COMMENTARY

Neanderthal symbolism and ornament manufacture: The bursting of a bubble?

Paul Mellars Department of , Cambridge University, Cambridge CB2 3DZ, United Kingdom

or the past 40 y, the range of the Châtelperronian industries themselves, grooved and perforated animal and in particular on their demonstrably F teeth, mammoth-ivory rings, and strong links with the final Mousterian other unmistakably “decorative” technologies of western France belonging or “ornamental” items discovered during to the so-called “Mousterian of Acheulian the excavations in the Châtelperronian tradition” group, for which the direct levels in the cave of Grotte du Renne in associations have never been north central France (1–4) (Fig. 1) have seriously contested (22). Even without the played a pivotal role in discussions of Fig. 1. Grooved and perforated “personal orna- support of the disputed skeletal remains the European Neanderthal populations, ments” and ivory ring from the Châtelperronian from the Châtelperronian levels at both suggesting that these populations were levels at the Grotte du Renne, France. [Repro- the Grotte du Renne and Saint-Césaire, actively engaged in a range of highly duced with permission from Randall White, New few authors now dispute that the makers York University, New York.] “symbolic” behaviors, which had pre- of the Châtelperronian industry in western viously been generally regarded as cultur- Europe were indeed the direct descend- ally and behaviorally diagnostic of the ern populations in western Europe (5–7, ants of the immediately preceding Homo sapiens Neanderthal populations, who are known earliest populations, who 15–19). Although for technical and ad- to have occupied effectively all regions of are known to have dispersed rapidly across ministrative reasons it was not possible to Europe (from their ultimately African Europe for the preceding 200,000 to secure any dates directly on the orna- origins) between ca. 45,000 and 35,000 y 250,000 y (7, 10–12, 17–24). mental items themselves, the results raise ago (5–9). Despite being effectively unique The question of whether the (final the strong possibility—if not probability— in the archaeological records of the Neanderthal) Châtelperronian popula- that they were stratigraphically intrusive European , the finds have tions ever manufactured the hauntingly into the Châtelperronian deposits from often been taken to demonstrate that the “modern”-looking animal-tooth pendants, these overlying Proto-Aurignacian levels— Neanderthals possessed highly symbolic either at the Grotte du Renne or at any an option already suggested by a number cultural and cognitive capacities (including other sites in western Europe, is perhaps of earlier workers on stratigraphic grounds language) closely similar if not identical to more debatable. The only other claims for those of the ensuing “modern” (2). One could perhaps raise similar animal-tooth pendants (or any similar populations (10–13). questions as to the provenance of some of decorative items) in association with ap- The PNAS article by Higham et al. (14) the diagnostically Neanderthal skeletal proximately 40 known Châtelperronian has now thrown substantial doubt on these remains (including 29 teeth and a tempo- sites in France and northern Spain come anatomical and cultural associations at the ral bone) recovered from the same Châ- from the site of Quinçay (2, 10, 24), where Grotte du Renne site, based on an in- telperronian levels (3, 4), but on the basis [as Higham et al. (14) point out] the as- tensive campaign of radiocarbon acceler- of the dated radiocarbon samples, there is sociated archaeological and stratigraphic ator dating measurements by the highly much less evidence for intermixture of evidence has never been adequately pub- “ respected Oxford Radiocarbon Accelera- material from the underlying Mouste- lished, and where the objects in question ” tor Unit. These measurements demon- rian levels in the sequence (with the ex- might again be intrusive from the overlying ± strate that the archaeological and faunal ception of a single date of 48,700 3,600 y levels on the site. On a more positive note, material originally recovered from the B.P.), which makes the doubts sometimes perhaps, the Oxford team did succeed in critical (presumed final Neanderthal) raised as to the provenance of the remains securing clear radiocarbon measurements Châtelperronian levels span an extraordi- (19) much less plausible on both strati- on two morphologically simple but un- narily wide range of ages (ranging between graphic and radiocarbon grounds. mistakably shaped bone awls from the ca. 49,000 and 21,000 y B.P. in radiocar- Exactly what implications these results Châtelperronian levels at Grotte du bon terms), which can only be explained by have for our current perceptions of the Renne, which demonstrate unambiguously a substantial degree of stratigraphic mixing nature of any social or cultural relation- that the Châtelperronian groups were of materials from several adjacent ar- ships between the final Neanderthal pop- involved in the manufacture of these chaeological levels that occurred either ulations of western Europe and the simple, shaped bone tools from at least during the original excavations on the site incoming anatomically modern human ca. 37,000 y B.P. onward, arguably making in the 1950s or as a result of earlier geo- populations is, perhaps, more open to these tools the earliest clear evidence for logical or human disturbance of the de- debate. As several authors have pointed systematically shaped bone tool manu- posits. As the dates themselves reveal, out (20–23), the case for thinking that facture in western Europe, and appar- a significant proportion of the newly dated the highly distinctive Châtelperronian in- ently surpassing anything demonstrably bone samples from these critical Châ- dustries were manufactured by the final produced by the preceding Neander- telperronian levels almost certainly derive Neanderthal populations in western thals (11, 12). from the immediately overlying “Proto- Europe hinges not only on the Neander- Aurignacian” level on the site, now re- thal skeletal remains recovered from the liably dated by the new radiocarbon Grotte du Renne (and those from the Author contributions: P.M. wrote the paper. measurements to approximately 35,000 y equally contested site of Saint-Césaire in The author declares no conflict of interest. B.P., and generally agreed to be the southwestern France), but also on several See companion article on page 20234. product of the earliest anatomically mod- of the specific archaeological features of E-mail: [email protected].

