From Tools to Symbols: From Early Hominids to Modern Humans [1 ed.] 1868144119, 1868144348, 9781868146376

A number of researchers have tried to characterise the anatomy and behavioural systems of early hominid and early modern

696 94 129MB

English Pages 606 [608] Year 2005

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

From Tools to Symbols: From Early Hominids to Modern Humans [1 ed.]
 1868144119, 1868144348, 9781868146376

  • Commentary
  • HDSS

Table of contents :
Cover
Half-title
Title
Copyright
Contents
Acknowledgements
Profile of Professor Tobias
List of participants
Foreword
Address
Keynote address
Searching for common ground in palaeoanthropology, archaeology and genetics
The history of a special relationship: prehistoric terminology and lithic technology between the French and South African research traditions
Essential attributes of any technologically competent animal
Significant tools and signifying monkeys: the question of body techniques and elementary actions on matter among apes and early hominids
Tools and brains: which came first?
Environmental changes and hominid evolution: what the vegetation tells us
Implications of the presence of African ape-like teeth in the Miocene of Kenya
Dawn of hominids: understanding the ape-hominid dichotomy
The impact of new excavations from the Cradle of Humankind on our understanding of the evolution of hominins and their cultures
Stone Age signatures in northernmost South Africa: early archaeology in the Mapungubwe National Park and vicinity
Vertebral column, bipedalism and freedom of the hands
Characterising early Homo: cladistic, morphological and metrical analyses of the original Plio Pleistocene specimens
Early Homo, ‘robust’ australopithecines and stone tools at Kromdraai, South Africa
The origin of bone tool technology and the identification of early hominid cultural traditions
Contribution of genetics to the study of human origins
An overview of the patterns of behavioural change in Africa and Eurasia during the Middle and Late Pleistocene
From the tropics to the colder climates: contrasting faunal exploitation adaptations of modern humans and Neanderthals
New neighbours: interaction and image-making during the West European Middle to Upper Palaeolithic transition
Late Mousterian lithic technology: its implications for the pace of the emergence of behavioural modernity and the relationship between behavioural modernity and biological modernity
Exploring and quantifying technological differences between the MSA I, MSA II and Howieson’s Poort at Klasies River
Stratigraphic integrity of the Middle Stone Age levels at Blombos Cave
Testing and demonstrating the stratigraphic integrity of artefacts from MSA deposits at Blombos Cave, South Africa
From tool to symbol: the behavioural context of intentionally marked ostrich eggshell from Diepkloof, Western Cape
Chronology of the Howieson’s Poort and Still Bay techno-complexes: assessment and new data from luminescence
Subsistence strategies in the Middle Stone Age at Sibudu Cave: the microscopic evidence from stone tool residues
Speaking with beads: the evolutionary significance of personal ornaments
Personal names index
Subject index

Citation preview

From Tools to Symbols From Early Hominids to Modern Humans

Edited by Francesco d’Errico and Lucinda Backwell

From Tools to Symbols From Early Hominids to Modern Humans

From Tools to Symbols From Early Hominids to Modern Humans edited by

Francesco d’Errico and Lucinda Backwell In honour of Professor Phillip V. Tobias

Witwatersrand University Press

Wits University Press 1 Jan Smuts Avenue Johannesburg 2001 South Africa http://witspress.wits.ac.za

Selection, compelation and introduction © 2005 by Francesco D’Errico and Lucinda Backwell Individual articles © 2005 by the authors First published in South Africa 2005 ISBN 1-86814-411-9 (soft cover) ISBN 1-86814-434-8 (hard cover) ISBN 978-1-86814-637-6 (ePDF) All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the express permission, in writing, of both the author and the publishers.

Designed & produced by Riaan de Villiers & Associates, Johannesburg, South Africa Printed and bound by Creda Communications, Cape Town, South Africa

Contents

Acknowledgements

ix

Profile of Professor Tobias

xi

List of participants

xv

Foreword Justice Edwin Cameron

xxiii

Address Bernard Malauzat

xxvii

Keynote address Phillip V. Tobias

xxxi

Searching for common ground in palaeoanthropology, archaeology and genetics Francesco d’Errico and Lucinda R. Backwell

1

The history of a special relationship: prehistoric terminology and lithic technology between the French and South African research traditions Nathan Schlanger

9

Essential attributes of any technologically competent animal Charles K. Brain

38

Significant tools and signifying monkeys: the question of body techniques and elementary actions on matter among apes and early hominids Frédéric Joulian

52

v

Contents

Tools and brains: which came first? Phillip V. Tobias

82

Environmental changes and hominid evolution: what the vegetation tells us Marion K. Bamford

103

Implications of the presence of African ape-like teeth in the Miocene of Kenya Martin Pickford and Brigitte Senut

121

Dawn of hominids: understanding the ape-hominid dichotomy Brigitte Senut

134

The impact of new excavations from the Cradle of Humankind on our understanding of the evolution of hominins and their cultures Lee R. Berger

152

Stone Age signatures in northernmost South Africa: early archaeology in the Mapungubwe National Park and vicinity Kathleen Kuman, Ryan Gibbon, Helen Kempson, Geeske Langejans, Joel Le Baron, Luca Pollarolo and Morris Sutton

163

Vertebral column, bipedalism and freedom of the hands Dominique Gommery

183

Characterising early Homo: cladistic, morphological and metrical analyses of the original Plio-Pleistocene specimens Sandrine Prat

198

Early Homo, ‘robust’ australopithecines and stone tools at Kromdraai, South Africa Francis Thackeray and José Braga

229

The origin of bone tool technology and the identification of early hominid cultural traditions Lucinda Backwell and Francesco d’Errico

238

vi

Contents

Contribution of genetics to the study of human origins Himla Soodyall and Trefor Jenkins

276

An overview of the patterns of behavioural change in Africa and Eurasia during the Middle and Late Pleistocene Nicholas J. Conard

294

From the tropics to the colder climates: contrasting faunal exploitation adaptations of modern humans and Neanderthals Curtis W. Marean

333

New neighbours: interaction and image-making during the West European Middle to Upper Palaeolithic transition David Lewis-Williams

372

Late Mousterian lithic technology: its implications for the pace of the emergence of behavioural modernity and the relationship between behavioural modernity and biological modernity Marie Soressi

389

Exploring and quantifying technological differences between the MSA I, MSA II and Howieson’s Poort at Klasies River Sarah Wurz

418

Stratigraphic integrity of the Middle Stone Age levels at Blombos Cave Christopher Henshilwood

441

Testing and demonstrating the stratigraphic integrity of artefacts from MSA deposits at Blombos Cave, South Africa Zenobia Jacobs

459

From tool to symbol: the behavioural context of intentionally marked ostrich eggshell from Diepkloof, Western Cape John Parkington, Cedric Poggenpoel, Jean-Philippe Rigaud and Pierre-Jean Texier

475

vii

Contents

Chronology of the Howieson’s Poort and Still Bay techno-complexes: assessment and new data from luminescence Chantal Tribolo, Norbert Mercier and Hélène Valladas

493

Subsistence strategies in the Middle Stone Age at Sibudu Cave: the microscopic evidence from stone tool residues Bonny S. Williamson

512

Speaking with beads: the evolutionary significance of personal ornaments Marian Vanhaeren

525

Personal names index

555

Subject index

563

viii

Acknowledgements This book is the tangible outcome of a conference entitled From Tools to Symbols: From Early Hominids to Modern Humans, held at the University of the Witwatersrand, 16–18 March 2003. This event was organised in honour of Professor Phillip Tobias by the editors of this volume, in recognition of his outstanding scientific contributions to the field of palaeoanthropology, his crucial role as mentor, the assistance he has constantly and generously offered to colleagues and students from around the word, his insights into the nature and history of humanity, and the effort he has made to disseminate this knowledge and craft it into an integral aspect of human consciousness. We thank him for contributing a keynote address and stimulating inspiring debate during the meeting. This conference and the publication of the proceedings would not have taken place without the encouragement, assistance and support of numerous individuals and institutions. First and foremost, we thank Bernard Malauzat, Counsellor for Science, Culture and Development at the Embassy of France in South Africa, who was instrumental in developing this initiative. The Trustees of the Palaeoanthropology Scientific Trust (PAST) encouraged this endeavour and their Operations Manager, Andrea Leenen, together with Christine Read, were particularly helpful in final preparations. Jennifer Oppenheimer kindly secured Jan Smuts House, an ideal venue for the conference. Justice Edwin Cameron, Chairperson of Council, University of the Witwatersrand, generously accepted our invitation to give the opening address on behalf of the University, and travelled some distance to do so. Khotso Mokhele, President of the National Research Foundation, kindly agreed to speak on the significance of scientific research in general and palaeoanthropology in particular in the context of the new South Africa. Jean-Marie Hombert, Director of the Department of Humanities, French Centre National de la Recherche Scientifique (CNRS), strove to encourage the success of the conference and spoke on the importance of South African/ French scientific collaboration. We thank Loyiso Nongxa and Richard Pienaar, Vice-Chancellor and Deputy ViceChancellor, Paul Dirks, Head of the School of Geosciences, and Bruce Rubidge, Head of the Bernard Price Institute for Palaeontology, University of the Witwatersrand, Patrick Buat-Menard, Vice-Chancellor, and Gerard Blanc, Dean of the Faculty of Geology,

ix

Acknowledgements

University of Bordeaux, and Monique Rivière, Scientific Attaché at the Embassy of France in South Africa, for their encouragement and support. Many friends, colleagues and members of the Wits community actively collaborated in the organisation of the event. Most notably Chrisandra Chetty and Iain Burns, who graciously oversaw the financial aspects of the conference. Lee Berger invited delegates to the Palaeoanthropology Unit for Research and Exploration for a welcome luncheon, provided the vehicles to transport the delegates and allowed students from the Unit to help with the organisation of the conference. In this regard we greatly appreciate the help of Rodrigo Lacruz, Christine Steininger, Barend van Rensburg, Headman Zondo, and Pedro Boshoff. The Rock Art Research Unit and the Department of Archaeology willingly agreed to open their doors and give tours to delegates on their arrival at Wits. The members of these institutions are sincerely thanked for their time and kindness. Our thanks also go to Matt Kitching and the Audiovisual Unit team at Wits for their competence in ensuring that things ran smoothly during the conference. Bob Brain and Kathy Kuman kindly led unforgettable post-conference excursions to Swartkrans and Sterkfontein. We are also grateful to Cathy Snow for creating and maintaining the conference website, and John Gurche for graciously permitting the use of his artwork as the conference logo. The conference was generously sponsored by the Embassy of France in South Africa, the Palaeoanthropology Scientific Trust, the Centre National de la Recherche Scientifique, the University Bordeaux 1, the Johannesburg branch of the Alliance Française, the University of the Witwatersrand, and the Eurocore programme ‘Origin of Man, Language and Languages’. The preparation of this volume was funded by the Embassy of France in South Africa and the Palaeoanthropology Scientific Trust. We gratefully acknowledge the invaluable help provided by Mike Raath in editorially revising some of the manuscripts and Françoise Lagarde for formatting the figures of the manuscripts. In particular we thank the conference participants for the lively exchange of ideas that took place during these memorable days, and for contributing papers to this book.

x

Profile of Phillip Vallentine Tobias Phillip Tobias is Professor Emeritus of Anatomy and Human Biology at the University of the Witwatersrand Medical School. He holds the positions of Honorary Professor of Palaeoanthropology, Honorary Professorial Research Associate, and Director of the Sterkfontein Research Unit. He is the Andrew Dickson White Professor-at-Large of Cornell University, Ithaca, New York, USA. He was born in Durban, Natal, in October 1925. At Wits University he obtained five degrees, including a medical degree, PhD in genetics and DSc in Palaeoanthropology. He served Wits University for over fifty years. He was a part-time teacher in the Anatomy Department (1946–1950), a lecturer and senior lecturer under Professor Raymond A. Dart (1951–1958), and Professor of Anatomy for thirty-five years, thirty-two of them as Head of the Department of Anatomy. He was Dean of the Wits Medical Faculty for three years and a Member of the University Council for fourteen years. Although he retired officially at the end of 1993, he continues to be actively involved in academic matters, supervising higher degree (PhD) students and Post-doctoral Fellows; he serves on numerous committees such as the planning committee for the World Heritage Site at Sterkfontein, Swartkrans, Kromdraai and the environs; the South African National Commission for UNESCO (Immediate Past Chairman); Institute for the Study of Mankind in Africa (Founder President); on the editorial boards of a number of scientific periodicals, and is an honorary member and member of numerous national and international scientific academies. Phillip Tobias has researched on the chromosomes of mammals, the living peoples of southern Africa, especially Kalahari San (Bushmen), miners on the Witwatersrand gold mines from southern African countries, and the Tonga people of Zambia; the anatomy, growth, physique and secular trends in southern African peoples; the meaning of race in human beings, and the implications of racism; the history and philosophy of anatomy, anthropology and biology. He is a world authority on human evolution and the analysis of early hominid fossils. His work on the evolution of the human brain and the origins of spoken language is internationally recognised. He has examined and in most cases described ancient hominid fossils from South Africa, Namibia, Zimbabwe, Zambia, Tanzania, Kenya, Libya, Israel, Spain and other parts of Europe, Indonesia and China. His work at Sterkfontein is especially important. He first worked there as an undergraduate student in 1945 and has run a major excavation there continuously

xi

Profile of Phillip V Tobias

Professor Phillip Tobias

since 1966, being responsible for the recovery of some six hundred hominid fossils; these have made Sterkfontein the world’s richest single deposit for ancient hominid remains. He has excavated also at Makapansgat, Cave of Hearths, Rainbow Cave, Mwulu’s Cave, Kromdraai, Gladysvale, Taung, and Rose Cottage Cave, Ladybrand, Free State, and has been a consultant to the World Heritage Centre of UNESCO on the Peking Man site of Zhoukoudian near Beijing, China. Professor Tobias was entrusted

xii

Profile of Phillip V Tobias

by Louis and Mary Leakey with the study of all the fossil hominids they recovered from Tanzania and Kenya, a working partnership that continued for some twenty years and resulted in three large volumes and numerous articles by Tobias. Under his direction, the Wits Anatomy Department became a world centre for research and teaching on fossil hominids and human evolution. It attracted PhD and other research students from Canada, Mexico, the United Kingdom, France, Italy, Portugal, Germany, China, the United States of America and southern Africa. Tobias has published over a thousand works, including approximately forty books and monographs and over ninety chapters in books. Internationally, Tobias is one of the most renowned South African scientists and probably the most highly honoured. He has received sixteen honorary degrees and was a recipient of the Balzan International Prize (the first time it was awarded for accomplishments in physical anthropology); the first to be awarded the L.S.B. Leakey Prize; the Charles Darwin Lifetime Achievement Award; the Anisfield-Wolf Award in Race Relations. He was the first South African living and working in South Africa to be honoured with the Fellowship of the Royal Society (London). He is a Foreign Associate of the National Academy of Sciences (USA) and is the only recipient in South Africa of this highest American honour. Among his many medals are the Huxley Memorial Medal, the Wood Jones Medal, the Ales Hrdlicka Medal, the South Africa Medal and the Rivers Memorial Medal. Various civil honours have been conferred upon him: the Order of Meritorious Service (Gold Class) of South Africa; the Order of the Southern Cross, Class II, of South Africa; Commander, National Order of Merit of France; Commander of the Order of Merit of the Republic of Italy and the Honorary Cross for Science and Arts, First Class, of Austria. Tobias was active in initiating the anti-apartheid campaign in the universities of South Africa from 1949, in his capacity as President of the non-racial National Union of South African Students. He kept up the fight against apartheid and for academic and human freedom in academia and in society at large, and played a prominent part in the struggle to keep the universities of South Africa open to all students and academic staff, irrespective of race. He was one of a small group of Wits and Cape Town professors who lodged a formal complaint with the South African Medical and Dental Council on the handling of Steve Biko by the ‘Biko doctors’, whose treatment of Biko was a critical factor in the events leading to his untimely death. On failing to gain satisfaction, Tobias and the other professors took the Medical Council to the Supreme Court and won the case. Numerous lectures and articles by Tobias document his sustained campaigning against the apartheid government’s policies, for over forty years. Among many socio-political activities, here are a handful: he negotiated on behalf of the Department of Arts, Culture,

xiii

Profile of Phillip V Tobias

Science and Technology (DACST) for the return to South Africa from France of the remains of Ms Saartjie Baartman; he was Founder of the African Medical Scholarships Trust Fund and Founder Chairman of MESAB – South Africa (Medical Education for South African Blacks); he is a Member of the Advisory Committee on Khoisan Identity and Heritage, DACST; Consultant on World Heritage Sites to the Gauteng and National governments, and President of the Education League of South Africa, a body set up to campaign against apartheid education.

xiv

List of Participants Graham Avery Natural History Division, Cenozoic Studies Iziko Museums of Cape Town Box 61, Cape Town 8000, South Africa. [email protected] Lucinda Backwell Institute for Human Evolution, School of Geosciences University of the Witwatersrand Private Bag 3, WITS 2050, Johannesburg, South Africa. UMR 5808 du CNRS, Institut de Préhistoire et de Géologie du Quaternaire Avenue des Facultés, 33405 Talence, France. [email protected] Marion Bamford Bernard Price Institute for Palaeontological Research University of the Witwatersrand Private Bag 3, WITS 2050, Johannesburg, South Africa. [email protected] Lee Berger Institute for Human Evolution School of Geosciences University of the Witwatersrand Private Bag 3, WITS 2050, Johannesburg, South Africa. [email protected] José Braga Laboratoire d’Anthropologie, PACEA/UMR 5199 du CNRS, Université Bordeaux 1, Avenue des Facultés, 33405 Talence Cedex, France. [email protected]

xv

List of participants

Bob Brain Transvaal Museum, Northern Flagship Institution, Paul Kruger Street P.O. Box 413, Pretoria, 0001, South Africa. [email protected] Edwin Cameron Supreme Court of Appeal P.O. Box 258, Bloemfontein 9300. School of Law, University of the Witwatersrand WITS 2050, Johannesburg, South Africa. Nicholas J. Conard Institut für Ur- und Frühgeschichte und Archäologie des Mittelalters Abteilung Ältere Urgeschichte und Quartärökologie Eberhard-Karls-Universität Tübingen, Schloss Hohentübingen, 72070 Tübingen, Germany. [email protected] Francesco d’Errico PACEA/UMR 5199 du CNRS Institut de Préhistoire et de Geologie du Quaternaire UFR de Geologie, Bat. B18, Avenue des Facultés, 33405 Talence, France. Department of Anthropology, The George Washington University, Washington DC. [email protected] Simon Donnelly Department of Linguistics, University of the Witwatersrand Private Bag 3, WITS 2050, Johannesburg, South Africa. [email protected] Dominique Gommery UPR 2147 CNRS 44 rue de l’Amiral Mouchez, 75014 Paris, France. [email protected]

xvi

List of participants

Christopher Henshilwood Centre for Development Studies, University of Bergen Nygårdsgaten 5, N- 5015, Bergen, Norway. African Heritage Research Institute 167 Buitenkant Street Gardens, Cape Town 8001, South Africa. [email protected] Jean-Marie Hombert Laboratoire Dynamique du Langage Institut des Sciences de l’Homme 14, avenue Berthelot 69363 Lyon Cedex 07. Département des Sciences de l’Homme et de la Société du CNRS 3, rue Michel Ange 75794 Paris cedex 16, France. [email protected] Jean-Jacques Hublin Department of Human Evolution Max Planck Institute for Evolutionary Anthropology Deutscher Platz 6, D-04103 Leipzig, Germany. [email protected] Zenobia Jacobs Council for Scientific and Industrial Research (CSIR) P.O. Box 395, Pretoria, 0001, South Africa. [email protected] Trefor Jenkins MRC/NHLS/Wits Human Genomic Diversity and Disease Research Unit National Health Laboratory Service and the University of the Witwatersrand, Johannesburg, South Africa. [email protected] Frederic Joulian Responsable du Programme de Recherches Interdisciplinaires ‘Evolution, Natures et Cultures’, SHADYC, Ecole des Hautes Etudes en Sciences Sociales, 2 rue de la Vieille Charité, 13002 Marseille, France. [email protected]

xvii

List of participants

Kathy Kuman School of Geography, Archaeology and Environmental Studies University of the Witwatersrand Private Bag 3, WITS 2050, Johannesburg, South Africa. [email protected] Kevin Kuykendall Department of Anatomical Sciences University of the Witwatersrand Medical School 7 York Road, Parktown 2193, Johannesburg, South Africa. [email protected] David Lewis-Williams Rock Art Research Institute University of the Witwatersrand Private Bag 3, WITS 2050, Johannesburg, South Africa. [email protected] Bernard Malauzat Service de coopération et d’action culturelle Ambassade de France en Afrique du Sud 250 Melk Street, Nieuw Muckleneuk 0181, Pretoria, South Africa. [email protected] Alan Mann Department of Anthropology, Princeton University, Princeton, New Jersey 08544. [email protected] Curtis Marean Institute of Human Origins Department of Anthropology P.O. Box 872402, Arizona State University Tempe, AZ 85287-2402. [email protected]

xviii

List of participants

Norbert Mercier Laboratoire des Sciences du Climat et de l’Environnement UMR 1572 Avenue de la Terrasse, F-91198 Gif-sur-Yvette Cedex, France. [email protected] Khotso Mokhele National Research Foundation P.O. Box 2600, Pretoria 0001, South Africa. John Parkington Department of Archaeology University of Cape Town Private Bag, Rondebosch 7700, South Africa. [email protected] Martin Pickford Laboratoire de Paleontologie UMR 8569 du CNRS, 8, rue Buffon, 75005, Paris. College de France, 11, Place Marcellin Berthelot, 75005, Paris, France. [email protected] Cedric Poggenpoel Department of Archaeology University of Cape Town Private Bag, Rondebosch 7700, South Africa. [email protected] Sandrine Prat UPR 2147 du CNRS, 44 rue de l’Amiral Mouchez 75014 Paris, France. [email protected] Jean-Philippe Rigaud UMR 5808 du CNRS, Institut de Préhistoire et de Géologie du Quaternaire Av des Facultés, 33405, Talence, France. [email protected]

xix

List of participants

Nathan Schlanger Archives of European Archaeology (AREA) Institut national d’histoire de l’art, 2, rue Vivienne, 75002 Paris, France. [email protected] Frank Senegas Transvaal Museum, Paul Kruger Street P.O. Box 413, Pretoria, 0001, South Africa. [email protected] Brigitte Senut Département Histoire de la Terre, USM 0203 du Muséum national d’Histoire naturelle & UMR 5143, PICS 1048 (CNRS), Case 38, 57, rue Cuvier, 75005 Paris, France. [email protected] Himla Soodyall MRC/NHLS/Wits Human Genomic Diversity and Disease Research Unit National Health Laboratory Service and the University of the Witwatersrand, Johannesburg, South Africa. [email protected] Marie Soressi Max Planck Institute for Evolutionary Anthropology Deutscher Platz 6, D-04103 Leipzig, Germany. PACEA/UMR 5199 du CNRS Institut de Préhistorie et de Géologie du Quartenaire, UFR de Géologie, Bat. B18 Avenue des Facultés, 33405, Talence, France. [email protected] Christopher Stringer Department of Palaeontology The Natural History Museum, Cromwell Road London, SW7 5BD, United Kingdom. [email protected]

xx

List of participants

Pierre-Jean Texier EP 2058 Préhistoire et Technologie 250 rue Albert Einstein Sophia-Antipolis, 06560 Valbonne, France. [email protected] Francis Thackeray Transvaal Museum P.O. Box 413, Pretoria, 0001, South Africa. [email protected] Phillip Tobias Sterkfontein Research Unit Department of Anatomical Sciences University of the Witwatersrand Medical School 7 York Road, Parktown, 2193, Johannesburg, South Africa. [email protected] Chantal Tribolo Laboratoire des Sciences du Climat et de l’Environnement UMR 1572, Avenue de la Terrasse F-91198 Gif-sur-Yvette Cedex, France. [email protected] Helene Valladas Laboratoire des Sciences du Climat et de l’Environnement UMR 1572, Avenue de la Terrasse F-91198 Gif-sur-Yvette Cedex, France. [email protected] Marian Vanhaeren CNRS UMR 7041 ArScAn, Ethnologie préhistorique 21 allée de l’Université, F-92023 Nanterre, France. [email protected]

xxi

From Tools to Symbols

Lyn Wadley Department of Archaeology University of the Witwatersrand, Private Bag 3, WITS 2050, Johannesburg, South Africa. [email protected] Bonny Williamson School of Geography, Archaeology and Environmental Studies University of the Witwatersrand, Private Bag 3, WITS 2050, Johannesburg, South Africa. [email protected] Sarah Wurz Department of Geography and Environmental Studies University of Stellenbosch Private Bag X1, Stellenbosch, 7602, South Africa. [email protected]

xxii

Foreword Opening address by Justice Edwin Cameron Supreme Court of Appeal Chairperson of Council, University of the Witwatersrand This conference celebrates and explores the origins of technology and the human brain, and the evolution of our species. It is particularly appropriate that it is held in South Africa. For more than a decade, South Africa has been at the centre of dynamic change that has reshaped our world and our ideas about ourselves as Africans and South Africans. We achieved, against all odds, a relatively peaceful transition from a racially oppressive system to a functioning democracy. Every person in our country has guaranteed rights of democracy and freedom of speech under our Constitution. In addition, through our Constitution – one of the most progressive in the world – we as South Africans make a series of promises to each other concerning treating each other with dignity, equality and non-discrimination, as well as providing each other with basic socio-economic rights. From the base premises of apartheid we have joined as South Africans in aspiring to create a non-racial society based upon the fundamentals of civilised conditions and mutual respect. In doing so we have done more than merely refuse to be the prisoners of our degrading racial past. We have in fact demonstrated to the world the innate potential of humanity – the qualities of restraint, coordination and respect that have helped make Homo sapiens such a successful species. Yet, we still struggle. As the scientists assembled here will almost certainly tell us, we are very far from perfect – we have flawed anatomies, we misuse and abuse our technologies, and we have a too-marked propensity for intra-species violence. At this time in history, almost more than at any other time, humankind as a species faces grave issues. Here in Africa and around the world we face a mass epidemic of HIV/AIDS – a disease that probably originated in Africa. Our continent struggles with wars and famine, with racism, ethnicity, tribalism and religious intolerance. Our technologies – the very subject this conference explores – have been a blessing and a burden. We have used them for good: curing diseases, uplifting human lives, and exploring our world around us. In doing so, we utilised the power of the human brain, evolved here in Africa, to its fullest. But we have also used technology for war, the destruction of our

xxiii

Foreword

environment and other actions that threaten every species on this planet. Yet six million years of evolution in Africa has made us a resilient, innovative and even caring species – and it is countries like South Africa that show us the great potential of humankind. South Africa has in the last decade been a persisting source of interest and inspiration to the rest of the world. The hope we have brought to our problems continues to inspire the world, and it is for this reason that we feel pride in welcoming you to our country and our University. As participants in this conference, you carry a surprising burden. Archaeologists, palaeontologists and palaeoanthropologists are the explorers of our species. You are the scientists who examine what attributes define our humanity, what makes us a destructive species – while at the same time also having qualities of caring and altruism. You explore our history as a species, and your research guides us in better understanding ourselves, in order to reach our true potential. So it is most appropriate that a conference such as this is held here in Africa, and more particularly South Africa – where we sit forty-five minutes from what has rightly been dubbed ‘the Cradle of Humankind’. It is also appropriate as a demonstration of human cooperation. This conference was made possible by strong support from the French and South African governments, the private sector in the form of the Palaeoanthropology Scientific Trust (PAST), and from the scientists themselves who have come from all corners of the world to be here. We thank the Embassy of France in South Africa for this wonderful initiative and their very generous sponsorship of this event. Thanks also to the other sponsors – notably PAST – a trust comprising South African men and women of business that for more than a decade has been dedicated to the support of human evolutionary studies in South Africa. Thanks to the University of Bordeaux and to the host, Wits. And if, in thanking Wits, we had to single out one human face, one extraordinary individual – he would not be hard to choose. This conference is of course also a celebration of the life’s work of a very special individual. A great thinker and humanitarian, and an individual who has worked his entire life for a better understanding of our species; Phillip Vallentine Tobias is a child of South Africa and an offspring of this University, but he is really a son of the World. Professor Tobias has obtained five degrees from Wits, including a medical degree, PhD in genetics and DSc in Palaeoanthropology. He has served Wits University for over fifty years, thirty-two of them as Head of the Department of Anatomy. He is the first and only person to hold three professorships at Wits – simultaneously! Internationally, he is one of the most renowned South African scientists and probably the most highly honoured, having received sixteen honorary degrees. Civil honours that have been conferred upon him include: The Order of Meritorious Service (Gold Class) of South Africa, Commander, National Order of Merit of France,

xxiv

Foreword

Commander of the Order of Merit of the Republic of Italy, The Honorary Cross for Science and Arts, First Class, of Austria. His many accolades include being selected as one out of a maximum of twenty permitted as an Honorary Life Member of the International Union of Anthropological and Ethnological Sciences; the first South African living and working in this country to be honoured with membership as a Fellow of the Royal Society of London; the only South African invited to become a Foreign Member of The American Philosophical Society; and the only South African ever to receive the Charles Darwin Lifetime Achievement Award from the American Association of Physical Anthropologists. Phillip Tobias has made an enormous – and probably incalculable – contribution to our understanding of the human species. To call his work multidisciplinary or multifaceted is only to begin to understand its impact. His work has not only been confined to the study of bones, nor has it been confined to the study of human anatomy, nor even to the more difficult study of the processes of evolution or the process of human thought itself. Reading any one of his numerous papers illustrates the depth and breadth of his knowledge and his unique ability to combine many different aspects of science and life into a cogent and lucid argument. Professor Tobias has published over a thousand works, including forty books and monographs and over ninety chapters in books. But at the end of last year he ventured into a new field, that of star of the small screen. He hosted a successful television series entitled Tobias’s Bodies. This looked at what it means to be human in the twenty-first century in the light of what we know about our distant past. Its flighting on national television was highly successful – not least because of the inimitably engaging presence of the narrator and steward of the series – and is currently being considered for syndication and translation worldwide. To his intellectual and academic achievements I want to add that, as a person, his qualities of warmth, the ambition of his intellect, the liberality of his spirit and his social politics have served as an inspiration not just to his own students, but to successive generations of executive and academic leaders at Wits. And so, it is particularly important that the title of this conference is so allencompassing: From Tools to Symbols – From Early Hominids to Modern Humans. This could well be a summary of the life achievement of Phillip Tobias. At this time in history, the exploration of our heritage is important to South Africa and, I believe, to the human species. The exchange of cultural and scientific ideas amongst nations builds awareness of others, and in this awareness-building is the key to making the world a better place for humankind. It is my hope that this conference will contribute in a substantial way to making us more aware of ourselves as humans, and in doing so make this world a better place for humankind and all of the remainder of life that shares this planet with us. I

xxv

Foreword

trust that your conference will be a success and that your discourses prove productive, and I remind you of the auspicious grounds that you conduct your business upon – the very ground that moulded the evolution of our species. Johannesburg, 16 March 2003

xxvi

Address

Address by Bernard Malauzat Counsellor for Science, Culture and Development Embassy of France in South Africa Our Embassy is deeply committed to the continuing development of a sound co-operation between France and South Africa in the fields of palaeontology and palaeoanthropology. There are several reasons for this. Subsidising the discovery of the past, our origins and our ancestors has been a well-established, almost centennial tradition of our department ever since we began archaeological missions to the Middle East. This tradition continues today and is fitting for a country where cultural research is a vital part of its transition to democracy. The French Ministry of Foreign Affairs devotes around one third of its budget to ‘cultural cooperation’, which is considered an important component of our diplomacy. By supporting our quest for the past through various missions throughout Africa, France reinforces her well-known commitment to the promotion of cultural diversity throughout the world and her respect for diversified forms of human fulfilment. In this regard, we share President Thabo Mbeki’s homage to the work of Phillip Tobias and Francis Thackeray at the inauguration of the Freedom Park Museum in Pretoria. Here he stressed how important for the new South Africa the work of these two eminent scholars and their colleagues are, because when investigating the past three million years of human evolution, they see beyond the divisions of race, colour and creed and stress instead the unifying aspects of our common ancestry. We share that belief and this homage to Professor Tobias, our guest of honour today. Another reason for our sustained efforts in this field lies in the already established research cooperation between France and South Africa, as the number of speakers in this seminar attests. Public funds to the tune of €1 500 000 per year are made available for the development of scientific cooperation between our two countries. We wish to maintain and if possible increase this level, particularly through support for the training of students, researchers or technical agents from historically disadvantaged communities. This cooperation is based on both a common respect for and interest in the exceptionally rich heritage of our two countries. The discovery of the Chauvet

xxvii

Address

and Cosquer Caves in the last decade dramatically increased the cultural heritage of the world, and in reply to Abbé Breuil’s visits to your country, the French government invited Professor David Lewis-Williams from Wits University to assess these discoveries in France. An interesting book resulted from the cooperation between Professor LewisWilliams and Professor Jean Clottes. By the same token, several French experts have been invited to study the palaeontological treasures of South Africa. The rich material in both countries and well-developed networks of laboratories and research naturally encourages the establishment of long-term partnerships. Having lived many years in Jordan, Egypt and Turkey, I know from experience that documenting our past also has many political and economic dimensions. Firstly, it contributes to consolidating the identity of the country concerned. Secondly, by attracting the sustained interest of the media, it is one of the few scientific pursuits that gains the widespread interest of the public at large. For now, in wishing success to this seminar, I would like to extend our warmest thanks to our main hosts, the people and the government of South Africa; to the University of the Witwatersrand and the South African Institute for International Affairs; to Jennifer Oppenheimer for her generous involvement; the Palaeoanthropology Scientific Trust for co-funding this event; Professor Hombert, Director of the CNRS Department of Humanities, and last but not least, to Lucinda Backwell and the students of her unit, who together with Francesco d’Errico, have mustered a lot of energy and work in making this moment possible. To all, thank you. Johannesburg, 16 March 2003 Notre Ambassade est profondément attachée au développement d’une coopération solide entre la France et l’Afrique du Sud dans les domaines de la paléontologie et de la paléo-anthropologie. Il y a à cela plusieurs raisons. Subventionner la découverte du passé, de nos origines et de nos ancêtres est une tradition bien établie, presque centenaire, de notre ministère depuis le temps où commencèrent nos missions archéologiques au Moyen Orient. Cette tradition continue aujourd’hui et convient particulièrement dans un pays, l’Afrique du Sud, où la recherche ‘culturelle’ est partie intégrante de la transition vers la démocratie. Le Ministère des Affaires Etrangères en France consacre près du tiers de son budget à ‘la coopération culturelle’, que nous considérons comme une composante importante de notre diplomatie. En soutenant la quête du passé au travers de diverses missions en Afrique, la France réaffirme son engagement bien connu pour la promotion de la diversité culturelle dans le monde et son respect pour diverses formes d’accomplissement

xxviii

Address

humain. A cet égard, nous partageons l’hommage du Président Mbeki à l’œuvre de Phillip Tobias et de Francis Thackeray lors de l’inauguration du ‘Freedom Park Museum’ à Pretoria. Il mit alors l’accent sur l’importance pour la nouvelle Afrique du Sud des travaux de ces deux éminents savants et de leurs collègues, car, en faisant des recherches sur les trois derniers millions d’années de l’évolution humaine, ils regardent par delà les divisions de races, de couleurs, de croyances, et mettent plutôt l’accent sur les aspects unificateurs de notre passé commun. Nous partageons cette conviction et cet hommage au Professeur Tobias, notre hôte d’honneur aujourd’hui. Une autre raison de nos efforts soutenus dans ce domaine réside dans le fait que la coopération entre la France et l’Afrique du Sud dans ce domaine de recherche est déjà bien établie comme en témoigne le nombre de intervenants à ce séminaire. Chaque année, des fonds publics de l’ordre de 1 500 000 Euros sont mis à la disposition de la coopération scientifique entre nos deux pays. Nous souhaitons maintenir, et si possible, accroître ce niveau, en particulier par le soutien à la formation d’étudiants, de chercheurs ou de techniciens issus des communautés historiquement désavantagées. Cette coopération est fondée à la fois sur un respect commun et sur un intérêt pour le patrimoine exceptionnellement riche de nos deux pays. La découverte de la grotte Chauvet, et de la grotte Cosquer durant ces dix dernières années a accru de façon spectaculaire l’héritage culturel du monde, et en écho aux visites de l’Abbé Breuil dans votre pays, le Gouvernement Français a invité le Professeur David LewisWilliams de l’Université Wits à évaluer ces découvertes en France. Il en est résulté un livre intéressant, fruit de la coopération entre le Professeur Lewis-Williams et le Professeur Jean Clottes. Dans le même esprit, plusieurs experts français ont été invités à participer à l’étude des trésors paléontologiques de l’Afrique du Sud. Le matériau très riche dans les deux pays, et des réseaux très développés de laboratoires ou de chercheurs encouragent naturellement l’établissement de partenariats à long terme. Ayant vécu de nombreuses années en Jordanie, en Egypte et en Turquie, je sais par expérience que le fait de documenter notre passé a aussi de nombreuses dimensions politiques ou économiques. Cela contribue d’abord à consolider l’identité du pays concerné. De plus, en attirant l’intérêt soutenu des médias, c’est l’une des rares aventures scientifiques qui mobilise largement l’intérêt du grand public. Pour le moment, en souhaitant beaucoup de succès à ce séminaire, je voudrais exprimer nos remerciements les plus chaleureux à nos hôtes, le peuple et le Gouvernement de l’Afrique du SUD ; à l’Université de Witwatersrand et à l’Institut sud africain des relations Internationales; à Jennifer Oppenheimer, par sa participation généreuse; à PAST, la Fondation scientifique pour la Paléo-anthropologie qui a co-

xxix

Address

financé cet événement; au Professeur Hombert, Directeur du Département des Sciences Humaines du CNRS, et, ‘last but not least’, à Lucinda Backwell et aux étudiants de son unité, qui avec Francesco d’Errico, ont mobilisé une énergie et un travail considérables afin que ce moment devienne possible. A tous, merci. Johannesburg, le 16 mars 2005

xxx

Keynote Address Phillip V Tobias This Round Table is a celebration of Franco–South African collaboration in the fields of palaeoanthropology and archaeology. In this cooperative endeavour, we who pursue our lives and researches in South Africa have been at the receiving end of extraordinary largesse from France, mediated with sagacity and imagination by the Embassy of France in South Africa and the CNRS. This conference is the latest manifestation of this fruitful inter-hemispheric interaction. It is our hope that the research opportunities and facilities which South Africa is able to offer in abundance and the ‘Open House’ policy towards visiting researchers and students which we have pursued for nearly half a century may effect a certain symmetry in this relationship. It is symbolic of this cross-pollination that this symposium has been organised by a French scientist, Francesco d’Errico, and a South African one, Lucinda Backwell. For their combined efforts, coupled with their manifest organisational skills, we who participated in the meeting are deeply indebted. To them and their helpers, a sincere expression of thanks is due. Personally, I convey my gratitude to the organisers for their very kind thought in dedicating the Round Table to myself, a most touching and generous gesture which I deeply appreciate. A fitting time was chosen for the holding of this conference: worthy of celebration is the fact that fifty years ago Francis Crick and James Watson published their historic paper announcing the double helix model for the structure of DNA; another cause of rejoicing is the award of a Nobel Prize for Physiology and Medicine for 2002 to Sydney Brenner, who obtained his first four degrees at the University of the Witwatersrand and is also a Doctor of Science honoris causa of this University. In part commemoration of both these historic events, a meeting is to be held next week, in tandem with this one, on the Human Genome in Africa, organised by Dr Wilmot James of the South African Human Sciences Research Council.

xxxi

Keynote address

Cette Table Ronde est une célébration de la collaboration franco-sud-africaine dans les domaines de la paléoanthropologie et de l’archéologie. Dans cet effort de coopération, nous avons, pour ceux qui vivent et font des recherches en Afrique du Sud, bénéficié d’une générosité extraordinaire de la part de la France, par l’intermédiaire de l’Ambassade de France en Afrique du Sud et du CNRS. La présente Table Ronde est la manifestation la plus récente de cet échange fructueux entre les deux hémisphères. Nous espérons que les opportunités de recherche et les installations que l’Afrique du Sud est à même d’offrir, ainsi que la politique de ‘Portes ouvertes’ aux chercheurs et aux étudiants de passage que nous menons depuis presque cinquante ans, puisse apporter une certaine symétrie dans cette relation. Notre collaboration scientifique est symbolisée e par le fait que deux universitaires et chercheurs d’Afrique du Sud, Lucinda Backwell, et de France, Francesco d’Errico, ont organisé la Table Ronde. Grâce à leurs aptitudes à l’organisation et à leurs efforts de collaboration, nous avons donc le plaisir de prendre part à cette réunion et nous les remercions de vive voix. Merci encore à Lucinda Backwell et à Francesco d’Errico. Pour ma part, j’aimerais exprimer ma gratitude aux organisateurs pour m’avoir dédié cette Table Ronde, un geste touchant et généreux que j’apprécie grandement. La tenue de cette Table Ronde n’est pas sans raison. En effet, nous célébrons aujourd’hui les cinquante ans de la publication de la communication historique de Francis Crick et James Watson qui annonçaient la structure en double hélice de l’ADN; nous célébrons aussi le Prix Nobel de Physiologie et de Médecine 2002 qu’a reçu Sydney Brenner, qui a obtenu ses quatre premiers diplômes de l’Université du Witwatersrand et qui est Docteur en Science honoris causa de la même Université. En commémoration de ces deux événements historiques, une réunion se tiendra la semaine prochaine sur le génome humain en Afrique, organisé par le Dr Wilmot James du HSRC. Johannesburg, 16 March 2003

xxxii

Searching for common ground in palaeoanthropology, archaeology and genetics Francesco d’Errico and Lucinda Backwell Sub-Saharan Africa provides an archaeological and palaeontological record that is crucial to understanding hominid anatomical and behavioural evolution. This cradle of humanity has attracted a number of international interdisciplinary research teams in search of answers as to what made us human. Collaboration with African scientists has been particularly fruitful, producing in the last few decades some of the most significant contributions to these fields of study. Since the 1920s French archaeologists and palaeoanthropologists have collaborated with South African colleagues to unearth and highlight this unique heritage. The primary aim of the symposium was to synthesise and debate results of current research on the origin of humankind and share thoughts on the future of this endeavour. More specifically, it provided the opportunity to highlight results of collaborative French–South African research projects in the framework of international research programmes conducted in southern Africa, and envision paths for future collaboration. Scholars interested in the human past are living in an exciting era, in which crossfertilisation between disciplines such as palaeoanthropology, primatology, genetics, archaeology, palaeoecology, climatology, linguistics, ethnography, evolutionary psychology and the neurosciences is producing novel integrated attempts at modelling biological–cultural interactions. The challenge is how to promote this dialogue in a manner that stimulates better comprehension of the human adventure without creating dogmatic paradigms or mainstream scenarios. In the past the concept of ‘culture’ played a crucial role in creating a conceptual barrier between humans and other primates. We now accept that chimpanzees possess rather complex cultural traditions that are independent of ecological constraints (Whiten et al., 1999). Partly as a consequence of this, it has become commonplace

1

From Tools to Symbols

to use the notion of ‘behavioural modernity’ rather than that of culture to indicate the range of ‘advanced’ traits that distinguish us and our recent ancestors from living primates and a fluctuating number of fossil hominid populations (Bar-Yosef, 2002; Klein, 1999; McBrearty & Brooks, 2000). However, it is argued that the definitions of ‘behavioural modernity’ proposed thus far are ambiguous and often represent ad hoc accommodative arguments to provide a theoretically grounded basis for the interpretation of archaeological evidence (d’Errico, 2003; d’Errico et al., 2003; Henshilwood & Marean, 2003). Can archaeology, palaeoanthropology and genetics tell us how and when human cultures developed the traits that make our societies different, in some respect, from those of our closest living relatives? In which cases are these differences substantial, and when do they simply reflect our definition of culture, of species (Holliday, 2003), the image we have of their evolution, and in the final analysis, of ourselves? When we address the question of the origin of behavioural modernity by relying exclusively on the archaeological and palaeontological record, the acquisition of these ‘modern’ traits appears to be the result of a long process that has affected different groups and hominid types, and not a sudden explosion coincidental with one or two momentary biological changes. The quest of archaeology and palaeoanthropology is to propose models that best fit the hard evidence. Many researchers have tried to characterise the anatomy and the behavioural systems of early hominids and early modern populations in an attempt to understand how we became who and what we are. This book records some of these endeavours and attempts to chronicle them in a coherent form. Cross-pollination between disciplines and research traditions has a long history, the narrative of which often sheds light on current epistemological approaches and allows us to look at future collaborations through the wise eyes of the past. Schlanger’s essay in this volume is an elegant illustration of how revisiting the history of archaeology may reveal unexpected turns of events and ideas. Terminological innovations in early South African archaeology are shown to have been the result of strategic scientific self-affirmation on both international and domestic levels. European archaeological traditions were instrumental in this endeavour as they provided an essential term of reference and ground for confrontation. Schlanger’s research resurrects another largely forgotten page of Franco–South African scientific dialogue. He shows that the interest of South African archaeological pioneers in lithic technology and raw material left its mark on eminent French prehistorians, leading to the development of the well-known French chaîne opératoire approach to the study of lithics. It is significant that this approach, to the emergence of which South African scholars seminally contributed, is becoming central to present-day collaboration between these two scientific communities, as demonstrated by a number of papers in this book.

2

Searching for common ground in palaeoanthropology, archaeology and genetics

Brain’s paper takes us to the heart of the matter that constitutes the focus of this book. He explores the significance of some of the more striking attributes that make us human (brain size, appendages, social organisation, complex communication, suitable birth canal), and speculates why other animals, which possess some of these traits, have not followed the same course. Building on his findings at Swartkrans, he elaborates on the mechanisms that may have stimulated the emergence of these features and their eventual evolutionary success. Joulian cautions us from the outset that human abilities and behaviours we are now ready to accept as shared with chimpanzees, including tool making, creation and transmission of distinct cultural traditions, and carnivory, might well be only the tip of an iceberg that researchers have thus far tackled using a clumsy anthropocentric approach. When analysed from the perspective of the elementary actions recorded in wild chimpanzees, or inferred from the archaeological record left by Plio-Pleistocene hominids, no significant differences appear between the two material cultures. The only exception is the use of sharp-edged tools by hominids, not found among chimpanzees. The implications of these observations for the emergence and definition of modern behaviour are certainly many and need to be explored in future research. Tobias recalls that chimpanzees, and to some extent baboon communities, are able to create and transmit a number of cultural behaviours that some archaeologists would consider suggestive of modern cognition. Also, a review of the archaeological record does not support an abrupt genetically driven transition to behavioural modernity. This evidence tells us that the story of human cognition was a gradual one; but how does one identify steps, if any, in this process? Tobias turns to comparative anatomy for an answer. Homo habilis unequivocally stands out as the first hominid exhibiting a disproportionate enlargement of the brain, a critical hallmark of humankind. Complex culture may be seen as the innovation, driven by increased encephalisation, which allowed our ancestors to overcome Mather’s paradox – that is, to simultaneously increase adaptability and adaptedness. Tobias argues that endocranial casts demonstrate that H. habilis was the first to manifest a human-like brain structure, and to display clear signs of Broca’s and Wernicke’s caps, features associated with language competence in modern humans. Has the time to accept this evidence finally arrived? One may wonder, in light of the thirtyyear saga surrounding this discovery that Tobias so eloquently relates in his paper. Environmental changes certainly played a role in this process. Their documentation has become an integral part of any scenario attempting to model early hominid evolution. Bamford’s contribution aims at synthesising the available South and East African botanical evidence, exploring how palaeobotanical data may be used for palaeoenvironmental reconstructions, and discusses the potential of this record for understanding the influence of the environment on hominid adaptation.

3

From Tools to Symbols

Early hominid and primate bones remain the primary source of information to estimate the timing and mechanisms of our evolution. Pickford and Senut’s contribution takes us to a crucial yet poorly understood part of history, that of the dichotomy between the chimpanzee and hominid lineage. Even if sparse, the palaeontological record is providing clear evidence, they argue, for a considerably earlier split between the gorilla, chimpanzee and hominid clades, thereby challenging the timing that most molecular biologists propose for this process. This observation, and the broader review of the evidence that Senut provides in her paper, stress the importance of looking further back in time for answers to the origin of hominids. The Miocene hominoid record certainly represents the new frontier in palaeoanthropological enquiry, and once decisive inferences are made, will constitute the ideal climate in which to establish a constructive dialogue with genetics. Gommery concentrates on a critical and long-debated aspect of human adaptation, the emergence of bipedalism and its possible link with the handling, transport and production of stone tools. His research shows that Plio-Pleistocene and perhaps even Upper Miocene hominids experimented with different modes of bipedalism, none of which prevented them from transporting or deliberately modifying stone tools. Contrary to the popular belief that views development of complex technology, hand dexterity, and bipedalism as closely related, Gommery makes the point, mostly relying on the South African palaeontological record, that technology and highly developed cognitive abilities probably had more to do with changes in the brain than in mode of locomotion. A detailed reappraisal of the taxonomic affiliation of twenty-three key hominid cranial specimens variably designated as ‘early Homo’ from East and South Africa leads Prat to identify two species of the same genus (Homo habilis and Homo rudolfensis) and to challenge their inclusion in the genera Australopithecus or Kenyanthropus. These interesting results confirm the original diagnosis proposed by Hughes and Tobias more than thirty years ago. Thackeray and Braga are also concerned with distinguishing Homo from australopithecine remains, albeit in the context of two close deposits, Kromdraai A and B. After discussing the dating of these deposits, they make a case, based on dental morphology, for the presence of Homo at Kromdraai B. The occurrence of very few stone tools in this deposit is attributed to the fact that the site may have functioned as a death trap. The relative abundance of stone tools, most of which are polyhedral in shape at Kromdraai A, a site with no hominid remains, is instead interpreted as an accumulation of large felid prey, occasionally scavenged by either Homo or Australopithecus. Berger’s contribution is also concerned with the acquisition of new data as a way to test hypotheses on early hominin evolution, adaptation and cultural behaviour. In the past decade, the Cradle of Humankind has been the object of numerous field

4

Searching for common ground in palaeoanthropology, archaeology and genetics

operations (e.g. Coopers, Gladysvale, Plover’s Lake) that are significantly increasing our knowledge of the nature and potential of sites preserved in this area. Berger correctly points out that this unique heritage requires the application of precise recording techniques. This is crucial to understanding site formation processes, the context of palaeontological/archaeological remains, and creating site databases that may be compared. Kuman and colleagues report on a similar endeavour, namely the systematic survey and excavation of open-air Early, Middle and Later Stone Age sites from the northernmost border of South Africa. This is a key area to relate technological changes to occupation of the landscape, and explore the geographic extent of Early Stone Age industries named after national traditions, but which in reality may well represent the same cultural phenomenon. The pioneering studies conducted by Brain have shown that stone tools are not the only archaeological expression of South African early hominid material culture. Bone tools in the form of elongated shaft fragments used in digging activities, so far found at three sites, reflect subsistence strategies that would have seemed beyond the grasp of science until a few years ago. Reappraisal of the South African evidence leads Backwell and d’Errico to identify a number of new bone tools from the key site of Swartkrans, explore the implications of their occurrence through members for the identification of the user (Homo vs Australopithecus), and propose a new functional interpretation based on wear pattern quantification. Contrasting this record with the result of their first-hand analysis of the Olduvai purported bone tools, they recognise the presence of two different bone tool industries, suggestive of distinct cultural traditions. Soodyall and Jenkins’s elegant review of the contribution of genetics to the study of human origins highlights the importance of an interdisciplinary approach to tracing our recent and distant past. The Out of Africa model for the emergence of modern humans, grounded on mtDNA variation, proposed twenty years ago by Cann and collaborators, has since been strengthened and refined, and undoubtedly represents one of the major scientific achievements to shed light on human history. While genetic variation among living peoples and analysis of ancient DNA are often seen as complementary means by which to trace genetic lineage, Jenkins and Soodyall caution that extraction of ancient DNA is a recent technique and not a trivial procedure. Accordingly, obtaining reliable results depends on the application of a strict protocol. Genetics is becoming a necessary friend of palaeontology, archaeology and linguistics. A number of syntheses contrasting the African and Eurasian archaeological records and documenting the emergence of behavioural modernity have recently been advanced. The synthesis offered here by Conard comes from a scholar who has extensive knowledge of both the European and the southern African evidence. He is

5

From Tools to Symbols

furthermore involved in excavation of sites such as Geissenklosterle, which are crucial for the understanding of the Middle–Upper Palaeolithic transition in Europe. He concludes that the emergence of modern behaviour is gradual and heterogeneous in space and time. The appearance of personal ornaments, figurative art and other classes of artefacts, including musical instruments, points to a punctuated development of fully modern behaviour during the middle of the Upper Pleistocene, and certainly no later than forty thousand years ago. This leads him to reject a strict monogenetic model in favour of a pattern of highly variable polygenetic development. Marean complements this picture by providing a model that resolves the apparent paradox of a successful species adapted to temperate/cold climate, the Neanderthals, replaced by another, ourselves, having evolved in the tropics. In his view the answer lies in substantially different subsistence strategies employed by each, one dependent on high-risk confrontational hunting, the other relying on more technically flexible hunting practices requiring complex cultural transmission. A second model for this contact is provided by Lewis-Williams’s intriguing contribution. This essay represents a precise and reasoned formalisation of the image that most researchers from outside of Europe, not primarily concerned with the archaeological evidence, have of Neanderthal cognition. This gives the freedom to explore the more recessed aspects of ancient minds and offers ground for discussion and future empirical testing. The interest of this perspective is that it comes from a scholar with a vast knowledge of the symbolic material culture produced by African traditional societies, and the way symbolic systems shape human behaviour. Resolutely engaged in a hypothesis-testing approach, Soressi reaches similar conclusions to those of Conard, and contradictory to those of Lewis-Williams. From her study of lithic technology produced by Neanderthal communities before the arrival of Modern Human in Europe (oxygen isotope stage 3), she identifies patterns of innovation, geographic variations indicative of regional identity, and longterm planning strategies that support the hypothesis of a gradual evolution toward behavioural modernity of Neanderthals at much the same time as similar changes are observed in the African Middle Stone Age. Documenting these technological changes is the main concern of Wurz, who recognises and quantifies at Klasies River technological differences between MSA I and MSA II lithics. These assemblages were previously considered indistinguishable, or part of a continuum, and interpreted as demonstrating pre-modern behaviour. Wurz appropriately points out that this is more a function of a lack of detailed studies than a real stasis in technological evolution, and that new analyses of multi-layered sequences and inter-site comparisons will be crucial to understand, if not interpret, this record in terms of the emergence of modernity.

6

Searching for common ground in palaeoanthropology, archaeology and genetics

Henshilwood’s contribution focuses on Blombos Cave, a Middle Stone Age site that has arguably provided one of the most important records for the origin of behavioural modernity in the last ten years. Questioned by some authors, the stratigraphic integrity of the Blombos sequence is discussed in detail, and the most outstanding recent finds described. The symbolic meaning of Blombos MSA material culture indicates modern cognition not previously associated with Middle Stone Age people. Jacobs’s results on the optically stimulated luminescence dating of MSA layers, and the sterile sand layer separating the LSA and MSA units at Blombos Cave, complement well the data presented by Henshilwood. They confirm the antiquity of the MSA layers and of the archaeological material therein, and provide a better understanding of the site formation process and phases of human occupation. Another currently excavated South African site yielding a remarkable record is Diepkloof in the Western Cape. Parkington and his colleagues report the discovery of the oldest known engraved ostrich eggshells, occurring in the Howieson’s Poort levels of this site. They focus on the stratigraphic and spatial provenance of this potentially symbolic evidence, and present the environmental and technological context in which this novel behaviour appears. The dating of these finds is critical for any attempt at modelling the emergence of behavioural modernity. This is the subject of the research undertaken at a number of MSA sites by Tribolo, Mercier and Valladas. Thermoluminescence dating of the Howieson’s Poort layers at Klasies River and the Still Bay layers at Blombos Cave, together with the review of dates previously obtained with this and other methods for South African sites, demonstrates the precedence of the Still Bay over the Howieson’s Poort techno-complex. It also contributes, together with OSL dating, to clarify the chronology of the emergence of modern behavioural traits in this region. Williamson’s contribution points to another striking feature of the Sibudu record, the good preservation of residues of plant and animal origin on lithics, and the identification of traces of ochre on a number of tools. Occurrence of these remains suggests intensive plant material and minimal animal processing, an observation that sheds light on stone tool function and provides a sound basis for future comparative studies. Finally, Vanhaeren dedicates her paper to the origin and evolutionary significance of personal ornaments, a behavioural trait consensually considered decisive in assessing the symbolic nature of archaeological societies. The recent discovery of shell beads in the Still Bay layers of Blombos Cave demonstrates that the use of the human body as a vessel for conveying messages is much older than previously thought, and appears to have originated in Africa. Were we successful in our attempt to stimulate a dialogue between disciplines and research traditions? The answer must be ‘yes’ in light of the thought-provoking and

7

From Tools to Symbols

lively debates that took place at the conference, continued during tea breaks, and at night around the dinner table. We see these discussions reflected in a number of contributions to this book, and are glad to see that some colleagues revised former attitudes following discussions at the Johannesburg conference. Young researchers had the opportunity to present their results to an international audience, and to benefit from the comments of leading scholars both in and outside of their field and representing different schools of thought. More mature researchers were introduced to novel approaches. Everybody took the opportunity to reinforce existing and establish new collaborations. No consensus was reached on what made us human and when this happened, although a number of participants seemed to agree on the factors that drove the process. All concurred that no definitive answer will be found without investing a common and concerted effort in Africa.

References Bar-Yosef, O. (2002). The Upper Paleolithic revolution. Annual Review of Anthropology 31, 363– 93. d’Errico, F. (2003). The invisible frontier. A multiple species model for the origin of behavioral modernity. Evolutionary Anthropology 12, 188–202. d’Errico, F., Henshilwood, C., Lawson, G., Vanhaeren, M., Tillier, A-M., Soressi, M., Bresson, F., Maureille, B., Nowell, A., Lakarra, J., Backwell, L. & Julien, M. (2003). Archaeological evidence for the emergence of language, symbolism, and music – an alternative multidisciplinary perspective. Journal of World Prehistory, 17, 1–70. Henshilwood, C.S. & Marean, C.W. (2003). The origin of modern human behavior: critique of the models and their test. Current Anthropology 44, 627. Holliday, T.W. (2003). Species, concepts, reticulation, and human evolution. Current Anthropology 44, 653. Klein, R.G. (1999). The Human Career. Chicago: University of Chicago Press. McBrearty, S. & Brooks, A.S. (2000). The revolution that wasn’t: a new interpretation of the origin of modern human behaviour. Journal of Human Evolution 39(5), 453–563. Whiten, A., Goodall, J., McGrew, W.C., Nishida, T., Reynolds, V., Sugiyama, Y., Tutin, C.E.G., Wrangham, R.W. & Boesch, C. (1999). Cultures in chimpanzees. Nature 399, 682–685.

8

The history of a special relationship: prehistoric terminology and lithic technology between the French and South African research traditions Nathan Schlanger Archives of European Archaeology (AREA) Institut national d’histoire de l’art, 2, rue Vivienne, 75002 Paris, France

Our European world, especially in its western part, is a cul-de-sac into which human waves, arriving from the east or the south under unknown impulses, have come to mix and superpose their sediments (Breuil, 1912, 170). Impression at the end of day: Europe as much a cul-de-sac and as confused as South Africa (C. Van Riet Lowe, 3. 10. 1931. Notebook ‘France 1931’, RARI archives n° 19/81).

Abstract The longstanding relationships that developed between the French and South African research traditions in prehistoric archaeology were mostly based on shared goals and perspectives, but also included moments of tension and misunderstanding. Focusing here on the crucial decades of the 1920s to 1940s, and on both the publications and the archival sources of such major actors as John Goodwin, C. Van Riet Lowe and Henri Breuil, my aim in this paper is to show that these interactions were essentially productive for both parties. In the case of the famous ‘African terminology’ devised by Goodwin and Van Riet Lowe, it appears that French scholars had actually welcomed these propositions as fully compatible with their own modes of archaeological designation and reasoning – and therefore that its presentation as a deliberate alternative to the European scheme was primarily a rhetorical disciplinary tactic. This differs from the case of lithic technology, where South African scholars of the times made genuinely innovative contributions to the study of stone artefact raw materials, production

9

From Tools to Symbols and use. The originality of these advances is due, in my view, to two factors – the presence of the Bushman, and the absence of flint. Though their specific origins have since largely been forgotten, these ideas left their mark on leading French prehistorians, and developed in subsequent generations into a full-fledged technological approach. The historical appraisal of these decades-long interactions can thus shed light on the production of archaeological knowledge, and also confirm the potential and scope for further collaborations between these two research traditions.

Résumé Les liens qui se sont développés au fil des décennies entre les traditions de recherches préhistoriques françaises et sud-africaines étaient principalement fondés sur des objectifs et des perspectives communs, mais ont connu aussi des moments de tensions et de discorde. En m’attardant ici sur la période clef des années 1920 à 1940, et sur les publications et les archives d’acteurs principaux tels J. Goodwin, C. Van Riet Lowe et H. Breuil, je tenterai de montrer que ces interactions étaient essentiellement et mutuellement fructueuses. En ce qui concerne la fameuse ‘terminologie africaine’, elle reçut en fait un accueil favorable auprès des savants français, qui la reconnaissaient comme pleinement compatible avec leurs démarches et leurs nomenclatures. La présentation de cette terminologie comme une alternative explicite au schéma européen s’apparente plutôt à la tactique disciplinaire. Le cas de la technologie lithique est bien différent, puisque les chercheurs sud-africains de l’époque ont apporté quelques contributions véritablement innovantes à l’étude des matières premières lithiques, de leur production et de leur usage. L’originalité de ces apports découle, à mon avis, de deux facteurs: la présence des Bushmen, et l’absence de silex. L’origine sud-africaine de ces idées a été depuis oubliée, mais elles ont fortement influencés plusieurs préhistoriens français, et se sont développées par la suite en une véritable approche technologique à part entière. La prise en compte de ces interactions nous apporte un éclairage historique et critique sur la production du savoir archéologique, et confirme aussi les perspectives ouvertes à des collaborations futures entre ces deux traditions de recherches.

Introduction From one ‘cul-de-sac’ to another – as they were often labelled during the first half of the twentieth century, with resigned realism when applied to themselves and rather more disparaging intent when directed at the other – there developed between the French and South African traditions of prehistoric research something of a special relationship, involving over the years both common goals and occasional misunderstandings, lengthy spells of closely-knit connivance enlivened by sporadic feelings of antagonism and strain. These interactions were particularly intense during the crucial decades of the 1920s to 1940s, the self-acknowledged ‘formative years’ of South African prehistoric archaeology which were years of important transformations in French prehistory as well. While South African science as a whole continued to entertain privileged linguistic and institutional links with Britain, the former imperial

10

The history of a special relationship

power and ‘mother-country’, prehistoric archaeology was then (still) a predominantly French discipline, with French research and publications a compulsory reference point for any investigation of the topic. As I address here several moments and aspects of this ongoing relationship, I will highlight the primarily productive and beneficial effects it has had on both parties – even if not always at the same time, or on the same grounds. This appraisal will amply confirm the need for us to take this history on board, as we seek today to reinforce the bonds between French and South African prehistoric research. Now archaeologists, by trade and by vocation, should really need no convincing that such a history matters. Without digressing at any length on the goals and methods of disciplinary history, we can agree at the onset that this history is not condemned to be a mere celebratory ‘warm-up’ act for present-day preferences and perspectives. Rather, the same sensitivity we routinely invest in the contextualisation and interpretation of unearthed remains from the past could profitably be directed to the theories and practices by which, historically, these remains have been and continue to be made sense of. The past is too rare and precious and rich in implications – for science and for society at large – for us to gloss over or ignore the process of its construction. On the contrary, we need to take on board the diverse scientific, cultural, ideological and political settings in which archaeology is being practised and deployed, with all their possible conflicts and contradictions. This in turn puts on us the onus to critically analyse and integrate all relevant bodies of evidence – the vast corpus of scientific publications, of course, but also the range of correspondence, diaries, notes, sketches, drafts and other such archived minutiae which do not necessarily represent formal, authoritative or measured statements intended to reach the public domain, but which nevertheless fully evidence science in the making.1 Archaeological actors, their deeds and dealings, loom particularly large in the scientific relationship under study here. Not only were the leading protagonists conveniently few in numbers – essentially Astley John Hilary Goodwin (1900–1959) and Clarence ‘Peter’ Van Riet Lowe (1894–1956) on the South African side, and in France the travelling Abbé Henri Breuil (1877–1961) and arguably also the ghost of Gabriel de Mortillet (1821–1898) – they also generated a vast output of relevant and readily accessible publications, and furthermore accumulated extremely rich and diversified archival holdings which are only now beginning to be explored.2 Upon these sources, I turn in the first part of this paper to the contentious topic of the ‘African terminology’ of prehistory, as propounded by Goodwin and Van Riet Lowe from the mid-1920s onwards. Here, contrary to expectations, we will see that French scholars were actually well acquainted with these propositions, which they readily endorsed as fully compatible with their own modes of archaeological designation and

11

From Tools to Symbols

reasoning. The terminological posturing emanating from South African scholars really becomes intelligible as a strategy of scientific self-affirmation on both international and domestic scenes – a worthwhile manoeuvre, no doubt, but one that needs to be acknowledged as such. Very different is the case of lithic technology, as discussed in the second part of this paper. Here, on the contrary, we are dealing with a genuinely original South African perspective on the study of stone artefact raw materials, production and use – a contribution that was taken on board by leading French prehistorians of the times (and brought to fruition in subsequent generations), and yet seems to have faded out of disciplinary consciousness, in South Africa as much as in France. As we can imagine, recovering this forgotten dimension will only enhance our prospects for studying, together, techniques and symbols in human evolution.

Figure 1

The Somme valley, 1931. C. Van Riet Lowe visiting Henri Breuil. Van Riet Lowe Papers. Courtesy of the Rock Art Research Institute, University of the Witwatersrand.

12

The history of a special relationship

The African terminology: breaking away from de Mortillet Actually the similarities have been very much overemphasised and over-stated, by European prehistorians who see things through eyes trained in Europe (…) The fact that France formed an eddy, a backwash, at that time does not merit their controlling what types we shall find in Africa. (Goodwin to Smuts, 4 April 1932, Smuts papers 49/107).

The rapid development from about 1925 onwards of a new terminological framework of prehistoric periods and cultures is recognised by many as the founding act of the distinctive South African research tradition – by its chief proponents Goodwin and Van Riet Lowe, of course, but also by most subsequent commentators in South Africa and beyond.3 This ‘African terminology’ has much to do in its conception and promotion with the ‘European terminology’ it was designed to replace, and a few words of background on the latter are therefore in order. Backlashes Upon the establishment of high human antiquity in the 1860s, Palaeolithic archaeology emerged in France as a fully-fledged scientific discipline, with its own journals, congresses, museum displays and indeed its terminology. By 1864, the Stone Age had been divided by John Lubbock into an older or ‘Palaeolithic’ period with chipped stone tools, and a younger, ‘Neolithic’ period with polished implements. As the time depths involved were further grasped and explored, the Palaeolithic came under the sustained attention of Gabriel de Mortillet, a geologist-engineer who became the curator for prehistoric archaeology at the Musée des antiquités nationales and later a professor at the Ecole d’anthropologie. Initially inspired by E. Lartet’s palaeontological classification (Mammoth, Cave Bear and Reindeer Ages), de Mortillet crucially believed that the history of human progress should be reflected in and reconstructed through human creations. This led him to promote a classification based on archaeological artefactual criteria involving different types of stone implements; bifaces (or ‘coupde-poing’ as they were then called), flake-tools, retouched blades, etc. (de Mortillet, 1872, 1883). Drawing on evidence from across France, he identified the succession of Acheulean, Mousterian, Aurignacian and Madgalenian epochs – so named, following geological practices of nomenclature, after the site or locality where the type fossil (effectively the dominant or distinctive form of stone implement) was first isolated and published. Over the years further epochs were gradually added (Chellean, Solutrean, etc.), the term ‘culture’ began to be used to describe them, and they were recast in a threefold partition of the Palaeolithic into Lower, Middle and Upper periods. By the end of the century, this terminological system became widely acknowledged and used,

13

From Tools to Symbols

often with marked rigidity and dogmatism. Combining transformist and materialist postulates with the evolutionist tenets of Victorian anthropology, it promoted the unilinear and immutable succession of inherently progressive stone industries on a global scale.4 In these terms, then, it made very good sense for Goodwin and Van Riet Lowe to question the universal applicability of this chrono-cultural framework, and to seek instead a specific scheme appropriate to the prehistoric sequences of southern Africa. Working in tandem or separately, they gradually identified an array of prehistoric phases, cultures or industrial variants called Stellenbosch, Victoria West, Fauresmith, Still Bay, Smithfield, Wilton, etc., set within a tripartite division of the African Stone Age into Early, Middle and Later (ESA, MSA, LSA). Prehistorians will have of course much more to say on this scheme and on its empirical grounding, details and developments. What primarily concerns us here, however, is the deliberate presentation of this ‘African terminology’ as an alternative designed to supersede or topple its European (read, French) counterpart. This radical ‘break-away’ position was proclaimed by both Goodwin and Van Riet Lowe, who presented their terminological ‘emancipation’ as an historical achievement, localised in time and space: At the annual session of the South African Association for the Advancement of Science, held in Pretoria in July 1926, the founding of this new system of describing local prehistoric periods was accepted by all local prehistorians, and it was decided definitely to abandon the direct use and application of European terminology (Van Riet Lowe, 1929, 152).

For Goodwin, the adoption of these terminological resolutions effectively brought an obsolete chapter to a close, and ‘paved the way’ for ‘a new period in the history of South African prehistory’ (Goodwin, 1935, 334 ff.). Welcomed as a ‘new dawn’ for scientific research, this terminological moment was also heralded as the rallying-cry of a dissident movement, indeed a deliberate defiance of the established order. Drawing a parallel with the late nineteenth-century recognition of the independence of cultural developments in North America ‘whether or not in accord with the Old World determination’, Van Riet Lowe (1929, 151–152) urged for a ‘complete break away from the European school’. Goodwin was even more forceful in his critique of European prehistory, as expressed in the (above quoted) 1932 exchange with J.C. Smuts, the leading statesman and champion of prehistory in southern Africa (see Schlanger, 2002b for full quote and background). As Goodwin further confirmed in one of his retrospective pieces, this vehemence derived at least in part from a heightened colonial sensitivity:

14

The history of a special relationship The classic system of archaeological terminology was evolved in a limited field (parts of France only) at a time when the ethnic intricacies of ethnology and prehistory were incompletely understood, and the problems that would eventually arise were by no means fully appreciated (…) By the chance spread of European culture and colonisation to distant lands, it was only natural that the same developments, the same sequence of incoming cultures and local evolutions found in the glacier-limited habitable patches of western Europe should be sought (and indeed found with great facility) in these outside areas. The conservatism of early training in the European field has much to do with this, and it is easily understood how discoveries made in distant parts of the world came to be fitted to the known pattern (Goodwin, 1945, 91–92).

These critiques and objections sound plausible enough – but were they really well directed, or at all valid? Had the African terminology really been such a ‘novel and provocative venture’ valiantly established against the reactionary resistance of French prehistorians, subjected, as Van Riet Lowe (1936, 199) claimed, to ‘much criticism, even scorn’? No doubt, this is the impression the founders wanted to leave us with, and virtually all subsequent commentators on their wake. Nevertheless, a more critical and attentive appraisal, taking into consideration a broader range of evidence – in France as well as in South Africa – casts these matters in a different light. Far from upholding a vanguard position, Goodwin and Van Riet Lowe rather seem driven by the zeal of latecomers, lagging behind by something like a generation, punching straw men of their own making. Indeed, we will see that the South African terminological innovations actually had many precedents, and also that they were positively received by French prehistorians, in line with the strong theoretical and practical affinities between these two research traditions. Antecedents Let us first acknowledge that prehistorians had not been waiting for the 1920s to question de Mortillet’s scheme and challenge its rationale. For all the acclaim it achieved in its heyday, some of its central tenets have from the onset given rise to misgivings. Besides alternative schemes – that of the palaeontologist Edouard Lartet, but also the parallel strands proposed by the Belgian Edouard Dupont (1872) – explicit critiques were also raised over its universalistic ambitions and its occlusion of ethnic specificities: already in 1872 de Mortillet was taken to task by Abbé Bourgeois and A.W. Franks of the British Museum (de Mortillet, 1872). These objections were amplified by de Mortillet’s colleague and successor at the Musée des antiquités nationales, the classical scholar Salomon Reinach, who severely questioned the assumption of uniform progress and the absence of historical considerations (Reinach, 1889).

15

From Tools to Symbols

By the first and second decades of the twentieth century, the culture-historical approach was gaining momentum in both archaeological and ethnographic circles. While it retained many evolutionist assumptions regarding the unequal development of civilisations or the relations between biology, race and culture, this diffusionist paradigm focused attention on the origins, movements and contacts of localised prehistoric and ethnographic cultures, whose specific traits were eagerly sought after in material remains, arts, languages or myths. These views left their mark on arguments of ancestry, interbreeding or side-branching in the field of human palaeontology, and they also dominated the practices and interpretations of palaeolithic archaeology, as evidenced in such popular syntheses as H. Obermaier’s Fossil Man in Spain (1924), M.C. Burkitt’s Prehistory: A Study of Early Cultures in Europe and the Mediterranean Basin (1921) or the later editions of W. J. Sollas’s Ancient Hunters and Their Modern Representatives. But above all, this ascending perspective owed its coherence and appeal to the labours of Henri Breuil, who notably marked with his omnipresent influence all the works listed here. A staunch opponent of de Mortillet’s school both on empirical grounds (over the succession of Upper Palaeolithic industries) and on conceptual issues (with his resolutely anti-materialist evolutionism), Breuil enlisted his formidable expertise on stone implements and quaternary stratigraphy to promote a global cultural-historical vision of prehistory. The thrust of his position was most eloquently expressed in his programmatic statements at the 1912 International Congress of Prehistoric Archaeology and Anthropology in Monaco. The times are over, he asserted, when one could dream of a simplistic evolution, everywhere identical (Breuil, 1912, 169). What had previously been taken to be a continuous and unilinear series of human development was in fact the product of contacts between different tribes, interacting through exchanges or invasions: Thus, just as in palaeontological studies, the phylogeny of stone industries forces us to constantly admit the existence of multiple roots; none of the civilisations which had developed in our Western Europe can be said to be autochthonous in the full sense of the term; all have roots on neighbouring continents (...) [and it is there that we shall find] the solution to all these problems of origins that cannot be resolved in Europe, this small peninsula clinging on to Asia and to Africa (Breuil, 1912, 238).

This resolute ‘de-centring’ of Europe, at least so far as origins and contacts were concerned, obviously encouraged renewed interest in ‘world prehistory’. Moreover, these general interpretative dispositions also had practical repercussions on classification and terminology. In this vein, several newly identified and named cultures or industries had been advanced in North Africa since the turn of the century.

16

The history of a special relationship

Paul Pallary identified in 1909 the ‘Ibero-Maurusian’ as a Microlithic industry in North Africa with Spanish affinities. The following year Jacques de Morgan and colleagues defined in Tunisia the ‘Capsian’ as an industry comparable to the European Aurignacian, albeit independently evolved. Maurice Reygasse proposed in 1919 a post-Acheulean ‘Aterian’ in Algeria, described in an aptly titled ‘Note au sujet de deux civilisations préhistoriques africaines pour lesquelles deux termes nouveaux me paraissent devoir être employés’.5 All these new industries had been classified and designated by researchers with previous archaeological experience in metropolitan France. Notwithstanding their direct contact with European science, which usually saw North Africa as part of its own circum-Mediterranean backyard, these inventors were not particularly inhibited or oppressed, nor indeed did they feel excessively daring or revolutionary, with the specification and naming of local cultures. Their proposals were initially made in regional publications, and the more sustainable were regularly advanced in metropolitan gatherings to be finally consecrated in l’Anthropologie, whose editor and reviewers proved to be particularly attentive (Roubet, 1979, 24 ff.) – as indeed they will be to the terminological propositions emanating from further south. Reception There is a double irony in the first appearance, in French and in France, of Goodwin’s early work – his 1926 Handbook for the South African Museum. First, while this handbook was extensively translated in the months following its publication, this was apparently done without Goodwin’s knowledge (Goodwin, 1935, 380): full acknowledgments of his authorship can spare us however a misguided trial of plagiarism (witness the ‘Burkitt affair’, Schlanger, 2003). Next, the scholar who recognised the importance of publishing this work in his journal, l’Homme préhistorique, was none other than Adrien de Mortillet, the son and intellectual heir of Gabriel de Mortillet. And he could not be more appreciative in his introductory paragraph, noting that Goodwin’s very interesting booklet: ‘[differed] from the somewhat arid form which this kind of work usually has, [and] constitutes an excellent small manual in which the author summarises the actual state of our knowledge on the stone ages of South Africa’ (A. de Mortillet, 1926, 305). When the issue of terminology was raised, A. de Mortillet clearly remained unfazed: Until now use was mostly made of terms adopted in Europe to designate the different industries distinguished among the finds recovered in South Africa. Goodwin proposes a new classification. He considers it necessary to abandon the use of European denominations, and to replace them with South African names (A. de Mortillet, 1926, 306).

17

From Tools to Symbols

While this publication alone secured Goodwin a far wider readership than he could expect in South Africa, he was even better served by l’Anthropologie. The world-leading journal proved particularly thorough in its appraisals of prehistoric research in South African, in both its rubrics Mouvement scientifique and Nouvelles et correspondance. Its eminent editor, Marcellin Boule, dedicated considerable space to reviewing basically all of Goodwin’s important publications (Boule, 1927; 1928; 1929). The Stone Age Cultures of South Africa received a remarkably comprehensive six-page review, which fully ‘testifies to the value I attach to this monograph’ (Boule, 1930, 149). These reviews included of course the new terminological propositions, which were reported with interest and approval, and in any case without feelings of antagonism, disdain or threat. Indeed, for Boule, Goodwin does convincingly ‘demonstrate the utility of using a new terminology for South Africa, carefully chosen and differing from the European one’ (Boule, 1928, 160; 1930, 145).6 Given this seal of approval, the new terminology was readily endorsed by French scholars in their presentations or discussions of southern African prehistory. This was the case with Breuil himself, who had long been weary of Johnson’s ill-named ‘Solutric’ culture (Breuil, 1911, 58). Hardly back from his four-month trip to South Africa (in conjunction with the 1929 Joint Meeting of the South African and British Associations for the Advancement of Science), the huge quantities of materials he brought with him still unpacked, his notes still rough, Breuil overviewed current knowledge on the prehistory of South Africa, which he obtained thanks to the research of numerous South African prehistorians and the terminological advances of Goodwin and Van Riet Lowe (Breuil, 1930, 211). This endorsement was expressed in other ways, public and private. Already on the steamer back to Europe, Breuil had scribbled to Burkitt in Cambridge a comprehensive report of his trip: For the industries, the very general lines are, I think, [always secure], but much remains to do – Stellenbosch will be divided at least in 3 or 4 levels. Fauresmith, probably, is two distinct things, the two with little coup de poings, one, more or less Micoque, other, more or less Combe Capelle. The middle stone age is a puzzle of many things (…) I have taken to France, with my Friend Kelly [sic], no less than 56 boxes of material, but I dont hope that will be put in order before Easter (Breuil to Burkitt, 17 October 1929, Burkitt papers, Box 2).

What is more, as part of his crucial scientific and political interventions in South African prehistory, Breuil had also been championing the new terminology on South African soil, lending his authoritative expertise to the interested parties themselves.7 Significantly enough, this is testified by none other than Goodwin himself, writing to Van Riet Lowe:

18

The history of a special relationship Leakey, Jones, Breuil and I think everyone at the [joint SA-BAAS] meeting, agreed on the terms Earlier, Middle and Later, all three, and even said Burkitt was inconsistent in using Lower Palaeolithic for Earlier Stone Age. Breuil actually said don’t attempt to use Upper Palaeo. and Epipalaeo., nor even Middle Palaeo. [Breuil] says no Upper Palaeo is present in the Union, Smithfield must be regarded as largely local in origin (Goodwin to Van Riet Lowe, 12 August 1929, Goodwin papers, Box 8).

Rhetorics This Franco-South African accord on terminological matters – and generalised feelings of goodwill – rested on some fundamental practical and theoretical affinities between the two research traditions. With regard to terminological practices, the conceptual structure of both schemes is fundamentally similar: a first order of nomenclature based on eponymous sites, named after localities where remains of the ‘culture’ or industry in question were first unearthed or identified (be it St Acheul in Northern France, or Stellenbosch in the Western Cape), has been fitted on a more abstract tripartite division, the one couched in compounds of Greek and the other in their English rendition. All parties may have claimed (not always consistently) that there were no one-to-one typological or chronological correlations between the constituent elements of the two schemes, but this obviously did not render them any less compatible and mutually intelligible – a compatibility itself underlain by a common set of ‘culture-historical’ assumptions regarding the methods and goals of prehistoric archaeology. As Van Riet Lowe once put it: Ethnographical and geological data combined will make it possible for us ultimately to determine the approximate centre or centres of diffusion of cultures, and to appreciate those numberless migrations that so influenced primitive man’s progress and development (Van Riet Lowe, 1930, 105; see also Goodwin, 1935, 344).

With ‘culture’ conceived as an undifferentiated expression of some ethnic community or ‘folk’, materialised in distinctive tool types or assemblages, scholars could address favoured questions of movements and dispersals, origins and off-shoots, firmly grasping distribution maps with blank spaces yearning to be filled in with dots, arrows and concentric circles. Incidentally, it is really only from such a perspective that the notion of cul-de-sac can be bandied about, and that it becomes possible to imagine that a given physical location in relation to a continent’s geographical edges can somehow attest to an ineluctable position – in the mainstream or on the sidelines – of the great flow of human history. From their respective peripheries, then, Breuil could perceive the repeated arrival of submerging ‘waves’ into Europe, while Goodwin

19

From Tools to Symbols

felt mired in a backwater ‘eddy’, never really doubting, from his early study on Capsian affinities to the very end of his career, the directionality of cultural flow from the north to the ‘distal end’ of the African landmass. Since it appears to have had so little grounding in fact – as we saw, European responses were not in the lines of ‘how dare they’, but rather ‘at last, well done’ – the excessive belligerence surrounding the African terminology has probably more to do with a ‘cul-de-sac’ angst felt by Goodwin and many fellow South Africans. Indeed, political, demographic and economic expansions after the turn of the century brought uncertainties regarding the country’s position and role on the world stage, separate from but still entwined with the British Empire, European in outlook but African in prospects. One way to elucidate this global destiny was through the ‘South Africanisation’ of science, as devised by J.C. Smuts and others in the mid-1920s, which broadly consisted of clamouring South Africa’s distinctiveness, equating it with the whole southern hemisphere and then setting it on par and in competition with Europe to highlight the antiquity, originality and importance of its scientific riches (Schlanger, 2002b; Dubow, 2000). In this climate of willed emancipation and decolonisation, when questions over ‘South Africa’s place in prehistory’ were intensely debated (Van Riet Lowe, 1930), it made sense to cast the African terminology as an explicit alternative which would secure this emerging discipline its operational basis and intellectual credentials. All the more so when, to broaden the usual focus, it is realised that the African terminology was not only designed for external consumption in hemispheric interactions, but had also to take root and thrive on South African soil. While some local researchers such as Johnson had indeed been making indiscriminate use of European terms, others (e.g. Péringuey, Dunn, Lebeltzer, Van Hoepen) had been advancing their own terminological propositions – some of them quite far-reaching and viable, and couched in English, German, or indeed Afrikaans (see Schlanger, 2003). In this competitive context, Goodwin and Van Riet Lowe’s presentation of their own version as the outcome of a victorious struggle against foreign opposition could only reinforce its appeal, as the one worth fighting for, on the domestic front as well. To conclude on this issue: in its practical and theoretical underpinnings, the African terminology was modelled after and compatible with the European scheme – which is notably why French scholars had no difficulty in accepting it. At another level however, that of scientific-ideological positioning, the promoters of the new terminology used its European predecessor – in many respects an obsolete rendition of de Mortillet’s terminology – as a foil or punching bag against which to assert the viability and legitimacy of their own. French prehistorians would probably have been bemused by this posturing language (had they taken notice of it), but they undoubtedly recognised and approved the contribution of this new terminology to the scientific and institutional establishment

20

The history of a special relationship

of South African prehistory. As we know, the African terminology soon became the most comprehensive scheme of its kind, and went on to expand its area of application to the whole of sub-Saharan Africa and beyond – a success which certainly deserves a dedicated account and assessment of its own. All this, however, should not be allowed to obscure or marginalise the no less crucial contribution of South African archaeology to another major field of prehistoric research – the study of lithic technology.

Lithic technology: on Bushman, flint and French inspirations Incidentally, you give the most intelligible account of flint working I have yet read (V. Gordon Childe to Goodwin, 4 October 1926, Goodwin papers, Box 18). As a matter of fact, European prehistorians do not yet fully understand the Levallois – just as they do not understand the Clacton. In my opinion these are not ‘cultures’ but ‘techniques’ – techniques that persist through several cultures (Van Riet Lowe to Smuts, 10 February 1932, Smuts papers 49/139).

As I see it, the technological approach developed in the early days of South African prehistoric research derived its distinctiveness and originality from two principal factors – the presence of Bushman hunter-gatherers and the absence of flint. To these ostensibly methodological and empirical features should be added some interpretative arguments (regarding the evolution of prehistoric industries), as well as the personal interest and acumen of the individual researchers concerned – the whole inevitably enmeshed in this ideologically inspired climate of hemispheric comparisons and competitions. And yet again, this tense atmosphere engendered paradoxical effects. The terminological unruliness manifested at the ‘distal end’ of Africa, notwithstanding its motivations or validity, did leave local archaeologists better disposed than their European colleagues to venture beyond issues of nomenclature and typology, and thus give a broader compass to prehistoric research. At the same time, their determination to be recognised as major players within the prevailing diffusionist paradigm probably explains why their technological contribution has since been insufficiently recognised, both at home and abroad – and this despite the fact that in its heyday it provided some crucial inspiration for … French studies of prehistoric technology. The presence of Bushman hunter-gatherers Needless to say, the presence of Bushman people has obvious and fundamental implications for South Africa in general, for its history, society and identity – in the first half of the twentieth century and even more so today. Pending a more thorough

21

From Tools to Symbols

discussion, I simply mention here some points of contact between Bushman huntergathers (or rather the knowledge and image of them) and lithic technology. So far as ethnoarchaeology in the strict sense is concerned, there appear to have been very few reliable or usable accounts of Bushman stone working. These percolated in time into the archaeological literature when for example Goodwin noted that ‘Dr Kannemayer states that that the duckbill was called !kuin, and was used for scraping skins’, or when Miss Bleek provided useful information on Bushman ‘pierced stones’ (Goodwin, 1925, 430; 1926a, 21, 27). Goodwin further recorded some ‘replication studies’ initiated by Wilhelm Bleek, who had the !Xam reproduce their arrowheads on glass, and more generally argued that ‘knowledge of our own Bushman, as they are to-day and as they were when they were first met by the European (…) should give us a number of clues as to our Later Stone Age deposits’ (Goodwin, 1953, 30). Nevertheless, the immediate interpretative or actualistic ‘clues’ provided by the Bushman people were far less significant than the wider conceptual and ideological impacts of their presence. Among its many distinctive features, South Africa was indeed the one place on earth that could simultaneously boast a rich and continuous prehistoric record (as in Europe or the Mediterranean) and the presence of people reputedly ‘still living in the Stone Age’ (as in Australia or Tierra del Fuego). This state of affairs inspired the somewhat contradictory representation of the country as being at once a ‘cradle’ of humanity and a ‘museum’ of its earlier representatives. Even Smuts, for all his holistic anti-positivism, could not resist vaunting South Africa as a great living laboratory, an idea further expressed in both popular and specialised venues by the likes of Van Riet Lowe, Goodwin, Burkitt and Dart. The postulated coexistence of different evolutionary rhythms notably helped to render stone artefacts more intelligible – in contrast to Europe (with its unfathomable antediluvian ceraunia) those of South Africa hardly presented any material or cultural alterity, neither for the collecting farmer nor for the professional scientist. It was indeed easy to envision a continuous lithic production beginning with the dawn of humanity at the ESA and progressing uninterrupted through to European conquest and the end of the LSA, ‘now known to have lasted until 1870 at Kimberly, where implements of bottle glass and of plate glass appear’ (Goodwin, 1926b, 785). Furthermore, the positioning of Bushman hunter-gatherers within the prevailing Theal-Stow hypothesis confirmed the notions of waves and cul-de-sacs, and added credence to their further extension backwards into the Stone Age – just as the white presence in South Africa served to render plausible continent-wide population movements. Lastly, besides serving to foster historical and ethnographic familiarity, the Bushman ‘folk’ directed attention to archaeological variability and its possible interpretation. In the spirit of Smuts’s legitimising ‘cheekby-jowl’ account of hominid diversity in the Johannesburg area (Schlanger, 2002b,

22

The history of a special relationship

207), Goodwin advanced an argument whose archaeological potential remains valid despite its manifest historical perversity: Just as typical flint techniques were confined to flint areas, so the distribution of other stones seems to have had a similar limiting effect. The people of Africa have always varied from one area to another. There is room to-day for Bushman, Bantu and European cultures to exist side by side, and similar contrasting cultural groups have always been typical of Africa (Goodwin, 1938, 246).

The absence of flint Rather more than an innocuous geological fact, then, the absence of flint in southern Africa became within the prevailing ideological ambiance a genuine question-raising and knowledge-generating factor. In the European Palaeolithic record, the ubiquity of flint had quite naturally led typological and technological assessments to be based on the physical properties of this raw material, both for the ancient knapper – a relatively controllable easy-to-work-with isotropic material – and for the modern analyst – a material on which are retained non-commutative and sequential traces of knapping actions, in the form of conchoidal waves, ripples and scars (see Inizan et al., 1999). By contrast, the various diorites, lydianites, quartzites, shales and such used throughout the southern African Stone Age have markedly different physical properties, with their own effects on the description and interpretation of lithic remains. Considered first as archaeological documents, these raw materials are overall less ‘readable’ than flint – or rather, like faded manuscripts, their reading requires much closer and systematic scrutiny. Directions, sequences and patterns of knapping, as evidenced on cores and debitage products, cannot be taken for granted and have instead to be pointed at and argued. This was notoriously the case with Victoria West, an ESA industry whose characteristic implement resembled a coup-de-poing on one face while displaying the negative scar of a large removal on the other. First described by Jansen (Jansen, 1926; Smith, 1919), these items gave rise to such controversy that a small expert commission was set up by the South African Association for the Advancement of Science. Once it was ascertained that these dolerite items were not, as some had claimed, merely the natural outcome of thermal fracturing, it remained to establish whether they were core tools and therefore desired end products as such – as argued by Jansen – or rather prepared cores from which a large flake had been detached – as opined by Goodwin, Heese, Van Hoepen, etc., in line with Reginald Smith’s initial association of Victoria West with the Levallois ‘tortoise core’. These unfolding arguments implicated such pertinent aspects as experimentation to assess the tractability of the material, weathering to account for its poor readability, and

23

From Tools to Symbols

taphonomic and behavioural factors to explain the dearth of large detached flakes in archaeological deposits (Goodwin & Van Riet Lowe, 1929, 53–69). Here and elsewhere, Goodwin, Van Riet Lowe, Neville Jones, Jansen and others had to use or devise a dedicated descriptive vocabulary and also a satisfactory ‘visual language’, including bold lines, cross-hatchings and arrows, to represent and interpret relevant technological features. This understanding was furthermore amenable to tangible display: ‘specimens showing this [Victoria West] technique are to be seen in Case C.3 [of the South African Museum], an arrow marking the point where the blow was struck removing the final flake’ (Goodwin, 1926a, 19). All these innovative practices no doubt contributed to the intelligibility of the account so applauded by V. Gordon Childe – notwithstanding the ironic caveat that even the celebrated Marxist archaeologist took it for granted that it referred to ‘flint working’! This commonplace assumption left its mark also on the actual interpretations of prehistoric cultures and industries. Overall, South African researchers concurred that their local rocks were lesser quality ‘substitutes’ for the raw material par excellence. This ‘second-best’ conception engendered a sustained reflection on the inhibiting or

Figure 2

Unworked and slightly worked flakes, showing the types of workmanship and of flakes used (after Goodwin, 1928, 420, Fig. 1).

24

The history of a special relationship

Figure 3

Early examples of primitive technology in the use of stone (after Goodwin, 1953, 59, Fig. IV).

Figure 4

Silcrete cores. The thicker lines indicate the areas from which implement flakes have been removed (after Jones, 1924, 282, Fig. 4).

25

From Tools to Symbols

enabling qualities of raw materials, and on their typological incidence. Van Riet Lowe thus contended that ‘if the settler here had any chert, agate or any truly flinty material he would, in addition to the Smithfield type, also have produced Pygmy types’ (Van Riet Lowe to Goodwin, 16 August 1925, Goodwin papers, Box 7). Elsewhere, he noted that ‘[Those folk of the Vaal Valley] begun using fragments of lydianite or indurated shale and their mastery over this flint-like material is, I feel, largely, if not wholly, responsible for the development of the Fauresmith Culture’ (Van Riet Lowe, 1932, 761). Goodwin too recognised connections between raw material, type and technique, but in some determined passages penned in the mid-1930s he undertook to overturn the prevailing valorisation of (European) flint, and effectively make of its absence a quintessential African virtue. The issue at hand was the development of the Levallois technique, which in his view did not take place in Europe, but in Africa (with the Victoria West and Tachnegit techniques). An ‘isolated’ glaciations-prone area, Europe was actually mired into ‘stagnation’ by ‘the persistent use of the flint nodule, which never allowed (or forced) sufficient variations on the coup-de-poing theme’. This was so because flint, though tough and dependable, generally occurred as awkward nodules with a thick and convoluted crust, which made it necessary to use a knapping method giving the minimum of wastage. ‘But directly a flint area is left, larger material is at man’s disposal, and is in fact actually forced upon him, and developments in technique become possible as soon as the strong need for economy is eliminated’ (Goodwin, 1933, 110). In another article, Goodwin took a different route to reach a broadly similar conclusion: The lack of flint throughout the greater part of the continent has forced man constantly to turn his hand to a great variety of other stones. This has affected the methods of making implements in two ways. There has been the effect of the different cleavage of the rock, but quite as important are the effects of the different sources of material. In Europe flint may generally be found in nodules or ‘rognons’ of a handy size, but in Africa standing rock must first be broken down to obtain the raw material for implement making. To work standing rock, a different primary technique is essential. Both the source and the substance of material have affected technique very profoundly. (...) From these few examples it should be realised that we may expect a greater variety of techniques from Africa, and, as a result, a far greater degree of inventiveness and adaptability on the part of prehistoric man (Goodwin, 1938, 345, emphasis added).

Thus, either way, Goodwin had the prehistoric folks of Africa drawing technological strength and dynamism from raw material adversity, and in the process chipping down flint-rich Europe to its proper size.

26

The history of a special relationship

Traits and traces The presence of the Bushman people (bringing ethnographic familiarity) and the absence of flint (raising issues of raw material type, quality and sources) also underlie Van Riet Lowe’s technological approach. More of a fieldworker than Goodwin, but also something of an archaeological administrator, these inputs led him to promote two distinct technological outlooks, which here I somewhat summarily label ‘traits’ and ‘traces’. With their different scales of application, theoretical grounding and practical implications, both were to have lasting impacts on French prehistoric research. The conception of techniques as cultural traits may be unpopular today, and with good reason, but in its time it was undoubtedly a considerable breakthrough of the culture-historical paradigm. Setting aside evolutionist notions of progress as their sole interest or measure, techniques were enlisted to address the then topical questions of cultural centres and dispersals – an issue to which South African scholars were particularly attentive, given their perception of their own country as a dead-end recipient of inflowing admixtures. Reminding his readers that local rocks were ‘erratic and frequently unreliable’ in comparison with typology-friendly flint, Van Riet Lowe added that: We are rapidly reaching a stage where typology may be said to be of little use and so we are turning to technique. (…) If affinities exist between widely separated cultures, I believe we are more likely to detect them in technique rather than in the types produced. In this regard changes are, I feel, imminent (Van Riet Lowe, 1937, 107–108).

So far as prehistoric archaeology was concerned, this represented indeed a certain threshold: this concept of ‘technical affinities’ is one that marked, for better or worse, the discipline for the subsequent decades. It is worth recalling, however, that already by the turn of the century techniques had begun to be considered markers of civilisations rather than milestones of progress. In this respect, techniques were seen as more veridical indicators than finished products, also because processes of manufacture were themselves beginning to be conceived in a more suggestive anthropological and sociological light, in terms of social habitus and ‘traditional efficient acts’ all the more revealing for being unreflected (on this, see Mauss, 1935; Schlanger, 1998; 2005b). Although these connections were not made by Van Riet Lowe, or for that matter by his French readers, this nascent sociological perspective on techniques had actually much closer affinities with his approach to technical ‘traces’. Consideration of such aspects as assemblage composition, raw material outcrops, knapping procedures and waste products had admittedly been suggested, on other grounds in other research traditions, but in Van Riet Lowe’s case it was probably his early encounter with ancient

27

From Tools to Symbols

sites as a public works engineer that spurred his attention to these issues. This transpires in one of his very first publications, a survey of the stone implement workshops of the Orange Free State which enabled him to establish a distinction between such sites as visitation sites, where but few implements are found, settlement sites, and factory sites where the whole process of manufacture can be studied (…) owing to the number of discards to be found. (…) The two varieties [of hammers found at factory sites] throw considerable light on the process of implement manufacture (Van Riet Lowe, 1925, 426).

Van Riet Lowe continued in subsequent works to address issues of site functions and their technological incidence, noting for example how difficult it might be ‘to detect the same hand and brain behind the factory sites debris and the finished homesite products’ (1937: 101). But it was in his well-known study of ‘The Evolution of the Levallois Technique in South Africa’ (1945) that his ‘traits’ and ‘traces’ outlooks were most effectively combined. The main objective here was apparently to situate Levallois within the evolution of the ‘Great handaxe culture’ and thus oppose the European notion of parallel ‘core-tool’ and ‘flake tool’ phyla.8 At the same time, his committed appreciation of Levallois as a technical process was both timely and welcome.9 He introduced into the debate issues of raw material effects and variability, as could be expected, but also questions of economy – ‘Man had now become a keen economizer and used his cores again and again – i.e. having struck one flake, he re-trimmed the core to strike another and smaller flake and often continued the process’ – and indeed considerations of knapping gestures and methods: Man seems to have reached a stage where he realised that the slenderness, length and breadth of a flake or blade and the degree of rippling depended not only on the force and direction of the blow, but also on the angle between the striking platform and the face of the desired flake. In other words, he applied and exercised a finer control over his flaking methods – a control that is best reflected in the waste-products rather than in the tools of his industry (Van Riet Lowe, 1945, 53).

In fact, Van Riet Lowe had stated his combined ‘traits’ and ‘traces’ credo at the onset of his paper: Typology should not be confined to the tools man made, but needs urgently to be extended to the waste products of human industry of the time; my belief being that it is safer to stress affinities on technological rather than on typological grounds where

28

The history of a special relationship typology is confined, as it all too frequently is, to the final objects of human industry and excludes the rejects and processes men practised in achieving those objects (Van Riet Lowe, 1945, 49).

Conclusions – French inspirations If these notions somehow sound familiar, this is probably because of the comprehensive endorsement they received at the hands of French prehistorians, who adopted them as their own. The young François Bordes, for one, then studying alluvial granulometry while dabbling in flintknapping, made of them the rationale of his ‘Principes d’une méthode d’étude des techniques de débitage et de typologie’. Not only did he reproduce in his seminal 1950 paper Van Riet Lowe’s above-quoted passage in its entirety – in the original English and in French translation – he also elaborated his own guiding hypothesis: Every Palaeolithic civilisation has had at its service one or usually several flintknapping techniques. These techniques may be common to the whole of those civilisations, or on the contrary particular to one of them. We may therefore think that it might be possible to characterise such or such civilisations by these techniques (Bordes, 1950, 19).

True to this principle, Bordes went on to focus on techniques as typological attributes (‘facetted platform’, ‘Levallois’, etc) within his famous Lower and Middle Palaeolithic type-list which long dominated Eurasian Palaeolithic research (Sackett, 1981; Julien, 1992). It was the aging Henri Breuil, surprisingly perhaps, who fully appreciated the technological insights emanating from South Africa. Admittedly, by 1954, when he penned his presidential address to the French prehistoric society, Breuil had already spent quite a few years of his life there. Still, South African evidence and ideas loomed particularly large in his overview of prehistoric classifications. Besides making numerous references to the region’s prehistoric remains (Sangoan picks from Natal, South African cleavers, Middle Stone Age), Breuil also wholeheartedly embraced the idea that ‘There is something more fundamental than forms, with regards to knapped stones; it is technique, that is the method followed to knapp a block of stone, a nodule or pebble, into usable flakes’ (Breuil, 1954, 9). Given the diffusionist climate, Breuil followed this claim with some developments on the ‘techniques as traits’ front – e.g. ‘The knapping technique with prepared striking platform, called ‘levalloisian’, was born in South Africa and thence propagated up to the Pyrenees and Hindustan’ (Breuil, 1954, 10) – but he was also fully convinced of the interest presented by the study of ‘technical traces’. Not only did he urge his audience to compare factory sites, strewn with waste products and unfinished implements, and habitation sites,

29

From Tools to Symbols

Figure 5

Lourenço Marques, 1944. Henri Breuil visiting C. Van Riet Lowe. Caricature signed by Vitela. Goodwin papers. Courtesy of the Manuscripts and Archives Department of the University of Cape Town.

‘richer in good pieces’ (1954, 8); he also considered it particularly important to draw on both experimental data and archaeological remains to reconstruct the ancient processes of stone tool manufacture. His plea in this matter is well worth quoting: The study of fracturing surfaces, of cones and bulbs of percussion, makes it often possible to recognise the procedure or procedures used by ancient humans. We should therefore study the core decortification flakes, the prepared cores which have not yet yielded their

30

The history of a special relationship flake, the unsuccessful flakes, the stages of shaping of tools and weapons, successful or not, or presenting signs of utilisation, the blunted tools on the way to repair, lastly the pieces rejected as no longer of use (Breuil, 1954, 10).

All this sounds very compatible, on the face of it, with the ideas and formulations soon to be expanded into the chaîne opératoire approach by A. Leroi-Gourhan, L. Balout, J. Tixier and subsequent generations of prehistorians and anthropologists. While a much-needed appraisal of the theoretical foundations and practical developments of this approach must remain beyond the scope of the present paper,10 it is nonetheless well worthy of note that contemporary French technological studies derive some of their thrust from specifically South African inspirations. Indeed, to complement the above-noted convergences, direct testimony to this effect is provided by the interested parties themselves, as when Goodwin wrote to Breuil: I am glad that you are tending to my view, that material is the essential factor in the earlier and middle Stone Age – perhaps even in the later, though not to the same extent. I shall send you a copy of some of my remarks on regioning in South Africa, a subject I am beginning to work on now. The difference between material, to my mind, lies not so much in material itself [sic.], but rather in the manner of its occurrence. You will remember in my paper on the Development of Technique (R.S.S.Af. XXI 1933) I brought out the point that what I called the Abbeville technique was related to smallish river pebbles in this country, the Tachnegit to large standing rocks and the Victoria West to bigger pebble material. The more I see of the position the more certain I am that this is true (Goodwin to Breuil, 16 January 1943, Goodwin papers, Box 11).

But then, as a final twist to round up this paper, Goodwin’s archives provide us with another tantalising piece of evidence to confirm that things are rarely as simple and straight-cut as they appear. In an autobiographical passage which he subsequently expunged from the published version (the ‘Formative years of our prehistoric terminology’, 1958), Goodwin recalled how, as a student of the English tripos in Cambridge, in 1920 he joined a vacation course in French phonetics at the University of Grenoble: The course did not take more than the morning, and in the afternoon I worked with Dr Muller, Curator of the Grenoble Museum. He was primarily interested in the Bronze Age and had a curious collection of skulls – Peruvian executed, French criminals etc. He put me through an intensive training in French archaeology – mainly his beloved Bronze Age – and in physical anthropology for two months. This brought me into the field of

31

From Tools to Symbols archaeology. On returning to Cambridge (…) I arranged to meet Dr A.C. Haddon (….) (Typescript, c. 1944, Goodwin papers, box 83).

As it happens, this museum curator from Grenoble did more than kindle and reorient the vocation of his young volunteer from literary to archaeological. Though almost forgotten today, Hippolyte Müller was also a maverick autodidact who, at the turn of the century, launched highly original and meticulously documented lithic experiments, including the ‘complete series of trials of fabrications, of hafting and use, carried out wholly with primitive means and without recourse to metal tools’. Drawing on his skills as a jeweller-craftsman, he undertook technical studies and in situ reconstructions seeking to reach the ‘operational procedures’ followed by the ancient artisans (Müller, 1903; Schlanger, 2004b, c). Whether Müller did impart to Goodwin some of his technological sensitivities, or helped open his eyes to the importance of materials and their fashioning, is a question that will have to remain open pending further evidence. After all, at the end of the day, what matters most is not necessarily the formal or specific attribution of anteriority or directionality, but rather the very fact that there has been such a variety of rich and productive contacts over the years between the ‘culde-sacs’ here in question. Having cleared the air in the preceding pages through some critical-historical discussions of prehistoric terminology and lithic technology, we can also move beyond a polarised geopolitical conception of the scientific enterprise. We can indeed better understand how, at personal, institutional and national levels, unfolding interactions between far-flung research traditions can have an impact on the construction of archaeological knowledge – an understanding which, we will agree, bodes well for the future prospects of this special relationship.

Endnotes 1

For general statements on the history of archaeology, analyses and case studies, see Trigger, 1989; Schnapp, 1996; Murray, 1999–2000; and Schlanger, 2002a.

2

The c. 100-box Goodwin papers (BC 290) at the Manuscripts and Archives Department of the University of Cape Town are an essential resource. Parts of the Van Riet Lowe papers are kept at the Rock Art Research Institute of the University of Witswatersrand, while other fonds remain to be localised. The extensive Breuil archives in France, South Africa and elsewhere are the subject of a dedicated research project funded by the French Ministry of Research (ACI ‘Archives Breuil’, http://www.mmsh.univ-aix.fr/iea/d_fichiers/ACIfauvelle.html). Other archives used in this research include the Smuts papers (microfilms in Cambridge University Library) and the Burkitt papers (Cambridge University Library). I thank the custodians of

32

The history of a special relationship all these archives for giving me access to this material and permission to use and reproduce selections of it. Warm thanks are also due to Francesco d’Errico and Lucinda Backwell for inviting me to participate in this conference. This research has been carried out in the framework of the AREA project (Archives of European Archaeology) with the support of the Culture 2000 programme of the European Commission. All translations from the French are my own. 3

Goodwin & Van Riet Lowe, 1929; Goodwin, 1945, etc. On the history of South African prehistoric research see Goodwin, 1935, 1958, as well as Schrire et al., 1986; Deacon, 1990; Shepherd, 2002.

4

On the establishment of human antiquity, the early prehistoric terminological framework, and nineteenth-century evolutionary prehistory, see Sackett, 1981; Grayson, 1983; Coye, 1997; and Stocking, 1987.

5

Cf. Goodwin 1925, 432 passim, and Roubet, 1979; Coye, 1993, and references therein.

6

l’Anthropologie provides further quantitative evidence of this sustained interest. In the name index for the period 1910–1956 no less than 20 items by Goodwin and 24 by Van Riet Lowe are mentioned. In the thematic index, something like 28 entries (including articles, reviews, notes, etc.) are dedicated to ‘South Africa’, some 30 to its human palaeontology, and 25 each to its prehistoric industries and its prehistoric art.

7

Breuil made decisive contributions to the institutional establishment of prehistoric archaeology in South Africa. Already during his 1929 visit (while receiving a doctorate Honoris Causa from UCT) he struck up a friendship with Smuts, networked and publicly called for the establishment of a University chair, an archaeological service and appropriate legislation. Breuil’s South African deeds are the subject of ongoing research within the ACI ‘Archives Breuil’.

8

Van Riet Lowe opposed the lack of distinction between these phyla (as proposed by Breuil in the 1930s) by stressing the existence of flake tools – and Levallois – in the ‘hand axe’ tradition. This drew an exasperated response from Raymond Vaufray in the South African Archaeological Bulletin: ‘It is really annoying that the same prehistorians to whom we owe the extraordinary progress of prehistory in South Africa should be immured (probably through lack of facility in our language) in such complete ignorance of French publications (…) they [should] know that no Frenchmen (and this certainly includes the Abbé Breuil) has ever had the absurd idea that there were no flakes in the so-called biface industries’ (Vaufray, 1950, 137).

9

On relevant aspects of Levallois research see Perpère, 1981; Schlanger, 1996; White & Aston, 2003.

10

See various aspects in Pelegrin et al., 1988; Julien, 1992, and Schlanger, 2004a, 2005a.

33

From Tools to Symbols

References Bordes, F. (1950). Principes d’une méthode d’étude des techniques de débitage et de typologie du Paléolithique ancien et moyen. L’Anthropologie 54, 19–34. Boule, M. (1927). (Review of Goodwin, 1925). L’Anthropologie 37, 169–70. Boule, M. (1928). (Review of Goodwin, 1926a). l’Anthropologie 38, 160–1. Boule, M. (1929). (Review of Goodwin, 1929). L’Anthropologie 37, 602. Boule, M. (1930). (Review of Goodwin & Van Riet Lowe, 1929). L’Anthropologie 40, 144–149. Breuil, H. (1911). (Review of Johnson, 1910). L’Anthropologie 22, 56–60. Breuil, H. (1912). Les subdivisions du paléolithique supérieur et leur signification. Congres international d’archéologie et d’anthropologie préhistorique, 14th session, Monaco, 165– 238. Breuil, H. (1930). Premières impressions de voyage sur la préhistoire sud-africaine. l’Anthropologie 40, 209–223. Breuil, H. (1954). Prolégomènes à une classification préhistorique. Bulletin de la Société Préhistorique Française 51, 7–15. Coye, N. (1993). Préhistoire et protohistoire en Algérie à la fin du XIXe siècle, les signification du document archéologique. In (G. Boetch & J-N. Ferrié, Eds) Le Savant et le Berbère, pp. 99–137. Cahiers d’Etudes Africaines 129. Coye, N. (1997). La préhistoire en parole et en acte. Méthodes et enjeux de la pratique archéologique (1830–1950). Paris: L’Harmattan. Deacon, J. (1990). Weaving the fabric of Stone Age research in Southern Africa. In (P. Robertshaw, Ed.) A History of African Archaeology, pp. 39–58. London: J. Currey. Dubow, S. (2000). A commonwealth of science: the British Association in South Africa, 1905 and 1929. In (S. Dubow, Ed.) Science and Society in Southern Africa, pp. 66–99. Manchester: Manchester University Press. Dupont, E. (1872). Classement des âges de la pierre en Belgique. Congrès international d’archéologie et d’anthropologie préhistorique, 6th session, Brussels, 459–479. Goodwin, A.J.H. (1925). Capsian affinities of South African Later Stone Age culture. South African Journal of Science 22, 428–436. Goodwin, A.J.H. (1926a). A Handbook of the Collections of the Stone Implements (Guide leaflet no. 2). Cape Town: South African Museum. Goodwin, A.J.H. (1926b). South African stone implements industries. South African Journal of Science 23, 784–788. Goodwin, A.J.H. (1928). Sir Langham Dale’s collection of stone implements. South African Journal of Science 25, 419–426. Goodwin, A.J.H. (1933). Some developments in technique during the Earlier Stone Age. Transactions of the Royal Society of South Africa 21, 109–123.

34

The history of a special relationship Goodwin, A.J.H. (1935). A commentary on the history and present position of South African prehistory with full bibliography. Bantu Studies 9, 291–417. Goodwin, A.J.H. (1938). Recent advances in South African prehistory. Scientia 62, 345–353. Goodwin, A.J.H. (1945). The terminology of prehistory. The South African Archaeological Bulletin 1, 91–100. Goodwin, A.J.H. (1953). Method in Prehistory. An Introduction to the Discipline of Prehistoric Archaeology with Special Reference to South African Conditions. Cape Town: The South African Archaeological Society. Goodwin, A.J.H. (1958). Formative years of our prehistoric terminology. The South African Archaeological Bulletin 13, 25–33. Goodwin, A.J.H. & Van Riet Lowe, C. (1929). The Stone Age Cultures of South Africa. Annals of the South African Museum 27. Grayson, D.K. (1983). The Establishment of Human Antiquity. New York: Academic Press. Inizan, M.-L., Reduron-Ballinger, M., Roche, H. & Tixier, J. (1999). Technology and Terminology of Knapped Stone. Nanterre: CREP. Jansen, F.J. (1926). A new type of stone implement from Victoria West. South African Journal of Science 23, 818–825. Johnson, J.P. (1910). Geological and Archaeological notes on Orangia. London: Longman. Jones, N. (1924). On the Palaeolithic deposits of Sawmills, Rhodesia. Journal of the Royal Anthropological Institute 54, 276–286. Julien, M. (1992). Du fossile directeur a la chaîne opératoire. In (J. Garanger, Ed.) La Préhistoire dans le Monde, pp. 163–193. Paris: PUF. Mauss, M. (1935). Techniques of the body. Republished in Marcel Mauss. Techniques, Technology and Civilisation. Oxford: Berghahn Press (2005). Mortillet, A. de (1926). Le préhistorique de l’Afrique du sud. L’Homme préhistorique 13, 305– 315. Mortillet, G. de (1872). Classement des diverses périodes de l’âges de la pierre. Congres international d’archéologie et d’anthropologie préhistorique, 6th session, Brussels, 432– 459. Mortillet, G. de (1883). Le préhistorique, antiquité de l’homme. Paris: Reinwald. Muller H. (1903). Essais de taille du silex. Montage et emploi des outils obtenus. l’Anthropologie 14, 417–436. Murray, T., Ed. (1999–2001). Encyclopedia of Archaeology (5 vols). The great Archaeologists, Histories and Discoveries. Oxford: ABC/Clio. Pelegrin, J., Karlin, C., Bodu, P. (1988). Chaînes opératoires: un outil pour le préhistorien. In (J. Tixier, Ed.) Technologie lithique. Paris: C.N.R.S. (CRA Notes et Monographie techniques, 25), 55–62.

35

From Tools to Symbols Perpère, M. (1981). A propos de quelques nucléus Levallois africains. In (C. Roubet et al., Eds) Préhistoire africaine. Mélanges offerts au Doyen L. Balout. Paris: A.D.P.F. Reinach, S. (1889). Antiquités nationales. Description raisonnée du musée de Saint-Germain-enLaye, v. 1. Epoque des alluvions et des caverns. Paris: Firmin-Didot. Roubet, C. (1979). Economie pastorale préagricole en Algérie orientale: le néolithique de tradition capsienne. Paris: CNRS. Sackett, J. (1981). From de Mortillet to Bordes: a century of French Paleolithic research. In (G. Daniel, Ed.) Towards a History of Archaeology, pp. 85–99. London: Thames & Hudson. Schlanger, N. (1996). Understanding Levallois: lithic technology and cognitive archaeology. Cambridge Archaeological Journal 6, 231–254. Schlanger, N. (1998). The study of techniques as an ideological challenge: technology, nation and humanity in the work of Marcel Mauss. In (W. James & N. Allen, Eds) Marcel Mauss: A Centenary Tribute, pp.192–212. Oxford, Berghahn Books. Schlanger, N. (2002a). Introduction. Ancestral archives. Explorations in the history of archaeology. Antiquity 76, 127–131. Schlanger, N. (2002b). Making the past for South Africa’s future: the prehistory of Field-Marshal Smuts (1920s–1940s). Antiquity 76, 200–209. Schlanger, N. (2003). The Burkitt affair revisited: colonial implications and identity politics in early South African prehistoric research. Archaeological Dialogues 10(1), 5–26. Schlanger, N. (2004a). ‘Suivre les gestes, éclat par éclat’: la chaîne opératoire de Leroi-Gourhan. In (F. Audouze & N. Schlanger, Eds) Autour de l’homme: contexte et actualité de Leroi-Gourhan. Antibes: Editions APDCA, pp. 127–147. Schlanger, N. (2004b). Une appréciation technologique des expérimentations d’Hippolyte Müller. In Aux origines de la préhistoire alpine. Hippolyte Müller, 1865–1933, Grenoble, Musée dauphinois, pp. 71–73. Schlanger, N. (2004c). L’éveil de la vocation de John Goodwin, préhistorien d’Afrique du Sud. In Aux origines de la préhistoire alpine. Hippolyte Müller, 1865–1933, Grenoble, Musée dauphinois, pp. 91–92. Schlanger, N. (2005a). The Chaîne opératoire. In (C. Renfrew & P. Bahn, Eds) Archaeology: The Key Concepts, pp. 25–31. London: Routledge. Schlanger, N. (2005b). Introduction. In Marcel Mauss. Techniques, Technology and Civilisation. Oxford: Berghahn Press. Schnapp, A. (1996). The Discovery of the Past. London: British Museum. Schrire, C., Deacon, J., Hall, M. & Lewis-Williams, D. (1986). Burkitt’s milestone. Antiquity 60, 123–131. Shepherd, N. (2002). Disciplining archaeology; the invention of South African prehistory, 1923– 1953. Kronos 28. Smith, R. (1919). Recent finds of the Stone Age in Africa. Man 19, 100–106.

36

The history of a special relationship Stocking, G.W. (1987). Victorian Anthropology. London: Macmillan. Trigger, B. (1989). A History of Archaeological Thought. Cambridge: Cambridge University Press. Van Riet Lowe, C. (1925). Stone implement workshops in the Orange Free State. South African Journal of Science 22, 425–427. Van Riet Lowe, C. (1929). In (A.J.H. Goodwin & C. Van Riet Lowe) The Stone Age Cultures of South Africa. Annals of the South African Museum 27. Van Riet Lowe, C. (1930). South Africa’s place in prehistory: A plea for organised research and the better preservation of prehistoric remains. South African Journal of Science 27, 100–116. Van Riet Lowe, C. (1932). The prehistory of South Africa in relation to that of Western Europe. South African Journal of Science 29, 756–767. Van Riet Lowe, C. (1936). Nomenclature of Palaeolithic finds. Man, 36, 199–200. Van Riet Lowe, C. (1937). The archaeology of the Vaal river basin. Geological Survey, Memoirs 35, 61–164. Pretoria: Department of Mines. Van Riet Lowe, C. (1945). The evolution of the Levallois technique in South Africa. Man 45, 49–59. Vaufray, R. (1950). Flake-using and Biface-using peoples. South African Archaeological Bulletin 5, 137–139. White, M. & Aston, N. (2003). Lower Palaeolithic core technology and the origins of the Levallois method in North-Western Europe. Current Anthropology 44, 598–609.

37

Essential attributes of any technologically competent animal C. K. Brain Curator Emeritus at the Transvaal Museum, Northern Flagship Institution, Paul Kruger Street, P.O. Box 413, Pretoria 0001, South Africa

Abstract One of the factors contributing to human dominance among animals is the ability to translate intelligence and imagination into tangible technology. During the last two million years, members of the human lineage have doubled their brain size, despite the fact that neural tissue is metabolically expensive, suggesting that strong selective pressures were driving the process. Some of these were clearly the advantages bestowed by technology, such as the management of fire and the use of more and more sophisticated tools and weapons. But if technology can bestow such valuable advantages, why have many other animals not followed this course to the same extent? It seems to me that there are at least five obvious constraints that would apply to any candidate for the role of a ‘technologically competent animal’. Although some are applicable only if the candidate were a mammal, they are: 1.

A critical minimum brain size. As was emphasised by Phillip Tobias as long ago as 1971, the human brain is characterised by a very large number of ‘extra neurons’, over and above those needed to handle the basic functions of the body. For human-style intelligence, it seems that a volume of at least 500 cc of ‘extra’ neural tissue-hardware is required, so a constraint is imposed on the minimum size of a mammal that could accommodate this. The constraint does not seem to be as rigidly applied in the case of birds.

2.

Suitable appendages for manipulating objects related to the technology in question.

3.

A social organisation that would promote the collective effort required for the successful fulfilment of any major technological undertaking.

4.

A language-like communication system that would allow the exchange of concepts between individuals.

5.

Appropriate birth-canal adjustments to allow the passage of the offspring’s large head in such technologically competent mammals. Various animals fulfil these requirements to some extent and could go on to become completely technologically

38

Essential attributes of any technologically competent animal competent in the course of future evolution, given the appropriate selective pressures. Some such animals are discussed in this paper.

Résumé Un des facteurs contribuant au succès de notre espèce réside dans sa capacité à traduire son intelligence et son imagination en une technologie performante. A cours des deux derniers millions d’années, les membres de la lignée humaine ont doublé la taille de leur cerveau en dépit du fait que le tissu neuronal est coûteux du point de vue métabolique, ce qui suggère que de fortes pressions sélectives dirigeaient le processus, dont les avantages offerts par la technologie, tels que la gestion du feu et l’utilisation d’outils et d’armes de plus en plus sophistiqués. Mais si la technologie peut offrir des avantages aussi précieux, pourquoi d’autres animaux n’ont-ils pas suivi cette même direction? Il semble que cinq conditions sont nécessaires pour atteindre le statut d’animal “technologiquement compétent”, Certaines de ces conditions se rencontrent exclusivement chez les mammifères. Il s’agit de: 1.

1.Une taille critique minimale du cerveau. Phillip Tobias l’avait souligné dès 1971, le cerveau humain est caractérisé par un grand nombre de “neurones supplémentaires”, en supplément de ceux nécessaires au traitement des fonctions de base du corps. Pour une intelligence de type humain, il semble qu’un volume d’au moins 500 cm3 de tissu neuronal supplémentaire soit nécessaire, ce qui impose des contraintes sur la taille minimale du mammifère qui pourrait posséder un tel cerveau. Le fait que cette contrainte ne semble pas s’appliquer de manière aussi stricte aux oiseaux.

2. 3.

2.Des appendices appropriés pour manipuler les objets liés à la technologie en question. 3.Une organisation sociale qui favoriserait l’effort collectif nécessaire à la réalisation de toute entreprise technologique d’envergure.

4.

4.Un système de communication comparable au langage humain qui permettrait l’échange de concepts entre individus.

5.

5.Une taille appropriée du canal pelvien permettant le passage d’une progéniture pourvue d’une tête de fortes dimensions. Divers animaux remplissent, dans une certaine mesure, ces conditions et pourraient donner naissance à des

sociétés “technologiquement compétentes” une fois soumises aux les pressions sélectives appropriées. Nous traiterons de ces animaux dans cette contribution.

Brain expansion in the human lineage By any standards, the increase in the size of the brain relative to that of the body in our human ancestors during the last two million years was a remarkable zoological event. When the earliest known members of the Homo lineage appeared on the scene, in the form of H. habilis or H. rudolfensis, their average brain capacity was about 650 cc; this had risen to about 850 cc in H. ergaster and H. erectus, and to 1 400 cc in archaic H. sapiens towards the end of the Middle Pleistocene. As Leslie Aiello and

39

From Tools to Symbols

Peter Wheeler (1995) pointed out, this event is all the more remarkable in view of the fact that a brain is built of ‘expensive tissue’. Although a human brain may only make up 2–3 per cent of the weight of the whole body, it uses 16–20 per cent of the energy consumed by the resting body. To double the size of the brain relative to that of the body would usually mean that the basic metabolic rate (BMR) of the animal would have to be substantially increased. Strangely enough, this has not been observed in humans, when compared to related primates, and Aiello and Wheeler concluded, therefore, that human brain expansion occurred at the expense of the size of the gut, which has apparently shrunk during the course of human evolution. To be able to function with a much smaller gut implies that ancestral humans changed to a diet of higher quality, such as one regularly including animal protein, and they would have done this by scavenging and active hunting. It has been customary to think of humans as being unique in so many ways, but this is not always true. When we look at brain weight relative to that of the body, our percentage is half that of squirrel monkeys of the genus Saimiri from South America. Turning to energy consumption of the brain relative to that of the whole body, we find that some mormyrid fishes show a figure three times higher than our own (Nilsson, 2000). Clearly, a greatly increased brain size is not a luxury to be acquired lightly. It is something that would only have evolved under strong selective pressure, but the question remains as to what this pressure was. For many years it has been suggested that brain expansion, and the benefits that it brings to humans, has been linked to the problems of making a living in the changed and more open habitats that characterised Africa during the last two million years. Frequently cited is the need to cope with more complex foraging strategies than had been the case when ancestral hominids lived in evergreen forests. I have no doubt that this need would have been one of the factors, but the one that I would like to focus on here was the need that early humans experienced to outlive the ever-present threat to their security posed by carnivores in those savanna habitats. It seems to me that predation could have constituted a large part of the selective pressure needed to promote the process of human brain expansion. It was the Swartkrans cave in particular that drew my attention to these issues, during more than thirty years that I spent on personal investigations there (Brain, 1993a). It became apparent that the very complex cave filling had resulted from repeated cycles of deposition and erosion, with each depositional member providing a glimpse of animal life in the vicinity of the cave at that particular time. The erosion of many metres of dolomite hillside above the cave, together with the upper part of the cave filling itself, had doubtless removed a good deal of the stratigraphic complexity

40

Essential attributes of any technologically competent animal

at Swartkrans. But what is left, and came to light during our excavation, indicated five depositional members, each separated from adjacent members by erosional discontinuities. The earliest of these is designated Member 1 Lower Bank, followed by Member 1 Hanging Remnant, Member 2 and Member 3. All of these have remains of robust australopithecines and are thought to vary in age from 1,8 to 1 million years. Member 4 is a Middle Stone Age deposit with abundant stone artefacts, but little bone, while Member 5 has been dated at about eleven thousand years old, with a fossil assemblage reminiscent of the food remains of leopards. Although the fossil sample from the Hanging Remnant that resulted from the early excavation of Robert Broom and John Robinson between 1948 and 1953 has many superb hominid fossils in it, its use is limited from the perspective of taphonomic interpretation. The extremely hard matrix was first dislodged with dynamite and the resulting blocks were then broken up with hammers. When an interesting-looking fossil was broken through, particularly if it was from a skull, it would be kept for subsequent preparation, but many other fossils were simply discarded onto the dump. Although we subsequently searched through these excavation dumps, recovering some specimens, many others had been carried away in the intervening years by casual visitors to the site. The final Hanging Remnant sample was therefore biased in favour of cranial pieces and, very probably, in favour of primates at the expense of less spectacular mammals. Described in detail elsewhere (Brain, 1981), the assemblage was found to consist of 2 381 fossil bones from 41 identified taxa. Primates – baboons and hominids – made up 47 per cent of the total number of individuals, followed in abundance by ungulates (35 per cent) and carnivores (12 per cent). By contrast, our excavation of the Lower Bank of Member 1 yielded a total of 153 784 pieces of fossil bone that were analysed in detail by Virginia Watson (1993). Of the animals that had contributed to this collection, she found that primates – baboons and hominids – made up 21 per cent, or less than half that in the Hanging Remnant sample. The figure for Member 2, based on a sample of 70 524 fossil pieces, was rather similar at 25 per cent. Concerning these four species of baboon and two of hominid, which naturally showed a wide range of body weights, it was interesting to observe that those species with the greatest body weights also had the highest proportion of juveniles represented among their fossils. Furthermore, cranial remains were disproportionately common in relation to other skeletal parts and we were faced with the ‘mystery of the missing bodies’. An obvious possibility was that we were dealing with the food remains of a carnivore, such as a leopard, with a preferred prey size, and this suspicion was confirmed by the specific damage that some of the bones had suffered (Newman, 1993). One well-known specimen, the calvaria of a hominid child, was found to have

41

From Tools to Symbols

two punctures in its parietal bones and the distance between these was matched by the spacing of the lower canines of a fossil leopard from the same part of the cave. The reconstruction that I suggested (Brain, 1981) was that the child had been killed by a leopard, perhaps by the usual throat-bite method, and that it had then been picked up by its head, as leopards are inclined to do, and dragged off to a feeding place within the dark recesses of the cave. This carrying behaviour, observed in contemporary leopards with monkey or baboon prey, results in the upper canines gripping the face of the prey, while the lower canines penetrate the back of the skull. The detailed taphonomic analysis of the fossil assemblages from Swartkrans Members 1 and 2 suggested to me that hominids and baboons came to shelter within the entrance area of the cave, or its well-wooded catchment area, on cold winter nights and that they were preyed upon there by leopards and sabre-toothed cats. The predators took their victims to the lower parts of the cave and ate them; what scraps survived their attention, and that of scavengers such as hyaenas, whose coprolites in the deposits testify to their presence there, contributed to the fossil assemblage. In broad perspective, my impression is that the life of hominids in environments like that of the Sterkfontein valley 1,5 million years ago would have been a hazardous one, calling for continual vigilance against a wide variety of predatory threats, day and night. Some of these, apart from leopards, would have been lions, false sabre-toothed cats (Dinofelis), true sabre-toothed cats (Megantereon), brown and spotted hyaenas, as well as the now extinct hunting hyaenas (Chasmoporthetes) and hunting dogs as we know them today.

A glimpse of how early humans started to overcome the dangers posed by predators It seems highly likely that humans eventually established their current dominance in the natural world through intelligence and its product, technology. But were the initial steps along this path also mediated in this way? Some of the evidence from the Swartkrans cave seems to confirm this possibility. Our excavation revealed that the Member 3 deposit accumulated in a roofed erosional gully, about twenty metres long and up to five metres wide, running between the west wall of the cave and a vertical bank of older sediments on the east side of the gully. Initially, I was not aware that the lightly calcified sediment in this gully was different from that further to the east, but when pieces of burnt bone started turning up with regularity, suspicions were aroused and a near-vertical unconformable contact became apparent between the contents of this gully and what surrounded it. The excavation proceeded to a depth of 850 cm and produced 59 488 pieces of fossil bone, including nine fossils of robust australopithecines and 270 pieces of fossil bone that showed signs of having been burnt

42

Essential attributes of any technologically competent animal

Figure 1

A selection of burnt bone pieces from the Member 3 deposit at Swartkrans.

Figure 2

An experimental fire of Celtis wood, made at Swartkrans in September 1985 and mentioned in the text. Temperatures attained in various parts of the fire were measured by Tim Brain and Virginia Watson with a digital thermometer attached to a long thermocouple probe, as shown here.

43

From Tools to Symbols

Figure 3

A thin-section of one of the fossil bone pieces from Swartkrans Member 3 that is thought to have been burnt. Darkening in the bone’s lamellar structure is caused by the presence of diffused carbon.

(Fig. 1). Following temperature-controlled experiments (Fig. 2), careful chemical and histological examinations (Fig. 3) confirmed that the bones had been exposed to fire and it was possible to estimate the approximate temperatures to which each had been subjected (Brain & Sillen, 1988; Sillen & Hoering, 1993; Brain, 1993b). One may expect that natural grass fires passing the entrance of a cave should burn any pieces of bone that might be lying on the surface there, and that these could later make their way back into the lower parts of the cave and be preserved. In fact, three pieces of fossilised burnt bone had come to light in the Lower Bank sediment of Swartkrans Member 1 and it could be argued that the burning occurred in this way. But when pieces of burnt bone made their appearance in seventeen excavation squares (1m x 1m) and in up to twenty-three vertical excavation spits (each 10 cm thick) in the newly exposed Member 3 deposit, one was clearly dealing with a different situation. The interpretation that we proposed was that fires had been tended in the entrance area of the Member 3 gully repeatedly during the accumulation of this sedimentary profile, and that pieces of bone heated in these fires had made their way down the talus slope to their final repository. There is no evidence that people at this time had mastered the technique of fire making, but, presumably they had collected burning branches from natural, lightning-induced grass fires, that are very much a feature of

44

Essential attributes of any technologically competent animal

the highveld grassland in early summer, and brought this fire back to their sleeping place. If Member 3 is about one million years old, hundreds of thousands of years would probably have had to pass before the deliberate making of fire became a reality. But whatever the source of the fire, its presence in the cave entrance would have given these early human groups some measure of protection from the ever-present danger of waiting leopards. Apart from the burnt bones, Member 3 has also provided many pieces of fossil bone with unmistakable cut- and chop-marks on them (Fig. 4). Such damage has not been seen on any of the very numerous fossils from Members 1 and 2, suggesting that hominid meat-eating at the cave, probably round a campfire, became a reality then. Presumably, without the protection provided by fire, it would have been too dangerous to bring meat to the cave, for fear of attracting other carnivores. In my opinion, fire-management of this kind must have represented a critical early step in human emancipation from subservience to more powerful carnivores, which eventually led to their domination. The recent discovery of hominin controlled use of fire at Gesher Benot Yaaqov (Goren-Inbar et al., 2004), an Acheulean site from Israel, confirms our argument for a much earlier use of fire than generally thought at the time of our findings at Swartkrans. As a result of further intelligence-mediated technology, humans then went on to become highly effective social hunters in their own right. The

Figure 4

Scanning electron microscope image of a bone flake (SKX 35444) from Swartkrans Member 3, with clear evidence of cut-marks caused by a sharp-edged stone implement. Scale bar = 1 mm.

45

From Tools to Symbols

selective pressures driving this process were presumably the same as those that had allowed the human emergence from a former subservient role. Among the variety of selective pressures that drove the evolution of the large human brain, it can be argued that the pressures of predation, first in surviving its dangers and later in its successful practice, were ever present and powerful in their effects.

Some constraints on any technologically competent animal By any standards, the human animal is a very successful one. Homo sapiens is probably the most dominant animal species that the world has seen thus far, even if it is also the most destructively invasive one. This dominance is the result of its intelligence, frequently translated into tangible technology, so it is interesting to enquire why other animals have not followed, to the same extent, the route of brainexpansion that has characterised this human lineage. It seems to me that there are at least five obvious constraints that would apply to any candidate for the role of a technologically competent animal. Although some of these would only apply if the candidate were a mammal, they are: 1. A critical minimum brain size. As was emphasised by Tobias (1971), the human brain is characterised by a very large number of ‘extra neurons’, over and above those needed to handle the basic functions of the body. For humanstyle intelligence, it seems that a volume of at least 500 cc of ‘extra’ neural tissue-hardware is required, so a constraint is imposed on the minimum size of a mammal that could accommodate this. The constraint does not seem to apply as rigidly in the case of birds, as will be mentioned shortly. 2. Suitable appendages for manipulating objects related to the technology in question. 3. A social organisation that would promote the collective effort required for the successful fulfilment of any major technological undertaking. 4. A language-like communication system that would allow the exchange of concepts between individuals. 5. Appropriate birth-canal adjustments to allow the passage of the offspring’s large head where live-bearing mammals are concerned. Various animals fulfil these requirements to some extent and could go on to become completely technologically competent in the course of future evolution, given the appropriate selective pressures. Before considering several such contemporary organisms, it is important to mention that many animals, both invertebrate and vertebrate, display remarkable technological ability in, for instance, the construction and stocking of their nests; such ability is innate and species-typical. Take the example, for instance, of African mud-wasps of the family Specidae that build elaborate nests of

46

Essential attributes of any technologically competent animal

mud. The shape and structure of each nest is typical of the species involved, and such nests are then stocked with the bodies of jumping spiders, each of which the wasp has immobilised with its sting, without killing it. The female wasp lays her eggs on one of the spiders and when these hatch, the larvae feed on the bodies of spiders that, through the action of the injected toxin, are unable to resist the process. This is remarkable technology to be sure, but it has evolved in a way different from imagination-driven technology, such as we see in humans. Other examples are the elaborate nests of weaver birds, the form of which is again typical of the species involved, and represents a different realm of technological competence. But, apart from such animals, there are many showing various degrees of intelligence-driven technological competence that could evolve further in the future, given appropriate circumstances and selective pressures. Often quoted are the living anthropoid apes, such as chimpanzees, gorillas and orang-utans. Much has been written about the skills of these primates (e.g. Rumbaugh, 1995; Whiten et al., 1999) and I do not plan to discuss them further here, but rather to concentrate on other, less familiar animals that also show the potential for technological competence. Let us start with an invertebrate, a molluscan cephalopod – an octopus, cuttlefish or squid. All of these have eight or ten highly mobile appendages, each equipped with numerous suction-cups. Anyone who has watched an octopus build its shelter in a rock pool from stones that it has selected, carried and positioned with care and precision will have no doubt as to the manipulatory skill of these animals. The sense organs of a cephalopod – including the beautiful eyes of cuttlefish with their W-shaped irises, which allow for vision forward and backward at the same time – feed into nervous systems that are very different from those of mammals. Though different, their series of ganglia and legendary giant axons are highly effective and allow these animals to communicate in various ways, such as through intricate patterns and colours of their skins (Holloway, 2000). Some of the squids are very large and it is quite conceivable that the neural hardware needed for a fully technologically competent squid could evolve in some of these animals, if selective pressures developed to drive the process. It would be fun to see such creatures ‘conquering the land’ in their water-filled ‘squidmobiles’. Among mammal candidates, dolphins and whales immediately come to mind. Dolphins particularly deserve attention, as they are very encephalised, social and make use of an intricate communication system (Herman et al., 1993, Leatherwood & Reeves, 1990). For a large brain, body size is not a problem, and their rudimentary pelvis would easily allow the birth of large-headed offspring. What I see as a serious disadvantage is that they do not have manipulative appendages, the forelimbs having become so highly specialised as swimming flippers that they may be beyond evolutionary recall. But just how evolution can ‘tinker’ with an already specialised forelimb is exemplified

47

From Tools to Symbols

by the well-documented case of the giant panda’s ‘pseudo-thumb’ (Gould, 1978; Endo et al., 1999), which came about after a dietary change from a bear-like omnivorous carnivore to a specialised feeder on bamboo. The necessity to manipulate bamboo shoots with hands already specialised for walking, saw the modification of both the radial sesamoid and accessory wrist bone to form ‘a double pincer-like apparatus in the medial and lateral sides of the hand, respectively, enabling the panda to manipulate objects with great dexterity’ (Endo et al., 1999). Among carnivores, this kind of objectmanipulation is not an isolated case – Californian sea-otters have long been known to pick up stones and then, while floating on their backs in the water, to rest them on their bellies. These stones are used as anvils, against which they break open mollusc shells. Given considerable increase in brain and body size, a variety of carnivores could conceivably become technologically competent in the human sense. I sometimes regret the fact that we humans did not in fact emerge from a good carnivore, rather than a primate, stock. I think that, as a species, our natures might then have been less devious. Finally, one should not overlook birds as candidates for technologically competent animals, as a number of avian species have higher brain-to-body weight ratios than is the case in humans. Irene Pepperberg (1999) has made a particularly detailed behavioural study of African Grey Parrots and has drawn attention to their remarkable ability to imitate human speech. But, in addition to this, they are also able to make up sentences of their own and to manipulate objects with their feet. In my opinion, they could well take these capabilities further, if their survival demanded it. As could crows, which have long been known to be exceptionally intelligent throughout their extensive geographic range, in a wide variety of circumstances. But those from the New Caledonian Islands in the Pacific Ocean north of New Zealand (Corvus moneduloides) are particularly remarkable, as demonstrated by the observations of Gavin Hunt, a biologist at the University of Auckland, who has been observing them for the last ten years. He found that the crows habitually use tools for the extraction of insect larvae from holes and cracks in the forest trees. Some of these are straight probes that the crows fashion and then carry around for days at a time. But others are hooks that are more efficient for the extraction of reluctant grubs from their holes. These are made by the crows in a variety of ways from a range of raw materials; one type, known as the ‘crochet hook’ is made by detaching a side twig from a larger one, leaving enough of the larger twig to shape into a hook (Pain, 2002). Another tool is known as a ‘stick with a hook’, made from the leaf of a forest vine that has a sturdy midrib, with paired leaflets, each with a thorn at its base. The crows select a piece of midrib, removing the leaflets and all but one of the thorns at the tip, so that they end up with a highly effective hook. But the tool that has created a great deal of interest in biological circles

48

Essential attributes of any technologically competent animal

is ‘the pandanus stepped-cut tool’, which is a ‘tapered probe, pointed at the tip, broad at the base and bearing a row of tiny hooks along one edge. With its sharp point and backward-facing barbs, it’s an ideal implement for winkling prey from tight spots’ (Pain, 2002). It is made from either the left or right edge of the long narrow leaves of pandanus trees. In his long-term New Caledonian island fieldwork, Gavin Hunt and his colleagues have collected a great deal of information on these tools, derived very largely from the stepped tool counterparts left by the crows on the leaves of the pandanus trees involved. He wrote: The crow’s use of left or right leaf edges depends in part on the direction in which the leaves spiral. Clockwise-spiralling leaves provide easier access to left edges, and anticlockwise-spiralling leaves provide easier access to right edges. This access effect was overridden, however, by an island-wide preference for manufacturing tools from left edges ... Although right-handedness is thought to be uniquely human, we show here that crows from different localities display a widespread laterality in making their tools, indicating that this behaviour is unlikely to be attributable to local social traditions or ecological factors. To our knowledge, this is the first demonstration of species level laterality in manipulatory skills outside humans. ... It has been proposed that righthandedness in humans may be a consequence of the evolution of language, which is also predominantly left-hemispheric. Our results favour the more general possibility that species level lateralization is an adaptation for the efficient neural programming of complex sequential processing, of which language and right-handedness in humans and stepped-tool manufacture in crows are examples (Hunt et al., 2001).

But the brain of a New Caledonian crow is small, devoid of the 500 cc of ‘extra neurons’ that are demanded by our central nervous systems to provide human-style intelligence. Could it be that birds, including the clever crows, have managed to miniaturise the neural structure of their central nervous systems, in a way similar to that whereby the size of human-made computers has been reduced during the last fifty years? If so, ‘bird brains’ might yet be the key to the next wave of technologically competent animals that could dominate the earth, once the human experiment has faded into the fossil record.

Acknowledgements I would like to thank Lucinda Backwell and Francesco d’Errico for inviting me to participate in their International Round Table From Tools to Symbols, From Early Hominids to Modern Humans, at the University of the Witwatersrand in March 2003, as well as their suggestion that I contribute a paper to the Proceedings.

49

From Tools to Symbols

My previous contact with these two people was an interesting one. The longterm excavations that I conducted at Swartkrans between 1965 and 1986 produced a large number of bone tools, the characteristic wear on which suggested that they had been used for digging. This conclusion was confirmed by digging experiments and the detailed scanning electron microscope studies of wear patterns undertaken by Pat Shipman and myself at the Johns Hopkins University in the United States during 1987. When Lucinda started work on her dissertation relating to bone tools from hominid sites, with Francesco as her supervisor, they re-examined the wear patterns on the Swartkrans tools, concluding that, although they had, indeed, been used for digging, this had been done specifically in termite mounds for the extraction of the termites themselves as food. This case strikes me as an example of how good science should proceed. Initially, a hypothesis is set up on the basis of available evidence; when other researchers re-investigate this evidence, bringing new techniques and insights to bear on the questions, valuable new information emerges, in this case, concerning the behaviour of early hominids. I am grateful to Jane Dugard for interesting discussions on species-typical technologies in a variety of animals.

