by direct ancestors of modern humans, as hunting had long been thought to be a prime mover in human evolution, if not a basic human trait. It should be noted that some archaeologists did get a little over-excited about the presence of cave bear bones in deep caves during this period, and assumed the presence of a Cave Bear Cult, rather than the more prosaic fact that some cave bears hibernate a little too permanently for their own good. New questions began to be asked of the data as cultural sequences became established and archaeologists became more interested in how hominins behaved in the Pleistocene. Questions regarding hunting (human) and scavenging (non-human) behaviors became topics of interest, propelled in part by new fossil discoveries outside Europe. The discovery of australopithecine fossils in Africa and the realization that these early hominins weighed “ninety pounds dripping wet” (L.R. Binford, lecture on australopithecines, November 1983, University of Southampton). This led to serious questions about the ability of early hominins to hunt big game, even if stone tools were found in association with carcasses of megafauna. Debate ensued as to hunting abilities of early and later presumed ancestors, and the point in time and space when scavenging was replaced by hunting. The identification with hunting as a modern behavior was not questioned. As a result, faunal analysis became central to questions about human behavior rather than peripheral to discussions of culture change. Even with this new focus, it was not uncommon for non-diagnostic material (i.e. material not recognized by the excavators) to be discarded at the site (Francine David, pers. comm.). As archaeologists became aware of the finite nature of archaeological resources, the collection and retention of all material became standard. 212 Conclusion Despite the lack of screening or excavation by sub levels, the excavators at Abri Cellier were unusual for their time period. The discard of non-diagnostic faunal and lithic material has reduced the amount of information that we can derive from the assemblage regarding site organization and subsistence practices. It appears that non-worked bone was collected, possibly for later use in the museum or as part of a teaching collection. The intended use of the Cellier material clearly had importance for the selection practices of some of the material. As a result the amount of bone in the collection is unusually large for an excavation of that date. While no spatial data are available to reconstruct activity areas or other details of site organization, it will be possible to examine prey choice and compare the supports used for bone tools with the broader pattern of prey exploitation. The presence of two layers, with differences in mobility patterns between the Lower and Upper Levels, also provides an opportunity to examine if there were any shifts in subsistence behavior between the Aurignacian I and Aurignacian II and to discover if these shifts are related to a change in environmental conditions, or to differences in foraging or collecting behavior. These in turn can be used to evaluate any differences in subsistence behavior and selection of bone tool supports with the material from the Grotte du Renne. In the next chapter, the results of the faunal analysis are presented. This will be followed by a discussion of the use of bone as a raw material and the selection practices utilized in the Early Upper Palaeolithic. 213 CHAPTER 10: FAUNAL ANALYSIS OF ABRI CELLIER Introduction The faunal assemblage from Abri Cellier contains a diversity of taxa and the proportions of taxa vary between levels. Faunal material from Abri Cellier is the product of human transportation to the site. Taphonomic analysis indicates that although carnivores are present, there is relatively little carnivore damage to the bones and, therefore, carnivores were minor agents in bone accumulation or destruction. As discussed in the site history, and as will become apparent in this chapter, the faunal assemblage has been subjected to strong selection by the excavators for identifiable or large bone fragments. No small bone fragments or esquilles are present in the collection. However, given the strong bias towards identifiable material, I believe that the faunal remains form a representative sample of the animals exploited for subsistence purposes. The total number of items catalogued in the Cellier collection at Beloit College as fauna is 1205. This included 4 lithics and two mollusk shells (one fossilized, the other a member of the genus Sipho) which were not included in this analysis. Therefore the total number of elements analyzed was 1199. Of this number, 832 were identifiable to genus and/or species and 369 were classified as unidentified mammal bone. A further 60 faunal elements from Abri Cellier are held at the Musée National de Préhistoire, Les Eyzies, These are largely teeth, plus four unidentified bone fragments. Peyrony also collected a sample of bone tools: 15 of antler, 1 of bone (a probable metapodial), a drilled incisor plus a mollusk shell (18 items in all). All the unworked fauna is from the Upper Aurignacian or Aurignacian II level. The worked items are labeled I (n=7) or II (n=10) which suggests that Peyrony did test both levels of the site. The selection criteria for this material are unclear, and the representative nature of the sample is extremely suspect. Therefore the Eyzies data will be mentioned in the general 214 discussion of taxa present in terms of MNI and NISP, but will form no part of the overall analysis of the faunal assemblage. In total, 1279 bone or antler elements remain from the excavations at Abri Cellier, of which the majority (n=1201) are held in the Logan Museum collection at Beloit. This analysis focusses on the unworked bone. In this chapter I will describe the taxa present. I will then discuss the Number of Identified Specimens (NISP), and the Minimum Number of Individuals (NMI) present by level. This will precede a description of the elements present within each taxon, which will consider the subsistence and transportation strategies of the Aurignacian occupants of the site, reflected in the Minimum Number of Elements (MNE) and Minumum Animal Units (MAU). Any differences in elements present by level will be noted. Butchery patterns will be discussed by taxon and by element. Taxa present at Abri Cellier The following taxa were identified in the collection: reindeer (Rangifer tarandus) horse (Equus caballus) red deer (Cervus elaphus) bovidae (Bos sp.), saiga (Saiga tataricus), ibex, (Capra ibex), wild boar (Sus scrofa), bear (Ursus sp), wolf (Canis lupus), fox, hare, mammoth (Mammuthus sp) waterfowl (Anseriformes) and fish (Pisces). Tables 10.1and 10.2 show totals for NISP by taxon for all collections and the material held at the Logan Museum, Beloit College. The fauna is dominated by herbivores reindeer, horse, red deer, and bison or auroch (Figures 10.1 and 10.2). Carnivores, small mammals and megafauna are present in very low numbers, and many of these specimens have been altered for use as personal ornament or as tools. Only reindeer, red deer, horse and ibex are present in the Eyzies fauna, which strongly suggests a bias in collection. This is particularly true if the fauna is derived from the Aurignacian II, because both red deer and bovids should be present in relatively high proportions, as will be seen below. 215 Upper Level Lower Level No Level Total Bird 0 5 0 5 Bear 0 4 0 4 Bovid 79 16 1 96 Cervidae 87 159 11 257 Fish 0 1 0 1 Fox 0 5 0 5 Hare 0 1 0 1 Horse 80 77 0 157 Ibex 1 0 0 1 Mammoth 5 0 0 5 Red deer 45 8 0 53 Reindeer 112 153 14 279 Rodent 0 2 0 2 Saiga 9 20 0 29 Wild boar 1 1 0 2 Wolf 1 4 0 5 U mammal 226 105 19 350 U large 16 0 1 17 U med 0 3 0 3 U small 0 2 0 61 Worked 1 2 0 3 Total Unid 243 112 20 376 Total NISP 593 563 46 1279 . Table 10.1: Total Number of Identified Specimens by count from all collections of fauna from Abri Cellier. 216 Upper Lower Unknown NISP %NISP NISP %NISP NISP %NISP Aves 0.00 0.00 5.00 1.11 0.00 0.00 Bear 0.00 0.00 4.00 0.89 0.00 0.00 Bovid 78.00 21.97 16.00 3.55 1.00 3.85 Cervidae 77.00 21.69 153.00 33.92 11.00 42.31 Fish 0.00 0.00 1.00 0.22 0.00 0.00 Fox 0.00 0.00 5.00 1.11 0.00 0.00 Hare 0.00 0.00 1.00 0.22 0.00 0.00 Horse 68.00 19.15 77.00 17.07 0.00 0.00 Mammoth 4.00 1.13 0.00 0.00 0.00 0.00 Mollusca 0.00 0.00 1.00 0.22 0.00 0.00 Red deer 44.00 12.39 8.00 1.77 0.00 0.00 Reindeer 73.00 20.56 153.00 33.92 14.00 53.85 Rodent 0.00 0.00 2.00 0.44 0.00 0.00 Saiga 9.00 2.54 20.00 4.43 0.00 0.00 Wild boar 1.00 0.28 1.00 0.22 0.00 0.00 Wolf 1.00 0.28 4.00 0.89 0.00 0.00 Total 355.00 100.00 451.00 100.00 26.00 100.00 Table 10. 2: Total Number of Identified Specimens for fauna from Abri Cellier curated at the Logan Museum. 217 Figure 10.1: Graph showing count and percentage of NISP by taxon for the Upper Level of Abri Cellier, held at the Logan Museum. Figure 10.2: Graph showing count and percentage of NISP by taxon for the Lower Level of Abri Cellier, held at the Logan Museum. 218 Figure 10.3: Graph showing proportions of NISP by taxon for the Abri Cellier faunal assemblage held at the Logan Museum. NISP and MNI Identifications were made using comparative material housed at the University of Iowa, Department of Anthropology, Department of Geology and Natural History Museum and published sources (Barone 1976; Gilbert 1990; Gilbert, et al. 1996; Olsen 1964, 1968; Pales and Garcia 1981a and b; Pales and Lambert 1971a and b). The proportion of species present varies by level in the Abri Cellier assemblage (Figure 10.3). As the majority of antler recovered is in the form of tools or tool fragments, transported to the site as part of a toolkit, I have omitted antler from specimen counts. In the Upper Level reindeer, horse and bovids form 62% of the NISP, followed by red deer (12%). A further 22% of the NISP is classified as Cervidae. All other taxa form a total of 4% of the Upper Level assemblage NISP. In the Lower Level reindeer forms approximately 33% of the NISP, and horse 17%, with cervidae forming a further 33% of the total NISP. The amount of bovid material drops to 3.5% and red deer to 1.75%, while saiga increases to 4.4%. 219 Logan Museum Upper Lower Les Eyzies Aurig II Total MNI Aves 0 1 0 1 Bear 0 1 0 1 Bovid juve 2 0 0 2 Bovid adult 5 2 1 8 Cervidae 1 1 0 2 Fish 0 1 0 1 Fox 0 2 0 2 Hare 0 1 0 1 Horse juve 1 1 1 3 Horse adult 3 2 1 6 Ibex 0 0 1 1 Mammoth 1 0 0 1 Mollusca 0 1 0 1 Red deer 6 1 1 8 Red deer juve 1 0 0 1 Reindeer juve 1 2 1 4 Reindeer adult 3 6 4 13 0 1 0 1 Rodent Table 10.3: Total Minimum Number of Individuals from all collections from Abri Cellier. 220 Logan Museum Les Eyzies Upper Aurig II Lower Total MNI Saiga 1 2 0 3 Wild boar 1 1 0 2 Wolf 1 1 0 2 Total 27 27 10 64 Table 10.3: concluded. Figure 10.4: Graph showing count and percentage of MNI by taxon for the Upper Level of Abri Cellier, held at the Logan Museum. 221 Upper Lower MNI %MNI MNI %MNI Aves 0 0.00 1 3.70 Bear 0 0.00 1 3.70 Bovid juve 2 7.41 0 0.00 Bovid adult 5 18.52 2 7.41 Cervidae 1 3.70 1 3.70 Fish 0 0.00 1 3.70 Fox 0 0.00 2 7.41 Hare 0 0.00 1 3.70 Horse juve 1 3.70 1 3.70 Horse adult 3 11.11 2 7.41 Ibex 0 0.00 0 0.00 Mammoth 1 3.70 0 0.00 Mollusca 0 0.00 1 3.70 Red deer 6 22.22 1 3.70 Red deer juve 1 3.70 0 0.00 Reindeer juve 1 3.70 2 7.41 Reindeer adult 3 11.11 6 22.22 Rodent 0 0.00 1 3.70 Saiga 1 3.70 2 7.41 Wild boar 1 3.70 1 3.70 Wolf 1 3.70 1 3.70 Total 27 100 27 100 Table 10.4: Total MNI for fauna from Abri Cellier held at the Logan Museum. 222 Figure 10.5: Graph showing count and percentage of MNI by taxon for the Lower Level of Abri Cellier, held at the Logan Museum. Figure 10.6: Graph showing proportions of NMI by taxon for the Abri Cellier faunal assemblage held at the Logan Museum. 223 The MNI show a distinct contrast in genera present by level within the shelter (Tables 10.3 and 10.4; Figures 10.4 through 10.6). The Upper Level is dominated by bovids (26%) and red deer (26%), followed by horse (14%) and reindeer (14%). The Lower Level is dominated by reindeer (31%), followed by horse (11%), then bovids, saiga and fox (all 7.7%). When the intra-taxon variation by level is examined, the pattern in enhanced: the proportions of bovids, red deer and horse increase from the Lower to the Upper levels, while the proportions of reindeer and saiga decrease (Figure 10.7). This suggests a cooler and more open environment in the Aurignacian I (theLower Level) with an increase in moisture, tree cover and mean annual temperature by the Aurignacian II (the Upper Level). I will return to shift in prey species at the end of the chapter. Figure 10.7: Graph showing intra-taxon variation between levels at Abri Cellier, where MNI is greater than 1, excluding cervidae. 224 The pattern for NISP shows a similar shift, which is probably a factor of excavator selection of identifiable material. As a result NISP and NMI are more closely related that would be expected for a complete faunal assemblage, where all material is collected. Of course, Grayson (1984) argues that NISP and MNI are correlated, but other authors disagree with this interpretation of the data. Total NISP for the two levels at Abri Cellier is shown in Table 10.6 for herbivores and Table 10.7 for all other taxa, below. In the Upper Level red deer, horse and bovids form 48.6% of the total NISP, but only 22% of the Lower Level. In contrast the proportion of reindeer rises from 26.7% in the Upper Level to 33.5% in the Lower Level. The amount of unidentified Cervidae (largely antler) also increases slightly between levels. The majority of the antler is worked and may have been transported to the site as finished tools. More importantly, there are two species of large antler bearing cervids in the assemblage: red deer and reindeer. While reindeer antler is preferred for projectile points, red deer antler is also used for a wide variety of tools (Guthrie 1983; MacGregor 1985). It was not possible to ascertain the source of the majority of the worked antler. If the unidentified cervid and the small numbers of ominivores or carnivores are removed from the calculations, reindeer represents slightly over half the total NISP of herbivores in the Lower Level (Table 10.5). Horse is at 28.1% with small amounts of saiga (7.3%), bovids (5.84%) and red deer (2.92%). In contrast, bovids (28.2%), reindeer (26.5%) and horse (24.64 %) are nearly equally dominant in the NISP for the Upper Level, with saiga reduced to 3.26%, while red deer rises to 15.94%. The proportions above refer to the percentage of total identified specimens. If the intra-taxon variation is examined, generally trends follow the pattern seen in the MNI, with the exception of horse, which increases slightly in the Lower Level for NISP. This may relate to taphonomic factors, particularly excavator selection. There are a large 225 Upper Lower %NISP Upper % NISP Lower Bovid 78 16 28.26 5.84 Horse 68 77 24.64 28.10 Mammoth 4 0 1.45 0 Red deer 44 8 15.94 2.92 Reindeer 73 153 26.45 55.84 Saiga 9 20 3.26 7.30 Total 276 274 100 100 Table 10.5: Herbivores at Abri Cellier, NISP counts and percentages, excluding unidentified cervidae and hare. number of horse teeth and teeth fragments in the Lower Level fauna, relative to limb bones. With the exception of the second premolar and third molar, horse teeth are very difficult to identify precisely. As a result the horse MNI for the Lower Level may be under-estimated, and the NISP suggests that horse may be a little more common in the environment during the earlier phase of occupation. The higher proportion of cold and open steppe fauna in the Lower Level is not negated however, as there is still a marked reduction in the number of woodland browsers or more temperate species, particularly red deer and bovids. As can be seen from the proportions of MNI and NISP, there appears to be a shift from a fauna associated with cold, open grasslands in the Lower Level, to a fauna that contains animals that represent a more temperate climate, and, possibly, an increase in woodland vegetation in the Upper Level. Before this shift can be confirmed, taphonomic analysis is necessary to determine which factors have operated on the formation and destruction of the material from Abri Cellier. Reindeer Horse Red deer Bovids Mammoth Cervidae Saiga Hare U L U L U L U L U L U L U L U L Antler/horn 8 16 0 0 0 0 1 2 0 0 71 149 0 0 0 0 Cranuim 0 1 0 1 0 0 0 0 0 0 0 0 0 0 0 0 Mandible 2 16 0 2 2 0 1 0 0 0 1 0 0 0 0 0 Maxilla 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Atlas 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Axis 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Vertebrae 0 0 0 0 2 0 0 0 0 0 0 0 0 0 0 0 Rib 0 0 0 0 1 0 2 0 0 0 0 0 0 0 0 0 Scapula 1 4 0 0 0 0 0 1 0 0 0 0 0 2 0 0 Humerus 4 8 5 0 4 0 4 1 0 0 0 1 1 2 0 1 Radius 1 4 5 0 1 0 8 0 0 0 0 0 1 0 0 Table 10 6: Summary table of NISP per element by level and taxon for herbivores from the Abri Cellier fauna held at Beloit College. (U=Upper Level; L= Lower Level). 226 Reindeer Horse Red deer U L U L U L U L U L U L U L U L Ulna 1 2 1 0 0 0 0 1 0 0 0 0 0 0 0 0 Radius/ulna 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Carpals 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Metacarpals 2 10 1 0 5 2 0 0 0 0 0 0 0 0 0 0 Inominate 2 0 2 1 1 1 0 1 0 0 0 0 1 0 0 0 Femur 2 4 2 0 1 1 5 1 0 0 0 0 0 0 0 0 Tibia 4 4 18 4 15 0 7 0 0 0 1 0 0 1 0 0 Astragalus 2 2 0 1 0 0 0 0 0 0 0 0 0 0 0 0 Calcaneus 2 2 1 0 0 0 1 1 0 0 1 0 0 0 0 0 Tarsals 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Metatarsals 3 10 0 4 2 0 3 0 0 0 0 1 3 2 0 0 Sesamoids 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 First Phalange 3 6 0 1 0 0 1 0 0 0 0 0 1 0 0 0 Mammoth Cervidae Saiga Hare 227 Table 10.6: continued. Bovids Reindeer Horse Red deer U L U L U L U L U L U L U L U L Second Phalange 3 2 0 4 0 0 1 0 0 0 0 0 0 0 0 0 Third Phalange 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 Metapodials 0 3 0 0 1 1 0 0 0 0 1 0 0 0 0 0 Residuals 0 8 0 5 0 0 0 0 0 0 0 0 0 0 0 0 total id bones 41 104 35 23 35 5 36 8 0 0 75 151 7 7 0 1 Incisors 6 5 6 24 0 0 3 0 2 0 0 1 0 5 0 0 Canines 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 Max teeth 6 7 8 19 3 0 8 1 0 0 0 0 0 0 0 0 Mand teeth 16 36 15 11 5 3 24 6 0 0 0 1 0 8 0 0 Unid teeth 3 0 3 0 0 0 7 1 0 0 2 0 0 0 0 0 Unid long bone 0 1 1 0 0 0 0 0 1 0 0 0 0 0 0 0 total teeth 31 48 32 54 9 3 42 8 2 0 2 2 0 13 0 0 TOTAL 72 152 68 77 44 8 78 16 3 0 77 153 7 20 0 1 Mammoth Cervidae Saiga Hare 228 Table10.6: concluded. Bovids Wolf Cave bear Aves Fox Pisces Rodent Sus U L U L U L U L U L U L U L Antler/horn 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Cranuim 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Mandible 1 0 0 0 0 0 0 0 0 0 0 0 0 0 Maxilla 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Atlas 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Axis 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Vertebrae 0 0 0 0 0 0 0 0 0 1 0 0 0 0 Rib 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Scapula 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Humerus 0 0 0 0 0 1 0 0 0 0 0 0 0 0 Radius 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Table 10.7: Summary table of NISP per element for non-herbivores from Abri Cellier fauna held at Beloit College. (U=Upper Level; L=Lower Level). 229 Wolf Cave bear Aves Fox Pisces Rodent Sus U L U L U L U L U L U L U L Ulna 0 1 0 0 0 0 0 0 0 0 0 0 0 0 Radius/ulna 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Carpals 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Metacarpals 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Inominate 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Femur 0 0 0 0 0 1 0 0 0 0 0 2 0 0 Tibia 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Astragalus 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Calcaneus 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Tarsals 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Metatarsals 0 0 0 0 0 0 0 2 0 0 0 0 0 0 Sesamoids 0 0 0 0 0 0 0 0 0 0 0 0 0 0 First Phalange 0 0 0 0 0 0 0 1 0 0 0 0 0 0 230 Table 10.7: continued. Wolf Cave bear Aves Fox Pisces Rodent Sus U L U L U L U L U L U L U L Second Phalange 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Third Phalange 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Metapodials 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Residuals 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Teeth 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Incisors 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Canines 0 3 0 4 0 0 0 2 0 0 0 0 0 1 Max teeth 0 0 0 0 0 0 0 0 0 0 0 0 1 0 Mand teeth 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Unid teeth 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Unid long bone 0 0 0 0 0 3 0 0 0 0 0 0 0 0 total teeth 0 3 0 4 0 0 0 2 0 0 0 0 1 1 TOTAL 1 4 0 4 0 5 0 5 0 1 0 2 1 1 231 Table 10.7: concluded. 232 Unlike the material from Level Xc at the Grotte du Renne, the majority of the fauna from Abri Cellier was identifiable. For the Upper Level, 60% of the fauna was identifiable to genera or species, and 80% of the Lower Level was identifiable. The difference in the proportions of identifiable material is largely the result of the incorporation of the previously excavated Merlan material in to the Upper Level. This contained more diaphysis fragments than the material catalogued from the Beloit excavation. It is clear that easily identifiable large bone fragments and teeth were preferred to small bone fragments of unidentifiable mammal bone. Unidentifiable bone Small Cancellous Flat Long Epiphysis Tooth Nutritive Bone Total bone bone bone foramen scrap 0 0 2 0 0 0 0 2 0 2 1 0 0 0 0 3 0 0 16 1 0 0 0 17 51 35 231 0 27 0 1 345 51 37 251 1 27 0 1 367 mammal Medium mammal Large mammal Unknown Mammal Total Table 10.8: Table showing the total amount of unidentified bone by category from Abri Cellier 233 Bone fragments were classed by size and thickness and categorized as flat, cancellous or long bone fragments. The majority of the bone fragments were assigned to the unidentified mammal taxon either because they were of a size and form that made it difficult to easily assign a size class, or because they had been altered by working which also meant that assigning a particular category was difficult. Unidentified bone formed 30.5% of the total assemblage. The unidentified material occurs in both levels. Approximately two-thirds (n=237) was recovered from the Upper Level or purchased from Fernand Merlan. The remaining 130 fragments were collected in the Lower Level (Table 10.8). The proportion of unidentified bone fragments larger than 2.5cm in length is far lower than the percentage of the same class of material from Level Xc, where unidentified bone (n=1544) forms 61.1% of the assemblage. The low proportion of unidentified bone indicates a strong bias towards identifiable specimens by the excavators at Abri Cellier. Figure 10.8: Graph showing the proportion of dry, fresh, recent and undetermined breaks on long bone fragments in the Abri Cellier assemblage (all levels). 234 The proportions of unidentified bone in the collection contrasts strongly with Nesbitt’s description of the proportions of bone fragments and presence of bone in general. According to Nesbitt, the Upper Level was almost sterile in terms of animal bone, while the majority of the faunal assemblage in the Lower Level was long bone fragments that had been broken for marrow. This suggests that he did not include any of the Merlan material in his analysis. A slight majority (51%) of the bone fragments has at least one fresh break with 20% of the long bone fragments showing dry breaks (Figure 10.8). The remainder weree undetermined, or were recent breaks that has occurred post-excavation. Excavator selection is a major taphonomic factor in the composition of the assemblage, but as this appears to be biased against unidentifiable bone. An analysis of subsistence behavior is therefore possible, with the caveat that little can be said with certainty regarding marrow or grease processing giving the absence of the majority the unidentified bone fragments. Taphonomy In contrast to the Grotte du Renne, excavator bias played a major role in the composition of the assemblage from the two levels of Abri Cellier. Excavation techniques were basic. Screening of material only occurred in the first week of excavation, and was abandoned as it was found unproductive (Nesbitt 1928:32). Excavator bias is evident in the proportions of unidentified bone in the current assemblage. As noted above, the Lower Level was reported to contain long bone fragments in relatively high frequency in relation to identifiable bones and teeth, while remains were far sparser in the Upper Level, but this is not apparent in the assemblage housed at Beloit. Although not explicitly stated, Nesbitt’s thesis implies that identifiable bone alone was removed for further study: “faunal remains susceptible to identification were composed chiefly of teeth, reindeer horn, ribs, log bones with distal or proximal ends present, vertebraes (sic) and the small bones of the feet” (Nesbitt 1928:62). 235 Nesbitt notes that reindeer remains were the most abundant, bison was plentiful, auroch rare, horse plentiful but chiefly teeth, ibex rare (mostly teeth), cave bear rare and fox frequent but less common than bison. He also records a single lion tooth (Nesbitt 1928:62). General condition of the assemblage The material from Abri Cellier has been used as a teaching collection and for museum and other educational purposes for approximately 80 years. The collection is now housed in a climate controlled environment in conditions that meet modern curatorial standards. The majority of the collection is relatively lightly weathered, with minor cracking or light longitudinal cracking (Figure 10.9). It appears that much of the drying occurred in the post-excavation period. A number of bones have recent spalls on the surface, which has removed cortical material. Figure 10.9: Graph showing the proportion of weathering for the Upper and Lower Level of Abri Cellier. 236 A total of 272 items in the Upper Level and 73 in the Lower Level showed evidence of weathering. Weathering stages are taken from Behrensmeyer (1978). The collection does not appear to have been severely impacted by post-depositional weathering. The majority of the specimens show Level 2 weathering. There is evidence that some of the assemblage was damaged by root etching. A total of 133 elements in the Upper Level and 19 elements in the Lower Level showed evidence of root etching or vermiculation (Figure 10.10). These items were deposited have been in locations where soil formed at a later date, and where plant growth was such that it resulted in the acid etching by roots on the material. This damage reflects periods when the cave environment was relatively stable allowing soil formation to occur. As no provenience data is available, no further inferences can be made, although it seems likely that the etched material would be from near the cave entrance or talus, where conditions would be more favorable to plant growth and soil formation. Figure 10.10: Graph showing the proportions of vermiculated elements in the Upper and Lower Levels. 237 The long period of storage resulted in some elements exhibiting drawer wear, with rounding and smoothing or removal of the external surface through contact with other bones or surfaces of storage containers. A total of 103 elements from the Upper Level and 29 from the Lower Level had damage consistent with drawer wear (Figure 10.11). The majority of the items retained over 50% of their surface. Figure 10.11: Graph showing proportions of drawer wear for the Upper and Lower Levels. Density and survivorship Density values were calculated for reindeer, horse, bovids and red deer in the Upper Level and for reindeer and horse in the Lower Level. Density calculations could not be undertaken for bovids and red deer in the Lower Level because these taxa lacked landmarks that corresponded to locations utilized by Lyman or Lam. Any study of density must be undertaken with the understanding that this collection lacks major portions of elements, particularly midshaft fragments that can be identified by foramena. 238 Figure 10.12: Bivariate plot of MNE of reindeer against density value for the Upper Level. Figure 10.13: Bivariate plot of MNE of reindeer against density value for the Lower Level 239 Bivariate plots of reindeer in the Upper and Lower levels show no correlation between MNE and density values (Figures 10.12 and 10.13). Statistical analysis confirmed this lack of correlation. In the Upper Level the relationship was not significant with a Spearman’s rho correlation coefficient of .095, p=0.114. In the Lower Level the relationship was not significant with a Spearman’s rho correlation coefficient of .063, p=0.091. In contrast to reindeer, it appears that the horse assemblage in the Upper Level has been mediated by density factors, or (more likely) excavator selection (Figure 10.14). The relationship between density and element presence/absence was significant at the 0.05 level with a correlation coefficient of .039 for Spearman’s rho, p=0.051 Figure 10.14: Bivariate plot of MNE of horse against density value for the Upper Level. 240 Figure 10.15: Bivariate plot of MNE of horse against density value for the Lower Level. Figure 10.16: Bivariate plot of MNE of bovids against density value for the Upper Level. 