www.pnas.org/cgi/doi/10.1073/pnas.1014588107 PNAS | November 23, 2010 | vol. 107 | no. 47 | 20147–20148 Downloaded by guest on September 29, 2021 In the light of all this new evidence, the from the Grotte du Renne is that the sin- language (25). One could equally ask in debates over the cultural capacities of the gle most impressive and hitherto widely this context why a population that had late Neanderthal populations in Europe, cited pillar of evidence for the presence of become so thoroughly and successfully and their social and cultural interactions complex “symbolic” behavior among the “adapted” to the glacial conditions of with the incoming anatomically modern late Neanderthal populations in Europe Europe over a span of more than 200,000 y populations, will no doubt rumble on. Two has now effectively collapsed. Whether any should have succumbed so rapidly (within scenarios have been postulated in the further evidence of advanced, explicitly a period of a few thousand years) to de- fi earlier literature. The rst is that the late mographic competition from the newly Neanderthal populations across Europe Evidence for the presence arrived modern human populations, who could conceivably have independently had developed all of their biological, “invented” several features of distinctively of complex “symbolic” technological, and other cultural adapta- “modern” human culture—including the tions in the vastly different environments production of both highly shaped bone behavior among the tools and explicitly “symbolic” ornamental of sub-Saharan Africa (8, 9). If the evolu- items—without any connection or cultural late Neanderthal tionary lineages that led to the Neander- input from the incoming modern human thal populations in Europe and the populations, who are known to have been populations in Europe anatomically and genetically modern dispersing progressively across Europe at (H. sapiens) populations in Africa initially precisely the same time as the allegedly has now effectively diverged at least 350,000 y ago [as all of “independent” Neanderthal innovations the recent genetic evidence strongly sug- (10–13), a scenario that has been de- collapsed. gests (8, 9)], this would amount cumula- scribed elsewhere as a seemingly “impos- tively to at least 700,000 y of separate sible coincidence” (23). The second evolutionary development in the two scenario is that certain features of dis- symbolic behavior of this kind can be re- regions—arguably providing ample time tinctively modern human culture and liably claimed from any other Neanderthal for significantly different cognitive and technology could simply have been dis- sites in Europe is still a matter of debate other biological adaptations and associ- persed by interpopulation cultural con- (13, 21). One crucial question that must ated genetic mutations to have occurred tacts and exchange mechanisms between inevitably be posed in this context is why, if within the two geographically and de- fi the use of explicitly symbolic behavior was the nal Neanderthals and incoming mographically separated populations an integral part of the cultural and be- modern human groups during the actual [a conclusion already strongly hinted at havioral repertoire of the European process of modern human expansion from in the draft Neanderthal genome study east to west across the continent (5, 10, 18, Neanderthals, there is so little actual (or reported recently by Green et al. (9)]. 21, 23)—a process that could be described even claimed) evidence for this across the Clearly, one should be cautious about as a cultural “bow-wave diffusion” or “ac- 250,000-y time span of the Neanderthal culturation” model (23). occupation in Europe, extending across pressing the evolutionary and cognitive These debates will no doubt continue, a wide range of sharply contrasting envi- implications of all this too far. But one especially in the wake of the recently ronments, and over a geographical span must confront the possibility that the hu- claimed genetic evidence for a small of more than 2,000 miles (21). This in turn man evolutionary implications of the new degree (approximately 1–4%) of de- raises questions as to not only the actual dating of the Grotte du Renne sequence mographic and genetic interbreeding cultural repertoire of the European are potentially even more significant and between the Neanderthals and the earliest Neanderthals, but also (inevitably) their far-reaching than the (admirably cau- genetically modern populations in Eurasia innate cultural and cognitive capacities for tiously expressed) conclusions advanced (9). However, the central and inescap- advanced symbolic thinking, including, in the paper by Higham et al. (14) able implication of the new dating results perhaps, the capacities for fully developed would suggest.