References Aiello, L.C. & Wheeler, P. (1995). The expensive tissue hypothesis. The brain and the digestive system in human and primate evolution. Current Anthropology 36(2), 199–221. Brain, C.K. (1981). The Hunters or the Hunted? An Introduction to African Cave Taphonomy. Chicago: University of Chicago Press. Brain, C.K., Ed. (1993a). Swartkrans: A Cave’s Chronicle of Early Man. Transvaal Museum Monograph no. 8. Pretoria: Transvaal Museum. Brain, C.K. (1993b). The occurrence of burnt bones at Swartkrans and their implications for the control of fire by early hominids. In (C.K. Brain, Ed.) Swartkrans: A Cave’s Chronicle of Early Man, pp. 229–242. Transvaal Museum Monograph No. 8. Pretoria: Transvaal Museum. Brain, C.K. & Sillen, A. (1993). Evidence from the Swartkrans cave for the earliest use of fire. Nature 336(6198), 464–466. Endo, H., Yamagiwa, D., Hayashi, Y., Koie, H., Yoshiki, Y. & Kimura, J. (1999). Role of the giant panda’s pseudo thumb. Nature 397, 309–310. Gould, S.J. (1978). The panda’s peculiar thumb. Natural History 87(9), 20–30. Goren-Inbar, N., Alperson, N., Kislev, M.E., Simchoni, O., Melamed, Y., Ben-Nun, A. & Werker, E. (2004). Evidence of hominin control of fire at Gesher Benot Yaaqov, Israel. Science 304, 725–727. Herman, L., Pack, A.A. & Morrel-Samuels, P. (1993). Representational and conceptual skills of dolphins. In (H.L. Roitblat, L.M. Herman & P.E. Nachtigall, Eds) Language and

50

Essential attributes of any technologically competent animal Communication: Comparative Perspectives, pp. 403–442. Hillsdale, NJ: Lawrence Erlbaum Associates. Holloway, M. (2000). Cuttlefish say it with their skin. Natural History 4(00), 70–76. Hunt, G.R., Corballis, M.C. & Gray, R.D. (2001). Laterality in tool manufacture by crows. Nature 414, 707. Leatherwood, S. & Reeves, R.R. (1990). The Bottlenose Dolphin. San Diego: Academic Press. Newman, R.A. (1993). The incidence of damage marks on Swartkrans fossil bones from the 1979–1986 excavations. In (C.K. Brain, Ed.) Swartkrans: A Cave’s Chronicle of Early Man, pp.217–228. Transvaal Museum Monograph No. 8. Pretoria: Transvaal Museum. Nilsson, G.E. (2000). The cost of a brain. Natural History 108(10), 66–73. Pain, S. (2002). Look, no hands! If making tools is so clever, how come New Caledonian crows can do it? New Scientist (17.8.02), 44–47. Pepperberg, I.M. (1999). The Alex Studies. Cognitive and Communicative Abilities of Grey Parrots. Cambridge, Mass.: Harvard University Press. Sillen, A. & Hoering, T. (1993). Chemical characterisation of burnt bones from Swartkrans. In (C.K. Brain, Ed.) Swartkrans: A Cave’s Chronicle of Early Man, pp. 243–250. Transvaal Museum Monograph No. 8. Pretoria: Transvaal Museum. Rumbaugh, D. (1995). Primate language and cognition: Common ground. Social Research 62, 711–730. Tobias, P.V. (1971). The Brain in Hominid Evolution. New York and London: Columbia University Press. Watson, V. (1993). Composition of the Swartkrans bone accumulations in terms of skeletal parts and animals represented. In (C.K. Brain, Ed.) Swartkrans: A Cave’s Chronicle of Early Man, pp. 35–74. Transvaal Museum Monograph No. 8. Pretoria: Transvaal Museum. Whiten, A., Goodall, J., McGrew, W.C., Nishida, T., Reynolds, V., Sugiyama, Y., Tutin, C.E.G., Wrangham, R.W. & Boesch, C. (1999). Cultures in chimpanzees. Nature 399, 682–685.

51

Significant tools and signifying monkeys: the question of body techniques and elementary actions on matter among apes and early hominids Frédéric Joulian Responsable du Programme de Recherches Interdisciplinaires ‘Evolution, Natures et Cultures’, SHADYC, Ecole des Hautes Etudes en Sciences Sociales, 2 rue de la Vieille Charité, 13002 Marseille, France

Abstract On the basis of the etho-archaeological research conducted within a team project coordinated by the author and entitled ‘Men and Primates in Perspective’, the different degrees of semiotisation at work in tool fabrication and use among chimpanzees (Pan troglodytes), and the implications that these might have for understanding behaviour and cognition of early hominids are discussed. Primatologists demonstrated several years ago not only that traditions or cultures exist among non-humans, but also that a significant number of phenomena believed to be special attributes of the genus Homo (tool making, carnivory, self-consciousness, meta-representation …) must henceforth be tackled in a more extensive taxonomic context: that of hominoids. We argue that the same goes for the existence of representations, defined as forms independent of content or content independent of functions, which authors tend now to examine within a broader specific and temporal framework than imagined thus far. In this contribution the question of human cultural modernity is broached on the basis of techniques, representation, and semiosis of primate societies, whether human or not, illustrating this theme with the ethological and archaeological data published in the literature and findings made in West Africa during the past four years by the author, on the question of body techniques and on elementary actions on matter, as research in these domains allows us to avoid the problems inherent in the tool and the language (or symbol) that have masked all debates for more than a century.

52

Significant tools and signifying monkeys

Resumé Sur la base des travaux étho-archéologiques que je dirige dans le cadre de l’équipe ‘Hommes et Primates en Perspective’, je montrerai les différents degrés de sémiotisation à l’oeuvre dans la fabrication et l’utilisation des outils chez les chimpanzés (Pan troglodytes) et les implications qu’ils peuvent avoir pour les Hominidés anciens. En effet, depuis quelques années nous avons pu démontrer l’existence de traditions ou de cultures chez des non-humains. De même, un grand nombre de phénomènes que l’on pensait être le propre du genre Homo (fabrication d’outils, carnivorie, conscience réfléchie, méta-représentations …) doivent désormais s’appréhender dans un espace d’espèces plus étendu: celui des Hominoïdes. Il en va de même de l’existence de représentations – autrement dit de formes indépendantes des contenus ou de contenus indépendants des fonctions – qui se conçoivent également dans un cadre spécifique et temporel plus large qu’on ne l’imagine jusqu’à présent. Dans cette conférence et dans cet article j’ai donc abordé la question d’une modernité culturelle humaine sur la base des techniques, de la représentation et de la sémiotisation des sociétés de primates, humains ou non, illustrant ce thème des découvertes éthologiques et archéologiques publiées ces dernières années et des découvertes que nous avons faites ces quatre dernières années en Afrique de l’Ouest. J’ai circonscrit mon propos à la question des techniques du corps et des actions élémentaires sur la matière qui constituent un moyen d’éviter le problème massif de l’outil et du langage (ou du symbole) comme propre de l’homme et de reposer de façon plus féconde cette question d’émergence du sens.

Introduction A title such as ‘From Tools to Symbols’ provides us with a choice. We can either consider it simply a metaphor, or we can take it seriously. In the latter case, it presents a real scientific challenge for both the archaeologist and the anthropologist. We have opted for the second course because we recognise one of the founding fixations in this issue: that of the change from the simple to the complex, from the evolution of primitive beings (primate or pre-human) endowed with rudimentary techniques, into evolved beings. How did primates move from a functional world, where action and accomplishment are equivalent, to a world where action and meaning have become separate, i.e. where meaning (in fact) saturates every moment of human life? Did semiotic evolution constitute a rupture with primates? Once or several times during history? Continuously? Is it linked to superior forms of cognition? Or to the appearance of tools and language? The observation of evolutionary progress may be evident on a scale of hundreds of thousands, or even millions of years, but as soon as we enter into detail about material facts and their social, cultural, or cognitive significance for today’s human societies, or for primate societies, the antecedence of tools over symbols or, more generally, of techniques over meaning systems becomes less clear. For an anthropologist working at the crossroads between the ethology of apes, prehistory, and social anthropology, and who wishes to fully and empirically integrate the methods and results of all three disciplines, the road is narrow. In this article we

53

From Tools to Symbols

aim to show how two concepts, that of body techniques, introduced by Marcel Mauss at the beginning of the twentieth century, and that of elementary actions on matter, developed by Leroi-Gourhan in the 1940s, allow us to rethink the frustrating question of tools as distinguishing emblems of humanity and facilitate the comparison between primates, pre-humans and modern humans. Before tackling these great questions we need to answer very simple (and concrete) ones such as: Why and how to compare chimpanzee and pre-Acheulean technical systems? This ‘why’ and this ‘how’ constitute the object of long-term research conducted since the middle of the 1980s (Joulian, 1986, 1993, 1994b; Beyriès & Joulian, 1990). Very briefly, let us recall that archaeology is a science constrained by preservation and taphonomic imperatives. It is also a social science mainly preoccupied with answering broad questions about social meaning and functioning. In that respect, prehistoric archaeology can interface with palaeoanthropology, ethology, art, linguistics and other disciplines. Depending on the disciplines, the nature of data can be very different and difficult to compare: a dynamic behaviour observed for ten years in a chimpanzee community and a stone tool knapped in ten minutes and abandoned by an australopithecine are very different by nature. In some cases Plio-Pleistocene and chimpanzee tools are identical (Joulian, 1993, 1996a), but the possible interpretation concerns functional aspects (nut-cracking) and nothing beyond (no sex differentiation in the use of tools, traditional variations, and so on). To understand at a higher level (adaptive, social, cultural) it is necessary to build larger theoretical constructs and raise the recurrent question of the meaning of assemblages of tools or behaviours for the scientist (whether ethologist or prehistorian), for the australopithecine and for the ape. To bring together primates and prehistoric humans without excluding their social and cultural dimensions, we proceeded by steps, first finding the common ground between ethologists and prehistorians, technologies and gestures (Joulian, 1996b, 1998). We then reintroduced the social and cultural dimensions of behaviours and techniques by calling on appropriate concepts forged by social anthropology. There are two different paths which allow such interdisciplinary interaction: first, the application of the concept of body techniques to non-human primates, which allows us to elucidate the necessary links between tradition, social representation, and techniques; second, the idea that it is (some) elementary actions on matter, and not tools, that are specific to prehistoric man and innovative in comparison to primates.

Do animals have ‘body techniques’? The notion of body technique as a cultural usage of one’s own body as an instrument, and which describes the way to walk, to swim, to dance, to give birth – all cultural uses of the body by people living in societies – clearly belongs to the field of anthropology,

54

Significant tools and signifying monkeys

and it may be surprising to find it transposed to the study of non-human primates. Nevertheless, in view of certain types of simian behaviour, the topic deserves to be explored. The use of findings in primatology to interpret early human behaviour also leads us to reflect on how these fields of research and thought can be brought together. It is in this perspective that interaction among the disciplines of anthropology, ethology, and prehistory is envisaged here. We know that Mauss defined a technique as ‘un acte traditionnel et efficace’ (‘a traditional and efficient act’) (Mauss, 1936), more commonly translated in English as ‘traditional effective actions’ (Schlanger, 1998). The concept of a body technique presupposes three conditions: tradition, efficiency, and embodiment (the inscription in the body and its movements of practices specific to a group, as well as their stabilisation or perpetuation over long periods). Concerning animals, at the time when Mauss proposed this definition, only the second condition – that of efficiency – had any meaning for most authors. The other two, tradition and incorporation, are nowadays observed among different species of primates. The existence of tradition and

Figure 1

An example of a body technique: a gorilla using its incisor teeth as scrapers for removing the bark of a branch in a specific task. Parc Zoologique de Beauval, France, 2003 (photo F. Joulian).

55

From Tools to Symbols

the transmission of acquired behaviour have now been confirmed (Whiten & Ham, 1992; Byrne, 1995; McGrew, in press). According to Mauss (1936), ‘There is no technique nor transmission if there is no tradition. It is above all through tradition that man distinguishes himself from animals, through the transmission of his techniques, and most probably through their oral transmission.’ In addition, for Mauss, techniques are embodied and internalised, and that is what distinguishes them from religious, symbolic or moral acts. It is also in this sense that body techniques are fundamental: the body is ‘the first and most natural technical object’ as well as a ‘technical means’. Primatology during the 1930s offered Mauss limited access to animal behaviour. Nevertheless, he mentions the primatologists Kohler and Yerkes, who became involved in a debate (Mauss, 1929a) with French psychologists Guillaume and Meyerson. All these scientists had performed experiments on captive animals. It goes without saying that the spontaneous use of tools by primates living in their natural setting was not only unheard of at the time, but unthinkable. Field primatology would not develop to its full extent, with its numerous implications for prehistory, anthropology and psychology, until the early 1960s. The discovery of such behaviour took place in three stages. The first corresponds to the blossoming of Japanese primatology in the 1950s and 1960s. The first synthesis of ‘cultural’ or ‘precultural’ behaviour among macaques was written in the mid-1960s by Kawai (1965). The second stage can be identified as the Oregon symposium on ‘Precultural Primate Behaviour’ (Menzel, 1973). This meeting was the first to clearly pose the question of culture among primates by tying together long-term studies of Japanese macaques and those of chimpanzees in the wild (Itani & Nishimura, 1973; Goodall, 1973). The third stage of the discovery is ongoing, and began with the work of ethologist William McGrew (McGrew & Tutin, 1978; McGrew, 1985, 1992). McGrew based his analyses on the comparison of behavioural techniques among different groups of chimpanzees and launched, in a provocative yet justified manner, the ‘cultural’ debate, which has since then swung dynamically between the fields of anthropology and primatology. In addition, he applied the concepts of anthropology to primates. Various studies in ethology followed the same lines, among them our own work comparing traditional behaviour of chimpanzees and of the first hominids (Joulian, 1994a, 1996a). The approach to variation adopted by McGrew is that of a naturalist. This approach is shared by numerous ethologists (Nishida, 1987; Sugiyama, 1993; Boesch, 1991; Boesch & Boesch-Achermann, 1994; Huffman & Wrangham, 1994), who study behavioural differences among groups in a systematic manner. They base their approach on Galef’s (1976) conceptions, stating that if a behaviour is not determined by genetic, environmental or physical factors, it must be cultural. Although this

56

Significant tools and signifying monkeys

type of approach is effective from a practical point of view and brings to light real choices in simian activities, it was not at first endorsed by experimental psychologists. Some psychologists were very critical of the use by primatologists of the notion of transmission, imitation, and innovation (Tomasello, 1990, 1994; Premack & Premack, 1994), and denied the existence of animal traditions and cultures. Nevertheless, several observations of chimpanzee behaviour in a natural setting – especially ‘leaf clipping’ and ‘leaf grooming’ – appear to contradict their objections (Boesch & Boesch-Achermann, 1994). Ironically, after that first battle, psychologists and ethologists came together to publish on chimpanzee and human culture, focusing mainly on transmission and cumulative aspects of culture (Boesch & Tomasello, 1998). In my view, however, much of the debate between psychologists and ethologists is based on a false question, more the consequence of the confrontation of opposing research paradigms and the parochial use of certain concepts (in this case that of imitation) than a truly scientific discussion. In several articles, Ingold (1988, in press) has shown that the Cartesian dualism between mind and body systematically underlies this debate, masking the complexity of the situation and leading us away from the true locus of behavioural variation. Yet this variation in behaviour is exactly what interests us when we study body techniques. Let us not forget, furthermore, that the context of the physical, natural and social environment ‘takes part in’ or ‘constitutes’ at least as much as it ‘determines’ the ways in which the body or an instrument are used. The purpose of these remarks is to point out that other less rigid ideas would perhaps allow us to escape the opposing frameworks of naturalism and experimentalism. Let us return now to the notion of animal traditions in relation to Mauss’s definition of a technique as a ‘traditional and efficient act’. We should ask ourselves whether the use of the term ‘traditional’ permits us to speak of body techniques at all. This is a most delicate question, situated at the very heart of the most advanced studies in modern ethology and animal psychology, for it implies what some have called ‘internalisation’. Contrary to humans, who mark group identity (and also material production) by acknowledging their traditions, animals do not show the capacity to attribute significance to their own traditions (Thierry, 1995, 2004). This point about the internalisation of tradition is crucial and we have raised it in other, more anthropological ways by considering the possibility of collective representations of knowledge among chimpanzees (Joulian, 1994b, 1995a). This line of questioning is an important key to understanding the nature and the scope of animal traditions, as well as to distinguishing more clearly the respective characteristics of animal and human tradition. If we follow Thierry’s hypotheses, we should, for example, not qualify as a tradition the use of the same sleeping cliffs by baboons over several generations. It seems that such a definition, in many cases even for humans, is too narrow.

57

From Tools to Symbols

When we talk about techniques, what exactly are we describing? On the one hand, a set of activities in a natural setting, and on the other, a set of more general cognitive abilities usually studied in experimental conditions. Though intergroup comparisons reveal behavioural or artifactual variations (sometimes traditions), knowledge (‘know-how’) and skills are rarely described in themselves. In other words, all that is intermediate between mind and product, all that constitutes manners, modes, and ways of acting, still eludes us, due perhaps to a lack of adequate theoretical and methodological frameworks. Still, some ethological studies may be quoted, in particular the works of Marchant and McGrew (Marchant et al., 1995; Marchant & McGrew, 1996), who have analysed gestures and laterality in chimpanzees and in some non-Western human populations. I should mention that these studies, which concern lateral activities not involving tools, conflict with results from similar studies by psychologists (Marchant et al., 1995, 256). Sugiyama and his colleagues have also studied laterality and learning in nut-cracking in a natural setting among Bossou chimpanzees (Sugiyama et al., 1993; Matsuzawa, 1994; Inoue et al., 1995). Matsuzawa’s observations reveal variations in nut-cracking gestures, in contrast with experimental studies. Unfortunately, their observations are focused more on the learning and transmission of daily motor activities than on the analysis of the differences in skills unique to each group. It should be stressed that it is only such differences that would allow us to envisage the existence of traditions (and thus of body techniques). While, over the last twenty years, ethology has largely been preoccupied with social and ecological aspects of animal societies (Smuts et al., 1987), the social dimensions of traditional behaviour have scarcely been explored. A large part of primatology is monopolised by socio-biological and socio-ecological paradigms in which the question of techniques remains secondary. In such an intellectual climate, the notion of ‘body technique’ offers a means to understand the complex relation between the body as an instrument and the body’s use of instruments. This notion allows us to circumvent the enormous problem of tools. In fact, it seems that the insistence on discriminating between animal and human tools has masked a number of important questions. The classical distinction between the tekhnè, considered as a divider between the human and the animal worlds, and an organicistic or zoological conception of technique (in which tools extend biological organs) can be overcome by concepts as ‘techniques without tools’, ‘body techniques’, and ‘body instrumentalisation’. This is necessary because such distinctions that use rupture and continuity conceptions are common in prehistory and in palaeoanthropology, but they are no longer compatible with ethological data. The initial question, ‘do animals have body techniques?’, draws, at one level, an affirmative response, in that animals do indeed develop practices unique to their own

58

Significant tools and signifying monkeys

group, and that these practices distinguish them from neighbouring groups. Mentioned earlier was the existence of animal traditions, such as those observed in tool use, but also in behaviour not involving tools and not strictly adaptive – handclasp grooming among chimpanzees, for example (McGrew & Tutin, 1978). As already suggested, if these observations are few in number, it is perhaps due to study topics and research conditions, which are rarely guided by such questions. For instance, despite several decades of continuous observation, it is still impossible to decide whether chimpanzees at the Gombe Stream site move around and mate in the same way as chimpanzees in Mahale. Though inter-group comparisons of posture and locomotion are beginning to multiply (Hunt, 1992; Doran, 1996), research questions and observation conditions in natural settings still allow limited access to the detailed information needed to address the problem of body techniques in the strong sense of the term. In the Maussian sense, body techniques ‘are not natural’. With regard to walking and dancing, Mauss (1936) said that ‘there is no such thing as a “natural manner” in adults’. This goes back to the fact that ‘“techniques are naturally human”, that is, “arbitrary”’ (Schlanger, 1991, citing Mauss, 1927, 1941), or that ‘the social domain is the domain of modality’ (Mauss, 1929b, 470). According to Mauss, these arbitrary, ‘traditional’ choices, similar to those involved in digging for termites or in cracking nuts among chimpanzees (McGrew, 1992; Boesch et al., 1994), must, in order to respond to all the implications of the concept, be invested with social value (along the lines of ‘It is impolite to eat with your elbows on the table!’) or even become a norm. Here we arrive at the outer limit of the extension of the concept, and the social value or meaning of a practice in a group of primates remains, at least for the time being, a problem beyond our reach. This having been said, and before examining the social meanings of techniques, another problem presents itself: that of the nature of technical knowledge in apes. A body technique presupposes specific knowledge concerning the technique. Or, to pose the problem on a more general level, it presupposes the existence of a ‘technical system’ as defined by Gille (1978) and Lemonnier (1992). My work raised the issue for chimpanzees (Joulian, 1994b) and showed that they do have techniques and technical systems, but that their skills as ‘technicians’ or ‘technologists’ need to be discussed. This is where the real question lies: where does the divergence between animal and human ‘technicity’ occur? This focus on the core questions of techniques and culture, explored in similar terms by anthropologist Guille-Escuret (1994), leads us logically to respond to the critique of internalisation advanced by Thierry, or, to use more anthropological terminology, to attempt a demonstration of the existence of social or collective representations of techniques among chimpanzees. We use here the expression ‘social representation of

59

From Tools to Symbols

techniques’ as defined by Lemonnier (1991, 1993). In fact, one often demands more of chimpanzees than of modern or prehistoric humans to prove the existence of collective representations of techniques that concern the very elements of variation, the elements of tradition, in which case tradition would certainly be represented consciously. The hypothesis of a social representation of techniques in chimpanzees is extremely strong and implies, in my view, the existence of specific knowledge of the material world and of other beings. It must be envisaged in comparison with other still-debated abilities of chimpanzees to attribute intentions (cf. Povinelli, 1994; Suddendorf & Whiten, 2001 for a discussion). Though field data for such a demonstration are still lacking, we have today several clues that lead in that direction. The intentional teaching of female chimpanzees to their offspring has been observed in Tai, Ivory Coast, by Boesch (1991); we have collected accounts of nut-cracking traditions (Joulian, 1995b); put together, these studies point to the need to re-examine the ability to represent tradition consciously, until now considered limited to modern humans. The observations carried out at the Tai site are exceptional, but some researchers have contested that they do not fulfil criteria defining transmission by imitation, in the strictest sense (McGrew, in press). It still remains that the female chimpanzees studied showed their offspring actively and through examples – and thus intentionally – how to proceed optimally to crack open Panda oleosa nuts. Three different methods have been distinguished which accelerate the acquisition of nut-cracking by young chimps observing their mothers at Tai. 1.

Stimulation. A mother chimp may stimulate experiments by her infant by leaving her hammer on the anvil while she collects nuts, something a childless female practically never does, lest the tool be stolen. (…)

2.

Facilitation. The mother may give her own hammer to her young, or she may let him take a few of the nuts she has collected while she continues to crack nuts (…)

3.

Active teaching. (…) In one case, the mother seized the tool and proceeded to give a veritable slow-motion demonstration of exactly how to hold the hammer and how to strike the nut. In the other case, the mother repositioned the nut correctly on the anvil, thus causing the three kernels to come out of the shell without breaking it.

(Boesch & Boesch-Achermann, 1994: 18–19, my translation).

Note, however, that observed cases of active teaching, as a formalised act related to practice, in the transmission of know-how among chimpanzees – and in traditional human societies – are rare but always related to different dimensions of beings in society (Gosselain, in press). The observations above attest to the existence of reflection, in the strict sense of the word, and of knowledge of the techniques used. The mother chimpanzees show not only ‘what to do’, but also ‘how to do it’. We find ourselves squarely at the level of ‘know-how’ (savoir-faire), but there is more. These observations

60

Significant tools and signifying monkeys

attest to the existence of an explicit representation of the tasks to be carried out and the way to carry them out, a representation integrated with a specific knowledge, with a broader group memory. The types of behaviour the mothers demonstrate to their young involve showing ‘the right tools’, ‘the right nuts’, ‘the right motions’, all essential elements in successfully cracking open the hard Panda oleosa nuts. If we compare this type of knowledge to the technical variants (or traditions) described for nut-cracking (Table 1), which include the type of tool, the material it is made of, and hand gestures (for example, with or without hammers, stone or wood hammer, held with one hand or two (Joulian, 1995b, 1996a)), we see that the elements ‘taught’ by the females to their offspring are elements that we have observed to vary among chimpanzee communities. This leads us to make the hypothesis of a social dimension of the existence of specific knowledge, and, second, confirm the social dimension of the representation of these activities. What is the significance of such technical traditions for chimpanzees? Does traditional variation have for them a meaning other than that defined by the task (obtaining an edible kernel)? The solution to these questions would require, without a doubt, many more years of intensive research. The main idea to retain here is Table 1 Tradition and transmission in Panda oleosa nut-cracking activity TRADITION

TRANSMISSION

(from Joulian 1993, 1995b, 1996)

(elements objectified by the female during transmission of Panda oleosa nut-cracking activity, from Boesch, 1991)

• type of nut eaten

• nut (type, size, etc.)

• type of nut cracked • technique used (with or without a hammer) • type of tool (mobile anvil, ...) • raw material • type of hammer (in case of Panda oleosa)

• tools (hardness, weight, size)

Panda oleosa nut-cracking: Hammers: variation in dimensions, pit numbers, utilised sides, type of anvil. Gestures: variation in holds (one hand, two hands)

• holding the hammer • accomplishing the gesture • repositioning the nuts

61

From Tools to Symbols

that the established link between ‘tradition’ and ‘transmission’ is, at present, the only one which allows the demonstration that the gestures and the representations associated with them are not only individual actions, sub-products of the large motor and intellectual skills of chimpanzees, but indeed the sign of the actualisation of ‘representations’ of instrumentalised techniques. Variation in the use of instruments, along with the fact that the animals distance themselves from their knowledge, offers us such a demonstration. Unfortunately, similar observations concerning incorporated techniques, or body techniques, have not yet been made. The problem of instrumentalisation is secondary to that of the existence of collective representations of traditional knowledge. If such collective representations do indeed exist, they make ‘techniques’ in general, and ‘body techniques’ in particular, all the more significant, and render vain the search for the precedence of one of these forms with respect to the other (i.e. body techniques vs tool-use techniques). In other words, the hypothesis that ‘primitive’ techniques – whether non-instrumented or involving instruments made of plant matter – came prior to the appearance of stone tools that

Figure 2

Male chimpanzee cracking Panda oleosa nut in a ballistic movement, in Mont Nieniokoué, Côte d’Ivoire 2000 (photo F. Joulian).

62

Significant tools and signifying monkeys

mark the rise of human technology, loses ground. The genealogy of the technical, cultural and symbolic capacities that we are trying to establish should not rest on a chronology of tools, but on a lineage of complex technical systems in which the ability of pre-human and non-human primates and of humans to interact with their environment or their co-specifics also enter.

Elementary actions on matter Exploring the practices of primates first requires an inventory of elementary actions on matter, which is the smallest units of action, the most elementary actions possible carried out towards a specific goal (to displace the body or a body part, to obtain an object, transform an object, etc.). This definition is a contraction of the extensive reflections of André LeroiGourhan in L’Homme et la Matière (1971) and La Mémoire et les Rythmes (1965). A rapid examination of the literature in primatology on this theme indicates that putting together such an inventory of elementary actions on matter is an extremely complicated task, for zoologists never think in these terms. Specialists in anatomy and biomechanics have, of course, described movement (for example, Napier & Walker, 1967; Tuttle & Watts, 1985), but their perspectives are associated with gross motor functions (locomotion, grasping, etc.). Analysed in the framework of evolutionary morphology, or, at best, ecology, the study of these gross motor functions has not yet been related to the analysis of intra- or inter-specific variation in action. Fortunately, some articles offer refined definitions and classifications of grasping (Marzke & Wullstein, 1996) and of posture and locomotion (Hunt et al., 1996). In the opposite corner of the field of primatology, the actions of primates have been explored by psychologists. But here again, even though these studies are closer to our concerns, the descriptions are inadequate because they are tied to approaches based on development, on ontogenesis in the broad sense, or, more rarely, on the cognitive abilities involved with skills and knowledge (‘know-how’, savoir-faire). The action is not analysed in and of itself, but serves, first and foremost, to gain access to the cognitive capabilities of different species (Vauclair, 1982; Torigoe, 1987; Takeshita & Walraven, 1996). It should also be remembered that since these studies were carried out on captive chimpanzees, they do not allow us to examine behaviour as a function of the context and surroundings (ecological, social, situational) in which traditions, if any, could have meaning. In order to fulfil the broader objectives of the analysis of body techniques, however, such analytical frameworks focusing on constraints which may involve efficiency of action are not sufficient, even if they call into play different sorts of universals – anatomical, psychological, and physical – or if they integrate the variability of actions in progress, mobilising the idea of affordance, initiated by Gibson (1979). To properly document body techniques and elementary actions on matter, the social bases of

63

From Tools to Symbols

actions, knowledge (know-how), and skills in all their dimensions must be taken into account. Unfortunately, the material available on these aspects is very scarce. This study called on the notion of elementary actions on matter in hope of its being applicable to prehistory (Joulian, 1993). A quick comparison of primate and human actions reveals that certain functions (cutting and slicing, for example) are not necessarily associated with tool-use among primates, yet they are observed in connection with the first stone tools used by humans as early as 2,6–2,3 million years ago (Merrick, 1976; Roche & Tiercelin, 1980; Semaw et al., 1997). In phylogenetic terms, common chimpanzees (Pan troglodytes) are our closest relatives. They are also the most remarkable tool-users in the animal world, and their tools are used in a wide variety of activities (defence, feeding, grooming). This variety in turn leads to a special interest in the absence of certain actions. Is cutting a sign of an important change and somehow involved with factors of hominisation? Do sharp cutting tools, which since the beginning of prehistory have been interpreted as belonging to humankind, have a greater influence than other tools (hammers or scrapers, for example) on human technical development? Does the very act of ‘cutting’, which lies behind the instrument, reflect or induce capacities and skills leading to hominisation? Since such questions cannot be answered directly from what we know about ancient prehistory, they have never been pursued. Only two major authors, André Leroi-Gourhan in France and Glynn Isaac in the USA, have investigated the subject. The conception of evolution developed by Leroi-Gourhan in Le geste et la parole (1964–1965 and 1993 for the English version) is one of the most interesting for its breadth and its richness. It covers very diverse fields of knowledge and describes techniques in a dual perspective, both evolutionary and comparative. The coherence of his view of hominisation is such that it has influenced several generations of prehistorians. Consider the great importance of tools in French prehistory and palaeontology, and, concurrently, the weakness of thought on non-tool techniques, or body techniques. Leroi-Gourhan, a student of Mauss, was aware of these concepts that he managed to include in his organicistic and continuistic idea of evolution. He was thoroughly acquainted with animal behaviour, but except for a few very particular uses – especially in his analysis of parietal figures (Leroi-Gourhan, 1979) – his empiricism kept him from integrating animal behaviour in his reasoning. On the other hand, this empiricism allowed him to counterbalance the overly philosophical positions of several of his colleagues. The influence of his ‘palethnologie’ should not be underestimated, which explained activities and ways of life (Leroi-Gourhan & Brézillon, 1972), but not behaviour, as true interfaces of evolution (contra Chance in the same period, 1974, and most modern ethologists, for example, Lee, 1988) (Joulian, 1995c). Contrary to the ideas common to the prehistory of his time, which attributed the first tools to reflexive knowledge and creative thought in the image of modern humans,

64

Significant tools and signifying monkeys

Leroi-Gourhan thought that the greatest part of human technical evolution was the expression of a biological fact. He even qualified ‘the tool as a veritable secretion of the body and brain of Anthropians’ (my translation). In all periods, even the most ancient, the artificial tool responds, according to him, to stereotypes and should be studied as ‘the product of intelligence integrated with matter and function’ (Leroi-Gourhan, 1964, 132–133; 1993, 91). Although Leroi-Gourhan found that the earliest human pebble industries corresponded to a particular mental stereotype, he differed from other typological conceptions of his time by proposing that the stereotype had a direct bearing not on the tools or on the products, but rather on operations and actions: ‘Strike a pebble at 90° with another pebble.’ He stated, furthermore, that though the goal is indeed to produce a sharp cutting edge, the action carried out by an early hominid is that of very simple percussion ‘which would also serve to break bones, crack nuts, or stun a beast with a club’ (Leroi-Gourhan, 1965). His analysis of gestures and motor functions in mammals is also important in a more general evolutionary perspective. Leroi-Gourhan showed, for example, the inversion of the hand/face relation when comparing primates with rodents or carnivores. He also demonstrated – and this is the main element of his analysis of human actions in comparison to those of primates – the subordination of freeing the hands to the appearance of bipedalism (Leroi-Gourhan, 1965; 1991). Even if human tools are ‘exuded’, even if they are ‘organs’ produced by slow evolution, there were, for Leroi-Gourhan, two very distinct levels: that of primates that have (practically) no tools, and that of humans, bipedal with hands freed from locomotive constraints and intervening in the world through the intermediary of artefacts. This difference in level between animal and human technicity was called into question thirty years after Le geste et la parole, when primates were observed to use tools in their natural environment. We should also note (McGrew, personal communication) that most tool use in humans and chimpanzees is performed sitting down, when the hands are not used in locomotion. The chart of elementary actions on matter proposed by Leroi-Gourhan that expressed, in his words, ‘the technical behaviour of primates and the technical capital of man, from his beginnings up to the dawn of Homo sapiens’ (Leroi-Gourhan, 1965, 38) can today be reread and completed. It is adapted here for its originality and its ambition to bring about a synthesis: it is the only such diagram that combines motor skills, actions actually carried out, and tools. Table 2 summarises, in the categories proposed by Leroi-Gourhan, the technical actions and instruments of wild chimpanzees described over the last thirty years. A first observation concerns the relatively high number of tools and their diversity. Indeed, they fulfil most of the basic functions of human tools. What apparently distinguishes chimpanzee tools from those of prehistoric man is that the former are made mostly

65

From Tools to Symbols Table 2 Types of relations to matter among primates (inspired by and adapted from Leroi-Gourhan, 1965).

Dental percussion Manual percussion Agression Aquisition Feeding Care Social function

Crushing Sectioning

Hammering

Percussion with nails Scraping Digging

Grasping Labio-dental

Relationship

Digito-palmar

Locomotion, leaping Brachiation, seizing Affective contact Kneading Cupping Snuggling, protection

Crusher Knife Awl Spike Piercing stick

Leaf gloves Chopper, hammer, club Spatula Sponge Hammer, anvil

Notcher Digging stick Pick, hoe Twigs Probe

Peeling Grooming Molding

Graver Punch Needle

Leaves

Probes Scraper

Spear

Stones Stone projectile Bola Sticks

Interdigital

Projection

Tearing

of vegetable matter rather than mineral and that they do not necessarily involve a transformation. But it must be understood that if prehistorians work principally with stone artefacts, it is because only durable vestiges (stone, bone, antlers, etc.) are available for analysis. This bias, due to principles of conservation, may seem obvious, yet it has not been sufficiently explored. Nothing proves, for example, the common belief among prehistorians that working with stone implies a higher level of cognitive skills than chimpanzees’ use of complicated tools or their transforming wood sticks into tools. Though fashioning stone and fashioning softer materials do, of course, involve different constraints, the demonstration of distinct or ‘superior’ cognitive skills has yet to be established. That said, the actions of crushing and splitting (column 1) are absent among chimpanzees (see above), and more generally, the only action among primates which more or less fits this category is that of piercing. It is found, for example, in opening up termite mounds in central Africa (Sabater-Pi, 1978), or beehives (McGrew, 1992) or palm heads (Yamakoshi & Sugiyama, 1995). If, as has been stated, this chart reflects an evolutionary logic proceeding from organs to instruments, and if it constrains the analysis within an excessively rigid framework, this is because Leroi-Gourhan’s schema rested on the extremely strong relation believed to exist in the 1960s between bipedalism, freed hands, and tool

66

Significant tools and signifying monkeys

manufacturing. At that time, these three characteristic ensembles seemed to be concomitant, whereas today bipedal locomotion is being pushed farther and farther back in time, to more than 2 million years before the appearance of the first tools. For example, the probably bipedal hominid found in Aramis in Ethiopia dates to more than 4 million years (White et al., 1994), and bipedal Australopithecus footprints discovered at Laetoli in 1978 are 3,2 million years old (Leakey, 1981). The oldest archaeological sites, however, date back to ‘only’ 2,6–2,5 million years (Roche, 1996). Direct causality can thus not be invoked to explain the origin of tools, yet it underlies a great number of palaeoanthropological works which still associate bipedalism and technical ability as the foundation of humankind. The socio-cultural and cognitive factors involved in technical innovation have not really been diffused in palaeoanthropology sensu stricto, with a few exceptions (Holloway, 1969; Tobias, 1983; 1995). If we now consider another research tradition, that of Anglo-American prehistory, this concept of the action being more important than the tool is rarely found. The prehistorian Glynn Isaac was an exception. In two articles in 1983 and 1989, he examined the adaptive significance of the first tools and the functions of cutting and carrying (‘Cutting and carrying: archaeology and the emergence of the genus Homo’). His approach was original in prehistory in that it compared the supposed functions of Oldowan tools to what we know of primate behaviour, in the chimpanzee in particular. He defined the elementary functions for which the Oldowan tools could have been made or employed, and explored the functional (adaptive) or cultural meanings of lithic assemblages by way of a parsimonious reasoning applying the method of residuals. This type of reasoning allowed him to show that, with regard to pre-Acheulian industries,

Figure 3

Plio-Pleistocene chopping tool from Comoé region, Côte d’Ivoire, 2001 (photo F. Joulian).

67

From Tools to Symbols

‘in spite of the diversity of forms, the early assemblages were fundamentally simple’ (Isaac, 1983, 15) and that they could be classed in two large categories, one of relatively massive instruments with robust edges, the other of relatively light instruments with fine, sharp edges, or pointed (flaked pieces vs detached pieces). Technical analysis showed that the production of stone flakes was much more important than had previously been thought and revealed the efficient simplicity of the means used to produce them. Microwear analyses of these sharp-edged tools (Keeley & Toth, 1981; Beyriès, 1993) also yielded diverse and unexpected results for such old tools (used for meat, wood, and plants). Studies in archaeo-zoology that describe in detail the tooth marks left on animal bones (gazelles, elephants, hippopotami, etc.) support the microwear results (Potts & Shipman, 1981; Bunn & Kroll, 1986; Potts, 1988). Some of the tooth marks can be associated with the dismemberment of the animal carcasses, others with the slicing off of the distal extremities of the limbs, and still others with the removal of meat from the diaphyses. In the early 1980s, what had long been mere speculation in prehistory (the use of the first tools as means of access to carcasses) was actually demonstrated as constituting ‘indeed one novel adaptation connected with the invention of shaped-edged tools’ (Isaac, 1983, 20). Experimental studies of the feasibility of the tasks which could be carried out with the tools of 2 million years ago (especially those by Toth, 1985; Toth & Schick, 1993)(Fig 4), however simple they may be, show that the tools match perfectly the different tasks which could have been carried out by hominids. Note that this table compresses more than 1,5 million years and associates Oldowan and old Acheulian technocomplexes. The functions appear to be similar for both, and this may be true for more recent periods. However the risk of simplification or trivialisation must be admitted at this stage of the analysis. The probable polyfunctionality of certain tools is not taken into account here. Rather, the table provides a baseline, and probably offers the most complete picture for these earlier periods. The stone tools of ancient hominids had obvious adaptive advantages in access to food. Signs of polish on bones at the Swartkrans site in South Africa (Brain, 1994) also attest that hominids could have used them as digging sticks and thus had access to all sorts of tubers inaccessible to other primates (for a synthesis of studies on vegetable food, see Sept, 1992). Recent studies carried out by Backwell and d’Errico have extended this range of interpretation by showing, based on an experimental approach and on a chimpanzee model, that lower Pleistocene hominids from South Africa had access to a much more diverse diet than previously thought (Backwell & d’Errico, 2001). In fact, they have introduced into our field of vision the most important biomass on earth, largely predated by primates: insects (Joulian, 2001a). The use of bone tools to dig for termites might also correspond to a new and specific type of action on matter (i.e. in actions related to

68

Significant tools and signifying monkeys

Figure 4

Possible functions for lower Pleistocene stone tools (from Toth, 1985).

69

From Tools to Symbols Table 3 Elementary actions on matter: comparative chart hominids/chimpanzees. Action

Chimpanzee (Pan t.)

Plio-pleisto. Hominid

without tool

tool

without tool

tool

Chew

X

_

X

_

Beat

X

_

X

_ _

Nibble

X

_

X

Grate/ teeth

?

_

?

Drink

X

X

X

X*

X

Blow

X

Hold

X

?

Hold/ hand

X

X

Pinch/ fingers

X

X

Squeeze

X

X

Lift

X

X

Shake

X

X

X?

Pull

X

captivity

X

Push

X

captivity

X

Turn over

X

X

Push down

X

X?

Grind - Smash

X

X

Knead

X

Rub

X

X

X?

Scratch

X

X

X?

X

Scrape

X

X

X?

X

Wipe

X

X

X?

Sponge

_

X

_

Pierce

X

X

X?

Insert

X

X

X?

X

X?

Probe

_

X

_

Fish

_

X

_

Dig

X

X

X?

X

X?

Pound

_

X

_

Hit

X

X

X?

X

Beat (repeated action)

X

X

X?

X

Slice

X

X?

X

70

Significant tools and signifying monkeys Action

Chimpanzee (Pan t.) without tool

Plio-pleisto. Hominid without tool

tool

Cut/ hands or teeth

X

tool

X?

X

Twist

X

X?

Screw

X

Lever

X

X

X?

Bridge

X

X

X?

X?

Hold (object, baby)

X

Drop

X

Throw

X

Move/ feet-handsmetacarpals

X

Run Jump Swing

X

Climb

X

X

X

Swim

_

_

X?

_

Copulate

X

_

X

_

X? _

X?

_

X?

X?

_

X

_

X

_

X

_

X

X

X X

X = present; X ? = probable or possible; – = impossible; blank = absent because it does not exist, or, for vegetal tools, because there is no preservation possible; * = in the ‘social tool-use’ case

perforation). The two authors have also shown a probable grinding action involved in the manufacture of bone tools from the Swartkrans site (d’Errico & Backwell, 2003). Let us return to the idea of cutting and Isaac’s thesis that ‘If meat scavenged from large carcasses was becoming an important potential food for some hominids in the late Pliocene 2–3 million years ago, it is conceivable that a strong need for cutting tools arose’. Faced with carnivores, ‘Tools would have enabled the hominids to make a quick foray, get hunks of meat and leave’ (Isaac, 1983, 21). Given the nature of available materials, statements such as these reveal the limits of archaeological interpretation, or even its tautological nature. Recourse to a comparative approach involving primates seems to be the best solution to ‘inform’ these old facts and to place such a practice, if it really has the importance it is said to have, in a broader history, in, as Leroi-Gourhan would have said, ‘a palaeontology of gestures and of tools’. The goal is also, as we have seen, to uncover the action, and perhaps the functioning, behind the tool. We would then be better equipped to distinguish the different ‘embodiments’ or ‘body techniques’, and, moreover, the different cognitive skills called into play by the first hominids in carrying out such activities. It thus seems essential to describe and inventory the actions of primates in the wild.

71

From Tools to Symbols

However, as was stated in the first part of this paper, the project would depend on a detailed, comparative analysis of movement. The task will be difficult and, considering the documentation currently available, impossible to carry out for the time being. Accordingly, the following synthetic chart is intended only to bring to the surface a few of the principal actions not immediately apparent to us. Indeed, it is meant only to point out actions that could be evidence of techniques specific to hominids or chimpanzees. For the Plio-Pleistocene, it is possible to speak only of attested actions, since available information does not allow us to state categorically that certain actions did not exist. Therefore, only chimpanzees are considered here, and, in particular, the actions that they lack as compared to actions present 2 million years ago (indicated by ‘x’ on the chart). This type of comparison is one of the most complex, due not only to our ignorance of actions existing in the past, but also to the degree of generality with which we analyse them. The scope of the analysis is thus quite limited. The pinching action, to take just one example, is sometimes observed among chimpanzees, and therefore appears in the affirmative in the chart, but it is not done very efficiently (in terms of strength) among anthropoids. Pinching forcefully is impossible for chimpanzees because of the anatomical configuration of the hand (Marzke & Wullstein, 1996). All of the actions listed on the chart should be described in detail in this way and evaluated in anatomical, environmental, social and psychological frameworks. Other

Figure 5

Termite fishing from Gorowi, Comoé Region, Côte d’Ivoire, 2001: (a) termite mound; (b) stick introduced and left by chimpanzees (photos F. Joulian, 2005).

72

Significant tools and signifying monkeys

contexts should also be taken into account, especially those of the activity and the way the tool works, but all this could hardly be done in the space of this short essay. It remains, nevertheless, that the absence of actions such as grating, pressing, grinding, and cutting among chimpanzees is immediately noticeable. In fact chimpanzees are able to carry out these actions as well as others such as kneading and cutting with body apparatus (hands, nails, and teeth), but not through the intervention of tools. Tools allow chimpanzees to act on matter in a repeated, rhythmical fashion (as in nut-cracking), but they do not maintain contact with the object ‘in diffused placed percussion’ (percussion posée diffuse, Leroi-Gourhan, 1971), such as when a wide active edge, a knife for instance, is used to cut or grate. Repetition and contact constitute the common characteristics of these two opposed types of action. The inability of chimpanzees to act according to the contact mode contrasts sharply with their ability to break and throw, in the ballistic mode. Chimpanzees are even able to crush (as with a pestle) the heads of palms (Yamakoshi & Sugiyama, 1995) and in doing so show different variations of the same type of action. In a Piagetian approach to action, we would probably interpret the absence of ‘diffused placed percussion’ in terms of cognitive development (Parker & Gibson, 1977), but we would not be any closer to explaining the significance of the action itself. It can also be pointed out that crushing with a pestle, as opposed to cutting (an action present in the earliest tool use), is not identified with certainty until the middle Palaeolithic (Beaune, 2000) with Homo sapiens, though crushing and cutting seem to be based on similar mechanical principles. We would add that the evolutionary perspective followed by Beaune (2004) and stressing different stages (i.e. a cracking one versus a knapping one) does not seem to be very productive because it is too synthetic and too orthogenetic. Without ruling out a possible bias in the documentation, ballistic motion in prehistoric cutting or in nut-cracking by chimpanzees implies a complexity radically different from that of using an instrument in a deliberate placed/posed and repeated fashion. In other words, explanations for the absence of an action founded on cognitive differences between chimpanzees and hominids are fragile and lead us rather to biomechanical or socio-cultural explanations (in which case we are faced with the problem of comparing species that are anatomically and socially very different). Or maybe, more simply, we should consider feeding modes and the transformation of food as factors that directly influence the existence or absence of such actions. The phylogenetic significance of the actions on, or relations with, matter that have been examined here is still elusive. What is the meaning of cutting, piercing, grinding, etc., with or without tools, depending on whether one is bipedal or moves on all fours? And what is the significance for evolutionary theory of these modes of action or changes in modes of action? These are a few of the questions that remain to be answered. If we

73

From Tools to Symbols

could succeed in showing the representational and collective aspect of actions, these questions could be directed towards the physical principles that govern the actions (traction, pressure, leverage, torsion, etc.) and that would allow us to describe the skills and knowledge of hominoids in a more coherent manner (Joulian, 2001b). In closing, should we continue to grant to the action of cutting and to sharp-edged tools the prominent value usually given them? Everything leads us to believe that cutting and sharp-edged tools play an important role in new feeding modes in prehumans, especially for exploiting meat resources, although they are absent in primates. Yet they should also be examined from the point of view of elementary actions, and thus in relation to other actions, such as rubbing, grating, grinding, etc., that are non-existent among hominids and pongids. It is in this structural relation between body techniques on the one hand and ‘unnatural’ tooled techniques that have nevertheless been stabilised into gestures and tools on the other, that non-functional meaning could emerge with the first prehuman cutting tools. Should the latter be given a founding status as a symbol of humanity? Because tools and techniques represent only a narrow window on the past, it seems unlikely.

Acknowledgements Thanks to Helene Roche, William McGrew, Blandine Bril, George Guille-Escuret, François Sigaut and Suzanne de Cheveigné for their attentive readings of the manuscript and suggestions given for improving it. I also thank the Foundation Fyssen which in 1985 supported my venturing into the marginal areas of prehistory and primatology at a time when that approach was far from being accepted or practised.

References Backwell, L.R. & d’Errico, F. (2001). Evidence of termite foraging by Swartkrans early hominids. Proceedings of the National Academy of Sciences 98(4), 1358–1363. Beaune, S.A. de (2000). Pour une archéologie du geste: broyer, moudre, piler, des premiers chasseurs aux premiers agriculteurs. Paris: CNRS. Beaune, S.A. de (2004). The invention of technology. Prehistory and cognition. Current Anthropology 45(2), 139–162. Beyriès, S. & Joulian, F. (1990). L’utilisation de l’outil chez les animaux: chaînes opératoires et complexité technique. Paléo. Revue d’Archéologie Préhistorique 2, 17–26. Beyriès, S. (1993). Peut-on déterminer la fonction des outils du paléolithique le plus ancien? In (A. Berthelet & J. Chavaillon, Eds.) The Use of Tools by Human and Non-human Primates, pp. 225–236. A symposium of the Fyssen Foundation. Oxford: Clarendon Press. Brain, C.K. (1994). The Swartkrans palaeontological research project in perspective: results and conclusions. South African Journal of Science 90, 220–223. Boesch, C. (1991). Teaching in wild chimpanzees. Animal Behaviour 41(3), 530–532.

74

Significant tools and signifying monkeys Boesch, C. & Boesch-Achermann, H. (1994). Technique et culture chez les chimpanzés sauvages. Techniques et Culture 23–24, 1–28. Boesch, C., Marchesi, P., Fruth, B. & Joulian, F. (1994). Is nut-cracking in wild chimpanzees a cultural behaviour? Journal of Human Evolution 26, 325–338. Boesch, C. & Tomasello, M. (1998). Chimpanzee and human cultures. Current Anthropology 39 (5), 591–614. Bunn, H. & Kroll, H.M. (1986). Systematic butchery by Plio-Pleistocene hominids at Olduvai Gorge, Tanzania. Current Anthropology 27(5), 431–452. Byrne, R. (1995). The Thinking Ape: Evolutionary Origins of Intelligence. Oxford: Oxford University Press. Chance, M. (1974). Une dimension absente en biologie: le comportement. In (E. Morin & M. Piattelli-Palmarini, Eds.) L’unité de l’Homme I. Le primate et l’homme, pp. 218–227. Colloque de la Fondation Royaumont. Paris: Point Seuil. d’Errico, F. & Backwell, L. (2003). Possible evidence of bone tool shaping by Swartkrans hominids. Journal of Archaeological Science 30, 1559–1576. Doran, D. (1996). Comparative positional behaviour of the African apes. In (W.C. McGrew, L.F. Marchant & T. Nishida, Eds.) Great Ape Societies, pp. 213–224. Cambridge: Cambridge University Press. Galef, B.G. (1976). Social transmission of acquired behaviour: a discussion of tradition and social learning in vertebrates. In (J. S. Rosenblatt, R. A. Hinde, E. Shaw & C. Beer, Eds.) Advances in the Study of Behaviour. Vol. 6, pp. 77–100. New York: Academic Press. Gibson, J. (1979). The Ecological Approach to Visual Perception. Boston: Houghton Mifflin. Gille, B. (1978). Histoire des techniques. Paris: Gallimard. Goodall, J. (1973). Cultural elements in a chimpanzee colony. In (E.W. Menzel, Ed.) Precultural Primate Behavior. Vol. 1, pp. 144–184. Symposium of the IVth International Congress of Primatology. Portland, Oregon, 1972. Bâle: S. Karger. Gosselain, O. (in press). Ça ne s’attrape pas, ça se pratique. Contour et dynamique des cultures potières en Afrique. In (F. Joulian, Ed.) How the Chimpanzee Stole Culture. Paris: Odile Jacob. Guille-Escuret, G. (1994). Outils, gestes et produits: ressemblances illusoires et contrastes instructifs entre les animaux et l’homme. Techniques et Culture 23–24, 63–90. Holloway, R. (1969). Culture: a human domain. Current Anthropology 10, 395–412. Huffman, M.A. & Wrangham, R.W. (1994). Diversity of medicinal plant use by chimpanzees in the wild. In (R.A. Wrangham, W.C. McGrew, F.B.M. de Waal and P.G. Heltne, Eds) Chimpanzee Cultures, pp. 129–148. Cambridge: Harvard University Press. Hunt, K.D., Cant, J.G., Gebo, D.L., Rose, M.D., Walker, S.E. & Youlatos, D. (1996). Standardized descriptions of primate locomotor and postural modes. Primates 37(4), 363–387.

75

From Tools to Symbols Hunt, K. (1992). Positional behaviour of Pan troglodytes in the Mahale Mountains and Gombe Stream National Parks, Tanzania. American Journal of Physical Anthropology 87, 83–105. Ingold, T., Ed. (1988). Introduction. In What is an Animal? London: Unwin Wyman. One World Archaeology, pp. 1–16. Ingold, T. (in press). Trois en un: sur la dissolution des distinctions entre corps, esprit et culture. In (F. Joulian, Ed.) How the chimpanzee stole culture. Paris: Odile Jacob. Inoué, N., Tonooka, R. & Matsuzawa, T. (1995). Developmental processes of nut-cracking skill among infant chimpanzees in the wild. Japanese Journal of Developmental Psychology 7(2), 148–158. Isaac, G. (1983). Early stages in the evolution of human behaviour: the adaptive significance of stone tools. In Zesde Kroon-Vroordracht Conference, Stichting Nederlands Museum voor Anthropologie en Praehistorie: Amsterdam, pp. 5–32. Itani, J. & Nishimura, A. (1973). The study of infrahuman culture in Japan. A review. In (E.W. Menzel, Ed.) Precultural Primate Behaviour, Vol. 1, pp. 26–50. Symposium of the IVth International Congress of Primatology. Portland, Oregon, 1972. Bâle: S. Karger. Joulian, F. (1986). ‘Pan faber’, bibliographie sélective à propos des évidences d’outils chez les singes supérieurs: problématiques anthropologiques. DEA. Université de Paris I, PanthéonSorbonne. Joulian, F. (1993). Application de l’éthologie des chimpanzés Ouest-africains au comportement des hominidés du Plio-pléistocène: le problème de la culture. Thèse de doctorat, Vol. 2. Université de Paris I, Panthéon-Sorbonne. Joulian, F. (1994a). Culture and material culture in chimpanzees and early hominids. In (J.J. Roeder, B. Thierry, J.R. Anderson & N. Herrenschmidt, Eds) Current Primatology. Vol. II, pp. 397–404. Social Development, Learning and Behaviour. Selected Proceedings of the XIVth Congress of the International Primatological Society. Strasbourg: Univ. Louis Pasteur. Joulian, F. (1994b). Peut-on parler d’un système technique chimpanzé? Primatologie et archéologie comparées. In (B. Latour & P. Lemonnier, Eds) De la préhistoire aux missiles balistiques: l’intelligence sociale des techniques, pp. 45–64. Paris: La Découverte. Joulian, F. (1995a). ‘Human and non-human primates’. Des limites de genre bien problématiques en préhistoire. Préhistoire et Anthropologie Méditérranéennes 4, 5–15. Joulian, F. (1995b). Mise en évidence de différences traditionnelles dans le cassage des noix chez les chimpanzés (Pan troglodytes) de la Côte d’Ivoire, implications paléoanthropologiques. Journal des Africanistes 65(2), 57–77. Joulian, F. (1995c). Hommes et primates, primatologues et préhistoriens: l’hominisation en question. Communication orale au Colloque ‘Geste technique, parole, mémoire: actualité scientifique et philosophique de Leroi-Gourhan’ (dir.) F. Audouze, B. Stiegler. Meudon-Bellevue: CNRS.

76

Significant tools and signifying monkeys Joulian, F. (1996a). Comparing chimpanzee and early hominid techniques: some contributions to cultural and cognitive questions. In (P.A. Mellars & K.R. Gibson, Eds) Modelling the Early Human Mind, pp. 173–189. Cambridge: McDonald Institute for Archaeological Research. Joulian, F. (1996b). Technique et société chez les Hominidés du Plio-Pléistocène et chez les Panidés: quelques réflexions théoriques et méthodologiques sur l’usage d’analogies. In (C. Peretto, F. Facchini, Eds) Congrès UISPP, Forli. ‘Les premiers hommes et leurs manifestations culturelles’. Lithic industries, language and social behaviour in the first human forms’, pp. 69– 75. Forli: ABACO. Joulian, F. (1998). Hominization and apes: an unnatural kinship. ‘Genes and humanity’s past. A renewed dialogue’. Diogenes 180, 73–96. Joulian, F. (2000). Techniques du corps et traditions chimpanzières. ‘Les animaux pensent-ils?’ Terrain 34, 37–54. Joulian, F. (2001a). Origine de l’homme, termites et chimpanzés. La Recherche. 347, 14–15. Joulian, F. (2001b). On elementary actions on matter and on their mechanical principles among Hominoids or on how to frame living and extinct primates? In (V. Roux & B. Bril, Eds) Knapping Stone: A Uniquely Hominid Behaviour? International workshop, November 2001, Abbaye des Prémontrés, Pont à Mousson, France. Joulian, F. (in press). Hommes et Primates en Perspective. Paris: Odile Jacob. Kawaï, M. (1965). Newly acquired precultural behaviour of the natural troop of Japanese monkeys on Koshima Island. Primates 6, 1–30. Keeley, L. & Toth, N. (1981). Microwear polishes on early stone tools from Koobi-Fora, Kenya. Nature 293, 464–465. Leakey, M.D. (1981). Tracks and tools. In (J.Z. Young, E.M. Jope & K.P. Oakley, Eds) The Emergence of Man. A joint symposium of the Royal Society and the British Academy. Philosophical Transactions of the Royal Society of London. 292, 95–102. Lee, P. (1988). Comparative ethological approaches in modelling hominid behaviour. Ossa 14, 113–126. Lemonnier, P. (1991). De la culture matérielle à la culture? Ethnologie des techniques et préhistoire. In 25 ans d’études technologiques en préhistoire. Bilan et perspectives, pp. 15–20. Actes des XIè rencontres internationales d’archéologie et d’histoire d’Antibes, Oct. 1990. Juan les Pins: APDCA. Lemonnier, P. (1992). Elements for an anthropology of technology. Anthropological Papers No. 88. Ann Arbor, Museum of Anthropology. Lemonnier, P. (1993). Technological Choices. Transformation in Material Cultures Since the Neolithic. London: Routledge, pp. 1–35. Leroi-Gourhan, A. (1964). Le Geste et la Parole. t. 1. Technique et langage. Paris: Albin-Michel. Leroi-Gourhan, A. (1965). Le Geste et la parole. t. 2. La mémoire et les rythmes. Paris: AlbinMichel.

77

From Tools to Symbols Leroi-Gourhan, A. (1971). Evolution et techniques. I. L’homme et la matière. Paris: Albin Michel. Leroi-Gourhan, A. & Brézillon, M. (1972). Fouilles de Pincevent. Essai d’analyse ethnographique d’un habitat magdalénien (la section 36). Vol 2. VIIè supplément à Gallia Préhistoire. Paris: CNRS. Leroi-Gourhan, A. (1979). Le réalisme de comportement dans l’art paléolithique d’Europe de l’ouest. In (H.-G. Baudi, W. Huber & B. Sitter, Eds) La contribution de la zoologie et de l’éthologie à l’interprétation de l’art des peuples chasseurs préhistoriques, pp. 75–90. 3è colloque de la Société suisse des Sciences Humaines. Leroi-Gourhan, A. (1993). Gesture and Speech. Cambridge, Massachusetts: The MIT Press, p. 431. Marchant, L., McGrew, W.C. & Eibl-Eibesfeldt, I. (1995). Is human handedness universal? Ethological analyses from three traditional cultures. Ethology 101, 239–258. Marchant, L. & McGrew, W.C. (1996). Laterality of limb function in wild chimpanzees of Gombe National Park: comprehensive study of spontaneous activities. Journal of Human Evolution 30, 427–443. Marzke, M.W. & Wullstein, K.L. (1996). Chimpanzee and human grips: a new classification with a focus on evolutionary morphology. International Journal of Primatology 17(1), 117–139. Matsuzawa, T. (1994). Field experiments on use of stone tools by chimpanzees in the wild. In (R.W. Wrangham, W.C. McGrew, F.B.M. de Waal & P.G. Heltne, Eds) Chimpanzee Cultures. Cambridge: Harvard University Press, pp. 351–370. Mauss, M. (1929a). Intervention à la suite d’un exposé de M.M. Guillaume et Meyerson à l’Institut français d’anthropologie [Débat sur l’origine de la technologie humaine], Anthropologie 39, 130. (Oeuvres III 1969, Paris Les Editions de Minuit, p. 257). Mauss, M. (1929b). Les civilisations: éléments et formes. In Civilisation. Le mot et l’idée. (La renaissance du livre, 2è fasc. 1930: 81–108. (Oeuvres II 1969, Paris: Les Editions de Minuit, pp. 456–479). Mauss, M. (1936). Sociologie et Anthropologie. Paris: Puf. McGrew, W.C. 1985. The chimpanzee and the oil palm. Patterns of culture. Social Biology and Human Affairs 50, 7–23. McGrew, W.C. 1992. Chimpanzee Material Culture. Implications for Human Evolution. Cambridge: Cambridge University Press, p. 277. McGrew, W.C. (in press). Comment le chimpanzé a volé la culture ou vingt leçons durement apprises en primatologie culturelle. In (F. Joulian, Ed.) Comment le chimpanzé a volé la culture. Paris: Balland. McGrew, W.C. & Tutin, C.E.G. (1978). Evidence for social custom in wild chimpanzees? Man 13, 234–251.

78

Significant tools and signifying monkeys Menzel, E.W. 1973. Préface. In (E.W. Menzel, Ed.) Precultural Primate Behaviour. Vol. 1, p. 6. Symposium, IVth International Congress on Primatology, Portland, Oregon, 1972. Bâle: S. Karger. Merrick, H.V. (1976). Recent archaeological research in the Plio-Pleistocene deposits of the lower Omo valley, southwestern Ethiopia. In (G. Isaac & E.R. McCown, Eds) Human Origins, pp. 461–481. Menlo Park: W.A. Benjamin. Napier, J.R. & Walker, A.C. (1967). Vertical climbing and leaping: a newly recognized category of primate locomotion. Folia Primatologica 6, 204–219. Nishida, T. (1986). Local traditions and cultural transmission. In (B.B. Smuts, D.L. Cheney, R.M. Seyfarth, R.W. Wrangham & T.T. Struhsaker, Eds) Primate Societies, pp. 462–474. Chicago: Chicago University Press. Parker, S. & Gibson, K.R. (1977). Object manipulation, tool use and sensorimotor intelligence as feeding adaptations in Cebus monkeys and great apes. Journal of Human Evolution 6, 623– 641. Potts, R. (1988). Early Hominid Activities at Olduvai. New York: Aldine de Gruyter, p. 396. Potts, R. & Shipman, P. (1981). Cutmarks made by stone tools on bones from Olduvai Gorge, Tanzania. Nature 291, 577–580. Povinelli, (1994). What chimpanzees (might) know about the mind. In (R.W. Wrangham, W.C. McGrew, F.B.M. de Waal & P.G. Heltne, Eds) Chimpanzee Cultures, pp. 285–300. Cambridge, Massachusetts: Harvard University Press. Premack, D. & Premack, A. (1994). Why animals have neither culture nor history. In (T. Ingold, Ed.) Companion Encyclopedia of Anthropology, pp. 350–365. London: Routledge. Roche, H. & Tiercelin, J.-J. (1980). Industries lithiques de la formation Plio-Pléistocène d’Hadar Ethiopie (campagne 1976). In (R.E.F. Leakey & B.A. Ogot, Eds) Actes du 8è Congrès Panafricain de Préhistoire et d’Etudes du Quaternaire, Nairobi, 1977, pp. 194-199. Roche, H. (1996). Remarques sur les plus anciennes industries en Afrique et en Europe. In (F. Facchini, Ed.) The First Humans and their Cultural Manifestations, pp. 55–68. Colloque VIII Congrès de l’UISPP. Forli: ABACO. Sabater-Pi, J. (1978). El chimpancé y los origenes de la cultura. Barcelona: Anthropos, Editorial del Hombre, p. 159. Schlanger, N. (1991). Le fait technique total. La raison pratique et les raisons de la pratique dans l’oeuvre de Marcel Mauss. Terrain 16, 114–130. Schlanger, N. (1998). The study of techniques as an ideological challenge: technology, nation and humanity in the work of Marcel Mauss. In (W. James and N.J. Allen, Eds) Marcel Mauss A Centenary Tribute, pp. 192–212. Oxford: Berghahn Books. Semaw, S., Renne, P., Harris, J.W.K., Feibel, C.S., Bernor, R.L., Fesseha, N. & Mowbray, K. (1997). 2.5-million-year-old stone tools from Gona, Ethiopia. Nature 385, 333–336.

79

From Tools to Symbols Sept, J. (1992). Archaeological evidence and ecological perspectives for reconstructing early hominid subsistence behaviour. In (M.B. Schiffer, Ed.) Archaeological Method and Theory, Vol. 4, pp. 1–56. New York: Academic Press. Smuts, B.B., Cheney, D.L., Seyfarth, R.M., Wrangham, R.W. & Struhsaker, T.T., Eds (1987). Primate Societies. Chicago: Chicago University Press. Suddendorf, T, & Whiten, A. (2001). Mental evolution and development: evidence for secondary representation in children, great apes and other animals. Psychological Bulletin 127, 629– 650. Sugiyama, Y. (1993). Local variation of tools and tool behaviour among wild chimpanzeee populations. In (A. Berthelet & J. Chavaillon, Eds) The Use of Tools by Human and Nonhuman Primates, pp. 175–190. Oxford: Oxford University Press. Sugiyama, Y., Fushimi, T., Sakura, O. & Matsuzawa, T. 1993. Hand preference and tool-use in wild chimpanzees. Primates 43(2), 151–159. Takeshita, H. & Walraven, V. (1996). A comparative study of the variety and complexity of object manipulation in captive chimpanzees (Pan troglodytes) and bonobos (Pan paniscus). Primates 37(4), 423–441. Thierry, B. (1995). Social transmission, tradition and culture in primates: from the epiphenomenon to the phenomenon. Techniques et Culture 23–24, 91–119. Thierry, B. (2004). Quand les primates ont divergé: aux prémices de l’évolution culturelle. In (F. Audouze, N. Schlanger, Eds) Autour de l’Homme. Contexte et actualité d’André Leroi-Gourhan, pp. 179-193. Antibes: APDCA. Tobias, P.V. (1983). Hominid evolution in Africa. Canadian Journal of Anthropology 3, 163–190. Tobias, P.V. (1995). The communication of the dead. Earliest vestiges of the origin of articulate language. Zeventiende Kroon-Vroordracht, Gehouden voor de Stichting Nederlands Museum voor Anthropologie en Praehistorie, Amsterdam, p. 70. Tomasello, M. (1990). Cultural transmission in the tool and communicatory signaling of chimpanzees? In (S. Parker & K. Gibson, Eds) Language and Intelligence in Monkeys and Apes. Comparative Developmental Perspectives, pp. 274–311. New York: Cambridge University Press. Tomasello, M. (1994). The question of chimpanzee culture. In (R.W. Wrangham, W.C. McGrew, F.B.M. de Waal & P.G. Heltne, Eds) Chimpanzee Cultures, pp. 301–317. Cambridge, Massachusetts: Harvard University Press. Torigoe, T. (1987). Object manipulation in a captive troop of Japanese monkeys (Macaca fuscata): a developmental analysis. Primates 28(4), 497–506. Toth, N. (1985). The Oldowan reassessed: a close look at early stone artefacts. Journal of Archaeological Science 12, 101–120. Toth, N. & Schick, K. (1993). Making Silent Stone Speak: Human Evolution and the Dawn of Technologies. New York: Simon & Schuster, p. 351.

80

Significant tools and signifying monkeys Tuttle, R. H. & Watts, D. P. (1985). The positional behaviour and adaptive complexes of Pan gorilla. In (S. Kondo, Ed.) Primate Morphophysiology, Locomotor Analysis and Human Bipedalism, pp. 261–288. Tokyo: Tokyo University Press. Van Schaik, C.P., Ancrenaz, M., Borgen, G., Galdikas, B., Knott, C.D., Singleton, I., Suzuki, A., Utami, S.S. & Merrill, M. (2003). Orangutan cultures and the evolution of material culture. Science 299, 102–105. Vauclair, J. (1982). Sensorimotor intelligence in human and non-human primates. Journal of Human Evolution 11, 257–264. White, T.D., Suwa, G. & Asfaw, B. (1994). Australopithecus ramidus, a new species of early hominid from Aramis, Ethiopia. Nature 371, 306–312. Whiten, A., Goodall, J., McGrew, W.C., Nishida, T., Reynolds, V., Sugiyama, Y., Tutin, C. E.G., Wrangham, R.W. & Boesch, C. (1999). Cultures in chimpanzees. Nature 399, 682–685. Whiten, A. & Ham, R. (1992). On the nature and evolution of imitation in the animal kingdom: reappraisal of a century of research. In (P. Slater, J. Rosenblatt & C. Beer, Eds) Advances in the Study of Behaviour, pp. 239–283. New York: Academic Press. Yamakoshi, G. & Sugiyama, Y. (1995). Pestle-pounding behaviour of wild chimpanzees at Bossou, Guinea: a newly observed tool-using behaviour. Primates 36(4), 489–500.

81

Tools and brains: which came first? Phillip V. Tobias Department of Anatomical Sciences, University of the Witwatersrand Medical School, 7 York Road, Parktown 2193, Johannesburg, South Africa

Abstract Many scholars have sought to relate the findings and interpretations of archaeology to the evolution of the brain and mind. In effect, such studies venture a statement that the techniques and symbols explicit and implicit in the archaeological record are related, more or less directly, to the cognitive abilities, mental competences and intelligences of evolving hominins, ancient and modern. It has however been evident for some time that it is not only from the archaeological record that we may glean evidence on the evolution of hominin intelligence. The size and form of endocranial casts of fossil hominins have added grist to the mill of those probing the evolution of hominin cerebration. The analogy and in a hopeful mood the homology between the brains and behaviours of human and non-human primates, and those inferred for our remote ancestors, have provided new pointers in the analysis of culture. Indeed they question the validity of the very concept of culture, as understood during most of the twentieth century. Ethological studies have shown some close resemblances between human and ape behaviours. Just over fifty years ago, the apparently human preserve of tool-using and tool-making led Kenneth Oakley to speak of Man the Tool-Maker, while he could write ‘… it is evident that man may be distinguished as the tool-making primate …’. Yet, in the 1980s and 1990s, Jane Goodall, Frédéric Joulian, William McGrew and C. and H. Boesch cast a flood of new light on the implemental activities of wild chimpanzees, just as H. Khroustov, in Moscow, did for chimpanzees in captivity in the early 1960s. Joulian went on to contest the longcherished paradigm that ‘culture’ is an exclusively human realm. The pursuit by various groups of West African chimpanzees of nut-cracking in some populations, but not in others of the same species, strongly suggested that such behavioural traits were transmitted by epigenetic means. In a word, they were learned behaviour of a kind which, conventionally, has been assigned to human cultural behaviour. Based on our analysis of H. habilis endocasts and on a review of the inferred cultural and social aspects of this hominin, it is argued here that H. habilis was able to speak.

Résumé Nombre de scientifiques ont cherché à rattacher découvertes et interprétations archéologiques à l’évolution du cerveau et de l’intellect. En fait, de telles études avancent que les techniques et les symboles présentes de façon explicite ou implicite dans le registre archéologique sont liés, plus ou moins directement, aux capacités cognitives, aux compétences mentales et à l’intelligence des hominidés anciens et modernes. Il est pourtant

82

Tools and brains: which came first? devenu clair depuis un certain temps que nous serions à même de recueillir des preuves sur l’évolution de l’intelligence des hominidés autrement qu’en nous basant sur les restes archéologiques. L’étude des dimensions et morphologie des moulages endocrâniens des hominidés fossiles ont apporté de l’eau au moulin de ceux mènent des recherches sur l’encéphalisation des hominidés. L’analogie et probablement l’homologie entre le cerveau et le comportement des primates humains et non humains et ceux déduits pour nos ancêtres lointains, ont des implications sur notre vision des changements culturels au cours de la préhistoire. En effet, ces études mettent en question la validité du concept même de culture tel qu’il a été défini durant la plus grande partie du XXème siècle. Des études éthologiques ont démontré les fortes ressemblances qui existent entre le comportement des hommes et celui des singes anthropomorphes. Il y a un peu plus de cinquante ans, la fabrication et l’utilisation d’outils, considérés à l’époque comme le propre de l’homme conduisait Kenneth Oakley à parler de notre espèce comme du Fabricant d’Outil’ (Man the Tool-Maker), et à écrire’... il est clair que l’homme est le primate qui fabrique des outils ...’. Dans les années 1980 et 1990, Jane Goodall, Frédéric Joulian, William McGrew ainsi que C. et H. Boesch ont jeté une lumière nouvelle sur les activités techniques des chimpanzés sauvages, tout comme H. Khroustov l’avait fait à Moscou pour les chimpanzés en captivité au début de années 1960. Joulian a décidé de contester le paradigme longuement entretenu que la ‘culture’ est du domaine de l’homme. Le fait que divers groupes de populations de chimpanzés d’Afrique occidentale cassent des noix alors que d’autres de la même espèce ne le font pas, indique que de tels traits de comportement ont été transmis par des voies épigénétiques. En un mot, il s’agit là de comportements appris comparables à ceux qui ont été souvent considérés caractéristiques de notre espèce. En nous basant sur notre analyse de moulages endocrâniens d’Homo habilis et sur un examen des traits culturels et sociaux attribués à cet hominidé, nous proposons ici que l’Homo habilis était capable de parler.

Introduction Just over fifty years ago, the apparently human preserve of tool-using and toolmaking led Kenneth Oakley (1949) to speak of Man the Tool-Maker, while he could write ‘… it is evident that man may be distinguished as the tool-making primate …’. There was an immediate reaction. Investigators drew attention to many other instances of tool-using and, in a few cases, of tool-modifying, which was a presumably rudimentary form of tool-making, in other members of the Animal Kingdom, aside from the hominids. Examples of implemental activities were furnished by some birds, notably the Egyptian geese and some of Darwin’s finches. In 1968, along with the somatotypologist Barbara Honeyman Heath, I stood on the rocks at Carmel, California, and watched a sea otter diving for abalone, bringing it to the surface, producing a stone it had secreted under its forelimb, and then, using its thorax as an anvil, smashing the stone down upon the abalone to gain access to its contents. Kenyon (1969) recorded this behaviour by sea otters. Similarly, it had been on record since the nineteenth century that capuchin monkeys used stones to break open walnuts and other thick-shelled

83

From Tools to Symbols

nuts: I recall taking my students to the Johannesburg Zoo and, with the help of the then Zoo Curator, Malcolm Lyall Watson, showing them such activities by capuchins in their zoo enclosure. In the 1980s and 1990s, Jane Goodall, Frédéric Joulian, William McGrew and C. and H. Boesch cast a flood of new light on the implemental activities of wild chimpanzees. Joulian even compared Oldowan stone tools made, most people hold, by Homo habilis about 2 million years ago (Mya), with what he dubbed analogous tools made by chimpanzees of today. Joulian went on to contest the long-cherished paradigm that ‘culture’ is an exclusively human realm. The pursuit by various groups of West African chimpanzees of nut-cracking in some populations, but not in others in the same species, strongly suggested that such behavioural traits were transmitted by epigenetic means. In a word, they were learned behaviour of a kind that conventionally has been assigned to human cultural behaviour. In Moscow, in the early 1960s, H.F. Khroustov (1964) made a series of experiments on chimpanzees in captivity, in order to determine, if he could, ‘the highest frontier of implemental activity of anthropoids’. Having heard a presentation and seen a film on this work during the XIIth International Congress of Anthropological and Ethnological Sciences in Moscow in 1964, and knowing that this important work had largely escaped the attention of scholars in the West, I gave an extended summary of it a year later (Tobias, 1965). Khroustov’s researches could be regarded as a latter-day extension of the earlier researches of W. Köhler (1924) on chimpanzees in captivity. Other studies worthy of mention here are those of Kortlandt and Kooij (1963) and J.B. Lancaster (1968). Frédéric Joulian (1996) went one step further in his studies on the behaviour of chimpanzees in West Central Africa. He drew attention to the shaping of stones and to group-specific behaviour, seemingly culturally transmitted on standard definitions of cultural behaviour and transmission. Recently, Carel van Schaik, who has been studying orang-utan in the wild for thirty years, and his colleagues (2003) reported on a number of behavioural traits among six widely separated bands in Kalimantan (Borneo) and Sumatra. Examples of behaviour which some groups upheld and not others were: When a ‘kiss-squeak’ noise is made by inhaling, all groups use a leaf to amplify the sound, but one Kalimantan group uses hands cupped over the mouth to produce an altered sound. A ‘raspberry’ sound made by exhaling was made habitually by one out of six groups. Other behavioural traits were the use of ‘gloves’ of leaves for the handling of thorny plants, modes of drinking, the placing of sticks in among the teeth, and face-wiping. Thus, they showed for orang-utan much the same pattern of cultural transmission among some but not other groups, as Joulian had revealed for chimpanzees. Culture, in this scientific context, is the ability to invent new behaviours that are adopted by the population group and passed on to succeeding generations. If culture

84

Tools and brains: which came first?

on this definition is fairly common among some African and Asian great apes, we may no longer arrogate it to humanity. Either we must change the definition; or use a qualifier, such as hominin culture, or human culture; or we must use another concept or word. Some suggest ‘modern behaviour’, or ‘behavioural modernity’, but that does not quite fit the bill. Baboons have ‘modern’ behaviour, as the South African pioneering ethologist, Eugene Marais, showed a long time ago, just over 100 kilometres from here. Even these terms need a qualifier, such as human behavioural modernity. I see it as a challenging task for this Round Table to resolve this tripartite question of culture, behaviour – and modernity!

The non-recency of modern human activities It has been widely bruited abroad that modern human activities arose quite suddenly a mere 40 000 years ago. This view undoubtedly results from the preoccupation of many Northern Hemisphere scholars with the European archaeological sequence, in which a relatively sudden transition seems to be evident at the dawn of the Upper Palaeolithic. Sally McBrearty and Alison Brooks (2000), in a searching analysis, demonstrate that the Eurocentric idea of a revolution at the onset of modern human cultural behaviour is a model which should not have been applied to Africa. In the African record they speak of ‘The revolution that wasn’t’. With the exception of a fairly abrupt transition from Middle Stone Age to Later Stone Age at the northern and southern margins of Africa, they find that, in the rest of the continent, ‘novel features accrued stepwise’. They adduce evidence that: ‘Distinct elements of the social, economic, and subsistence bases changed at different rates and appeared at different times and places.’ Evidence from the African MSA, they hold, supports ‘the contention that both human anatomy and human behaviour were intermittently transformed from an archaic to a more modern pattern over a period of more than 200 000 years.’ Their work illustrates this point: if cultural humanisation was a long, slow and intermittent process, it becomes most unlikely that the entire human species underwent a simultaneous, punctuated, genetically encoded event, such as the development of modern capacities for language, as Richard Klein avers.

What does the brain have to say? Many scholars have sought to relate the findings and interpretations of archaeology to the evolution of the brain and mind. In effect, such studies venture a statement that the techniques and symbols explicit and implicit in the archaeological record are related, more or less directly, to the cognitive abilities, mental competences and intelligences of evolving hominins, ancient and modern. It has however been evident for some time that it is not only from the archaeological record that we may glean

85

From Tools to Symbols

evidence on the evolution of hominin intelligence. For example, the size and form of endocranial casts of fossil hominins have added grist to the mill of those probing the evolution of hominin cerebration. Let us look first at the absolute size of endocranial capacity in a variety of hominoids (Tables 1 & 2). Table 1 Mean endocranial capacities of pooled male-plus-female samples of extant great apes (cm3) Pan troglodytes

(n = 363)

383,4

Tobias, 1971a,b

Pan paniscus

(n = 11)

343,7

Tobias, 1994

Gorilla gorilla gorilla

(n = 668)

504,6

Tobias, 1971a,b

Pongo pygmaeus

(n = 402)

404,8

Tobias, 1971a,b

For male and female samples considered separately, see Tobias, 1971a, 1971b, 1994. Table 2 Mean endocranial capacities of australopithecine series expressed as percentages of pooled male-plusfemale means of extant great apes Hominin value as percentage of mean for Fossil taxon

Mean

Pan troglodytes

G. gorilla

Pongo pygmaeus

A. afarensis

c. 413,5

?108

?82

?102

A. africanus

451,0

117,6

89,4

111,4

A. boisei

463,3

121

92

114

The mean value for 18 australopithecines of various species, based on the newest individual estimates, is c. 457 cm3. Estimates for several different australopithecine taxa for which endocranial capacity values are available irrespective of sex are as follows: A. afarensis c. 428 cm3 (n = 3) specimens from Hadar A. africanus 451 cm3 (n = 7) specimens from Sterkfontein, Taung and Makapansgat A. robustus 530 cm3 (n = 1) specimen from Swartkrans A. boisei c. 463 cm3 (n = 7) specimens from Olduvai, Koobi Fora, West Lake Turkana, Omo It is interesting to note that, irrespective of sex, the mean value for A. africanus is 17,6 per cent greater than that for chimpanzee (Pan troglodytes), 31,2 per cent greater than that for the bonobo (Pan paniscus), 10,6 per cent smaller than the mean for Gorilla and 11,4 per cent greater than the mean for orang-utan. With the exception of P. paniscus, the samples of great ape values are all very large, while, as is to be expected,

86

Tools and brains: which came first?

the samples for four australopithecine species are small: the best samples are those for A. africanus and A. boisei, for each of which seven values are available. It should be noted that no estimates are yet available for the endocranial capacity of the ‘Little Foot’ cranium of Australopithecus sp., nor for the newly announced partial cranium from the Jacovec Cavern at Sterkfontein (Partridge et al., 2003). Table 2 gives australopithecine/ ape comparisons for three fossil taxa and three ape taxa. The mean value for six specimens of Homo habilis is 640,2 cm3 (Table 3). The sample comprises three putative males (mean 688,0 cm3) and three putative females (592,3 cm3). The male-plus-female mean value for H. habilis is 42 per cent greater than the combined-sex mean for A. africanus, 49,6 per cent greater than the mean for A. afarensis, and 38 per cent greater than for A. boisei. When the value for H. habilis is compared with the value for eighteen australopithecine specimens, it is a clear 40 per cent larger. It is patent that H. habilis had an absolute capacity bigger by two-fifths than the Table 3 Summary of available data for fossil hominins, grouped into various series and species Endocranial capacity values for various fossil hominin series (cm3)1 Coefficient of variation (%)

95% population limits (to nearest cm3) 352–?5002

Taxon

n

Mean

Standard deviation

A. afarensis

3

?413,5

?77,10

?18,65

A. africanus

7

451,0

34,96

7,75

A. robustus

1

530,0







A. boisei

7

463,3

53,66

11,58

332–595

H. habilis

6

640,2

82,23

12,85

429–852

H. erectus erectus

3

7

895,6

93,57

10,45

667–1125

H. erectus erectus

4

363–538

6

929,8

91,67

9,86

694–1165

H. erectus pekinensis

5

1043,0

112,51

10,79

731–1355

H. erectus (Asia and Africa)

15

937,2

135,48

14,46

647–1228

H. sapiens soloensis

5

6

1090,8

75,39

6,91

897–1285

H. sapiens soloensis 6

5

1151,4

99,51

8,64

896–1407

1

In this table no attempt has been made to separate the series into presumptive male and female sub-sets

2

Observed range

3

Based on Tobias’s (1975) estimate, but with the incorporation of the author’s new value for Trinil 2, based on Holloway’s (1975) new value for Sangiran 2.

4

Based on Holloway’s (1981) new values for six Indonesian specimens

5

Based on Weidenreich (1943)

6

Based on Holloway (1980)

87

From Tools to Symbols

values for A. africanus and other australopithecines. Up to the present we do not have corresponding values for other early hominins such as H. rudolfensis, A. anamensis, A. garhi, Ardipithecus ramidus, Orrorin tugenensis or Sahelanthropus tchadensis. Is the difference in capacity between the early hominins and the extant great apes ‘real’ or is it simply the consequence of differing body sizes? Brain size in absolute terms tells us only part of the story. If one looks at the Animal Kingdom in general, one finds that smaller animals have smaller brains, while larger animals have larger brains. Even within modern humankind, it has been shown that tall people have bigger brains than short people. So the increases in brain size in ancient and modern hominins would be biologically meaningful only if we could estimate the body sizes of the ancient hominins. Body size is sometimes expressed as stature, more commonly as body mass or weight. G. Cuvier first introduced the concept of relative brain weight – that is, the weight of the brain expressed as a fraction of the weight of the body. When we are dealing with fossil taxa, body weight is generally calculated from postcranial bones. A number of methods have been devised whereby the investigator may do this. One is based on the size of the vertebrae, especially the cross-sectional areas. These data in Australopithecus and in H. habilis provide a means for computing the probable body weights of individuals in these species. Similarly, the lengths of the limb bones, especially those of the lower limb, provide a basis for estimating body length or stature. It may be noted in passing that when cranial, dental, and postcranial remains are found scattered, as is commonly the case, and where more than one hominin species is known to have coexisted at the time in question, as in the entire known span of H. habilis, there are obvious difficulties in the ascribing of isolated postcranial bones to species. The problem is compounded by the consideration that the species are defined, exclusively or predominantly, on cranial and dental evidence. However, with the accumulation of more and more bony remains, the study of associations on living floors, morphological patterns, and palaeodemographic data, the provisional assignment of postcranial bones to species has reached a degree of consensus for at least some of these skeletal elements. Of the Olduvai postcranial bones, for instance, Day (1977), Campbell (1978), and Tobias (1991) agree in provisionally assigning OH 8 (a foot) and OH 35 (a tibia and fibula) to H. habilis, while the first and third authors agree similarly in respect of OH 48 (a clavicle) and OH 49 (a partial radial shaft). Similar provisional allocations have been made for other East African postcranial bones. On the basis of such provisional identifications for other hominin taxa, strengthened by those instances from Ethiopia, Kenya, Tanzania and Sterkfontein where partial skeletons, some including cranial parts, have been found, numerous attempts have been made to estimate body weight in the fossil taxa under consideration

88

Tools and brains: which came first?

– for example by Genet-Varcin (1966, 1969), Lovejoy and Heiple (1970), Robinson (1972), Wolpoff (1973), McHenry (1974, 1982, 1991), Pilbeam and Gould (1974), Krantz (1977), Steudel (1980). Other investigators have used body height (stature) in the computation of relative brain size or relative cranial capacity (see, for example, Grüsser & Weiss, 1985). We shall here use one recent set of estimates of body weight of H. habilis and of other taxa, namely those of McHenry (1975, 1976, 1982, 1984, 1994). At the same time even those critical estimates may be seen as not necessarily the last word on body weight estimates. At least they provide a provisional range of estimates of body size to which to relate the endocranial capacity values. A second important consideration is the immense literature devoted recently to ‘scaling’ – that is, the structural and functional consequences of differences in size (or scale) among organisms of more or less similar design (Jungers, 1984, 1985). Studies by Martin, Armstrong, and Hofman have stressed that there are metabolic constraints in brain enlargement (Martin, 1980, 1981, 1982; Armstrong, 1981, 1983, 1984; Hofman, 1982); others have stressed the problem at which systematic level comparisons of brain scaling are most meaningful (Harvey & Mace, 1982; Holloway & Post, 1982). Lande (1985) has made an important study on the quantitative genetic aspects of the problem of brain size/body size. He observes that ‘genetic uncoupling’ of brain and body sizes in primates would have facilitated encephalisation in primates, because natural selection for larger brain size would then not necessarily have favoured correlated uneconomical increased body sizes: ‘If the genetic correlation between brain and body size within populations in the human lineage was … low as suggested by data on primates, hominins would have been enabled to rapidly increase brain size in response to selection for more complex behaviour without the cost of antagonistic selection to prevent the evolution of gigantism’ (Lande, 1985: 30). A number of different techniques have been proposed to determine the degree of encephalisation, when body size is taken into account. EQ stands for Jerison’s Encephalisation Quotient, which is the ratio of actual brain size to expected brain size (a kind of average for living mammals that takes body size into account) (Jerison, 1970, 1973). As obtained by Jerison, expected brain size is derived from body weight by the formula 0,12 (body weight)0,667. The scaling coefficient of 0,667 has been claimed to fit the relationship between brain weight and body weight in a large sample of living mammals (Jerison, 1973; Gould, 1975; McHenry, 1982). On a much larger sample of species, however, Martin (1982) has obtained an exponent closer to 0,75, rather than 0,67. A scaling coefficient of about 0,75 has been found to apply to primates (Bauchot & Stephan, 1969), though within the primates these authors report coefficients ranging in various groups from 0,58 to 0,80 in round figures, as cited by Holloway and Post (1982).

89

From Tools to Symbols

CC is Hemmer’s Constant of Cephalisation and is derived by dividing the endocranial capacity by the body weight scaled to the power of 0,23. Table 4 gives results for two estimates of relative brain size. The values obtained by the two equations used here for EQ and CC confirm that (a) the various australopithecine species were slightly more encephalised than the chimpanzee; and (b) H. habilis was clearly much more encephalised than any of the australopithecine series and represented a major step, indeed the first such, in brain aggrandisement. Its values reveal that it had attained some 50 per cent of the H. sapiens degree of encephalisation. More marked encephalisation followed from H. habilis to H. erectus, the latter species reaching some 70–80 per cent of the degree of encephalisation shown by H. sapiens. These coefficients, like Jerison’s Nc, reveal that H. habilis is appreciably advanced in its degree of encephalisation as compared with the Hadar hominins and with A. africanus. Since the estimated body size is built into the formulae for EQ and CC, it is clear that the larger endocranial capacity of H. habilis is not to be explained solely as the result of its larger estimated body mass or, for that matter, of a higher estimated stature (Grüsser et al., cited by Grüsser & Weiss, 1985). It clearly represents an advance in encephalisation over the small-brained hominins, the australopithecines. Table 4 Mean endocranial capacity, estimated body mass, and coefficients of encephalisation for a series of hominoids (modified after McHenry, 1982). Mean endocranial capacity (cm3)

Estimated body mass (kg)

EQ (actual value)

EQ (as % of H. sapiens value)

CC (actual value)

CC (as % of H. sapiens value)

P. troglodytes

395,0

45,0

2,6

34

33,6

31

A. afarensis

413,5

37,1

3,1

41

36,8

34

A. africanus

451,0

35,3

3,5

46

40,6

37

A. robustus

530,0

44,4

3,5

46

45,2

42

H. habilis

640,2

48,0

4,0

53

53,6

49

H. e. erectus (PVT)

895,6

53,0

5,3

70

73,4

67

H. e. erectus (RLH)

929,8

53,0

5,5

72

76,2

70

H. erectus (Asia and Africa)

937,2

53,0

5,5

72

76,8

71

H. e. pekinensis

1 043,0

53,0

6,1

80

85,4

78

H. sapiens

1 350,0

57,0

7,6

100

108,8

100

Note: EQ, encephalisation quotient; CC, constant of cephalisation; PVT, the author’s estimate for H. erectus erectus; RLH, Holloway’s (1981b) estimate for H. erectus erectus. McHenry’s value of 53,0 kg for H. erectus has been used in the above table for all of the subsets of H. erectus. Although mass is given in kilograms in the table, it is expressed in grams for the calculations.

90

Tools and brains: which came first?

The data presented here on relative brain size show that while the australopithecines were encephalised slightly more than the chimpanzee, H. habilis had unequivocally begun the remarkably ‘uncoupled’ or disproportionate enlargement of the brain that is a critical hallmark of humankind.

Natural selection and encephalisation It was fashionable at one time to dismiss studies on brain size, or more precisely, endocranial capacity, as a rather crude measure of the evolution of the hominin brain. Recent studies have related the increase of hominin brain size to a variety of factors, such as life-cycles, demographic profiles, ecology, metabolic and nutritional constraints, especially the availability of DHA (docosahexaenoic acid), brain-cooling, and the implications of brain enlargement for blood-supply and venous drainage. By these means, the search for context and causality has deepened brain-size studies and given them new conceptual worth. Whatever selective agencies were operating, they must have continued influencing the hominin brain throughout the assumption of a bewildering array of new lifestyles and environments. One suggestion that was offered to account for the sustained duration of encephalisation was an autocatalytic positive feedback system (cf. Mayr, 1963; Bielicki, 1964, 1969; Tobias, 1971 & 1981). The questions one needs to answer are these: does the occurrence of a greater relative brain size in humans connote advantages that might have been favoured by natural selection? If so, what manner of advantage did the larger brain size confer? On Lashley’s (1949) and Jerison’s (1963, 1970, 1973) approach, ‘improved adaptive capacities’ provide the key to the selective advantage conferred by increased encephalisation. In Jerison’s (1991) newer work this view is maintained: he places hominin encephalisation in the context of a specific environmental niche, namely the shrinking of the African wet forest and spread of the savanna region at the forest’s edge. These conditions accompanied the emergence of the hominins over 6 Mya. In this setting, he suggests, adaptive changes occurred in the nervous systems of the ancestors of the hominins. We may paraphrase ‘improved adaptive capacities’ as increased adaptability or greater evolutionary flexibility. Is this the advantage of greater encephalisation? The problem is more exacting than it might seem. Long ago, Mather (1943) showed that adaptedness and adaptability in evolution were inversely proportional to each other. In other words the more highly adapted an animal is to its present environment, the less evolutionary plasticity it has retained for adaptation to a new environment should conditions change. If the attainment by humankind of maximal encephalisation implies that humans have attained maximal adaptability, we might expect that, on

91

From Tools to Symbols

Mather’s analysis, present adaptedness had been sacrificed. It is doubtful whether this deduction would stand up to close scrutiny. Moreover, encephalisation is present in a number of other mammalian lineages, for example, the cetaceans, Old World and some New World monkeys, the tree shrews or Tupaioidea, and certain rodents: the maximally encephalised end-products of these evolutionary lineages certainly appear to be highly adapted in their respective econiches. The author suggests that increased encephalisation provided a means by which the organisms concerned could rise above the constraints suggested by Mather’s paradox. It is hypothesised here that the relative enlargement of the hominin brain was a mechanism by which enhanced adaptability might be furnished and on which natural selection could go to work, while adaptation – that is, concurrent adaptedness – was not sacrificed and could even have been improved. In a word, it is proposed that encephalisation enhanced adaptability while permitting organisms to maintain adaptation, or even to manifest more efficient adaptation. This general proposition should have applied to all mammalian lineages characterised by progressive encephalisation. In the hominin lineage it has perhaps attained its pinnacle of evolution, as reflected by modern humanity’s remarkable degree of encephalisation. Along this lineage, the particular property ‘secreted’ by the expanding brain was the cognitive faculty, of such quality and degree as to generate culture. Probably no more puissant force has yet appeared on earth in its capacity to potentiate adaptation and to widen dramatically the evolutionary flexibility of its possessors. In the hominin line, particularly, culture may provide the means by which Mather’s paradox has been addressed and surmounted. In this situation, adaptability and adaptedness are not inversely proportional.

The surface of the brain The patterns of convolutions and sulci on the surface of the cerebrum differ appreciably between living apes and humans. It is not necessary here to elaborate on the detailed differences. Crucial is the question: when did the human pattern emerge? Endocranial casts, whether natural or artificial, permit one to approach some answers. In Australopithecus, with one or two exceptions, the sulcal pattern is for the most part ape-like. In H. habilis, it is human-like (Falk, 1983; Tobias, 1975, 1987); and so it is with all later members of the genus Homo, as Grimaud-Hervé (1991) has shown. It is a striking fact that the brains of H. habilis were the first to show marked enlargement, absolute and relative, but also they were the first to manifest a human-like structural pattern of the brain.

92

Tools and brains: which came first?

If there was a correlation between brain evolution and cultural behaviour, as manifested by the succession of archaeological cultural stages, and if the emergence of language was correlated with culture, it is meet to inquire with which cultural stage or stages the emergence of spoken, cognitive language was correlated. On this question of the emergence of spoken language, archaeologists tend to favour recency, while biological anthropologists and anatomists incline to favour an earlier emergence, but there are many exceptions to both generalisations. It seems to me that there would be an element of arbitrariness if one chose a single cultural phase or industry. Even the Oldowan involves complex visuo-spatial concepts.

The speech areas and language: a personal odyssey The various functional modalities that are represented in the cerebral cortex are localised in specific parts of the cerebral hemispheres, the left or the right or both. Each such area constitutes a mass of distinctive nerve-cells, a cyto-architectonic area. Those cortical areas that lie beneath the outer surface of a hemisphere are commonly related to a specific configuration of overlying superficial convolutions and sulci. The cellular masses or nuclei underlie a ‘cap’ or bulge on the surface. Two such areas in the brain of modern humans are those that have to do with spoken language: Broca’s area above and behind the eye, usually on the left alone, and Wernicke’s area, above and behind the ear, usually only on the left side (Fig. 1). A Broca’s cap and a Wernicke’s cap overlie these two cell masses. As these areas are on the outer surface of the hemisphere, each cap reflects itself by a matching hollow or valley on the inner surface of the cranial vault. Thus, they are easily detectable on endocranial casts. While the Broca and Wernicke caps occur on the surface of the brain (and endocast) of modern humans, they are lacking on the brains of apes and, with one or two exceptions, of australopithecines. Apes do not have the capacity for spoken language and it was assumed neither did australopithecines. In late 1972 and early 1973, I was preparing a symposium contribution to be presented at the XIth International Congress of Anthropological and Ethnological Sciences at Chicago in September 1973. The Symposium was organised jointly by Russell Tuttle, Becky Sigmon and myself. In the course of my preparations, I reexamined the artificial endocasts of the Olduvai H. habilis which had been made in Nairobi by Ron Clarke and myself. I was amazed to find signs of Broca’s and Wernicke’s caps in these endocasts, especially that of the female H. habilis Olduvai hominid 24 (‘Twiggy’) (Fig. 1). I presented an account of my observations in Chicago, early in September 1973. The significance of my discovery was overlooked by those at the Chicago meeting. Not only our students but, it seems, scientists are fearful of the brain. (It was in an attempt

93

From Tools to Symbols

to obviate this craven attitude that Arnold and I, in our textbook Man’s Anatomy (1963–1990), put the brain back into the cadaveric body, hoping thereby to help our students see it as a part of the body, not to be separated on a pedestal!) The fear of Figure 1 Left: skull of Homo habilis (OH 24). Centre-right and right: location of the language areas in the brain.

the brain persists today – even among some distinguished colleagues who omit all but the most passing reference to it from their books and monographs on fossil man. Re-reading the record of the discussion following the Chicago symposium on brain evolution, I find that in over eleven pages of discussion no mention is made of the speech areas in the brain of H. habilis – although points on the evolution of the brain are addressed by discussants Milford Wolpoff, Philip Lieberman, Russel Tuttle, Fred Szalay, Bob Eckhardt, George Sacher, Morris Goodman and Bill Howells. My Chicago article and the Discussion in question were published in 1975. Up to February 1979, I made no claim that H. habilis used its language centres for the purpose of speaking. I assumed, with some others, that the first speaking primates were H. erectus. Then, between March and September 1979, I underwent a change of mind. This was occasioned by two sets of factors, archaeological and contextual. Archaeologically, the works of Glyn Isaac, Mary Leakey, Sue Parker and Kathleen Gibson adumbrated a concept world of the makers of the Oldowan culture, most probably H. habilis, that was far more complex than had previously been conceived. There were ‘living floors’ that Mary Leakey had excavated at Olduvai; Isaac’s concept that some of these were ‘home bases’; a set of new adaptive strategies, including toolmaking, transport of food and materials, eating of meat and sharing of food. Only recently, Blumenschine et al. (2003) have supplemented Isaac’s hypothesis with a new eco-selecting propensity, by which some of the Olduvai H. habilis groups made ‘irregular, seasonal forays to the western basin streams from the ecologically more

94

Tools and brains: which came first?

productive southeastern basin’. One should not forget Mary Leakey’s famous circle of stone walling on one of her living floors of which the colleagues were recently reminded (Tobias, 2003a). The behavioural flexibility and the range of concepts implicit in such a lifestyle could, it seemed to me, be transmitted down the generations only by spoken language. Contextually, I was struck by the near-contemporaneity of the following: (a) the earliest appearance of the modern human cerebral sulcal pattern; (b) the first appearance of Broca’s and Wernicke’s caps; (c) the first appreciable enlargement of the brain; (d) the first appearance of deliberately fashioned stone tools to a set and consistent pattern; (e) signs that about 2,6–2,5 Mya Africa underwent drying, cooling, and elevation, with concomitant faunal and floral changes. When the cultural and social aspects of the inferred lifestyle of H. habilis were added to the testimony of the endocasts and of those other synchronic, morphological and ecological happenings, I proposed the hypothesis that H. habilis not only possessed the neural basis, but used it, for spoken language. I recognised that such language might initially have been rudimentary in phonetic repertoire, phonemic range, syntactic versatility and cognitive substance. In September 1979, I posted an article to La Recherche in Paris. It appeared in March 1980: it was the centenary of the death of Broca! When my French article appeared, it was the first published version of my claim that H. habilis had a mastery of spoken language. My hypothesis was developed further over the ensuing three years, at the Royal Society (London), in the Abbie Memorial Lecture (Adelaide), at the International Anatomical Congress (Mexico City), at the Congress of Human Genetics (Jerusalem), at the Pontifical Academy of Sciences (Vatican City), and in a symposium at the University of Alberta (Edmonton). By 1983, the claim had appeared in print seven times, but it was not until then that any support was forthcoming. At first, my view was unsupported by any colleagues. It looked as though the claim was turning out to be another example of a premature discovery in the sense of Gunter Stent (1971, 1972), much as the first announcement of A. africanus by R.A. Dart (1925) and of H. habilis by L.S.B. Leakey, Tobias and J.R. Napier (1964) had been. The first supporter was Dean Falk (1983) when her study of a Koobi Fora H. habilis endocast led her to the same conclusion as I had reached in 1973. Indeed, she claimed later that ‘Phillip Tobias and I independently concluded that [Homo habilis] may have been capable of some rudimentary form of language’ (Falk 1992, p. 145). The support was pleasing except that the dates of my and her claims were years apart! The idea was next adopted by Sir John Eccles and supported at a Study Week on

95

From Tools to Symbols

the Design and Operation of the Brain, held at the Pontifical Academy of Sciences in October 1988 (Eccles & Creutzfeldt, 1990). In his last book, Evolution of the Brain: Creation of the Self, Eccles (1989) accepted that Homo habilis was capable of spoken language. More recently a rising tide of acceptance became evident at the hands of Jeff Laitman, Peter Andrews, Christopher Stringer, Terry Deacon, Wendy Wilkins and Jennie Wakefield and even Philip Lieberman, a long-time supporter of the recency of acquisition of spoken language. Lieberman shifted the goalposts somewhat when he declared, ‘The evolution of fully modern speech occurred fairly recently’ (note the qualification fully modern speech), but he conceded ‘… though Homo habilis may have had some degree of speech and syntactic ability, it was not fully modern’. His shift of the goalposts distinguished between ‘fully modern’ and ‘not fully modern’ language. Indeed, it seems to me that there were probably not just two stages of linguistic evolution. Rather there might have been a succession of phases of increasing complexity of the conceptual and syntactic modes of languages, just as there might have been a widening with time of the range of phonemes in the repertoire of language sounds. Clearly, however, the earlier forms and the later, more evolved modes of language would all qualify to be classified as human spoken language.

Conclusion It is a breathtaking aspect of the finds of Louis and Mary Leakey that specimens assignable to Homo should have existed as long ago as 2 Mya or more, for that is the approximate age of the earliest fossils of H. habilis (and herein I include H. ergaster cf. Blumenschine et al., 2003; Tobias, 2003a; Falk, 1983) at Omo, East Turkana (Koobi Fora), Olduvai and Sterkfontein Member 5. This revelation of the high antiquity of the genus Homo is one of the greatest contributions of the Leakey family. Add to that the fact that Homo did not begin with the big-brained H. erectus, as was widely believed before 1964, but with a more modest antecedent, H. habilis, which held its own in the face of competition from its robust and hyper-robust australopithecine contemporaries. The distinguished evolutionist, Theodosius Dobzhansky, sometimes dubbed ‘the Charles Darwin of the twentieth century’, in his posthumous Raymond Dart Lecture (Dobzhansky & Ayala, 1977) recognised two great steps forward in the development of life. His so-called First Transcendence referred to the Origin of Life itself. The Second Transcendence was the coming of Man with his futuristic survival equipment. Dobzhansky saw the Second Transcendence as marked by the advent of the Hominidae (or Hominini in today’s preferred parlance). I have proposed that, instead of the coming of the hominins, it was Homo habilis, the small, meek hominin, that announced the Second Transcendence to the world (Tobias, 1980). Neither the hominins nor the world would ever be the same again.

96

Tools and brains: which came first?

Time does not allow me to explore the interesting possibility – first advanced by Schepers (1946), pressed more recently by Eccles (1989) and flirted with by myself (Tobias, 1990, 1995), at least in respect of ‘advanced A. africanus’ – that there might have been spoken language before the emergence of H. habilis. If the ability for cognitive spoken language were confined to Homo, the neurological basis of spoken language might be regarded as an autapomorphy (uniquely derived trait) of the genus Homo. However, if there were evidence that any australopithecines (of whatever genus, subgenus or species) had the capacity for spoken language, even although rudimentary, this view would have to be changed. At present the only direct evidence bearing on this question is the presence of a modest Broca’s cap in some A. africanus endocasts (Schepers, 1946); but Falk (1983) has shown that the gyral and sulcal pattern of this area in A. africanus is not human-like but ape-like. It remains conceivable that derived A. africanus was the first hominin to have been capable of a proemial spoken language, before H. habilis came to depend on language for survival (Tobias, 2003b). It is quite likely that some of the many archaeologists at this meeting adhere to the concept of the late or recent origin of language. What I have presented and urged here is the fact that there is an alternative view! The organisers of this meeting, and the participants, will not, I am sure, say the last word on the first word. However, I have no doubt that your deliberations will throw much light on the humanising of the human hominins!

Acknowledgements The valued help of my personal assistant, Heather White, and of Peter Faugust and Terry Borain is warmly appreciated. My work continues to be generously supported by the Ford Foundation, the PAST Fund, the National Research Foundation, the Embassy of France in South Africa, and the Department of Arts, Culture, Science and Technology. The University of the Witwatersrand remains the host, patron and advocate of the research endeavours of the Sterkfontein Research Unit and of myself personally, as it has done for 58 years.

97

From Tools to Symbols

References Armstrong, E. (1981). A look at relative brain size in mammals. Neuroscience Letters 34, 101– 104. Armstrong, E. (1983). Relative brain size and metabolism in mammals. Science 220, 1302– 1304. Armstrong, E. (1984). Allometric considerations of the adult mammalian brain with special emphasis on primates. In (W.L. Jungers, Ed.) Size and Scaling in Primate Biology, pp. 115– 147. New York: Plenum. Bauchot, R. & Stephan, H. (1969). Encephalisation et niveau evolutif chez les simians. Mammalia 33, 225–275. Bielicki, T. (1964). Evolution of the intensity of feedbacks between physical and cultural evolution from man’s emergence to present times. UNESCO Expert Meeting on Biological Aspects of Race, Moscow, August 1964, pp. 1–3. Bielicki, T. (1969). Deviation-amplifying cybernetic systems and hominid evolution. Materialy iPrace Antropologiczne 77, 57–60. Blumenschine, R.J., Peters, C.R., Masao, F.T., Clarke, R.J., Deino, A.L., Hay, R.L., Swisher, C.C., Stanistreet, I.G., Ashley, G.M., McHenry, L.J., Cushing, A.E., Sikes, N.E., Van der Merwe, N.J., Tactikos, J.C., Deocampo, D.M., Njau, J.K. & Ebert, J.I. (2003). Late Pliocene Homo and hominid land use from western Olduvai Gorge, Tanzania. Science 299, 1217–1221. Campbell, B.G. (Ed.) (1978). W.E. LeGros Clark: The Fossil Evidence for Human Evolution, 3rd ed. Chicago, London: University of Chicago Press. Dart, R.A. (1925). Australopithecus africanus: The man-ape of South Africa. Nature 115, 195– 199. Day, M.H. (1977). Guide to Fossil Man, 3rd ed. London: Cassell. Dobzhansky, T. & Ayala, F.J. (1977). Humankind: A Project of Evolutionary Transcendence, pp. 1–15. Johannesburg: Institute for the Study of Mankind in Africa and University of the Witwatersrand Press. Eccles, J.C. (1989). Evolution of the Brain, Creation of the Self. New York: Routledge Eccles, J.C. & Creutzfeldt, O. (Eds.) (1990). The Principles of Design and Operation of the Brain. Vatican City: Pontifical Academy of Sciences. Falk, D. (1983). Cerebral cortices of East African early hominids. Science 221, 1072–1074. Falk, D. (1992). Braindance. New York: Henry Holt. Genet-Varcin, E. (1966). Conjectures sur l’allure générale des Australopithèques. Bulletin de la Société Prehistorique Francaise 63, 106–107. Genet-Varcin, E. (1969). Structure et comportement des Australopithèques d’après certains os post-craniens. Annales de Paléontologie (Vertebrata) 55, 137–148. Gould, S.J. (1975). Allometry in primates, with emphasis on scaling and the evolution of the brain. In (F.S. Szalay, Ed.) Approaches to Primate Paleobiology, pp. 244–292. Basel: Karger.

98

Tools and brains: which came first? Grimaud-Hervé, D. (1991). L’évolution de l’encéphale chez l’Homo erectus et l’Homo sapiens. Paris: Université de Provence. Grüsser, O.-J. & Weiss, L.-R. (1985). Quantitative models on phylogenetic growth of the hominid brain. In (P.V. Tobias, Ed) Hominid Evolution: Past, Present and Future, pp. 457–464. New York: Alan R. Liss. Harvey, P.H. & Mace, G.M. (1982). Comparisons between taxa and adaptive trends: problems of methodology. In (King’s College Research Group, Eds.) Current Problems in Sociobiology, pp. 343–361. Cambridge: Cambridge University Press. Hofman, M.A (1982). Encephalisation in mammals in relation to the size of the cerebral cortex. Brain, Behavior and Evolution 20, 24–96. Holloway, R.L. (1975). Early hominid endocasts: volumes, morphology and significance for hominid evolution. In (R.H. Tuttle, Ed.) Primate Functional Morphology and Evolution, pp. 393–416. The Hague: Mouton. Holloway, R.L. (1980). Indonesian ‘Solo’ (Ngandong) endocranial reconstructions: some preliminary observations and comparisons with Neandertal and Homo erectus groups. American Journal of Physical Anthropology 53, 285–295. Holloway, R.L. (1981). The Indonesian Homo erectus brain endocasts revisited. American Journal of Physical Anthropology 55, 503–521. Holloway, R.L. & Post, D.G. (1982). The relativity of relative brain measures and hominid mosaic evolution. In (E. Armstrong & D. Falk, Eds) Primate Brain Evolution: Methods and Concepts, pp. 57–76. New York, London: Plenum. Jerison, H.J. (1963). Interpreting the evolution of the brain. Human Biology 35, 263–291. Jerison, H.J. (1970). Gross brain indices and the analysis of fossil endocasts. The Primate Brain 1, 225–244. Jerison, H.J. (1973). Evolution of the Brain and Intelligence. New York, London: Academic. Jerison, H.J. (1991). Brain Size and the Evolution of Mind. New York: American Museum of Natural History. Joulian, F. (1966). Comparing chimpanzee and early hominids. In (P.A. Mellars & K. Gibson, Eds) Modelling the Early Human Mind, pp. 173–189. Cambridge: McDonald Institute for Archaeological Research. Jungers, W.L. (1984). Aspects of size and scaling in primate biology with special reference to the locomotor skeleton. Yearbook of Physical Anthropology 27, 73–97. Jungers, W.L. (Ed.) (1985). Size and Scaling in Primate Biology. New York: Plenum. Kenyon, K.W. (1969). The Sea Otter in the Eastern Pacific Ocean. Washington: United States Government Printing Office. Khroustov, H.F. (1964). Formation and highest frontier of the implemental activity of anthropoids. VII International Congress of Anthropological and Ethnological Sciences, Moscow, August 1964. Köhler, W. (1924). The Mentality of Apes (English Edition, 1959). New York: Vintage Books.

99

From Tools to Symbols Kortlandt, A. & Kooij, M. (1963). Protohominid behaviour in primates. Symposia of the Zoological Society of India, No. 10, 61–87. Krantz, G.S. (1977). A revision of australopithecine body sizes. Evolutionary Theory 2, 65–94. Lancaster, J.B. (1968). On the evolution of tool-using behavior. American Anthropologist 70, 56–66. Lande, R. (1985). Genetic and evolutionary aspects of allometry. In (W.L. Jungers, Ed.) Size and Scaling in Primate Biology, pp. 21–32. New York: Plenum. Lashley, K.S. (1949). Persistent problems in the evolution of mind. Quarterly Review of Biology 24, 28–42. Leakey, L.S.B., Tobias, P.V. & Napier, J.R. (1964). A new species of the genus Homo from Olduvai Gorge. Nature 202, 7–9. Lovejoy, C.O. & Heiple, K.F. (1970). A reconstruction of the femur of Australopithecus africanus. American Journal of Physical Anthropology 22, 33–40. Martin, R.D. (1980). Adaptation and body size in primates. Zeitschrift für Morphologie und Anthropologie 71, 115–124. Martin, R.D. (1981). Relative brain size and basal metabolic rate in terrestrial vertebrates. Nature 293, 57–60. Martin, R.D. (1982). Allometric approaches to the evolution of the primate nervous system. In (E. Armstrong & D. Falk, Eds) Primate Brain Evolution: Methods and Concepts, pp. 39–56. New York: Plenum. Mather, K. (1943). Polygenic inheritance and natural selection. Biological Reviews 18, 32–64. Mayr, E. (1963). Animal Species and Evolution. Cambridge: Harvard University Press. McBrearty, S. & Brooks, A. (2000). The revolution that wasn’t: a new interpretation of the origin of modern human behavior. Journal of Human Evolution 39: 453–563. McHenry, H.M. (1974). How large were the australopithecines? American Journal of Physical Anthropology 40: 329–340. McHenry, H.M. (1975). Fossil hominid body weight and brain size. Nature 254, 686–688. McHenry, H.M. (1976). Early hominid body weight and encephalisation. American Journal of Physical Anthropology 45, 77–83. McHenry, H.M. (1982). The pattern of human evolution; studies on bipedalism, mastication and encephalisation. Annual Reviews of Anthropology 11, 151-173. McHenry, H.M. (1984). Relative cheek-tooth size in Australopithecus. American Journal of Physical Anthropology, 64, 297-306. McHenry, H.M. (1991). Petite bodies of the ‘robust’ australopithecines. American Journal of Physical Anthropology 86, 445-454. McHenry, H. (1994). Early hominid postcrania: phylogeny and function. In (R .S. Corruccini & R.L. Ciochon, Eds) Integrative Paths to the Past: Paleoanthropological Advances in Honor of F. Clark Howell, pp. 251–268. Englewood Cliffs, New Jersey: Prentice Hall.

100

Tools and brains: which came first? Oakley, K.P. (1949). Man the Tool-Maker. London: British Museum (Natural History). Partridge, T.C., Granger, D.E., Caffee, M.W. & Clarke, R.J. (2003). Lower Pliocene hominid remains from Sterkfontein. Science, 300, 607–612. Pilbeam, D. & Gould, S.J. (1974). Size and scaling in human evolution. Science 186, 892–901. Robinson, J.T. (1972). Early Hominid Posture and Locomotion. Chicago: University of Chicago Press. Schepers, G.W.H. (1946). The endocranial casts of the South African apemen. In (R. Broom & G.W.H. Schepers, Eds) The South African Fossil Apemen: The Australopithecinae. Transvaal Museum Memoirs 2, 153–272. Stent, G.S. (1971). Molecular Genetics: An Introductory Narrative. San Francisco: W.H. Freeman. Stent, G.S. (1972). Prematurity and uniqueness in scientific discovery. Scientific American 227, 84–93 Steudel, K. (1980). New estimates of early hominid body size. American Journal of Physical Anthropology 52, 63–70. Tobias, P.V. (1965). Australopithecus, Homo habilis, tool-using and tool-making. South African Archaeological Bulletin 20, 167–192. Tobias, P.V. (1971a). The Brain in Hominid Evolution. New York, London: Columbia University Press. Tobias, P.V. (1971b). The distribution of cranial capacity values among living hominoids. Proceedings of the 3rd International Congress of Primatology, Zurich 1970, 1, 18–35. Tobias, P.V. (1975). Brain evolution in the Hominoidea. In (R.H. Tuttle, Ed.) Primate Functional Morphology and Evolution, pp. 353–392. The Hague: Mouton. Tobias, P.V. (1980). L’evolution du cerveau humain. La Recherche 11, 282–292. Tobias, P.V. (1981). The Evolution of the Human Brain, Intellect and Spirit. Adelaide: University of Adelaide Press. Tobias, P.V. (1987). The brain of Homo habilis: a new level of organization in cerebral evolution. Journal of Human Evolution 16, 741–761. Tobias, P.V. (1990). Some critical steps in the evolution of the hominid brain. Pontifical Academy of Sciences, Scripta Varia, 78, 1–16. Tobias, P.V. (1991). Olduvai Gorge, Vols. 4A and 4B: The Skulls, Endocasts and Teeth of Homo habilis. Cambridge: Cambridge University Press. Tobias, P.V. (1994). The craniocerebral interface in early hominids. In (R.S. Corruccini & R.L. Ciochon, Eds) Integrative Paths to the Past: Paleoanthropological Advances in Honor of F. Clark Howell, pp. 185–203. Englewood Cliffs, New Jersey: Prentice Hall. Tobias, P.V. (1995). The Communication of the Dead: Earliest Vestiges of the Origin of Articulate Language. 17th Kroon Lecture. Amsterdam: Stichting Nederlands Museum Voor Anthropologie en Praehistorie.

101

From Tools to Symbols Tobias, P.V. (2003a). Encore Olduvai. Science 299, 1193–1194. Tobias, P.V. (2003b). Twenty questions about human evolution. Keynote Address to XV ICAES, Florence, Italy. In: (B. Chiarelli, Ed.) Humankind/Nature Interaction: Past, Present and Future, pp. 9–64. Van Schaik, C.P., Ancrenaz, M., Borgen, G., Galdikas, B., Knott, C.D., Singleton, I., Susuki, A., Utami, S.R. & Merrill, M. (2003). Orangutan cultures and the evolution of material culture. Science 299, 102–105. Weidenreich, F. (1943). The skull of Sinanthropus pekinensis: a comparative study of a primitive hominid. Palaeontologia Sinica 10, 1–298. Wolpoff, M.H. (1973). Posterior tooth size, body size and diet in South African gracile australopithecines. American Journal of Physical Anthropology 39, 375–393.

102

Environmental changes and hominid evolution: what the vegetation tells us Marion K. Bamford Bernard Price Institute for Palaeontological Research, School of Geosciences, University of the Witwatersrand, Private Bag 3, WITS 2050, Johannesburg, South Africa

Abstract Vegetation and climate influence the lifestyles and behaviour of the fauna in any region, so much so that animals with certain requirements can only live in suitable areas. Modern humankind can manipulate the vegetation to suit his needs but early humankind was at the mercy of the elements. Before we can postulate how and when evolutionary changes, modifications or adaptations occurred in the hominids, we need to know what the vegetation and climate were like. Comparative studies of fossil faunas with modern faunal distributions have been used to predict the palaeoclimate and vegetation. Pollen, phytolith and light isotope studies are also used, but the most direct method is to look at the actual fossil plants. Unfortunately these are not often preserved with the faunal remains but where they do occur an interesting picture emerges. Four case studies will be presented here, from East and South Africa, from short periods within a long time range. From Laetoli (Tanzania) fossil woods have been collected, between 4,3 and 3,8 Ma (million years), which show a complex diversity of species. Many seeds have been preserved just below the Foot Print Tuff, dated at 3,56 Ma. Detailed research has just begun on these plant remains. A multi-disciplinary project at Olduvai Gorge, Tanzania, has revealed many fossil plants: wood of Guibourtia coleosperma, sedges, grasses and other woody plants still to be identified. These plants come from upper Bed I and lower Bed II, approximately 1,8–1,7 Ma. Together with the tephrostratigraphy and sedimentology, these show that there have been dramatic and frequent fluctuations in the vegetation and climate. The saline lake has expanded and contracted, and faulting has also changed the drainage pattern, so the fluvial systems and wetlands have shifted a number of times. Fossil woods from the Sterkfontein Cave site in South Africa show that there was gallery forest during Member 4 times. One piece of wood has survived from the Florisbad site in South Africa. This is much younger, 259–125 Ka (thousand years), and appears to have been worked into a tool. The wood, Zanthoxylum chalybeum, is not a local one, and shows that there has been a climatic shift and the area is much drier today. The data and interpretation by other researchers and from other sites are discussed here. These are fragments of information but gradually the vegetation will be reconstructed and the palaeoanthropologists and archaeologists will be able to use the information for their evolutionary models.

103

From Tools to Symbols

Résumé La végétation et le climat déterminent en large mesure la biogéographie des faunes vivant sur notre planète. L’homme moderne peut manipuler la végétation pour convenir à ses besoins mais cela n’était pas le cas pour les premiers hominidés. La reconstitution des paléoenvironnements est donc nécessaire pour comprendre le rôle joué par des changements climatiques sur des processus évolutifs ou adaptatifs. L’étude comparative de faunes fossiles et de la distribution des faunes actuelles ont permis de reconstituer le paléoclimat et la végétation. Le pollen, les phytolithes et les analyses isotopiques offrent également des informations précieuses, mais la méthode la plus directe reste l’examen des restes de plantes fossiles. Malheureusement, celles-ci ne sont pas toujours conservées avec les restes fauniques mais lorsque c’est le cas, une image intéressante apparaît. Quatre études de cas sur l’Afrique orientale et l’Afrique du Sud seront présentées ici. Elles concernent des périodes courtes. . Les bois fossiles découverts à Laetoli (Tanzanie), vieux de 4,3 à 3,8 Ma, révèlent une diversité complexe d’espèces. Les nombreuses graines préservées juste au-dessous du Foot Print Tuff, daté à 3,56 Ma font l’objet d’une recherche en cours. Un projet pluridisciplinaire à Olduvai Gorge en Tanzanie a permis de trouver de nombreux restes de plantes: du bois de Guibourtia coleosperma, des cypéracées, de graminées et autres végétaux ligneux encore à identifier. Ces plantes viennent du sommet de l’Ensemble I (upper Bed I) et de la zone inférieure de l’Ensemble II (lower Bed II) , datés approximativement de 1,8–1,7 Ma. Un fois mis en relation avec les données de la téphrostratigraphie et de la sédimentologie ces résultats montrent qu’il y a eu des fluctuations spectaculaires et fréquentes dans la végétation et le climat. Le lac salé s’est étendu puis contracté et la formation de failles a aussi transformé la trace du réseau hydrographique, de manière que le système fluvial et l’emplacement des terrains marécageux ont changé à plusieurs reprises. Les bois fossiles de la Grotte de Sterkfontein en Afrique du Sud montrent qu’il y avait des forêts-galeries lors de la l’accumulation de l’ensemble 4 (Member 4). Un fragment de bois a survécu dans le site de Florisbad en Afrique du Sud. Il est plus récent, 259–125 Ka, et semble avoir été façonné pour être utilisé comme un outil. L’espèce utilisée, Zanthoxylum chalybeum, n’est pas présente actuellement dans la région révèle qu’il y a eu un changement climatique et que la région est beaucoup plus sèche aujourd’hui. Nous discutons également les données et interprétations proposées par d’autres chercheurs sur d’autres sites. Bien qu’encore fragmentaires ces informations produiront à terme des reconstitutions paléonvironementales permettant aux paléoanthropologues et aux archéologues de prendre en compte les facteurs climatiques et environnementaux dans leurs scénarios.

Introduction Only since the research and teachings of Glynn Isaac in the 1980s has the holistic approach to studying the environment and fauna become commonplace. Studies on modern ecosystems in which the combinations of animals, plants and climate are noted form the basis for reconstructing past ecosystems where some of this information is missing. The interplay of these components, the effects of changes and ultimately the forcing factors can eventually be determined, particularly for hominid evolution, behavioural changes, use of tools and the resultant ‘modern’ humans.

104

Environmental changes and hominid evolution

Reconstructions of hominid palaeoecology at the terminal Pliocene have been attempted by the application of various indirect methods such as faunal and isotope analyses. Bovids studied by Leakey (1971) and Gentry and Gentry (1978a, b) at Olduvai Gorge showed a shift from predominantly reduncine antelopes which live close to water, to alcelaphine antelopes which inhabit drier open country in Bed I. Later some researchers took the information from bovids a step further. The ratios of bovid groups (Vrba’s Antelopine-Alcelaphine Criterion or AAC, 1980) have been used by Potts (1988), Shipman and Harris (1988), Kappelman (1984), Kappelman et al. (1997) to show the general drying out of a wetland environment to more open grassland from lower to upper Bed I times. Those methods have since lost popularity. Plummer and Bishop (1994) used the bovid metapodial functional analysis and habitat preferences (a non-taxonomic approach) to show a similar drying-out trend, but they postulated the presence of more of the intermediate and closed habitats than was concluded from the AAC methods. Jaeger (1976) used the micro-mammalian fauna of Olduvai Gorge to infer a wetter climate and more closed vegetation during early Bed I times, succeeded by a drier, more open habitat during middle Bed I times. Fernández-Jalvo et al. (1998) studied not only the rodents but also their taphonomy and predators, and so inferred that middle Bed I had a very rich closed woodland environment. By upper Bed I times the woodland was less rich and then gave way to more open and seasonal woodlands. At Laetoli and Koobi Fora, for example, similar approaches have been used. Analysis of specific modern vegetation types (e.g. woodland, shrubland, desert, forest) and their proportions of faunal types (e.g. arboreal, fossorial, terrestrial) has been used by Andrews (Andrews et al. 1979; Andrews, 1989) and Reed (1997) to infer the palaeoenvironment of hominid sites. Such analyses give an indication of the vegetation structure and general climatic conditions over time. Another indirect method is the use of stable carbon isotopes (Vogel, 1978; Cerling, 1984; 1992). Plants which have the C3 photosynthetic pathway are more tolerant of cooler temperatures, higher atmospheric CO2 concentrations and shading, and include all trees, most shrubs and most temperate, high-altitude and moist habitat grasses. Plants which have the C4 photosynthetic pathway can tolerate higher temperatures, lower atmospheric CO2 concentrations, water stress and greater salinity. C3 plants preferentially take up 13C over 12C and so their 13 C/12 C isotopic ratio is smaller than that for C4 plants. The ratios, measured against an international standard, are –35 to –22 ppm, with an average of –26 ppm for C3 plants and –16 to –8 ppm, average –11 ppm, for C4 plants. CAM plants use both pathways, include many succulents and euphorbias and have intermediate delta 13C values. Soil organic carbon derived from plants, soil carbonates and tooth enamel reflect these differences, although somewhat modified. The delta 13C ratios are, therefore, used to distinguish between drier open grassland habitats and cooler more closed woodland habitats. The implied

105

From Tools to Symbols

diets of various browsers (C3) and grazers (C4) from Lothagam (Leakey et al., 1996, late Miocene, Kenya), when compared to the diets of modern taxa, do not always correlate. The interpretation of the vegetation needs to be checked as the picture is more complex, especially as the animals in question may not be consumers of pure C3 or C4 foods. Consequently scavengers and carnivores could also give mixed signals. Identification of the fossil vegetation is the most direct approach to reconstructing the palaeoenvironment. Pollen and spores are easily transported and so represent the regional as well as the local flora. Comparisons with modern analogues enable the distinction of local and regional taxa. The study of assemblages of fossil macroplants or wood provides a good indication of the local vegetation and when the taphonomy is taken into account it is possible to distinguish between autochthonous, parautochthonous and allochthonous floral components (Spicer, 1991; Gastaldo et al., 1996). Phytoliths accumulate in sediments and can also give an indication of the vegetation over a period of time. Good reference material is needed for identification of the taxa. Unfortunately, fossil floras in hominid sites are rare. Here I will present data from my own research at two East African sites and two South African sites and then discuss other data from selected sites in Africa.

Laetoli, Tanzania Fossils were first collected from the Laetoli (Fig. 1) area by Mary and Louis Leakey in 1935 and preliminary mapping was done by Kent (1941). This area had been brought to the attention of Mary and Louis Leakey by an old Masai man while they were excavating at Olduvai Gorge, some 60 km to the NW. It was not until the discovery of gracile australopithecine jaws and teeth in 1974 that the site was recognised for its importance and potential in studying hominids. Laetoli is probably best known for the footprints of hominids and animals in the sediments and tuffs. The Laetolil Beds, dated 4,3–3,8 Ma for the Lower Bed and 3,8–3,5 Ma for the Upper Bed (Hay, 1987) are aeolian tuffs, and are exposed on both sides of the Garusi River valley (Fig. 1). They represent fully terrestrial conditions. From the sedimentology and fauna (bovids, giraffids, rhinocerotids and lagomorphs) it has been deduced that this open terrain with a dry savanna environment was exploited by hominids 3,5 million years ago. The palynological studies of Bonnefille and Riollet (1987) showed a variety of trees and shrubs in the Upper Laetolil beds belonging to the families Anacardiaceae, Burseraceae, Capparidaceae, Combretaceae, Compositae, Ebenaceae, Euphorbiaceae, Meliaceae, Mimosaceae, Oleaceae, Rosaceae, Sapindaceae, Sapotaceae, Simaroubaceae, Ulmaceae, Palmae, and the conifers Podocarpus and Juniperus. This vegetation is interpreted to have been typical of a higher altitude (1 500–1 800 m) open savanna with a very diverse herbaceous component with abundant grasses. The

106

Environmental changes and hominid evolution

climate would have been arid with a pronounced dry season but not necessarily with higher temperatures (Bonnefille & Riollet, 1987). Scott has re-sampled for pollen but without success (pers. comm.). Today the area is exceptionally arid for the equatorial zone (3ºE 13' S, 35ºE 13' E), even though the mean annual rainfall is 500–1 000 mm, with a bimodal distribution. The rainfall is very irregular, however, and the mean annual evaporation very high (2 000 mm). Acacia-Commiphora steppe is the vegetation type with short and medium grasslands. Along the main drainage lines are low or high woodlands with Acacia lahai, Acacia seyal, Acacia nilotica and others. Acacia drepanolobium and Balanites aegyptiaca are the common trees growing on the heavy cotton soils or mbuga. In the narrow gorge of Naisuri, protected from fire but used by the Masai for watering their goats, the woody vegetation is tall and includes Acacia xanthophloea, Albizia gummifera, Combretum molle, Croton macrostachys, Dombeya rotundifolia, Ekebergia capensis, Grewia similis, Phyllanthus sp., Rhus natalensis and Ziziphus mucronata, to name a few (Peter Andrews,

Figure 1 Map of East Africa showing hominid localities that have palaeobotanical remains.

107

From Tools to Symbols

pers. comm. and own observation). Silicified fossil woods are abundant at Noiti 1 and Noiti 2 in the Laetolil Beds (4,3–3,8 Ma), varying in size from small twigs to trunk pieces over 20 cm in diameter. Preliminary work shows that there are at least ten types of fossil wood, whereas there are only four main types of extant wood in the immediate area. No conclusions can be drawn about the woody palaeovegetation at this early stage of the project except to add that there are several species of Acacia and members of the Dichapetalaceae, Euphorbiaceae, Oleaceae and Sterculiaceae. Seeds are round and unadorned, silicified and still to be identified. They are numerous and occur just below Tuff 7, the Footprint tuff which has been dated at 3,56 Ma (Drake & Curtis, 1987). Collections of fauna and flora are continuing under the leadership of Terry Harrison with the aim of better understanding the whole community and community change over time.

Olduvai Gorge, Tanzania Fossils were first collected from Olduvai Gorge (Fig. 1) in the south-eastern part of the Serengeti in 1911 by a German entomologist, Prof. Kattwinkel. In 1913 Hans Reck began mapping and collecting fossils. Louis Leakey saw some of these in a museum in Berlin, recognised their significance and started excavating at Olduvai in 1931. He and Mary had many productive field seasons there from 1932 to 1972, when he died, and 1984, respectively, when she retired from fieldwork. Her discovery of the skull Zinjanthropus boisei (Paranthropus boisei) in 1959 brought them world attention and funding for more excavations. Early hominids such as A. robustus, Homo habilis and Homo erectus occurred there too. Early stone tools were studied by Mary Leakey and these ‘Oldowan’ tools are recognised worldwide. Richard Hay described the sedimentology and stratigraphy in detail (1976) and currently he and other sedimentologists and tephrostratigraphers are further refining the data. Palynological studies have been done by Raymonde Bonnefille (1984) and she has given a general outline of the local and highland vegetation changes over time. During middle Bed I times the local vegetation was wooded grassland while the highland had montane evergreen forest. By upper Bed I time, with climatic drying, locally there was sub-desert steppe with Acacia spp. and Commiphora spp. and the highlands supported reduced forest vegetation. In lowermost Bed II times the climate became more mesic with denser woodland occurring, as well as sedges and Typha (Bonnefille, 1984). More recently the multidisciplinary team, Olduvai Landscape Palaeoanthropology Project (OLAPP) has, among its other objectives, concentrated on trying to understand the palaeoenvironment of Olduvai Gorge when the hominids occupied the area. More trenches have been excavated in key areas. Numerous fragments of silicified sedges, grasses, twigs, wood and unidentifiable plant material have been collected from the

108

Environmental changes and hominid evolution

sites HWKE, HWKEE, MCK, FLK and VEK. These sites represent Bed I and lowermost Bed II on the eastern margin of the palaeolake. Petrified wood has successfully been used to reconstruct the vegetation and palaeoenvironment from sites all over the world. This is done taxonomically and anatomically. Some extant trees have very restricted distributions and once the fossil taxon has been determined and compared with modern taxa, the climatic tolerance of the fossils can be deduced. From a purely anatomical point of view, much information has been gathered (Carlquist, 1975; Wheeler & Baas, 1991) on the adaptations of wood to various climatic conditions, so the fossil wood structure can give an indication of the climatic conditions under which the tree grew. Dendroclimatology is a useful application for reconstructing past climates, but only where growth rings are clear, which is not often the case in tropical woods. From HWK East, in lowermost Bed II, two of the small woody fragments have been identified as Guibourtia coleosperma (Caesalpiniaceae) (Bamford, submitted), (Figure 3a). The anatomy of the fossil wood was compared with modern material housed in the xylarium of the Musée Royal de l’Afrique Centrale in Tervuren, Belgium. The extant tree is tall, more or less evergreen, reaches up to 20 metres high with a drooping and somewhat rounded crown, and grows on deep Kalahari sands, in woodland (Palmer & Pitman, 1972). It is typical of the Sudano-Zambezian phytochorion (Werger & Coetzee, 1978). Today Guibourtia coleosperma occurs in Angola, Botswana, Mozambique, Congo, Zambia and Zimbabwe (Léonard, 1950). According to Palmer and Pitman (1972) the red arils enclosing the seeds of this tree are edible. Local people eat them raw, mix them into a porridge or make them into a nourishing drink. The tree also has medicinal uses: the leaves are used for coughs and the roots are applied to wounds. Wood from this tree is hard and heavy, so has many uses, while the bark is used to cure hides (Palmer & Pitman, 1972). Although the woody fragments are small, they indicate the presence of trees, and by association, some woodland during lower Bed II times. This is further supported by the presence of hard woody stem bases, still to be identified. As the size and shape of the modern trees is known we can reconstruct a woodland with trees about 20 metres high and virtually evergreen, thus providing shade and shelter all year round, and food for part of the year (June–August in southern Africa, but this may be different near the equator where the seasons are less pronounced). Grass fragments also occur in the Upper Bed I and Lower Bed II sediments, indicating the presence of grassland. A picture of the complex and changing vegetation will emerge as this and other fossil plants are identified.

109

From Tools to Symbols

Sterkfontein Cave Site, South Africa Fossils were first discovered in the breccias of Sterkfontein Caves (Fig. 2) by miners excavating the calcium carbonate for the gold mining industry in 1936. The skull and cast of Mrs Ples were recognised by Robert Broom as Australopithecus africanus (1936). Since then various people have worked the site, including Prof. Phillip Tobias and Prof. Ron Clarke. The cave stratigraphy is complicated by talus cone deposits, roof collapse and even floor collapse into chambers below. The ages of the Members has recently been disputed. Member 4 was dated at 2,6–2,8 Ma (Partridge, 1978; Clarke, 1984) but this age was challenged by Berger et al. (2002) as being too old and more likely 1,5–2,5 Ma. New dating of Member 2 by Partridge et al (2003) puts Member 2 at approximately 4 Ma and Member 4 older than 3 Ma. Whatever the precise age of Member 4, it contained over 300 small pieces of calcified wood, ranging in size from a few millimetres in diameter to 30 mm in diameter. These were carefully excavated by Alun Hughes in the 1980s. Other noteworthy fossils from

Figure 2 Map of South Africa showing hominid localities that have palaeobotanical remains.

110

Environmental changes and hominid evolution

this member are Australopithecus africanus, bovids, rodents, cercopithecoids, hyaenids, felids, canids, an elephantid, hyracoids, equids and suids (Berger et al., 2002). The fossil woods are Dichapetalum mombuttense (Dichapetalaceae) and Anastrabe integerrima (Scrophulariaceae) (Bamford, 1999), (Figs 3b, 3c). Identifications were verified using modern material from the xylarium of the Musée Royal de l’Afrique Centrale. Dichapetalum is a large genus of tropical and subtropical shrubs and lianas with a mainly African distribution (Hauman, 1958). Only one species of this genus grows today in South Africa, Dichapetalum cymosum, which is a suffrutex and poisonous to livestock. The fossil is clearly a liana with the characteristic very tall rays, short and numerous vessel elements, and narrow tracheary elements with numerous bordered pits. It has no fibre (the tissue that gives wood strength and rigidity), which is usually lacking in lianas. Dichapetalum mombuttense grows only in central Africa today, in dense humid forest and gallery forests (Hauman, 1958; Bamford, 1999), and, as is the case with extant lianas, would have relied on forest trees for support, enabling the long, thin, flexible stems to reach the light so the leaves could photosynthesise. The second wood preserved in Member 4 is the shrub Anastrabe integerrima, which still grows on the edge of rocky outcrops, in forest margins in bush and along streams from East London to Zululand (Palmer & Pitman, 1972). It would seem likely that the lianas grew up and over gallery forest trees around the cave entrance in the Sterkfontein Valley. It is possible that bits of liana fell into the cave and were preserved along with the bones of the prey from the leopards feeding in the trees above. From Member 4 there are very few arboreal animals and Reed (1997) concluded that the surrounding vegetation during Member 4 times was bushland and medium density woodland. In contrast the rodent fauna indicates an open grassland environment with no indication of a gallery forest fauna (Avery, 2001). Although the modern descendants of the fossil rodents live in the present-day Grassland, Arid Savanna, and Moist Savanna biomes (Avery, 2001; Rutherford & Westfall, 1994) the tacit assumption is that the modern vegetation is undisturbed. This is not so. Occupation of the area for at least the past 50 Ka by humans who have grazed livestock, collected food plants and used hardwoods for fires, particularly for iron smelting (Deacon & Deacon, 1999), has resulted in the degraded and less diverse vegetation that we see today. Pollen records from Wonderkrater, Tswaing and Rietvlei Dam, for example, show the presence of Kalahari Thornveld during the early and middle Holocene (Scott, 2002). Thus it is very important to distinguish between natural grasslands and secondary grasslands when using the modern vegetation analogues.

111

From Tools to Symbols

Florisbad, South Africa The Pleistocene site at the Florisbad freshwater spring site (Fig. 2 ) is well known for fauna and stone tools (Kuman & Clarke, 1986; Kuman et al., 1999) and the early modern human skull, Homo helmei (Clarke, 1986). Apart from various stone tools, several wooden pieces were recovered, studied and chemically treated (Clark, 1955), but only one survived the treatment. Working of one end of the wooden piece implies that is has been fashioned into a tool of some unknown application. This piece has been identified as Zanthoxylum chalybeum L. (Fig. 3d), a tree belonging to the Rutaceae (Bamford & Henderson, 2003). The identification was also verified using modern woods from the xylarium at the Musée Royal de l’Afrique Centrale, Tervuren, Belgium. Other species of this genus occur in South Africa today but this particular species only occurs farther north, in Zimbabwe (Coates Palgrave, 2002). It seems unlikely that humans would have walked 1 000 km to collect or trade wood which has the same properties as the local species (Zanthoxylum davyi, Palmer & Pitman, 1972). We concluded that the wood grew locally. This implies that the climate and vegetation have shifted and the Free State was warmer and wetter, like central Zimbabwe today, at some time between 259 Ka and 125 Ka when the Florisbad Spring was an occupation site (Bamford & Henderson, 2003).