241 However, in the Lower Level, the horse assemblage shows no statistical relationship between density value and survival of elements (Figure 10.15). Spearman’s rho has a correlation coefficient of .86, p=.074. It seems unusual that the less dense reindeer elements should not be impacted by density-mediated attrition while the denser horse bones in the Upper Level have a statistically significant relationship between density and survival. This may be the result of excavator selection processes, or possibly relate to differential preservation as the result of crushing of skeletal elements in parts of the cave by the collapse of the shelter vault. The bovid assemblage from the Upper Level shows no statistical correlation between density and survivorship (Figure 10.16). The Spearman’s rho correlation coefficient was .086, p=.153. Figure 10.17: Bivariate plot of MNE for red deer against density value for the Upper Level. 242 In contrast, the red deer assemblage in the Upper Level did demonstrate a statistically significant relationship between density and survivorship (Figure 10.17). The relationship was strong, significant at the .01 level with a Spearman’s rho correlation of .237, significance .008, p=.007. As with the horse assemblage, it seems odd that density would impact these larger and more robust elements to a greater extent than the reindeer assemblage. One possible explanation of the density patterns, other than accidental selection of denser elements by the excavators for later analysis, is to consider the impact of roof collapse on the assemblage. Nesbitt notes that some sections of the former cave roof were so large that dynamite was necessary to remove the blocks. Fauna in these areas of the excavation may be more seriously impacted (no pun intended) by post-depositional cave collapse than fauna from other areas where the roof survived or simply spalled off. Unfortunately the absence of any provenience data regarding the relationship of the roof fall and faunal remains means that no further investigation can be made into the possible relationship between the location of the horse and red deer bones in the Upper Level, roof collapse and density-mediated attrition. Carnivore gnawing There is little evidence of carnivore ravaging. Only 12 elements from the Upper Level and 11 from the Lower Level show evidence of damage by carnivores or rodents. In the Upper Level, the majority of the elements damaged were fragments of unidentified mammal long bones, plus two antlers, a bovid femur and a saiga metacarpal (Figure 10.18). 243 Figure 10.18: Graph showing the number of elements damaged by gnawing per taxon for the Upper Level. Damage patterns in the Upper Level and Lower Levels are consistent with gnawing by large and small carnivores (Figures 10.18 and 10.19). The level of carnivore damage is relatively low and does not indicate any major role for carnivores as agents of attrition. In the Lower Level, only eleven items showed evidence of damage by carnivores or rodents. The majority of bones damaged by gnawing were reindeer phalanges, which exhibited puncture marks consistent with the canines of a large carnivore (Figures 10.20 and 10.21). A reindeer calcaneum and a scapula fragment were also gnawed. A bovid scapula, a red deer metapodial, a saiga tibia and two unidentified mammal long bone fragments were also damaged. The pattern of damage suggests occasional use of the site as a den, particularly in the Lower Level, where the number of tooth punctures on phalanges increases. This could represent occasional scavenging of material abandoned by the human occupants of the cave (Figure 10.21). 244 Figure 10.19: Graph showing the number of elements damaged by gnawing per taxon for the Lower Level. Figure 10.20: Graph showing carnivore damage patterns on elements in the Upper Level. 245 Figure 10.21: Graph showing carnivore damage patterns on elements in the Lower Level Some elements have had the epiphyses completely removed, a classic pattern of carnivore gnawing to access fat and marrow. Wolf and fox may have used the cave on occasion – both taxa are represented by more elements than simply isolated canines – which may indicate occupation or scavenging when the site was abandoned by humans. Staining Six elements in the Upper Level and eight elements in the Lower Level were stained by the surrounding cave sediments. One element has concreted sediment attached and the rest are stained by dark or pinkish cave sediments. Many of the worked items are stained by later preservation treatments, but the majority of the unworked bone does not appear to have been stained or treated by the excavators. Burning The excavators recorded hearths in both levels of the site. However the amount of burnt bone is quite low, with two burnt elements collected from the Upper Level and 246 eight bone fragments from the Lower Level. In the Upper Level a horse tooth and a probable red deer tibia show signs of considerable heat alteration. In the Lower Level four fragments of cancellous bone, two vertebra fragments and two long bone fragments are completely carbonized. Summary of taphonomy The major taphonomic factor operating on the assemblage is collector bias. The absence of long bone shaft fragments and small bone fragments reduces the potential to make meaningful statements about the degree of bone damage related to processing for fat and marrow. It appears that the excavators collected items that were easily identifiable, particularly teeth. The discrepancy in density attrition between horse in the Upper and Lower Levels, and the apparent density-mediated attrition of red deer is hard to explain. If the items were better provienienced, spatial data would allow an examination of the location of these elements. It is known that the excavators had to remove large fragments of fallen cave roof from the excavation area. If the horse and red deer were concentrated in that area, it might explain the differential patterns of density attrition. Unfortunately this cannot be established, as no spatial data survives. Another, highly likely, explanation for the pattern is the selection behavior of the excavators, who may have taken denser items purely by chance. The absence of large amounts of carnivore damage suggests that carnivores were not major agents of bone accumulation or destruction at the site. The damage patterns suggest that they may have scavenged abandoned carcass parts. While acknowledging that excavator selection bias is operating on the assemblage, it still can be argued that humans were the main agents of accumulation and destruction of bones. We will now turn to the evidence for subsistence practices and tool blank selection present in the data set. 247 Herbivores Reindeer (Rangifer tarandus) A total of twelve individuals, 9 adults and three juveniles, were identified in the Logan collection, based on the presence of erupted third molars, nutritive foramena, deciduous teeth and unfused long bones. Three adults and one juvenile were present in the Upper Level and six adults and two juveniles were present in the Lower Level. A total of 224 specimens were identified as reindeer (79 teeth and 145 bone/bone fragments and antler). The collection is highly fragmented, with 93% of the bone represented in fragments. Unbroken bones present are all small, dense bones: 1 carpal, 4 astragali, 2 calcaneuii, 5 phalanges, and 1 residual phalange. All appendicular elements are present but axial elements such as vertebrae and ribs are under-represented. Given the presence of all elements, it seems likely that carcasses were transported whole to the site and butchered in or near the shelter. One set of incisors was found in anatomical connection (mistakenly or optimistically thought to be a necklace by the excavators), indicating that at least part of a lower jaw was deposited intact. The MNE, % survivorship and MAU are shown for the Upper and Lower Levels in Tables 10.9 and 10.10, below. Because the reindeer elements were collected in a nonrandom fashion, there is a high proportion of one side or another in the collection – for example all the adults in the Lower Level are identified by 6 left third molars, and there are no right left molars in the collection. As a result, the MNE and MAU calculations do not match the expected number of individuals as the data is skewed in a non-random fashion. 248 Cellier Upper Expected MNE % survival MAU %MAU MGUI Antler 6 1 16.7 0.5 16.67 1.02 Cranium 3 0 0.0 0 0.00 17.47 Mandible 18 2 11.1 3 100.00 30.26 Atlas 3 1 33.3 1 33.33 9.79 vt Cervical 15 0 0.0 0 0.00 35.71 vt Thoracic 39 0 0.0 0 0.00 45.53 vt Lumbar 18 0 0.0 0 0.00 32.05 Rib 78 0 0.0 0 0.00 49.77 Scapula 6 1 16.7 0.5 16.67 43.47 Pelvis 3 1 33.3 1 33.33 47.89 P Humerus 6 0 0.0 0 0.00 43.47 D Humerus 6 0 0.0 0 0.00 36.52 P Radius 6 0 0.0 0 0.00 26.64 D Radius 6 0 0.0 0 0.00 33.23 P Ulna 6 1 16.7 0.5 16.67 D Ulna 6 0 0.0 0 0.00 Carpals 36 0 0.0 0 0.00 15.53 P Metacarp 12 1 8.3 0.25 8.33 12.18 D Metacarp 12 1 8.3 0.25 8.33 10.5 Table 10.9: Summary of expected, observed and survival rate of reindeer elements in the Upper Level and their representation as Minimum Animal Units. 249 Cellier Upper Expected MNE % survival MAU %MAU P Femur 6 1 16.7 0.5 16.67 D Femur 6 0 0.0 0 0.00 P Tibia 6 3 50.0 1.5 50.00 64.73 D Tibia 6 0 0.0 0 0.00 47.09 Patella 6 0 0.0 0 0.00 P Fibula 6 0 0.0 0 0.00 D Fibula 6 0 0.0 0 0.00 Calcaneum 6 2 33.3 1 33.33 31.66 Astragalus 6 2 33.3 1 33.33 31.66 Tarsals 18 0 0.0 0 0.00 31.66 P Metatars 6 1 16.7 0.5 16.67 29.93 D Metatars 6 1 16.7 0.5 16.67 23.93 Ph 1 12 3 25.0 0.75 25.00 13.72 Ph 2 12 0 0.0 0 0.00 13.72 Ph 3 12 0 0.0 0 0.00 13.72 Table 10.9: concluded. MGUI 250 Cellier Lower Expected MNE % survival MAU %MAU MGUI Antler 12 0 0.00 0 0.00 1.02 Cranium 6 1 16.67 1 16.67 17.47 Mandible 36 12 33.33 6 100.00 30.26 Atlas 6 1 16.67 1 16.67 9.79 vt Cervical 30 0 0.00 0 0.00 35.71 vt Thoracic 78 0 0.00 0 0.00 45.53 vt Lumbar 36 0 0.00 0 0.00 32.05 Rib 156 0 0.00 0 0.00 49.77 Scapula 12 2 16.67 1 16.67 43.47 Pelvis 6 0 0.00 0 0.00 47.89 P Humerus 12 2 16.67 1 16.67 43.47 D Humerus 12 2 16.67 1 16.67 36.52 P Radius 12 1 8.33 0.5 8.33 26.64 D Radius 12 0 0.00 0 0.00 33.23 P Ulna 12 0 0.00 0 0.00 D Ulna 12 0 0.00 0 0.00 Carpals 72 0 0.00 0 0.00 15.53 P Metacarp 24 2 8.33 0.5 8.33 12.18 D Metacarp 24 2 8.33 0.5 8.33 10.5 Table 10.10: Summary of expected, observed and survival rate of reindeer elements in the Lower Level and their representation as Minimum Animal Units. 251 Cellier Lower Expected MNE % survival MAU %MAU MGUI P Femur 12 0 0.00 0 0.00 100 D Femur 12 1 8.33 0.5 8.33 100 P Tibia 12 1 8.33 0.5 8.33 64.73 D Tibia 12 1 8.33 0.5 8.33 47.09 Patella 12 0 0.00 0 0.00 P Fibula 12 0 0.00 0 0.00 D Fibula 12 0 0.00 0 0.00 Calcaneum 12 2 16.67 1 16.67 31.66 Astragalus 12 2 16.67 1 16.67 31.66 Tarsals 36 1 2.78 0.167 2.78 31.66 P Metatars 12 0 0.00 0 0.00 29.93 D Metatars 12 2 16.67 1 16.67 23.93 Ph 1 24 6 25.00 1.5 25.00 13.72 Ph 2 24 2 8.33 0.5 8.33 13.72 Ph 3 24 0 0.00 0 0.00 13.72 Table 10.10: concluded. Breakage patterns Reindeer long bone fragments vary in size– the average size ranges from 2.5cm to 13.8cm for long bones; with average length ranging from 4.1cm to 9.3cm (Table 10.11, Figure 10.22). The majority of the fragments are from proximal or distal fragments of long bones, with few shaft fragments. 252 femur tibia humerus metacar. metatar. radius metapod. mean 73.68 90.3 68.65 53.84 75.25 72.22 41.2 median 69.75 104.0 64.1 52.7 75.25 67.6 41.2 n/a n/a n/a n/a n/a n/a n/a longest 100.4 116.1 124.9 73.4 133.8 98.9 32.9 shortest 86.2 24.6 33.3 34.8 38.7 54.8 49.5 mode Table 10.11: Table showing reindeer appendicular bone fragment lengths, both levels (in millimeters). Figure 10.22: Graph showing lengths of reindeer long bones, in millimeters, all levels. 253 % Dry % Fresh % Indeterminate Antler 22.22 0.00 28.57 Cranium 2.78 0.00 0.00 Mandible 25.00 0.00 33.33 Vertebrae 2.78 0.00 0.00 Rib 0.00 0.00 0.00 Scapula 5.56 0.00 0.00 Humerus 0.00 22.22 0.00 Radius 2.78 5.56 0.00 Ulna 5.56 0.00 4.76 Metacarpals 11.11 8.33 9.52 Inominate 0.00 0.00 4.76 Femur 2.78 13.89 0.00 Tibia 5.56 16.67 0.00 Astragalus 0.00 0.00 0.00 Calcaneum 0.00 0.00 0.00 Metatarsals 5.56 19.44 14.29 Sesamoids 0.00 0.00 0.00 Ph 1 0.00 0.00 0.00 Ph2 0.00 0.00 0.00 Ph 3 0.00 0.00 0.00 Residuals 8.33 13.89 4.76 Table 10.12: Table showing proportions of dry, fresh and undetermined breaks for elements for reindeer in all levels of Abri Cellier. 254 Breakage patterns were similar for both levels of Abri Cellier (Table 10.12, Figure 10.23). Dry breaks were present on antler, mandible fragments and vertebrae. All other elements had a mixture of fresh and dry breaks. Dry breaks were less common on appendicular elements, but were present. This might indicate less intensive processing of elements for marrow and fat, or a higher proportion of post-depositional breakage as a result of roof collapse. Figure 10.23: Graph showing the proportion of dry, fresh and undetermined breaks by element for reindeer at Abri Cellier. Butchery Cutmarks are present on reindeer long bones: the humerus, radius, tibia, metacarpals, metatarsals, a residual phalange and a phalange. Cutmarks associated with disarticulation are present on the proximal edge of the wing of the atlas and the proximal metacarpal. A series of cutmarks on the lateral shaft of a metatarsal may also relate to 255 disarticulation of the skeleton. All other cutmarks are associated with meat removal and are placed near points of muscle attachment on the femoral shaft, the humerus, radius and tibia (Appendix, Figures A.28-A.31). Tools Antler tools and modified reindeer antler occur in both levels, but bone tools were only collected from the Lower Level (n=11). Reindeer metacarpals and metatarsals were used as supports for tools, plus an ulna, a fragment from a radius and a residual phalange. All these bones could be described as “preforms”. Two other tools on supports from reindeer were in the unknown level catalog numbers: an ulna fragment and another residual metatarsal. Worked and unworked antler is also common. A single shed antler base was identified in the Lower Level assemblage. This is from a male reindeer, but unfortunately was in too poor a state of conservation for any measurements to be taken. Horse (Equus caballus) Seven individuals are present (5 adults and 2 juveniles) based on landmarks and dentition. The axial skeleton is represented by cranial fragments (primarily mandible and maxilliary fragments) and fragments of the innominate. The majority of post cranial elements are from the appendicular skeleton, including metapodials and phalanges. The most common identified specimen was the tibia. As with the reindeer, only dense small bones are unbroken, and these form 10.3% of the bones present (1 astragalus, 1 calcaneum and 4 second phalanges). Tables 10.13 and 10.14 summarize the MNE, survival rate and MAU in relation to MGUI for the Upper and Lower Levels. 256 Cellier Upper Expected MNE % survival MAU %MAU MGUI Antler/horn core 0 0 0 0 0 Cranium 3 1 33.33 1 50 17.9 Mandible 6 0 0 0 0 7.4 Atlas 3 0 0 0 0 7.8 vt Cervical 15 0 0 0 0 45.2 vt Thoracic 39 0 0 0 0 100 vt Lumbar 18 0 0 0 0 22.4 Rib 78 0 0 0 0 Scapula 6 0 0 0 0 15 Pelvis 3 1 33.33 1 50 53 P Humerus 6 1 16.667 0.5 25 15 D Humerus 6 0 0 0 0 14.1 P Radius 6 1 16.67 0.5 25 8.7 D Radius 6 4 66.67 2 100 6 P Ulna 6 0 0 0 0 D Ulna 6 0 0 0 0 n/a Carpals 36 0 0 0 0 3.1 P Metacarpal 6 1 16.67 0.5 25 1.6 D Metacarpal 6 0 0 0 0 0.7 P Femur 6 0 0 0 0 45.4 D Femur 6 0 0 0 0 45.4 Table 10.13: Summary table of expected, observed and survival rate of horse elements in the Upper Level and their representation as Minimum Animal Units 257 Cellier Upper Expected MNE % survival MAU %MAU MGUI P Tibia 6 0 0 0 0 25.3 D Tibia 6 4 66.67 2 100 15.2 Patella 6 0 0 0 0 P Fibula 6 0 0 0 0 D Fibula 6 0 0 0 0 Calcaneum 6 0 0 0 0 7.6 Astragalus 6 0 0 0 0 7.6 Tarsals 18 0 0 0 0 7.6 P Metarsal 6 0 0 0 0 3.8 D Metatarsal 6 0 0 0 0 1.8 P metapdial 12 0 0 0 0 3.5 D metapodial 12 0 0 0 0 2.45 1st Phalange 12 0 0 0 0 0.9 2nd Phalange 12 0 0 0 0 0.9 3rd Phalange 12 0 0 0 0 0.9 Table 10.14: concluded. 258 Cellier Lower Expected MNE % survival MAU %MAU MGUI Antler/horn core 0 0 0 0 0 Cranium 2 1 50 1 66.67 17.9 Mandible 4 2 50 1 66.67 7.4 Atlas 2 0 0 0 0 7.8 vt Cervical 10 0 0 0 0 45.2 vt Thoracic 26 0 0 0 0 100 vt Lumbar 12 0 0 0 0 22.4 Rib 52 0 0 0 0 Scapula 4 0 0 0 0 15 Pelvis 2 0 0 0 0 53 P Humerus 4 0 0 0 0 0 D Humerus 4 0 0 0 0 14.1 P Radius 4 0 0 0 0 8.7 D Radius 4 0 0 0 0 6 P Ulna 4 0 0 0 0 D Ulna 4 0 0 0 0 Carpals 24 0 0 0 0 3.1 P Metacarpal 4 0 0 0 0 1.6 D Metacarpal 4 0 0 0 0 0.7 Table 10.15: Summary of expected, observed and survival rate of horse elements in the Lower Level and their representation as Minimum Animal Units. 259 Cellier Lower Expected MNE % survival MAU %MAU MGUI P Femur 4 0 0 0 0 45.4 D Femur 4 0 0 0 0 45.4 P Tibia 4 0 0 0 0 25.3 D Tibia 4 1 25 0.5 33.33 15.2 Patella 4 0 0 0 0 P Fibula 4 0 0 0 0 D Fibula 4 0 0 0 0 Calcaneum 4 3 75 1.5 100 7.6 Astragalus 4 1 25 0.5 33.33 7.6 Tarsals 12 0 0 0 0 7.6 P Metarsal 4 0 0 0 0 3.8 D Metatarsal 4 0 0 0 0 1.8 P metapdial 8 0 0 0 0 3.5 D metapodial 8 0 0 0 0 2.45 1st Phalange 8 1 12.5 0.25 16.67 0.9 2nd Phalange 8 4 50 1 66.67 0.9 3rd Phalange 8 0 0 0 0 0.9 Table 10.14: concluded. Breakage The majority of bones and some teeth are fragmented (Table 10.15, Figure 10.24). Bone fragments ranged in sized form 4.3cm to 14 cm in size, and the majority of bone fragments were from the shaft. The majority of bone fragments were retrieved from the 260 Upper Level, only 13 bone fragments were present in the Lower Level. All other elements from the Lower Level were teeth or teeth fragments. femur tibia humerus metacarpal metatarsal radius mean n/a 140.5 93.83 n/a 77.77 122.36 median n/a 140.8 92.4 n/a 68.75 141.3 mode n/a n/a n/a n/a n/a n/a longest n/a 210.4 105.2 n/a 130.7 155.4 shortest n/a 80.3 83.9 n/a 42.9 70.5 Table 10.16: Horse appendicular element fragment lengths, both levels, in millimeters. Both the fresh or indeterminate breaks are the most common on long bones in the Upper Level (each approximately 45% of the total) (Table 10.16, Figure 10.25). In the Lower Level, the degree of breakage on the teeth results in indeterminate breaks being the most common. Fresh breaks occur on long bones or long bone fragments, suggesting that some processing of horse occurred at the site. However the low number of elements makes any more definitive or robust statement difficult. 261 Figure 10.24: Graph showing the long bone length for horse elements from both levels of Abri Cellier. Figure 10.25: Graph showing the proportion of dry, fresh and undetermined breaks by element for horse at Abri Cellier. 262 Dry Fresh Unknown Cranium 8.33 0.00 0.00 Mandible 8.33 0.00 0.00 Vertebrae 0.00 0.00 0.00 Rib 0.00 0.00 0.00 Scapula 0.00 0.00 0.00 Humerus 0.00 19.23 0.00 Radius 16.67 7.69 33.33 Ulna 0.00 3.85 0.00 Carpals 0.00 0.00 0.00 Metacarpals 0.00 3.85 0.00 Inominate 8.33 0.00 33.33 Femur 0.00 7.69 0.00 Tibia 50.00 57.69 33.33 Astragalus 0.00 0.00 0.00 Calcaneum 0.00 0.00 0.00 Tarsals 0.00 0.00 0.00 Metatarsals 0.00 0.00 0.00 Sesamoids 0.00 0.00 0.00 Ph 1 8.33 0.00 0.00 Ph2 0.00 0.00 0.00 Ph 3 0.00 0.00 0.00 Residual 0.00 0.00 0.00 Table 10.17: Proportions of dry, fresh and undetermined breaks by element for horse at Abri Cellier. 263 Butchery Three horse specimens have cutmarks – a femur, a tibia, and a second phalange. Cutmarks on the tibia and femur indicate meat removal. The cutmark on the second phalange may indicate disarticulation or tendon removal. Tools One metapodial fragment and one stylet (a residual metapodial) were used as blanks for tools. These appear to be used as awls. Both were retrieved from the Lower Level. Bovids The majority of bovid elements (both teeth and bones) were present in the Upper Level. A total of 94 elements (50 teeth and 44 bone fragments) represent seven adults and 2 juveniles. Axial and appendicular elements are present in both levels, suggesting transportation of near complete carcasses (absent the vertebrae). It is possible that both bison and auroch are represented in the assemblage, as a number of relatively large bovids are present in the Upper Level, which suggests the presence of auroch. These tend to be larger than bison. The more wooded, more temperate environment suggested by the shift in herbivore populations would be consistent with the presence of auroch in the Upper Level. Tables 10.17 and 10.18 summarize the MNE, expected and survivorship of elements and the MGUI. The MGUI calculations here use Emerson’s (1993) values for bison. 264 Cellier Upper Expected MNE % Survival MAU %MAU MGUI Horn core 10 2 20.00 1 66.67 Cranium 5 0 0.00 0 0.00 Mandible 10 1 10.00 0 0.00 Atlas 5 0 0.00 0 0.00 6.4 vt Cervical 25 0 0.00 0 0.00 56.6 vt Thoracic 65 0 0.00 0 0.00 84.7 vt Lumbar 30 0 0.00 0 0.00 82.9 Rib 130 0 0.00 0 0.00 100 Scapula 10 0 0.00 0 0.00 31.6 Pelvis 5. 0 0.00 0 0.00 54.7 P Humerus 10 3 30.00 1.5 100.00 31.6 D Humerus 10 0 0.00 0 0.00 25.1 P Radius 10 0 0.00 0 0.00 16.5 D Radius 10 0 0.00 0 0.00 12.1 P Ulna 10 0 0.00 0 0.00 D Ulna 10 0 0.00 0 0.00 Carpals 60 0 0.00 0 0.00 6.6 P Metacrp 10 0 0.00 0 0.00 3.9 D Metacrp 10 0 0.00 0 0.00 2.6 14.20 Table 10.18: Observed and expected elements for bovids from the Upper Level of Abri Cellier, MAU and MGUI. 265 Cellier Upper Expected MNE % Survival MAU %MAU MGUI P Femur 10.00 1.00 10.00 0.50 33.33 69.40 D Femur 10.00 1.00 10.00 0.50 33.33 69.40 P Tibia 10.00 0 0.00 0 0.00 40.8 D Tibia 10.00 2.00 20.00 1.00 66.67 25.5 Patella 10.00 0 0.00 0 0.00 P Fibula 10.00 0 0.00 0 0.00 D Fibula 10.00 0 0.00 0 0.00 Calcaneum 10.00 1.00 10.00 0.50 33.33 Astragalus 10.00 0 0.00 0 0.00 Tarsals 30.00 0 0.00 0 0.00 P Metartars 10.00 0 0.00 0 0.00 7.5 D Metatars 10.00 0 0.00 0 0.00 7.5 Ph 1 20.00 1.00 5.00 0.25 16.67 2.40 Ph 2 20.00 1.00 5.00 0.25 16.67 2.40 Ph 3 20.00 1.00 5.00 0.25 16.67 2.40 Table 10.17: concluded. 13.60 266 Cellier Lower Expected MNE % Survival MAU %MAU MGUI Horn core 4.00 1.00 25.00 0.50 50.00 Cranium 2.00 0 0.00 0 0.00 Mandible 4.00 0 0.00 0 0.00 Atlas 2.00 0 0.00 0 0.00 6.4 vt Cervical 10.00 0 0.00 0 0.00 56.6 vt Thoracic 26.00 0 0.00 0 0.00 84.7 vt Lumbar 12.00 0 0.00 0 0.00 82.9 Rib 52.00 0 0.00 0 0.00 100 Scapula 4.00 1.00 25.00 0.50 50.00 31.6 Pelvis 2.00 1.00 50.00 1.00 100.00 54.7 P Humerus 4.00 0 0.00 0 0.00 31.6 D Humerus 4.00 0 0.00 0 0.00 25.1 P Radius 4.00 0 0.00 0 0.00 16.5 D Radius 4.00 0 0.00 0 0.00 12.1 P Ulna 4.00 1.00 25.00 0.50 50.00 D Ulna 4.00 0 0.00 0 0.00 Carpals 24.00 0 0.00 0 0.00 6.6 P Metacarp 4.00 0 0.00 0 0.00 3.9 D Metacarp 4.00 0 0.00 0 0.00 2.6 14.20 Table 10.19: Observed and expected elements for bovids from the Lower Level of Abri Cellier, MAU and MGUI. 267 Cellier Lower Expected MNE % Survival MAU %MAU MGUI P Femur 4.00 1.00 25.00 0.50 50.00 69.40 D Femur 4.00 0 0.00 0 0.00 69.40 P Tibia 4.00 0 0.00 0 0.00 40.8 D Tibia 4.00 0 0.00 0 0.00 25.5 Patella 4.00 0 0.00 0 0.00 P Fibula 4.00 0 0.00 0 0.00 D Fibula 4.00 0 0.00 0 0.00 Calcaneum 4.00 0 0.00 0 0.00 Astragalus 4.00 0 0.00 0 0.00 Tarsals 12.00 0 0.00 0 0.00 P Metartars 4.00 0 0.00 0 0.00 7.5 D Metatars 4.00 0 0.00 0 0.00 7.5 Ph 1 8.00 0 0.00 0 0.00 2.40 Ph 2 8.00 0 0.00 0 0.00 2.40 Ph 3 8.00 0 0.00 0 0.00 2.40 13.60 Table 10.18: concluded. In a modern excavation, the absence of scapulae would suggest that these meat rich elements were processed during initial butchering, and only the meat returned to the site. Only four elements, three phalanges and a sesamoid, remained intact. The remaining 92.2% of bones are fragmented. Bone fragments ranged in size from 4.5cm to 20.8cm (Table 10.19, Figure 10.26). 268 femur tibia mean 105.82 108.6 116.58 0 97.63 135.61 median 100.75 114.0 117.18 0 84.7 113.2 n/a n/a n/a n/a n/a n/a longest 155.6 140.4 147.9 0 126 208.64 shortest 77.7 44.9 77.1 0 82.2 106.7 mode humerus metacarpals metatarsal radius Table 10.20: Lengths of bone fragments for bovids from Abri Cellier, in millimeters. Figure 10.26: Graph showing the lengths of bovid appendicular elements in the assemblage, in millimeters. Breakage patterns. Fresh and dry breaks formed 34.1% of the entire appendicular assemblage each. All other breaks were indeterminate or modern. Fresh breaks were most common on the appendicular elements, suggesting processing for fat or marrow. 269 Butchery Cutmarks on bovid elements indicate meat removal and occur on a femur, a tibia and a humerus. Tools Two worked items were present: a horn core shaped into a punch and a pierced incisor that does not seem to have been recorded by White and Knecht (1992). Red deer (Cervus elaphus) Red deer shows a similar pattern to bovids, with far higher proportion of elements and individuals in the Upper Level. Of the 8 individuals present (7 adults, 1 juvenile), 7 (6 adults and the juvenile) were present in the Upper Level and one adult was identified in the Lower Level. The MNI was determined by fusion rates and landmarks son the tibia. NISP shows a similar pattern, with 44 specimens in the Upper Level and only 8 in the Lower Level. Elements present indicate transportation of both axial and appendicular elements. The MNE, survivorship, MAU and MGUI of elements present is tabulated below in Table 10.20 for the Upper Level, and Table 10.21 for the Lower Level. The MGUI values for reindeer calculated by Binford (1978) have been used in the chart. These values were selected because they were derived from another cervid and meat, marrow and fat proportions are probably similar for the two genera. The MGUI values for some red deer elements, particularly the upper forelimbs may be slightly undervalued. This is because red deer are considerably larger than reindeer, and the males carry relatively large antler racks when mature, which would increase the amount of muscle, and hence meat, in the forelimbs. 270 Cellier Upper Expected MNE % survivor MAU %MAU MGUI Antler 12 0 0.0 0 0.00 1.02 Cranium 6 1 16.7 1 20.00 17.47 Mandible 12 1 8.3 0.5 10.00 30.26 Atlas 6 0 0.0 0 0.00 9.79 vt Cervical 30 0 0.0 0 0.00 35.71 vt Thoracic 78 1 1.3 0 0.00 45.53 vt Lumbar 36 0 0.0 0 0.00 32.05 Rib 156 1 0.6 0 0.00 49.77 Scapula 12 0 0.0 0 0.00 43.47 Pelvis 6 0 0.0 0 0.00 47.89 P Humerus 12 0 0.0 0 0.00 43.47 D Humerus 12 0 0.0 0 0.00 36.52 P Radius 12 0 0.0 0 0.00 26.64 D Radius 12 0 0.0 0 0.00 33.23 P Ulna 12 1 8.3 0.5 10.00 D Ulna 12 0 0.0 0 0.00 Carpals 72 0 0.0 0 0.00 15.53 P Metacarp 12 0 0.0 0 0.00 12.18 D Metacarp 12 0 0.0 0 0.00 10.5 Table 10.21: Observed and expected elements for red deer, MAU and MGUI for the Upper Level of Abri Cellier. 271 Cellier Upper Expected MNE % survivor MAU %MAU MGUI P Femur 24 0 0.0 0 0.00 100 D Femur 24 0 0.0 0 0.00 100 P Tibia 12 10 83.3 5 100.00 64.73 D Tibia 12 1 8.3 0.5 10.00 47.09 Patella 12 0 0.0 0 0.00 P Fibula 12 0 0.0 0 0.00 D Fibula 12 0 0.0 0 0.00 Calcaneum 12 0 0.0 0 0.00 31.66 Astragalus 12 0 0.0 0 0.00 31.66 Tarsals 36 0 0.0 0 0.00 31.66 P Metatars 12 0 0.0 0 0.00 29.93 D Metatars 12 0 0.0 0 0.00 23.93 Ph 1 48 0 0.0 0 0.00 13.72 Ph 2 48 0 0.0 0 0.00 13.72 Ph 3 48 0 0.0 0 0.00 13.72 Table 10.20: concluded. 272 Cellier Lower Expected MNE % survivor MAU %MAU MGUI Antler core 2 0 0.00 0 0 1.02 Cranium 1 1 100.00 1 100.00 17.47 Mandible 2 1 50.00 0.5 50.00 30.26 Atlas 1 0 0.00 0 0 9.79 vt Cervical 5 0 0.00 0 0 35.71 vt Thoracic 13 0 0.00 0 0 45.53 Vt Lumbar 6 0 0.00 0 0 32.05 Rib 26 0 0.00 0 0 49.77 Scapula 2 0 0.00 0 0 43.47 Pelvis 1 1 100.00 1 100.00 47.89 P Humerus 2 0 0.00 0 0 43.47 D Humerus 2 0 0.00 0 0 36.52 P Radius 2 0 0.00 0 0 26.64 D Radius 2 0 0.00 0 0 33.23 P Ulna 2 0 0.00 0 0 D Ulna 2 0 0.00 0 0 Carpals 12 0 0.00 0 0 15.53 P Metacarp 2 1 50.00 0.25 25.00 12.18 D Metacarp 2 1 50.00 0.25 25.00 10.5 Table 10.22: Observed and expected elements for red deer, MAU and MGUI for the Lower Level of Abri Cellier. 273 Cellier Lower Expected MNE % survivor MAU %MAU MGUI P Femur 4 0 25.00 0 0 100 D Femur 4 1 25.00 0.5 50.00 100 P Tibia 2 0 0.00 0 0 64.73 D Tibia 2 0 0.00 0 0 47.09 Patella 2 0 0.00 0 0 0 P Fibula 2 0 0.00 0 0 0 D Fibula 2 0 0.00 0 0 0 Calcaneum 2 0 0.00 0 0 31.66 Astragalus 2 0 0.00 0 0 31.66 Tarsals 0 0 0.00 0 0 31.66 P Metatars 2 0 0.00 0 0 29.93 D Metatarl 2 0 0.00 0 0 23.93 Ph 1 8 0 0.00 0 0 13.72 Ph 2 8 0 0.00 0 0 13.72 Ph 3 8 0 0.00 0 0 13.72 Table 10.21: concluded. Breakage All bone was fragmented, but the items were relatively large in size, ranging in length from 6.3cm to 15.4cm. This suggests selection by the excavators for larger pieces of bone (Table 10.22, Figure 10.27). The majority of the breaks were fresh (60%). 274 femur tibia humerus metacarpals metatarsal radius metapodial median n/a 100.6 81.2 109.05 n/a n/a 93.3 mode n/a n/a n/a n/a n/a n/a n/a longest n/a 150 90.2 154.9 n/a n/a 106.5 shortest n/a 63 66.3 81.4 n/a n/a 80 Table 10:23: Bone fragment lengths for red deer elements from Abri Cellier, all levels. Figure 10.27: Graph showing the mean, median, longest and shortest length for red deer bone. Butchery Cutmarks consistent with disarticulation are present on an ilium, and cutmarks consistent with meat or tendon removal are present on a humerus and tibia fragment. 