1. Taborin Y (1998) Les prémices de la parure. Paléolithi- 10. D’Errico F, et al. (1998) Neanderthal acculturation in 18. Stringer C, Gamble C (1993) In Search of the Neanderthals que Moyen Récent et Paléolithique Supérieur Ancien western Europe? A critical review of the evidence (Thames & Hudson, London). en Europe, ed Farizy C (Musée de Préhistoire d’Ile de and its interpretation. Curr Anthropol 39:S1–S44. 19. Bar-Yosef O, Bordes JG (2010) Who were the makers ’ France, Nemours, France), pp 335–344. 11. D Errico F (2005) The invisible frontier: A multiple spe- of the Châtelperronian culture? J Hum Evol 59:586– 2. White R (2001) Personal ornaments from the Grotte du cies model for the origin of behavioural modernity. 593. – Renne at Arcy-sur-Cure. Athena Review 2:41–46. Evol Anthropol 12:188 202. 20. Pelegrin J, Soressi M (2009) Le Châtelperronien et ses 12. D’Errico F, et al. (2003) Many awls in our argument. 3. Hublin JJ, Bailey SE (2006) Revisiting the last Neandertals. rapports avec le Moustérien. Les Néandertaliens: Biol- Bone tool manufacture and use in the Châtelperronian When Neanderthals and Modern Met,ed ogie et Cultures, eds Vandermeersch B, Maureille B Conard NJ (Kerns Verlag, Tübingen), pp 105–128. and Aurignacian levels of the Grotte du Renne at Arcy- sur-Cure. The Chronology of the Aurignacian and of (Editions du Comité des Travaux Historiques et Scienti- 4. Vandermeersch B, Hublin JJ (2009) Les derniers Néander- fi – the Transitional Complexes, eds Zilhao J, d’Errico F (In- ques, Paris), pp 283 296. thaliens. Les Néandertaliens: Biologie et Cultures,eds stituto Português de Arqueologia, Lisbon), pp 240–270. 21. Mellars P (1996) The Neanderthal Legacy (Princeton Vandermeersch B, Maureille B (Editions du Comité des 13. Zilhão J, et al. (2010) Symbolic use of marine shells and Univ Press, Princeton). Travaux Historiques et Scientifiques, Paris), pp 109–116. mineral pigments by Iberian Neandertals. Proc Natl 22. Soressi M, Jones HL, Rink WJ, Maureille B, Tillier AM 5. Mellars P (2006) Archeology and the dispersal of mod- Acad Sci USA 107:1023–1028. (2007) The Pech-de-l’Azé I Neandertal child: ESR, ura- ern humans in Europe: deconstructing the Aurigna- 14. Higham T, et al. (2010) Chronology of the Grotte du nium-series, and AMS 14C dating of its MTA type B cian. Evol Anthropol 15:167–182. Renne (France) and implications for the context of or- context. J Hum Evol 52:455–466. 6. Mellars P (2006) A new radiocarbon revolution and the naments and human remains within the Châtelperro- 23. Mellars P (2005) The impossible coincidence: A single dispersal of modern humans in Eurasia. Nature 439: nian. Proc Natl Acad Sci USA 107:20234–20239. – species model for the origins of modern human behav- 931 935. 15. Churchill SE, Smith FH (2000) The makers of the early ’ ior in Europe. Evol Anthropol 14:12–27. 7. Zilhao J, d Errico F (2003) The Chronology of the Auri- Aurignacian in Europe. Am J Phys Anthropol 43 (suppl 24. Harrold FB (1989) Mousterian, Châtelperronian and gnacian and of the Transitional Complexes (Instituto 31):61–115. early Aurignacian in western Europe: continuity or Português de Arqueologia, Lisbon). 16. Bar-Yosef O, Zilhao J, eds (2006) Towards a Definition 8. Mellars P (2006) Why did modern human populations of the Aurignacian. (Instituto Português de Arqueolo- discontinuity?. The Human Revolution, eds Mellars P, – disperse from Africa ca. 60,000 years ago? A new gia, Lisbon). Stringer C (Edinburgh Univ Press, Edinburgh), pp 714 model. Proc Natl Acad Sci USA 103:9381–9386. 17. Vandermeersch B, Maureille B, eds (2009) Les Néander- 742. 9. Green RE, et al. (2010) A draft sequence of the Neandertal taliens: Biologie et Cultures (Editions du Comité des 25. Mithen S (1996) The of the Mind (Thames genome. Science 328:710–722. Travaux Historiques et Scientifiques, Paris). & Hudson, London).

20148 | www.pnas.org/cgi/doi/10.1073/pnas.1014588107 Mellars Downloaded by guest on September 29, 2021