Other palaeobotanical evidence from East African hominid sites West Rift Semliki, Zaire The late Pliocene fossil woods from the Upper Semliki River Valley (Fig. 1) indicate that there was a variety of vegetation types in close proximity to each other, from lowland and montane dense forest, to swamp forest, riverine gallery forest, savanna gallery forest, savanna and even desert steppe woody vegetation (Dechamps & Maes, 1990). At the Plio-Pleistocene time there was dense shady forest much like what occurs in the Virunga National Park today (Dechamps & Maes, 1990). The fossil woods were compared with a large reference collection in Tervuren and a detailed analysis of the present-day climate was carried out for each species. Since the whole assemblage was considered, the interpretation of the climate from the woods is as accurate as can be expected for palaeoclimate reconstructions. Woods occurring throughout the Lusso beds (2–2,3 Ma; Boaz et al., 1992) showed an increase in humidity in the Late Pliocene (Dechamps, 1987; Dechamps & Maes, 1990, Boaz et al.,1992). East Rift Omo, Ethiopia

In the Omo Valley there are fossils from about 3,6–1 Ma and a good fossil wood record (Dechamps, 1976; de Heinzelin, 1983). The vegetation and climate has fluctuated over time and is summarised below (de Heinzelin, 1983):

112

Environmental changes and hominid evolution

Figure 3

Photomicrographs of transverse sections of fossil woods: (a) Guibourtia coleosperma (Caesalpiniaceae), BP/16/824b, Olduvai Gorge. Scale bar = 300 Fm (b) Dichapetalum mombuttense (Dichapetalaceae), BP/16/462, Sterkfontein caves. Scale bar = 200 Fm (c) Anastrabe integerrima (Scrophulariaceae), BP/16/460, Sterkfontein caves. Scale bar = 300 Fm (d) Zanthoxylum chalybeum (Rutaceae), Florisbad. Scale bar = 165 Fm

113

From Tools to Symbols Mursi Formation, more than 4 Ma: dry woody savanna and humid gallery forest, together with grass pollen and the similar woodland types (Bonnefille, 1974, in Dechamps & Maes, 1985). Usno Formation and lower Shungura Formation, between 3.4 and about 2.6 Ma: welldeveloped gallery forest surrounded by dry savanna and shrubs. Middle to upper Shungura Formation, between about 2.6 and 1 Ma: shift from dominance of dry savannas with a gallery forest component to more gallery forest and back to slightly drier conditions.

Hadar, Ethiopia From the hominid site at Hadar in Ethiopia (Fig. 1) remains of Australopithecus afarensis have been recovered, as well as a rich fauna. Bonnefille et al. (1987) have analysed the pollen and found that the montane evergreen bushland extended down to 500 metres in the Awash Basin during the Pliocene (3,3–2,9 Ma) compared with today’s restricted distribution to above 1 500 metres. This bushland was associated with a humid montane forest. After this time there was a significant drying out and a shift towards more arid grasslands. There are woods of the same taxa as some of the pollens and so the interpretation is corroborated (Dechamps & Maes, 1985; Bonnefille et al., 1987). Lake Turkana, Kenya: (Koobi Fora project) Pollen analyses of 30 recent surface soil samples and 34 lake deposit surface samples were used to interpret 17 fossil samples in the region of Lake Turkana (Fig. 1) by Vincens (1982) She was able to deduce a climatic succession of fluctuating humidity but general cooling: 2,7–2,1 Ma: climate warm and humid; 2,1–1,8 Ma: climate dry with a tendency toward cooling; 1,8–1,5 Ma: climate more humid and still cooler. Her studies on the modern pollens and vegetation show that montane pollen does not travel far by air but travels much farther by water or rivers. Further palynological analyses showed that the vegetation surrounding the lake between 2,0 and 1,4 Ma was sub-arid type, probably Acacia-Commiphora steppe, with more woody vegetation along the streams (Vincens, 1987). The grassland expanded around 1,9 Ma and there was some regression of the montane forests (Vincens, 1987).

114

Environmental changes and hominid evolution

Other palaeobotanical evidence from South African hominid sites Taung Within the breccias and tuffs at Taung there are numerous leaf impressions but these have been mined out of the deposit and dumped so the provenance has unfortunately been lost (pers. observation). I am unaware of any successful palynological studies from this site in the northeastern part of South Africa. Makapansgat Although several attempts have been made to reconstruct the vegetation at Makapansgat from palynological samples taken from the Limeworks (Cadman & Rayner, 1989; Zavada & Cadman, 1993), the nature of the deposit, porous cave breccias, means that there is contamination and reworking within the deposit as well as from outside rainwater percolating through. Numerous other approaches have been used to reconstruct the palaeoclimate but the results are contradictory. Nonetheless work is still in progress. Gladysvale Fossil seeds of Phoenix reclinata less than 1 Ma have been reported from Gladysvale, but have not described (Lacruz et al., 2002). Kromdraai, Sterkfontein, Swartkrans The hominid sites in the Sterkfontein valley are also limestone caves with problems of contamination, low concentrations of pollen and difficulties in extracting pollen from the carbonate-impregnated sediments (Scott & Bonnefille, 1986; Carrion & Scott, 1999; Scott, 2002). Any interpretations from these sites should be treated with caution and closely compared with palaeoclimatic interpretations from other methodologies.

Discussion The fossil plant evidence from hominid sites is rather scanty, but it does give direct evidence of the vegetation. Other sites may well have plant material that has been overlooked or underestimated in its usefulness. In general the palynology samples are larger and give a regional and local vegetation signal, but it not always possible to identify the pollen to a particular species, usually only to the genus level. Fossil woods are usually less abundant but can be identified to species level and then close climatic correlations can be established. It is important, however, to be certain of the provenance of the wood and to look for indications in the deposit of any reworking or long-distance transportation, before and after fossilisation. The first four sites described here represent different settings and different types of

115

From Tools to Symbols

deposits. Laetoli is a dry, open setting and the flora, fauna and footprints have been preserved in tuffs. The wood deposits were capped by the Ogal lavas. Beds I and II of Olduvai Gorge are lacustrine deposits which have also been covered by ashfalls and preserved. Sterkfontein is a limestone cave deposit where plant and animal debris has been preserved on flowstone and calcified. Florisbad is a spring deposit where the bones and wood have been preserved in a peat deposit. They also represent different vegetation types. Palaeobotanical evidence from other sites is briefly discussed. As the areas are geographically widespread and cover approximately the last 5 Ma, it not feasible to compare them closely. The overall impression, however, is that the climate and vegetation have fluctuated considerably in the past and the present-day vegetation in each particular site is mostly different from when the early hominids occupied the area. More research still needs to be done in order to interpret how the vegetation affected the hominids and their distribution, but at least we are gradually getting a picture of what the vegetation was like at certain times.

Acknowledgements I would like to thank Phillip Tobias and Ron Clarke for inviting me to work on the Sterkfontein woods; the leaders of the OLAPP team, Rob Blumenschine, Charles Peters and Fidelis Masao, for inviting me to join the team and work on the Olduvai fossil plants; Terry Harrison for the Laetoli woods, and Zoe Henderson and the National Museum, Bloemfontein, for access to the Florisbad wood. Financial assistance from the University of the Witwatersrand, NSF Grant to OLAPP (number 0109027), a Leakey Foundation Grant and a grant from the Palaeoanthropological Scientific Trust, South Africa, are gratefully acknowledged. The organisers and sponsors of this conference are also warmly thanked. My thanks to José Carrión for reviewing this paper.

References Andrews, P. (1989). Palaeoecology of Laetoli. Journal of Human Evolution 18, 173–181. Andrews, P., Lord, J.M. & Nesbitt-Evans, E.M. (1979). Patterns of ecological diversity in fossil and modern mammalian faunas. Biological Journal of the Linnean Society 11, 177–205. Avery, D.M. (2001). The Plio-Pleistocene vegetation and climate of Sterkfontein and Swartkrans, South Africa, based on micromammals. Journal of Human Evolution 41, 113–132. Bamford, M.K. (1999). Pliocene wood from an early hominid cave deposit, Sterkfontein, South Africa. South African Journal of Science 95, 231–237. Bamford, M.K. (submitted). Plio-Pleistocene woods from Olduvai Gorge, Tanzania. Quaternary International Bamford, M.K. & Henderson, Z. (2003). A reassessment of the wooden fragment from Florisbad, South Africa. Journal of Archaeological Sciences 30, 637–650.

116

Environmental changes and hominid evolution Berger, L.R., Lacruz, R. & de Ruiter, R.J. (2002). Revised age estimates of Australopithecus-bearing deposits at Sterkfontein, South Africa. American Journal of Physical Anthropology 119, 192–197. Boaz, N.T. et al. (1992). A new evaluation of the significance of the Late Neogene Lusso Beds, Upper Semliki Valley, Zaire. Journal of Human Evolution 22, 505–517. Bonnefille, R. (1984). Palynological research at Olduvai Gorge. National Geographic Society Research Reports 17, 227–243. Bonnefille, R. & Riollet, G. (1987). Palynological spectra from the Upper Laetolil Beds. In (M.D. Leakey & J.H. Harris, Eds) Laetoli: A Pliocene Site in Northern Tanzania, pp. 52–61. Clarendon Press: Oxford. Bonnefille, R., Vincens, A. & Buchet, G. (1987). Palynology, stratigraphy and palaeoenvironment of a Pliocene hominid site (2.9–3.3 M.Y.) At Hadar, Ethiopia. Palaeogeography, Palaeoclimatology, Palaeoecology 60, 249–281. Cadman, A. & Rayner, R.J., (1989). Climatic change and appearance of Australopithecus africanus in the Makapangat sediments. Journal of Human Evolution 18, 107–113. Carlquist, S. (1975). Ecological Strategies of Xylem Evolution. Berkeley: University of California Press. Carrión, J.S. & Scott, L. (1999). The challenge of pollen analysis in palaeoenvironmental studies of hominid beds: the record from Sterkfontein caves. Journal of Human Evolution 36, 401–408. Cerling, T.E., 1984. The stable isotopic composition of modern soil carbonate and its relationship to climate. Earth and Planetary Science Letters, 329, 237–239. Cerling, T.E., 1992. Development of grasslands and savannas in East Africa during the Neogene. Palaeogeography. Palaeoclimatology. Palaeoecology. 97, 241–247. Clark, J.D. (1955). A note on a wooden implement from the level of Peat 1 at Florisbad, Orange Free State. Researches of the National Museum, Bloemfontein 1, 574–577. Clarke, R.J. (1984). On some new interpretations of Sterkfontein stratigraphy. South African Journal of Science 90, 211–214. Clarke, R.J. (1986). A new reconstruction of the Florisbad cranium, with notes on the site. In (E. Delson, Ed.) Ancestors: The Hard Evidence, pp. 301–305. New York: Alan R. Liss. Coates Palgrave, K. (2002). Trees of Southern Africa. Revised. Cape Town: Struik. Deacon, H.J. & Deacon, J. (1999). Human Beginnings in South Africa. Cape Town: David Phillip. Dechamps, R. (1976). Résultats preliminaires de l’étude des bois fossiles de la basse vallée de l’Omo (Ethiopie sud occidentale). Musée Royal de l’Afrique Centrale, Tervuren, (Belgique) Département Géologique et Minerale, Rapport Annuel. 1975, 59/65. Dechamps, R. (1987). Premiers résultats du Semliki Research Project (Parc National des Virunga, Zaïre). II Détermination des bois fossiles. Musée Royal de l’Afrique Centrale Tervuren, (Belgique) Département Géologique et Minerale, Rapport Annuel. 1985/86, pp. 145-148. Dechamps, R. & Maes, F. (1985). Essai de réconstitution des climats et des végétations de la

117

From Tools to Symbols basse vallée de l’Omo au Plio-Pléistocène à l’aide de bois fossiles. In (Y. Coppens, Ed.) L’environnement des Hominidés au Plio-Pléistocène. Fondation Singer-Polignac, pp. 175–222, Paris: Masson. Dechamps, R. & Maes, F. (1990). Woody plant communities and climate in the Pliocene of the Semliki Valley, Zaire. Virginia Museum of Natural History Memoir 1, 71–94. de Heinzelin, J., Ed. (1983). The Omo Group. Archives of the International Omo Research Expedition. Musée Royal de l’Afrique Centrale , ser. 8, Science et Géologique. 85, 209–229. Drake, R. & Curtis, G.H. (1987). K-Ar geochronology of Laetoli fossil localities. In (M.D. Leakey & J.M. Harris, Eds) Laetoli: A Pliocene Site in Northern Tanzania, pp. 48–52. Oxford: Clarendon Press. Fernández-Jalvo, Y., Denys, C., Andrews, P., Williams, T., Dauphin & Y., Humphrey, L., (1998). Taphonomy and palaeoecology of Olduvai Bed-I (Pleistocene, Tanzania). Journal of Human Evolution 34, 137–172. Gastaldo, R.A., Ferguson, D.K., Walther, H. & Rabold, J.M. (1996). Criteria to distinguish parautochthonous leaves in Tertiary alluvial channel-fills.Review of Palaeobotany and Palynology. 91, 1–21. Gentry, A.W. & Gentry, A. (1978a). Fossil Bovidae (Mammalia) of Olduvai Gorge, Tanzania. Part I. Bulletin of the British Museum of Natural History, Geological Series 29(4), 289–446. Gentry, A.W. & Gentry, A. (1978b). Fossil Bovidae (Mammalia) of Olduvai Gorge, Tanzania. Part II. Bulletin of the British Museum of Natural History, Geological Series 30(1), 1-83. Hauman, L. (1958). Dichapetalaceae. Flore du Congo-Belge et du Ruanda-Urundi. Spermatophyte, Vol. VII. INEAC. Hay, R.L. (1976). The Geology of Olduvai Gorge. Berkeley: University of California Press. Hay, R.L. (1987). Geology of the Laetoli area. In (M.D. Leakey & J.M. Harris, Eds) Laetoli: A Pliocene Site in Northern Tanzania, pp. 23–47. Oxford: Clarendon Press. Jaeger, J.-J. (1976). Les rongeurs (Mammalia, Rodentia) du Pléistocène inférieur d’Olduvai Bed I (Tanzania). Première partie: Les Muridés. In (R.J.G. Savage & S. Coryndon, Eds) Fossil Vertebrates of Africa, Vol. 4, pp. 58–120. New York: Academic Press. Kappelman, J. (1984). Plio-Pleistocene environments of Bed I and Lower Bed II, Olduvai Gorge, Tanzania. Palaeogeography. Palaeoclimatology. Palaeoecology 48, 171–196. Kappelman, J., Plummer, T., Bishop, L., Duncan, A., Appleton, S. (1997). Bovids as indicators of Plio-Pleistocene palaeonvironments in East Africa. Journal of Human Evolution 32, 229– 256. Kent, P.E. (1941). The recent history and Pleistocene deposits of the plateau north of Lake Eyasi, Tanganyika. Geological Magazine 78, 173–184. Kuman, K. & Clarke, R.J. (1986). Florisbad – new investigations at a Middle Stone Age hominid site in South Africa. Geoarchaeology, 1, 103–125. Kuman, K., Inbar, M. & Clarke, R.J. (1999). Palaeoenvironments and sequence at the Florisbad

118

Environmental changes and hominid evolution Middle Stone Age hominid site, South Africa. Journal of Archaeological Sciences 26, 1409– 1426. Lacruz, R.S., Brink, J.S., Hancox, P.J., Skinner, A.R., Herries, A., Schmid, P. & Berger, L.R. (2002). Palaeontology and geological context of a middle Pleistocene faunal assemblage from the Gladysvale Cave, South Africa. Palaeontologica Africana 38, 9–114. Leakey, M.D. (1971). Olduvai Gorge Excavations in Beds I and II, 1960–1963, Vol. 3. Cambridge: Cambridge University Press. Leakey, M.G., Feibel, C.S., Bernor, R.L., Harris, J.M., Cerling, T.E., Stewart, K.M., Storrs, G.W., Walker, A., Werdelin, L. & Winkler, A.J. 1996. Lothagam: a record of faunal change in the late Miocene of East Africa. Journal of Vertebrate Palaeontology 16, 556–570. Léonard, J. (1950). Notulae systematicae IX. Nouvelles observations sur le genre Guibourtia (Caesalpiniaceae). Bullétin du Jardin Botanique de Bruxelles 20, 269–284. Palmer, E. & Pitman, N. (1972). Trees of Southern Africa, Vol. III. Cape Town: A.A. Balkema. Partridge, T.C. (1978). Re-appraisal of the lithostratigraphy of Sterkfontein hominid site. Nature, 275, 282–287. Partridge, T.C., Granger, D.E., Caffee, M.W. & Clarke, R.J. (2003). Lower Pliocene hominid remains from Sterkfontein. Science 300, 607–612. Plummer, T.W. & Bishop, L.C. 1994. Hominid paleoecology at Olduvai Gorge, Tanzania as indicated by antelope remains. Journal of Human Evolution 27, 47–75. Potts, R. 1988. Early Hominid Activities at Olduvai Gorge. Hawthorne, NY: Aldine de Gruyter. Reed, K.E. (1997). Early hominid evolution and ecological change through the African PlioPleistocene. Journal of Human Evolution 32, 289–322. Rutherford, M.C. & Westfall, R.H. (1994). Biomes of southern Africa: an objective categorization. Memoirs of the Botanical Survey of South Africa 63, 1–94. Scott, L. (2002). Grassland development under glacial and interglacial conditions in southern Africa: review of pollen, phytolith and isotope data. Palaeogeography Palaeoclimatology Palaeoecology 177, 47–57. Scott, L. & Bonnefille, R. (1986). Search for pollen from the hominid deposits of Kromdraai, Sterkfontein, and Swartkrans: some problems and preliminary results. South African Journal of Science 82, 380–382. Shipman, P. & Harris, J.M. 1988. Habitat preference and paleoecology of Australopithecus boisei in Eastern Africa. In (F.E. Grine, Ed.) Evolutionary History of the ‘Robust’ Australopithecines, pp. 343–381. Hawthorne, NY: Aldine de Gruyter. Spicer, R.A. (1991). Plant taphonomc processes. In (P.A. Allison & D.E.G. Brigs, Eds) Taphonomy: Releasing the Data Locked in the Fossil Record, pp. 71–113. New York: Plenum Press. Vincens, A. (1982) Recherches palynologiques dans la région du lac Turkana (Kenya). Palaeoecology of Africa 14, 85–91. Vincens, A. (1987). Environnements botanique et climatiques des hominidés de l’est Turkana,

119

From Tools to Symbols Kenya, entre 2.0 et 1.4 millions d’années: apport de la palynologie. Palaeoecology of Africa 18, 257–269. Vogel, J.C. (1978). Isotopic assessment of dietary habits of ungulates. South African Journal of Science 74, 298–301. Vrba, E.S. 1980. The significance of bovid remains as an indicator of environment and predation patterns. In (A.K. Behrensmeyer & A.P. Hill, Eds), Fossils in the Making, pp. 247–272. Chicago: University of Chicago Press. Werger, M.J.A. & Coetzee, B.J. (1978). The Sudano-Zambezian region. In (M.J.A. Werger, Ed.), Palaeogeography and Ecology of Southern Africa. The Hague: W. Junk bv Publishers. Wheeler, E.A. & Baas, P. (1991). A survey of the fossil record for dicotyledonous wood and its significance for evolutionary and ecological wood anatomy. IAWA Bulletin, n.s. 12, 275– 332. Zavada, M.S. & Cadman, A. (1993). Palynological investigations at the Makapansgat limeworks: an australopithecine site. Journal of Human Evolution 25, 337–350.

120

Implications of the presence of African ape-like teeth in the Miocene of Kenya Martin Pickford and Brigitte Senut Laboratoire de Paleontologie, UMR 8569 du CNRS, 8, rue Buffon, 75005, Paris College de France, 11, Place Marcellin Berthelot, 7505, Paris, France Département Histoire de la Terre, USM 0203, du Muséum National d’Histoire naturelle & UMR 5143, PICS 1048 (CNRS), Case 38, 57, rue Cuvier, 75005, Paris, France

Abstract This paper deals with some of the implications of the discovery of four ape-like teeth from the Middle Miocene (12,5 Ma) and Late Miocene (6–5,9 Ma) of Kenya. An unworn, isolated lower molar from Member B of the Ngorora Formation (12,5 Ma), Tugen Hills, Kenya, differs markedly from lower molars of Middle and Early Miocene large hominoids but is closer in morphology to chimpanzee molars (peripheralised cusps, buccolingually compressed lingual cusps, thin enamel, large and deep occlusal basin, reduced buccal cingulum). If the tooth is part of the chimpanzee clade then it is important for estimating the timing of the dichotomy between chimpanzees and hominids and suggests that this event would have occurred several million years earlier than is currently estimated by most researchers. An incomplete, unworn isolated upper molar, an upper central incisor and a lower molar from the Lukeino Formation (6–5,9 Ma), Tugen Hills, Kenya, are morphologically closer to those of Gorilla gorilla than to any other fossil or extant hominoid with which they were compared. The upper molar is a large tooth (mesio-distal length 14 mm) with peripheralised cusps, bucco-lingually wide distal fovea, fairly voluminous trigon basin and high dentine penetrance, all features which suggest affinities with gorillas. Its enamel thickness (1,6–1,7 mm on the hypocone) is similar to that of gorilla molars. It differs markedly from molars of the early hominid, Orrorin tugenensis, which occurs at the same site, which are smaller, have more centralised cusps, smaller trigon basin, reduced distal fovea and low dentine penetrance. The Kapsomin molar differs from teeth of australopithecines for much the same reasons, even if its dimensions overlap with those of Australopithecus antiquus and Praeanthropus africanus. It is highly divergent from chimpanzee teeth, not only in its dimensions, but also in its morphology. An upper central incisor from Kapsomin is large and wedge-shaped in lateral view without the lingual fossa that characterises teeth of hominids and chimpanzees. It is close in size and morphology to those of gorillas.

121

From Tools to Symbols If the species from which the Kapsomin and Cheboit teeth came is part of the gorilla clade, then there are important implications for the timing of events in gorilline evolution, and they make it less likely that European genera such as Ouranopithecus are ancestral to African apes or hominids. The four ape teeth from Ngorora and Lukeino suggest that the extant African apes evolved in Africa, and did not immigrate into the continent from Europe or Asia. Orrorin suggests the same for hominids, meaning that the lineages leading to both of the extant African ape genera as well as the hominids are more likely to have originated in Africa rather than Eurasia.

Résumé Cet article présente les implications de la découverte de quatre dents de grands singes dans le Miocène moyen (12,5 Ma) et le Miocène Supérieur (6–5,9 Ma) du Kenya. Une molaire inférieure isolée et non usée provenant du Membre B de la Formation de Ngorora (12,5 Ma) dans les Collines Tugen présente des différences importantes avec les molaires inférieures des hominoïdes connus de grande taille du Miocène inférieur et moyen, mais se rapproche morphologiquement de celles des chimpanzés (cuspides périphériques, cuspides linguales comprimées bucco-lingualement, émail fin, bassin occlusal grand et profond, cingulum buccal réduit). Si la dent appartient au clade des chimpanzés, elle s’avère extrêmement importante pour estimer la date de divergence entre les chimpanzés et les hominidés. En effet, elle suggère que la dichotomie aurait eu lieu plusieurs millions d’années plus tôt que ce que les chercheurs acceptent généralement. Une molaire supérieure isolée et non usée, une incisive centrale supérieure et une molaire inférieure provenant de la Formation de Lukeino (6–5,9 Ma) dans les Collines Tugen au Kenya, sont plus proches morphologiquement de celles de Gorilla gorilla que de celles de n’importe quel autre hominoïde actuel. La molaire supérieure est une dent de grande taille (longueur mésio-distale: 14 mm) aux cuspides périphériques, présentant une grande fovea distale, un bassin du trigonide assez volumineux et une haute pénétration de dentine. Tous ces caractères suggèrent des affinités avec les gorilles. L’épaisseur de l’émail (1,6 à 1,7 mm sur l’hypocone), est similaire à celle des molaires de gorilles. La dent se différencie fortement de celles de l’hominidé ancien, Orrorin tugenensis représenté sur le même site. Les dents de ce dernier sont plus petites, leurs cuspides plus centrales, le bassin du trigone plus petit, la fovea distale réduite et la pénétration de dentine faible. La molaire de Kapsomin se differencie des dents d’Australopithèques pour les mêmes raisons évoquées précédemment, même si ses dimensions recoupent celles d’Australopithecus antiquus et de Praeanthropus africanus. Elle est très différente des dents de chimpanzés, pas seulement en dimensions, mais aussi en morphologie. Une incisive centrale supérieure provenant de Kapsomin est grande et une forme “en coin” en vue latérale; elle ne présente pas la fosse linguale typique des dents d’hominidés et de chimpanzés. Elle ressemble en taille et en morphologie à celle des gorilles. La dent de Cheboit, une m/2 ou m/3, présente un grand bassin occlusal, et ses cuspides sont localisées à la périphérie de la dent. Si l’espèce à laquelle appartiennent les dents de Kapsomin et de Cheboit sont incluses dans le clade des gorilles, le résultat est d’importance pour dater les événements de l’évolution des gorilles, et rend plus improbable que des genres européens comme Ouranopithecus, par exemple, descendent des grands singes africains ou des hominidés.

122

Implications of the presence of African ape-like teeth in the Miocene of Kenya Les dents des hominoïdés de Ngorora et de Lukeino indiquent que les grands singes africains modernes ont évolué en Afrique et qu’ils n’ont pas immigré sur le continent à partir de l’Europe ou de l’Asie. Orrorin suggère la même chose pour les hominidés, montrant ainsi que les lignées menant aussi bien aux grands singes africains qu’aux hominidés ont plus de chance d’avoir eu une origine africaine qu’eurasiatique.

The Ngorora lower molar In 1998, an unworn, isolated lower right molar crown (Fig. 3a) was found by Mr Kiptalam Cheboi at locality 2/1, Kabarsero, Tugen Hills, Kenya. This site is in Member B of the Ngorora Formation (Bishop & Pickford, 1975; Pickford, 1986) in fluvial deposits aged c. 12,5 Ma (Deino et al., 1990). The description and comparisons of the tooth appear in Pickford and Senut (in press). Suffice to say here that it most closely resembles Dryopithecus among the known Miocene apes of Eurasia and Africa, and Pan among the extant apes. Implications of the Ngorora ape lower molar for the molecular clock Until now, no fossil chimpanzees and gorillas have been reported. Some authors (Senut et al., 2001; Wolpoff et al., 2002) have suggested that Ardipithecus and Sahelanthropus are related to chimpanzees and gorillas respectively, but this is currently a minority view. Because of the lack of palaeo-gorillas and palaeo-chimpanzees, there has been no fossil evidence to test the molecular clock estimates of the divergence times of African apes from hominids. If the chimp-like features of the Ngorora molar represent homologies shared with chimpanzees, then it would indicate that the Pan clade has its phylogenetic roots in the latter part of the Middle Miocene some time prior to 12,5 Ma (Fig. 1). The dichotomy between chimpanzees and humans is usually estimated by molecular biologists to have occurred later than 6 Ma (Fig. 1) (Gagneux et al., 1999; Stauffer et al., 2001) and the split between chimpanzees and gorillas has been estimated at 8–9 Ma by Wrangham and Pilbeam (2001) (Fig. 2) and 7,7 Ma by Gagneux et al. (1999). The only molecular biologists who have proposed an earlier age for African ape origins are Arnason et al. (1996, 1998, 2000; Janke & Arnason, 2002) (Fig. 2), but their results are usually considered suspect by others who appear to favour appreciably later divergence times (Adachi & Hasegawa, 1995; Bailey et al., 1992; Gagneux et al., 1999; Pilbeam, 2002; Stauffer et al., 2001). The Ngorora molar is important because it provides fossil evidence that by 12,5 Ma the chimpanzee clade may already have existed as an entity distinct from the other Early and Middle Miocene African hominoids (afropithecines, proconsulines, kenyapithecines) (Harrison, 2002; Ward & Duren, 2002) and Late Miocene Asian ones (sivapithecines,

123

From Tools to Symbols

lufengpithecines) (Kelley, 2002), although it is possible that it is related to, or even belongs to the same group as European dryopithecines (Begun, 2002). Furthermore, the recognition of a possible proto-chimpanzee in the late Middle Miocene of tropical Africa suggests that the chimpanzee clade may have been present in the continent long before any of the various Late Miocene Eurasian hominoid genera that have, at one time or another, been claimed as ancestral to extant African apes and humans including Dryopithecus (Begun, 1994, 2002), Sivapithecus (and the female which used to be considered a separate genus Ramapithecus (Pilbeam, 1966)), Graecopithecus (sometimes called Ouranopithecus (de Bonis et al., 1981), Lufengpithecus (Wu, 1987) and even Oreopithecus (Hürzeler, 1960). It seems increasingly less likely that there was a re-invasion of Africa by Eurasian apes in the Late Miocene as proposed on various occasions by Begun (1994, 2002; Begun & Gülec, 1998; Heizmann & Begun, 2001) and others (de Bonis et al., 1981; Stewart & Disotell, 1998). Such hypotheses were only attractive whilst there remained huge chronological and morphological gaps Gageneux et al, 1999

Humans

Orrorin 6 Ma

Pan troglodytes central & eastern 4,7

Ngorora ape 12,5 Ma

1,6

Pan troglodytes western

2,5

7,7

Pan paniscus 12,5

Gorilla gorilla eastern

Kapsomin ape 5,9 Ma 2,5

Gorilla gorilla western Pongo pygmaeus Sumatra Calibration age not specified

1,7

Pongo pygmaeus Borneo

Figure 1 Molecular clock estimates by Gagneux et al. (1999) of the timing of dichotomies between African apes and humans, compared with key fossils from Kenya.

124

Implications of the presence of African ape-like teeth in the Miocene of Kenya

in the African fossil record between Middle Miocene African hominoids on the one hand, and extant African apes on the other. The Ngorora tooth differs markedly from the lower molars of gorillas which appear to have originated from a more bunodont, large-cusped species such as Samburupithecus kiptalami or Sahelanthropus tchadensis in which the main cusps were voluminous and were not peripherally located (Ishida & Pickford, 1997; Pickford & Ishida, 1998; Brunet et al., 2002; Wolpoff et al., 2002). This suggests that the dichotomy between chimpanzees and gorillas was probably more remote than is generally accepted by molecular biologists. Gagneux et al. (1999) estimated an age of 7,7 Ma, and Wrangham and Pilbeam (2001), an age of 8–9 Ma for this event, but the Ngorora molar would indicate a split some time prior to 12.5 Ma, but perhaps not as far back in time as the estimates of 17–18 Ma published by Arnason et al. (1996, 1998, 2000) (Fig. 2). Phylogenetic implications of the Ngorora molar In addition, the derived morphology shared by the Ngorora tooth and chimpanzee molars (peripheralised cusps, enlarged and deep occlusal basin, thin enamel) distances them both from australopithecine and human teeth, which, in general features of the Wranghan & Pillbeam, 1999

Arnason et al, 1996, 1998, 2000

Pan Gorilla: 8–9 Ma

Pan Gorilla: 17–18 Ma

Pan Homo: 6 Ma

Pan Homo: 13,5 Ma

G

KNM LU 335 6 Ma H

C Bar 91’99 12,5 Ma

Bar 1757’02 5,9 Ma

6 Ma

8–9 Ma

13,5 Ma

17–18 Ma

Figure 2 Two recent molecular clock estimates of dichotomies between African apes and humans compared with some African ape-like teeth.

125

From Tools to Symbols

crown (voluminous cusps, restricted foveae, narrow and shallow occlusal basins, thick enamel) are little different from Middle Miocene hominoids such as Kenyapithecus. In the past, the basic similarities between Plio-Pleistocene and recent hominid teeth on the one hand, and those of Kenyapithecus and Ramapithecus on the other, were employed as evidence in favour of the latter genera being hominids as well (Leakey, 1962, 1967; Pilbeam, 1966, 1972). It seems unlikely that the shared morphology observed in the Ngorora specimen and chimpanzee molars would give rise to the morphology that occurs in hominids, as this would require a reversion to a more primitive pattern. While evolutionary reversal is not impossible, it is more parsimonious to consider that the basics of the australopithecine and human molar morphological pattern were inherited from one of the Middle Miocene forms with similar molar construction, rather than to postulate a chimpanzee-like stage in between. If this is so, then the chimpanzee clade would already have been distinct from the human one by 12,5 Ma. The Ngorora specimen thus runs counter to the recent ideas of Pilbeam (1996), who published that ‘the common ancestor of humans and chimpanzees was probably chimpanzee-like, a knuckle-walker with small thin-enameled cheek teeth’ and Wrangham and Pilbeam (2001) who postulated that the ‘6 mybp ancestor ... would have been thin-enameled, knuckle-walking, and females would have had black body coats’. It is already known that 6 Ma hominids such as Orrorin tugenensis had thickenamelled molars with restricted occlusal basins and were fully bipedal (Senut et al., 2001; Pickford et al., 2002). Instead the Ngorora fossil ape tooth accords with the scenario published by Arnason et al. (1996, 1998, 2000) based on molecular evidence, of an early divergence (c. 13,5 Ma) between Pan and Homo. Biogeographic implications of the Ngorora molar In Europe, Dryopithecus occurs in terminal Middle Miocene and basal Late Miocene deposits from Spain in the west to Georgia in the east, the oldest specimen (MN 7/8, Mein, 1986) being about the same age as the Ngorora tooth. Begun (2002) reported a gap between the latest occurrence of the Eurasian hominoid (Griphopithecus), which he described as ranging in age from c. 17–15 Ma, and the earliest Dryopithecus, which he thought ranged in age from c. 11,5–9,5 Ma. Despite major morphological differences between Griphopithecus and Dryopithecus this gap allowed him to propose a rather direct ancestor–descendent relationship between the two genera. Some comments need to be made about the evolutionary scenario and the chronology proposed by Begun (2002). The age of the European dryopithecines given by Begun (2002) seems to be at variance with the results of other researchers. Heizmann and Begun (2001) position Pasalar and Candir (Turkey) in MN 5, but other authors have correlated these sites to MN 6 (Pickford et al., 2000). Thus the earliest well-dated large hominoid in Europe

126

Implications of the presence of African ape-like teeth in the Miocene of Kenya

is from MN 6, which is less than 14 Ma. Even if Engelswies is in MN 5, as reported by Begun (2002) (most authors put it in MN 6), this would make the arrival time of large hominoids in Europe about 15,5 Ma at the earliest (Pickford, 1998) and not 16,5–17 Ma as published by Begun (2002). The youngest Griphopithecus is from MN 7/8 or perhaps even MN 9 (Mein, 1986). The earliest remains attributed to Dryopithecus are from MN 7/8, aged about 12,6 Ma, and thus somewhat earlier than the age of 11,5 Ma given by Begun. Begun’s 3,5 million year gap between Griphopithecus and Dryopithecus thus seems to be an overestimate, and in fact the chronological ranges of the two genera overlap in time (Mein, 1986). This leaves almost no time for thickenameled Griphopithecus with non-peripheralised cusps to evolve into thin-enameled Dryopithecus with peripheralised cusps, as envisaged by Begun (2002, Fig. 20-16). It thus seems unlikely that Griphopithecus gave rise to Dryopithecus as proposed by this author. The Ngorora molar suggests a different scenario, in which the dryopithecine ancestor may have evolved in Africa and then migrated to Europe at the end of the Middle Miocene about 12,5 Ma.

The Kapsomin ape In 2000 an isolated hominoid central upper incisor (Fig. 3d) was found in situ at Kapsomin, the site that yielded a humerus and proximal femur of Orrorin. The tooth was initially attributed to Orrorin (Senut et al., 2001). However, comparisons with australopithecine and human teeth reveal that it differs from these in possessing a thick wedge-shaped crown in lateral view, with no sign of the deep lingual fossa that characterises the upper central incisors of Praeanthropus africanus, Australopithecus antiquus and early Homo. Nor does the tooth recall those of chimpanzees, but it does resemble upper central incisors of Gorilla. In 2002, a partial, unworn upper molar (Fig 3b) was recovered during excavation of the channel deposit that yielded the most complete femur of Orrorin (Senut et al., 2001; Pickford et al., 2002). Detailed descriptions and comparisons of these teeth are presented separately (Pickford & Senut, in press). Here we concentrate on the phylogeny and the implications of the specimens for the molecular clock estimates of the ape-human divergence. The fragment of unworn molar from Kapsomin (Bar 1757’02) (Fig. 3b) represents a species distinct from Orrorin tugenensis. Apart from its greater dimensions, it has morphology that is different from the bunodont crown with inflated main cusps, restricted trigon basin and foveae of the latter taxon. Whereas the molars of Orrorin recall those of later hominids (Australopithecus, Praeanthropus and even Homo) in overall crown shape, bunodonty, cusp inflation and basin restriction, Bar 1757’02 stands out as anomalous, with its more peripheralised cusps, widely separated metacone and hypocone, wide, obliquely oriented distal fovea, vertical and flat buccal surface, high

127

From Tools to Symbols

dentine penetrance, and thinner enamel with light wrinkling. In most of these features it is closest to Gorilla gorilla, yet it is by no means a perfect fit with this species. Metrically, the tooth is appreciably larger than any of the Orrorin molars, but it falls within the range of metric variation of the gorilla. It also falls within the range of variation of Pliocene fossil hominids, including Ardipithecus, Australopithecus and Praeanthropus. Its relationships to Sahelanthropus are not clear, as published photographs of the latter do not reveal enough detail about crown morphology. What can be said is that the known upper molars of the Chad genus are smaller than the Kapsomin tooth. The upper central incisor (Bar 1001’00) (Fig. 3d) was initially attributed to Orrorin tugenensis (Senut et al., 2001) because at the time of the discovery it was assumed that only a single hominoid was represented at the site. With the discovery of Bar 1757’02, it became clear that a second hominoid taxon was present at Kapsomin, leading to a

Figure 3 (a) Bar 91’99, right lower third molar, ?Dryopithecus (sp.) Ngorora Formation, Kenya, stereo occlusal view of cast; (b) Bar 1757’02, upper molar, large ape, Lukeino Formation, Kenya, occlusal view of cast; (c) Bar 2000’03, ape right lower molar from Cheboit, Lukeino Formation, (left) stereo occlusal and (right) mesial views; (d) Bar 1001’00, upper central incisor, large ape, Lukeino Formation, Kenya, (left) occlusal and (right) distal views of cast.

128

Implications of the presence of African ape-like teeth in the Miocene of Kenya

re-evaluation of all the specimens from the site. It now appears likely that the upper central incisor belongs to this second hominoid rather than to Orrorin. Bar 2000’03 from Cheboit, Lukeino Formation, is an unworn right lower molar lacking the roots, but preserving much of the cervical line (Fig. 3c). The apices of the two lingual cusps are bucco-lingually compressed and peripherally located. The tips of the protoconid and metaconid are 5,4 mm apart and the tooth is 10,5 mm broad at this level. The two buccal cusps are slightly in advance of the lingual ones, and there is minor buccal flare. Because of the peripheral positions of the main cusps, the occlusal basin is large and elongated. The mesial fovea is wide but mesio-distally short. The hypoconulid is small and is located slightly to the lingual side of the centre line of the tooth and in a very distal position, and as a result the tooth has an elongated trapezoidal outline (Fig 3c). Because of this the tooth could be a lower third molar. The distal fovea is thus small, but is not separated from the main occlusal basin by the crests from the hypoconulid or entoconid, as these do not reach each other. The tooth measures 12,7 mm mesio-distal by 11,1 mm bucco-lingual. This tooth is morphologically compatible with Bar 1757’02, the upper molar from Kapsomin. We consider it likely that the two specimens belong to a single taxon. The importance of these discoveries is that they reveal the presence of a second large hominoid in the Late Miocene of East Africa at the same time as early bipedal hominids. If the attribution of the Kapsomin and Cheboit teeth to a large African ape related to gorillas is correct, then this would represent the first discovery of a fossil member of the extant African ape clade.

Discussion and conclusions Four ape-like teeth from the Miocene of Kenya reveal greater similarities to extant chimpanzee and gorilla teeth than to Miocene apes and Mio-Plio-Pleistocene to recent hominids. The specimen from Ngorora (12,5 Ma) is close in size and some morphological details to Pan but also has resemblances to the European Miocene genus Dryopithecus, with which it could be congeneric, whereas the Lukeino specimens (6– 5,9 Ma), recall, but are not identical to, the teeth of gorillas. The morphology of the Ngorora tooth suggests that the Dryopithecus lineage may have evolved in Africa and then invaded Europe about 12–12,5 Ma, rather than evolving within Europe from a thick-enamelled lineage such as Griphopithecus (Begun, 2002). If it is part of the Pan clade, then it would push back the split between hominids and African apes to the Middle Miocene (Fig. 1). If this is so, then thick-enamelled apes such as Kenyapithecus possibly take on a renewed significance for throwing light on the earliest stages in the evolution of hominids, as thought by L. Leakey in the 1960s (Leakey, 1962, 1967, 1969, 1970), even though the supposedly hominid features

129

From Tools to Symbols

employed by Leakey in his proposals have subsequently been interpreted as being related to sexual dimorphism and to plesiomorphic features found in several Middle Miocene hominoids rather than to derived morphology shared with hominids (Pickford, 1985). It is more parsimonious to consider that thick-enamelled hominids descended from thick-enamelled precursors rather than to hypothesize a thin-enamelled intermediate stage, as has become fashionable (Wrangham & Pilbeam, 2001). What is required is a fresh look at the problem, including the relationships between diet on the one hand and enamel thickness and dentine penetrance on the other. If the Kapsomin and Cheboit ape teeth belong to the gorilla clade, then they would indicate that by about 6 Ma, and probably somewhat earlier, the lineage was a separate entity from the Pan + Homo clade. Taken together, the Ngorora and Kapsomin ape teeth and those of the early bipedal hominid Orrorin, plead for considerably earlier split times between the gorilla, chimpanzee and hominid clades than most molecular biologists have considered possible for the past three decades (Gagneux et al., 1999; Stauffer et al., 2001) but more in accord with the results of Arnason and his colleagues (Arnason et al., 1996, 1999, 2000; Janke & Arnason, 2002) (Fig. 2). If, however, the Kapsomin and Cheboit teeth represent a hominid rather than an African ape – a possibility that we consider unlikely – then it would indicate the presence of a second hominid in the Lukeino Formation, and would plead for a precocious dichotomy in the Hominidae prior to 6 Ma, a suggestion already made by Senut et al. (2001) on other criteria, and before the discovery of the Kapsomin ape teeth. These four ape teeth from Baringo District, Kenya, although somewhat tantalising, considering their isolation and, in one case, incomplete preservation, nevertheless reveal the presence of ape-like hominoids in East Africa during the latter part of the Middle Miocene and the Late Miocene. They thus refute the recent statement by Begun (2002) that: ‘In actual fact, none of the many late Miocene African fossil localities has any hominoids.’ When we add them to Samburupithecus from the Late Miocene of Samburu Hills, Kenya (9,5 Ma) (Ishida & Pickford, 1997), Orrorin from Lukeino, Kenya, (6–5,7 Ma) (Senut et al., 2001), Sahelanthropus from Toros-Menalla, Chad (c. 7–6 Ma) (Brunet et al., 2002), and Ardipithecus from Ethiopia (White et al., 1994), it is clear that Late Miocene Africa was not devoid of hominoids until they reintroduced themselves from Europe (Begun, 2002). Rather, it is more likely that chimp-sized Dryopithecus was originally an African lineage that invaded Western Europe about 12,5–12 Ma, and while the evidence is scanty, some of the large gorilla-sized hominoids from the Late Miocene of Greece and possibly Turkey could also be of African origin rather than autochthonously evolved descendents of Dryopithecus or Sivapithecus as envisaged by Begun (2002). Indeed, despite the relative poverty of the African Late Miocene fossil record, the new discoveries reveal that hominoids were more diverse in

130

Implications of the presence of African ape-like teeth in the Miocene of Kenya

Africa than they were in Europe (five genera now known in Africa compared to three or perhaps four in Europe).

Acknowledgements We thank members of the Kenya Palaeontology Expedition for their help in the field, in particular Mr Kiptalam Cheboi. Research permission was accorded by the Kenya Ministry of Education, Research and Technology. Funds were provided by the Collège de France (Prof. Y. Coppens), the Département Histoire de la Terre (Prof. Ph. Taquet), the French Ministry of Foreign Affairs (Commission de Fouilles) and the CNRS (Projet PICS). We are particularly keen to thank the Community Museums of Kenya (Mr E. Gitonga) for their help and cooperation and Prof. H. Ishida for inviting MP to spend time in his laboratory as visiting Professor at Kyoto University. We thank Yutaka Kunimatsu for discussions.

References Adachi, J. & Hasegawa, M. (1995). Improved dating of the human-chimpanzee separation in the mitochondrial DNA tree: heterogeneity among amino acid sites. Journal of Molecular Evolution 40, 622–628. Arnason, U., Gullberg, A., & Janke, A. (1998). Molecular timing of Primate divergences as estimated by two nonprimate calibration points. Journal of Molecular Evolution 47, 718– 727. Arnason, U., Gullberg, A., Janke, A. & Xiufeng Xu (1996). Pattern and timing of evolutionary divergences among hominoids based on analyses of complete mtDNAs. Journal of Molecular Evolution 43, 650–651. Arnason, U., Gullberg, A., Schweitzer Burguette A. & Janke A., 2000. Molecular estimates of Primate divergences and new hypotheses for Primate dispersals and the origin of modern humans. Hereditas 133, 217–228. Bailey, W.J., Hayasaka, K., Skinner, C.G., Kehoe, S., Sieu, L.C., Slightom, J.L. & Goodman, M. (1992). Reexamination of the African hominoid trichotomy with additional sequences from the primate β-globin gene cluster. Molecular Phylogenetics and Evolution 1, 97-135. Begun, D. (1994). Relations among the great apes and humans: new interpretations based on the fossil great ape Dryopithecus. Yearbook of Physical Anthropology 37, 11–63. Begun, D. (2002). European hominoids. In (W.C. Hartwig, Ed.) The Primate Fossil Record, pp. 339–368. Cambridge: Cambridge University Press. Begun, D. & Gülec, E. (1998). Restoration of the type and palate of Ankarapithecus meteai: taxonomic and phylogenetic implications. American Journal of Physical Anthropology 105, 279–314. Bishop, W.W. & Pickford, M. (1975). Geology, fauna and palaeoenvironments of the Ngorora

131

From Tools to Symbols Formation, Kenya Rift Valley. Nature 254, 185–192. Brunet, M., Guy, F., Pilbeam, D., Mackaye, H., Likius, A., Ahounta, D., Beauvilain, A., Blondel, C., Bocherens, H., Boisserie, J.-R., de Bonis, L., Coppens, Y., Dejax, J., Denys, C., Duringer, P., Eisenmann, V., Fanone, G., Fronty, P., Geraads, D., Lehmann, T., Lihoreau, F., Louchar, A., Mahamat, A., Merceron, G., Mouchelin, G., Otero, O., Campomanes, P., Ponce de Leon, M., Rage, J.-C., Sapanet, M., Schuster, M., Sudre, J., Tassy, P., Valentin, X., Vignaud, P., Viriot, L., Zazzo, A. & Zollikofer, C. (2002). A new hominid from the Upper Miocene of Chad, Central Africa. Nature 418, 145–151. de Bonis, L., Johanson, D., Melentis, J. & White, T. (1981). Variations métriques de la denture chez les Hominidés primitifs: comparaison entre Australopithecus afarensis et Ouranopithecus macedoniensis. Comptes Rendus de l’Académie des Sciences de Paris, 292, 373–376. Deino, A., Tauxe, L., Monaghan, M. & Drake, R. (1990). Single crystal 40Ar/39Ar age calibration of the litho- and paleomagnetic stratigraphies of the Ngorora Formation, Kenya. Journal of Geology, 98, 567–587. Gagneux, P., Wills, C., Gerloff, U., Tautz, D., Morin, P., Boesch, C., Fruth, B., Hohmann, G., Ryder, O. & Woodruff, D. (1999). Mitochondrial sequences show diverse evolutionary histories of African hominoids. Proceedings of the National Academy of Science 96, 5077–5082. Harrison, T. (2002). Late Oligocene to Middle Miocene catarrhines from Afro-Arabia. In (W.C. Hartwig, Ed.) The Primate Fossil Record, pp. 311–333. Cambridge: Cambridge University Press. Heizmann, E. & Begun, D. (2001). The oldest Eurasian hominoid. Journal of Human Evolution 41, 463–481. Hürzeler, J. (1960). The significance of Oreopithecus in the genealogy of man. Triangle 4, 164– 174. Ishida, H. & Pickford, M. (1997). A new late Miocene hominoid from Kenya: Samburupithecus kiptalami gen. et sp.nov. Comptes Rendus de l’Académie des Sciences de Paris 325, 823–829. Janke, A. & Arnason, U. (2002). Primate divergence times. In (B. Galdikas, N. Briggs, L. Sheeran, G. Shapiro & J. Goodall, Eds) All Apes Great and Small. Vol. 1: African Apes, pp. 18–25. New York: Kluwer Academic/Plenum Publishers. Kelley, J. (2002). The hominoid radiation in Asia. In (W.C. Hartwig, Ed.) The Primate Fossil Record, pp. 369–384. Cambridge: Cambridge University Press. Leakey, L.S.B. (1962). A new lower Pliocene fossil primate from Kenya. Annals and Magazine of Natural History, Series 13, 4, 689–696. Leakey, L.S.B., 1967. An early Miocene member of Hominidae. Nature 213, 155–163. Leakey, L.S.B. (1969). Fort Ternan hominid. Nature 222, 1202. Leakey, L.S.B. (1970). The relationships of African apes, man and Old World monkeys. Proceedings of the National Academy of Science 67, 746–748. Mein, P. (1986). Chronological succession of hominoids in the European Neogene. In (J. Else &

132

Implications of the presence of African ape-like teeth in the Miocene of Kenya P. Lee, Eds) Primate Evolution, pp. 59–70. Cambridge: Cambridge University Press. Pickford, M. (1985). A new look at Kenyapithecus based on recent collections from Western Kenya. Journal of Human Evolution 14, 113–143. Pickford, M. (1986). The geochronology of Miocene higher primate faunas of East Africa. In (J. Else & P. Lee, Eds) Primate Evolution, pp. 19–45. Cambridge: Cambridge University Press. Pickford, M. (1998). Dating of the Neogene Old World anthropoid fossil record: essential base for phylogenetic analysis, biogeography and palaeoecology. Primatologie 1, 27–92. Pickford, M. & Ishida, H. (1998). Interpretation of Samburupithecus, an upper Miocene hominoid from Kenya. Comptes Rendus de l’Académie des Sciences de Paris 326, 299–306. Pickford, M., Senut, B., Gommery, D. & Treil, J. (2002). Bipedalism in Orrorin tugenensis revealed by its femora. Comptes Rendus Palevol 1, 191–203. Pilbeam, D. (1966). Notes on Ramapithecus, the earliest known hominid, and Dryopithecus. American Journal of Physical Anthropology 25, 1–5. Pilbeam, D. (1972). The Ascent of Man. New York: Macmillan. Pilbeam, D. (1996). Genetic and morphological records of the Hominoidea and hominid origins: A synthesis. Molecular Phylogenetics and Evolution 5, 155–168. Pilbeam, D. (2002). Perspectives on the Miocene Hominoidea. In (W.C. Hartwig, Ed.) The Primate Fossil Record, pp. 303–310. Cambridge: Cambridge University Press. Senut, B., Pickford, M., Gommery, D., Mein, P., Cheboi, K. & Coppens, Y. (2001). First hominid from the Miocene (Lukeino Formation, Kenya). Comptes Rendus de l’Académie des Sciences deParis 332, 137–144. Stauffer, R.L., Walker, A., Ryder, O., Lyons-Wailer, M. & Hedges, S. (2001). Human and ape molecular clocks and constraints on paleontological hypotheses. The American Genetic Association 92, 469–474. Stewart, C.B. & Disotell, T.R. (1998). Primate evolution: in and out of Africa. Current Biology 8, 582–588. Ward, S. & Duren, D. (2002). Middle and Late Miocene African Hominoids. In (W.C. Hartwig, Ed.) The Primate Fossil Record, pp. 385–397. Cambridge: Cambridge University Press. White, T., Suwa, G. & Asfaw, B. (1994). Australopithecus ramidus, a new species of early hominid from Aramis, Ethiopia. Nature 371, 306-312. Wolpoff, M., Senut, B., Pickford, M. & Hawks, J. (2002). Palaeoanthropology (communication arising): Sahelanthropus or Sahelpithecus? Nature 419, 581–582. Wrangham, R. & Pilbeam, D. (2001). African apes as time machines. In (B. Galdikas, N. Briggs, L. Sheeran, G. Shapiro & J. Goodall, Eds) All Apes Great and Small. Vol. 1: African Apes, pp. 5–17. New York: Kluwer Academic/Plenum Publishers. Wu, R., 1987. A revision of the classification of the Lufeng great apes. Acta Anthropologica Sinica 6, 263–271.

133

Dawn of hominids: understanding the ape-hominid dichotomy Brigitte Senut Département Histoire de la Terre, USM 0203 du Muséum national d’Histoire naturelle & UMR 5143, PICS 1048 (CNRS), Case 38, 57, rue Cuvier, 75005 Paris, France

Abstract Until 2000, several scientists considered Ardipithecus ramidus from Ethiopia, at 4,5 Ma old (million years) to be a very early ancestor of later hominids, australopithecines and hominines. Based on molecular clocks, the dichotomy between apes and humans was supposed to be situated around 6 Ma, and this discovery was viewed as supporting the molecular data. But, at the time, the hominoid fossil record between 9,5 and 4,5 Ma was very poor (only a few fragments of bones, maxillae and mandibles were known – all from Kenya) and new material was clearly needed. In 2000, field work by the Kenya Palaeontology Expedition (a Franco-Kenyan cooperative project) led to the discovery of Orrorin tugenensis in the Tugen Hills (Kenya), which dates to between 6,0 and 5,7 Ma. The species was originally represented by eleven specimens: a mandible in two pieces, several isolated teeth and postcranial bones (three fragmentary femora, a distal humerus and a proximal manual phalanx). Since 2000, the number of specimens has doubled, including more teeth and postcranials. The teeth show a complex mixture of primitive ape-like features (such as the presence of a low distal shoulder on the upper canine crown, a mesiodistally elongated upper canine) and derived hominid features (the molars – small, squarish with thick enamel and with almost vertical lingual walls – recall those of hominids, a lower canine with a distal tubercle and a mesial marginal ridge, and the absence of a C-P3 diastema). The femoral features show clear evidence of adaptation to bipedalism (presence of an obturator externus groove, an elongated femoral neck, an anteriorly twisted head, and the pattern of the cortical distribution in the femoral neck among others). However, Orrorin exhibits some differences from the australopithecines in the morphology of the femoral neck, the position of the lesser trochanter and projection of the femoral head; but also shares some features with them (asymmetrical cortical distribution in the neck, length of the neck, presence of the m. obturator externus groove, among others). The humerus and phalanx imply arboreality in Orrorin. Dentally, Orrorin appears to be a very early hominid, a conclusion reinforced by the adaptation to bipedalism. However, its bipedalism appears to be different from that of australopithecines, which we consider not to be direct human ancestors. Orrorin is microdont (small teeth associated with a large skeleton) and australopithecines are megadont (large teeth associated with small skeleton). In July 2001 a new sub-species of Ardipithecus ramidus, A. r. kadabba (5,7 Ma) was announced in Ethiopia, recently raised to species rank (A. kadabba); however, the evidence for bipedalism does not appear to be

134

Dawn of hominids conclusive. A. ramidus is reported to be a biped, but the evidence in support of this conclusion has not been published in the scientific literature. The only published lower limb bone of A. kadabba is a pedal phalanx, which matches the morphology of that of Lucy. Its strong curvature suggests a possible adaptation to climbing. Later the same year, the discovery of a hominoid skull in Chad 6–7 Ma old, Sahelanthropus tchadensis, was announced, the hominid status of which has been challenged. All these new discoveries show that to understand the dichotomy between the African great apes and hominids it is no longer possible to concentrate our research on modern hominoids (including humans) and Pliocene fossils, but the Miocene apes must be considered; it also becomes clear that further evidence is required from Upper Miocene deposits of Africa (11–8 Ma).

Résumé Jusqu’à l’an 2000, plusieurs scientifiques considéraient que Ardipithecus ramidus trouvé en Ethiopie dans des terrains vieux de 4,5 Ma représentait un ancêtre très lointain des hominidés ultérieurs, Australopithécinés et Homininés. La dichotomie entre les grands singes et les hommes basée sur les horloges moléculaires était censée se situer vers 6 Ma et cette découverte confortait les données moléculaires. Mais, à l’époque, les données fossiles sur la période comprise entre 9,5 et 4,5 Ma étaient pratiquement inexistantes: seulement quelques fragments d’os, de maxillaire et de mandibules étaient connus au Kenya. En 2000, lors des travaux de terrain de la Kenya Palaeontology Expedition (un projet de coopération franco-kenyan), furent découverts les restes d’Orrorin tugenensis dans les Collines Tugen (Kenya), dont l’âge est compris entre 6,0 et 5,7 Ma. L’espèce était représentée par 11 spécimens: une mandibule en deux morceaux, plusieurs dents isolées et des restes postcrâniens (3 fragments fémoraux, une extrémité distale d’humérus et une phalange proximale de la main). Depuis 2000, le nombre de spécimens a doublé, incluant plusieurs autres dents et une phalange distale de pouce. Les dents présentent un mélange complexe de caractères primitifs de grands singes (comme la présence d’un épaulement distal bas à la canine supérieure, une canine supérieure allongée mésiodistalement) et de caractères dérivés d’hominidés (les molaires – petites et carrées à l’émail épais aux murs linguaux verticauxrappellent celles des hominidés, une canine inférieure avec un tubercule distal et une crête mésiale marginale et l’absence d’un diastème C-P3). Les caractères fémoraux apportent la preuve d’une adaptation à la bipédie (présence d’une gouttière pour le m. obturateur externe, col fémoral allongé, tête tordue vers l’avant, mode de distribution de l’os cortical dans le col fémoral entre autres). Toutefois, Orrorin montre des différences avec les Australopithèques dans la morphologie du col fémoral, la position du petit trochanter et la projection de la tête fémorale; mais il présente aussi quelques ressemblances (distribution asymétrique de l’os cortical dans le col, longueur du col, présence d’une gouttière pour le muscle obturateur externe, parmi d’autres). Les caractères de l’humérus et de la phalange indiquent qu’Orrorin était aussi arboricole. Ainsi, Orrorin est un hominidé très ancien par ses traits dentaires, conclusion renforcée par son adaptation à la bipédie. Toutefois, cette bipédie apparaît différente de celle des Australopithèques, que nous ne considérons pas comme des ancêtres directs des hommes. Orrorin est un être microdonte (petites dents associées à un grand squelette) et les Australopithèques des êtres mégadontes (grandes dents associées à un petit squelette). En juillet 2001, une nouvelle sous-espèce d’Ardipithecus ramidus, A. r. kadabba (5,7 Ma) fut annoncée en

135

From Tools to Symbols Ethiopie, élevée récemment au rang d’espèce (A. kadabba); cependant, les caractères présentés ne permettent pas de conclure avec certitude sur la présence d’une bipédie chez ce taxon. A. ramidus est considéré comme bipède, mais les données qui le confirmerait n’ont pas encore été publiées. Le seul reste de membre postérieur d’A. kadabba publié est une phalange de pied, qui ressemble beaucoup à celle de Lucy. Sa forte courbure suggère une possible adaptation au grimper. Plus tard, la même année, la découverte d’un crâne d’hominoïde vieux de 6 à 7 millions d’années était annoncée au Tchad, Sahelanthropus tchadensis, dont le statut d’hominidé est discuté. Toutes ces nouvelles découvertes montrent que pour comprendre la divergence entre les grands singes africains et les hommes, il n’est plus possible aujourd’hui de concentrer nos recherches sur les hominoïdes modernes (incluant les hominidés) et les fossiles pliocènes, mais il faut prendre en considération les grand singes miocènes. Il apparaît aussi évident que de nouveaux fossiles doivent être découverts dans les dépôts du Miocène supérieur de l’Afrique (11 à 8 millions d’années).

Introduction In the 1960s and 70s, Ramapithecus and Kenyapithecus from the Middle Miocene were considered by most scientists to be the earliest hominids (Leakey, 1961/1962, 1967; Simons, 1961, 1969; Simons & Pilbeam, 1965; Andrews, 1971; Aguirre, 1972, 1975). However, some scholars disagreed (Genet-Varcin, 1969) and studies of the sexual dimorphism and new discoveries of kenyapithecids in Maboko (Kenya) led to the conclusion that Miocene apes were highly dimorphic (de Bonis & Melentis, 1977, 1978; Pickford, 1986a; Pickford & Chiarelli, 1986; Benefit & McCrossin, 1989; Kelley, 1993) and Ramapithecus and its African counterpart Kenyapithecus were, in fact, females of apes and not hominids (Pickford, 1985, 1986a, 1986b). Up to 1994, very few fossil hominoids aged between 13 and 4,5 Ma were known from Africa: a few from East Africa (Senut, 1998, and see the bibliography included), and only one, Otavipithecus namibiensis, from southern Africa (Conroy et al., 1992). This is one reason why the quest for the earliest human ancestor has been in such turmoil; all the scenarios on human evolution published up to then being based solely on the australopithecines and later hominids, the oldest Pliocene fossils being systematically regarded as the stem species of our lineage. However, to understand the evolution of human beings, it is crucial to consider the Miocene data, even if they are limited, because they provide information on hominoid history, including our own. During that time Eurasian fossil hominoids were usually considered to be relatives, but not necessarily ancestors, of hominids. With the development of molecular biology and of the molecular clock, it has been assumed for many years that the divergence between the African apes and the hominids was around 6 Ma or younger (Pilbeam, 1996; Wrangham & Pilbeam, 2001, and see bibliography). However, several new data (Arnason et al., 2001; Marks, 2002,

136

Dawn of hominids

and see bibliography) and new field discoveries (Pickford & Senut, this volume; Senut & Pickford, 2004) suggest that this date is too young and that the common ancestor to African apes and humans should be searched for in the period from 10 to 8 Ma, and maybe even earlier.

Upper Miocene and basal Pliocene hominoids known before 1994 Few specimens were known before 1994, and their taxonomic status has changed considerably. Ngorora Five hominoid specimens were discovered at Ngorora in the 1970s: an unworn upper molar, two canines, a premolar and an upper incisor (Bishop & Chapman, 1970; Bishop & Pickford, 1975). The upper M2 has been considered by Hill (1994) to be close to chimpanzees and he suggested that it should be placed in a new genus. However the shape and the enamel thickness suggest that the specimen could belong to Kenyapithecus, as it cannot be distinguished from that genus. This was already suggested by Leakey (1970) and later by Ishida and Pickford (1998). Samburu Hills In 1982, Ishida and his team discovered a fragmentary hominoid left maxilla (Fig. 1) in the 9,5 Ma old Namurungule Formation, Samburupithecus kiptalami (Ishida et al., 1984; Pickford et al., 1984; Ishida & Pickford, 1998; Pickford & Ishida, 1998). It has the alveolus of the canine, and bears the two premolars and three molars. The specimen was originally called Motopithecus, which is an invalid name as it was published without any diagnosis and no type-species designated. Several of its features recall Gorillinae: the size, the anterior position of the zygomatic arch, the pneumatisation of the maxilla. Other features differ from gorillas: zygomatic arch located low above the M2; deep and arched palate; thick enamel; puffy cusps on the molars with a well-marked cingulum on the lingual side of the teeth; mesio-distally elongated premolars; M3 larger than M2 which is larger than M1. Samburupithecus has been considered either as an ancestor of Hominidae, or as an ancestor of African great apes, the latter hypothesis being more in favour today. Hominoidea indet. from Lukeino A lower molar originally said to be an M1 or M2, but in fact an M3, discovered in 1974 in 6 Ma old sediments from Lukeino (Kenya), was attributed to Hominidae by Andrews (in Pickford, 1975). Corruccini and McHenry (1980) pointed out resemblances to chimpanzees. But in 1988, Hill and Ward (1988) considered the isolated tooth as a very early hominid that could be a common ancestor to humans and chimpanzees;

137

From Tools to Symbols

Figure 1 Drawing of the Samburupithecus kiptalami maxilla, Namurungule Formation, Kenya (after Ishida et al., 1984).

however given the limitations of the sample, its classification was difficult. Mabaget (= Chemeron Northern Extension) This site in the Baringo area, which belongs to the Mabaget Formation (5.1 Ma), yielded a proximal partial humeral extremity in 1980. Originally attributed to Australopithecus afarensis (Pickford et al., 1983), the morphology of the damaged bicipital groove does not permit us to conclude in favour of an australopithecine or a hominoid sensu lato (Senut, 1983). However, Hill and Ward (1988) maintain the attribution to Australopithecus afarensis. Tabarin At Tabarin, deposits belonging to the Mabaget Formation yielded a fragmentary mandible known as the Tabarin mandible in 1984 (Hill, 1985), 4,5 Ma old (Hill et al., 1985). The wide and low base of the mandibular corpus and the morphology of its lingual surface led Hill (1985) and Hill and Ward (1988) to attribute the specimen to Australopithecus afarensis. But the mandible is smaller than the ones usually attributed to Australopithecus afarensis; moreover, its metrics seem closer to M2 of Proconsul africanus (Hill, 1994). In 1989, Ferguson selected the mandible as the holotype of

138

Dawn of hominids

Australopithecus praegens. Hill (1998) suggested that the specimen might belong to Ardipithecus ramidus on the basis of the thin enamel and the narrowness of the molars. The mandible of Tabarin was thus taken to be the earliest hominid ancestor as it is slightly older than the specimens from Aramis in Ethiopia. Lothagam In 1967, at Lothagam (South-West Turkana), a fragmentary hominoid mandible with the alveolus of P4, the M1 and the distal root of the M2, was found by Patterson (Patterson et al., 1970). For many years it was said to be c. 7 Ma old, but it is actually aged between 4,2 and 5 Ma (McDougall & Feibel, 2003). It was named Australopithecus cf. africanus on the basis of its size and morphology. In his study, Eckhardt (1977) concluded that the specimen was close to the dryopithecines. Later, White (1986) suggested that the mandible might belong to Australopithecus afarensis on the basis of the morphology of the inferior transverse torus and the depression on the lateral surface of the bone. The same assignment was later suggested by Hill et al. (1992). However, the specimen being so fragmentary, it appears more cautious to attribute it to Hominoidea indet. Kanapoi In 1966, Patterson discovered a hominid distal humerus at Kanapoi in Kenya. At this time, very little was known in East Africa about the upper limb bones of early hominids and, its general aspect being human, it was logically attributed to Australopithecus (Patterson & Howells, 1967). When the specimens from Koobi Fora and Hadar were found, better comparisons could be undertaken and it appeared that the morphology of the Kanapoi bone was quite human-like (Senut, 1979, 1992). More recently, on the basis of biometrical studies, several authors have concluded that it was closer to australopithecines (Lague & Jungers, 1996; Bacon, 1999); however, these works deal mainly with statistical means of values. The human-like morphology was, however, confirmed by Leakey et al. (1995) with the new field discoveries, described below.

The new discoveries The sample of hominoids from the time period between 10 and 4 Ma ago was thus very small up to 1994. Subsequently, new discoveries have been made in East Africa, namely in Ethiopia and Kenya.

139

From Tools to Symbols

Ardipithecus ramidus (= Ardipithecus ramidus ramidus) In the fall of 1994, White and his co-authors published a new species – Australopithecus ramidus – of an earliest hominid interpreted as the ancestor of the later hominids (Australopithecus and Homo). A few months later they published a corrigendum, referring their material to a different genus, Ardipithecus. The material was discovered at Aramis in Ethiopia in 4,4 Ma old deposits (White et al., 1994, 1995; WoldeGabriel et al., 1994). It differs noticeably from Australopithecus by its lesser postcanine megadontia, upper and lower canines larger relative to postcanine teeth, lower first deciduous molar narrow and obliquely elongated with large protoconid, small and distally placed metaconid, no anterior fovea, absolutely and relatively thinner enamel on canines and molars, P3 strongly asymmetrical, and temporo-mandibular morphology. The species exhibits several hominid features such as the height of the shoulder on the canines and the wear pattern. However, when all the features are taken into account, and especially as suggested by its creators, most of the comparisons fall within the variation of modern Pan paniscus (White et al., 1994) Ardipithecus is reported to be a bipedal creature, but up to now very few postcranial studies have been published and no good evidence of bipedality in this species has been evinced to date. The data gained from sedimentological, botanical and faunal evidence at Aramis suggests a woodland to forest setting: the fauna is dominated by colobine monkeys (30 per cent of collected vertebrates) and a medium-sized kudu, Tragelaphus sp. The plant remains collected consist of thousands of Canthium seeds, which is a common plant in African woodlands and forests (WoldeGabriel et al., 1994). Australopithecus anamensis (= Praeanthropus africanus) In the summer of 1995, a new species of Australopithecus, A. anamensis, was published (Leakey et al., 1995, 1998) which we consider to be a synonym of Praeanthropus africanus (Senut, 1995, 1996). The material discovered at Kanapoi and Allia Bay is 4,2–3,2 Ma old, the specimens found at Kanapoi being older. Several postcranial remains have been found and they are human-like, just as with the humerus found in 1966 by Patterson. Orrorin tugenensis In the fall of 2000, in the Tugen Hills in Kenya, the remains of several individuals (Fig. 2) attributed to a new genus and species, Orrorin tugenensis, were discovered (Senut et al., 2001). They were found in 6 Ma old deposits (Pickford & Senut, 2001; Sawada et al., 2002) and they provide evidence of bipedalism in remote times (Pickford

140

Dawn of hominids

et al., 2002). The most interesting aspect of the discovery is that in its posterior teeth, the specimen appears to be hominid-like, whereas in its anterior teeth it appears more ape-like; but these ape-like features are primitive, as they are retained from a Miocene precursor. The Kapsomin remains exhibit a mosaic of primitive and derived features. Among the primitive ones may be mentioned the presence of two transverse offset

Figure 2 Remains of Orrorin tugenensis discovered in 2000 (Lukeino Formation, Tugen Hills, Kenya).

141

From Tools to Symbols

roots in the lower P4, a low distal shoulder on the upper canine and a mesiodistally elongated upper canine. But in the morphology of the molars, Orrorin is closer to humans, especially in the verticality of the lingual walls. There is no strong molar crenulation and it has a mesial crown shoulder at midcrown of the lower canines, two traits which differ from chimpanzees. Orrorin shares some derived features with Homo such as a vertical symphysis, no diastema between the canine and the first premolar in the mandible, a lower canine with a distal tubercle and a mesial marginal ridge. Postcranially, Orrorin exhibits adaptations to bipedality. The best evidence is yielded by the femur (Pickford et al., 2002). It shows a combination of plesiomorphic as well as apomorphic hominid features, but no apomorphic ape-like features. In particular it does not exhibit a deeply penetrating trochanteric fossa as in chimpanzees. Among the ape-like features are the platymeria and the position and course of the spiral line below the lesser trochanter. These ape-like features can be found in humans and australopithecines, but are not exclusive to them and are considered to be plesiomorphic. Femoral platymeria is usually considered to be a human feature, but it is often seen in Miocene apes such as Ugandapithecus (Gommery et al., 1998) and Proconsul nyanzae (Ward et al., 1993). Medial projection of the lesser trochanter also occurs in several Miocene apes such as Ugandapithecus (Gommery et al., 1998), Proconsul (Bacon, 2001; Ward et al., 1993) and Kenyapithecus africanus (Le Gros Clark & Leakey, 1951); the morphology seen in australopithecines (posterior projection) would be a derived condition. The femur of Orrorin also exhibits clear human features, some of which occur in australopithecines: presence of an obturator externus groove, shallow and wide superior notch, well-marked gluteal tuberosity, elongated neck, antero-posteriorly compressed femoral neck, asymmetric distribution of the cortical bone in the femoral neck typical of the pattern seen in hominids, cortex thinner superiorly and thicker inferiorly as in humans, but not like chimpanzees, in which the cortex is thick in all directions (Ohman et al., 1997). Even if the pattern of cortex distribution in Orrorin is not identical to that of humans and australopithecines, it is closer to them and quite different from that of chimpanzees. Moreover the femoral head is twisted anteriorly, more like the morphology seen in humans, which differs from australopithecines, where it is posteriorly twisted. The size of the femoral head relative to the shaft diameter also recalls the morphology seen in hominids. The combination of all these features evidences the adaptation to bipedalism of Orrorin. However, other parts of the skeletal remains suggest climbing adaptations (Senut et al., 2002), especially in the humeral morphology and the phalanges. The humeral shaft shows a clear distal antero-posterior flattening, a vertical rectilinear supraepicondylar crest and an asymmetry of the distal pillar. These features are present in australopithecines and apes such as chimpanzees, and have been related

142

Dawn of hominids

to climbing (Coppens & Senut, 1991; Senut, 1981; Susman et al., 1984; Schmid, 1983). The proximal phalanx found at Kapcheberek is curved supero-inferiorly and elongated, two features which reflect an arboreal life. The distal thumb phalanx resembles those of South African Plio-Pleistocene hominids and those of humans (Gommery & Senut, 2002); in its general dimensions it is close to the human variation, based on a comparison with Susman’s (1988) data. It is more slender than the distal phalanx of Plio-Pleistocene hominids. However, the morphology of the phalanx, which has been previously related to tool-making behaviour, is probably not due to this behaviour but is probably related to a precision grip in an arboreal environment. Altogether, the data gained from the postcranial elements suggest arboreal adaptation such as climbing. The association between bipedalism and climbing is not new in the fossil record – it has already been seen in australopithecines, and could be considered an ancestral behaviour. Ardipithecus kadabba (= Ardipithecus ramidus kadabba) A few months after the publication of Orrorin, Haile-Selassie (2001) created a sub-species of Ardipithecus ramidus, A. r. kadabba, older than the other species, A. r. ramidus. It has recently been promoted to the specific level (Haile-Selassie et al., 2004). The specimens were found in strata 5,54–5,77 Ma old, slightly younger than the beds in which Orrorin was found (WoldeGabriel et al., 2001). The preserved material exhibits some features close to the previously described species, and some that are different. Postcranial elements are poorly represented, and the deduction that A. kadabba and A. ramidus were bipeds is not supported by clear evidence. Bipedalism in the recently discovered sub-species is evidenced only by a pedal phalanx supposedly similar to that of Lucy. However, several authors have shown that the morphology of the pedal phalanx suggests climbing adaptations. The reconstruction of the environment of Ardipithecus kadabba based on oxygen isotope ratios indicates cool, high altitude and/or humid habitats, and the fossil assemblages suggest mainly wet and closed woodland/forest habitats (WoldeGabriel et al., 2001). Sahelanthropus tchadensis The most recently announced discovery comes from Chad; in 2002, Brunet and his team announced the discovery of the earliest hominid, Sahelanthropus tchadensis, found in strata 6–7 Ma old (Brunet et al., 2002; Vignaud et al., 2002). The features on which they base their assignment of the species to a hominid are: the flattening of the face, the smallness of the canine, the position of the foramen magnum, and the apical wear in the canines. The sexual status is based on the size of the supraorbital torus. The first three features were classically used to determine the hominid status of a species, but it has

143

From Tools to Symbols

been shown by several authors since the early 1980s that these traits are also related to the sex of a specimen. Basically, the females of apes (modern and Miocene) exhibit shorter faces and smaller canines than males; this means that these features are not necessarily a hominid mark (Wolpoff et al., 2002). The mandible shows markedly offset roots for the P3 and a wide space for a big canine, which are classical features for male apes. The upper incisor is also quite gorilla-like. Finally, the anterior position of the foramen magnum is not due exclusively to bipedal locomotion as it varies with increase in brain size (Biegert, 1963).

New hominoids from the middle and upper Miocene of Kenya More recently, a lower molar was found in the Ngorora Formation (Pickford & Senut, 2004a; this volume). It exhibits several ape-like features incompatible with the anatomy of the upper molar found in 1970. It appears more like Dryopithecus and modern chimpanzee, Pan troglodytes. This implies that the ancestors of chimpanzees may have been present in the Middle Miocene and not in the Pliocene, and that molecular phylogenies need to be dramatically revised. An incomplete upper molar of a hominoid found in October 2002 at Kapsomin cannot be attributed to Orrorin tugenensis on the basis of its size (much bigger than the hominid) and morphology (occlusal basin volume, cusp morphology, presence of a buccal slit) (Pickford & Senut, 2004a; this volume). It is close in size to gorilla molars. In the previous hypodigm of Orrorin tugenensis an upper central incisor was included which is very large compared with the other teeth. Its morphology is also very different from that of the hominid. In distal view the wedge-shape aspect recalls clearly the morphology seen in gorilla. Did a gorilla-like creature coexist with Orrorin? The evidence being limited, we will probably have to wait until more material is recovered to answer this question.

Conclusions The past decade has seen a great improvement in knowledge of fossils found in the so-called ‘black hole’, the time gap between 10 and 4,5 Ma ago. It is clear today that to be able to understand the earliest stages of the hominid lineage, we cannot restrict our studies exclusively to the Plio-Pleistocene hominids and modern humans and chimpanzees. It is essential to look at the Miocene hominoid record as well. If we use only neontological evidence to define the features, then we eliminate the wealth of evidence that can be gleaned from the Miocene fossils, which after all are likely to contain the ancestors of hominids. When the Miocene apes are taken into account, several features which have traditionally been said to be highly derived in humans or in modern apes appear to be primitive and retained by all or retained in the chimp

144

Dawn of hominids

line or human line. Several genera are well placed chronologically to claim the title of ancestral hominid (see Fig. 3 for a tentative phylogeny); however, only Orrorin is known to display clear adaptations to bipedalism. Despite the fact that Ardipithecus has been claimed to be hominid, on the basis of the published data we have to remain cautious. Remains of Sahelanthropus interpreted as belonging to an early hominid appear to be more ape-like (Wolpoff et al. 2002). However, all the known Late Miocene material is fragmented and/or deformed, and this is one of the reasons why it is so difficult to clarify the dichotomy between apes and humans. A clear conclusion emerges: if we want to understand the ape/human dichotomy, we have to prospect more in the African Upper Miocene, a fact which was neglected for a long time due to belief in the claims of the molecular biologists who envisaged a Pliocene dichotomy. It appears that the dichotomy is much more ancient than generally thought. In terms of environmental approaches, it appears that the environment in which the very early hominids lived was more forested than thought by many researchers; and finally bipedalism, a key hominid feature (Coppens & Senut, 1991), did not emerge in a dry environment, but in a more humid one (Pickford & Senut 2001, 2004b; WoldeGabriel et al. 2001; Haile-Selassie et al. 2004). Finally, the possible European origins of hominids (Begun, 2002, de Bonis & Koufos, 1994) are

Figure 3 Proposed phylogeny of Mio-Plio-Pleistocene hominoids (after Senut & Pickford, 2004).

145

From Tools to Symbols

not confirmed as yet with the new African data. Only new discoveries will permit a better understanding of the modalities of the divergence between humans and apes.

Acknowledgements I thank Lucinda Backwell and Francesco d’Errico for inviting me to the Round Table ‘From Tools to Symbols’, and for their dedicated help in editing the manuscript. Sincere thanks are due to Dr Martin Pickford for his suggestions and discussions. Special thanks to Prof. Yves Coppens who has been supporting my research for so many years.

Bibliography Aguirre, E. (1972). Les rapports phylétiques de Ramapithecus et de Kenyapithecus et l’origine des Hominidés. L’Anthropologie (Paris) 76 (5–6), 501–523. Aguirre, E. (1975). Kenyapithecus and Ramapithecus. In (R.H. Tuttle, Ed.) Paleoanthropology – Morphology and Palaeoecology, pp. 99–104. The Hague: Mouton. Andrews, P. (1971). Ramapithecus wickeri mandible from Fort Ternan, Kenya. Nature 230, 192– 194. Arnason, U., Gullberg, A., Schweitzer, A. Burguette & Janke, A. (2001). Molecular estimates of primate divergences and new hypotheses for primate dispersals and the origin of modern humans. Hereditas 133, 217–228. Bacon, A-M. (1999). Analyse de la variabilité de taille et de forme de l’extrémité distale de l’humérus chez les grands singes africains par la morphométrie géométrique. Comptes rendus de l’Académie des Sciences de Paris, IIa, 329, 675–681. Bacon, A-M. (2001). La locomotion des primates du Miocène d’Afrique et d’Europe. Cahiers de Paléoanthropologie. Paris: CNRS, 1–126. Begun, D. (2002). European hominoids. In (W.C. Hartwig, Ed.) The primate fossil record. Cambridge: Cambridge University Press, pp. 339–368. Benefit, B. & McCrossin, M.L. (1989). New primate fossils from the Middle Miocene of Maboko Island, Kenya. American Journal of Physical Anthropology 18, 493–498. Biegert, J. (1963). The evaluation of characters of the skull, hands and feet for primate taxonomy. In (S.L. Washburn, Ed.) Classification and human evolution. Chicago: Aldine, pp. 116–145. Bishop, W.W. & Chapman, G.R. (1970). Early Pliocene sediments and fossils from the northern Kenya Rift Valley. Nature 226, 914–918. Bishop, W.W. & Pickford, M. (1975). Geology, fauna and palaeoenvironments of the Ngorora Formation, Kenya Rift Valley. Nature 254, 185–192. Bonis, L. de & Melentis, J. (1977). Les primates hominoïdes du Vallésien de Macédoine Grèce). Etude de la mâchoire inférieure. Geobios 10, 849–885. Bonis, L. de & Melentis, J. (1978). Les primates hominoïdes du Vallésien de Macédoine Grèce).

146

Dawn of hominids Etude de la mâchoire supérieure. Annales de Paléontologie (Vertébrés-Invertébrés) 64, 185– 202. Bonis, L. de & Koufos, G. (1994). Our ancestor’s ancestor: Ouranopithecus is a Greek link in human ancestry. Evolutionary Anthropology 3 (3), 75–83. Brunet, M., Guy, F., Pilbeam, D., Mackaye, H., Likius, A., Ahounta, D., Beauvilain, A., Blondel, C., Bocherens, H., Boisserie, J-R., De Bonis, L., Coppens, Y., Dejax, J., Denys, C., Duringer, P., Eisenmann V., Fanone, G., Fronty, P., Geraads, D., Lehmann, T., Lihoreau, F., Louchar, A., Mahamat, A., Merceron, G., Mouchelin, G., Otero, O., Campomanes, P., Ponce De Leon, M., Rage, J-C., Sapanet, M., Schuster, M., Sudre, J., Tassy, P., Valentin, X., Vignaud, P., Viriot, L., Zazzo, A. & Zollikofer, C. (2001). A new hominid from the Upper Miocene of Chad, Central Africa. Nature 418, 145–151. Coppens, Y. & Senut, B., Eds (1991). Origine(s) de la bipédie chez les Hominidés. Cahiers de Paléoanthropologie. Paris: CNRS. Conroy, G.C., Pickford, M., Senut, B., Van Couvering, J.A. & Mein, P. (1992). Otavipithecus namibiensis, first Miocene hominoid from Southern Africa. Nature, 356, 144–148. Corruccini, R.S. & McHenry, H.M. (1980).Cladometric analysis of Pliocene Hominids. Journal of Human Evolution 9, 209–221. Eckhardt, R.B. (1977). Hominid origins: problem of Lothagam. Current Anthropology 18, 356. Ferguson, W.W. (1989). Taxonomic status of the hominid mandible KNM TH 13150 from the Middle Pliocene of Tabarin in Kenya. Primates 30, 383–387. Genet-Varcin, E. 1969. A la recherche du primate ancêtre de l’homme. Paris: Boubée. Gommery, D. & Senut, B. 2002. Orrorin tugenensis distal thumb phalanx. In (E. Gitonga, B. Senut, H. Ishida, Y. Sawada & M. Pickford, Eds) From Samburupithecus to Orrorin : origins of hominids. Geological and palaeontological background. International Workshop at Bogorio, Abstracts, 2. Gommery, D., Senut, B. & Pickford, M. (1998). Nouveaux restes postcrâniens d’Hominoidea du Miocène inférieur de Napak (Ouganda). (Hommage à W.W. Bishop). Annales de Paléontologie (Vertébrés-Invertébrés) 84, 3–4, 287–306. Haile-Selassie, Y. (2001). Late Miocene hominids from the Middle Awash. Nature 412, 178– 181. Haile-Selassie, Y., Suwa, G. & White, T.D. (2004). Late Miocene teeth from Middle Awash, Ethiopia, and early hominid dental evolution. Science 303, 1503–1505. Hill, A. (1985). Early hominid from Baringo, Kenya. Nature 315, 222–224. Hill, A. (1994). Late Miocene and Early Pliocene hominoids from Africa. In (R.S. Corruccini & R.L. Ciochon, Eds) Integrative paths to the past. Englewood Cliffs, New Jersey: Prentice Hall, pp. 123–145. Hill, A. (1998). Very early plausible Hominids from the Tugen Hills, Kenya. Abstracts, Dual Congress 1998. Johannesburg, Organising Committee of Dual Congress, p. 20.

147

From Tools to Symbols Hill, A., Drake, R., Tauxe, L., Monaghan, M., Barry, J.C., Behrensmeyer, A.K., Curtis, G., Fine Jacobs, B., Jacobs, L., Johnson, N. & Pilbeam, D. (1985). Neogene palaeontology and geochronology of the Baringo Basin, Kenya. Journal of Human Evolution 14, 749–773. Hill, A. & Ward, S. (1988). Origin of the Hominidae: The record of African large hominoid evolution between 14 My and 4 My. Yearbook Physical Anthropology 31, 49–83. Hill, A., Ward, S. & Brown, B. (1992). Anatomy and age of the Lothagam mandible. Journal of Human Evolution 22, 439–451. Ishida, H. & Pickford, M. (1998). A new Late Miocene hominoid from Kenya: Samburupithecus kiptalami gen. et sp. nov. Comptes rendus de l’Académie des Sciences de Paris, IIa, 325, 823– 829. Ishida, H., Pickford, M., Nakaya, H. & Nakano, Y. (1984). Fossil anthropoids from Nachola and Samburu Hills. African Studies Monograph, Supplementary Issue, 2, 73–85. Kelley, J. (1993). Taxonomic implications of sexual dimorphism in Lufengpithecus. In (W.H. Kimbel & L.B. Martin, Eds) Species, Species Concepts, and Primate Evolution. New York: Plenum Press, pp. 429–458. Lague, M.R. & Jungers, W.L. (1999). Morphometric variation in Plio-Pleistocene hominid distal humeri. American Journal of Physical Anthropology 101 (3), 401–427. Leakey, L.S.B. (1961/1962). A new lower Pliocene fossil primate from Kenya. Annual Magazine of Natural History 13 (4), 689–696. Leakey, L.S.B. (1967). An early Miocene member of Hominidae. Nature 213, 155–163. Leakey, L.S.B. (1970). Early Pliocene sediments and fossils from northern Kenya Rift Valley, Kenya. Nature 226, 914–918. Leakey, M.G., Feibel, C.S., McDougall, I. & Walker, A. (1995). A new four million-year old hominid species from Kanapoi and Allia Bay, Kenya. Nature 376, 565–571. Leakey, M.G., Feibel, C.S., McDougall, I. & Walker, A. (1998). New specimens and confirmation of an early age for Australopithecus anamensis. Nature 393, 62–66. Le Gros Clark, W.E. & Leakey, L.S.B. (1951). The Miocene Hominoidea of East Africa. Fossil Mammals of Africa (British Museum of Natural History) 1–117. Marks, J. (2002). What it means to be 98% chimpanzee? Berkeley: California Press. McDougall, I. & Feibel, C.S. (2003). Numerical age control for the Miocene-Pliocene succession at Lothagam, a hominoid-bearing sequence in the Northern Kenya Rift. In (M.G. Leakey & J.M. Harris, Eds) Lothagam: the dawn of humanity in East Africa. New York: Columbia University Press, pp. 43–64. Ohman, J.V., Krochta, T.J., Lovejoy, O.C., Mensforth, R.P. & Latimer, B. (1997). Cortical distribution in the femoral neck of hominids: implications for the locomotion of Australopithecus afarensis. American Journal of Physical Anthropology 104, 117–131. Patterson, B., Behrensmeyer, A.K. & Sill, W.D. (1970). Geology and fauna of a new Pliocene locality in North-Western Kenya. Nature 226, 918–921.

148

Dawn of hominids Patterson, B. & Howells, W.W. (1967). Hominid humeral fragment from early Pleistocene of Northwestern Kenya. Science 156, 64–66. Pickford, M. (1975). Late Miocene sediments and fossils from the Northern Kenya Rift Valley. Nature 256, 279–284. Pickford, M. (1985). A new look at Kenyapithecus based on recent discoveries in Western Kenya. Journal of Human Evolution 14, 113–143. Pickford, M. (1986a). Sexual dimorphism in Proconsul. In (M. Pickford & B. Chiarelli, Eds) Sexual Dimorphism in Primates. Florence: Il Sedicesimo, pp. 133–170. Pickford, M. (1986b). Hominoids from the Miocene of East Africa and the phylogenetic position of Kenyapithecus. Zeitschrift für Morphologie und Anthropologie 76, 115–130. Pickford, M. & Chiarelli B., Eds, 1986. Sexual dimorphism in primates. Florence: Il Sedicesimo. Pickford, M. & Ishida H. (1998). Interpretation of Samburupithecus, an Upper Miocene hominoid from Kenya. Comptes rendus de l’Académie des Sciences de Paris, IIa, 326, 299–306. Pickford, M., Johanson, D.C., Lovejoy, C.O., White, T.D. & Aronson, J.L. (1983). A hominid humeral fragment from the Pliocene of Kenya. American Journal of Physical Anthropology 60, 337–346. Pickford, M., Nakaya, H., Ishida, H. & Nakano, Y. (1984). The biostratigraphic analyses of the faunas of the Nachola area and Samburu Hills, northern Kenya. African Study Monographs, Supplementary Issue 2, 67–72. Pickford, M. & Senut, B. (2001). The geological and faunal context of Late Miocene hominid remains from Lukeino, Kenya. Comptes rendus de l’Académie des Sciences de Paris, série IIa, 332, 145–152. Pickford, M. & Senut, B. (2004a). Hominid teeth with chimpanzee- and gorilla-like features from the Miocene of Kenya: implications for the chronology of the ape-human divergence and biogeography of Miocene hominoids. Anthropological Science, online publication, July 2004. Pickford, M. & Senut, B. (2004b). Early Pliocene Tragulidae and peafowls in the Rift Valley, Kenya: evidence for rainforest in east Africa. Comptes Rendus Palevol 3, 179–189. Pickford, M., Senut, B., Gommery, D. & Treil, J. (2002). Bipedalism in Orrorin tugenensis revealed by its femora. Comptes Rendus Palevol 1, 191–203. Pilbeam, D. 1996. Genetic and morphological records of the Hominoidea and hominid origins: a synthesis. Molecular Biology and Evolution 5, 155–168. Sawada, Y., Pickford, M., Senut, B., Itaya, T., Hyodo, M., Miura, T., Kashine, C., Chujo, T. & Fuji, H. (2002). The age of Orrorin tugenensis, an early hominid from the Tugen Hills, Kenya. Comptes Rendus Palevol 1, 293–303. Schmid, P. (1983). Eine Rekonstruktion des Skelettes von A.L. 288-1 (Hadar) und deren Konsequenzen. Folia Primatologica 40, 283–306. Senut, B. (1979). Comparaison des Hominidés de Gomboré IB et de Kanapoi: deux pieces du

149

From Tools to Symbols genre Homo? Bulletins et Mémoires de la Société d’Anthropologie de Paris 6, XIII, 111–117. Senut, B. (1981). L’humérus et ses articulations chez les Hominidés plio-pléistocènes. Cahiers de Paléontologie (Paléoanthropologie). Paris: CNRS, pp. 1–181. Senut, B. (1983). Quelques remarques à propos d’un humérus d’hominoïde pliocène provenant de Chemeron (bassin du lac Baringo, Kenya). Folia Primatologica 41, 267–276. Senut, B. (1992). French contribution to the study of human origins: the case of Australopithecus afarensis. Human Evolution 7 (4), 15–24. Senut, B. (1995). D’Australopithecus à Praeanthropus ou du respect du Code de Nomenclature. Annales de Paléontologie (Vertébrés – Invertébrés) 81 (4), 279–281. Senut, B. (1996). Pliocene hominid and systematics. South African Journal of Science 92, 165– 166. Senut, B. (1998). Grands singes fossiles et origine des hominidés. Primatologie, 1, 91–134. Senut, B. & Pickford, M. (2004). La dichotomie grands singes – homme revisitée. Comptes Rendus Palevol 3, 265–276. Senut B., Gommery D. & Pickford, M. (2002). The locomotion of Orrorin tugenensis: implications for the origins of bipedalism. In (E. Gitonga, B. Senut, H. Ishida, Y. Sawada & M. Pickford, Eds) From Samburupithecus to Orrorin: Origins of Hominids. Geological and Palaeontological Background. International Workshop at Bogorio. Abstracts, 18. Senut, B., Pickford, M., Gommery, D., Mein, P., Cheboi, K. & Coppens, Y. 2001. First hominid from the Miocene Lukeino Formation, (Kenya). Comptes rendus de l’Académie des Sciences de Paris, IIa, 332, 137–144. Simons, E.L. (1961). The phyletic position of Ramapithecus. Postilla Yale Peobody Museum 57, 1–9. Simons, E.L. (1969). Late Miocene hominid from Fort Ternan, Kenya. Nature 221, 449. Simons, E.L. & Pilbeam, D.R. (1965). Preliminary revision of the Dryopithecinae (Pongiae, Anthropoidea). Folia Primatologica 3, 81–152. Susman, R. (1988). Hand of Paranthropus robustus from Member 1, Swartkrans: fossil evidence for tool behaviour. Science 240, 781–784. Susman, R.L., Stern, J.T. Jr. & Jungers, W.L. (1984). Arboreality and bipedality in the Hadar hominids. Folia Primatologica 43, 113–156. Vignaud P., Duringer, P., Mackaye, H.T., Likius, A., Blondel, C., Boisserie, J-R., de Bonis, L., Eisenmann, V., Etienne, M.E., Geraads, D., Guy, F., Lehmann, T., Lihoreau, F., LopezMartinez, N., Mourer-Chauviré, C., Otero, O., Rage, J-C., Schuster, M., Viriot, L., Zazzo, A. & Brunet, M. (2002). Geology and palaeontology of the Upper Miocene Toros-Menalla hominid locality, Chad. Nature 418, 152–155. Ward, C.V., Walker, A.C., Teaford, M.F. & Odhiambo, I. (1993). Partial skeleton of Proconsul nyanzae from Mwfangano Island. American Journal of Physical Anthropology 90, 77–111. White, T.D. (1986). Australopithecus afarensis and the Lothagam mandible. Anthropos, 23, 73–

150

Dawn of hominids 90. White, T.D., Suwa, G. & Asfaw, B. (1994). Australopithecus ramidus, a new species of early hominid from Aramis, Ethiopia. Nature 371, 306–312. White, T.D., Suwa, G. & Asfaw, B. (1995). Australopithecus ramidus, a new species of early hominid from Aramis Ethiopia: corrigendum. Nature 375, 88. WoldeGabriel, G., White, T.D., Suwa, G., Renne, P., de Heinzelin, J., Hart, W.K. & Heiken, G., (1994). Ecological and temporal placement of early Pliocene hominids at Aramis, Ethiopia. Nature 371, 330–333. WoldeGabriel, G., Haile-Selassie Y., Renne P.R., Hart W.K., Ambrose S.H., Asfaw, B., Heiken G. & White T. (2001). Geology and palaeontology of the Late Miocene Middle Awash Valley, Afar rift, Ethiopia. Nature 412, 175–178. Wolpoff, M.H., Senut, B., Pickford, M. & Hawks, J. (2002). Sahelanthropus or Sahelpithecus? Nature 419, 581–582. Wrangham, R. & Pilbeam, D. (2001). African apes as time machines. In (B. Galdikas, N. Briggs, L. Sheeran, G. Shapiro & J. Goodall, Eds) All Apes Great and Small. Vol. 1: African Apes, pp. 5–17. New York: Kluwer Academic/Plenum Publishers.

151

The impact of new excavations from the Cradle of Humankind on understanding the evolution of hominins and their cultures Lee Berger Institute for Human Evolution, School of Geosciences, University of the Witwatersrand, Private Bag 3, WITS 2050, Johannesburg, South Africa

Abstract Excavations of poorly known or previously understudied sites within the Sterkfontein region over the past several years have revealed an abundance of new information concerning the mode and tempo of hominin evolution and culture, faunal variability and faunal change through time, and the chronology of sites. They have also increased our understanding of cave formation processes, which have a bearing on the taphonomy of fossil assemblages. As excavations have extended, it has become clear that early hominin cultural remains are more prevalent than has been previously hypothesised and the presence or absence of cultural remains appears to be more closely related to the excavations’ proximity to cave entrances than to other factors. Additionally, the increased diversity of work has offered considerable insight into supposedly rare faunal forms and the frequency of their occurrence in the South African assemblages. The extension of research into these ‘new’ sites has also yielded information about the chronological ‘windows’ preserved in the region. The application of new technologies, in particular GIS, promises to allow greater understanding of these assemblages.

Résumé Des fouilles dans des sites peu connus ou qui n’avaient pas fait l’objets d’analyses systématiques dans la région de Sterkfontein ont révélé au cours des dernières années des nouvelles informations sur les populations d’hominidés qui ont vécu dans cette région, sur leurs cultures, sur la variabilité et les changements des faunes et sur la chronologie des sites. Ils nous ont également permis d’accroître notre compréhension du mode de formation des cavités et des dépôts, fait crucial pour comprendre la taphonomie des assemblages fossiles. En élargissant les surfaces fouillées est apparue une abondance insoupçonné de restes et la concentration de ceux-ci à l’entrée des cavités. Ces nouvelles fouilles ont également produit des nouvelles informations sur la présence et fréquence de certains taxons dans les assemblages fauniques sud-africains et sur la chronologie des gisements. L’application de techniques SIG offre une nouvelle clef pour la compréhension de ces gisements.

152

The impact of new excavations from the Cradle of Humankind

Introduction Within the Sterkfontein valley area – now designated the Cradle of Humankind World Heritage site by UNESCO – cave sites bearing stone tools have been considered rare. Until recently, only three sites had been shown to possess in situ stone tool assemblages where there is definitive evidence of on-site or near-site hominin occupation – Sterkfontein, Swartkrans and Kromdraai A. Within these three sites is evidence of the Earliest Stone Age (Oldowan), the Acheulian, the Middle (MSA) and Later (LSA) Stone Ages, and the Iron Age. However, over the past several years exploratory work and more extensive excavations at other dolomitic cave sites within the region by the Palaeoanthropology Unit for Research and Exploration at the University of the Witwatersrand have shown that not only are there a far greater number of fossil-bearing caves in the region than previously recognised (Berger et al., 2003a, b), but that stonetool bearing assemblages in caves and other types of archaeological sites are not as rare as had previously been thought. This paper discusses newly excavated sites in the Witwatersrand region that now record in situ stone tool assemblages which may add to our ability to understand and assess hominin cultural tradition and cognition, hominin movement in the landscape and their methods of sourcing raw materials. Specifically, Coopers D, a newly discovered fossil and archaeological site located adjacent to the described sites of Coopers A and B (Berger et al., 2003a), contains significant evidence of the Earliest Stone Age in the presence of Paranthropus remains (Berger et al., 2002) and evidence of the MSA and Iron Age in overlying deposits. Gladysvale has recently been shown to have the Acheulian in situ (Berger et al., in prep.), while a cave adjacent to the previously described Plovers Lake site which has been provisionally named Plovers Lake 2, has exhibited significant evidence of the MSA in association with the remains of anatomically modern humans (Berger et al., 2003c).

Coopers D In early 2001, excavations of the Coopers B deposit were re-opened to continue exploring the small exposure of in situ fossiliferous breccia described and sampled by Brain in the 1960s, and expose an area immediately to the north of this site under the assumption that a larger area of cave breccia had been covered by miners’ waste. In May 2001, a small area of exposed fossiliferous breccia approximately 80 metres to the south of Coopers B was sampled for in situ fossils. Almost immediately the excavation teams recovered hominin remains and work at Coopers was focused on this area, dubbed Coopers D. Continuing excavations along an approximately 2 metre wide by 20 metre long north–south trending exposure of breccia have resulted in the recovery of more than 7 000 identifiable faunal remains including more than a dozen hominin remains, as well as over 50 artefacts.

153

The Coopers D site represents the first definitive report of in situ archaeological remains from this site and at the time of its discovery was only the second site in the area to have definitive stone tools in association with hominins. Artefacts are common in the northern aspects of the excavation, in both the decalcified and calcified breccia; geological analysis and our projections indicate that the ancient entrance must have been here. At present, only artefacts from the decalcified areas have been extracted pending more precise geological study of the calcified in situ sediments. Artefacts from the decalcified sediments are largely undiagnostic and a significant part of the assemblage comprises numerous un-retouched quartz flakes, hammer stones and at least one spheroid (Fig. 1). Developed Oldowan or early Acheulian tools have also been recovered in lower numbers. At present, more than 54 definitive artefacts and flakes have been recovered, although many more remain in situ in calcified breccia. The artefactual assemblage also includes one purported tool that is unusual in its morphology and material – a very elongated biconical fragment of chert with evidence of impact at both ends, which will be described in detail in another publication. To date, the artefacts have only been recovered in the presence of Paranthropus robustus. Transport-abraded MSA artefacts and Iron Age material have been found in the superficial parts of the excavation, where no fauna are recovered, and there is evidence of disturbance, indicating that this material almost certainly derives from hill wash, but it does not contaminate the underlying deposits. The fauna from Coopers D are surprisingly diverse, given the small excavation area and the limited period of work. In content it corresponds well with sites such as Swartkrans and Kromdraai A. It differs from these sites in that it preserves numerous examples of what are generally considered rare taxa at other Witwatersrand sites. In particular, Coopers shows an abundance of suids and canids, both of which are already represented by many dozens of specimens. The diversity of taxa, as well as the presence of more closed habitat taxa (Berger et. al., 2003b) in direct association with open habitat taxa, suggests that Coopers D may have been situated near a transitional environment between riverine woodland and more open habitats. At present, indications are that the fauna are largely accumulated by non-human taphonomic processes – including but not limited to carnivore activity and entrapment. Nevertheless, the lack of condition of the stone tools, including lack of evidence of surficial erosion or abrasion, indicates that they probably originated in the immediate vicinity.

Gladysvale Gladysvale (Berger, 1992, 1993; Berger et al., 1993; Berger & Tobias, 1994; Lacruz et al., 2003) has yielded, for the first time, in situ evidence of the Acheulian in a newly opened area of the excavations in the external aspect of the cave (Fig. 2). The material

154

The impact of new excavations from the Cradle of Humankind

Figure 1 Stone artefacts from the Coopers D site. Top: purported tool made of shale, showing signs of battering at either end, and striations towards the distal end and medial sections. Centre: quartz spheroid with extensive battering on its surface, originally a polyhedral core. Bottom: left to right, quartz flake with edge damage due to utilisation on the right-hand edge; quartz flake, possibly from a polyhedral core; quartz flake with signs of retouch on the distal dorsal surface.

155

Figure 2 Mid-Acheulean handaxe from Gladysvale made on quartzite. Scale = 1cm

originates from a debris cone below what would have been an ancient entrance. Given the geological circumstances and rich fauna, excavations in the immediate area of the debris cone hold promise for the recovery of more hominin cultural remains. No hominin fossils have as yet been found in the immediate vicinity of the artefact nor in temporally contemporaneous deposits in other areas of the site, although early and later hominin remains do occur in several areas.

Plovers Lake 2 The Plovers Lake 2 site offers a unique window into the MSA of the region. Although the MSA is represented in the Witwatersrand area (Member 4 of Swartkrans and at Sterkfontein), no hominin fossils have yet been recovered from these localities. In fact, human skeletal remains associated with MSA assemblages in southern Africa in general are rare and tend to be quite fragmentary, such as the well-known material from Klasies River Mouth and Hoedjiespunt. The hominin fossils recovered from the in situ decalcified deposits of Plover’s Lake 2 (Berger et al., 2003a), although

156

The impact of new excavations from the Cradle of Humankind

fragmentary, represent the first hominins directly associated with an MSA industry in the Witwatersrand area. Recovered to date are three mandibular teeth (RP4-RM2) that have been refitted via interproximal contact, indicating that a single individual is represented. These fossils are consistent with modern Homo sapiens, although taxonomic attribution based on isolated teeth must be considered tentative. The faunal assemblage associated with the hominin fossils is rich and varied, displaying moderate to good preservation. Over 11 000 fossils have been recovered to date. Although predominantly modern forms are represented at Plovers, at least six extinct taxa are also recognised. Carnivores are plentiful, being dominated by smaller body size forms, in particular the Canidae. The Bovidae are the most abundant group in the assemblage, accounting for five of the six extinct taxa at the site. The presence of tooth-marked specimens and coprolites in the assemblage indicates that carnivores were responsible for at least a portion of the bone accumulation. However, human

Figure 3 Stone artefacts from Plovers Lake. Left to right: slate flake blade with possible hafting notch on the proximal dorsal surface; quartzite flake blade; slate side-scraper with retouch on right-hand edge; slate flake blade. Top: distally broken quartz flake blade (left) and quartz flake retouched on the right edge (right).

157

occupation of the cave or its surroundings is indicated by the presence of MSA flakes, retouched tools, and debitage made largely of locally available quartz and quartzite (Fig. 3). Cut-marked bones are present but rare, suggesting that humans were also at least partially responsible for the accumulation of the later Quaternary deposits at Plovers Lake. At present over 2 500 artefacts have been recovered from the in situ excavations, making the ratio of artefacts to fossils very high at this site.

Research perspectives Ongoing research in the Witwatersrand area has shown that archaeological remains in the area are not as uncommon as had previously been thought and that we should not underestimate the potential of the Witwatersrand caves to yield significant evidence of the mode and tempo of cultural evolution over the last c. 2,5 million years. Specifically, the three additional stone-tool bearing sites and one bone-tool bearing site that have been discovered in the Cradle of Humankind area demonstrate the commonness of archaeological assemblages in the region. This number is further placed in context when it is understood that significant excavations have only taken place at nine locales – implying that far from being uncommon, on a site-by-site basis, artefactual remains are more common than not at fossil-bearing sites in the region. It remains to be seen whether artefacts are less abundant, relative to faunal remains, than at East African sites or occupation sites of a younger age, as these recent excavations show what appears to be a correlation between proximity to the ancient entrance and tool densities, indicating that there may be significant location and sampling bias in our present data set. Furthermore, our work is showing that at these sites and others (Nigro et al., 2002, 2003), there appears to be a very close and obvious link between artefactual assemblages found in situ and the proximity of excavations to ancient entrances. These entrances are clearly associated with fault lines in the region. This leads one to suggest that a closer examination of the relationship between such deposits as Sterkfontein Member 5 and Member 4 might be in order, bearing in mind that much of their traditional separation has been based upon the presence/absence of artefacts with only limited geological and faunal support (Kuman & Clarke, 2000). Additionally, the new sites have highlighted the fact that excavations of new sites and fossil-bearing localities in the area hold the promise of revealing windows into previously poorly sampled temporal periods and ecological situations.

The application of GIS to sites in the area While the expansion of work at new localities in the World Heritage site has led to the recognition of the promise that the area holds for further important discoveries, it has

158

The impact of new excavations from the Cradle of Humankind

also led to the need to apply novel technologies and methods to best extract maximum information from the complex geological situations of the Witwatersrand cave sites. Our experimentation with GIS (Geographic Information Systems) technologies for mapping the three-dimensional positions of fossil and archaeological remains and for comparing and contrasting geological and stratigraphic information has shown this to be a promising tool for all aspects of palaeoanthropological work, particularly in exploring questions of complex relationships in dolomitic cave. In brief, the GIS acts as a spatial database containing attribute information in which specific data themes can be integrated and explored cumulatively in order to derive new information on a particular question or questions. The strength of a GIS lies in its ability to facilitate the identification and comprehension of complex patterns and associations between spatial phenomena that may otherwise remain undetectable by conventional means of analysis. The highly accurate plotting capabilities of laser theodolites combined with the novel software developed by our staff, students and colleagues has allowed us to examine the caves and their contents in more sophisticated ways than previous plotting methods allowed (Fig. 4). We have also, for the first time, applied GIS analysis to laser theodolite data for intra-site comparisons. The 3-D plotting capabilities of laser theodolites allows for more effective excavation methods to be employed, methods superior to grid-based excavation methods in the Witwatersrand caves, which are generally inferior in their plotting accuracy and which prove vastly inferior in situations where overhangs are encountered or where the calcified and decalcified breccias are present (Nigro, 2002; Nigro et al., 2002; Lacruz et al., 2003). Additionally, laser theodolites are not affected by constant exposure to the elements and can adapt readily to changing conditions in the excavation. I feel strongly that the future of excavation methods in the dolomitic caves of South Africa must move towards the use of these more precise methods of plotting and analysis and away from traditional methods of solely grid-based (tape measure and plumb bob) excavations. While grid-based excavations may produce a greater number of ‘high profile’ fossils due to the vastly greater amount of earth being moved, the loss in spatial resolution that occurs, resolution that is so critical to understanding context and relationship within these systems, is in my opinion unacceptable. Finally, we must greatly expand our efforts in the area of understanding the formation processes of these cave systems and their infills if we are to have any hope of accurately dating these fossils and deposits. These two areas should be seen as a priority by South African researchers.

159

Figure 4 The 3-D programs allow advanced search capabilities that can be used to show relationships. 3-D buffer zones can show, for example, spatial relationships of all hominin fossils within a given distance of a scatter, or all recorded lithic flakes within a given distance of a core (after Nigro, 2002).

160

The impact of new excavations from the Cradle of Humankind

Acknowledgements The author would like to thank the staff and students of the Palaeoanthropology Unit for Research and Exploration at the University of the Witwatersrand, the School of Geosciences at the University of the Witwatersrand, and staff and students at the University of Arkansas, Duke University and the University of Zürich who collaborate on the various projects. The South African Heritage Resource Agency for issuing the necessary permits. The author gratefully acknowledges the Fossil Trackers for their invaluable assistance in excavation and mapping. Funding was made available for these projects by the Palaeoanthropology Scientific Trust, the National Geographic Society and the University of the Witwatersrand. The author would like to thank the hosts of the conference and in particular, the French Embassy of South Africa for its support.

References Berger, L.R. (1992). Early hominid fossils discovered at Gladysvale Cave, South Africa. South African Journal of Science 88, 362. Berger, L.R. (1993). A preliminary estimate of the age of the Gladysvale australopithecine site. Palaeontologica Africana 30, 51–55 Berger, L.R., Keyser, A.W. & Tobias, P.V. (1993). Gladysvale: first early hominid site discovered in South Africa since 1948. American Journal of Physical Anthropology 92, 107–111. Berger, L.R. & Tobias, P.V. (1994) New discoveries at the early hominid site of Gladysvale. South African Journal of Science 90, 223–226. Berger, L.R., Churchill, S. & de Ruiter, D. (2003a). Plover’s Lake: a hominin-bearing Middle Stone Age site in the Witwatersrand area, South Africa. Proceedings of the 72nd Annual Meeting of the American Association of Physical Anthropologists. Berger, L.R., de Ruiter, D. J. & Steininger, C. M. (2003b). Preliminary results of excavations at the newly discovered Coopers D deposit, Gauteng, South Africa. South African Journal of Science 99, 276–278. Kuman, K. & Clarke, R.J. (2000). Stratigraphy, artefact industries and hominid associations for Sterkfontein, Member 5. Journal of Human Evolution 38: 827–847. Lacruz, R., Ungar, P., Hancox, P.J., Brink, J.S. & Berger, L.R. (2003) Gladysvale: Fossils, strata and GIS analysis. South African Journal of Science 99: 283–285. Nigro, J. (2002). The Swartkrans GIS project. Using Geographic Information Systems to explore cave taphonomy. M.Sc. thesis. University of Arkansas. Nigro, J., Limp, F., Kvamme, K., DeRuiter, D. & Berger, L. (2002). The creation and potential applications of 3-dimensional GIS for the early hominin site of Swartkrans, South Africa. In (G. Burenhult, Ed.) Archaeological Informatics: Pushing the Envelope; Computer Applications and Quantitative Methods in Archaeology. BAR International Series 1016. Oxford: Archaeopress.

161

Nigro, J.D., Ungar, P.S., de Ruiter, D. & Berger, L.R. (2003). Developing a Geographic Information System (GIS) for mapping and analysing fossil deposits at Swartkrans, Gauteng Province, South Africa. Journal of Archaeological Science 30, 317–324.

162

Stone Age signatures in northernmost South Africa: early archaeology in the Mapungubwe National Park and vicinity Kathleen Kuman, Ryan Gibbon, Helen Kempson, Geeske Langejans, Joel Le Baron, Luca Pollarolo* and Morris Sutton School of Geography, Archaeology and Environmental Studies, University of the Witwatersrand, Private Bag 3, WITS 2050, Johannesburg Laboratorio di Antropologia, Via del Proconsolo No. 12, Firenze 50122, Italy

*

Abstract The oldest archaeological sites currently known in northernmost South Africa are found in the Mapungubwe National Park (formerly known as the Vhembe-Dongola National Park) and neighbouring farms, where there is a widespread distribution of open-air sites in deflated contexts. They are sealed by Holocene sands, which at some of the sites contain Later Stone Age (LSA) artefacts. The industry to which the older assemblages are most comparable is final Earlier Stone Age (ESA) in character, with parallels to the Sangoan Industry, or what has locally been proposed as the Charaman from Zimbabwe. A developed phase of the Middle Stone Age (MSA) with segments and retouched points is also represented on one landscape. Rockshelter sites are being investigated to locate stratified deposits to which the open sites may be compared. In the interim, the material provides a form of ‘archaeological signature’ that can contribute to the overall evaluation of Stone Age occupations in northernmost South Africa. Large-scale climatic fluctuations during the course of the Pleistocene have influenced occupations across southern Africa. The archaeology of the Mapungubwe area appears to have more in common with developments north of the Limpopo than it does with the South African sequence.

Résumé Les plus anciens sites préhistoriques de la région septentrionale de l’Afrique du Sud se trouvent dans le Parc National de Mapungubwe (appelé auparavant Parc National de Vhembe-Dongola) et dans les fermes

163

From Tools to Symbols voisines. Il s’agit de gisements de plein air localisés dans des plaines et recouverts de formations sableuses datant de l’Holocène qui contiennent, dans certains des sites, des objets du Later Stone Age. Les industries les plus anciennes de la région sont comparables à celles de la fin du Early Stone Age (ESA), avec quelques ressemblances avec le Sangoan ou le Charaman du Zimbabwe. Une phase évoluée du Middle Stone Age (MSA), avec des segments et des pointes retouchées, a également été identifiée dans une zone. Des abris sous roche font actuellement l’objet de prospections avec l’objectif de localiser des dépôts archéologiques stratifiés auxquels rattacher les sites de plein air. En attendant, le matériel recueilli dans ces derniers présente des aspects originaux qui permettent de caractériser l’occupation humaine préhistorique de la région. Des variations climatiques très importantes ont sans doute influencé le peuplement de l’Afrique australe au cours du Pléistocène. L’archéologie de la zone du Vhembe-Dongola semble avoir plus en commun avec ce que l’on connaît au nord du Limpopo qu’avec les séquences sud-africaines.

Introduction For many years we have been aware of the widespread surface distribution of Stone Age artefacts in the Mapungubwe National Park and surrounding farms, which fringe the northernmost border of South Africa shared with Botswana and Zimbabwe. Here concentrations of Earlier Stone Age (ESA) artefacts were observed on the Samaria farms, close to the confluence of the Limpopo and Shashe Rivers (Fig. 1). In 1985, the owner of the Samaria 1 farm, Michael Moerdyk, invited archaeologists from Wits University to see the artefacts that littered his landscape, and sixteen years later, in 2001, we initiated a Stone Age research programme. While the initial focus of this programme was to study the Earlier Stone Age in the region, it soon became clear that documentation of the entire Stone Age cultural record has important information to add to our understanding of human occupations of southern African during the Stone Age. Such occupations have been strongly influenced by the oscillating climates of the Pleistocene, when significant fluctuations in temperature resulted in shifts in air circulation patterns that impact moisture and vegetation regimes throughout southern Africa (Tyson & Partridge, 2000). This paper describes the cultural record we have been documenting and its relevance to the emergence of modern human populations in the late Pleistocene.

The Earlier Stone Age – Hackthorne and Keratic Koppie Hackthorne At the heart of the research has been the ESA site of Hackthorne (Site 2229AB231; 22° 13' 47" S, 29°18' 56" E; 619 m altitude; Kuman et al., 2005). The site is situated atop an ancient terrace of the Limpopo River that has been cut back in this location, forming an escarpment 20–30 metres above the surrounding landscape and three kilometres

164

Stone age signatures in northernmost South Africa

Figure 1 Location of the study area in the northern Limpopo Province, immediately south of the Limpopo River. The Mapungubwe National Park is the area south of the Limpopo River enclosed by the dashed line.

Figure 2 Detail of the study area, showing the location of the escarpment (hatched line) formed by a Miocene terrace of the Limpopo River. ESA artefacts are distributed along the escarpment.

165

From Tools to Symbols

long, crossing the Machete, Hackthorne and Samaria farms (Fig. 2). ESA artefacts are distributed along the length of this escarpment. Although there are no fossils preserved in the terrace sediments, the comparative geomorphology of the area indicates that the terrace formed in the Tertiary, when a number of the major river systems in the region were active (de Wit, 1996; de Wit et al., 2000). More specifically, the terrace sediments at Hackthorne are thought to be Miocene in age; they became cemented in the late Miocene or early Pliocene during an arid period and were subsequently cut back during the Plio-Pleistocene (T.C. Partridge, pers. comm.; de Wit et al., 2000). Gravels within the terrace deposits provided toolmakers with a variety of good raw materials: quartz, quartzite, sandstone, rhyolite, chert, chalcedony, and occasionally banded ironstone, dolerite and hornfels. The Hackthorne site is situated only 20 metres from the eroding edge of the terrace. At the bottom of this escarpment, lower ground leads to the floodplains of the Limpopo River, the present course of which lies about four kilometres to the north. Artefact accumulations are frequently found at lower elevations in the study area, but their contexts are poor, disturbed by water or slope erosion and lacking the kind of sedimentation that creates buried sites. On the escarpment, stone tools are even more common, but their original host sediments have been deflated and assemblages are buried by younger sands. Sites on the escarpment lie at elevations of 600–620 metres’ altitude and 80–100 metres above the modern Limpopo River. Hackthorne was chosen because of its accessible location and indications that it would be productive. Sediments consist of a uniform, unstratified sand cover that rests directly on the cemented Miocene terrace sediments. Investigations by Le Baron (2003) have shown that the upper levels are calcrete, while the fine, unconsolidated sand derives from weathering of the local bedrock, which is Clarens sandstone (Bordy & Catuneanu, 2002; Bordy et al., 2004). Calcrete is exposed at the escarpment edge, but the depth of the sand cover increases southwards. Within the site, it varies in thickness across the eroded, highly irregular buried calcrete substrate, ranging from an average of about 0,5 metres to well over 1,5 metres within deep solution pits formed within the calcrete. The irregular surface of the calcrete has been created by humic acids produced by tree roots and other vegetation growing within the modern (and presumably prehistoric) sediments that mantle the terrace. Artefacts are concentrated in the lowest levels of the sand close to the contact with calcrete and in many of the solution pits. Some have also become cemented into the hard laminated surface of the calcrete through its dissolution and regrowth, and no bone is preserved in these sediments. (See Kuman et al., 2005 for further detail.) The unstratified nature of the sand and the concentration of artefacts towards its base suggest that the sand cover is younger than the tool assemblage. This was

166

Stone age signatures in northernmost South Africa

confirmed when Woodbourne (2003) obtained a Holocene date for the sand by the optically stimulated luminescence (OSL) method. It indicates that the artefact assemblage, which is predominantly late ESA in technology and typology, was deflated onto or close to the calcrete substrate and was subsequently covered with Holoceneage sands. During extended dry phases, dynamic winds undoubtedly deflated the site, winnowing it of the original sediments, perhaps more than once. A sample of 275 flakes was studied microscopically by Kempson (2004) to investigate the number of weathering states present for each raw material. The results show that 30 per cent of this sample is very weathered and appears to be an older component that has undergone considerable exposure to sun and the elements. The majority of the assemblage, however, is relatively fresh in appearance (33 per cent ‘fresh’, 27 per cent slightly weathered). Further work on a larger sample of artefacts in relation to characteristics of each raw material is under way, which may help with a more detailed assessment of the deflation episodes in this time-averaged accumulation. It is interesting that the current study suggests that the most weathered component is dominated by quartz, the hardest rock, most resistant to weathering. These results will be confirmed with a larger sample, and Kempson will also conduct a technological comparison of the artefacts across each weathering state to search for any differences that relate to the conflation of material through time. Some upward movement of artefacts has occurred within the sand, and this will also be documented with three-dimensional point plots taken on artefacts > 20 mm in size. The Hackthorne assemblage presently consists of over 4 000 pieces from an excavated area of about 30 m2 with the majority of the site dug to calcrete. Owing to the lack of significant slope atop the escarpment, the assemblage is well preserved, even including some cores with conjoining flakes. Only the smallest fraction of tools < 10 mm in size is under-represented (Gibbon, 2002), due to run-off and erosion at the edge of the escarpment. Despite its deflated context and the long-term accumulation that this assemblage represents, it is remarkable that there are no formal tool types present to indicate significant mixing with much later industries. In other words, tools diagnostic of the LSA and developed phases of the MSA are absent. Instead, the most diagnostic types are small bifaces and picks, a cleaver and cleaver-like unifacial pick, denticulates, and denticulated scrapers (Figs 3–5). Such types fit with a final phase of the Earlier Stone Age known elsewhere in Africa as the Sangoan (discussed below), while denticulates are also characteristic of the early MSA. Prepared cores are present (Fig. 6), but these are not time-diagnostic as this technology originates in the Acheulean and is well represented in some Sangoan assemblages in good context (e.g. McBrearty, 1988). Such a time-limited assemblage could have accumulated if the site had been used only during the late ESA and early MSA, with climate change causing the area to be little used afterwards. There

167

From Tools to Symbols

is some indication of classic and late MSA occupation at other sites in the area, but no clear signs of this at Hackthorne. It is, indeed, the time-restricted appearance of this deflated assemblage that makes this site worthy of further study as a means of documenting the period/s when the area was most habitable. Kempson is now doing a comparative study of the Hackthorne technology with two assemblages from Kudu Koppie (discussed below), where both Sangoan and early MSA assemblages occur in a stratified context. This work will thus show if there is a mixed assemblage of final ESA and early MSA phases at Hackthorne. An additional sand sample for OSL dating has been taken from a solution pocket that undercuts the calcrete. This should determine if pockets of older sand may have been preserved in such hollows, thus evading deflation events. Keratic Koppie Keratic Koppie is a second site similar to Hackthorne, with over 2 000 pieces currently excavated (Site 2229AB407; 22o13'51.9"S, 29o19'48.3"E; 605 m altitude; Le Baron, 2004). In contrast with Hackthorne, it is located at a hilly outcrop of rocks (a koppie) and overlies sandstone rather than Miocene terrace calcrete. It was discovered in 2003 during a landscape test pit programme conducted by Le Baron and worked in 2003–2004. Diagnostic types include a cleaver and a uniface (Fig. 7), some small crude bifaces, and denticulates. Like Hackthorne, the assemblage is deflated as all the artefacts are contained within the bottom-most 30 cm of a two-metre-thick sand unit overlying sandstone bedrock. Three-quarters of the assemblage consists of small flaking debris < 20 mm in size, and unlike Hackthorne, a full proportion of artefacts < 10 mm is present (Le Baron 2004). As the site is not located on the escarpment, material is well preserved within a matrix of decayed sandstone blocks and has not been subjected to run-off. The two metres of sand cover reflects how koppies have acted as sediment traps, building up substantial deposits of wind-blown sands. Sediment build-up in the area exceeds 3–5 metres in some locations, with depths varying as a result of bedrock topography (Le Baron, 2004). Several sets of artefacts from Keratic Koppie have been conjoined, which further demonstrates the good preservation of material. Although the assemblage is in a deflated context, we are taking point plots of artefacts for spatial analysis. With deflation, the movement of material on such a flat surface should be largely vertical, with very limited horizontal movement (e.g. Kandel et al., 2003). Point plots of conjoining artefacts thus have the potential to provide spatial information on behaviour at the site. OSL samples have been taken from the sand cover to date the sediment and determine if it accumulated in stages or as the result of one major climatic event.

168

Stone age signatures in northernmost South Africa

a

b

c

d Figure 3 Formal tools from Hackthorne: (a) and (b) small bifaces or pygmy picks; (c) a knife-like bifacial handaxe; (d) a trihedral pick.

169

From Tools to Symbols

a

b

Figure 4 Formal tools from Hackthorne: (a) a cleaver with bifacial base and side; (b) a unifacial pick possibly used like a cleaver.

Other Stone Age signatures—Kudu Koppie and the Parma Farms Kudu Koppie While Hackthorne and Keratic Koppie appear to be primarily ESA-component sites, at least four components are present at Kudu Koppie, an open site adjacent to a very large, hilly outcrop of quartzite (Site 2229AB415; 22º 13'40.5"S, 29º 20''21.6"E; 604 m altitude). The site was discovered by Le Baron in his 2003 test pit programme. Here two phases of Holocene LSA are contained in the sand cover – an upper level with pottery and a lower level with microlithic segments. The upper level is diagnostic of the period of contact between LSA hunter-gatherers and Iron Age people, who entered the area less than two thousand years ago. This level, however, need not be very old as hunter-gatherers continued to live in the area for centuries after the appearance of

170

Stone age signatures in northernmost South Africa

a

d

c

b

e

f h

g i

Figure 5 Retouched tools from Hackthorne: (a)–(e) denticulates and denticulated scrapers; (f) convergent scraper; (g) retouched flake; (h) a steep, denticulated scraper; (i) a heavy-duty scraper made on a core, retouched at bottom left.

a

b Figure 6 Prepared cores from Hackthorne: (a) centripetal; (b) bipolar; (c) unipolar. A number of prepared

c

and unifacial radial cores appear to be made on split cobble surfaces.

171

From Tools to Symbols

Figure 7 Large tools from Keratic Koppie. Left: a uniface. Right: a cleaver.

Iron Age inhabitants. The sand cover lacks stratigraphy, lost through root growth and bioturbation, but the cultural material is contained within two discrete levels. This unit is then underlain by a Pleistocene-aged assemblage in deflated context, lying on and within a thick unit of decayed sandstone cobbles weathered from the koppie. This unit is dominated by early MSA artefacts, with denticulates and denticulated scrapers the dominant formal tool types (Fig. 8c). A few other later MSA types are also present: a large segment from the excavation (not illustrated), and two retouched points from a test pit a few metres away (Fig. 8b and Fig. 9). Prepared flakes and cores are also present (Fig. 9 and Fig. 11) and bifaces thus far are absent in the very large sample. Fragmentary fauna is preserved in all of the levels.. The proportions of diagnostic types suggest that this unit should for now be interpreted as a palimpsest of MSA material accumulated predominantly in the early MSA. A few LSA microlithic segments have also worked their down into this rocky deposit (Fig. 8a). Below this unit, however, lies another horizon with Sangoan-like picks and other large formal tools absent in the overlying rocky palimpsest. This is a closely packed lag deposit about 30 cm thick, lying directly on bedrock. The excavations and detailed study of the assemblages by level are being undertaken by

172

Stone age signatures in northernmost South Africa

b

a c Figure 8 Formal tools from Kudu Koppie: (a) an LSA segment; (b) a broken retouched point; (c) a denticulate.

Figure 9 Kudu Koppie artefacts. Top left: retouched point; others: unretouched convergent flakes.

173

From Tools to Symbols

Pollarolo. He has recovered the complete sequence from 2 m2 and plans to complete a further 5 m2 in 2005. This will greatly enlarge the existing sample, which contains many thousands of artefacts. Sangoan-like artefacts have also been retrieved from 6 m2 of test pits close to the koppie and from bulldozed ground within 20 metres of the excavation – picks, broken bifaces and a crude, heavy cleaver are illustrated in Figure 10 – and additional Sangoan types have been recovered in recent excavations. The four-phase cultural sequence that we find at Kudu Koppie is key to interpreting sites like Hackthorne and Keratic Koppie, which consist of a single deflated unit. The density of the two lower units at Kudu is very high, concentrated through deflation, but also indicating regular occupation of the site over time. Because the enormous sandstone outcrop acted as an effective sediment trap, artefacts were accumulated in phases that must once have been buried in substantial sand covers, which only later were subjected to fierce winds that deflated the original host sediments. The timing of

a c

b

d

e

Figure 10 Kudu Koppie artefacts. (a) and (b) broken bifaces; (c) and (e) picks; (d) a heavy, roughly formed cleaver. (a)–(c) are from excavated test pits; (d) and (e) are from disturbed ground that had been bulldozed by farmers quarrying for calcrete.

174

Stone age signatures in northernmost South Africa

c a

b

Figure 11 Kudu Koppie prepared core artefacts: (a) a prepared core for preferential flake removal; (b) a prepared core for centripetal removals; (c) a prepared core flake with facetted platform.

such deflation events cannot be told, but the typological integrity of the basal Sangoan unit suggests that ensuing early MSA people occupied a site with good sediment cover. The interlacing of MSA artefacts with the weathered sandstone cobbles from the koppie is another interesting environmental question that needs to be addressed. Elsewhere in Africa, the Sangoan is considered to be c. 300 000 years old, while early MSA assemblages are c. 250 000–130 000 years in age (Volman, 1984; McBrearty & Brooks, 2000). S. Woodbourne has taken sand samples for OSL dating from Kudu Koppie, and he and Z. Jacobs will be able to date the accumulation period/s of the sand cover during the Holocene. Le Baron’s first test pit dug on the northwest side of the koppie reached a depth of 3,6 metres before the narrow trench was stopped well above bedrock. Other test pits dug on the south-eastern side have sediment cover up to two metres thick. The dating of such large sand mobilisation events, probably all within the Holocene, will be of considerable interest. The Parma Farms In 2002, Sutton conducted an intensive survey for MSA on the Parma farms near Pont Drift, to the west of our core research area (Fig. 1). This work consisted of seven weeks of fieldwalking that covered a 15 km2 area. The Parma farms were chosen because this was the recorded location of a collection of refined MSA artefacts donated to the Wits Archaeology Dept in 1935 (Fig. 12). The collection was labelled as coming from ‘sub-surface grit’ and was biased by selection for the most interesting pieces, such as MSA cores, retouched points, and Howieson’s Poort-like segments. Despite the intensity of his survey, Sutton (2003) was not able to locate any similar

175

From Tools to Symbols

tools. He covered 85 per cent of the land, with the remaining 15 per cent consisting of homesteads and planted fields. Sections of cultivated fields were also surveyed without results. It is likely that the 1935 collection was found during the construction of foundations for buildings or latrines. A large number of other MSA and some LSA find spots were recorded, but unfortunately none of these was in a good context as erosion and stream activity have had an extensive impact on the landscape (Sutton, 2003). Nevertheless, the Parma collection indicates that a developed MSA industry is present in the region, which is younger than the phases represented at Hackthorne, Keratic Koppie and Kudu Koppie.

Discussion: archaeological assessments The deflated Hackthorne and Keratic Koppie assemblages do not appear to be classic MSA but have more affinity with the final ESA industry known as the Sangoan. However, Kudu Koppie has a Sangoan assemblage overlaid by MSA. Comparisons with

Figure 12 Parma farms artefacts: points and segments diagnostic of the MSA.

176

Stone age signatures in northernmost South Africa

this sequence will allow us to evaluate the single component sites for mixing, which will be more apparent in the technological details than in the typology. Preliminary analysis suggests that prepared core technology is more prominent in the MSA than the Sangoan levels. Overall, the striking aspects of the local assemblages are the prevalence of crude bifaces and picks of variable size, denticulates and denticulated scrapers of both small and large size, chopper-cores and prepared core technology. These traits resemble what Cooke (1966) termed the Charaman Industry from Zimbabwe, or in older terminologies, the Rhodesian Proto-Stillbay, which Cooke considered to be transitional between the final ESA (i.e. the Sangoan) and the MSA. However, the Charaman has not yet been established as a separate, transitional industry. While Cooke claimed to have found Sangoan levels that underlay his Charaman in three sites, his samples are considered too small to make accurate comparisons (Sampson, 1974; Volman, 1984). North of the Limpopo, the Sangoan is best described in Zambia and Kenya, where it is characterised by a heavy-duty component of picks, core-axes, and corescrapers, with picks often having a high-backed, plano-convex profile (Clark, 1970, 2001; McBrearty, 1988, 1991). Clark has long argued that the Sangoan was a regional adaptation to woodland habitats, and although some Sangoan sites are reconstructed as having more open habitats, woodworking may still have been an important function. At some sites notched and denticulated scrapers also suggest this function. In Kenya, large bifacially worked lanceolate points are recorded at Sangoan-Lupemban Complex sites (McBrearty, 1988, 1991). Some researchers place assemblages with such large lanceolates in a separate Lupemban Industry, which is argued to follow the Sangoan in time and to represent an early MSA (Clark, 2001). While the Lupemban is prominent in Zambia and central Africa, no lanceolate points have been found thus far in our study area, and we should not expect to see exact parallels. Given the proximity of our sites to Zimbabwe, it is not surprising that Hackthorne is most similar to Cooke’s proposed Charaman. The Limpopo River today is neither large nor perennial enough to create a persistent barrier to the north–south movement of people or animals, and it is likely that the situation was similar in mid- to late Pleistocene times. There may also be strong parallels with the Botswanan sequence, but there has been little Stone Age research in that country and no information exists on the late ESA or early MSA periods. However, interesting MSA sites occur in the northwest Makgadikgadi Pans of Botswana, and with these are associated refined handaxes that grade in size down into retouched points (Kuman, personal observation of sites and collections made by Ralph and Jack Bousfield). Clearly there is a great deal of potential for research in Botswana, Zimbabwe and northern South Africa which can elucidate our understanding of the southern African regional pattern of industries in

177

From Tools to Symbols

the late ESA and the early MSA. The question looms as to whether we should expect to see a Sangoan industry here, or whether regional variability is great enough to define a separate industry. One needs to bear in mind, however, that industrial names are classificatory schemes for the archaeologist’s convenience. The ‘transition’ from the final ESA to the early MSA was a process during geological time and not an event. In essence, the ‘Middle Stone Age’ is a historical and relative term, created to define the period between the Earlier and the Later Stone Ages at a time when the true antiquity of these developments was not understood. Just as McBrearty (1988) has described the Sangoan/Lupemban Complex material from Kenya as technologically similar to the MSA, but typologically distinct, the same may be true of our assemblages. In other words, radial and prepared core flaking techniques show continuity between the ‘Charaman/Sangoan’ and the MSA, but the classic formal tools of the MSA have yet to become prominent. The sequence at Kudu Koppie will be key to characterising these industries in our region. The LSA presence in this landscape has been researched by S. Hall and his students, who have found that most sites date to the last 2 000 years, with the greatest number in the first millennium AD (Hall & Smith, 2000). The oldest LSA currently known in the study area is dated to 11 000 ±90 years at Balerno Main Shelter, at the western side of the park (Simon Hall, pers. comm.), and B. van Doornum is currently working other rockshelter sites that are less than 3 000 years old in the same area. During our research, LSA deposits have been excavated not only at Kudu Koppie, but also at Mbere Shelter on the Machete farm by Sutton and Langejans (Site 2229AD96; 22º15’30"S, 29º17’30"E). At this site, LSA is associated with pottery throughout most of 65 cm of deposit. This contrasts with Kudu Koppie, where pottery was found only in the topmost level of the LSA deposit. It is also interesting that LSA artefacts in our area are associated mainly with rockshelters and koppies and are largely absent at open-air locations. At Hackthorne they are virtually absent, but at some koppies, small LSA flaking debris is often present in sands that cover the deflated surfaces. This pattern was noted by Le Baron in twenty-seven 1 m2 test pits dug across a 10 km2 area.

Conclusion The earliest prehistoric occupations in the Vhembe-Dongola National Park are preserved in deflated contexts at Hackthorne, Keratic Koppie and Kudu Koppie, and in selected collections from Parma. However, the material does provide a set of ‘archaeological signatures’ that contribute to the overall evaluation of the Stone Age occupations in the northernmost part of South Africa. The Hackthorne material most closely resembles the Zimbabwean industry referred to as Charaman. However, the Charaman has not yet been demonstrated to be a separate industry distinct from the

178

Stone age signatures in northernmost South Africa

Sangoan, which occurs at sites in nearby Zambia and Zimbabwe. Thus, for the present, Hackthorne should be grouped with the Sangoan Complex, and Keratic Koppie may belong to the same industry. However, both sites are being evaluated to determine if the assemblages may consist of a mix of Sangoan and early MSA artefacts. The stratified sequence from Kudu Koppie will provide the means to evaluate such mixing. This site has a basal layer of Sangoan material, overlaid by a palimpsest of artefacts dominated by the early MSA. The sporadic occupation of the area by later MSA groups with a more developed technology is evident from the selected collection of artefacts from the Parma farms. The Sangoan (like the Fauresmith) has been classed as a final ESA industry, possibly about 300 000 years old (McBrearty & Brooks, 2000), although some researchers would classify it as a transitional industry, or even an early MSA (Clark, 2001). In South Africa, the MSA has traditionally been defined by the appearance of pointed flakes and the disappearance of handaxes and cleavers, with the earliest MSA only broadly dated at Florisbad to about a quarter of a million years (Grun et al., 1996). Classifications schemes are historical definitions which sometimes do not reflect reality so much as the gaps in our knowledge at the time – gaps due to a sporadically preserved and incompletely studied archaeological record. Although they show marked typological differences, the parallels between the Sangoan and the Fauresmith are striking. The Fauresmith represents the final phase of the ESA in South Africa and is defined by small, well-made handaxes, points, blades, and prepared cores (Sampson, 1974; Klein, 2000). The Sangoan, on the other hand, is characterised by a heavy-duty, less elegant component of picks and steep, sometimes denticulated scrapers. Both the Sangoan and the Fauresmith have long been considered to mark the beginning of regional cultural specialisations towards the end of the ESA (Clark, 1970). Such developments can arguably be linked to the appearance of a more evolved hominid species about 200 000–500 000 years ago, variously referred to as Homo rhodesiensis, Homo heidelbergensis, or archaic Homo sapiens. This was a critical period both in hominid and cultural evolution, which immediately precedes the development of anatomically modern humans after 150 000 years. By 100 000 to 60 000 years ago, classic MSA technology is well developed, yet this does not reflect a change so much in the actual technology as a shift toward a lighter, more mobile toolkit which incorporates hafting. Deciphering of the local Sangoan vs Charaman industries in the northern Limpopo Province is now a focus of this project, as is documentation of the post-ESA sequence preserved in our sites. Although the ESA cultural record of the Mapungubwe National Park is not ideal in its context, it nonetheless has the potential to put on record those phases of prehistory when this region was inhabited. The area today is classed as

179

From Tools to Symbols

Mopane Veld and is semi-arid and drought-prone, with an annual precipitation range of 250–750 mm and an average of only 320 mm (Smith, 2004). With the large-scale climatic fluctuations that occurred during the Pleistocene, significant gaps in prehistoric occupations south of the Limpopo appear to have occurred. During extreme cold-dry periods, Stone Age peoples probably moved to regions further north that experienced more habitable conditions. Regardless of whether we achieve good dates for the Stone Age industries of the park, dates on the younger aeolian events will help address the issue of climatic change in the region. Documentation of deflation episodes and the existing cultural phases will help to create an overall picture of climatic and cultural developments over the last few hundred thousand years.

Acknowledgments KK owes a great debt to Professor P.V. Tobias for his support of her work on the Stone Age and for embracing her within the Sterkfontein Research Unit since 1991. The complicated archaeology of Sterkfontein has provided her with invaluable experience on site formation and transformation, as well as a drive to explore for new sites to expand the archaeological record in South Africa in a territory where Prof. Tobias worked as a young man. Fieldwork has been supported by the L.S.B. Leakey Foundation, the De Beers Fund Educational Trust and the University of the Witwatersrand. We thank the landowners, Frans and Sartjie Venter, for their support and encouragement of work on Hackthorne, and Hazel and Duncan MacWhirter for their help and hospitality on Machete. The directors and staff of the Mapungubwe National Park and the Venetia Limpopo Nature Reserve of the De Beers Ecology Division (De Beers Consolidated Mines) have greatly assisted the project, and C. Deschamps assisted in the landscape archaeology work and field studies at Hackthorne. We thank our colleagues for direct assistance or helpful discussions: F. Netterberg, E. Bordy, J. McNabb, A. Sinclair, S. Grab, J. Hancox, A. Delagnes, M. de Wit, L. Wadley, T.C. Partridge, T.N. Huffman and B. Moon. The figures have been drawn by W. Voorvelt and M. Clarke assisted with the images. The Moerdyk and Heynes families have been of great help on the Samaria farms. Special thanks are due to Michael Moerdyk, who introduced us to the archaeology on Samaria many years ago, and to his widow Alida Moerdyk and his daughter, Anna-Marie Friedrich, who continue to support our efforts.

References Barham, L. (2002). Backed tools in Middle Pleistocene central Africa and their evolutionary significance. Journal of Human Evolution 43: 585–603. Bordy, E.M. & Catuneanu, O. (2002). Sedimentology and palaeontology of upper Karoo aeolian strata (Early Jurassic) in the Tuli Basin, South Africa. Journal of African Earth Sciences 35, 301–314.

180

Stone age signatures in northernmost South Africa Bordy, E.M., Bumby, A., Catuneanu, O. & Eriksson, P.G. (2004). Advanced early Jurassic termite (Insecta: Isoptera) nests: Evidence from the Clarens Formation in the Tuli Basin, Southern Africa. Palaios 19, 68–78. Clark, J.D. (1970). The Prehistory of Africa. London: Thames & Hudson. Clark, J.D. (2001). Kalambo Falls Prehistoric Site, Vol. III. London: Cambridge University Press. Cooke, C.K. (1966). Re-appraisal of the industry hitherto named the Proto-Stillbay. Arnoldia (Rhodesia) 2(22), 1–14. De Wit, M.C.J. (1996). The distribution and stratigraphy of inland alluvial diamond deposits in South Africa. Africa Geoscience Review, Special Edition 3, 19–33. De Wit, M.C.J., Marshall, R.R. & Partridge, T.C. (2000). Fluvial deposits and drainage evolution. In (T.C. Partridge & R.R. Maud, Eds) The Cenozoic of Southern Africa. New York: Oxford University Press. Gibbon, R. (2002). Preliminary investigation into the context and integrity of an Earlier Stone Age site in the Shashe-Limpopo basin. Honours project, School of Geography, Archaeology and Environmental Studies, University of the Witwatersrand, Johannesburg. Grun, R., Brink, J.S., Spooner, N.A., Taylor, L., Stringer, C.B, Franciscus, R.G. & Murray, A.S. (1996). Direct dating of Florisbad hominid. Nature 382: 500–501. Hall, S. & Smith, B. (2000). Empowering places: rockshelters and ritual control in farmerforager interactions in the Northern Province. South African Archaeological Society Goodwin Series 8: 30–46. Kandel, A.W., Felix-Henningsen, P. & Conard, N.J. (2003). An overview of the spatial archaeology of the Geelbek Dunes, Western Cape, South Africa. In (G. Füleky, Ed.) Papers of the 1st International Conference on Archaeology and Soils. BAR International Series 1163. Kempson, H. (2004). The Hackthorne Earlier Stone Age site: a technological analysis and evaluation. Honours project, School of Geography, Archaeology and Environmental Studies, University of the Witwatersrand, Johannesburg. Klein, R.G. (2000). The Earlier Stone Age of southern Africa. South African Archaeological Bulletin 55, 107–122. Kuman, K., Gibbon, R.J. & Le Baron, J.C. (2005). Earlier Stone Age archaeology of the VhembeDongola National Park and vicinity. Quaternary International, 23–32. Le Baron, J.C. (2003). The geoarchaeology of the Hackthorne 1 site. Biennial Meeting of the Southern African Society for Quaternary Research, March, Johannesburg. Le Baron, J.C. (2004). A landscape of Stone Age archaeology: a primary investigation near Hackthorne 1, Limpopo Province, South Africa. Annual Meeting of the Paleoanthopology Society, McGill University, March, Montreal. McBrearty, S. (1988). The Sangoan-Lupemban and Middle Stone Age sequence at the Muguruk site, western Kenya. World Archaeology 19, 388–420.