275 Tools. No tools were identified on red deer bone fragments. One red deer ulna (unfortunately from an unknown level and not included in this analysis) shows evidence of raclage and may have been shaped into a tool. Saiga (Saiga tataricus) A smaller hypsodont herbivore is present at Abri Cellier. Dentition and tooth size do not fit well with ibex (Capra ibex) or chamois (Rupicapra rupicapra). The post crania are almost identical to those of the pronghorn antelope specimen held at the University of Iowa. While the pronghorn is in a different order to the saiga (Artilodactylae,) the similarities in habitat have resulted in convergent evolution, suggesting that the elements in the collection are from a grassland animal. Given these affinities, this small hypsodont is assigned to saiga. This identification is also congruent with the environmental reconstruction. Saiga is most common in the Lower Level, where the environment was open and probably cooler and possibly drier than that of the Upper Level. This is an environment (steppe/grassland/prairie) favored by modern antelope and pronghorns. Small cervidae, such as roe deer (Capreolus capreolus) occupy temperate woodland habitats. Other small bovids, such as chamois or ibex, also occupy forested environments or high altitude and rocky environments. A minimum of three saiga are present, two from the Lower Level and one from the Upper Level. This reflects the proportions of NISP, with nine elements in the Upper Level and 20 in the Lower Level. Teeth, axial and appendicular elements are present, suggesting transport of whole carcasses to the site for processing. Seven elements had at least one fresh break, and three exhibited at least one dry break. Butchery Cutmarks are present on two metatarsals and a scapula. 276 Cervidae This taxon includes antler or post crania that could not be assigned with precision to reindeer or red deer. The majority of the group is represented by antler (both tools and unworked fragments) plus four bone fragments. The far higher proportion of antler in the Lower Level (153: 77) suggests that this material is reindeer antler, given the dominance of reindeer in the Lower Level and the absence or near absence of other large cervids. The antler is largely antler tools or points and, as such, is likely a good representative sample of the material collected, given the Logan Museum excavators’ interest in the tools of “Early Man”. Other herbivores Other herbivores are present in very low frequencies –a hare humerus in the Lower Level and mammoth ivory in the Upper Level. There are no butchery marks on the hare humerus, nor any signs of animal gnawing. The hare could have been introduced to the site by humans or carnivores, or by chance. The mammoth ivory was likely collected rather than hunted. One large unidentified bone also likely came from megafauna, based on cortical thickness. Summary for herbivores In summary, elements for herbivores indicate that entire or nearly complete carcasses of reindeer, and saiga were transported to the site for processing. The evidence for butchery and transportation practices for larger herbivores (red deer, bovids and horse) is less clear. The presence of teeth indicates transportation of crania, and long bone fragments suggest transportation for processing for marrow of the lower limbs or red deer and bison. The low frequency of elements and the clear distortion of the assemblage through excavator selection make any further statements about processing and transportation problematic. Vertebrae and scapulae are underrepresented, but this is more 277 likely a result of taphonomic factors (including collection practices) than prey transportation decisions. Non-herbivores Small amounts of non-herbivores were present in the Logan collections, largely represented by teeth (canines), many of which were modified for suspension. Three carnivores are present: wolf, fox and bear. Non-carnivores present are wild boar, rodent (a modern intrusion), bird and fish. All taxa apart from the rodent were transported to the site by humans. This statement is based on the absence of any incisors or molars of cave bear, fox and wolf. The high proportion alteration of carnivore teeth also argues strongly for human transportation and accidental loss of items. Wolf (Canis lupus) Two wolves are present in the collection, with a NISP of five. The taxon occurs in both the Upper and Lower levels. A wolf mandible is present in the Upper Level. An ulna and four teeth (all canines) were retrieved from the Lower Level. One canine was modified for suspension, but the most interesting item is the ulna. This has been used as a tool – the distal end is rounded, the inter-osseous crest has been smoothed down and the medial surface (opposite the crest) has been smoothed. This will be discussed in the following chapter. Fox Two foxes are present, with an NISP of five. All are from the Lower Level, and include two metatarsals, a phalange and two canines. One canine is altered for suspension. There is not enough data to determine species with any confidence. Modern red and arctic fox have a considerable overlap in range, with red fox ranges extending into the Arctic Circle in Europe and Artic fox extending south into the boreal forest. 278 Bear (Ursus speleaus) Four bear canines were present in the Lower Level. One is altered for suspension. Wild boar (Sus scrofa) A wild boar tusk was found in the Lower Level and a wild boar second molar in the Upper Level. Wild boar is reported as a minor component in a number of Upper Palaeolithic faunas in Eurasia, and it is possible that this specimen was hunted. Bird (Aves) Five fragments of bird bone were recovered from the Lower Level, all are long bones, and all but one are worked. The unworked bone is a humerus comparable in size to a Brent Goose, but has some morphological dissimilarities, Two other long bone tubes (worked) are similar in size suggesting exploitation of larger, possibly migratory, waterfowl. Fish (Pisces) A single fish vertebra was collected from the Lower Level, with a deep cutmark across the centrum indicating processing by humans. (The cynic in me would argue that it was mistaken for a bead, and hence collected). The presence of the vertebra indicates occasional exploitation of fish from the nearby river. Summary for non-herbivores Non-herbivore material present in the collection shows the transportation and alteration of carnivore teeth by humans. Bird bones at the site are also worked. There appears to be a strong selection for teeth and identifiable items that might have been decorative items. 279 Prey selection As noted in this chapter, the role of excavator bias makes statements regarding subsistence behavior at Abri Cellier tentative. The presence of axial and appendicular material for reindeer and saiga suggests that these animals were transported to the site and processed on site. It is less clear if larger mammals were transported in their entirety to the site or partially butchered elsewhere. Density data makes any statement regarding a gourmet or generalist strategy inappropriate for red deer and horse. The relatively few elements from bovids and saiga make any attempt at defining a relationship problematic. Reindeer does not appear to be impacted by density issues, but excavator selection may have skewed the data, particularly with their emphasis on collecting teeth. Plotting MNE against the MGUI shows a slight indication of a generalist strategy for the Upper Level, but the Lower Level shows no particular relationship between MGUI and MNE (Figures 10.28 and 10.29). Figure 10.28: Bivariate plot of MAU and MGUI values for reindeer from the Upper Level of Abri Cellier. 280 Figure 10. 29: Bivariate plot of MAU and MGUI values for reindeer from the Lower Level of Abri Cellier. The intensity of processing seems less than that of the Neanderthals at the Grotte du Renne, based on the larger lengths of the long bone fragments and the lower frequency of green breaks. Nevertheless, long bones do appear to have been fractured to obtain marrow. This may relate to site function, or to the season of occupation. One shed antler base in the Upper Level indicates some use of the site in the late Fall or early Winter, after male deer have shed their antlers. Other seasonality data is less clear. Likewise, cutmarks are not common in the assemblage. Only 38 elements exhibit traces of cutmarks and of these only 27 are on specimens identifiable to genus or species. Cutmarks on reindeer are shown in the appendix. Reindeer are shown in this manner because this is not a density mediated assemblage, unlike the horse and red deer. The reindeer assemblage is also large enough to assume that the cutmark patterns are representative. 281 Figure 10 30: Graph showing wear stages for reindeer mandibular molars from both levels of Abri Cellier. Tooth wear patterns were examined for reindeer, as these were the most common and more likely to be representative, based on mandibular molars and using Grant’s (1982) wear stages. Two peaks are shown in the data (Figure 10.30), suggesting that relatively young animals are being taken alongside mature individuals. One individual has unworn deciduous teeth, which suggests an individual that is recently weaned, possibly in the first year of life, and at least one individual has unworn permanent third molar (Grant stage a). An unweaned or partially weaned individual suggests an occupation in the late summer or fall, which contrasts with the presence of shed male antler, which indicates a visit in the late fall or early winter. This suggests that Abri Cellier was utilized as a base camp or habitation at different seasons of the year, probably as part of an overall foraging or encounter-based subsistence strategy. The change in proportions of open, cold adapted species with taxa associated with milder more forested environments between the Lower Level and the Upper Level indicates that the Aurignacian occupants of Abri Cellier pursued an encounter rather than 282 an intercept strategy to obtain game. There is no focus on a particular species, and, if the tooth eruption and wear patterns are any indicators, the site may have been used at different times during a single seasonal round. There is also the possibility that the milder climate and more varied available species resulted in a reduction in mobility. Woods (2011) found that there was a lower percentage of exotic flint and chert in the Upper Level, which suggests that the Upper Level occupants did not travel over as wide a territory to obtain lithic raw materials and, presumably game. The amelioration in climate and greater availability of prey may have resulted in less need to travel over a wide territory. Another possibility is that infilling of the region had occurred between the Aurignacian I and Aurignacian II, resulting in a larger population with smaller band or group territories. Tool blanks There are 167 worked antler items present, plus antler tines cut from the beam, beam fragments and an unworked shed antler base which indicates that some antler was collected and processed at the site. Bone tools or worked bone fragments form 10% of the Upper Level worked assemblage, (in total, 3 long bone fragments), and 38% of the Lower Level assemblage. Tools from the Lower Level include lissoirs (defleshers) made on split ribs from large mammals; poinçons/awls on reindeer ulnae, metartarsals and metcarpals; plus a range of tools on pieces of large and medium mammal cortical bone. Two horse stylets and a metatarsal were worked. Bird long bones served to make fine awls and tubes. A bovid horn core was shaped into a wedge or chisel and a wolf ulna used as a tool. The tools present suggest that there was no particular selection for, or transportation of, skeletal elements to make bone tools. The raw material was obtained as part of quotidian subsistence practices: using reindeer and horse lower limb bones when these species are available and long bones when bovids and red deer dominate the fauna. 283 Is this pattern present elsewhere? In the next chapter we will examine the evidence for choice of tool blanks at other Châtelperronian and Aurignacian sites and compare these practices to the evidence from Abri Cellier and Level Xc of the Grotte du Renne. It may be that the selection of tool blanks is entirely pragmatic. We will also examine if there is any evidence that the introduction of antler as a raw material in the Aurignacian would require a shift to a more logistical pattern of collection to obtain adequate supplies of the raw material. The focus on antler tools, particularly armatures, in the archaeological literature may overstate the importance of this raw material. Conclusion The faunal assemblage in the Upper and Lower Levels at Abri Cellier does not show any evidence for a logistical subsistence strategy. The proportions of animals present suggest an opportunistic or foraging strategy. The amelioration of the climate that occurred between the Lower and Upper Levels probably resulted in a more productive environment with higher biomass. This might result in easier access to subsistence items, and a reduction in overall mobility, which is suggested by the lithic raw material at the site. There is no apparent preferential transportation of particular elements to the site. Raw material for bone tools appears to be derived from the skeletal elements available at Abri Cellier. 284 CHAPTER 11: DISCUSSION: A BONE TO PICK, OR SCRAPE WITH Introduction Bone tool use is known in the Lower and Middle Palaeolithic (e.g. Burke and d'Errico 2008; Tartar 2009). The use of bones as digging tools has been argued to date back into the Early Pleistocene (d'Errico and Backwell 2003). Bone tools are known from Middle Stone Age contexts in South Africa, but appear and disappear from the archaeological record during the Pleistocene in African and Eurasia. The emergence of osseous technology in the Upper Palaeolithic in Europe and the Middle Stone Age in Africa is frequently argued to be a marker of modern human behavior. However, formal studies of bone tools and manufacturing processes emerged relatively recently in archaeology. This chapter will provide an overview of the development of bone tool studies in archaeology, a general discussion of bone formation and mechanics and a review of bone tool use and selection in the ethnographic and archaeological record. Both the Neanderthals at the Grotte du Renne, Level Xc and the modern humans who occupied Abri Cellier discarded bone tools at the sites. Both sites demonstrate the use of ivory and bone in the daily subsistence pattern. At the Grotte du Renne, the excavators argue for the use of mammoth tusks as supports for the three cabins along the north wall of the shelter. At Cellier, unworked ivory fragments are present, and a single ivory pendant has been described by White and Knecht (1992). The osseous industries are consistent with previously analyzed materials for the Châtelperronian and Aurignacian: no antler tools are present in the Châtelperronian levels at the Grotte du Renne, but there are numerous antler points and other tools in the Cellier material. Two questions arise: first, are there any differences in the types of bone used or the degree of bone use between the two cultures? Second, given the use of antler only in the Aurignacian, did the Aurignacian inhabitants at Abri Cellier need antler in such quantities that the collection of antler had to be scheduled into the annual round? 285 Osseous material culture studies: a brief history The study of bone tools in the archaeological record, especially the technology of bone tool production, considerably post-dates the technological studies of lithic production (Averbouh 2000a). While archaeologists in the early twentieth century did discuss bone tools (e.g. Breuil and Crawford 1938), particularly the use of broken bones as informal tools (Le Moine 2007) or Dart’s arguments for an early osteo-dentokeratic culture among the australopithecines, few formal studies were undertaken. A significant milestone in bone tool studies in Europe was the publication of an analysis of antler working based on the excavated material from Star Carr (Clark and Thompson 1953). Despite interest in bone tools in the Palaeolithic record from the nineteenth century onwards and their use as chronological markers, formal studies, replication experiments and a formalization of terminology only began in French Palaeolithic studies in the 1970s (Averbouh 2000a:20). The study of bone tool production and creation of a chaîne opératoire was hampered by issues of conservation, a lack of production debris and difficulty in determining the original element, or even raw material, of some heavily modified tools. While formal bone, antler and ivory tools were recognized and collected by archaeologists, informal tools or bones or tool fragments were frequently deleted from the archaeological record as part of the once-standard practice of discarding unidentifiable bone during or shortly after excavation. As a result, these items remain unrecognized in the archaeological record. An example is my identification of four additional tool fragments from the Grotte du Renne Xc diaphysis fragments. The bone tool manufacturing techniques employed during the Upper Palaeolithic are relatively simple and the osseous artifact assemblage is not particularly elaborate, when compared with later worked bone assemblages from the Mesolithic through the Middle Ages (Ramseyer 2004). Bone tools serve as proxies, e.g. “bone tools, more than any other types of artifact, have also been used as markers for other, missing elements of technology” (Le Moine 2007:16), particularly as an indicator of clothing manufacture. In 286 the Early Upper Palaeolithic, this becomes particularly important as a means of examining Neanderthal adaptations or innovations during a period of unstable and increasingly cool climate at Arcy-sur-Cure. The existance of clothing in the Aurignacian is not a matter of debate, but the number and types of tools present could indicate the degree of investment in the preparation of clothing and shelters. Bone formation and structure Bone tissue comprises collagen and mineral crystals (hydroxyapatite) which surround the collagen fibrils. At the microscopic level these organic and inorganic compounds are organized into woven bone and lamellar bone. Woven bone, as the name suggests, shows a lack of orientation in the structure of the collagen fibers and the apatite crystals and has a higher mineral/organic ration than lamellar bone. In lamellar bone the collagen fibrils are grouped and organized in distinct domains within separate layers or lamellae. The lamellae are separated by inter-lamellar bone, with occasional fibrils passing from one lamella to another (Currey 1979, 1990, 2002; Francillon-Viellot, et al. 1990; MacGregor 1985; MacGregor and Currey 1983; Semenov 1964). Blood vessels run through both types of bone structure - randomly in woven bone and in the general orientation of the lamellae in lamellar bone. Either structural system can be modified into a type of bone characterized by Haversian systems within the bone. In a Haversian system, an erosion cavity forms around the blood vessel and subsequently increases in size. The surface of the cavity is smoothed off, and new bone is laid down on the interior of the erosion cavity, resulting in a Haversian system with collagen fibrils organized in a spiral within the channel. Haversian bone, therefore, is a collection of Haversian systems. Haversian bone has important mechanical properties and forms the majority of bone in a mammal skeleton, forming both cancellous and cortical bone. Another bone structure that develops is laminar bone, which incorporates both woven and lamellar bone. This is form of bone structure is created by the growth of 287 woven bone around blood vessels on the surface of the bone, under or in the periosteum (not within the bone as with Haversian systems), engulfing the blood vessels which are left in a series of cavities. Lamellar bone forms these cavities while further laminae are being created on the bone surface (Currey 2002; Francillon-Veillot, et al, 1990; MacGregor 1985; MacGregor and Curry 1983). Bone formation and renewal continues through the life of an animal. MacGregor (1985) differentiates between membrane bone and cartilage bone in the foetus depending on the origin of the soft material that is ossified during pregnancy through the action of osteoblasts (bone producing cells). Some bone-forming membranes are retained into adulthood - the periosteum, on the outer bone surface and endosteum within certain bones serve to rework and/or repair bone over the lifetime of the organism. The deposition of lamellar tissue will convert cancellous bone to cortical bone, and cortical bone can be resorbed into a cancellous form. Cortical and cancellous bone structures are found in bones with medullary cavities. Thick cortical bone surrounds the medullary cavity, with cancellous tissue with a thin covering of cortical bone found at the epiphyses. All long bones have this structure. Bones that lack medullary cavities have a superficial layer of compact bone around a thick, spongy tissue; for example vertebrae or elements of the wrist and ankle (MacGregor 1985). The bone structures discussed begin forming in utero and continue to develop during the life of the animal. Juvenile bone formation includes ossification of cartilage bones - cartilage grows and is replaced by bone until maturity. Membranal bone growth occurs around the peripheries of epiphyses, diaphyses and bone plates, forming sutures that fuse in maturity. Woven bone forms the majority of the embryonic skeleton and at the margins of growing bones (it also forms at fracture sites). Woven bone is then converted into lamellar bone, which forms the majority of the adult skeleton. As the individual ages, bone renewal slows, and the higher proportion of old tissue results in increasing brittleness. After bone formation is completed, bone is reworked through 288 resorption in response to dietary stress, physiological stress, such as pregnancy and lactation, physical stress and loading, damage and disease. Bone quality varies in an animal as a result of aging and stressing. Older animals’ bones become increasingly brittle as bone renewal slows. Because the skeleton serves as reservoir for minerals, even young animal bone can be poor in quality. The overall health and age of an animal will therefore be reflected in the quality and strength of its bone (Currey 2002; FrancillonVeillot et al, 1990; MacGregor 1985; MacGregor and Curry 1983). The elasticity of bone is an important factor in the choice of support for bone tools (Averbouh 2000a; Burr 1980; MacGregor and Currey 1983; O'Connor 1987; Scheinsohn 2010; Scheinsohn and Ferretti 1995; Semenov 1964; Steinbring 1966; Tartar 2009). Elasticity refers to resistance to elastic deformation (or stress), and failure can result in sudden breakage, permanent deformation or splitting of the bone. The underlying bone structure and degree of mineralization are major factors in elasticity and reaction to stress. Woven bone is less strong but has more reliable performance in applications in which it is stressed in different directions, while highly mineralized bone has low elasticity and resistance to impact stress. Bones with a higher elasticity would be preferred for tasks where pressure was applied consistently along the bone, while less mineralized elements, such as antler, would be preferred for tasks that required greater resistance to impact (MacGregor 1985:24). Antler and ivory formation Antler and ivory are two other sources of raw material for tool and ornament manufacture. Antler is an annually forming outgrowth of bone, carried by most male cervids. The primary function of antler is for display, but a secondary function (noted by the author while vacationing in Norway) is to break up the line of the animal, acting as a form of camouflage (particularly effective when a troop is standing on highways!). Antler forms annually on pedicules (permanent protuberances on the frontal bone) (MacGregor 289 1985, Semenov 1964). Growth can be rapid, with blood vessels carried both internally from the frontal bone and in the velvet that covers the forming antler. Bone tissue forms around the latter vessels as the antler enlarges, producing a distinctive channel or gutter pattern that can indicate genus or species. The blood supply diminishes as the base of the antler ossifies, resulting in the death of the antler and the velvet. Antler formation results in mineral resorption from the skeleton during the growing period. Antlers increase in size and complexity as an animal matures, but are impacted by disease, age, damage and dietary stress. Antler is shed when osteoclast cells erode the tissue at the junction of the antler burr and the pedicule. As a result of the rapid growth of antler, the ossified tissue comprises compact outer layer (grooved with the marks of the velvet blood system), with a spongy core of cancellous tissue. Growth patterns and structure have implications for the strength of the antler, and its ability to withstand pressure or shock. This varies by species within the cervidae (Guthrie 1983). Ivory is derived from teeth, usually from large land mammals, such as elephants, or sea mammals, such as walrus or narwhals (Carlson 1990). Enamel forms from ameloblasts, which form into a honeycomb pattern interspersed with secretory vessels (Tomes processes). As apatite crystals form they increase in size and force the organic matrix to the border. Dentine, which underlies enamel is formed from odontoblasts, and varies in structure depending on the degree of calcification. Cement, the third mineralized tissue present in teeth, is structurally similar to bone. Growth takes place in the tooth through the action of odontoblasts that line the pulp cavity after the enamel cap is formed. While growth generally ceases upon maturity, some mammals have continuously growing teeth. These include elephant and walrus incisors, where growth continues to maintain the length of the tooth in reaction to wear. Walrus and elephant ivory are derived from the overdeveloped upper canines (walrus) or upper incisors (elephants), which are largely composed of dentine. Elephant ivory has a unique internal structure of tubules that is thought to be a response to the weight of the tusk and the stresses put on 290 tusks by elephants. As ivory desiccates, it fractures along these lines. This pattern of fracturing was exploited during the Upper Palaeolithic, where Gravettian artisans preferentially selected partially fossilized ivory to manufacture tools and decorative items (Goutas, pers. comm.). Bone tool manufacture and use: archaeological, ethnographic and experimental data The properties of bone are important in the selection of skeletal elements for tool manufacture. Both the mechanical properties of the element and the intended function of the tool impact the selection process (MacGregor 1985; MacGregor and Currey 1983). For example, Natufian bone tools are largely made on the shaft portions of ungulate long bone where the osseons are parallel and bones are chosen for the longest grain - ungulate metapodials and tibiae (Campana 1989:23). While Natufians appear to select for the longest grained bones, other studies suggest that regular manufacturers of bone tools had a clear understanding of the properties of the different bones used to manufacture tools for different tasks (Buc 2011; Le Moine 1991; Scheinsohn 2010). In Tierra del Fuego Scheinsohn and Ferrreti (1995) found a high rate of recurrence in the use of camelid, pinniped, cetacean and avian bones for different tools. Tools for penetration without impact, such as awls, should have a high elasticity modulus to withstand the loading. Bones used for leverage (such as bark removal) would require high moment of inertia for stiffness and strength relative to length. Bone tools for pressure flaking require good geometric and structural properties to withstand both impact and loading. Bones used to make tools for hunting or for use as wedges (impact mediated penetration) require high elasticity and inertia points for fixed points, while flexibility would be favored for detachable points. Wedges would require high energy absorbance in elastic conditions (Scheinsohn and Ferretti 1995). Experimental studies and analysis of artifacts from Tierra del Fuego found a correlation with the different 291 categories: tools for penetration without impact, such as awls, were made on cormorant humeri or guanaco metapodials (Scheinsohn 2010). Tools for leverage, such as bark removal to make containers or boats, required stiffness and strength and were only manufactured from guanaco metapodials. The same material was exclusively used for pressure flaking stone tools. Fixed harpoons were made from the metapodials of guanaco or cetacean clavicles, while detachable points were made from cetacean bone, which has a low stiffness compared to guanaco bone. Wedges were made from cetacean clavicles or pinniped radii or cubitii (Scheinsohn 2010). Bone tool utilization in Tierra del Fuego relates to the availability of raw materials and the local ecology. Scheinsohn (1995) argues that while wood has similar properties to bone, bone was preferred for tools in Tierra de Fuego because of the local climate. The extremely high humidity (in a temperate rain forest) results in rapid decay of wood, while bone is more damp resistant and widely available from both terrestrial and marine resources. Clearly procurement costs for raw material and longevity of formal tools should be considered in the adoption of bone tools. The issues of seasonal availability of different types of skeletal material are not addressed. Marine mammals may be available at certain times of year (breeding season or molt), and terrestrial mammals also migrate to summer