181

From Tools to Symbols McBrearty, S. (1991). Recent research in western Kenya and its implications for the status of the Sangoan industry. In (J.D. Clark, Ed.) Cultural Beginnings. Bonn: Dr. Rudolf Habelt GMBH, pp. 159–176. McBrearty, S. & Brooks, A. (2000). The revolution that wasn’t: a new interpretation of the origin of human behaviour. Journal of Human Evolution 39, 453–563. Sampson, C.G. (1974). The Stone Age Archaeology of Southern Africa. New York: Academic Press. Smith, J. (2004). Pers. comm. from Jeanette Smith of the School of Geography, Archaeology and Environmental Studies, University of the Witwatersrand. Smith has made a compilation of the 1975–1998 S.A. Weather Bureau data from the rain stations at Alldays, Musina and Pontdrift for her PhD research project. Sutton, M. (2003). Survey for Middle Stone Age Sites in the Limpopo River Valley, South Africa. MSc Research Report, School of Geography, Archaeology and Environmental Studies, University of the Witwatersrand, Johannesburg, South Africa. Tyson, P.D. & Partridge, T.C. (2000). Evolution of Cenozoic climates. In (T.C. Partridge & R.R. Maud, Eds) The Cenozoic of Southern Africa. New York: Oxford University Press, pp. 371–387. Volman, T.P. (1984). Early prehistory of Southern Africa. In (R.G. Klein, Ed.) Southern African Prehistory and Paleoenvironments. Rotterdam: A.A. Balkema, pp. 169–220. Woodbourne, S. (2003). Personal communication from Dr Stephan Woodbourne, Quaternary Dating Unit, Council for Scientific and Industrial Research, Pretoria, South Africa.

182

Vertebral column, bipedalism and freedom of the hands Dominique Gommery UPR 2147 du CNRS, 44 Rue de l’Amiral Mouchez, 75014 Paris, France Human Origin and Past Environments Research Programme (HOPE), Transvaal Museum, P.O. Box 413, Pretoria 0001, South Africa

Abstract The popular literature on human evolution suggests that trunk erectness and the appearance of permanent bipedalism implies a freeing of the hands. The hands, no longer having locomotor constraints, acquired a more important dexterity in the handling of objects, including tools. Observations of wild chimpanzees show, however, that tool use does not correlate with the acquisition of trunk erectness and permanent bipedalism. Manipulating objects can be done in a sitting position. But trunk erectness and the acquisition of permanent bipedalism will undoubtedly increase the capacity to handle objects, either standing still or moving. Permanent bipedality in extinct species can only be studied using postcranial skeletons of Plio-Pleistocene or upper Miocene hominids. South Africa has one of the most important fossil hominid postcranial collections. The site of Sterkfontein alone has yielded three partial skeletons of australopithecines (Australopithecus africanus). That of Sts 14 has been employed to reconstruct the vertebral column and pelvis of ‘Lucy’ (AL288-1) (Australopithecus antiquus seu afarensis) from Ethiopia. Sts 14 was discovered in 1947 by R. Broom and J.T. Robinson, pioneers of palaeoanthropology in South Africa. South Africa has also yielded another type of PlioPleistocene hominid, Paranthropus robustus. They show many resemblances to Australopithecus africanus but have anatomical characteristics in their postcrania by which they differ. These characteristics might be useful in revealing a specific adaptation to their mode of locomotion and thus to their type of bipedality. Australopithecus africanus and paranthropines are divergent from other types of hominid from East Africa. These anatomical and functional differences must have implications for their behaviour, as in the way they handled or made objects and/or tools. Some authors consider the South African paranthropines capable of making tools. This hypothesis, in

183

From Tools to Symbols particular, has been supported by the study of a phalanx of the thumb from Swartkrans. The discovery of phalanges of other hominids, like that reported for Orrorin tugenensis, makes it possible to review this conclusion.

Résumé Classiquement, on trouve dans les ouvrages de vulgarisation que le redressement du corps et donc de l’apparition de la bipédie permanente ont entraîné une libération des mains. Ces mains n’ayant plus de contraintes locomotrices ont acquis une dextérité plus importante dans la manipulation d’objets et d’outils. L’observation de chimpanzés vivant en milieu naturel montre cependant que l’emploi d’outils ne nécessite pas le redressement du tronc et la bipédie permanente. La manipulation peut se faire en position assise. Mais le redressement du tronc et l’acquisition de la bipédie permanente vont favoriser l’accroissement de cette capacité à manipuler les objets, ceci en position statique ou lors de déplacements. La bipédie permanente, caractéristique biologique de l’homme, ne peut-être étudiée qu’à partir du squelette postcrânien pour les hominidés Plio-Pléistocène ou du Miocène supérieur. L’Afrique du Sud présente l’une des plus importantes collections de restes postcrâniens d’hominidés Le site de Sterkfontein a, à lui seul, livré trois squelettes partiels d’australopithèque (Australopithecus africanus). Celui de Sts 14 a été à plusieurs reprises utilisé pour reconstituer le rachis et le bassin de Lucy (AL288-1) (Australopithecus antiquus seu afarensis). Sts 14 a été découvert en 1947 par R. Broom et J.T. Robinson, deux pionniers de la paléoanthropologie sudafricaine. L’Afrique du Sud a livré un autre type d’hominidé plio-pléistocène, le paranthrope (Paranthropus robustus). Ce dernier révèle plusieurs ressemblances avec Australopithecus africanus mais présente aussi quelques spécificités anatomiques dans le squelette postcrânien. Ces spécificités pourraient être révélatrices d’une bipédie particulière. Australopithecus africanus et Paranthropus robustus se distinguent des hominidés d’Afrique orientale. Ces différences anatomiques et fonctionnelles ont sans doute eu des implications sur leur comportement comme dans la façon d’utiliser ou de fabriquer des objets et/ou des outils. Certains auteurs considèrent que les paranthropes étaient capables de fabriquer des outils. Soutenue par l’étude d’une phalange du pouce découverte à Swartkrans, cette hypothèse est actuellement nuancée, après la découverte de phalanges d’autres d’hominidés, comme celles attribuées à Orrorin tugenensis.

Introduction Humans are not the only animals who handle tools or objects in the living world. Some manipulate stones or pieces of wood (examples include birds, otters and some non-human primates, amongst others). In the popular literature one reads that trunk erectness and the emergence of permanent bipedalism implied a freeing of the hands. The hands, no longer having locomotor constraints, acquired a more important dexterity in the handling of objects including tools. In the wild, chimpanzees handle objects in a sitting position. Bipedalism and handling or making of tools are not necessarily linked, although trunk erectness and the acquisition of permanent bipedalism might be expected to increase the capacity to handle objects, both while standing still and while moving.

184

Vertebral column, bipedalism and freedom of the hands

The first purpose of this article is to show the importance of postcranial bones in understanding the emergence of bipedalism and associated characteristics. Several lineages of Plio-Pleistocene hominids are known. Do they all present similar characteristics in the trunk? Did only one type of bipedalism evolve, or several? Each of these anatomical characteristics of the first hominids would have had consequences in terms of their behaviour. The second point is to show from the example of the distal phalanx of the thumb that the characteristics known to be human and classically extrapolated to relate to the manufacture of tools, are perhaps specifically hominid characters. The handling of objects typical of humans could be an indirect effect related to the anatomical and functional characteristics of a hominid. Most of the anatomical characteristics discussed here were observed on fossils from South African collections, which have some exceptional specimens, including partial skeletons. Indeed, the study of some parts of the trunk requires the observation of an anatomical segment (succession of several consecutive vertebrae) or of a complex articulation.

The trunk and pelvis of hominids Characteristics of erectness and non-erectness in the trunk of hominoids Homo sapiens is associated with an orthograde posture and permanent bipedal locomotion. All extant non-human hominoids also show both these characteristics, but only occasionally. African great apes have a semi-erect trunk and a locomotor habit, with a preference for knuckle-walking. The human trunk is characterised by three curvatures: cervical lordosis, thoracic cyphosis and lumbar lordosis; the cervical component has an ‘erect’ atlas-axis, the lumbar segment is long and the sacrum is wide and short. By contrast, the trunk in chimpanzees has only two curvatures: cervical lordosis and thoracic-lumbar cyphosis (Gommery, 1998). The atlas is tilted on the axis, the lumbar segment is short and the sacrum is narrow and long. Plio-Pleistocene hominid material and reconstruction of the trunk Many australopithecine reconstructions use ‘Lucy’ (AL288-1), as a frame of reference for the trunk and pelvis (e.g. Schmid, 1983, 1991), but this partial skeleton is probably not the best example to understand the postcranial morphology of the australopithecines. ‘Lucy’ was discovered by M. Taieb, Y. Coppens and D. Johanson at Hadar in Ethiopia in 1974, and described as A. afarensis (Johanson et al., 1982). More recently the skeleton has been attributed to Australopithecus antiquus seu afarensis (Senut, 1995, 1996). It is dated to 3,2 Mya (Walter, 1994).

185

From Tools to Symbols

The partial skeleton Sts 14 from Sterkfontein (Broom & Robinson, 1950; Robinson, 1972; Thackeray & Gommery, 2002) has been employed to help reconstruct the vertebral column and pelvis of ‘Lucy’, although it is usually considered a full adult (Abitbol, 1995; Berge & Gommery, 1999; Häusler & Schmid, 1995). Sts 14 has an age of c. 2,5 My and is attributed to Australopithecus africanus. It was discovered in 1947 by R. Broom and J.T. Robinson. This skeleton is thought to be associated with ‘Mrs Ples’ (Sts 5), representing a sub-adult (Thackeray et al., 2002a, 2002b). South African hominid postcrania (Australopithecus africanus and Paranthropus robustus) are important in terms of sample size and quality of preservation. For example, Sterkfontein has two other partial skeletons, notably Stw 431, discovered by A. R. Hughes (Benade, 1990) and Stw 573, discovered by R. Clarke, S. Motsumi & N. Molefe (Clarke, 1998, 2002), which has the nickname ‘Little Foot’. Anatomical characteristics of the trunk and pelvis Three anatomical segments were used for this study. The cervical segment

The partial skeletons referred to above have no upper cervical vertebrae but one characteristic of the vertical trunk is the ‘erect’ nature of the atlas-axis. This feature is dependent on the morphology of the axis. In the Plio-Pleistocene hominid samples available for study thus far, we have only two axis pieces (Gommery, 1995, 1997): 1-AL 333.101 (Fig.1) is an axis discovered in the levels DD2 and DD3 of the Denen Dora Member of the locality AL 333 from Hadar in Ethiopia and is dated at 3,3 Mya (Lovejoy et al., 1982; Walter, 1994); 2-SK 854 (Fig.1) is an axis discovered in Member 1 of Swartkrans in South Africa and is dated at circa 1,8 Mya (Robinson, 1972; Brain, 1993). The two specimens differ in the orientation and morphology of the superior articular facet. The morphology of the superior articular facets in SK 854 is cone-shaped near the odontoid process, while the Hadar specimen is platform-shaped, as found in humans. This cone-shaped morphology suggests that the atlas would tilt a little on the axis in SK 854. Delattre (1924) observed that the orientation of the odontoid process and the morphology of the superior articular facets are linked. In the Ethiopian axis, the odontoid process is straight as in humans, whereas in the Swartkrans specimen, just the beginning of angulation of this process is visible; it seems that the odontoid process was slightly angulated. The association of the superior articular facet with the odontoid process makes a functional plan of rotation but also an angulation with the vertebral body of the axis. The angulation of the odontoid is responsible for one component of the lordosis curvature of the spinal column. In humans, the atlas is erect on the axis, in comparison with the condition in chimpanzees, where it is tilted on

186

Vertebral column, bipedalism and freedom of the hands

Figure 1 Plio-Pleistocene Axes: (a-b) ventral and lateral view of SK 854 from Swartkrans, South Africa, (c-d) ventral and lateral view of AL 333.101 from Hadar, Ethiopia.

the axis. In the Ethiopian specimen, the atlas would be erect on the axis, while in the South African specimen, the atlas would be slightly tilted on the axis, but less so than in chimpanzees. Another significant difference is the presence of a well-developed and sharp ventral crest on the vertebral body (crista ventralis) in the South African axis. This crest ends in a prominent tubercle (tuberculum anterius). The ventral face of the body is very similar to the morphology of the bonobo, but unlike that of the Ethiopian fossil. The latter is more human in form. These results are supported by specific angles: the homologous angle of the superior articular facets of the axis (Fig. 2). This angle was defined in previous studies (Gommery, 1995, 1999, 2000) after observations on the superior articular facet of the atlas (fovea articulares superiores atlantis) (Gommery, 1995, 1996), and especially the posterior part of this facet, the retro-glenoid tubercle. The retro-glenoid angle corresponds to the orientation of curvature of the two retro-glenoid tubercles in the same atlas. The orientation of the posterior part of the inferior articular facets of the atlas and of the superior articular facets of the axis are dependent on the orientation of the retro-glenoid tubercle. The orientation of the posterior part of the inferior articular facets of the atlas and of the superior articular facets of the axis are defined by the homologous angle of the atlas and the axis. The value of the three angles are nearly similar for the same specimen. The homologous angle of SK 854 has a value (85°), lying between the ranges of variability of apes (73°–84°) and humans (87°–96°). The Ethiopian axis angle is 90°, lying within the human range of variability. These homologous angles of the superior articular facets of the axis have implications for movement between atlas and axis.

187

From Tools to Symbols

The morphology of the vertebral body and superior articular surface of the axis reveals certain biomechanical constraints. The platform morphology, as seen in AL 333.101, corresponds with compression forces typical of upright posture and permanent bipedality, as is the case in humans. The cone-shaped axis corresponds with tensional forces associated with flexion-extension movements. The Swartkrans specimen suggests the presence of more flexion-extension movement than in humans and less than in great apes. This axis is perhaps more typical of bipedalism associated with climbing, in the same way as some other Plio-Pleistocene hominid postcranial bones (Senut, 1978; Senut & Tardieu, 1985). During the Plio-Pleistocene, different types of axis existed. The Hadar specimen has been attributed to Praeanthropus (Coppens, 1995; Gommery, 1997, 2000; Senut, 1995, 1996). The lumbar segment

The morphology of lumbar vertebrae is important to demonstrate whether or not lordosis existed. The morphology of the vertebral body is clearly different in chimpanzees and humans. The vertebral bodies of lumbar vertebrae are shorter, low and very broad in humans, while in chimpanzees they are longer, high and narrow. The South African Plio-Pleistocene hominid specimens Sts 14 and Stw 8/41, for example, have short, low and broad vertebral bodies (Broom & Robinson, 1950; Tobias, 1980).

Figure 2 Homologous angle of the superior articular facets of axis (1) of Primates with the retro-glenoid angle (2) and homologous angle of the inferior facets of atlas (3): (a) Strepsirhini: Propithecus verreauxi (Pv), Propithecus diadema (Pd), Indri indri (Ii). (b) Platyrrhini: Aotus trivirgatus (Ao), Alouatta seniculus (Ats), Alouatta palliata (Alp), Ateles geoffroyi (Age), Ateles paniscus (Apa), Saimiri oerstoedii (Soe), Saimiri sciureus (Sse), Cebus nigrivittatus (Cni), Cebus apella (Cap), Cebus capucinus (Cca); Cercopithecoidea: Nasalis larvatus (Nl), Presbytis cristata (Pc), Presbytis melalophos (Pm), Presbytis rubicundra (Pr), Colobus badius (Cb), Colobus guereza (Cg), Mandrillus sphinx (Ms), Papio hamadryas (Ph), Papio sp. (Psp), Papio anubis (Pa), Cercopithecus nictitans (Cn), Cercopithecus cephus (Cc), Cercopithecus ascanius schmidti (Ca), Cercocebus torquatis (Ct). (c) Hominoidea non-human: Symphalangus syndactylus (Sy), Hylobates leuciscus (Hle), Hylobates lar lar (Hla), Pongo pygmaeus (Py), Pan paniscus (Pn), Pan troglodytes (Pt), Gorilla gorilla beringei (Ggb), Gorilla gorilla gorilla (Ggg). (d) Human: Homo sapiens sapiens (Hss). (e) SK 854 (XXX).

188

Vertebral column, bipedalism and freedom of the hands

a

b a

c b

d c

e

d

189

From Tools to Symbols

Other morphological features associated with systematic characters or locomotion are also found in such hominids: The position of lumbar transverse processes relative to the dorsal face of the vertebral body: The position of the transverse processes relative to the dorsal face of the vertebral body (Fig. 3) varies in primates (Shapiro, 1993). The transverse processes are dorsal in extant great hominoids such as chimpanzees and humans, Sts 14 and other PlioPleistocene hominids. This feature is associated with a shorter lumbar segment and a muscular organisation that is different from classical quadrupedal primates. The difference in dorsal musculature between monkeys and great hominoids is associated with a modification of thoracic cage shape (Ward, 1993). The dorsal position of the transverse processes increases the distance between the muscle m. erector spinae and the axis of arc created by the lumbar segment. Moreover, it greatly increases the power of these muscles to counteract flexion. The geometrical forms of the four zygapophyses of the last three lumbar vertebrae: The major difference between chimpanzee and hominid is the geometrical shape formed by the summits of the four zygapohyses of the last three lumbar vertebrae in dorsal view (Fig. 4). In chimpanzees, the shapes are trapezoidal and the last one is the smaller. The shapes are rectangular for Sts 14 and man and are bigger for Sts 14 than human. The morphology of the last three lumbars of Sts 14 proves that lumbar lordosis exists.

a

b

e

c

f

d

g

Figure 3 Position of lumbar transverse processes relative to the dorsal face of the vertebral body: (a) Cebus; (b) Ateles; (c) Cercopithecus; (d) Papio; (e) Hylobates; (f) Pan; (g) Sts 14.

190

Vertebral column, bipedalism and freedom of the hands

a

b

c

d

Figure 4 The last three lumbar vertebrae in chimpanzee (a) and Sts 14 (Australopithecus africanus) (c) and the geometrical forms of the four zygapophyses in chimpanzee (b) and Sts 14 (d). (Modified after Benade, 1990.)

The sacrum and sacro-iliac joint

Next to the last lumbar, the sacrum makes the transition between the vertebral column and the pelvis. Plio-Pleistocene sacral remains are represented by four fossils: AL 288-1 (Lucy) (Johanson et al., 1982); Sts 14q (Robinson, 1972); Stw 431 (Benade, 1990); and DNH 43A (Gommery et al., 2002). The collection of PlioPleistocene sacra is poor compared with the iliac bone collection of the same period (AL 288.1, Sts 14s, Sts 14r, Sts 65, Stw 431, TM 1605, SK 45, SK 3155b and DNH 43B, for example). These sacra are broad and short. The vertebral body is broad for the first sacral vertebra. The superior articular processes are lateral, as are the inferior articular processes of the last lumbar. The transverse processes are well developed. The auricular facets are curved and well developed. The broad sacrum with the large iliac tuberosities provides increased leverage for the muscles of the back that balance the spine over the pelvis (m. erector spinae). These features are associated with an erect trunk and bipedality. A recent study on the sacro-iliac joint of Paranthropus robustus (DNH 43 from Drimolen, dated to nearly 2 Mya), shows some features specific to this taxon (Fig. 5). One upper lateral angle is preserved on this fossil. Stern and Susman (1983) did not observe this feature on the sacrum of ‘Lucy’. A well-developed transverse process with upper lateral angles provides attachment areas for ligaments that stabilise the sacrum. On DNH 43, the auricular surface is very well developed and complex. The sacro-iliac joint is very strong between the auricular surface of the sacrum and the ilium. These features might represent a particular adaptation to their own type of bipedalism.

Pollical distal phalanx The hand is the major organ implicated in handling and making tools or other objects. For the hand, we focus on the pollical distal phalanx. The morphology of

191

From Tools to Symbols

a

b

Figure 5 Sacro-iliac joint of DNH 43 (Paranthropus robustus) from Drimolen (South Africa): (a) position of upper lateral angle; (b) auricular surface of iliac bone.

this bone was used to test the possibility of making tools among paranthropines (Shrewsbury & Johnson, 1983; Susman, 1988a, 1988b, 1998). This bone is very different in humans and chimpanzees, in terms of both size (larger in humans than in chimpanzees) and morphology. Chimpanzees regularly handle objects or tools, but not in the same way as humans. Humans make more use of the thumb, and especially of the pollical distal phalanx. The thumb is very strong in humans. The oldest hominid pollical distal phalanx, BAR 1901’01 (Fig. 6) from the site of Kapsomin in the Lukeino Formation in Kenya, is dated to nearly 6 Mya (Pickford & Senut, 2001; Sawada et al., 2002), and has been recently discovered (Gommery & Senut, 2002a, 2002b). The Kapsomin site has yielded one new hominid, Orrorin tugenensis (Senut et al, 2001). This hominid is more bipedal than some Plio-Pleistocene hominids, such as ‘Lucy’, but it also shows climbing adaptations (Pickford et al., 2002; Senut et al, 2001). In comparing this fossil with humans, it is noted that this phalanx has a strong and broad apical tuft, a ‘horseshoe-shaped’ apical tuft, as in the posterior view of the human homologue. In chimpanzees, the apical tuft is very small with a spherical morphology. In anterior view, the pollical distal phalanx has a very deep ventral depression for the m. flexor digitis pollicis longus and an asymmetrical apical tuft as in humans. In chimpanzees, this depression is sometimes present but is much less developed (Shrewsbury et al, 2003). The pollical distal phalanx of hominids is rare in the fossil record; two South African

192

Vertebral column, bipedalism and freedom of the hands

Figure 6 Dorsal view of Bar 1901’01, a pollical distal phalanx of Orrorin tugenensis.

fossils are available for comparison. The first is from Sterkfontein, SKX 5016 (Susman, 1988a, 1988b, 1998), and is dated to 1,8 Mya; it has been attributed to Paranthropus robustus. The second, Stw 294 from Sterkfontein, is dated to 2,6–2,5 Mya and has been attributed to Australopithecus africanus (Ricklan, 1987, 1988). BAR 1901’01 shows both resemblances and differences to the South African fossil hominids. The resemblances are the lateral developments of the proximal part, the proximal overhang of the tuft, the ‘horseshoe-shaped’ apical tuft, a deep depression for the m. flexor digitis pollicis longus, and a shaft that tapers slightly towards the distal end. The main difference is that the pollical distal phalanx has a robust apex on its distal part and a reduced apical breadth. The anatomical features of the pollical distal phalanx used by some authors (e.g. Shrewsbury & Johnson, 1983; Susman, 1988a, 1988b, 1998) to suggest that paranthropines made tools are also found in Australopithecus africanus and Orrorin tugenensis. These characteristics are probably plesiomorphic hominid features which are not necessarily associated with the manufacture of tools.

Conclusions The South African collections contribute substantially to facilitating an understanding of the emergence of bipedalism and the use of the hand in hominid evolution. During the Plio-Pleistocene and probably during the upper Miocene, different modes of bipedalism developed in hominids. It is probable that in most of these the hand was used in locomotion, but less so than in modern African great apes. Many hominids may have handled objects, not only in a sitting position. Paranthropines had a stabilised trunk on the pelvis and may have been more bipedal than other australopithecines. These anatomical and functional characteristics have consequences for the behaviour of these hominids.

193

From Tools to Symbols

But were paranthropines responsible for the stone tools discovered at sites such as Kromdraai or Swartkrans? The anatomical morphology of the pollical distal phalanx unfortunately does not permit an answer this question. Orrorin tugenensis, dated at 6 Mya, displays the same morphology. Perhaps all hominids had the potential to handle objects, but not the same ability to make tools. If these characteristics meet at the first hominids, they probably all have functional consequences like the potentiality of arboricolism. To make tools, a design plan is required, and that itself requires organisation of the brain.

Acknowledgements I would like to thank Professor Phillip V. Tobias for welcoming me into his laboratory at the University of the Witwatersrand on several occasions since my first visit in 1995. I thank the two organisers, L. Backwell and F. d’Errico for the invitation to participate in the International Round Table, ‘From Tools to Symbols. From Early Hominids to Modern Humans – Des outils aux symboles. Des premiers Hominidés aux hommes modernes’, University of Witwatersrand, South Africa, 16–18 March 2003. I also wish to express my thanks to F. Thackeray, B. Senut and M. Pickford for their assistance.

References Abitbol, M. (1995). Reconstitution of the Sts 14 (Australopithecus africanus) pelvis. American Journal of Physical Anthropology 96, 143–158. Benade, M. (1990). Thoracic and lumbar vertebrae of African hominids ancient and recent: Morphological and functional aspects with special reference to upright posture, Unpublished Masters Dissertation, University of Witwatersrand, Dept. of Anatomy, Thesis 47. Berge, C. & Gommery, D. (1999). Le sacrum de Sterkfontein (Australopithecus africanus) : Nouvelles données sur la croissance et sur l’âge osseux du spécimen (Hommage à R. Broom et J. T. Robinson). Comptes Rendus de l’Académie des Sciences de Paris 329, IIa, 227–232. Brain, C. K. (1993). Structure and stratigraphy of the Swartkrans Cave in the light of the new excavations. In (C.K. Brain, Ed.) Swartkrans: a Cave’s Chronicle of Early Man, Transvaal Museum Monograph, 8, 23–33. Pretoria: Transvaal Museum. Broom, R. & Robinson, J. (1950). Further evidence of structure of the Sterkfontein Ape-Man Plesianthropus. Annals of the Transvaal Museum 4, 1, 1–83. Clarke, R. (1998). First ever discovery of a well-preserved skull and associated skeleton of Australopithecus. South African Journal of Science 94, 460–463. Clarke, R. (2002). Newly revealed information on the Sterkfontein Member 2 Australopithecus skeleton. South African Journal of Science 98, 523–526. Coppens, Y. (1995). Paléoanthropologie et préhistoire. Annales du Collège de France, 1994–1995, 595–627.

194

Vertebral column, bipedalism and freedom of the hands Delattre, A. (1924). Essai sur l’anatomie comparée et la mécanique fonctionnelle de l’axis de Mammifères. Armentières: Watrelot. Gommery, D. (1995). Le rachis cervical des primates actuels et fossiles, aspects fonctionnel et évolutif. Thèse de Doctorat de l’Université de Paris 7 – Denis Diderot, nouveau régime; UFR: Biologie – Sciences de la Nature. 2 Vols. Gommery, D. (1996). Nouvelle approche de la morphologie des cavités glénoïdes de l’atlas (foveae articulares superiores atlantis) chez les primates actuels. Comptes Rendus de l’Académie des Sciences de Paris 323, IIa, 1067–1072. Gommery, D. (1997). Les atlas et les axis des hominidés du Plio-Pléistocène: morphologie et systématique. Comptes Rendus de l’Académie des Sciences de Paris 325, IIa, 639–642. Gommery, D. (1998). Axe vertébral, hominoidea fossiles et posture orthograde: préambule à la bipèdie. Primatologie 1, 135–160. Gommery, D. (1999). Les angles rétro-glénoïdiens et homologues du rachis cervical supérieur des primates actuels. Comptes Rendus de l’Académie des Sciences de Paris 329, IIa, 527–531. Gommery, D. (2000). Superior cervical vertebrae of a Miocene hominoid and a plio-pleistocene hominid from Southern Africa. Palaeontologia Africana 36, 139–145. Gommery, D. & Senut, B. (2002a). Orrorin tugenensis distal thumb phalanx. From Samburupithecus to Orrorin: origins of hominids. Geological and palaeontological background, 28–30 September 2002, International Workshop at Bogorio, Kenya, p.2. Gommery, D. & Senut, B. (2002b). L’extrémité du pouce de l’ancêtre du millénaire (Orrorin tugenensis-Kenya). XIVème colloque de la SFDP, 23–25 octobre 2002, Doué-la-Fontaine, p.15. Gommery, D., Senut, B. & Keyser A. (2002). Un bassin fragmentaire de Paranthropus robustus du site plio-pléistocène de Drimolen (Afrique du Sud). Geobios 35, 265–281. Häusler, M. & Schmid, P. 1995. Comparison of the pelves of Sts 14 and AL 288-1: implications for birth and sexual dimorphism in australopithecines. Journal of Human Evolution 29, 363-383. Johanson, D. C., Lovejoy, C. O., Kimbel, W. H., White, T. D., Bush, M. E., Latimer, B. M. & Coppens, Y. (1982). Morphology of the Pliocene partial hominid skeleton (AL.288-1) from the Hadar formation, Ethiopia. American Journal of Physical Anthropology 57, 403–451. Lovejoy, C. O., Johanson D. C. & Coppens Y. (1982). Elements of the axial skeleton recovered from the Hadar formation: 1974–1977 Collections. American Journal of Physical Anthropology 81, 131–135. Pickford M. & Senut B. (2001). The geological and faunal context of Late Miocene hominid remains from Lukeino, Kenya, Comptes Rendus de l’Académie des Sciences de Paris 332 (Série IIa), 145–152. Pickford, M., Senut, B., Gommery, D. & Treil, J. (2002). Bipedalism in Orrorin tugenensis revealed by its femora. Comptes Rendus Palevol 1, 191–203. Ricklan, D. (1987). Functional anatomy of the hand of Australopithecus africanus. Journal of Human Evolution 16, 643-664.

195

From Tools to Symbols Ricklan, D. (1988). A functional and morphological study of the hand bones of early and recent South African hominids. Ph D Thesis, University of the Witwatersrand. Robinson, J. T. (1972). Early Hominid Posture and Locomotion. London: University of Chicago Press. Sawada, Y., Pickford, M., Senut, B., Taya, T., Hyodo, M., Miura, T., Kashine, C., Chujo, T. & Fujii, H. (2002). The age of Orrorin tugenensis, an early hominid from the Tugen Hills, Kenya. Comptes Rendus Palevol 1, 293–303. Schmid, P. (1983). Eine Rekonstruktion des Skelettes von AL 288-1 (Hadar) und deren Konsequenzen. Folia Primatologia 40, 283–306. Schmid, P. (1991). The trunk of the australopithecines. In (Y. Coppens & B. Senut, Eds) Origine(s) de la Bipédie chez les Hominidés, pp. 225–234. Paris: Cahiers de Paléoanthropologie, Editions du CNRS. Senut, B. (1978). Contribution à l’étude de l’humérus et de ses articulations chez les Hominidés du Plio-Pléistocène. Thèse de 3ème cycle, Université de Paris, 6-Pierre et Marie Curie, 2 Vols. Senut, B. (1995). D’Australopithecus à Praeanthropus ou du respect du code de nomenclature. Annales de Paléontologie 81(4), 279–281. Senut, B. (1996). Pliocene hominid systematics and phylogeny. South African Journal of Science 92, 165–166. Senut, B. & Tardieu, C. (1985). Functional aspects of Plio-Pleistocene Hominid limb bones: implications for taxonomy and phylogeny. In (E. Delson, Ed.) Ancestors: The Hard Eevidence, pp. 193–201. New York: Alan R. Liss. Senut, B., Pickford, M., Gommery, D., Mein, P., Cheboi, K. & Coppens, Y. (2001). First Hominid from the Miocene (Lukeino formation, Kenya). Comptes Rendus de l’Académie des Sciences de Paris 332, IIa, 137–144. Shapiro, L. (1993). Functional morphology of the vertebral column in Primates. In (D.L. Gebo, Ed.) Postcranial Adaptation in Nonhuman Primates, pp. 121–149. De Kalb: North Illinois University Press. Shrewsbury, M. & Johnson, R. (1983). Form, function, and evolution of the distal phalanx. Journal of Hand Surgery 8(4), 475–479. Shrewsbury, M., Marzke, M., Linscheid, R. & Reece, S. (2003). Comparative morphology of the pollical distal phalanx. American Journal of Physical Anthropology 121, 30–47. Stern, J. & Susman, R. (1983). The locomotor anatomy of Australopithecus afarensis. American Journal of Physical Anthropology 60, 279–317. Susman, R. (1988a). Hand of Paranthropus robustus from Member 1, Swartkrans: fossil evidence for tool behavior. Science 240, 781–784. Susman, R. (1988b). New postcranial remains from swartkrans and their bearing on the functional morphology and behavior of Paranthropus robustus. In (F.E. Grine, Ed) Evolutionary History of the Robust Australopithecines, pp. 149–172. New York: Aldine de Gruyer.

196

Vertebral column, bipedalism and freedom of the hands Susman, R. (1998). Hand function and tool behavior in early hominids. Journal of Human Evolution 35, 23–46. Thackeray, F., Braga, J., Treil, J., Niksch, N. & Labuschagne, J. (2002a). ‘Mrs Ples’ (Sts 5) from Sterkfontein: an adolescent male? South African Journal of Science, 98, 21–22. Thackeray, F. & Gommery, D. (2002). Spatial distribution of australopithecine specimens discovered at Sterkfontein between 1947 and 1949. Annals of the Transvaal Museum 39, 70–72. Thackeray, F., Gommery, D. & Braga, J. (2002). Australopithecine postcrania (Sts 14) from the Sterkfontein Caves, South Africa: the skeleton of ‘Mrs Ples’? South African Journal of Science 98, 211–212. Tobias, P. V. (1980). ‘Australopithecus afarensis’ and A. africanus: Critique and alternative hypothesis. Palaeontologica Africana 23, 1–17. Walter, R. C. (1994). Age of Lucy and the first family: single-crystal Ar40/Ar39 dating the Denen Dora and lower Kada Hadar members of the Hadar formation, Ethiopia. Geology 22, 6–10. Ward, C. (1993). Torso morphology and locomotion in Proconsul nyanzae. American Journal of Physical Anthropology 92, 291–328.

197

Characterising early Homo: cladistic, morphological and metrical analyses of the original Plio-Pleistocene specimens Sandrine Prat UPR 2147 du CNRS, 44 rue de l’Admiral Mouchez, 75014 Paris, France

Abstract Since the discovery in 1959 of the first specimens allocated to Homo habilis in the Olduvai Gorge, no consensus has been achieved concerning the status of the species Homo habilis, and the taxonomic allocation of the specimens of early Homo. Four hypotheses have been expressed: (1) the specimens from Olduvai, East Turkana and Omo belong to the same palaeo-species: Homo habilis sensu lato; (2) the hypodigm is heterogeneous: two species could be defined in that group, Homo habilis sensu stricto and Homo rudolfensis; (3) these species do not belong to the genus Homo but to the genus Australopithecus; or (4) it would be more appropriate to include the specimens of Homo rudolfensis in the genus Kenyanthropus. The goal of this study is to re-evaluate the hypotheses concerning the taxonomy of the specimens attributed to early Homo, and to test whether they belong anatomically to the genus Homo or to another genus. A morphological comparative study, a craniofacial variation study and numerical cladistic analyses were carried out on the original Plio-Pleistocene specimens. The Operational Taxonomic Unit used in this analysis is defined by the specimen and not by the species (as often used) in the absence of consensus on the content of the hypodigm of the species Homo habilis. The results of this analysis show that based on the cranial specimens: (1) two species can be distinguished: habilis and rudolfensis; (2) the specimens belonging to these two taxa are included in the Homo clade; (3) the conclusions concerning the revision of the genus Homo and the inclusion of the specimens of Homo habilis and Homo rudolfensis in the genus Australopithecus or Kenyanthropus are questionable.

198

Characterising early Homo

Résumé Il n’existe actuellement aucun consensus concernant la taxinomie et la constitution de l’hypodigme de Homo habilis Leakey et al., 1964. Quatre hypothèses majeures sont aujourd’hui avancées à partir des restes crâniens et mandibulaires attribués classiquement à Homo habilis sensu lato: (1) tous les spécimens appartiennent à un seul et même taxon : Homo habilis; (2) deux espèces peuvent être identifiées dans ce groupe : Homo habilis sensu stricto et Homo rudolfensis; (3) ces spécimens n’appartiennent pas au genre Homo mais au genre Australopithecus; (4) les spécimens de l’espèce rudolfensis devraient être mis dans le genre Kenyanthropus. Le but de cette étude est de réévaluer de façon critique la variabilité au sein de cet ensemble fossile du Plio-Pléistocène et de tester si les spécimens appartiennent bien au genre Homo. Trois types d’approches ont été entrepris sur les restes crâniens: une comparaison morphologique, une étude de la variabilité et une analyse cladistique. Dans cette dernière analyse, l’Unité Taxinomique Opérationnelle (OTU) est constituée par le spécimen et non l’espèce, en l’absence de consensus concernant l’attribution taxinomique des spécimens fossiles étudiés. Ces analyses portent sur les pièces originales provenant d’Ethiopie, du Kenya, de Tanzanie, du Malawi et de la République d’Afrique du sud. L’étude morphologique permet de mettre en évidence une variabilité considérable au sein des premiers représentants du genre Homo. Deux espèces peuvent être définies au sein de cet ensemble: habilis et rudolfensis. L’analyse cladistique des 122 caractères crâniens montre l’appartenance de ces deux espèces au genre Homo et non aux genres Australopithecus ou Kenyanthropus.

Introduction Since the discovery of the first specimens attributed to Homo habilis in the Olduvai Gorge in 1959, no consensus has been reached on the taxonomic allocation of specimens of early Homo, and on the hypodigm constitution of the species defined by Leakey, Napier and Tobias in 1964. Ninety dental, cranial and mandibular fragments have been allocated to this taxon. These specimens have been discovered in Ethiopia (Hadar, Omo), Kenya (East Turkana, Chemeron), Tanzania (Olduvai), Malawi (Uraha) and the Republic of South Africa (Sterkfontein, Swartkrans, Drimolen) (Fig. 1). They are dated between 2,45 and 1,55 million years. Concerning the taxonomic allocation of the specimens of early Homo, four main hypotheses have been expressed: 1. All the specimens belong to the same species, Homo habilis sensu lato (Howell, 1978; Tobias 1978a, 1991, 2003; Blumenshine et al., 2003). 2. This group is heterogeneous and two taxa can be distinguished, Homo habilis sensu stricto and Homo rudolfensis (Groves & Mazak, 1975; Alexeev, 1978, 1986; Stringer, 1986; Chamberlain, 1987; Groves, 1989; Wood, 1991, 1992, 1996; Rightmire, 1993; Ferguson, 1995; Lieberman et al., 1996; Prat, 1997). The specimens belong to the genus Homo. 3. The two species habilis and rudolfensis do not belong to the genus Homo but to the genus Australopithecus (Wood & Collard, 1999a, b).

199

From Tools to Symbols

Figure 1 Map of distribution of early Homo specimens.

4. It would be more appropriate to put the holotype of the species rudolfensis (KNM-ER 1470) into the genus Kenyanthropus (Leakey et al., 2001). The goal of this paper is to re-evaluate the hypotheses concerning the taxonomy of early Homo, and their generic attribution. For this purpose, a craniofacial variation analysis, a morphological comparative study and cladistic analysis have been carried out. These analyses focus on specimens, not on species. Furthermore, an original subspecies approach on great apes and modern humans is also proposed to test if the traits commonly used to differentiate the specimens of early Homo could be related to sexual attribution or to anatomical age. Previous studies have been based on morphometric analysis (Prat, 1997, 2000a), using cluster analysis, ratio diagram (as used by Rightmire, 1998) and size and shape components. Many three-dimensional methods could be employed (for example Procrustes analyses, EDMA, etc.), but they were not used in our previous studies because of the fragmentary nature of some specimens. The method developed by Darroch & Mosimann (1985) has been employed, the size component being expressed by the geometric mean, and the shape component by the value divided by the geometric mean.

Material and methods Material Ninety-five cranial specimens were examined (Prat, 2000a), of which thirtytwo were selected in the present study. They represent the best-preserved available

200

Characterising early Homo

specimens. The fossil sample comprises the original cranial specimens discovered in Ethiopia, Kenya, Tanzania, and the Republic of South Africa (Table 1). The specimens from Hadar (Ethiopia) AL 666-1 (Kimbel et al., 1996) and from the Olduvai Gorge OH 65 (face and palate) recently described by Blumenshine and colleagues (Blumenshine et al. 2003) are not included in this study. The sample of humans and great apes studied comprised 226 individuals: 49 Gorilla gorilla (20 Gorilla gorilla gorilla, 20 Gorilla beringei graueri, and 9 Gorilla beringei beringei), 55 Pan troglodytes (30 Pan troglodytes troglodytes, 25 Pan troglodytes schweinfurthi), 20 Pan paniscus, 47 Pongo pygmaeus (27 Pongo pygmaeus pygmaeus, 20 Pongo pygmaeus abelii) and 55 modern humans (Prat, 2000a). The age of the specimens was estimated by the degree of eruption of the third molar and the sex was determined on the living animals. Table 1 List of the best-preserved early Homo cranial remains. Site

Crania

Olduvai (Tanzania)

OH 7, 13, 16, 24, 62

East Turkana (Kenya)

KNM-ER 807, 1470, 1478, 1590, 1805, 1813, 3732, 3735, 3891

Chemeron (Kenya) Sterkfontein (South Africa) Omo (Ethiopia)

KNM-BC 1 Stw 53, Sts 19 L 894-1

OH: Olduvai Hominid; KNM: Kenya National Museums; ER: East Rudolf; BC: Baringo Chemeron; Stw: Sterkfontein Witwatersrand; Sts: Sterkfontein site; L: Omo locality

Methods Morphological comparisons and phylogenetic analyses

For these analyses the trait list, comprising 122 cranial features (Table 2), was compiled from several different studies (Chamberlain & Wood, 1987; Lieberman et al., 1996; Skelton et al., 1986; Skelton & McHenry, 1992; Strait et al., 1997; Stringer, 1987; Tobias, 1991; Zeitoun, 2000) and our own observations on the original material. The features are described and illustrated in Prat (2000a).

201

From Tools to Symbols Table 2 Trait list and character states. 1.

Cranium shape: (0) ovoid, (1) intermediate, (2) long

2.

Cranial shape in norma occipitalis: round (0), (1) trapezoidal, (2) lateral walls divergent on the upper part

3.

Maximum cranial breadth: (0) mastoid, (1) parieto-temporal (2) parietal

4.

Maximum biparietal breadth: (0) low, (1) medium, (2) high

5.

Supraorbital torus: (0) absent, (1) present, continuous, (2) present, not continuous

6.

Supraorbital torus prominence: (0) high, (1) medium, (2) weak

7.

Supraorbitalis sulcus: (0) absent, (1) incomplete, (2) complete

8.

Glabellar region in norma facialis: (0) straight, (1) rounded, (2) depression

9.

Glabellar region in norma lateralis : (0) straight, (1) rounded, (2) depression

10.

Prominence of the glabella: (0) not prominent , (1) medium, (2) prominent

11.

Supraorbital depth: (0) shallow, (1) medium, (2) deep

12.

Lateral postorbital depression: (0) absent, (1) present

13.

Supratrigonal depression: (0) absent, (1) present

14.

Fronto-temporale tubercle : (0) absent, (1) present

15.

Frontal eminence : (0) absent, (1) present

16.

Convexity of the frontal squama: (0) absent or weak, (1) strong

17.

Metopic crest: (0) absent, (1) present

18.

Postorbital constriction: (0) strong, (1) weak

19.

Superior temporal lines on the supraorbital torus: (0) absent, (1) present

20.

Parietal eminence: (0) absent, (1) present, medium position, (2) present, high position

21.

Prelambdoid depression: (0) absent, (1) present

22.

Torus angularis (Weidenreich, 1943): (0) absent, (1) present

23.

Sagittal crest on the anterior part of the line bregma-lambda: (0) present, (1) absent

24.

Development of the sagittal crest: (0) weak, (1) medium, (2) strong

25.

Maximal length of the parietal bone: (0) high, (1) low

26.

Position of the temporal lines: (0) high, (1) medium, (2) low

27.

Temporal lines or crests: (0) lines, (1) crests

28.

Temporal lines position/parietal lines: (0) superior, (1) same level, (2) inferior

29.

Shape of the temporal squama: (0) triangular and low, (1) rounded and high

30.

Orientation of the anterior part of the temporal squama: (0) vertical, (1) anterior (2) posterior

31.

Shape of the superior part of the temporal squama: (0) horizontal, (1) posterior

32.

Lateral development of the supramastoid crest: (0) < 5 mm, (1)5 < d < 10 mm, (2) > 10 mm

33.

Supramastoid development: (0) weak, (1) intermediate, (2) strong

34.

Supramastoid development at porion: (0) absent, (1) present

35.

Mastoid crest individualisation: (0) absent, (1) weak, (2) strong

36.

Supramastoid tubercle: (0) absent, (1) present

202

Characterising early Homo 37.

Junction between the mastoid and the supramastoid crests: (0) absent, (1) present

38.

Slope of the zygomatic process of the temporal/ FH: (0) anterior, (1) parallel

39.

Angle between the zygomatic process and the temporal process: (0) present, (1) absent

40.

Shape of the posterior root of the zygomatic process: (0) elliptic, (1) flat, (2) circular

41.

Position of the posterior root of the zygomatic process: (0) postglenoid process, (1) between the postglenoid process and the anterior zygomatic tubercle, (2) porion

42.

Inflection of mastoids beneath cranial base (0) medial, (1) vertical, (2) lateral

43.

Lateral projection of the mastoid with distinct sulcus: (0) absent, (1) present

44.

Juxtamastoid eminence: (0) absent, (1) weak, (2) strong

45.

Occipitomastoid suture/ juxtamastoid eminence : (0) middle, (1) lateral, (2) medial

46.

Shape of the external meatus auditory: (0) rounded, (1) elliptic SA-IP, (2) elliptic AP, (3) elliptic SI, (4) elliptic AI-SP

47.

Shape of the temporomandibular joint: (0) rectangular, (1) diamond-shaped

48.

Development of the articular eminence: (0) ML=AP, (1)ML > AP, (2) ML >> AP, (3) ML < AP

49.

Shape of the articular eminence: (0) straight, (1) 2 joint surfaces

50.

Continuity between the posterior slope of the articular eminence and tympanic plate: (0) present, (1) absent

51.

Entoglenoid process position: (0) entirely on the squamosal, (1) in part on the sphenoid

52.

Fissure between the tympanic plate and the postglenoid process: (0) present, (1) absent

53.

Position of the entoglenoid process apex and the corotid and ovale foramina axis: (0) lateral, (1) in the axis

54.

Postglenoid process shape: (0) symmetrical, triangular, (1) asymmetrical laterally orientated, (2) rectangular

55.

Position of the postglenoid process relative to the lateral part of the tympanic part of the temporal: (0) lateral, (1) same position, (2) medial

56.

Position of the postglenoid apex relative to the middle of the articular eminence: (0) lateral, (1) middle

57.

Existence of a preglenoid tubercle: (0) absent, (1) present, anterior part of the articular eminence, (2) present, forward the articular eminence

58.

Mandibular fossa depth/ FH: (0) under, (1) same level, (2) above

59.

Continuity of the mastoid fissure: (0) present, (1) absent

60.

Mastoid process/ tympanic plate : (0) same surface, (1) fissure

61.

Paramastoid process: (0) absent, (1) present

62.

Petrous crest development: (0) absent, (1) present

63.

Vaginal process: (0) absent, (1) present

64.

Shape of the tympanic canal: (0) straight, (1) tubular, (2) crest

65.

Eustachian process: (0) absent, (1) present

66.

Shape of the foramen magnum: (0) oval, (1) heart-shaped

67.

Position of the foramen magnum relative to the bitympanic line: (0) posterior, (1) same level, (2) anterior

203

From Tools to Symbols 68.

Position of the basion relative to the PPG-PPG line: (0) same level, (1) posterior, close, (2) posterior, far

69.

Inclination of nuchal plane: (0) near vertical, (1) near FH

70.

Inclination of foramen magnum: (0) posterior, (1) anterior

71.

Dimension nuchal plane/ occipital plane: (0) shorter, (1) longer

72.

Inion / opisthocranion: (0) behind, (1) same level, (2) above

73.

External occipital protuberance: (0) absent, (1) present

74.

Occipital torus: (0) absent, (1) present, development in the medial part, (2) present, development in the medial and lateral parts

75.

External occipital crest:(0) absent, (1) present, not continuous, (2) present, continuous

76.

Linea suprema: (0) absent, (1) small lines, (2) developed

77.

Compound temporal/ nuchal crest: (0) absent, (1) present

78.

Asterionic notch: (0) absent, (1) present

79.

Temporoparietal overlap of occipital at asterion: (0) absent, (1) present

80.

Occipital bun: (0) absent, (1) present

81.

Nasion approaches glabella : (0) yes, (1) no

82.

Nasoalveolar clivus convexity: (0) concave, (1) straight, (2) convex

83.

Maxillary trigon: (0) absent, (1) present

84.

Maxillary groove: (0) absent, (1) present

85.

Shape of the dental arch: (0) uspuloid, (1) parabolic

86.

Canina fossa: (0) absent, (1) present

87.

Infraorbital depression: (0) absent, (1) present

88.

Alignment of canines and incisors: (0) yes, (1) only incisors, (2) no

89.

Convexity of the alveolar part at the incisors level: (0) absent, (1) present

90.

Anterior pillars: (0) absent, (1) present

91.

Incisor region independent of pyriform aperture: (0) no, (1) yes

92.

Facial prognathism: (0) weak, (1) strong

93.

Position of the projection of the temporal process / pyriform aperture: (0) low, (1) middle, (2) high

94.

Lateral zygomatic prominence: (0) absent, (1) present

95.

Zygomaticomaxillary fossa: (0) absent, (1) present

96.

Projection of the zygomatic bone: (0) anterior, (1) vertical, (2) posterior

97.

Frontomaxillary position: (0) P4-M1, (1) M1-M2, (2) M2-M3

98.

Orbital shape: (0) rectangular ML elongation, (1) rectangular SI elongation, (2) square

99.

Superior orbital shape: (0) rectangular, (1) rounded, (2) ovoid

100.

Orientation of the lateral margin of the orbits: (0) vertical, (1) medial

101.

Orientation of the medial margin of the orbits: (0) vertical, (1) lateral

102.

Position of the superior margin/ inferior margin of the orbit: (0) anterior, (1) same level, (2) posterior

103.

Inferior orbital margin rounded laterally: (0) no, (1) yes

104.

Multiple infraorbital foramen: (0) present, (1) absent

204

Characterising early Homo 105.

Anterior nasal spine: (0) absent, (1) present

106.

Nasal eversion: (0) absent, (1) present

107.

Shape of the lateral margin of the nasal aperture: (0) rounded, (1) sharped

108.

Lateral expansion of the superior part of the nasal bones: (0) absent, (1) present

109.

Location of the greatest width of the nasal bones: (0) inferior, (1) superior

110.

Projection of the nasal bones relative to the frontomaxillary suture: (0) above, (1) same level, (2) under

111.

Divided hypoglossal canal: (0) absent, (1) present

112.

Paracondylar process: (0) absent, (1) present

113.

Spine of the ligament of the dens of the axis: (0) absent, (1) present

114.

Styloid process: (0) absent, (1) present

115.

Pterygoid bridge: (0) absent, (1) present

116.

1Endocranial processus of the jugular foramen: (0) absent, (1) present

117.

Exocranial processus of the jugular foramen: (0) absent, (1) present

118.

Spinosum and ovale foramina: (0) distinct, (1) not distinct

119.

Infraorbital fissure: (0) absent, (1) present

120.

Metopic suture: (0) absent, (1) present

121.

Incisive suture closure (anterior part): (0) present, (1) absent

122.

Alveolar prognathism: (0) strong, (1) weak

For the phylogenetic study, a numerical cladistic analysis was carried out on these 122 cranial traits. The outgroup taxa used in the present analysis are mature specimens of Pan troglodytes and Gorilla gorilla. Twenty-five males and twenty-five females of Pan and fifteen males and fifteen females of Gorilla from different populations were sampled from the Musée Royal d’Afrique Centrale de Tervuren (Belgium) and the Powell Cotton Museum (United Kingdom). The hominid ingroup includes twenty-two original specimens for the first analysis, and the data from Kenyanthropus platyops for the second analysis (Table 3). State assignments were based on observations of original fossils except for the Kenyan specimen KNM-WT 40000 (holotype of Kenyanthropus platyops), for which the data came from descriptions in the literature. The scoring for each original fossil is set out in Table 4. All 122 traits were used independently to maximise information, and an effort was made to eliminate the characters which redundantly describe the same morphological feature. No dental data have been included in this analysis because of the presence of many edentulous specimens in the sample. Moreover, since cranial capacity has been estimated for only a small number of specimens, it was not included in this study. In addition, this analysis differs from those of prior studies in the procedures used to define the Operational Taxonomic Unit by the specimen rather by the species (as often used). Indeed, because there is no consensus on the hypodigm of

205

From Tools to Symbols

the species Homo habilis, the creation of OTU on the basis of shared anatomy would have introduced circularity in our analysis. This approach has been used by Capparos (1997), Zeitoun (2000), Prat (2000a), and Gilbert et al. (2003). Character polarity has been determined by rooting the outgroup. The polymorphism is coded as multiple states (0/1, 0/2, 0/1/2) with the polymorphism option of the Paup 3.1 software (Swofford, 1993). The quantitative characters were coded using the method proposed by Thiele (1993) (after a logarithmic transformation of data, these latter being standardised using the formula xs = ((x-min)/(max-min))*n; n = maximum number of ordered states allowable by the algorithm used (32 for PAUP). There has been considerable debate concerning methods of coding quantitative characters. The method of Thiele (1993) was used because it allows the coding of all characters in a similar manner. The data were computed in a non-arbitrary way to avoid any preconceived phylogenetic hypothesis. The most parsimonious trees were obtained using the heuristic algorithm with the options ‘general, starting trees, stepwise addition and branch swapping’. This tree is presented with its length and with its consistency and retention indices. These indices are measures of the amount of homoplasy present. The consistency index Table 3 Chronology and geographical provenance of the specimens used in the cladistic analysis. Specimens

Sites

Age (million years)

AL 333-45 KNM-ER 406 KNM-ER 730 KNM-ER 1470 KNM-ER 1805 KNM-ER 1813 KNM-ER 3732 KNM-ER 3733 KNM-ER 3883 KNM-WT 17000 KNM-WT 40000 OH 5 OH 9 OH 13 OH 16 OH 24 OH 62 Sts 5 Sts 71 Stw 53 Stw 505 SK 46 SK 48

Hadar (Ethiopia) East Turkana (Kenya) East Turkana East Turkana East Turkana East Turkana East Turkana East Turkana East Turkana West Turkana (Kenya) West Turkana (Kenya) Olduvai (Tanzania) Olduvai Olduvai Olduvai Olduvai Olduvai Sterkfontein (Republic of South Africa) Sterkfontein Sterkfontein Sterkfontein Swartkrans Swartkrans

3,22–3,18 (Walter, 1994) 1,7 (Feibel et al., 1989) 1,7 (Feibel et al., 1989) 1,9–1,88 (Feibel et al., 1989) 1,85 (Feibel et al., 1989) 1,9–1,88 (Feibel et al., 1989) 1,9–1,88 (Feibel et al., 1989) 1,78 (Feibel et al., 1989) 1,57+/– 0,08 (Feibel et al., 1989) 2,5 (Feibel et al., 1989) 3,5 (Leakey et al., 2001) 1,85–1,8 (Hay, 1976) 1,25 (Hay, 1976) 1,6 (Hay, 1976) 1,67 (Hay, 1976) 1,88 (Hay, 1976) 1,85–1,75 (Johanson et al., 1987) 2,8–2,4 (Schwartz et al., 1994) 2,8–2,4 (Schwartz et al., 1994) 2,0 (Schwartz et al., 1994) 2,8–2,4 (Schwartz et al., 1994) 1,8–1,5 (Vrba, 1985) 1,8–1,5 (Vrba, 1985)

206

Characterising early Homo

(CI) is calculated as the minimum possible tree-length divided by the observed treelength (Kluge & Farris, 1969; Farris, 1989). If there is no homoplasy in a tree, then its observed length equals the minimum tree-length, and the CI equals one. If homoplasy is present, then the CI is less than one. The retention index (RI) is calculated by subtracting the observed tree-length from the maximum possible tree-length, and then dividing this value by the difference between the maximum and the minimum lengths (Archie, 1989; Farris, 1989). Character–sex dependence, character–age dependence With regard to the significance of the morphological traits, an original sub-species approach on great apes and modern humans is proposed to test if the traits commonly used to differentiate the specimens of early Homo could be related to sexual attribution or to anatomical age. The statistical approaches of the Chi square and Fisher tests provide an exact probability of the correlation between the tested groups. If the probability is p < 0,05, then we can reject the null hypothesis that means that the distribution of the trait is independent of the sexual or anatomical age allocation. Size variation

As noted by Kramer et al. (1995), the most commonly used univariate measurement for testing multiple-species hypotheses and quantifying variation, is to employ the Coefficient of Variation (CVs = (1+1/4n)CV) (Sokal & Braumann 1980), for small sample size n < 30. Furthermore, to take into account fossil sample size, we used a resampling method (bootstrapping) to access size variation in the early Homo sample. Different methods based on the same principles have been used in hominid samples (Cope & Lacy, 1995; Kramer et al., 1995; Lockwood, 1999). Random samples of n (n = fossil sample size) have been generated from extant comparative groups in order to establish a distribution of variation that can be expected from modern analogues. This analysis has been performed for the comparisons of the CVs of the geometric mean for each anatomical region. The geometric mean is the variable used to express the size component (Mosimann, 1970; Jungers et al., 1995). For each anatomical complex, the geometric mean is calculated as the nth root of the product of the n measurements. We compared the CVs of the fossil sample and the frequency histograms of the CVs of 1 000 random samples of extant hominoid (Pan, Gorilla, Pongo). The most sexually dimorphic species has been used as a comparative sample for each anatomical region (Prat, 2000a). The hypothesis is that if the variability of the early Homo sample is greater than that of the random sample, this is evidence for multiple species in Homo habilis, and the single-species hypothesis is rejected. In this present paper, we present only the results of the maxillo-facial complex, for which Gorilla gorilla represents

207

Pan Gorilla

ER 1470 ER 1805 ER 1813 OH 24 OH13 OH16 OH62 STW53 ER406 ER3732 ER730 OH9 OH5 ER3733 ER3883 STW505 STS5 SK48 SK46 STS71 AL333-45 WT17000 WT40000

1 0 0 1 2 1 0 ? 2 ? ? 0 0 ? 2 0 1 1 0 2 ? ? ? ? 2 0

2 0 1 1 0 1 ? ? 0 ? ? 0 0 ? 0 ? 1 1 ? 0 ? ? ? 0 0 ?

3 0&1 0 0 1 0 0 ? ? ? ? 0 ? ? 1 1 0 0 ? 0 ? ? ? ? 0 ?

4 0 0&1 0 0 0 0 ? ? ? ? 0 ? ? ? 0 0 0 ? 0 ? ? ? 0 0 ?

5 2 2 1 ? 1 ? ? ? ? 1 2 1 1 1 1 1 1 1 ? ? ? 0 ? 1 1

6 1 1 1 ? 1 ? ? ? ? ? 1 1 ? ? 0 1 0 1 1 2 1 1 ? 0 1

7 0 0 2 ? 1 2 ? 1 ? 1 1 1 2 2 1 1 1 1 2 2 1 2 ? 1 ?

8 0 1 2 ? 2 1 ? 0 ? ? 0 2 ? 2 1 0 0 1 1 0 ? 0 ? 0 ?

9 1 1 0 ? 0 1 ? 0 ? ? 1 0 ? 2 1 0 0 1 0 1 ? 1 ? 2 ?

10 0 1&2 0 ? 0 1 ? 0 ? ? 2 0 ? 0 1 0 0 2 2 2 ? 0 ? 0 ?

11 1 1&2 1 0 1 1 ? 1 ? 0 0 1 1 2 0 1 1 0 0 2 0 0 ? 1 1

12 0 0 0 ? 0 1 ? 1 ? 1 0 0 0 1 0 0 0 0 1 0 1 1 ? 0 1

13 0 0 0 ? 1 0 ? 1 ? 1 0 0 0 1 0 0 0 0 ? 0 0 1 ? 1 0

14 0 0 0 0 0 ? ? 0 ? 0 0 1 ? 0 0 0 0 0 0 0 0 0 ? 0 0

15 0 0 0 0 0 1 ? 0 ? 0 0 0 0 0 0 0 0 0 0 0 ? 0 ? 0 0

16 1 1 1 1 1 ? ? 1 ? ? 0 1 ? 0 1 0 ? 0 1 1 1 0 ? 1 ?

17 0 0 0 1 0 0 ? 0 ? 0 0 0 ? 0 0 0 0 0 0 0 ? 0 ? 0 0

18 1 1 ? 0 1 ? ? ? ? ? 0 ? ? ? ? 1 1 ? 0 0 ? ? ? 1 1

19 0 0 1 ? 1 1 ? 1 ? 0 1 1 ? 1 1 0 0 0 2 0 0 0 ? 0 0

20 0 0 1 ? 0 1 1 0 ? 1 0 0 ? 0 0 1 1 1 0 0 0 1 0 0 ?

21 0 0 0 1 0 0 0 0 ? ? 0 0 ? ? 0 1 1 ? 0 ? ? 0 0 0 0

From Tools to Symbols

208

Table 4 Taxon and character matrix.

23 1 0&1 1 0 1 1 1 1 ? 1 0 1 ? 1 0 1 1 0 1 0 0 1 0 0

0

24 2 2 2 1 2 2 2 2 ? 2 1 2 ? 2 0 2 2 1 2 1 0 2 1 0 0

25 0&1 0 0 0 0 0 0 0 ? ? ? 1 ? 0 ? 1 1 ? 1 0 0 ? 1 ? ?

26 0&2 0 1 0 1 0 0 0 ? 0 0 1 ? 1 0 1 1 0 0 0 0 0 0 0 ?

27 0 0&1 0 1 0 0 0 1 ? 0 0 0 ? 0 1 0 0 0 0 1 1 0 1 1 0

28 2 2 2 2 1 2 2 2 ? 2 2 2 ? 2 2 2 2 2 2 ? 2 2 2 2 2

29 0 0 1 0 0 0 ? ? ? 0 0 ? ? 0 0 0 0 0 0 0 0 1 0 0 ?

30 0 0&1 1 0 0 0 0 ? ? 0 0 ? ? 0 ? 0 0 0 1 0 ? 1 0 1 ?

31 0&1 0&1 1 0 1 ? ? ? ? 0 1 ? ? 0 ? 1 1 1 0 1 1 1 0 0 ?

32 0&1 1&2 ? 1 0 0 ? ? ? 0 2 ? ? 0 1 1 0 0 1 1 1 1 2 2 ?

33 1&2 1 ? 1 0 ? ? ? ? 2 1 ? ? 0 2 1 1 2 1 0 2 1 1 1 ?

34 0 0 0 0 1 1 ? ? ? 1 0 ? ? 0 1 0 1 0 0 0 0 1 0 0 ?

35 0 0 2 2 0 1 ? ? ? 3 ? ? ? 2 3 3 3 3 0 ? 3 1 1 2 ?

36 0 0 0 0 1 0 ? ? ? 0 1 ? ? 0 0 0 0 1 0 ? 0 1 0 0 ?

37 0 1 1 1 0 ? ? ? ? 0 ? ? ? 0 1 0 0 1 1 ? ? 1 0 1 ?

38 0 0 ? ? ? 0 ? ? ? 0 0 ? ? ? ? 0 ? 0 0 1 ? 0 ? 0 ?

39 0 0 ? ? ? ? ? ? ? 1 1 ? ? ? 1 1 ? ? 1 0 ? 1 ? ? ?

40 0 0 1 ? 0 1 ? 0 ? 1 1 ? ? ? 1 0 0 ? 0 0 0 0 ? 0 2

41 1&2 0 0 ? 0 0 ? ? ? 0 0 ? ? ? 1 0 1 ? 1 1 0 1 ? 0 2

42 0 0 ? 1 1 ? ? ? ? 1 ? ? ? 0 0 0 0 0 1 ? 0 0 0 2 ?

Characterising early Homo

209

Pan Gorilla ER 1470 ER 1805 ER 1813 OH 24 OH13 OH16 OH62 STW53 ER406 ER3732 ER730 OH9 OH5 ER3733 ER3883 STW505 STS5 SK48 SK46 STS71 AL333-45 WT17000 WT40000

22 0 0 0 0 0 0 0 0 ? 0 1 0 ? 0 0 1 0 0 0 ? ? 0 0 0 0

Pan Gorilla

ER 1470 ER 1805 ER 1813 OH 24 OH13 OH16 OH62 STW53 ER406 ER3732 ER730 OH9 OH5 ER3733 ER3883 STW505 STS5 SK48 SK46 STS71 AL333-45 WT17000 WT40000

43 0 0 ? 0 0 ? ? ? ? 1 ? ? ? 0 1 1 1 1 0 ? 1 1 0 1 ?

44 0 0 ? 2 1 ? ? ? ? 1 1 ? ? 1 2 1 1 ? 2 ? ? 1 1 ? ?

45 0 0 ? 0 0 ? ? ? ? 0 2 ? ? 0 0 2 ? 0 1 ? ? 1 2 ? ?

46 0 0 ? 1 0 ? ? ? ? 4 1 ? ? 2 3 2 2 2 2 1 1 2 2 ? 3

47 0&1 0 1 1 0 0 ? ? ? 0 0 ? ? 0 0 0 0 0 0 0 ? 0 ? 0 ?

48 0&1 1&2 1 1 1 ? 1 1 ? ? 1 ? ? 2 2 1 2 0 1 2 ? 1 3 2 ?

49 0&1 0 0 1 ? ? 1 1 ? 1 ? ? ? 0 1 1 1 1 0 1 ? 1 0 0 ?

50 0 0 0 1 1 ? 1 1 ? 1 1 ? ? 1 0 1 1 1 0 1 1 1 0 0 ?

51 0 0 0 0 0 ? 0 0 ? 0 0 ? ? 0 0 1 1 1 0 0 0 0 0 0 ?

52 0 0 ? 0 0 ? ? ? ? 1 0 ? ? 0 0 0 0 0 0 0 ? 1 0 0 ?

53 0 0 ? ? ? ? ? ? ? 0 0 ? ? 1 0 0 0 ? 0 ? ? ? ? 0 ?

54 0 0&1 1 2 2 ? 1 1 ? 1 0 ? ? 1 1 1 0 1 0 0 ? 0 0 0 ?

55 0 0 0 2 1 1 0 0 ? 1 0 ? ? 0 1 0 0 0 0 0 1 0 0 1 ?

56 0 0 ? 0 0 0 ? ? ? 0 0 ? ? 0 0 0 0 0 1 0 0 0 0 1 ?

57 0 0 0 0 0 0 ? ? ? ? 0 ? ? 0 0 0 0 0 0 0 ? 0 ? 1 ?

58 0 0 1 1 0 2 2 ? 1 1 2 ? ? ? ? 1 1 1 0 1 1 1 2 1 ?

59 0 0 ? 0 0 ? ? ? ? 0 0 ? ? 0 0 0 0 0 0 0 0 ? 0 1 ?

60 0 0 ? 1 1 ? ? ? ? 1 0 ? ? 1 0 1 1 1 1 1 1 ? 1 0 ?

61 0 0 ? ? 1 ? ? ? ? 0 0 ? ? ? 1 1 0 0 0 0 0 ? 1 ? ?

62 0 0 ? ? ? 1 ? ? ? 1 0 ? ? 1 ? 1 1 ? 1 ? ? ? ? 0 0

From Tools to Symbols

210

Table 4 Taxon and character matrix (cont).

Pan Gorilla

64 1 1 ? ? 2 2 2 ? ? 1 0 ? ? 2 1 1&2 2 2 2 2 2 ? 0&1 2 1

65 0 0 ? ? ? ? 0 ? ? 0 0 ? ? 0 ? 1 ? ? 1 ? ? ? ? 0 ?

66 0 0 ? 0 ? ? ? ? ? ? 2 ? ? ? 2 0 0 ? 0 ? ? ? 0 ? 0

67 0 0 ? 2 ? ? ? ? ? ? 0 ? ? 0 2 0 1 ? 2 ? ? ? 2 1 ?

68 0 0 ? ? ? ? ? ? ? ? 0 ? ? ? 0 1 1 ? 1 ? ? ? 2 1 ?

69 0 0 0 ? 1 1 ? ? ? ? 1 ? ? ? ? 1 1 ? 1 ? ? ? ? ?

?

70 0 0 ? ? 1 1 ? ? ? ? 1 ? ? ? ? 1 1 ? 1 ? ? ? ? ? ?

71 0 0 ? ? 1 ? ? ? ? ? 1 ? ? ? ? ? ? ? 1 ? ? 0 1 ? ?

72 0 0 ? ? 1 ? ? ? ? ? 1 ? ? ? ? ? ? ? 0 ? ? 1 ? 0 ?

73 0 0 0 1 1 0 ? ? ? 0 0 ? ? 1 1 0 0 0 1 ? ? 1 0 1 ?

74 0 0 0 1 1 ? ? 1 ? 0 0 ? 1 2 2 2 2 0 0 ? ? 1 ? 0 ?

75 0&2 0&1 ? 1 2 2 2 ? ? ? 1 ? ? 2 2 2 1 ? 2 ? ? 1 2 ? ?

76 0 0 1 2 2 0 0 1 ? 1 2 ? 2 2 2 1 1 2 2 ? ? 1 1 1 ?

77 0 0&1 0 1 0 0 ? ? ? 0 1 ? ? 0 1 0 0 0 0 ? ? 0 0 1 0

78 0 0 0 0 0 0 0 0 ? ? 0 ? ? 0 0 1 0 ? 1 ? ? 1 0 ? ?

79 0 0 0 0 0 0 ? ? ? ? 1 ? ? 0 0 0 0 ? 0 ? ? 0 0 ? ?

80 0 0 0 0 0 ? 0 0 ? 0 0 ? 0 1 0 1 1 0 0 ? ? 0 0 0 ?

81 1 0 ? ? 1 ? ? ? ? 1 0 ? ? ? ? ? ? 1 ? ? ? 1 ? ? ?

82 1 0 1 2 1 1 ? ? 1 1 ? ? ? ? 1 1 ? 2 1 1 ? 2 ? 2 1

Characterising early Homo

211

ER 1470 ER 1805 ER 1813 OH 24 OH13 OH16 OH62 STW53 ER406 ER3732 ER730 OH9 OH5 ER3733 ER3883 STW505 STS5 SK48 SK46 STS71 AL333-45 WT17000 WT40000

63 0 0 ? ? ? 1 ? ? ? 1 0 ? ? 1 1 1 1 ? 1 ? ? ? ? 0 ?

Pan Gorilla ER 1470 ER 1805 ER 1813 OH 24 OH13 OH16 OH62 STW53 ER406 ER3732 ER730 OH9 OH5 ER3733 ER3883 STW505 STS5 SK48 SK46 STS71 AL333-45 WT17000 WT40000

83 0 0 0 0 0 0 ? ? 0 ? 0 ? ? ? 0 0 ? 1 1 1 1 1 ? 0 0

84 0 0 0 0 0 0 ? ? ? 0 0 ? ? ? ? 0 ? 1 1 1 ? ? ? ? 0

85 0 0 0 0 0 0 ? ? ? 0 0 ? ? ? 0 0 ? 0 0 0 0 0 ? 0 0

86 0 0 0 0 0 0 ? ? 0 1 0 ? ? ? 0 0 ? 0 1 1 1 1 ? 0 0

87 0 1 0 ? 1 0 ? ? ? 1 0 ? ? ? 0 0 ? 1 0 0 1 1 ? 1 0

88 1&2 1 ? 2 2 2 ? ? 0 0 0 ? ? ? 0 2 ? 2 2 0 0 2 ? 0 1

89 0 0 0 0 0 0 ? ? 0 0 ? ? ? ? 1 0 ? 1 0 0 0 ? ? 0 ?

90 0 0 0 0 0 0 ? ? 0 0 0 ? ? ? 0 0 ? 1 1 1 1 1 ? 0 0

91 1 1 0 0 1 1 ? ? ? 0 0 ? ? ? 0 1 ? 0 0 0 ? 0 ? ? 1

92

0 0 0 ? 0 1 ? ? ? ? 0 ? ? ? ? ? ? ? 1 1 1 1 ? 1

0

93 2 2 ? ? 1 0 ? ? ? ? 1 ? ? ? 1 ? ? 1 2 2 2 2 ? 1 0

94 0 0 ? ? 0 0 ? ? ? 0 1 ? ? ? 1 1 ? 1 1 1 ? 1 ? ? 0

95 0 0 0 ? 1 1 ? ? ? 0 ? ? ? ? 0 0 ? 0 1 0 1 1 ? 1 0

96 1&2 1&2 0 ? 1 0 ? ? ? ? 0 ? ? ? 0 0 ? 1 0 0 0 0 ? 0 1

97 1 0&1 0 0 0 1 ? ? 1 ? 1 ? ? ? 2 1 ? 1 1 0 0 1 ? 1 0

98 0 1 2 ? 1 2 ? ? ? ? 1&2 1 ? ? 0 2 2 0 ? 1 1 0 ? 0 0

99 0&1 0 2 ? 1 1 ? 1 ? ? 1 1 ? ? 2 0 0 1 0 1 1 1 ? 0 0

100 0&1 0 1 ? 1 1 ? ? ? ? 1 0 ? ? 0 0 0 ? 0 0 ? 1 ? 1 0

101 0 0 1 ? 1 0 ? ? ? 0 1 ? ? ? 1 0 0 ? 0 0 ? 0 ? 0 1

102 0 0 1 ? 2 2 ? ? ? ? 0 2 ? ? 0 2 2 0 0 2 ? 0 ? 0 0

From Tools to Symbols

212

Table 4 Taxon and character matrix (cont).

104 0&1 1 ? ? 1 1 ? ? ? 1 1 ? ? ? 1 ? 1 1 1 1 1 1 ? 1 ?

105 0 0 ? 0 1 1 ? ? 1 1 1 ? ? ? 1 1 ? 0 0 0 0 1 ? 1 0

106 0 0 1 0 0 ? ? ? 1 1 0 ? ? ? 0 1 ? 0 0 1 ? 0 ? ? ?

107 0 0 0 0 1 0 ? ? ? 0 0 ? ? ? 0 1 ? 0 0 ? 0 0 ? ? ?

108 109 110 0&1 0&1&2 0 0&1 0 0 0 0 0 0 0 1 0 0 0 1 2 0 ? ? ? ? ? ? ? ? ? ? 0 0 1 2 0 ? ? ? ? ? ? ? ? ? 1 2 0 0 0 0 0 0 0 0 0 0 0 0 0 1 2 0 ? ? ? 1 2 0 ? ? ? ? 0 2 ? ? ?

111 0 1 ? ? ? ? ? ? ? 1 ? ? ? ? 0&1 ? ? ? ? ? ? ? 0 ? ?

112 0 0 ? ? 0 0 ? ? ? 0 0 ? ? ? 0 1 1 ? 0 0 ? ? 1 ? ?

113 0 0 ? ? ? 0 ? ? ? ? 0 ? ? ? 1 0 0 ? 0 0 ? ? ? ? ?

114 0 0 ? 1 1 ? ? ? ? 1 ? ? ? ? 0&1 1 1 ? 0&1 1 ? ? 1 ? ?

115 0 0 ? 0 ? 0 ? ? ? ? 0 ? ? ? ? ? ? ? 0 0 ? ? ? ? ?

116 0 0 ? ? 0 ? ? ? ? 0 ? ? ? ? 0 0 0 ? ? ? ? ? 0 ? ?

117 0 0 ? ? 0 ? ? ? ? 0 ? ? ? ? 0 0 0 ? ? ? ? ? 0 ? ?

118 0 0 ? 1 1 1 0 0 ? ? 1 ? ? ? 1 1 1 ? 1 0 ? ? 1 ? ?

119 0 0 ? 0 1 0 ? ? ? 0 0 ? ? ? 0 ? ? ? 0 0 0 ? ? ? ?

120 0 0 ? ? 1 1 ? ? ? ? 1 ? ? ? 0 1 1 ? 1 0 ? ? ? ? ?

121 0 0 ? 1 1 1 ? ? 1 0 1 ? ? ? 1 ? ? 0 1 1 1 ? ? ? ?

122 0 0 0 ? 0 0 ? ? 0 0 0 ? ? ? ? 1 ? ? 0 0 0 0 ? 0 1

Characterising early Homo

213

Pan Gorilla ER 1470 ER 1805 ER 1813 OH 24 OH13 OH16 OH62 STW53 ER406 ER3732 ER730 OH9 OH5 ER3733 ER3883 STW505 STS5 SK48 SK46 STS71 AL333-45 WT17000 WT40000

103 1 1 0 ? 0 0 ? ? ? ? 1 ? ? ? 1 0 0 ? 1 1 ? 1 ? 1 0

From Tools to Symbols

the most dimorphic species (Prat, 2000b). Twenty-three measurements of the facial skeleton were taken when available (Table 5). Table 5

Craniofacial measurements.

1.

Upper facial height

2.

Alveolar height

3.

Superior facial breadth

4.

Biorbital breadth

5.

Bijugal breadth

6.

Bizygomatic breadth

7.

Bimaxillary breadth

8.

Orbital breadth

9.

Orbital height

10.

Nasal aperture breadth

11.

Nasal height

12.

Upper nasal bones breadth

13.

Lower nasal bones breadth

14.

Maxillo-alveolar length

15.

Palatal length

16.

Palatal breadth

17.

Breadth at M3

18.

Intercanine distance

19.

Distance between nasion and rhinion

20.

Distance between rhinion and subnasal

21.

Zygomatic height

22.

Nasospinal to bimaxillary distance (projection)

Results Morphological comparisons This study supports the recognition of multiple species, based on cranial morphological differences. Two groups are distinguished: 1. KNM-ER 1813, 3735, 3891, BC1 from Kenya; L 894-1, from Ethiopia; OH 7, 13, 16, 24, 62 from Tanzania, and Stw 53 from the Republic of South Africa, named Homo habilis sensu stricto by reference to OH7, holotype of Homo habilis Leakey et al., 1964.

214

Characterising early Homo

Figure 2 Most parsimonious cladogram. Consensus tree, 122 traits, 22 mature specimens (L= 431 steps, IC = 0,452; IR = 0,431).

2. The group defined by KNM-ER 1470, 1590, 3732 from Kenya, named Homo rudolfensis, in reference to KNM-ER 1470, holotype of the species Homo rudolfensis (Alexeev, 1978). Twenty-three morphological traits can differentiate between these two species. No traits are linked to the sexual allocation of the individual (Table 6). The main traits which differentiate the specimens of Homo habilis from those of Homo rudolfensis are located on the maxillo-facial complex. These specimens are distinguished by the proportion between the upper and the lower face, the alveolar prognathism and the size of the zygomatic bone. Cladistic analysis: Homo/Australopithecus Three equally parsimonious trees were obtained, regardless of the option used (general, starting trees, stepwise addition or branch swapping) based on 122 cranial unordered characters, taken from 22 mature fossil individuals. A consensus tree (Fig. 2) was constructed based on the topologies of all trees. The length of this tree is 431 steps with a consistency index of 0,452 and a retention index of 0.431.

215

From Tools to Symbols

The result of the analysis reveals the existence of two monophyletic groups: the Homo clade defined at node A, and the Australopithecus clade defined at node B. The Homo clade is defined by (((((KNM-ER 1470((KNM-ER 1813(OH 16, OH 9))KNMER 730))(OH 24, OH 13)KNM-ER 3732)(KNM-ER 3733, KNM-ER 3883))OH 62)Stw 53) at node A by six unambiguous characters. The synapomorphies at node A are: the asymmetrical shape of the postglenoid process, a uniform petrous crest, an articular eminence with two joint areas (angle greater than 90°) and the angle between the articular eminence and the preglenoid planum (parallel with clade C), a medial position of the parietal prominence and the presence of an anterior nasal spine. The Australopithecus clade is defined at node B by ((((KNM-ER1805((KNMER406,OH5)(SK48, SK46)))(Stw505(Sts5,Sts71)))KNM-WT17000)AL333-45) by four unambiguous characters. The synapomorphies are the presence of temporal and sagittal crests with a moderate expansion and the anterior position of the foramen magnum relative to the bi-tympanic line. Concerning the specimens of early Homo, this analysis suggests that: 1. The specimens KNM-ER 1470, KNM-ER 1813, KNM-ER 3732, OH 24, OH 62 and Stw 53 belong to the genus Homo and not to the genus Australopithecus. 2. The taxonomic attribution of the Kenyan specimen KNM-ER 1805 requires discussion (Prat, 2001, 2002). This specimen, considered by many authors as a Homo habilis specimen, is not grouped with the specimens belonging to the Homo clade defined at node A. It links to node D with the specimens KNM-ER 406 and OH 5 attributed to A. boisei (Paranthropus boisei) and with SK 46 and SK 48, all considered to be A. robustus (P. robustus). Cladistic analysis: Homo/Kenyanthropus The new cladistic analysis (Fig. 3) includes data from the description of the new Kenyan specimen KNM-WT 40000, holotype of the species Kenyanthropus platyops. The result shows that the specimens (KNM-ER 730, 1470, 1813, 3732, 3733, 3883, OH 9, 16, 13, 24, 62, Stw 53) belong to the clade Homo (defined at node A). They are not linked with the specimen of the species Kenyanthropus platyops (KNM-WT 40000). The consensus tree of the two most parsimonious trees has a length equal to 442 steps with a consistency index of 0,442 and a retention index of 0,431. Synapomorphies at node A are a flat glabellar region in norma lateralis, a lateral postorbital depression, the absence of a sagittal crest, weak development of the supramastoid crest, no junctions between the mastoid and supramastoid crests, a uniform petrous crest, a canine region independent of the piriform aperture, no lateral prominence of the zygomatic bone, the inferior part of the orbital region posteriorly located relative to the superior part and the nasal bone eversion.

216

Characterising early Homo

Character–sex dependence and character–age dependence Numerous traits which are used to distinguish the two groups named in this study and in other papers (Lieberman et al., 1996; Strait et al., 1997) could be related to the sex or to the ontogenetic age of the specimen. Examples of features influenced by the ontogenetic age are: the glabellar prominence (for Gorilla gorilla; χ2 = 14,83, p < = 0,0006), the convexity of the nasoalveolar clivus (for Pan paniscus; χ2 = 6,28, p < = 0,01, and Pongo pygmaeus χ2 = 12,37, p < = 0,0009), the nuchal plane inclination (for Gorilla gorilla, χ2 = 12,62, p < = 0,0003), the asterionic notch (for Pongo pygmaeus, χ2 = 9,7, p < = 0,001), and the presence of the sagittal crest (for Gorilla gorilla, χ2 = 25,16, p < = 0,00001). Some other traits are related to the sexual attribution of the individual, for example, the orientation of the zygomatic bone (for Pongo pygmaeus, χ2 = 14,93, p < = 0,005), and the frontal prominence (for Homo sapiens, χ2 = 7,02, p < = 0,0072). Size variation The variability within the early Homo sample (CVs = 0,2995, n = 6, fossils included in the analysis, KNM-ER 1470, 1805, 1813, 3732, OH 24 and Stw 53) exceeds the CVs of the Gorilla gorilla random sample. This latter species has the highest degree of

Figure 3 Most parsimonious cladogram. Consensus tree, 122 traits, 23 mature specimens (L = 442 steps, IC = 0,442; IR = 0,431).

217

and Homo rudolfensis (KNM-ER 1470, KNM-ER 1590 & KNM-ER 3732).

Homo habilis sensu stricto

Homo rudolfensis

Supratrigonal depression

present

absent

Lateral postorbital depression

present (Stw 53, KNM-ER 3891, OH 16, 24)

absent (KNM-ER 3732)

Shape of the temporal squama

triangular and low (KNM-ER 1813, OH 24)

round and high

Orientation of the anterior part of the temporal squama

vertical (OH 13, 24, KNM-ER 1813)

anterior (KNM-ER 1470)

Supramastoid crest at porion

medium

weak

Confluence between the mastoid and supramastoid crests*

absent (KNM-ER 1813, Stw 53, L894-1)

present (KNM-ER 1470)

Shape of the posterior root of the zygomatic process of the temporal bone

elliptic (KNM-ER 1813, OH 16, KNM-BC 1, Sts 19); plane (OH 24, Stw 53)

straight (KNM-ER 1470)

Articular eminence shape *

2 joint surfaces (L894-1, Stw 53, KNM-BC 1, KNM-ER 3891, 3735, OH 13, 16)

straight (KNM-ER 1470)

Position of the postglenoid process relative to the lateral part of the tympanic part of the temporal

same level

medial

high (OH 7, 13, 16, 24)

median (KNM-ER 1470, 1590, 3732)

Frontal bone

Temporal bone

Parietal bone Position of the temporal lines *

From Tools to Symbols

218

Table 6 Comparison of Homo habilis sensu stricto (KNM-ER 1813, KNM-ER 3735, KNM-ER 3891, KNM-BC 1, OH 13, OH 16, OH 24, OH 62, L 894-1 & Stw 53)

Homo habilis sensu stricto

Homo rudolfensis

Nuchal plane inclination *, #

high (KNM-ER 1813, OH 24)

weak (KNM-ER 1470)

Occipital torus #

present, developed in the medial part (KNM-ER 1813, OH 16)

absent (KNM-ER 1470, 1590)

Superior facial breadth/midfacial breadth

greater

less

Alveolar prognathism

reduced

low

Size of the zygomatic bone

Small

strong

Zygomatico-maxillary fossa

present

absent

Orientation of the maxillary zygomatic process relative to the Francfort plane*

vertical

posterior

Position of the frontomaxillary point

M1-M2 (OH 24, 62)

P4-M1 (KNM-ER 1470)

Individualisation of the incisor region relative to the nasal aperture

present

absent

Orbital shape *

rectangular (KNM-ER 1813)

square (KNM-ER 1470)

Shape of the superior orbital margin

round

ovoid

Position of the superior orbital margin relative to the inferior one

anterior

same level

Palatal breadth

narrow

large

Occipital bone

Maxillo-facial complex

Characterising early Homo

219

* trait assignment related to the developmental age of the specimen # high polymorphism in extant great apes

From Tools to Symbols

Figure 4 Frequency distribution of coefficient of variation of 1 000 random samples of Gorilla gorilla geometric means for the face (n = 6).

sexual dimorphism among extant hominoids, especially for the maxillo-facial complex (Prat, 2000b) (Fig. 4). It is therefore probable that the early Homo sample includes more than one species. These results are consistent with those of Lieberman et al. (1988), Kramer et al. (1995) and Grine et al. (1996). However these latter studies, contrary to this study, are based in part on pairwise comparisons, which produce lower probability estimations than those of the bootstrap methods, and are not linked to the question of sample variation.

Discussion and conclusions The results of this study propose two taxa within the hypodigm of Homo habilis sensu lato: H. habilis sensu stricto and H. rudolfensis. Their attribution to the genus Australopithecus or Kenyanthropus is questionable. The conclusions relating to the variability of Homo habilis sensu lato confirm those of some authors (Stringer, 1987; Chamberlain, 1987; Groves, 1989; Wood 1991, 1992, 1996; Rightmire, 1993; Kimbel & Rak, 1993; Lieberman et al., 1996; Strait et al., 1997), but the list of the specimens allocated to that species is different, notably for the specimens KNM-ER 1805, Sts 19, and Stw 53 (Table 7). The taxonomic allocation of the specimen KNM-ER 1805 has been the subject of debate since its description by Richard Leakey in 1974 (e.g. Kimbel et al., 1984; for the complete list of the different attributions, see Prat, 2002). This specimen exhibits many traits not seen in early Homo, such as the presence of a weak posteriorly positioned sagittal crest, a temporo-nuchal crest and the projection of the nasal bones. Morever, its

220

Characterising early Homo Table 7 Specimen attributions (the best preserved crania) Authors

Homo habilis

Homo rudolfensis

Groves (1989)

OH7, 13, 16, 24

KNM-ER 1470, 1590

Wood (1991, 1992, 1996)

OH7, 13, 16, 24, 62, KNM-ER 1805, 1813

KNM-ER 1470, 1590, 3732

Rightmire (1993)

OH7, KNM-ER 1470, 1590, 3732

OH13, 24, KNM-ER 1813 (Homo sp. nov)

Kimbel & Rak (1993)

KNM-ER 1813, 3735, L894-1, Stw 53, Sts 19

KNM-ER 1470, 1590, 3732

Lieberman et al. (1996)

OH 7, 13, 16, 24, 62, KNM-ER 1805, 1813

KNM-ER 1470, 1590, 3732

Strait et al. (1997)

OH 7, 13, 24, 62, KNM-ER 1805, 1813, 3735, Sts 19, Stw 53, SK 847, L894-1

KNM-ER 1470, 1590, 3732

Prat (2000a)

OH7, 13, 16, 24, 62, KNM-ER 1813, 3735, BC1, L 894-1, Stw 53

KNM-ER 1470, 1590, 3732

basicranial morphology, especially the position of the foramen magnum is compatible with its inclusion either in A. boisei or Homo (Dean & Wood, 1982; Prat, 2001, 2002). According to White et al. (1981, p. 456) its persistent metopic suture suggests the possibility of some growth abnormality. However, this feature is not exceptional and occurs frequently in Pliocene and Pleistocene hominids (e.g. Sts 5, KNM-ER 1813, KNM-ER 3733, KNM-ER 3883, KNM-WT 15000 and OH 24). The cladistic analysis also shows that this specimen (KNM-ER 1805) does not belong to the Homo clade. This Kenyan specimen is not an appropriate morph for early Homo, and should not be considered as an average male of Homo habilis for the cladistic analyses and the morphological comparisons (Prat, 2002). The specimen Sts 19 from Sterkfontein exhibits, for the traits taken into account, more affinity with A. africanus (especially Sts 5) than with Homo habilis (Prat, 2000a), as for example in the position of the postglenoid process, and the inflection of mastoids beneath the cranial base. These results do not support the contention that Sts 19 should be excluded from the Australopithecus africanus hypodigm. Our conclusions are similar to those of Ahern (1998) but contrast with those of Kimbel & Rak (1993). However, supplementary data from the temporal bone would be necessary to confirm our hypotheses. The taxonomic position of the South African specimen Stw 53 from Sterkfontein is also the subject of much debate. This specimen was first allocated to Homo by Hughes & Tobias (1977), and recently an allocation to the genus Australopithecus has been advanced (Clarke, 1998; Braga, 1998; Thackeray et al., 2000; Kuman & Clarke, 2000). Spoor et al (1994) have pointed out that Stw 53 has semicircular canal proportions not seen in the other hominid fossils. Our results show that, according the traits taken into account in this analysis, Stw 53 shows closer affinity with Homo habilis than with Australopithecus

221

From Tools to Symbols

africanus. Stw 53 exhibits elevation of the nasal bones; this feature is also observed in OH 62 and KNM-ER 1470, but no specimen allocated to A. africanus (Stw 505, Sts 5) exhibits such a trait. Strong similarity is also observed between Stw 53, OH 24 and KNM-ER 1813 for the maxillary groove, the morphology of the anterior pillar, the anterior nasal spine, the convexity of the alveolar part in the incisive region and the lack of a lateral prominence of the zygomatic (KNM-ER 1813, OH 24, OH 62). The teeth are very large, and the size of the molar teeth increases from front to back. Kuman & Clarke (2000) consider that this morphology is typical for Australopithecus, but this feature is also observed in OH 16. In basal view, Stw 53 exhibits similarities with specimens allocated to early Homo and differences with the fossil attributed to A. africanus (Sts 5) for the shape of the entoglenoid process and the articular eminence (OH 13, OH 16), the development of the petrous crest (OH24), the vaginal (OH24) and eustachian processes (OH13). Contrary to the conclusions of Hughes and Tobias (1977), a styloid process is observed in A. africanus as in the specimen MLD 37/38, and this trait is not specific to early Homo. However, the morphology of Stw 53 is similar to the specimens of A. africanus for the relative position between the canine region and the nasal aperture, the angle at the inion and the position of the temporal lines close to the midline suture (Prat, 2000d). Following the results of the cladistic analysis, Stw 53 belongs to the Homo clade. These results are consistent with our previous analysis based on phenetic distances using size and shape components (Prat, 2000c). These analyses confirm the results of Tobias (1978b, 1980), retaining the specimen Stw 53 in the genus Homo. The results of this cladistic analysis show that the genus Homo is monophyletic even if the specimens of the species habilis and rudolfensis are included in it. These results differ from those proposed by Wood & Collard (1999a, b). The inclusion of the maxillary specimens AL 666-1 and OH 65 should be necessary to confirm these results, but these fragmentary specimens lack many traits used in this analysis. However, one important methodological problem concerns the comparisons between the different cladistic results. The differences between the topologies could be due to the species attribution of the specimen or to the traits included in the analysis. The allocation, for example, of the Kenyan specimen KNMER 1805 to the species Homo habilis has some important consequences for the character state assignment for this species. Indeed, the traits ‘presence of a sagittal crest and presence of a temporo-nuchal crest’ are coded ‘ yes’ for the species habilis (Skelton & McHenry, 1992, p. 321; Strait et al., 1997, p. 26), because of the attribution of KNM-ER 1805 to this species (for the temporo-nuchal crest ‘the H. habilis state assignment is based on KNM-ER 1805’, Strait et al. 1997, p. 69). However, if this individual is not regarded as an average male of Homo habilis (Prat, 2002), then these traits are coded ‘no’ and all the topology of the consensus tree could be modified. Concerning the significance of the morphological traits used in the comparative and cladistic studies, it is essential to test if the traits are homologous (Lieberman,

222

Characterising early Homo

1999; Lockwood & Fleagle, 1999) and independent (not functionally or structurally related, Skelton & McHenry, 1998; Strait & Grine, 1998; Strait, 2001). Moreover, it seems also important to test if the traits can be related to the developmental age or to the sexual attribution of the specimen (Prat, 2000a; Prat & Thackeray, 2001). Indeed, numerous traits which are used to distinguish the two groups named in this study and in other papers (Lieberman et al., 1996; Strait et al., 1997) could be related to the sexual allocation or to the age of the specimen. A re-examination of the significance of the traits is necessary for all comparative and phylogenetic studies. Thus, the definition of the genus Homo becomes more and more difficult.

Acknowledgements I gratefully acknowledge Lucinda Backwell and Francesco d’Errico for inviting me to the International Round Table ‘From Tools to Symbols. From Early Hominids to Modern Humans’ in honour of Professor Phillip V. Tobias. I would also like to thank P.V. Tobias, L. Berger, F. Thackeray, M.G. Leakey, the Office of the President, the Cultural Ministry of Tanzania, the National Museums of Tanzania, National Museums of Kenya and National Museums of Ethiopia, W. Van Neer, H. Hartman and C. Smeenk for permission to study the great apes and fossil hominids in their care. This study was supported by the French Ministry of Foreign Affairs, the Singer-Polignac Foundation and the Fyssen Foundation.

References Ahern, J.C.M. (1998). Underestimating intraspecific variation: the problem with excluding Sts 19 from Australopithecus africanus. American Journal of Physical Anthropology 105, 461–480. Alexeev, V.P. (1978). Paleoantropologia zemnogo shara i formirovanie chelovecheskikh ras (The palaeoanthropology of the world and the forming of the human races). Moscow: Paleolit, Nauka. Alexeev, V.P. (1986). The Origin of Human Race. Moscow: Progress. Archie, J. (1989). Homoplasy excess ratios: new indices for measuring levels of homoplasy in phylogenetic systematics and a critique of consistency index. Systematic Zoology 38, 239–252. Blumenshine, R.J., Peters, C.R., Masao, F.T., Clarke, R.J., Deino, A.L., Hay, R.L., Swisher, C.C., Stanistreet, I.G., Ashley, G.M., McHenry, L.J., Sikes, N.E., van der Merwe, N.J., Tactikos, J.C., Cushing, A.E., Deocampo, D.M., Njau, J.K. & Ebert, J.I. (2003). Late Pliocene Homo and hominid land use from western Olduvai gorge, Tanzania. Science 299, 1217–1221. Braga, J. (1998). Chimpanzee variation facilitates the interpretation of the incisive suture closure in South African Plio-Pleistocene hominids. American Journal of Physical Anthropology 105, 121–135. Capparos, M. (1997). Homo sapiens archaïques: un ou plusieurs (taxons) espèces? Analyse cladistique et analyse morphométrique. Thèse de Doctorat, Museum National d’Histoire Naturelle de Paris.

223

From Tools to Symbols Chamberlain, A.T. (1987). A taxonomic review and phylogenetics analysis of Homo habilis. Ph. D. thesis, University of Liverpool, Liverpool. Chamberlain, A.T. & Wood, B.A. (1987). Early hominid phylogeny. Journal of Human Evolution 16, 119–133. Clarke, R.J. (1998). Sterkfontein and its several australopithecine species. In (M.A. Raath, H. Soodyall, D. Darkban, K.K. Kuykendall & P.V. Tobias, Eds.) Abstracts of contribution to the dual congress, 28 June–4 July 1998, Sun City, South Africa, p. 69. Cope, D.A. & Lacy, M.G. (1995). Comparative application of the coefficient of variation and range-based statistics for assessing the taxonomic composition of fossil samples. Journal of Human Evolution 29, 549–576. Darroch, J.N. & Mosimann, J.E. (1985). Canonical and principal components of shape. Biometrika 72, 241–252. Dean, M.C. & Wood, B.A. (1982). Basicranial anatomy of plio-pleistocene hominids from East and South Africa. American Journal of Physical Anthropology 59, 159–174. Farris, J.S. (1989). The retention index and the rescaled consistency index. Cladistics 5, 417–419. Feibel, C.S., Brown, H. & McDougall I. (1989). Stratigraphic context of fossil hominids from the Omo group deposits: northern Turkana Basin, Kenya and Ethiopia. American Journal of Physical Anthropology 78, 595–622. Ferguson, W.W. (1995). A new species of the genus Homo (Primates: hominidae) from the Plio/ Pleistocene of Koobi fora, Kenya. Primates 36, 69–89. Gilbert, W.H., White, T.D. & Asfaw, B. (2003). Homo erectus, Homo ergaster, Homo ‘cepranensis’, and the Daka cranium. Journal of Human Evolution 45, 255–259. Grine, F.C., Jungers, W.L. & Schultz, J. (1996). Phenetic affinities among early Homo crania from East and South Africa. Journal of Human Evolution 30, 189–225. Groves, C.P. (1989). A Theory of Human and Primate Evolution. Oxford: Clarendon Press. Groves, C.P. & Mazak, V. (1975). An approach of taxonomy of the hominidae: gracile Villafranchian hominids in Africa. Casopis pro Mineralogii geologii 20, 225–247. Hay, R.L. (1976). The setting of the gorge and history of the geological investigations. In (R.L. Hay, Ed.) The Geology of Olduvai Gorge, pp. 25–28. Berkeley: University of California Press. Howell, F.C. (1978). Hominidae. In (V.J Maglio & H.B.S. Cooke, Eds) Evolution of African Mammals, pp. 190–194. Washington: Harvard University Press. Hughes, A.R. & Tobias, P.V. (1977). A fossil skull probably to the genus Homo from Sterkfontein, Transvaal. Nature 265, 310–312. Johanson, D.C., Masao, F.T., Eck, G.G., White, T.D., Walter, R.C., Kimbel, W.H., Asfaw, B., Manega, P., Ndessokia, P. & Suwa, G. (1987). New partial skeleton of Homo habilis from Olduvai Gorge. Nature 327, 205–209. Jungers, W.L., Falsettti, A.A. & Wall, C.E. (1995). Shape, relative size, and size-adjustements morphometrics. Yearbook of Physical Anthropology 38, 137–161.

224

Characterising early Homo Kimbel, W.H. & Rak, Y. (1993). The importance of species taxa in Paleoanthropology and an argument for the phylogenetic concept of the species category. In (W.H. Kimbel & B. Martin, Eds) Species, Species Concepts, and Primate Evolution, pp. 461–484. New York: Plenum Press. Kimbel, W.H., White, T.D. & Johanson D.C. (1984). Cranial morphology of Australopithecus afarensis: a comparative study based on a composite reconstruction of the adult skull. American Journal of Physical Anthropology 64, 337–388. Kimbel,W.H., Walter, R.C., Johanson, D.C., Reed, K.E., Aronson, J.L., Assefa, Z., Marean, C.W., Eck, G.G., Bobe, R., Hovers, E., Rak, Y., Vondra, C., Yemane, T., York, D., Chen, Y., Evensen, N.M. & Smith, P.E. (1996). Late Pliocene Homo and Oldowan tools from the Hadar formation (Kada Hadar member), Ethiopia. Journal of Human Evolution 31, 549–561. Kluge, A.C. & Farris, J.S. (1969). Quantitative phyletics and the evolution of the anurans. Systematic Zoology 18, 1–32. Kramer, A., Donnelly, S.M., Kidder, S.D., Ousley, S.D. & Olah, S.M. (1995). Craniometric variation in large-bodied hominoids: testing the single-hypothesis for Homo habilis. Journal of Human Evolution 29, 443–462. Kuman, K. & Clarke, R.J. (2000). Stratigraphy, artefact industries and hominid association for Sterkfontein, Member 5. Journal of Human Evolution 38, 827–847. Leakey, L.S.B., Tobias, P.V. & Napier, J.R. (1964). A new species of the genus Homo from Olduvai Gorge. Nature 202, 7–9. Leakey, M.G., Spoor, F., Brown, F.H., Gothogo, P.N., Kiairie, Ch., Leakey, L.N. & McDougall, I. (2001). New hominin genus from eastern Africa shows diverse middle Pliocene lineages. Nature 410, 433–439. Lieberman, D.E. (1999). Homology and hominid phylogeny: problems and potential solutions. Evolutionary Anthropology 7, 147–151. Lieberman, D.E., Pilbeam, D.R. & Wood, B.A. (1988). A probabilistic approach to the problem of sexual dimorphism in Homo habilis: a comparison of KNM-ER 1470 and KNM-ER 1813. Journal of Human Evolution 17, 503–511. Lieberman, D.E., Wood, B.A. & Pilbeam, D.R. (1996). Homoplasy and early Homo: an analysis of the evolutionary relationships of H. habilis sensu stricto and H. rudolfensis. Journal of Human Evolution 30, 97–120. Lockwood, C.A. (1999). Sexual dimorphism in the face of Australopithecus africanus. American Journal of Physical Anthropology 108, 97–127. Lockwood, C.A. & Fleagle J.G. (1999). The recognition and evaluation of homoplasy in primate and human evolution, Yearbook of Physical Anthropology 42, 189–232. Mosimann, J.E. (1970). Size allometry: size and shape variables with characterizations of the lognormal and generalized Gamma distributions. Journal of the American Statistical Association 65, 930–945. Prat, S. (1997). Problème taxinomique des premiers représentants du genre Homo. Etudes

225

From Tools to Symbols crâniennes des individus d’Olduvai et de Koobi Fora. Bulletins et Mémoires de la Société d’Anthropologie de Paris 9, 251–266. Prat, S. (2000a). Origine et taxinomie des premiers représentants du genre Homo. Université Bordeaux I: Thèse de doctorat. Prat, S. (2000b). Intraspecific variability within the genus Gorilla. A pattern for the taxonomic study of the fossil hominids. Folia Primatologica 71, p. 265. Prat, S. (2000c). Who could be Stw 53? 11th Biennal Conference of the Palaeontological Society of Southern Africa, 4–8 September 2000, Pretoria, p. 28. Prat, S. (2000d). Problems concerning the taxonomic allocation of the South African hominid: Stw 53. 12th Congress of the European Anthropological Association, 8–11 September 2000, Cambridge, p. 55. Prat, S. (2001). Comparison between KNM-ER 1805 and other specimens of early Homo: a reevaluation of its taxonomic allocation. Am. J. Phys. Anthropol. suppl. 32, 121. Prat, S. (2002). Anatomical study of the skull of the Kenyan specimen KNM-ER 1805: a reevaluation of its taxonomic allocation? Comptes Rendus Palevol 1, 27–33. Prat, S. & Thackeray, J.F. (2001). Position des lignes temporales sur le cranium de ‘Mrs’ Ples (A. africanus): une attribution sexuelle est-elle possible? Comptes Rendus de l’Académie des Sciences de Paris, Série II a. 332, 403–409. Rightmire, G.P. (1993). Variation among early Homo crania from Olduvai Gorge and the Koobi Fora region. American Journal of Physical Anthropology 90, 1–33. Rightmire G.P. (1998). Evidence from cranial morphology for similarity of Asian and African representatives of Homo erectus. American Journal of Physical Anthropology 106, 61–85. Schwartz, H.P., Grün, R. & Tobias, P.V. (1994). ESR dating studies of the Australopithecine site of Sterkfontein, South Africa. Journal of Human Evolution 26, 175–181. Skelton, R.R. & McHenry, H.M. (1992). Evolutionary relationships among early hominids. Journal of Human Evolution 23, 309–349. Skelton, R.R. & McHenry, H.M. (1998). Trait list bias and a reappraisal of early hominid phylogeny. Journal of Human Evolution 34, 109–113. Skelton, R.R., McHenry, H.M. & Drawhorn G.M. (1986). Phylogenetic analysis of early hominids, Current Anthropology 27, 50–66. Sokal, R.R. & Braumann C.A. (1980). Significance tests for coefficients of variation and variability profiles. Systematic Zoology 29, 50–66. Spoor, F., Wood, B.A. & Zonneveld F. (1994). Implications of early hominid labyrinthine morphology for evolution of human bipedal locomotion. Nature 369, 645–648. Strait, D.S. (2001). Integration, phylogeny, and the hominid cranial base. American Journal of Physical Anthropology 114, 273–297. Strait, D.S. & Grine, F.E. (1998). Trait list bias? A reply to Skelton and McHenry. Journal of Human Evolution 34, 115–118.

226

Characterising early Homo Strait, D.S., Grine, F.E. & Moniz, M.A. (1997). A reappraisal of early hominid phylogeny. Journal of Human Evolution 32, 17–82. Stringer, C.B. (1986). The credibility of Homo habilis. In (B.A. Wood., L. Martin & P. Andrews, Eds) Major Topics in Primate and Human Evolution, pp. 266–294. New York: Cambridge University Press. Stringer, C.B. (1987). A numerical cladistic analysis for the genus Homo. Journal of Human Evolution 16, 135–146. Swofford, D.L. (1993). PAUP Phylogenetic Analysis Using Parsimony, version 3.1.s, Computer program distributed by the Illinois Natural History Survey, Champaign, Illinois, USA. Thackeray, J.F., Loth, S.R., Laing, M., Swanepoel, E., Dayal, M.R. & Lubbe, K. (2000). Comparison of Sts 5 (‘Mrs Ples’) and Stw 53 (‘early Homo’) from Sterkfontein, South Africa. South African Journal of Science 96, 21. Thiele, K. (1993). The holy of the perfect character: the cladistic treatment of morphometric data. Cladistics 9, 275–304. Tobias, P.V. (1978a). Position et rôle des Australopithécinés dans la phylogénèse humaine, avec étude particulière d’Homo habilis et des théories controversées avancées à propos des premiers hominidés fossiles de Hadar et de Laetoli. In (Fondation Singer Polignac, Ed.) Les origines humaines et les époques de l’intelligence, pp. 38–74. Paris: Masson. Tobias, P.V. (1978b). The earliest Transvaal members of the genus Homo with another look at some problems of hominid taxonomy and systematics. Zeitschrift fur Morphologie und Anthropologie 69, 225–265. Tobias, P.V. (1980). The natural history of the helicoidal occlusal plane and its evolution in early Homo. American Journal of Physical Anthropology 53, 173–187. Tobias, P.V. (1991). Olduvai Gorge, Vols IVa and IVb: Homo habilis, skulls, endocasts and teeth. Cambridge: Cambridge University Press. Tobias, P.V. (2003). Encore Olduvai. Science, 299, 1193–1194. Vrba, E.S. (1985). Ecological and adaptative changes associated with early hominid evolution. In (E. Delson, Ed.) Ancestors: The Hard Evidence, pp. 63–71. New York: Alan R. Riss. Walter, R.C. (1994). Age of Lucy and the first family single cristal 40Ar/ 39Ar dating of the Denen Dora and lower Kada Hadar members of the Hadar Formation, Ethiopia. Geology 22, 6–10. White, T.D., Johanson, D.C. & Kimbel W.H. (1981). Australopithecus africanus: its phyletic position revisited. South African Journal of Science 77, 445–470. Wood, B.A. (1991). Koobi Fora: Research Project Vol. 4: Hominid Cranial Remains. Oxford: Clarendon Press. Wood, B.A. (1992). Origin and evolution of the genus Homo. Nature 355, 783–790. Wood, B.A. (1996). The origin and evolution of the genus Homo. In (W.E. Meikle, F.C. Howell & N.G. Jablonshy, Eds.) Issues in Human Evolution, Memoir 21, pp.105–114. San Francisco: California Academy of Sciences.

227

From Tools to Symbols Wood, B.A.& Collard, M. (1999a). The human genus. Science 284, 65–71. Wood, B.A. & Collard M. (1999b). The changing face of the genus Homo. Evolutionary Anthropology 8, 195–207. Zeitoun, V. (2000). Révision de l’espèce Homo erectus (Dubois, 1893). Bulletins et Mémoires de la Société d’Anthropologie de Paris 12, 1–200.

228

Early Homo, ‘robust’ australopithecines and stone tools at Kromdraai, South Africa Francis Thackeray and Jose Braga Transvaal Museum, P.O. Box 413, Pretoria 0001, South Africa Laboratoire d’Anthropologie, PACEA/UMR 5809 CNRS, Université Bordeaux 1, Avenue des Facultés, 33405 Talence Cedex, France

Abstract Kromdraai A and B are two Plio-Pleistocene sites within thirty metres of each other and within two kilometres from Sterkfontein in the Cradle of Humankind World Heritage Site, South Africa. Hominid remains representing nine individuals have been recovered from Kromdraai B, and until recently all of these were attributed to Paranthropus robustus. No hominid remains have been recovered as yet from Kromdraai A (KA), but stone tools have been recovered from both Kromdraai A and Kromdraai B, representing Developed Oldowan and early Acheulean industries. Recent work suggests that early Homo is represented at Kromdraai B. The possibility that robust australopithecines used artefacts is not excluded. A hypothesis is presented to suggest that hominids at KA scavenged from carcasses of animals killed by large carnivores such as Dinofelis.

Résumé Kromdraai A et B sont deux sites d’âge Plio-Pléistocène, distants de trente mètres, et situés à environ deux kilomètres de Sterkfontein, en Afrique du Sud, dans une zone inscrite sur la liste du Patrimoine Mondial de l’Unesco: le Cradle of Humankind World Heritage Site. Des restes d’hominidés fossiles représentant neuf individus ont été découverts à Kromdraai B (KB) et étaient, encore récemment, attribués à Paranthropus robustus. Aucun reste d’hominidé n’a encore été découvert à Kromdraai A (KA). Des outils en pierre proviennent des deux sites et correspondent à de l’Oldowayen évolué et de l’Acheuléen ancien. Une étude récente suggère la présence d’Homo habilis à KB. Nous proposons que les hominidés de KA pratiquaient le charognage sur des carcasses d’animaux tués par des grands carnivores tels que Dinofelis et n’excluons pas la possibilité que Paranthropus robustus soit l’artisan des outils découverts dans ce site.

229

From Tools to Symbols

Introduction In April 1965, Professor Phillip Tobias delivered his Presidential Address to the South African Archaeological Society of South Africa, on the subject ‘Australopithecus, Homo habilis, tool-using and tool-making’. He described newly discovered hominid fossils in the context of morphology and phylogeny, and stated a case ‘for associating the new hominine species, Homo habilis, with the Oldowan culture’. On the basis of evidence from Olduvai Bed I, Olduvai Bed II, Peninj and Garusi in East Africa, as well as from Taung, Sterkfontein Member 4, Sterkfontein Member 5, Swartkrans and Kromdraai in South Africa, he noted inter alia that stone tools had been found in cases where both early Homo and Australopithecus were represented, but artefacts were present at every site which had yielded remains of early Homo. He concluded by presenting a hypothesis that ‘Australopithecus was not the maker of the stone implements, but that more advanced hominids almost certainly were’ (Tobias, 1965). When it was formulated, this hypothesis was consistent with the apparent lack of cases in which stone tools were found in association with an australopithecine only. Before the recognition of the ‘advanced hominid’ which Leakey et al. (1964) described as Homo habilis, Leakey (1959) had considered Zinjanthropus as the Oldowan tool maker, just one year after Brain (1958) had reported stone tools from the site of Kromdraai B. This solution cavity, situated within thirty metres of Kromdraai A and less than two kilometres east of Sterkfontein, had yielded the type specimen of Paranthropus (Australopithecus) robustus discovered in 1938 (Broom & Schepers, 1946), similar in many respects to the type specimen of Zinjanthropus (OH 5) described by Tobias (1967) as Australopithecus boisei. The first stone artefacts from Kromdraai mentioned by Brain (1958) were not known to be directly associated with hominids, and it is probably for this reason that Tobias (1965) did not consider Kromdraai as a possible site where stone tools might have been linked to the robust australopithecine only, in the apparent absence of early Homo. Referring to Kromdraai B, Tobias (1965) stated: ‘Australopithecus robustus present as far as is known, no stone tools, no trace of a more advanced hominid.’ Since 1938, nine hominid individuals have been discovered at Kromdraai B, but none from Kromdraai A (Thackeray et al., 2001). Within the past ten years, stone artefacts have been reported from both Kromdraai A and B (Kuman, Field & Thackeray, 1997). In a review of all hominids found from Kromdraai B, only P. robustus was recognised (Thackeray et al., 2001). Likewise, Kaszycka (2002) identified only P. robustus from the Kromdraai B deposits. To some this has appeared curious, given the presence of stone tools. The possibility that P. robustus might at least potentially have been a tool-maker had

230

Early Homo, ‘robust’ australopithecines and stone tools at Kromdraai

Figure 1

Polyhedral artefact from Kromdraai B (KB 5501, drawn by W. Voorvelt).

been considered by Susman (1988, 1993), based on his study of hominid handbones from Swartkrans and Olduvai. In particular, Susman (1993) examined the pollical distal phalanx of a Swartkrans hominid (SKX 5016) from Member 1, and noted that morphologically it corresponded closely to the thumb associated with OH 7, the type specimen of H. habilis from Olduvai. There certainly are differences in size, and it could be considered that SKX 5016 falls within the range of variability expected for Paranthropus robustus rather than that of early Homo. However, considering the morphological similarity between these two specimens, it would not seem impossible that SKX 5016 could potentially represent a large specimen of the species represented by OH 7, bearing in mind that Susman (1993) stated that the closest morphological counterpart of SKX 5016 is the pollical distal phalanx of the type specimen of H. habilis. Here we look again at questions concerning hominid identification and stone tool technology in the light of work that has been undertaken at Kromdraai within the past ten years. We ask three principal questions: 1. What are the dates for Kromdraai A and Kromdraai B, using not only associated mammalian fauna but also palaeomagnetism? 2. Is there evidence of the presence of early Homo as well as an australopithecine at Kromdraai B, and how does this contrast with the situation at Kromdraai A? 3. Is it possible to identify the tool makers at Kromdraai with certainty?

The dating of Kromdraai deposits Recent palaeomagnetic data obtained from oriented cores of flowstone and calcified deposits from Kromdraai B (Thackeray et al., 2002) have been assessed in the context of mammalian fauna which was used to place this site within the period 1,5–2,0 Mya (Vrba, 1985; Delson 1988). A palaeomagnetic reversal within the Kromdraai B

231

From Tools to Symbols

sequence has been identified with the onset of the Olduvai Event, c. 1,95 Mya. This is regarded as a minimum date for the type specimen of P. robustus, TM 1517 (Thackeray et al., 2002). Palaeomagnetic data are available for Kromdraai A, and together with mammalian fauna, suggest that this assemblage postdates the Olduvai Event (Thackeray et al., in preparation). Seriation of faunal assemblages indicates that both Kromdraai A and B postdate Sterkfontein Member 4 (McKee et al., 1995). McKee et al. (1995, Table 3B, column EXTA) calculated a faunal resemblance index associated with the degree to which fossil assemblages resemble extant fauna from southern Africa. This index (here designated FRI-EXTA), in combination with bestestimate dates for certain assemblages from Plio-Pleistocene southern African sites, is potentially useful for calibrating the results of seriation of faunal assemblages. For example, if Makapansgat Member 3 is considered to date to c. 3 Mya, Swartkrans Member 1 to 2 Mya, Kromdraai B to c. 1,9 Mya, Klasies River Mouth and Border Cave to c. 100 000 years ago, we obtain the following regression equation for the relationship between AGE and the FRI-EXTA index: AGE = –0,038 FRI-EXTA +3,784 Mya (r = 0,9) (standard error of the AGE estimate: 0,184 Mya) where AGE is the age-estimate in millions of years ago (Mya). Application of this equation to the FRI-EXTA value for the faunal assemblage from Sterkfontein Member 5 gives an age estimate of 1,69 Mya (associated with 95 per cent confidence limits spanning from 1,51 to 1,87 Mya), similar in age to the faunal assemblage from Swartkrans Member 2 here dated at 1,65 Mya (associated with a 95 per cent confidence range between 1,47 and 1,84 Mya), and slightly older than the assemblage from Swartkrans Member 3, here dated at 1,42 (associated with 95 per cent confidence limits ranging between 1,24 and 1,61).

Hominids at Kromdraai The taxonomy of Kromdraai specimens has been the subject of much debate, partly because most of the sample consists of isolated teeth. At least nine hominids are represented at Kromdraai B. Since their discovery, additional specimens from both South Africa (Drimolen, Swartkrans) and East Africa (Koobi Fora and Olduvai) have been discovered, and several of these have been fully described. Recently, Braga and Thackeray (2003), after considering the diagnostic features that have been used in the past for lower dentition, suggested that KB 5223 (including M1 and dm2 antimeres, all the permanent lower incisors, the left dc, dm1 and lower permanent canine) from in situ Member 3 breccia of the Kromdraai B East Formation, represents early Homo, distinct from P. robustus. KB 5223 is morphologically and metrically indistinguishable

232

Early Homo, ‘robust’ australopithecines and stone tools at Kromdraai

Figure 2

Occlusal view of KB 5223 M1 (mesiosdistal diameter = 13,8 mm). Photograph by J. Braga.

from early Homo and does not group with P. robustus. Morphological assessments of KB 5223 were made with reference to Robinson’s (1956) odontological terminology, and four quantitative traits were used: MD and BL crown diameters (in mm), the calculated MD/BL index and occlusal area (MD x BL). The analysis was done by a two-step resampling procedure, with replacement, avoiding assumptions about the underlying distributions of the qualitative traits. Before examining KB 5223 values, Braga and Thackeray (2003) tested hypotheses of difference between, on the one hand, early Homo and P. robustus samples, and on the other, P. robustus and P. boisei samples. Significant differences were found between certain variables of early Homo and P. robustus (not in permanent incisors). KB 5223 clearly differs from P. robustus in the following respects: all M1 values except the absolute MD diameter, and all dm2 values except the MD/BL index. Notably, the KB 5223 M1 MD/BL index, one of the most diagnostic features that have been used in the past to identify early Homo (Grine, 1989; Tobias, 1991a, b), fall well outside the P. robustus distribution and within the distribution of early Homo. Together this constitutes metrical evidence for KB 5223 being attributed to early Homo. The absence of C6, combined with the presence of C7, sensu lato, in the M1 of the 2,4–2 Mya East African hominids, has been considered indicative of a ‘non-robust’ species (Wood & Abbot, 1983; Suwa et al., 1996). This morphological pattern is found in most specimens attributed to early Homo from both South and East Africa. Suwa et al. (1996) included the absence of C6 combined with the presence of C7 in a total morphological pattern, leading them to consider that ‘by circa 2,4 myr, the postcanine dentition of the East African nonrobust lineage phenetically approximates the early Homo condition but lacks any specific affinities with Australopithecus africanus.’ (p. 275). Besides this morphological pattern, other important morphological features distinguishing ‘robust’ species from early Homo are evident in KB 5223 deciduous

233

From Tools to Symbols

molars: the absence of the hypoconulid and the mesial accessory cuspule on the fourcusped dm1, the slight postmetaconulid combined with the lack of C6, and distal marginal ridge on the dm2. These features indicate close affinities between KB 5223 and some specimens from Koobi Fora attributed to early Homo (e.g. ER 1507). Because these important features are also lacking in A. africanus, they can be considered as derived, even when KB 5223 dm2 displays a ‘doubled’ anterior fovea, a ‘primitive’ trait also seen in early Homo representatives (e.g. DNH 35). The absolute enamel thickness values of KB 5223 M1 (Grine & Martin, 1988) fall within the 95 per cent confidence limits obtained for early Homo (Beynon & Wood, 1986). Two absolute values of KB 5223 (LT and OT) fall well outside the 95 per cent confidence limits obtained for P. boisei (Beynon & Wood, 1986). Four indices of average or relative enamel thickness have also been used. If we compute z-scores to compare absolute or relative enamel thickness of KB 5223 (corrected values) (Grine & Martin, 1988) with either early Homo (values in Beynon & Wood, 1986) or extant humans (values in Grine & Martin, 1988), we find no significant differences to reject the null hypothesis of absence of differences. In other words, from the data published so far, we see no reason to consider absolute or relative enamel thickness in KB 5223 M1 different from early Homo or extant humans. The enamel accretionary pattern, as expressed by the length and orientation of the Brown striae of Retzius, is also of interest. KB 5223 M1 displays Brown striae of Retzius that tend to become more highly angled relative to the enamel–dentine junction in the cervical region, than do the molars of P. boisei and Swartkrans P. robustus specimens (Grine & Martin, 1988). In terms of lower incisor spacing and distribution of perikymata, KB 5223 does not appear to be different from early Homo specimens (Dean & Reid, 2001).

Conclusions The Olduvai Event has been identified at Kromdraai B (type locality of Paranthropus robustus) as well as at Olduvai Bed I (type locality of Homo habilis and Paranthropus/ Zinjanthropus/Australopithecus boisei). If we apply criteria that Tobias (1965) used to try to identify tool makers at early Pleistocene sites in South Africa, we could say that there is a case for the use of stone tools by early Homo at Kromdraai B, but we cannot exclude the possibility that P. robustus made and used artefacts. In the case of Kromdraai B, the relatively low frequency of stone tools may be attributed in part to the fact that it was a death trap. At Kromdraai A, we have stone tools, including tools attributed to the Developed Oldowan. However, as far as hominids are concerned, we do not have any remains of either early Homo or P. robustus. One scenario which might apply in this case is that Kromdraai A was used to a large extent by large carnivores (Vrba & Panagos, 1982). The sabre-toothed

234

Early Homo, ‘robust’ australopithecines and stone tools at Kromdraai

cat Dinofelis is represented at Kromdraai A together with a relative abundance of alcelaphines. On the basis of crown heights of alcelaphine molars, it has been suggested that Dinofelis was preying preferentially on vulnerable young wildebeest, just as lions in the Serengeti do today (Thackeray & von Leuvan-Smith, 2001). Although hominids from Kromdraai A are not as yet known, the artefacts including polyhedral cores indicate a hominid presence. One hypothesis that can be offered is that hominids at Kromdraai A were occasionally present, without using it habitually as a living site, instead using it as a cave where they occasionally scavenged from carcasses of wildebeest and other animals. Cut marks on bovid bone flakes appear to be absent. At present there is no reason to discount the possibility that hominids at Kromdraai A used polyhedral cores to break open long bones of animals such as wildebeest for marrow. In the absence of any hominid remains from Kromdraai A, it is not possible to determine whether the artefacts were made and used by early Homo and/or an australopithecine.

Acknowledgements This work has been supported by the National Research Foundation, South Africa (GUN 2065329), the French Ministry of Foreign Affairs, and the French Embassy in South Africa (Co-operation and Cultural Service).

235

From Tools to Symbols

References Beynon, A. & Wood, B. (1986). Variation in enamel thickness and structure in East African hominids, American Journal of Physical Anthropology 70, 177–193. Braga, J. & Thackeray, J.F. (2003). Early Homo at Kromdraai B: probabilistic and morphological analysis of the lower dentition. Comptes Rendus Palevol 2, 269–279. Brain, C.K. (1958). The Transvaal ape-man-bearing cave deposits. Transvaal Museum Memoir 11. Pretoria: Transvaal Museum. Broom, R. & Schepers, G.W.H. (1946). The South African ape-men: the Australopithicinae. Transvaal Museum Memoir 2. Pretoria: Transvaal Museum. Dean, M.C. & Reid, D. (2001). Perikymata spacing and distribution on hominid anterior teeth, American Journal of Physical Anthropology 116, 209–215. Delson, E. (1988). Chronology of South African australopith site units. In (F.E. Grine, Ed.) Evolutionary History of the ‘Robust’ Australopithecines, pp. 317–324. New York: A. de Gruyter. Grine, F. (1989). New hominid fossils from the Swartkrans Formation (1979–1986 excavations): craniodental specimens. American Journal of Physical Anthropology 79, 409–449. Grine, F. & Martin, L. (1988). Enamel thickness and development in Australopithecus and Paranthropus. In (F.E. Grine, Ed.) Evolutionary History of the ‘Robust’ Australopithecines, pp. 3–42. New York: Aldine de Gruyter. Leakey, L.S.B., Tobias, P.V. & Napier, J.R. (1964). A new species of the genus Homo from Olduvai Gorge. Nature 202, 7–9. Robinson, J. (1956). The dentition of the Australopithecinae. Transvaal Museum Memoir 9. Pretoria: Transvaal Museum. Susman, R.L. (1988). Hand of Paranthropus robustus from Member 1, Swartkrans: fossil evidence for tool behaviour. Science 240: 781–784. Susman, R.L. (1993). Hominid postcranial remains from Swartkrans. In Swartkrans, A Cave’s Chronicle of Early Man. Transvaal Museum Monograph 8, pp. 117–136. Pretoria: Transvaal Museum. Suwa, G., White, T. & Howell, F.C. (1996). Mandibular postcanine dentition from the Shungura formation, Ethiopia: Crown morphology, taxonomic allocations, and Plio-Pleistocene hominid evolution. American Journal of Physical Anthropology 101, 247–282. Thackeray, J.F. & von Leuvan-Smith, T. (2001). Implications of crown height measurements of alcelaphine molars from Kromdraai A, South Africa. Annals of the Transvaal Museum 38, 9–12. Thackeray, J.F., Kirschvink, J.L. & Raub, T.D. (2002). Palaeomagnetic analyses of calcified deposits from the Plio-Pleistocene hominid site of Kromdraai, South Africa. South African Journal of Science 98: 537–540.

236

Early Homo, ‘robust’ australopithecines and stone tools at Kromdraai Tobias, P.V. (1965). Australopithecus, Homo habilis, tool-using and tool-making. South African Archaeological Bulletin. Tobias, P.V. (1967). Olduvai Gorge, Vol 2: The cranium and maxillary dentition of Australopithecus (Zinjanthropus) boisei. Cambridge: Cambridge University Press. Tobias, P.V. (1991a). Olduvai Gorge, Vol. 4, Parts I-IV. The skulls, endocasts and teeth of Homo habilis. Cambridge: Cambridge University Press. Tobias, P.V. (1991b). Olduvai Gorge, Vol. 4, Parts V-IX. The skulls, endocasts and teeth of Homo habilis. Cambridge: Cambridge University Press. Vrba, E.S. 1985. Early hominids in southern Africa: updated observations on chronological and ecological background. In (P.V. Tobias, Ed.) Hominid Evolution: Past, Present and Future, pp. 195–200. New York: Alan R. Liss. Vrba, E.S. & Panagos, D.C. 1982. New perspectives on taphonomy, palaecology and chronology of the Kromdraai ape-man. In Palaeoecology of Africa and the surrounding islands 15, 13–26. Wood, B. and Abbott, S. (1983). Analysis of the dental morphology of Plio-Pleistocene hominids. I. Mandibular molars: crown area measurements and morphological traits. Journal of Anatomy 136, 197–219.

237

The origin of bone tool technology and the identification of early hominid cultural traditions Lucinda Backwell and Francesco d’Errico Institute for Human Evolution, School of Geosciences, University of the Witwatersrand, Private Bag 3, WITS 2050, Johannesburg, South Africa UMR 5808 du CNRS, Institut de Préhistoire et de Géologie du Quaternaire, Avenue des Facultés, 33405 Talence, France PACEA/UMR 5199 du CNRS, Institut de Préhistoire et de Geologie du Quaternaire, UFR de Geologie, Bat. B18, Avenue des Facultés, 33405 Talence, France Department of Anthropology, The George Washington University, Washington DC

Abstract A number of natural processes occurring during the life of an animal or after its death can produce pseudotools, mimics of human-made objects. A number of purported bone tools from Lower and Middle Palaeolithic sites have been published without any validating microscopic analysis of the bone surfaces showing possible traces of manufacture and use. This paper discusses the evolutionary significance of bone tool technology and summarises results of research on the use of bone tools by early hominids between one and two million years ago (Mya). It attempts to establish formal criteria for the identification of minimally modified bone tools by characterising the modifications produced by known human and non-human agents, and applying these criteria to the purported bone tool collections from Swartkrans, Sterkfontein and Olduvai Gorge. A number of experiments involving a variety of tasks were conducted in order to increase the range of diagnostic features available. New analytical techniques have been developed for the quantification of microscopic use-wear, and a wide range of taphonomic and morphometric variables have been used to isolate idiosyncratic populations of specimens for which a robust argument can be made for their identification as tools. South and East African early hominid

238

The origin of bone tool technology sites dated to between 1,8 Mya and 1 Mya have yielded what appear to be very different types of bone tools. The former are characterised by long bone shaft fragments and horn-cores of medium to large-sized bovids, collected by hominids after weathering, and possibly used in specialised digging activities. Most fragments were used as such, though a few horn-cores were modified by grinding the tips to points on sandstone or compact abrasive sediment. Those from East Africa mainly consist of freshly broken, or more rarely, complete irregular bones from very large mammals, used as such, or modified by flaking. Irregular bones or epiphyses appear to have been used as hammers, while the others were apparently involved in a variety of light- and heavy-duty activities. Based on the bone tool manufacturing techniques recorded in the two regions, there appear to be no significant differences between the cognitive abilities of the hominid users. Evidence of intentional flaking by knapping seen on the Olduvai bone tools, and traces of grinding on those from South Africa, suggests that the makers of the tools had a clear understanding of the properties of bone, could anticipate the end product, and conceived shaping techniques specific to this raw material in order to achieve optimal efficiency in the tasks for which they were used.

Résumé Un certain nombre de phénomènes naturels se produisant au cours de la vie d’un animal ou après sa mort peuvent produire des pseudo outils en os, imitant les objets façonnés par l’homme. Plusieurs fragments d’os provenant de sites du Paléolithique inférieur et moyen ont été interprétés comme des outils en os sans que cette interprétation soit validée par une analyse microscopique documentant des traces de modification intentionnelle et d’utilisation. Ce chapitre traite des implications d’une technologie de l’os pour l’évolution cognitive des hominidés et récapitule les résultats de nos recherches sur les outils en os utilisés par les hominidés ayant vécu en Afrique australe entre un et deux millions d’années. Nous tentons également d’établir des critères pour l’identification d’outils en os faiblement modifiés en caractérisant les modifications produites par des agents humains et non humains connus. Ces critères sont appliqués à l’analyse du matériel de Swartkrans, Sterkfontein et Olduvai Gorge. Une approche expérimentale est adoptée dans certains cas pour augmenter le nombre et vérifier la pertinence des critères diagnostiques. Des nouvelles techniques d’analyse ont été élaborées pour quantifier les traces d’utilisation et une gamme de variables taphonomiques et morphométriques ont été utilisées pour isoler des populations d’objets ayant pu être utilisés comme outils. Nos résultats indiquent que les sites de premiers hominidés du sud et de l’est de l’Afrique datés entre 1,8 et 1 millions d’années livrent des outils en os différents. Les premiers consistent en des éclats d’os longs et des chevilles osseuses provenant de bovidés de taille moyenne à large, que des hominidés ont ramassés à même le sol, déjà altérés par les agents atmosphériques, et utilisés comme des bâtons à fouir. Certaines chevilles osseuses ont été appointées par abrasion sur du grès ou du sol compacté. Les outils d’Afrique de l’est consistent en des éclats issus de la fracturation d’os frais ou des os complets de grands mammifères qui ont été utilisés tels quels, ou modifiés par percussion. Certains os entiers ou épiphyses semblent avoir été utilisés comme des percuteurs, les autres ont servi dans des activités variées (découpe, raclage...). Les techniques de façonnage et le mode d’utilisation des outils provenant des deux régions ne peuvent pas être utilisées pour proposer que ces hominidés avaient des capacités cognitives différentes. L’analyse des stigmates du façonnage par percussion sur les outils en os de Olduvai, et de celui par abrasion sur

239

From Tools to Symbols les outils en os d’Afrique du Sud, suggèrent que les auteurs de ces outils avaient une claire compréhension des propriétés de l’os, qu’ils pouvaient prévoir le résultat de leur actions sur la matière et qu’ils ont, dans les deux cas, été capables de développer des traditions techniques bien adaptées à la matière première disponible dans le but d’obtenir une efficacité optimale dans les tâches pour lesquelles ces outils étaient utilisés.

Introduction The earliest use of bone tools is a topic of ongoing debate among researchers interested in early human culture and the emergence of modern cognition. This debate concerns the implications of bone tools for assessing hominid cognitive abilities, and the criteria one must use to firmly identify potentially used or minimally modified bone tools, such as those that one may expect to find associated with, or that have been reported from early hominid sites. Unlike stone tools, the morphological predetermination of which is limited by the constraints imposed by the fracture of isotropic materials, the final shape and size of a bone tool produced with techniques such as grinding, scraping and grooving may be determined with a high degree of accuracy. It is probably for this reason that bone tool industries have been considered as particularly appropriate in characterising technical systems, identifying regional patterns, disentangling style from function, tracking changes in time, and inferring from these observations the degree of complexity of a human culture. Klein (1999) has made this point, defining as ‘formal’ bone tools that were ‘cut, carved or polished to form points, awls, borers, and so forth’. McBrearty and Brooks (2000) list the use of bone and antler and their shaping into task-specific tools among the features they consider diagnostic of behaviourally modern humans. The absence in ancient prehistory of labour-intensive techniques specifically conceived to modify bone material is consistent with the traditional view that early hominid technological behaviour was essentially immediate, and involved only a short series of single-stage operations, and thus a lower degree of conceptualisation than did Upper Palaeolithic tools, which often involved several stages of manufacture (Dennell, 1983; Noble & Davidson, 1996). It is also consistent with the view that the development of technology was a gradual process that proceeded in parallel with biological evolution. It comes as no surprise to such authors that bones used as hammers to retouch stone tools, or bone tools shaped by knapping, are reported from Lower and Middle Palaeolithic sites (Radmilli, 1985; Radmilli & Boschian, 1996), as they see these behaviours as the simple transfer of percussion flaking from stone to bone, and proof that early humans were incapable of developing sophisticated techniques specifically conceived for bone. One may wonder, however, how ‘formal’ a formal bone tool must be to tell us

240

The origin of bone tool technology

something about the cognitive abilities of its maker and user. Most of the techniques used to manufacture bone tools do not seem to require a particularly high level of dexterity or cognition, nor do they seem difficult to transmit from one generation to another. Among the tool types listed as reflecting modernity (awls, bores, points), some are in fact in the techno-complexes where they are preserved in abundance, minimally modified, and do not fall within precise morpho-technological standardised categories as one might have expected if tools made with this raw material were the quintessential reflection of modernity. The degree of manufacture of bone tools used by ethnographically documented societies is highly variable, and ranges, as in prehistory, from minimally modified to highly sophisticated artefacts. We also observe in societies of anatomically modern humans, contemporaneous with and postdating the European Upper Palaeolithic, bone technology reaching a high level of sophistication in some, while others make little or no use of bone tools. We see four means by which to move a step forward in addressing this issue. Although we do not have any direct analogy for evaluating ancient bone technologies in terms of cognition, variability in the use of bone material by ethnographically known and recent archaeological societies on the one hand, and the technical traditions and related motions performed by chimpanzees and bonobos on the other, may provide a suitable frame of reference. Chimpanzees in the wild are known to perform a wide range of technical activities, some requiring a high degree of dexterity, such as food-pounding, nut-hammering, pestle-pounding, termite and ant-fishing, fluiddipping, bee-probing, marrow-picking and expel/stirring (McGrew, 1996; Whiten et al., 1999, Joulian, this volume). However, with the exception of recently observed ‘food smearing’ at the Madrid Zoo (Fernandez-Carriba & Loeches, 2000a, FernandezCarriba et al, 2000) they do not seem to perform motions such as scraping or grinding, nor the shaping of objects by reducing them through other wearing techniques. The recognition of such techniques and motions in the archaeological record, whether applied to bone or other raw materials, is an observation that requires explanation, and may be indicative of differences between chimpanzee and hominid cognition. Does the difference lie in the motion itself, in the duration required by the action to achieve the goal, or in the conceptualisation of the desired morphological outcome? Experimentation with captive chimpanzees may answer these questions and establish whether ‘formal’ bone tools should still be considered as a hallmark of modernity. A second approach that can certainly provide useful insight is the reconstruction of the process from inception of the tool to its disposal and incorporation in the archaeological record. This approach, known as the study of the chaîne opératoire, seeks to read material culture in the form of an ordered chain of actions, gestures, and

241

From Tools to Symbols

processes in a production sequence that led to the transformation of a given material to the finished product (Lemonnier, 1986; Schlanger, 1994, this volume). The concept, linked in France to Andre Leroi-Gourhan (1964), is significant in that it allows the archaeologist to infer from the finished artefact the procedures involved, intentionality in the production sequence, and arguably the conceptual template of the maker. Linked to this approach, we see the need to consider each instance of bone use as an independent cultural adaptation to environmental conditions. This view seeks to elaborate systemic models, providing best-fit explanations as to the role of a bone technology within a specific subsistence strategy, and does not assume gradual patterns of evolution in technology. Finally, we believe that it is crucial to systematise the above inferences in time and space. This allows researchers to identify possible patterns of innovation within multistratified/membered deposits, and contemporaneous sites. It also provides a means by which to demonstrate geographic variations suggestive of distinct cultural traditions. The second, no less important, aspect of early bone technology concerns the criteria used to firmly identify true tools. A number of natural processes occurring during the life of an animal or after its death can produce pseudotools, mimics of human-made objects. These include surface features resulting from vascular grooves (Shipman & Rose, 1984; d’Errico & Villa, 1997), teeth use-wear (Gautier, 1986), breakage and wear of deer antler (Olsen, 1989) and elephant tusk tips (Haynes, 1991; Villa & d’Errico, 1998), gnawing or digestion by carnivores, rodents or herbivores (Pei, 1938; Sutcliffe, 1973, 1977; Binford, 1981; Villa & Bartram, 1996; d’Errico & Villa, 1997), fracture for marrow extraction by hominids or carnivores (Bunn, 1981, 1982; Gifford-Gonzalez, 1989), trampling (Haynes, 1988), root etching (Binford, 1981), weathering (Brain, 1967), and the action of different sedimentary environments (Brain, 1981; Lyman, 1994). As suggested by these and other authors (Bonnichsen & Sorg, 1989; Shipman, 1988; Shipman & Rose, 1988), in order to distinguish between pseudo-tools and true tools, it is necessary to adopt an interdisciplinary approach, combining taphonomic analysis of the associated fossil assemblages, microscopic studies of possible traces of manufacture and use, and the experimental replication of the purported tools. It is by applying this approach, for example, that Dart’s (1957) theory for an early hominid ‘Osteodontokeratic’ culture has strongly been challenged and largely refuted (Klein, 1975; Shipman & Phillips, 1976; Brain, 1981; Maguire et al., 1980). What do we know about early bone tools? The early use of bone as a raw material for retouching stone artefacts is evidenced at a number of Middle and Upper Pleistocene sites in Europe (Henri-Martin, 1907; Chase, 1990; Pitts & Roberts, 1997; Malerba & Giacobini, 1998). Acheulean-type bifaces flaked on elephant long bones and tusks are known from three Middle Pleistocene sites in Italy (Cassoli et al., 1982; Radmilli,

242

The origin of bone tool technology

1985; Biddittu & Bruni, 1987; Radmilli & Boschian, 1996), and an increasing number of shaped bones are reported from Middle Stone Age sites in Africa (Beaumont et al., 1978; Singer & Wymer, 1982; Henshilwood & Sealey, 1997; Deacon & Deacon, 1999; Henshilwood et al., 2002). These include a dagger-like object and barbed and unbarbed points dated to between 90 and 60 thousand years ago (Kya) from the Congo (Brooks et al., 1995; Yellen et al., 1995), and tools dated 70 Kya from Blombos Cave, South Africa. Late Neanderthal sites in Europe such as Arcy-sur-Cure and Quincay in France, dated to between 40 and 35 Kya, have yielded clear evidence of complex bone technology, including personal ornaments, and shaped and decorated awls and bone tubes (d’Errico et al., 1998; Zilhão & d’Errico, 1999a, b; d’Errico et al., 2003). Many other putatively used or modified bone, antler, and ivory tools are reported from a large number of Lower (Breuil, 1932, 1938; Breuil & Barral, 1955; Dart, 1957; Bonifay, 1974; Cahen et al., 1979; Biddittu & Segre, 1982; Howell & Freeman, 1983; Mania & Weber, 1986; Aguirre, 1986; Justus, 1989; Dobosi, 1990) and Middle Palaeolithic sites in Africa and Europe (Kitching, 1963; Debenath & Duport, 1971; Freeman, 1978, 1983; Vincent, 1988; Stepanchuk, 1993; Gaudzinsky, 1998, 1999). However, most of these pieces have been published without validating microscopic analyses of the bone surfaces to document possible traces of manufacture and use, and in isolation of their taphonomic contexts. Our aim here is to synthesise the results that we have obtained during the last five years in assessing the evidence for bone tool utilisation at South and East African early hominid sites, explore the significance of this evidence to identify early cultural traditions, and evaluate the cognitive abilities of early hominids.

The South African evidence Background In 1959 Robinson published a single bone tool from Sterkfontein Member 5 West (c. 1,7–1,4 Mya) consisting of a pointed metapodial shaft fragment with evidence of use on the tip. In the course of 24 years of excavation at Swartkrans, Brain (Brain et al., 1988; Brain, 1989; Brain & Shipman, 1993) identified 68 bones, bovid horn cores and one equid mandible from Members 1–3 (c. 1,8–1 Mya) bearing similar modifications. Comparative microscopic analysis of the wear pattern on the smoothed tips of these bones, and on modern shaft fragments used experimentally to dig up tubers and work skins, suggested to Brain and Shipman (1993) that the surface modifications were not natural, and that the activities they tested experimentally were indeed those in which the Swartkrans tools were involved. Although Brain and Shipman’s work was based on microscopic analysis of a number of specimens, their interpretation of these bones as tools used for digging up tubers and working skins was not supported by a

243

From Tools to Symbols

systematic comparison of the purported tool morphology and wear pattern with those produced by natural processes known to mimic anthropic modifications. Brain and Shipman did not consider alternative functional interpretations, nor did they test them experimentally by using appropriate analytical methods. Other potentially relevant data (species, type of bone used, fracturing patterns, degree of weathering, bone flake morphometry, spatial distribution) were not collected or discussed by Brain and Shipman in the context of the site’s taphonomy. We have reappraised the function of the South African bone tools using a multiple approach study based on data provided by microscopic, taphonomic and morphometric analysis of the purported bone tools, faunal material from the remainder of the assemblages, and experimentally and naturally modified bone (Backwell, 2000; Backwell & d’Errico, 1999a, b, 2000, 2001, 2002a, b, c, 2003; d’Errico & Backwell, 2000, 2001, 2003; d’Errico et al., 2001).

Methodology Swartkrans and Sterkfontein Material High-resolution dental impression material (Coltene President microSystem light body surface activated silicone paste for moulds, and Araldite M resin and HY 956 Hardener for casts) were used to replicate the one Sterkfontein (SE) and 68 Swartkrans (SKX) purported bone tools, and optical and scanning electron microscopy was used to identify their surface modifications. Microscopic images of the transparent resin replicas were digitised at 40x magnification on a sample of 18 fossils from Swartkrans. The orientation and dimension of all visible striations was recorded by using MICROWARE image analysis software (Backwell & d’Errico, 2000, 2001). The collection of 23 000 bone fragments from Swartkrans was then taphonomically studied and examined for specimens with a wear pattern similar to that recorded on the purported bone tools from the same site (Fig. 1). Comparative taphonomic analysis was conducted on Swartkrans because all but one of the putative tools come from this site, and because the stratigraphic provenance of both tools and faunal remains is reliable. In the course of research, 16 additional specimens (Fig. 2) from Swartkrans Members 1–3 that had wear comparable to that of the 69 previously described specimens were identified, bringing the total to 85. After investigation of the content and context of the Swartkrans material, the next step involved the examination of 35 reference collections of modern and fossil bones from open air and cave contexts (13 301 specimens) modified by 10 non-human agents (hyaena, dog, leopard, cheetah, porcupine, river gravel, spring water, flood plain, wind, and trampling) without evidence of human involvement. At a macroscopic scale, 24 of the pieces examined appeared similar to the SKX/SE specimens. Resin replicas of these pseudotools were made and examined microscopically. A comparison was then

244

The origin of bone tool technology

Figure 1 Typical wear pattern recorded on the tips of the Swartkrans bone tools consisting of sub-parallel individual striations: (1–2) SKX 1142; (3–4) two aspects of SKX 35196; (5–6) close-up views of the same tool. Notice how striations affect concave areas of the spongy bone (6) indicating that fine loose abrasive particles were responsible for the wear pattern; (7–8) SKX 47045; (9) SKX 38830. Scale = 5 mm in 1, 3, 7, 9 and 1 mm in 2, 5, 6, 8 (d’Errico & Blackwell, 2003).

245

From Tools to Symbols

Figure 2 Sixteen newly identified Swartkrans bone tools: (a) SKX 8741; (b) SKX 30568; (c) SKX 19845; (d) SKX 39364; (e) SKX 36969; (f) SKX 8954; (g) SKX 47046; (h) SKX 34370; (i) SKX 29434; (j) SKX b; (k) SKX 47045; (l) SKX 2787; (m) SKX 39365; (n) SKX 9123; (o) SKX SEM; (p) SKX 5847. Scale = 1 cm.

246

The origin of bone tool technology

made between the wear patterns on the SKX/SE fossils, those on two antelope long bone shaft fragments used by Brain to dig up bulbs of Scilla marginata and Hypoxis costata, and those on bone tools experimentally used by us. This last sample included 11 antelope limb bone shaft fragments and horn cores used to dig for tubers in a wide range of soil types, scrape and pierce animal hides, and excavate termites from termite mounds found in the Sterkfontein Valley today. The worn tips of these bones were each replicated with dental impression material after 5, 15, 30, and 60 minutes of use. Resin replicas of the SKX/SE fossils and experimental tools were made, and then examined under transmitted light. Image analysis was conducted on digitised images of the wear patterns on 18 SKX/SE fossils, 9 of our experimental tools, and both of the experimental tools used by Brain to dig up bulbs. Microscopic analysis of all of the experimental tools was conducted to verify that they would have provided comparable results. Quantification of striation width and orientation comprising the wear pattern on the SKX/SE tools suggested they were not used to extract tubers or work skins. The wear pattern more closely fits that created experimentally when bone is used to excavate in a fine-grained sedimentary environment, such as that found in the presorted sediment constituting termite mounds present in the Sterkfontein area (Fig. 3 and Fig. 4). This led us to propose that the main, if not exclusive, function of the Sterkfontein and Swartkrans bone tools, and of the similar 23 undescribed specimens from Drimolen (c. 2–1,5 Mya) (Keyser, 2000), was that of extracting termites. We also showed that the wear on the bone tools does not represent an extreme in variation of a taphonomic process affecting to a lesser degree the rest of the assemblage. In addition, taphonomic analysis of the breakage patterns and size of the bone tools from this site, compared with the remainder of the faunal remains, indicated that early hominids selected heavily weathered, elongated and robust bone fragments for use as tools. Evidence of grinding After identifying possible evidence of grinding on the tips of six horn cores and a bone shaft fragment from Swartkrans (Fig. 5), we re-examined the 198 bovid horn cores found in Members 1–3 to study the preservation of their tips. (Comparative natural and anthropic traces were examined at microscopic level following the methods described below.) One hundred and ninety-eight horn cores from Swartkrans were compared with a sample of those recovered from the southern African Plio-Pleistocene sites of Makapansgat (Maguire et al., 1980), Sterkfontein (Kuman & Clarke, 2000), and Gondolin (Menter et al., 1999) to check whether modifications similar to those observed at Swartkrans occur on the horn cores from these sites, and to characterise the natural alterations affecting these pieces. We also examined the horn cores and sheaths

247

From Tools to Symbols

Figure 3 Wear pattern on Swartkrans and experimental bone tool tips photographed in transmitted light using transparent resin replicas: (a) bone tool from Swartkrans Member 3 (SKX 38830); (b) tip of a tool used in Brain’s experiment to dig up Scilla marginata bulbs; (c) experimental bone tool used to dig the ground in search of tubers and larvae; (d) experimental bone tool used to dig in a termite mound. Note the similarity in the orientation and the width of the striations in (a) and (d). Scale bar = 2 mm (Backwell & d’Errico, 2001).

248

The origin of bone tool technology

Figure 4 Image analysis of the wear patterns on the Swartkrans fossils and on experimental bone tools: (a) variability (top graph) and mean (bottom graph) in the orientation of the striations on the Swartkrans tools (S), on experimental tools used to dig termite mounds (T), to excavate the ground in search of tubers and larvae (G), and to extract bulbs (B) (Brain’s experimental tools). An unpaired t-test has shown the orientation of the striations on the Swartkrans and termite digging tools to be the most similar, and significantly different from the other experimental tools; (b) striation width as measured at 40x magnification on all the striations visible. A non-parametric statistical test has shown the striation width on all the experimental tools to be significantly different from each other, but with the closest similarity recorded between the Swartkrans and termite-digging tools.

of various African bovid skulls housed at the Bernard Price Institute, University of Witswatersrand. Later Stone Age arrow points, awls and fish gorges shaped by abrasion from Nelson Bay Cave (Deacon & Brett, 1993), Die Kelders (Avery et al., 1997; Klein, 1994), Goergap (Van der Ryst, 1998), Olieboomspoort (under analysis) and Rose Cottage (Wadley, 1997), as well as worked Iron Age bone from the Mapungubwe Complex and Kleinfontein sites, was also studied for comparative purposes. San arrow points and link-shafts shaped through grinding, including an Australian aboriginal piece used for

249

From Tools to Symbols

Figure 5 Swartkrans bone tools bearing possible traces of grinding: (a) SKX 12383; (b) SKX 7068; (c) SKX 28876B; (d) SKX 30215; (e) SKX 39364; (f) SKX 15536; (g) SKX 28437. Lines identify ground facets. Scale = 1 cm.

magical ritual and manufactured using the same technique, were analysed. Experimental material included long bone shaft fragments and horn cores experimentally ground on granite, sandstone, and the surfaces of termite mounds of the genus Trinevitermes found in the Sterkfontein Valley, South Africa. Our results show that grinding is characterised by wide parallel striations orientated oblique to the bone main axis (Fig. 6). These striations are morphologically different from those produced by use-wear in that they have a fusiform (spindle-like) shape. Also, longitudinal striations produced by use are recorded on concave surfaces while

250

The origin of bone tool technology

Figure 6 SEM photos of traces of grinding on an aboriginal Australian bone point (a); Later Stone Age bone tools from Kasteelberg B (b); and the Hunterian Museum collection (c); grinding on a termite mound (d); the Swartkrans horncore SKX 15536 (e); and ulna SKX 39364 (f). Scale in (a) = 1 mm.

fusiform striations are restricted to facets. Comparative analyses confirm the existence of intentional shaping by grinding on the Swartkrans pieces, indicating that southern African early hominids had the cognitive ability to modify the functional area of bone implements with a technique specific to bone material, in order to achieve optimal efficiency in digging activities (d’Errico & Backwell, 2003).

251

From Tools to Symbols

Stability in southern African early hominid bone tool use In order to test whether the bone tools represent a continuous cultural tradition that persisted unchanged for nearly a million years, or reflect a more flexible practice, subjected to pressures determined by adaptation to changing environments and/or cultural evolution, we analysed Swartkrans bone tools in search of patterns of variation between Members. A range of variables was used for the study of the bone tools. These related to the species, animal size, type of bone used, fracturing patterns, degree of weathering, shape and morphometry of the bone flake and of the worn area. Our results showed no significant differences between the bone tools from Swartkrans Members 1–3. In the three assemblages the majority of the specimens derive from the medial portion of long bone shafts from mammal size classes II–III/III–IV. Though restricted to a few specimens, the use of horn cores persists throughout all of the Members. The high proportion of weathered bones selected to be used as tools also remains stable throughout the stratigraphy. Only 5 of the 85 specimens comprising the enlarged collection are complete, i.e. without post-depositional breakage, making it difficult to establish whether significant variation occurs in the size or shape of the tools between the Members. However, analysis of the breadth and thickness of complete tips at 5, 10, 15 and 20 mm from the tip reveals a remarkable dimensional similarity between tools from Members 1 and 3, and a slight preference for more robust blanks among tools from Member 2 (Table 1). The length of the wear ranges for the large majority of the tools from the three Members, between 20 and 40 mm, and the frequency distribution of this variable is virtually the same in the three assemblages. Based on our digging experiments, this suggests comparable motions and a similar time-span for which the tools were used.

Table 1 Comparison between the width, compact bone thickness, length of the bone tools and a representative sample of long bone shaft fragments from Swartkrans Members 1–3.

Width

Thickness

Length

n

Mean (mm)

SD

n

Mean (mm)

SD

n

Mean (mm)

SD

Bone tools

41

19,1

9,6

67

7,8

3,3

75 *

52,6

26,9

Unmodified shaft fragments

614

14,3

7,4

614

4,5

2,3

614

37,7

23,8

* broken bone tools

252

The origin of bone tool technology

Spatial distribution of lithic and organic artefacts Analysis of the spatial distribution of bone and stone tools in Members 1–3 was conducted to provide a better understanding of the catchment basin and the activities carried out by early hominids at or near the site. These data provided a means by which to address why these artefacts were there, for how long the subsistence strategy they reflect was in existence, and by which hominid type(s) they were used (Backwell & d’Errico, 2003). Analysis of the stone tool assemblages from Swartkrans Members 1–3 shows that they are not the result of in situ knapping activities (Clark, 1993; Field, 1999). The range of flaking debris that one may expect to find in a flaking area is absent and there is no refitting of pieces, suggesting lithics, as faunal and hominid remains gravitated down the entrance shafts from the hillside exterior. This is consistent with the hypothesis of a relatively low sedimentation rate, with material of different nature falling into the cave from the hillside, but in Clark’s view does not exclude the possibility that artefacts may from time to time have been introduced into the cave by hominids, an event which is more likely to have occurred in Member 3 where a consistent amount of burnt bone, a number of faunal remains with clear cut-marks, and evidence suggesting the presence of a flat area were found. Field’s (1999) study of the stone tool collection indicates that the proportion of bone tools versus lithics remains roughly similar in the three members. Thus, considering the homogeneity in time of both categories, this proportion is likely to reflect stability in the artefact catchment basin and in the distance from the cave entrance of the activity area where the artefacts were discarded. This proportion may also depend, if these activities were carried out very close to the entrance, on a similar intensity of production and use of these two categories of artefacts through time near the site. The excellent state of preservation of the wear pattern on the bone tools suggests that these artefacts were incorporated in the deposit relatively quickly and were discarded relatively close to the site. In contrast, the stone tools show different degrees of weathering. The different degrees of alteration indicate that, unlike bone tools, some stone remained exposed to alteration processes in the landscape for longer. This suggests that the catchment basin for the bone tools was smaller than that of the stone tools, incorporating in the deposit bone artefacts discarded close to the entrance before being altered or destroyed by taphonomic agents. The virtual absence of bone tools with poorly preserved wear patterns suggests that if present in this larger area, the bone tools were unable to reach the entrance of the cave before falling victim to taphonomic processes. The bone tool spatial distribution (Fig. 7A) reveals that in each member the tools come from a different area, with very little overlapping. The plot of Member 1 shows two concentrations in the northeast quadrant, as well as three isolated instances in the

253

From Tools to Symbols

Figure 7 Spatial distribution of (A) bone and (B) stone tools in Members 1–3. The grey area represents the unexplored in situ Member 1 Hanging Remnant.

southern quadrants. With the exception of one piece, the bone tools from Member 2 are scattered along a north–south axis in bands 4–6. As is to be expected, all the bone tools from Member 3 lie within the restrictive Member 3 gully. The depth at which the bone tools occur reveals a north–south slope in the vertical distribution of the bone tools from Members 1 and 2, and an opposite trend in those from Member 3 (Fig. 8). Interesting differences appear when we compare bone and stone tool distribution patterns (Fig. 7A, B). Lithic artefacts from Member 1 cluster mainly in two areas located in the northeast and southeast quadrants. The northern concentration, which has the highest density of artefacts, corresponds in area and depth to the main concentration of bone tools seen in this Member. However, no bone tools come from the 11 m2 making up the southern concentration. Also, no lithic artefacts come from the 2 m2 near the northern limit of the excavation where three bone tools were found at a considerable depth. Whilst occurring in roughly the same area, bone and stone tools from Member

254

The origin of bone tool technology

Figure 8 Vertical distribution of the Swartkrans bone tools from Members 1–3. Depths are means of the depth interval at which the pieces were found.

2 differ significantly in their distribution density. The squares which have yielded the highest number of lithics have yielded no bone tools, and the highest concentration of bone tools comes from squares where comparatively few or no lithics were found. Still a different situation is observed in Member 3, where the highest number of bone tools was found. In spite of a significant overlap between the concentrations of the two categories of artefacts, that of the bone tools appears skewed towards the northwest of the quadrant. To this difference also corresponds a difference in the depth of the objects, five bone tools having been found in W3 and W5 at a lower depth (550–700 cm) than any of the lithics from this member. It is interesting that the bone tools do not share the same spatial and vertical distribution as the stone tools. This indicates that in a number of instances the two types of artefacts did not enter the cave at the same time. This difference suggests that the two types of tools were used in different tasks, possibly reflecting seasonal activities conducted at a slightly different place or time by most of the members of a hominid group, different members of the same hominid type (male, female, juvenile) or different hominid taxa visiting the site.

255

From Tools to Symbols

Our spatial data demonstrate that bone tools were incorporated in the Swartkrans deposits during the whole accumulation time of Members 1–3. This is significant in that many of these pieces come from the Member 1 Lower Bank, and their proportion relative to the rest of the faunal assemblage remains constant through all of the members. This eliminates the possibility that the bone tools are recent intrusions in substantially older deposits, as this would result in a decrease of their occurrence from the top to the bottom of the sequence, which is not the case. Based on faunal analysis, Member 1 falls within the time span 1,8–1,0 Mya. Members 2 and 3 do not differ significantly from Member 1, and are assumed to be between 1,5 and 1,0 million years old (Brain, 1993). However, evidence of mixing suggested by a reappraisal of the palaeontological evidence (deRuiter, 2003) and by Electron Spin Resonance dating (Curnoe et al., 2001), may however limit this time-span to between 1,8 and 1,5 Mya (Member 1). If the faunal dating methods applied at Swartkrans are correct, it implies that a bone tool culture existed unchanged in this region for nearly a million years.

The East African evidence Background Mary Leakey (1971) reports 125 artificially modified bones and teeth from Olduvai Beds I and II bearing evidence of intentional flaking, battering and abrasion (Fig. 9). These specimens derive from massive elephant, giraffe and Libytherium limb bones, and to a lesser extent from equids and bovids, as well as from hippopotamus and suid canines. In a comprehensive reappraisal of this material, Shipman (1989) correctly points out that Leakey’s identification of Olduvai bone tools was not based on explicit criteria, and lacked analogies that would allow the ruling out of alternative interpretations. In her reappraisal of the Olduvai material, Shipman (1984, 1989) used a control sample consisting of scanning electron microscope-analysed resin replicas of bones submitted to a number of natural phenomena (weathering, chewing, licking, digestion, wind, etc.), and experimental or ethnographic bone tools used for butchering, digging, grinding, or hide and meat processing. Microscopic analysis of these collections provided criteria (Shipman & Phillips-Conroy, 1977; Shipman et al., 1984; Shipman & Rose, 1988) by which to identify the material on which bone tools were used (hides, meats, soft vegetables), the kinesis and function (digging, bark-working, grinding hard grains, butchery), and the duration (brief, moderate, extensive) for which they were used. Shipman’s ability to distinguish between unused and used bones, and to identify their main function, was verified through blind tests. The control sample also includes experimental reproduction of wind abrasion through the use of an abrasion gun driven by pressurised air. Sedimentary abrasion was mimicked using a tumbling barrel with

256

The origin of bone tool technology

Figure 9 Olduvai bone tools proposed by Leakey: (a) BKII 068-6668; (b) BKII 068-6666; (c) DKI 067-4259; and by Leakey and by Shipman (d) MNKII 068-6676; (e) FCII 068-6679; (f) SHKII 068-6688. Scale = 1 cm.

257

From Tools to Symbols

different types of sediments, with and without the addition of water. According to Shipman, utilisation produces differential wear between functional and non-functional zones of the tool, and at a microscopic scale, between more exposed and recessed/ concave areas, while aeolian and sedimentary abrasion with no water create a pitted or pebbly texture, homogeneously altering the entire surface. Pits caused by striking harder particles may occur on areas worn by utilisation, but they are irregularly spaced and sized. Also, experimental abrasion only rarely creates scratches, while utilisation on mixed substances produces glassy polish crossed by striations. Shipman stresses, however, that these criteria are provisional and that further experimental studies of abrasion are needed to firmly identify distinctive features. Application of these criteria to 116 of the 125 pieces described by Leakey (teeth were excluded from Shipman’s analysis) led her to conclude that 41 were utilised by hominids and the remainder bore ambiguous traces or evidence of abrasion by sediment. Four of the tools bearing punctures – a patella, astragalus, femoral condyle and magnum – are interpreted as anvils due to the triangular or diamond shape of the impressions, which are different from those produced by carnivores due to the absence of counter-bites; large size of the bones difficult to bite; location of the marks consistent with their proposed use, and their apparent antiquity. Shipman, following Leakey, proposes that the marks on these tools may have been produced by stone awls found in the same localities, and that they were used to pierce leather/hide. Among the remaining 37 specimens diagnosed as implements, 35 are described as bones broken and shaped by flaking prior to use. Twenty-six are interpreted as light-duty implements used on soft substances (hide-working), and the remaining 11 described as heavy-duty tools utilised on mixed substances, perhaps in butchery or digging activities. According to Shipman, wear patterns cannot be confused with sedimentary abrasion or weathering, since bone tools show, with the exception of three cases, a low degree of natural alteration. Variables such as taxon, body part, breakage (location, orientation, type and number) and type of surface alteration (weathering, abrasion) were recorded by Shipman on the 41 tools and on 350 randomly selected bones from Olduvai and a few other sites. Comparison of these parameters indicated that the bone tools had a significantly higher occurrence of flaked fractures, flake scars and punctures, and a lower presence of stepped, jagged, or smooth fractures, suggesting that the bone tools were broken shortly after the death of the animal. It also showed that humeri, scapulae and femora, particularly from giraffids and elephants – relatively rare taxa at Olduvai – are overrepresented among the bone tools.

258

The origin of bone tool technology

Methodology Olduvai material The Olduvai bone tool collection housed in the Department of Archaeology at the National Museums of Kenya in Nairobi consists of 125 specimens. These include the pieces designated as tools by Leakey (1971), minus seven specimens described by her, but that could not be located in the museum, and seven specimens not considered by her as tools but that we considered relevant for the analysis (HWKEII 368; HWKEII 886; MNKII 23369; MNKII 1099; BKII 2494; BKII 068-6688; BKII 3240). Annotated line drawings comprising 2–4 aspects of each specimen were made. These recorded the location of macro- and microscopic modifications such as original or post-depositional breakage, flake removals, punctures, carnivore traces, cut-marks, trampling and polish. Recorded variables also included taxon, body part, bone region involved, dimensions of each specimen, the weathering stage according to Behrensmeyer (1978), and location, number, association and length of flake scars according to fracture axis. While some of these variables have already been recorded by Shipman, others – such as the number, location on the bone flake, occurrence on the periosteal versus medullar face, and dimension of removals, possibly due to intentional shaping – were recorded in the framework of the present study for the first time. The same variables were recorded on a control sample of 86 randomly selected limb bone shaft fragments from the FLKI, FLKNI, FLKII, BKII, MNKII and DKI Olduvai sites. This was to establish whether the modifications recorded on the purported bone tools did not represent an extreme in variation affecting, to a lesser degree, the remainder of the Olduvai assemblage. Colour slides and digital images of 2–4 aspects of each piece were also taken, in order to document the collection. The same methods described for Swartkrans were used to make 76 replicas from different areas of the purported tools and the control sample, which consisted of shaft fragments from the FLKI, FLKII and MNKII Olduvai sites. Cast areas included the edges of the tools, whether described by Shipman as utilised or not, regions located away from the purported functional zones, and similar areas on the control specimens. All puncture marks and some cut marks were also moulded. Transparent replicas were examined in transmitted light and 300 digital micrographs were captured. Forty-one replicas were analysed with a Scanning Electron Microscope (Bromage 1987; d’Errico 1988) and 380 SEM micrographs were taken at 15x to 350x magnification. The presence of striations (either single or multiple, parallel or intersecting) and evidence of smoothing, polishing, pitting, and possible residues were recorded.

259

From Tools to Symbols

Comparative collections Thirty-five non-human reference collections of known taphonomic history were examined and studied using the same microscopic techniques described above. These represent nine damage categories derived from both modern and fossil contexts, including animals (hyaena, dog, leopard, cheetah, porcupine) and geological processes (river gravel, spring, flood plain, wind, trampling). Experimental material Nine modern elephant limb bones, ranging between 9 and 22 kg each in weight, were experimentally broken by 26 university students male and female. Eight of the bones originated from a young adult about 20 years old that had died five months before the experiment. Only one bone originated from a teenage individual and was weathered. The experiment was conducted at Plovers Lake in the Sterkfontein Valley, South Africa. The students were asked to work in groups of 3–5 in order to break the bones and produce flakes, employing only resources available in the environment. Knapping of bone flakes was attempted by one of us (FD) using elongated pebbles to replicate the flake removals recorded on the Olduvai purported bone tool collection. Un-retouched flakes were used for flaying and cutting the fresh meat from an adult male eland, working fresh hides with the addition of sand, drying hides with the addition of salt, and digging in soil to extract tubers and grubs, as well as removing bark from trees.

Results Comparative microscopic analyses of the purported tool edges, areas far from the potential functional zones, and edges of bone pieces from the remainder of the Olduvai assemblage, show that the modifications recorded on all of them can be attributed to post-depositional abrasion. Apart from two pieces bearing traces of repeated percussion, a probable bone wedge, and one flake with a macroscopically worn tip, the remainder of the Olduvai purported bone tools do not provide unambiguous evidence of utilisation. However, analysis of the number, location and length of flake scars in the Olduvai bone tool collection reveals that a reduced proportion of purported bone tools bear invasive, contiguous, often bifacially arranged removals not seen in the remainder of the Olduvai assemblage, nor on our experimentally broken elephant bones, elephant bones broken by other researchers, or flaked bones from hyaena dens. This makes these pieces good candidates for having been intentionally modified and used, probably in the butchering of large mammals. One large flake resulting from experimental breakage of elephant bones is noteworthy in that it has a remarkable ‘hand axe’-like morphology with contiguous pseudo-removals on both ends that

260

The origin of bone tool technology

mimic pseudo-bifacial shaping at its base (Fig. 10). In spite of its general resemblance to an Acheulean stone hand axe or to one of the Acheulean elephant bone hand axes from the Italian sites (Radmilli, 1985; Radmilli & Boschian, 1996), this piece has no invasive contiguous bifacial scars. Only two of the four pieces interpreted by Leakey and by Shipman as anvils, a giraffe astragalus (BKII 2933) and an elephant patella (FLKII 884), were located in the National Museums of Kenya (Fig. 11). Our reappraisal of these pieces has taken into account criteria proposed by other authors for identifying the causes of impressions on bone, as well as observations made on our experimentally broken elephant limb bones. Our analysis confirms Leakey’s and Shipman’s diagnosis of these bones as anthropically modified. We believe, however, that an interpretation of these objects as hammers used on intermediate stone tools, rather than anvils on which to pierce

Figure 10 Bone flake resulting from experimental breakage of elephant limb bones with a hand-axe-like morphology.

261

From Tools to Symbols

Figure 11 Top: Astragalus from Olduvai (BKII 2933) with close-up view showing puncture marks. Bottom: Elephant patella from Olduvai ( FLKII 884) with punctures on the articular surface.

skins, fits the evidence better. Experimental piercing of leather (d’Errico et al., 2003) shows that a rotating motion is needed to effectively perforate this material and leave a suitable non-tearing hole. If exerted against a bone surface, this motion results in circular or semicircular impressions with curved internal striations, not seen on the Olduvai specimens. Also, striking motions are unsuitable for piercing skin at precise locations, as generally required by this activity. Piercing a skin by striking it against a bone anvil requires a relatively large and stable bone. Neither of the bones appears large enough, and the patella is particularly unstable. The dispersed location of the punctures on the patella and the location of some impressions near the edge also cast doubt on the anvil interpretation, since the bone would have been destabilised by the striking force. The morphology of these bones, which fit comfortably in the hand, and their use in single-session hammering tasks, is instead consistent with their

262

The origin of bone tool technology

designation as hammers used on intermediate stone tools, most likely wedges to split bones, fruit or wood.

Contrasting South and East African evidence There are two means by which to establish the artefactual nature of potential bone tools showing ambiguous traces of manufacture and use. The first entails the documentation of possible evidence of utilisation, and the demonstration that the recorded modifications, if interpreted as resulting from use, cannot be the outcome of other taphonomic processes. Using this approach, our results suggest not only that the SKX/SE specimens were real tools, but also that they were predominantly used to dig in termite mounds. The same interpretation may apply to the 23 undescribed, but similarly worn bone tool pieces recently found at the Drimolen early hominid site (Keyser et al., 2000), suggesting that bone tool-assisted termite extraction was a persistent component of the subsistence behaviours of early hominids in this area. It is clear that termites were present in this region during the deposition of Swartkrans Members 1–3 by the direct evidence of termite-feeding taxa such as Proteles sp. (aardwolf; Members 1 and 3), Orycteropus afer (antbear; Members 1, 2, and 3), and Manis sp. (pangolin; Member 3) represented in the Swartkrans faunal collection (Watson, 1993). Circumstantial evidence is provided by termite damage identified on some fossils in the Swartkrans faunal collection (Newman, 1993). Using chimpanzees to model early hominid behaviour, we argue for an implement-assisted termite-foraging cultural tradition among southern African hominids, and the role of insectivory in the early hominid diet. We also propose tool utilisation by robust australopithecines, based on the absence of Homo remains in Swartkrans Member 3 (where the largest collection of bone tools was found), and the abundance of Paranthropus robustus remains at Drimolen (found in association with many bone tools and only two possible stone tools). This hypothesis is consistent with independent isotope analyses that show a significant proportion of protein in the diets of both Homo and Paranthropus robustus – the latter traditionally considered a vegetarian. Comparative microscopic analysis of different areas of the purported Olduvai tools, and of the edges of bone pieces from the rest of the bone assemblage (control sample), suggests that possible modifications due to utilisation are indistinguishable from features attributed to post-depositional abrasion. This conclusion is reached after a systematic microscopic survey of the purported bone tools and control sample from Olduvai. Experimental and comparative non-human-modified bone collections were similarly surveyed, involving optical and Scanning Electron Microscopic inspection of hundreds of specimens. Additionally, further visual comparison and the recording

263

From Tools to Symbols

of features on a comparable amount of SEM micrographs were conducted. We cannot exclude the possibility that similar research by Shipman has made her more adept than we are in the identification of anthropic use-wear, as distinct from other causes. If this is the case, however, one has to acknowledge that her criteria for making this distinction and differentiating between task-specific tools are not clear. Robust criteria are essential if inferences from this type of archaeological evidence are to be made, accepted by a scientific community, and become shared knowledge reinforced by repeatable results. Future analyses of the identification of anthropic use-wear should include the quantification of possible worn areas and the development of appropriate analogues. At present, the SEM is perfectly suited to documenting microscopic features; however, if it is the only diagnostic tool used, it may provide deceptive results for this site, in that gentle, mechanical sedimentary abrasion appears to have affected most of, if not the entire Olduvai assemblage, overprinting potential evidence of use-wear. It is noteworthy that experimentally used bone tools show that tasks involving a high degree of mechanical abrasion, such as digging in soil or working hide with sand, produce distinct localised macroscopic modifications on the active zone of the tool. Considering the excellent state of preservation of the more probable Olduvai tools, one would expect that the presence of use-wear generated by these aggressive tasks should be easily detected on the edges of tools. With the possible exception of two pieces (BKII 201, MNKII 1741), no evidence of localised macro-wear is observed on the probable tools. This suggests that they may have been used in activities such as butchering, which do not significantly alter the tool edge. The second means by which to identify ambiguous bone tools is through the recognition of intentional modifications for the purpose of shaping the artefact, and the demonstration that such modifications cannot be ascribed to natural agents, or be the by-product of other subsistence activities. Our identification of possible traces of grinding on seven Swartkrans bone tools led to a comprehensive description of this type of modification, and analysis of a wide range of comparative material for verification. Our results show that grinding is characterised by wide parallel striations orientated oblique to the bone main axis. These striations are morphologically distinct from those produced by use-wear, are limited to facets only, and do not occur as natural alterations in large collections of modern and fossil horn cores. The fusiform striations recorded on the tips of some Swartkrans specimens closely match those observed on archaeological and experimental material where grinding was used as a shaping technique. Villa and Bartram (1996) correctly caution against the interpretation of flaked bones as evidence of bone shaping without the support of contextual and taphonomic analysis of the bone assemblage. They report on bones of medium-sized to large herbivores from the Pleistocene hyaena den of Bois Roche in France bearing continuous scars that

264

The origin of bone tool technology

in some cases mimic scaling retouch. A carnivore origin for the flake scars on the more convincing Olduvai bone tools cannot be advocated for a number of reasons. Almost all of the Bois Roche ‘flaked’ bones show clear signs of hyaena damage in the form of heavy gnawing of articular ends, and pitting and scoring on shafts, features that are rare at Olduvai and virtually absent on the specimens interpreted as tools. Instead, the majority of these pieces record diagnostic stone-induced percussion marks, in a number of cases clearly associated with flake scars. Additionally, pseudo-retouch at Bois Roche is small relative to bone size and does not invade the surface of bones from large mammals by more than 15 mm on the pieces illustrated by Villa and Bartram. This is in stark contrast to the more invasive removals recorded on the Olduvai bone tools. If carnivores were responsible for the production of flake scars recorded on the bone tool collection at Olduvai, we should find the same number and proportion of contiguous removals on bone from medium-sized to large mammals in the Olduvai control sample, but this is not the case. Our results indicate that Mary Leakey was right in isolating a collection of bones that in her opinion looked different from the others emerging at Olduvai, and in proposing their interpretation as tools. This was mainly intuitive, relying on morphological similarities between flake scars on stone and putative bone tools. Our results show that many pieces comprising her original collection do not differ significantly from the control sample, and may be similarly interpreted as intentionally shaped tools or the result of marrow extraction. We also identify a reduced number of specimens that confirm her contention that the bones were tools used by hominids. In order to differentiate between marrow extraction and intentional shaping, future research will focus on the experimental breakage and knapping of extremely fresh bone from very large mammals. Recorded differences in the morphology of the flake scars produced on experimentally broken elephant bones suggest that those on the Olduvai specimens were produced immediately after the animals’ death. The breakage of large bones in the same condition can provide an appropriate analogue by which to gather more informed inferences on early bone tool use by East African hominids.

Conclusion In sum, South and East African early hominid sites dated to between 1,8 and 1 Mya have yielded what appear to be very different types of bone tools. The former are characterised by long bone shaft fragments and horn cores of medium-sized to large bovids, collected after weathering, and possibly used in specialised digging activities. Bone tools of similar shape and size, bearing the same wear patterns and spanning approximately the same time period, occur at Sterkfontein and Drimolen, confirming that a southern African bone tool culture existed for possibly as much as a

265

From Tools to Symbols

million years. Some pieces record intentional shaping of the tips through grinding – a technique peculiar to this raw material. Those from East Africa consist mainly of freshly broken, or more rarely, complete irregular bones from very large mammals, used as such or modified by flaking. Irregular bones or epiphyses appear to have been used as hammers, while the others were apparently involved in a variety of light- and heavy-duty activities. Usage of the end product may be related to multiple tasks, but was most probably restricted to large mammal carcass processing. What are the reasons for such differences? Were these bones used by the same or by different hominid species, if not taxa? If the first applies, do they reflect different cultural traditions? One may expect, if this is the case, to find additional differences between these two regions in other aspects of material culture and adaptation. Although the Oldowan is associated with sites from both regions, this lithic technology appears to occur in East Africa at least more than half a million years earlier than in South Africa (Kibunjia, 1994; Semaw et al., 1997, 2003; Kuman, 1994, 2003; Kuman & Clarke, 2000). This gap may be due to a time lag in the diffusion of this behaviour, staggered independent invention, or a scarcity of late Pliocene deposits in South Africa. Since few studies (Petraglia & Korisettar, 1998) have tried to address this question through detailed comparative technological analysis of contemporaneous lithic assemblages, as currently conducted by Roche’s team on East African sites (Roche et al., 1999), it is problematic at present to know whether what is generally called Oldowan in these two regions corresponds to a single cultural tradition, or is the expression of distinct regional trends. However, our identification of two distinct bone tool cultural traditions in East and South Africa demonstrates that variability in bone tool manufacture may provide a means independent of lithic technology to address crucial behavioural issues and the characterisation of early hominid cultural traditions. The hand-axe-like morphology of one of the flaked bone tools from Olduvai (FCII 068-6679; Fig. 9e) may be taken as an indication that bone shaping by knapping is associated with an Early Acheulean Industry traditionally assigned to Homo erectus. Broken stone bifaces are reported from the same Olduvai locality where the hand-axe-like bone tool was found, but this does not exclude other hominids such as Australopithecus boisei or Homo habilis as the potential makers and users of these tools in East Africa, nor does it exclude Paranthropus robustus as the maker of the South African bone tools. If the bone tool implements in both East and South Africa are purely extensions of the Early Acheulean Industry, then they are presumably adapted to different regions and slightly variable resources. They may also be simply an extension of a single species’ behaviour (possibly Homo erectus). However, the presence of bone digging tools in South Africa might be directly associated with a specific type of epigeal termite mound, or a resource

266

The origin of bone tool technology

specific to this region, and this may account for their atypical Acheulean morphology. The absence of knapped bone flakes in South African sites, and of South African-type digging implements in East Africa, suggests that two distinct bone tool cultures existed in Africa during the same time period, either as extensions of a single species’ behaviour as noted above, or due to manufacture by two different hominid taxa. Based on the bone tool manufacturing techniques recorded in East and South African sites, there appear to be no significant differences between the cognitive abilities of the hominid users, despite their being geographically separated. Evidence of intentional flaking by knapping, seen on the Olduvai bone tools, and traces of grinding on those from South Africa, suggests that the makers of the tools had a clear understanding of the properties of bone, could anticipate the end product, and conceived shaping techniques specific to this raw material in order to achieve optimal efficiency in the tasks for which they were used. Evidence of grinding on the South African bone tools spans Members 1–3 at Swartkrans, indicating that this technique did not appear as an innovation within an existing bone tool culture, but rather represents an integral component of this long-standing tradition. The emergence of bone tool use is clearly not coincidental with the emergence of the genus Homo, but does correspond with the emergence of Homo erectus. In southern Africa, it is also coincident with the emergence of Paranthropus robustus. This suggests, in light of the virtual absence of bone tools in the later African Acheulean and early Middle Stone Age, that early bone tool industries do not represent, as postulated in the past, the first step in a process of increasing sophistication, the beginning of which has been viewed as the behavioural counterpart of the emergence of our genus. In addition, results presented here show that the use of bone and its shaping into taskspecific tools need not imply modern cognitive abilities, and should not, as recently proposed by other authors, be considered as a hallmark of behavioural modernity.

Acknowledgements We would like to thank Francis Thackeray and Heidi Fourie for facilitating access to the Swartkrans material, and Bob Brain and Darryl de Ruiter for helpful information and discussions on the Swartkrans site formation process and taphonomic context. We thank P. Bushozi of the Ministry of Natural Resources and Tourism in Tanzania and A.G. Kaaria of the Ministry of Education, Science and Technology in Kenya for permission to study the Olduvai material. We are most grateful to Meave Leakey, Mary Muungu and Karega Munene of the National Museums of Kenya for facilitating access to the collections. We also thank Pat Shipman for her assistance at the start of the project, and Cathy Snow for carefully reading a first draft of the manuscript. This research was funded by the Ernest Oppenheimer Memorial Trust, the Palaeoanthropological

267

From Tools to Symbols

Scientific Trust, the Cultural Service of the French Embassy in South Africa, the French Ministry for Education and Science, OMLL/ESF Program, Human Sciences Research Council and Nedcor Foundation.

References Aguirre, E. (1986). Format et technique dans la fracturation d’ossements à Torralba (SoriaEspagne). In (M. Patou, Ed.) Artefacts 3. Outillage peu élaboré en os et en bois de Cervidés, pp. 81–89. Treignes: CEDA. Avery, G., Cruz-Uribe, K., Goldberg, P., Grine, F.E., Klein, R.G., Lenardi, M.J., Marean, C.W., Rink, W.J., Schwarcz, H.P., Thackeray, A.I. & Wilson, M.L. (1997). The 1992–93 excavations at Die Kelders Middle and Later Stone Age cave site, South Africa. Journal of Field Archaeology 24, 263–91. Backwell, L.R. (2000). A Critical Assessment of Southern African Early Hominid Bone Tools. MSc. Dissertation. University of the Witwatersrand, South Africa. Backwell, L.R. & d’Errico, F. (1999a). The application of SEM, confocal microscopy and image analysis to assess the first use of bone by early hominids. Microscopy Society of Southern Africa 38th Annual Conference, 1–3 December 1999, University of the Free State, South Africa. Backwell, L.R. & d’Errico, F. (1999b). A reassessment of ‘bone tools’ from the early hominid sites of Swartkrans and Sterkfontein. Archaeological Sciences 99 Conference, 20–22 September, University of Bristol, United Kingdom. Backwell, L.R. & d’Errico, F. (2000). A new functional interpretation of the Swartkrans early hominid bone tools. Abstract for the Paleoanthropology Society Meeting. Journal of Human Evolution 38(3), A4–A5. Backwell, L.R. & d’Errico, F. (2001). Evidence of termite foraging by Swartkrans early hominids. Proceedings of the National Academy of Sciences 98(4), 1358–1363. Backwell, L.R. & d’Errico, F. (2002a). Possible evidence of deliberate shaping on the Swartkrans early hominid bone tools. Meeting of the American Association of Physical Anthropologists, Book of Abstracts, p. 39. 10–14 April 2002. New York. Backwell, L.R. & d’Errico, F. (2002b). What the bones tell us. New information on the Swartkrans early hominid bone tools. Palaeontological Society 12th Biennial Meeting, Book of Abstracts, p. 8. 3–7 October, Bloemfontein. Backwell, L.R. & d’Errico, F. (2002c). Microscopic evidence for deliberate shaping of bone tools by South African early hominids. 15th International Congress on Electron Microscopy, Proceedings 2. Biology and Medicine, pp. 625–626. 1–5 September 2002. Durban. Backwell, L.R. & d’Errico, F. (2003). Additional evidence on the early hominid bone tools from Swartkrans with reference to spatial distribution of lithic and organic artefacts. South African Journal of Science 99, 259–267. Beaumont, P., de Villiers, H. & Vogel, J. (1978). Modern man in sub-Saharan Africa prior to

268

The origin of bone tool technology 49,000 BP: a review and evaluation with particular reference to Border Cave. South African Journal of Science 74, 409–419. Behrensmeyer, A. K. (1978). Taphonomic and ecological information from bone weathering. Paleobiology 4, 150–162. Biddittu, L. & Segre, A. G. (1982). Utilizzazione dell’osso nel Paleolitico inferiore Italiano, Atti della XXIII Riunione Scientifica dell’Istituto Italiano di Preistoria e Protost, pp. 89–105. Florence. Biddittu, I. & Bruni, L. (1987). Il rinvenimento di una amigdala in osso del Paleolitico inferiore a Fontana Ranuccio (Anagni). Latium 4, 3–10. Binford, L.R. (1981). Bones: Ancient Men and Modern Myths. New York: Academic Press. Bonifay, E. (1974). Outils en os et os utilisés dans le Paléolithique ancien du Mas des Caves à Lune-Viel (Héraut) (Note préliminaire). In (H. Camps-Fabrer, Ed.) L’industrie de l’os dans la Préhistoire, pp. 157–167. Aix-en-Provence: Université de Provence. Bonnichsen, R. & Sorg, M.H. (1989). Bone Modification. Orono: University of Maine Centre for the Study of the First Americans. Thompson-Shore Inc. United States of America. Brain, C.K. (1967). Bone weathering and the problem of bone pseudo-tools. South African Journal of Science 63, 97–99. Brain, C.K. (1981). The Hunters or the Hunted? An Introduction to African Cave Taphonomy. Chicago: The University of Chicago Press. Brain, C.K. (1989). The evidence of bone modification by early hominids in southern Africa. In (R. Bonnichsen & M.H. Sorg, Eds) Bone Modification, pp. 291–297. Orono: University of Maine. Brain, C.K., ed. (1993). Swartkrans: a Cave’s Chronicle of Early Man. Transvaal Museum Monograph 8. Pretoria: Transvaal Museum. Brain, C. K. & Shipman, P. (1993). The Swartkrans bone tools. In (C.K. Brain, Ed.) Swartkrans: a Cave’s Chronicle of Early Man, pp. 195–215. Transvaal Museum Monograph 8. Pretoria: Transvaal Museum. Brain. C.K., Churcher, C.S., Clark, J.D., Grine, F.E., Shipman, P., Susman, R.L., Turner, A. & Watson, V. (1988). New evidence of early hominids, their culture and environment from the Swartkrans cave, South Africa. South African Journal of Science 84, 828–835. Breuil, H. (1932). Pointe d’épieu en os du Moustérien de la caverne du Castillo (Santander). L’anthropologie 42, 679–680. Breuil, H. (1938). Bone and antler industry of the Chou-Koutien Sinantropus site. Paleontologia Sinica 6, Beijing. Breuil, H. & Barral, L. (1955). Bois de cervidés et autres os travaillés sommairement au Paléolithique ancien du vieux monde et au Mustérien des grottes de Grimaldi et de l’Observatoire de Monaco. Bulletin du Musée d’Anthropologie Préhistorique de Monaco 2, 3–34. Bromage, T. G. (1987). The SEM/replica technique and recent applications to the study of fossil bone. Scanning Electron Microscopy 1987, 607–613.

269

From Tools to Symbols Brooks, A.S., Helgren, D.M., Cramer, J.S., Franklin, A., Hornyak, W., Keating, J.M., Klein, R.G., Rink, W.J., Schwarcz, H., Leith Smith, J.N., Stewart, K., Todd, N.E., Verniers, J. & Yellen, J.E. (1995). Dating and context of three Middle Stone Age sites with bone points in the Upper Semliki Valley, Zaïre. Science 268, 548–553. Bunn, H.T. (1981). Archaeological evidence for meat-eating by Plio-Pleistocene hominids from Koobi Fora and Olduvai Gorge. Nature 291, 574–577. Bunn, H.T. (1982). Meat-eating and Human Evolution: Studies on the Diet and Subsistence Patterns of Plio-Pleistocene Hominids in East Africa. Ph.D. Dissertation, Princeton University. Cahen, D., Haesaert, P., van Neer, W. & van Pamel, P. (1979). Un outil en os du Paléolithique inférieur dans la nappe alluviale de Mesvin. Hélinium 19, 105–127. Cassoli, P.F., De Guilii, C., Radmilli, A.M. & Segre, A.G. (1982). Giacimento del Paleolitico inferiore a Malagrotta (Rome). Atti della XXIII Riunione Scientifica dell’Istituto Italiano di Preistoria e Protostoria, pp. 531–549. Florence. Chase P.G. (1990). Tool-making tools and Middle Paleolithic behaviour. Current Anthropology 31(4), 443–447. Clark, J.D. (1993). In Swartkrans. A Cave’s Chronicle of Early Man, ed. C.K. Brain, pp. 167–194. Transvaal Museum Monograph 8. Pretoria: Transvaal Museum. Curnoe, D., Grün, R., Taylor, L. & Thackeray, F. (2001). Direct ESR dating of a Pliocene hominin from Swartkrans. Journal of Human Evolution 40, 379–391. Dart, R.A. (1957). The Osteodontokeratic culture of Australopithecus prometheus. Transvaal Museum Memoirs 10. Pretoria: Transvaal Museum. Deacon, J. & Brett, M. (1993). Peeling away the past: The display of excavations at Nelson Bay Cave. South African Archaeological Bulletin 48, 98–104. Deacon, H.J. & Deacon, J. (1999). Human Beginnings in South Africa: Uncovering the Secrets of the Stone Age. Cape Town: David Philip. Debenath, A. & Duport, L. (1971). Os travaillés et os utilisés de quelques gisements préhistoriques charentais. Mémoires de la Société Archéologique et Historique de la Charente, pp. 183–201. Dennell, R. (1983). European Economic Prehistory: A New Approach. New York: Academic Press. d’Errico, F. (1988). The use of resin casts for the study of usewear. In (S.L. Olsen, Ed.) Scanning Electron Microscopy in Archaeology, British Archaeological Report, pp. 155–167. Oxford. d’Errico, F. & Villa, P. (1997). Holes and grooves. The contribution of microscopy and taphonomy to the problem of art origins. Journal of Human Evolution 33, 1–31. d’Errico, F. & Backwell, L.R. (2000). Bone tool use and the identification of early hominid cultural traditions. New data from Swartkrans and Sterkfontein. 11th Biennial Conference of the Palaeontological Society of Southern Africa. 4–8 September 2000. Pretoria: Transvaal Museum. d’Errico, F. & Backwell, L.R. (2001). Early hominid cultural traditions and technical systems: the

270

The origin of bone tool technology full picture. Possible evidence of bone tool shaping from Swartkrans, South Africa. Knapping Stone. A Unique Hominid Behaviour? International Workshop. 21–24 November 2001. Abbaye des Prémontrés, Paris. d’Errico, F. & Backwell, L.R. (2003). Possible evidence of bone tool shaping by Swartkrans early hominids. Journal of Archaeological Science 30, 1559–1576. d’Errico, F., Zilhão, J., Baffier, D., Julien, M. & Pelegrin, J. (1998). Neanderthal acculturation in western Europe? A critical review of the evidence and its interpretation. Current Anthropology 39, 1– 44. d’Errico, F., Backwell, L.R., & Berger, L.R. (2001). Bone tool use in termite foraging by early hominids and its impact on understanding early hominid behaviour. South African Journal of Science 97, 71–75. d’Errico, F., Julien, M., Liolios, M., Vanhaeren, M. & Baffer, D. (2003). Many awls in our argument. Bone tool manufacture and use from the Chatelperronian and Aurignacian layers of the Grotte du Renne at Arcy-sur-Cur. In (J. Zilhão & F. d’Errico, Eds) The Chronology of the Aurignacian and of the Transitional Technocomplexes. Dating, Stratigraphies,Cultural Implications. Lisbon: Instituto Português de Arqueologia. deRuiter, D. (2003). Revised faunal lists for Members 1–3 of Swartkrans, South Africa. Annals of the Transvaal Museum 40, 29–41. Dobosi, V.T. (1990). Description of the archaeological material. In (M. Kretzoi & V.T. Dobosi, Eds) Vértesszölös, Site, Man and Culture, pp. 311–395. Budapest: Akadémiai Kiadó. Fernández-Carriba, S. & Loeches, A. (2000a). Fruit smearing by captive chimpanzees: a newly observed food processing behavior. Current Anthropology 42(1), 143–161. Fernández-Carriba, S., Loeches, A. & Morcillo, A. (2000b). Fruit smearing: a case of food transformation in captive chimpanzees. Animal Behavior Society Meeting. Atlanta, GA, August 2000. Field, A.S. (1999). An Analytical and Comparative Study of the Earlier Stone Age Archaeology of the Sterkfontein Valley. Unpublished M.Sc. Dissertation. University of the Witwatersrand. Freeman, L.G. (1978). Mousterian worked bones from Cueva Morin (Santander, Spain). A preliminary description. In (L.G. Freeman, Ed.) Views of the Past, pp. 29–51. The Hague: Mouton. Freeman, L.G. (1983). More on the Mousterian flaked bone from Cueva Morin. Current Anthropology 24(3), 366–377. Gaudzinski, S. (1998). Middle Palaeolithic bone tools from the open-air site Salzgitter-Legensted (Germany). Journal of Archaeological Science 26, 125–141. Gaudzinski, S. (1999). The faunal record of the Lower and Middle Palaeolithic of Europe: remarks on human interference. In (W. Roebroeks & C. Gamble, Eds) The Middle Palaeolithic Occupation of Europe, pp. 215–233. Leiden: University of Leiden. Gautier, A. (1986). Une histoire de dents: les soi-disant incisives travaillées du Paléolithique moyen de Sclayn. Hélinium 26, 177–181.

271

From Tools to Symbols Gifford-Gonzalez, D. (1989). Ethnographic analogues for interpreting modified bones: some cases from east Africa. In (R. Bonnichsen & M.H. Sorg, Eds) Bone Modification, pp. 179–246. Orono: University of Maine Centre for the Study of the First Americans. Thompson-Shore Inc. United States of America. Haynes, G.A. (1988). Longitudinal studies of African elephant death and bone deposits. Journal of Archaeological Science 15, 131–157. Haynes, G. (1991). Mammoths, Mastodonts, and Elephants: Biology, Behaviour and the Fossil Record. Cambridge: Cambridge University Press. Henri-Martin, G. (1907). Similitude de certains os naturels et de quelques pièces osseuses préhistoriques. Bulletin de la Société Préhistorique Française 4, 432–436. Henshilwood, C.S. & Sealey, J. (1997). Bone artefacts from the Middle Stone Age at Blombos Cave, southern Cape, South Africa. Current Anthropology 38(5), 890–95. Henshilwood, C.S., d’Errico, F., Marean, C.W., Milo, R.G. & Yates, R. (2002). An early bone tool industry from the Middle Stone Age at Blombos Cave, South Africa: implications for the origins of modern human behaviour, symbolism and language. Journal of Human Evolution 41(6), 631–678. Howell, C., & Freeman, L.G. (1983). Ivory points from the earlier Acheulean of the Spanish Meseta. In Homenaje al Prof. Martin Almagro Basch, pp. 41–61. Madrid: Ministerio de Cultura. Justus, A. (1989). Der Altpaläolitische fundplatz Dmanisi in Georgien (Kaukasus), (Knochenund geweihbearbeitung). Jahrbuch des Römisch-Germanischen Zentralmuseums Mainz 36, 107–108. Keyser, A. (2000). New finds in South Africa. National Geographic May, 77–83. Keyser, A., Menter, C.G., Moggi-Cecchi, J., Pickering, T.R. & Berger, L.R. (2000). Drimolen: a new hominid-bearing site in Gauteng, South Africa. South African Journal of Science 96, 193–197. Kibunjia, M. (1994). Pliocene archaeological occurrences in the Lake Turkana basin. Journal of Human Evolution 27, 159–171. Kitching, J.W. (1963). Bone, Tooth and Horn Tools of Palaeolithic Man: An Account of the Osteodontokeratic Discoveries in Pin Hole Cave. Derbyshire. Manchester: Manchester University Press. Klein, R.G. (1975). Palaeoanthropological implications of the non-archaeological bone assemblage from Swartklip I, south-western Cape Province, South Africa. Quaternary Research 5, 275–288. Klein, R.G. (1994). Southern Africa before the Iron Age. In (R.S. Corruccini & R.L. Ciochon, Eds) Integrative Paths to the Past: Paleoanthropological Advances in Honor of F. Clark Howell, pp. 471–519. Englewood Cliffs, New Jersey: Prentice-Hall. Klein, R.G. (1999). The Human Career, 2nd ed. Chicago: University of Chicago Press.

272

The origin of bone tool technology Kuman, K. (1994). The archaeology of Sterkfontein – past and present. Journal of Human Evolution 27, 471–495. Kuman, K. (2003). Site formation in the early South African Stone Age sites and its influence on the archaeological record. South African Journal of Science 99(5/6), 251–254. Kuman, K. & Clarke, R.J. (2000). Stratigraphy, artifact industries and hominid associations for Sterkfontein, Member 5. Journal of Human Evolution 38, 827–847. Leakey, M.D. (1971). Olduvai Gorge, Volume 3: Excavations in Beds I and II, 1960–1963. Cambridge: Cambridge University Press. Leroi-Gourhan, A. (1964). Le geste et la parole I – technique et langage. Paris: Albin Michel. Lemonnier, P. (1986). The study of material culture today: toward an anthropology of technical systems. Journal of Anthropological Archaeology 5, 147–186. Lyman, R.L. (1994). Vertebrate Taphonomy. Cambridge Manuals in Archaeology. Cambridge: Cambridge University Press. Maguire, J.M., Pemberton, D. & Collett, M.H. (1980). The Makapansgat Limeworks Grey Breccia: Hominids, hyaenas, hystricids or hillwash ? Palaeontologia Africana 23, 75–98. Malerba, G. & Giacobini, G. (1998). Les retouchoirs sur éclats diaphysaires du Paléolithique moyen et supérieur de trois sites de l’Italie nord orientale (Grotte de San Bernardino, Abri de Fumane et Abri Tagliente). In Atti del XIII Congresso U.I.S.P.P, Forlí 1996 6(I), 167–171. Mania, D. & Weber, T. (1986). Bilzingsleben III. Deutscher Verlag der Wissenschaften: Berlin. McBrearty, S. & Brooks, A.S. (2000). The revolution that wasn’t: a new interpretation of the origin of modern human behaviour. Journal of Human Evolution 39, 453–563. McGrew, W.C. (1996). The material of culture. In (R.W. Wrangham, W.C. McGrew, F.B.M. de Waal & P.G. Heltne, Eds) Chimpanzee Cultures. Cambridge, Mass.: Harvard University Press. Menter, C.G., Kuykendall, K.L., Keyser, A.W. & Conroy, G.C. (1999). First record of hominid teeth from the Plio-Pleistocene site Gondolin, South Africa. Journal of Human Evolution 37, 299–307. Newman, R.A. (1993). The incidence of damage marks on Swartkrans fossil bones from the 1979–1986 excavations. In (C.K. Brain, Ed.) Swartkrans: a Cave’s Chronicle of Early Man, pp. 217–228. Transvaal Museum Monograph 8. Pretoria: Transvaal Museum. Noble, W. & Davidson, I. (1996). Human Evolution, Language and Mind. Cambridge: Cambridge University Press. Olsen, S.L. (1989). On distinguishing natural from cultural damage on archaeological antler. Journal of Archaeological Science 16, 125–135. Pei, W.C. (1938). Le rôle des animaux et des causes naturelles dans la cassure des os. Palaeontologia Sinica. Peiping: Geological Survey of China, 7, 1–16. Petraglia, M.D. & Korisettar, R.I. (1998). Early Human Behavior in Global Context: The Rise and Diversity of the Lower Palaeolithic Record. One World Archaeology, No. 28. New York: Routledge.

273

From Tools to Symbols Pitts, M. & Roberts, M. (1997). Fairweather Eden: Life in Britain Half a Million Years Ago as Revealed by Excavations at Boxgrove. London: Century. Radmilli, A.M. (1985). Scavi nel giacimento del Paleolitico inferiore di Castel di Guido presso Roma. In (A.M. Bietti Sestieri, Ed.) Preistoria e Protostoria nel territorio di Roma, pp. 75–85. Rome: De Luca. Radmilli, A.M. & Boschian, G. (1996). Gli scavi a Castel di Guido. Florence: Instituto Italiano di Preistoria e Protostoria, ETS. Roche, H., Delagnes, A., Brugal, J.-P., Feibel, C., Kibunjia, M., Mourre, V. & Texier, P.-J. (1999). Early hominid stone tool production and technical skill 2.34 Myr ago in West Turkana, Kenya. Nature 399, 57–60. Schlanger, N. (1994). Mindful technology: unleashing the chaine operatoire for an archaeology of mind. In (C. Renfrew and E.B. Zubrow, Eds) The Ancient Mind: Elements of Cognitive Archaeology, pp. 143–151. Cambridge: Cambridge University Press. Semaw, S., Renne, P. Harris, J.W.K. Feibel, C.S., Bernor, R.L. Fesseha, N. & Mowbray, K. (1997). 2.5-million-year-old stone tools from Gona, Ethiopia. Nature 385, 333–336. Semaw, S., Rogers, M.J., Quade, J., Renne, P., Butler, R.F., Dominguez-Rodrigo, M., Stout, D., Hart, W.S., Pickering, T.R. & Simpson, S. (2003). 2.6-Million-year-old stone tools and associated bones from OGS-6 and OGS-7, Gona, Afar, Ethiopia. Journal of Human Evolution 45, 169–177. Shipman, P. (1984). The earliest tools: re-assessing the evidence from Olduvai Gorge. Anthroquest 29, 9–10. Shipman, P. (1988). Actualistic studies of animal resources and hominid activities. In (S.L. Olsen, Ed.) Scanning Electron Microscopy in Archaeology, pp. 261–285. BAR International Series 452, Oxford. Shipman, P. (1989). Altered bones from Olduvai Gorge, Tanzania: Techniques, problems, and implications of their recognition. In (R. Bonnichsen & M.H. Sorg, Eds) Bone Modification, pp. 317–334. Orono: University of Maine Centre for the Study of the First Americans. Thompson-Shore Inc. United States of America. Shipman, P. & Phillips, J. (1976). On scavenging by hominids and other carnivores. Current Anthropology 17(1), 170–172. Shipman, P. & Phillips-Conroy, J. (1977). Hominid tool-making versus carnivore scavenging. American Journal of Physical Anthropology 46, 77–86. Shipman, P. & Rose, J. (1984). Cutmark mimics on modern and fossil bovid bones. Current Anthropology 25(1), 116–117. Shipman, P. & Rose, J. (1988). Bone tools: An experimental approach. In (S.L. Olsen, Ed.) Scanning Electron Microscopy in Archaeology, pp. 303–335. BAR International Series 452. Oxford. Shipman, P., Fisher, D.C. & Rose, J.J. (1984). Mastodon butchery: microscopic evidence of carcass processing and bone tool use. Paleobiology 10(3), 358–365.

274

The origin of bone tool technology Singer, R. & Wymer, J. (1982). The Middle Stone Age at Klasies River Mouth in South Africa. Chicago: Chicago University Press. Stepanchuk, V.N. (1993). Prolom II, a Middle Paleolithic site in the Eastern Crimea with nonutilitarian bone artefacts. Proceedings of the Prehistoric Society 59, 17–37. Sutcliffe, A.J. (1973). Similarity of bones and antlers gnawed by deer to human artefacts. Nature 246, 428–430. Sutcliffe, A.J. (1977). Further notes on bones and antlers chewed by deer and other ungulates. Deer 4(2), 73–82. Van der Ryst, M.M. (1998). The Waterberg Plateau in the Northern Province, South Africa, in the Later Stone Age. British Archaeological Reports, International Series 43. Oxford: Oxford University Press. Villa, P. & Bartram, L. (1996). Flaked bone from a hyaena den. Paleo 8, 143–159. Villa, P. & d’Errico F. (1998). Bone and ivory points in the Lower and Middle Paleolithic: a review of the evidence from Western Europe. Abstract for the Paleoanthropology Society Meeting. Journal of Human Evolution 36(4), 25–26. Vincent, A. (1988). Rémarques préliminaires concernant l’outillage osseux de la Grotte Vaufrey. In (J.P. Rigaud, Ed.) La Grotte Vaufrey à Cenac et Sain-Julien (Dordogne). Paleoenvironnement, chronologie et activités humaines, pp. 529–533. Mémoires de la Société Préhistorique Française 19. Wadley, L. (1997). Excavations at Rose Cottage Cave. South African Journal of Science 93, 439– 444. Watson, V. (1993). Composition of the Swartkrans bone accumulations, in terms of skeletal parts and animals represented. In (C.K. Brain, Ed.) Swartkrans: a Cave’s Chronicle of Early Man, pp. 35–73. Transvaal Museum Monograph 8. Pretoria: Transvaal Museum. Whiten, A., Goodall, J., McGrew, W.C., Nishida, T., Reynolds, V., Sugiyama & Tutin, C.E.G. (1999). Cultures in chimpanzees. Nature 399, 682–685. Yellen, J.E., Brooks, A.S., Cornelissen, E., Mehlman, M.J. & Stewart, K. (1995). A Middle Stone Age worked bone industry from Katanda, Upper Semliki Valley, Zaïre. Science 268, 553–556. Zilhão, J. & d’Errico, F. (1999a). Reply to Straus, Mellars and Otte. Current Anthropology 40(3), 355–364. Zilhão, J. & d’Errico, F. (1999b). 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.

275

Contribution of genetics to the study of human origins Himla Soodyall and Trefor Jenkins MRC/NHLS/Wits Human Genomic Diversity and Disease Research Unit, National Health Laboratory Service and the University of the Witwatersrand, Johannesburg, South Africa

Abstract We can reconstruct human history using a number of different methods. In the absence of written records, scholars have made use of information from disciplines including linguistics, archaeology, physical anthropology, cultural anthropology, history and palaeoanthropology to reconstruct their prehistory. The most direct account of our past is inferred from the fossil record. Skeletal remains have been instrumental in establishing the evolution of human ancestors in Africa, and they have also provided important information about the evolution of modern Homo sapiens. Study of the genetic variation of humans, the concern of the field of molecular anthropology, attempts to produce objective data with which to provide new insights about human history. From a molecular-genetic perspective, it is clear that the DNA found in contemporary individuals has been passed down to them from previous generations. It is also clear that in every generation, some DNA sequences are not passed on because some individuals have no children or the sequence fails to be transmitted during meiosis. Therefore, the genealogy of a DNA sequence will trace back to fewer and fewer ancestors until it comes together in one common ancestor. Genetic studies, including those of mitochondrial DNA and Y chromosome DNA studies, have suggested that this ancestor lived in Africa, about 100–150 Kya (thousand years ago). Genetic studies are also providing insights into what makes humans different from our closest primate relative, the chimpanzee. It is becoming more evident that in addition to genomic differences, differences in the level of expression of certain genes could be responsible for producing the morphological and adaptive changes found between humans and chimpanzees.

276

Contribution of genetics to the study of human origins

Résumé Il est possible de retracer l’histoire de l’humanité en utilisant différentes méthodes. Pour retracer la préhistoire les scientifiques ont fait usage, en l’absence de traces écrites, d’informations provenant de disciplines telles que la linguistique, l’archéologie, l’anthropologie physique, l’anthropologie culturelle, l’histoire et la paléoanthropologie. La manière la plus directe pour faire état de ce cheminement est de faire appel aux fossiles. Les fossiles ont joué un rôle décisif pour déterminer l’origine africaine de notre lignée et nous informer sur l’origine de notre espèce. L’étude de la variabilité génétique humaine, qui fait partie du domaine de l’anthropologie moléculaire, se propose de produire des données objectives nous donnant un nouvel aperçu de l’histoire de l’humanité. La génétique moléculaire nous enseigne que Nos contemporains ont hérité leur ADN des générations précédentes. A chaque génération certaines séquences d’ADN n’ont pas été transmises parce que certains individus n’ont pas eu d’enfants ou parce qu’une séquence n’a pas été transmise lors de la méiose. Nous pouvons ainsi retracer la généalogie d’une séquence d’ADN sur un nombre de plus en plus réduit d’ancêtres, pour en arriver à un ancêtre commun. Plusieurs études génétiques et en particulier ceux sur compris l’ADN mitochondrial et sur le chromosome Y, indiquent que cet ancêtre vivait en Afrique il y a 100 000 à 150 000 ans. Les études génétiques nous donnent aussi un aperçu de ce qui rend les humains différent du chimpanzé, le primate le plus proche. Il devient de plus en plus évident que, en plus des différences génomiques, des différences dans le niveau d’expression de certains gènes pourraient être responsables des changements morphologiques et adaptatifs observés chez les hommes et les chimpanzés.

Phillip Tobias – a personal tribute by Trefor Jenkins I feel privileged to have been invited to speak at this International Round Table ‘From tools to symbols: from early hominids to modern humans’, which is being held to honour Professor Emeritus Phillip Tobias. The discovery of the structure of DNA by Watson and Crick fifty years ago, almost to the day, is a happy coincidence – although Phillip had already shown an interest in genetics a few years before that, completing a PhD entitled ‘Chromosomes, Sex-Cells and Evolution of the Gerbil’, in 1953. When I joined his staff (as a very junior demonstrator in anatomy) in February 1963, his reputation as an anatomist, a cytogeneticist, a palaeontologist and a human geneticist was already well established (Fig. 1). He had already made important contributions to the study of the living peoples of southern Africa, and even saw the occasional family for genetic counselling. I had been working in Durban as a learner surgeon for only a few weeks when a colleague of mine there told me about his eminent teacher of anatomy at Wits, where he had qualified, and I was intrigued to hear that Phillip had conducted fieldwork among the Tonga people of the Gwembe (Zambezi) Valley on the northern side of Lake Kariba. The Tonga had only a few years earlier been moved from the Valley to the plateau, prior to the construction of the

277

From Tools to Symbols

Kariba Dam and the subsequent flooding of the valley. As a medical officer working on the southern side of Lake Kariba I had, in a very amateurish way, carried out some serological genetics research on the Tonga. This happy coincidence resulted in my journey to Johannesburg in mid-1962 to meet Phillip and to discuss the possibility of working with him at Wits. This life-changing experience for me led to my entry into academia and the discovery that I liked teaching and research. With encouragement, guidance and generous support from Phillip Tobias and colleagues in his department, I carried out field work in Zambia, Botswana and South Africa, researching the genetics and health of the peoples of subSaharan Africa and, later, those of Namibia, Papua New Guinea and Madagascar. More recently still, there have been forays into Central African Republic, the Democratic Republic of Congo, Uganda and Mozambique. What a privileged life I have led. And I shall be eternally grateful to Phillip for welcoming me into the world of academics which, in the field of anthropology and human genetics, he has continued to inspire.

Figure 1 Photograph featuring the relatively recently appointed head, Phillip V. Tobias, sitting centre stage, first row (1963).

278

Contribution of genetics to the study of human origins

I have been privileged to see, at close quarters, a man of high principle and idealism; courageous and fearless, a strong opponent of racism in a country where race discrimination was a part of the way of life. He brought his astute incisive mind to bear on the debate over the pernicious policy of apartheid, convincingly arguing that the policy was morally bankrupt and had to be dismantled as soon as possible. He gives tangible expression to his strong convictions in the field of education and research. He remains an example to us all.

Introduction During the first twenty years or so of the career of one of us (TJ), research in population genetics was carried out in South Africa at a leisurely pace. Some of the research questions we asked were: ‘What (if any) are the genetic differences between San (“Bushmen”) and Khoi (“Hottentots”)?’ ‘What are the relationships between the Khoisan people and Bantu-speakers?’ ‘What (if any) are the genetic differences between the San who speak different languages?’ ‘How do the different chiefdoms of the Bantuspeakers differ genetically?’ ‘What is the genetic constitution of different, so-called “Coloured” populations of South Africa?’ ‘What proportion of African genes have been assimilated by the Caucasoid populations of South Africa?’ To answer these and other similar questions, we had at our disposal a wide-range of classical gene markers: blood groups, serum proteins and red cell enzymes, as well as sensory polymorphisms like colour blindness and the ability to taste phenyl thio carbamide, and pharmacogenetic markers like the rapid/slow acetylation characteristic and hereditary persistence of intestinal lactase, to mention part of our repertoire during those years. From 1980 onwards, armed with recombinant-DNA technology, and later, soon after the advent of the polymerase chain reaction (PCR), we were able to elevate population genetic studies to a new plane of sophistication, facilitating the possible examination of hundreds to thousands of genetic polymorphisms. It was the study of mitochondrial DNA (mtDNA) and later Y-chromosome DNA polymorphisms which led to the acknowledgement that humans had originated in Africa, approximately 100 Kya, and had subsequently migrated ‘Out of Africa’ to populate the rest of the world. All ‘races’/populations of human beings belong to one species, Homo sapiens sapiens, a relatively young species. There has, of course, been some divergence between the present-day populations, the differences being brought about by adaptations to the different environmental circumstances in which these populations continued to live, as well as by the operation of chance effects, like the particular characteristics which their founding fathers (and mothers) happened to possess. So-called ‘races’ are thought to be the result of such influences. For all the apparent morphological differences which some observers believe

279

From Tools to Symbols

characterise these different populations (e.g. skin colour, shape of nose, colour of eyes, type of hair), the human species remains strikingly one. All the populations of the world are inheritors of the same genome and are united by the same physical abilities and cultural characteristics, like the ability to learn a complex language, to be capable of abstract thought and to conceptualise ideas and to posses a consciousness and self-awareness to which no other species can compare. These genetically determined abilities/characteristics are possessed to equal degree by all human populations. In spite of the overwhelming similarity in the DNA sequence of the genome in all humans, there is, on average, one difference per thousand nucleotides between any two individuals; these differences are known as single nucleotide polymorphisms or SNPs (pronounced ‘snips’). To understand fully the global picture of genetic variation at any locus in the human genome, which is approximately 3 billion nucleotides (or building blocks) in length and which carries about 30 000 genes in total, we have to understand variation in Africa. The peoples of sub-Saharan Africa are endowed with more genetic variation than all the other populations of the world. Thus, the unique genetic resources within our region, together with a history that traces back to the very beginnings of modern humans, and in fact of hominids, provides us with an ideal opportunity to reconstruct human history and evolution.

Reconstructing history – the molecular tool kit Genealogy is the study of family history and descent from ancestors. People usually trace their ancestry using information from family records, e.g. birth certificates, marriage certificates, death certificates and other historical archival material. The concept of ancestry is deeply rooted in our different cultures. We all identify with the family unit – our parents, brothers and sisters, grandparents, great-grandparents, and so on. We sometimes recognise certain physical features (hair colour, nose shape, etc.) or some behavioural characteristics (temper, excellence in sport, art, music, etc.), as well as some hereditary diseases (e.g. haemophilia in the royal families of Europe) that we have inherited from some family member. The thread that connects us biologically with our ancestors is stored in the DNA found in every cell of our body. Tracing one’s ancestry can now be done incorporating new genetic technology using the information in DNA that makes up the human genome. The total genetic complement of humans contains some 3 billion of these bases in different combinations controlling the development of the organism from conception to birth, to death, and producing the genetic variation that distinguishes one individual from every other one. Humans have 46 chromosomes, half of them inherited from our mothers and the other half from our fathers. Chromosomal DNA is found in the nucleus of the cell and is referred to as nuclear DNA. In addition to

280

Contribution of genetics to the study of human origins

nuclear DNA, the mitochondria, the energy-producing organelles in the cytoplasm of all cells, also contain DNA that is referred to as mitochondrial DNA (mtDNA). MtDNA is inherited only from our mothers and only females can pass it on to their children. The Y chromosome is found in the nucleus and is transmitted exclusively from father to son.

Tracing genetic lineages Since mtDNA is passed on exclusively by females and the Y chromosome is passed on exclusively by males, both types of DNA can be used to trace maternal and paternal lineages (Fig. 2). A lineage is the direct line of descent from an ancestor. There are many different mtDNA and Y chromosome lineages present among humans living throughout the world. The different lineages are the result of changes (mutations) introduced in the DNA during the course of human evolution. It is possible to reconstruct the relationships of all lineages found in present-day people and to trace them to a single common mtDNA and Y chromosome ancestor, and to estimate the time when these ancestors would have lived. The limitation of using mtDNA and Y chromosome DNA for genealogical testing is that this DNA will trace only two genetic lines on a family tree in which branches double with each preceding generation. For example, Y chromosome tracing will connect a man to his father but not his mother, and it will connect him to only one of his four grandparents: his paternal grandfather. In the same way it will connect him to one of his eight great grandparents. Continue back in this manner for 14

Figure 2 Schematic representation of mtDNA and Y chromosome DNA inheritance across four generations.

281

From Tools to Symbols

generations and the man will still be connected to only one ancestor in that generation. Y chromosome DNA testing will not connect him to any of the other 16 383 ancestors in that generation to whom he is also related in equal measure. The same scenario applies when using mtDNA. The genetic variation found among living peoples offers another way of studying human evolution. The underlying principle of this approach is to reconstruct the history of mutations found in the DNA of contemporary individuals, and to trace their origins to a common ancestor who would have lived at some point in the past. The age of this ‘ancestor’ can be estimated using a molecular clock, i.e. the rate at which mutations occur, calibrated for the segment of DNA under investigation. It has been shown that major demographic events such as population migrations, bottlenecks and expansions leave imprints, in the form of altered gene frequencies, on the collective human genome. Because these imprints are transmitted to succeeding generations, the modern human genome contains an indelible record of our evolutionary past.

MtDNA variation mtDNA is particularly useful when studying human evolution because of its unique pattern of inheritance. Unlike nuclear DNA, it is strictly maternally inherited and does not undergo recombination. Therefore differences in mtDNA are the direct result of mutations, and the ‘history’ of these mutational events can be reconstructed from contemporary divergent lineages. Also, mtDNA evolves about 10–15 times faster than nuclear DNA, thus facilitating the discrimination between closely related populations. For these reasons, mtDNA has been exploited as a genetic marker to study the transmission of inherited traits passed on exclusively by females. In 1987, Rebecca Cann, Mark Stoneking and the late Alan Wilson from the University of California, Berkeley fuelled the debate concerning modern human origins by suggesting that the mtDNA found in all living humans could be traced to a single ancestor who lived in Africa about 200 Kya (Cann et al, 1987). Later, Vigilant et al. (1991) extended the mtDNA studies conducted by Cann and colleagues, by sequencing the control region of the mtDNA molecule from a sample of individuals from different parts of the world. Once again, the data and newer methods of analysis supported the theory that modern humans originated in Africa. More recently Ingman et al. (2000) succeeded in studying the entire sequence of the mtDNA molecule in a worldwide sample of 53 individuals. As with previous studies, they found that Africans were more genetically diverse, but they also estimated that the divergence between African and non-African populations took place as recently as 52 Kya, instead of the 100–150 Kya date estimated from other mtDNA studies. One of the most significant findings to emerge from mtDNA studies is that non-

282

Contribution of genetics to the study of human origins

Africans often show genetic signs of a severe reduction in population size, a ‘bottleneck,’ at some time in the past, followed by a population expansion (Ingman et al., 2000). This bottleneck and expansion is presumed to have occurred when a branch of the early modern human population from Africa split off to form a small subpopulation, which then expanded in size as it spread out to colonize Eurasia (Brown, 2001). The distribution of mtDNA types among populations from different regions of the world is consistent with the Out of Africa theory. The mtDNA subhaplogroups (groups of lineages derived from a common ancestor) showing greatest antiquity are still retained in African populations (L1, L2 and L3), and all other subhaplogroups found in nonAfrican populations can be traced to subhaplogroup L3 (Fig. 3). More importantly, Khoisan populations from southern Africa harbour some of the oldest mtDNA lineages found in living peoples. Another significant advance towards understanding our direct history has been the successful extraction of DNA from Neanderthals, and how information about their gene pool (albeit mtDNA) can shed light on human evolution. To date, ancient DNA (aDNA) studies have been conducted on Neanderthal specimens from three regions – the Feldhofer specimen from the Neander Valley in Germany (Krings et al., 1997, 1999), the Mezmaiskaya specimen from the northern Caucasus, and specimens from the Vindija Cave in Croatia (Ovchinnikov et al., 2000). A comparison of mtDNA from modern humans with mtDNA from these specimens revealed that modern humans and Neanderthals did not exchange genes, but rather that they diverged from a common

Figure 3 Geographic distribution of mtDNA haplogroups (Mitomap:http//mitomap.org).

283

From Tools to Symbols

ancestor about 650 Kya. These data added additional support to the Out of Africa model concerning human origins. More recently, Gregory Adcock and colleagues (Adcock et al., 2001) claimed that a specimen from Lake Mungo (LM3) in Australia was older than the most recent common ancestor (MRCA) of modern humans. This new evidence was used to argue against an African origin of modern humans. Adcock and colleagues analysed mtDNA from the remains of ten ancient Australians: the LM3 specimen dated to ~60 Kya; three other gracile individuals from Holocene (< 10 Kya) deposits at Willandra Lakes, and six Pleistocene/early Holocene individuals (15 Kya to < 8 Kya) from Kow Swamp that displayed robust morphological features outside the range of modern humans. Although these authors claimed to have adopted the strict recommended guidelines for work on aDNA, and felt confident about the authenticity of their data, their protocols have been widely criticised (Cooper et al., 2001). Reanalysis of their data, in conjunction with other published mtDNA sequences from other modern Australian Aboriginal individuals, did not support the claim that LM3 was older than the MCRA (Cooper et al., 2001). Instead, the LM3 sequence clustered with contemporary mtDNA and ancient Australian mtDNA types and so argued against LM3 diverging prior to the MRCA (Cooper et al., 2001). While aDNA data are a very powerful tool in evolutionary genetic studies, it is not a trivial procedure to obtain DNA from specimens. DNA is a chemical and is susceptible to changes in the environment. Two types of damage usually occur – (1) hydrolytic, which will result in deamination of bases (loss of the exocyclic amino acid) and in depurination and depyrimidination (hydrolysis of the glycosyl bond between the base and the sugar), and (2) oxidative, caused by the direct interaction of ionising radiation with DNA, as well as damage mediated by free radicals created from water molecules by ionising radiation (Höss et al., 1999). Contamination is an additional problem, particularly from exogenous contemporary human sources. Thus, for aDNA studies to be authentic, very strict control in specific environments has to be maintained and the data have to be reproducible (Höss et al., 1999). Thus, in the fifteen years since the first claim was made by Cann et al. (1987), mtDNA data still provide strong evidence for the African origin of modern humans.

Y chromosome DNA studies The Y chromosome is the paternally inherited equivalent to mtDNA. Most of the Y chromosome is non-recombining – that is, it does not undergo crossing-over between itself and the X chromosome during meiosis. Variation is brought about only by mutation. Once thought to lack variation when compared with mtDNA, recent studies have identified a number of useful microsatellite markers (Seielstad et al., 1999), as well

284

Contribution of genetics to the study of human origins

Figure 4 Global distribution of Y chromosome haplogroups based on biallelic markers (Underhill et al., 2001).

285

From Tools to Symbols

as biallelic markers on the non-recombining region of the Y chromosome (Hammer et al., 2001; Underhill et al., 2000; 2001). Such studies enhanced our understanding of Y chromosome variation. Using over 180 single nucleotide polymorphisms (SNPs), Underhill et al. (2001) have shown that the Y chromosome lineages found among contemporary humans could be assigned to ten (I-X) haplogroups (i.e. groups of different Y chromosome haplotypes). The Y chromosome consortium (YCC) have recently introduced a new nomenclature making use of alphabetical letters to refer to the ten haplogroups as follows: I (A), II (B), III (E), IV (D), V (C), VI (FG, H, I, J), VII (O), VIII (KNM, L), IX (R), and X (PQ). The deepest lineage in the tree was found in African populations within haplogroup A. Haplogroups A, B and E were found exclusively among African populations, with the remaining haplogroups found at varying frequencies in nonAfrican populations, and to a lesser extent in Africans (Fig. 4). Thus, Y chromosome data are also consistent with the greater antiquity of Y chromosome lineages in Africa (~150 000 years), and with the Out of Africa theory, in explaining the present-day distribution of Y chromosome lineages.

Autosomal and X chromosome studies Various autosomal DNA markers (e.g. Alu insertion/deletion polymorphisms, microsatellites or short tandem repeats (STRs), RFLPs, haplotypes, the HLA complex and ß-globin gene polymorphisms), have been used to assess the genetic affinities of human populations. Using RFLP analysis, Bowcock et al. (1991) suggested that African and non-African populations diverged from a common ancestor about 100 Kya. Microsatellite data confirmed this observation (Goldstein et al., 1995; Péréz-Lézaun et al., 1997), providing additional support for the Out of Africa theory concerning modern human origins. Haplotype data from several genetic loci (CD4, DM, DRD2, PAH and PLAT) showed that the number of haplotypes at each locus is greatest in Africa, and that in each case only a small subset of these haplotypes were found in non-African populations (Tishkoff et al., 1996; Tishkoff et al., 2000). Alu elements are a group of genetic markers that have been exploited in studies on human evolution, since they are stably inherited and the ancestral state is known (Stoneking et al., 1997). The patterns of variation observed at these loci are also consistent with a population tree rooted in Africa, with an estimated effective time of separation between Africans and non-Africans of about 137 Kya. Some autosomal data do not support the African origin theory concerning the origins of the gene pool of modern humans. Using sequence variation within the ßglobin gene, Harding et al. (1997) estimated that the present-day variation detected at this locus is due to an evolutionary history dating to about 800 Kya. These data

286

Contribution of genetics to the study of human origins

are in agreement with haploid markers (mtDNA and Y chromosome data), since the coalescence times for autosomal markers are expected to be approximately four times greater than the coalescence times for haploid markers. Harding et al. (1997) concluded that variation in the β-globin gene could also be due to contributions to the modern gene pool from Asian sources. The suitability of this locus for such studies has been questioned, since variation within the β-globin region may be influenced by natural selection to endemic malaria (Seielstad et al., 1999). Analysis of genes on the X chromosome has yielded results similar to other DNA marker studies. Using data from two X chromosome loci, the MRCA for modern humans has been estimated to have existed between 178 Kya (Kaessmann et al., 1999) and 200 Kya (Harris & Hey, 1999). X chromosome data, therefore, also supports the ‘Out of Africa’ theory concerning modern human origins.

Primate evolution It was DNA studies (Miyamoto et al., 1987) as well as earlier protein studies (Wilson & Sarich, 1969), comparing humans and great apes, that first challenged the view that the human–chimpanzee divergence had taken place about 30 Mya. There is consensus now that the human lineage diverged from the chimpanzee lineage about 4–6 Mya, from that of the gorillas about 6–8 Mya, and from that of the orang-utans about 12–16 Mya (Chen et al., 2001). Molecular genetic studies have proved that the current taxonomic classification of the great apes established by G.G. Simpson in 1963 is wrong: chimps are not closer to gorillas than to humans. The common chimpanzees and the bonobos (the two species making up the chimp genus, Pan) are much closer to humans than to gorillas. A recent study, comparing nearly 100 human genes with their homologous genes, or counterparts, in other primates, carried out by a group of researchers at Wayne State University, Detroit, Michigan, led by the same Morris Goodman who was responsible for some of the earliest molecular studies on primate evolution, has confirmed this claim (Wildman et al., 2003). The authors suggest that both chimpanzee species ought to be included in the genus Homo, which would then include the subgenera Homo (Homo) and Homo (Pan). Acceptance of such a classification might change the way in which we think about ourselves – and chimpanzees. It might become universally unacceptable to use chimpanzees for research involving invasive experiments. The hunting of the chimpanzee is decimating their numbers at an alarming rate and urgent steps are needed to halt this crime against humanity’s sister sub-genus Homo (Pan) without detracting from protest against crimes against humanity, i.e., against the subgenus Homo (Homo). Genetic studies are surely going to provide information on saltational changes that have occurred in the hominid line subsequent to its divergence from the common

287

From Tools to Symbols

ancestor. Such changes have been referred to as the ‘Great Leap Forward’ by Jared Diamond (1991), which he defines as ‘the stage in human history when innovation and art at last emerged’. Answering the question ‘What tiny change in genes could have had such enormous consequences?’, Diamond suggests that it was the anatomical basis for spoken language, brought about by some modifications of the proto-human vocal tract. Such modifications would give humans finer control, which would facilitate the production of a much greater variety of sounds. The fine modification of muscles may well not be detectable in fossil skulls, but would modify function so that refined communication would result. It would have taken tens of thousands of years to perfect the structure of language with rules about word order, tenses and the development of vocabulary. ‘It was the spoken word that made us free’, according to Diamond, and probably contributed to the eclipsing of Neanderthals and possible Neanderthal/Cro-Magnon hybrids by Homo sapiens. The anatomy of modern humans was evident by 40 Kya, and, intellectually, H. sapiens of that time were very little different from today’s humans – they could be taught to fly a jet airplane. After the ‘Great Leap Forward’, cultural development no longer depended on genetic changes; we were ‘the first species, capable of destroying all life’.

Evolution of language: what can we learn through genetics? It seems unlikely that the study of fossil hominids will reveal the basis for the genetic changes that produced the ‘Great Leap Forward’. However, a breakthrough has come from the study by medical molecular geneticists of a family in England (Fisher et al., 1998; Lai et al., 2001). The family, known as the KE family, shows, across three generations, approximately half of its members affected with a severe speech and language disorder, which is transmitted as an autosomal dominant Mendelian trait. Every aspect of grammar and of language is affected. Those affected have a severe orofacial dyspraxia, and their speech is largely incomprehensible to the naïve listener. A genetic linkage study in this family revealed that a region of chromosome 7 segregates with or is linked with the speech and language disorder. The relevant region contains a gene, named FOXP2 (Forkhead box P2) which encodes a protein belonging to the forkhead class of transcription factors. It is variation in this gene which, it is claimed, is responsible for the severe speech problem in the KE family. Lai et al. (2001) also found that a single unrelated patient suffering from a speech and language disorder strikingly similar to the one occurring in the affected individuals in the KE family (Fisher et al., 1998) has a de novo translocation involving chromosome number 7. The breakpoint disrupts the functioning of FOXP2 gene on the translocated chromosome number 7.

288

Contribution of genetics to the study of human origins

The story of the FOXP2 gene has become even more interesting with the studies of the gene in non-human primates by Enard et al. (2002). These workers showed that the human FOXP2 protein differs in only three amino-acid positions from its orthologue or equivalent gene in the mouse; in a comparison of 1 880 human–rodent gene pairs it is among the most conserved proteins. The FOXP2 proteins of the chimpanzee, gorilla and rhesus macaque are all identical to each other, and show only one difference from the mouse but two differences from the human protein. The orang-utan carries two differences from the mouse and three from humans. What is particularly striking is the observation that, although the FOXP2 protein is highly conserved, two of the three amino-acid differences between humans and mouse occurred on the human lineage after humans separated from the common ancestor with the chimpanzee (Fig. 5). Although the FOXP2 protein is extremely conserved between mammals, it acquired two amino acid changes on the human lineage, at least one of which, it can be predicted, is likely to have functional consequences. Enard et al. (2002) sequenced exon 7 of FOXP2, the exon carrying the mutations for the unique human characteristics, from 44 human chromosomes representing the major continents and found no polymorphism which might have led to a change in amino acid sequence. Another 91 unrelated individuals, mainly of European descent, had the coding region of FOXP2 sequenced and no functional replacements were discovered. The fact that the two amino acid variants specific to humans were found to occur in 226 human chromosomes, suggests that they are fixed in Homo sapiens. Such findings suggest that after the divergence of humans and chimpanzees, some 4–6 Mya, two mutations have occurred along the human line – in exon 7 of the FOXP2 gene.

Figure 5 Silent and replacement substitutions mapped on a phylogeny of primates (Enard et al., 2002).

289

From Tools to Symbols

Enard et al. (2002) make a convincing case that the fixation of the mutation causing the amino acid substitution(s) in humans occurred very recently in human evolution – during the last 200 000 years of human history, at or subsequent to the emergence of modern humans. It is possible that the actual expansion of modern humans was brought about by the relatively sudden appearance of a more proficient spoken language. When the normal functions of both the human and the chimpanzee FOXP2 proteins have been established we might have a clearer idea of the importance of speech in human evolution, and the molecular explanation for this peculiarly human attribute. One of the early studies on the KE family (Vargha-Khadem et al., 1995) showed that a predominant feature of the affected individuals is ‘an impairment of selection and sequencing of fine orofacial movements’. Such an ability is typical of all humans but is not present in the great apes. The mutation seen in the affected members in the KE family adversely affects the individual’s ability to control orofacial movements and thus to acquire proficient spoken language.

Conclusion Questions concerning population affinities and human origins have long held the fascination of palaeoanthropologists, archaeologists, anthropologists, linguists and historians. Despite this intense interest, many questions remain unanswered. Studies of molecular genetic variation offer quantitative insights into the genetic structure and evolutionary history of human populations that are generally unavailable from other lines of evidence. The genomes of living peoples have been shaped by the continuous appearance of new mutations, ancient human migrations, selection by climate and infection for genetic variants that conferred a selective advantage, and mating patterns determined by cultural norms. Thus, in order for us to get a better understanding of our past, we must take a multidisciplinary approach that integrates genetic data with other historical, linguistic, archaeological, anthropological and palaeoanthropological data.

Acknowledgements We wish to thank the National Health Laboratory Service, the South African Medical Research Council, the National Research Foundation and the University of the Witwatersrand for their support in making this research possible.

290

Contribution of genetics to the study of human origins

References Adcock, G.J., Dennis, E.S., Easteal, D., Huttley, G.A., Jermiin, L.S., Peacock, W.J. & Thorne, A. (2001). Mitochondrial DNA sequences in ancient Australians: implications for modern human origins. Proceedings of the National Academy of Science U.S.A. 98, 537–542. Bowcock, A.M., Hebert, J.M., Mountain, J.L., Kidd, J.R., Rogers, J., Kidd, K.K. & Cavalli-Sforza, L.L. (1991). Study of an additional 58 DNA markers in five human populations from four continents. Gene Geography 5, 151–73. Brown, S.J. (2001). Genetic evidence (present-day DNA). http://www.neanderthal-modern.com/ genetic3.htm Cann, R.L., Stoneking, M. & Wilson, A.C. (1987). Mitochondrial DNA and human evolution. Nature 325: 31–36. Chen, F.C., Vallender. E.J., Wang, H., Tzeng, C.S. & Li, W.H. (2001). Genomic divergence between human and chimpanzee estimated from large-scale alignments of genomic sequences. Journal of Heredity 92, 481–489. Cooper, A., Rambaut, A., Macaulay, V., Willerslev, E., Hansen, A.J. & Stringer, C. (2001). Human origins and ancient human DNA. Science 292, 1655–1656. Diamond, J. (1991). The Rise and Fall of the Third Chimpanzee: How Our Animal Heritage Affects the Way We Live. London: Radius Enard, W., Przeworski, M., Fisher, S.E., Lai, C.S.L., Wiebe, V., Kitano, T., Monaco, A.P. & Pääbo, S. (2002). Molecular evolution of FOXP2, a gene involved in speech and language. Nature 418, 869–872. Fisher, S.E., Vargha-Khadem, F., Watkins, K.E., Monaco, A.P. & Pembrey, M.E. (1998). Localisation of a gene implicated in a severe speech and language disorder. Nature Genetics 18, 168–170. Goldstein, D.B., Linares, A.R., Cavalli-Sforza, L.L. & Feldman, M.W. (1995). Genetic absolute dating based on microsatellites and the origin of modern humans. Proceedings of the National Academy of Science U.S.A. 92, 6723–6727. Hammer, M.F., Karafet, T.M., Redd, A.J., Jarjanazi, H., Santachiara-Benerecetti, S., Soodyall, H. & Zegura, S.L. (2001). Hierarchical patterns of global human Y-chromosome diversity. Molecular Biology and Evolution 8, 1189–1203. Harding, R.M., Fullerton, S.M., Griffiths, R.C., Bond, J., Cox, M.J., Schneider, D.S., Moulin, D.S. & Clegg, J.B. (1997). Archaic African and Asian lineages in the genetic ancestry of modern humans. American Journal of Human Genetics 60 772–789. Harris, E.E. & Hey, J. (1999). X chromosome evidence for ancient human histories. Proceedings of the National Academy of Science U.S.A. 96, 3320–3324. Höss, M., Jaruga, P., Zastawny, T.H., Dizadaroulu, M. & Pääbo, S. (1999). DNA damage and DNA sequence retrieval from ancient tissues. Nucleic Acid Research 24, 1304–1307.

291

From Tools to Symbols Ingman, M., Kaessmann, H., Pääbo, S. & Gyllensten, U. (2000). Mitochondrial genome variation and the origins of modern humans. Nature 408, 708–713. Kaessmann, H., Heißig, F., von Haeseler, A. & Pääbo, S. (1999). DNA sequence variation in a non-coding region of low recombination on the human X chromosome. Nature Genetics 22, 78–81. Krings, M., Stone, A., Schmitz, R.W., Krainitzki, H., Stoneking, M. & Pääbo, S. (1997). Neanderthal DNA sequences and the origin of modern humans. Cell 90, 19–30. Krings, M., Geisert, H., Schmitz, R.W., Krainitzki, H. & Pääbo, S. (1999). DNA sequence of the mitochondrial hypervariable region II from the Neanderthal type specimen. Proceedings of the National Academy of Science U.S.A. 96, 5581–5585. Lai, C.S.L., Fisher, S.E., Hurst, J.A., Vargha-Khadem, F. & Monaco, A.P. (2001). A forkheaddomain gene is mutated in a severe speech and language disorder. Nature 413, 519–523 Miyamoto, M.M., Slightom, J.L. & Goodman, M. (1987). Phylogenetic relations of humans and African apes from DNA sequences in the psi eta-globin region. Science 238, 369–373. Ovchinnikov, I.V., Götherström, A., Romanova, P., Kharitonov, V.M., Lidén, K. & Goodwin, W. (2000). Molecular analysis of Neanderthal DNA from the northern Caucasus. Nature 404, 490–493. Péréz-Lézaun, A., Calafell, F., Mateu, E., Comas, D., Bosch, E. & Bertranpetit, J. (1997). Allele frequencies for 20 microsatellites in a worldwide population survey. Human Heredity 47, 189–196. Seielstad, M., Bekele, E., Ibrahim, M., Touré, A. & Touré, M. (1999). A view of modern human origins from Y chromosome microsatellite variation. Genome Research 9, 558–567. Stoneking, M., Fontius, J.J., Clifford, S.L., Soodyall, H., Arcot, S.S., Saha, N., Jenkins, T., Tahir, M.A., Deininger, P.L. & Batzer, M.A. (1997). Alu insertion polymorphisms and human evolution: evidence for a larger population size in Africa. Genome Research 7, 1061–1071. Tishkoff, S.A., Dietzsch, E., Speed, W., Pakstis, A.J., Kidd, J.R., Cheung, K., Bonné-Tamir, B., Santachiara-Benerecetti, A.S., Moral, P., Krings, M., Pääbo, S., Watson, E., Risch, N., Jenkins, T. & Kidd, K.K. (1996). Global patterns of linkage disequilibrium at the CD4 locus and modern human origins. Science 271, 1380–1387. Tishkoff, S.A., Pakstis, A.J., Stoneking, M., Kidd, J.R., Destro-Bisol, G., Sanjantila, A., Lu, R.B., Deinard, A.S., Sirugo, G., Jenkins, T., Kidd, K.K. & Clark, A.G. (2000). Short tandem-repeat polymorphism/Alu haplotype variation at the PLAT locus: implications for modern human origins. American Journal of Human Genetics 67, 901–925. Underhill, P.A., Shen, P., Lin, A.A., Jin, L., Passarino, G., Yang, W.H., Kauffman, E., Bonné-Tamir, B., Bertranpetit, J., Francalacci, P., Ibrahim, M., Jenkins, T., Kidd, J.R., Mehdi, S.Q., Seielstad, M.T., Wells, R.S., Piazza, A., Davis, R.W., Feldman, M.W., Cavalli-Sforza, L.L. & Oefner, P.J. (2000). Y chromosome sequence variation and the history of human populations. Nature Genetics 26, 358–361.

292

Contribution of genetics to the study of human origins Underhill, P.A., Passarino, G., Lin, A.A., Shen, P., Mirazon Lahr, M., Foley, R.A., Oefner, P.J. & Cavalli-Sforza, L.L. (2001). The phylogeography of Y chromosome binary haplotypes and the origins of modern human populations. Annals of Human Genetics 65, 43–62. Vargha-Khadem, F., Watkins, K., Alcock, K., Fletcher, P. & Passingham, R. (1995). Praxic and nonverbal cognitive deficits in a large family with a genetically transmitted speech and language disorder. Proceedings of the National Academy of Science U.S.A. 92, 930–933. Vigilant, L., Stoneking, M., Harpending, H., Hawkes, K. & Wilson, A.C. (1991). African populations and the evolution of human mitochondrial DNA. Science 253, 1503–1507. Wildman, D.E., Uddin, M., Liu, G., Grossman, L.I.& Goodman, M. (2003). The role of natural selection in shaping 99.4% identity between humans and chimpanzees at nonsynonymous DNA sites: implications for enlarging the genus Homo. Proceedings of the National Academy of Science U.S.A. 96, 5581–5585 Wilson, A.C. & Sarich, V.M. (1969). A molecular time scale for human evolution. Proceedings of the National Academy of Science U.S.A. 63, 1088–1093

293

An overview of the patterns of behavioural change in Africa and Eurasia during the Middle and Late Pleistocene Nicholas J. Conard Institut für Ur- und Frühgeschichte und Archäologie des Mittelalters, Abteilung Ältere Urgeschichte und Quartärökologie, Eberhard-Karls-Universität Tübingen, Schloss Hohentübingen, 72070 Tübingen, Germany

Abstract This paper examines some large-scale patterns of behavioural change that are often viewed as indicators for the advent of cultural modernity and developed symbolic communication. Using examples from Africa and Eurasia, the paper reviews patterns of lithic and organic technology, subsistence and settlement as potential indicators of modern behaviour. These areas of research produce a mosaic picture of advanced technology and behavioural patterns that come and go during the late Middle and Late Pleistocene. Based on these data the emergence of modern behaviour, as seen in the archaeologically visible material record, appears to be gradual and heterogeneous in space and time. The evidence for the use of pigments is consistent with these data. During the early part of the Late Pleistocene personal ornaments in the form of sea shells are documented in south-western Asia and southern Africa. By about 40 thousand years ago (Kya) a diverse array of personal ornaments is documented across the Old World in association with Neanderthals and anatomically modern humans in Europe. These include both modified natural objects and fully formed ornaments. The timing and distribution of the appearance of figurative art and other classes of artefacts including musical instruments point to a more punctuated development of fully modern behaviour during the middle of the Late Pleistocene at approximately 40 Kya. Due perhaps in part to the long and intense history of research much, but by no means all, of the relevant data come from Europe. Early figurative art from the Aurignacian of south-western Germany, northern Italy, Austria and southern France provides undisputed evidence for fully developed symbolic communication and behavioural modernity. This paper also discusses some of the hypotheses for the development and spread of cultural modernity

294

An overview of the patterns of behavioural change in Africa and Eurasia and rejects a strict monogenetic model in favour of a pattern of historically contingent, polygenetic development within a dynamic equilibrium between archaic and modern humans. The paper highlights the need for new refutable, regional and super-regional hypotheses for the advent and spread of behavioural modernity.

Résumé Le présent article examine quelques-uns des aspects du changement comportemental généralement considéré comme indicateur de l’apparition d’une modernité culturelle et d’un système de communication symbolique élaboré. Sur la base d’exemples tirés d’Afrique et d’Eurasie nous passerons en revue la technologie lithique et en matière dure animale, les modes de subsistance et d’habitat comme vecteurs potentiels d’un comportement moderne. Cette révision met en évidence une image en mosaïque avec des technologies et des comportements complexes émergeant et disparaissant à des multiples reprises au cours du Pléistocène moyen tardif et du Pléistocène récent. A partir de ces données, l’apparition d’un comportement moderne, tel qu’il est perçu à travers le matériel archéologique qui nous est parvenu, semble avoir été graduelle et hétérogène dans l’espace et le temps. Les informations disponibles sur l’emploi de pigments au cours de cette période semblent confirmer cette vision. Au début du Pléistocène récent, des éléments de parure sont attestés sous forme de coquillages perforés en Asie du sud-ouest et en Afrique du Sud. A partir d’environ 40 000 ans en chronologie calendaire une grande variété d’objets de parure est documentée en Europe, en association tant avec des Néandertaliens qu’avec des hommes anatomiquement modernes. Ces parures comprennent des éléments naturels perforés et des objets entièrement façonnés. La datation et répartition géographique de l’apparition d’un art figuratif, de même que d’autres classes d’artéfacts, tels que les instruments de musique, indiquent un développement plus ponctuel du comportement moderne au cours de la moitié du Pléistocène récent, il y a environ 40 000 ans en chronologie calendaire. En raison, en partie, d’une longue et intense histoire de la recherche, la majeure partie des données provient d’Europe. L’art figuratif aurignacien du sud-ouest de l’Allemagne, d’Italie du nord, d’Autriche et du sud de la France, constituent autant d’éléments indiscutables d’une communication symbolique élaborée et d’une modernité comportementale pleinement atteinte. Cet article discute également certaines des hypothèses proposées par le passée sur l’émergence et la diffusion de la modernité culturelle et rejette un modèle monogénétique strict en faveur d’un scénario polygénique, soumis aux contingences historiques dictées par un équilibre dynamique entre populations archaïques et modernes Enfin, cet article souligne l’importance d’établir des hypothèses sur l’apparition et l’expansion de la modernité comportementale d’ordre régional et supra-régional qui puissent être vérifiées à travers l’analyse du registre archéologique.

Introduction The question of when in the course of human evolution hominids became like ourselves has been at the centre of several decades of productive debate in palaeoanthropology. Reduced to the most fundamental level, the appearance of

295

From Tools to Symbols

anatomical and behavioural modernity is a question of where and when hominid anatomy and behaviour fall within the variability documented in recent societies. The key component of fully modern cultural behaviour is communication within a symbolically organised world and the ability to manipulate symbols in diverse social contexts. This paper will not address the development of modern human anatomy; here I consider some of the key evidence for the evolution of complex behavioural systems. While there is no consensus about when modern behaviour can first be identified in the archaeological record, by no later than about 40 Kya diverse evidence for the production of ornaments, musical instruments and figurative art provides undisputed evidence for cultural modernity. These and other archaeologically visible indicators of cultural modernity point to a patchy development of complex cultural behaviour and symbolic communication across the Old World. While some regional patterning is becoming visible (Delporte, 1998; McBrearty & Brooks, 2000; Bon, 2002; Conard & Bolus, 2003; Conard, 2005), the current data on this topic are generally a hodgepodge of evidence that has been put through a selective taphonomic filter and a geopolitically defined history of research, and do not allow us to define convincing centres of origin and dispersal for many of the key features considered here. At present we see diverse points of view regarding the origins of behavioural modernity, and current interpretations include but are not limited to the following models: (1) gradual African origin (McBrearty & Brooks, 2000); (2) coastal origin in connection with new dietary patterns during the early Late Pleistocene (Parkington, 2001); (3) punctuated late African origin (Klein, 1999; Klein & Edgar, 2002); (4) gradual origins across multiple human taxa and multiple continents (d’Errico et al., 2003); (5) relatively late origins among multiple human taxa including ‘Neanderthals’ own Upper Palaeolithic revolution’ (Zilhão, 2001: 54). Here I argue for gradual polygenetic origins of behavioural modernity within a dynamic equilibrium between archaic and modern humans. This dynamic equilibrium manifests itself in the form of shifting geographic distributions of anatomically modern and archaic humans, and in the form of occasional contact between these populations. In this model neither Neanderthals nor other archaic Eurasian populations should be viewed as behaviourally or biologically homogeneous. Similarly, the evolution of behavioural modernity in Africa must be viewed as a complex process that affected culturally diverse and physically variable populations. The evolution toward behavioural modernity accelerated in the middle of the Late Pleistocene, and culturally modern behaviour with diverse regional signals and local innovations can be seen in many parts of Africa, Europe, Asia and Australia between 30 and 40 Kya. While archaic and modern humans must have interacted in many regions in the context of diverse social

296

An overview of the patterns of behavioural change in Africa and Eurasia

and ecological settings, ultimately modern humans were at a demographic advantage in all regions and replaced archaic humans with relatively little if any interbreeding. The ultimate demographic collapse of anatomically archaic humans does not necessarily imply that they were universally archaic in their behaviour. Regional records need to establish specific pattern of evolution, since it is becoming increasingly clear that different regions preserve unique palaeoanthropological signatures, and that schematic models of replacement or continuity are already inadequate to address the empirical record. This paper reviews some of the evidence for advanced cultural behaviour and argues for a highly variable pattern of development depending on specific historical and evolutionary contingencies. The development of modern behaviour does not in my view represent a one-time-only quantum leap, but a complex pattern of innovation, spread and local extinction of new traits through cultural selection and social reproduction. Social, technological and linguistic reproduction through learning are fostered by the biological success of the members of societies, but are not only driven by demographic growth. Demographic trends and complex patterns of intra- and inter-societal contacts led to mosaic patterns of cultural development that result from specific historical and ecological occurrences during the Pleistocene. The current archaeological record provides glimpses of these evolutionary processes, but it would be naïve to think that our current data on the fleeting material remains of the development and spread of behavioural modernity provide a one-to-one indication of where and when advanced technology, highly developed patterns of settlement and subsistence, ornaments, music, abstract and figurative representation evolved. The question of why fully modern cultural behaviour evolved is still more difficult to answer, but recent years have begun to see attempts to address the thorny questions of causality (Klein, 1999; Parkington, 2001; Shennan, 2001; Lewis-Williams, 2002; Conard & Bolus, 2003). Much more work is needed that addresses the potential causes of cultural evolution and develops testable hypotheses. In this context monogenetic and polygenetic models need to be formulated and tested explicitly. Turning to the more mundane aspects of archaeology, it is necessary to stress the ambiguities and problems with dating sites in excess of 30 000 years. Radiocarbon dating, the strongest tool for dating Later Stone Age (LSA) and Upper Palaeolithic assemblages, begins to reach its limits in the period before 30 Kya. Here several factors come into play. In this period in excess of five radiocarbon half-lives, contamination becomes a serious problem. The isolation of preserved collagen in bones and similar problems related to sample preparation become more problematic than in younger periods. Also the physics of accelerator mass spectrometry (AMS) and beta counting become more challenging as minimal contamination begins to affect the results more

297

From Tools to Symbols

strongly and the uncertainties related to the chemistry and instrument background become significant. Equally important is the wealth of evidence that there are major fluctuations in radiocarbon levels, probably in connection with variations in production due to magnetic excursions (Voelker et al., 2000; Beck et al., 2001; Conard & Bolus, 2003; Hughen et al., 2004). These factors tend to make radiocarbon measurements underestimate the calendar age of archaeological materials in excess of 30 000 years. In this paper the abbreviation Kya refers to calendar years ago, while ages in radiocarbon years are explicitly referred to as such. Other methods, including luminescence dating, have great potential for sorting out the chronology of the emergence of modern human anatomy and behaviour, but have yet to find broad application in addressing these issues. Fortunately, this situation is improving rapidly and the prospects for gaining improved chronological control for the later stages of human evolution are excellent (Richter et al., 2000; Jacobs, Wintle & Duller, 2003; Jacobs, Duller & Wintle, 2003) This presentation will of necessity be brief and in no way attempts to be encyclopaedic. Instead, this paper considers examples to illustrate the overall pattern of behavioural evolution. These examples are often drawn from regions where I have worked and know the data best. The subject matter is divided into two main sections. The first deals with the nuts and bolts of Palaeolithic archaeology and focuses on lithic and organic artefacts, and patterns of subsistence and settlement. The second section deals more with data that provide more direct access to Palaeolithic world of symbols, beliefs and communication, and reviews dates for burials, ornament, figurative and non-figurative representation and music as means of defining modern cultural patterns. In general, the results from a review of the latter kinds of evidence give a better indication of the origins of behavioural modernity. My concern here is not in developing trait lists or single signatures for modernity, but rather to look at the evolutionary contexts of diverse classes of data that may help us to identify patterns of behavioural evolution. Other similar reviews of this evidence at different geographic scales can be found in a number of recent publications and should be consulted along with the primary references for further details (Klein, 1999; Deacon & Deacon, 1999; McBrearty & Brooks, 2000; d’Errico, 2003; d’Errico et al., 2003; Conard, 2004a). Finally, many of the papers in this volume present up-to-date information that is of central importance for defining the evolution of modern behaviour.

Nuts and bolts approaches to defining modernity Lithic technology Lithic artefacts provide the most physically robust class of artefacts and are in many settings able to survive the numerous potential forms of taphonomic destruction.

298

An overview of the patterns of behavioural change in Africa and Eurasia

In this regard they are the ideal source of data on early human behaviour. In many Palaeolithic settings stone artefacts are the most abundant class of anthropogenically altered material. These attributes of lithic artefacts make them the most important means of defining Palaeolithic cultural groups. Thus if specific lithic artefacts provided an indication of modern cultural behaviour, scholars could use such finds as indicators of modernity. Despite attempts to define linear or even cladistic systems for the evolution of stone tools (Foley, 1987; Foley & Lahr, 2003), lithic technology is based on learned behaviour and is not directly transmitted biologically. Thus it comes as little surprise that new forms of lithic technology come and go over the more than two million year old Palaeolithic record. Oldowan technology is the most common form of flint knapping at the pyramids of Giza (Conard, 2000), and this simplest of knapping approaches comes and goes throughout the Stone Age. Many other knapping technologies also come and go over the last several hundred thousands of years that are the backdrop for the development of anatomical and cultural modernity. It is also important to remember that ethnographically documented hunter-gatherers using Stone Age technology, despite being undeniably behaviourally modern, would not necessarily leave an archaeologically visible record of lithic artefacts that would distinguish them from archaic humans. Handaxes, Levallois technology and other elements of stone knapping come and go during the Pleistocene and do not provide certain indicators of modernity. Even blades, which were once seen as clear indicators for culturally modern, Upper Palaeolithic and Later Stone Age cultures, have been documented in diverse contexts in Africa, the Near East and Europe (Rust, 1950; McBurney, 1967; Singer & Wymer, 1982; Conard, 1990; Révillion, 1994; McBrearty & Brooks, 2000) (Fig. 1). These blade-based assemblages date to the second half of the Middle Pleistocene and the Late Pleistocene, and include technologies based on Upper Palaeolithic platform cores, non-Levallois and Levallois blade production. Lithic assemblages document a heterogeneous pattern of development with forms coming and going across the Old World. While in Europe there is no doubt a difference between Middle and Upper Palaeolithic assemblages, many forms typically associated with the Upper Palaeolithic appear in earlier periods, and it is becoming increasingly clear that the variability documented by Bordes (1961) in the Middle Palaeolithic of south-western France reflects only a small portion of the overall lithic variability. Many regions of Europe (Bosinski, 1967, 1982; Conard & Fischer, 2000) show a diverse pattern of cultural development that is analogous to that documented in Africa (Clark, 1982, 1988; McBrearty & Brooks, 2000). Also in the Near East, the early Middle Palaeolithic includes lithic assemblages such as Yabrudian and Humallian, and the

299

From Tools to Symbols

Figure 1 Kapthurin Formation, Kenya. Late Middle Pleistocene blades c. 250 000 years old (after McBrearty & Brooks, 2000).

later Middle Palaeolithic is characterised by Levalloisian assemblages that were made by both Neanderthals and anatomically modern humans (Shea, 2003). As Bosinski (1982), Clark (1982, 1988) and others have long pointed out, the Middle Stone Age (MSA) and Middle Palaeolithic are marked by the growth and increased visibility of local traditions. The frequently made suggestion that lithic technology from these periods is static or even boring, is difficult to defend. In many areas where high quality data are available, MSA and Middle Palaeolithic assemblages show considerable diversity. The development of local traditions appears to increase with time in some areas of Africa and Eurasia (Bosinski, 1967; Conard & Fischer, 2000; Wadley, 2001; Jöris, 2002), but these trends are, to a certain extent, a reflection of the improved quality of data that reflect both better chronological control and more numerous assemblages per unit time. Researchers who try to define variability must consider the quality and density of the available data. In general, early periods of the MSA and Middle Palaeolithic have provided less data to address these questions than the later phases of these periods or the LSA or Upper Palaeolithic. Thus it is not surprising that, in general, assemblages from more recent periods document more technological and typological variation than samples from earlier periods. The complexity of Middle Palaeolithic and MSA lithic technology is highly variable, but at times advanced. Hafting and composite tools have been documented directly and indirectly in many regions. In Africa we can consider the standardised backed forms from Howieson’s Poort assemblages to be strong candidates for hafting, as well

300

An overview of the patterns of behavioural change in Africa and Eurasia

Figure 2 Klasies River Mouth, South Africa. Highly standardised lithic artefacts from the Howiesons Poort assemblage (after Singer & Wymer, 1982).

301

From Tools to Symbols

as numerous point assemblages of the Upper Pleistocene and perhaps the Middle Pleistocene (Singer & Wymer, 1982; McBrearty & Brooks, 2000) (Fig. 2). In southwestern Asia, Shea (1988, 1993, 1998) has long argued for hafting based on patterns of damage to artefacts and use wear. Mastic attached to Middle Palaeolithic artefacts at Umm-el-Tlel in central Syria also demonstrate the use of hafting and provide evidence for composite tools (Boëda, Connan & Muhesen, 1998). In Europe a similar pattern is present with small backed artefacts that almost certainly required hafting being recovered at Tönchesberg (Conard, 1992). European chipped stone points would have required hafting as on other continents, and mastic has been recovered, for example, at Middle Palaeolithic sites of Königsaue (Mania & Toepfer, 1973) and Neumark-Nord (Mania et al., 1990; Meller, 2003) in north-eastern Germany. Lithic assemblages of the MSA and Middle Palaeolithic do not provide the evidence needed to define precisely when modern patterns of human behaviour developed. They do, however, clearly show a heterogeneous pattern of technological development and transmission that does not indicate that the beginnings of the LSA and Upper Palaeolithic saw fundamental revolutionary changes in technology across the Old World. This transition saw change and the further development of new technologies, but while more advanced forms of lithic technology came into broader use in the LSA and Upper Palaeolithic, most of these technologies have well-documented precursors in earlier periods. Organic technology The development of organic technology shows a pattern analogous to that of lithic technology. While the LSA and Upper Palaeolithic are defined on the basis of new artefact forms that occur in easily detectable numbers, these organic artefacts have antecedents extending into the ESA and Lower Palaeolithic. Thus the beginnings of the LSA and Upper Palaeolithic reflect legitimate archaeological divisions, but the changes represent a further elaboration and intensification of technologies that in some cases existed earlier. In regard to this question the most important discoveries of the last decade are the finds from Schöningen in northern Germany, where Thieme’s excavations have yielded eight wooden spears and numerous other wooden tools (Thieme, 1997, 1999) (Fig. 3). These tools are of the highest workmanship and lend support to the importance of wooden tools from Clacton-on-Sea (Oakley et al., 1977) and Lehringen (Thieme & Veil, 1985). Unless we postulate that this part of eastern Niedersachsen enjoyed a privileged position in human cultural evolution, we must conclude that organic technology and diverse well-made wooden tool assemblages were a part of daily life of the Lower and presumably Middle Palaeolithic inhabitants of Europe. These sites provide a highly

302

An overview of the patterns of behavioural change in Africa and Eurasia

favourable setting for preservation that cannot be matched in other sedimentary settings, but occasional finds of preserved wood in Africa and the Near East leave room for optimism that future work may uncover comparable wooden artefacts. Much has been made of the development and elaboration of bone, ivory and antler tools in recent years (d’Errico, 2003; d’Errico et al., 2003; Gaudzinski, 1999). MSA assemblages from sites including Apollo 11 (Vogelsang, 1998), Klasies River (Singer & Wymer, 1982), and Blombos (Henshilwood et al., 2001) have produced a wealth of

Figure 3 Schöningen, Germany. Lower Palaeolithic wooden spear and horse bones (photo: N.J. Conard).

303

From Tools to Symbols

Figure 4 Klasies River Mouth, South Africa. Bone artefacts from Middle Stone Age deposits (after Singer and Wymer, 1982).

bone artefacts (Fig. 4). Many examples are sharpened bones and bone splinters. Other bone tools show series of notches or more enigmatic forms. An exceptional case is the elaborately made harpoons from Katanda in D. R. Congo (Brooks et al., 1995); these finds would be remarkable if they are of early Late Pleistocene age. Certainly by the Late Pleistocene simple bone tools were widespread in the MSA. The European Lower Palaeolithic also documents early examples of bone tools including carefully manufactured handaxes (e.g. Segre & Ascenzi, 1984; Gatti, 1993). Similarly, bone tools are well documented at Middle Palaeolithic sites including Salzgitter-Lebenstedt (Gaudzinski, 1999), Große Grotte (Wagner, 1983) and Vogelherd (Riek, 1934). Bone tools are by no means as common or complex as those of the Upper Palaeolithic, but they no doubt existed in Middle Palaeolithic assemblages. Bone tools were clearly used by late Neanderthals in many settings, and they have occasionally been documented in large numbers (d’Errico et al., 2003). These tools tend to be less standardised and less elaborate that the organic tools of the Aurignacian. Here split base points, for example, form marker artefacts for the early Aurignacian over much of Europe (Albrecht, Hahn & Torke, 1972; Hahn, 1977). These standardised forms occur in Europe in significant numbers starting nearly 40 Kya. 304

An overview of the patterns of behavioural change in Africa and Eurasia

Subsistence Patterns of subsistence vary in time and space due to changing environmental conditions and changes in technology combined with changing social and settlement strategies. Although most sites do not contain preserved botanical remains, there is every reason to assume that plants played an important part in the diet of all hominids, just as they do for all ethnographically documented societies (Owen, 2003). The diet of Neanderthals as reflected in stable isotope data indicates a relatively high component of animal resources (Bocherens et al., 1999, 2001), but these results do not preclude the use of plants in the diet, and even in the harshest arctic and desert environments, plants are seasonally available and nutritionally important. This is not the appropriate place to summarise the history of research on this question, but recent decades have seen a shift from assuming that archaic and early modern humans practised fully developed systems of hunting and food sharing to a critical assessment and rejection of the earlier interpretations by many Anglophone colleagues. More recently, many case studies have provided convincing evidence that both later archaic and anatomically modern humans practised systematic hunting of large, medium and small game. These data by no means suggest that patterns of subsistence are homogeneous over whole continents or subcontinents, but the advocates of subsistence forms based on scavenging or ineffective forms of hunting (Binford, 1989; Stiner, 1990, 1994) seem to have overstated the case against the existence of reliable hunting economies within MSA and Middle Palaeolithic societies (Marean & Kim, 1998; Marean & Assefa, 1999). Again in this context the finds from Schöningen are of central importance and have redefined the discourse on Lower Palaeolithic subsistence. Thieme’s (1997, 1999) team recovered eight spears from Schöningen in direct association with the bones of over twenty horses in deposits dating to between 300 Kya and 400 Kya. These discoveries of the mid-1990s brought the more extreme assessment of Lower and Middle Palaeolithic subsistence to an end, and as far as I am aware the implications of these remarkable finds for documenting hunting of large game by archaic hominids and the implications of the recovery of a yew wood spear with the skeleton of an Eemian ages forest elephant at Lehringen have not been questioned in recent years. These finds do not demonstrate that hunting large game was a universal phenomenon in the late Middle and Late Pleistocene, but they do document the existence of wellplanned and successfully executed hunting of large and fast game. More mundane sources of information tend to support this view. Numerous faunal assemblages indicate that late archaic and early modern humans had frequent early access to game. In most settings, the possibility of scavenging cannot be completely excluded, but active hunting is the most parsimonious explanation for the faunal

305

From Tools to Symbols

assemblages at sites including, for example, Salzgitter-Lebenstedt (Gaudzinski & Roebroeks, 2000), Tönchesberg (Conard, 1992) and Wallertheim (Schmidtgen & Wagner, 1929; Gaudzinski, 1995; Conard & Prindiville, 2000). In other contexts in many parts of Eurasia and Africa, similar evidence for the role of mammals in the diet of Middle Palaeolithic and MSA people has begun to accumulate (Gaudzinski, 1996; Marean & Kim, 1998; Marean & Assefa, 1999; Burke, 2000; Bocherens et al., 1991, 2001). Finally, it must be stressed that scavenging fresh carcasses is an attractive economic option in contemporary hunting and gathering societies (O’Connell, Hawkes & Blurton Jones, 1988). Thus there is no reason to stigmatise Palaeolithic scavenging as a pre-modern adaptation. In southern Africa Klein and Parkington have developed new approaches and hypotheses for the development of subsistence practices during the MSA. Parkington (2001) stresses the key role of the exploitation of coastal resources for brain development and the origin of cultural modernity in coastal settings. He has also suggested that similar processes may have driven human evolution in other coastal environments, including the circum-Mediterranean region. Klein (1999) has looked at small game such as tortoises and marine resources as playing an important role in MSA and LSA subsistence. He argues that until c. 50 Kya, hunting was limited to comparatively easily hunted game and that people only started systematically hunting dangerous animals including suids and buffalo in the late MSA and LSA. Klein sees this shift in subsistence as an indication of the rise of cultural modernity in connection with genetic mutations and the appearance of fully developed language. Both Parkington’s and Klein’s hypotheses have been received with considerable scepticism, but both hypotheses present important and entirely welcome refutable models for the rise of cultural modernity. Given the general lack of clearly formulated models that provide a causal explanation for the rise of behavioural modernity, these hypotheses, even if they are later demonstrated to be incorrect, have fostered considerable new research. This is certainly the case of the critical assessment of the early evidence for hunting by Binford and colleagues in the 1980s and 1990s. Much like the other data we have considered thus far, the evidence on subsistence during the Middle and Upper Pleistocene shows a pattern of advanced adaptations at an early date. With the possible exception of Parkington’s model for increased use of marine resources in the Late Pleistocene, the data on subsistence tend to argue against a revolutionary change in economic and social behaviour that defines the appearance of cultural modernity.

306

An overview of the patterns of behavioural change in Africa and Eurasia

Settlement Reconstructing patterns of settlement and the organisation of space is one of the more elusive ways of trying to define modern patterns of behaviour. This relates to the general difficulty of reconstructing settlement dynamics in any period and particular problems associated with Palaeolithic periods, where the amount and quality of data are generally poorer than in later periods. The analysis of Palaeolithic settlement in the contexts of defining modern behavioural forms has two major approaches, an intersite and an intra-site or regional approach. Binford (1998), Wadley (2001) and others have argued that spatial organisation within a find horizon can be used to define cultural modernity. Binford, for example, sees repetitive modular units of hearths and bedding areas in rockshelters as a hallmark of modern spatial organisation. In his view this pattern of spatial organisation is not present before the LSA or Upper Palaeolithic. Wadley sees a marked increase in spatial organisation during the late MSA of Rose Cottage Cave in the Free State of South Africa as a further indication that the final stages of the MSA may reflect the period in which cultural modernity developed. In Europe Kolen (1999) has pointed to the lack of clear evidence for architecture as an indication that neither Lower nor Middle Palaeolithic groups regularly built shelters as centres of social and economic interaction, as are known in many later archaeological periods. Instead archaic humans used what Kolen refers to as ‘nests’ to provide shelter. If correct, this would indicate that settlement dynamics of archaic humans, including Neanderthals, fell outside the range of culturally modern people. Several researchers have questioned this model and suggest that even if clear architectural features other than hearths are generally lacking before the Upper Palaeolithic, late Middle Palaeolithic sites document spatially structured activity areas similar to those one would expect in sites of modern hunters and gatherers (Conard, 2001a; Vaquero et al., 2001; Vaquero, Rando & Chacón, 2004). As with many of the criteria we are considering, it is unclear to what extent taphonomic factors and the quality of data affect our interpretations. Kolen, however, is certainly correct to note that clear evidence for anthropogenic shelters and dwellings is extremely rare prior to the Upper Palaeolithic. At a larger scale of analysis, we see more tantalising, yet largely inconclusive evidence for the use of space and distant resources as indicators of behavioural modernity. Important work by Geneste (1988), Roebroeks, Kolen and Rensink (1988), Floss (1994) and others examines the use of distant raw materials as a source of information on Palaeolithic economic and spatial organisation. Especially in the context of the continental European approaches to the study of patterns of lithic reduction and technology (Hahn, 1988; Geneste, 1988; Boëda, Geneste & Meignen, 1990) much research has been aimed at linking patterns of lithic technology to systems

307

From Tools to Symbols

of mobility and settlement. These and other studies show the nearly universal pattern that more distant raw materials are present at sites in more reduced form than local raw materials. This applies for all Palaeolithic periods. With time, more distant raw material are transported to sites, but there is no point where one could define a pattern of behaviour as modern or non-modern. Also, the ‘provisioning of place’ (Kuhn, 1995) – that is, the movement of quantities of raw material to sites for future use – is documented on sites of both modern and archaic hominids (Conard & Adler, 1997). Examination of the abundance of distant raw materials as a reflection of the size of territories and long-distance economic and social relationships also provided ambiguous results. Middle Palaeolithic assemblages document the use of raw materials from 100 or more kilometres away (Floss, 1994; Feblot-Augustins, 1997). Such longdistance transport of tools and raw materials is still more common in the Upper Palaeolithic, but the difference is more one of degree than of kind, and so far this kind of data has not led researches to devise a means of distinguishing between archaic and modern behaviour forms. These lithic data also suggest mosaic, context-dependent systems of adaptations with considerable variability, rather than a black-and-white world of unilinear evolution, in which quantum leaps between archaic and modern behaviour can be readily identified. Finally, an analysis of site types and links between sites within settlement systems shows considerable diversity in MSA and Middle Palaeolithic systems of settlement, but no easily recognisable criterion for defining behavioural modernity (Conard, 2001b; 2004b). Here, as in other areas, one wonders whether the search for a holy grail of cultural modernity is a productive way of defining a research programme. Scholars continue to struggle to identify the origins of a settlement system that reflects a symbolically mediated landscape inhabited by culturally modern people. Furthermore, if our definition of behavioural modernity includes all ethnographically documented patterns of settlement, we must concede that a nearly endless diversity of adaptations among sub-recent hunters and gatherers are by definition modern even if they would not always be identifiable as such archaeologically.

Beyond technology, subsistence and settlement As the discussion above suggests, identifying clear criteria for behavioural modernity is perhaps more likely in the realms of ideology and symbolic communication than in the nuts and bolts archaeology of chipped stone and faunal remains. Now this paper turns to several lines of argument and sets of data that lie outside the pragmatic economic concerns of day-to-day subsistence.

308

An overview of the patterns of behavioural change in Africa and Eurasia

Burials Most of the more complete human skeletons from before the Middle Palaeolithic and Middle Stone Age appear to be the result of extraordinarily favourable taphonomic contexts. No evidence for deliberate burials exists for the Lower Palaeolithic or ESA. Despite arguments to the contrary by Gargett (1989, 1999) and other colleagues, there are a wealth of Middle Palaeolithic human skeletons that seem to have been buried deliberately (Solecki, 1971; Trinkaus, 1983; Defleur, 1993). Such burials could be motivated by purely practical factors like the need to dispose of undesirable cadavers, but it is more likely that the numerous burials of Neanderthals and anatomically modern humans of the Middle Palaeolithic reflect the deliberate burial of kin and are linked to personal and emotional ties between the living and the dead. Defleur (1993) has summarised much of the evidence for Middle Palaeolithic burials and points to a number of convincing cases in Europe and the Levant. The question of the deliberate inclusion of grave goods and the identification of specific ritual practices is more contentious and difficult to demonstrate beyond doubt. In the Upper Palaeolithic the data are unambiguous, and many burials preserve opulent grave goods that reflect the status of the individuals and the needs of the dead in the afterlife. Examples of burials from Sungir’, Dolní Ve˘stonice, the Grimaldi Caves and other sites suggest that the systems of beliefs in association with death in Upper Palaeolithic societies were much more elaborate than those of Middle Palaeolithic people, whether they were anatomically archaic or modern. These Upper Palaeolithic burials are universally accepted as indicators of cultural modernity. As far as I am aware, aside from somewhat enigmatic cases like the highly fragmented and partially burnt assemblage from Klasies River Mouth in South Africa (Singer & Wymer, 1982; White, 1987), the MSA and early LSA have not produced sufficient data for burials to allow conclusions to be drawn about practices and beliefs in sub-Saharan Africa. Pigments In recent years there have been a number of reports of early occurrences of pigments and discussions of the importance and meaning of the use of pigments (Barham, 1998; McBrearty & Brooks, 2000; d’Errico & Soressi, 2002; Hovers et al., 2003). Based on this work, it has become clear that pigments were used in some MSA contexts during the later Middle Pleistocene and in numerous MSA and Middle Palaeolithic settings of the Late Pleistocene (Watts, 1998). Southern Africa has yielded particularly abundant evidence for the use of pigments during the MSA. Barham’s (1998) work at Twin Rivers in Zambia is a noteworthy example of the presence of many pieces of pigments in Middle Pleistocene contexts, and numerous MSA sites including Klasies River (Singer & Wymer, 1982), Peers Cave, Hollow Rock Shelter (Watts, 2002), Apollo 11 (Vogelsang, 1998),

309

From Tools to Symbols

Blombos (Henshilwood et al., 2001), and Diepkloof (Parkington et al., this volume) have produced much evidence for grinding of pigments. Parkington and colleagues (this volume) have argued that the use of pigments provides additional indications of the advent of behavioural modernity in the MSA, particularly in more coastal settings, where Howieson’s Poort and Still Bay assemblages are concentrated. Watts (1998, 2002) has reviewed the evidence for the use of pigments in the MSA and concludes that they are extremely common at many MSA sites and reflect a widespread ability to structure the world into a symbolically organised whole. Watts rejects the hypothesis that pigments were primarily used for strictly utilitarian purposes, including tanning hides. In the Levant and Europe Hovers, d’Errico, Soressi and colleagues also see strong evidence for the use of ochre at Middle Palaeolithic sites including Qafzeh (Vandermeersch, 1969; Hovers et al., 2003) and Pech de l’Azé (Bordes, 1972; d’Errico & Soressi, 2002). The potential uses of ground ochre include body painting, rock painting, drawing, ritual and medicinal purposes. Although we rarely have reliable information on the specific use of these early occurrences of ochre, they are presumably, at least in some settings, such as in Middle Palaeolithic burials, connected with religious beliefs that speak for a high level of cultural development and a significant degree of symbolic communication (Hovers et al., 2003). As with other potential indicators of advanced cultural attributes discussed above, the use of ochre does not appear to reflect a quantum leap signifying the shift from archaic to modern patterns of behaviour. Both anatomically modern and archaic humans used pigments and presumably attached symbolic meaning to red, black and perhaps other ground mineral pigments. Given the well-documented use of mineral pigments, the use of organic pigments is likely, even if difficult to demonstrate with direct archaeological observations. Decorated objects and non-figurative representation There is a long history of claims for deliberate non-utilitarian modification of objects in Palaeolithic contexts. These include finds from the Lower Palaeolithic, such as incised bones from Bilzingsleben (Mania, 1990; Steguweit, 2003), and many finds from later periods. These objects are often controversial and are usually not accepted as demonstrating cultural modernity. Following other lines of arguments, colleagues have suggested that the perfect symmetry of some handaxes indicates an advanced aesthetic development, but Wynn (1995) and Haidle (2004) argue that handaxes do not necessarily reflect complex, symbolically-based communication or language. Over the course of the Middle Palaeolithic and MSA larger numbers of enigmatic objects have been published, including the cross-incised stone and modified fragment of a mammoth tooth from Tata, Hungary (Vértes, 1964), and the so-called ‘mask’ from La Roche-Cotard (Lorblanchet, 1999). Some researchers have included evidence for

310

An overview of the patterns of behavioural change in Africa and Eurasia

collected fossils or curated natural products as indicators of advanced aesthetic and behavioural patterns (Schäfer, 1996). Particularly in recent years, the MSA has produced a number of incised objects that have been taken as evidence for symbolic communication and a high degree of cultural development. Important examples of these finds include engraved linear and cross-hatched patterns on pieces of ochre from Still Bay deposits at Blombos dating to c. 75 Kya (Henshilwood et al., 2002), and incised pieces of ochre from, for example, Peers Cave. Current excavations at Diepkloof have produced examples of decorated ostrich eggshell and a piece of an ostrich eggshell flask from Howieson’s Poort contexts (Parkington et al., this volume). Similar finds have also been recovered from MSA contexts at sites including Apollo 11 (Vogelsang, 1998). These finds are the result of deliberate manufacture and probably reflect the desire of the craftsperson to convey symbolic content and aesthetically meaningful information to members of his or her social group. There can be little doubt that such carefully produced decorated objects and the non-figurative representations they carry communicated information from the maker to people who used or saw these objects. Deciphering the specific meaning broadcast through these finds is not easy, and few specific explanations for their meaning have been presented. With increasing amounts of carefully executed fieldwork during the MSA, there is reason for optimism that contextual information will help archaeologists to develop hypotheses to explain the meaning of these finds. Some colleagues accept these finds as definitive evidence of cultural modernity with fully developed symbolic communication, modern cognitive abilities including language (Henshilwood et al., 2002; d’Errico et al., 2003), while others are less convinced that these finds demonstrate proof of behavioural modernity. Ornament The manufacture and use of ornaments conveys social information about individual identity and group affiliation (Wiessner, 1983). This potential for assertive individual style or emblemic style reflecting social affiliation within a larger demographic group is an important characteristic of modern behavioural patterns and has been the focus of much recent research (Kölbl & Conard, 2003; Vanhaeren, 2002). The archaeological distribution of ornaments provides a clearer signal than many of the classes of finds considered above. Early evidence for the use of marine shells as ornaments comes from burial contexts from Qafzeh Cave in Israel and dates to about 100 Kya (Bar-Yosef & Vandermeersch, 1993). Younger examples of perforated marine shell ornaments come from Still Bay deposits at Blombos Cave dating to about 75 Kya (Henshilwood et al., 2004). Starting roughly 40 Kya ago, personal ornaments have been documented in many parts of the

311

From Tools to Symbols

Old World. Early ornaments include ostrich eggshell beads from early LSA contexts in Enkapune Ya Muto rockshelter, Kenya, with associated radiocarbon measurements between 30 Kya and 40 Kya (Ambrose, 1998). AMS radiocarbon dates directly on ostrich eggshell beads from deposits representing the transition from the MSA to LSA at Mumba Cave in Tanzania (Fig. 5) (Weiß, 2000; Conard, 2004a) have yielded multiple dates between 29 Kya and 33 Kya and lend support to the early dates from Enkapune Ya Muto. There is every reason to assume that these East African ornaments were made by anatomically and presumably culturally modern people. Excavations at Ksar Akil in Lebanon (Azoury, 1986) and at Üçagizli in the Hatay Province of Turkey (Kuhn, Stiner & Güleç, 1999; Kuhn et al., 2001) have produced rich assemblages of perforated marine shells from Initial Upper Palaeolithic contexts dating to about 40 Kya (Fig. 6). Similar finds have been recovered from other Mediterranean early Upper Palaeolithic contexts, including Riparo Mochi on the Ligurian Coast of Italy (Kuhn & Stiner, 1998; Stiner, 1999).

Figure 5 Üçagizli Cave, Turkey. Perforated marine shell ornaments dating to c. 40 000 radiocarbon years ago (after Kuhn et al., 2001).

Figure 6 Mumba Cave, Tanzania. Ostrich eggshell beads radiocarbon dated between 29 000 and 33 000 radiocarbon years ago (photo: H. Jensen).

312

An overview of the patterns of behavioural change in Africa and Eurasia

Elsewhere in Europe there is considerable evidence for a rapid spread in the use of ornaments with the beginning of the Upper Palaeolithic. Neanderthals apparently created a wide range of perforated and incised ornaments in Châtelperronian contexts such as at Grotte du Renne at Arcy-sur-Cure (Leroi-Gourhan & Leroi-Gourhan, 1964; Baffier, 1999; d’Errico et al., 1998). At more or less the same time, numerous examples of early Aurignacian ornaments have been recovered from several diverse regions including the Swabian sites such as Hohlenstein-Stadel, Geißenklösterle and Hohle Fels (Conard, 2003a) (Fig. 7). In addition to incised and perforated natural forms such as teeth, these artefacts include diverse ornaments made of mammoth ivory. It is noteworthy that many of the oldest forms of ornaments in Europe are not only perforated shells or teeth, but also completely carved, three-dimensional ivory beads, pendants, and figurines for which the maker completely dictated the form of the artefact. Although earlier examples of personal ornament are known, by around 40 Kya examples of ornaments are well documented across much of the Old World. These data are consistent with the hypothesis that modern cultural behaviour spread rapidly between roughly 30 Kya and 50 Kya. Shell beads from Mandu Mandu Creek Rock Shelter in Western Australia dating to more than 30 Kya (Morse 1993) suggest that the use of personal ornaments was indeed widespread at an early date. Although Australia lies outside the scope of this review, the colonisation of Sahul is an event in prehistory that requires crossing the vast open water of Wallacea with rafts or other forms of boats. The best available dates for the colonisation lie in the range of c. 42– 45 Kya and fit with the pattern suggesting the rapid spread of advanced behavioural patterns at about this time (O’Connell & Allen, 1998, 2004). Figurative representations The presence of figurative art has been universally accepted as an indication of behavioural modernity. As far as I am aware, no one has disputed that figurative representations are a hallmark of modern cultural behaviour. Mann (2003) has gone so far as to argue that representational art is the ‘gold standard’ by which behavioural modernity can be identified and measured. In Africa the earliest figurative art is from the late MSA of Apollo 11, dating between 25 500 and 27 500 radiocarbon years ago (Vogelsang, 1998). These examples of painted mobile art depict a number of animals, geometric forms and a possible therianthrope (Fig. 8). The Middle Pleistocene-aged, anthropomorphic-shaped stone from Tan Tan, Morocco (Bednarik, 2003), much like a similar object from Berekhat Ram, Israel (Goren-Inbar, 1986; Goren-Inbar & Peltz, 1995; d’Errico & Nowell, 2000), appears to be a modified natural form rather than deliberately carved figurine. In the Levant there is little or no evidence of figurative art before 30 Kya.

313

From Tools to Symbols

Figure 7 Sirgenstein, Bockstein Cave, Sirgenstein, Hohlenstein-Stadel, Vogelherd, Bockstein-törle, Germany. Personal ornaments from the Aurignacian dating to c. 36 000 – 30 000 radiocarbon years ago (after Conard, 2003).

314

An overview of the patterns of behavioural change in Africa and Eurasia

The situation in Europe is very different in that several sites have provided evidence of figurative representation between 30 and 40 Kya. The earliest examples of figurative art are the mammoth ivory figurines from four caves in Swabia in south-western Germany (Hahn, 1986; Schmid, 1989; Conard, 2003b; Conard & Bolus, 2003) and several red monochrome paintings from Fumane in northern Italy (Broglio, 2002). The Swabian Caves of Vogelherd, Hohlenstein-Stadel, Geißenklösterle and Hohle Fels have produced about 20, mostly very small, ivory figurines and figurative representations in bone and stone dating well in excess of 30 000 radiocarbon years. Due to the noisy radiocarbon signal in this period and above-average 14C production, the radiocarbon ages at the Swabian Caves and the similarly aged deposits from Fumane significantly underestimate the age of these artworks. The Swabian ivory figurines include depictions of lions, mammoths, horses, bison, bears, a water bird and two or perhaps three therianthropes that combine features of lions and humans (Hahn, 1986; Conard, 2003b) (Fig. 9 and Fig. 10). These

Figure 8 Apollo 11 Cave, Namibia. Figurative painting from Middle Stone Age deposits dated with radiocarbon to c. 27 Kya.

315

From Tools to Symbols

artworks are small and beautifully carved. They stand in sharp stylistic contrast to the highly schematic paintings of animals, unknown forms and a possible therianthrope from Fumane (Broglio, 2002). Geißenklösterle has also produced a painted rock from this period that preserves traces of red, yellow and black pigments (Hahn, 1986). Most of the spectacular paintings from Grotte Chauvet in the Ardèche region of southern France appear to slightly postdate the examples of figurative art from Swabia and Fumane (Clottes, 2001) (Fig. 11). Here depictions of animals date back as far as 32 000 radiocarbon years ago. The selection of animals in Chauvet, with an emphasis

Figure 9 Hohle Fels, Germany. Waterfowl made from mammoth ivory, c. 32 000 radiocarbon years old (photo: H. Jensen).

Figure 10 Hohle Fels, Germany. Lion-man made from mammoth ivory, c. 32 000 radiocarbon years old.

316

An overview of the patterns of behavioural change in Africa and Eurasia

on dangerous, strong and large animals, shows similarities to the Aurignacian figurines from Swabia and no stylistic similarities to the simple depictions from Fumane. Other important sites in this context include Stratzing in Lower Austria, where a human figurine of stone dated to between 30 000 and 32 000 radiocarbon years ago has been recovered (Neugebauer-Maresch, 1989), as well as Abri Cellier, La Ferrassie, Abri Blanchard and Abri Castanet in south-western France, which have produced representations of animals and vulvas dating to about 30 000 radiocarbon years ago (Leroi-Gourhan, 1995). These figurative depictions from European contexts are the oldest known worldwide. They all date to the early Upper Palaeolithic and were presumably made by modern humans; however, Neanderthals still occupied parts of Europe at this time, nearly 40 Kya. Noteworthy in this context are the recent dates of the human bones from Vogelherd, which clearly demonstrate that they are of Holocene age. At present there is no concrete evidence for a direct association between modern humans and early figurative art in Swabia. Thus the hypothesis that Neanderthals created the figurative art and other remarkable finds of the early Aurignacian cannot be refuted (Conard, Grootes & Smith, 2004).