or winter territories. Selection of raw material may also relate to forward planning, anticipating needs for tools while the raw material is readily available. The choice of raw material therefore depends on the mechanical properties of the element in terms of its elasticity and/or resistance to torsion; its suitability for the proposed task; the aesthetic properties, including comfort of handling, and, most importantly, the ease of access to the materials (Averbouh 2000a:105). In a study of the material culture from Kangiguksuk, Alaska, Hall (1971) demonstrates a strong preference for antler as a raw material for arrowheads, picks, scraper hafts, hammers, adze heads, and wood splitting wedges. Bone tools included awls and bodkins, retouchers, a reindeer 292 radius used as a beamer and bone arrowheads. The bone tools were largely manufactured from by-products of marrow processing (Hall 1971:57). A similar suite of formal and expedient tools was excavated at Hawikah, New Mexico where a variety of awls and other tools were manufactured on long bones of mammals, birds and reptiles for use in basketry and hide clothing manufacture (Hodge 1920). No preference was shown for any particular element for awls or bone chisels; but ribs were preferred for knives and antlers were preferred for punches and handles. The only strong preference for a particular element was the use of jack rabbit metapodials for needles and turkey or chicken longbones for tubes for beads or whistles. Similar exploitation of available raw materials is seen in Patagonia, where a tool made from a camelid tibia at La Olla 1 was manufactured by breaking the bone with an anvil facture, used and discarded on site; and there was little investment in refurbishment, in a manner similar to the production and use of lithics at the site (Johnson, et al. 2000). Le Moine (1991) found that the MacKenzie Inuit made many tools on reindeer metapodials that had been processed for marrow. Metapodials appear to be favored for tool manufacture in many different archaeological and ethnographic cultures. In the postRoman period in the Old World metapodials from sheep, cattle and horse were the most commonly used bones. In this case it is hypothesized that the bones were supplied as a by-product of slaughter for meat or from hide processing, as the bones are of low meat utility (McGregor 1985). The preference for metapodials could also relate to bone resilience and ease of handling. The raw material for bone tools appears to be a by-product of processing animal carcasses for meat and hides across a wide range of economic systems. But what of antler? This is a seasonally available raw material, a result of the annual growth and shedding of antler by all male cervids and by female reindeer. In the Post-Roman and Medieval periods in Europe, antler was derived from shed antler, and energy had to be invested in collection and construction of storage facilities (MacGregor and Currey 293 1983). Among modern and recent hunter-gatherers antler could be obtained through hunting or by collection while foraging. Stenton documents the logistical behavior of the Eastern Inuit, a group who focused on maritime resources for subsistence, in the Arctic. This group traveled to the reindeer breeding grounds in the fall to obtain shed antler to manufacture armatures and other equipment (Stenton 1991). The Eastern Inuit hunted reindeer in the fall to obtain pelts suitable for the manufacture of parkas warm enough to withstand life on the ice in winter. This logistical behavior was clearly a product of both reindeer ethology and the subsistence organization and energetic requirements of the Inuit themselves. Bone tool manufacture in the Upper Palaeolithic and the industries at Arcy-sur-Cure and Abri Cellier In the Upper Palaeolithic there is a clear choice of antler for armatures, and bone for tools used in what is sometimes referred to as the “domestic” sphere ; the bone toolkit remains relatively unchanged from the Early Upper Palaeolithic, unlike armature or ornaments (Tartar 2009; Tartar, et al. 2006). Bone tool manufacturing processes are reductive, requiring the shaping of bones into the desired tool, whether formal or informal. Hominins would have become familiar with the breakage patterns of bones as a result of splitting or smashing bones to acquire marrow. Bone has similar properties to stone (it is solid, hard and resistant) and to wood (fibers aligned along longitudinal axes) (Averbouh and Provenzano 1998:9). The use of manufacturing techniques similar to lithic tool manufacture is seen in the Mousterian, where bone tools are flaked (and this technique occasionally occurs in later phases – three flaked bone tools were identified among the Level Xc diaphysis fragments during this research project). Other methods (grooving and snapping, sawing, shaving, abrasion, polishing and incising) appear in the Early Upper Palaeolithic, including the Châtelperronian. These techniques may be derived or transferred from woodworking (Liolios 2003:221). Late stages of tool 294 manufacturing can be identified on the surface of the tool (for example, shaving to remove periosteum or polishing). Unfortunately, the early stages of bone tool manufacture may be subsumed into analyses of butchery practices, given the breakage of the element to provide a support (Tartar 2009:56). With antler, it is easier to identify debitage from the reductive process, as the deliberate breakage of this material is unlikely to be related to acquisition of subsistence material. The same applies to ivory. Unfortunately, prior to the introduction of modern excavation and recovery techniques (and even after this) much unidentifiable bone, antler and ivory material was not collected. Bone tools from Châtelperronian contexts are reported from Quinçay, although these remain unpublished (d’Errico, et al. 2003:267), and were also recorded at sites excavated by the Abbé Parat at Arcy-sur-Cure in the late early twentieth century (as described above). To date, the site with the highest number of bone tools, and, as a result, an extremely controversial site, is the Grotte du Renne, Arcy-sur-Cure. The published data will be discussed below. A full report on the Châtelperronian worked bone assemblage is currently in progress. In her review of bone tools from four Aurignacian sites (Castanet Sud, Castnet Nord, Brassempuouy and Gatzarria); Tatar notes that in three of the four sites, bone tools were more common than antler tools (2009:55). Many of the supports (blanks) were from meat-rich bones, therefore the whole carcass would not necessarily be transported to the site (220). Both Castanet Nord and Castanet Sud are Aurignacian Ancien (Aurignacian 1). Castanet Nord was excavated by M. Castanet under the supervision of Denis Peyrony. The tools are a representative sample of the worked bone from the site, but the collection is not complete, a situation analogous with Abri Cellier. Castanet Sud is the focus of ongoing excavations by Randall White, using modern data-recovery techniques. Castanet Sud has produced evidence of significant amounts of bone burning, which may skew the NISP counts tabulated below. 295 Tool type Castanet Nord Castanet Sud Brassempouy Gatzarria 42 0 30 9 Awl 43 3 11 11 Retouchers 42 83 57 62 Small, pointed 21 0 37 0 Multi-use 35 5 30 9 Batonnet 13 0 46 0 Points 131 8 25 42 Other 39 11 5 3 Total 366 110 241 136 Bone tools Lissoir Antler tools Table 11.1: Bone and antler tools from four Early Aurignacian sites in southwest and southern France. Table 11.1 summarizes the bone tool assemblages from the four sites and Table 11.2 summarizes the NISP data for the four sites. All data in the tables are adapted from Tatar (2009). Data from Brassempouy are derived from a series of excavations at a number of locations within the cave system. The lower levels of the Aurignacian occupations are dominated by reindeer, and the upper levels by horse, but with a more even distribution of other large herbivores (Tartar 2009:48). The fauna is anthropogenic but the site was also used as a den by hyenas, resulting in damage and possibly fragmentation of the faunal remains. Tartar does not state what evidence there is for carnivore ravaging or carnivores as actors in bone accumulation. In contrast to Castanet Nord and Sud, which 296 are winter occupations, the site of Brassempouy reflects a series of occupations from spring through fall with no winter occupation identified to date. Fauna Castanet Castanet Brassempouy Brassempouy %NISP Nord* Sud Lower levels** Upper levels Gatzarria Reindeer 90 90 50 nd 2.7 Horse 5 5 nd 30 9 Bovidae >1 >1 nd nd 43 Red deer 0 0 0 0 22 Roe deer 0 0 0 0 16 Other 4 4 nd nd 7.3 * No data from Peyrony, test excavations of the back dirt indicate similar proportions of the fauna ** Tartar does not provide further data Table 11.2: Percentage of NISP for herbivores at four Early Aurignacian sites in southwest and southern France. Tartar also examined a sample of tools from the Pyrenean site of Gatzarria. This dataset was still under analysis at the time of her research, and had not been fully analyzed. The fauna from this site was dominated by bison, followed by red deer, roe deer and horse with very few reindeer. This faunal composition suggests a local environment similar to that of the Upper Level from Abri Cellier. No site produced any formal bone working debitage, in contrast to the antler debitage recovered. This was largely because the bone breakage techniques were the same as those used for marrow processing. The simple reduction sequences (largely 297 shaving the tools) also reduced the likelihood of identifying unfinished or roughed-out bone tools (Tartar 2009:62). Choices of support for the tool types present (lissoirs and awls) were consistent across the samples surveyed. Ribs from medium to large mammals were preferred for lissoirs. Use-wear indicates use for scraping or polishing hides and very few were repaired after breakage. Awls were made on metapodials, with an epiphysis as a proximal end (handle) or on long bones which lack marrow cavities (ulnae, residual metapodials and fibulae) again with the articular surface used as a handle. Others were made on esquilles (splinters or fragments) of long bones, or from broken tools. In some cases, the tools were completely worked and a support could not be identified. At Castanet, awls were made on the residual metapodials (stylets) from horse, which are ideal ‘preforms;’ for awls. Awls at Brassempouy were also made on residual phalanges of reindeer and on an ulna (Tartar 2009:99). In contrast, at Gatzarria awls were made on fragments of longbones. Tool use and manufacture at the Grotte du Renne At the Grotte du Renne, the lowest Aurignacian level (level VII) is dominated by horse and reindeer, with chamois, red deer and megafauna also present. Carnivores include cave bear, hyena, wolf and a large felid (David 2002). Horse and reindeer supplied the majority of the supports for bone tools: awls were made on reindeer ulnae, metatarsals and metapodials, and horse stylets. Tools were also made on long bone fragments and on large and medium sized mammal ribs (Julien, et al. 2002). In the Châtelperronian levels, published data from Levels X and IX show a similar preference from reindeer and horse ulnae and auxiliary bones for awls, which were also manufactured on medium and large mammal longbones (d’Errico, et al. 2003:255) (Table 11.3). Châtelperronian Horse Residual Reindeer Hyena Carnivore Auringancian Indet Total Horse Reindeer Carnivore Total 2 0 0 0 0 2 2 0 0 2 Metapodial 2 3 0 0 0 5 1 4 0 5 Fibula 0 0 3 1 0 4 0 0 0 0 Radius 0 2 0 0 0 2 0 0 0 0 Tibia 0 1 0 0 0 1 0 0 0 0 Ulna 0 1 0 0 0 1 0 1 1 2 Indet 0 2 0 0 33 35 0 0 0 0 Total 4 9 3 1 33 50 3 5 1 9 metapodial Table 11.3: Awls and sources of tool supports from the Châtelperronian and Aurignacian levels at the Grotte du Renne. 298 299 Figure 11.1: Drawing showing the tool fragments and areas of polish, and their location on the left proximal tibia shaft. Carnivore fibulae were also utilized for awls. In addition to published data, I identified two fragments of proximal reindeer tibia that were possibly used as lissoirs from Level Xc (Figure 11.1, base image from Pales and Lambert 1971a: Plate 87). These are both made on the same portion of the element, which indicates a consistent pattern of manufacture and use. The proximal articular surface had been removed to expose the cancellous tissue. The cortical and cancellous bone exhibited a high polish consistent with use as a scraper, possibly a hide scraper. Both items have dry breaks on the distal end, suggesting that they were discarded after breakage. These two tools show deliberate 300 selection and use of one particular element for a particular purpose. This suggests that the tibia served as a support for hide scrapers. The redundancy in the tools indicates some form of consistency in selection of element and tool use. Other tools identified during the analysis included a possible ad hoc tool formed on the diaphysis of a reindeer humerus (artifact 61.A6(168)). This had been formed by flaking one end to a point to serve as an awl. Another possible tool from an identifiable element was a fragment of a reindeer ulna exhibiting heavy polish (artifact B6, no number). Three diaphysis fragments were also identified as tools. These were flaked, a manufacturing technique reported for Mousterian bone tools (Figure 11.2). Figure 11.2: Sketch of three scrapers made on unidentified mammal bone fragments 61.63.A6; 63.C9; A5. Actual size. All were made on the diaphysis of large mammals and all appear to have been used as side-scrapers with retouch along one or more edges (61.63.A6; 63.C9; A5). These items are currently under study by Michelle Julien, CNRS, as part of a larger report on the Châtelperronian levels of the Grotte du Renne. 301 With the exception of the flaked bone items, manufacturing techniques in the Châtelperronian and Aurignacian were the same. Both cultures made use of the natural morphology of the bones, which were then scraped or fractured to make tools (the latter case for the carnivore fibulae), or limb bones were fractured, leaving an epiphyseal end, and then shaped by scraping. Some elongated tools may have been split by grooving, and were polished in addition to shaping by scraping (d’Errico, et al. 2003; Julien, et al. 2002). There are some differences in how elongated blanks are treated – in the Châtelperronian there seems to be a more opportunistic use of bone fragments that resulted from the extraction of marrow, in contrast to the selection of reindeer metapodials for more standardized blanks in the Aurignacian. Châtelperronian awls are also more variable in morphology than the more standardized Aurignacian tools (d’Errico, et al. 2003:266). This may be a product of the intensive use and reworking of Châtelperronian awls at the Grotte du Renne, an issue not addressed in the published data. Tool raw material selection clearly reflects knowledge of the mechanical properties of bone. Awls and other elements for piercing or scraping would require both strength and elasticity both to pierce any fabric and also to resist the pressure applied by the tool user. There is a preference for non-marrow bearing long bones, a pattern also seen in the Aurignacian, as discussed by Tartar. This pattern of use is also consistent with ethnographic examples discussed earlier. The occupants of the Grotte du Renne, both Neanderthals and modern humans, were choosing supports deliberately to make use of their mechanical properties. Tool manufacture at Abri Cellier At Abri Cellier, bone or worked bone fragments were present in the Upper and Lower Levels. Only three worked bone fragments were present in the Upper Level, all made from fragments of long bone of a large or medium-sized mammal. This is too small 302 a sample to make any definitive statements, but it seems that the occupant of this level were following a similar practice seen at Gatzarria (within a similar environment) of using available by-products of bone breaking for marrow. Bone tools or worked bone fragments form 10% of the Upper Level worked assemblage, and 38% of the Lower Level assemblage. The remaining tools were antler points and worked teeth, which have been described by White and Knecht (1992). Table 11.4, below, summarizes all the bone tools recovered to date from Abri Cellier using both White and Knecht’s 1992 published data and data from this research project. Bone tools from the lower level include lissoirs, poinçons/awls and a range of tools on pieces of large and medium mammal cortical bone, also described by White and Knecht (1992). The choice of supports from the Lower Level suggests that both the mechanical properties of the bone and the size and shape of the bone were factors in selection. The identifiable bones chosen are “preforms” and require minor modification. Further, they are comfortable to handle. Curation and resharpening indicate that these were used for some length of time. The number of tools relative to the number of animals or elements present again indicates that supports were obtained as part of quotidian subsistence practices: using reindeer and horse lower limb bones when these species are available and long bones when bovids and red deer dominate the fauna. Tools from the Lower Level of Abri Cellier were made on a diverse range of supports. Lissoirs were made on split ribs from large mammals, awls on reindeer ulnae, metatarsals and metcarpals; and two horse stylets and a metatarsal were worked. Bird long bones served to make fine awls and tubes. A number of tools were made on long bone fragments of large and medium sized mammals. It seems that, as at the Grotte du Renne, bones were selected largely for penetration without impact and their ability to withstand the application of pressure. As consistent choices are made (ribs for lissoirs, tibia for defleshers) at both sites, I would argue that both Neanderthals at the Grotte du 303 Renne, and modern humans at Abri Cellier were aware of the properties of elements and utilized them accordingly. Figure 11.3: Detail of shaped horn core from the Lower Level of Abri Cellier. Two unusual tools from Abri Cellier were a bovid horn-core modified into a chisel or pressure flaker, and a highly polished and modified wolf ulna (Figures 11.3 and 11.4). Neither type of bone blanks have been reported previously from an Aurignacian context (Goutas, pers. comm.). The horn core appears to be an ad-hoc tool, with minimal shaving on the distal end of the core to thin the tip to a flat, even edge. In contrast, the wolf ulna had clearly been curated and used for some time. The element was heavily worn, resulting in a rounded distal end of the bone exposing the medullar cavity. The inter-osseous crest was partially worn away towards the distal end, and the obverse side of the one was very smooth and flat. The dry break at the proximal end may indicate failure of the tool during use. The function of this item is unknown, and usewear analysis might be of assistance in defining its purpose. The rounded end and amount of wear suggest that this might be some form of digging stick rather than a hide-working tool. If 304 this is the case, it could provide evidence for foraging for roots and tubers as part of the subsistence activities at Abri Cellier Figure 11.4: Worked wolf ulna, from the lower level of Abri Cellier,showing rounded distal end and dry break. The range of supports from the Lower Level indicates that both the mechanical properties of bone and the size and shape of the bone were factors in selection. The identifiable bones required only minor modifications to be functional tools, although some bones were clearly curated and reshaped over time. Again, the occupants of Abri Cellier are choosing bone tools and bone fragments for their properties of resistance to stress and high elasticity, such as metapodials. The use of non-marrow bearing axial bones is consistent with the data from other Aurignacian sites describe by Tartar. As with the Châtelperronian material, the occupants of the site of Abri Cellier are using material that is already available in the form of carcass fragments, and not transporting particular elements to the site. Reindeer Red deer Bovid Equus Wolf Aves Indet Total Crania 0 0 1 0 0 0 0 1 Residual 0 0 0 1 0 0 0 1 Metpodial 6 0 0 1 0 0 0 1 Ulna 1 1 0 0 1 0 0 3 Residual 1 0 0 0 0 0 0 1 Rib/longbone 0 0 0 0 0 2 42 44 Total 8 1 1 2 1 2 42 51 metapodials phalanges Table 11.4: Sources of tool supports from the Upper and Lower Levels of the Aurignacian occupation at Abri Cellier, excluding antler. 305 306 Raw material for tools at Abri Cellier was obtained as part of quotidian subsistence practices, with herbivore prey species forming both the basis of the diet and the supports for bone tools. This may include the bird bones, from larger anseriforms. The carnivore used for a tool, the wolf, could be the by-product of hunting for furs. So far, I have discussed the mechanics of the bones chosen for use as tools. But other factors can influence the choice of raw material. The comfort of the bone used (i.e. the shape and size of the handle) may also be a factor in support choice (Griffitts 2007). In addition to tools, worked bone items at Abri Cellier included two bird longbones with grooved incisions (similar to tally-sticks), a worked bone “anthropomorph”, and pierced and grooved teeth utilized as pendants. These items are described in White and Knecht, as are the antler tools recovered from the site (White and Knecht 1992). Whilst the bone tools present indicate that the occupants of Abri Cellier, like the occupants of the Grotte du Renne, utilized available elements for tools, is there any evidence to indicate deliberate selection of deer to obtain antler for the spear points found in the Aurignacian levels at Abri Cellier? Antler supplies – logistical behavior or simple collection? No antler tools are present in level Xc of the Grotte du Renne, and the introduction of antler points is a major shift in armament technology in the Aurignacian. The excavators at Abri Cellier focused strongly on the collection of tools. There are 167 worked antler items in the collection, plus antler tines cut from the beam, beam fragments and an unworked shed antler base. Recent analysis of the points in the collection has identified two different chaînes opératoires (Luc Doyon, pers. comm.). This is part of an ongoing graduate research project undertaken by Luc Doyon, Université de Montréal into the curation and resharpening of Aurignacian points. The two methods of production are present in both levels of the site, and further analysis may elucidate the underlying factors that result in these patterns. 307 The more basic question is how was the antler obtained? Shed antler and antler from female or young adult males and mature males are present at the site. Antler was clearly transported to the site and processed in situ. Antler can simply be a by-product of carcass acquisition in late summer and fall (known as bois de massacre in the literature), brought to the site as part of the carcass. Shed antler could be collected from males in winter, and from females in the late winter/early spring. Obtaining cast male antler from the rutting grounds would suggest a logistical strategy similar to obtain this raw material. Collecting female antler would require proximity to the female herds in the late winter or early spring. Antler does not survive for a long time in the open. It is destroyed by natural taphonomic agents and is also sought out and consumed by cervids in a rather basic form of recycling. Does the impressive number of tools represent a logistical or foraging strategy? In other words, how many antlers are actually represented at the site, and were these the product of deliberate procurement by selective hunting, or of deliberate collecting at reindeer aggregation localities to collect cast antler, as documented in the ethnographic record of the Arctic (Stenton 1991)? Or do we see another by-product of subsistence behavior where cast antler or bois de massacre were acquired as part of an encounter strategy? The total weight of all antler in the collection is 1.91kg. The antler is very dry, so weight loss through desiccation had to be estimated. No data could be found on desiccation rates for reindeer antler. Red deer antler loses between 1% and 8% of its weight through desiccation in the first four weeks after loss (Currey, et al. 2009). Weight loss was estimated at 16% to factor in post-excavation desiccation. Based on calculations of antler by body weight (Hall 2005:105) and the average weight of prime age male antler in velvet for reindeer (Prichard, et al. 1999) the total weight of antler recovered represents a minimum of 2-4 male reindeer or 1-4 red deer antlers. In other words, one or two individual animals could have supplied all the material for the tools collected by the excavators at Abri Cellier. 308 Calculations by weight of antler may under-estimate the number of antlers used because, in general, only the beam of the antler is used for tool making and the tines and palmate sections are discarded. Experimental data for red deer antler working (Tejero, et al. 2012) found that inexperienced antler workers could produce predetermined antler blanks ranging in size from 8.6cm to 21.2cm using grooving or cutting and splitting techniques common in the Aurignacian. The beam was first cut into sections, then split longitudinally to produce blanks of cortical and cancellous tissue. In total, 13 blanks were obtained from two shed antler beams. Reindeer antler is more time consuming and somewhat harder to work, as the cortical bone is denser (Liolios 1999; Guthrie 1983), but predictably sized blanks can be produced using the Aurignacian technique of cutting the beam into sections and then splitting the segments (Averbouh 2000). There are 65 tools or antler items greater than 5cm in length in the collection. Based on experimental data, these tools represent a minimum of five antlers, which could be obtained from killing 3 adult males, or from a short foraging trip to an area used by cervids after the rut, to collect shed antler. The amount of antler required to produce the tools present in the Abri Cellier collection does not indicate that it was necessary for the Aurignacian occupants to utilize a logistical strategy to obtain this raw material. The sporadic occupation of the site (implied by the different lenses of occupation reported by Collie in the Lower Level) clearly indicates that more than the minimum number of antlers calculated here were in fact utilized. Nevertheless, the data suggest that a relatively small amount of antlers from a few individuals could supply an adequate number of blanks for tool production. The minimum number of individuals of reindeer or red deer in either level would be able to supply the antler required. Reindeer have an MNI of three from the Upper Level, and six from the Lower Level in the collection held at Cellier. Red deer have an MNI of four in the Upper Level and one in the Lower Level. Therefore there are seven large cervids with 309 antler suitable for tool manufacture in both levels, more than the minimum needed to supply the antler required for all tools excavated from the site. I would argue that antler was acquired as part of a generalized subsistence strategy, and was a by-product of standard subsistence practices. Shed antler was collected and transported to the site, but antler is workable while in the late stages of velvet and prior to shedding. This provides a long window of opportunity for acquisition, from late July (as the blood supply ceases) through the early winter for adult male reindeer. The movements of reindeer herds in the Pleistocene are not well understood, but the seasonal displacement of modern herds is the product of local ecological conditions, including population size, density and resource availability. Many models of reindeer exploitation take the Barren Ground herds as a model, assuming large annual movements of herds. The actual movements of reindeer is much more variable from year to year and the site of Abri Cellier would be well positioned for its occupants to exploit male troupes or mixed herds as they moved between summer and winter grounds. Red deer are not migratory, and, as with reindeer, and could also be monitored from the site and antler acquired during the late summer and early fall. Red deer males are relatively easy to locate during the rut, as part of the dominance display includes roaring or bugling to intimidate other males competing to form harems. Antler could therefore be obtained as part of an opportunistic foraging or encounter strategy and as a by-product of meat and hide acquisition. Conclusion The exploitation of animal bones as supports for tools in the Châtelperronian and Aurignacian did not require any major changes in the transportation of animal parts to a site. The raw material for bone tools was a by-product of subsistence behavior by Neanderthals and modern humans. There is a clear preference for the use of cortical bone for tools, either in the form of fragments of long bones, or use of non-marrow containing 310 appendicular elements such as the ulna or residual metapodials or phalanges. Tools are shaped by either modification of the bone, or by splitting and shaping the element by scraping and/or polishing. Both modern humans and Neanderthals are using elements that have been demonstrated archaeologically to be favored for penetration without impact and the majority of tools appear to be associated with processing hides, either as defleshers (lissoirs) or manufacturing items made on soft materials (awls). The major difference between the two sites is the adoption of antler for projectile points and other tools. These patterns are consistent with the data collected by Tartar and other researchers (Averbouh 2000; Liolios 2003; Peterkin 2001; Tartar 2009; Tartar, et al. 2006) for the Early Upper Palaeolithic. The addition of antler to hunting technology represents an addition to the existing exploitation of osseous materials, and the extent to which this provides an adaptive advantage to modern humans in the Aurignacian remains an area of debate. Bone tools for hide fabrication continue in use across the transition from the Châtelperronian to the Aurignacian. There are some differences in the shape and regularity of the awls manufactured in the Châtelperronian and Aurignacian, which may relate to intensity of use, or of a more established pattern of tool manufacture and use. Is this a product of a longer-established industry, where forms have become more standardized, or does it reflect a higher degree of investment in bone tools? Further research may answer this question. 311 CHAPTER 12: CONCLUSIONS AND SUGGESTIONS FOR FURTHER RESEARCH Introduction This thesis has examined the faunal remains from Level Xc, the lowest Châtelperronian level of the Grotte du Renne, Arcy-sur-Cure, and from the Aurignacian I and II levels (Upper and Lower Levels) of Abri Cellier, Dordogne. This analysis was undertaken with reference to the use of animals as a source of raw material within the broader subsistence strategies of Neanderthals and modern humans. Before presenting the data and results of the faunal analysis, this thesis examined the current archaeological record to discuss the degrees of differences and similarities between Neanderthals and Modern Humans in the Early Upper Palaeolithic, particularly the Châtelperronian and Aurignacian cultures of western Europe. There is considerable debate regarding the interaction of the two populations of hominins. The appearance of the Châtelperronian has been interpreted as either the results of contact between the two populations, where Neanderthals adopted new osseous and lithic technologies, or that the Neanderthals developed new osseous technologies as a response to their socio-ecological environment. Little attempt has been made by palaeoanthropologists to model the circumstances under which acculturation of either group would take place. At present, the two populations may have been contemporaneous in terms of occupation of parts of eastern, central and western Europe, but there is no clear indication of any prolonged episode of culture contact. The appearance of bone tools, associated with the manufacture of containers ranging in size from bags to windbreaks or tents, and including clothing, may relate to a response by both groups to an increasingly unstable and intemperate climate at the beginning of the last glacial maximum. Even in interglacial periods in Europe, Neanderthals and modern humans required some form of clothing in the temperate and higher latitudes of the continent. 312 Neanderthals and contemporary human groups show considerable similarities in their lithic technologies and subsistence strategies. Both populations were capable of manufacturing both flake and blade based lithic technologies. Both groups were foragers, and effective predators encountering and taking prime age individuals from local herbivore populations. Both groups show similar patterns of mortality and injury level. There are no data concerning diet or life history to explain why Neanderthals become extinct. There may be differences in the social networks occupied by Neanderthals and modern humans, indicated by different levels of symbolic communication and the transport of non-local lithic raw material. Lithic raw material data suggest that Neanderthals were more closely tied to small, local territories, and had weaker chains of communication with other groups in their larger regions. Modern humans, in contrast, show a higher degree of long distance transportation of lithic raw material and other items that indicate larger communication networks. This may be an advantage when both populations had to respond to rapid environment shifts resulting from climatic instability. Neanderthals may have been less entrained in larger social networks, but their ontogeny and comparison with modern hunter gather groups suggests that Neanderthal bands or groups, of whatever size, had a similar social structure to modern humans. The relatively slow growth of Neanderthal children mirrors the growth and development of modern human children. Modern humans do not acquire full competence in hunting or foraging skills until after the age of first reproduction. A certain degree of group provisioning or alloparenting is necessary to provide adequate nutrition for both children and the offspring of young parents. There are few major differences between Neanderthals and modern humans in the Early Upper Palaeolithic. The adoption of osseous technology has been argued as a major technological development, particularly when associated with the production of personal ornaments. This research project has focused on the acquisition of osseous raw material as part of the overall subsistence strategy. What, if any, differences are there in the 313 acquisition practices of Neanderthals at the Grotte du Renne or modern humans at Abri Cellier? Three null hypotheses were proposed to test for similarities and differences between the two hominins in terms of subsistence behavior. Testing the null hypotheses Null Hypothesis 1: The faunal remains at Abri Cellier and Level Xc, Grotte du Renne are solely the product of hominin behavior. Both assemblages are largely the product of human hunting behavior. Both assemblages are dominated by prime age herbivores. Butchery marks and fresh breaks on long bones confirm the role of hominins in carcass processing. Carnivores are present at both locations, but some of the carnivore assemblage, particularly in level Xc of the Grotte du Renne is the result of hominin behavior. It appears that Neanderthals used carnivores such as hyenas and wolf for pelts. Adult cave bear lower limbs and feet also show evidence of hide removal. The presence of cave bear at the Grotte du Renne is more equivocal, as this location clearly served as a hibernation and birthing den. However, cutmarks on bear phalanges attest to a role for Neanderthals in the incorporation of this omnivore into the faunal assemblage. It also suggests use of the phalanges as decorative items, but this cannot be proven. The amount of carnivore damage is low at both sites, indicating a minor role for carnivores as agents of accumulation or destruction of both assemblages. Puncture marks on four phalanges in the Abri Cellier assemblage and removal of epiphyses three long bones attest to carnivore destruction by gnawing. Carnivore damage at the Grotte-duRenne is also low. There are few digested bones in the assemblage. Edge damage on many of the gnawed bone fragments may be associated with the teething processes of cave bear cubs. As there is little evidence of carnivore damage or major occupations by carnivores at either site, it is not possible to reject the Null Hypothesis These assemblages are the product of human behavior. 314 Null Hypothesis 2: There is no difference in subsistence behavior in terms of exploitation of animals for food between the Châtelperronian and Aurignacian. Both cultures followed the same subsistence practices over time and space. Any apparent differences will be explained by environmental change, not differences in social organization. There was little evidence for non-hominin accumulation of the faunal material. It was determined that both levels of Abri Cellier are the result of foraging for resources within the local environment. The faunal assemblage in the Lower Level of Abri Cellier was dominated by reindeer both in terms of NISP and MNI. The reindeer assemblage included prime age and juvenile reindeer, as indicated by tooth wear patterns. Other herbivores in the Lower Level were horse, saiga and a large bovid. The percentage of reindeer (both in terms of MNI and NISP) dropped in the Upper Level. The Upper Level assemblage was dominated by red deer and bovids, with horse and saiga also present in lower numbers. Carcass transportation patterns suggests that smaller herbivores (reindeer and saiga) were transported as carcasses to the site for processing, while larger herbivores, such as horse and bovids may have been butchered near the site and only selected elements transported to the site. The change in species present indicates an adaptation to an amelioration of the climate between the Lower and Upper Level occupations. There are no data to suggest that a logistical hunting strategy, focused on a single species, was practiced by the occupants of either level of Abri Cellier. Similarly, at the Grotte du Renne, Level Xc, reindeer dominates the assemblage but horse, bovids (bison and possibly auroch) and red deer are also present. Neanderthals were practicing an encounter strategy similar to that of modern humans in the Aurignacian. Reindeer carcasses were processed at the site, but horse, bovid and red deer carcasses were butchered at the kill site and only elements requiring further processing were transported to the site. The high proportion of fresh breaks and the relatively small fragments of marrow, fat and grease rich long bones indicate a high investment in 315 harvesting the fat and marrow from the carcasses and carcass fragments. This may be more intense than the processing of bones for marrow and fat at Abri Cellier, but the collection practices of the excavators at the latter site make any statements about bone processing tentative. The deliberate selection of relatively large identifiable bone fragments by the excavators at Abri Cellier resulted in severe underrepresentation of the shaft fragments that would indicate the processing of bones for marrow, fat or grease. In summary there is no difference in the subsistence behaviors of the two hominins in terms of prey acquisition. All hominin occupants of Abri Cellier and the Grotte du Renne, Level Xc practiced an opportunistic encounter or foraging strategy, taking game on an encounter basis. In the cooler environments of Level Xc and the Lower Level of Abri Cellier, hominins hunted locally available species adapted to cold, open environments predominate. In the Upper Level of Abri Cellier, modern humans took species which indicate a milder climate in greater numbers. Null Hypothesis 2 cannot be rejected. There is no difference in terms of exploitation of animals for subsistence at the two sites. Null Hypothesis 3: There is no difference in the selection and use of bone for tools between the Châtelperronian and Aurignacian. Selection for raw materials will be the same and both cultures will use tools for a similar suite of manufacturing and subsistence behaviors. Null Hypothesis 3 is partially rejected because antler points and tools only occur in the Aurignacian levels of Abri Cellier. No antler tools occur in Level Xc of the Grotte du Renne. Both the Aurignacian and Châtelperronian assemblages produced bone tools: awls and hide scrapers, items associated with hide processing and container manufacture. From the antler fragments it is clear that antler was worked at Abri Cellier. Little identifiable bone-working debris was present. Unfortunately bone-working debris does not differ to any great degree from the breakage patterns found when fresh bone is processed for marrow or fat. 316 not differ to any great degree from the breakage patterns found when fresh bone is processed for marrow or fat. Antler points only occurred in the Aurignacian assemblage from Abri Cellier. This reflects a difference in hunting strategies and armature. It should be remembered that antler points appear after bone tools in the Aurignacian and reflect a different set of behaviors. Bone tools in the Aurignacian are used for manufacturing purposes, not hunting. There is a greater range of tool blanks in the Aurignacian from Abri Cellier – awls, on bird bone, on reindeer metapodials, on reindeer residual phalange and on horse residual metapodials, bovid horn cores and a wolf ulna. Small points (hameçons) and decorative items are also present. This coincides with Tartar’s arguments for a greater investment in tools or range of processes by modern humans in the Aurignacian. However, the Châtelperronian assemblage from the Grotte du Renne is rich in awls, and also contained two hide scrapers on reindeer tibiae and three small bone side scrapers made on long-bone fragments. In fact there are more bone tools in the Châtelperronian levels of the Grotte du Renne than in the Aurignacian. This might suggest a greater investment in bone tools by Neanderthals at this particular site. Both groups show a preference for non-marrow bearing long bones, or use long bone shaft fragments to make tools. Both groups are aware of the mechanical properties of the bones utilized for raw material and select items that have a certain degree of plasticity to withstand the loading placed on them as part of normal use. There is no evidence of deliberate transportation of long bones for use specifically as raw material for tools. The pattern of using available bones for raw material recurs at other Aurignacian sites, for example Castanet Nord, Castanet Sud, Brassempouy and Gatzarria. Neanderthals at the Grotte du Renne and modern humans at Abri Cellier were using raw material that was a by-product of meat and marrow procurement. The use of bone tools infers the need to manufacture items from fragile material such as hide or intestine. These items were containers and clothing. Both groups required 317 effective clothing to survive in a cold climate. The simple bone tool kit of both groups indicates that neither group was using needles to produce fully tailored clothing, as is seen in the Later Upper Palaeolithic. Based on the osseous assemblage we cannot assume that Neanderthal clothing was any less effective than the clothing worn by modern humans. Both groups of hominins clearly invested time in hide processing and the manufacture of clothing. There are differences in the use of antler, but it is not clear how or if the adoption of antler for armaments would provide a major adaptive advantage to modern humans. A shed antler base at Abri Cellier indicated that shed antler was collected and transported to the site for manufacture into points or other items. Calculations of the amount of raw material represented by the antler tools at Abri Cellier suggest that the occupants of the site could meet the requirements for “tooling-up” through their standard subsistence behavior by foraging for shed antler as part of a collector strategy, or by acquiring antler as part of their basic subsistence strategy. Antler is workable while in the late stages of velvet, or after the velvet is shed. As this occurs in July for reindeer, and the antler is not shed until late fall or early winter (for males) there are approximately 5 months of the year to obtain antler as a by-product of hunting. The evidence does not suggest that a logistical strategy was necessary to obtain adequate supplies of raw material. Conclusion and further research There is no apparent difference in the use of animals as raw material for bone tools by Neanderthals in Level Xc of the Grotte du Renne, Arcy-sur-Cure, or the modern humans who occupied the Upper and Lower Levels of Abri Cellier. Both groups used the available animals as sources of raw material, and there is no evidence for transportation of any post-cranial elements specifically for the purpose of bone tool manufacture. There is a difference in the use of antler, which Neanderthals did not use for tool-making 318 purposes, but the amount of antler required to produce the antler points and tools present in Abri Cellier indicates that this could be acquired as part of a general foraging strategy. A number of avenues for further research are open. One question is how early in the Mousterian did Neanderthals and modern humans start to use bone tools? And where? It is entirely possible that bone tool use was an innovation that occurred multiple times in multiple regions during the Pleistocene. Further research examining long bone fragments, for example from the Mousterian levels at the Grotte du Renne, or the nearby Grotte du Bison, may aid in solving this problem. The bone tools from Level Xc and manufacturing techniques suggest that this new toolkit developed out of an earlier, less formal industry. By examination of entire faunal assemblages, including all the bone shaft fragments, it may be possible to identify less formal, more ad-hoc tools that were replaced by the more formal Châtelperronian tools. It should be remembered that the apparent absence of bone tools from the European Mousterian may be a product of the simple fact that the search for them only began recently. This relates back to assumptions about Neanderthals’ ability to adapt and innovate, or lack thereof. The adoption of formal bone tools may not have occurred throughout the entire Neanderthal range. If this novel technology is associated with the need for more effective clothing, we should expect to see this develop most strongly in the northern, or less temperate, parts of their range. The adoption of clothing and containers, expressed through the appearance of bone tools could also explain the continued existence of Neanderthals in the more northerly parts of their range in France until relatively late. They now had the extrasomatic means of adaptation to remain in cooler environments. It would be fruitful to model how small bands would react to unstable environments in terms of stochastic fluctuations in temperature and available resources. This could be contrasted with the Aurignacian groups, who had more developed social networks, which could provide a major adaptive advantage in the long term as a means of buffering unpredictable subsistence resources. 319 A major issue is the nature of contact and information flow between Neanderthals and modern humans. To better understand the potential for interaction between the two populations, palaeoanthropologists must begin to build robust models for predicting when and how transfers of knowledge occurred. Neanderthals are the indigenous population in Europe, and modern humans represent a group that had to adapt to new environments. Under these circumstances it would be expected that the non-native population would be as likely to adopt or adapt local subsistence practices as the native population would be to adopt new technology. Palaeoanthropologists also need to question how much emphasis we place on apparent differences in behavior between the two populations. The development of new analytical techniques, such as phytolith analysis; or reexamination of subsets of the overall faunal assemblage, for example, bird bones in Mousterian assemblages, is producing a more nuanced picture of Neanderthal behavior in terms of spatial organization, behavioral organization and possible symbolic behavior. Perhaps our major goal as palaeoanthropologists should be to try to understand how the small differences in a suite of behaviors may or may not have contributed to the demise of the Neanderthals. Extinction is a complex process. It is easy to see the end result but much harder to understand the underlying causes. 320 APPENDIX: CUTMARK LOCATIONS ON ELEMENTS FROM LEVEL RXC, GROTTE DU RENNE, AND ABRI CELLIER Figure A.1: Level Xc, Grotte du Renne. Cutmark locations on reindeer skull and vertebrae. Based on Pales and Garcia (1981a): Plate 26 and Plate 86. 321 Figure A.2: Level Xc, Grotte du Renne. Cutmark locations and impact marks on reindeer right humeri. Based on Pales and Lambert (1971a): Plate 2. 322 Figure A.3: Level Xc, Grotte du Renne. Cutmark locations and impact marks on reindeer left humeri. Based on Pales and Lambert (1971a): Plate 2. 323 Figure A.4: Level Xc, Grotte du Renne. Cutmark locations and impact marks on reindeer right radii and ulnae. Based on Pales and Lambert (1971a): Plate 6. 324 Figure A.5: Level Xc, Grotte du Renne. Cutmark locations and impact marks on reindeer left radii and ulnae. Based on Pales and Lambert (1971a): Plate 6. 325 Figure A.6: Level Xc, Grotte du Renne. Cutmark locations and impact marks on reindeer carpals (right) and right metacarpals (left). Based on Pales and Lambert (1971a): Plates 9 and 16. 326 Figure A.7: Level Xc, Grotte du Renne. Cutmark locations and impact marks on reindeer indetrminate metacarpals. Based on Pales and Lambert (1971a): Plate 16. 327 Figure A.8: Level Xc, Grotte du Renne. Cutmark locations on reindeer right femora (left) and indeterminate femora (right). Based on Pales and Lambert (1971a): Plate 21. 328 Figure A.9: Level Xc, Grotte du Renne. Cutmark locations and impact marks on reindeer right tibia. Based on Pales and Lambert (1971a): Plate 27. 329 Figure A.10: Level Xc, Grotte du Renne. Cutmark locations and impact marks on reindeer left tibia. Based on Pales and Lambert (1971a): Plate 27. 330 Figure A.11: Level Xc, Grotte du Renne. Cutmark locations and impact marks on reindeer tarsals. Based on Pales and Lambert (1971a):Plate 31 331 Figure A.12: Level Xc, Grotte du Renne. Cutmark locations on reindeer right metatarsals. Based on Pales and Lambert (1971a): Plate 31. 332 Figure A.13: Level Xc, Grotte du Renne. Cutmark locations on reindeer left metatarsals. Based on Pales and Lambert (1971a):Plates 31. 333 Figure A.14: Level Xc, Grotte du Renne. Cutmark and impact locations on reindeer indeterminate metatarsals. Based on Pales and Lambert (1971a):Plate 31. 334 Figure A.15: Level Xc, Grotte du Renne. Cutmark locations and impact fractures on reindeer phalanges. Based on Pales and Lambert (1971a): Plate38. 335 Figure A.16: Level Xc, Grotte du Renne. Cutmark locations on horse right humerus. Based on Barone (1976): 262. 336 Figure A.17: Level Xc, Grotte du Renne. Cutmark locations on horse indeterminate radius. Based on Barone (1976): 278. 337 Figure A.18: Level Xc, Grotte du Renne. Impact locations on horse right tibia. Based on Barone (1976): 386. 338 Figure A.19: Level Xc, Grotte du Renne. Impact locations on horse left tibia. Based on Barone (1976): 387. 339 Figure A.20: Level Xc, Grotte du Renne. Cutmark and impact locations on horse indeterminate metapodials. Based on Barone (1976): 423. 340 Figure A.21: Level Xc, Grotte du Renne. Cutmarks and impact locations on bear indeterminate humeri. Based on Pales and Lambert (1971b): Plate 3. 341 Figure A.22: Level Xc, Grotte du Renne. Cutmark and impact locations on bear indeterminate femora. Based on Pales and Lambert (1971b): Plate 26. 342 Figure A.23: Level Xc, Grotte du Renne. Cutmark and impact locations on bear indeterminate tibia. Based on Pales and Lambert (1971b): Plate 32. 343 Figure A.24: Level Xc, Grotte du Renne. Cutmark locations on bear phalanges. Based on Pales and Lambert (1971b): Plate 37. 344 Figure A.25: Level Xc, Grotte du Renne. Cutmark and impact locations on hyena radius (right) and possible cutmarks on hyena fibula (left). Based on Pales and Lambert (1971b): Plate 9 and Plate 33. 345 Figure A.26: Level Xc, Grotte du Renne. Cutmark and impact locations on hyena phalanges and tarsals. Based on Pales and Lambert (1971b): Plate 38 346 Figure A.27: Level Xc, Grotte du Renne. Cutmark location on felid third phalange. Based on Pales and Lambert (1971b): Plate 40 347 Figure A.28: Abri Cellier. Cutmark locations on reindeer atlas. Based on Pales and Garcia, (1981a): Plate 6. 348 Figure A.29: Abri Cellier. Cutmark locations on reindeer humeri. Based on Pales and Lambert, (1971a): Plate 2. 349 Figure A.30: Abri Cellier. Cutmark locations on reindeer right radius and metacarpal. Based on Pales and Lambert, (1971a): Plate 6 and Plate 14. 350 Figure A.31: Abri Cellier. Cutmark locations on reindeer right femur and tibia Based on Pales and Lambert, (1971a): Plate 21 and Plate 27 351 Figure A.32: Abri Cellier. Cutmark locations on reindeer right metatarsal. Based on Pales and Lambert, (1971a): Plate 36 352 REFERENCES CITED Adler, D. S., G. Bar-Oz, A. Belfer-Cohen and O. Bar-Yosef 2006 Ahead of the game: Middle and Upper Palaeolithic hunting behaviors in the southern Caucasus. Current Anthropology 47(1):89-118. Adler, D.S., O. Bar-Yosef, A. Belfer-Cohen, N. Tushabramishvili, E. Boaretto, N. Mercier, H. Valladas, and W. J. Rink 2008 Dating the demise: Neandertal extinction and the establishment of modern humans in the southern Caucasus. Journal of Human Evolution 55(5): 817-833 Aiello, L.C. and J.C.K. Wells 2002 Energetics and the evolution of the genus Homo. Annual Review of Anthropology 31(1): 323-338. Aiello, L. C. and P. Wheeler 1995 The expensive tissue hypothesis: the brain and digestive system in human and primate evolution. Current Anthropology 36(2):199-221. Altuna, J. 1989 Subsistence d'origine animale pendant la Moustérien dans la région Cantabrique. In L'Homme de Néandertal 6: La Subsistence, edited by M. Otte, pp. 31-43. ERAUL, Liege. Andrews, P. 1980 Owls, Caves and Fossils. University of Chicago Press, Chicago. Averbouh, A. 2000a Technologie de la Matière Osseuse Travaillé et Implications Paléthnologiques: Vol 1.Unpublished PhD Dissertation, Department of Préhistoire-Ethnologie-Anthropologie, Paris I-Pantheon-Sorbonne, Paris. 2000b Technologie de la Matière Osseuse Travaillé et Implications Paléthnologiques: Vol 2. Unpublished PhD Dissertation. Department of Préhistoire-Ethnologie-Anthropologie, Paris I-Pantheon-Sorbonne, Paris. Averbouh, A. and N. Provenzano 1998 Propositions pour une terminololgie du travail préhistorique des matières osseuse: I - les techniques. Préhistoire Anthropologie Mediterrannée 7-8:5-25. Baffier, D. and M. Girard 1998 Les Cavernes d'Arcy-sur-Cure. Maison des Roches, Paris. Baffier, D. and M. Julien 1990 L'outillage en os des niveaux chatelperronens d'Arcy-sur-Cure (Yonne). In Paléolithique Moyen Recent et Paléolithique Supérieur Ancien en Europe, edited by C. Farizy, pp. 329-334. Musee de Prehistoire d'Ile de France, Nemours. Bailey, G, and N. Flemming 2008 Archaeology of the continental shelf: Marine resources, submerged landscapes and underwater archaeology. Quaternary Science Reviews: 27 (23-24): 2153-2165. 353 Bailey, S. E. and J.-J. Hublin 2008 Did neanderthals make the Châtelperronian at Arcy-sur-Cure? In Neanderthals Revisited: New Approaches and Perspectives, edited by K. Harvati, and T. Harrison, pp. 191-210. Springer, New York. Bar-Oz, G., D. S. Adler, A. Vekua, T. Meshveltani, N. Tushabramishivili, A. Belfer Cohen and O. Bar-Yosef 2004 Faunal exploitation patterns along the southern slopes of the Caucasus during the Late Middle and Early Upper Palaeolithic. In Colonization, Migration and Marginal Areas: A Zooarchaeological Approach, edited by M. Mondini, S. Munoz, and S. Wickler, pp. 46-54. Proceedings of the 9th Conference of the International Council of Zooarchaeololgy, Durham, August 2002. Oxbow Books, Oxfod. Bar Yosef, O. 2006 Neanderthals and modern humans: a different perspective. In When Neanderthals and Modern Humans Met: A Different Perspective, edited by N. J. Conard, pp. 467-482. Kerns, Tubingen. 2007 The dispersal of modern humans in Eurasia: a cultural interpretation. In Rethinking the Human Revolution : New Behavioural and Biological Perspectives on the Origin and Dispersal of Modern Humans, edited by P. Mellars, K. Boyle, O. Bar-Yosef, and C. Stringer, pp. 207-217. McDonald Institute Monographs. McDonald Institute for Archaeological Research, Cambridge. Bar Yosef, O. and S. L. Kuhn 1999 The big deal about blades: laminar technologies and human evolution. American Anthropologist 101(2):322-338. Bar Yosef, O. and J. Zilhão 2006 Preface. In Towards a Definition of the Aurignacian: Proceedings of the Symposium Held in Lisbon, Portugal, June 25-30, 2002; edited by O. Bar Yosef and J. Zilhão, pp 7-8. Instituto Português de Arqueologia and the American School of Prehistoric Research, Peabody Museum, Harvard University, Lisbon Barone, R. 1976 Anatomie Comparée des Mammifères Domestiques. Second Edition.Vigot Frères, Editeurs, Paris. Bartram, L. E. and C. W. Marean 1999 Explaining the "Klasies Pattern": Kua ethnoarchaeology, the Die Kelders Middle Stone Age archaeofauna, long bone fragmentation and carnivore ravaging. Journal of Archaeological Science 26:9-29. Balter, V., and L. Simon 2006 Diet and behavior of the Saint-Césaire Neanderthal inferred from biogeochemical data inversion. Journal of Human Evolution 51(4): 329-338 Behrensmeyer, A. K. 1978 Taphonomic and ecological information from bone weathering. Paleobiology 4(2):150-162. 