Figure 11 Grotte Chauvet, France. Early Upper Palaeolithic parietal art radiocarbon dated to c. 30 000 years ago (after J. Clottes, 2001).

317

From Tools to Symbols

The specific context in which figurative art developed has been the subject of considerable discussion of late and will not be elaborated upon here (Lewis-Williams, 2002; Conard & Bolus, 2003). Regardless of the specific social-cultural mechanisms that led to the development and spread of figurative art, there is a consensus among archaeologists and palaeoanthropologists that the makers of these early artistic traditions were culturally modern people (Churchill & Smith, 2001). While many other advanced behavioural forms have precursors in earlier periods, there is no convincing evidence for figurative depictions prior to the beginnings of the European Upper Palaeolithic. Immediately after the appearance of figurative art in Europe several distinctive regional artistic traditions developed across Europe (Conard, 2005). Music Perhaps because of the long research tradition and favourable taphonomic conditions, the earliest examples of musical instruments have been recovered in early Upper Palaeolithic deposits in Europe (Hahn & Münzel, 1995; d’Errico et al., 2003; Conard et al., 2004). As is the case with figurative representations, evidence for music and musical instruments can be seen as an indication of fully developed cultural forms based on symbolic communication. The assumption in this context is that where there is figurative art and music, there must have been fully developed language, by which Palaeolithic people assigned specific concrete and abstract meaning to words and could efficiently communicate information about the past, present and future. Thus, where there is figurative art and music, there must have been behaviourally modern people. While speech, song, music and dance presumably existed still earlier, the oldest musical instruments known are two bone flutes and one mammoth ivory flute from archaeological horizon II at Geißenklösterle (Hahn & Münzel, 1995; Conard et al., 2004). This deposit has been dated by thermoluminescence to about 37 Kya and to several thousand years younger with radiocarbon (Richter et al., 2000). The better preserved of the bone flutes is made of the radius of a swan (Fig. 12). Reconstructions of the instrument produce a high-pitched but pleasing music. Friedrich Seeberger (2003) of Ulm has recently recorded a CD of music played on a flute made from a swan’s radius with dimensions similar to the one from Geißenklösterle. The bone flutes and the ivory flute can be played without a reed and are clearly flutes rather than a reed- or trumpet-voiced instrument as suggested by d’Errico and colleagues (2003). While Aurignacian musicians may have played very different-sounding music, Seeberger’s playing provides a striking impression of what this early Upper Palaeolithic music may have sounded like. Other sites, most notably Isturitz in the French Pyrenees, have produced additional flutes and indicate that wind instruments were in fairly wide use during the early Upper

318

An overview of the patterns of behavioural change in Africa and Eurasia

Palaeolithic (Buisson, 1990). Of course, there are numerous other less conspicuous forms of percussion and other instruments that could have existed during the early Upper Palaeolithic or still earlier, yet they remain to be identified. Claims for earlier examples of Middle Palaeolithic flutes have generally been met with scepticism in archaeological circles, as was the case with recent claims for a Middle Palaeolithic flute made from a cave bear bone from Divje Babe in Slovenia (Turk, 1997; Albrecht et al., 1999).

Conclusions This overview has touched on some, but by no means all, of the evidence for the development of behavioural modernity. I have mentioned some of the main data sets and lines of reasoning that play a role in the discussions and debates about the origins of modern behaviour. This leads to the question of by what means, where and under what circumstances did behavioural modernity arise and which of the hypotheses for the origins of behavioural modernity lies closest to the mark? The answers to these questions depend on how the evidence is weighed and interpreted. From my point of view there can be no doubt that European Aurignacian societies by roughly 40 Kya had all of the hallmarks of modern behaviour including Mann’s (2003) ‘gold standard’ of figurative art as well as musical instruments. The best evidence for early figurative art and music comes from the caves of the Swabian Jura. While one could argue that some important Upper Palaeolithic artefact forms developed in the Upper Danube basin in the period around the time of the arrival of modern humans, naming this region as the single global centre for the origin of cultural modernity would be a radical and naïve interpretation. The contemporary finds of figurative art from Fumane, and slightly later finds from southern France, indicate that the beginnings of the Upper Palaeolithic reflect a time in which archaic behavioural forms were replaced across the board by behavioural forms that lie within the range

Figure 12 Geißenklösterle, Germany. Aurignacian flute made from the bone of a swan, c. 33 000 radiocarbon and 37 000 thermoluminescence years old (photo H. Jensen).

319

From Tools to Symbols

of modern variability. This transition appears to have begun across much of Europe about 40 Kya when modern humans entered a continent inhabited by Neanderthals. Based on the presence of late Neanderthals in several regions of Europe (Hublin et al., 1995; Smith et al., 1999), it appears that there must have been a period in which both archaic and modern humans coexisted in Europe, and contact between the two forms of people must have occurred. Given the poor chronostratigraphic resolution and lack of human fossil material during this key period between roughly 30 Kya and 40 Kya, it is difficult to specify exactly how long both hominids coexisted in specific regions (Conard & Bolus, 2003). Data from the Levant clearly demonstrate that Neanderthals had more successful adaptations and more biological success in previous interactions between Late Pleistocene anatomically archaic and modern humans, but this time modern humans arrived with better developed cognitive skills or behavioural advantages that led to demographic success relative to the indigenous Neanderthals. Thus in western Eurasia a dynamic equilibrium between Neanderthals and anatomically modern populations existed, in which moderns presumably profited from the knowledge and cultural practices of the archaics and vice versa. There is little reason to postulate a violent, rapid, regional advance of Neanderthals into the Levant replacing indigenous anatomically modern humans in the middle of the Late Pleistocene. Similarly there is no reason to assume that the arrival and spread of modern humans into Europe was either universally rapid or brutal. On the contrary, the transition from the Middle to the Upper Palaeolithic and the infiltration and eventual complete dominance of Homo sapiens sapiens in Eurasia probably took on countless local ecologically and historically dictated variants in which there was considerable give and take between archaic and modern humans. This pattern is reflected by observations from nearly every region that has produced relevant data for this transition. These data show very different archaeological signatures depending on the environmental and social-cultural setting encountered by incoming populations (Conard, 1998; Conard & Bolus, 2003). Evidence from the sites occupied by late Neanderthals indicates that they too manufactured and used ornaments (Baffier, 1999; d’Errico et al., 1998), and as discussed above, there is little that separated the patterns of lithic technology, subsistence and settlement reflected in Middle Palaeolithic assemblages from those of the MSA or early Upper Palaeolithic. Still, some time presumably in the early and middle parts of the Late Pleistocene and certainly no later than 40 Kya, people began producing material cultural remains that allow us to identify behavioural modernity. This pattern of behaviour was carried primarily, but perhaps not exclusively, by anatomically modern humans. Many characteristics of modern behaviour can be found across much of the Old

320

An overview of the patterns of behavioural change in Africa and Eurasia

World, and the distribution of advanced cultural traits is significantly determined by the intensity of research in different regions. The recent trend of important discoveries being made in MSA contexts in Southern Africa will no doubt continue as more work is done. Results from excavations at Klasies River, Apollo 11, Rose Cottage Cave, Blombos, Sibudu, and Diepkloof clearly show the enormous potential of the subcontinent. Elsewhere, a similar intensification of research would perhaps produce a similar increase in data relevant to the definition of cultural modernity. While western Eurasia also has considerable potential, there is less reason to assume that the archaeological record will be so radically transformed by further work. Instead important gaps will be filled and, presumably, gradually a more complete picture of the highly variable behavioural patterns during the Lower, Middle and Upper Palaeolithic will emerge. With time we will be better able to develop and test new hypotheses for the evolution and spread of cultural modernity. Based on the data presented above, a strict monogenetic model for the evolution of behavioural modernity appears less likely than a pattern of polygenetic development. These data suggest that MSA and Middle Palaeolithic societies generally existed within regionally specialised social groups with variable material culture. Whether anatomically modern or archaic, these people lived at a similar level of technological and cultural development. Perhaps by about 80 Kya or possibly as few as 40–50 Kya full behavioural modernity developed in Africa and in Eurasia. Most anatomically archaic humans appear not to have completely mastered the repertoire of new behaviours including fully developed symbolic communication. If, however, late archaic humans, including Neanderthals, were culturally fully modern, their behavioural patterns still put them at a reproductive and demographic disadvantage in comparison with the anatomically and culturally modern social groups that propagated across the Old World. The extinction of Neanderthals does not necessarily mean that they were not culturally modern, just as historically documented extinction of local groups of Homo sapiens sapiens does not mean that they were not culturally modern. The main characteristic of Homo is that our cultural development can and does vary independent of our biological morphology (Conard, 1990). Thus late anatomically archaic peoples may have been behaviourally modern, just as early anatomically modern humans may well have been behaviourally archaic (Zilhão, 2001). In the coming years archaeologists and palaeoanthropologists need to work to establish high-quality regional databases while building specific local scenarios and hypotheses for the evolution of modern patterns of behaviour (e.g. Hublin et al., 1996; Parkington, 2001; Lewis-Williams, 2002; Conard & Bolus, 2003). As work progresses, researchers should be able to test these hypotheses and better define the diverse regional scenarios to create new models that come closer to reflecting the evolutionary

321

From Tools to Symbols

reality that a nuanced history of our species warrants. This work should proceed using multiple analytical paradigms and shifting scales of analysis (Conard, 2001c). There are certainly multiple approaches to this complex problem, and all contextually informed explanatory models for the rise and spread of cultural modernity are welcome, regardless of whether they originate from the natural sciences, social science or humanities.

Acknowledgements I am grateful to Lucinda Backwell and Francesco d’Errico for inviting me to participate in the symposium ‘From Tools to Symbols’. If the scientific quality of this volume approaches the level of hospitality and good cheer that welcomed the participants in Johannesburg, the volume will be a great success of lasting value. I also thank Michael Bolus for editorial assistance.

References Albrecht, G., Hahn, J. & Torke, W. G. (1972). Merkmalanalyse von Geschoßspitzen des mittleren Jungpleistozäns in Mittel- und Osteuropa. Archaeologica Venatoria 2. Stuttgart: Verlag W. Kohlhammer. Albrecht, G., Holdermann, C.-S., Kerig, T., Lechterbeck, J. & Serangeli, J. (1998). „Flöten” aus Bärenknochen – die frühesten Musikinstrumente? Archäologisches Korrespondenzblatt 28, 1–19. Ambrose, S. H. (1998). Chronology of the Later Stone Age and food production in East Africa. Journal of Archaeological Science 24, 377–392. Azoury, I. (1986). Ksar Akil, Lebanon. A Technological and Typological Analysis of the Transitional and Early Upper Palaeolithic Levels of Ksar Akil and Abu Halka. Vol. I: Levels XXV – XII. With an introduction by C. Bergman and L. Copeland. Oxford: BAR International Series 289. Baffier, D. (1999). Les derniers Néandertaliens. Le Châtelperronien. Paris: La maison des roches. Bar-Yosef, O. & Vandermeersch, B. (1993) Modern humans in the Levant. Scientific American 268, 94–99. Barham, L.S. (1998). Possible early pigment use in south-central Africa. Current Anthropology 39, 703–720. Beck, J.W., Richards, D.A., Edwards, R.L., Silverman, B.W., Smart, P.L., Donahue, D.J., HererraOsterheld, S., Burr, G.S., Calsoyas, L., Jull, A.J.T. & Biddulph, D. (2001). Extremely large variations of atmospheric

C concentration during the last glacial period. Science 292,

14

2453–2458. Besançon, J., Copeland, L., Hours, F., Muhesen, S. & Sanlaville, P. (1981). Le Paléoloithique d’el Kowm. Rapport Préliminaire. Paléorient 7, 33–55. Bednarik, R.G. (2003). A figurine from the African Acheulian. Current Anthropology 44, 405–413. 322

An overview of the patterns of behavioural change in Africa and Eurasia Binford, L.R. (1989). Isolating the transition to cultural adaptations: an organizational approach. In (E. Trinkaus, Ed.) The Emergence of Modern Humans. Biocultural Adaptations in the Later Pleistocene, pp. 18–41. Cambridge: Cambridge University Press. Binford, L.R. (1998). Hearth and home: the spatial analysis of ethnographically documented rock shelter occupations as a template for distinguishing between human and hominid use of sheltered space. In (N. J. Conard & F. Wendorf, Eds) Middle Palaeolithic and Middle Stone Age Settlement Systems, pp. 229–239. Proceedings of the XIII Congress of the International Union of Prehistoric and Protohistoric Science, Forli, Italy. Bocherens, H., Billiou, D., Mariotti, A., Patou-Mathis, M., Otte, M., Bonjean, D. & Toussaint, M. (1999). Palaeoenvironmental and palaeodietary implications of isotopic biogeochemistry of last interglacial Neanderthal and mammal bones in Scladina Cave (Belgium). Journal of Archaeological Science 26, 599–607. Bocherens, H., Billiou, D., Mariotti, A., Toussaint, M., Patou-Mathis, M., Bonjean, D. & Otte, M. (2001). New isotopic evidence for dietary habits of Neandertals from Belgium. Journal of Human Evolution 40, 497–505. Boëda, E., Geneste J.-M. & Meignen L. (1990). Identification de chaînes opératoires lithiques du Paléolithique ancien et moyen. Paléo 2, 43–80. Boëda, E, Connan, J. & Muhesen, S. (1998). Bitumen as hafting material on Middle Palaeolithic artifacts from the el-Kowm Basin, Syria. In (T. Akazawa, K. Aoki & O. Bar-Yosef, Eds) Neanderthals and Modern Humans in Western Asia, pp. 181–204. New York: Plenum Press. Bon, F. (2002) L’Aurignacien entre Mer et Océan : Réflexion sur l’unité des phase anciennes de l’Aurignacien dans le sud de la France. Société Préhistorique Française Mémoire XXIX. Bordes, F. (1961). Mousterian cultures in France. Science 134, 803–810. Bordes, F. (1972). A Tale of Two Caves. New York: Harper and Row. Bosinski, G. (1967). Die mittelpaläolithischen Funde im westlichen Mitteleuropa. Fundamenta A/4. Köln/Graz: Böhlau-Verlag. Bosinski, G. (1982). The transition Lower/Middle Palaeolithic in northwestern Germany. In (A. Ronen, Ed.) The Transition from Lower to Middle Palaeolithic and the Origin of Modern Man, pp. 165–175. BAR International Series 151, Oxford. Brooks, A.S., Helgren, D.M., Cramer, J.M., Franklin, A., Hornyak, W., Keating, J., Klein, R. G., Rink, W.J., Schwarcz, H.P., Smith, J.N.L., Steward, K., Todd, N.E., Verniers, J. & Yellen, J.E. (1995). Dating the context of three Middle Stone Age sites with bone points in the upper Semliki Valley, Zaire. Science 268, 548–553. Broglio, A. (2002). Palaeolitico e Mesolitico. In (A. Aspes. Ed.) Preistoria Veronese, pp. 11–56. Verona: Memorie del Museo Civico di Storia naturale di Verona. Buisson, D. (1990). Les flûtes paléolithiques d’Isturitz (Pyrénées atlantiques). Bulletin de la Société Préhistorique Française 87, 420–433.

323

From Tools to Symbols Burke, A. (Ed.) (2000). Hunting in the Middle Palaeolithic. International Journal of Osteoarchaeology 10(5), Special Issue, 281–406. Churchill, S.E. & Smith, F.H. (2001). Makers of the Early Aurignacian of Europe. Yearbook of Physical Anthropology 43, 61–115. Clottes, J. (Ed.) (2001). La Grotte Chauvet. L’art des origines. Paris: Éditions du Seuil. Clark, J.D. (1982). The cultures of the Middle Palaeolithic/Middle Stone Age. In (J.D. Clark, Ed.) The Cambridge History of Africa, Vol. 1: From the Earliest Times to 500 B.C., pp 248–341. Cambridge: Cambridge University Press. Clark, J.D. (1988). The Middle Stone Age of East Africa and the Beginnings of Regional Identity. Journal of World Prehistory 2, 235–305. Conard, N.J. (1990). Laminar lithic assemblages from the last interglacial complex in Northwestern Europe. Journal of Anthropological Research 46, 243, 262. Conard, N.J. (1992). Tönchesberg and its Position in the Palaeolithic Prehistory of Northern Europe. Monographien des Römisch-Germanischen Zentralmuseums Mainz 20. Bonn: Habelt. Conard, N.J. (1998). Comment on: Neanderthal Acculturation in Western Europe? A Critical Review of the Evidence and Its Interpretation by F. d’Errico et al. Current Anthropology 39, S22–S24. Conard, N.J. (2000). Flint artifacts from the 1988/1989 Excavations at Giza. Mitteilungen des Deutschen Archäologischen Instituts, Abteilung Kairo 56, 21–41. Conard, N.J. (2001a). River terraces, volcanic craters and Middle Palaeolithic settlement in the Rhineland. In (N.J. Conard, Ed.) Settlement Dynamics of the Middle Palaeolithic and Middle Stone Age, pp. 221–250. Tübingen Publications in Prehistory. Tübingen: Kerns Verlag. Conard, N.J. (ed.) (2001b). Settlement Dynamics of the Middle Palaeolithic and Middle Stone Age. Tübingen Publications in Prehistory. Tübingen: Kerns Verlag. Conard, N.J. (2001c). The future of archaeology. In (Z. Kobylin´ski, Ed.) Quo vadis archaeologia? Whither European archaeology in the 21st century? Proceedings of the European Science Foundation Exploratory Workshop, Ma˘ dralin near Warsaw, 12–13 October 2001, pp. 106– 117. Warsaw: Institute of Archaeology and Ethnology, Polish Academy of Sciences. Conard, N.J. (2003a). Eiszeitlicher Schmuck auf der Schwäbischen Alb. In (S. Kölbl & N. J. Conard, Eds.) Eiszeitschmuck. Status und Schönheit, pp. 15–49. Museumsheft 6. Blaubeuren: Urgeschichtliches Museum. Conard, N.J. (2003b). Palaeolithic ivory sculptures from southwestern Germany and the origins of figurative art. Nature 426, 830–832. Conard, N.J. (2004a). Die Entstehung der kulturellen Modernität. In (N.J. Conard, Ed.) Woher Kommt der Mensch?, pp. 188–218. Tübingen: Attempto Verlag. Conard, N. J. (Ed.) (2004b). Settlement Dynamics of the Middle Palaeolithic and Middle Stone Age II. Tübingen Publications in Prehistory. Tübingen: Kerns Verlag. Conard, N. J. (2005). Aurignacian art in Swabia and the beginning of figurative representations in Europe, In (A. Broglio & G. Dalmeri, Eds) Pitture paleolitiche nelle Prealpi venete. Grotta

324

An overview of the patterns of behavioural change in Africa and Eurasia di Fumane e Riparo Dalmeri, pp. 66-72. Editors: Memorie Museo Civico di Storia Naturale di Verona, Sez. Scienze dell’Uomo, n. 7, Preistoria Alpina, nr. speciale, 2005. Conard, N.J. & Adler, D.S. (1997). Lithic reduction and hominid behavior in the Middle Palaeolithic of the Rhineland. Journal of Anthropological Research 53, 147–175 Conard, N.J. & Bolus, M. (2003). Radiocarbon dating the appearance of modern humans and timing of cultural innovations in Europe: new results and new challenges. Journal of Human Evolution 44, 331–371. Conard, N. J. & Fischer, B. (2000). Are there recognizable cultural entities in the German Middle Palaeolithic? In (A. Ronen & M. Weinstein-Evron, Eds) Toward Modern Humans: Yabrudian and Micoquian, 400–50 kyears ago, pp. 7–24. BAR International Series 850, Oxford. Conard, N.J. & Prindiville, T.J. (2000). Middle Palaeolithic hunting economies in the Rhineland. International Journal of Osteoarchaeology 10, 286–309. Conard, N.J., Grootes, P.M. & Smith, F.H. (2004). Unexpectedly recent dates for human remains from Vogelherd. Nature 430, 198–201. Conard, N.J., Malina, M., Münzel, S. C. & Seeberger, F. (2004). Eine Mammutelfenbeinflöte aus dem Aurignacien des Geißenklösterle: neue Belege für eine Musikalische Tradition im Frühen Jungpaläolithikum auf der Schwäbischen Alb. Archäologisches Korrespondenzblatt 34, 447–462. Deacon, H.J. & Deacon, J. (1999). Human Beginnings in South Africa: Uncovering the Secrets of the Stone Age. Cape Town: David Philip Publishers. Defleur, A. 1993. Les sépultures moustériennes. Paris: CNRS Éditions. Delporte, H. (1998). Les Aurignaciens – premiers hommes modernes. Paris: La maison des roches. d’Errico, F. (2003). The invisible frontier. A multiple species model for the origin of behavioral modernity. Evolutionary Anthropology 12, 188–202. d’Errico, F. & Nowell, A. (2000). A new look at the Berekhat ram figurine: implications for the origins of symbolism. Cambridge Archaeological Journal 10, 123–167. d’Errico, F. & Soressi, M. (2002). Systematic use of pigment by Pech-de-l’Azé Neandertals: implications for the origin of behavioral modernity. Palaeoanthropology Society Meeting, 19–20 March, Denver. Journal of Human Evolution 42, A13. d’Errico, F., Henshilwood, C., Lawson, G. Vanhaeren, M., Tillier, A.-M., Soressi, M., Bresson, F., Maureille, B., Nowell, A., Lakarra, J., Backwell, L. & Julien, M. (2003). Archaeological evidence for the emergence of language, symbolism, and music – an alternative multidisciplinary perspective. Journal of World Prehistory 17, 1–70. d’Errico, F., Zilhão, J., Julien, M., Baffier, D. & Pelegrin, J. (1998). Neanderthal acculturation in Western Europe? A critical review of the evidence and its interpretation. Current Anthropology 39, S1–S44.

325

From Tools to Symbols Féblot-Augustins, J. (1997). La circulation des matières premières au Paléolithique. ERAUL 75. Liège: Université de Liège. Floss, H. (1994). Rohmaterialversorgung im Paläolithikum des Mittelrheingebietes. Monographien des Römisch-Germanischen Zentralmuseums Mainz 21. Bonn: Habelt. Foley, R. (1987). Hominid species and stone-tool assemblages: how are they related? Antiquity 61, 380–392. Foley, R. & Lahr, M.M. (2003). On stony ground: lithic technology, human evolution and the emergence of culture. Evolutionary Anthropology 12, 109–122. Gargett, R.H. (1989). Grave shortcomings. The evidence for Neandertal burial. Current Anthropology 30, 157–190. Gargett, R. H. (1999). Middle Palaeolithic burial is not a dead issue: The view from Qafzeh, Saint-Césaire, Kebara, Amud and Dederiyeh. Journal of Human Evolution 37, 27–40. Gatti, S. (ed.) (1993). Dives Anagnia – Archeologia nella valle del Sacco. Roma: ‘L’Erma’ di Bretschneider. Gaudzinski, S. (1995). Wisentjäger in Wallertheim. Zur Taphonomie einer mittelpaläolithischen Freilandfundstelle in Rheinhessen. Jahrbuch des Römisch-Germanischen Zentralmuseums Mainz 39/1992 (1995), 245–423. Gaudzinski. S. (1996). On bovid assemblages and the consequences for the knowledge of subsistence patterns in the Middle Palaeolithic. Proceedings of the Prehistoric Society 62, 19–39. Gaudzinski, S. (1999). Middle Palaeolithic bone tools from the open-air site Salzgitter-Lebenstedt (Germany). Journal of Archaeological Science 26, 125–141. Gaudzinski, S. & Roebroeks, W. (2000). Adults only. Reindeer hunting at the Middle Palaeolithic site Salzgitter-Lebenstedt, Northern Germany. Journal of Human Evolution 38, 497–521. Geneste, J.- M. (1988). Les industries de la grotte Vaufrey: technologie du débitage, économie et circulation de la matière première lithique. In (J.-P. Rigaud, Ed.) La grotte Vaufrey à Cénac et Saint- Julien (Dordogne). Paléoenvironnements, chronologie et activités humaines, pp. 441–517. Mémoires de la Société Préhistorique Française 19. Goren-Inbar, N. (1986). A figurine from the Acheulean site of Berekhat Ram. Mitekufat Haeven 19, 7–12. Goren-Inbar, N. & S. Peltz (1995). Additional remarks on the Berekhat Ram figurine. Rock Art Research 12, 131–132. Hahn, J. (1977). Aurignacien, das ältere Jungpaläolithikum in Mittel- und Osteuropa. Köln/Wien: Böhlau-Verlag. Hahn, J. (1986). Kraft und Aggression. Die Botschaft der Eiszeitkunst im Aurignacien Süddeutschlands? Tübingen: Verlag Archaeologica Venatoria. Hahn, J. (1988). Die Geißenklösterle-Höhle im Achtal bei Blaubeuren I. Fundhorizontbildung und Besiedlung im Mittelpaläolithikum und im Aurignacien. Stuttgart: Konrad Theiss Verlag.

326

An overview of the patterns of behavioural change in Africa and Eurasia Hahn J. & Münzel, S. (1995). Knochenflöten aus dem Aurignacien des Geißenklösterle bei Blaubeuren, Alb-Donau-Kreis. Fundbeichte aus. Baden-Württemberg 20, 1–12. Haidle, M. N. (2004). Menschenaffen? Affenmenschen? Menschen! Kognition und Sprache im Altpaläolithikum. In (N.J. Conard, Ed.) Woher Kommt der Mensch?, pp. 69–97. Tübingen: Attempto Verlag. Henshilwood, C.S., Sealy, J.C., Yates, R., Cruz-Uribe, K., Goldberg, P., Grine, F.E., Klein, R.G., Poggenpoel, C., van Nierkerk, K. & Watts, I. (2001). Blombos Cave, Southern Cape South Africa: Preliminary report on the 1992–1999 excavations of the Middle Stone Age levels. Journal of Archaeological Science 28, 421–448. Henshilwood, C.S., d’Errico, F., Yates, R., Jacobs, Z., Tribolo, C., Duller, G.A.T, Mercier, N., Sealy, J.C., Valladas, H., Watts, I. & Wittle, A. (2002). The emergence of modern human behavior: Middle Stone Age engravings from South Africa. Science 295, 1278–1280. Henshilwood, C.S., d’Errico, F., Vanhaeren, M., van Niekerk, K. & Jacobs, Z. (2004). Middle Stone Age shell beads from South Africa. Science 304, 404. Hovers, E., Ilani, S., Bar-Yosef, O. & Vandermeersch, B. (2003). An early case of color symbolism: ochre use by modern humans in Qafzeh Cave. Current Anthropology 44, 491–522. Hublin, J.-J., Barroso Ruiz, C., Medina Lara, P., Fontugne, M. & Reyss, J.-L. (1995). The Mousterian Site of Zafarraya (Andalucia, Spain): dating and implications on the palaeolithic peopling processes of Western Europe, Comptes Rendus de l’ Academie des Sciences Paris 321, sér. IIa, 931–937. Hublin, J.-J., Spoor, F., Braun, M., Sonneveld, F. & Condemi, S. (1996). A late Neanderthal associated with Upper Palaeolilithic artifacts. Nature 381, 224–226. Hughen, K, Lehman, S., Southon, J., Overpeck, J., Marchal, O., Herring, C. & Turnbull, J. (2004). C activity and global carbon cycle changes over the past 50,000 years. Science 303, 202–207.

14

Jacobs, Z., Wintle, A.G. & Duller, G.A.T. (2003). Optical dating of dune sand from Blombos Cave, South Africa: I – multiple grain data. Journal of Human Evolution 44, 599–612. Jacobs, Z., Duller, G.A.T. & Wintle, A.G. (2003). Optical dating of dune sand from Blombos Cave, South Africa: II – single grain data. Journal of Human Evolution 44, 613–625. Jöris, O. (2002). Die aus der Kälte kamen... Von der Kultur Später Neandertaler in Mitteleuropa. Mitteilungen der Gesellschaft für Urgeschichte 11, 5–32. Klein R. G. (1999). The Human Career: Human Biological and Cultural Origins, 2nd Edition. Chicago: University of Chicago Press. Klein, R.G. & Edgar, B. (2002). The Dawn of Human Culture. New York: Nevramont. Kölbl, S. & Conard, N.J. (eds.) (2003). Eiszeitschmuck. Status und Schönheit. Museumsheft 6. Blaubeuren: Urgeschichtliches Museum. Kolen, J. (1999). Hominids without homes: on the nature of Middle Palaeolithic settlement in Europe. In (W. Roebroeks & C. Gamble, Eds) The Middle Palaeolithic Occupation of Europe, pp. 139–175. Leiden: University of Leiden.

327

From Tools to Symbols Kuhn, S. L. (1995). Mousterian Lithic Technology: An Ecological Perspective. Princeton, N.J.: Princeton University Press. Kuhn, S. L. & Stiner, M.C. (1998). The Earliest Aurignacian of Riparo Mochi (Liguria, Italy). Current Anthropology 39, Supplement, S175–S189. Kuhn, S.L., Stiner, M.C. & Güleç, E. (1999). Initial Upper Palaeolithic in south-central Turkey and its regional context: a preliminary report. Antiquity 73, 505–517. Kuhn, S.L., Stiner, M.C., Reese, D.S. & Güleç, E. (2001). Ornaments of the earliest Upper Palaeolithic: New insights from the Levant. Proceedings of the National Academy of Science of the USA 98, 7641–7646. Leroi-Gourhan, A. (1995). Préhistoire de l’art occidental. Paris: Éditions d’art Lucien Mazenod. Leroi-Gourhan, Arl. & Leroi-Gourhan, A. (1964). Chronologie des grottes d’Arcy-sur-Cure (Yonne). Gallia Préhistoire 7, 1–64. Lewis-Williams, D. (2002) The Mind in the Cave. London: Thames and Hudson. Lorblanchet, M. (1999). La naissance de l’art. Genèse de l’art préhistorique dans le monde. Paris: Éditions Errance. Mania, D. (1990). Auf den Spuren der Urmenschen. Die Funde aus der Steinrinne von Bilzingsleben. Berlin: Deutscher Verlag der Wissenschaft. Mania, D. & Toepfer, V. (1973). Königsaue. Gliederung, Ökologie, mittelpaläolithische Funde der letzten Eiszeit. Berlin: Deutscher Verlag der Wissenschaften. Mania, D., Thomae, M, Litt, T. & Weber, T. (1990). Neumark – Gröbern. Beiträge zur Jagd des mittelpaläolithischen Menschen. Berlin: Deutscher Verlag der Wissenschaften. Mann, A. (2003). Paper presented at the international symposium From Tools to Symbols. Johannesburg, 2003. Marean, C.W. & Kim, S.Y. (1998). Mousterian large-mammal remains from Kobeh Cave. Behavioral implications for Neanderthals and early modern humans. With comments. Current Anthropology 39, Supplement, S79–S113 Marean, C.W. & Assefa, Z. (1999). Zooarchaeological evidence for the faunal exploitation behavior of Neandertals and early modern humans. Evolutionary Anthropology 8, 22–37. McBrearty, S. & Brooks, A.S. (2000). The revolution that wasn’t: a new interpretation of the origin of modern human behavior. Journal of Human Evolution 39, 453–563. McBurney, C.B.M. (1967). The Haua Fteah (Cyrenaica) and the Stone Age of the South-East Mediterranean. Cambridge: Cambridge University Press. Meller, H. (ed.) (2003). Geisteskraft. Alt- und Mittelpaläolithikum. Begleithefte zur Daueraussstellung im Lamndesmuseum für Vorgeschichte Halle 1. Halle (Saale): Landesamt für Archäologie – Landesmuseum für Vorgeschichte Sachsen-Anhalt. Morse, K. (1993) Shell beads from Mandu Mandu Creek rock-shelter, Cape Range peninsula, Western Australia, dated before 30,000 b.p. Antiquity 67, 877–883. Neugebauer-Maresch, C. (1989). Zum Neufund einer weiblichen Statutte bei den

328

An overview of the patterns of behavioural change in Africa and Eurasia Rettungsgrabungen an der Aurignacien-Station Stratzing/Krems-Rehberg, Niederöstereich. Germania 67, 551–559. Oakley, K.P., Andrews, P., Keeley, L.H. & Clark, J.D. (1977). A reappraisal of the Clacton spearpoint. Proceedings of the Prehistoric Society 43, 13–30. O’Connell, J.F. & Allen, J. (2004). Dating the colonization of Sahul (Pleistocene Austalia-New Guinea). A review of recent research. Journal of Archaeological Science 31, 835–853. O’Connell, J.F. & Allen, J. (1998). When did humans first arrive in Greater Australia and why is it important to know? Evolutionary Anthropology 6, 132–146 O’Connell, J.F., Hawkes, K. & Blurton Jones, N.G. (1988). Hadza scavenging: Implications for Plio/Pleistocene hominid subsistence. Current Anthropology 29, 356–363. Owen, L.R. (2003). Gender and the division of labor in the European Upper Palaeolithic. Habilitation thesis: University of Tübingen. Parkington, J. (2001). Milestones: the impact of systematic exploitation of marine foods on human evolution. In (P. V. Tobias, M. A. Raath, J. Moggi-Cechi & G. A. Doyle, Eds) Humanity from African Naissance to Coming Millennia, pp. 327–336. Johannesburg: Witwatersrand University Press. Parkington, J., Poggenpoel, C., Rigaud, J.-P. & Texier, P.-J. (this volume). The behavioural context of intentionally marked ostrich eggshells from Diepkloof, western Cape. Révillion, S. (1994). Les industries laminaires du Paléolithique moyen en Europe septentrionale. Publications du Cerp 5. Lille: Université des Sciences et Technologies de Lille. Richter, D., Waiblinger, J., Rink, W.J. & Wagner, G.A. (2000). Thermoluminescence, electron spin resonance and

C-dating of the Late Middle and Early Upper Palaeolithic site of

14

Geißenklösterle Cave in Southern Germany. Journal of Archaeological Science 27, 71–89. Riek, G. (1934). Die Eiszeitjägerstation am Vogelherd im Lonetal I: Die Kulturen. Tübingen: Akademische Buchhandlung Franz F. Heine. Roebroeks, W., Kolen, J. & Rensink, E. (1988). Planning depth, anticipation and the organization of Middle Palaeolithic technology: the ‘archaic natives’ meet Eve’s descendants. Helinium 28, 17–34. Rust, A. (1950). Die Höhlenfunde von Jabrud (Syrien). Neumünster: Karl Wachholz Verlag. Schäfer, J. (1996). Die Wertschätzung außergewöhnlicher Gegenstände (non-utilitarian objects) im Alt- und Mittelpaläolithikum. Ethnographisch-Archäologische Zeitschrift 36, 173–190. Schmid, E. (1989). Die altsteinzeitliche Elfenbeinstatuette aus der Höhle Stadel im Hohlenstein bei Asselfingen, Alb-Donau-Kreis. With contributions by J. Hahn and U. Wolf. Fundberichte aus Baden-Württemberg 14, 33–118. Schmidtgen, O. & Wagner, W. (1929). Eine altpaläolithische Jagdstelle bei Wallertheim in Rheinhessen. Notizblatt des Vereins für Erdkunde und der Hessischen Geologischen Landesanstalt zu Darmstadt für das Jahr 1928, V/11. Darmstadt 1929.

329

From Tools to Symbols Seeberger, F. (2003). Klangwelten der Altsteinzeit. CD edited by Urgeschichtliches Museum Blaubeuren. Segre, A. & Ascenzi, A. (1984). Fontana Ranuccio: Italy’s Earliest Middle Pleistocene Hominid Site. Current Anthropology 25, 230–233. Shea, J.J. (1988). Spear points from the Middle Palaeolithic of the Levant. Journal of Field Archaeology 15, 441–450. Shea, J.J. (1993). Lithic use-wear evidence for hunting in the Levantine Middle Palaeolithic. In (P. C. Anderson, S. Beyries, M. Otte & H. Plisson, Eds) Traces et fonction: les gestes retrouvés. Actes du colloque international de Liège 8–9–10 décembre 1990, pp. 21–30. ERAUL 50. Liège: Université de Liège. Shea, J.J. (1998). Neandertal and Early Modern Human Behavioral Variability. A Regional-Scale Approach to Lithic Evidence for Hunting in the Levantine Mousterian. Current Anthropology 39, Supplement, S45–S78. Shea, J.J. (2003). Neandertals, competition, and the origin of modern human behavior in the Levant. Evolutionary Anthropology 12, 173–187. Shennan, S. (2001). Demography and cultural innovation: a model and its implications for the emergence of modern human culture. Cambridge Archaeological Journal 11, 5–16. Singer, R. & Wymer, J. (1982). The Middle Stone Age at Klasies River Mouth in South Africa. Chicago: University of Chicago Press. Smith, F.H., Trinkaus, E., Pettitt, P.B., Karavanic´, I. & Paunovic´, M. (1999). Direct radiocarbon dates for Vindija G1 and Velika Pec´ina Late Pleistocene hominid remains. Proceedings of the National Academy of Science of the USA 96, 12281–12286. Solecki, R.S. (1971). Shanidar. The First Flower People. New York: Alfred A. Knopf. Steguweit, L. (2003). Gebrauchsspuren an Artefakten der Hominidenfundstelle Bilzingsleben (Thüringen). Tübinger Arbeiten zur Urgeschichte 2. Rahden/Westf.: Verlag Marie Leidorf. Stiner, M.C. (1990). The use of mortality patterns in archaeological studies of hominid predatory adaptations. Journal of Anthropological Archaeology 9, 305–351. Stiner, M.C. (1994). Honor among Thieves: A Zooarchaeological Study of Neandertal Ecology. Princeton: Princeton University Press. Stiner, M.C. (1999). Palaeolithic mollusc exploitation at Riparo Mochi (Balzi Rossi, Italy): food and ornaments from the Aurignacian through Epigravettian. Antiquity 73, 735–754. Thieme, H. (1997). Lower Palaeolithic hunting spears from Germany. Nature 385, 807–810. Thieme, H. (1999). Altpaläolithische Holzgeräte aus Schöningen, Lkr. Helmstedt. Germania 77, 451–487. Thieme, H. & Veil, S. (1985). Neue Untersuchungen zum eemzeitlichen Elefanten-Jagdplatz Lehringen, Ldkr. Verden. Die Kunde N. F. 36, 11–58. Trinkaus, E. (1983). The Shanidar Neandertals. London: Academic Press.

330

An overview of the patterns of behavioural change in Africa and Eurasia Turk, I. (ed.) (1997). Moustérienska „košcˇena pišcˇal“ in druge najdbe iz Divjih bab I v Sloveniji / Mousterian ‘bone flute’ and other finds from Divje babe I cave site in Slovenia, Ljubljana. Opera Instituti Archaeologici Sloveniae 2. Ljubljana: Znanstvenoraziskovalni center SAZU. Vandermeersch, B. (1969). Découverte d’un objet en ocre avec traces d’utilisation dans le Moustérien de Qafzeh (Israel). Bulletin de la Société Préhistorique Française 66, 157–158. Vanhaeren, M. (2002). Les fonctions de la parure au Paléolithique supérieur: de l’individu à l’unité culturelle. Unpubl. PhD thesis: Université de Bordeaux I. Vaquero, M., Chacón, G., Fernández, C., Martínez, K. & Rando, J.M. (2001). Intrasite spatial patterning and transport in the Abric Romaní Middle Palaeolithic site (Capellades, Barcelona, Spain). In (N.J. Conard, Ed.) Settlement Dynamics of the Middle Palaeolithic and Middle Stone Age, pp. 376–392. Tübingen Publications in Prehistory. Tübingen: Kerns Verlag. Vaquero, M., Rando, J.M. & Chacón, G. (2004). Neanderthal spatial behavior and social structure: hearth-related assemblages from the Abric Romaní Middle Palaeolithic Site. In (N.J. Conard, Ed.) Settlement Dynamics of the Middle Palaeolithic and Middle Stone Age II, pp. 573–595. Tübingen Publications in Prehistory. Tübingen: Kerns Verlag. Vértes, L. (ed.) (1964). Tata. Eine mittelpaläolithische Travertin-Siedlung in Ungarn. Budapest: Akademie der Wissenschaften. Voelker, A.H.L., Grootes, P.M., Nadeau, M.-J. & Sarnthein, M. (2000). Radiocarbon levels in the Iceland Sea from 25-53 kyr and their link to the earth’s magnetic field intensity. Radiocarbon 42, 437–452. Vogelsang, R. (1998). Middle Stone Age Fundstellen in Südwest-Namibia. Africa Praehistorica 11. Cologne: Heinrich-Barth-Institute. Wadley, L. (2001). What is cultural modernity? A general view and a South African perspective from Rose Cottage Cave. Cambridge Archaeological Journal 11, 201–221. Wagner, E. (1983): Das Mittelpaläolithikum der Großen Grotte bei Blaubeuren (Württemberg). Stuttgart: Konrad Theiss Verlag. Watts, I. (1998). The origin of symbolic culture: the Middle Stone Age of southern Africa and Khoisan ethnography. Unpubl. Ph.D. thesis, University of London. Watts, I. (2002). Ochre in the Middle Stone Age of southern Africa: ritualised display or hide preservative? South African Archaeological Bulletin 57 (175), 1–14. Weiß, C. (2000). Die Artefakte aus Straußenei der Mumbahöhle, Tansania (Schicht III). Unpubl MA thesis, University of Tübingen. White, T. D. (1987). Cannibals at Klasies: cutmarks on a fragment of human skull joins a growing body of clues to suggest that the Klasies River Caves may hold dark secrets about their early inhabitants. Sagittarius (Magazine of the South African Museum) 6–9. Wiessner, P. (1983). Style and social information in Kalahari San projectile points. American Antiquity 48, 253–276. Wynn, T. (1995). Handaxe enigmas. World Archaeology 27, 10–24.

331

From Tools to Symbols Zilhão, J. (2001). Anatomically Archaic, Behaviorally Modern: The Last Neanderthals and Their Destiny. 23. Kroon-Voordracht. Amsterdam: Joh. Enschedé.

332

From the tropics to the colder climates: contrasting faunal exploitation adaptations of modern humans and Neanderthals Curtis W. Marean Institute of Human Origins, Department of Anthropology, P.O. Box 872402, Arizona State University, Tempe, AZ 85287-2402 USA

Abstract There now seems little doubt that Neanderthals were replaced by modern humans from Africa. The counterintuitive character of this stems from the fact that Neanderthals were a highly successful species specially adapted to these cold temperate and cold environments, but were replaced by a species evolved in the tropics. Explaining this evolutionary event mandates the integration of the ecological conditions for hominin evolution in western Eurasia and tropical Africa wedded to a bio-behavioural perspective that seamlessly joins the evidence for archaeology, physical anthropology, and human biology. Drawing on ecological theory and evidence for physiological and behavioural differences between modern humans and Neanderthals, I construct a model that argues that Neanderthals evolved a bio-behavioural faunal exploitation strategy that was high risk, high cost, high return and was focused on the pursuit of larger mammals than later appearing modern humans. Modern humans evolved in Africa a strategy that was more low risk, low cost, and focused on more consistent returns, overall more generalised, and based on technological flexibility coupled to knowledge transmission through language. Its routes lie in the development of a strategy to cope with the high diversity of plant foods in Africa, and their spatial and temporal variations. Neanderthals and modern humans evolved distinct adaptational paths characterised by distinct faunal exploitation strategies that, when juxtaposed together in initial sympatry after the migration of modern humans into Eurasia, resulted in modern humans usurping the niche space of Neanderthals and forcing them into extinction.

333

From Tools to Symbols

Résumé Il y a désormais peu de doute sur le fait que l’homme de Néandertal a été remplacé en Europe par des populations modernes provenant d’Afrique. Le caractère contre intuitif de ce constat vient du fait que les Néandertaliens étaient une espèce particulièrement adaptée aux milieux tempérés et froids, mais qu’elle a été remplacée par une espèce qui a évolué dans les tropiques. L’explication de cet événement demande que l’on conjugue les conditions écologiques de l’évolution des hominidés en Eurasie occidentale et en Afrique tropicale avec une perspective sur le comportement biologique qui permette de relier progressivement les données de l’archéologie, de l’anthropologie physique et de la biologie humaine. En nous servant de la théorie écologique et des données sur les différences physiologiques et de comportement entre l’homme moderne et l’homme de Néandertal, nous proposons un scénario qui voit l’homme de Néandertal comme une espèce ayant développé une stratégie d’acquisition du gibier qui demandait de grands risques, de forts coûts, qui offrait un grand rendement et qui se centrait sur la poursuite d’animaux de taille plus grande que ceux chassés par les hommes modernes. L’homme moderne a développé en Afrique une stratégie qui demandait moins de risques, dont les coûts étaient moindres, qui impliquait sur un rendement plus régulier et qui se basait sur une flexibilité technologique combinée à la transmission du savoir par le langage. Cette stratégie trouve son origine dans le besoin d’exploiter la grande diversité d’aliments végétaux qu’offre l’Afrique et, à la fois, faire face à leur variation spatiale et temporelle. L’homme de Néandertal et l’homme moderne ont développé des stratégies adaptatives distinctes, caractérisées par des stratégies d’acquisition du gibier différentes. La sympatrie produite par la migration de l’homme moderne en Eurasie et la conséquente juxtaposition de ces deux stratégies a eu pour conséquence la conquête par l’homme moderne de la niche écologique occupée jusqu’alors par de l’homme de Néandertal et la conséquente extinction de ce dernier.

Introduction There are two general patterns in human evolution that culminate with the origin of modern humans. One is the continual reduction in hominin species diversity that follows an initial radiation in the Pliocene, ultimately leading to a single surviving species. A second pattern is Africa as a consistent and potent engine for human evolution, while the rest of the world saw local hominin evolution terminating with replacement from Africa. While there is some disagreement about the ultimate fate of the final cousins to modern humanity, the final result is clear: Neanderthals as a recognisable population are gone, and modern humans, likely from Africa, remain. I believe the final reduction in hominin diversity to one species has one most parsimonious explanation: the lineage of hominins on the path to modern humans moved out of the tropics and, wherever that population went, usurped the fundamental niche space available to hominins, making foraging resources unavailable or too low in productivity for other competing hominins. It seems most parsimonious to argue that the critical elements being usurped were food, though one could argue that there

334

From the tropics to the colder climates

were other important, but less critical resources under competition, such as shelter and raw materials. However, the loss of either of these seems insufficient to result in the extinction of a hominin. Foods edible to hominins are limited, and particularly so where the final extinction event occurred (the cold temperate and cold regions of Eurasia). The somewhat counterintuitive correlate of this suggestion is that modern humans quickly and decisively became better than Neanderthals at procuring food in temperate (south-west Asia), cold temperate and cold (Europe) environments, despite the fact that modern humans evolved in tropical environments and were competing with Neanderthals on their home field. Neanderthals likely had several biological and behavioural advantages by means of their long evolutionary history in these environments. We know they had a physique properly adapted to these conditions, while early modern humans in Europe were tropical in morphotype (Pearson, 2000). The long evolutionary history in these environments likely resulted in Neanderthals having a variety of physiological adaptations to cold climates, as do modern humans in these climates (Frisancho, 1993). There are myriad technological and knowledgebased advantages a resident population has over an invading population. These include knowledge of the behaviour patterns of the local fauna, and the edibility and best exploitation strategies for the local flora. Stating that modern humans out-competed Neanderthals for food may seem like an overly simple conclusion, and while I think it must be correct, it still leaves the most interesting aspects of the question unexplained, and these directly address the problem I identified above: how is it that an animal evolved in the tropics quickly outcompetes for food an animal evolved in temperate and cold environments in its own territory? Here I introduce a model that attempts to explain this by positing separate adaptational paths for Neanderthals and modern humans. These adaptational paths were characterised by distinct faunal exploitation strategies that, when juxtaposed in initial sympatry after the migration of modern humans into Eurasia, resulted in modern humans usurping the niche space of Neanderthals and forcing them into extinction. The model posits the following. Neanderthals evolved in Western Eurasia as their evolutionary core area a bio-behavioural faunal exploitation strategy that was high risk, high cost, high return and more focused on pursuit of larger mammals than laterappearing modern humans. Modern humans evolved in Africa a strategy that was more generalised and based on technological flexibility coupled to knowledge transmission through language. When these populations came into contact, Neanderthals may have attempted to mimic technological advances brought in by modern humans (Mellars, 1995) or were evolving toward complexity (d’Errico et al., 1998; d’Errico, 2003; Zilhão,

335

From Tools to Symbols

2001), but their behavioural flexibility was simply insufficient to adapt this new system of behaviour in its entirety. Because of that, they failed to compete effectively for food and went extinct. The critical part of this model is the manner in which these hominids attained their food, and given that the final quarter of the game was probably played in Upper Pleistocene Western Europe and perhaps Central Asia (which would have been temperate to cold and thus lacking plant foods a significant portion of the year), my focus will be on hunting. However, it is essential to go beyond a simple focus on hunting and place the problem more broadly in a bio-behavioural framework. By this I mean that life strategies are integrations of biological and behavioural adaptation, and thus we need to link the empirical and theoretical knowledge of human biology and behaviour and ground it in environmental contexts to deduce how these differing bio-behavioural strategies may have evolved. To do this effectively, we need a far richer empirical record in Africa (more excavated sites and a better fossil record). However, I think there are some areas where we can look profitably for distinctions, and as I will argue, I think there are some clear ones between Africa and its modern humans and western Eurasia and its Neanderthals. Jelinek (1994) made a similar point in his equally bio-behavioural perspective, and while some of my points overlap with his, our conclusions are distinct and my perspective is somewhat more Africanist.

The evidence for MP/MSA faunal exploitation abilities Much of the recent debate over Neanderthals and early modern human faunal exploitation revolves around what is often termed the ‘effectiveness’ of their faunal exploitation abilities. There are some good reasons for this historical focus. Whether one accepts either a weak or strong version of the replacement models for the origins of modern humans, there seems little doubt that Neanderthals either disappeared or were genetically swamped by an invading population of modern humans. While this may have occurred as a result of direct confrontation, there are other competitionbased scenarios that make equal sense. For example, if modern humans entered Europe and employed a strategy of large mammal hunting that was ‘more effective’, Neanderthals may have suffered a critical depletion in foraging returns, eventually resulting in extinction. Several models exist that describe faunal exploitation during the Middle Stone Age (MSA) and Middle Paleolithic (MP), and these have been reviewed elsewhere (Marean & Assefa, 1999). These models have been the subject of regular debate in the recent literature, and this debate has come a long way in reaching some consensus. Several researchers have argued that Neanderthals relied always (Binford, 1981, 1984, 1985) or sometimes (Stiner, 1991a, b, 1993, 1994) on scavenging large mammal

336

From the tropics to the colder climates

prey. The distinction between obligate and opportunistic scavenging helps define the differences between these arguments. Obligate scavenging is when the predator lacks the ability to effectively kill the prey. Opportunistic scavengers scavenge when they can, but can hunt. The ethnographic record shows us that modern humans equipped with tipped spears have hunted effectively every extant terrestrial mammal, even up to the size of elephants. For this reason, obligate scavenging is clearly behaviourally non-modern, a point made long ago by Binford (1985). If Neanderthals were in some manner limited to scavenging, then we have a potent explanation for their eventual demise. To date researchers have been rather vague about what they mean by ‘hunting effectiveness’. Within the realm of hunting, there are at least three areas where abilities can vary. One is hunting efficiency, which is a measure of the amount of energy gained minus the amount of energy expended (net return). There are several parameters that can cause variation in efficiency. Success rate is the number of times a predator has a positive result for each hunting attempt. Low success rates are likely to be inefficient. However, two predators could have equal rates of success, but if predator A expends on average one hour of effort searching and predator B expends three hours of search for similar gross returns, then A has a substantially more efficient strategy. A second area is killing ability as measured by whether or not a predator can kill particular prey. A third is trauma rate as measured by the amount of trauma you receive per encounter, or perhaps the severity of trauma per unit of energy gained. A predator could be very successful for each pursuit, but if there is a relatively high rate of trauma then fitness is substantially reduced. All of these variables in hunting effectiveness could be significant in the issue of replacement of Neanderthals by modern humans. However, the literature to date has focused on just a sub-set of these, mostly the significance of scavenging and killing ability, and has for the most part ignored the other two. At this point I will review that literature and see where we stand. The competing models Binford’s obligate scavenger model argued that Neanderthals and early modern humans were obligate scavengers of large antelope and cervids (Binford, 1981, 1984, 1985, 1988). His main data sets were Grotte Vaufrey and Combe Grenal from France and Klasies River from South Africa. Binford’s argument was based on several basic types of evidence. First, carnivore tooth marks were argued to be abundant on the bones of the large ungulates. Second, stone tool cut marks were less abundant and generally concentrated on low utility foot bones. Third, and most important, large ungulate skeletal elements were best represented by heads and feet.

337

From Tools to Symbols

The basic argument, and there is good data to support this (Blumenschine, 1986a, 1987), is that the first feeders at a carcass will take the highest ranking parts first and leave their signature in the form of tooth marks or cut and percussion marks on the bones. Thus if we have an archaeological site with only the lowest ranking parts, and carnivore tooth marks are abundant, then the hominids must have been scavengers. Figure 1 summarises the skeletal element data marshalled by Binford. Note the abundance of low-ranked parts, and particularly the dominance of heads. We can project this as a scatterplot, and a hunted assemblage should show a positive relation between element abundance and food value. Binford’s data displays a negative relation, or what is typically called a reverse utility curve, and is clearly consistent with scavenging.

a

b

c

d

e

f

Figure 1 A sample of the skeletal element data used by Binford to argue for scavenging during the MSA and MP. In the bar charts, (a) through (c), very low-utility elements are indicated by open bars, while darkened bars are elements with significant amounts of flesh. Note the abundance of low utility parts. When plotted against utility, (d) through (f), a reverse relationship occurs. ‘Std’ on the lower charts abbreviates ‘Standardised’, and the values are drawn from Metcalfe and Jones (1988).

338

From the tropics to the colder climates

Stiner (1990, 1991a, b, 1993, 1994) built on Binford’s results and developed what Marean and Assefa (1999) have called the opportunistic scavenger-hunter model. This model argues that Neanderthals were opportunistic scavengers who could both hunt and scavenge, but shifted strategies relative to changing contexts. Stiner argues that

a

,

b

, , ,

c

d

Figure 2 A sample of the skeletal element data used by Stiner to argue for scavenging during the MP of Italy. Very low utility elements are indicated by open bars, while darkened bars are elements with significant amounts of flesh. Note the abundance of low-utility parts, and particularly the abundance of head elements, and that some assemblages have head parts and virtually nothing else ((c) and (d), Moscerini). Stiner argued this was due to a focus on these parts by Neanderthals, but it likely results from the heavily biased nature of the assemblages resulting from excavators selecting for retention those elements they considered most taxonomically identifiable.

339

From Tools to Symbols

during times of plant food abundance, Neanderthals switched to a plant food-dominated diet and would only take animals as they opportunistically encountered carcasses. During environmental conditions when plants were less abundant, Neanderthals switched to more active hunting. Her argument is based on two basic patterns. First, the chronologically older collections are dominated by head parts, which she sees as being a scavenger signal, like Binford (Fig. 2). Second, her older samples show a pattern where the animals present tend to be the old individuals, while in the later assemblages they tend to be individuals in the prime of their life. She also notes that a reasonable alternative hypothesis is that Neanderthals evolved into effective hunters over time. It is now clear that both scavenger models are unsupported (Bartram & Marean, 1999; Marean, 1998; Marean & Assefa, 1999; Marean & Kim, 1998; Mussi, 1999), primarily because the evidentiary record underlying the interpreted patterning was compromised by poor excavation procedures and/or selective zooarchaeological methods. All of the faunal assemblages used to argue for scavenging – except one, Grotte Vaufrey – received heavy excavator selection either during excavation or shortly after. The result of that procedure is that only the very best faunal fragments were retained. For example, Mussi (1999) describes alarmingly poor recovery procedures for the Italian sites that comprise Stiner’s scavenged pattern, while Klasies River (Klein, 1976) and Combe Grenal (Chase, 1986) were both sorted after excavation into identifiable and unidentifiable fragments, with the latter being discarded. The problem with this procedure is that when sorting collections field archaeologists traditionally save head parts and more complete articular fragments of the postcrania, discarding shafts and other parts perceived as being less valuable. For reasons described elsewhere (Bartram & Marean, 1999; Grayson & Delpech, 1994; Marean, 1998; Marean & Assefa, 1999; Marean & Frey, 1997; Marean & Kim, 1998), such a procedure inevitably results in a head (heavily selected, Fig. 2c–d) or head and foot (less heavily selected, Fig. 1a–c, 2a–b) pattern of skeletal element abundance. Grotte Vaufrey was not selected, but Binford followed well-worn zooarchaeological method and quantified only the articular pieces, a procedure that forces the same pattern to arise. When shafts are included, these ‘scavenger patterns’ evaporate (Fig. 3 and Fig. 4). Several researchers have argued that MSA (Klein, 1995, 1998, 1999, 2000; Klein & Cruz-Uribe, 1996) and MP (Mellars, 1973, 1996) hunters were less effective hunters than fully modern hunter-gatherers. Both argue that Neanderthals and early modern humans were primarily hunters, but just were not as good at it as modern humans that existed after 40 000 years BP. Klein raises several forms of evidence in support of this model. During the MSA the faunal assemblages have a lower frequency of dangerous animals relative to Later Stone Age (LSA) assemblages from similar habitats. The dangerous animals like pigs and buffalo are rare in the MSA but abundant in the

340

From the tropics to the colder climates

a

b

c

d

e

f

Figure 3 The Kobeh Mousterian faunal assemblage Size 2 bovids projected without shaft fragments included (a, c, e) and with shaft fragments included (b, d, f). In the bar charts (a, b, e, f), very low utility elements are indicated by open bars, while darkened bars are elements with significant amounts of flesh. Note the abundance of low-utility parts in the ‘without shafts’ set, and the reverse utility curve produced by the data. When shafts are included, meaty elements dominate the profile and the reverse utility curve breaks down. ‘Std’ on the lower charts abbreviates ‘Standardised’, and the values are drawn from Metcalfe and Jones (1988).

341

From Tools to Symbols

a

0

0,5

1

1,5

b

2

c

e

0 0,5 1 1,5 2 2,5 3 3,5

d

f

Figure 4 The Die Kelders Cave 1 faunal assemblage Size 4 bovids projected without shaft fragments included (a, c, e) and with shaft fragments included (b, d, f). In the bar charts (a, b, e, f), very low-utility elements are indicated by open bars, while darkened bars are elements with significant amounts of flesh. Note the abundance of low utility parts in the ‘without shafts’ set, and the reverse utility curve produced by the data. When shafts are included, meaty elements dominate the profile and the reverse utility curve breaks down. ‘Std’ on the lower charts abbreviates ‘Standardised’, and the values are drawn from Metcalfe and Jones (1988).

342

From the tropics to the colder climates

LSA. Klein’s argument is that the MSA hunters avoided the dangerous animals and sought out the ones more easily killed, and the dangerous animals that are present are represented by the very young only. Mellars (1973, 1996, 2004) argues that MP faunal assemblages do not show the same specialised hunting patterns of later UP assemblages in that MP have broader diversity while UP assemblages show a pattern of being dominated by a single taxon. The implication is that UP people had developed specialised single-species focus strategies. Grayson and Delpech have challenged this interpretation (Grayson et al., 2001, Grayson & Delpech, 2003), while Stiner (Stiner et al., 2000; Stiner, 2001) has found the opposite pattern in Italy with diversity steadily increasing over time, starting in the MP. Klein’s ‘less-effective’ model is targeting what I have called ‘killing ability’ – MSA people could not hunt the most dangerous prey. There are several problems with this argument (Marean & Assefa, 1999), with its application to both the African MSA and the MP. The African data are explained more parsimoniously as resulting from an expansion of diet breadth associated with increasing population pressure. Furthermore, more recent taphonomically informed analyses suggest that at Die Kelders Cave 1 (DK1) the hunters focused on the largest antelope on the landscape, the eland, and ignored smaller, lower-ranking prey items. This suggests a focused faunal exploitation strategy at DK1 not inconsistent with the abilities of modern hunters (Marean et al., 2000). During a time when most researchers were arguing that the evidence for faunal exploitation in the MSA suggested either scavenging or less effective hunting, Chase (1986, 1988, 1989; Chase et al., 1994) was one of the few to argue that Neanderthals were as effective at hunting as later modern humans. He used much of the same data as Binford but interpreted it differently. More recent analyses have come to similar conclusions. A re-analysis of the Grotte Vaufrey collection concluded that the cut-mark data suggest butchery of fully-fleshed carcasses (Grayson & Delpech, 1994). Milo (1994; 1998) re-examined the Klasies River assemblage and has argued from cut and tooth mark patterns that fully modern hunting is indicated. While I agree with the conclusion, the underlying data have several problems. First, the sample is the same biased sample relied on by Binford, and such biased samples provide unreliable estimates of skeletal element abundance and surface modification (Marean, 1998). Second, the patterning Milo described differs substantially from what we would anticipate. For example, the frequencies of percussion marks reported by Milo range from 3 per cent to 6,4 per cent for long bones, while experimental control assemblages show values between 28 per cent and 50 per cent for long bones (see Marean et al., 2000). As one can see, most authors have tended to evaluate Neanderthal hunting in terms of ‘killing ability’. Binford has been the most extreme, seeing Neanderthals (and early

343

From Tools to Symbols

modern humans in Africa) as capable of hunting only the smallest ungulates and/or the youngest of large antelopes, and thus being ‘obligate scavengers’. Stiner’s position is more moderate, arguing that earlier Neanderthals were ‘opportunistic scavengers’, capable of hunting, but having practised a strategy divergent from the modern Upper Paleolithic in that ‘they [Neanderthals] did not hunt as consistently as Upper Paleolithic foragers’ (Stiner, 1991a: 180). However, since no ethnographically documented huntergatherer ever focused on opportunistic scavenging, her interpretations strongly suggest a less-than-modern ‘killing ability’. Klein and Mellars both accept hunting as the primary faunal exploitation strategy, but Klein sees ‘killing ability’ as markedly less lethal prior to 40 000 years BP, while Mellars sees the lack of a specialised strategy as indicating less effective organisational skills. While Chase, Grayson and Delpech, Marean, and Milo have all rebutted the scavenger and less effective hunter models, their fallback position of fully modern hunting still leaves unexplained the critical specifics (efficiency and trauma rates) of this hunting strategy. For example, Grayson and Delpech (2003) have recently argued for a lack of behaviourally significant changes in zooarchaeological data across the Mousterian and Aurignacian boundary at Grotte XVI. They argue that these data may not be sensitive to some important differences in hunting strategies, such as energetic returns, and I return to this point below. However, first I develop a model that specifies how the foraging strategies of Neanderthals and moderns may have differed, and how one came to replace the other. I do so in two steps, first setting the background by describing the environmental differences, and nutritional consequences, between tropical versus cold temperate and cold environments, and second, linking this to our knowledge of Neanderthal physiology and skeletal biology.

Ecological differences between Western Eurasia and Africa and the nutritional physiological consequences Environmental background The Neanderthal lineage and the mosaic of features that define it appear to have evolved in an accretionary manner beginning as early as 450 000 years BP (Hublin, 1998). By the beginning of isotope stage 5 (127 000 years BP) there is widespread agreement that Neanderthal populations existed throughout western Eurasia. The last appearance of Neanderthals is a point of dispute centering on the interpretation of the radiocarbon dates (for example Conard & Bolus, 2003; Zilhão & d’Errico, 2003), but some time in the middle of isotope stage 3 (perhaps 36 000–33 000 BP) classic Neanderthals are replaced by modern humans in western Eurasia. Climates over the span of Neanderthal evolution were generally far colder than today (Van Andel & Tzedakis, 1996). Stage 6 was long and very cold, and while stage 5e may

344

From the tropics to the colder climates

have been as warm as current conditions, even the less cold sub-stages of 5 (5c and 5a) were colder than current conditions. Stages 5d and 5b were substantially cooler than current conditions and stages 4 through 3 much colder. Locations that are today temperate woodland were steppe and/or tundra during stages 4 and 3, and the colder stages of stage 5 saw the development of coniferous forests that could tolerate colder conditions than are present today. If Neanderthals evolved in stage 6, as is widely thought, then they evolved in conditions that were some of the coldest that Europe has ever experienced. Modern people evolving in Africa faced a very different environment. Most of Africa currently falls within the tropics, while the far northern and southern areas are in subtropical to mild temperate climates (Grove, 1988). While all of Africa experienced temperature drops during glacial periods, they were far less dramatic than those in the temperate and cold latitudes, probably more in the range of 3–5º C (Deacon & Lancaster, 1988) versus as much as 14º C in France (Gentry et al., 2003). With some regional exceptions, most of Africa was drier during glacial periods, and saw dramatic reductions in the coverage of forest and woodland, and vast expansions of grasslands, arid grasslands, and deserts (Hamilton, 1982). Temperate and cold latitudes, unlike tropical environments, have prolonged cold seasons, generally lasting from late fall through early spring (Strahler & Strahler, 1992). This difference has a profound impact on floral and faunal diversity and biomass, and therefore food for evolving populations of hominins. Holding rainfall constant, biomass and net primary production is greater in the tropics and decreases steadily from the temperate to the cold latitudes (Coupland, 1979; De Vos, 1969; Krebs, 1978; Pianka, 1966; Rosenzweig, 1992; Whittaker & Likens, 1973). The result is that tropical African ecosystems have a greater biomass and diversity of plants and animals, and because of the less seasonal nature of the environment, less seasonal punctuation to these resources. Reflecting these differences, modern hunter-gatherers in tropical environments rely more on plants foods for subsistence (Lee, 1968; Kelly, 1995) and harvest far more species of plant foods (Marean, 1997) than hunter-gatherers in temperate and cold environments. Figure 5 shows the differences in diversity of harvested plant foods from a sample of hunter-gatherers spanning these environments (though note that the temperate climate sample is from environments warmer and drier than Pleistocene Western Europe). While tropical environments have seasonal rainfall that results in seasonal shifts in plant food abundance, plant seasonality is far less severe than in temperate and cold latitudes where temperature shifts typically depress those parts of plants (fruits, berries, nuts, and tubers) that are edible to people (Archibold, 1995). Several studies of modern African hunter-gatherers show that collectible plant foods

345

From Tools to Symbols

can be found in abundance all year round even in the most arid of African habitats, mostly as a consequence of the abundance of below-ground tuberous plants that are still harvestable in dry seasons. However, these plant foods typically occur in spatially diverse patches, often in short bursts of productivity, scattered in a complicated way across the seasons (Sept, 1986, 1990, 1992, 1994; Tanaka, 1976; Vincent, 1984, 1985). Faunal resources in the tropics are also more abundant and less seasonally punctuated than in cold temperate and cold environments. Tropical Africa differs significantly from temperate and cold environments by having large numbers of both migratory and residential ungulates. For example, grassland ecosystems in Africa typically have 15–30 species of large herbivores, far exceeding any known ecosystem

Figure 5 The number of edible plant species used by several hunter-gatherer groups that inhabited or inhabit cold, temperate, and tropical grasslands. Data derived from Marean (1997).

346

From the tropics to the colder climates

in cold temperate and cold environments. Most importantly, roughly 25 per cent of these tropical species are non-migratory, and these are available all year round. In contrast, cold temperate and cold environments have far fewer resident species, and thus prey is only available for short periods of time in any one location; albeit often at high densities (see discussion in Marean, 1997). The tropics also have far more species of small animals, particularly reptiles and amphibians, as well as edible insects such as locusts and termites. Consequences for food availability The result of these environmental facts is that early modern humans evolving in Africa had access to animal prey that were more diverse, more abundant, and more predictable temporally and spatially than Neanderthals evolving in the cold temperate and cold environments. Furthermore, early modern humans in Africa had access to plant foods that were more diverse, more abundant, and less seasonal. The result is that Neanderthals in western Eurasia faced a critical problem that was not a key stressor for early modern humans in Africa: from autumn through early spring the majority of calories must come from animal tissues. This has many consequences, two of which I mention briefly. First, the discussion above identifies an interesting possible distinction between the goals of faunal exploitation in the tropics versus the temperate and colder latitudes. Faunal exploitation in the temperate and colder latitudes is primarily about food and provisioning, since carcasses are the primary source for both calories and protein. In distinction, faunal exploitation in the tropics is less clearly about food and provisioning since plant foods are so productive and predictable. Hunting may be undertaken primarily for attaining status (O’Connell et al., 2002). Second, given the significance of fauna to the diet in colder latitudes, we must ask the question whether a scavenging adaptation in these environments is viable. There is only one study that directly and quantitatively addresses the productivity of passive scavenging opportunities for hominins (Blumenschine, 1986b, 1987). That study, conducted in one of the richest tropical grasslands in the world (northern Tanzania), found that scavengeable food available to hominins is low most of the year and only becomes abundant during migrations and dry seasons when animals die from predation and natural causes. Being an aggressive scavenger increases return rates (O’Connell, 1988), but even then returns are limited by availability. Availability is an unknown variable for cold temperate and cold environments. Since these environments have lower large mammal diversity and biomass than tropical systems, it seems unlikely that scavenging could function as an effective foraging strategy, particularly given the fact that plant foods are absent the majority of the year.

347

From Tools to Symbols

The result is that there is almost certainly an ecological barrier to obligate and focused opportunistic hominin scavenging in colder environments, and thus hominins living in these environments must have been able to hunt with sufficient success to support their populations on only faunal resources from autumn through spring. These results, of course, fit perfectly with isotopic data that show that European Neanderthals had high-protein diets (Richards et al., 2001). However, this leads to a series of paradoxes with Neanderthals that cuts across archaeology, hunter-gatherer ethnography, skeletal biology, energetics, and the biology of skin colour.

A series of Neanderthal paradoxes The nutritional problem Carbohydrates from plants are generally the most efficient source of energy, while fats from animals follow a close second. Transforming protein into energy is the least efficient path, and has numerous problems that make it an untenable strategy for the long term (Speth & Spielmann, 1983). During seasonal droughts in plant food availability, huntergatherers must develop a strategy to overcome this energy drought, and there are three that are well documented in the ethnographic literature for modern hunter-gatherers: (1) storing plant foods through the energy drought, (2) replacing carbohydrates with fat as the main source of energy, and (3) trade with others who have access to fats or carbohydrates, such as with the Nunamiut – Taremiut relationship (Spencer, 1959). Strategy 2 is used by virtually all-modern hunter-gatherers in the cold temperate and cold environments, such as the Northern Athapaskans and Inuit, respectively. Neanderthals faced the same nutritional hurdle that modern ethnographically documented cold temperate and cold environment hunters and gatherers face – a long drought in carbohydrate-based food resources. What strategy did Neanderthals use to overcome this energy drought? Storing plant foods for long periods of time from autumn through spring is certainly possible. However, to date there is no Neanderthal record of plant food storage technology of the type needed for this type of strategy (Mellars, 1996). Furthermore, the isotopic data suggest that at least some Neanderthals had a diet low in plant food (Richards et al., 2001). Trading for energy-rich foods seems unsupportable for Neanderthals, for several reasons. First, there is no evidence that Neanderthals anywhere were regular hunters of fat-rich sea mammals, and there is no evidence that this material was traded to more inland locations (Mellars, 1996). Second, such a strategy, if it existed, is only viable for those reasonably close to the coast, and leaves the vast majority of Neanderthal-occupied land without a source of fat. Megafauna, such as mammoths, may have been a key source of fat, and there is some isotopic evidence to suggest that Neaderthals may have been targeting mammoth (Bocherens et al., 2001).

348

From the tropics to the colder climates

Of the three strategies discussed above, it would seem that switching to fats is the most parsimonious explanation. Wild ungulates have very low fat levels in their flesh (Speth & Spielmann, 1983), but they do have significant amounts of fat in their bones in the form of marrow and bone grease trapped in the cancellous portions of bone. All modern human hunter-gatherers living in temperate to cold environments attack these sources of fat in two ways, the first rather easy and the second more labour-intensive. Stripping the bone of flesh and breaking it by hammerstone percussion removes marrow from bones fairly easily. For any animal in size class 3 (red deer size) or lower, this is easy and has significant returns (Binford, 1978; Lupo, 1998; Madrigal & Holt, 2002). Animals at the size 4 (buffalo size) range and above require substantially more effort, but the gross returns are also very high. We know that Neanderthals regularly processed bones for marrow by hammerstone percussion due to the abundance of percussion marks on their faunal assemblages (Marean & Kim, 1998). Rendering grease from cancellous bone is a far more difficult undertaking (Lupo & Schmitt, 1997). Some carnivores, such as hyaenas, have evolved specially integrated anatomical and physiological functional suites to overcome the difficulty of this process. Modern humans have adapted technologically. The ethnographic record for modern human hunter-gatherers inhabiting cold temperate and cold environments shows that all use some type of bone boiling technology to extract this fat from the spongy bone. A-ceramic hunter-gatherers typically use a ‘hot rock’ technology for bone boiling, an indirect approach to boiling water where rocks are brought to a high temperature in a fire, and then are thrown into water to bring it to the boil. An ethnographic survey shows that the lean prey that dominated Neanderthal environments were generally boiled by modern people, not cooked over an open flame, because boiling retains the fat (in the broth) and returns moisture to the meat (Wandsnider, 1997). Given this, Neanderthals should have been using hot rock technology. Hot-rock technology has a clear signature – pavements of fire-cracked rock and large hearths with fire-cracked rock in association. As noted elsewhere (Marean & Assefa, 1999), the MP in Europe lacks evidence of hot-rock technology, while it is clearly evident in the Upper Palaeolithic, suggesting that Neanderthals lacked a technological adaptation utilised by modern humans living in similar environments. This paradox becomes more glaring when we examine Neanderthal anatomy and other behavioural indicators. The anatomical paradox As noted above, Neanderthals in Europe inhabited a range of environments, but the classic Neanderthals almost certainly evolved during the long and cold OIS 6, and spent much of their later history in cold glacial conditions. For this reason, researchers often use Inuit as a control-group of modern humans for comparison to Neanderthals

349

From Tools to Symbols

(e.g. Pearson, 2000). There are interesting similarities and contrasts between Inuit and Neanderthals in behaviour and biology that lead to several bio-behavioural paradoxes. Neanderthals and Inuit both have polar-adapted body forms (Pearson, 2000), but in some traits Neanderthals have a ‘hyperpolar’ body form that exceeds that seen in modern humans (Holliday, 1997). They also differ in a variety of critical anatomical features that have direct implications for diet and foraging strategies. Relative to modern humans, Neanderthals have short limbs, particularly in the distal parts, and appear to have had a large body mass (Ruff, 1994; Trinkaus, 1981, 1983). The morphology of the bony labyrinth of Neanderthals is highly derived, differing markedly from both modern humans and H. erectus, and is more consistent with a locomotor adaptation that was less agile and cursorial (Spoor, 2003), which is consistent with their overall body form. Various authors have noted that Neanderthal muscle insertions are enlarged and prominent relative to modern humans, suggesting a more heavily muscled body (Thoma, 1975; Trinkaus, 1983, 2000; Trinkaus et al., 1998). The Neanderthal right humerus has a polar moment of inertia that exceeds modern humans, (BenItzhak et al., 1988), indicating that their right arm was far more powerful (see also Trinkaus, 2000; Trinkaus et al., 1998). The high rate of trauma and early mortality of Neanderthals have been interpreted to mean that Neanderthals practised a high-contact and dangerous approach to hunting (Berger & Trinkaus, 1995; Trinkaus, 1995), thus putting a premium on retaining massive but expensive muscle. The greater amounts of muscle mass on the Neanderthal body translate into greater caloric burn (Leonard, 2003; Sorensen & Leonard, 2001); thus it seems clear that Neanderthals were an energy-expensive animal in an energy-poor environment, and that is the first paradox. Given the expensive nature of muscle mass, it must have been critical for their survival. While not minimising the physical stress of Inuit existence, the genius behind the Inuit adaptation is high performance technologies linked to sophisticated strategies, not muscle. This allows them to lower muscle mass and activity levels to minimise energetic expenditures, which likely allows them to divert energy to various specialised heat-production physiological responses to cold such as non-shivering thermogenesis, elevated BMR, and elevated peripheral temperatures (Frisancho, 1993; Moran, 1982). The impact of the sophisticated technology and behavioural strategies of the Inuit is reflected in various other aspects of life history and skeletal biology. Despite a hunting strategy targeting large and often dangerous prey, 30–40 per cent of Inuit and Aleut live past the age of forty (Laughlin, 1972) while only 10 per cent of Neanderthals lived past forty (Trinkaus, 1995). Inuit cortical bone is some of the least dense among modern populations (Lazenby, 1997), and they have cortical thinning, high porosity,

350

From the tropics to the colder climates

and secondary osteon growth that exceeds Caucasian control groups (Thompson & Gunness-Hey, 1981). In contrast, European Neanderthals have thick cortices (BenItzhak et al., 1988; Kennedy et al., 1990) that exceed modern humans. It seems highly unlikely that the cortically thin and low density skeleton of Inuit could have sustained the heavily muscled, high-contact life-style indicated among Neanderthals, and thus there appears to be a fundamental divergence in overall adaptation between Inuit and Neanderthals. The nutritional demands of Neanderthal biology make the lack of evidence for hot-rock technology glaring, and create another paradox. Bone grease is critical to the survival of hunter-gatherers living in cold temperate and cold environments, yet Neanderthals do not seem to have possessed this technology, despite their higher energy burn. Along the same lines, environments typical of glacial Europe during the time of the Neanderthals generally require travel over significant distances since the main prey, ungulates, are typically migratory or highly mobile. Modern hunter-gatherers in such environments typically intercept these large concentrations of ungulates and kill for surplus, which is stored for use through the long periods when hunting returns are low (Burch, 1972). These hunts are typically carried out communally, or at least with several hunters, often using natural or modified features of the landscape (what I have called the tactical landscape method, see review in Marean, 1977). Neanderthal sites do not indicate large group organisation of the type typical for these communal hunts, and there is no evidence of meat storage (Mellars, 1996). This would suggest that Neanderthals hunted in a more solitary manner. The lack of evidence for storage is particularly odd, but then they may have been using raised wooden platforms and tree crotches that may not preserve archaeologically. If they were killing very large mammals like mammoth, which seems likely (see below), then some type of storage must have been used. However, if Neanderthals did not practise storage, then it seems unlikely that they could have survived without high-mobility herd-following strategies, since most of the year migratory animals would not be present within a home range. Herd-following strategies are not used by modern humans (Burch, 1972). This integration of the ecological, archaeological, and anatomical data raises several clear bio-behavioural distinctions between Neanderthals and what we expect of a hominin in these environments, and suggests that Neanderthals had an adaptation unlike any modern human in similar environments. Further paradoxes To summarise, there are some key bio-behavioural distinguishing features between Neanderthals and their modern counterparts in similar environments. Neanderthals were heavily muscled high-caloric burn mammals, probably did not practise tactical

351

From Tools to Symbols

landscape strategies to mass kill ungulates, lacked long-term storage, and did not exploit hot-rock bone-boiling technologies. The latter is particularly troubling to me, since I fail to see how a human can live in cold temperate and cold environments without extracting bone grease. My guess then is that they did, but how? Hyaenas and large canids also extract grease from bone, but as noted above their strategy differs from that of modern humans in that they consume greasy cancellous bone and let their digestive systems do the work. Perhaps Neanderthals practised a modified bone consumption strategy. While Neanderthals lacked the bone cracking anatomy of hyaenids, they do have access to hammerstones, and the abundance of hammerstone percussion marks clearly shows that they broke bones for marrow. One possible strategy is that Neanderthals crushed spongy bone into bits, swallowed it, digestively extracted bone grease, and then passed the remains. It would be useful to know the costs of grease rendering through the human gut so that a net return-rate could be calculated, and the viability of the strategy estimated. This brings us back to the differences between Inuit and Neanderthal skeletal biology. As noted above, adult Inuit skeletons have thin bone cortices and low bone density, resulting in high rates of age-related osteoporosis. The most common causal explanation is a high-protein ‘acid-ash’ diet – serum acidosis overcome by bone calcium resorption (Pfeiffer & Lazenby, 1994). More recently, this argument has been questioned with an alternative suggested (Lazenby, 1997): non-shivering thermogenesis creates higher production and utilisation of thyroid hormones T3 and T4, ultimately resulting in bone loss. There is a third, and simple, hypothesis. Inuit live at latitudes where the dosage of UVB radiation is insufficient, all year, to prompt vitamin D synthesis in the skin. Furthermore, Inuit skin colour is darker than expected of populations living in such areas, meaning that even less vitamin D synthesis occurs (Jablonski & Chaplin, 2000). Vitamin D is essential to calcium absorption in the intestines, and thus essential to normal development and maintenance of a healthy skeleton (see review in Frisancho, 1993). There are nutritional paths to vitamin D, including fish and marine mammals, but these are available only on a seasonal basis to the Inuit, and rare in many inland populations. There are few other natural sources, and even the best of those, such as beef liver and eggs, provide less than 10 per cent of RDA for a serving. It is likely that vitamin D deficiency resulting from insufficient UVB is the primary cause of Inuit low bone density and thin cortices. This creates another Neanderthal paradox. European Neanderthals lived at latitudes currently at the border between those where UVB is insufficient all year and those where it is insufficient for one month out of the year (Jablonski & Chaplin, 2000). With the decrease in radiation intensities during glacial periods (Imbrie & Imbrie, 1979), UVB intensity in

352

From the tropics to the colder climates

Europe was probably solidly in the ‘insufficient all year’ range. The MP archaeological record in Europe lacks evidence for regular exploitation of fish and marine mammals (Mellars, 1996; Richards et al., 2001); certainly nothing approaching Inuit or Northern Athapaskan levels, and European Neanderthals ate very high protein diets (Bocherens et al. 2001). Given the explanations for Inuit density and cortical thinness, Neanderthals should have low bone density and thin cortices like the Inuit. Yet they have thick cortices, and the one study of density suggests it is similar to modern humans, though there may be complications related to the fossilisation of the specimen (Senut, 1985). Where did they get their calcium for bone growth, and where did they get their vitamin D to metabolise it? Given the long evolutionary history of Neanderthals in these environments, Neanderthals must have had very light skins (contra typical media representations) to maximise UVB intake, and it seems very likely that they possessed other climatic adaptations (nonshivering thermogenesis, elevated BMR, increased peripheral blood flow, optimised vasoconstriction/dilation, and ones we have not observed among modern people). This leaves unexplained the differences in Inuit and Neanderthal bone density and cortical thickness. Inuit have inhabited these cold northern habitats for several thousand years, and despite that short time, have evolved several genetically-based adaptations to it. Perhaps the long (more than 100 000 years) evolutionary history of Neanderthals in these cold environments resulted in some physiological adaptations among Neanderthals that are not present among modern humans. One source for calcium is the same as that I have suggested for grease extraction: they could have crushed greasy cancellous bone and consumed it. The same process would provide plenty of calcium for bone growth, and would be an added benefit to ingestion of greasy spongy bone, partially off-setting the digestion costs. This hypothesis of bone consumption for calcium intake has several direct test implications: (1) Neanderthal coprolites should show a high occurrence of bone and by-products of digested bone, (2) spongy bone portions should be relatively rare in Neanderthal assemblages because of the crushing, (3) bits and pieces of unconsumed epiphyseal ends of long bones should have hammerstone percussion marks at frequencies that exceed that found in assemblages where long bones have been hammerstone broken only for marrow. I know of no data that address test implication (1), but the Kobeh cave assemblage (Marean & Kim, 1998) shows massive destruction of spongy bone relative to cortical bone. My experience with Zagros and South African assemblages suggests an intensity of bone fragmentation in the Zagros ones that far exceeds that typical in South Africa. I interpreted this to mean that carnivores removed that bone after consumption, but perhaps Neanderthals pulverised it for bone consumption.

353

From Tools to Symbols

A model for divergent adaptational paths As discussed above, modern humans entering Eurasia had several critical disadvantages relative to the resident Neanderthals, and these were in both biological and behavioural realms. Modern humans were dark-skinned in a low UVB environment, while Neanderthals were likely light-skinned, placing early modern humans in danger of vitamin D deficiency resulting in rickets, osteomalacia, osteopenia, and ultimately osteoporosis. The biological response would have been strong selection for lighter skin, while the behavioural response would have been incorporation of fish in the diet to offset the vitamin D deficiency. Modern humans were tall and thin, while Neanderthals were shorter with relatively shorter extremities. The implication is that modern humans had a physique that was less energy-efficient at these upper latitudes than that of Neanderthals. The biological response was selection for a more coldadapted form, and the behavioural response was the development of technologies for producing and retaining heat. Modern humans had no physiological adaptations to cold when they entered Europe, and it is likely that Neanderthals did. The main behavioural disadvantages of modern humans were a lack of knowledge of local plant availability and toxicity, as well as a lack of knowledge of local animals and their behavioural patterns. Modern humans lacked a technology evolved for these environments. Despite these disadvantages, modern humans either completely replaced Neanderthals or genetically swamped them. How did they overcome these disadvantages? I think they did so with a suite of bio-behavioural traits evolved during isotope stage 6 in the tropics of Africa. Table 1 sets out the basics of a model for contrasting adaptational paths for Neanderthals and modern humans. The model is based on the ecological differences between Europe and Africa discussed above, and incorporates the known archaeological, anatomical, and biological data and theory previously discussed. The model addresses three behavioural systems: Neanderthals evolving in their core area of western Eurasia at temperate to cold latitudes during primarily glacial conditions, modern humans evolving in Africa during dry arid (=glacial) conditions, and then modern humans transplanted to Europe. It addresses four key behavioural variables: prey choice, encounter technique for those prey, organisational technique for those encounters, and killing technique. For each, it posits a strategy, the behavioural implications of that strategy, and the archaeological and physical anthropological empirical expectations for those behaviours. Those expectations that have already been met in the literature, and discussed in the preceding text, are italicised. This model posits that Neanderthals evolved as a specialised predator with a highrisk, high-return, but energetically costly and physically risky (relative to modern humans) foraging strategy. They focused on pursuit of the largest mammals on the

354

From the tropics to the colder climates Table 1

A model for the evolution of contrasting adaptational paths for Neanderthals and modern humans in the cold temperate and cold environments of Western Europe versus the warm tropics of Africa.

Neanderthals in Europe

Foraging Strategy

Behavioral Implications

Archaeological/ Anatomical Empirical Expectations

Prey Choice

Narrow, high return

Targeted focus on largest prey on landscape, less selection for ability to map on to many highly variable food resources

Narrow species diversity in faunal assemblages, isotopic signal narrow and reflective of just several major prey

Encounter Technique

Long dogged pursuit, perhaps using a herd following search pattern if no storage

High search rates per day, high residential mobility, little time for more social and symbolic activities

Skeleton showing extreme physical activity, little evidence for social and symbolic activities

Organizational Technique

Small group, little aggregation, little economic input from women

Relatively infrequent opportunities for extra-core group social activities, non-egalitarian, women socially weak

Sites showing short-term occupation with little structured use of space, small sites focused around a single hearth, lack of evidence for longrange trade

Killing technique

Close combat spear thrusting

High trauma, early mortality, strong selection for powerful anatomy and premium on muscles involved in spear-assisted killing

Patterns and frequencies of bone breakage and early mortality outside norm for modern humans in these environments, powerful thrusting muscles

Modern Humans in Africa

Foraging Strategy

Behavioral Implications

Archaeological/ Anatomical Empirical Expectations

Prey Choice

Broad, focused on plants supplemented by animals small and large

Flexible strategy with strong economic significance on women, strong selection on ability to map on to many variable food resources

Broad species diversity, isotopic signal broad

Encounter Technique

Plants – routed foraging linked to high residential mobility that maps onto plant distributions Animals – routed foraging, taking virtually anything encountered

Low search rates per day, high residential mobility, lots of time for social and symbolic activities

Skeletons showing activity levels similar to modern hunter-gatherers, some evidence for social and symbolic activity

355

From Tools to Symbols Modern Humans in Africa (cont)

Foraging Strategy

Behavioral Implications

Archaeological/ Anatomical Empirical Expectations

Organizational Technique

Variable pattern of dispersal and aggregation

Frequent opportunities for extra-core group activities, egalitarian, women socially powerful

Both small and larger sites reflecting dispersal and aggregation patterns, evidence for long-range trade

Killing technique

Distant kill employing thrown spears, traps, deadfalls, poison

Low trauma, late mortality, selection for speedy, stealthy, smart, and technically adept hunter and gatherer

Patterns and frequencies of bone trauma and morbidity in range of modern hunter-gatherers in these environments, technology flexible and fairly light

Modern Humans entering Europe

Foraging Strategy

Behavioral Implications

Archaeological/ Anatomical Empirical Expectations

Prey Choice

Seasonally varying between broad and narrow switching rapidly with conditions

Flexible strategy with shift toward men in economic significance, continues selection for ability to map onto highly variable resources

Species diversity and isotopic signature reflecting a broader diet relative to Neanderthals

Encounter Technique

Tactical landscape and herd intercept for migratory mammals, small animals such as fish and birds take with residential moves to locations of availability

Search rates lower than Neanderthals but seasonally variable, lower residential mobility than Neanderthals with lots of logistical mobility, lots of time for social and symbolic activities

Skeletons more gracile than Neanderthals, evidence for social and symbolic activity

Organizational Technique

Large group to communal for large mammals, dispersal with small group exploitation of small animals

Frequent opportunities for extra-core group activities, weakened egalitarian structures, women’s social power lessoned

Skeletons showing activity levels similar to modern hunter-gatherers, sites showing both long-term residential occupations and specialized logistical occupations

Killing Technique

Distant low risk killing for large mammals, technologically sophisticated exploitation of small animals if needed

Low trauma, late mortality, selection for technically and socially adept hunter, technology focused on high performance and often labor intensive to produce

Patterns and frequencies of bone trauma and mortality in range of modern hunter-gatherers in these environments, technology sophisticated, labor intensive, and high performance

356

From the tropics to the colder climates

landscape with the highest return rates, and eschewed smaller mammals like birds, fish, and small mammals. They employed an energy-costly strategy reliant on sustained search (dogged search across the landscape) and small group size. When an encounter began, Neanderthals employed an ambush strategy with a risky close contact capture that resulted in high trauma rates and early mortality, and put a premium on great strength and physical endurance despite the energetically costly nature of these adaptations. Modern humans in Africa evolved a low-risk, regular-return strategy similar to that seen among modern tropical hunter-gatherers in arid environments. Prey choice was broad and focused on plants supplemented by animals, including everything from small to large items. This put strong selection on a flexible strategy with the ability to map on to the spatial and temporal appearance of hundreds of differing food resources, and it put an economic emphasis on women. Modern humans in Africa relied on routed foraging strategies that resulted in fairly low search rates per day, high residential mobility, frequent dispersal and aggregations of groups. Encounters with animals were based on safe-distance killing with selection for a stealthy, smart, and technically adept hunter. Evolution within this context produced an organism that would respond very differently than the Neanderthals to cold temperate and cold environments. Moving into the upper latitudes of western Eurasia, modern humans practised a strategy that was, relative to Neanderthals, more generalised and low risk. By necessity the economic emphasis switched from plants to animals, but modern humans incorporated a broader range of food items, including fish and small mammals when these were available. Large mammal hunting focused on tactical landscape strategies, communal hunts and herd intercept facilitated by high-performance weaponry targeting the production of surplus and its storage, all closely timed to the appearance of migrating animals. Residential mobility was reduced while logistical mobility was employed to attack mobile prey, and search costs were reduced. The result was substantial opportunity for core group and inter-core group social and symbolic activity, and long-range economic contact. This new strategy allowed modern humans to displace Neanderthals by increasing their hunting success rates with their improved tactics and technology, along with lowering their risk and trauma rates. Overall, their net return rate exceeded that of Neanderthals. The result was that modern humans out-bred Neanderthals and usurped the narrow range of food items targeted by Neanderthals, depressing the productivity of those food items, and Neanderthals lacked the ability to expand their diet breadth and adapt.

357

From Tools to Symbols

Discussion and speculations In Table 1, I have tried to identify very specific characteristics to the model that have test implications that can either be employed with current palaeoanthropological data, or could be testable in the future without overly optimistic projections of our methodological abilities. Equally importantly, there are several empirical implications that have the ability to either falsify outright the hypotheses that make up the model, or at least force a major reconfiguration of the model. I address some of those now. Neanderthals and targeting of high-risk prey Grayson and Delpech (2003) have recently published on the faunal assemblage from Grotte XVI in the Dordogne of France. Grotte XVI has a rather continuous stratigraphy from the late Mousterian through to the later Upper Palaeolithic. Their data show some substantial changes in species representation through time and across the MP and UP boundary, but they argue these changes are due to climatic forcing, not shifts in species selection by hominins. The model presented here posits several differences between Neanderthal and modern human faunal exploitation, and at least one of these argues that Neanderthals tended to focus their pursuit on larger, higher-risk, higher-return animals. According to the interpretations of Grayson and Delpech (2003), that posited difference is not revealed in the Grotte XVI species representation, so either the test implication is not met with the Grotte XVI data, or there are complicating factors to the interpretation of the Grotte XVI data. Mellars (2004) has recently contested the Grayson and Delpech interpretation of the record, so there is some room for debate over the faunal patterning alone. One striking issue that is raised by the Grotte XVI data is its inconsistency with the isotopic data on Neanderthal and modern human diet (Bocherens et al., 1999; Bocherens et al., 2001; Richards et al., 2000). The Bocherens study samples three Neanderthal individuals from three separate sites in the Maas valley in Belgium that date between 35 000 and 40 000 BP. While these sites are further north than Grotte XVI, the Grotte XVI Mousterian layers date to around 65 000 BP, a period somewhat colder than that of the Belgian samples (Van Andel, 2002). The Belgian isotopic data show a ‘Neandertal diet characterised by high proportions of open environment herbivore proteins, even during forested conditions’ (Bocherens et al., 2001: 503). Even more intriguing, the three Neanderthals display 15N values that surpass mammoths, and mammoth 15N values surpass those of the range of mammals well represented in the Grotte XVI assemblage, or in other Neanderthal cave assemblages, for that matter. The most parsimonious explanation is that the Neanderthals ingested high quantities of mammoth flesh, though other explanations need to be ruled out. A geographically broader study of Neanderthals shows a result equally inconsistent with the Grotte XVI

358

From the tropics to the colder climates

data (Richards et al., 2001). The isotopic data show a significant broadening of the diet breadth across the Neanderthal/modern human boundary, and a greater evenness in the exploitation of resources, neither of which is reflected in the Grotte XVI data. In fact, ungulate evenness decreases with time in the Grotte XVI data. This is not the only known instance where the archaeological data, and particularly the zooarchaeological data, do not match the isotopic results. Perhaps the most dramatic contradiction comes from the south-western Cape near Elands Bay. Here, a wide range of data (far more robust than the Neanderthal record) strongly suggested a model where populations moved between the coastline and interior, subsisting on a broad diet in both areas (Parkington, 1976, 1981, 1987). Isotopic analyses demonstrated this to be incorrect – skeletons near the coast had diets based nearly entirely on marine resources, while skeletons from the interior had the broader diet (Sealy & van der Merwe, 1987; Sealy & Sillen, 1988). The faunal assemblages from the coastal caves and rockshelters, like Grotte XVI, are far more diverse than that suggested by the isotopic values for the skeletons at the coast (Parkington, 1981). Why this disparity? I can think of several, not mutually exclusive, explanations that could be specific to the Neanderthal case here, or more generally applicable to faunal assemblages from caves and rockshelters. First, it is possible, even likely, that these caves and rockshelters sample a very narrow period of time in the annual cycle of a hominin’s life. The slice of time specific to cave occupation could include a rather broader diet than is typical, and the open-air sites on the landscape could display a different pattern that is as yet not well documented. Second and related to the first, it is possible that we (zooarchaeologists) have failed to tease apart the relative contributions of faunal collecting agents among the taxa present in caves and rockshelters. Thus, if a cave or rockshelter is inhabited for just one month per year, then the other eleven months pose the potential for accumulation by other agents. If Neanderthals brought in mostly reindeer, but other agents dragged in other taxa, then our taphonomic analysis needs to tease this apart. For example, taphonomic analysis at DK1 reveals that small bovids were accumulated primarily by raptors, while sizes 2–4 were accumulated by people, in the same layers (Marean et al., 2000). Third, in the specific case of the Neanderthals, perhaps the bones of the large taxa (such as mammoth) were not transported to the sites. This, of course, is what we would expect, given ethnographic data that show large taxa are less completely transported, particularly if group size is small (Bunn et al., 1988; Monahan, 1998), as has been argued to be the case for Neanderthals (Mellars, 1996). So, for each red deer in a cave, there could have been ten mammoths that were filleted and consumed out on the landscape.

359

From Tools to Symbols

Thus, these cave and rockshelter faunal assemblages may not be telling us much about dietary composition, at least with the way that we are studying them. Neanderthals, long searches, and high mobility The model posits that Neanderthals practised a high-energy search strategy reliant on long periods of search by small groups. I noted that they may even have practised herd following, though this is not a critical part of the model, and is only triggered into being important if Neanderthals did not store large amounts of food. If they did store food, then herd following would not have been needed as they could have stored sufficient food to make it through the periods when herds were outside their home range. There are several forms of data that are relevant to this issue, including patterns of raw material procurement and anatomical morphology. Neanderthal lithic assemblages show a pattern where the exploited raw materials are dominated by local sources (see review in Mellars, 1996). The majority come from 5–6 km distance and the next most frequent are those within 6–12 km. On the face of it, this would seem to be inconsistent with herd following, and that may be the case, but Neanderthal sites do, intriguingly, frequently have raw materials from 20–100 km distance. However, this inconsistency is based on the assumption that high mobility across great distances would result in large quantities of exotic raw materials. This is an assumption untested in ethnographic contexts since we do not have ethnographic data on lithic raw material procurement and mobility. It is perfectly reasonable to propose that highly mobile people did not carry large quantities of raw material with them for prolonged periods of time, but rather exploited locally known raw materials as they moved through regions. Raw material procurement probably speaks even less decisively about the intensity of search strategies. A pattern of long versus short search, provided it took place within similarly sized home ranges, has no obvious implications for raw material procurement. Some final speculations Clearly, hunter-gatherer foraging strategies have complex relations with other aspects of the behavioural system, and if this model is even remotely close to reality it is likely to have many consequences. I take this opportunity to speculate on what those might be. The hunter-gatherer ethnographic record clearly shows us that in tropical environments women provide most of the plant foods and men do most of the hunting (Kelly, 1995). Since plant foods comprise the majority of the diet, women are economically extremely important. Their significance to the food-getting task decreases as one moves to higher

360

From the tropics to the colder climates

latitudes, so it seems reasonable to speculate that women were a critical part of the economic equation in the tropics of Africa, and perhaps less so among Neanderthals, unless the link between work and sex differed fundamentally from that seen in modern humans. That of course is a possibility, and presumably can be tested by sex-based studies of skeletal robusticity and trauma. While one might intuitively speculate that women’s status in plant-food based systems would be higher, at least one study has not found that to be demonstrably the case (Hayden et al., 1986). However, another study has shown that women’s status is typically lower when men spend significant periods of time away and the environment is perceived as harsh or hostile (Sanday, 1981). If this pattern holds for Neanderthals and the presented model is correct, then women may have had very low status. Hunter-gatherers focusing on plant foods in the tropics tend to have egalitarian social structures (Woodburn, 1982). While these do not always result in higher status for women, they do have complex rule sets of checks and balances that level status for everyone (see discussion in Kelly, 1995). At the beginning of this paper I asked a simple question: ‘how is it that an animal evolved in the tropics quickly out-competes for food an animal evolved in temperate and cold environments in its own territory?’ It seems likely to me that the complexity of commanding and communicating the temporal and spatial distribution of hundreds of plant foods, and the maintenance of egalitarian social structures, could be a powerful force for language development in tropical hominins. I think it likely that this tropical evolutionary context resulted in a hominin whose adaptation was rooted in complex language coupled to a highly flexible and adaptable mental capacity grounded in long-term planning. While Neanderthals may have shared some of these abilities (d’Errico, 2003), I think they were rudimentary relative to modern humans and existed only as a result of a shared common ancestor who also had these abilities at a comparatively rudimentary level. These differences allowed modern humans from Africa to overcome the severe handicaps mentioned above when they entered the new European environment and out-compete Neanderthals on their home ground. Some implications for technology The crux of this model is that modern humans entered Europe with a far more adaptable technology and flexible behavioural repertoire than was present among Neanderthals. While there is a tendency to minimise the differences between African and Eurasian lithic technology during the MSA and MP (Klein, 1999, 2000), I think there are some critical differences between what I will call the ‘African technological core area’ and the ‘Eurasian technological core area’ during the MSA and MP that may be technological markers of these differing technologies (Marean & Assefa, 2005). I believe the former is associated with the core area for the evolution of modern humans,

361

From Tools to Symbols

and the latter with Neanderthals, with blending at the boundaries such as the Levant, and diffusion as a wave front as modern humans entered Europe. The explosion of material cultural complexity seen with the introduction of the Aurignacian into Western Europe (Conard & Bolus, 2003) reflects this flexible African behavioural system reacting to the new pressures of glacial and peri-glacial environments. Ultimately, it would be appropriate and useful to drop the term Middle Paleolithic for Northern Africa, and expand the use of the term Middle Stone Age to include the entirety of the African technological core area (Marean & Assefa, 2005). Throughout Africa, MSA lithic technologies are regularly based on blade production, and these blades are often made on special blade cores. Some of these blade technologies are Levallois, but others are not and are clearly designed for regular blade production. Blade-dominated assemblages occur in the Levant, but this is what one would expect at the edge of an African blade technology core area. Outside this boundary lithic assemblages are flake-based and while blades are present (Kuhn & Bar-Yosef, 1999), they are rarely the dominant technology as they often are in Africa. The African technological core area also has a more regular occurrence of specialised highly curated bifaces, often made with sophisticated flaking techniques. Some of the large and thin Still Bay bifaces appear to have been pressure-flaked (Minichillo, 2004), though systematic studies have yet to be published. These include the Aterian, Lupemban, and Still Bay, as well as others. Variants of these technologies are present all over Africa. In contrast, the Eurasian technological core area rarely, if ever, has these types of technologies. Coupled to this greater technological sophistication in Africa is a raw material extraction pattern that utilises high-quality raw materials attained from sources far away, while virtually all raw material extraction in the Eurasian technological core area is more local (McBrearty & Brooks, 2000). As has recently been demonstrated, the African technological core area incorporated the use of raw materials other than stone when needed, such as the production of sophisticated bone tools (Brooks et al., 1995; Yellen et al., 1995; Henshilwood et al., 2001), while the Eurasian technological core area does not. Wedded to these higher-grade technologies is the encoding of symbols into common technology in the African core area. For example, both the Aterian and Stillbay include bifaces that are far too thin and fragile for the purposes of hunting, perhaps for any functional task, though the vast majority are clearly useable as either spear/dart tips or knives. These rarer, more fragile bifaces likely took on social and religious roles and demonstrate an ability to morph technology into a variety of non-utilitarian purposes. This is not known from the Eurasian technological core area. I believe this technological distinction began to develop at the beginning of the MSA as far back as 300 000 years BP, and the post-isotope stage 6 differences between

362

From the tropics to the colder climates

Africa and Europe reflect an intensifying of the different adaptational paths evolving for Neanderthals and modern humans during the harsh conditions of isotope stage 6. As noted above, the diets of tropical African hunter-gatherers are far more diverse than hunter-gatherers in cold temperate and cold environments and are based on plant foods, which tend to be highly seasonal and often require very special types of processing to remove toxicities. During the aridity of stage 6 plants would have taken on an even greater role in Africa. People would have been forced to learn to map their mobility strategies on to the spatio-temporal expression of these plants, and wed that to a strategy for intercepting, conserving, and moving water. All of this would have put heavy selection on the ability to store and transmit knowledge of the timing and location of this complex suite of resources. While large mammal hunting would not have been as significant as it was among Neanderthals, when modern humans entered environments where large mammal hunting became essential, they had the intellectual equipment to develop very sophisticated strategies to exploit those resources. Modern humans were pre-adapted to the expression of technological and cultural complexity given the proper environmental stimuli, and these stimuli were provided by the cold and harsh conditions of western Eurasia, explaining the sudden florescence of complex material culture with the Upper Palaeolithic, but also explaining the relative lack of such florescence in Africa all the way to the Holocene.

Conclusions The crux of this model is that modern humans in Africa entered Europe with abilities that allowed them to hunt more effectively than Neanderthals. These abilities were biologically-based behavioural differences that evolved as a consequence of differing evolutionary trajectories in radically different environments. My argument is that Neanderthals and modern humans were fundamentally different species with distinct behavioural and anatomical traits. The growing body of physical anthropological data shows this to be the unarguable case. Neanderthals and modern humans were different bio-behavioural packages that, when juxtaposed in Eurasia, resulted in Neanderthal extinction. This is no way implies that Neanderthals were poorly adapted to their environment – clearly, they were a successful occupant of Eurasia for thousands of years. But that does not detract from the fact that a better occupant can arrive simply through historical contingency, as has happened so many times in evolution, usurping the niche space and resulting in extinction. How do we investigate this model? As I noted above, hunting effectiveness has been and can be conceived of in a variety of ways. Binford and Stiner have focused on the distinctions between hunting and scavenging. Klein and Stiner have targeted hunting effectiveness as reflecting the ability to hunt animals of varying age classes, and

363

From Tools to Symbols

Klein has extended this to animals of differing dangerousness. These may be important distinctions, but I think that by the time Neanderthals and early modern humans evolved they could kill any animal on the landscape. The key issue we should be targeting is the overall energetic efficiency of capture, and the reduction of risk (both failure and bodily harm). When Neanderthals went hunting, how many attempts were required for a success, and how much energy and time were exerted searching; what was the net energy return? How many times for each effort did they get injured? This poses extreme challenges on zooarchaeological method and theory, and calls for an integration of archaeological and physical anthropological data and theory that is not currently the norm, but which I have attempted to initiate here.

Acknowledgements I thank Francesco d’Errico and Lucinda Backwell for inviting me to participate in the conference and contribute to the published volume. The conference was outstanding, a tribute to their thoughtfulness and organisational abilities. I thank the following for helpful comments on this paper: Francesco d’Errico, Donald Grayson, Jim O’Connell, and John Shea. The financial support of the National Science Foundation (USA) (grant # BCS-9912465 and BCS-0130713) and the Hyde Family Trust are gratefully appreciated.

References Archibold, O.W. (1995). Ecology of World Vegetation. New York: Chapman and Hall Bartram, L.E. & Marean, C.W. (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. Ben-Itzhak, S., Smith, P. & Bloom, R.A. (1988). Radiographic study of the humerus in Neandertals and Homo sapiens sapiens. American Journal of Physical Anthropology 77, 231–242. Berger, T.D. & Trinkaus, E. (1995). Patterns of trauma among the Neandertals. Journal of Archaeological Science 22, 841–852. Binford, L.R. (1978). Nunamiut Ethnoarchaeology. New York: Academic Press. Binford, L.R. (1981). Bones: Ancient Men and Modern Myths. New York: Academic Press. Binford, L.R. (1984). The Faunal Remains from Klasies River Mouth. New York: Academic Press. Binford, L.R. (1985). Human ancestors: changing views of their behavior. Journal of Anthropological Archaeology 4, 292–327. Binford, L.R. (1988). Etude taphonomique des restes fauniques de la Grotte Vaufrey, couche VIII. In (J.-P. Rigaud, Ed.), La Grotte Vaufrey: paleoenvironnement, chronologie, activités humaines, Memoires de la Societe Prehistorique Francaise, pp. 535–564.

364

From the tropics to the colder climates Blumenschine, R.J. (1986a). Carcass consumption sequences and the archaeological distinction of scavenging and hunting. Journal of Human Evolution 15, 639–659. Blumenschine, R.J. (1986b). Early Hominid Scavenging Opportunities: Implications of Carcass Availability in the Serengeti and Ngorongoro Ecosystems. BAR International Series 283, Oxford. Blumenschine, R.J. (1987). Characteristics of an early hominid scavenging niche. Current Anthropology 28, 383–407. Bocherens, H., Billiou, D., Mariotti, A., Toussaint, M., Patou-Mathis, M., Bonjean, D. & Otte, M. (2001). New isotopic evidence for dietary habits of Neandertals from Belgium. Journal of Human Evolution 40, 497–506. Bocherens, H., Billiou, D., Mariotti, A., Patou-Mathis, M., Otte, M., Bonjean, D. & Toussaint, M. (1999). Palaeoenvironmental and palaeodietary implications of isotopic biogeochemistry of last interglacial Neanderthal and mammal bones in Scladina Cave (Belgium). Journal of Archaeological Science 26, 599–607. Brooks, A.S., Helgren, D.M., Cramer, J.S., Franklin, A., Hornyak, W., Keating, J.M., Klein, R.G., Rink, W.J., Schwarcz, H.P., Smith, J.N.L., Stewart, K., Todd, N., Verniers, J. & Yellen, J.E. (1995). Dating and context of three Middle Stone Age sites with bone points in the upper Semliki Valley, Zaire. Science 268, 548–553. Burch Jr., E.S. (1972). The caribou/wild reindeer as a human resource. American Antiquity 37, 339–368. Bunn, H.T., Bartram, L. & Kroll, E.M. (1988). Variability in bone assemblage formation from hadza hunting, scavenging, and carcass processing. Journal of Anthropological Archaeology 7, 412–457. Chase, P.G. (1986). The Hunters of Combe Grenal: Approaches to Middle Paleolithic Subsistence in Europe. BAR International Series 286, Oxford, England. Chase, P.G. (1989). How different was Middle Paleolithic subsistence? A zooarchaeological perspective on the Middle to Upper Palaeolithic transition. In (P. Mellars & C. Stringer, Eds), The Human Revolution: Behavioural and Biological Perspectives on the Origins of Modern Humans, pp. 321–337. Edinburgh: Edinburgh University Press. Chase, P.G., Armand, D., Debenath, A., Dibble, H. & Jelinek, A.J. (1994). Taphonomy and zooarchaeology of a Mousterian faunal assemblage from La Quina, Charente, France. Journal of Field Archaeology 21, 289–305. Chase, P.G. (1988). Scavenging and hunting in the Middle Paleolithic. In (H.L. Dibble & A. Montet-White, Eds), Upper Pleistocene Prehistory of Western Eurasia, pp. 225–232. Philadelphia: The University Museum. Conard, N.J. & Bolus, M. (2003). Radiocarbon dating the appearance of modern humans and timing of cultural innovations in Europe: new results and new challenges. Journal of Human Evolution 44, 331–371.

365

From Tools to Symbols Coupland, R.T. (1979). Grassland Ecosystems of the World. New York: Cambridge University Press. Damas, D. (1968). The diversity of Eskimo societies. In (R.B. Lee & I. DeVore, Eds) Man the Hunter, pp. 111–117. Chicago: Aldine. d’Errico, F. (2003). The invisible frontier. A multiple species model for the origin of behavioral modernity. Evolutionary Anthropology 12, 188–202. d’Errico, F., Zilhão, J., Julien, M., Baffier, C. & Pelegrin, J. (1998). Neanderthal acculturation in western Europe. Current Anthropology 39, S2–S44. De Vos, A. (1969). Ecological conditions affecting the production of wild herbivorous mammals on grasslands. Advances in Ecological Research 6, 137–183. Deacon, J. & Lancaster, N. (1988). Late Quaternary Paleoenvironments of Southern Africa. Oxford: Clarendon Press. Frisancho, A.R. (1993). Human Adaptation and Accommodation. Ann Arbor: University of Michigan Press. Gentry, D., Blamart, D., Ouahdi, R., Gilmour, M., Baker, A., Jouzel, J. & Van-Exter, S. (2003). Precise dating of Dansgaard-Oeschger climatic oscillations from stalagmite data. Nature 421, 833–837. Grayson, D.K. & Delpech, F. (1994). The evidence for Middle Paleolithic scavenging from Couche VIII, Grotte Vaufrey (Dordogne, France). Journal of Archaeological Science 21, 359–375. Grayson, D.K., Delpech, F., Rigaud, J.-P., & Simek, J.F. (2001). Explaining the development of dietary dominance by a single ungulate taxon at Grotte XVI, Dordogne, France. Journal of Archaeological Science 28, 115–125. Grayson, D.K. & Delpech, F. (2003). Ungulates and the Middle-to-Upper Paleolithic transition at Grotte XVI (Dordogne, France). Journal of Archaeological Science 30, 1633–1648. Grove, A.T. (1988). The Changing Geography of Africa. Oxford: Oxford University Press. Hamilton, A.C. (1982). Environmental History of East Africa. New York: Academic Press. Hayden, B., Deal, M., Cannon, A. & Casey, J. (1986). Ecological determinants of women’s status among hunter/gatherers. Human Evolution 1, 449–474. Henshilwood, C.S., D'Errico, F., Marean, C.W., Milo, R.G. & Yates, R.J. (2001). An early bone tool industry from the Middle Stone Age, Blombos Cave, South Africa: implications for the origins of modern human behaviour, symbolism and language. Journal of Human Evolution 41, 631–678. Holliday, T.W. (1997). Postcranial evidence of cold adaptation in European Neanderthals. American Journal of Physical Anthropology 104, 245–258. Hublin, J.-J. (1998). Climatic changes, paleogeography, and the evolution of the Neandertals. In (T. Akazawa, K. Aoki & O. Bar-Yosef, Eds) Neandertals and Modern Humans in Western Asia, pp. 295–310. New York and London: Plenum Press. Imbrie, J. & Imbrie, K.P. (1979). Ice Ages: Solving the Mystery. Cambridge, Mass.: Harvard University Press.

366

From the tropics to the colder climates Jablonski, N. G. & Chaplin, G. (2000). The evolution of human skin coloration. Journal of Human Evolution 39, 57–106. Jelinek, A.J. (1994). Hominids, energy, environment, and behavior in the Late Pleistocene. In (M.H. Nitecki & D.V. Nitecki, Eds) Origins of Anatomically Modern Humans, pp. 67–92. New York: Plenum Press. Jones, K. & Metcalfe, D (1988). Bone marrow indices and efficiency. Journal of Archeological Science 15, 415-423. Kennedy, G.E., Gauld, S.C. & Nelson, A.J. (1990). Bone thickness in fossil hominids. In (A.D. Dixon, B.G. Sarnat & D.A.N. Hoyte, Eds) Fundamentals of Bone Growth: Methodology and Applications, pp. 587–593. Boston: CRC Press. Klein, R.G. (1976). The mammalian fauna of the Klasies River Mouth sites, southern Cape Province, South Africa. South African Archaeological Bulletin 31, 75–98. Klein, R.G. (1995). Anatomy, behavior, and modern human origins. Journal of World Prehistory 9, 167–198. Klein, R.G. (1998). Why anatomically modern people did not disperse from Africa 100,000 years ago. In (T. Akazawa, K. Aoki & O. Bar-Yosef, Eds) Neanderthals and Modern Humans in Western Asia, pp. 509–521. New York: Plenum Press. Klein, R.G. (1999). The Human Career: Human Biological and Cultural Origins. Chicago: University of Chicago Press. Klein, R.G. (2000). Archeology and the evolution of human behavior. Evolutionary Anthropology 9, 17–36. Klein, R.G. & Cruz-Uribe, K. (1996). Exploitation of large bovids and seals at Middle and Later Stone Age sites in South Africa. Journal of Human Evolution 31, 315–334. Krebs, C. J. (1978). Ecology: The Experimental Analysis of Distribution and Abundance. New York: Harper and Row. Laughlin, W.S. (1972). Ecology and population structure in the Arctic. In (G. Harrison & A. Boyce, Eds) The Structure of Human Populations. Oxford: Clarendon Press. Lazenby, R.A. (1997). Bone loss, traditional diet, and cold adaptation in Arctic populations. Journal of Human Biology 9, 329–341. Lee, R.B. (1968). What hunters do for a living, or, how to make out scarce resources. In (R.B. Lee & I. DeVore, Eds) Man the Hunter, pp. 30–48. Chicago: Aldine. Leonard, W.R. (2003). Food for thought. Scientific American 13, 62–71. Lupo, K.D. (1998). Experimentally derived extraction rates for marrow: implications for body part exploitation strategies of Plio-Pleistocene Hominid scavengers. Journal of Archaeological Science 25, 657–676. Lupo, K.D. & Schmitt, D.N. (1997). Experiments in bone boiling: nutritional returns and archaeological reflections. Anthropozoologica 25–26, 137–144.

367

From Tools to Symbols Madrigal, T.C. & Holt, J.Z. (2002). White-tailed deer meat and marrow return rates and their application to Eastern Woodlands archaeology. American Antiquity 67, 745–759. Marean, C.W. (1997). Hunter-gatherer foraging strategies in tropical grasslands: model building and testing in the East African Middle and Later Stone Age. Journal of Anthropological Archaeology 16, 189–225. Marean, C.W. (1998). A critique of the evidence for scavenging by Neanderthals and early modern humans: new data from Kobeh Cave (Zagros Mousterian) and Die Kelders (South Africa Middle Stone Age). Journal of Human Evolution 35, 111–136. Marean, C.W., Abe, Y., Frey, C.J. & Randall, R.C. (2000). Zooarchaeological and taphonomic analysis of the Die Kelders Cave 1 Layers 10 and 11 Middle Stone Age larger mammal fauna. Journal of Human Evolution 38, 197–233. Marean, C.W. & Assefa, Z. (1999). Zooarchaeological evidence for the faunal exploitation behavior of Neandertals and Early Modern Humans. Evolutionary Anthropology 8, 22–37. Marean, C.W. & Assefa, Z. (2005). The Middle and Upper Pleistocene African record for the biological and behavioral origins of modern humans. In (A.B. Stahl, Ed.) African Archaeology, pp. 93–129. New York: Blackwell. Marean, C.W. & Frey, C.J. (1997). The animal bones from caves to cities: reverse utility curves as methodological artifacts. American Antiquity 62, 698–711. Marean, C.W. & Kim, S.Y. (1998). Mousterian faunal remains from Kobeh cave (Zagros Mountains, Iran): behavioral implications for Neanderthals and early modern humans. Current Anthropology 39, S79–S114. McBrearty, S. & Brooks, A.S. (2000). The revolution that wasn’t: a new interpretation of the origin of modern human behavior. Journal of Human Evolution 39, 453–563. Mellars, P.A. (1996). The Neanderthal Legacy: An Archaeological Perspective from Western Europe. Princeton: Princeton University Press. Mellars, P.A. (1973). The character of the middle-upper Paleolithic transition in south-west France. In (C. Renfrew, Ed.) The Explanation of Cultural Change, pp. 255–276. London: Duckworth. Mellars, P.A. (2004). Reindeer specialization in the early Upper Palaeolithic: the evidence from south west France. Journal of Archaeological Science 31, 613–617. Milo, R.G. (1994). Human-Animal Interactions in Southern African Prehistory: a Microscopic Study of Bone Damage Signatures. Unpublished Ph.D. dissertation, University of Chicago. Milo, R.G. (1998). Evidence for hominid predation at Klasies River Mouth, South Africa, and its implications for the behavior of early modern humans. Journal of Archaeological Science 25, 99–133. Minichillo, T. (2004). The Still Bay in context. Paper delivered at the 2004 Annual Meeting of the Paleoanthropology Society, 30–31 March 2004, Montreal, Canada.

368

From the tropics to the colder climates Monahan, C.M. (1998). The Hadza carcass transport debate revisited and its archaeological implications. Journal of Archaeological Science 25, 405–424. Moran, E.F. (1982). Human Adaptability. Boulder: Westview Press. Mussi, M. (1999). The Neanderthals in Italy: A tale of many caves. In (W. Roebroeks & C. Gamble, Eds) The Middle Paleolithic Occupation of Europe, pp. 49–80. Leiden: University of Leiden. NIH Clinical Nutrition Service (2002). Vitamin D Fact Sheet. htp://www.cc.nih.gov/ccc/ supplements/vitd.html. O’Connell, J.F., Hawkes, K., Lupo, K.D. & Blurton Jones, N.G. (2002). Male strategies and PlioPleistocene archaeology. Journal of Human Evolution 43, 831–872. Parkington, J. (1976). Coastal settlement between the mouths of the Berg and Olifants Rivers, Cape Province. South African Archaeological Bulletin 31, 127–140. Parkington, J. (1981). The effects of environmental change on the scheduling of visits to the Elands Bay Cave, Cape Province, S.A. In (I. Hodder, G. Isaac & N. Hammond, Eds) Pattern of the Past, pp. 341–359. Cambridge: Cambridge University Press. Parkington, J. (1987). Changing views of prehistoric settlement in the Western Cape. In (J. Parkington & M. Hall, Eds) Papers in the Prehistory of the Western Cape, South Africa, BAR International, Oxford, pp. 4–23. Pearson, O.M. (2000). Activity, climate, and postcranial robusticity: implications for modern human origins and scenarios of adaptive change. Current Anthropology 41, 569–607. Pfeiffer, S.K. & Lazenby, R.A. (1994). Low bone mass in past and present aboriginal populations. In (H.H. Draper, Ed.) Advances in Nutritional Research, pp. 35–51. New York: Plenum Press. Pianka, E.R. (1966). Latitudinal gradients in species diversity: a review of concepts. American Naturalist 100, 65–75. Richards, M.P., Pettitt, P.B., Stiner, M.C. & Trinkaus, E. (2001). Stable isotope evidence for increasing dietary breadth in the European mid-Upper Paleolithic. Proceedings of the National Academy of Sciences 98, 6528–6532. Rosenzweig, M.L. (1992). Species diversity gradients: we know more and less than we thought. Journal of Mammalogy 73, 715–730. Ruff, C.B. (1991). Climate and body shape in hominid evolution. Journal of Human Evolution 21, 81–105. Sanday, P. (1981). Female Power and Dominance: On the Origins of Sexual Inequality. Pittsburg: University of Pittsburg Press. Sealy, J.C. & Sillen, A. (1988). Sr and Sr/Ca in marine and terrestrial foodwebs in the Southwestern Cape, South Africa. Journal of Archaeological Science 15, 425–438. Sealy, J.C. & van der Merwe, N.J. (1987). Stable carbon isotopes, Later Stone Age diets and seasonal mobility in the South-western Cape. In (J. Parkington & M. Hall, Eds) Papers in the Prehistory of the Western Cape, South Africa, BAR International, Oxford, pp. 262–268.

369

From Tools to Symbols Senut, B. (1985). Computerized tomography of a Neanderthal humerus from Le Regourdou (Dordogne, France): Comparisons with modern man. Journal of Human Evolution 14, 717–723. Sept, J. (1992). Archaeological evidence and ecological perspectives for reconstructing early hominid subsistence behavior. Archaeological Method and Theory 4, 1–56. Sept, J.M. (1986). Plant foods and early hominids at site FxJj 50, Koobi Fora, Kenya. Journal of Human Evolution 15, 751–770. Sept, J.M. (1990). Vegetation studies in the Semliki Valley, Zaire as a guide to paleoanthropological research. Virginia Museum of Natural History Memoirs 1, 95–121. Sept, J.M. (1994). Beyond bones: archaeological sites, early hominid subsistence, and the costs and benefits of exploiting wild plant foods in East African riverine landscapes. Journal of Human Evolution 27, 295–320. Sorensen, M.V. & Leonard, W.R. (2001). Neandertal energetics and foraging efficiency. Journal of Human Evolution 40, 483–496. Spencer, Robert F. (1959). The North Alaskan Eskimo: A Study in Ecology and Society. Washington: Smithsonian Institution Bureau of American Ethnology. Speth, J.D. & Spielmann, K.A. (1983). Energy source, protein metabolism, and hunter-gatherer subsistence strategies. Journal of Anthropological Archaeology 2, 1–31. Stiner, M.C., Munro, N.D. & Surovell, T.A. (2000). The tortoise and the hare: small-game use, the broad-spectrum revolution, and Paleolithic demography. Current Anthropology 41, 39–73. Spoor, F., Hublin, J.J., Braun, M. & Zonneveld, F. (2003). The bony labyrinth of Neanderthals. Journal of Human Evolution 44, 141–165. Stiner, M.C. (1990). The use of mortality patterns in archaeological studies of hominid predatory adaptations. Journal of Anthropological Research 9, 305–351. Stiner, M.C. (1991a). An interspecific perspective on the emergence of the modern human predatory niche. In (M.C. Stiner, Ed.) Human Predators and Prey Mortality, pp. 149–185. Boulder: Westview Press. Stiner, M.C. (1991b). Food procurement and transport by human and non-human predators. Journal of Archaeological Science 18, 455–482. Stiner, M.C. (1993). Modern human origins – faunal perspective. Annual Review of Anthropology 22, 55–82. Stiner, M.C. (1994). Honor Among Thieves: A Zooarchaeological Study of Neandertal Ecology. Princeton: Princeton University Press. Stiner, M.C. (2001). Thirty years on the ‘Broad Spectrum Revolution’ and paleolithic demography. Proceedings of the National Academy of Sciences 98, 6993–6996. Strahler, A.H. & Strahler, A.N. (1992). Modern Physical Geography. New York: Wiley. Tanaka, J. (1976). Subsistence ecology of central Kalahari San. In (R.B. Lee & I. DeVore, Eds) Kalahari Hunter-Gatherers, pp. 98–119. Cambridge, Mass.: Harvard University Press.

370

From the tropics to the colder climates Thoma, A. (1975). Were the Spy fossils evolutionary intermediates between classic Neandertal and modern man? Journal of Human Evolution 4, 387–410. Thompson, D.D. & Gunness-Hey, M. (1981). Bone mineral-osteon analysis of Yupik-Inupiaq skeletons. American Journal of Physical Anthropology 55, 1–7. Trinkaus, E. (1981). Neanderthal limb proportions and cold adaptation. In (C.B. Stringer, Ed.) Symposia of the Society for the Study of Human Biology, pp. 187–224. London: Taylor and Francis. Trinkaus, E. (1983). The Shanidar Neandertals. New York: Academic Press. Trinkaus, E. (1983). Neandertal postcrania and the adaptive shift to modern humans. In (E. Trinkaus, Ed.) The Mousterian Legacy: Human Biocultural Change in the Upper Pleistocene, BAR International Series, Oxford, pp. 165–200. Trinkaus, E. (1995). Neanderthal mortality patterns. Journal of Archaeological Science 22, 121–142. Trinkaus, E. (2000). The ‘robusticity transition’ revisited. In (C.B. Stringer, R.N.E. Barton & J.C. Finlayson, Eds) Neanderthals on the Edge, pp. 227–236. Oxford: Oxbow Books. Trinkaus, E., Ruff, C.B. & Churchill, S.E. (1998). Upper limb versus lower limb loading patterns among Near Eastern Middle Paleolithic hominids. In (T. Akazawa, K. Aoki & O. Bar-Yosef, Eds) Neandertals and Modern Humans in Western Asia, pp. 391–404. New York and London: Plenum Press. Van Andel, T.H. (2002). The climate and landscape of the middle part of the Weichselian Glaciation in Europe: the Stage 3 Project. Quaternary Research 57, 2–8. Van Andel, T.H. & Tzedakis, P.C. (1996). Palaeolithic landscapes of Europe and environs, 150,000-25,000 years ago: an overview. Quaternary Science Reviews 15, 481–500. Vanstone, J.W. (1974). Athapaskan Adaptations. Chicago: Aldine Publishing. Vincent, A.S. (1984). Plant foods in savanna environments: a preliminary report of tubers eaten by the Hadza of northern Tanzania. World Archaeology 17, 131–147. Vincent, A.S. (1985). Underground Plant Foods and Subsistence in Human Evolution. Unpublished Ph.D. dissertation, University of California at Berkeley. Wandsnider, L. (1997). The roasted and the boiled: food composition and heat treatment with special emphasis on pit-hearth cooking. Journal of Anthropological Archaeology 16, 1–48. Whittaker, R.H. & Likens, G.E. (1973). Primary production: the biosphere and man. Human Ecology 1, 357–369. Woodburn, J. (1982). Egalitarian societies. Man 17, 431–451. Yellen, J.E., Brooks, A.S., Cornelissen, E., Mehlman, M.J. & Stewart, K. (1995). A Middle Stone Age worked bone industry from Katanda, Upper Semliki Valley, Zaire. Science 268, 553–556. Zilhão, J. (2001). Anatomically Archaic, Behaviorally Modern: The Last Neanderthals and Their Destiny. Amsterdam: Stichting Nederlands Museum.

371

New neighbours: interaction and imagemaking during the West European Middle to Upper Palaeolithic transition David Lewis-Williams Rock Art Research Institute, University of the Witwatersrand, Private Bag 3, WITS 2050, Johannesburg, South Africa

Abstract The Middle to Upper Palaeolithic Transition in Western Europe presents evidence for two species of Homo living side by side. The anatomically archaic species, the Neanderthals, borrowed certain practices from their Homo sapiens neighbours, but did not borrow others. It is hypothesised that the reason for their selectivity was their type of consciousness. While Homo sapiens people had higher-order consciousness, Neanderthals had a form of primary consciousness that did not permit them long-term symbolic memory or the ability to conceive of a spirit world. Coping with the shifting nature of higher-order consciousness, anatomically modern people necessarily divided the spectrum of mental states into evaluated segments. In this way, social discrimination and religion originated in tandem. Image-making in subterranean caverns was both a religious ritual and an instrument for social discrimination.

Résumé La transition Paléolithique moyen – Paléolithique supérieur en Europe occidentale se caractérise par la présence de deux espèces d’Homo vivant l’une à côté de l’autre. L’espèce anatomiquement archaïque, l’homme de Néandertal, a emprunté certaines pratiques de son voisin l’Homo sapiens, mais n’en a pas emprunté d’autres. Nous proposons comme hypothèse que la raison de cet emprunt sélectif réside dans le type de conscience dont étaient pourvus les néanderthaliens. Alors que l’Homo sapiens avait une conscience d’ordre supérieure, l’homme de Néandertal avait une forme de conscience primaire qui ne lui offrait pas une mémoire symbolique à long terme ni la capacité de concevoir un monde spirituel. Faisant face à la nature changeante d’une conscience d’ordre supérieur, les hommes anatomiquement modernes divisaient inévitablement l’éventail des états mentaux

372

New neighbours en segments évalués. En suivant ce model, la stratification sociale et la religion sont apparues ensemble. La production d’images dans la les grottes profondes était à la fois l’expression d’un rituel religieux et un moyen de différenciation sociale.

Introduction The West European Middle to Upper Palaeolithic Transition (c. 35 000–40 000 BP) is richly evidenced. During that time, Homo neanderthalensis groups, makers of the longenduring Mousterian technocomplex, lived in proximity to in-coming Aurignacian Homo sapiens communities. The Neanderthals took over some practices from the new arrivals to create the Châtelperronian technocomplex. Nevertheless, by the end of the period, the Neanderthals were extinct. During contact times H. sapiens produced an efflorescence of cave and portable imagery and sometimes elaborately furnished graves. These practices, together with diversifying technologies, continued to characterise H. sapiens communities until the end of the Upper Palaeolithic – and beyond.1 That skeletal account raises numerous issues. Here, I am concerned with only two, the way in which interaction between the two species of Homo may have contributed to the generation of image-making, and how the neurological mechanism necessary for image-making was integral to the formation of human communities as we know them in all their diversity. I use ‘image’ and ‘image-making’ instead of ‘art’ because I argue that art and aesthetics arose after the first images had been made. There is no innate, adaptive human drive to make things beautiful or ‘special’, as evolutionary theories of the origins of art have argued (e.g. Dissanayake, 1995). On this point, I follow the art historian Ernst Gombrich (1972: 4) – though recent discoveries show that the earliest images were not, as he puts it, merely ‘roughed out’: There really is no such thing as Art. There are only artists. Once these were men who took coloured earth and roughed out the forms of a bison on the wall of the cave; today some buy their paints, and design posters for the hoardings; they did and do many other things. There is no harm in calling all these activities art as long as we keep in mind that such a word may mean very different things in different times and places, and as long as we realize that Art with a capital A has no existence.

Borrowing from neighbours Table 1 summarises what Neanderthals took over from in-coming H. sapiens Aurignacian communities and what they did not. Some researchers question the notion of borrowing and argue that Neanderthals may have independently developed the apparent borrowings before they were in contact with Aurignacians. A point to bear

373

From Tools to Symbols

in mind here is what I call the ‘advance ripple effect’, or diffusion. New technologies which Neanderthals were mentally equipped to take up could have long preceded faceto-face contact with Aurignacians. Ideas can move swiftly through hunter-gatherer communities scattered over hundreds of kilometres. This phenomenon has been observed in the Kalahari Desert of southern Africa. I therefore accept the view that, after thousands of years of adequate living, Neanderthals of the terminal Mousterian adopted certain features of Aurignacian life, but not others. This selectivity needs to be explained. Table 1 Artefacts and practices that Châtelperronian Neanderthals borrowed and did not borrow from Aurignacian H. sapiens communities. What the Châtelperronian Neanderthals… …borrowed

…did not borrow

stone tool techniques advanced hunting strategies blades end-scrapers burins burials personal ornament

painting bone & antler engraving burials with elaborate grave goods

It is reasonable to assume that the Neanderthals employed stone tools made by new technologies for the same purposes as the Aurignacians did. Without any difficulty, they used flint scrapers for scraping and blades for cutting. They were, after all, familiar with those tasks, even if they did not employ them in advanced hunting and subsistence strategies. Personal ornaments were a different matter altogether. Some, but not many, items of this kind have been found in Neanderthal levels. Whether they were made by Neanderthals or whether they obtained them from Aurignacians by borrowing or stealth is disputed. I am inclined to think that Neanderthals did not make them themselves, but a more pertinent question concerns what they meant to the Neanderthals. Today it is widely accepted that decorations of the human body refer to and reproduce social distinctions that may vary from situation to situation. People don decorations to say something about their relations with other people in specific contexts. Few, if any, researchers would argue that the Neanderthals had exactly the same social and kinship systems as H. sapiens communities. The archaeological evidence for Neanderthal social and subsistence strategies suggests that they probably did not recognise complex discriminations and classifications, such as differences

374

New neighbours

between consanguineal and affinal relations, multiple generations, and other often quite subtle distinctions that exist today in all human communities. That being so, the ornaments found in Châtelperronian levels must have meant something different to them from the social statuses they signified for Aurignacians. The same distinction probably applied to (sometimes disputed) Neanderthal burials (Gargett, 1989, 1999; Riel-Salvatore & Clark, 2001). If the later Neanderthals did bury (some of) their dead, they did not include the elaborate grave goods that are characteristic of many H. sapiens graves. This omission implies that the Châtelperronians understood something different about buried corpses from the beliefs that H. sapiens people entertained. As with personal ornaments, it was the form of burials rather than their meaning and detailed symbolism that Neanderthals borrowed. To these differences between Neanderthals and H. sapiens communities we must add what is the most striking distinction, the one that has given rise to most discussion. In-coming Aurignacians and their successors made images of animals and, more rarely, anthropomorphs on cave walls and in open rock shelters, as well as finely carved pieces of art mobilier. They also made positive and negative handprints, a seemingly more direct and mechanical practice than image-making (but see Lewis-Williams, 2002: 216–220). There is no evidence that Neanderthals took over the simple process of making handprints, let alone image-making. How can Neanderthal borrowing and not-borrowing be explained?

Intelligence and consciousness Accounts of human evolution almost invariably emphasise intelligence and ignore human consciousness. For most researchers, hominids became more and more intelligent and hence more able to adapt to different environments and to make more complex artefacts. Intelligence and adaptation are the cornerstones of evolutionary theory, and they are routinely applied to the West European Middle to Upper Palaeolithic Transition. But a difference in intelligence alone does not sufficiently explain the distinctions between what the Neanderthals borrowed and what they ignored. Without questioning the importance of intelligence in many contexts, I point to the centrality of consciousness. It is differences in consciousness that most adequately explain the differences between the Neanderthal and H. sapiens communities of the Transition. To make this point, I outline the nature of modern human consciousness and then distinguish between two types of consciousness.

The spectrum of consciousness Human consciousness is not a unitary mental state. Rather, it is better thought of as a spectrum that grades from alert states through more meditative conditions, to

375

From Tools to Symbols

day-dreaming, to dreaming, and on to ‘unconsciousness’ (e.g. Martindale, 1981). The spectrum has alert and autistic (inward-directed) ends. Daily, all people move back and forth along this spectrum. In the very nature of communal life, all communities are obliged to divide up the spectrum of consciousness into recognised and, importantly, evaluated sections. In modern Western life it is the alert end of the spectrum that is most valued, for it is in alert states that scientific and technological advances are believed to be generated. Day-dreaming and fantasising are treated with scepticism. Dreams are generally regarded as amusing, or, in some instances, terrifying, but no great significance is accorded them, except by a couple of schools of psychology. If an answer to a problem comes to a researcher in a dream, its origin is no argument for its value: the answer has to be tested under rigorous conditions. This pattern of evaluation of the spectrum of consciousness was not always embraced in the West, nor is it universally accepted today. In mediaeval times dreams were believed to be one of the ways in which God spoke to his chosen, or one of the routes by which the Devil infiltrated himself into human souls. Certainly, other cultures place different values on dreams. They may be believed to be the voices of ancestors or intimations of hidden witchcraft. Consequently, we cannot assume that Upper Palaeolithic people evaluated the spectrum of consciousness in the same way that Westerners do today. For whatever reasons, people in all times and cultures interfere with the spectrum of normal consciousness and launch it on an intensified trajectory. This trajectory leads through stages of what we call altered consciousness to a state in which subjects experience visual, somatic, aural, gustatory and olfactory hallucinations (LewisWilliams, 2002). It may be entered upon by the ingestion of psychotropic substances or by audio and rhythmic driving, intense concentration and meditation, sensory deprivation, prolonged pain, or pathological conditions such as temporal lobe epilepsy and schizophrenia. The intensified trajectory is also evaluated; it is socialised and people in a given community agree on what it signifies, though there are often those who contest the generally accepted evaluation. The process of socialisation entails parts of the trajectory being marked off as the preserve of special people, the ‘seers’, those who ‘see’ and thereby have access to realms that ordinary people can only glimpse in their dreams. The seers often control access to the intensified trajectory, and becoming a seer involves learning to experience and understand the stages of altered consciousness that may, but not necessarily, lead to full hallucinations. The spectrum of consciousness thus becomes a foundation for social discrimination that cross-cuts stratifications based on sex, age and brute strength.

376

New neighbours

Two types of consciousness Gerald Edelman (1992; Edelman & Tononi, 2000) distinguishes two kinds of consciousness, primary and higher-order consciousness. Primary consciousness is not a single, unitary condition. Animals other than human beings experience it to varying degrees; early hominids had it, also in different degrees. It entails an awareness of the environment and the entertaining of mental images in the present. Edelman likens it to being in a dark room with a flashlight. As the beam moves around, it illuminates parts of the room and then lets them slip back into darkness. Primary consciousness thus permits the construction of an integrated mental scene in the present that does not require language or a true sense of self. This is what Edelman calls ‘a remembered present’. As a result of its essential restriction to the present, animals with primary consciousness do not have a sense of a person with a past and a future. They have some long-term memory, but they are unable to plan an extended future based on memory. Some animals may have what may be called a protolanguage, but one without past and future tenses. In short, they can have no socially constructed self. A consequence of primary consciousness is that animals with it dream (sleep is a condition that leads to the manufacture of proteins in the brain) but are unable to remember and socialise their dreams. Autistic altered states of consciousness of the intensified trajectory may be induced in them, but they do not remember and later act upon socially constructed understandings of those experiences. Higher-order consciousness is the kind of consciousness experienced today by all human beings. According to Edelman, it evolved out of primary consciousness not just by gross changes in brain morphology but, more importantly, by the establishment of what he calls ‘reentry’ circuitry. This new ‘wiring’ within the brain contributed substantially to a new order of complexity (he uses the phrase ‘meshwork of the thalamocortical system’) and to the integration of conscious experience (Edelman & Tononi, 2000: 216). Because Upper Palaeolithic H. sapiens communities were anatomically fully modern, we can assume that their neurological wiring was the same as it is today. We therefore have a neurological bridge to the Upper Palaeolithic. Higher-order consciousness is characterised by recognition of one’s own acts and emotions, concepts of a deep past, a future model of the world, a socially constructed self and long-term storage of socially constructed symbolic relations. This kind of consciousness is founded on complex language that embraces past, present and future tenses and the utterance of never-before articulated sentences that can be understood by members of the language community. It makes possible long-term planning and strategising, the maintenance of complex kinship systems, and the manipulation of symbols to express and impact upon complex social relations. In addition, higherorder consciousness permits people to remember and to socialise dreams and visions.

377

From Tools to Symbols

This kind of consciousness probably evolved in Africa, so that by the time H. sapiens communities reached Western Europe they had the potential to achieve all of its functions. Discussions of ‘modern human behaviour’, a concept so difficult to define, usually omit any mention of consciousness. Yet it is higher-order consciousness that makes modernity possible. Modern human behaviour came together piecemeal and spasmodically in Africa (McBrearty & Brooks, 2000) on, I argue, a foundation of higher-order consciousness. It cannot be understood without an appreciation of that sort of consciousness.

The consciousness hypothesis The hypothesis that Middle to Upper Palaeolithic Transition H. sapiens communities had higher-order consciousness and that the Neanderthals had a form of primary consciousness explains why Neanderthals were able to borrow certain things but not others, and why H. sapiens supplanted the Neanderthals. I focus on burials and imagemaking. Without higher-order consciousness Neanderthals would have been unable to conceive of a spirit realm or an afterlife, a state of being far removed in the future or, rather, outside of time. They would therefore have been incapable of understanding the placing of valuable items in H. sapiens graves; grave goods could have had no meaning for them. Nor could they have had any clear idea of why some H. sapiens people were given elaborate burials and others were not. Without any notion of social distinctions based on criteria other than age, sex and strength they would have been baffled as to why someone who seemed to have so little going for him or her would be accorded such treatment. They may have been able to see some purpose in placing bodies beneath the ground, if only to emulate H. sapiens behaviour, but they could not have entertained any religious concepts about death and burial. Neanderthals were congenital atheists. Upper Palaeolithic image-making poses more complex problems than burials. First, we need to ask how people came to believe that small, static marks on a twodimensional surface could represent a huge, live, moving, three-dimensional bison or horse. The conventions of two-dimensional representations are not inherited; they have to be learned. This is what the anthropologist Anthony Forge (1970) found in New Guinea. The Abelam, among who he worked, did not understand photographs. When he explained the conventions to them, they readily learned to ‘see’ photographs, but the ability was not in-built. It is this point that exposes the inadequacy of some of the old explanations of image-making. People could not have fortuitously discerned the outline of, say, a bison in natural or random human-made marks on cave walls without first having a concept of two-dimensional pictures.

378

New neighbours

The notion that an especially intelligent individual noticed such an outline and then told others about it, or simply invented image-making, is therefore flawed. If the origin of image-making was idiosyncratic in this way, we cannot explain why, right from the beginning of image-making over the whole of Western Europe, the makers confined themselves to a fairly restricted set of animals: bison, horses, felines, aurochs. Other creatures, such as birds, mammoths and reindeer, were less frequently depicted; still other motifs, such as human beings, are extremely rare. Proportions between depicted species varied through the Upper Palaeolithic, but the fundamental set of motifs remained the same. The most plausible explanation is that the vocabulary of motifs existed before people started making images of them. There was a shared, socialised bestiary of animals with symbolic associations that informed the beginnings of image-making. Along with stone tool technologies, the in-coming Aurignacians brought an established religion and symbology, though not highly developed and with no tradition of subterranean image-making.