354 Behrensmeyer, A.K. Time resolution in fluvial vertebrate assemblages. Paleobiology 8(3): 211-227. Behrensmeyer, A. K. and D. E. Boaz 1980 The recent bones of Amboseli Park, Kenya in relation to East African paleoecology. In Fossils in the Making edited by A. K. Behrensmeyer and A. P. Hill, pp. 72-93. University of Chicago Press, Chicago. Belfer-Cohen, A. and A. N. Goring-Morris 2007 From the beginning: Levantine Upper Palaeolithic cultural change and continuity. In Rethinking the Human Revolution : New Behavioural and Biological Perspectives on the Origin and Dispersal of Modern Humans, edited by P.Mellars, K. Boyle, O. Bar-Yosef, and C. Stringer, pp. 199-205. McDonald Institute for Archaeological Research, Cambridge. Berger, T. D. and E. Trinkaus 1995 Patterns of trauma among the Neandertals. Journal of Archaeological Science 22(6):841-857. Berry, J. W. 2003 Conceptual approaches to acculturation. In Acculturation: Advances in Method and Theory, edited by P. B. O. K. Chun, and G. Marin, pp. 17-37. American Psychology Association, Washington DC. Bettinger, R. L. 1991 Hunter-Gatherers: Archaeological and Evolutionary Theory. Plenum Press, New York. Beyries, S. 2008 Modélisation du travail du cuir en ethnologie: proposition d'un système ouvert à l'archeologie. Anthropozoologia 43(1):9-43. Binford, L. R. 1978 Nunamuit Ethnoarchaeology. Academic Press, New York. 1981 Bones: Ancient Man and Modern Myths. Academic Press, New York. 1984 Butchering, sharing and the archaeological record. Journal of Anthropological Archaeology 3:235-257. 1988 Etude taphonomique sur des restes fauniques de la Grotte Vaufrey. In La Grotte Vaufrey, edited by J.-P. Rigaud, pp. 535-568. MSPF, Paris. 2001 Constructing Frames of Reference. University of California Press, Berkeley and Los Angeles. Binford, L. R. and J. B. Bartram 1977 Bone frequencies and attritional processes. In For Theory Building in Archaeology, edited by L. R. Binford, pp. 77-153. Academic Press, New York. Binford, L. R. and S. Binford 1966 A preliminary analysis of functional variability in the Mousterian of Levallois facies. American Anthropologist 68:238-295. 355 Bird, D. W. and R. Bliege Bird 2001 The ethnoarchaeology of juvenile foragers: shell fishing strategies among Meriam children. Journal of Anthropological Archaeology 19:461-476. 2005 Martu children's hunting strategies in the Western Desert, Australia. In Hunter-Gatherer Childhoods, edited by B. S. Hewlett and M. E. Lamb, pp. 129146. Aldine, New Brunswick, NJ. Bird, D. W., R. Bliege Bird and B. F. Codding 2009 In pursuit of mobile prey: Martu hunting strategies and archaeofaunal interpretation. American Antiquity 74(1):2-29. Bisson, M. S. 2000 Nineteenth century tools for twenty-first century archaeology? Why the Middle Paleolithic typology of François Bordes must be replaced. Journal of Archaeological Method and Theory 7(1):1-48. Bliege Bird, R. and Douglas W. Bird 2008 Why women hunt: risk and contemporary foraging in a Western Desert Aboriginal Community. Current Anthropology 49(4):655-693. Blockley, S. P. E., C. Bronk Ramsey and T. F. G. Higham 2008 The Middle to Upper Paleolithic transition: dating, stratigraphy, and isochronous markers. Journal of Human Evolution 55(5):764-771. Blumenschine, R. J., C. W. Marean and S. D. Capaldo 1996 Blind tests of inter-analyst correspondence and accuracy in the identification of cut marks, percussion marks and carnivore tooth marks on bone surfaces. Journal of Archaeological Science 23:493-507. Blurton Jones, N., K. Hawkes and P. Draper 1994 Differences between Hadza and !Kung children's work: original affluence or practical reason? In Key Issues in Hunter-Gatherer Research, edited by E. S. Burch, Jr., and L. J. Ellanna, pp. 189-215. Berg, Oxford, Providence. Blurton Jones, N., K. Hawkes and J. F. O'Connell 1996 The global process and local ecology: how should we explain differences between the Hadza and the !Kung. In Cultural Diversity Among Twentieth Century Foragers: An African Perspective, edited by S. Kent, pp. 159-187. Cambridge University Press, Cambridge. Bochenski, Z. 2005 Owls, diurnal raptors and humans: signatures on avian bones. In From Biosphere to Lithosphere: New Studies in Vertebrate Taphonomy, edited by T. O'Conner, pp. 31-45. Oxbow Books, Oxford. Bocherens, H. 2001 New isotopic evidence for dietary habits of Neandertals from Belgium. Journal of Human Evolution 40(6): 497-505. 2011 Diet and ecology of Neanderthals: implications from C and N isotopes. . In Neanderthal Lifeways, Subsistence and Technology: One Hundred and Fifty Years of Neanderthal Study., edited by N. J. Conard and J. Richter, pp. 73-85. Springer, New York. 356 Bocherens, H., and D. Drucker 2004 Dietary competition between Neanderthals and modern humans: insights from stable isotopes. In When Neanderthals and Modern Humans Met: A Different Perspective, edited by N. J. Conard, pp. 129-143. Kerns, Tubingen. Bocherens, H., D. Drucker, D, Bonjean, A. Bridault, N.J. Conard, C. Cupillard, M. Germonpré, M. Höneisen, S. Münzel, H.Napierala, M. Patou-Mathis, E. Stephan, H-P. Uerpmann and R. Ziegler 2011 Isotopic evidence for dietary ecology of cave lion (Panthera spelaea) in North-Western Europe: Prey choice, competition and implications for extinction. Quaternary International 245(2): 249-261 Bock, J. 2005 What makes a competent adult forager? In Hunter-Gatherer Childhoods, edited by B. S. Hewlett and M. E. Lamb, pp. 109-128. Aldine, New Brunswick, NJ. Bodenhorn, B. 1990 I am not the Great Hunter, my wife is: Inupiat and anthropological models of gender. Etudes Inuit/Inuit Studies 14(1/2):55-70. Bodu, P., 1990 L’application de la méthode des remonates à l’étude du matière lithique des premières niveaux châtelperronian à Arcy-sur-Cure. In Paléolithique Moyen Recent et Paléolithique Supérieur Ancien en Europe, edited by C. Farizy, pp 309312. Musée de Préhistoire d’Ile de France, Nemours. Bordes, F. 1972 A Tale of Two Caves. Harper & Row, New York. Bordes, J.-G. 2003 Lithic taphonomy of the Châtelperronian/Aurignacian interstratifications in Roc de Combe and Le Piage (Lot, France). In The Chronology of the Aurignacian and of the Transitional Technocomplexes: Dating, Stratigraphies, Cultural Implications. proceedings of Symposium 6.1 of the XIVth Congress of the UISPP, Liege, 2001, edited by J. Zilhão and F. d'Errico, pp. 223-244. Trabalhos de Arqueologia. Instituto Português de Arqueologia., Lisbon. Boule, M. 1911-1913 L'homme fossile de La Chapelle aux Saints. Annales de Paleontologie:1911, 1111-1172, 1912: 1921-1192, 1913: 1911-1967. Boyle, K. V. 1990 Upper Palaeolithic Faunas from South-West Europe: A Zoogeographic Perspective. British Archaeological Reports International Series 557. British Archaeological Reports, Oxford. Bradtmöller, M. A.Pastoors, B.Weninger, and G. C.Weniger 2012 The repeated replacement model – Rapid climate change and population dynamics in Late Pleistocene Europe. Quaternary International 247:38-49. 357 Brain, C. K. 1980 Some criteria for the recognition of bone-collecting agencies in African caves. In Fossils in the Making, edited by A. K. Behrensmeyer and A. P. Hill, pp. 108-130. University of Chicago Press, Chicago. 1981 The Hunters or the Hunted? An Introduction to African Cave Taphonomy. University of Chicago Press, Chicago. 2004 Fifty years of fun with fossils: some cave taphonomy-related ideas and concepts that emerged between 1953 and 2003. In Breathing Life Into Fossils: Taphonomic Studies in Honor of C.K. Brain, edited by T. R. Pickering, K. Schick, and N. Toth, pp. 1-24. Stone Age Institute Press, Gosport, Indiana. Brandišauskas, D. 2010 Hide tanning and its use in the taiga: the case of the Orochen-Evenki reindeer herders and hunters of Zabailkalye (East Siberial). Journal of Ethnology and Folklorisitics 4(2):97-114. Breuil, H. and O. G. S. Crawford 1938 The use of bone implements in the Old Palaeolithic. Proceedings of the Prehistoric Society 12(45):56-67. Bribiescas, R. G., P. T. Ellison and P. B. Gray 2012 Male life history, reproductive effort, and the evolution of the Genus Homo. Current Anthropology 53(S6):S424-S435. Britton, K., V. Grimes, L. Niven, T. E. Steele, S. McPherron, M. Soressi, T. E. Kelly, J. Jaubert, J. J. Hublin and M. P. Richards 2011 Strontium isotope evidence for migration in late Pleistocene Rangifer: implications for Neanderthal hunting strategies at the Middle Palaeolithic site of Jonzac, France. Journal of Human Evolution 61(2):176-185. Brumbach, H.-J. and R. Jarvenpa 1997 Ethnoarchaeology of subsistence space and gender: a Subarctic Dene case. American Antiquity 62(3):414-436. Bruner, E. M. 1953 Primary group experience and the process of acculturation. American Anthropologist 58:605-623. Buc, N. 2011 Experimental series and use-wear in bone tools. Journal of Archaeological Science 38(3):546-557. Bunn, H. T. 1993 Bone assemblages at base camps: a first consideration of carcass transport and bone destruction by the Hadza. In From Bones to Behavior, edited by J. M. Hudson, pp. 156-168. Center for Archaeological Investigations Occasional Paper No 21 , Southern Illinois University, Carbondale. 358 Burgess, G. R. 1999 The bones of the beast: actualistic taphonomic research and its role in understanding hominid exploitation of bison and other larger mammals. In Le Bison: Gibier et Moyen de Subsistence du Paléolithque aux Paléoindiens de Grandes Plaines, edited by J. G. Enloe and F. David, pp. 1-17. Editions APDCA, Antibes. Burkart, J. M., S. B. Hrdy and C. P. Van Schaik 2009 Cooperative breeding and human cognitive evolution. Evolutionary Anthropology: Issues, News, and Reviews 18(5):175-186. Burke, A. 2006 Neanderthal settlement patterns in Crimea: a landscape approach. Journal of Anthropological Archaeology 25:510-523. 2012 Spatial abilities, cognition and the pattern of Neanderthal and modern human dispersals. Quaternary International 247:230-235 Burke, A. and F. d'Errico 2008 A Middle Palaeolithic bone tool from Crimea (Ukraine). Antiquity 82:843852. Burr, D. B. 1980 The relationships among physical, geometrical and mechanical properties of bone, with a note on the properties of non-human primates. Yearbook of Physical Anthropology 23:109-146. Byrne, R. 2004 Understanding culture across species. Trends in Cognitive Sciences 8(8):341-346. Cabanes, D., C. Mallol, I. Expósito and J. Baena 2010 Phytolith evidence for hearths and beds in the late Mousterian occupations of Esquilleu cave (Cantabria, Spain). Journal of Archaeological Science 37(11):2947-2957. Campana, D. V. 1989 Natufian and Protoneolithic Bone Tools: The Manufacture and Use of Bone Implements in the Zagros and Levant . British Archaeological Reports International Series 434. British Archaeological Reports, Oxford. Carlson, S. J. 1990 Vertebrate Dental Structures. In Skeletal Biomineralization: Patterns, Processes and Evolutionary Trends,Volume 1, edited by J. G. Carter, pp. 531-556. Van Nostrand Reinhold, New York. Caron, F., F. d’Errico, P. Del Moral, F. Santos and J. Zilhão 2011 The reality of Neanderthal symbolic behavior at the Grotte du Renne, Arcy-sur-Cure. PloS ONE 6(6):e21545. Chase, P. G. 1986 The Hunters of Combe Grenal. British Archaeological Reports International Series 286. British Archaeological Reports, Oxford. 359 Chase, P. G. 2007 The significance of "acculturation" depends on the meaning of "culture". In Rethinking the Human Revolution, edited by P. Mellars, K.Boyle, O. BarYosef, and C. Stringer, pp. 55-65. McDonald Institute Monographs. McDonald Institute for Archaeological Research, Cambridge. Chase, P. G., D. Armand, A. Debenath, H. Dibble and A. J. Jelinek 1994 Taphonomy and zooarchaeology of a Mousterian faunal assemblage from La Quina, Charente, France. Journal of Field Archaeology 21(3):289-305. Chase, P. G. and H. L. Dibble 1987 Middle Palaeolithic symbolism: a review of current evidence and interpretations. Journal of Anthropological Archaeology 6:623-296. Churchill, S. E., R. G. Franciscus, H. A. McKean-Peraza, J. A. Daniel and B. R. Warren 2009 Shanidar 3 Neandertal rib puncture wound and Paleolithic weaponry. Journal of Human Evolution 57(2):163-178. Churchill, S. E. and F. H. Smith 2000 Makers of the Early Aurignacian of Europe. Yearbook of Physical Anthropology 43:61-115. Clark, G. A. 2002 Neandertal archaeology: implications for our origins. American Anthropologist 104(1):50-67. Clarke, G. A. and J.M. Lindley 1991 On paradigmatic biases and Paleolithic research traditions. Current Anthropology 32(5): 577-587. Clark, J. G. D. and M. W. Thompson 1953 The groove and splinter technique of working antler in Upper Palaeolithic and Mesolithic Europe. Proceedings of the Prehistoric Society 6:148-161. Cleghorn, N. and C. W. Marean 2004 The destruction of skeletal elements by carnivores: the growth of a general model for skeletal element destruction and survival in zooarchaeological assemblages. In Breathing Life Into Fossils: Taphonomic Studies in Honor of C.K. Brain, edited by T. R. Pickering, K. Schick, and N. Toth, pp. 37-66. Stone Age Institute Press, Gosport, Indiana. Collie, G. L. 1928 The Aurignacians and Their Culture. The Logan Museum, Beloit College, Beloit. Conard, N., S. Walker and A. Kandel 2008 How heating and cooling and wetting and drying can destroy dense faunal elements and lead to differential preservation. Palaeogeography, Palaeoclimatology, Palaeoecology 266(3-4):236-245. Conkey, M. W. and J. M. Gero 1991 Tensions, pluralities and engendering archaeology: an introduction to women and prehistory. In Engendering Archaeology, edited by J. M. Gero and M. Conkey, pp. 3-30. Blackwell, Oxford. 360 Cosgrove, R., and A. Pike-Tay 2004 The Middle Palaeolithic and late Pleistocene Tasmania hunting behavior: a reconsideration of the attributes of modern behavior. International Journal of Osteology 14(34): 321-332. Costamagno, S., I. Théry-Parisot, J.-P. Brugal and R. Guiber 2005 Taphonomic consequences of the use of bones as fuel. Experimental data and archaeological applications. In From Biosphere to Lithosphere: New Studies in Vertebrate Taphonomy edited by T. O'Conner, pp. 52-61. Oxbow Books, Oxford. Crowther, J. 2002 The experimental earthwork at Wareham, Dorset after 33 years: retention and leaching of phosphate released in the decomposition of buried bone. Journal of Archaeological Science 29(4):405-411. Currat, M., and L. Excoffier 2011 Strong reproductive isolation between humans and Neanderthals inferred from observed patterns of introgression. Proceedings of the National Academy of Sciences 108(37): 15129-34 Currey, J. D. 1990 Biomechanics of mineralized skeletons. In Skeletal Biomineralization: Patterns, Processes and Evolutionary Trends Volume 1, edited by J. C. Carter, pp. 11-25. Van Nostrand Reinhold, New York. 1979 Mechanical properties of bone tissues with greatly differing functions. Journal of Biomechanics 12:313-319 2002 Bones: Structures and Mechanics. Second Edition, Princeton University Press, Princeton. Currey, J. D., T. Landete-Castillejos, J. Estevez, F. Ceacero, A. Olguin, A. Garcia and L. Gallego 2009 The mechanical properties of red deer antler bone when used in fighting. Journal of Experimental Biology 212(Pt 24):3985-3993. Cusick, J. G. 1998 Historiography of acculturation: an evaluation of concepts and their application in archaeology. In Studies in Culture Contact: Interaction, Culture Change, and Archaeology, edited by J. G. Cusick, pp. 126-145. Occasional Paper No. 25. Center for Archaeological Investigations, Southern Illinois University, Carbondale. Cutler, A., A.K. Behrensmeyer and R. E. Chapman 1999 Environmental information in a recent bone assemblage: roles of taphonomic processes and ecological change. Palaeogeography, Palaeoclimatology, Palaeoecology 149: 359-372 d'Errico, F. and L. Backwell 2003 Possible evidence of bone tool shaping by Swartkrans early hominids. Journal of Archaeological Science 30 (12):1559-1576. 361 d’Errico, F., M. Julien, D. Liolios, M. Vanhaeran and D. Baffier 2003 Many awls in our argument. Bone tool manufacture and use in the Châtelperronian and Aurignacian levels of the Grotte du Renne at Arcy-sur-Cure. In The Chronology of the Aurignacian and of the Transitional Technocomplexes: Dating, Stratigraphies, Cultural Implications. edited by J. Zilhão and F. d’Errico. Trabalhos de Arqueologia. vol. Proceedings of symposium 6.1 of the XIVth Congress of the UISPP (University of Liege, Belgium, September 2-8, 2001). Trabalhos de arqueologia, Lisbon. d’Errico, F., J. Zilhao, M. Julien, D. Baffier and J. Pelegrin 1998 Neanderthal acculturation in Western Europe? A critical review of the evidence and its intepretation. Current Anthropology 39(S):1-44. de Quiros, F.B., V. Cabrera, M. Lloret and A. Pike-Tay 2001 New kids on the block: some comments on the Middle to Upper Palaeolithic transition in Cantabrian Spain. In Questioning the Answers: ReSolving Fundamental Problems in the Early Upper Palaeolithic, edited by M. Hayes, P. T. Thacker, pp. 27-38. Archaeopress, Oxford. Daniel, G. E. 1976 A Hundred and Fifty Years of Archaeology. Harvard University Press, Cambridge, Mass. David, E. 2007 Technology of bone and antler industries: a relevant methodology for characterising early post-glacial societies (9th-8th millenium BC). In Bones as Tools, edited by C. Gates St-Pierre and R.B. Walker, pp. 35-50.British Archaeological Report International Series 1622. Archaeopress, Oxford. David, F. 2002 Les mammifères. In L'Aurignacien de la Grotte du Renne, edited by B. Schmider, pp. 51-94. Gallia Prehistoire Suppelement XXXIV. CNRS, Paris. 2012 Personal communication to Clare Tolmie regarding the standard practise of discarding non-identifiable bone fragments until the 1960s or later, March 2012. David, F., N. Connet, M. Girard, V. L'Homme, J.-C. Miskovsky and A. Roblin-Jouve 2001 Le Châtelperronian de la Grotte du Renne à Arcy-sur-Cure (Yonne). Données sedimentologiques et chronostratigraphiques. Bulletin de la Société Préhistorique Française 98(2):207-230. David, F., V. D'Iatchenko, C. Karlin and Y.V. Tchenekov 1998 Du traitment des peaux en Siberie: Dolanges et autres nomades du Nord. Boreales 74/77: 111-137. David, F., V. D'Iatchenko, J. G. Enloe, M. Girard, M. Hardy, V. Lhomme, A. RoblinJouve, A.-M. Tillier and C. Tolmie 2009 New Neandertal remains from the Grotte du Bison at Arcy-sur-Cure, France. Journal of Human Evolution 57(6):805-809. 362 David, F. and C. Farizy 1994 Les vestiges osseux: étude archaéozoologique. In Hommes et Bisons du Paléolithique Moyen à Mauran (Haute Garonne), edited by C. Farizy, F. David and J. Jaubert, pp. 177-324. Gallia Préhistoire Supplement. CNRS, Paris. David, F. and T. Poulain 1990 La faune de grands mammifères des niveaux XI et XC de la Grotte du Renne à Arcy-sur-Cure, Yonne. In Paléolithique Moyen Recent et Paléolithic Supérieur Ancien en Europe, edited by C. Farizy, pp 319-324. Musée de Préhistoire d’Ile de France, Nemours. Davies, W. 2007 Rethinking the Aurignacian as an expression of modern human mobility and dispersal. In Rethinking the Human Revolution, edited by P. Mellars, K. Boyle, O. Bar-Yosef, and C. Stringer, pp. 263-274. McDonald Institute Monographs. McDonald Institute for Archaeological Research, Cambridge. Delaporte, Y., M. Roué and A. Cooper 1980 Lapp fur coats: clothing adaptation in a pastoral Society. Arctic Anthropology 17(2):64-67. Delpeche, F. 1983 Les Faunes du Paléolithique Supérieur dans le Sud-Ouest de la France. Cahiers du Quaternaire 6. CNRS, Paris. 1993 The fauna of the Early Upper Palaeolithic: biostratigraphy of large mammals and current problems in chronostratigrapy. In Before Lascaux: The Complex Record of hte Early Upper Palaeolithic, edited by H. Knecht, A. PikeTay, and R. White, pp. 71-84. CRC Press, Boca Raton. Dibble, H. 1987 The interpretation of Middle Paleolithic scraper morphology. American Antiquity 52:109–117. Dohrenwend, B. P. and R. J. Smith 1962 Toward a theory of acculturation. Southwestern Journal of Anthropology 18(1):30-39. Domínguez, R M. and T. R. Pickering 2003 Early hominid hunting and scavenging: A zooarcheological review. Evolutionary Anthropology: Issues, News, and Reviews 12(6):275-282. Dominguez, R. M. and J. Yravedra 2009 Why are cut mark frequencies in archaeofaunal assemblages so variable? A multivariate analysis. Journal of Archaeological Science 36(3):884-894. Doyon, Luc 2013 Personal communication to Clare Tolmie regarding results of ongoing research on the Abri Cellier antler points as part of his MA thesis, April 2013. 363 Drucker, D. and H. Bocherens 2004 Carbon and nitrogen stable isotopes as tracers of change in diet breadth during Middle and Upper Palaeolithic in Europe. International Journal of Osteoarchaeology: 14 (34):162-177. Drumheller, Stephanie 2009 Personal communication to Clare Tolmie regarding carnivore ravaging and possible role of crocodilians in the formation of East African Pleistociene faunal assemblage, March 2009. Dusseldorp, G. L. 2010 Studying Pleistocene Neanderthal and cave hyena dietary habits: combining isotopic and archaeozoological analyses. Journal of Archaeological Method and Theory 18(3):224-255 Eerkens, J. W. and C. P. Lipo 2007 Cultural transmission theory and the archaeological record: providing context to understanding variation and temporal changes in material culture. Journal of Archaeological Research 15(3):239-274. Efremov, I. A. 1940 Taphonomy: an new branch of paleontology. The Pan-American Geologist 74(2):81-93. Ehrhardt, K. L. 2009 Contextualizing Indigenous Copper Consumption and ‘Hybridization’ in Technological Style in the Early Contact Period in North American Midcontinent. Paper presented at the 26th Annual Visiting Scholar Conference, Center for Archaeological Investigation, Southern Illinois University, Carbondale. Eisley, L. 1957 Neanderthal man and the dawn of human paleontology. Quarterly Review of Biology 32(4):323-329. El Zaatari, S., F. E. Grine, P. S. Ungar and J.-J. Hublin 2011 Ecogeographic variation in Neandertal dietary habits: Evidence from occlusal molar microwear texture analysis. Journal of Human Evolution 61(4):411-424. Ellis, B. J., A. J. Figueredo, B. H. Brumbach and G. L. Schlomer 2009 Fundamental dimensions of environmental risk. Human Nature 20(2):204268. Emerson, A.M. 1993 The role of body part utility in small-scale hunting under two stragies of carcass recovery. In From Bones to Behavior, edited by J. M. Hudson, pp. 138157. Center for Archaeological Investigations Occasional Paper No 21, Southern Illinois University, Carbondale. 364 Enloe, J. G. 1993 Subsistence organization in the Early Upper Palaeolithic: reindeer hunters of the Abri du Flageolet, Couche V. In Before Lascaux: The Complex Record of the Early Upper Palaeolithic, edited by H. Knecht, A. Pike-Tay, and R. White, pp. 101-115. CRC Press, Boca Raton. 1997 Seasonality and age structures in remains of R. tarandus: Magdalenian hunting strategies at Verberie. Anthropozoologia 25/26:95-102. 2001 Investigating human adaptations in the Early Upper Palaeolithic. In Questioning the Answers: Re-Solving Fundamental Problems in the Early Upper Palaeolithic, edited by M. Hayes and P. T. Thacker, pp. 21-26. Archaeopress, Oxford. 2006 Geological processes and site structure: assessing integrity at a Late Paleolithic open-air site in northern France. Geoarchaeology 21(6):523-540. In Press Neanderthal to Neanderthal evolution: preliminary observations of faunal exploitation from Mousterian to Châtelperronian at Arcy-sur-Cure. In Zooarchaeology and Modern Human Origins: Human Hunting Behavior During the Later Pleistocene, edited by J.L. Clark and J. Speth. Springer Publishing, New York. Enloe, J.G., and F. David 1995 Rangifer herd behavior: seasonality of hunting in the Magdalenien of the Paris Basin. In Caribou and Reindeer Hunters of the Northern Hemisphere edited by L. Jackson and P. Thacker, pp 47-63. Avebury Press, Aldershot. Enloe, J. G. and F. Lanoë 2012 Spatial organization in the Terminal Mousterian: taphonomy and Neanderthal behavior at the Grotte du Bison,Arcy-sur-Cure. Paper presented at the 77th Annual meeting of the Society for American Archaeology, Memphis, Tennessee. Enloe, J. G. and E. Turner 2006 Methodological problems and biases in age determinations: a view from the Magdalenian. In Recent Advances in Ageing and Sexing Animal Bones, edited by D. Ruscillo, pp. 129-143. Oxbow Books, Oxford. Estabrook, V. H. 2009 Sampling Biases and New Ways of Addressing the Significance of Trauma in Neandertals. Upublished PhD Dissertation, Department of Anthropology, University of Michigan, Ann Arbor. Ewers, J. C. 1945 Blackfeet Crafts. Indian Handcrafts. United States Indian Service, Education Division, Washingtgon DC. Faith, J. and A. Behrensmeyer 2006 Changing patterns of carnivore modification in a landscape bone assemblage, Amboseli Park, Kenya. Journal of Archaeological Science 33(12):1718-1733. 365 Faith, J., C. Marean and A. Behrensmeyer 2007 Carnivore competition, bone destruction, and bone density. Journal of Archaeological Science 34(12):2025-2034. Farizy, C. 1990a Du mousterien au châtelperronien à Arcy-sur-Cure: un état de la question. In Paléolithique Moyen Récent et Paléolithique Supérieur Ancien en Europe, edited by C. Farizy, pp. 281-290. Musée de Préhistoire d’Ile de France, Nemours. 1990b The transition from Middle to Upper Palaeolithic at Arcy-sur-Cure (Yonne, France): technological, economic and social aspects. In The Emergence of Modern Humans, edited by P. Mellars, pp. 303-326. University of Edinburgh Press, Edinburgh. Farizy, C., F. David and J. Jaubert (editors) 1994 Hommes et Bisons du Paléolithique Moyen à Mauran (Haut Garonne). Gallia Prehistoire Supplement 30. CNRS, Paris. Fedigan, L. M. 1986 The changing role of women in models of human evolution. Annual Review of Anthropology 15:15-26. Fernández-Laso, M., M.G. Chacón Navarro, M.D. García-Antón, and F. Rivals 2011 Territorial mobility of Neanderthal groups: a case study from Level M of Abric Romaní (Capellades, Barcelona, Spain). In Neanderthal Lifeways: Subsistence and Technology. 150 Years of Neanderthal Studies, edited by N.J. Conard and J. Richter, pp. 187-202. Springer, New York. Fink, C. 2004 Optimal foraging theory and cognitive archaeology: Cup'ik cultural perceptions in Southwestern Alaska. In Hunter-Gatherers in Theory and Archaeology, edited by G. M. Crothers, pp. 279-298. Center for Archaeological Investigations Occasional Paper No. 31. Southern Illinois University, Carbondale. Finlayson, C. 2004 Neanderthals and Modern Humans : An Ecological and Evolutionary Perspective. Cambridge Studies in Biological and Evolutionary Anthropology. Cambridge University Press, New York. Finlayson, C., K. Brown, R. Blasco, J. Rosell, J.J.Negro, G.R. Bortolotti, G. Finlayson, A. Sanchez Marco, F. Giles Pacheco, J. Rodriguez Vidal, J.S. Carrion, D.A. Fa, and J.M. Rodriguez Llanes 2012 Birds of a feather: Neanderthal exploitation of raptors and corvids. PloS ONE 7(9): e45927. doi10.1371/pone.0045927. Francillon-Vieillot, H., V. de Buffrenil, J. Castanet, J. Geraudie, F. J. Meunier, J. Y. Sire, L. Zylberberg and A. de Riqles 1990 Microstructure and mineralization of vertebrate skeletal tissues. In Skeletal Biomineralization: Patterns, Processes and Evolutionary Trends, Volume 1, edited by J. C. Carter, pp. 471-530. Van Nostrand Reinhold, New York. 366 Friemand, C. 2009 ‘The Truth is Rarely Pure and Never Simple’: Interpreting skeuomorphism in the archaeological record. Paper presented at the 26th Annual Visiting Scholar Conference, Center for Archaeological Investigation, Southern Illinois University, Carbondale. Gamble, C. S. 1999 The Palaeolithic Societies of Europe. Cambridge University Press, Cambridge. 2007 Origins and Revolutions: Human Identity and Earliest Prehistory. Cambridge University Press, Cambridge. 2011 The social and material life of Neanderthals. In Neanderthal Lifeways, Subsistence and Technology: One Hundred and Fifty Years of Neanderthal Study, edited by N. J. Conard and J. Richter, pp. 157-166. Springer, New York. Gargett, R. H. 1989 Grave shortcomings: the evidence for Neanderthal burial. Current Anthropology 30:157-190. Gaudelli, J.-L. and H. Laville 1990 L'environnement climatique de la fin de la Moustérien à Combe Grenal et à Camiac: confrontation des donées naturalistes et implications. In Paléolithic Moyen Recent et Paléolithic Supérieur Ancien en Europe, edited by C. Farizy, pp. 43-48. Musée Prehistorique d'Ile de France, Nemours. Gero, J. M. 1985 Sociopolitics and the woman-at-home ideology. American Antiquity 50(2):342-350. 1991 Genderlithics: women's roles in stone tool production. In Engendering Archaeology, edited by J. M. Gero, Meg Conkey, pp. 163-193. Blackwell, Oxford. Gifford, D. P. 1981 Taphonomy and paleaoecology: a critical review of archaeology’s sister disciplines. Advances in Archaeological Method and Theory 4:365-438. Gilbert, B. M. 1990 Mammalian Osteology. Missouri Archaeological Society, Columbia, MO. Gilbert, B. M., L. D. Martin and H. G. Savage 1996 Avian Osteology. Missouri Archaeological Society, Columbia, MO. Gilligan, I. 2007 Neanderthal extinction and modern human behaviour: the role of climate change and clothing. World Archaeology 39(4):499-514. 2008 Clothing and climate in Aboriginal Australia. Current Anthropology 49(3):487-495. 2010 The prehistoric development of clothing: archaeological implications of a thermal model. Journal of Archaeological Method and Theory 17(1):15-80. 367 Gilmore, K. P. 2005 These boots were made for walking. In Gender and Hide Production, edited by L. Frink and K. Weedman, pp. 13-36. Altamira Press, New York. Girard, C. 1980 Les industries mousteriènnes de la Grotte du Renne à Arcy-sur-Cure (Yonne). Gallia Prehistoire 23:1-36. Girard, M., J.-C. Miskovsky and J. Evins 1990 La fin du Wurm moyen et le debut du Wurm Supérieur à Arcy-sur-Cure. In Paléolithique Moyen Recent et Paléolithique Supérieur Ancien en Europe, edited by C. Farizy, pp. 295-303. Musée de Prehistoire d'Ile de France, Nemours. Goldberg, P., H. Dibble, F. Berna, D. Sandgathe, S. J. P. McPherron and A. Turq 2010 New evidence on Neandertal use of fire: examples from Roc de Marsal and Pech de l'Azé IV. Quaternary International: doi: 10.1016/ j.quaint. 2010. 11.015. Gouedo, J.M., 1990 Quelques reflexions sur la technologie lithique du châtelperronien de la Couche X de la Grotte du Renne d’Arcy-sur-Cure. In Paléolithique Moyen Recent et Paléolithique Supérieur Ancien en Europe, edited by C. Farizy, pp. 305-308. Musée de Prehistoire d'Ile de France, Nemours. Goutas, N. 2012 Personal communication to Clare Tolmie regarding bone tool raw materials, Aurignacian tool types and Gravettian site locations, November 2012. Gowlett, J., C. Gamble and R. Dunbar 2012 Human evolution and the archaeology of the social brain. Current Anthropology 53(6):693-722. Grant, A., 1982 The use of toothwear as a guide to the age of domestic ungulates. In Aging and Sexing Animal Bones from Archaeological Sites, edited by B. Wilson, C. Grigson and S. Payne, pp. 91-108. British Archaeological Report 109, BAR, Oxford. Grayson, D., K., 1984 Quantatative Zooarchaeology. Academic Press, New York. 1991 Alpine faunas from the White Mountains, California: adaptations in the late prehistoric Great Basin? Journal of Archaeological Sciences 18: 483-506. 2003 Ungulates and the Middle-Upper Palaeolithic transition at Grotte XVI (Dordogne, France). Journal of Archaeological Science 30(12): 1633-1648. 1986 Eoliths, archaeological ambiguity and the generation of "Middle Range" research. In American Archaeology, Past and Future, edited by D. J. Meltzer, Don D. Fowler, Jeremy A. Sabloff, pp. 77-153. Society for American Archaeology, Washington D.C. 368 Green, R. E., J. Krause, A. W. Briggs, T. Maricic, U. Stenzel, M. Kircher, N. Patterson, H. Li, W. Zhai, M. H. Y. Fritz, N. F. Hansen, E. Y. Durand, A. S. Malaspinas, J. D. Jensen, T. Marques-Bonet, C. Alkan, K. Prufer, M. Meyer, H. A. Burbano, J. M. Good, R. Schultz, A. Aximu-Petri, A. Butthof, B. Hober, B. Hoffner, M. Siegemund, A. Weihmann, C. Nusbaum, E. S. Lander, C. Russ, N. Novod, J. Affourtit, M. Egholm, C. Verna, P. Rudan, D. Brajkovic, Z. Kucan, I. Gusic, V. B. Doronichev, L. V. Golovanova, C. Lalueza-Fox, M. de la Rasilla, J. Fortea, A. Rosas, R. W. Schmitz, P. L. F. Johnson, E. E. Eichler, D. Falush, E. Birney, J. C. Mullikin, M. Slatkin, R. Nielsen, J. Kelso, M. Lachmann, D. Reich and S. Paabo 2010 A draft sequence of the Neandertal genome. Science 328(5979):710-722. Green, R. E., J. Krause, S. E. Ptak, A. W. Briggs, M. T. Ronan, J. F. Simons, L. Du, M. Egholm, J. M. Rothberg, M. Paunovic and S. Pääbo 2006 Analysis of one million base pairs of Neanderthal DNA. Nature 444 (7117):330-336. Green, W.R. 2008 Personal communication to Clare Tolmie regarding the changing research focus of the Logan Museum, March 2008. Griffitts, J. B. 2006 Bone Tools and Technological Choice: Change and Stability on the Northern Plains. Unpublished PhD Thesis, Department of Anthropology, University of Arizona. 2007 Stability and change in bone tool use along the Middle Missouri, North Dakota. In Bones as Tools: Current Methods and Interpretations in Worked Bone Studies edited by C. Gates St-Pierre and R.B Walker, pp. 93-106. Oxbow Books, Oxford. 2009 Bone Tool Technology on the Middle Missouri: Experiments and Innovations. Paper presented at the 26th Annual Visiting Scholar Conference, Center for Archaeological Investigation, Southern Illinois University, Carbondale. Guatelli-Steinberg, D. 2009 Recent studies of dental development in Neandertals: implications for Neandertal life histories. Evolutionary Anthropology: Issues, News, and Reviews 18(1):9-20. Gurven, M. and K. Hill 2009 Why do men hunt? Current Anthropology 50(1):51-74. Gurven, M. and H. Kaplan 2006 Determinants of time allocation across the lifespan. Human Nature 17(1):1-49. Gurven, M., H. Kaplan and M. Gutierrez 2006 How long does it take to become a proficient hunter? Implications for the evolution of extended development and long life span. Journal of Human Evolution 51(5):454-470. 369 Guthrie, R. D. 1983 Osseous projectile points: biological considerations affecting raw material selection and design among Paleolithic and Paleoindinan peoples. In Animals and Archaeology. Volume 1: Hunters and Their Prey, edited by J. Clutton Brook and C. Grigson, pp. 273-294. British Archaeological Reports International Series 163, BAR, Oxford. Hall, E. S. J. 1971 Kangigugksuk: a cultural reconstruction of a sixteenth century Eskimo site in Northern Alaska. Arctic Anthropology 8(1):1-101. Hames, R. and P. Draper 2004 Women's work, child care, and helpers-at-the-nest in a hunter-gatherer society. Human Nature 15(4):319-341. Hanson, C. B. 1980 Fluvial taphonomic processes: models and experiments. In Fossils in the Making, edited by A. K. Behrensmeyer and A. P. Hill, pp. 156-181. University of Chicago Press, Chicago. Hardy, B., 2010 Climatic variability and plant food distribution in Pleistocene Europe: Implications for Neanderthal diet and subsistence. Quaternary Science Reviews 29 (5-6) 662-679. Harrold, F. B. 1988 The Châtelperronian and Early Aurignacian in France. In The Early Upper Palaeolithic: Evidence from Europe and the Near East, edited by J. F. Hoffecker, and C.A. Wolf, pp. 157-193. British Archaeological Reports International Series 437. Archaeopress, Oxford. 2002 The Châtelperronian in historical context. Journal of Anthropological Research 56(1):59-75. Harrold, F. B. and M. Otte 2001 Time, space and cultural process in the European Middle-Upper Palaeolithic transition. In Questioning the Answers: Re-solving Fundamental Problems of the Early Upper Palaeolithic, edited by M. Hayes and P. T. Thacker, pp. 3-11. British Archaeological Reports International Series 1005. Archaeopress, Oxford. Hatt, G. and K. Taylor 1969 Arctic skin clothing in Eurasia and America an ethnographic study. Arctic Anthropology 5(2):3-132. Hawkes, K., J. F. O'Connell and N. G. Blurton Jones 1995 Hadza children's foraging: juvenile dependency, social arrangements and mobility among hunter-gatherers. Current Anthropology 36(4):688-700. 1997 Hadza womens's time allocation, offspring provisioning, and the evolution of long postmenopausal life spans. Current Anthropology 38(4):551-577. 2001 Hunting and nuclear families: some lessons from the Hadza about men's work. Current Anthropology 41(5):681-709. 370 Haynes, G. 1983 A guide for differentiating mammalian carnivore taxa responsible for gnaw damage on herbivore limb bones. Paleobiology 9(2):164-172. 1988 Mass deaths and serial predation: comparative studies of modern large animal death sites. Journal of Archaeological Science 15: 219-235. 2004 Some rather odd detective stories: a view of some actualistic and taphonomic trends in Paleoindian studies. In Breathing Life Into Fossils: Taphonomic Studies in Honor of C.K Brain, edited by T. R. Pickering, K. Schick, and N. Toth, pp. 25-35. Stone Age Institute Press, Gosport, Indiana. Henry, A. G., A. S. Brooks and D. R. Piperno 2010 Microfossils in calculus demonstrate consumption of plants and cooked foods in Neanderthal diets (Shanidar III, Iraq; Spy I and II, Belgium). Proceedings of the National Academy of Sciences 108(2):486-491. Henshilwood, C. S. and C. W. Marean 2003 The origins of modern human behavior: critique of the models and their test implications. Current Anthropology 44(5):627-651. Higham, T., R. Jacobi, M. Julien, F. David, L. Basell, R. Wood, W. Davies and C. B. Ramsey 2010 Chronology of the Grotte du Renne (France) and implications for the context of ornaments and human remains within the Châtelperronian. Proceedings of the National Academy of Sciences 107(47):20234-20239. Hill, A. P. 1980 Early post-mortem damage to the remains of some contemporary East African mammals. In Fossils in the Making, edited by A. K. Behrensmeyer and A. P. Hill, pp. 131-152. University of Chicago Press, Chicago. Hill, K.R., R.S. Walker, M. Božičević, J. Elder. T. Headland, B. Hewlett, A. M. Hurtado, F. Marlow, P. Wiessner and B. Wood 2011 Co-residence patterns in hunter-gatherer societies show unique human social structure. Science 331:1286-1289 Hilton, C., E. and R. D. Greaves 2008 Seasonality and sex differences in travel distance and resource transport in Venezuelan foragers. Current Anthropology 49(1):144-153. Hockett, B., and J.A. Hawes 2005 Nutritional ecology and the human demography of Neandertal extinction. Quaternary International 137(1): 21-34. Hodge, F. W. 1920 Hawikuk Bonework. Indian Notes and Monographs III. Museum of the American Indian, Heye Foundation, New York. 371 Hoffecker, J. F., V. T. Holliday, M. V. Anikovich, A. A. Sinitsyn, V. V. Popov, S. N. Lisitsyn, G. M. Levkovskaya, G. A. Pospelova, S. L. Forman and B. Giaccio 2008 From the Bay of Naples to the River Don: the Campanian Ignimbrite eruption and the Middle to Upper Paleolithic transition in Eastern Europe. Journal of Human Evolution 55(5):858-870. Hoffman, J. 1980 Scapula skin-dressing and fiber-processing tools. Plains Anthropologist 25(88):135-141. Hofmann, H. 1989 Mammifères d'Europe. Nathan, Paris. Howell, F. C. 1999 Paleo-demes, species clades, and extinctions in the Pleistocene hominin record. Journal of Anthropological Research 55(2):191-243. Hrdlička, A. 1929 The Neanderthal phase of man. Journal of the Royal Anthropological Institute 57:249-274. Hrdy, S. B. 2005 Comes the child before man: how cooperative breeding and prolonged post-weaning dependence shaped human potential. In Hunter-Gatherer Childhoods, edited by B. S. Hewlett and M. E. Lamb, pp. 65-91. Aldine, New Brunswick, NJ. Hublin, J.-J. 2000 Modern-nonmodern hominid interactions: a Mediterranean perspective. In The Geography of Neandertals and Modern Humans in Europe and the Greater Mediterranean, edited by O. Bar Yosef, and D. Pilbeam, pp. 157-182. Peabody Museum of Archaeology and Ethnology, Harvard University, Cambridge, Mass. Hublin, J.-J., F. Spoor, M. Brun, F. Zonneveld and S. Condemi 1996 A late Neanderthal associated with Upper Palaeolithic artefacts. Nature 381:224-226. Hublin, J. J., S. Talamo, M. Julien, F. David, N. Connet, P. Bodu, B. Vandermeersch and M. P. Richards 2012 Radiocarbon dates from the Grotte du Renne and Saint-Césaire support a Neandertal origin for the Châtelperronian. Proceedings of the National Academy of Sciences of the United States of America 109(46):18743-18748. Hurel, A. 2007 La France Préhistorienne de 1789 a 1941. CNRS Editions, Paris. Hurtado, A. M., H. Kaplan, K. Hill and I. Hurtado 1992 Trade-offs between female food acquisition and child care among Hiwi and Ache foragers. Human Nature 3(3):185-216. Isler, K. and C. P. van Schaik 2012 Allomaternal care, life history and brain size evolution in mammals. Journal of Human Evolution 63(1):52-63. 372 Jarvenpa, R. and H. J. Brumbach 2009 Fun with Dick and Jane: ethnoarchaeology, circumpolar toolkits and gender "inequality". Ethnoarchaeology 1(1):57-78. Jochim, M. A. 1988a Early Upper Palaeolithic research: a review. In The Early Upper Palaeolithic: Evidence from Europe and the Near East, edited by J. F. Hoffecker, and C.A. Wolf, pp. 273-277. vol. British Archaeological Report International Series 437. Archaeopress, Oxford. 1988b Optimal foraging and the division of labor. American Anthropologist 90(1):130-136. Johnson, E., G. Politis and M. Gutierrez 2000 Early Holocene bone technology at the La Olla I site, coast of the Argentine Pampas. Journal of Archaeological Science 27:463-477. Julien, M., D. Baffier and D. Liolios 2002 L'outillage en matières dures animals. In L'Aurignacien de la Grotte du Renne: Les fouilles d'Andre Leroi-Gourhan, edited by B. Schmider, pp. 217-250. CNRS, Paris. Kaplan, H. and K. Hill 1985 Food sharing among Ache foragers: test of explanantory hypotheses. Current Anthropology 26(2):233-246. Kaufman, D. 2002 Redating the social revolution: the case for the Middle Paleolithic. Journal of Anthropological Research 58(4):477-492. Kent, S. 1993 Variability in faunal assemblages: the influences of hunting skills, sharing, dogs and modes of cooking on faunal remains on a sedentary Kalahari community. Journal of Anthropological Archaeology 12:323-385. 1996 Hunting variability at a recently sedentary Kalahari village. In Cultural Diversity Among Twentieth Century Foragers, edited by S. Kent, pp. 123-156. Cambridge University Press, Cambridge. Kilberger, M. and J. G. Enloe 2005 Hunting Rangifer in the Late Upper Paleolithic: Developing Methodologies for Constructing Mortality Profiles from Maxillary Dentition. Paper presented at the Paleoanthropology Society Meeting, Milwaukee. Klein, R. G. and K. Cruz-Aribe 1984 The Analysis of Animal Bones from Archaeological Sites. University of Chicago Press, Chicago. Klein, R. G., K. Allwarden, and C. Wolf 1983 The calculation and interpretation of ungulate age profiles from dental crown heights. In Hunter-Gatherer Economy in Prehistory, edited by G. Bailey, pp. 47-57. Cambridge University Press, Cambridge. 373 Klippel, W. E. and J. A. Synstelien 2007 Rodents as taphonomic agents: bone gnawing by brown rats and gray Squirrels. Journal of Forensic Sciences 52(4):765-773. Knecht, H. 1991 Technological Innovation and Design During the Early Upper Palaeolithic: A Study of Organic Projectile Points. Unpublished PhD Dissertation, Department of Anthropology, New York University, New York. Knecht, H. 1993 Splits and wedges: the techniques and technologies of Early Aurignacian antler working. In Before Lascaux: the complex record of the Early Upper Palaeolithic, edited by H. Knecht, A. Pike-Tay, and R. White, pp. 137-162. CRC Press, Boca Raton. Konner, M. 2005 Hunter-gatherer infancy and childhood: the !Kung and others. In HunterGatherer Childhoods, edited by B. S. Hewlett and M. E. Lamb, pp. 19-64. Aldine, New Brunswick, NJ. Kozlowski, J. K. 2007 The significance of blade technologies in the period 50-35kyr BP for the Middle-Upper Palaeolithic in central and eastern Europe. In Rethinking the Human Revolution, edited by P. Mellars, K. Boyle, O. Bar-Yosef, and C.Stringer, pp. 317-328. McDonald Institute Monographs. McDonald Institute for Archaeological Research, Cambridge. Krings, M., A. Stone, R. W. Schmitz, H. Krainitzki, M. Stoneking and S. Paabo 1997 Neandertal DNA Sequences and the origin of modern humans. Cell 90(July 11):19-30. Kuhn, B., L. Berger and J. Skinner 2009 Variation in tooth mark frequencies on long bones from the assemblages of all three extant bone-collecting hyaenids. Journal of Archaeological Science 36(2):297-307. Kuhn, S. L. 2011 Neanderthal technoeconomics: an assessment and suggestions for future developments. In Neanderthal Lifeways, Subsistence and Technology: One Hundred and Fifty Years of Neanderthal Study, edited by N. J. Conard and J. Richter, pp. 99-110. Springer, New York:. 2012 Questions of complexity and scale in explanations for cultural transitions in the Pleistocene: a case study from the Early Upper Paleolithic. Journal of Archaeological Method and Theory. doi 10.1007/s10816-012-9146-3. Kuhn, S. L. and A. Bietti 2000 The late Middle and Early Upper Palaeolithic in Italy. In The Geography of Neanderthals and Modern Humans in Europe and the Greater Mediterranean, edited by O. Bar Yosef and D. Pilbeam, pp. 49-76. Peabody Museum of Archaeology and Ethnography, Harvard University, Cambridge, MA. 374 Kuhn, S. L. and M. C. Stiner 2001 The antiquity of hunter-gatherers. In Hunter Gatherers: An Interdisciplinary Approach, edited by C. Painter Brick, R. H. Layton, and P. Rowley Conwy, pp. 99-142. Cambridge University Press, Cambridge. Lalueza-Fox, C., H. Römpler, D. Caramelli, C. Stäubert, G. Catalano, D. Hughes, N. Rohland, E. Pilli, L. Longo, S. Condemi, M. d. l. Rasilla, J. Fortea, A. Rosas, M. Stoneking, T. Schöneberg, J. Bertranpetit and M. Hofreiter 2007 A melacortonin 1 receptor allele suggests varying pigmentation among Neandertals. Science 318(5855):1453-1455. Lam, Y. 2003 Bone density studies in zooarchaeology. Journal of Archaeological Science 30(12):1701-1708. Lam, Y. M., X. Chen and O. M. Pearson 1999 Intertaxonomic variability in patterns of bone density and the differential representation of bovid, cervid, and equid elements in the archaeological record. American Antiquity 64(2):343-362. Larounlandie, V. 2005 Anthropogenic versus non-anthropogenic bird bone assemblages: new criteria for their distinction. In From Biosphere to Lithosphere: New Studies in Vertebrate Taphonomy edited by T. O'Conner, pp. 25-30. Oxbow Books, Oxford. Lazuén, T. 2012 European Neanderthal stone hunting weapons reveal complex behavior long before the appearance of modern humans. Journal of Archaeological Science 39(7):2304-2311. Le Moine, G. 1991 Experimental Analysis of the Manufacture and Use of Bone and Antler Tools Among the Mackenzie Inuit. Unpublished PhD Dissertation, Department of Archaeology, University of Calgary, Calgary. 2007 Bones tools and bone technology: a brief history. In Bones as Tools: Current Methods and Interpretations in Worked Bone Studies, edited by C. Gates St-Pierre and R. B. Walker, pp. 9-22. British Archaeological Reports International Series 1622. Archaeopress, Oxford. Lee, R. B. and I. DeVore (editors) 1969 Man The Hunter. Aldine Publishing Co., Chicago. Leroi-Gourhan, A., 1964 La Geste et La Parole. Albin Michel, Paris. Leroi-Gourhan, A. and A. Leroi-Gourhan 1965 Chronologie des grottes d'Arcy-sur-Cure (Yonne). Gallia Prehistoire VIII:1-64. Leroyer, C. and A. Leroi-Gourhan 1993 Pollen analysis at St. Césaire. In Context of a Late Neanderthal, edited by F. Lévèque, A. M. Backer and M. Guilbaud, pp. 59-70. Monographs in World Prehistory. Prehistory Press, Madison. 375 Lévèque, F., A. M. Backer and M. Guilbaud 1993 Context of a Late Neanderthal. Monographs in Prehistory 16. Prehistory Press, Madison. Liapunova, R. and N. N. Miklukho 1996 Essays on the Ethnography of the Aleuts. Translated by J. Shelest. The Rasmuson Library Historical Translation Series. University of Alaska Press, Fairbanks. Lieberman, D.E., and J. J. Shea 1994 Behavioral differences between archaic and modern humans in the Levantine Mousterian. .American Anthropologist 96(2): 300-332. Linton, S. E. 1971 Women the gatherer: male bias in anthropology. In Woman in Cross Cultural Perspectives, edited by S. E. Jacobs. University of Illinois, Champaign. Liolios, D. 2003 L'apparitions de l'industrie ossuese au debut du Paleothithique Supérieur. Un transfert de techniques de travail du vegetal sur les matieres osseuses. Proceedings of the Congres Nationaux des Sociétés Historiques et Scientifiques:219-225. Lille. Liolios, D. 2006 Reflections on the role of bone tools in the definition of the Early Aurignacian. In Towards a Definition of the Aurignacian : Proceedings of the Symposium Held in Lisbon, Portugal, June 25-30, 2002, edited by O. Bar-Yosef and J. Zilhão, pp. 37-51. Instituto Português de Arqueologia and the American School of Prehistoric Research, Peabody Museum, Harvard University, Lisbon. Lord, T. C., T. P. O'Connor, D. C. Siebrandt and R. M. Jacobi 2007 People and large carnivores as biostratinomic agents in Late Glacial cave assemblages. Journal of Quaternary Science 22(7):681-694. Lothrop, S. K. 1928 The Indians of Tierra del Fuego. Museum of the American Indian, Reye Foundation, New York. Lubinski, P. M. and C. J. O'Brien 2001 Observations on seasonality and mortality from a recent catastrophic death assemblage. Journal of Archaeological Science 28(8):833-842. Lupo, K. 2002 Cut and tooth mark distributions on large animal bones: ethnoarchaeological data from the Hadza and their implications for current ideas about early human carnivory. Journal of Archaeological Science 29(1):85-109. Lyman, R. L. 1984 Bone density and differential survivorship of fossil classes. Journal of Anthropological Archaeology 3:259-299. 1985 Bone frequencies: differential transport, in situ destruction and the MGUI. Journal of Archaeological Science 12:221-236. 376 Lyman, R. L. 1987 Hunting for evidence of Plio-Pleistocene hominid scavengers. American Anthropologist 89(3):710-715. 1989 Taphonomy of cervids killed by the 18th May 1980 volcanic eruption of Mount St. Helens, Washington. In Bone Modification, edited by R. Bonnischen and M. Sorg, pp. 149-167. Center for the Study of the First Americans, University of Maine, Orano. 1993 Density mediated attrition of bone assemblages: new insights. In From Bones to Behavior, edited by J. M. Hudson, pp. 324-341. Center for Archaeological Investigations Occasional Paper No 21, Southern Illinois University, Carbondale. 1994 Vertebrate Taphonomy. Cambridge Manuals in Archaeology. Cambridge University Press, Cambridge. 2003 The influence of time averaging and space averaging on the application of foraging theory to zooarchaeology. Journal of Archaeological Science 30(5):595610. 2005 Analyzing cutmarks: lessons from artiodactyl remains in the northwestern United States. Journal of Archaeological Science 32(12):1722-1733. Lyman, R.L. and K. Ames 2007 On the use of species-area curves to detect the effects of sample size. Journal of Archaeological Science 34(12):1985-1990. MacDonald, K., 2007 Cross-cultural comparison of learning in human hunting. Human Nature 18 (4):386-402. MacGregor, A. G. 1985 Bone, Antler, Ivory and Horn: The Technology of Skeletal Materials Since the Roman Period. Croom Helm, London. MacGregor, A. G. and J. D. Currey 1983 Mechanical properties as conditioning factors in the bone and antler industry of the 3rd to 13th Century AD. Journal of Archaeological Science 10:7177. Mangerud, J. 1991 The Last Glacial cycle in Europe. In Quaternary Landscapes, edited by L. C. K. Shane and E. J. Cushing, pp. 38-75. University of Minnesota Press, Minneapolis. Marchant, L. M. 1991 Primate reproductive biology and models of human origins. In The Archaeology of Gender, edited by D. Walde and N. D. Willows, pp. 50-55. University of Alberta, Calgary. Marean, C. W. and Z. Assefa 1999 Zooarcheological evidence for the faunal exploitation behavior of Neandertals and early modern humans. Evolutionary Anthropology 8(1):22-37. 377 Marean, C. W. and C. J. Frey 1997 Animal bones from caves to cities: reverse utility indices as methodological artifacts. American Antiquity 62(4):698-711. Marean, C. W. and S. Y. Kim 1998 Mousterian large-mammal remains from Kobeh Cave: behavioral implications for Neanderthals and Early Modern Humans. Current Anthropology 39(2):S79-S113. Marean, C. W. and L. M. Spencer 1991 Impact of carnivore ravaging on zooarchaeological measures of element abundance. American Antiquity 56(4):654-658. Marín Arroyo, A., P. Fosse and J. Vigne 2009 Probable evidences of bone accumulation by Pleistocene bearded vultures at the archaeological site of El Mirón Cave (Spain). Journal of Archaeological Science 36(2):284-296. Marks, A. E. 1988 The Middle to Upper Palaeolithic transition in the southern Levant: technological change as an adaptation to increasing mobility. In L'Homme de Néandertal, Volume 8: La Mutation, edited by M. Otte, pp. 109-124. ERAUL, Liege. Marquet, J.-C. 1993 Paléoenvironnement et Chronololgie des Sites du Domaine Atlantique Français d'Age Pleistocene Moyen et Supérieur, d'apres l'Etude des Rongeurs. Cahiers de la Claise, Toulouse. Martin-Gonzalez, J. A., A. Mateos, I. Goikoetxea, W. R. Leonard and J. Rodriguez 2012 Differences between Neandertal and modern human infant and child growth models. Journal of Human Evolution 63(1):140-149. McBrearty, S. 2000 The revolution that wasn't: a new interpretation of the origin of modern human behavior. Journal of Human Evolution 39(5):453-563. McCreedy, M. 1994 The arms of the Dibouka. In Key Issues in Hunter-Gatherer Research, edited by E. S. Burch, Jr., and L. J. Ellanna, pp. 15-34. Berg, Oxford and Providence. McGraw, W. S., C. Cooke and S. Shultz 2006 Primate remains from African crowned eagle (Stephanoaetus coronatus) nests in Ivory Coast's Tai Forest: Implications for primate predation and early hominid taphonomy in South Africa. American Journal of Physical Anthropology 131(2):151-165. Meighan, C. W. 1953 Acculturation in California awl forms. Kroeber Anthropological Society Papers:61-68. 378 Mellars, P. 1996 The Neanderthal Legacy: An Archaeological Perspective from Western Europe. Princeton University Press, Princeton. 1999 The Neanderthal problem continued. Current Anthropology 40:341-364. 2000a The archaeological records of the Neanderthal-modern human transition in France. In The Geography of Neandertals and modern humans in Europe and the Greater Mediterranean, edited by O. Bar Yosef, and D. Pilbeam, pp. 35-47. Peabody Museum of Archaeology and Ethnology, Cambridge, Mass. 2000b Châtelperronian chronology and the case for Neanderthal/modern human "acculturation" in Western Europe. In Neanderthals on the Edge, edited by C. B. Stringer, R.N.E. Barton, and J.C. Finlayson, pp. 33-39. Oxbow Books, Oxford. Mellars, P. and B. Gravina 2008 Châtelperron: theoretical agendas, archaeological facts, and diversionary smoke-screens. Palaeoanthropology 2008:43-64. Mellars, P., B. Gravina and C. Bronk Ramsey 2007 Confirmation of Neanderthal/modern human interstratification at the Châtelperronian type-site. Proceedings of the National Academy of Sciences 104(9):3657-3662. Metcalfe, D. and K. T. Jones 1988 A reconsideration of animal body-part utility indices. American Antiquity 53(3):486-504. Mithen, S. J. 2005 The Singing Neanderthals. Weidenfield & Nicholson, London. Monahan, C. M. 1998 The Hadza carcass transport debate revisited and its archaeological implications. Journal of Archaeological Science 25:405-424. Mondini, M. 2005 Magnitude of faunal accumulations by carnivores and humans in the South American Andes. In From Biosphere to Lithosphere: New Studies in Vertebrate Taphonomy edited by T. O'Conner, pp. 16-24. Oxbow Books, Oxford. Morant, G. M. 1927 Studies of Palaeolithic man: a biometric study of Neanderthal skulls and of their relationships to modern races. Annals of Eugenics 2:318-381. Moirier, C. 2013 Personal communication to Clare Tolmie regarding butchery of large raptors in the Mousterian of the Grotte du Bison, Arcy-sur-Cure, June 2013. Morin, E. 2004 Late Pleistocene Population Interaction in Western Europe and Modern Human Origins: New Insights Based on the Faunal Remains from Saint-Césaire, Southwestern France. Unpublished PhD Dissertation, Department of Anthropology, University of Michigan, Ann Arbor. 379 Morin, E. 2007 Fat composition and Nunamiut decision-making: a new look at the marrow and bone grease indices. Journal of Archaeological Science 34(1):69-82. Morin, E. and V. Laroulandie 2012 Presumed symbolic use of diurnal raptors by Neanderthals. In PloS ONE 7 (3): e32856. doi:10.1371/journal.pone.0032856 Mouton, A. P. and R. Joffroy 1958 Le Gisement Aurignacien des Rois à Moutheir (Charente). CNRS, Paris. Movius, H. 1969 The Châtelperronian in French archaeology: the evidence of Arcy-surCure. Antiquity 43:111-123. Nesbitt, P. H. 1928 A Study of the Aurignacian Site of La Ruth (France). Unpublished MA Dissertation, Sociology and Anthropology, University of Chicago, Chicago. Niven, L. B. and M. G. Hill 1995 Seasonality Studies and Plains Archaic Bison Hunting. Paper presented at the 53rd Annual Plains Conference, Laramie, WY. Noss, A. J. and B. S. Hewlett 2001 The contexts of female hunting in Central Africa. American Anthropologist 103(4):1024-1040. O'Connell, J. F., K. Hawkes and N. Blurton Jones 1988 Hazda hunting, butchering and bone transport and their archaeological implications. Journal of Anthropological Research 44(2):113-161. 1996 Reanalysis of large animal body part transportation among the Hadza. Journal of Archaeological Science 17(3): 301-316. O'Connor, T. P. 1987 On the structure, chemistry and decay of bone, antler and ivory. In Archaeological Bone, Antler and Ivory, edited by K. Starling and D. Watkinson, pp 6-8. United Kingdom Institute for Conservation, Occasional Paper 5. UKIP, London. Oakes, J. 1991 Skin preparation techniques. In Copper and Caribou Inuit Skin Clothing Production, edited by J. Oakes, pp. 102-118. Canadian Museum of Civilization, Hull, Quebec. Olsen, S. J. 1964 Mammal Remains from Archaeological Sites. Peabody Museum of Archaeology and Ethnohistory, Harvard University, Cambridge. 1968 Fish, Amphibian and Reptile Remains from Archaeological Sites. Harvard University, Cambridge. 380 Otárolla-Castillo, E. 2010 Differences between NISP and MNE in cutmark analysis of highly fragmented faunal assemblages. Journal of Archaeological Science 37(1):1-12. Owen, L. 1999 Questioning stereo-typical notions of prehistoric tool functions - ethnoanalogy, experimentation and functional analysis. Urgeschichtliche Materielhelte 14:17-30. 2005 Distorting the Past: Gender and the Division of Labor in the European Upper Paleolithic. Kerns Verlag, Tubingen. Owen-Smith, N. and M. G. Mills 2008 Predator-prey size relationships in an African large-mammal food web. The Journal of Animal Ecology 77(1):173-183. Padilla, A. M. 1980 The role of cultural awareness and ethnic loyalty in acculturation. AAAS Selected Symposium 39:47-84. Palés, L. and M. Garcia 1981a Atlas Ostéologique des Mammifères du Quaternaire Les Herbivores 2: Tête, Rachis. CNRS, Paris. 1981b Atlas Ostéologique des Mammifères du Quaternaire. Les Carnivores 2: Tête, Rachis. CNRS, Paris. Palés, L. and C. Lambert 1971a Atlas Ostéologique des Mammifères du Quaternaire. Les Carnivores 1: Membres. CNRS, Paris. 1971b Atlas Ostéologique des Mammifères du Quaternaire. Les Herbivores 1: Membres. CNRS, Paris. Pasda, K. 2005 Some taphonomic investigations on reindeer (Rangifer tarandus) in West Greenland. In From Biosphere to Lithosphere: New Studies in Vertebrate Taphonomy, edited by T. O'Conner, pp. 4-15. Oxbow Books, Oxford. Pastoureau, M. 2007 L’Ours: Histoire d’un Roi Dechu. Editions Lavoisier, Paris. Patou-Mathis, M. 2000 Neanderthal subsistence behaviours in Europe. International Journal of Osteoarchaeology 10:379-395. Patou, M. 1989 Subsistence et approvisionnement au Paléolithique Moyen. In L'Homme de Néandertal Vol 6: La Subsistence, edited by M. Otte, pp. 11-18. ERAUL, Liege. 