An origin of image-making The answer to the question of how people came to understand and make twodimensional images of this bestiary lies in features of higher-order consciousness and the intensified trajectory (Lewis-Williams & Dowson, 1988; Lewis-Williams, 2002).2 Neuropsychological research has shown that hallucinations experienced on the intensified trajectory are projected onto plane surfaces, such as walls or ceilings. Subjects liken this experience to a slide or film show (Klüver, 1926: 505, 506; Siegel & Jarvik, 1975: 109; Siegel, 1977: 134). Non-veridical mental imagery can thus come to be associated with a surface in front of the subject. Two implications flow from this observation. First, if some Upper Palaeolithic people experienced projected imagery under certain circumstances, as they had the neurological potential to do, their world would already have been invested with two-dimensional ‘pictures’. They did not have to invent two-dimensional imagery. Ordinary, daily mental imagery (the kind of ‘imagining’ that everyone knows) was wired into their brains, as it is into the brains of all H. sapiens. Projected mental imagery (hallucinations) would necessarily have been associated with altered states; it could not have derived from ‘normal’, alert consciousness or been directly associated with experiences in the countryside, such as hunting. Secondly, if the vocabulary of motifs was established before image-making started, as the archaeological evidence shows it was, the projected images of the selected animals must have had symbolic significance beyond being ‘pictures’ of creatures seen in daily life. The projected images must have had some value or ‘power’. It would then have been a short step for people to reach out to touch their projected images and thus to

379

From Tools to Symbols

fix them on the walls with a finger in soft surfaces or with pigment. Or they may have made images in an attempt to recall, to re-create, mental images on the surfaces on which they had had been projected while they were in a deeper state. The first images were therefore not ‘pictures of real life’ but rather fixed, projected mental images of important animals that were part of an established symbolic bestiary. As Forge was easily able to teach the Abelam the conventions of two-dimensional imagery, so too would Upper Palaeolithic seers have been able to teach others to comprehend the projected imagery that they fixed on the rock walls. This explanation for the origin of two-dimensional image-making is supported by characteristics of the painted and engraved images in the caves of south-west Europe. Researchers have long noticed that they appear to float on the walls without any suggestion of a ground line or other context – there are no trees, grass or hills. Moreover, the images are often integrated with the convolutions of the surface; it is as though the projected mental image locked into features of the rock wall. As a result, some images are in vertical positions, not as animals would have been seen outside the cave. In other instances, it is necessary to hold one’s lamp in a given position so that the shadows cast across the wall form, say, the dorsal line of a bison. The image-fixer then added legs and perhaps horns to complete the figure. If the viewer moves the lamp back and forth, the image disappears and re-appears. An impression of ‘appearing’ is also created by images that are positioned so that they seem to be coming out of cracks or fissures in the rock walls. The intimacy of the relation between images and surfaces leads to inferences about the caves and their topography.

Subterranean realms Perhaps the most striking feature of Upper Palaeolithic cave images is their location in deep underground chambers, passages and small niches. In some instances, people walked, waded, crawled and squeezed through narrow openings for more than a kilometre underground before they made images. These were not ‘art galleries’ for leisurely contemplation of objets d’art. Some of the remote images may never have been seen by anyone apart from their makers; others, by contrast, are in large chambers that could have accommodated a number of people. What did ‘viewing images’ mean to Upper Palaeolithic people? The answer to this question lies again in the spectrum of human consciousness and its intensified trajectory. Two hallucinatory experiences are wired into the brain. As subjects move along the intensified trajectory, they may experience a sensation of rising up, attenuation and flying; or they may experience a vortex, a constricting tunnel, through which they pass and emerge on the other side into a realm of

380

New neighbours

hallucinations where they ‘see’ monsters, people and bizarre events. They may feel themselves transformed, partially or completely, into animals. It is, I argue, these hardwired experiences that give rise to worldwide beliefs in spirit realms above and below the level of daily life. The Christian concept of Heaven above and Hell below is but one example. In some instances, the upper and nether realms are subdivided into multiple levels. There is, of course, no evidence in everyday life for the existence of such spheres of existence, yet people everywhere believe in them. The only persuasive explanation for this universality is that the experiences are hard-wired (Fig. 1). A belief in a tiered cosmos explains why Upper Palaeolithic people went underground to make images of a set of symbolic animals and why those images are intimately integrated with the walls of the caves. Physical entry into the caves paralleled, was perhaps indistinguishable from, mental entry into the neurologically generated vortex that leads to deeply altered states and hallucinations. The caves were the nether world, a realm inhabited by powerful animals. These animals lived behind the walls of the cave. They were sought by sight, touch, the interplay of light and shadow, and probably by responding to aural hallucinations of animal sounds. It was the aim of the deeply penetrating quester to ‘see’ these animals, sometimes to draw

Tiered Cosmos Neuropsychology

Three Tiers

Weightlessness

Spirit world above

Dissociation Attenuation

(Flight)

Altered states of consciousness

Vortex Difficulty in breathing Sound in ears

Daily life

Spirit realm below (underground, underwater)

Figure 1 The three-tiered cosmos, as generated by hard-wired experiences of altered states of consciousness.

381

From Tools to Symbols

them through the ‘membrane’ of the walls and to ‘fix’ them there. They were fixed visions of special, powerful animals whose essence image-seekers and image-makers desired. The context of the images was not ‘Nature’ but rather the nether realm and the membrane of the cave wall. The image-makers went into the cave to seek visions of animals; they did not take recollections of real animals and incidents with them so that they could paint or engrave them. What, then, was the relationship between the ‘fixed’ spirit animals and their lookalikes that roamed outside the cave? Worldwide hunter-gatherer ethnography offers some clues. For instance, the nineteenth-century /Xam San of southern Africa believed that some of their !gi:ten (ritual specialists, shamans) possessed spirit animals that could be caused to go among a herd of springbok and lead them in the direction of the hunters’ ambush. One !gi:xa described such an animal as her ‘heart’s springbok’ (Bleek, MS L.V.10.4729 rev.). Unless they behaved in some peculiar way, spirit animals were indistinguishable from real animals, and an informant told how one was accidentally shot, with unfortunate consequences (Bleek, 1935: 44–47). Another /Xam person said that her husband had hunted a giraffe that unexpectedly turned out to be a spirit animal; the man then acquired supernatural potency from it (Biesele, 1993: 68–69). Such beliefs are a reflection of the immanence and simultaneous transcendence of the spirit world. Generated in a person’s head, the spirit world is both ‘with’ one and ‘beyond’ one. Further hints come from the South American Desana, who speak of Vaí-mashë, Lord of the Animals, a being who has control of animals that he keeps in spiritual form in his maloca, or house (Reichel-Dolmatoff, 1971: 80–86). He is associated with isolated rock formations that rise out of the Amazonian forest ‘like dark islands on the horizon’. In their ‘caverns and dark recesses’ there are painted images made by shamans, who alone are strong enough to visit such dread places. They do so in reality and in trance (induced by inhaling the narcotic powder of vihó, the plant Piptadenia). Once in the presence of Vaí-mashë, they negotiate with him to release animals for the benefit of hunters. In addition to being the place where supernatural animals and beings reside, the rock formations are said to contain illness and their dark and inhospitable aspect indicates danger. The cracks, caverns, and tunnels are the entrances to the interior of the hills, to the great malocas of the animals. There, within their dark interior, the gigantic prototypes of each species exist and thousands of animals are kept (Reichel-Dolmatoff, 1971: 81).

In some West European caves, such as Enlène, people pushed small pieces of bone into cracks in the walls. In the small Chamber of the Lions, in the adjacent cave Les Trois

382

New neighbours

Frères, a cave bear tooth was placed in a small niche. The pieces of bone in Enlène are so small and the cracks so narrow that intense concentration and close scrutiny of the cave wall must have been necessary. In the total darkness of the chambers the person’s flickering tallow lamp must have sharply focused his or her attention. People were carefully passing pieces of animals back through the membrane, perhaps propitiating the spirits behind the rock wall, perhaps hoping that the pieces would reconstitute themselves into spirit animals which some guiding being, perhaps a Lord of the Animals, would later release into the outside world as real animals that could be hunted. Whilst direct, one-to-one parallels with ethnographically recorded practices, such as the San and Desana reports I have cited, are naive and potentially misleading, they do give us something of the flavour of the relationship that Upper Palaeolithic people may have perceived between real animals, spirit animals, and the caves. Upper Palaeolithic society and belief almost certainly have no present-day ethnographic analogues. But, given the hard-wired experiences of human consciousness, we can formulate generalities that provide a multi-component context for tentative interpretation of Upper Palaeolithic image-making. In some instances, the ethnography records the hallucinogen that people used to access the spirit world, but in others, such as the San, no psychotropic substances are used today. In the West European Upper Palaeolithic the altered state of consciousness necessary to induce projected mental imagery could have been generated by any of the mechanisms I listed earlier. But it is worth noting that sensory deprivation of the kind experienced in the totally dark, silent caves induces altered states of consciousness. To expectant minds, isolated for long enough in the bowels of the earth, images would have appeared. Some images are, however, so large that they could have been made only by cooperating people. Those in the Hall of the Bulls in Lascaux are examples. Elaborate images such as these are usually in large chambers and are often fairly close to the cave entrance or easily accessed. These images, I suggest, were communally produced rather than the result of the sort of individual experiences that led to the limning of the more briefly made images in the spatially constricted depths.

Social differentiation There were thus what I call ‘activity areas’. These were selected places in the caves where images of different kinds were executed by means of different techniques (some by a few solitary, deft strokes, others by pooled labour) or where other kinds of ritual, such as chanting and dancing, took place. An implication of this conclusion is that the caves were templates for social discriminations (Lewis-Williams, 1997). Temples are always social templates.

383

From Tools to Symbols

An idealised cave helps to understand this proposition (Fig. 2). It seems likely that the entire community was associated in multiple ways with the land beyond the entrance to a cave. A large entrance chamber (if there was one) near or easily accessible from the entrance contained communally made, striking images that prepared novices, or even experienced seers, for what they would ‘see’ in the isolated depths. The content of hallucinations is always culturally informed and situated; people hallucinate what they expect to hallucinate. Large communally produced images, dramatically revealed, perhaps enveloped in a nimbus of chanting, would have informed the mental imagery of subsequent altered states. Such images may not have originated in a single vision; they were probably composite, socially stabilised visions, the co-operative manufacture of which expressed and constructed their significance not just for individuals but for society as a whole. Certainly, they were not made by people in deeply altered states of consciousness. But they none the less derived from the symbolic bestiary and contributed to the reproduction of that bestiary. A select group that exercised control over the embellishment of the subterranean world was probably associated with the large chamber. Beyond that, the narrow passages and small niches ensured that only a few people could penetrate far underground – far into the nether spirit realm. The intensified trajectory was thus paralleled by the topography of the caves: different states of consciousness were associated with different

Figure 2 An idealised Upper Palaeolithic cave, showing how the topographic form, in concert with altered states of consciousness, acted as a template for social discriminations (from Lewis-Williams, 2002: Fig. 61).

384

New neighbours

activity areas. Outside the cave everyone could experience dreams, euphoria, awe or fear; deeper altered consciousness was experienced by a group in front of impressive images; in the remote parts of the cave a few experienced deeply altered consciousness and its vivid visions. Both the cave and the parallel intensified trajectory of altered consciousness could be protected and made available to only certain people. Social distinctions were thus expressed and constructed neurologically and spatially. Each cave is different in its topography. There must have been a two-way interaction, a reciprocity, between peoples’ ideas about the nether realm and the actual chambers and passages in any given cave. People adapted each cave to their needs and, in doing so, their understanding of the nether realm was modified. When they embellished the chambers and passages, they were elaborating a spirit world, and, at the same time, bringing more and more of the supernatural realm into contact with selected people. Image-making and its topographical locations necessarily had social consequences. Moreover, the spirit realm and its elaborations were verifiable by those who were permitted to enter the cave itself and, in some cases, to experience altered states of consciousness. Even ordinary people could glimpse the spirit world in their dreams. But complete consensus is seldom obtained in such matters. Both the intensified trajectory and its concomitant social and topographic distinctions were probably contested. The discriminations and significances associated with them almost certainly changed during the Upper Palaeolithic. Both the caves and human consciousness were sites of contestation and active social negotiation. As Stephen Shennan rightly points out, ‘[A] key locus for the generation of social inequality in forager societies was the cultural transmission of ritual knowledge, even in the absence of material inequalities’ (Shennan, 2002: 224). Roy Rappaport adds the foundation for ‘ritual knowledge’: ‘The relationship between alterations of the social condition and alterations of consciousness is not a simple one, but it is safe to say that they augment and abet each other’ (Rappaport, 1999: 219).

A nexus of origins This brief account of a more wide-ranging argument (Lewis-Williams, 2002) points to a period in which some of the entities that we today consider to be different were in fact integrated. As H. sapiens communities became aware of the implications of the differences in consciousness between themselves and Neanderthals, they emphasised those distinctions. They consolidated their own corporate identity by creating social and conceptual distance between themselves and the Neanderthals, whose form of primary consciousness, though more advanced than that of other animals, prevented them from entertaining mental imagery, remembering their dreams and visions, conceiving of

385

From Tools to Symbols

spiritual worlds, and making images. We do not know exactly what the circumstances of contact were. Perhaps the Neanderthals were able to learn a ‘pidgin Aurignacian’ language and were thus able to establish some linguistic contact with their new neighbours. But they were shut out from mental and fixed imagery. Simultaneously, the making of imagery to distinguish themselves from the Neanderthals had effects within H. sapiens communities. The practice of making images itself entrenched social distinctions that probably already existed in the incoming communities and that were linked to the bestiary of mental imagery. The manifestation of the spirit world became more and more important not only amongst themselves and their Neanderthal neighbours but also within their own communities. At the Transition, social and mental mechanisms were initiated, together with their ritual and image-making concomitants. Those mechanisms continued to play a social role as the Upper Palaeolithic unfolded. People were making their own history in a fashion that the Neanderthals did not. In this way, cosmology, religion, image-making and social discrimination came together. At that time, I argue, it was impossible to distinguish between them: • A tiered cosmology was fashioned out of hard-wired flying and vortex experiences. • Religion was traversal of the tiered cosmology by means of altered states and with concomitant social and psychological effects. • Image-making was a religious ritual aimed, at least initially, at fixing and controlling powerful visions and thereby creating social distinctions. • Social differences were more and more founded on criteria that went beyond age, sex and strength and that derived from differential access to mental and subterranean experiences. Society as we know it in all its complexity, cross-cutting discriminations, rituals and ‘art’ was burgeoning, an extended, complex process begun long before in Africa. The hypothesis I have adumbrated suggests that it is futile to look for the ‘origins of art’. Art and an aesthetic sense are always socially and historically situated concepts. We should rather seek circumstances in which people began to make images. Following Gombrich, I conclude that there is no such thing as Art; there are only artists – or, during the Upper Palaeolithic, image-makers. Instead of looking for the origins of an indefinable phantom, researchers should seek evidence for activities performed by people in social contexts.

Notes 1. This article derives from Lewis-Williams (2002), where full references and contexts will be found. See also Clottes & Lewis-Williams (1996). For

386

New neighbours

overviews of the European Middle to Upper Palaeolithic Transition see, amongst many others, Hublin (2000), Mellars (1990, 1996, 2000), Stringer & Gamble (1993). 2. Here I deal with two-dimensional images only. For discussion of the comparable origin of three-dimensional carvings, such as those from Vogelherd, see LewisWilliams (2002: 196–202).

Acknowledgements I am grateful to colleagues who kindly commented on drafts of this article: Geoff Blundell, David Pearce and Ben Smith. Table 1 and Figure 1 were prepared by David Pearce. Figure 2 was drawn by Sam Challis. The Librarian, Jagger Library, University of Cape Town, permitted quotation from the Bleek and Lloyd Collection of nineteenthcentury manuscripts. The Rock Art Research Institute is funded by the University of the Witwatersrand, the National Research Foundation (grant numbers: 2053693 and 2053470), and Anglo-American. Views expressed here are not necessarily those of the funding agencies.

References Biesele, M. (1993). Women like Meat: The Folklore and Foraging Ideology of the Kalahari Ju/’hoan. Johannesburg: Witwatersrand University Press. Bleek, D.F. (1935). Beliefs and customs of the /Xam Bushmen. Part VII: Sorcerers. Bantu Studies 9, 1–47. Clottes, J. & Lewis-Williams, J.D. (1996). Les Chamanes de la Préhistoire: Transe et Magie dans les Grottes Ornées. Paris: Le Seuil. English edition (1998) The Shamans of Prehistory: Trance and Magic in the Painted Caves. New York: Harry Abrams. Dissanayake, E. (1995). Chimera, spandrel, or adaptation: conceptualizing art in human adaptation. Human Nature 6, 99–117. Edelman, G.M. (1992). Bright Air, Brilliant Fire: On the Matter of the Mind. Harmondsworth: Penguin. Edelman, G.M. & Tononi, G. (2000). Consciousness: How Matter Becomes Imagination. Harmondsworth: Penguin. Forge, A. (1970). Learning to see in New Guinea. In (P. Meyer, Ed.) Socialization: The Approach from Social Anthropology, pp. 269–290. Gargett, R.H. (1989). Grave shortcomings: the evidence for Neanderthal burial. Current Anthropology 30, 157–190. Gargett, R.H. (1999). Middle Palaeolithic burial is not a dead issue: the view from Qafzah, SaintCésaire, Kebara, Amud and Dederiyeh. Journal of Human Evolution 37, 27–90. Gombrich, E.H. (1972). The Story of Art. London: Phaidon.

387

From Tools to Symbols Hublin, J.J. (2000). Modern-nonmodern hominid interactions: a Mediterranean perspective. In (O. Bar-Yosef & D.R. Pilbeam, Eds) The Geography of Neanderthals and Modern Humans in Europe and the Greater Mediterranean, pp. 157–182. Cambridge, Mass.: Peabody Museum of Archaeology and Ethnology. Klüver, H. (1926). Mescal visions and eidetic vision. American Journal of Psychology 37, 502– 515. Lewis-Williams, J.D. (1997). Art, agency and altered consciousness: a motif in French (Quercy) Upper Palaeolithic parietal art. Antiquity 71, 810–830. Lewis-Williams, J.D. (2002). The Mind in the Cave: Consciousness and the Origins of Art. London & New York: Thames & Hudson. Lewis-Williams, J.D. & Dowson, T.A. (1988). The signs of all times: entoptic phenomena in Upper Palaeolithic art. Current Anthropology 29, 201–245. Martindale, C. (1981). Cognition and Consciousness. Homewood, Illinois: Dorsey Press. McBrearty, S. & Brooks, A.S. (2000). The revolution that wasn’t: a new interpretation of the origin of modern human behaviour. Journal of Human Evolution 39, 453–563. Mellars, P.A. (Ed.) (1990). The Emergence of Modern Humans: An Archaeological Perspective. Edinburgh: Edinburgh University Press. Mellars, P.A. (1996). The Neanderthal Legacy: An Archaeological Perspective from Western Europe. Princeton: Princeton University Press. Mellars, P.A. (2000). The archaeological records of the Neanderthal-modern human transition in France. In (O. Bar-Yosef & D.R. Pilbeam, Eds) The Geography of Neanderthals and Modern Humans in Europe and the Greater Mediterranean, pp. 35–47. Cambridge, Mass.: Peabody Museum of Archaeology and Ethnology. Reichel-Dolmatoff, G. (1971). Amazonian Cosmos: The Sexual and Religious Symbolism of the Tukano Indians. Chicago: University of Chicago Press. Riel-Salvatore, J. & Clark, G.A. (2001). Middle and early Upper Palaeolithic burials and the use of chronotypology in contemporary Palaeolithic research. Current Anthropology 42, 449–479. Shennan, S. (2002). Genes, Memes and Human History: Darwinian Archaeology and Cultural Evolution. London: Thames & Hudson. Siegel, R.K. (1977) Hallucinations. Scientific American 237, 132–140. Siegel, R.K. & Jarvik, M.E. (1975) Drug-induced hallucinations in animals and man. In (R.K. Siegel & L.J. West, Eds) Hallucinations: Behaviour, Experience and Theory, pp. 81–161. New York: Wiley. Stringer, C. & Gamble, C. (1993). In Search of the Neanderthals. London & New York: Thames & Hudson. Rappaport, R.A. (1999). Ritual and Religion in the Making of Humanity. Cambridge: Cambridge University Press.

388

Late Mousterian lithic technology: its implications for the pace of the emergence of behavioural modernity and the relationship between behavioural modernity and biological modernity Marie Soressi Max Planck Institute for Evolutionary Anthropology, Deutscher Platz 6, D-04103 Leipzig, Germany PACEA/UMR 5199 du CNRS, Institut de Préhistoire et de Géologie du Quaternaire, UFR de Géologie, Bat. B18, Avenue des Facultés 33405 Talence, France

Abstract An analysis is presented of several Mousterian industries of Acheulian tradition from Western Europe dated to the first half of IOS 3 and manufactured by Neanderthals before the arrival of anatomically modern humans in Europe. It is shown that some behaviours previously thought to be characteristic of recent behaviours associated with anatomically modern humans were in fact shared with another species. Among those are: the variability of Mousterian technologies across time and space; the use of Upper Palaeolithic methods of production immediately prior to the arrival of anatomically modern humans in Europe; and the long-term planning of knapping activities across the territory. This paper also demonstrates that some of these specific behaviours (the scheduling of lithic tool production within the territory) might eventually have been abandoned by Neanderthals, while others (the use of a volumetric method of producing blanks) were kept alive by them. These results show that models of the development of behavioural ‘modernity’ have to take into consideration every line of evidence, including the testimony of the behaviour of anatomically non-modern humans. We do not have to consider a priori that anatomically modern humans were better suited or were the only ones capable of behavioural ‘modernity’. On the contrary, it is necessary to demonstrate how they were better adapted than Neanderthals. Evolutionary trajectories might be punctuated, and resulting from a combination of biological and contingent events which created a patchwork of changes.

389

From Tools to Symbols

Résumé Le lien causal entre la ‘modernité’ culturelle, synthèse des changements comportementaux qui deviendront la norme à la fin du Pléistocène, et la modernité biologique doit être discuté à partir de l’ensemble des documents dont nous disposons. L’étude des comportements des Néandertaliens avant l’arrivée des hommes anatomiquement modernes en Europe, en s’appuyant sur les industries lithique MTA de la première moitié du stade isotopique 3, montre que des comportements d’abord jugés spécifiques de notre espèce ont en fait été partagés avec les Néandertaliens. Elle montre également que certains de ces comportements, comme l’organisation à long terme de l’activité de taille dans le territoire, ont été finalement abandonnés par les Néandertaliens, tandis que d’autres, comme l’utilisation d’une méthode de taille volumétrique, ont été conservés par les mêmes Néandertaliens. Nous ne devons pas a priori exclure les Néandertaliens d’une ‘modernité’ comportementale. Au contraire, nous devons rechercher, par une analyse comparée précise des comportements des Néandertaliens et des premiers hommes anatomiquement modernes en Europe, la nature des avantages de ce dernier groupe. L’évolution humaine pourrait bien être ponctuée et résulter d’une combinaison d’évènements biologiques et contingents qui ont créé un patchwork de changements.

Introduction By 30 000 years ago, almost all humanity had adopted a set of new behaviours generally defined as ‘modern’. These behaviours are qualified as ‘modern’1 because there is clear continuity between them and the set of behaviours of historic huntergatherers, and because of the apparent coincidence in Europe of these behavioural innovations with the expansion of anatomically modern human populations, while Neanderthals were in the process of disappearing (Klein, 2000). Behavioural modernity is a key question regarding the behavioural differences between us, anatomically modern humans, and our earlier ancestors. It is tackling the relationship between behavioural modernity and anatomical modernity (McBrearty & Brooks, 2000; Klein, 1998, 2000; Mellars, 1989, 1998). Yet, speaking of behavioural modernity certainly does not imply that behaviourally modern people have to be anatomically modern ones (see Chase & Dibble, 1990; Zilhão, 2001; d’Errico, 2003). We have to consider biological evolution and behavioural evolution separately, at least at the initial stage. At a later stage, we should progress towards interpreting the relationship between them. Then, is behavioural modernity linked to modern biology, or could it have arisen within the Neanderthal lineage? Furthermore, does it mean that ‘behavioural modernity’ is a dead concept if it is not specific to our species? Three main models have been proposed to reconstruct the development of behavioural modernity. 1. After a selectively advantageous genetic mutation within anatomically modern humans (Klein, 1998, 2000), ‘modern’ behaviours would have developed rapidly

390

Late Mousterian lithic technology

and then dispersed from Africa to Asia and Europe at around 50–40 Ky BP (Klein, 1998, 2000; Ambrose, 1998). Characterised by the use of symbols and a fully elaborated language, the sudden development of behavioural modernity can be considered a ‘revolution’ (Klein, 1989, 1994, 1995, 2000; Mellars & Stringer, 1989; Diamond, 1992; Mellars, 1996a, b; Noble & Davidson, 1991; Bar-Yosef, 1998; Wadley, 2001; Henshilwood & Marean, 2003). 2. Others support an earlier development of modern human behaviours (Deacon, 1989; Brooks et al., 1995; Knight et al., 1995; Barham, 1998; Watts, 1999; Henshilwood et al., 2002). The development of behavioural modernity would have been gradual, as the gradual appearance of several distinct modern behaviours between 250 Ky and 40 Ky BP demonstrates (Deacon & Deacon, 1999; McBrearty & Brooks, 2000; Deacon & Wurz, 2001; Barham, 2001). Both of these models imply that behavioural modernity arose only within anatomically modern humans. Indeed genetic and fossil evidence seems to favour a single origin – or an ‘Out of Africa’ model – over the continuity model for the appearance of modern humans in Europe and Asia (e.g. Stringer, 2003). The Out-of-Africa model is thought to imply that there were behavioural differences between the expanding modern humans and the indigenous population outside of Europe and Asia. This consequently suggests that European Neanderthals were not behaviourally modern – or at least less so than the incoming behaviourally modern humans. This idea must not be accepted without precise knowledge of the behavioural differences between the colonisers and the local populations. 3. The third model has been recently put forward by d’Errico and his colleagues (d’Errico, 2003; d’Errico et al., 2003). Extensive review of available data on Neanderthal behaviours would not support the single-species origin of behavioural modernity. Neanderthals would have contributed to the development of behavioural modernity, as well as anatomically modern humans. Analyses of European Neanderthal behaviour then provides a point of comparison of the behaviour of yet another species. It aids in defining species-specific behaviours, and in testing the tempo of acquisition of new behavioural traits that became the norm after 30 Kya. This paper focuses on two of the traits defining behavioural modernity according to the synthesis made by McBrearty and Brooks (2000): technological innovativeness and planning depth. It provides data on the archaeological signature of those traits within the Mousterian of Acheulian tradition (MTA) from the south-west of France. No anatomically modern human remains have ever been found associated with this industry, only the remains of Neanderthals (Maureille & Soressi, 2000). MTA assemblages have a precise geographic location, centred on the south-west of France

391

From Tools to Symbols

(Bordes, 1984; Mellars, 1973; and see a recent synthesis in Soressi, 2002: 6–7), and a precise time duration, enduring briefly either side of 50 Ky BP (Figs 1, 2). MTA is characterised by small and finely retouched cordiform bifaces and by backed elongated flakes as well as well formed end-scrapers and borers (Bordes, 1984; Fig. 3). The assemblages used for this analysis were the eponymous assemblages: Le Moustier layers G and H, Pech-de-l’Azé layers 4 to 7, as well as two others – La Rochette layer 7 and La Grotte XVI layer C. These were all dated by radiometric methods (see references for the dating in the legend of Fig. 2).

Stability of Mousterian technology? According to some authors, among them several who defend the Revolution model of the development of behavioural modernity, Mousterian lithic technology would be internally static and consistent through the Middle and the Late Pleistocene compared

Figure 1 Geographic spread of the Mousterian of Acheulian tradition with cordiform bifaces of the late Micoquian of Central Europe, and location of sites mentioned in the text.

392

Late Mousterian lithic technology

Figure 2 Dating on the MTA from south-west of France and Late Micoquian sites from Central Europe (after Guibert et al., 1999; Valladas et al., 1987; Soressi, 2002: 30–32, 36, and Soressi et al., in preparation; Boëda et al., 1996; Vogel & Zagwijn, 1967; Richter et al., 2000; Veil et al., 1994; Mook, 1988; and Rink et al., 1996). ESR dates are means of ages given by EU (early uptake) and LU (linear uptake) age models.

with the succeeding Upper Palaeolithic (e.g. Binford, 1989; Foley, 1997). Then, frequent technological changes would be characteristic of behavioural modernity as shown by the ethnography. Also, frequent technological changes are related to a high degree of innovation and only anatomically modern populations would have been sufficiently organised in social structures and mental capacity to allow a high degree of innovation (e.g. Wynn & Coolidge, 2004) The stability of Mousterian technology is spectacularly exemplified by the Levallois technology known in Europe since at least IOS 8 and used continuously until IOS 3. The other method frequently used in the European Mousterian, the discoid method (Boëda, 1993; Peresani, 2003), is not specific to Mousterian industries, and would have been used in Western Europe since IOS 6 at least (Jaubert & Mourre, 1996; Texier, 1996), if not since more than 780 Ky (Vaquero & Carbonell, 2003). The Quina technology (Turq, 1989; Bourguignon, 1996, 1997) occurs more frequently in Europe during IOS 4/3 and would have been present since IOS 5 (Bourguignon, 1997: 37). Yet some technologies were used in Europe for shorter periods: for instance, a volumetric

393

From Tools to Symbols

Figure 3 MTA industry: (a) small and finely retouched bifaces, (f, g, i) backed elongated flakes, (h) end-scrapers, (c) borers, (b,d) convex side-scrapers, a, b, d: Pech-de-l’Azé I, layer 4 MTA A; c: La Rochette layer 7 MTA B; e, f, g, h, i: Pech-de-l’Azé I, layer 6 or 7, MTA B (drawings by J.G. Marcillaud except (c) by S. Pasty; use-wear analysis after Anderson-Gerfaud, 1981: 112).

394

Late Mousterian lithic technology

395

From Tools to Symbols

blade production was in fashion around stage 5c in the north of France/Germany (Révillion & Tuffreau, 1994; Delagnes, 2000; and see Delagnes & Meignen, in press for a recent synthesis).2 To address the stability across time and uniformity across space of Mousterian technology, I compared the methods used to shape bifaces within the MTA and those of a contemporaneous and neighbouring industry: the late Micoquian of Central Europe (also named the industry of the Keilmesser group). The MTA and the late Micoquian are both characterised by their bifaces and both these industries show use of other technologies such as that of the Levallois or discoid (Richter, 1997; Soressi, 2002: 240–241). The late Micoquian of Central Europe is centred on Germany. The late Micoquian industry is radiometrically dated to about 50 Ky BP (Fig. 2). 3 The MTA method of shaping bifaces is characterised by the shaping of a bi-convex transverse section (Table 1), which progressively becomes plano-convex by retouching and resharpening of the edges (Table 2). The removals used to create this volume are generally struck from the lateral sides of the bifaces (Soressi, 2002: 113). Although the late Micoquian (LM) method of shaping bifaces is more variable, it never matches the MTA method. The transverse section of the LM bifaces often appears flat (Fig. 4a). These bifaces are manufactured with a special technique involving removals with a flat profile instead of the MTA convex removals. The transverse section may also involve a combination of flat and convex removals (Fig. 4b). Another LM method involves convex removals highly inclined on a flat or even concave surface (Fig. 4c). The biface longitudinal section is generally flat and convex, while the MTA one is bi-convex. Removals are frequently struck from the point (Fig. 5c), which is almost never the case on MTA bifaces. Table 1

Position of the mean plane of intersection of the two faces at the biface point relative to the volume of the transverse section of the bifaces abandoned during production stage at Le Moustier and La Rochette.

396

Moustier, G n = 22 %

Rochette, MAT n = 18 %

Central

68

61

Low

32

39

Total

100

100

Late Mousterian lithic technology Table 2

Position of the mean plane of intersection of the two faces at the biface point relative to the volume of the transverse section on bifaces abandoned during retouch stage at Pech-de-l’Azé I (layer 4). Pech-de-l’Azé I,4 n = 70 % Low

61

Central

6

Central inclined

31

Foliate

1

Total

100

Figure 4 Method of shaping bifaces used within the Late Micoquian of Central Europe, view in section (after Boëda, 1995 and Bosinski, 1967).

a a a

b b b

3 3 23 2 2

1 1 41 4 4

c c c 3 3 2 3 2 2

1 4 1 4 1 4

Might those differences between MTA and late Micoquian methods be related to the manufacture of tools designed for different uses? Available use-wear analysis shows that the MTA bifaces were used to perform variable tasks on variable materials, such as scraping wood, butchering or scraping hide (Anderson-Gerfaud, 1981: 85, Soressi & Hays, 2003), and there are examples of an MTA biface Figure 4 used to perform several different Figure bifaces 4 tasks (Fig. 6). Use-wear analyses of Micoquian are rare, but when available Figure 4 they show that some of them were used as meat knives (Veil et al., 1994). As use-wear analyses of MTA and Micoquian bifaces are still not common, another useful way to extract information on the functionality of the bifaces is to study their

397

From Tools to Symbols

Figure 5 Late Micoquian of Central Europe industry. (a) Keilmesser from Klausennische (Bavière, Germany: after Debenath & Dibble, 1994: 158); (b) Keilmesser from Sesselfelsgrotte (Saxe-Anhalt, Germany; Richter, 1997: 383); (c) Keilmesser from Lichtenberg (Basse-Saxe, Germany; Veil et al., 1994); (d) Faustkeilblätter from Klausennische (Bavière, Germany: after Debenath & Dibble, 1994: 157; (e) Faustkeilblätter from Sesselfelsgrotte (Saxe-Anhalt, Germany; Richter, 1997: 301); (f) Halbkeile from Bockstein (Rhénanie, Germany: after Debenath & Dibble, 1994: 155).

morphology. As some specialised tasks require tools of specialised morphology, studying the morphology of the bifaces would help to understand if these bifaces were specialised tools (like sharp knives or efficient scrapers) or if they were multifunctional tools.

398

Late Mousterian lithic technology

Figure 6 Bifaces from Grotte XVI layer C used on two different materials. On the smaller one, use on hide or meat probably came after the use on wood, according to the chronology of removals shaping the biface (after Soressi & Hays, 2003).

The angulation and the delineation of edges situated on both sides of the point are different from the ones of the biface bases. Point edges are acute and regular while basal edges are blunt, often abraded and irregular. The contrast between the angulation and delineation of those edges as well as the geometric opposition between them allow the inference that the edges on both sides of the point are the active edge

399

From Tools to Symbols

Figure 7 Position of active and prehensile edge on MTA bifaces, and possible multiple uses, implying the point or the long edge, according to the direction of the force applied to the biface.

while the basal edge is the passive – or prehensile – one. This hypothesis is verified by the available use-wear analyses (Anderson-Gerfaud, 1981: 85; Soressi & Hays, 2003). André Leroi-Gourhan’s (1943: 47–64) ethnographic referential demonstrates that for a tool to be utilised in a punctiform way (e.g. to bore), it merely needs a polyhedral point and a prehensile edge opposed to this point. It also shows that to be used in a linear movement (e.g. to cut or to whittle down), a tool needs a prehensile edge allowing a prehension perpendicular to the long axis of the active edge. Actually, the MTA bifaces, with their initially symmetrical active edges, allow a punctiform action. Simultaneously their morphology also permits a linear action, because the longest active edge here is sufficiently extended to allow prehension perpendicular to its long axis (Fig. 7). As many ethnographic studies have shown, the possibility of using tools to cut-in material (i.e. a vertical incision), or to cut-out (i.e. a horizontal incision), or to do both is related to the angulation of the edges. Edges close to 35° and more acute than 35°

400

Late Mousterian lithic technology

would be useful mainly for in-cutting, wider edges exceeding 65° would be useful for out-cutting only, while intermediate angulation would allow both directions of cutting (Soressi, 2002: 61–62 and references therein). In fact, the intermediate angulation of biface active edges, about 53°, allows both directions of cutting. Furthermore, the continuity of this angle toward the centre of the piece allows deep as well as superficial cuts while an abrupt change of angle toward the centre of the tool would allow only a superficial cut (continuity of the angle is present on 85 per cent of the bifaces at Pechde-l’Azé I, n = 55). So each of the available lines of evidence, use-wear analyses and analyses of their morphology, shows the multifunctionality of MTA bifaces. The variety of shape and the angulation along the acute edges of the Micoquian bifaces, ranging from very acute edges to more open edges, would probably have allowed a variety of uses (Fig. 5). So, the MTA bifaces as well as the late Micoquian bifaces were probably used in the same way: to perform a variety of tasks. These neighbouring groups were using two different technologies to produce tools used to satisfy equivalent needs during contemporaneous time, according to the time resolution given. These technologies were used for a restricted period of time relative to the duration of the Middle Palaeolithic (less than one tenth of its duration). To conclude on this point, the MTA provides one archaeological signature of innovation within the European Mousterian.

Acculturation responsible for innovation by the end of the Mousterian? The use of an Upper Palaeolithic volumetric method of debitage by the last Neanderthals in South-western Europe is considered by some to be a result of acculturation from contact with anatomically modern humans (AMH) after 40 Ky (Demars & Hublin, 1989; Mellars, 1989a, 1989b; Graves, 1991; Wynn & Coolidge, 2004). Châtelperronian blade production would be one example of this acculturation. In fact, as mentioned earlier, volumetric methods of debitage are known in the north of France during ISO stage 5 (Révillion & Tuffreau, 1994; Delagnes, 2000). Yet none had previously been described within the period of IOS 4 and IOS 3, which precede the Châtelperronian, and this volumetric method of producing blanks is thought to have disappeared until the influence of anatomically modern humans in Western Europe. Several methods of debitage had been used within the MTA. One of them was used to produce elongated blanks, of a 1,76 ± 0,53 elongation ratio (mean calculated on three assemblages, n total = 505; Fig. 8a). Actually, my analysis of MTA from Pech-del’Azé layer 5–7 and La Rochette layer 7 shows that this method is of Upper Palaeolithic type because the debitage (i.e. the flaking of the end-products) occurs not only on the broad faces of the core but also on the narrow ones. It actually rotates around part of the volume of the core (Table 3; Fig. 8b). The direction of the removals shaping

401

From Tools to Symbols

a Figure 8 Flakes (a) and core (b) from MTA type B of La Rochette layer 7 (drawings S. Pasty, chronology of removals is indicated).

402

Late Mousterian lithic technology

b

403

From Tools to Symbols

the platform also indicates that the platform is more often prepared for a lateral flake removal instead of an axial flake, which in turn indicates that the debitage has systematically been removed in a circular fashion around part of the circumference of the core (Table 4). Table 3

Position of the mean plane of intersection of the two faces at the biface point relative to the volume of the transverse section of the bifaces abandoned during production stage at Le Moustier and La Rochette. Rochette, 7 n = 38 %

Pech-de-l’Azé I, 7 n = 34 %

Pech-de-l’Azé I, 6 n = 37 %

Debitage on the narrow side(s) of the core

73

81

57

Debitage only on broad side(s) of the core

27

19

43

100

100

100

Total

Table 4

Position of the mean plane of intersection of the two faces at the biface point relative to the volume of the transverse section on bifaces abandoned during retouch stage at Pech-de-l’Azé I (layer 4). Rochette, 7 n = 43 %

Pech-de-l’Azé I, 6 n = 36 %

Lateral

81

87

Axial

19

13

Total

100

100

Conclusively, MTA people were using the volumetric Upper Palaeolithic method of producing blanks shortly before the arrival of anatomically modern humans in Europe. MTA is the only Mousterian industry within south-western Europe with an emphasis on elongated backed artefacts (cortical backing, retouch backing or backing produced during the debitage process) as important as the one characterising the Châtelperronian (Soressi, 2002: 277–284; Fig. 9). Considering that Pelegrin (1995: 260–265) has already demonstrated that the Châtelperronian does not mimic Aurignacian technology, there is no reason to believe that the Châtelperronian Upper Palaeolithic method of producing blades is derived from acculturation whilst in contact with anatomically modern humans. These were actually modifying habits that were already in use before the arrival of anatomically modern humans. Once again Mousterian innovativeness is demonstrated.

404

Late Mousterian lithic technology

Lack of planning of Mousterian technology? McBreartry and Brooks (2000) define planning depth as the ability to formulate strategies based on past experience and to act upon them in a group context. Tangible traces of this could be seen, for example, in the scheduling of resource exploitation (McBreatry & Brooks, 2000: 402). One way of interrogating the planning of Mousterian technology is by looking at the different location(s) used during the production process of stone tools. It would be exclusively in the context of raw material scarcity that Mousterian tools and blanks would be transported far away from their place of manufacture, but no planning of the knapping process would be seen in context of abundant raw material. The entire manufacture process would then have taken place at the site whereas Upper Palaeolithic anatomically modern humans would use sites with local abundant raw material as workshops. They would only have processed the initial stages of the manufacture at those sites, and the unfinished blanks would have been exported out of those sites to a further destination (Feblot-Augustins, 1997: 223–228). The MTA assemblages I studied revealed that only the first stages of the reduction sequence were done in situ at sites like Le Moustier and La Rochette, and bifaces were in most cases abandoned at the production stage (Fig. 10a). At other sites like Pech-del’Azé I, abandoned retouched bifaces were the norm and the reduction sequence would have been entirely completed at the site (Fig. 10b). Finally, at sites like the Grotte XVI, only imported retouched bifaces are found (Fig. 10c). Yet the sites of Le Moustier, La Rochette and Pech-de-l’Azé I rest directly on the rich source of good flint which was used to make the bifaces. Le Moustier and La Rochette have to be considered as biface workshops, where tools were manufactured in advance to be brought eventually to other sites, e.g. Grotte XVI. The massive retouch seen at Pech-de-l’Azé I cannot be due to a lack of local raw material compared to Le Moustier or La Rochette, as the size of the archaeological assemblage at Pech is its testimony. At Pech-de-l’Azé I, more than 30 000 lithic artefacts larger than 3 cm had been manufactured from the local flint during the MTA. Acquiring raw material, then, presented no problem, without even considering the thousands of contemporaneous artefacts produced at Pech-de-l’Azé IV eighty metres from Pech-de-l’Azé I (McPherron & Dibble, 2000). Then, even in the context of abundant raw material, which is true of the Périgord area where these four sites are located, a scheduling of the knapping process is observed. This is one example of long-term planning of knapping activities, not driven by rawmaterial scarcity, previously thought to have been used in Europe only by anatomically modern humans.

405

Elongation (length/width)

From Tools to Symbols

Figure 9 Percentage of backed flakes (pseudo-Levallois points, flakes with a cortical back, débordants flakes) and elongation of production flakes within MTA type B assemblages (Rochette layer 7, Pech-de-l’Azé 6 & 7: Soressi, 2002: 218–230) and within other Mousterian assemblages (Beauvais: Locht & Swinnen, 1993; Locht et al., 1995; Fumane: Peresani, 1998; Meillers: Pasty, 2000; Corbehem, Combe-Greanl 35 & 38, Suard 51, Bourgeois-Delaunay 9, Mesvin IV: Delagnes, 1992; Les Canalettes: Meignen, 1993; Vaufrey VIII, Fonseigner Dsup: Geneste, 1985; Hauteroche C, La Quina 3, Petit Puymoyen 2: Bourguignon, 1997; Le Moustier G: Soressi, 1999).

A progressive improvement of Mousterian behaviours? Two successive episodes can be distinguished within the MTA. MTA Type A is characterised by the production and use mainly of bifaces, MTA type B by the production and use mainly of backed knives and elongated flakes. The chronology of these episodes is based on their relative stratigraphic position at key sites (Bordes, 1984: 149; Delporte, 1970), as the resolution of radiometric dating has not yet allowed the determination of the precise duration of each event (Fig. 2). The later stage of the MTA, MTA type B, is characterised by: • the less frequent use of bifaces, knowing that MTA bifaces were long-lasting tools, resharpened (Soressi, 2002: 127–134, 2004) and transported from site to site (Soressi, 2002: 80, 163; Soressi, 2004). • the less frequent hafting of tools, indicated by use-wear analysis on Pech-del'Azé I stone tools. Denticulates, the more frequent MTA B tool, are significantly less hafted than scrapers and end-scrapers, more frequently found in the MTA A (Soressi, 2002: 262, 2004 after Anderson-Gerfaud, 1981:77–85).

406

Late Mousterian lithic technology

• the less elaborate planning of the production of stone tools. The organisation within the territory mentioned above is used only during the first stage of the MTA: MTA type A. Analyses of MTA type B assemblages at the same sites (Pechde-l'Azé I layer 6 and 7, La Rochette layer 7) indicate that, contrary to what is seen in MTA type A, entire reduction sequences are completed at the sites (Soressi, 2002: 205–206, 2004). The production of MTA type B was not constrained by a sudden lack of raw material. At Le Moustier and La Rochette, several other assemblages were produced after the MTA, still using the same raw material (there are three successive layers above the MTA at Le Moustier and at La Rochette – Peyrony, 1930; Delporte & David, 1966). Scheduling of resource exploitation in respect of lithic raw material was practised during the MTA type A, but was later abandoned. It is one example of a diachronic behavioural change towards less planning.

Discussion This paper provides new elements to show that some behaviours, earlier thought to be characteristic of recent behaviours associated with anatomically modern humans, were in fact shared with another biological species. This paper also demonstrates that some of these behaviours (the scheduling of lithic tool production within a territory) were eventually abandoned by Neanderthals, while others (the use of a volumetric method of producing blanks) were kept alive by the same Neanderthals. Should we refuse to qualify these behaviours as ‘modern’ because they were performed by Neanderthals? It is not necessary to do so if we accept that biological and behavioural changes occur autonomously, or if we accept that some behavioural changes included within the ‘modernity’ package might have occurred without a

b

c

Figure 10 Proportion of the two main stages of the reduction sequence of bifaces at different MTA sites.

407

From Tools to Symbols

biological change. Transmitted by imitation or teaching, behavioural changes do not necessarily require biological changes, as the (often used) example of the Neolithic revolution has demonstrated. This leads me to conclude that: 1. We have to accept that Neanderthals did behave in ways that would have been the norm after their disappearance. The dominance over the last years of the Revolution model of the emergence of behavioural ‘modernity’ might have induced the transformation of its initial assumption of the behavioural advantages of the last surviving species to the new assumption that Neanderthals must have always behaved in a peculiar and different way (see Chase & Dibble, 1990). We need to be aware that teleological or goal-direct explanations are opening the way to circular arguments (Renfrew, 1996). There is general agreement that anatomically modern humans represent the last species of hominid, and that our species emerged some 40–200 Kya – according to the different settings of the molecular clock and to the observed phenotype of the human fossil remains. But it is still hotly debated whether or not this genetic change gave rise to a quick, massive and permanent behavioural change. One has to then demonstrate and not to assume (cf. Gould, 1989: 304–306) that the last surviving hominid species, our species, did acquire some new, rapidly expanding and persistent behavioural changes after their speciation. Only this procedure would establish a firm link between biological changes and behavioural changes for this period of time. Consequently, the behaviour of Neanderthals has to be as accurately and thoroughly investigated as that of early anatomically modern humans. We also know that depending on historical contingencies, actual or sub-actual societies developed in different ways. Sub-actual groups were living in Stone Age times, whereas others were living in the time of the industrial age. So to distinguish between biologically, environmentally or historically induced changes, we would probably have to check for the universality and the persistence of those changes. Would symbolism fulfil those conditions? Symbolic behaviour would certainly be recognised as a universal and persistent change after 40 Ky. But then the question still remains about the link between the adoption of this behaviour and biology. Symbolic behaviour by Neanderthals has been demonstrated in the Châtelperronian of Southwestern Europe (d’Errico et al., 1998; Granger & Lévêque, 1997). Furthermore, some of the potentially critical evidence of interaction or acculturation of Neanderthals by Aurignacian groups has recently been withdrawn (e.g. by the invalidation of technical influence between those two groups – Pelegrin, 1995: 261–262 ; d’Errico et al., 1998, and by the refutation of interstratifications between Châtelperronian and Aurignacian layers – Bordes, 2002, 2003). Yet, at the same time, evidence for contemporaneity from radiometric dating is increasing, at least on a broad geographic scale (e.g. Conard &

408

Late Mousterian lithic technology

Bolus, 2003), while noting that other authors regard it as controversial (e.g. Zilhão & d’Errico, 2003). 2. The other conclusion that I am driven to is that behaviours that became the norm or that were eventually generalised after 40 or 30 Ky would have developed in a patchy way with regard to time and space (see d’Errico, 2003 for the development of a similar idea). This implies that according to the historical and environmental circumstances, some groups might have adopted, at an early date, behaviour that would have been generalised later. They might have then abandoned this behaviour (as exemplified by the MTA type B). Later the same group or another one would have re-invented and re-adopted it, and eventually it would become generalised. In the end, the picture of the emergence of behavioural ‘modernity’ we get now is probably over-simplified because it was obtained mainly by considering AMH behaviours and not always using comparable data about Neanderthals, and because it over-emphasises either the revolutionary idea or a gradual progression toward ‘modernity’, though actually this trajectory might be multi-punctuated, and even reversing from time to time. Finally, the actual picture is also not enough considering that behavioural changes might be related not to a single cause, but to a combination of biological and contingent events which created a patchwork of changes.

Notes 1

Jean-Jacques Hublin argued at the conference that the term ‘modern’ should be avoided when referring to Palaeolithic biology or behaviour. He and other discussants proposed that ‘recent’ and ‘earlier’ Homo sapiens would better suit the reality of the fossils remains we are dealing with (Morris, 2003). Concerning behavioural modernity, it is true that using this term to define a way of life which disappeared more than 7 Kya in Europe is probably not appropriate, especially when addressing the general public. Nonetheless, I will for now continue to use the term ‘behavioural modernity’, until a more appropriate word is established in the literature.

2

Yet, the low resolution of dating techniques of sites older than 35 Ky (with a very large sigma error) does not allow knowing the precise duration of the blade production or of the MTA episode.

3

Since the writing of this paper, Joris (2004) published a synthesis of the chronostratigraphy of the late Middle Palaeolithic assemblages in Central Europe.

Acknowledgements I am very grateful to Lucinda Backwell and Francesco d’Errico for their kind invitation to the meeting, as well as for their comments on an early draft of this paper.

409

From Tools to Symbols

C. Henshilwood gave me the opportunity to enter the ‘network’ of collaborations between France and South Africa by inviting me to analyse Blombos lithic industries. Data presented here were collected while I was working on my PhD dissertation which benefited from many discussions with J-M. Geneste, P. Villa and H. Dibble. Nikki Berington helped me to write this paper in better English. My gratitude extends to the CNRS/ECLIPSE project lead by M. Sanchez-Goñi for funding the AMS 14C dating of La Rochette, to the French Minister of Culture, the Conseil General de la Dordogne and the IPGQ for their support toward the ESR dating of Pech-de-l’Aze I. This paper was prepared while I was financially supported by the Fyssen Foundation. The IZIKO South African Museum provided working space and was a place of stimulating discussions while working in Cape Town.

Bibliography Ambrose, S.H. (1998). Late Pleistocene human population bottlenecks, volcanic winter, and differentiation of modern humans. Journal of Human Evolution 34(6): 623-652. Anderson-Gerfaud, P. (1981). Contribution méthodologique à l’analyse des micro-traces d’utilisations sur les outils préhistoriques. Thèse de 3ème cycle de l’Université de Bordeaux I. Antoine, P., Auguste, P., Munaut, A.-V., Frechen, M., Englemann, A. & Michel, V. (2002). Etudes géologiques, environnementales et datations. In (J.-L. Locht, Ed) Bettencourt-saintOuen (Somme). Cinq occupations paléolithiqus au début de la dernière glaciation, pp. 15–55. Documents d’Archéologie française, No. 90. Paris: Maison des Sciences de l’Homme. Barham, L.S. (1998). Possible early pigment use in south-central Africa. Current Antropology, 39, 703–710. Barham, L.S. (2001). Central Africa and the emergence of regional identity in the Middle Pleistocene. In (L. Barham & K. Robson-Brown, Eds) Human Roots. Africa and Asia in the Middle Pleistocene, pp. 65-80. Bristol: Western Academic & Specialist Press Limited. Bar-Yosef, O. (1998). On the nature of transitions: the Middle to Upper Palaeolithic and the Neolithic revolution. Cambridge Archaeological Journal 8, 141–163. Binford, L. (1989). Isolating the transition to cultural adaptations: an organizational approach. In (E. Trinkaus, Ed.) The Emergence of Modern Humans: Biocultural Adaptations in the Late Pleistocene, pp. 18–41. Cambridge: Cambridge University Press. Boëda, E. (1993). Le débitage discoïde et le débitage Levallois récurrent centripète. Bulletin de la Société Préhistorique Française 90(6), 392-404. Boëda, E. (1995). Caractéristiques techniques des chaînes opératoires lithiques des niveaux micoquiens de Külna (Tchécoslovaquie). In Les industries à pointes foliacées d’Europe centrale, Actes du colloque: Les premières découvertes de Paléolithique à Miskolc et la

410

Late Mousterian lithic technology question des industries à pièces foliacées de l’Europe centrale dans leur cadre chronologique, paléoécologique, paléontologique, Miskolc (Hongrie), 10–15 September 1991, Paléo, suppl. 1, 57–72. Boëda, E., Fontugne, M., Valladas, H. & Ortega, I. (1996). Barbas III. Industries du paléolithique moyen récent et du Paléolithique supérieur ancien. In (E. Carbonell & M. Vaquero, Eds) The last Neandertals, the first anatomically modern Humans: a take about the Human diversity. Cultural change and Human evolution: the crisis at 40 ka BP, pp. 147–156. Tarragona: Universitat Rovira i Virgili Press. Bon, F., Gambier, D., Ferrier, C. & Gardère, P. (1998). Gisement de Brassempouy (Landes) : les recherches de 1995 à 1997, bilan et perspectives. Bulletin de la Société Borda 449, 203–222. Bordes, F. (1984). Leçons sur le Paléolithique. Paris: eds. du C.N.R.S., Cahiers du Quaternaire 7. Bordes, J.-G. (2002). Les interstratifications Châtelperronien/Aurignacien du Roc-de-Combe et du Piage (Lot, France). Analyse taphonomiques des industries lithiques, implications archéologiques. Thèse de l’Université Bordeaux I. 365 p. Available on the web at: http:// www.u-bordeaux1.fr/bx1/p1_recherche.html Bordes, J.-G. (2003). Lithic taphonomy of the Châtelperronian/Aurignacian interstratifications at Roc de Combe and Le Piage (Lot, France). In (J. Zilhão & F. d’Errico, Eds) The Chronology of the Aurignacian and of the Transitional Technocomplexes. Dating, Stratigraphies, Cultural Implications, pp. 223–244. Lisbon: Instituto Português de Arqueologia. Bosinski, G. (1967). Die Mittelpalaolithischen funde in westlichen mitteleuropa. Kohn: Fundamenta Reihe A/4. Bourguignon, L. (1996). La conception de débitage Quina. Quaternaria Nova, VI, 149–166. Bourguignon, L. (1997). Le Moustérien de type Quina: Nouvelle définition d’une entité technique. Thèse de l’Université Paris X. Brooks, A.S., Helgren, D.M., Cramer, J.M., Franklin, A., Hornyak, W., Keating, J.M., Klein, R.G., Rink, W.J., Schwarcz, H.P., Smith, J.N.L., Stewart, K., Todd, N.E., Verniers, J. & Yellen, J.E. (1995). Dating and context of three Middle Stone Age sites with bone points in the upper Semliki Valley, Zaire. Science 268, 548–553. Chase, P.G. & Dibble, H.L. (1990). On the emergence of modern humans. Current Anthropology 31, 58–59. Conard, N.J. & Bolus, M. (2003). Radiocarbon dating the appearance of Modern Humans and timing of cultural innovations in Europe: new results and new challenges. Journal of Archaeological Science 44, 331–371. Deacon, H.J. & Wurz, S. (2001). Middle Pelistocene populations of southern Africa and the emergence of modern behaviour. In (L. Barham & K. Robson-Brown, Eds) Human Roots. Africa and Asia in the Middle Pleistocene, pp. 55–63. Bristol: Western Academic & Specialist Press Limited.

411

From Tools to Symbols Deacon, H.J. (1988). The origins of anatomically modern people and the South African evidence. Paleoecology of Africa 19, 193–200. Deacon, H.J. & Deacon, J. (1999). Human Beginnings in South Africa: Uncovering the Secrets of the Stone Age. Cape Town: David Philip. Debénath, A. & Dibble, H.L. (1994). The Handbook of Palaeolithic Typology. Vol I. The Lower and Middle Palaeolithic of Europe. Philadelphia: University Museum, University of Pennsylvania. Delagnes, A. (1992). L’organisation de la production lithique au Paléolithique moyen, approche technologique à partir de l’étude des industries de la Chaise-de-Vouthon. Thèse de l’Université de Paris X. Delagnes, A. (2000). Blade production during the Middle Palaeolithic in North-western Europe. In (W. Dong, Ed.) Proceedings of 1999 Beijing International Symposium on Paleoanthropology. Acta Anthroplogica Sinica 19 supplement Bejing, 181–188. Delporte, H. (1970). Le passage du Moustérien au Paléolithique supérieur. In (G. Camps & G. Olive, Eds) L’homme de Cro-Magnon: anthropologie et archéologie, pp.129–139. Paris: Arts et métiers graphiques. Delporte, H. & David R. (1966). L’évolution des industries moustériennes à la Rochette, Commune de Saint-Léon-sur-Vézère (Dordogne). Bulletin de la Société Préhistorique Française 42, 48–62. Demars, P.-Y. & Hublin, J.-J. (1989). La transition Néandertaliens/Hommes de type modernes en Europe occidentale. In (B. Vandermeersch, Ed.) L’Extinction, Vol 7. ERAUL 34: 23–38. d‘Errico F., Zilhão J., Julien, M., Baffier, D. & Pelegrin, J. (1998). Neanderthal acculturation in Western Europe? A critical review of the evidence and its interpretation. Current Anthropology 39, supplement June 1998, S1–S44. d’Errico, F. (2003). The invisible frontier. A multiple species model for the origin of behavioural modernity. Evolutionary Anthropology 12, 188–202. d’Errico, F., Henshilwood, C., Lawson, G., Vanhaeren, M., Tillier, A.-M., Soressi, M., Bresson, F., Maureille, B., Nowell, A., Backwell, L., Lakarra, J.A. & Julien, M. (2003). Archaeological evidence for the emergence of language, symbolism and music: an alternative multidisciplinary perspective. Journal of World Prehistory 17(1), 1–70. Diamond, J.M. (1992). The Rise and Fall of the Third Chimpanzee. New York: Harper Collins. Feblot-Augustins, J. (1997). La circulation des matières premières au Paléolithique. Liège: ERAULT, No. 75. Foley, R.A. (1997). Humans Before Humanity: An Evolutionary Perspective. Malden, Massachusetts: Blackwell. Geneste, J.-M. (1985). Analyse d’industries moustériennes du Périgord: une approche technologique du comportement des groupes humains au Paléolithique moyen. Thèse de l’Université de Bordeaux I.

412

Late Mousterian lithic technology Gould, S. J. (1998). Wonderful Life. New York: W.W. Norton. Granger, J.-M. & Lévêque F. (1997). Parure castelperronienne et aurignacienne: étude de trois séries inédites de dents percées et comparaisons. Compte-rendu de l’Académie de Sciences, Paris, 325, 537–543. Graves, P. (1991). New models and metaphors for the Neanderthal debate. Current Anthropology 32, 513–42. Guibert, P., Bechtel, F., Schvoerer, M., Rigaud, J.-Ph. & Simek, J.F. (1999). Datations par thermoluminescence de sédiments chauffés provenant d’une aire de combustion moustérienne (Grotte XVI, Cénac et St-Julien, Dordogne, France). Revue d’Archéométrie 23, 163–175. Hedges, R.E.M., Housley, R.A., Law, I.A. & Bronk, C.R. (1990). Radiocarbon dates from the Oxford AMS system: archaeometry datelist 10. Archaeometry 32, 101-108. Henshilwood, C., Sealy, J., Yates, R., Cruz-Uribe, K., Goldberg, P., Grine, F.E., Klein, R.G., Poggenpoel, C., Van Niekerk, K. & Watts, I. (2001). Blombos cave, southern Cape, South Africa : preliminary report on the 1992-1999 excavations of the Middle Stone age levels. Journal of Archaeological Science 28, 421–448. Henshilwood, C. & Marean, C. (2003). The origin of modern human behavior. Critique of the models and their test implications. Current Anthropology 44, 627–652. Jaubert J. & Mourre, V. (1996). Coudoulous, Le Rescoundudou, Mauran: diversité des matières premières et variabilité des schémas de production d’éclats. Quaternaria Nova VI, 313–341. Joris O. (2004). Zur chronostratigraphischen Stellung der spätmittelpaläolithischen Keilmessergruppen. Bericht des Römisch-Germanischen Kommission 84, 49–153. Klein, R.G. (2000). Archaeology and the evolution of human behavior. Evolutionary Anthropology 9(1), 17–36. Klein, R.G. (1989). Biological and behavioural perspectives on Modern Human origins in southern Africa. In (P. Mellars & C. Stringer, Eds) The Human Revolution, pp. 529–546. Edinburgh: Edinburgh University Press. Klein, R.G. (1994). The problem of Modern Human origins. In (M. Nitecki & D. Nitecki, Eds) Origins of Anatomically Modern Humans, pp. 3–22. New York: Plenum Press. Klein, R.G. (1995). Anatomy, behavior, and Modern Human origins. Journal of World Prehistory 9(2), 167–198. Klein, R.G. (1998). Why anatomically modern people did not disperse from Africa 100,000 years ago. In (T.A. Akazawa & O. Bar-Yosef, Eds) Neandertals and Modern Humans in Western Asia, pp. 509–521. New York: Plenum Press. Knight, C.D., Power, C. & Watts, I. (1995). The human symbolic revolution: a Darwinian account. Cambridge Archaeological Journal 5, 75–114. Leroi-Gourhan, A. (1943). Evolution et technique. Tome I, L’Homme et la Matière. Paris: Albin Michel.

413

From Tools to Symbols Locht, J.-L, Swinnen, C., Antoine, P., Auguste, P., Patou-Mathys, M., Depaepe, P., Falguères, C., Laurent, M. & Bahain, J.-J. (with collaboration of P. Mathys) (1995). Le gisement Paléolithique moyen de Beauvais (Oise). Bulletin de la Société Préhistorique Française 92(2), 213–226. Locht, J.-L. & Swinnen C. (1993). Le débitage discoïde du gisement de Beauvais (Oise) : aspects de la chaîne opératoire au travers de quelques remontages. Paléo 6, 89–104. Maureille, B. & Soressi, M. (2000). A propos de la position chrono-stratigraphique de l’enfant du Pech-de-l’Azé (commune de Carsac, Dordogne): la résurrection du fantôme. Paléo 12, 339–352. McBreartry, S. & Brooks, A. (2000). The revolution that wasn’t: a new interpretation of the origin of modern human behavior. Journal of Human Evolution 39, 453–563. McPherron, S.P. & Dibble, H.L. (2000). The lithic assemblages of Pech de L’Aze IV (Dordogne, France). Préhistoire Européenne 15, 9–43. Meignen, L. (1993). Les industries lithiques de l’abri des Canalettes: couche 2. In (L. Meignen, Ed.) L’abri des Canalette: un habitat moustérien sur les grands Causses (Nant, Aveyron): fouilles 1980–1986, pp. 239–328. Monographies du CRA 10. Paris : Ed. CNRS. Mellars, P. (1973). The character of the middle-upper Palaeolithic transition in south-west of France. In (C. Renfrew, Ed.) The Explanation of Culture Change, pp. 255–276. London: Duckworth. Mellars, P. (1989a). Technological changes at the Middle-Upper Palaeolithic Transition: economic, social and cognitive perspectives. In (P. Mellars & C. Stringer, Eds) The Human Revolution, pp. 338–365. Edinburgh: Edinburgh University Press. Mellars, P. (1989b). Major issues in the emergence of modern humans. Current Anthropology 30, 349–385. Mellars, P. (1996a). The Neanderthal Legacy. An Archaeological Perspective from Western Europe, p. 471. Princeton: Princeton University Press. Mellars, P. (1996b). Symbolism, language, and the Neanderthal mind. In (K. Gibson & P. Mellars, Eds) Modelling the early human mind, pp. 15–32. McDonald Institute Monographs, McDonald Institute for Archaeological Research, University of Cambridge, Cambridge. Mellars, P. (1998). Neandertals, modern humans and the archaeological evidence for language. In (N.G. Jablonski & L.C. Aiello, Eds) The Origin and Diversification of Language, pp. 89– 115. Memoirs of the California Academy of Sciences 24. Mellars P. (1999). The Neanderthal problem continued. Current Anthropology 30, 341–350 Mellars, P. & Stringer, C. (Eds) (1989). The Human Revolution: Behavioral and Biological Perspectives on the Origins of Modern Humans. Edinburgh: Edinburgh University Press. Mercier, N., Valladas, H, Joron, J.-L. & Reyss, J.-L. (1993). Thermoluminescence dating of the prehistoric site of La Roche à Pierrot, Saint-Césaire. In (F. Lévêque, A. Backer & M. Guilbaud, Eds) Context of a Late Neandertal, pp. 15–22. Madison: Prehistory Press.

414

Late Mousterian lithic technology Mook, W.G. (1988). Radiocarbon Daten aus der Kulna. Anthropos, Brno 24, 285–286. Morris, D. (2003). From Tools to Symbols. From Early Hominids to Modern Humans. Report on the International Round Table in honour of Professor P. Tobias, University of the Witwatersrand, 16–18 March 2003. The Digging Stick 20(1), 12. Noble, W. & Davidson, I. (1991). The evolutionary emergence of Modern Human behaviour: language and its archaeology. Man 26, 223 –253. Pasty, J.-F. (2000). Le gisement Paléolithique moyen de Meillers (Allier): un exemple de la variabilité du débitage discoïde. Bulletin de la Société Préhistorique Française 97(2), 165–190. Pelegrin, J. (1995). Technologie lithique: le Châtelperronien de Roc-de-Combe (Lot) et de la Côte (Dordogne), Cahiers du Quaternaire 20. Paris: CNRS éditions. Peresani, M. (1998). La variabilité du débitage discoïde dans le grotte de Fumane (Italie du Nord). Paléo 10, 123–146. Peresani, M. (Ed.) (2003). Discoïd Lithic Technology. Advances and Implications. BAR International Series 1120. Peyrony, D. (1930). Le Moustier: ses gisements, ses industries, ses couches géologiques. Revue Anthropologique 40, 3–76 and 155–176. Renfrew, C. (1996). The sapient behaviour paradox: how to test for potential?. In (P. Mellars & K.R. Gibson, Eds) Modelling the Early Human Mind, pp. 11–14. Cambridge: McDonald Institute Monographs. Révillion, S. & Tuffreau, A. (Eds) (1994). Les industries laminaires au Paléolithique moyen. Actes de la table ronde internationale organisée par l’ERA 37 du CRA-CNRS à Villeneuve d’Ascq, 13–14 November 1991. Dossier de documentions archéologiques No. 18, CNRS, p. 191. Richter, D., Mauz, B., Böhner, U., Weissmüller, W., Wagner, G.A., Freund, G., Rink, W.J. & Richet, J. (2000). Luminescence dating of the Middle/Upper Palaeolithic sites ‚Sesselfelsgrotte‘ and ‚Abri I AM chulerloch‘, Altmühltal, Bavaria. In (J. Orschiedt & Weniger G.-C., Eds) Neandertals and Modern Humans – Discussing the transition. Central and Eastern Europe from 50,000 – 30,000 B.P., pp. 30–41. Neanderthal Museum (Wissenschaftliche schriften des Neanderthal Museum, Bd. 2). Richter, J. (1997). Sesselfelsgrotte III. Der G-Schichten-Komplex der Sesselfelsgrotte. Zum Verständnis des Micoquien. Quartär-Blibliotek, Band 7. Rink, W.J., Schwarcz, H.P., Valoch K. & Stringer, C. B. (1996). Dating of the Neanderthal site of Kulna, Czech Republic. Journal of Archaeological Science 23, 889–901. Soressi, M. & Hays, M. (2003). Manufacture, transport and use of Mousterian bifaces. A case study from the Perigord (France). In (M. Soressi & H. L. Dibble, Eds) Multiple Approaches to the Study of Bifacial Technologies, pp. 125–147. Philadelphia: Publication of The University of Pennsylvania Museum Press, Monograph 115. Soressi, M. (1999). Stabilité technique au Moustérien. L’exemple du débitage du MTA A du Moustier (Dordogne, France). Paléo 11, 111–134.

415

From Tools to Symbols Soressi, M. (2002). Le Moustérien de tradition acheuléenne du sud-ouest de la France. Discussion sur la signification du faciès à partir de l’étude comparée de quatre sites : Pech-de-l’Azé I, Le Moustier, La Rochette et la Grotte XVI. Thèse de l’Université Bordeaux I, 344 p. Available on the web at: http://www.u-bordeaux1.fr/bx1/p1_recherche.html Soressi, M. (2004). From Mousterian of Acheulian tradition type A to type B: a change in technical tradition, raw material, task, or settlement dynamics? In (N. Conard, Ed.) Settlement Dynamics of the Middle Palaeolithic and Middle Stone Age, Vol. II, pp. 343–366. Tübingen: Kerns Verlag. Soressi, M., Jones, H., & Rink, J. (in preparation). ESR and Uranium Series dating of teeth from F.Borde’s excavation at Pech-de-l’Azé Ib, Dordogne, France. Stringer, C. (2003). Out of Ethiopia. Nature 423, 692–695. Texier, P.-J. (1996). Evolution and diversity in flaking techniques and methods in the Palaeolithic. In Oltre la pietra – Modelli e tecnologie per capire la Préistoria – Forli, XIII UISPP Congress, A.B.A.C.O. eds., pp. 281–321. Turq, A. (1989). Approche technologique et économique du faciès Moustérien de type Quina. Bulletin de la Société Préhistorique Française 86(8), 244–256. Valladas, H., Chadelle, J.-P., Geneste, J.-M., Joron, J.-L., Meignen, L. & Texier, P.-J. (1987). Datations par la thermoluminescence de gisements moustériens du Sud de la France. L’Anthropologie 91, 211–216. Vaquero, M. & Carbonell, E. (2003). A temporal perspective on the variability of the Discoïd method in the Iberian Peninsula. In: Discoïd Lithic Technology. Advances and implications. Peresani, M. (Eds.), BAR International Series 1120, pp. 67-81. Veil, S., Breest, K., Höfle, H.-C., Meyer, H.-C., Plisson, H., Urban-Küttel, B., Wagner, G.A., & Zöller, L. (1994). Ein mittelpaläolithischer Fundplatz aus der Weichsel-Kaltzeit bei Lichtenberg, lkr. Lüchow-Dannenberg. Zwischenbericht über die archäologischen und geowissenschaftlichen Untersuchungen. Germania 72, 1–66. Vogel, J.C. & Waterbolk H.T. (1967). Groningen Radiocarbon Dates VII. Radiocarbon 9, 107–155. Vogel, J. & Zagwijn, W.H. (1967). Groningen Radiocarbon Dates VI. Radiocarbon 9, 63–106. Wadley, L. (2001). What is cultural modernity ? A general view and a South African perspective from Rose Cottage Cave. Cambridge Archaeological Journal 11(2), 201 –221. Watts, I. (1999). The origin of Symbolic culture. In (R. Dunbar, C. Knight & C. Power, Eds) The Evolution of Culture, pp. 113–146. Edinburgh: Edinburgh University Press. Wynn, T. & Coolidge F.L. (2004). The expert Neandertal mind. Journal of Human Evolution 46(4), 467–487. Zilhão, J. (2001). Anatomically Archaic, Behaviourally Modern: The Last Neandertals and Their Destiny. Amsterdam: Stichting Nederlands Museum voor Anthopologie en Praehistorie.

416

Late Mousterian lithic technology Zilhão, J. & d‘Errico F. (2003). An Aurignacian ‘garden of Eden’ in southern Germany? An alternative interpretation of the Geissenklösterle and a critique of the Kulturpumpe model. Paléo 15, 69–86.

417

Exploring and quantifying technological differences between the MSA I, MSA II and Howieson’s Poort at Klasies River Sarah Wurz Department of Geography and Environmental Studies, University of Stellenbosch, Private Bag X1, Stellenbosch, 7602, South Africa

Abstract Inferences drawn from the study of the variability in sets of stone artefacts are central to the debate on the emergence of modern behaviour in the Middle and Late Pleistocene in Africa. Some argue that the Middle Stone Age demonstrates little variability while others emphasise the clear temporal and regional patterning. These contrary perceptions result from using methodologies that highlight different aspects of the variability. How the methodologies influence the detection and interpretation of variability is discussed with reference to the Klasies River site. This artefact sequence has been assessed as reflecting typological and technological stasis with marked change in the sequence only recognisable in the Howieson’s Poort sub-stage. The study reported here suggests that technological variables indicate that the MSA I, MSA II and Howieson’s Poort represent distinct technological conventions or technocomplexes aimed at the production of different end-products. In the MSA I, a blade strategy dominates, in the MSA II a Levallois-like production strategy is inferred while the Howieson’s Poort again represents a blade reduction strategy with a more extended chaîne opératoire than the other sub-stages. To clarify the differences between the MSA I and MSA II, univariate and multivariate statistical analyses of the continuous variables of the end-products, points and blades were undertaken. This confirmed and quantified that the MSA I can be distinguished from the MSA II in terms of technological characteristics. It is evident that the platform characteristics of the end-products are time-sensitive indicators of differences between the sub-stages. It is important that the changes in the sequence at Klasies River are correlated with other Middle Stone Age occurrences. It is only when regional patterning is clarified that the issue of modern behaviour can be addressed in terms of the variability in Middle Stone Age artefacts.

418

Exploring and quantifying technological differences

Résumé La variabilité des assemblages lithiques est au cœur du débat sur l’émergence des comportements moderne au cours du Pléistocène moyen et supérieur en Afrique. Certains auteurs considèrent que le Middle Stone Age (MSA) se caractérise par une faible variabilité alors que d’autres identifient des différences diachroniques et régionales. Ces visions opposées sont la conséquence des méthodes d’analyse, chacune mettant en évidence certains aspects de la variabilité entre assemblages. En nous basant sur les séries de Klasies River, nous examinons dans ce travail de quelle façon les méthodes choisies identifient et interprètent cette variabilité. Les assemblages lithiques de ce site ont été interprétés comme le reflet d’une stase dans la technologie et les types d’outils produits qui serait interrompue par l’émergence de l’Howieson’s Poort. Notre étude technologique indique au contraire que le MSA I, MSA II et Howieson’s Poort se caractérisent par des conventions techniques distinctes visant à la production de différents supports. Dans le MSA l, le débitage laminaire domine, dans le MSA II, un débitage de type Levallois est identifié alors que Howieson’s Poort présente à nouveau un débitage laminaire avec cependant une chaîne opératoire plus complexe que dans les phases précédentes. Pour explorer les différences entre le MSA I et le MSA II, nous avons réalisé des analyses statistiques univariées et multivariées de variables continues enregistrées sur les outils, les pointes et les lames. Cela a permis de montrer que le MSA I peut être différencié du MSA II à partir de caractères technologiques. Il est clair que la morphologie du talon est un indicateur particulièrement utile pour identifier des différences entre phases culturelles. Il est remarquable d’observer que les changements détectés à Klasies River s’observent dans d’autres séquences MSA. C’est seulement quand ces changements synchrones créeront un système régional que la question du comportement moderne pourra être abordée en utilisant l’industrie lithique.

Introduction The term ‘Middle Stone Age’ describes assemblages dating from c. 22–300 thousand years ago (Kya) in Africa south of the Sahara. Originally (Goodwin & Van Riet Lowe, 1929), the Middle Stone Age was viewed within a ‘lithi-cultural’ evolutionary paradigm that saw retouched types and size of artefacts as indicators of ‘advancement’. The ‘advancement’ in the Middle Stone Age was represented by the first appearance of ‘pure flake industries’, lance-head and point types, faceted butts and longitudinal flaking (Goodwin & Van Riet Lowe, 1929: 147). Since those formative years many more occurrences have been investigated, but the paradigm from which inferences on behaviour are made has hardly changed. However, some progress has been made in understanding the chrono-stratigraphic patterning within the Middle Stone Age. In this regard the Klasies River sequence has played an important role. Klasies River represents one of the largest Middle Stone Age occurrences in South Africa. Singer and Wymer (1982) sub-divided the artefact sequence there into the MSA I, MSA II, Howieson’s Poort, MSA III and MSA IV on the basis of typological changes. Volman (1981, 1984) has used similar sub-divisions, but with altered labels, to describe

419

From Tools to Symbols

the variability in the Middle Stone Age on a regional scale. The presumption is that only the Howieson’s Poort represents a separate industry (Singer & Wymer, 1982; Volman, 1984; Thackeray, 1992), because it is typologically different from the other sub-stages. The presence of backed artefacts in the Howieson’s Poort makes regional correlations (e.g. Deacon, 1992) possible while the lack of technological and typological variation in the other sub-stages precludes such correlations. This perceived lack of variability in the Middle Stone Age that is associated with early modern humans has been one of the reasons for excluding such humans from the grade of modern human behaviour (e.g. Klein, 1992; Thackeray, 1992; Mithen, 1996; Noble & Davidson, 1996; Wadley, 2001). How effective variability is recognised depends on the potential of the methodology used. Assessment of variability at Klasies River and elsewhere is conventionally based on the typological study of retouched types and metrical attributes of the flake-blades. This methodology is capable of capturing only a limited degree of variability in the Middle Stone Age, with its low frequencies of unstandardised retouch on artefacts. The study reported here explores another approach by including technological variables. Multivariate discriminate analyses and graphical representation by means of Canonical Variate Analysis (CVA) biplots and alpha bags (Wurz et al., 2003) are used to further explore and clarify patterning at Klasies River. The MSA II is sub-divided into MSA II Upper (MSA II U) and MSA II Lower (MSA II L) because this provides the opportunity to investigate within sub-stage variability. This methodology allows the recognition of more extensive differences between the MSA I, MSA II and Howieson’s Poort at Klasies River than previously possible. The MSA at Klasies, similar to other African Middle Stone Age assemblages (cf. Mc Brearty & Brooks 2000: 456), has point and blade industries with prepared cores as a component. In the Klasies River sequence described here two blade industries, at 110 Kya and 65 Kya, are separated by a point industry.

The site and sample The Klasies River site (34.06ºS, 24.24ºE) on the southern Cape coast (Fig. 1) consists of a group of caves and overhangs eroded into a sandstone cliff face. Caves 1, 1A, 1B and 2 served as a sheltered place between 110 Kya and 60 Kya and deposits have built up against the cliff face to a height of more than twenty metres. Singer & Wymer (1982) first excavated the site in 1967/8 and H.J. Deacon (Deacon & Geleijnse, 1988; Deacon, 1995) has directed excavations since 1984. The Deacon excavations were designed to complement the extensive excavations of Singer & Wymer by describing the stratigraphy (Fig. 2) and temporal resolution in more detail. Multiple dating estimates (e.g. Deacon et al., 1988; Grün et al., 1990; Vogel, 2001; Feathers, 2002) provide firm chronological control for the sub-stages at Klasies River. The MSA

420

Exploring and quantifying technological differences

Figure 1 Later Stone Age sites in the Western and Eastern Cape (after Deacon & Deacon 1999).

Figure 2 Scheme of sections showing units and members at Klasies River.

421

From Tools to Symbols

I dates to around 110 Kya, the MSA II to 80–100 Kya, and the Howieson’s Poort to approximately 65 Kya (Table 1, Wurz, 2002). The sample reported on here includes all the artefacts from the Deacon excavations (D-sample), including those studied by Thackeray (1989) and Thackeray & Kelly (1988). In some cases the D-sample is enlarged with artefacts from the Singer and Wymer sample (SW-sample) – cores from the MSA I (layer 37) and Howieson’s Poort (Layers 10–21) and the backed artefacts from the Howieson’s Poort are included.

Method Recent developments that take us beyond Bordian typology in the study of assemblage variability and inter-site comparison is the chaîne opératoire approach. ‘Chaîne opératoire’ is a ‘fuzzy but flexible’ concept (Audouze, 1999: 169) that encompasses many analytical procedures. In the context of lithic analysis it refers to the unfolding of the technical act (Chazan, 1997) from the initial stage of raw material procurement to the final stage of the discarding of the used artefacts. The emphasis on core analysis within the chaîne opératoire or technological approach provides an opportunity to improve Middle Stone Age analytical approaches. One of the major insights provided by technological studies is that cores are ‘set up’ to consist of a number of volumes that interact in a specific way to predetermine the characteristics of end-products (Boëda, 1995). The volumetric conceptualisation of a core follows a set of rules that differs according to the technical schemes followed. For example, the volumetric conception that guides core reduction in Levallois point and blade schemes is different from that in an Upper Palaeolithic blade scheme. The rules by which Middle Stone Age cores were reduced are not normally described because the chaîne opératoire method has not been widely applied in South Africa. However, there is general agreement that the object of Middle Stone Age flaking is the production of elongated parallel or convergent flake-blades, here termed blades and points. In the European research traditions such end-products are considered to have been produced within either Levallois or non-Levallois systems. In Levallois schemes the core consists of two asymmetrical volumes, a lower and upper volume (Boëda, 1988a, 1995). The lower volume functions as an inactive under-surface (striking platform) and the upper volume as the active surface from which end-products are struck. Preparation of the lateral, distal and proximal parts of the upper surface control the characteristics of the end-products. End-products are removed from the upper surface at a plane parallel to the intersection of the two volumes. A recurrent strategy, when a restricted number of end-products are removed before repreparation, or a preferential strategy, when a single end-product removes most of the prepared surface, can be used.

422

Exploring and quantifying technological differences

In non-Levallois blade schemes (Révillion, 1994; Boëda 1988b; Inizan et al., 1999) blades and points are produced by reducing the volume of the core continuously without repreparation of the productive surface. However, management of the lateral and distal convexities is essential to maintain the transverse convexity necessary to produce blades in a series. Cores are reduced around the full or half of the circumference of the core in a recurrent fashion, using the ridges of previous removals to guide subsequent removals. These ideas have influenced thinking in the study of the Klasies River artefacts. The challenge is how to adapt these concepts to a South African context without necessarily using the standard European methods like refitting, technical reading of dorsal scars or replication and experimentation. These methods are not suitable for the majority of Middle Stone Age assemblages, since Middle Stone Age artefacts are usually from lag-deposits, making refitting problematic, and artefacts in coarse grain quartzite make reading the direction of flaking difficult. Developments from elsewhere can be used for guidance, but unique geomorphological and site formation processes demand a unique approach. The ideal is to develop an ‘indigenous’ technological body of theory on the basis of replication and refitting for the analysis of South African artefacts. The approach followed here was to use technological principles to evaluate the volumetric conception of cores and the technical properties of the end-products. The cores from Klasies River are assessed in terms of the following technological principles: 1. method of initiation; 2. volumetric composition – i.e. how many volumes can be discerned and the hierarchical relationship of the volumes to each other; 3. characteristics of the active and inactive surfaces, including an evaluation of the hierarchical relationship of the platforms to each other and measurement of the angles of the platforms; 4. if a two-volume system is used, whether the axis of the flaking surface to the striking platform is parallel or tangential; 5. end-product removal strategy (recurrent, preferential). About half of the cores from the Klasies River sample allow investigation of reduction methods (Table 1). Given the complexities involved in linking end-state cores to technical schemes (Marks & Volkman, 1983; Boëda, 1995; Van Peer, 1995), these informative cores are described in terms of the scar pattern on the active face – either as cores with mainly point scars, or as cores with blade scars. This preliminary measure was taken to avoid premature designation to technical schemes. Here such high-level inferences are made on the basis of core as well as end-product characteristics.

423

From Tools to Symbols Table 1

Core classification.

MSA I

Lower MSA II Upper MSA II

SW-sample

D-sample

D-sample

Howieson’s Poort SW-sample

Cave 1A,

Cave 2,

D-sample

D-sample

%

%

n=

%

n=

%

n=

Point

35

34

48

67

33

13

0

0

0

0

0

0

Blade

18

18

8

11

13

5

49

186

47

15

34

22

4

4

7

10

3

1

11

45

3

1

15

10

Core fragment

n=

Howieson’s Poort

%

Irregular

%

Howieson’s Poort

n=

n=

38

37

36

51

49

19

24

103

44

14

51

33

Pre-form

5

5

1

2

3

1

7

31

3

1

0

0

‘Bladelet’

0

0

0

0

0

0

2

7

3

1

0

0

‘Microcore’

0

0

0

0

0

0

7

28

0

0

0

0

100

98

100

141

100

39

100

400

100

42

100

65

Total

One of the major problems in using European technological principles to analyse the Klasies River sample is how to decide which flakes are end-products. End-products are sometimes regarded as those flakes that are symmetrical, have an organised disposition of dorsal scars, and have carefully prepared platforms (Van Peer, 1992; Boëda 1995), but refitting is considered to be the ideal way to determine which are end-products. In this study all the elongated products, blades and points, are described, because researchers may apply the criteria for end-product recognition differently. This measure ensures that the methods used here can be reproduced, even though it is clear that not all the elongated pieces are end-products. When inferring reduction sequences, only the characteristics of the symmetrical blades and points are taken into account. The length, width and thickness of the blades and points are in Tables 2–4. Table 2

End product dimensions (average mm) – D-sample. Length Blades

Width Points

Blades

X

n=

X

n=

X

MSA I

81,0

84

71,0

60

28,3

MSA II lower

75,9

454

65,0

414

MSA II upper

68,8

244

59,0

246

HP (cave 1A)

43,9

75



424

n=

Thickness Points

Blades

X

n=

X

472

33,5

71

30,2

1 791

34,6

26,9

1 074

31,6

18,8

714



Points

n=

X

n=

8,2

472

9,3

71

545

9,6

1 791

11,0

545

298

8,7

1 074

10,3

298

4,9

714



Exploring and quantifying technological differences Table 3

Summary statistics for the blades.

Variable Length

Width

Thickness

Platform thickness

Statistic

MSA I

MSA II L

Mean

80,86

75,88

67,9

SD

23,52

23,4

22,78

N

83

455

218

Mean

28,27

30,18

26,87

SD

8,21

8,84

8,48

N

472

1792

1037

Mean

8,27

9,55

8,7

SD

3,43

3,65

3,41

N

472

1972

1073

Mean

6,83

9,99

9,12

SD

3,03

3,4

4,35

N

315

1244

658

19,61

24,47

21,43

6,6

7,85

7,74

Mean Platform width

SD N

Length: Platform thickness ratio

Platform angle

MSA II U

315

1244

659

Mean

14,85

8,37

8,23

SD

13,83

4,9

5,02

77

396

212

Mean

N

80,44

84,67

82,53

SD

8,15

6,4

7,82

N

315

1246

660

Technological investigation of end-products entails the study of the butts or platforms as they reflect the technique used, or the way in which energy or force is transmitted, in making artefacts (Knuttson, 1988; Chazan, 1997). While experimentation forms an integral part of investigating technique (Pelegrin, 2000), it was beyond the scope of this study. The platforms are tentatively interpreted as being manufactured with either a soft hammer or hard hammer on the basis of diagnostic criteria only. The characteristics used to infer soft hammer platforms include a small platform size of several millimetres, the presence of a lip, a platform angle of less than 80º, abrasion or thinning to remove the overhangs of the previous removals, a diffuse platform and the absence of marks that indicate the impact of percussion (Inizan et al., 1999; Pelegrin, 1995, 2000). The use of a hard hammer is inferred from a splintered bulb of percussion and a ‘ring-crack’ at the point of impact, and, more ambiguously, from a prominent

425

From Tools to Symbols Table 4

Summary statistics for points (multivariate analysis).

Variable Length

Statistic

MSA I

MSA II L

MSA II U

Mean

70,63

65,36

58,76

SD

15,92

16,6

15,79

N

60

413

246

Mean Width

Thickness

Platform thickness

Platform width

33,55

34,67

31,6

SD

6,1

7,69

7,2

N

71

545

298

Mean

9,34

10,97

10,31

SD

2,26

3,89

2.85

N

71

545

298

Mean

8,61

11,32

10,37

SD

2,79

3,05

2,85

N

70

528

293

Mean

25,29

29,74

27,19

SD

5,81

7,68

7,41

N Length\Platform thickness ratio

Platform angle

70

528

293

Mean

10,2

6,25

6,09

SD

7,45

2,29

2,11

59

407

245

Mean

N

82,71

84,57

82,9

SD

7,55

6,06

7,06

N

70

528

293

bulb of percussion (Inizan et al., 1999; Pelegrin, 2000). Replication using local raw materials is necessary to confirm the validity of the link between these variables and percussor type. The platforms are further classified according to shape (Figs 3 and 4), and platform thickness, platform width and platform angle measurements are in Tables 3–5. The ratio of length to platform thickness (Tables 3 and 4) is calculated to provide a measure of the platform in relation to the piece as a whole. The technological study of the end-products is complemented by statistical analyses of these products. The length, width, thickness, platform width, platform thickness, exterior platform angle and ratio of length to platform thickness of the blades and points have been analysed. Univariate statistical indices such as the mean and standard deviation are in Tables 3–5. In view of the fact that the Howieson’s Poort is readily distinguished from the MSA I and MSA II on the basis of retouched types and smaller

426

Exploring and quantifying technological differences

size in all variables, further multivariate statistical analyses are undertaken exclusively on the MSA I and MSA II (Wurz et al., 2003). Multivariate statistical methods are preferred to investigate the differences between the MSA I and MSA II end-products, because the inter-correlation between the variables limits the potential of univariate statistical methods to distinguish between these sub-stages. Only those blades and points with all seven variables recorded are included. This resulted in a reduced blades

Figure 3 Platform type for blades, D-sample.

Figure 4 Platform type for points, D-sample.

427

From Tools to Symbols

data set consisting of 77 observations for MSA I, 395 for MSA II L and 212 for MSA II U. The reduced points data set consists of 59 MSA I observations, 407 MSA II L and 245 MSA II U observations. Multivariate analysis of variance, a one-way MANOVA with all seven variables treated simultaneously as dependent variables, was undertaken. To uncover which particular variables and sub-stages caused the differentiation between the means of the three groups, 95 per cent simultaneous confidence intervals for all pairwise contrasts for mean effects of the three groups were obtained. CVA biplot methodology in combination with alpha bags was used to graphically display the overlap and separation among the MSA I, MSA II L and MSA II U groups. To evaluate the success of the discriminating procedures, a CVA biplot classification procedure was undertaken (see Wurz et al., 2003 for a detailed discussion on the methodology and relevant tables and figures). The inferences drawn from these procedures and the CVA biplots with 95 per cent alpha bags are discussed and illustrated here.

Comparing the MSA I, MSA II and Howieson’s Poort Raw material selection The majority of the Klasies River artefacts were made in quartzite from beach cobbles that occur close to the site. In the MSA I and MSA II more than 98 per cent of the artefacts are in quartzite (Wurz, 2002) and it is only in the Howieson’s Poort that appreciable numbers of artefacts, approximately 30 per cent, are in non-quartzite fine-grained raw materials. The raw material composition of the Howieson’s Poort distinguishes it from the MSA I and MSA II. Cores The few preforms or early-stage cores (Table 1) indicate that reduction was initiated either by striking a thick large flake from a cobble by hard hammer percussion to produce a core blank, or by splitting a cobble. On a superficial level, all the informative cores from the MSA I, MSA II and Howieson’s Poort appear to be of a similar volumetric conception since they consist of an inactive volume opposed by a flat active surface (Fig. 5). However, closer examination shows that two types of volumetric conceptions can be inferred. The most numerous kind of core has a radially shaped under-surface, often with cortex, and an upper surface with a slightly concave profile. Such cores with blade scars are more numerous in the MSA I and Howieson’s Poort than in the MSA II. In the MSA II the majority of the cores evidence point removals (Table 1). The proximal platforms of the two-volume cores in the MSA II are prepared intensively with the distal platforms often much smaller. In the MSA II, end-products are almost always struck from the proximal platform. Flaking from the distal platform

428

Exploring and quantifying technological differences

is rare and if present, seems to have been for the preparation of the distal convexity or renewal of the upper surface (Fig. 5). In the Howieson’s Poort and MSA I, end-products have been struck from both the proximal and distal platform. There are more doubleplatformed cores in the Howieson’s Poort than the MSA I and this reflects a more intensive reduction strategy. On the two-volume cores of all the sub-stages, two types of lateral preparation are evident. In most cases the laterals were shaped removing overshot naturally or cortically backed blades triangular in section, but in a few cases centripetal flaking from the margins can be seen. The other type of core, much less numerous (e.g. Fig. 5a and d), has a pyramidal shape and has been reduced around half the circumference of the core. The inactive faces of these cores are elongated and evidence little effort in shaping. The pyramidal cores have minimally prepared single platforms. The active faces have been flaked recurrently and usually carry blade scars. This type of core only occurs in the Howieson’s Poort and MSA I, and not in the MSA II. It is problematic to relate the cores to technical schemes. The indications are that the MSA II cores are uniformly patterned and that a parallel two-volume Levalloislike scheme was followed. The patterning in the MSA I is more ambiguous, since two-volume cores as well as pyramidal cores, both with blade and point scars, occur. In the Howieson’s Poort the two-volume cores and pyramidal cores were used to produce blades only. It is not certain whether the two-volume cores of the MSA I and Howieson’s Poort represent Levallois or non-Levallois strategies. Some of them may be worked-out non-Levallois blade cores, or they may be similar to Levallois blade cores. The relationship between the pyramidal cores and the two-volume cores in the MSA I and Howieson’s Poort requires further investigation by means of refitting and experimentation. The trends in the end-products are clearer and can be used to make further inferences on technical schemes that were followed. End-products Blades and points and sections thereof constitute about 20–25 per cent of the assemblages (Wurz, 2000: 177). The Howieson’s Poort differs from the other substages in having smaller blades that were retouched into formal tools, the backed artefacts. In the MSA I there are blade as well as point end-products (Fig. 6). Contrary to the Howieson’s Poort, the size range of the regular MSA I ‘end-product’ blades varies from Howieson’s Poort-size to large. The points are all regularly shaped and are within a more restricted size range. The blades in the MSA II are generally irregularly shaped and do not have well prepared platforms – examples similar to Fig. 6(i) are few. The points, on the contrary, fulfil all the qualitative criteria of end-products. Obvious technological differences between the sub-stages become evident when

429

From Tools to Symbols

the platform variables are compared. In the Howieson’s Poort and MSA I, platforms distinct from those in the MSA II occur and these may have been produced with a soft hammer (Figs 3 & 4). In the Howieson’s Poort the majority of the platforms (87 per cent, n = 337) are lipped and associated with diffused bulbs while 41 per cent (n = 133) of the points and blades of the MSA I have the same characteristics. It is difficult to define ‘small’ platforms adequately. The platforms of the Howieson’s Poort are small in comparison with those of the other sub-stages (Table 5). Although the MSA I platforms seem substantially larger than the Howieson’s Poort platforms, the mean platform thickness of the ‘soft hammer’ pieces (4,8 mm, n = 133) is similar to the mean platform thickness of the Howieson’s Poort blades (3,7 mm, n = 383, Wurz, 2000: 188). The differences between the platform widths of these pieces are more apparent. It is interesting to note that when platform thickness is expressed in terms of the piece size, using the ratio of length to platform thickness, the average values for the Howieson’s Poort and MSA I (12:1) are the same (Wurz, 2000: 73). Aspects peculiar to the MSA I and Howieson’s Poort are more acute platform angles (Wurz 2000, 2002) and rubbing to smooth the dorsal edge of the platform (Fig. 6e–g). This is often associated with the removal of small trimming flakes on the dorsal edge of the platform. Table 5

Platform width and platform thickness (mm) of blades and points, MSA I, MSA II & Howieson’s Poort, D-sample. Blades Width

Thickness

MSA I

19,6

6,8

MSA II lower

24,5

MSA II upper

21,4

HP

11,6

Points n=

Width

Thickness

n=

323

25,3

8,6

70

10,0

1 338

29,7

11,3

527

9,2

655

27,2

10,4

293

3,7

383







Almost all the end-products of the MSA II have platforms that are regularly faceted and thick (Table 5; Fig. 6h–j) associated with prominent bulbs of percussion and ‘ring cracks’, reflecting the use of direct hard hammer percussion. Many of the MSA II platforms are ‘skewed’ or asymmetrical in that the bulb of percussion and point of impact are not symmetrical to the longitudinal axis of the point of blade. This has been related to the recurrent mode of flaking in a convergent strategy (Meignen, 1995). The multivariate discrimination procedures on the MSA I and MSA II end-products (but excluding the Howieson’s Poort end-products) allow more extensive differences to be identified between the MSA I and MSA II. The results of the MANOVA and

430

Exploring and quantifying technological differences

a b

c

e

g

d

f

h

Figure 5 Cores, MSA Il, MSA II and Howieson’s Poort (in quartzite unless stated otherwise). (a) MSA I, cave 1, Layer 19 (b) MSA I, cave 1B, PP38 (c) MSA II, cave 1B, PP38 (d) Howieson’s Poort, cave 2, sieving platform (silcrete) (e) MSA II, cave 1A, O50 (f) Howieson’s Poort, cave 1A, E50 (g) MSA II, cave 1, SASU (h) Howieson’s Poort, cave 2, strip 2

431

From Tools to Symbols

a

b

c

e

d

f

h

i

g

j

Figure 6 End-products MSA I, MSA II and Howieson’s Poort (in quartzite unless stated otherwise).

432

(a) Howieson’s Poort, cave 1A, H51

(f) MSA I, cave 1, Layer 19

(b) Howieson’s Poort, cave 1A, H51 (hornfels)

(g) MSA I, cave 1/1A, AA43

(c) Howieson’s Poort, cave 1A, H51 (silcrete)

(h) MSA II, cave 1, SASU

(d) Howieson’s Poort, cave 2

(i) MSA II, cave 1, SASU

(e) MSA I, cave 1/1A, AA43

(j) MSA II, cave1, SASU

Exploring and quantifying technological differences

Figure 7 CVA biplot of MSA I, MSA II Lower and MSA II Upper blades.

discriminant analysis indicate that the blade and point platform thickness are significantly smaller in the MSA I than the MSA II, that the ratio of blade and point length to platform thickness is significantly larger in the MSA I than the MSA II and that the points of the MSA II Upper are significantly shorter (see also Thackeray, 1989) than those of the MSA II Lower and MSA I. The statistically significant differences between the MSA I and MSA II from this analysis involve the platform variables, and not length, as previously considered from a univariate perspective (Wurz, 2002). The CVA biplots (Fig 7 and Fig 9) with 95 per cent alpha bags (Fig 8 and Fig 10) offer a comprehensible summary of how the MSA I and MSA II sub-stages compare at a glance. The CVA biplots accentuate the same differences as the discriminant procedures,

433

From Tools to Symbols

but in addition, they indicate the relative degree of differentiation in every variable. Whereas the blade and point ratios of platform thickness to length cause maximum separation between the MSA I and MSA II groups, the MSA II Upper and MSA II Lower are almost identical in terms of this ratio. On a more subtle level, blade thickness and, to a lesser extent, blade length also separate the MSA I from the MSA II. The MSA II groups are very similar, but the MSA II Upper has somewhat smaller values for most of the variables indicating an overall piece size reduction. The separation between the MSA I and MSA II is more pronounced in the points data than the blade data. This may be because almost all the points conform to the notion of an end-product while the

Figure 8 95 % alpha bag plots of MSA I, MSA II Lower and MSA II Upper blades.

434

Exploring and quantifying technological differences

Figure 9 CVA biplot of MSA I, MSA II Lower and MSA II Upper points.

blades data set contains flakes that are clearly preparatory products. The CVA biplot classification procedure (Wurz et al., 2003) indicates that elongated products can be classified as belonging to either MSA I or MSA II with an overall success rate of approximately 80 per cent. The classification procedure could not satisfactorily discriminate between MSA II U and MSA II L elongated products. This result is potentially significant in the context of South African Middle Stone Age studies because elongated products of unknown provenance can be placed in MSA I or MSA II with confidence, using routinely recorded variables. These multivariate analytical

435

From Tools to Symbols

Figure 10 95 % alpha bag plots of MSA I, MSA II Lower and MSA II Upper points.

procedures can be extended by including descriptive variables such as dorsal scar patterning and platform shape (Gardner, 2003).

Discussion and conclusion The goal of this analysis was to assess the variability within the Klasies River sequence by exploring different methods to discriminate between the MSA I, MSA II and Howieson’s Poort. Typological analysis is useful when retouched artefacts occur in some frequency and it allows some distinction between the Howieson’s Poort and MSA I and MSA II. However it does not provide a way to describe the technological

436

Exploring and quantifying technological differences

characteristics of the Howieson’s Poort, or of discriminating between unretouched end-products with similar dimensions such as those of the MSA I and MSA II. By incorporating technological variables of the cores and end-products a larger degree of discrimination is achieved. Technological consideration of the cores indicates that the MSA II differs from the MSA I and Howieson’s Poort with respect to volumetric conception and direction of end-product removal. In the MSA II, two-volume cores with parallel end-product removal from one preferential platform occur almost to the exclusion of other kinds of cores. In the other two sub-stages two-volume cores, often double-platformed, occur in combination with pyramidal single-platform cores. The patterning in the endproducts is clearer and reinforces the trends for cores. Again there are technological commonalities between the MSA I and Howieson’s Poort. Distinctive small platforms occur in the MSA I and Howieson’s Poort, but not in the MSA II. In the MSA I and Howieson’s Poort, some of the platforms were intentionally thinned, sometimes rubbed and perhaps struck using a soft hammer. In contrast, the MSA II has thick, hard, hammer-produced points, and blades with well-faceted large platforms. Clarifying patterning through univariate summaries of metrical attributes has proved less effective than multivariate approaches involving the same variables. Multivariate statistical analysis coupled with graphical representation gives an indication of trends and relative differences between variables whilst taking intercorrelation into account. The more effective pattern searching techniques show that the metrical and technological attributes of the platform are sensitive indicators of differences between the MSA I and MSA II and that blades and points can be assigned to the correct sub-stage with a high degree of accuracy. These methods enlarge the scope of techniques that can be used to compare Middle Stone Age assemblages from less than ideal contexts. The clearly observable typological, technological and quantitative differences can be used to argue that different technological standards were followed in the MSA I, MSA II and Howieson’s Poort. In the Howieson’s Poort, as in some Upper Palaeolithic blade strategies, similar-sized (standardised) blades were produced, perhaps using a soft hammer (Boëda, 1988a, b; Inizan et al., 1999). In the MSA I the end-products of different shapes and sizes are more typical of a Levallois-like Middle Palaeolithic blade strategy (Boëda, 1988b). The possibility cannot be excluded that a blade as well as a non-blade strategy was followed in the MSA I. In the MSA II the two-volumes cores have been set up to remove perhaps up to three thick end-products before repreparation, using a unipolar convergent Levallois point production scheme. These results are contrary to the conventional notion that the sequence at Klasies River shows little meaningful change through time. A larger degree of variability at

437

From Tools to Symbols

Klasies River, and possibly other South African Middle Stone Age sites, necessitates reconsideration of arguments that relate a lack of variability in the Middle Stone Age to non-modern behaviour. The debate on modern behaviour is marred by ambiguous definitions of what constitutes modern behaviour. Moreover, modern behaviour is often discussed in terms of individual sites even though collective early human behaviour is at issue. It is imperative that inter-site comparison, using more powerful methodologies, becomes the rule rather than the exception in South African Middle Stone Age studies. This would provide a way forward in the study of modern behaviour.

Acknowledgements Hilary Deacon is thanked for comments and Liezl van Pletzen Vos for the drawings. My thanks go to Niel le Roux and Sugnet Gardner for the statistical analysis. The University of Stellenbosch is thanked for supporting this study through a post-doctoral research fellowship. This paper is based partly on a Ph.D. study for which financial assistance was provided by the NRF (Centre for Scientific Development). The Struwig Trust facilitated research in the field and the Iziko Museums of Cape Town made collections available for the study. The organisers and sponsors are thanked for the opportunity to take part in the conference.

Bibliography Audouze, F. (1999). New advances in French prehistory, Antiquity 73: 167–75. Boëda, E. (1988a). Le concept Levallois et evaluation de son champ d’application. L’Homme Levallois. l’Homme de Néanderthal, Vol. 8, La Mutation, Liège, 1988, pp. 13–26. Boëda, E. (1988b). Le concept Laminaire; rupture et filiation avec le concept Levallois. L’Homme de Néanderthal, Vol. 8, La Mutation, Liège, 1988, pp. 41–59. Boëda, E. (1995). Levallois: a volumetric construction, methods, a technique. In (O. Bar-Yosef & H.L. Dibble, Eds). The definition and interpretation of Levallois Technology, pp. 41–68. Monographs in World Archaeology 23. Wisconsin: Prehistory Press. Chazan, M. (1997). Lead Review – redefining Levallois. Journal of Human Evolution 33, 719– 735. Deacon, H.J. (1992). Southern Africa and modern human origins. Philosophical Transactions of the Royal Society, London B 337, 177–183. Deacon, H.J. & Deacon, J. (1999). Human Beginnings in South Africa. Uncovering the Secrets of the Stone Age. Cape Town: David Philip. Deacon, H.J. (1995). Two Late Pleistocene-Holocene archaeological depositories from the southern Cape, South Africa. South African Archaeological Bulletin 50,121–131. Deacon, H.J. & Geleijnse, V.B. (1988). The stratigraphy and sedimentology of the main site sequence, Klasies River, South Africa. South African Archaeological Bulletin 4, 5–14.

438

Exploring and quantifying technological differences Deacon, H.J., Talma, A.S. & Vogel, J.C. (1988). Biological and cultural development of Pleistocene people in an Old World southern continent. In (J.R Prescott, Ed.) Early Man in the Southern Hemisphere, pp. S23–S31. Adelaide: Department of Physics and Mathematical Physics, University of Adelaide. Feathers, J. (2002). Luminescence dating in less than ideal conditions: case studies from Klasies River main site and Duinefontein, South Africa. Journal of Archaeological Science 29(2), 177–194. Gardner, S. & le Roux, N.J. (2003). Exploring differences and overlap between MSA artefacts using MCA and biplot methodology. Conference of Correspondence Analysis and Related Methods, 29 June–2 July, Barcelona. Goodwin, A.J.H. & Van Riet Lowe, C. (1929). The Stone Age cultures of South Africa. Annals of the South African Museum 27. Grün, R, Shackleton, N.J. & Deacon, H.J. (1990). Electron-spin resonance dating of tooth enamel from Klasies River Mouth Cave. Current Anthropology 31, 427–432. Inizan, M., Reduron-Ballinger, M., Roche, H. & Tixier, J. (1999). Technology and terminology of knapped stone. Nanterre: CREP. Klein, R.G. (1992). The archaeology of modern human origins. Evolutionary Anthropology 1, 5–14. Knutsson, K. (1988). Making and using stone tools - the analysis of the lithic assemblages from Middle Neolithic sites with flint in Västerbotten, Sweden. Ph.D Dissertation, University of Uppsala. Marks, A.E. & Volkman, P. (1983). Changing core reduction strategies: a technological shift from the Middle to the Upper Palaeolithic in the Southern Levant. In (E. Trinkhaus, Ed.) The Mousterian Legacy: human biocultural change in the Upper Pleistocene, pp. 13–34. British Archaeological Reports, International series, 164. McBrearty, A. & Brooks, A.S. (2000). The revolution that wasn’t: a new interpretation of the origin of modern human behavior. Journal of Human Evolution 39, 453–563. Mithen, S. (1996). The Prehistory of the Mind: A Search for the Origins of Art, Religion and Science. London: Thames & Hudson. Meignen, L. (1995). Levallois lithic production systems in the Middle Paleolithic of the Near East: the case of the unidirectional method. In (H.L. Dibble & O. Bar-Yosef, Eds) The definition and interpretation of Levallois Technology. Monographs in World Archaeology 23, pp. 31–380. Wisconsin: Prehistory Press. Noble, W. & Davidson, I. (1996). Human Evolution, Language and Mind – A Psychological and Archaeological inquiry. Cambridge: Cambridge University Press. Pelegrin, J. (1995). Technologie lithique: le Châtelperronien de Roc-De Combe (Lot) et de La Côte (Dordogne). Cahiers du Quarternaire 20. Paris: CNRS Éditions. Pelegrin, J. (2000). Les techniques de débitage laminaire au Tardiglaciaire: critères de diagnose

439

From Tools to Symbols et quelques réflexions. L’Europe Centrale et Septentrionale au Tardiglaciaire. Table-ronde de Nemours, 13–16 May 1997. Mémoires du Musée de Préhistoire d’Ile de France, 7. Révillion, S. (1994). Les industries laminaires du Paleolithique Moyen en Europe Septentrionale. L’Exemple des Gisements de Saint-Germain-des-Vaux/Port-Racine (Manche), de Seclin (Nord) et de Riencourt-le-Bapaume (Pas-de-Calais). Lille: Universite des Sciences et Technologies de Lille-Flandres-Artois. CERP 5. Singer, R. & Wymer, J. (1982). The Middle Stone Age at Klasies River Mouth in South Africa. Chicago : Chicago University Press Thackeray, A.I. (1989). Changing fashions in the Middle Stone Age: The stone artefact sequence from Klasies River main site, South Africa. African Archaeological Review 7, 33–57. Thackeray, A.I. & Kelly, A.J. (1988). A technological and typological analysis of Middle Stone Age assemblages antecedent to the Howieson’s Poort at Klasies River main site. South African Archaeological Bulletin 43, 5–26. Thackeray, A.I. (1992). The Middle Stone Age south of the Limpopo River. Journal of World Prehistory 6, 385–440. Van Peer, P. (1992). The Levallois reduction strategy. Monographs in world archaeology No. 13. Wisconsin: Prehistory Press. Van Peer, P. (1995). Current issues in the Levallois Problem. In (H.L. Dibble & O. Bar-Yosef, Eds) The definition and interpretation of Levallois Technology, pp. 1–10. Monographs in World Archaeology 23. Wisconsin: Prehistory Press. Vogel J.C. (2001). Radiometric dates for the Middle Stone Age in South Africa. In (P.V. Tobias, M.A. Raath, J. Maggi-Cecchi & G.A. Doyle, Eds). Humanity from African Naissance to Coming Millennia – Colloquia in Human Biology and Palaeoanthropology, pp. 261–268. Florence: Florence University Press. Volman, T.P. (1981). The Middle Stone Age in the southern Cape. Ph.D. Dissertation. University of Chicago. Volman, T.P. (1984). Early prehistory of southern Africa. In (R.G. Klein, Ed.) Southern African Prehistory and Palaeoenvironments, pp. 169–220. Rotterdam: Balkema. Wadley, L. (2001). What is cultural modernity? A general view and a South African perspective from Rose Cottage Cave. Cambridge Archaeological Journal 11(2), 201–221. Wurz, S. (2000). The Middle Stone Age sequence at Klasies River, South Africa. D.Phil Dissertation. University of Stellenbosch. Wurz , S. (2002). Variability in the Middle Stone Age lithic sequence, 115 000 – 60 000 years ago at Klasies River, South Africa. Journal of Archaeological Science 29(9),1001–1015. Wurz, S., Le Roux, N.J, Gardner, S. & Deacon, H.J. (2003). Discriminating between the endproducts of the earlier Middle Stone Age sub-stages at Klasies River using biplot methodology. Journal of Archaeological Science 30(9), 1107–1126.

440

Stratigraphic Integrity of the Middle Stone Age Levels at Blombos Cave Christopher S. Henshilwood Centre for Development Studies, University of Bergen, Nygårdsgaten 5, N- 5015, Bergen, Norway African Heritage Research Institute, 167 Buitenkant Street Gardens, Cape Town 8001, South Africa

Abstract Over the past twelve years Blombos Cave has yielded a well-preserved sample of faunal and cultural material in Middle Stone Age (MSA) levels. The MSA phases are separated from the < 2 thousand years (Ka) Later Stone Age (LSA) levels by a blanketing aeolian dune sand 5–50 cm thick dated at c. 70 Ka by optically stimulated luminescence (OSL). Careful examination of sediments and anthropogenically derived deposits within individual levels over the past four years has allowed us to sub-divide the MSA levels into three major phases, namely (1) a Still Bay, or M1, phase dated at c. 75 Ka by OSL and thermoluminescence, (2) a middle M2 phase provisionally dated by OSL at c. 78 Ka, (3) a lower M3 phase provisionally dated by OSL at > 100 Ka. Artefacts unusual in a Middle Stone Age context have been recovered from all three phases. These include marine shell beads, bifacial ‘laurel-leaf’ points, bone tools, engraved bone and engraved ochre in M1, and bone tools in M2. The likely symbolic significance of these finds suggests levels of cognitively modern behaviour not previously associated with Middle Stone Age people. Key issues discussed in this paper are the possibility of admixture of older and younger deposits at Blombos Cave and whether some MSA artefacts derive from the LSA levels (also see Klein, 2000: 29).

Résumé Au cours des douze dernières années, la grotte de Blombos a livré des restes fauniques et culturels bien conservés dans les niveaux du Middle Stone Age (MSA). Les couches du MSA sont séparées de celles du Later Stone Age (LSA), datés de moins de 2 Ka, par une dune éolienne épaisse de 5 à 50cm, datée aux alentours de 70 Ka par luminescence stimulée optiquement (OSL). Un examen détaillé du dépôt a permis au cours des quatre dernières années de subdiviser les couches du MSA en trois phases principales, soit (1) une phase Still Bay ou M1 datée par OSL et thermoluminescence à environ 75 Ka, (2) une phase moyenne M2 datée provisoirement

441

From Tools to Symbols par OSL à environ 78 Ka, (3) une phase ancienne M3 provisoirement datée par OSL à plus de 100 Ka. Des trois phases proviennent des artefacts inhabituels pour l’MSA, tels que des objets de parure en coquillage, des pointes bifaciales en forme de feuille de laurier, des outils en os, des gravures sur os et un fragment ocre gravé dans M1, des outils en os dans M2. La signification symbolique probable de ces objets semble indiquer un degré de modernité cognitive qui par le passée n’était pas associée aux populations MSA. Dans cette contribution nous discutons de l’hypothèse, évoquée par exemple par Klein (2000 : 29) d’un mélange de matériel provenant de couches différentes qui aurait pu amener dans les couches MSA des objets déposés originellement dans les couches LSA.

Introduction Blombos Cave (BBC), situated near Still Bay in the southern Cape (34º25'S, 21º13'E), is some 100 metres from the coast and 35 metres above sea level (Fig. 1a). The interior of the cave contains 55 m2 of visible deposit (Fig. 1b, 2) with a depth of about 4–5 metres at the front and 3 metres toward the rear. Chambers within the cave on the western and eastern sides contain further archaeological deposits but the extent of these cannot yet be estimated. When test excavations at BBC commenced in June 1991 the cave entrance was almost totally sealed by dune sand, also about 20 cm of undisturbed aeolian sand overlay the surface of the LSA, indicating no disturbance of the cave’s contents since the final LSA occupation c. 290 years ago. The LSA deposits are less than 2 Ka, not as deep as the MSA, and are more massively bedded and undistorted. In addition, burned layers tend to be thicker and several appear to preserve their original hearth-like structures. In the MSA levels the matrix is composed mainly of aeolian, marine-derived dune sand, blown in through the cave entrance, and is intercalated with marine shell, decomposed humic materials and limestone, and wind-borne halites. Ground waters rich in CaCO3 percolate through the cave roof and walls, creating an environment suited to the preservation of bone and shell, particularly near hearths and ash deposits. Carbonised partings represent occupation horizons and separate major units. The MSA deposits undulate considerably from the back to front of the cave due to subsidence that produces a ‘wrapping effect’ over the rock falls and occasional slump faults into gaps between rocks (Henshilwood et al., 2001a).

The MSA phases Sterile yellow dune sand c. 5–50 cm thick named BBC Hiatus blew into the unoccupied cave during lowered sea levels at c. 70 Ka (Fig. 2, 3a, b). Shortly afterwards the cave entrance was blocked by a dune over 40 metres high. It is likely that the cave only reopened after the mid-Holocene when higher sea levels (Miller et

442

Strategic Integrity of the Middle Stone Age Levels at Blombos Cave

Figure 1a Location of Blombos Cave and view of cave entrance.

Figure 1b View of interior of Blombos Cave looking seawards.

443

From Tools to Symbols

Figure 2 Blombos Cave layout and stratigraphic section.

al., 1993) eroded the base of this dune 30 metres below the cave causing the dune at the entrance to subside. Fossilised stalactites that had formed atop this dune when the cave entrance was closed are visible on the cliff face about 5 metres above the cave entrance. Relic dune sections that date to this event are evident below and to the east of the cave and date to the same time period as the Hiatus phase in the cave (Jacobs, pers. comm.). BBC Hiatus separates the LSA and MSA across more than 95 per cent of the excavated area and provides visible evidence that the LSA occupation did not disturb the underlying MSA deposits (Fig. 3). Possible exceptions are squares E2/F2 and E3/F3 (Fig. 2), where the sterile dune layer is relatively thin (< 5 cm) (Fig. 4b), probably due to clearing or the excavation of bedding hollows by LSA people. Anthropogenically turbated LSA deposit is clearly visible in the D1/D2 section (Fig. 4c). However, even where the BBC Hiatus level is thinnest there is no visual evidence that LSA people disturbed MSA deposits (Fig. 4b). The five uppermost layers below BBC Hiatus are assigned to the M1 phase (Fig. 2). Small basin-shaped ash and carbon hearths are common in this phase. Carbonised sand and organic ‘partings’, a few millimetres thick, act as visual markers for the separation of discrete occupation layers. M1 phase lithics are typified by Still Bay type

444

Strategic Integrity of the Middle Stone Age Levels at Blombos Cave

Figure 3 West face showing separation of LSA and MSA phases by Hiatus level.

Figure 4 (a) Possible area of charcoal penetration through Hiatus level; (b) Hiatus level; (c) turbated LSA deposits in Square D1/D2.

445

From Tools to Symbols

bifacial foliate points, the fossile directeurs of the Still Bay Industry (Goodwin & Van Riet Lowe, 1929) (Fig. 5a), and end and side scrapers (Fig. 5b) (Henshilwood et al., 2001a). Two slabs of engraved ochre (Henshilwood et al., 2002), forty-one shell beads, formal bone tools and an engraved bone (d’Errico et al., 2001) come from this phase. Four levels typified by carbonised deposits, large hearths and shellfish comprise the M2 phase. Few bifacials were recovered in the M2 phase. Shaped bone tools, possibly used as awls and projectile points (Fig. 6), came mainly from the CFA/CFB layers in M2, but also from the M1 phase. Dominant in the M3 phase are shellfish deposits and a high density of ochre pieces; the lithic assemblage has yet to be analysed but a preliminary study indicates it does not conform to the typical MSA I or MSA II pattern observed at Klasies River (Soressi, pers. comm.).

Dating the MSA levels Luminescence dating Two luminescence-based dating methods were applied to the BBC MSA layers (Henshilwood et al. 2002). Thermoluminescence (TL) dates were obtained for five burnt lithic samples from the MSA phase M1. The mean age for the lithic samples is 77 ± 6 Ka. To confirm the stratigraphic integrity, optically stimulated luminescence (OSL) dating was applied to the aeolian dune (Hiatus) (Fig. 2) that separates the LSA

Figure 5 Bifacial points and end and side scrapers from M1 and M2 phases.

446

Strategic Integrity of the Middle Stone Age Levels at Blombos Cave

and MSA layers (cf. Henshilwood et al. 2002: Jacobs et al., 2003a, b). Multiple grain measurements using a single aliquot regenerative procedure yielded a depositional age of 69 ± 5 Ka. A second approach was combining OSL signals from sand grains to generate synthetic aliquots. These provide a depositional age of 70 ± 5 Ka and confirm the antiquity of the upper MSA layers (Henshilwood et al. 2002; Jacobs et al., 2003a, b). The M2 levels have been similarly dated by the OSL method. Preliminary dates for these levels are c. 78 Ka (Jacobs, pers. comm.). A probable date of > 100 Ka for the M3 levels is suggested (Jacobs, pers. comm.). Independent evidence against migration of LSA artefacts into the MSA levels across the site is confirmed by the OSL dating program (Jacobs et al., 2003a, b). Although several data handling processes were applied to dating the sterile sand layer (BBC Hiatus) that separates the MSA and LSA levels, all of them provided ages with similar precision. Jacobs et al. (2003a) report that the lack of intrusive young or old grains indicates that the Hiatus sand layer is not disturbed, thus confirming the integrity of the underlying MSA levels. C dating In Square E2 some fragments of well-preserved charcoal have penetrated up to 30 cm into MSA levels (Fig. 4a). These are an unusual find as charcoals in the MSA levels at BBC are mostly poorly preserved due to the antiquity and alkaline nature of these sediments. Charcoal is better preserved in acidic sediments that conversely do not favour the preservation of shell and bone. Four charcoal samples recovered from Square E2 are 14C dated to c. 2 Ka (Henshilwood et al., 2001b: 637). In one case this is due to a sample from an LSA layer mistaken, due to an acronym error, for an MSA layer. The three remaining dates are statistically identical to dates obtained from the lowermost LSA layer (cf. Henshilwood, 1996; Henshilwood et al., 2001b: 637), suggesting these three charcoal fragments derive from the overlying LSA. The probable reason for the charcoal contamination in this area can be attributed to a slumped burrow, digging stick hole or post hole (Fig. 4a). One charcoal piece and five marine shell opercula (Turbo sarmaticus) from the same area and depth provide infinite dates > 32 Ka. There is no further evidence of LSA derived artefacts penetrating the MSA in this area (for a fuller explanation cf. Henshilwood et al., 2001a, b). 14

Stratigraphic integrity Potential causes of an admixed assemblage at BBC are (1) human and nonhuman digging during the historical or LSA periods; (2) geological processes, such as sedimentary erosion and re-deposition, faulting and slumping; (3) and cross-cutting of LSA and MSA levels during excavation (Henshilwood et al., 2001b).

447

From Tools to Symbols

1. If large-scale burrows were dug through the base of the LSA they would be readily visible during excavation because of clear differences in colour between the bright yellow sterile dune sand and surrounding dark grey to black layers (Fig. 3). No burrows penetrated the yellow dune sand and MSA levels in the excavated area along grid lines 4, 5 and 6 (Fig. 2) where the sand is particularly thick. There is evidence of a small vertical burrow or possible posthole- or digging stick-like feature in square E2 that might be responsible for the young charcoal in the MSA levels. 2. Over most of the excavated area draping and slumping of sediments affects only the MSA and not the LSA (Fig. 3). The basal LSA sediments do not reflect the marked undulations of the MSA surface that underlies the sterile aeolian sand. Compression and draping of the MSA opened up a 5–10 cm gap between the rear cave wall and the MSA sediments in squares F2, F3, G3, G4 and H5 (Fig. 2). Dune sand and LSA material percolated into this narrow band and contaminated parts of the MSA here. Material recovered from this area has been excluded from all BBC analyses despite the fact that most of the recovered material is clearly MSA and the contaminated material accounts for only a small proportion of these deposits. (Henshilwood et al., 2001a). Bone and shell recovered from the MSA levels is considerably darker in colour and hue than that from the LSA and experienced excavators at BBC are readily able to distinguish these materials by colour and hue alone. Fish bone from the LSA levels is the most common contaminant in the squares near the cave wall, probably as the bones are mostly small in size and can easily slide down the small gap between the deposits and cave wall. The LSA fish bone looks fresh and is light in colour as opposed to the rich coffee colour of the MSA fish bone. Similarly, the LSA shells are very fresh looking while the MSA shells are less well preserved, many are chemically eroded and have lost their nacre. 3. Distortion within MSA layers was only properly appreciated after the 1992 and 1997 field seasons. During these two seasons there was some crosscutting of layers within the MSA but not between the LSA and MSA. After 1997, individual layers were systematically redefined according to basal markers and content, and a new nomenclature was introduced. This strategy has been successful and provides confidence that MSA materials recovered from the 1998 and subsequent excavations are temporally and spatially secure within an MSA context (Henshilwood et al., 2001a).

448

Strategic Integrity of the Middle Stone Age Levels at Blombos Cave

Later Stone Age artefacts Ostrich eggshell beads No ostrich eggshell (OES) beads have been recovered from the MSA levels at BBC or from the Hiatus, yet they are relatively common in the LSA and 86 were recovered from squares E2/E3/E4 in the first excavation season (Henshilwood, 1995). Twentyone of these beads come from the lowermost LSA layers that lie directly above the sterile dune layer where it is thinnest (Fig. 4b), yet none penetrated into the Hiatus or MSA. OES beads, given their small size (< 0,5 cm), are more likely to be transported downwards into lower levels than larger artefacts such as bone tools (minimum 6 cm). No beads were recovered from the MSA levels in 1992. During 1997–2000 a further 140 OES beads were excavated from LSA levels, and some from the layer directly above the Hiatus phase, yet again none were recovered from Hiatus or MSA phases. The only area in which beads were found at MSA depths is from an area of known mixed context against the cave wall in square G4 (Fig. 2) and these deposits are excluded from analyses. Bone tools Seven bone tools were recovered from the LSA levels. Three came from the upper LSA, two from the middle phase and two from the lower phase. Bone tools are considerably less common in the LSA levels (0,8 per m3) compared to the MSA (2,3 m3). Since no displacement between these two main horizons is observed amongst typical LSA and MSA finds (e.g. eggshell beads, bifacial points) a hypothesis for admixture would need to explain the selective displacement of the majority of purported ‘LSA’ bone tools into MSA levels. The sum of evidence lends no credence to this argument (see Henshilwood et al., 2001b). Spatial distribution of the LSA bone tools is restricted mainly to one square (E4) (Fig. 2) while those from the MSA levels are spatially widespread and particularly abundant in squares E3/4 and F3, where only one LSA bone tool is found (Fig. 2). Many of the MSA bone tools were recovered from a 7 m2 area where the MSA and LSA layers are distinctly separated by thick Hiatus level (Fig. 2 and Fig. 3). Evidence of intrusion of bone tools from the LSA layers above would be clearly visible in the clean sand of the Hiatus but this is not the case. On technical grounds we have demonstrated that the LSA bone tools are distinctly different to the MSA bone tools (Henshilwood et al., 2001b), both in manufacturing techniques and in use wear analyses. On colour alone, the MSA and LSA bone tools are visibly distinct (Henshilwood et al., 2001b). If the argument for mixing is that some bone tools moved from the LSA into the MSA yet others remained, then why should only the distinct group that is found in the MSA levels move, yet the other distinct set of tools found in the LSA remain in situ? Clearly, this is not a reasonable argument.

449

From Tools to Symbols

Middle Stone Age cultural artefacts Bifacial points Bifacial points, the fossile directeur of the Still Bay Industry are confined mostly to the uppermost MSA layers at BBC (Fig. 5a; Henshilwood et al., 2001a). More than four hundred bifacial points were recovered from the M1 phase and a few from the M2 phase. The majority came from layers CA and CB (Fig. 2) located directly below the Hiatus. Despite the relative proximity of the bifacial rich layers to the lowermost LSA (5–50 cm), no bifacial points have been recovered from LSA layers since formal excavations commenced in November 1992 (Henshilwood et al., 2001a). It is clear that bifacials did not penetrate upwards through the Hiatus level. If there was turbation and mixing of LSA and MSA deposits, then why are bifacials not found in LSA levels? Bone tools Thirty shaped bone tools were recovered from the M2 phase at BBC during 1992– 2002 (Fig. 6; Henshilwood et al., 2001b). The stratigraphic integrity of these tools has been challenged (e.g. Klein, 2000, 29), with later suggestions that at least some of the ‘formal’ bone points derive from the LSA levels. Burrows One possible avenue of intrusion is via burrows. First, there is no visible evidence of burrows in the Hiatus (e.g. Fig. 3) and second, if elongated objects such as bone tools were to fall down such burrows, they would most likely lie in a vertical rather than horizontal position. All the MSA bone tools, when recovered, were oriented in the same plane as the natural bedding, generally horizontal or near horizontal. Intrusion of bone tools into the MSA by means of burrows can thus be excluded. Furthermore, the sand samples extracted from the Hiatus level for OSL dating show no evidence of mixing in this level (Jacobs et al., 2003a, b), as would be expected if there was extensive burrowing. Chemical testing Two shaped and polished bone tools, 8954 and 8947 (Fig. 7), from MSA levels and a random selection of eleven animal bones recovered from MSA and LSA levels in 1992/7 were tested for relative percentages of carbon and nitrogen (C, N). Bone protein is known to degrade over time, resulting in the loss of carbon and nitrogen; hence bones from the LSA and MSA levels will have considerably different concentrations of these two elements (Henshilwood & Sealy, 1997). One of these bone tools (8947) came from square E3, adjacent to E2, the source of the contaminated 14C charcoal dates. The results show that both the bone tools tested have the same C/N ratios as the

450

Strategic Integrity of the Middle Stone Age Levels at Blombos Cave

MSA bone and are unambiguously from the MSA levels (Fig. 7; also see Henshilwood & Sealy, 1997 for a fuller discussion). This independent test conducted on two of the formal bone tools, said to derive from the LSA levels (Klein, 2000, 29), provides absolute evidence that these tools cannot derive from the < 2 Ka levels. Size

The cortical thickness of the bone used for MSA tools is significantly thicker than those for the LSA, and MSA tools are generally longer and wider (see Henshilwood et al., 2001b). The larger size of the MSA tools is the reverse of what would be expected if these were actually LSA tools displaced into the MSA layers. Smaller items are more likely to travel down rodent burrows and other intrusions. Similarly, the much greater age of the MSA deposits puts the MSA tools at higher risk of being post-depositionally fragmented, and thus their final state would be expected to be smaller. Despite this greater age, they are larger (see Henshilwood et al., 2001b). The three independent lines of evidence above demonstrate clearly that the MSA bone tools from the > 70 Ka levels are not intrusive (also see Henshilwood et al., 2001b: 638–642).

Figure 7 Carbon/nitrogen % in LSA and MSA bones from Blombos Cave and C/N % in bone tools SAM – AA 8947 & 8954.

451

From Tools to Symbols

Engraved ochre During excavations in 1999 and 2000 respectively two engraved ochre pieces (SAM – AA 8937, see Fig. 8a, b; SAM-AA 8938, see Fig. 8c, d) were recovered in situ from layers CC (Square E6) and CD (Square H6) (Fig. 2). Both engraved pieces were surrounded by intact MSA hearths and located in a matrix of undisturbed and wellconsolidated mixed ash and sand (Henshilwood et al, 2002). In Square H6 the Hiatus dune sand is more than 30 cm thick (Fig. 8), and over 10 cm thick in Square E6. There is no sign that the Hiatus layer has been disturbed in either of these squares. In the overlying MSA levels (CA, CB) the lithic artefacts are all typologically MSA with no evidence of intrusive LSA artefacts. Neither of these engraved ochre pieces derives from the LSA levels.

Shell beads During 1997–2000, thirty-nine Nassarius kraussianus perforated shell beads were recovered from the M1 phase and two from the M2 phase (Henshilwood et al., 2004; d’Errico et al., in press). Commonly called the tick shell, it is a scavenging gastropod

Figure 8 Engraved ochre SAM – AA 8938 in situ in Square H6a, Level CD. a–b. Engraved ochre SAM – AA 8937. c–d. Engraved ochre SAM – AA 8938

452

Strategic Integrity of the Middle Stone Age Levels at Blombos Cave

adapted to estuarine environments. The distribution of the age classes of the shell beads in the MSA levels is evidence they were not introduced accidentally. All the recovered MSA shells are deliberately perforated (Fig.9 a, c), and most have unique medium-size perforations located near the lip (Fig. 9c). A number of the shells show traces of ochre both inside the shells and on the outer surface (Fig. 9 a, b). Microscopic analysis of MSA tick shells reveals a distinct use-wear consisting of facets which flatten the outer lip or create a concave surface on the lip (Fig. 9d) and provides the most important evidence that the shells were strung as beads, possibly for necklaces or clothing decoration.

Figure 9 Nassarius kraaussianus shell beads from the MSA at Blombos Cave: (a) perforated hole opposite to the shell aperture with ochre traces on shell (b); (c) perforated shell with wear traces on perforation due to stringing; (d) wear facets on shell aperture due to use wear.

453

From Tools to Symbols

Although tick shells are also found in the < 2 Ka LSA levels at the cave we know that the MSA beads do not derive from these levels. First, the size distribution of the MSA shells is significantly different, as confirmed by a Mann-Whitney U-test (p < 0,0001), from that of the LSA shells (d’Errico et al., in press). Second, N. kraussianus beads from LSA levels have no wear facets similar to those on the MSA beads; shell beads from the MSA are dark orange or black in colour, while those from the LSA are white or pale beige; 52 per cent of the LSA shell beads have broken lips while this is observed on only two MSA specimens; almost all MSA shells were found in groups of 2–17 clustering in the same or neighbouring 50 × 50 cm excavation squares, with each group being recovered in a single excavation day. Within a group, shells display a similar size, shade, use-wear pattern and type of perforation. Each cluster may represent beads coming from the same beadwork item, lost or disposed of during a single event (Henshilwood et al., 2004; d’Errico et al., in press).

Discussion The shell beads, bone tools, engraved ochres and engraved bone from the MSA levels at BBC provide strong support to the argument that modern human behaviour developed in Africa at least 70 Ka ago (Henshilwood et al., 2002, 2004). To demonstrate that these artefacts derive with certainty from the MSA levels, and are not intrusive from the LSA, is essential for the human behaviour debate to develop further. This is only possible if the focus is directed at published factual evidence, and not based on hearsay, red herrings or intuition. Evidence presented above and in recent journal articles (see Henshilwood & Sealy, 1997; Henshilwood et al., 2001a, b, 2002, 2004; d’Errico et al., 2001, Jacobs et al., 2003a, b; d’Errico et al., in press) indicates that the artefacts from BBC, central to the human behavioural debate, were indubitably recovered in situ and date from levels older than 70 Ka (Henshilwood et al., 2002; Jacobs et al., 2003a, b). Increasingly, evidence is being recovered from MSA sites in Africa (e.g. Yellen et al., 1995; Deacon, 1998; Watts, 1999; McBrearty & Brooks, 2002; Parkington et al., this volume; Thompson et al., 2004; White et al., 2003; Wurz, 2000, 2003) and Middle Paleolithic sites in Eurasia (Hovers et al., 2003; Taborin, 2003) pointing to an early origin for behavioural modernity, long before the Eurasian Upper Paleolithic ‘Revolution’ or the African LSA ‘Squib’ at c. 45 Ka. What is this early evidence telling us? The capacity for modern human cognitive behaviour is likely to have been physically in place at about the time anatomically modern humans evolved (Donald, 1991), perhaps by 200 Ka. This is not the same as saying modern human behaviour was mediated by symbolism at 200 Ka; this might have come much later. However, it is probable that there was no single trajectory that led to all H. sapiens in Africa

454

Strategic Integrity of the Middle Stone Age Levels at Blombos Cave

appearing behaviourally modern, as reflected in their material culture, at one single point in time. Rather, a mosaic pattern of development towards cultural modernity with periods of innovation, stasis and even regression might be appropriate. This, in fact, is what the material record of the MSA in Africa reflects, despite the relative paucity of excavations on this vast continent compared to that of the geographically restricted Franco-Cantabrian Middle Palaeolithic. We can speculate that there was a broad Pan-African cognitive system operational in the African MSA with regional variation in socially mediated behaviours due to, among others, ecological variability (e.g. coastal vs inland habitats), cultural and/or social variability, demography (e.g. a likely population decline after 60 Ka in southern Africa) and the effects of an adequate diet on brain growth and development (e.g. omega 3 and 6 fatty acids present in some marine-derived foods). In the southern Cape at 75 Ka it is possible that the emergence of symbolically mediated behaviour was driven by population growth or demographic change in coastal areas during mild climatic periods in the latter stages of oxygen isotope (OI) stages 5b and the start of OI 5a (Fig. 10). During the latter stages of OI 5a and 4 desiccation

Figure 10 Oxygen Isotope Stages showing chronological location of Blombos Cave Hiatus, M1, M2 & M3 phases.

455

From Tools to Symbols

and environmental degradation is likely in the interior Karoo regions that may have forced populations to move over the Cape Folded Belt mountain range towards the environmentally benign coastal plains. The southern Cape coast may have been one of the areas subject to demographic pressure as a result of this. At about the same time a rapidly retreating coastline during the cold OI 4 must also have resulted in increased competition for dwindling coastal resources. We can speculate this was a time when exchange and mediation mechanisms (hxaro) were important to maintaining good social relations (Wiessner, 1999; Henshilwood & Marean, 2003), leading to a growth in symbolically imbued material culture. Apart from BBC, no Still Bay period sites containing well preserved deposits have yet been discovered, apart from Peers Cave (Skildergat), which was haphazardly excavated during the 1920s–1960s (e.g. Peers, 1929; Jolly, 1947, 1948; Malan, 1955). At Dale Rose Parlour on the Cape Peninsula (Schirmer, 1975), and at Hollow Rock Shelter in the Cederberg (Evans, 1994), large numbers of Still Bay bifacial points were recovered from levels that may date to c. 75 Ka (OI Stage 5a), but no organic material survived. There is also the possibility that some sites occupied during this period (OI 5a–OI 4) of lowered sea-levels are now underwater. During southern Cape coastal surveys, particularly in the De Hoop Nature Reserve, it was noticed that a number of caves, presently just above sea level, are scoured of MSA deposits by previous higher sea levels, probably during the mid-Holocene (+ 2–3 metres above sea level) (Miller et al., 1993). Evidence of these MSA deposits adheres to the walls of the caves and in these remnant sections MSA lithic artefacts are clearly visible. The paucity of MSA sites dating to the Still Bay phase hampers our efforts at present to further build on a human behaviour model for southern Africa. Continuing excavations at BBC, and at MSA sites in the Cape (e.g. Nilssen & Marean, 2002; Parkington et al., this volume) and elsewhere in Africa (Thompson et al., 2004) are likely to better inform us about this vital period in the development of modern humans.

References d’Errico, F., Henshilwood, C.S., & Nilssen, P. (2001). An engraved bone fragment from ca. 75 ka Middle Stone Age levels at BBC, South Africa: implications for the origin of symbolism . Antiquity 75, 309–18. d’Errico, F., Henshilwood, C.S., Vanhaeren, M, & van Niekerk, K.L. (2005). Nassarius kraussianus shell beads from Blombos Cave: evidence for symbolic behaviour in the Middle Stone Age. Journal of Human Evolution 48, 3–24. Deacon, H.J. (1998). Modern Human emergence: an African archaeological perspective. Dual Congress Proceedings, Colloquim 17, The Archaeology Of Modern Human Origins. Sun City, South Africa. Donald, M. (1991). Origins of the Modern Mind. Cambridge, Mass.: Harvard University Press.

456

Strategic Integrity of the Middle Stone Age Levels at Blombos Cave Evans, U. (1994). Hollow rock shelter: a Middle Stone Age site in the Cederberg. Southern African Field Archaeology 3: 63–73. Henshilwood, C.S. (1996). A revised chronology for the arrival of pastoralism in southernmost Africa: new evidence of sheep at ca. 2000 b. p. from BBC, South Africa. Antiquity 70, 945– 949. Henshilwood, C.S. & Marean, C.W. (2003). The origin of modern human behaviour: a review and critique of models and test implications. Current Anthropology 44(5), 627–651. Henshilwood, C.S. & Sealy, J.C. (1997). Bone artefacts from the Middle Stone Age at Blombos Cave, southern Cape, South Africa. Current Anthropology 38(5), 890–895. Henshilwood, C.S. (1995). Holocene archaeology of the coastal Garcia State Forest, Southern Cape, South Africa. Ph.D. thesis. University of Cambridge. Henshilwood, C.S., Sealy, J.C., Yates, R.J., Cruz-Uribe, K., Goldberg, P., Grine, F.E., Klein, R.G., Poggenpoel, C., van Niekerk, K.L. &, Watts, I. (2001a). Blombos Cave, southern Cape, South Africa: Preliminary report on the 1992 – 1999 excavations of the Middle Stone Age levels. Journal of Archaeological Science 28(5), 421–448. Henshilwood, C.S., d’Errico, F.E., Marean, C.W., Milo, R.G. & Yates, R. (2001b). An early bone tool industry from the Middle Stone Age at Blombos Cave, South Africa: implications for the origins of modern human behaviour, symbolism and language. Journal of Human Evolution 41, 631–678. Henshilwood, C.S., d’Errico, F., Yates, R., Jacobs, Z., Tribolo, C., Duller, G.A.T., Mercier N., Sealy, J.C., Valladas, H., Watts, I. & Wintle, A.G. (2002). Emergence of modern human behaviour: Middle Stone Age engravings from South Africa. Science 295, 1278-1280. Henshilwood, C.S., d’Errico, F., Vanhaeren, M., van Niekerk, K. & Jacobs, Z. 2004. Middle Stone Age shell beads from South Africa. Science, 384: 404. Hovers, E., Ilani, S., Bar-Yosef, O. & Vandermeersch, B. (2003). An early case of color symbolism: ochre use by modern humans in Qafzeh cave. Current Anthropology 44 (August–October), 491–522. Jacobs, Z., Wintle, A.G. & Duller, G.A.T. (2003a). Optical dating of dune sand from Blombos Cave, South Africa: I – multiple grain data. Journal of Human Evolution 44, 599–612. Jacobs, Z., Duller, G.A.T. & Wintle, A.G. (2003b). Optical dating of dune sand from Blombos Cave, South Africa: II – single grain data. Journal of Human Evolution 44, 613 – 625. Jolly, K. (1947). Preliminary note on new excavations at Skildergat, Fish Hoek. South African Archaeological Bulletin 2: 11–12. Jolly, K. (1948). The development of the Cape Middle Stone Age in the Skildergat Cave, Fish Hoek. South African Archeological Bulletin 3:106–107 Klein, R.G. (1999). The Human Career, 2nd Ed. Chicago: Chicago University Press. Klein, R.G. (2000). Archeology and the evolution of human behavior. Evolutionary Anthropology 9: 17–36.