381 Pearson, O. M. 2000 Activity, climate and post-cranial robusticity: implications for modern human origins and scenarios of adaptive change. Current Anthropology 41(4):569-607. Pearson, O. M., R. M. Cordero and A. M. Busby 2008 How different were Neanderthals’ habitual activities? A comparative analysis with diverse groups of recent humans. In Neanderthals Revisited: New Approaches and Perspectives, edited by K. Harvarti, T. Harrison, pp. 89-111. Springer, New York. Pelegrin, J. and M. Soressi 2007 Le châtelperronien et ses rapports avec le mousterien. In Les Néandertaliens: Biologie et Culture, edited by B. Vandermeersch and B. Maureille, pp. 283-296. CTHS, Paris. Perkins, D., and P. Daly 1968 A hunters’ village in Neolithic Turkey. Scientific American 219(5):97-105. Peresani, M. 2011 The end of the Middle Paleolithic in the Italian Alps. In Neanderthal Lifeways, Subsistence and Technology: One Hundred and Fifty Years of Neanderthal Study, edited by N. J. Conard and J. Richter, pp. 249-259. Springer, New York. Persani, M. , I. Fiore, M. Gala, M. Romandini, and A. Tagliocozzo 2011 Late Neanderthals and the intentional removal of feathers as evidenced from bird bone taphonomy at Fumane Cave, 44,000 BP, Italy. Proceedings of the National Academy of Sciences 108(10):3888-3893. Peterkin, G. L. 2001 Early Upper Palaeolithic hunting technology and techniques in southwestern France. In Questioning the Answers: Re-Solving Fundamental Problems of the Early Upper edited by M. Hayes and P.T. Thacker, pp. 171-186. Archeopress, Oxford. Pettitt, P. B. 2000 Neanderthal lifecycles: developmental and social phases in the lives of the last archaics. World Archaeology 31(3):351-366. Pettitt, P. B. and A. W. G. Pike 2001 Blind in a cloud of data: problems with the chronology of Neanderthal extinction and anatomically modern human expansion. Antiquity 75:415-420. Peyrony, D. 1946 Le gisement préhistorique de l’Abri Cellier au Ruth, Commune de Tursac (Dordogne).Gallia 4:291-301. Phoca-Cosmetatuou, N. 2005 Bone weathering and food procurement strategies: assessing the reliability of our behavioral inferences. In From Biosphere to Lithosphere: New Studies in Vertebrate Taphonomy edited by T. O'Conner, pp. 137-145. Oxbow Books, Oxford. 382 Pike-Tay, A. 1993 Hunting in the Upper Perigordian - a matter of strategy or expedience? In Before Lascaux : The Complex Record of the Early Upper Paleolithic., edited by H. Knecht, A.Pike-Tay, and .White, pp. 85-99. CRC Press, Boca Raton. 1995 Variability and synchrony of the seasonal indicators in dental cementum microstructure of the Kaminuriak caribou population. Archaeofauna 4:273-284. Pike-Tay, A., C. A. Marcom and M. O'Farrell 2000 Reconsidering the quadratic crown height model of age estimation for Rangifer from archaeological sites. Archaeozoologia 11(1-2):145-174. Ponce de Leon, M. S., L. Golovanova, V. Doronichev, G. Romanova, T. Akazawa, O. Kondo, H. Ishida and C. P. E. Zollikofer 2008 Neanderthal brain size at birth provides insights into the evolution of human life history. Proceedings of the National Academy of Sciences 105(37):13764-13768. Power, R. C., M. J. Walker, D. C. Salazar-García and A. G. Henry 2012 Neanderthal Plant Food Consumption and Environment Use at Sima de las Palomas, Southeastern Iberia. Paper presented at the Palaeoanthropology Society Meeting, Memphis, Tennessee. Prichard, A. K., G. L. Finstad and D. H. Shain 1999 Factors affecting velvet antler weights in free-ranging reindeer in Alaska. Rangifer 19(2):71-76. Radloff, F. G. T. and J. T. du Toit 2004 Large predators and their prey in a southern African savanna: a predator's size determines its prey size range. Journal of Animal Ecology 73:410-423. Ramseyer, D. (editor) 2004 Fiches de la Commission de Nomenclature sur l’Industrie de l'Os Préhistorique Cahier XI: Matières et Technique. Editions de la Société Préhistorique Française, Paris. Raynal, J.-P. and J.-L. Gaudelli 1990 Milieux physiques et biologiques: quels changements entre 60,000 et 30,000 ans à l'ouest de l'Europe. In Paléolithique Moyen Recent et Paléolithique Superiur Ancien en Europe, edited by C. Farizy, pp. 53-61. Musee de L'Ile de France, Nemours. Reed, D. 2005 Taphonomic implications of roosting behavior and trophic habits in two species of African owl. Journal of Archaeological Science 32(11):1669-1676. Reitz, E. J. and E. S. Wing 1999 Zooarchaeology. Cambridge Manuals in Archaeology. Cambridge University Press, Cambridge. Rendu, W. 2010 Hunting behavior and Neanderthal adaptability in the Late Pleistocene site of Pech-de-l'Azé I. Journal of Archaeological Science 37(8):1798-1810. 383 Rhodes, J. and S. Churchill 2009 Throwing in the Middle and Upper Paleolithic: inferences from an analysis of humeral retroversion. Journal of Human Evolution 56(1):1-10. Richards, M. P. and R. W. Schmitz 2008 Isotope evidence for the diet of the Neanderthal type specimen. Antiquity 82:533-559. Richards, M. P., G. Taylor, T. Steele, S. P. McPherron, M. Soressi, J. Jaubert, J. Orschiedt, J. B. Mallye, W. Rendu and J. J. Hublin 2008 Isotopic dietary analysis of a Neanderthal and associated fauna from the site of Jonzac (Charente-Maritime), France. Journal of Human Evolution 55(1):179-185. Riel-Salvatore and G. A. Clark 2001 Grave markers: Middle and Early Upper Palaeolithic burials and the use of chronotypology in contemporary Palaeolithic research. Current Anthropology 42(4):449-479. Riel-Salvatore, J., G. Popescu and C. Barton 2008 Standing at the gates of Europe: Human behavior and biogeography in the Southern Carpathians during the Late Pleistocene. Journal of Anthropological Archaeology 27(4):399-417. Robson, S. L. and B. Wood 2008 Hominin life history: reconstruction and evolution. Journal of Anatomy 212(4):394-425. Roebroeks, W. 2008 Time for the Middle to Upper Paleolithic transition in Europe. Journal of Human Evolution 55(5):918-926. Rogers, J. D. 1993 The social and material implications of culture contact on the Northern Plains. In Ethnohistory and Archaeology: Approaches to Postcontact Change in the Americas, edited by J. D. Rogers and S. M. Wilson, pp. 73-88. Plenum, New York. Rowley Conwy, P. 2001 Time, change and the archaeology of hunter gatherers: how original is the "Original Affluent Society?". In Hunter Gatherers: An Interdisciplinary Perspective, edited by C. Painter Brick, R.H. Layton, and P. Rowley Conwy, pp. 39-72. Cambridge University Press, Cambridge. Sandgathe, D. M., H. L. Dibble, P. Goldberg and S. P. McPherron 2011 The Roc de Marsal Neandertal child: a reassessment of its status as a deliberate burial. Journal of Human Evolution 61(3):243-253. Scheinsohn, V. 2010 Hearts and Bones: Bone Raw Material Exploitation in Tierra de Fuego. BAR International Series 2094, Archaeopress. Oxford. 384 Scheinsohn, V. and J. L. Ferretti 1995 The mechanical properties of bone materials in relation to the design and function of prehistoric tools from Tierra del Fuego, Argentine. Journal of Archaeological Science 22:7-11. Schiffer, M. B. 1987 Formation Processes of the Archaeological Record. University of New Mexico Press, Albuquerque. Schmider, B. 2002 Avant-propos. In L'Aurignacien de la Grotte du Renne: Les Fouilles d'Andre Leroi-Gourhan a Arcy-sur-Cure (Yonne). edited by B. Schmider, pp. 7-9. CNRS Editions, Paris. Schmitt, D. 2003 Experimental evidence concerning spear use in Neandertals and Early Modern humans. Journal of Archaeological Science 30(1):103-114. Schmitt, D. N. and K. E. Juell 1994 Toward the identification of coyote scatalogical faunal accumulations in archaeological contexts. Journal of Archaeological Science 21:249-262. Schmitt, D. N. and K. D. Lupo 1995 On mammalian taphonomy, taxonomic diversity and measuring subsistence data. American Antiquity 60(3):496-514. Schuppli, C., K. Isler and C. P. van Schaik 2012 How to explain the unusually late age at skill competence among humans. Journal of Human Evolution 63(6):843-850. Schuyler, R. L. 1998 Culture contact in evolutionary perspective. In Studies in Culture Contact: Interaction, Culture Change, and Archaeology, edited by J. G. Cusick, pp. 67-76. Center for Archaeological Investigations Occasional Paper No 25, University of Southern Illinois, Carbondale. Sear, R. and R. Mace 2008 Who keeps children alive? A review of the effects of kin on child survival. Evolution and Human Behavior 29(1):1-18. Semenov, S. A. 1964 Prehistoric Technology. Translated by M. W. Thompson. Cory, Adams & MacKay, London. Serre, D., A. Langaney, M. Chech, M. Teschler-Nicola, M. Paunovic, P. Mennecier, M. Hofreiter, G. Possnert and S. Paabo 2004 No evidence of Neanderthal mtDNA contribution to early modern humans. PLoS Biology 2(3):313-317. Shaw, C. N., C. L. Hofmann, M. D. Petraglia, J. T. Stock and J. S. Gottschall 2012 Neandertal humeri may reflect adaptation to scraping tasks, but not spear thrusting. PloS ONE 7(7): e40349. doi:10.1371/journal.pone.0040349. 385 Shea, J. 1998 Neanderthal and early modern human behavioral variability. Current Anthropology 39(Supplement):45-78. 2007 The boulevard of broken dreams: evolutionary discontinuity in the late Pleistocene Levant. In Rethinking the Human Revolution, edited by P. Mellars, K. Boyle, O. Bar-Yosef and C. Stringer, pp. 219-232. McDonald Institute Monographs. McDonald Institute for Archaeological Research, Cambridge. Shortman, E. M. and P. A. Urban 1998 Culture contact, structure and process. In Studies in Culture Contact: Interaction, Culture Change and Archaeology, edited by J. G. Cusick, pp. 102125. Center for Archaeological Investigations Occasional Paper No 25. Southern Illinois University, Carbondale, Illinois. Shott, M. J. 1991 Archaeological implications of revisionism in ethnography. In Foragers in Context, edited by P. T. Miracle, L. E. Fisher and J. H. Brown, pp. 31-40. Michigan Discussions in Anthropology Volume 10. University of Michigan, Ann Arbor. 2004 Hunter gatherer aggregation in theory and evidence: the eastern North American Paleoindian case. In Hunters and Gatherers in Theory and Archaeology, edited by G. M. Crothers, pp. 68-102. Center for Archaeological Investigations Occasional Paper No 31, Southern Illinois University, Carbondale. Silberbauer, G. B. 1996 Neither your ways are my ways. In Cultural Diversity Among Twentieth Century Foragers, edited by S. Kent, pp. 21-64. Cambridge University Press, Cambridge. Simek, J. F. and L. M. Snyder 1988 Changing assemblage diversity in Perigord archaeofaunas. In Upper Pleistocene Prehistory of Western Eurasia, edited by H. Dibble and A. MontetWhite, pp. 321-332. The University Museum, University of Pennsylvania, Pittsburgh. Sisk, M. L. 2011 Settlement and Site Location in the Middle and Upper Paleolithic of the Vézère Valley, France. Unpublished PhD Dissertation, Department of Anthropology, State University of New York, Stoney Brook. Slimak, L. and Y. Giraud 2007 Circulations sur plusieurs centaines de kilomètres durant le Paléolithique moyen. Contribution à la connaissance des sociétés néandertaliennes. Comptes Rendus Palevol 6(5):359-368. Soffer, O. 2004 Recovering perishable technologies through use-wear on tools: preliminary evidence for Upper Palaeolithic weaving and net making. Current Anthropology 45(3):407-413. 386 Soffer, O., J. M. Adovasio and D. C. Hyland 2000a The "Venus" figurines: textiles, basketry, gender and status in the Upper Paleolithic. Current Anthropology 41(4):511-537. Soffer, O., J. M. Adovasio, J. S. Illingworth, H. A. Amirkhanov, N. D. Praslov and M. Street 2000b Palaeolithic perishables made permanent. Antiquity 74:812-821. Sorensen, M. 2001 Neandertal energetics and foraging efficiency. Journal of Human Evolution 40(6):483-495. Spector, J. 1991 What this awl means: towards a feminist archaeology. In Engendering Archaeology: Women and Prehistory, edited by J. M. Gero and M. W. Conkey, pp 388-406. Basil Blackwell, Oxford. Speth, J. D. 2004 News flash: negative evidence convicts Neanderthals of gross mental incompetence. World Archaeology 36(4):519-526. 2012 Middle Palaeolithic subsistence in the Near-East: zooarchaeaological perspectives past, present and future. Before Farming 2012(2):1-45. Speth, J. D. and K. A. Spielman 1983 Energy source, protein metabolism and hunter-gatherer subsistence strategies. Journal of Anthropological Archaeology 2:1-31. Stallibrass, S. 1982 The use of cement layers for absolute ages of mammalian teeth: a selective review of the literature with suggestions for further studies and alternative applications. In Ageing and Sexing Animal Bones from Archaeological Sites, edited by B. Wilson, C. Grigson, and S. Payne, pp. 109-206. British Archaeological Reports, Oxford. Steele, T. E. 2002 Red deer: Their Ecology and How They Were Hunted by Late Pleistocene Hominids in Western Europe. Unpublished PhD. Dissertation, Department of Anthropological Sciences, University of Stanford, Palo Alto. Steele, T. and T. D. Weaver 2002 The modified triangular graph: a refined method for comparing mortality profiles in archaeological samples. Journal of Archaeological Science 29(3):317322. Steinbring, J. 1966 The manufacture and use of bone defleshing tools. American Antiquity 31(4):575-581. Stenton, D. R. 1991 The adaptive significance of Caribou winter clothing for arctic huntergatherers. Etudes Inuit/Inuit Studies 15:3-28. 387 Stettler, H. 2000 Upper Paleolithic transitions: evidence from organic artifacts of Cantabrian Spain. Journal of Anthropological Research 56(1):112-128. Stiner, M. C. 1990 The uses of mortality patterns in archaeological studies of human predatory adaptations. Journal of Archaeological Anthropology 9:305-351. 1991a Food procurement and transport by human and non-human predators. Journal of Archaeological Science 18:445-482. 1991b A taphonomic perspective on the origins of the faunal remains at Grotta Guitari (Latium, Italy). Current Anthropology 32:103-117. 1993a The place of hominids among predators: intraspecific comparisons of food procurement and transportation. In From Bones to Behavior: Ethnoarchaeological and Experimental Contributions to the Interpretation of Faunal Remains, edited by J. M. Hudson, pp. 38-61. Center for Archaeological Investigations, Occasional Paper No21. Southern Illinois University Press, Carbondale. 1993b Small mammal exploitation and its relation to hunting, scavenging and gathering in the Italian Mousterian. In Hunting and Animal Exploitation in the Later Palaeolithic and Mesolithic of Eurasia, edited by G. L. Peterkin, H. Bricker and P. Mellars, pp. 101-119. Archaeological Papers of the American Anthropological Association. American Anthropological Association, Washington. 1994 Honor Among Thieves: A Zooarchaeological Study of Neanderthal Ecology. Princeton University Press, Princeton. 1998 Comment on Marean et al. Current Anthropology 39(2):S100-S103. 2002a Carnivory, coevolution and the geographic spread of the genus Homo. Journal of Archaeological Research 10(1):1-63. 2002b On in situ attrition and vertebrate body part profiles. Journal of Archaeological Science 29(9):979-991. Stiner, M.C., S.L. Kuhn and E. Güleç 2013 Early Upper Paleolithic shell beads at Üçağizli Cave I (Turkey): technology and the socioeconomic context of ornament life-histories. Journal of Human Evolution 64: 380-398. Stoneking, M. and R. L. Cann 1989 African origins of human mitochondrial DNA. In The Human Revolution edited by P. Mellars and C. Stringer, pp. 17-30. University of Edinburgh Press, Edinburgh. 388 Straus, L.G. 2002 Notes on the role of American archaeologists in the study of the European Upper Paleolithic. In The Role of American Archaeologists in the Study of the European Upper Paleolithic, edited by L. G. Straus, pp 54-57. British Archaeological Reports International Series 1048, Oxbow Books, Oxford. Straus, W. L. and A. J. E. Cave 1957 Pathology and the posture of Neanderthal man. Quarterly Review of Biology 32(4):348-369. Stringer, C. B. 2008 The Neanderthal-H.sapiens interface in Eurasia. In Neanderthals Revisited: New Approaches and Perspectives, edited by K. Harvarti and T. Harrison, pp. 315-324. Springer, New York. Stringer, C.B. and P. Andrews 1988 Genetic and fossil evidence for the origin of modern humans. Science 239:1263-1268. Stringer, C.B. and C. S. Gamble 1993 In Search of the Neanderthals. Thames & Hudson, London. Svoboda, J. A. 2006 The Aurignacian and after: chronology, geography and cultural taxonomy in the Middle Danube region. In Towards a Definition of the Aurignacian : Proceedings of the Symposium Held in Lisbon, Portugal, June 25-30, 2002, edited by O. Bar-Yosef and J. Zilhão, pp. 259-274. Trabalhos de arqueologia Instituto Português de Arqueologia, Lisbon. 2007 On modern human penetration into northern Eurasia: the multiple advances hypothesis. In Rethinking the Human Revolution, edited by P. Mellars, K. Boyle, O. Bar-Yosef, and C. Stringer, pp. 329-339. McDonald Institute Monographs. McDonald Institute for Archaeological Research, Cambridge. Symmons, R. 2005 Bone density variation between similar animals and density variation in early life: implications for future taphonomic analysis. In From Biosphere to Lithosphere: New Studies in Vertebrate Taphonomy edited by T. O'Conner, pp. 87-94. Oxbow Books, Oxford. Tanner, N. M. 1981 On Becoming Human. Cambridge University Press, Cambridge. Tartar, E. 2009 De l’Os à l’Outil - Caractérisation Technique, Economique et Sociale de l’Utilisation de l’Os à l’Aurignacien Ancien. Étude de Trois sites : l’Abri Castanet (Secteurs Nord et Sud),Brassempouy (Grotte des Hyènes et Abri Dubalen) et Gatzarria. Unpublished PhD Dissertation, Department of PréhistoireEthnologie Anthropologie, Paris I Pantheon-Sorbonne, Paris. 2012 The recognition of a new type of bone tools in Early Aurignacian assemblages: implications for understanding the appearance of osseous technology in Europe. Journal of Archaeological Science 39(7):2348-2360. 389 Tartar, E., N. Teyssandier, F. Bon and D. Liolios 2006 Equipement de chasse, equipement domestique: une distinction efficace? Reflexion sur la notion d'investissement technique dans les industries aurignaciennes. In Normes Techniques et Pratiques Sociales de la Smplicité des Outillages Pré- et Protohistoriques, edited by L. Astruc, F. Bon, V. Lea, P-Y Milcent, and S. Philibert, pp. 107-117. Rencontres Internationals d'Archéologie ed d'Histoire d'Antibes. APDCA, Antibes. Tejero, J.-M., M. Christensen and P. Bodu 2012 Red deer antler technology and early modern humans in Southeast Europe: an experimental study. Journal of Archaeological Science 39(2):332-346. Texier, J.-P., C. Lemorinin, J.-P. Brugal and L. Wilson 1996 Une activité de tratitement des peaux dans l'habitat mousterien de La Combette (Bonnieux, Vaucluse, France). Quatarnaria Nova VI:369-372. Teyssandier, N. 2007 En Route Vers l'Ouest: les Debuts de l'Aurignacien en Europe. BAR International Series 1638. John & Erica Hodges, Ltd, Oxford. Todd, L. C. 1991 Seasonality and Paleoindian subsistence strategies. In Human Predator and Prey Mortality, edited by M. C. Stiner. Westview Press, Boulder. Tolmie, C. 2009 Faunal remains from Abri Cellier, France. Paper presented at the 74th Annual Meeting of the Society for American Archaeology, Atlanta, GA. in press The Châtelperronian: hybrid culture or independent innovation. In The Archaeology of Hybrid Material Culture, edited by J.J. Card. pp. 279-294. Center for Archaeological Investigations Occasional Paper No. 39, Southern Illinois Univerisity, Carbondale. Torrence, R. 2001 Technology: macroscale and microscale approaches. In Hunter Gatherers: An Interdisciplinary Approach, edited by C. Painter Brick, Robert H. Layton, Peter Rowley Conway, pp. 73-98. Cambridge University Press, Cambridge. Tostevin, G. B. 2007 Social intimacy, artifact visibility and acculturation models of Neanderthal-modern human interaction. In Rethinking the Human Revolution, edited by P. Mellars, K.Boyle, O. Bar-Yosef, and C. Stringer, pp. 341-357. McDonald Institute Monographs. McDonald Institute for Archaeological Research, Cambridge. Trigger, B. D. 1989 A History of Archaeological Thought. Cambridge University Press, Cambridge. Trinkaus, E. 1983 Neanderthal post-crania and the adaptive shift to modern humans. . In The Mousterian Legacy, edited by E. Trinkaus, pp 165-200. BAR International Series 164, British Archaeological Reports, Oxford. 390 Trinkaus, E. 1987 Neanderthal upper limb morphology and manipulation. In Hominidae: Proceedings of the 2nd International Congress of Human Palaeontology, edited by G. Giacobini, pp. 331-338. Jaca Book, Milan. 2011 Late Pleistocene adult mortality patterns and modern human establishment. Proceedings of the National Academy of Sciences 108(4):12671271. 2012 Neandertals, early modern humans, and rodeo riders. Journal of Archaeological Science 39(12):3691-3693. Trinkaus, E., S. E. Churchill, I. Villemeur, K. G. Riley, J. A. Heller and C. B. Ruff 1991 Robusticity "versus" shape: the functional interpretation of Neandertal appendicular morphology. Zinruigaku Zassi 99(3):257-278. Trinkaus, E. and P. Shipman 1993a The Neanderthals. Knopf, New York. 1993b Neandertals: images of ourselves. Evolutionary Anthropology 2(6):194201. Vallverdu, J., M. Vaquero, I. Caceres, E. Allue, J. Rosell, P. Saladié, C. Gema, A. Olle, A. Canals, R. Sala, M. A. Courty and E. Carbonell 2010 Sleeping activity area within the site structure of archaic human groups: evidence from Abric Romai, Level N combustion activity areas. Current Anthropology 51(1):137-145. van Andel, T. 2002 The climate and landscape of the middle part of the Weichselian Glaciation in Europe: The Stage 3 Project. Quaternary Research 57(1):2-8. von Fürer-Haimendorf, C. 1943 The Chenchus: Jungle Folk of the Deccan. The Aboriginal Tribes of Hyderabad. MacMillan& Co., London. Waguespack, N. M. 2005 The organization of male and female labor in foraging societies: implications for early Paleoindian archaeology. American Anthropologist 107(4):666-676. Wales, N. 2012 Modeling Neanderthal clothing using ethnographic analogues. Journal of Human Evolution 63(6):781-795. Walker, R. 2002 Age-dependency in hunting ability among the Ache of Eastern Paraguay. Journal of Human Evolution 42(6):639-657. Washburn, S. L. and C. S. Lancaster 1968 The evolution of hunting. In Man the Hunter, edited by R. B. Lee, I. De Vore, pp. 293-308. De Gruyter, New York. 391 Weaver, T. D. 2009 The meaning of Neandertal skeletal morphology. Proceedings of the National Academy of Sciences 106(38):16028-16033. Weaver, T. D. and C. C. Roseman 2005 Ancient DNA,late Neandertal survival, and modernhuman–Neandertal genetic admixture. Current Anthropology 46(4):677-683. Weaver, T. D. and K. Steudel-Numbers 2005 Does climate or mobility explain the differences in body proportions between Neandertals and their Upper Paleolithic successors? Evolutionary Anthropology: Issues, News, and Reviews 14(6):218-223. Weidenreich, F. 1943 The "Neanderthal Man" and the ancestors of "Homo sapiens". American Anthropologist 45(1):39-48. Weigelt, J. 1989 Recent Vertebrate Carcasses and Their Paleobiological Implications. Translated by Judith Schreiber. University of Chicago Press, Chicago. (Originally published Rezente Wirbeltierleichen und ihre paläobiologische Bedeutung. Verlag von Mag Weg, Leiden 1927). Wells, J. C. K. 2012 The Capital Economy in Hominin Evolution. Current Anthropology 53(S6):S466-S478. Whallon, R. 2006 Social networks and information: non-"utilitarian" mobility among huntergatherers. Journal of Anthropological Archaeology 25:259-270. White, M. J. 2006 Things to do in Doggerland when you're dead: surviving OIS3 at the northwestern-most fringe of Middle Palaeolithic Europe. World Archaeology 38(4):547-575. White, M. J. and P. B. Pettitt 2011 The British late Middle Palaeolithic: an interpretative synthesis of Neanderthal occupation at the northwestern edge of the Pleistocene world. Journal of World Prehistory 24(1):25-97. White, R. 1985 Upper Palaeolithic Land Use in the Perigord: A Topographic Approach to Subsistence and Settlement. British Archaeological Reports International Series 253, British Archaeological Report., Oxford. 1995 Ivory personal ornaments of Aurignacian age: technological, social and symbolic perspectives. In Le Travail et l'Usage d'Ivoire au Paléolithique Supérieur, edited by J. Hahn, M. Menu, Y. Taborin, Ph. Walter, F. Widemann, pp. 29-62. Actes de la table ronde, Ravello, 29-31 Mai 1992. Istituto poligrafico e Zecca dello Stato, Libreria dello Stato, Rome. 1998 Comment on d'Errico et al. Current Anthropology 39:S531. 392 White, R. 2001 Personal ornaments from the Grotte du Renne, Arcy sur Cure. Athena Review 2:41-46. 2002 The historic and legal context of foreign acquisitions of Palaeolithic artifacts from the Perigord: 1900-1941. In The Role of American Archaeologists in the Study of the European Upper Paleolithic, edited by L. G. Straus, pp. 71-83. BAR International Series 1048. Oxbow Books, Oxford. 2009 Personal communication to Clare Tolmie regarding the representative nature of early excavated tool assemblages from Castanet Nord, June 2009. White, R,. and L.B. Breitbord, editors 1992 French Paleolithic Collections in the Logan Museum of Anthropology, edited by R. White and L. B. Breitbord, Logan Museum Bulletin (New Series). Vol 1(2). Museums of Beloit College, Beloit, Wisconsin. White, R. and H. Knecht 1992 The Abri Cellier (or La Ruth), Commune de Tursac, (Dordogne, France): results of the 1927 Beloit College excavations. . In French Paleolithic Collections in the Logan Museum of Anthropology, edited by R. White and L. B. Breitbord, pp. 39-96. Logan Museum Bulletin (New Series). Vol 1(2). Museums of Beloit College, Beloit, Wisconsin. Wiessner, P. 1983 Style and social information in Kalahari San projectile points. American Antiquity 48(2):253-276. Wilson, S. M. and J. D. Rogers 1993 Historical dynamics in the contact era. In Ethnohistory and Archaeology: Approaches ot Postcontact Change in the Americas, edited by S. M. Wilson and J. D. Rogers, pp. 3-18. Plenum, New York. Winterhalder, B. and E. A. Smith (editors) 1981 Hunter-Gatherer Foraging Strategies: Ethnographic and Archaeological Analyses. University of Chicago Press, Chicago. Wismer, M. 2009 A New Analysis of Bison Subsistence at Cherokee Sewer, Iowa. Paper presented at the 74th Annual Meeting of the Society for American Archaeology, Atlanta, GA. Wobst, M. H. 1977 Stylistic behavior and information exchange. In For the Director: Research Papers in Honor of James B. Griffin, edited by C. B. Cleland, pp. 317342. Anthropology Papers vol 16. Museum of Anthropology, University of Michigan, Ann Arbor. Wood, R.E., C. Barroso Ruiz, M. Caparrós, J.F. J. Pardo, B.G. Santos, and T. F. G. Higham 2013 Radiocarbon dating casts doubt on the late chronology of the Middle to Upper Palaeolithic transition in southern Iberia. Proceedings of the National Acadamy of Science 110(8): 2781-2786. 393 Woods, A. 2011 The Effects of Lithic Raw Material Quality on Aurignacian Blade Production at Abri Cellier. Unpublished PhD Dissertation, Department of Anthropology, University of Iowa, Iowa City. Wylie, A. 1991 Gender theory and the archaeological record: why is there no archaeology of gender? In Engendering Archaeology, edited by J. M. Gero, and M. Conkey, pp. 31-54. Blackwell, Oxford. Yellen, J. E. 1977 Cultural patterning in faunal remains: evidence from the !Kung bushmen. In Experimental Archaeology, edited by D. Ingersoll, J.E. Yellen, W. McDonald, pp. 271-331. Columbia University Press, New York. Zilhão, J., D. E. Angelucci, E. Badal-Garcia, F. d'Errico, F. Daniel, L. Dayet, K. Douka, T. F. G. Higham, M. J. Martinez-Sanchez, R. Montes-Bernardez, S. Murcia-Mascaros, C. Perez-Sirvent, C. Roldan-Garcia, M. Vanhaeren, V. Villaverde, R. Wood and J. Zapata 2010 Symbolic use of marine shells and mineral pigments by Iberian Neandertals. Proceedings of the National Academy of Sciences 107(3):1023-1028. Zilhão, J. and F. d'Errico 1999 The chronology and taphonomy of the earliest Aurignacian and its implications for the understanding of Neandertal extinction. Journal of World Prehistory 13(1):1-68. Zilhão, J., F. d'Errico, J.-G. Bordes, A. Lenoble, J.-P. Texier and J.-P. Rigaud 2008a Grotte des Fées (Châtelperron): history of research, stratigraphy, dating and archaeology of the Châtelperronian type-site. Palaeoanthropology 2008:1-42. Zilhão, J., F. d'Errico, J.-G. Bordes, A. Lenoble, J.-P. Texier and J.-P. Rigaud 2008b Like Hobbes' chimney birds: response to Mellars and Gravina. Palaeoanthropology 2008:65-67. Zilhão, J., F. d'Errico, J.-G. Bordes, A. Lenoble, J.-P. Tixier and J.-P. Rigaud 2006 Analysis of Aurignacian interstratification at the Châtelperronian-type site and implications for the behavioral modernity of Neandertals. Proceedings of the National Academy of Sciences 103(33):12643-12648. Zollikofer, C. P., M. S. Ponce De Leon, B. Vandermeersch and F. Leveque 2002 Evidence for interpersonal violence in the St. Césaire Neanderthal. Proceedings of the National Academy of Sciences of the United States of America 99(9):6444-6448.

RELATED PAPERS

RELATED TOPICS

Log In

Animals for food, animals for tools: fauna as a source of raw material at Abri Cellier, Dordogne, and the Grotte du Renne, Arcy-sur-Cure

Animals for food, animals for tools: fauna as a source of raw material at Abri Cellier, Dordogne, and the Grotte du Renne, Arcy-sur-Cure

Related Papers

RELATED PAPERS

RELATED TOPICS