457

From Tools to Symbols Malan, B.D. (1955). The archaeology of the Tunnel Cave and Skildergat Kop, Fish Hoek, Cape of Good Hope. South African Archaeological Bulletin 10: 3–9. McBrearty, S. & Brooks, A. (2000). The revolution that wasn’t: a new interpretation of the origin of modern human behaviour. Journal of Human Evolution 38, 453–563. Miller, D.E., Yates, R.J., Parkington, J.E. & Vogel, J.C. (1993). Radiocarbon-dated evidence relating to a mid-Holocene relative high sea-level on the south-western Cape coast, South Africa. South African Journal of Science 89: 35–44. Nilssen, P.J. & Marean, C.W. (2002). Background and results from test excavations of Middle Stone Age sites at Pinnacle Point, Mossel Bay. QUARC Newsletter 10, 1–2. Peers, B. (1929). Preliminary report on the archaeology of the Fish Hoek – Noord Hoek valley. Unpublished Manuscript, South African Museum. Schirmer, G.R. (1975). An analysis of lithic material from Dale Rose Parlour, Trappies Kop, Kalk Bay, Cape Peninsula. Unpublished Archaeology Additional Report, University of Cape Town. Taborin, Y. (2003). La mer et les premiers hommes modernes. In (B. Vandermeersch, Ed.) Echanges et diffusion dans la préhistoire méditerranéenne, pp. 113–122. Paris: CTHS. Thompson, J.C., Bower, J.R.F., Fisher, E.C., Mabulla, A.Z.P., Marean, C.W., Stewart, K. & Vondra, C.F. (2004). Loiyangalani: Behavioral and Taphonomic Aspects of a Middle Stone Age site in the Serengeti Plain, Tanzania. Conference Proceedings of the Paleoanthroplogy Society, 30–31 March, Montreal, Canada. Watts, I. (1999). The origin of symbolic culture. In (R. Dunbar, C. Knight & C. Power, Ed.), pp. 113–146. Edinburgh: Edinburgh University Press. White, T.D., Asfaw, B., Degusta, D., Gilbert, H., Richards, G.D., Suwa, G. & Clark Howell, F. (2003). Pleistocene Homo Sapiens from Middle Awash, Ethiopia. Nature 423, 742–747. Wiessner, P. (1999). Indoctrinability and the evolution of socially defined kinship. In (I. EiblEibesfeldt & F. K. Salter, Eds) Indoctrinability, Ideology, and Warfare: Evolutionary Perspectives, pp. 133–150. Providence: Berghahn Books. Wurz, S. (2000). The Middle Stone Age At Klasies River, South Africa. Ph D Thesis, University Of Stellenbosch. Wurz, S., le Roux, N.J., Gardner, S. & Deacon, H.J. (2003). Discriminating between the end products of the Earlier Middle Stone Age sub-stages at Klasies River using biplot methodology. Journal of Archaeological Science 30: 1107–1126. Yellen, J.E., Brooks, A.S., Cornelissen, E., Mehlman, M.J., & Stewart, K. (1995). A Middle Stone Age worked bone industry from Katanda, Upper Semliki Valley, Zaire. Science 268, 553– 556.

458

Testing and demonstrating the stratigraphic integrity of artefacts from MSA deposits at Blombos Cave, South Africa Zenobia Jacobs Council for Science and Industrial Research (CSIR), P.O. Box 395, Pretoria, 0001, South Africa

Abstract Excavations at Blombos Cave have produced early evidence for advanced cognitive behaviour associated with the Middle Stone Age (MSA) in South Africa. An important question posed at this site was whether these significant artefacts may perhaps represent intrusions from the younger Later Stone Age (LSA) levels. Here it is demonstrated how optically stimulated luminescence (OSL) measurements of individual quartz grains can provide experimental data from which mixing between adjacent sedimentary layers can be assessed. By association, the likelihood of movement of artefacts through these sedimentary layers can be inferred. The stratigraphic sequence at Blombos Cave is ideal for such a study, since a thick archaeologically sterile aeolian dune layer separates the c. 2 Ka old LSA from the c. 70 Ka old MSA occupation layers. Based on the known age estimates from the LSA and MSA, an artificial sediment mixture was produced and measured in the laboratory to simulate what measured De values for grains from the dune layer would look like if mixing of LSA and MSA sediments had occurred. This mixture is then used as a guideline against which the measured distribution of De values from the natural sediment sample obtained from the dune layer itself can be compared. OSL measurements on a large number of quartz grains from the dune layer found no evidence for mixing and confirmed the stratigraphic integrity of the MSA artefacts.

Résumé La fouille de la grotte de Blombos, en Afrique du Sud, a permis la découverte, dans des couches du Middle Stone Age, de preuves particulièrement anciennes de comportements cognitifs modernes. Une question cruciale concernant ce site est de savoir si les objets permettant de proposer l’hypothèse d’une origine précoce des comportements modernes pourraient en fait représenter des intrusions des niveaux plus récents, datés du Later

459

From Tools to Symbols Stone Age (LSA). Nous montrons dans cette contribution de quelle façon les mesures faites par luminescence stimulée optiquement (OSL) sur des grains de quartz singuliers peuvent fournir des données expérimentales permettant d’évaluer le degré de mélange entre couches sédimentaires adjacentes. Par analogie, nous pouvons déduire la probabilité que des objets se soient déplacés à travers les couches archéologiques. La séquence stratigraphique de la grotte de Blombos est particulièrement adaptée à une telle étude puisqu’une épaisse couche de sable éolienne, stérile du point de vue archéologique, sépare la couche du Later Stone Age (LSA), datée de 2 Ka, de la couche MSA, déposée il y a environ 70 Ka. En se basant sur l’âge connu du LSA et du MSA de ce site, un mélange artificiel de sédiments a été produit puis mesuré en laboratoire pour simuler ce à quoi ressembleraient les valeurs De des grains contenus dans la couche dunaire, si un mélange de sédiments LSA et MSA s’était produit. Ces valeurs ont été ensuite comparées à ceux mesurés sur le sédiment qui compose la couche dunaire. Des mesures OSL faites sur une grande quantité de grains de quartz provenant de cette couche n’a relevé aucun signe de mélange et a permis de confirmer l’intégrité stratigraphique des artefacts du MSA.

Introduction Blombos Cave is situated on the southern Cape coast of South Africa and is one of many cave sites that contain evidence of MSA occupation (see Henshilwood et al., 2001). The MSA deposits at Blombos Cave hold large numbers of artefacts and much has been written about the symbolic significance of the deliberately engraved ochres (Henshilwood et al., 2002a), the regular production and use of formal bone tools (Henshilwood et al., 20012b, the finely made Still Bay bifacial points, the many shell beads (Henshilwood et al., 2004) as well as the MSA peoples’ broad subsistence base and likely ability to fish (Henshilwood et al., 2001). These strands of evidence have been used to suggest levels of cognitively modern behaviour that are conventionally not associated with MSA people (e.g. Klein, 2000, 2001). One of the main issues surrounding the significance of the Blombos Cave MSA artefacts, and in particular the significance of the formal bone tools, shell beads and engraved ochres, is whether any of these artefacts excavated from the MSA layers could have intruded from the c. 2 000 year old Later Stone Age (LSA) occupation layers (Henshilwood et al., 2001). Klein (2000) noted that ‘art objects that antedate 50 ky ago are probably younger intrusions’. It is of utmost importance to challenge such statements with concrete and experimental evidence in order for the artefacts from Blombos Cave to participate in the origins of modern human behaviour debate, without their context being questioned. Optically stimulated luminescence (OSL) measurements on individual quartz grains can provide a means of investigating and confirming the stratigraphic integrity of artefacts belonging to the MSA in Blombos Cave. The aim of this paper is to demonstrate how this can be achieved and to test the method using both a laboratory-

460

The stratigraphic integrity of artefacts from Blombos Cave

controlled dataset and data for grains from dune sand both inside and outside of Blombos Cave.

Blombos Cave deposits In an archaeological site there is always the potential for admixture that is beyond the control of the excavator. At Blombos Cave, in particular, the MSA deposits are complex because the deposits rise and fall from the back to the front of the cave due to subsidence that produces a ‘wrapping effect’ over the rockfalls (Henshilwood et al., 2001) (Fig. 1). Potential causes of admixture may include not only geological processes such as the observed slumping and faulting, but also animal burrowing or even anthropogenic practices such as digging or the construction of sleeping hollows by LSA people. It is conceivable that any of these factors may cause displacement of LSA artefacts into the older MSA occupation layers. Also, because of the difficulty of tracing these layers, excavators may cross-cut through layers and cause artificial mixing. During the 1992 excavation at Blombos Cave, charcoal and shell samples were recovered from excavated squares near the rear of the cave and were submitted for 14C dating in 1993. Four of the charcoal samples gave ages of c. 1.5–2.1 Ka and a single charcoal piece and five marine shell opercula (Turbo sarmaticus) samples gave age estimates of c. 32–39 Ka. These 14 C ages suggested admixture of LSA and MSA materials (Henshilwood et al., 2001). It was only during the 1998 excavation season that the reason for this admixture was identified. A 5–10 cm wide gap that had formed between two large basal roof blocks was recognised near the back wall of the cave; into this gap LSA derived sediments and materials most probably had slumped. Although the reasons for the ambiguous 14 C ages had been identified, it was still important to test and demonstrate that none of the other MSA artefacts that were recorded from other locations inside the cave could have intruded from the LSA.

Optically stimulated luminescence (OSL) measurements Possible intrusion of LSA material into the MSA occupation layers at Blombos Cave can be investigated and tested through the use of OSL measurements on single quartz grains. A quartz grain has its own internal energy ‘clock’. The energy is deposited in the quartz crystal as a result of ionising radiation (e.g. alpha, beta and gamma radiation) from its surrounding environment as well as a small contribution from cosmic rays (e.g. Aitken, 1998). The energy is stored as electrons trapped at defects in the crystal structure of the quartz grains. In nature each individual quartz grain undergoes repeated cycles of erosion, transport and deposition (Fig. 2). If transport involves exposure of the quartz grain to

461

From Tools to Symbols

Figure 1 Photograph of the main stratigraphic section at Blombos Cave showing clearly how the LSA and MSA are separated by a thick sterile aeolian dune sand (BBC HIATUS) containing OSL sample ZB15. Also evident is the ‘wrapping’ of the sedimentary layers around the rock fall.

462

The stratigraphic integrity of artefacts from Blombos Cave

daylight (e.g. aeolian transport), the energy stored is released and the luminescence ‘clock’ is zeroed. After the quartz grains are deposited, they will be covered by more grains; thus the zeroed grains are hidden from sunlight and the luminescence ‘clock’ will start ticking as the energy stored increases through time. The rate of increase is determined by the amount of ionising radiation to which the quartz grain is exposed. To obtain the depositional age of the buried sediments, a sediment sample is collected in the dark; this prevents exposure of the sample to light, a process that will lead to zeroing of the OSL signal. Once in the laboratory, the trapped electrons can be released by stimulating the quartz grains with a light source. This release causes luminescence that results in a characteristic OSL decay curve under conditions of constant stimulation (Fig. 3). This measured natural luminescence is proportional to the number of trapped electrons that have accumulated since the previous time the traps were emptied and therefore is proportional to the last time the grains were exposed to sunlight. The equivalent dose (De) is the amount of radiation required in the laboratory to produce a luminescence signal that matches the natural luminescence level. The De is determined by comparing the natural OSL signal with those OSL signals that result from a range of artificial irradiations administered in the laboratory using a 90Sr/90Y beta source. The resulting plots are known as dose response curves (Fig. 4); they show the relationship between the natural luminescence measured and the dose administered. Through projection of the natural signal onto this dose response curve, De can be estimated.

Figure 2 Schematic display of the process of energy build-up due to ionising radiation from the surrounding environment, signal zeroing by daylight representative of the last depositional event and growth of signal over a time scale of thousands of years.

463

From Tools to Symbols

Figure 3 An OSL decay curve for a moderately bright grain, indicating the decay of the signal as a result of exposure to the focused laser of the Risø single grain instrument.

Figure 4 An example of a dose response curve constructed using a range of laboratory regeneration doses, ranging between 0 and 80 Gy. Also indicated is the Natural measurement (LN/TN) and the equivalent dose De that can be calculated by projection onto the dose response curve.

464

The stratigraphic integrity of artefacts from Blombos Cave

De distributions and radial plots It is these values of De that form the basis of the investigation into possible admixture of a sedimentary layer. Developments in four different areas of luminescence dating have led to improvements in the ability to use De values to make this possible. This includes (1) the development of a measurement procedure to obtain an accurate value of De (Murray & Wintle, 2000), (2) the technological ability to measure many hundreds or thousands of single grains in a relatively routine manner (Duller et al., 1999), (3) the use of a graphical display of De values (Galbraith, 1990) from which one can visually assess the possibility of mixing or the lack thereof, and (4) rigorous statistical procedures to determine the presence or absence of more than one population of grains (Galbraith & Laslett, 1993). The development of the single aliquot regenerative-dose (SAR) measurement procedure (Murray & Wintle, 2000) was important because it provides an accurate means of determining the burial dose (De) received by the sample; its advantage over previous procedures is that it compensates for changes in luminescence sensitivity that occur as a result of laboratory procedures. The procedure also contains some internal tests that can be used to demonstrate the reliability of the derived De values (e.g. Jacobs et al., 2003a, b). Although named a single aliquot procedure, this same measurement procedure is applied to the measurement of individual quartz grains. However, making single grain measurements using conventional luminescence equipment is extremely time-consuming and laborious and this has resulted in the development of a purposebuilt single grain luminescence reader. This instrument uses a laser beam that is focused to a spot less than 30 μm in diameter (Duller et al., 1999; Bøtter-Jensen et al., 2000). One hundred single grains can be mounted in a specially manufactured sample holder (Fig. 5) and in any one measurement sequence up to 48 such holders can be measured, giving a total capacity of 4,800 grains. The development of this instrument results in a far more efficient throughput of samples and the quicker acquisition of a statistically representative number of De values, necessary to assess the potential of admixture in a sedimentary sample. The luminescence properties of individual grains can vary significantly, as was demonstrated by Adamiec (2000), Duller et al. (2000) and Jacobs et al. (2003b), with at least three orders of magnitude difference in the brightness of grains from the same sample. These differences in luminescence properties lead to large differences in the precision with which De for each individual quartz grain can be measured. Using a histogram to display such a data set loses the information on precision, and radial plots (Galbraith, 1990) are now accepted as the most appropriate graphical display method for multiple De values. On such plots (Fig. 6), each grain is shown as a separate point. The more precisely known data are plotted to the right, and the less well known to the

465

From Tools to Symbols

Figure 5 The single grain sample holder is 1 cm in diameter and has a 10 by 10 array of holes precision drilled into the surface of the disc. Each hole is 300 μm in diameter and can hold a 200 μm grain.

left. All the points lying on a straight line drawn from the origin to the radial scale on the right have the same De value (expressed in the SI unit of absorbed dose, Grays). A mathematical consequence of such a graph is that if all the results are consistent with a mean value within two standard deviations, then all the points will lie within the two broken lines originating from the +2 and –2 (two standard deviations) indicated on the y-axis (Galbraith, 1990). Assuming that all grains had the same burial history and that no mixing between different sedimentary units with different burial histories has occurred, one would expect that 96 per cent of the grains will fall within two standard deviations from the mean value. However, if post-depositional mixing occurred, grains of different ages will become mixed, with either younger or older material introduced. Mixing younger material into an older deposit will lead to a radial plot where two discrete populations of De values are observed, one centred on a line representative of the younger material and the other relating to the older grains (Fig. 6c). Such behaviour would suggest that the stratigraphic unit being measured consists of a mixture of grains with different depositional ages.

Importance of the dune layer in Blombos The stratigraphic sequence at Blombos Cave (Fig. 1) provides an ideal situation for a single grain luminescence study from which the stratigraphic integrity of the MSA layers can be tested. A large Quaternary sand dune, assigned to the Waenhuiskrans Formation (Malan et al., 1994), formed against the coastal cliff and the remains thereof

466

The stratigraphic integrity of artefacts from Blombos Cave

Figure 6 Single grain De values plotted on radial plots: (a) about 300 grains given a 4 Gy beta dose in the laboratory to simulate grains from the LSA occupations level, (b) c. 300 grains given a 49 Gy beta dose in the laboratory to simulate grains from the dune layer (BBC HIATUS); and (c) a composite radial plot obtained by combining the two data sets presented in (a) and (b) to simulate a mixture of grains from the LSA (filled circles) and dune layer (closed circles).

467

From Tools to Symbols

can be found at Blombos Cave. Sand from this dune is also preserved inside the cave, where it forms a thick (5–60 cm) and continuous archaeologically sterile layer that covers more than 90 per cent of the excavated MSA deposits. This aeolian sand clearly separates the younger LSA from the older underlying MSA occupation layers (Fig. 1). The LSA occupation layers were 14C dated to < 2 Ka (Henshilwood et al., 2001) and the dune layer was dated via OSL dating of single quartz grains to 67,3 ± 3,8 Ka (Jacobs et al., 2003b). The dune layer thus provides a reliable minimum age for the underlying MSA occupation layers. If any of the artefacts that were excavated from the MSA were intrusions from the more recent LSA, these artefacts would have had to travel through this aeolian sand layer. Such an occurrence would have resulted in the mixing in of younger grains from the LSA with the grains from the much older dune layer. Therefore in the absence of a mixture of sand grains representing sand grains with an LSA age and sand grains of an MSA age, it is not possible to suggest intrusions of artefacts present in the MSA from the LSA. On the contrary, where the dune layer represents a mixture, two clear and discrete De populations will be visible on the radial plot. The stratigraphic position of the dune layer makes it, therefore, a key unit. Its aeolian origin renders it a good candidate for optical dating because it is with this type of deposit that there is the greatest confidence in the sufficiency of depositional zeroing of the OSL signal, which will result in grains giving the same De (Aitken, 1998; Bray & Stokes, 2003). The large time gap between the LSA units and the dune layer, in addition to the similarly low environmental dose rates for the two units, means that grains from these two units should exhibit very large differences in their measured Des, because the De is proportional to the time that the sand grains have been buried. This increases the potential of differentiating between grains derived from the LSA and the dune layer if the deposit is mixed, even at a level where their absolute precisions are low.

Laboratory-controlled data set – an artificial mixture To simulate this situation where LSA and MSA grains are mixed, two laboratorycontrolled datasets were measured. These laboratory-controlled experiments involve zeroing the OSL signals of all grains under known and controlled conditions (532 nm laser light at 50W/cm2) followed by application of an identical beta dose for each individual grain using a calibrated 90Sr/90Y irradiation source (e.g. Roberts et al., 2000). Seven thousand grains were measured, of which half were given a 4 Gy beta dose and the other half a 49 Gy beta dose. A beta dose of 4 Gy, when taking into consideration the measured environmental dose rate, is equivalent to that from a sample c. 2 000 years old. A beta dose of 49 Gy was what was measured in the OSL dating study (see Jacobs et al., 2003a, b) to obtain a luminescence age of c. 67 Ka for the dune layer

468

The stratigraphic integrity of artefacts from Blombos Cave

inside Blombos Cave. The applied beta doses therefore simulate the buried dose, or De value, for the LSA layer and dune layer from Blombos Cave. Although 7 000 individual quartz grains were measured, not all of the grains are usable. Jacobs et al. (2003b) and Jacobs (2004) have demonstrated that methodological and experimental factors, especially the inclusion of non-quartz grains, can produce erroneous results. If these grains are not accounted for, their presence in a natural sample can be erroneously used to suggest mixing. As a necessary first step towards the reliable use of single grain OSL measurements as a tool to detect mixing in archaeological deposits, one needs to ensure that all experimental and methodological factors have been eliminated. Because of the large variability that can be observed on a grain-to-grain basis (e.g. Adamiec, 2000; Roberts et al., 1999; Duller & Murray, 2000; Duller et al., 2000; Jacobs et al., 2003b; Jacobs, 2004), these factors will be grain-dependent and require the application of a number of rejection criteria to each individual grain measurement. These rejection criteria are explained in Jacobs et al. (2003b) and Jacobs (2004) and are based upon recycling ratio characteristics, reduction in the OSL signal on exposure to infrared (indicating that the grains were not composed solely of quartz) and natural signals that did not intersect the dose response curve, as well as grains for which there is not a significant OSL signal. These rejection criteria do not reject De values that are indicative of mixing. After application of the rejection criteria, only c. 300 grains were usable in roughly equal proportions between the grains that received the two different beta doses. The measured De values are displayed as radial plots in Fig. 6a and b. The dose estimates for the two De populations representative of the LSA (4 Gy) and the dune layer (49 Gy) at Blombos Cave are in good agreement with the applied beta doses. The fact that we can measure an applied beta dose accurately in the laboratory confirms the applicability of the single aliquot regenerative (SAR) dose dating protocol (Murray & Wintle, 2000) that we routinely use to obtain an estimate of the De. Also, these two data sets indicate what it would look like if all the grains from a sedimentary layer had the same De values and therefore contained no evidence of mixing between adjacent stratigraphic layers. To simulate what a mixed dune layer would look like if there was large-scale mixing between the LSA and dune layer, a composite radial plot (Fig. 6c) has been produced by combining the two data sets presented in Fig. 6a and b. This radial plot in effect illustrates the expected dose distribution of De values from a sample that contains a mixture consisting of equal numbers of grains from the LSA and the dune layer in Blombos Cave. This graph provides a visual guideline to the expected pattern and dispersion if the dune layer is a mixed deposit. The measured De distributions from the Blombos dune layer sample can therefore be compared against this composite graph to assess the likelihood of mixing between the LSA and MSA at Blombos Cave.

469

From Tools to Symbols

This artificial mixture also indicates the necessity for measuring large numbers of grains so that there are at least some grains from which high precision estimates can be obtained. Where precision is good, discrimination between grains with different burial histories is optimised because of the radial scale. However, because of the large difference in the De values, precision of only 20 per cent (or better) is required to be able to use the single grain De values to assess possible admixture.

Comparing the laboratory-based mixture with the dune from Blombos Cave To assess whether the dune layer inside Blombos Cave consists of a mixture of LSA and MSA grains, 1 892 (ZB15) individual grains were measured using the SAR measurement procedure. Since the sterile dune layer found within the cave can also be traced outside the cave entrance, two additional dune samples (ZB13 and ZB20) were taken from outside the cave. A comparison of the dune sample from inside the cave with those two samples obtained from outside the cave would provide additional information on the degree of post-depositional mixing of the dune layer inside the cave; the two samples taken from outside the cave cannot be contaminated by grains from the LSA levels. Mixing of the dune layer inside the cave can therefore be assessed by investigating the similarity or differences in De distributions presented as radial plots between the three samples. For the two samples that were taken from outside the cave, 4 332 (ZB13) and 2 737 (ZB20) grains were measured. Experimental details are given in Jacobs et al. (2003b). The De distributions for the three dune samples are presented in Figure 7a–c, and exclude the grains that were rejected after they underwent the stringent rejection criteria explained and tested in Jacobs et al. (2003b). Only 34, 164 and 120 grains remained for each sample, respectively. The results for the three dune samples (Fig. 7a–c) can now be compared to those for the artificial mixture presented in Figure 6c. The radial plot representative of the dune layer inside the cave (ZB15) does not look at all similar to that of the artificial mixture that simulates large-scale mixing between the LSA and MSA. However, there is more over-dispersion in the De values than what one would expect. Instead of 96 per cent of the De values falling within two standard deviations of a central value (Galbraith et al., 1999), only c. 87 per cent of the values fall within the 2σ band. The grains that fall outside are shown as open circles in Figure 7a. The same degree of over-dispersion is also present in the two dune samples from outside the cave (Fig. 7b and c), emphasising the similarity of the De distributions where mixing with LSA is not possible. This suggests that the over-dispersion is not an indication of mixing between grains of LSA and MSA age, but rather small-scale differences in the dose received by individual grains while buried because of inhomogeneities in the surrounding environment that will lead to inhomogeneous beta irradiation in nature (Nathan et al., 2003).

470

The stratigraphic integrity of artefacts from Blombos Cave

Figure 7 Single grain De values plotted on radial plots: (a) grains from the dune layer inside Blombos Cave that separates the LSA from the MSA occupation layers, (b) grains of the cemented dune sand (ZB13) outside Blombos Cave, up against the cliff at the same height as the cave opening; and (c) grains from cemented dune sand at sea level (ZB20) in line with the cave opening.

471

From Tools to Symbols

The similarity of the dose distributions for samples from inside and outside the cave and the fact that none of the grains from the dune layer inside the cave had a De value as low as 4 Gy suggests that the dune layer inside the cave is not contaminated by the younger LSA material. The fact that no younger LSA grains could be detected in the dune layer suggests that the dune layer is not mixed. The most important implication is that it confirms the stratigraphic integrity of the underlying MSA deposits and the antiquity of the artefacts excavated from the MSA levels.

Conclusions OSL measurements on individual quartz grains were used to assess the possibility of LSA intrusions into the MSA occupation layers at Blombos Cave. Recent developments in the measurements procedures, technology and graphical display methods have increased our ability to use OSL measurements to assess the possibility of admixture. At Blombos Cave, the LSA and MSA occupation layers are clearly separated by a thick and continuous archaeologically sterile aeolian sand layer. The large difference in age between the dune layer (and by implication the MSA deposits) and the LSA provides an optimum situation in which possible mixing can be assessed using OSL measurements on individual quartz grains. It was demonstrated, using artificially irradiated grains to produce a laboratory-controlled data set, how De values for a mixed dune layer would appear, if large-scale mixing had occurred, as one would expect if the artefacts excavated from the MSA deposits had travelled through this layer. These results were then compared to the measurements on a natural sedimentary sample from the dune layer inside the cave, as well as from two natural samples of the equivalent dune sand outside the cave. The results suggested that there was no observable mixing between the LSA and the dune layer in Blombos Cave and that the radial plots present De distributions indicative of grains with the same burial history. The ability to measure De values from single mineral grains has thus provided the opportunity to investigate the distribution of grains with different doses within a sample. For samples where post-depositional mixing is conceivable, this approach is invaluable in demonstrating whether mixing has occurred – or, as in the case of Blombos Cave, demonstrating that is has not.

Acknowledgements The author wishes to thank Professor Ann G. Wintle and Dr Geoff Duller for their invaluable contribution to the OSL study performed at Blombos Cave. Thanks are also extended to Professor Chris Henshilwood for providing access to Blombos Cave. A maintenance grant from the Sir Henry Strakosch Memorial Trust, an Overseas Research Studentship award and tuition fees from the Institute of Geography and Earth Sciences are gratefully acknowledged.

472

The stratigraphic integrity of artefacts from Blombos Cave

References Adamiec, G. (2000). Variations in luminescence properties of single quartz grains and their consequences for equivalent dose estimation. Radiation Measurements 32, 427–432. Aitken, M. J. (1998). An introduction to optical dating. Oxford: Oxford University Press. Bøtter-Jensen, L., Bulur, E., Duller, G.A.T. & Murray, A.S. (2000). Advances in luminescence instrument systems. Radiation Measurements 32, 523–528. Bray, H.E. & Stokes, S. (2003). Chronologies for Late Quaternary barchan dune reactivation in the southeastern Arabian Peninsula. Quaternary Science Reviews 22, 1027–1033. Duller, G.A.T., Bøtter-Jensen, L., Kohsiek, P. & Murray, A.S. (1999). A high-sensitivity optically stimulated luminescence scanning system for measurement of single sand-sized grains. Radiation Protection Dosimetry 84, 325–330. Duller, G.A.T., Bøtter-Jensen, L. & Murray, A.S. (2000). Optical dating of single sand-sized grains of quartz: sources of variability. Radiation Measurements 32, 453–457. Duller, G.A.T. & Murray, A.S. (2000). Luminescence dating of sediments using individual mineral grains. Geologos 5, 88–106. Galbraith, R.F., Roberts, R.G., Laslett, G.M., Yoshida, H. & Olley, J.M. (1999). Optical dating of single and multiple grains of quartz from Jinmium rock shelter, northern Australia: Part I, Experimental design and statistical models. Archaeometry 41, 339–364. Galbraith, R.F. (1990). The radial plot: graphical assessment of spread in ages. Nuclear Tracks and Radiation Measurements 17, 207–214. Galbraith, R.F. & Laslett, G.M. (1993). Statistical models for mixed fission track ages. Nuclear Tracks and Radiation Measurements 21, 459–470. Henshilwood, C.S., d’Errico, F., Yates, R., Jacobs, Z., Tribolo, C., Duller, G.A.T., Mercier, N., Sealy, J.C., Valladas, H., Watts, I. & Wintle, A.G. (2002a). Emergence of modern human behaviour: Middle Stone Age engravings from South Africa. Science 295, 1278–1280. Henshilwood, C.S., d’Errico, F., Marean, C.W., Milo, R.G. & Yates, R. (2002b). An early bone tool industry from the Middle Stone Age at Blombos Cave, South Africa: implications for the origins of modern human behaviour, symbolism and language. Journal of Human Evolution 41, 631–678. Henshilwood, C.S., D’Errico, F., Vanhaeren, M., van Niekerk, K. & Jacobs, Z. (2004). Middle Stone Age shell beads from South Africa. Science 304, 404. Henshilwood, C.S., Sealy, J.C., Yates, R., Cruz-Uribe, K., Goldberg, P., Klein, R.G., van Niekerk, K. & Watts, I. (2001). Blombos Cave, Southern Cape, South Africa: preliminary report on the 1992–1999 excavations of the Middle Stone Age levels. Journal of Archaeological Science 28, 421–448. Jacobs, Z. (2004). Development of luminescence techniques for dating Middle Stone Age sites in South Africa. Unpublished Ph.D. thesis, University of Wales, Aberystwyth.

473

From Tools to Symbols Jacobs, Z., Wintle, A.G. & Duller, G.A.T. (2003a). Optical dating of dune sand from Blombos Cave, South Africa: I – multiple grain data. Journal of Human Evolution 44, 599–612. Jacobs, Z., Duller, G.A.T. & Wintle, A.G. (2003b). Optical dating of dune sand from Blombos Cave, South Africa: II – single grain data. Journal of Human Evolution 44, 613–625. Klein, R.G. (2000). Archaeology and the evolution of human behaviour. Evolutionary Anthropology 9, 17–36. Klein, R.G. (2001). Southern Africa and modern human origins. Journal of Anthropological Research 57, 1–16. Malan, J.A., Viljoen, J.H.A., Siegfried, H.P. & Wickens, H de V. (1994). Die geologie van die gebied Riversdal. Pretoria: Council for Geoscience. Murray, A.S. & Wintle, A.G. (2000). Luminescence dating of quartz using an improved singlealiquot regenerative-dose protocol. Radiation Measurements 32, 57–73. Nathan R.P., Thomas P.J., Jain, M., Murray A.S. & Rhodes E.J. (2003). Environmental dose rate heterogeneity of beta radiation and its implications for luminescence dating: Monte Carlo modelling and experimental validation. Radiation Measurements 37, 305–313. Roberts, R.G., Galbraith, R.F., Yoshida, H., Laslett, G.M. & Olley, J.M. (2000). Distinguishing dose populations in sediment mixtures: a test of single-grain optical dating procedures using mixtures of laboratory-dosed quartz. Radiation Measurements 32, 459–465. Roberts, R.G., Galbraith, R.F., Olley, J.M., Yoshida, H. & Laslett, G.M. (1999). Optical dating of single and multiple grains of quartz from Jinmium rock shelter, northern Australia: Part II, Results and implications. Archaeometry 41, 365–395.

474

From tool to symbol: the behavioural context of intentionally marked ostrich eggshell from Diepkloof, Western Cape John Parkington, Cedric Poggenpoel, Jean-Philippe Rigaud and Pierre-Jean Texier Department of Archaeology, University of Cape Town, Private Bag, Rondebosch 7700, South Africa UMR 5808 du CNRS, Institut de Préhistoire et de Géologie du Quaternaire, Avenue des Facultés, 33405, Talence, France EP 2058 Préhistoire et Technologie, 250 rue Albert Einstein, Sophia-Antipolis, 06560 Valbonne, France

Abstract Diepkloof is a large rock shelter overlooking the lower reaches of the Verlorenvlei in the Western Cape Province of South Africa. Earlier excavations have shown that a series of Middle Stone Age (MSA) assemblages underlie a shallow Later Stone Age (LSA) occupation of the shelter. Current excavations under the direction of the authors have resolved the cultural stratigraphy and generated valuable associations between stone tool sets, faunal, plant and wood charcoal remains and a large series of intentionally marked ostrich eggshell fragments, including at least one demonstrable water flask mouth. Stone tool assemblages of characteristically Howieson’s Poort forms are overlain by MSA assemblages with unifacial points and quite different raw materials and underlain by MSA assemblages that are as yet hard to classify. The intentionally marked ostrich eggshell fragments are found in the upper part of the Howieson’s Poort series.

Résumé Diepkloof est un grand abri qui domine l’estuaire du Verlorenvlei, situé dans la Province du Western Cape, en Afrique du Sud. Les premières fouilles du site ont mis en évidence une séquence comprenant une fine couche du

475

From Tools to Symbols Later Stone Age (LSA) sus-jacent plusieurs couches contenant des assemblages du Middle Stone Age (MSA). Les fouilles en cours, menées sous la direction des auteurs, ont précisé la stratigraphie du site et étudié la relation entre les industries lithiques, les restes fauniques, les fragments de bois et de charbon d’un côté et une série de fragments d’œufs d’autruche gravés de l’autre. Ces analyses révèlent la présence d’une industrie Howieson’s Poort sous-jacente à un MSA à pointes unifaciales et utilisant des matières premières différentes. La couche inférieure contient une industrie MSA encore à définir. Les fragments d’œuf d’autruche gravés proviennent du sommet de la couche Howieson’s Poort.

Introduction Diepkloof is one of two very high overhangs, or abris, that overlook the Verlorenvlei River about 18 km upstream from its mouth in the southern corner of Elands Bay (Fig. 1). One of these has an abundance of rock paintings and a shallow (less than a metre) deposit mostly comprising Middle Stone Age materials. It is an interesting site that demands more investigation. The other overhang, on which we concentrate here, has fewer rock paintings but a much deeper set of deposits and has been the focus of several excavations since the early 1970s. We refer to this site as Diepkloof and explain here the stratigraphic and behavioural context of several dozen fragments of, in our view, intentionally marked ostrich eggshell, recovered from our recent excavations. These are more than 55 000 and probably not more than 75 000 years old. Where do they fit along the implied continuum from tool to symbol? More complete analyses of materials and stratigraphy will appear later. The shelter is a massive indentation in the local quartzite, caused by the falling of several very large blocks, some of which still dominate the front of the shelter. Initially the floor must have been very rock-strewn and irregular, but as deposit accumulated some 150 m2 of usable space became available in the lee of these large blocks (Fig. 2). To judge by the level of outcropping bedrock at the front of the shelter, and based on some electronic surveying (Lenoble & Martinaud, 2003), at least some of the deposits are more than 5 metres deep. Facing almost due east, and about 100 metres above the river level, Diepkloof offers an extensive view across the reeds and open water of the Verlorenvlei and the adjacent Sandveld landscape.

Excavation history The earliest excavations took place in 1973 when two of us (JP and CP) removed several cubic metres of the mostly superficial Later Stone Age bedding and ash deposits as part of a regional excavation programme (Parkington, 1977; Parkington & Poggenpoel, 1987). It was clear from the outset that LSA remains were confined to the uppermost 30 cm of deposit, sometimes much less, although occasional pits

476

Intentionally marked ostrich eggshell from Diepkloof

Figure 1 The location of Diepkloof in relation to the south-western Cape.

into lower units were found. We have concluded that at least some of these pits may have been dug by LSA people searching for large, unretouched flakes to re-utilise as adze blanks. A significant result of our excavations has been the substantial number of domestic sheep bones that dominate the LSA faunal assemblage (Klein, 2003, pers comm.), making Diepkloof an important site for the understanding of local herder history. All the LSA material dates to the last 1 800 years, prior to which there appears to have been a massive occupational hiatus. In cleaning out one of the LSA pits in grid squares P10 and P11, we excavated a mixed stone tool assemblage that included some recognisable LSA forms, but also many large backed segments of the kind usually referred to as Howieson’s Poort (Deacon, 1995). This assemblage type is the most recognisable and, arguably, innovative, of all Middle Stone Age assemblages (Wurz, 2002). Although some early radiocarbon dates

477

From Tools to Symbols

Figure 2 Plan of Diepkloof rockshelter showing location and chronology of excavations.

on charcoals from the uppermost Howieson’s Poort levels initially suggested otherwise (Parkington, 1990), we now recognise that the whole of the MSA depositional volume at Diepkloof is earlier than the limits of this form of direct dating. In 1986 we (JP, CP and colleagues Royden Yates and Stephan Woodborne) returned to Diepkloof to resample the uppermost Howieson’s Poort levels and extended our excavations toward the mouth of the shelter. In doing this we encountered stone tool assemblages with fewer segments but associated with some apparently intentionally marked ostrich eggshell fragments. We also cleaned up a small disturbance a few metres away from our main excavation in grid square C5. The surprise here was an assemblage of bone, charcoal and stone tools in every way different from those of P10 and P11. The bone of C5 included many large fragments, as did the charcoal

478

Intentionally marked ostrich eggshell from Diepkloof

assemblage, both in contrast to what we had found in P10, where most bone was fragmentary and most charcoal comminuted. More surprisingly, both the stone raw materials and tool forms of C5 were also quite unlike those of P10 and P11. Although we recognised that spatially variable preservation conditions might account for the differences in organic materials, we could not explain the stone tool contrasts in this way. In C5 we had no segments, an assemblage not nearly as rich in blades as the Howieson’s Poort and evidence of more quartz and quartzite than in P10 and P11. It seemed very unlikely that this represented substantial spatial variability in a single contemporary occupation horizon. More likely, we thought, the C5 assemblages must be either earlier or later than those of P10 and P11, despite the almost horizontal depositional surface. Following these excavations, we initiated a programme of luminescence dating of sediments, which has resulted in estimates of between 60 000 and 70 000 years for the Howieson’s Poort assemblages in the P10 area (Feathers et al., in press). We also realised that the definition of the Howieson’s Poort (Wurz, 2002) and, indeed, the historical rather than typological understanding of local MSA behaviour, depended upon adequately sampling and dating the superimposed stack of assemblages of stone, bone, charcoal and ostrich eggshell that appeared to be unevenly distributed across the Diepkloof shelter. The apparently marked ostrich eggshell fragments, for example, seemed to be associated with stone tool assemblages that might or might not fit an analyst’s definition of Howieson’s Poort. Our current field programme, a joint project of the Universities of Bordeaux and Cape Town and the CNRS in Valbonne, is organised partly around the enigmatic hints thrown up by previous excavations, and partly around the potential of Diepkloof to contribute to the ‘modern human origins’ debate. Among the former are the following questions: what is the stratigraphic relationship between the assemblages of C5 and P10? What kinds of typological and technological changes are reflected in the stone tool assemblages at Diepkloof? What are the stratigraphic relationships between segments and marked fragments of ostrich eggshell? How old are these ostrich eggshell fragments? What can the faunal and charcoal assemblages tell us about the environmental contexts of these technological events? What are the palaeo-environmental implications of the analyses of sediments? In relation to the latter, we were interested in comparing the transition from MSA to LSA in the Cape with that from Middle Palaeolithic to Upper Palaeolithic in Acquitaine. Despite the great distances involved, it is notable that some common aspects of tool technology pervade these transitions, not least those of the prevalence of curved backed knives, the production of blades and the appearance of intentional markings on organic materials. Some of these may be only superficially similar.

479

From Tools to Symbols

Our strategy has been to excavate an exploratory trench northward from C5 towards the more extensive excavation near to P10, whilst opening up a larger area mostly towards the mouth of the shelter from P10. We are now in a position to link, typologically if not physically, the different assemblages confidently into a common chronological order and to use this to develop a preliminary narrative of events through the Diepkloof deposits. This confidence comes largely from the superpositioning of stone tool assemblages in the different excavated squares, but extends to very coherent and valuable patterns in the associations between stone, bone, charcoal and ostrich eggshell, pegged to a series of radiocarbon and luminescence dates. We describe these now as the framework for an understanding of the intentionally marked ostrich eggshell fragments.

Archaeological sequence Clearly the most recent of the MSA deposits are distributed only towards the mouth in the south-eastern parts of the shelter, including the area of C5. These have no segments but are best characterised as having regular, but neither numerous nor ubiquitous, unifacially retouched points (Fig. 3), mostly made from silcrete, some from quartz. These are markedly the most quartz-rich MSA assemblages at this site. There are no marked ostrich eggshell fragments from these layers, and an AMS radiocarbon date of > 55 000 (GifA 102381, Tribolo, this volume) from some remarkably wellpreserved wood in square E6 indicates their approximate age. In the excavated level immediately below the lowermost unifacial point in C5 and C6, we have recovered nineteen marked ostrich eggshell fragments and three segments. These lower stone tool assemblages have about equal proportions of quartz and silcrete. The assemblages from which marked ostrich eggshell fragments come seem to be more extensive than the unifacial point MSA ones above, but also do not extend as far to the north-west as those with segments. The zone of greatest thickness of these deposits seems to be in the north-eastern part of the shelter, toward its mouth. Below the LSA bedding and ash in P10 and P11 lie assemblages that are rich in segments but poor in marked ostrich eggshell. Observations by one of us (P-J T) show that blades with the characteristics of soft hammer technique are common in these Howieson’s Poort assemblages (Fig. 4). Silcrete is the preferred raw material for soft hammer blade removal and segment manufacture (Tables 1 and 2). We note that several of the segments have small burin-like removals along the chord or arc.

480

Intentionally marked ostrich eggshell from Diepkloof

Figure 3 Unifacial point. Scale = 1 cm.

Figure 4 Blades showing platform and bulb characteristics of soft hammer technique.

481

From Tools to Symbols Table 1 Characteristics of the main raw materials flaked from Howieson’s Poort layers of Diepkloof rock shelter Quartz

Quartzite

Silcrete

Hornfels

Local

Local

Neighbouring or remote

Neighbouring or remote?

Secondary

Primary

Secondary

Secondary

Morphology

Pebbles

Beds and angular rocks

Pebbles or blocks

Pebbles

Localisation

Access slopes to the rock shelter

Rock shelter and access slopes to the rock shelter

Unkown

Unkown

Origin / distance Position

Module

< 10 cm

Unlimited

< 10 cm

< 10 cm

Availability

Considerable

Unlimited

Limited

Limited

Suitability for flaking

Mediocre to medium

Mediocre

Good to excellent

Good to excellent

Table 2

Débitage chaînes opératoires (COD) identified within the Howieson’s Poort layers from Diepkloof rock shelter

Preparation Exploitation

Flakes & bladelets Centripetal COD

Flakes Discoid COD

Blades Laminar COD

Unifacial Centripetal exp.

Unifacial Centripetal exp.

Uni- or bifacial Centripetal exp.

Unidirectional exp.

Main raw material

Quartz Silcrete

Silcrete

Quartzite Quartz

Silcrete

Secondary raw material

Hornfels

Quartz

Silcrete

Quartz

Expected products

Thin flakes

Lamellar flake

Ordinary flakes

Blades & bladelets

Chaîne opératoire

First flake Technical flakes

Technical flakes

Technical flakes

Technical flakes

Full débitage core

Full débitage core

Full débitage core

Full débitage cores are missing

Direct percussion with a hammerstone

Direct percussion with a hammerstone

Direct percussion with a hammerstone

Direct percussion with a soft hammer

Phases represented

Technique

482

Intentionally marked ostrich eggshell from Diepkloof

A general pattern of change from early segment-rich assemblages, through those with both (fewer) segments and numerous marked ostrich eggshell fragments, and on to those without either but with unifacial points, is a useful chronological ordering at Diepkloof. Interestingly, faunal and charcoal assemblages show some correlated change along this same ordering. Most of the charcoal assemblages analysed by Caroline Cartwright (1999: pers. comm.) include several afro-montane tree species such as Celtis, Kiggelaria, Podocarpus and Halleria, as do those from the last glacial maximum from Elands Bay Cave (Cartwright & Parkington, 1997; Cowling et al., 1999; Parkington et al., 2000). These species thrive today only in areas with year-round moisture. Two stratigraphically consecutive samples, however, from near the top of the C5 excavations, the most recent MSA-associated charcoal assemblages studied so far by Cartwright, show the thicket pattern that is typical of the terminal Pleistocene levels at EBC. Afro-montane species have now disappeared, perhaps indicating lower moisture availability. Lower down in C5, both with unifacial points and with segments and marked ostrich eggshell, the charcoal assemblages reported (Cartwright, 1999: pers. comm.) are of afro-montane character. The nature of available firewood species seems to have changed near the very top of the MSA stack. This might reflect a shift to drier conditions, though this need not mean a significant change in precipitation. The faunal identifications provided by Richard Klein (Klein & Cruz-Uribe pers. comm.;

Figure 5 Stratigraphic context of Howieson’s Poort assemblages.

483

From Tools to Symbols

Klein, pers. comm.) indicate some changes toward the top of the MSA levels. Specifically, although the numbers of faunal identifications associated with unifacial points in C5 and C6 are much lower than those with segments in P10 and nearby squares, almost all large forms and almost all grazing forms come from this area. All of the Cape horse (Equus capensis), equid (Equus sp.), black wildebeest (Connochaetes gnou), rhinoceros (genus and species unknown), giant buffalo (Pelorovis antiquus) and eland (Taurotragus oryx), and most of the hippopotamus (Hippopotamus amphibius) bones and teeth are associated with unifacial points. The assemblages with a significant segment component (see also Costamagno, 2001) are dominated by small forms such as the dune mole rat (Bathyergus suillus), the dassie (Procavia capensis) and the hare (Lepus sp.). This distinction cannot, of course, be explained by variable preservation circumstances. We are currently interested in whether this signals significant changes in animal resource availability or different components of an overall unchanging subsistence strategy. The variable nature of the depositional matrix seems also to be correlated with shifts in faunal and stone tool assemblages. At the rear of the cave in P10, Howieson’s Poort assemblages are located in a complex set of lenticular deposits in which bedding results from the stacking of charcoal-rich levels (black and dark grey), phosphate-rich levels (dark brown, brown and yellowish brown) and (rare) ash-rich levels, sandy lenses and gypsum patches (Fig. 5). There are also lenses of red silty sand, some with tiny (1–2 cm) quarzitic platelets. All lenses are of limited extent, boundaries are often transitional, and deposits dip towards the shelter mouth. This accumulation is visually very reminiscent of that described by Singer and Wymer (1982) from the Howieson’s Poort levels in Cave 2 and Shelter 1A at Klasies River. We referred initially to the frequent halite-cemented layers in P10 as crusts, wrongly, as they are subsurface postdepositional phenomena. Regular if localised redistribution of materials by percolating water characterises this phase of deposition. The gypsum patches are evidence of solution and precipitation events associated with water percolation. Lenses are packed one upon another, though the edges are fuzzy and indistinct, an impression no doubt exaggerated by the disappearance of much organic material and the action of percolating water (Lenoble & Texier, 2001). By contrast, in the upper parts of C5 associated with unifacial points, ashy patches (hearths?) are much more clearly defined and gypsum virtually absent. There are no halite-cemented horizons here. It is no coincidence that uncharred plant material, such as the wood dated from E6, is restricted to these parts of the shelter, as they appear to have been much less impacted by percolating moisture. This might be taken to point to drier conditions toward the end of the MSA occupations in this site, though Lenoble and Texier (2001) argue that no differences between Howieson’s Poort and current conditions can be posited. They add, moreover, that the prevalence of gypsum and halite

484

Intentionally marked ostrich eggshell from Diepkloof

mark conditions of relative aridity and evaporation. The complex relations between rainfall, cloud cover, lower carbon dioxide content of the atmosphere, lower ambient temperature and changes in percolating moisture clearly remain to be unravelled. Ostrich eggshell fragments have been separated from other materials (bone, stone, charcoal, marine shell) in the archaeology laboratory at UCT, prior to any washing. All fragments have been visually inspected and any possibly marked pieces have been examined under a low-power binocular microscope. Almost all of the MSA eggshell pieces have been altered by fire or the chemical conditions of the depositional matrix, so that they are ochreous, maroon, dark brown or black in contrast to the pale yellow colour of the LSA pieces. There are, perhaps surprisingly, no marked LSA ostrich eggshell fragments at Diepkloof. Although it was not clear until recently (Feathers et al., in press), we can now demonstrate that all pieces of marked ostrich eggshell at Diepkloof are associated with segments and underlie levels with unifacial points.

Figure 6 Intentionally marked ostrich eggshell fragments. Scale = 1 cm.

485

From Tools to Symbols

There are, however, segment-bearing levels that lie below the lowermost marked ostrich eggshell.

Intentionally marked ostrich eggshell What, then, is the form of these apparently marked pieces, and what is their significance for the ‘modern human origins’ debate? We have found it useful to make a number of distinctions. First, there are many pieces of ostrich eggshell that preserve intricate sets of apparently randomly oriented light scratch marks. Scratch marks are often short and usually do not cross the eggshell pieces from edge to edge. These

a

b

c

d

Figure 7 (a–c) Intentionally marked ostrich eggshell fragments; (d) intentionally marked ostrich eggshell water flask rim fragment. Scales = 1 cm.

486

Intentionally marked ostrich eggshell from Diepkloof

we do not claim to be intentional markings, although they may result from human interactions with the eggs. We return to them later, because there is an intriguing correspondence between the provenance of these pieces and those we do consider to have been intentionally marked. More specifically, not a single piece with light scratch marks has been found in association with unifacial points. More significantly, there are more than eighty pieces of ostrich eggshell, almost all of them less than 25 mm in maximum dimension, that we believe to have been intentionally marked. Two kinds of marking are apparent, although there is a significant diversity of actual marking form. There are some cases of fairly deep U-shaped gouging of the surface leading to the removal of the uppermost ostrich eggshell layer. The edges of gouges are often marked by spalling or splintering. Contrasting with this are cases where finer V-shaped incisions have been made into the surface with little removal of material from the egg surface. We refer to these markings at present as, respectively, gouges and incisions. Gouges seem to have been produced by a blunter point. Rather than reify this distinction, we now describe some of the pieces and illustrate them as representative of the sample range. Two fragments, darkened by heat or the effects of soil chemistry, from near to P10 (Fig. 6) are characterised by sets of parallel gouges and incisions, with other gouges intersecting these at low angles (30–40º). The organisation of these markings surely precludes any explanation other than that they were produced by human intent. These have already been published (Parkington, 1999), but not yet properly described (Parkington et al., in prep). Another fragment shows possibly more organisation, in that diagonally intersecting incisions appear to be placed within bounded spaces above one another (Fig. 7a). The surface of this piece is somewhat damaged. Even more interesting are two fragments (Fig. 7b–c) that undoubtedly show that the maker intended to delineate zones or bands and then infill these spaces with hatched lines. The sixth piece illustrated here (Fig. 7d) is probably the most significant of all in that incisions radiate out from what we believe to have been an intentionally made water bottle mouth. More detailed analyses of these pieces are in progress. Comparison of these fragments with LSA decorated ostrich eggshell pieces raises the question of whether the MSA marked pieces were all parts of ostrich eggshell water flasks. Mindful of Kandel’s (2004, in press) observations on the kinds of markings produced on ostrich eggs by hyaenas, we have searched the Diepkloof sample for water flask rim fragments. There are many questionable but potential rim fragments and two that we consider quite persuasive, one of them marked. The polished, rounded cross-section of these fragments is very similar to those observed on LSA water flasks. Moreover, the opening is formed by multiple removals from the outer surface, unlikely to be the result of hyaena mouthing. It is here that the significance of the light scratch

487

From Tools to Symbols

marks may lie. Intentionally marked and lightly scratched eggshell fragments come from the same excavated units but both marks rarely appear on the same pieces. This raises the possibility that the scratches were produced in the course of intentional marking, but on different parts of the eggs. We are very keen to decide whether it was whole eggs or fragments that have been intentionally marked. We are also interested in determining what part of the egg carries the marking. Certainly the gouges and incisions extend to the very edges of most of the marked pieces, suggesting at least that larger egg fragments than those we have were marked. Although we have refitted several pieces of eggshell, so far none of the marked pieces fit with a rim. This underlines, then, the significance of the intentionally marked rim (Fig. 7d). At least some of the intentional marking is located around the rim of a water flask, but is that the only context? We have established from photogrammetric analysis of whole eggs (Ruther, 2003: pers. comm.) that it is possible to distinguish the curvature of ‘polar’ pieces from that of ‘equatorial’ pieces so long as pieces are more than about 30 mm in size. It remains necessary to refit pieces that are marked so that they become large enough to establish their location on the original egg. An issue we plan to pursue in future excavations and analyses is the relationship between the chronology of these intentionally marked pieces and other technological and environmental evidence from Diepkloof. We note that at Apollo 11 cave intentionally marked ostrich eggshell is also associated with backed stone artefacts very similar to those at Diepkloof (Vogelsang, 1998, 81). Seemingly, the Howieson’s Poort assemblage type in southern Africa is not simply a manifestation of stone tool technological innovations that subsequently disappear. It appears to be related to marking behaviours associated with storage devices that also disappear from the MSA record. Further, if Diepkloof is reflective of general patterns, the disappearance of Howieson’s Poort behaviours is associated with bio-archaeological evidence for environmental and subsistence change. Although this remains to be demonstrated, the transition from Marine Isotope Stage 4 to 3 might be the temporal context of this disappearance. The correlations between changes in the sediment, charcoal, bone and artefactual assemblages are clues to the causal relationships between behaviour and environments. We are confident that Diepkloof can contribute significantly to the understanding of these relationships.

From tool to symbol We do not claim at this point that all marked pieces were once parts of water flasks, but it is possible that they were. If this possibility is strengthened by refits and further finds, marking appears to reflect the investment of energy in distinguishing one egg (water flask) from another. This marking would then be linked to one of the earliest

488

Intentionally marked ostrich eggshell from Diepkloof

demonstrable artefacts designed for storage. Admittedly, it might be argued that all artefacts represent some form of symbolic storage in future use and rewards, but a water flask stores both in this way and, more literally, in its capacity as a vessel. To the average viewer one ostrich egg looks very much like the next. This presumably doesn’t matter if the function of the egg is to provide food for the collector, after which it is soon discarded. But if the egg is transformed from foodstuff to container, some discrimination between eggs might be necessary. Now an egg becomes that particular egg, my egg. The egg is now a container, useful for transport, but also for storage. An eggshell water flask has a life expectancy that requires that it becomes distinguishable from other similar flasks, essentially to denote ownership. Although only one intentionally marked eggshell fragment is clearly from a flask, the rest are very varied in their ‘design’. The intentional marking has had the effect of making all eggs distinctive. What can we draw from this? In reviewing the literature on the ‘earliest’ or the ‘origins of’ intentional marking (or art, or decoration), we are struck by two areas of potential debate. First, it appears that most usages of terms such as decoration or marking exclude artefacts of flaked stone from consideration. Why is a symmetrically flaked handaxe not eligible for the designation art? Flaking of stone seems to be perceived as part of the making rather than the marking of an artefact. The thinking seems to be that flaked form is functional rather than arbitrary and, thus, does not count as carrying symbolic information. We wonder whether this exclusion is reasonable. Second, when according significance to marking, it seems reasonable to expect that some significance may have been given to objects naturally, rather than culturally, marked. Discrimination between similar objects can certainly reside in pre-existing, rather than intentionally produced, properties. These two observations remind us that what we are trying to reconstruct here are patterns of thought, not all of which manifest themselves unambiguously in the material archaeological record. The intentionally marked ostrich eggshell fragments from Diepkloof may reflect a pattern of thinking long practised on other materials or with respect to natural markings. Nevertheless, it seems to us that the first recognisable application of intentional marking to ostrich eggshell, especially of water storage flasks, is certainly worth celebration. We propose to avoid an ‘earlier than thou’ debate, for which the discoverydriven practice of archaeology is not well suited, and to focus instead on the excavated associations of the marked pieces, an approach more suited to the nature of this kind of field research. Contexts of discovery are recorded in our notes, contexts of use may be reflected in empirical associations, but contexts of significance remain elusive. What can we reconstruct of the authorship and behavioural context of these marked eggs? Who made them? How were they made? The co-occurrence of intentionally marked ostrich eggshell fragments and backed

489

From Tools to Symbols

segments at Diepkloof is suggestive but not exact. It is, though, tempting to wonder whether the purpose of the blunting of the curved edge was not to allow the segments to be used in a pencil grip to score the egg surfaces. This might explain the burinlike spalls driven off parallel to the main dimension of some of the segments, though fewer removals along the chord and more along the arc might have been expected. As an analogue, we refer here to perhaps the only Later Stone Age segment preserved complete with hafting, or in this case holding, device. From an early Holocene burial dated about 8 000 years ago at Elands Bay Cave (Parkington, in prep.), this segment, slightly smaller than most MSA ones, was clearly held resting in a mastic holder intended to be gripped between thumb and the first two fingers. Another parallel with Later Stone Age patterns follows from this and other observations. Most terminal LSA ostrich eggshell water flask rims are not intentionally marked: indeed ‘decorated’ ostrich eggshell is rare along the Cape west coast after about 3 000 years ago (Parkington, in prep.; Halkett & Hart, 2003: pers. comm.). In an important, but yet unpublished, survey of Namaqualand coastal surface shell middens, Dave Halkett and Tim Hart have observed an increase in decorated ostrich eggshell alongside both shellfish and stone tool changes prior to this time, presumably in the late middle Holocene period. At this time, as elsewhere in the Cape, but not everywhere, LSA segments like that from Elands Bay Cave are common. Although possibly coincidence, this repeated association between small curved backed geometric artefacts and intentionally marked ostrich eggshell water flasks requires explanation. If the function of the Diepkloof segments can be taken as domestic marking of water storage flasks, rather than as inserts in hunting equipment, then we can look at other domestic evidence from these same levels. Our presumption is that these flasks were used to transport fresh water from the river below the cave, to be stored and used during visits. The very light scratch marks may come from twirling the flasks into the gritty sand around a hearth in order to stabilise the position of the flask. Alternatively, the scratches may result from setting the egg in sandy deposit during the marking. Carbonaceous lenses are densely packed in Howieson’s Poort levels at Klasies River main site (Singer & Wymer, 1982; Deacon & Geleijnse, 1988), a circumstance that may characterise all such stratified occurrences. At the very least these observations remind us that hunting, whilst important, need not have been the behavioural context responsible for changes in stone tool forms. There may also be gender implications in such a stance. Richard Klein’s identification of faunal remains at Diepkloof underlines the rarity of large forms and dominance of small animals such as dassie, dune mole rat and hare, beyond the ubiquitous tortoise, in segment-rich levels. Larger forms, though rare, are present in these excavated units (Costamagno, 2001), so that the contrast with later levels may well reflect the logistical variation in resource exploitation or domestic organisation as

490

Intentionally marked ostrich eggshell from Diepkloof

much as any significant alteration in animal populations. Above these levels intentionally marked ostrich eggshell disappears, soft hammer blank removal disappears and segments are replaced, numerically if not functionally, by unifacial points. More significantly, perhaps, we can ask why the intentional marking apparently ceased, along with some technological and typological changes and evidence for different faunal associations? Diepkloof is emerging as an important archive for investigating stratigraphic associations between different kinds of evidence. Obviously, despite surface appearances, the sediments at this shelter have accumulated unevenly over the available space and preserve a record of artefact manufacture, intentional marking of ostrich eggshell and domestic behaviour associated with many kinds of palaeoenvironmental signals. The significance of wood charcoal, unburnt plant remains, bone, ostrich eggshell and sediments as indicators of past climatic and environmental circumstances is still to be determined. The opportunity of enriching our understanding of mid-Upper Pleistocene hominid behaviour beyond the merely technological is one we are keen to grasp and develop in ongoing field and laboratory research. We recognise that the potentially symbolic nature of the intentional markings we discuss here requires us to understand the reference system that allowed the markings to ‘mean something’. Diepkloof offers opportunities to investigate the domestic context of the marking and the utilisation of marked objects. It also appears to cover time periods during which the marking first appears and after which it apparently ceased. These historical frameworks may allow us to go beyond claims for early symbolism into a fuller investigation of the behavioural circumstances in which marking of ostrich eggs, probably water flasks, was developed.

References Cartwright, C. & Parkington, J. (1997). The wood charcoal assemblages from Elands Bay Cave, Southwestern Cape: principles, procedures and preliminary interpretation. South African Archaeological Bulletin 52: 59–72. Costamagno, S. (2001). Etudes des Restes Faunique. In Rigaud, J.-P. (Ed) Rapport sur les travaux effectués au cours de la Campagne 2001 dans l’abri de Diepkloof. Unpublished report, Institut de Prehistoire et de la Geologie du Quaternaire, pp. 33–50. Cowling, R.M., Cartwright, C.R., Parkington, J.E. & Allsopp, J.C. (1999). Fossil wood charcoal assemblages from Elands Bay Cave, South Africa: implications for Late Quaternary vegetation and climates in the winter rainfall fynbos biome. Journal of Biogeography 26: 367–378. Deacon, H.J. & Geleijnse, V.B. (1988). The stratigraphy and sedimentology of the main site sequence, Klasies River, South Africa. South African Archaeological Bulletin 43: 5–14. Deacon, J. (1995). An unsolved mystery at the Howieson’s Poort name site. South African Archaeological Bulletin 50, 110–120.

491

From Tools to Symbols Feathers, J., Woodborne, S., Vogel, J., Parkington, J., Yates, R. & Poggenpoel, C. (in press). Luminescence dating of Howieson’s Poort deposits at Diepkloof Rock Shelter, Western Cape, South Africa. South African Journal of Science. Kandel, A.W. (2004). Modification of ostrich eggs by carnivores and its bearing on the interpretation of archaeological and paleontological finds. Journal of Archaeological Science. Lenoble, A. & Martinaud, M. (2003). Apports du penetrometre à la connaissance d’une site préhistorique: le cas de l’abri de Diepkloof ( Province du Cap, Afrique du Sud). Unpublished poster. Congres GMPCA, Bordeaux, 16–19 April 2003. Lenoble, A. & Texier, J.-P. (2001). Rapport concernant les travaux géologiques réalisés en 2001. In (J.-P. Rigaud, Ed.) Rapport sur les travaux effectués au cours de la Campagne 2001 dans l’abri de Diepkloof. Unpublished report, Institut de Prehistoire et de la Geologie du Quaternaire, pp. 8–25. Parkington, J.E. (1977). Follow the San. Unpublished PhD thesis, Department of Archaeology and Anthropology, Cambridge University. Parkington, J. (1990). A critique of the consensus view on the age of Howieson’s Poort assemblages in South Africa. In (P. Mellars, Ed.) The Emergence of Modern Humans: An Archaeological Perspective. Edinburgh: Edinburgh University Press. Parkington, J. (1998). Resolving the past. In (S. Kent, Ed.) Gender in African Prehistory. New York: Altamira Press. Parkington, J.E. (1999). Western Cape landscapes. In (J. Coles, R. Bewley & P. Mellars, Eds) World Prehistory: Studies in Memory of Grahame Clark. Proceedings of the British Academy 99: 25–35. Oxford: Oxford University Press. Parkington, J.E., Cartwright, C., Cowling, R.M., Baxter, A. & Meadows, M. (2000). Palaeovegetation at the last glacial maximum in the Western Cape, South Africa: wood charcoal and pollen evidence from Elands Bay Cave. South African Journal of Science 96, 543–546. Parkington, J. & Poggenpoel, C. (1987). Diepkloof Rock Shelter. In (J. Parkington & M. Hall, Eds) Papers in the Prehistory of the Western Cape, South Africa. BAR International Series 332 (ii), 264–293. Parkington, J. (in prep.). Elands Bay Cave: a view on the past. Parkington, J., Yates, R. & Poggenpoel, C. (in prep.). Intentionally marked ostrich eggshell fragments from Diepkloof Shelter, Western Cape Province, South Africa. Singer, R. & Wymer, J. (1982). The Middle Stone Age at Klasies River Mouth in South Africa. Chicago: University of Chicago Press. Vogelsang, R. (1998). Middle-Stone-Age-Fundstellen in Sudwest-Namibia. Köln: HeinrichBarth-Institut. Wurz, S. (2002). Variability in the Middle Stone Age lithic sequence, 115000-60000 years ago at Klasies River, South Africa. Journal of Archaeological Science 29, 1001–1015.

492

Chronology of the Howieson’s Poort and Still Bay techno-complexes: assessment and new data from luminescence Chantal Tribolo, Norbert Mercier and Hélène Valladas Laboratoire des Sciences du Climat et de l’Environnement (UMR 1572), Avenue de la Terrasse, F-91198 Gif-sur-Yvette Cedex, France

Abstract The chronologies of the Still Bay and the Howieson’s Poort techno-complexes have been hotly debated for the last three decades, because though these technologies belong to the South African Middle Stone Age (MSA) they show affinities with those of the Late Stone Age (LSA) and are found associated with remains that reflect modern human behaviour. The aim of this article is to sum up and discuss the various published chronological data. In the first part are summarised the hypotheses derived from multidisciplinary (geological, palaeontological, stable-isotopic…) studies at Border Cave and Klasies River Mouth which place the Howieson’s Poort within the last glacial cycle. In the second part a summary and evaluation of the radiometric and biochemical dates obtained since 1990 for the remains associated with these two techno-complexes is followed by a presentation of the thermoluminescence ages recently calculated for the burnt stones from the Howieson’s Poort levels of the Klasies River Mouth (56 ± 3 thousand years (Ka) and the [Still Bay] levels of the Blombos Cave (74 ± 5 Ka).

Résumé La chronologie des techno-complexes Still Bay (SB) et Howieson’s Poort a fait l’objet de nombreuses discussions depuis une trentaine d’années car, bien qu’attribués au Middle Stone Age (MSA) de l’Afrique australe, ces techno-complexes présentent des affinités avec ceux du Later Stone Age (LSA) et sont associés à des vestiges qui reflètent des comportements modernes. L’objectif de cet article est de faire le bilan des données chronologiques disponibles et de les discuter. La première partie résume le résultat d’études pluridisciplinaires (géologie, paléontologie, isotopes stables...) réalisées dans les sites de Border Cave et de Klasies River Mouth. Ces études situent l’Howieson’s Poort au sein du dernier cycle glaciaire. La seconde partie dresse le bilan

493

From Tools to Symbols des datations radiométriques ou biochimiques réalisées depuis les années 90 sur des vestiges associés à ces deux techno-complexes. Puis elle présente les âges moyens obtenus récemment par la méthode de la thermoluminescence appliquée aux pierres chauffées du niveau Howieson’s Poort de Klasies River Mouth (56 ± 3 Ka) et du niveau Still Bay de Blombos Cave (74 ± 5 Ka).

Introduction The Howieson’s Poort and Still Bay techno-complexes include certain characteristics usually attributed to Late Stone Age: the backed geometric tools of Howieson’s Poort are reminiscent of Wilton and the bifacial foliate points of Still Bay have been compared to those of the Solutrean (Fig. 1). These tools are relatively standardised and are manufactured from exotic materials, while typical Middle Stone Age artefacts, representing a greater variability, are made from local rocks. Moreover, these technocomplexes are sometimes associated with artefacts far rarer in the MSA than in the LSA. For example, engraved ochres are associated with the Still Bay at Blombos Cave, and incised ostrich eggshell with the Howieson’s Poort at Diepkloof Rock Shelter. Stratigraphic observations showing that these techno-complexes are indeed South African MSA entities instead of belonging to the MSA/LSA transition (Beaumont et al., 1978; Singer & Wymer, 1982; Thackeray, 1992; Wadley, 1997), have given rise to debate about their possible relations with the emergence of ‘modern’ cognitive behaviour and capacities (Deacon & Wurz, 1996; Klein, 1999; McBrearty & Brooks, 2000; Henshilwood et al., 2001a, b). Even if it is difficult to find a way out of this debate, especially since the notion of modern behaviour is controversial, it cannot be denied that Howieson’s Poort and Still Bay are significant phases in the behavioural evolution of anatomically modern human. Yet, understanding this evolution automatically requires knowledge of where and when the different behaviours appeared. The chronological position of Howieson’s Poort and, to a lesser extent, of Still Bay, has been the subject of many studies. It is our intention in this article to assess these studies and to present new data obtained through the luminescence methods applied to heated stones collected in the Still Bay layers of Blombos Cave, and in the Howieson’s Poort layers of Klasies River Mouth Cave. We do not intend to dwell on the methodological aspects of the dating, but wish to discuss the implications of adding the new dates to the existing data.

494

Chronology of the Howieson’s Poort and Still Bay techno-complexes

Chronological data for the Howieson’s Poort Vogel and Beaumont (1972) showed that most of the MSA was beyond the application limits of the radiocarbon dating method. However, the numerous finite ages obtained in addition to infinite ones (a few examples are given in Table 1), created some confusion. Moreover, Howieson’s Poort has for a long time been (and still is) perceived as a synchronous and short-lived phenomenon, and such different ages are not compatible with this hypothesis, which brought certain authors to consider that finite dates were the result of contamination by ‘young’ carbon, something impossible to verify. Table 1

C dates for Howieson’s Poort or post-Howieson’s Poort sites.

14

Sites

Dates BC

Note

Reference

Howieson’s Poort

3 940 ± 60 9 380 ± 210 11 120 ± 160 9 700 ± 90 19 070 ± 190 9 540 ± 100 18 740 ± 320 19 600 ± 220

J Deacon, 1995

Nelson Bay Cave

24 120 ± 660 17 600 ± 195 22 400 ± 340

Fairhall et al., 1976

Diepkloof

Umhlatuzana

Parkington and Poggenpoel, 1987

29 400 ± 675 > 24 400 > 45 270 40 800 ± 1 400 42 400 ± 1 600

Parkington, 1990

45 200 ± 3 200

Boomplaas

> 40 000

Border Cave

> 48 700 36 000 ± 1 000 > 47 500 > 48 350 > 42 600 42 000 + 3 000 42 000 – 2 000 > 48 5000

Rose Cottage

> 50 200 36 100 ± 2 000 > 39 000 > 42 500 > 48 400 > 50 000

Kaplan, 1989 Post HP

Fairhall et al., 1978 Butzer et al., 1978 Beaumont, 1980

Post HP Post HP Post HP Post HP Post HP Post HP

ref. in Wadley, 1991

495

From Tools to Symbols

a

b

c

d

Figure 1 Typical Howieson’s Poort (a, b) and Still Bay (c, d) stone tools. (a) trapeze and segment from Border Cave (after Beaumont et al. 1978); (b) triangle, trapezoid and segment from Klasies River (after Singer & Wymer, 1982); (c) bifacial point from Hollow Rock Shelter (after Evans, 1994); (d) bifacial point from Blombos Cave (after Henshilwood et al. 2001).

With few exceptions, other dating methods have only been applied in South Africa in the last fifteen years. Previously, researchers used palaeoenvironmental data supplied by archaeological sites (fauna, sediments, etc.) to infer local climatic conditions and place their evolution within the context of global climatic evolution (marine isotopic scale), which is itself relatively well dated (e.g. Martinson et al., 1987). The Border Cave deposits and especially those of Klasies River Mouth played a predominant role in these studies (Fig. 2).

496

Chronology of the Howieson’s Poort and Still Bay techno-complexes

Hypotheses drawn from palaeoenvironmental studies Following the works of Butzer (1978, 1982, 1984), Klein (1974, 1976, 1977) and Avery (1982, 1987, 1992), the prevailing hypothesis was at first the association of the Howieson’s Poort with isotopic stage 5. In Klasies River Mouth in particular, the three authors argued that stage 4 was probably not recorded in the sequence, various clues suggesting the proximity of the coast (cf. also Hendley & Volman, 1986) and thus the lack of an important marine regression that was a priori expected for that stage. However, these authors disagreed about the characteristics of the prevailing climate during Howieson’s Poort times. Butzer (1984), for example, associated it with a relatively cold period (stage 5b): rock fragments (éboulis secs) present in the Border Cave and Diepkloof sedimentary layers have indeed been interpreted as a consequence of frost, and the presence of a ferruginous soil in Nelson Bay Cave has also been interpreted as an indication of climatic cooling. D.M. Avery (1982, 1987, 1992), on the other hand, on the basis of micro-mammal biodiversity and vegetal reconstructions, associated the Howieson’s Poort with a relatively mild period, whether in Border Cave or Klasies River Mouth, and proposed stages 5c or 5a. The hypothesis that the Howieson’s Poort is associated with stage 3 has also been put forward but has remained more marginal (Parkington, 1990). After analysing the isotopic ratios of oxygen in Klasies River seashells, Shackelton (1982) concluded that these results corresponded to an environment slightly colder for the sample from the Howieson’s Poort level than for the other levels. Although he proposed stage 5b, he did not exclude stage 3. In 1992, J.F. Thackeray (cf. also Thackeray, 1987; Thackeray & Avery, 1990) conducted a multivariate statistical study on the micro-fauna of Klasies River. Complementing his analysis with that of the seashells’ frequency variation in the sequence, he inferred a temperature evolution curve and adjusted it on to the global marine isotopic scale. The Howieson’s Poort was thus associated with the warm phases of stage 3, c. 58–48 Ka. The third and last hypothesis developed is the ‘Klasies Regression Hypothesis’: contrary to what the above-mentioned authors suggested, Deacon wrote that the Howieson’s Poort at Klasies River is associated with the major regression that took place during the transition from stage 5a to stage 4 (Deacon et al., 1986, 1988; Deacon & Geleijnse, 1988; Deacon, 1989, 1995). His arguments are based at the same time on new isotopic measurements in seashells, on the ratio and nature of land and marine species, on the evolution of micro- and macro-fauna and, finally, on the nature of the sediment. According to this hypothesis, the carbonised structures of the Howieson’s Poort levels are interpreted as residues from the voluntary burning of geophytes, and therefore as indicating a terrestrial (instead of marine) economy.

497

From Tools to Symbols

498 Figure 2 Radiometric dates obtained for the Howieson’s Poort and the Still Bay. U/Th: Uranium/Thorium; ESR: Electron Spin Resonance; AAR: Racemisation of the Acidamines. Each point corresponds to a dated sample, AAR are means. The date interval for Diepkloof is preliminary and comes from five samples (after Miller et al., 1999; Vogel, 2000; Grün & Beaumont, 2001; Feathers, 2002; Grün et al., 2003; Jacobs, Valladas & Grün, pers. comm.).

Chronology of the Howieson’s Poort and Still Bay techno-complexes

It must be noted that the various authors who followed the Klasies Regression Hypothesis have interpreted the Howieson’s Poort as a consequence of the strong climatic deterioration at the beginning of stage 4, with very different explanations of the underlying mechanisms, and have also put forward very diverse conclusions regarding the cognitive capacities of the makers of Howieson’s Poort artefacts (Deacon, 1989, 1993, 1995; Deacon & Wurz, 1996; Ambrose & Lorenz, 1990; see also Avery, 1987). Critique of the hypotheses At the end of the 1980s, several authors criticised the use of the palaeoenvironmental data for chronological purposes by highlighting existing contradictions between various palaeoclimatic interpretations of observed geological phenomena and biological data (Van Andel, 1989; Parkington, 1990; A.I. Thackeray, 1992). One example concerns the éboulis secs and their supposed association with frost as proposed by Butzer, which has been disputed by other authors (Goede, in Deacon et al., 1984; Texier, in Rigaud et al., 2000). Furthermore, the reliability of the correlations established between palaeoclimatic indicators and the global isotopic curve has also been questioned because of the discontinuous nature of the archaeological data and the very oscillating nature of the marine isotopic curve. The discussion also concerns the interpretation of radiocarbon dates. Thus, Parkington (1990) argues that those derived from stage 3 should be given more consideration. He notes indeed that the only reason for rejecting these dates is based on a presumed age for the Howieson’s Poort and on the synchronism hypothesis for this techno-complex, which is not unequivocally confirmed by palaeoenvironmental data. A.I. Thackeray (1992) concludes: ‘it is the attempt to explore ways of obtaining absolute dates that can only contribute to solving the chronological chasm in which studies find themselves at present’. Radiometric and chemical data Radiometric dating methods other than radiocarbon have been used at South African MSA sites, but before 1990 were never directly applied to samples associated with the Howieson’s Poort. Published data for this techno-complex come from only four sites, with a limited number of samples for each (Fig.2). Grün et al. (1990a, 1990b, 2003), and Grün and Beaumont (2001), have performed dating using the electron spin resonance (ESR) method on teeth from the Border Cave and Klasies River Mouth deposits. These studies have been the subject of several updates following improvements in dose rate calculations and technological advances, such as combining ESR with Laser Induced Coupled Plasma Mass Spectrometry (LAICP-MS) for detailed studies of radioisotope distributions in samples. The latest results

499

From Tools to Symbols

obtained for the Howieson’s Poort at Border Cave gave six mean ages (for six teeth) falling between 58±2 and 76±4 Ka, and for Klasies River Mouth, a single age of 53±3 Ka with ages of 53 ±3 and 64+5–4 Ka for pre- and post-Howieson’s Poort, respectively. Vogel (2000) applied the Uranium-Thorium (U-Th) method on stalagmitic floors located in the Howieson’s Poort levels at the Klasies River Mouth and Boomplaas sites (Deacon et al., 1984). The two ages obtained were 65.6±5.3 Ka and 62.4±2.0 Ka, after correction of initial pollutions in Th232. Several authors have applied the amino acid racemisation (AAR) method to ostrich eggshell. Miller et al. (1999) published results for the Boomplaas (56 or 65 Ka, according to the calibration used) and Apollo 11 (63±6 and 69±7 Ka) sites. Brooks et al. (1993) obtained older ages for the Boomplaas (72–80 Ka) and Klasies River Mouth (65–80 Ka) deposits, but the details of the dating have not yet been published (McBrearty & Brooks, 2000; Brooks et al., 1993). Finally, by combining results from the Thermo-Luminescence (TL) and Optical Stimulated Luminescence (OSL and IRSL) techniques applied to quartz and feldspar grains, Feathers (2002) obtained ages of 52.4±4.0 and 46.7±3.3 Ka for the Howieson’s Poort levels at Klasies River Mouth. He foresaw the possibility that these results might be slightly under-estimated because of the increase of dose rate with time (because of sediment compaction after dissolution of the carbonated seashells). Taking this evolution into account, he proposed the interval of 55–60 Ka as the most likely one. To sum up, the few available results incline us to reject the association of Howieson’s Poort with stage 5, but they do not allow us to draw any conclusion regarding the transition of stages 5a to 4 or the beginning of stage 3. Yet in the literature of the last decade, it has often been reported that Howieson’s Poort was dated to 70 Ka or more, with the rare available data pointing rather in favour of the Klasies Regression Hypothesis (A.I. Thackeray, 1989; Ambrose & Lorenz, 1990; G. Avery, 1995; Deacon & Wurz, 1996; Avery et al., 1997; Klein, 1999; Carrion et al., 2000; Wurz, 2000). To be more precise in placing the Howieson’s Poort chronologically, it was therefore necessary to extend the application of the radiometric methods to the sites already mentioned and to initiate new dating on other deposits. In this quest, it is interesting to begin with Klasies River Mouth, since the study of this site has greatly influenced MSA research in South Africa. New TL ages for the Klasies River deposit We selected heated quartzites weighing between 8 g and 152 g and originating from the Howieson’s Poort layers at Klasies River Mouth, with an almost even distribution from top to bottom (squares E50 and H51; Fig. 3). The TL technique was used to evaluate the dose accumulated by the samples – named equivalent dose

500

Chronology of the Howieson’s Poort and Still Bay techno-complexes

or palaeodose – and the external dose rate was measured with dosimeters implanted in the archaeological sections (Aitken, 1985; Valladas, 1992). The internal dose rate was deduced from radioisotopic contents determined by neutron activation analyses (Joron, 1974), and by taking into account the size of the quartz grains making up the stones (Tribolo, 2003). The thirteen ages obtained being indistinguishable at one sigma level (Table 2), a weighted mean age of 56±3 Ka has been calculated for the whole of the Howieson’s Poort levels. Such an age is concordant with those of Grün and Feathers (cf. above) and places the Howieson’s Poort at the beginning of isotopic stage 3, but conflicts with the Klasies Regression Hypothesis since the limit of stages 5 and 4 is dated to 74±3 Ka (Martinson et al., 1987). As Feathers (2002) suggested, it is possible that TL ages are underestimated because of the evolution of external dose rates with time. However, in the case of the Klasies River heated stones, the TL ages are closely clustered despite internal dose rates and palaeodoses that vary greatly from one sample to another (from 97±6 to 849±49 µGy/a and from 32.0±1,1 to 84.1±3,8 Gy, respectively). Such coherence indicates that if the external dose rates did vary, the ages were not significantly affected.

Table 2 Thermoluminescence dates for the Howieson’s Poort levels at Klasies.

Internal dose rate

External dose rate

Annual dose rate

Ed

Age

Gamma Cosmic µGy/a µGy/a

± µGy/a

± Gy

± ka

Sample

Gamma µGy/a

Alpha µGy/a

Beta µGy/a

KRM114

10

6

266

428

110

819

88

46,4

3,7

57

8

KRM115

18

9

284

423

110

845

86

48,8

2,4

58

7

KRM116

53

38

811

423

110

1 435

98

84,1

3,8

59

6

KRM112

10

9

193

428

110

749

86

40,9

1,5

55

7

KRM120

18

15

407

419

110

969

89

49,6

1,7

51

6

KRM121

12

3

116

405

110

646

81

37,5

2,1

58

8

KRM81

18

13

268

487

35

820

51

47,4

2,6

58

5

KRM76

6

4

93

458

35

595

46

33,9

1,2

57

5

KRM86

19

6

497

458

35

1 014

61

63,3

4,8

62

7

KRM102

16

11

247

458

35

766

49

42,3

1,5

55

5

KRM96

8

4

134

453

35

634

46

33,9

4,8

54

9

KRM97

9

4

141

453

35

642

46

32,0

1,1

50

4

KRM103

8

4

139

420

35

606

61

35,8

2,2

59

7

501

From Tools to Symbols

Figure 3 Schematic rendering of the Klasies site section in squares E50 and H51. The ellipses indicate the location of dated samples (stratigraphy courtesy of H. J. Deacon).

502

Chronology of the Howieson’s Poort and Still Bay techno-complexes

The mean TL age of 56±3 Ka also fits well with the hypotheses of Shackelton (1982) and Thackeray (1992) based on the analysis of the seashells and micro-fauna of Klasies River Mouth. Finally, note that recent data on other sites place the Howieson’s Poort closer to 60 Ka than 70 Ka. Indeed TL analyses carried out for the Diepkloof Rock Shelter deposits (Parkington et al., this volume) lead to a preliminary age interval of 55–65 Ka (Tribolo, 2003), and TL results for three burnt lithics from the upper Howieson’s Poort levels at Rose Cottage gave a mean age of 57± 4 Ka (Valladas et al., submitted). Jacobs (2004, this volume) have slightly older ages for Howieson’s Poort or postHowieson’s Poort at Sibudu and Rose Cottage, but they still range in the 60–70 Ka interval and are believed to be linked with the transition from stage 4 to stage 3. All these data therefore suggest that at least part of the Howieson’s Poort industries is associated with the beginning of isotopic stage 3.

Still Bay Hypotheses based on stratigraphic correlations There is less to be said about Still Bay than Howieson’s Poort. Still Bay is known from far fewer sites, located only in the Western Cape Province, and this technocomplex, acknowledged during former excavations (at the eponymous site of Still Bay, Peers Cave, Dale Rose Parlour), was almost forgotten until the discovery of Blombos Cave. Chronological data are therefore few. In addition, the stratigraphic sequences are not well developed, as with Hollow Rock Shelter (Evans, 1994) and limit the possible correlations. As a result, Henshilwood and Sealy (1997) at first suggested an age of 50–60 Ka for the Still Bay deposits of Blombos Cave. This estimate was based on an argument taking into account infinite radiocarbon ages, the presence of many remains of fish and seashells, suggesting the proximity of the coast and therefore a relatively warm phase of the last ice age, and the fact that Still Bay could be placed a priori at the end of the MSA. However, a few years later, Henshilwood et al. (2001a) proposed an older age for the Still Bay deposits, remarking that the bifacial points (Still Bay or not) are found below the Howieson’s Poort levels but never above. The Still Bay would then be earlier than the Howieson’s Poort. Available radiometric data An OSL study of Blombos Cave deposits gave an age of 103±9,8 Ka for Still Bay by using the subtraction method (Vogel et al., 1999), but doubts were raised as to its reliability because of possible contamination of the sample by quartz grains originating from the cave walls. Indeed, the presence of grains unexposed to light would lead to an

503

From Tools to Symbols

overestimation of the equivalent dose and thus of the age (Woodborne, in Henshilwood et al., 2001b, p. 426). Efforts to date the Blombos Cave deposits were thus continued (Henshilwood et al., 2002). The single-grain OSL technique was applied by the luminescence team of the University of Aberystwyth (Jacobs et al., 2003a, b; Jacobs, 2004) in order to avoid contaminating (badly bleached) grains. Ages of 65,8± 2.8, 67,3±3.8 and 68,8±3.0 Ka (69,6±3.5, 69,2±3.9 and 70,9± 2.8 Ka for the same samples with the Multiple Grain technique) were thus calculated for the sterile dune covering the Still Bay levels, and an age of 74,9 ±3,8 Ka was proposed for the anthropogenic Still Bay levels. The ESR technique was also used (Jones, 2001) to date seven teeth, giving mean ages of 62±6 Ka (early uptake) and 80±6 Ka (linear uptake). New TL ages Two quartzites and three silcretes weighing 5–150 g (Table 3) were selected for dating by the TL technique (the methodology used is detailed in Tribolo, 2003 and Tribolo et al., ongoing). As was the case for the Klasies River Mouth samples, the five ages obtained for Blombos Cave samples are compatible at one sigma, allowing the calculation of a weighted average of 74±5 Ka. This TL mean age thus reinforces those obtained by ESR (linear model) and is in good agreement with the OSL mean age, associating the Still Bay levels with the end of isotopic stage 5a. The Still Bay content of Blombos Cave is at present the only one dated. Miller et al. (1999) proposed an age of 83 Ka for this techno-complex at Apollo 11, but its presence in the deposit is still debated (Henshilwood, pers. comm.).

Table 3 Thermoluminescence dates for the Still Bay levels at Blombos cave.

BZB

83 ± 8

410

BB23

G5c

CAB

47 ± 3

BB20

E7b

CAC

77 ± 6

BB15

E6a

CCh1

BB12

E6a

CCh1

504

187

Age

E6b

µGy/a µGy/a

Annual dose rae

BB24

Alpha

µGy/a µGy/a

Beta

Gamma internal

µGy/a

S-alpha

Gamma External

Gy

Cosmic

Ed

Layer

Ref

Square

Dose rate

µGy/a

ka

1 031 ± 55

81 ± 10

16

45

5,6

373

507

9

45

6,4

122

22

705 ± 52

67 ± 7

489

19

45

5,4

328

118

999 ± 56

77 ± 8

45 ± 2

400

16

45

6,1

192

11

664 ± 42

68 ± 6

77 ± 6

387

38

45

6,4

331

129

929 ± 50

82 ± 8

Chronology of the Howieson’s Poort and Still Bay techno-complexes

Discussion Age and duration of Howieson’s Poort A comparison of the ages obtained for Howieson’s Poort at Border Cave and Klasies River Mouth Cave (Fig. 2) shows a non-negligible difference in the periods of occupation of these sites. Howieson’s Poort would have appeared much earlier at Border Cave, perhaps as early as 80 Ka, and would have lasted until 60 Ka; only from this period or slightly before would it have appeared at Klasies River Mouth and in other sites such as Rose Cottage or Boomplaas. This apparent difference is based on the three ages greater than 70 Ka obtained in Border Cave using the ESR method. Several explanations can be proposed for this shift. The first would be that the three ESR dates are overestimated. The results of Miller et al., using the AAR method (1989, 1992, 1993, 1999), giving an age of 69±7 Ka for the post-Howieson’s Poort phase of this site, by falling half-way between the extremes do not bring any determining elements on this issue. A second explanation would be that, contrary to what has been said to date, Howieson’s Poort was not a synchronous phenomenon but actually appeared in the Border Cave region first and then spread throughout South Africa a few millennia later. Finally, it must be noted that certain authors have questioned the typo-technological unity of what is termed Howieson’s Poort (Parkington, 1990), which could also explain age differences. From Still Bay to Howieson’s Poort Although it is still too early to generalise, Blombos Cave being the only securely dated site which contains the Still Bay industry, it seems that this techno-complex is older than Howieson’s Poort (apart from at Border Cave), as Henshilwood et al. (2001a; b) proposed. One is tempted to associate the Still Bay techno-complex with the transition from climatic stage 5a to 4 (73±4 Ka – Martinson et al., 1987), and the Howieson’s Poort with the transition from stage 4 to 3 (59±6 Ka – Martinson et al., 1987). Consequently, the period corresponding to stage 4 could have been marked by an arid phase with seaside dwellings being abandoned: recall in this connection the presence of a sterile dune in Blombos Cave covering the Still Bay levels, dated to between 65,8±2.8 Ka and 68,8±3 Ka. Furthermore, in Klasies River Mouth Cave, the practically sterile Rock Fall Member (RF), located under the Howieson’s Poort levels made of sand and rock fragments, would also indicate a phase of abandonment; it could therefore date from the same period. Research carried out on other sites and deposits of the south coast of the Cape Province should help to resolve this issue.

505

From Tools to Symbols

Climatic changes and interpretations of Howieson’s Poort While one might be tempted to associate the ages obtained for the different technocomplexes with climatic transitions, it is more difficult to establish a cause-and-effect relationship. Indeed, South Africa has a wide variety of biomes reacting a priori differently to global climatic evolution. However, we can presume that if Howieson’s Poort is close to 60 Ka, it appeared in the majority of sites during the climatic amelioration of the beginning of stage 3. At Klasies River Mouth, this hypothesis is supported by microfaunal analyses by Avery (1987), who observed among other things an increase in biodiversity, and those of J.F. Thackeray (1992) who calculated an increase in average temperature. This would again call into question interpretations that Howieson’s Poort is a result of palaeoenvironmental stress.

Conclusion Radiometric and chemical dating methods are often criticised because of their relatively poor precision. It is true that the uncertainties are generally between 7 and 10 per cent and thus represent 5–7 Ka for an age of 70 Ka, leading to an overlap of the dates. In addition the reliability of certain parameters is often questioned (evolution of the uranium and water content of sediments for TL, OSL and ESR, or evolution of the average temperature for AAR). Therefore, the cross-dating of different materials is essential and must be developed. Despite the difficulties mentioned above, radiometric data have enabled researchers to identify the periods during which the Howieson’s Poort and Still Bay technocomplexes were produced, as well as to ask new questions. Among others, the TL dates on burnt stones presented in this article reinforce the hypothesis of a Still Bay industry older than Howieson’s Poort, and throw the Howieson’s Poort Synchronism and Klasies Regression Hypotheses back into question. Finally, the chronological framework within which Howieson’s Poort and Still Bay have been placed, even though still relatively incomplete and imprecise, allows us to date vestiges indicating the behavioural evolution of anatomically modern human: thus, engraved ochres of more than 70 Ka associated with the Still Bay levels of Blombos are changing our ideas as to how old symbolism is and perhaps also as to its origin (Henshilwood et al., 2002). As to dating Howieson’s Poort in South Africa, it is bringing a new element to the chronology of the appearance of backed tools found also in the structures of Mumba Rock Shelter in Tanzania (McBrearty & Brooks, 2000) and of Twin River and Kalambo Falls in Zambia (Barham, 2002). To be able to refine this chronology in future will allow us to understand better how this type of behaviour could have appeared and diffused.

506

Chronology of the Howieson’s Poort and Still Bay techno-complexes

Acknowledgements We would like to thank the various researchers who helped us to complete this study: H.J. Deacon, C.S. Henshilwood, J.E. Parkington, J.-P. Rigaud, J.C. Sealy, S. Woodborne and R. Yates. We would also like to thank J.-L. Joron, J.-L. Reyss and M. Sélo for their analyses. We also extend our warmest thanks to F. d’Errico and L. Backwell for inviting us to attend the Johannesburg conference. This work was funded by the CEA through a research training contract, and by the CNRS through the OHLL program.

References Aitken, M.J. (1985). Thermoluminescence dating. Studies in archaeological science. London: Academic Press.. Ambrose, S.H. & Lorenz, K.G. (1990). Social and ecological models for the Middle Stone Age in Southern Africa. In (P. Mellars, Ed.) The Emergence of Modern Humans: An Archaeological Perspective, pp. 3–33. Edinburgh: Edinburgh University Press. Avery, D.M. (1982). Micromammals as paleoenvironmental indicators and interpretation of the late Quaternary in the southern Cape Province, South Africa. Annals of the South African Museum 85, 183–374. Avery, D.M. (1987). Late Pleistocene coastal environment of the Southern Cape Province of South Africa: micromammals from Klasies River Mouth. Journal of Archaeological Science 14, 405–421. Avery, D.M. (1992). The environment of early modern humans at Border Cave, South Africa: micromammalian evidence. Palaeogeography, Palaeoclimatology, Palaeoecology 91, 77–87. Avery, D.M. (1995). Physical environment and site choice in South Africa. Journal of Archaeological Science 22(3), 343–353. Avery, G., Cruz-Uribe, K., Goldberg, P., Grine, F.E., Klein, R.G., Lenardi, M.J., Marean, C.W., Rink, W.J., Schwarcz, H.P., Thackeray, A.I. & Wilsond, M.L. (1997). The 1992–93 excavations at Die Kelders Middle and Later Stone Age cave site, South Africa. Journal of Field Archaeology 24, 263–291. Barham, L. (2002). Backed tools in Middle Pleistocene central Africa and their evolutionnary significance. Journal of Human Evolution 43, 585–603. Beaumont, P.B., de Villiers, H. & Vogel, J.C. (1978). Modern man in sub-Saharan Africa prior to 49000 BP: a review and evaluation with particular reference to Border Cave. South African Journal of Science 74, 409–419. Brooks, A.S., Hare, P.E. & Kokis, J.E. (1993). Age of anatomically modern human fossils from the cave of Klasies River Mouth, South Africa. Carnegie Institution of Washington, Year Book 92, pp. 95–96.

507

From Tools to Symbols Butzer, K.W. (1978). Climate patterns in an un-glaciated continent. Geographical Magazine, 201–208. Butzer, K.W. (1982). Geomorphology and sediment stratigraphy. In (R. Singer & J. Wymer, Eds) The Middle Stone Age at Klasies River Mouth in South Africa, pp. 33–42. Chicago: Chicago University Press. Butzer, K.W. (1984). Archaeology and quaternary environment in the interior of southern Africa. In (R.G. Klein, Ed.) Southern African Prehistory and Paleoenvironment, pp. 1–64. Rotterdam: Balkema. Carrion, J.S., Brink, J.S., Scott, L., & Binneman, J.N.F. (2000). Palynology and palaeoenvironment of Pleistocene hyena coprolites from an open-air site at Oyster Bay, Eastern Cape coast, South Africa. South African Journal of Science 96, 449–453. Deacon, H.J. (1989). Late Pleistocene paleoecology and archaeology in the southern Cape, South Africa. In (P. Mellars & S. Stringer, Eds) The Human Revolution, pp. 547–564. Edinburgh: Edinburgh University Press. Deacon, H.J. (1993). Southern Africa and modern human origins. In (A.J. Aitken, C.B. Stringer & P.A. Mellars, Eds) The Origin of Modern Humans and the Impact of Chronometric Dating: A Discussion, pp. 104–117. Princeton: Princeton University Press. Deacon, H.J. (1995). Two late Pleistocene-Holocene archaeological depositories from the southern Cape, South Africa. South African Archaeological Bulletin 50, 121–131. Deacon, H.J., Deacon, J., Scholtz, A., Thackeray, J.F., Brink, J.S. & Vogel, J.C. (1984). Correlation of paleoenvironmental data from the late Pleistocene and Holocene deposits at Boomplaas Cave, southern Cape. In (J.C. Vogel, Ed) Late Cainozoic Paleoclimates of the Southern Hemisphere, pp. 339–351. Rotterdam: Balkema. Deacon, H.J. & Geleijnse, V.B. (1988). The stratigraphy and sedimentology of the main site sequence, Klasies River, South Africa. South African Archaeological Bulletin 43, 5–14. Deacon, H.J., Geleijnse, V.B., Thackeray, A.I., Thackeray, J.F., Tusenius, M.L. & Vogel, J.C. (1986). Late Pleistocene cave deposits in the southern Cape: current research at Klasies River. Palaeoecology of Africa 17, 31–37. Deacon, H.J., Talma, A.S. & Vogel, J.C. (1988). Biological and cultural development of Pleistocene people in an Old World southern continent. In (J.R. Prescott, Ed.) Early Man in the Southern Hemisphere (Supplement to Archaeometry: Australian Studies), pp. S23-S31. Adelaide: University of Adelaide. Deacon, H.J. & Wurz, S. (1996). Klasies River Mouth main site, cave 2: a Howiesons Poort occurence. In (G. Pwiti & R. Soper, Eds) Aspects of African Archaeology, pp. 213–218. 10th congress of the Pan African Association for Prehistory and Related Studies. Evans, U. (1994). Hollow Rock Shelter, a Middle Stone Age site in the Cederberg. South African Field Archaeology 3, 63–73.

508

Chronology of the Howieson’s Poort and Still Bay techno-complexes Feathers, J.K. (2002). Luminescence dating in less than ideal conditions: case studies from Klasies River main site and Duinefontein, South Africa. Journal of Archaeological Science 29, 177–194. Grün, R. & Beaumont, P.B. (2001). Border Cave revisited: a revised ESR chronology. Journal of Human Evolution 40, 467–482. Grün, R., Beaumont, P.B. & Stringer, C.B. (1990a). Electron spin resonance dating evidence for early modern humans at Border Cave in South Africa. Nature 344(6266), 537–539. Grün, R., Shackleton, N.J. & Deacon, H.J. (1990b). Electron spin resonance dating of tooth enamel from Klasies River Mouth cave. Current Anthropology 31(4), 427–432. Grün, R., Beaumont, P., Tobias, P.V. & Eggins, S. (2003). On the age of Border Cave 5 human mandibule. Journal of Human Evolution 45, 155–167. Hendey, Q.B. & Volman, T.P. (1986). Last interglacial sea levels and coastal caves in the Cape Province, South Africa. Quaternary Research 25, 189–198. Henshilwood, C. & Sealy, J. (1997). Bone artefacts from the Middle Stone Age at Blombos Cave, Southern Cape, South Africa. Current Anthropology 38(5), 890–895. Henshilwood, C.S., d’Errico, F., Marean, C.W., Milo, R.G. & Yates, R. (2001a). An early bone tool industry from the Middle Stone Age at Blombos Cave, South Africa: implications for the origins of modern human behaviour, symbolism and language. Journal of Human Evolution 41, 631–678. Henshilwood, C.S., Sealy, J., Yates, R., Cruz-Uribe, K., Goldberg, P., Grine, F.E., Klein, R.G., Poggenpoel, C.A., van Niekerk, K. & Watts, I. (2001b). Blombos Cave, Southern Cape, South Africa: preliminary report on the 1992–1999 excavations of the Middle Stone Age levels. Journal of Archaeological Science 28, 421–448. Henshilwood, C.S., d’Errico, F., Yates, R., Jacobs, Z., Tribolo, C., Duller, G.A.T., Mercier, N., Sealy, J.C., Valladas, H., Watts, I. & Wintle, A.G. (2002). Emergence of modern human behavior: Middle Stone Age engravings from South Africa. Science 295, 1278. Jacobs, Z. (2004). Development of luminescence techniques for dating Middle Stone Age sites in South Africa. Unpublished Ph D. thesis, University of Wales, Aberystwyth. Jacobs, Z., Wintle, A.G. & Duller, G.A.T. (2003a). Optical dating of dune sand from Blombos Cave, South Africa: I-multiple grain data. Journal of Human Evolution 44(5), 599–612. Jacobs, Z., Wintle, A.G. & Duller, G.A.T. (2003b). Optical dating of dune sand from Blombos Cave, South Africa: II-single grain data. Journal of Human Evolution 44(5), 613–625 . Jones, H.L. (2001). Electron spin resonance dating of tooth enamel at three palaeolithic sites. Unpublished M.Sc. thesis, McMaster University. Joron, J.-L. (1974). Contribution à l’analyse des éléments en traces dans les roches et les minéraux par activation neutronique. Application à la caractérisation d’objets archéologiques. Unpublished PhD thesis, Université Paris Sud.

509

From Tools to Symbols Klein, R.G. (1974). Environment and subsistence of prehistoric man in the Southern Cape Province, South Africa. World Archaeology 5, 249–283. Klein, R.G. (1976). The mammalian fauna of the Klasies River Mouth sites, Southern Cape Province, South Africa. South African Archaeological Bulletin 31, 75–98. Klein, R.G. (1977). The mammalian fauna from the Middle and Later Stone Age (Later Pleistocene) levels of Border Cave, Natal Province, South Africa. South African Archaeological Bulletin 32, 14–27. Klein, R.G. (1999). The Human Career: Human Biological and Cultural Origins (2nd Ed.). Chicago: University of Chicago Press. Martinson, D.G., Pisias, N.G., Hays, J.D., Imbrie, J., Moore, T.C. & Shackelton, N.J. (1987). Age dating and the orbital theory of the Ice Ages: development of a high-resolution 0 to 300000years chronostratigraphy. Quaternary Research 27, 1–29. McBrearty, S. & Brooks, A. (2000). The revolution that wasn’t: a new interpretation of the origin of modern human behavior. Journal of Human Evolution 39, 453–563. Miller, G.H., Beaumont, P.B., Jull, A.J.T. & Johnson, B. (1992). Pleistocene geochronology and the paleothermometry from protein diagenesis in ostrich eggshells: implication for the evolution of modern humans. In (M.J. Aitken, C.B. Stringer & P.A. Mellars, Eds) The Origin of Modern Humans and the Impact of Chronometric Dating, pp. 149–157. Philosophical transactions of the Royal Society of London; series B. Miller, G.H., Beaumont, P.B., Jull, A.J.T. & Johnson, B. (1993). Pleistocene geochronology and the paleothermometry from protein diagenesis in ostrich eggshells: implication for the evolution of modern humans. In (M.J. Aitken, C.B. Stringer & P.A. Mellars, Eds) The Origin of Modern Humans and the Impact of Chronometric Dating, pp. 49–68. Princeton: Princeton University Press. Miller, G.H., Beaumont, P.B., Deacon, H.J., Brooks, A.S., Hare, P.E. & Jull, A.J.T. (1999). Earliest modern humans in southern Africa dated by isoleucine epimerization in ostrich eggshell. Quaternary Science Review 18, 1537–1548. Parkington, J.E. (1990). A critique on the consensus view on the age of the Howieson’s Poort assemblages in South Africa. In (P. Mellars, Ed.) The Emergence of Modern Humans: An Archaeological Perspective, pp. 34–55. Edinburgh: Edinburgh University Press. Rigaud, J.-P., Texier, J.-P. & Texier, P.-J. (2000). Eland/Elan, rapport scientifique sur les travaux effectués en 1999. Unpublished internal report. Shackelton, N.J. (1982). Stratigraphy and chronology of the Klasies River Mouth deposits: oxygen isotope evidence. In (R. Singer & J. Wymer, Eds) The Middle Stone Age at Klasies River Mouth in South Africa, pp. 194–199. Chicago: Chicago University Press. Singer, R. & Wymer J. (1982). The Middle Stone Age at Klasies River Mouth in South Africa. Chicago: Chicago University Press.

510

Chronology of the Howieson’s Poort and Still Bay techno-complexes Thackeray, A.I. (1989). Changing fashions in the Middle Stone Age: the stone artefact sequence from Klasies River main site, South Africa. African Archaeological Review 7, 33–57. Thackeray, A.I. (1992). The Middle Stone Age south of the Limpopo River. Journal of World Prehistory 6(4), 385–440. Thackeray, J.F. (1987). Late quaternary environmental changes inferred from small mammalian fauna, Southern Africa. Climatic Change 10, 285–305. Thackeray, J.F. (1992). Chronology of late Pleistocene deposits associated with Homo sapiens at Klasies River Mouth, South Africa. In (K. Heine, Ed.) Palaeoecology of Africa and the Surrounding Islands, 23, 177-191. IXth biennal conference of the Southern African Society for Quaternary Research, University of Durban. Thackeray, J.F. & Avery, D.M. (1990). A comparison between temperature indices for late Pleistocene sequences at Klasies River and Border Cave, South Africa. In (K. Heine, Ed.) Palaeoecology of Africa and the Surrounding Islands, pp. 311–315. IXth biennal conference of the Southern African Society for Quaternary Research, University of Durban. Tribolo, C. (2003). Apport des méthodes de la luminescence à la chronologie de technofaciès du Middle Stone Age associés aux premiers Hommes Modernes du sud de l’Afrique. Unpublished PhD thesis, Université Bordeaux I. Tribolo, C., Mercier, N. & Valladas, H. (2003). Attempt at using the single-aliquot regenerativedose procedure for the determination of equivalent doses of Upper Palaeolithic burnt stones. Quaternary Science Review 22, 1251–1256. Valladas, H. (1992). Thermoluminescence dating of flint. Quaternary Science Reviews 11, 1–5. Van Andel, T.H. (1989). Late Pleistocene sea levels and the human exploitation of the shore and shelf of southern South Africa. Journal of Field Archaeology 16(2), 133–155. Vogel, J.C. (2000). Radiometric dates for the Middle Stone Age in South Africa. In (P.V. Tobias, M.A. Raath, J. Moggi-Cecchi & G.A. Doyle, Eds) Humanity – From African Naissance to Coming Millennia. Florence: Florence University Press and Johannesburg: Witwatersrand University Press. Vogel, J.C. & Beaumont P.B. (1972). Revised radiocarbon chronology for the Later Stone Age in South Africa. Nature 237, 50–51. Vogel, J.C., Wintle A.G. & Woodborne, S.M. (1999). Focus: luminescence dating of coastal sands: overcoming changes in environmental dose rate. Journal of Archaeological Science 26, 729–733. Wadley, L. (1997). Rose Cottage Cave: Archaeological work 1987 to 1997. South African Journal of Science 93, 439–444. Wurz, S. (2000). The Middle Stone Age sequence at Klasies River, South Africa. Unpublished PhD thesis, University of Stellenbosch.

511

Subsistence strategies in the Middle Stone Age at Sibudu Cave: the microscopic evidence from stone tool residues Bonny S. Williamson School of Geography, Archaeology and Environmental Studies, University of the Witwatersrand, Private Bag 3, Johannesburg, 2050, South Africa

Abstract Microscopic residue analysis was conducted on a number of stone tools from Sibudu Cave to obtain more information on specific use of certain tool types at the site. Plant residues were preserved better than residues of animal origin. The presence of ochre as a use-related residue and use-wear traces were well represented. The soils were found to be slightly acidic, a factor that may have contributed favourably to the exceptional preservation of organic remains at the site. Although culturally modern behaviour has not been specifically inferred from this data, the potential exists to answer questions relating to patterned tool use and specific activity areas within the site.

Résumé Une analyse de résidus microscopiques présents sur un certain nombre d’outils en pierre provenant de la Grotte de Sibudu a été réalisée en vue d’obtenir des informations sur la fonction de certains types d’outils. Les résidus végétaux sont mieux conservés que ceux d’origine animale. Les traces d’utilisation sont bien conservées sur le pièces analysées et souvent associées à des traces d’ocre Nous avons constaté que les sols étaient légèrement acides, facteur qui pourrait avoir contribué favorablement à la conservation exceptionnelle de restes organiques sur ce site. Même si la question de l’émergence de la modernité culturelle n’est pas directement abordée à partir de ces données, celles-ci recèlent néanmoins des informations pertinentes pour répondre à des questions en relation avec cette problématique telles que la fonction des outils et l’existence de zones d’activités spécialisées au sein du site.

512

Subsistence strategies in the Middle Stone Age at Sibudu Cave

Introduction Residue analysis has been conducted on stone tools from a number of southern African sites to date (Lombard, 2001; Tomlinson, 2001; Williamson, 1997) as well as sites elsewhere in the world (Fullagar et al., 1997; Hardy et al., 2001; Tuross & Dillehay, 1995; Loy & Dixon, 1998). The applicability of persistent residues in determining site use and subsistence strategies of prehistoric people is finding wider acceptance. Residue types identified in this study included animal tissues, blood, hair, collagen and plant tissue remains, cellulosic fibres, resins and exudates, starch grains and starchy residues, as well as mineral deposits like ochre. Sibudu Cave has a long occupation sequence with an interesting and complex stratigraphy (Wadley, 2001a) that provides a relatively closed environment ideal for the preservation of microscopic organic residues. The association of the residues with the last use-event of the tool can be confidently established due to the careful excavation and handling methods and an examination of the associated soils from the deposit. The residues that are attributed to tool use differ in appearance from those that are interpreted as resulting from incidental contact with the source material. The tools submitted for residue analysis were carefully excavated and post-depositional contamination was kept to a minimum. Organic remains as evidence of human diet is elusive at most archaeological sites and stone tool residue analysis provides us with at least some information about subsistence practices in prehistory. It may not always be possible to directly infer human agency on other organic remains that are found at a site, but the identification of plant or animal remains on a stone tool surface indicates that those substances were at least processed by the occupants of the site. Even at sites where organic preservation is poor, microscopic organic remains can still be found on the lithic artefact surfaces because the supporting medium for the residue (i.e. the stone tool) is not porous and the residues often dry relatively quickly and frequently in thin films. This allows little time for bacterial degradation and putrefaction, as one would find with masses of fresh tissue (plant or animal). These factors make the study of stone tools invaluable for inferring modern human behaviour as it pertains to (for example) patterned tool use, spatial use of the living area of the site for specific activities, changes in food procurement strategies and symbolic vs utilitarian uses of colouring materials (e.g. ochre and manganese). The usefulness of the information gained from a microscopic analysis of stone tools is, therefore, dependent largely on the questions being asked by the palaeoanthropologist or archaeologist.

513

From Tools to Symbols

The artefacts In Sibudu cave, 412 stone tools were analysed from various squares and levels (Wadley, 2001a). The surfaces of the tools were examined at magnifications ranging from 50× to 800× under incident light. Cross-polarised light was used to identify residues such as starch grains, and to distinguish between cellulose and collagen. The Hemastix® test (Williamson, 2000) was used to identify blood films and to distinguish between blood residues and resin deposits. Detailed sketches were made of the positions of residues on the formal retouched and the other tools. The relative percentages for each residue type on the tools are given in Table 1. The percentages do not add up to 100 because more than one residue type is often found on a single tool. Points include unifacial and bifacial points. Hollow-based points are included in the bifacial category, which accounts for 25 per cent of the assemblage. Included in the scraper category are sidescrapers, endscrapers, end-and-sidescrapers and straight scrapers. (Straight scrapers are otherwise known as ‘knives’ and comprise 20 per cent of the tools analysed from Sibudu Cave.) The ‘other’ retouched category comprises mainly broken retouched tools, but also includes miscellaneous retouched tools and adzes. This category makes up 19 per cent of the tools analysed. Flakes (27 per cent) usually make up a large proportion of the debitage, or waste, of an assemblage. Endstruck, sidestruck, end-and-sidestruck, convergent and broken flakes are all included here. Chunks make up 9 per cent of the analysed assemblage and are also often considered to be part of the debitage from stone tool manufacture.

Plant residues Starch grains in the range 1–6 µm in diameter were found on 36 per cent of all tool types. The occurrence of fibrous plant material is low (9 per cent) but plant tissue occurs on 26 per cent of all tool types. Thus, not much fibrous plant material was being processed. In all instances, scrapers have the highest percentages of all kinds of plant residues. In Table 1 it can be seen that the ‘other’ retouch and flake categories have the lowest percentages of plant tissue residues (17 per cent and 18 per cent respectively) compared to points, scrapers and chunks of which at least 32 per cent had plant tissue residues. Plant tissue residues may result from the processing of plant foods, plant-derived artefacts (hunting implements, digging sticks, bedding and fuel for fires), plant material used for medicinal or ritual purposes, and sometimes the hafting of a stone component into a wooden haft (Loy & Dixon, 1998). The plant species from which the residue derived cannot be established at this point but such information may become available in the future.

514

Subsistence strategies in the Middle Stone Age at Sibudu Cave

Animal residues Animal residues on the tools amounted to < 1 per cent, with collagen (combined incidence of collagen fibres (five tools) and sheet collagen (six tools)) occurring on about 3 per cent of the tools (Table 1). Collagen was found on most tools that had other kinds of animal residue. Red blood cells were found on only a single quartz chunk in the whole assemblage. Blood films were found on only two retouched artefacts, both of which also had traces of animal tissue (Fig. 1) and collagen. Collagen fibres derive from the connective tissue of sinews and tendons and may result from the disarticulation or primary butchery of a carcass. Sheet collagen, on the other hand, often comes from the periosteum (Davis, 1987) of wet bone that has been cut or scraped, or from the removal of the subcutaneous layer of a hide during scraping. The sheet-like fragments are usually thin and flat but can also appear rumpled. Collagen is distinguishable from cellulose because it does not display the characteristic of birefringence under crosspolarised light that cellulose does (Williamson, 1997).

Figure 1 A tissue residue of animal origin with blood found on a broken retouched hornfels artefact that was 30 mm long and excavated from level BBYA. The image was taken at 100× magnification. The artefact had worn edges and traces of ochre.

515

From Tools to Symbols Table 1 Overall percentages of residue types for general tool categories.

Points

Scrapers

‘Other’ retouch

Flakes

Chunks

Total

Sample size

104

83

77

113

35

412

Percentage of total

25

20

19

27

9

Residue type: plant tissue

32

34

17

18

34

plant fibre

9

16

5

7

9

9

starch

42

47

45

29

37

36

white starchy

29

29

22

14

17

23

resin

13

36

16

22

23

19

exudate

15

22

8

4

3

11

3

0,24

blood film

1

1

3

9

1,70

animal tissue

1

collagen fibres

1

collagen sheets

2

Tool type

red blood cells

bone hair

1 1

26

0,48 2

2

3

1

1

1

3

1,21 1,46 0,73

1

0,24

ochre - u/h/i*

14

8

16

7

6

11

charcoal

7

2

9

4

14

6

ash

3

worn edges

59

52

61

3 34

3

1 48

polish

28

42

36

12

9

26

scratches

14

16

10

6

3

11

utilisation

4

6

9

8

6

7

re-used

7

4

19

2

hafted

24

39

21

12

6

20

used no res**

13

17

17

16

6

15

mineral crystals

13

27

12

19

11

17

4

4

5

1

vivianite/MnO2 mycohyphae

9

1

rootlets

8

2

no residues

14

3

3

16

27

*u = use-related; h = possible hafting evidence; i = incidental contact ** no use-related residues found on the tools

516

7

2 6

4

31

20

4

Subsistence strategies in the Middle Stone Age at Sibudu Cave

Significantly, chunks had the highest incidence of animal processing residues. The artefacts could have been carried back from a kill site or the carcasses could have been processed in the cave. Of greater significance is the fact that animal-related residues occur in such low frequencies in proportion to plant processing residues.

Ochre Points and ‘other’ retouch had much higher percentages of the use-related ochre residues than the other categories of tools. It has been suggested (Velo, 1984) that ochre was sometimes used to cure hides and scrapers are thought to have been used to scrape hides. The variety of other residues on the scrapers and the high incidence of polishes and worn edges on the scrapers means that they may well have been used for some scraping purposes but that this possibly did not involve only hide-working. Recent investigations of ethnographic stone scrapers from the Konso region in southwestern Ethiopia showed that blood and collagen do persist as residues on a stone scraper even when very dry hides were processed. Scrapers also have the highest percentage for tools with putative signs of hafting. Hafting traces are difficult to characterise unequivocally as they pertain to the distribution and patterning of use-residues on a tool in relation to the butt, bulb of percussion or retouched edges (Hardy et al., 2001; Wadley et al., 2004). In the case of chunks, none of these features are discernable, but a sharp workable edge or the distributions of the residues may suggest that the artefact was hafted. Points and ‘other’ retouch have comparable proportions of hafted elements (24 per cent and 21 per cent) with the figures for flakes and chunks being considerably lower. These figures may be an indication of the investment of time and energy in making and retouching the scrapers, points and ‘other’ retouch, as opposed to the expedient use of flakes and chunks. Conversely, it could be a function of the fact that hafting traces are easier to identify on retouched artefacts.

Use-wear traces Use-wear or microwear analysis as understood by most archaeologists (Vaughn, 1985) was not conducted in this study. The presence of polishes and striations visible on uncleaned tools under incident dark field light was recorded as an additional clue to the duration and direction of tool use. This was not a comprehensive microwear study in the strict sense of the term. Almost half of the 412 tools examined had worn edges (48 per cent) and a quarter (26 per cent) had polished edges or ridges (Fig. 2). Points had the highest incidence of worn edges (61 per cent) while scrapers had the highest proportion of polishes (42 per cent). The presence of polish on scrapers in particular corresponds with the

517

From Tools to Symbols

high proportion of plant residues observed for this tool category. Plants very often leave ‘silica polish’ (Fullagar, 1991) on stone tool surfaces. The phenomenon can be investigated further with microwear analysis. Many of the artefacts displaying worn edges and polishes had no other residues (15 per cent of the entire assemblage). This is significant because it shows that use-related residues can be absent even when other evidence suggests that an artefact was used. The absence of residues may result from any of a number of causes, namely prehistoric curation, depositional conditions or post-excavation handling. The identification of microscopic use-wear on stone tools in the absence of contact residues emphasises the importance of the maxim that an absence of evidence does not constitute evidence of absence. The tools were used even though no use-residues persist. Of equal importance, however, is the fact that half the assemblage had edges that were not worn or abraded and that the dulling of the edges of an artefact cannot be attributed exclusively to abrasion while in the deposit. Post-excavational removal of the residues was minimised with careful handling of the artefacts. Other signs of utilisation such as edge chipping were not common and amounted to only 7 per cent, with scratches, either on the tool surface itself or in the residue matrix, amounting to 11 per cent.

Figure 2 Polish and traces of ochre on the edge of an end-scraper from level SPCA made from dolerite. The image was taken at 50× magnification.

518

Subsistence strategies in the Middle Stone Age at Sibudu Cave

Post-breakage use Seventeen per cent of all the broken lithics were not used at all and had no useresidues or use-wear traces, while 21 per cent of all broken artefacts had been used after breakage (Table 2). It was noticed during microscopy that some of the residues and much of the use-wear and edge damage occurred along the broken edges of the tools. Curation and re-use of broken tools by prehistoric people shows expedient use of the sharp broken edges. The retouched tools with ochre have the highest incidence of re-use (9 of the 15 broken tools). This feature was also noticed by Hardy et al. (2001) on tools from the Crimean Palaeolithic. If the presence of ochre is indicative of hafting, as suggested elsewhere (Wadley et al., 2004), then the confirmed use of a hafted tool after breakage seems justified. Table 2 Artefact types divided into those with vs those without ochre, shown against the number of broken tools within each category and the frequency of those that had been used, those that had not been used at all and those that had been used after breakage. (None of the chunks were classified as broken.) Tool type

Total

Broken

% Broken

Not used at all

Used after breakage

Points with ochre

29

9

31



3

Points without ochre

75

32

43

5

4

Scrapers with ochre

30

2

7



1

Scrapers without ochre

53

1

2





‘Other’ Retouch with ochre

23

15

65



9

‘Other’ Retouch without ochre

54

37

69

8

5

Flakes with ochre

26

5

19



1

Flakes without ochre

87

16

18

7

1

412

117

28

17%

21%

Total and %

Soil analysis Soil samples from twelve levels (arranged in descending order) at Sibudu were analysed microscopically for soil organics and the presence of starch grains (Table 3). An average sample of 7.1 grams was removed from a larger soil sample taken during the field season. The samples were placed individually in sample vials; 20 ml of distilled deionised water was added to each; the vials were tightly capped, vigorously shaken and left to stand overnight. A Hemastix® test was conducted on each to determine the presence of mineral ions in the soil that could cause false positive results with the residues on the tools (Williamson, 2000). In all cases the results were negative. A pH reading for each soil sample was taken directly in the vial with an electronic pH meter at 20° C. The results show an increasing tendency towards acidity as the depth

519

From Tools to Symbols

of the deposit increases. The organic preservation at Sibudu is remarkably good, with plant remains and seeds being found in the lowest levels (Tuross & Dillehay, 1985). The slight acidity of the soil may have contributed to this. Foth and Turk (1984) state that ‘nitrifying organisms become inhibited when the pH is less than 5.5’. Courty et al. (1989) maintain that a ‘low pH increases the potential for preservation of pollen in the soil, although acid conditions are not favourable for mollusc or bone preservation’. Table 3 Analysis of soils from Sibudu for the presence of starch and soil organics. #

Level

1

OMOD

C3a

7,6

7,64

No starch grains, plant tissue and fibre fragments

2

ORE

D3b

7,3

7,37

No starch grains, lots of clear crystal grains

3

RSP

B4b

7,6

6,41

Few starch grains (1–2 µm)

4

BSP

D4c

7,2

7,10

Few starch grains (1–2 µm), some plant fragments

5

CH-SPCA

C5c

6,4

6,40

No starch grains

6

GR-SPCA

C5c

6,6

5,90

No starch grains, some carbonised plant tissue fragments

7

SPS

B5d

6,6

5,53

No starch grains

8

MEY

B5c

7,2

5,56

No starch grains

9

MUS

B6b

7,4

5,79

No starch grains

10

CHOC2

C5c

7,1

5,49

No starch grains

11

BGM2

B5c

7,2

5,45

Few starch grains (1–2 µm), abundant plant fragments and carbonised plant tissue

12

YA1

B5a

6,6

5,50

No starch grains

7,1

6,18

Average

Sqr Mass/g

pH

Starch grains and other organics

A micropipette was used to ‘puff’ up and remove 20 µl of the fine fragment from the settled soil sample. This smaller sample was placed on a microscope slide and allowed to air dry before being examined under incident light at various magnifications and lighting conditions. Hydrating the sample overnight allows organics and starch grains in the soils to float to the surface of the settled sample where they can be removed and examined easily (Loy, pers. comm.) This was not a quantitative assay of the starch concentration in the soil. Small ‘transient’ starch grains of 1–2 µm were found in three of the twelve soil samples. These were not the lowest levels, which suggests that there was no settling and no detectable movement of the starch grains within the deposit (Therin, 1998). Artefacts with and without starch grains were found in all these levels, suggesting that their presence on the artefacts can be interpreted as resulting from the use of the tool to process plant materials. Small starch grains in this size range are treated circumspectly

520

Subsistence strategies in the Middle Stone Age at Sibudu Cave

because they are regarded as transient, meaning either that they are short-lived within the plant or that they are not found in abundance in the food storage organs of most plants (Courty et al., 1989). The same applies to fragments of plant tissue and plant fibres; not all the artefacts from levels that contained fragmented plant tissues had plant tissue remains on them.

Contamination Contaminating residues on the artefacts are those that are post-depositional or post-excavational and do not relate to use of the tool at all. Fortunately, these kinds of deposits on the tool surfaces are readily distinguishable from use-related residues. It has been found that they may vary from site to site, and should be carefully characterised for each site. The presence of mycohyphae (fungal structures (Hufford, 1979)) and rootlets on the tools is very low, indicating that biological activity in the soil is minimal and the deposit is relatively undisturbed. Many of the Sibudu artefacts had a crystalline mineral deposit on their surfaces. This did not appear to interfere with the detection and interpretation of use-residues in any way, but was recorded on the original data sheets. The site formation processes that account for the crystalline mineral deposits are still to be explained. Another type of mineral deposit seen on the tools is manganese dioxide (MnO2) staining (Shahack-Gross et al., 1997), which has a thin metallic blueblack appearance and can be very reflective. Post-excavational contaminants include synthetic fibres from clothing, lipids or ‘finger-grease’ from handling, ink or other labels used on the tools and scratches that can be attributed to the use of metal trowels. There were relatively few tools displaying any of these traces, indicating that the excavation techniques at Sibudu Cave were sufficiently cautious. Artefacts destined for residue analysis should be stored individually, unwashed and unlabelled, in clean dry plastic sample bags.

Discussion The activities practised at Sibudu were specific and centred on the processing of plant foods that were generally starchy and/or succulent. Tools that had signs of use but no use residues suggest that some tools may have been cleaned after being used and then discarded, suggesting that the intention was to re-use them at a later date. If this were not the case then one would have expected the tools to be discarded with the last use-residue still adhering to the surface. Many broken tools were used after being broken, suggesting expedient use of sharp edges created by the breakage, or an economic consideration with respect to the availability of raw material. Broken artefacts were not re-sharpened before being used again. The use of ochre was present on tools

521

From Tools to Symbols

from all categories, although points and ‘other’ retouch had the highest incidents of ochre residues, which implies hafting. Limited butchery and animal processing was carried out at the site, mainly with chunks that may have been used for dismemberment or primary butchery of carcasses. The fact that some animal residues were found means that poor preservation cannot account solely for the dearth of animal residues on the tools. Formal tool categories were not the only tool types with residues. Use-residues can be distinguished from other contaminating residues on the Sibudu artefacts and there was no doubt that the residues interpreted as relating to specific activities at the site are authentic. Soil analysis showed that starch grains and fragmented plant remains in the soil did not mirror the types of residues found on the tools. Residues can show that tool categories do not relate to actual tool use. Categories are useful for detecting change though time and for comparing assemblages (Wadley, 2001b). Just as the presence of artefacts in themselves cannot be taken as symbolic (Wadley, 2001b), so too, the presence of ochre cannot be interpreted as symbolic in itself. The incidence of ochre on the tools may have a more mundane, functional explanation. The positioning of the ochre residues on the tools suggests that it constituted part of or facilitated the hafting of the tool. The excavation of Sibudu Cave provides a tight framework within which the use of a defined space can be assessed through time. Differential use of specific tool types can be investigated and the subsistence strategies of the occupants of the cave can be discerned to an extent. The occupants of Sibudu Cave processed plant foods and plant materials extensively throughout the occupation of the cave, and animal processing within the cave was minimal – although preserved bone is fairly abundant. Microscopic residue analysis provides us with a magnified window into the diet and practices of the Middle Stone Age people living next to the Tongaat River in South Africa. Residues are imperative for recovering plant remains and plant exploitation practices that would otherwise be lost in the archaeological record. Detailed analysis of the surfaces of artefacts takes us at least one step further away from pure conjecture and a step closer to actual evidence of specific tool function and subsistence practices, and from there to possible indicators of cultural modernity.

Acknowledgements This research was supported by the NRF, the University of the Witwatersrand and the ACACIA Programme. Lyn Wadley is thanked for her ongoing assistance. I wish to thank Lucinda Backwell and Francesco d’Errico for inviting me to present this paper at the International Symposium ‘From Tools to Symbols’.

522

Subsistence strategies in the Middle Stone Age at Sibudu Cave

References Courty, M.A., Goldberg, P. & Macphail, R. (1989). Soils and Micromorphology in Archaeology. Cambridge: Cambridge University Press, p. 20. Davis, S.J.M. (1987). The Archaeology of Animals. London: B.T. Batsford. Foth, H.D. & Turk, L.M. (1984). Fundamentals of Soil Science, 7th Ed. New York: John Wiley and Sons, p. 186. Fullagar, R., Furby, J. & Hardy, B. (1996). Residues on stone artifacts: state of a scientific art. Antiquity 70, 740–745. Fullagar, R. (1991). The role of silica in polish formation. Journal of Archaeological Science 18, 1–24. Hardy, B.L., Kay, M., Marks, A.E. & Monigal, K. (2001). Stone tool function at the palaeolothic sites of Starosele and Buran Kaya III, Crimea: behavioral implications. Proceedings of the National Academy of Sciences 98(19), 10972–10977. Hufford, T.L. (1979). Botany: basic concepts in plant biology. New York: Harper Row Publishers, p. 93. Lombard, M. (2001). The lithic rings of Honingklip, Goergap and Windsorton: A functional interpretation of a type of bored stone. BA Honours Project, University of the Witwatersrand. Loy, T.H. & Dixon, E.J. (1998). Blood residues on points from eastern Beringia. American Antiquity 63, 21–46. Shahack-Gross, R., Bar-Yosef, O. & Weiner, S. (1997). Black-coloured bones in Hayonim Cave, Israel: differentiating between burning and oxide staining. Journal of Archaeological Science 24, 439–446. Therin, M. (1998). The movement of starch grains in sediments. In (R. Fullagar, Ed.) A Closer Look: Recent Australian Studies of Stone Tools. Sydney University Archaeological Methods Series 6, University of Sydney, Australia, pp. 61–72. Tomlinson, N.E. (2001). Residue analysis of segments, backed and obliquely backed blades from the Howiesons Poort layers of Rose Cottage Cave, South Africa. MA project, University of the Witwatersrand. Tuross, N. & Dillehay, T. (1995). Mechanism of organic preservation at Monte Verde and one use of biomolecules in archaeological interpretation. Journal of Field Archaeology 22, 97–110. Vaughn, P.C. (1985). Use-wear analysis of flaked stone tools. University of Arizona Press, USA. Velo, J. (1984). Ochre as medicine: A suggestion for the interpretation of the archaeological record. Current Anthropolgy 25, 674. Wadley, L. (2001a). Preliminary report on excavations at Sibudu Cave, KwaZulu-Natal. Southern African Humanities 13, 1–17. Wadley, L. (2001b). What is cultural modernity? A general view and a South African perspective from Rose Cottage cave. Cambridge Archaeological Journal 11: 201–221.

523

From Tools to Symbols Wadley, L., Williamson, B.S. & Lombard, M. (2004). Ochre in hafting material in Middle Stone Age South Africa: a practical role. Antiquity 78 (301): 661–675. Williamson, B.S. (1997). Down the microscope and beyond: microscopy and molecular studies of stone tool residues and bone samples from Rose Cottage Cave. South African Journal of Science 93, 458–464. Williamson, B.S. (2001). Direct testing of rock painting pigments for traces of haemoglobin at Rose Cottage Cave, South Africa. Journal of Archaeological Science 27, 755–762.

524

Speaking with beads: The evolutionary significance of personal ornaments Marian Vanhaeren CNRS UMR 7041 ArScAn, Ethnologie préhistorique, 21 allée de l’Université, F-92023 Nanterre, France.

Abstract Personal ornaments have come to play an important role in the debate on the origin of behavioural modernity and the evolution of our ancestors’ cognitive abilities. All authors agree in categorising beads as one of the hallmarks of cultural modernity, i.e. the cultural traits that underlie societies similar to ours. Divergent opinions exist, however, as to the dating of the first evidence of bead use and the taxonomic status of the first bead makers. Recent data seem to suggest that the most ancient personal ornaments were not put on in Europe but rather in Africa or in the Near East, and they can therefore no longer be seen as tied to the Aurignacian. It seems equally established in Europe that not only anatomically modern humans but also late Neanderthals produced and used such objects. Undoubtedly these discoveries have important implications for the question regarding the unique or multiple emergence(s) of symbolic thought and its association with one or more human types. But beads may offer more. In traditional societies they play at least fourteen different and often multiple roles (e.g. they may be used to beautify the body, function as ‘love letters’ in courtship, or as amulets, exchange media, expressions of individual and group identity, markers of age, class, gender, wealth or social status) which offer varied and rich information on the individuals, social groups, and societies that used them. This paper focuses on the variety of functions that personal ornaments have in human societies and on the methods we may use to understand the role beadwork played in the earliest symbolic cultures. Application of such methods to a review of the earliest evidence for bead use from Europe, the Near East, Australia and Africa suggests that the motives that stimulated the creation of beadwork traditions in the different areas were different. The main function of the earliest African beadworks seems to be that of circulating in an exchange system to reinforce reciprocity networks ensuring the survival of hunter-gatherer groups in times of stress. In Europe beads seem rather to have been used to strengthen affiliation to a group and visualise social and individual roles within the group.

525

From Tools to Symbols

Résumé La parure occupe une place de plus en plus importante dans le débat sur l’origine de la pensée symbolique et l’évolution des capacités cognitives de nos ancêtres. Pour tous les auteurs, la parure est un des halos de la modernité culturelle, c’est à dire de l’ensemble de caractères sur lesquels se fondent des sociétés comparables aux nôtres. Les avis diffèrent toutefois quant à la datation des premiers témoignages de son utilisation et au statut taxinomique de l’auteur ou des auteurs des premiers objets de parure. De nouvelles données semblent indiquer que les premiers objets de parure n’ont pas été produits en Europe mais plutôt en Afrique ou au ProcheOrient et ne sont pas, de ce fait, à rattacher à l’Aurignacien. Il semble également établi qu’en Europe, les derniers Néandertaliens produisaient et utilisaient de tels objets. Il est certain que ces découvertes ont des implications sur la question d’une origine unique ou multiple de la pensée symbolique et sur son association avec un ou plusieurs types humains. Mais les parures pourraient bien nous offrir plus. Dans les sociétés traditionnelles, les objets de parure remplissent au moins quatorze fonctions différentes et souvent multiples (par ex. expression esthétique, «messages d’amour», amulettes, objets d’échange, marqueurs d’appartenance à un groupe ethnique, une classe d’âge, un genre, un statut individuel, objets de richesse) offrant des informations riches et variées sur les individus, sur leur statut et sur leur société. Cet article portera sur la variété des fonctions de la parure dans les sociétés humaines et sur les méthodes qui pourraient permettre de comprendre le rôle de ces objets dans les plus anciennes cultures symboliques. L’application de telles méthodes à un bilan des plus anciens témoignages d’utilisation d’objets de parure en Europe, au Proche Orient, en Australie et en Afrique suggère que les raisons qui ont stimulé la création de traditions des ornements personnels ont été différentes selon les régions. La fonction principale des plus anciennes parures africaines semble celle de circuler dans une logique du don pour consolider des réseaux d’entre aide garantissant la survie des groupes de chasseurs-cueilleurs dans des conditions précaires. En Europe, les parures semblent plutôt être utilisées pour renforcer l’affiliation à un groupe et visualiser des rôles sociaux et individuels à l’intérieur du groupe.

Introduction Personal ornaments are a polysemantic component of the archaeological record. Due to their inherent symbolic meaning, they represent an unambiguous hallmark of behavioural modernity which is frequently used to suggest the modern character of a Palaeolithic culture (Chase & Dibble, 1987; Mellars, 1989, 1996; Klein, 1989, 1992, 1995; Chase, 1991; Ambrose, 1998; Shennan, 1989, 2001; Lindley & Clark, 1990; Hahn, 1992; Bar-Yosef, 1992, 2001; Davidson & Noble, 1992; Thackeray, 1992; Taborin, 1993; Hayden, 1993; Stringer & Gamble, 1994; White, 1995, 2000; Bednarik, 1995, 1997; Noble & Davidson, 1996; Mithen, 1996; d’Errico et al., 1998, 2003; Zilhão & d’Errico, 1999; McBrearty & Brooks, 2000; Kuhn et al. 2001; Wadley, 2001; Conard & Bolus, 2003; d’Errico, 2003; Shea, 2003). Due to the multiple roles they play in human societies, personal ornaments found at archaeological sites are used to address issues such as ethnicity, social organisation, and to understand the place of the individual in

526

Speaking with beads

prehistoric societies (Bar-Yosef Mayer, 1989, 1991; Newell et al., 1990; Taborin, 1993; Fritz & Simmonet, 1996; d’Errico & Vanhaeren, 2000, 2002; Vanhaeren & d’Errico, 2001, 2003a–c; Vanhaeren, 2002). Items of trade par excellence, ornaments may also substantially contribute to the identification of exchange networks and home ranges (Fisher, 1876, 1896; Rivière, 1887, 1904a, b; Jackson, 1917; Cordier, 1956; Bosinski & Hahn, 1973; Reid & Wilson, 1981; Gramble, 1982; Masson, 1982; Bahn, 1982; Torti, 1983; Sacchi 1986; Taborin, 1993, 1996; Floss, 2000; Alvarez Fernandez, 2001, 2002; Vanhaeren et al., 2004). However, archaeologists hold differing views as to the dating of the first evidence for bead use, the taxonomic status of the first bead makers, and the place or places where this behaviour would have arisen. On the other hand, no comprehensive methodology has been proposed to identify the main function played by beads in a given prehistoric society or to evaluate the implications of bead exchange networks for cultural hybridisation. This paper focuses on the variety of functions that personal ornaments have in human societies and on the methods we may use to understand the role beadworks played in the earliest symbolic cultures. Identification of such methods and the review of the earliest evidence for bead use from Europe, the Near East, Australia, and Africa will allow us to explore the mechanism(s) that triggered our ancestors to become symbolic, and propose hypotheses on the role personal ornamentation may have played in the creation and maintenance of early symbolic traditions.

The function of personal ornaments in traditional societies A recent study, based on a survey of the ethnographic literature (Vanhaeren, 2002) has revealed that in hunter-gatherer and small-scale societies personal ornaments may have fourteen different functions and suggested their analysis may be of use to study prehistoric beads. We review below these functions, discuss their social implications, and provide actual ethnographic examples. Aesthetical expression and self-assertion According to many scholars, the primary function of bead use is that of beautifying the human body (Dubin, 1987; Price, 1991; Coles & Budwig, 1998). It has been proposed that this function would be linked to a deep aesthetic drive common to all humans (Sciama, 1998). Unfolding the history of beads around the globe, Dubin (1987) notes that beads are universal and must therefore express a fundamental human need. Price (1991) suggests that humans would use personal ornaments to distinguish themselves from the animal world. According to Coles and Budwig (1998), individuals would find in personal ornaments a means to assess their ego which would

527

From Tools to Symbols

provide them with a deep sense of self-satisfaction. Some psychiatrists explain the predilection for round beads by a subconscious memory of the sight of the mother’s eyeball (Erikson, 1969) or the touch of her nipples (Bross, in Dubin, 1987), these being the first senses engaging a feeling of security and pleasure to the child. A possible interest for body aesthetics has been observed by several primatologists among chimpanzees and orang-utans living in captivity and in the wild (Lethmate & Dücker, 1973; Wrangham et al., 1994; de Waal, 1996). Lethmate and Dücker (1973) have described a female orang-utan from the München Zoo (Germany) which, confronted with a mirror, used it for observing herself and taking care of her looks by cleaning her fur of food crumbs and paint spots previously applied without her knowing. Using the same mirror, this female has also been seen placing leaves and other plant elements on her head and other body parts. Similar behaviours are reported from wild orang-utans (McKinnon, 1971) and chimpanzees (Wrangham et al., 1994). In sum, some apes seem to consciously modify their appearance by covering themselves with different kinds of objects. These actions are, however, momentary, rare, and do not seem to have any tangible social implication for simian societies. No consensus exists among primatologists on the most likely explanation for these behaviours. Some regard them as a form of self-decoration (Köhler, 1921; Reynolds & Reynolds, 1965; Harrisson, 1961; Lethmate & Dücker, 1973), others as a game (Schenkel, 1964; Schaller, 1965; Jantschke, 1972). Courtship Beadworks have been interpreted as an adaptation developed by our species to attract members of the opposite sex (Price, 1991; Miller, 2001). Samburu men from Kenya consider desirable women adorned with necklaces that entirely cover their necks (Cole, 1975). Unmarried warriors from quite a few African societies invest an important amount of time in the manufacture of their beadworks to impress young females (Fisher, 1984). This hypothesis finds support in the fact that we are not the only species using artificial means for enhancing our chances of reproduction (Miller, 2001). Male bowerbirds from Australia and New Guinea represent one of the better known and striking examples of this. Males attract females by collecting branches and building bowers that may reach three metres in height and present bilateral or radial symmetry (Borgia, 1987). Some bowerbird species also ‘paint’ the interior of their constructions with a mixture of berries, bark, charcoal, clay and saliva. Others ‘decorate’ the entrance with colourful and shiny objects such as flowers, fruits, feathers, snails, ribbons, bottle-caps, pieces of glass, and even plastic items and batteries. To create a visual impact, these objects are heaped up and sorted by colour. The quality of the bower is directly related to the male’s reproductive success. Bad builders attract no females, while the best may copulate with up to ten a day.

528

Speaking with beads

Male competitors were observed to steal from their neighbours the items, apparently the most rare and difficult to obtain, that seemed to guarantee their reproductive success. Given that after copulation the female bowerbird leaves to build a simple nest, lay her eggs and raise her offspring on her own, the only function of the bower constructions seems that of attracting females. Local variations in the morphology and colours of the bowers attest to the existence of regional traditions. Ethnic marker Ethnographic studies (Kinietz, 1972; Heizer, 1978; Trigger, 1978) have shown that personal ornaments, such as body painting, scarification, tattooing, garment and headdress (Thompson, 1972; Reynolds, 1978; Burch, 1980), may be perceived by members of a society as powerful indicators of their ethnic identity, defined as the conscious and vindicated sense of belonging to a community (Jones, 1990). In such contexts, ornamentation enhances within-group cohesion and fixes boundaries with neighbouring groups. However, the link between personal ornamentation and ethnic identity is subtle as it may reflect connections of very different nature and strength (ideological, linguistic, religious, political, economical …) between individuals. It has been observed that ornaments are used to sustain a cultural heritage in periods of communal stress. In cases of forced population displacements, for example, people often pursue the use of traditional ornaments to preserve their cultural identity (Geary, 1994). If this identity is lost, as happened with the Uduk in Sudan, manufacture and use of characteristic beads are abandoned (Ichon, 1973; James, 1996). Social marker In traditional societies beads often reflect the affiliation of an individual to one or more social groups (Roach & Eicher, 1965; Strathern & Strathern, 1971; Faris, 1972; Kuper, 1973; Ray, 1975; Tainter, 1978; Corwell & Schwarz, 1979; Brain, 1979; Hodder, 1979, 1991; Turner, 1980; O’Shea, 1984; Wiessner, 1984; Carey, 1986, 1991; Dubin, 1987; Preston-Whyte, 1994; Sciama & Eicher, 1998). According to the society, such affiliations are determined by lineage (clan, moieties), wealth (aristocrats, slaves), gender (O’Hear, 1998; Meisch, 1998), age class (child, adolescent, mature, old age), biological (puberty, menopause) and relational (single, married, widow) stages. Married Ndebele women in South Africa, for example, wear distinctive aprons decorated with white beads (Powell, 1995). Among the Turkana in Kenya, children, adolescents and married women wear ostrich eggshell beads. For children, however, these are threaded on a string and worn as necklaces, belts, bracelets and anklets. For girls they are embroidered on triangular goatskins that become longer and longer as girls grow up. After marriage they are instead embroidered on rectangular aprons

529

From Tools to Symbols

(Dubin, 1987). Among the Kalabari in Nigeria, the use of particular beads identifies the members of a prestigious lineage and the loss of such beads is considered enough reason to be excluded from the family (Eicher, 1998). In Kenya, Masai are divided into two moieties, one composed of pastoralists living exclusively by rearing livestock, the other by farmers also practising agriculture. Each group wears distinct beadworks (Klump & Kratz, 1993). Individual marker Beadworks may characterise individuals that have acquired or inherited a unique social status (shaman, chief, king, pope, etc.). In most shamanistic societies shamans wear distinct every-day dresses and ornaments (James, 1988; Vitebsky, 1995). Kings ubiquitously use ornaments; take as an example the necklaces worn by the king of the Anuaks, Sudan (Seligman & Seligman, 1932), or the crown jewels of European monarchies, as tangible insignia of absolute power. Ritual objects Beads may play a role in rituals (Beckwith & Fisher, 1999). Ornaments may be used to identify the ceremony leaders and attendants. In South America some shamans identify themselves, when performing rituals, with a jaguar and wear on that occasion teeth and skins of this animal (Vitebsky, 1995). Among the Uduk of Sudan, for example, fathers wear female beads to celebrate a birth or the integration of an adopted child in the kin (James, 1988). Beads may also be used during rites of passage connected with birth, initiation, marriage, healing or death. In these cases they can be given to the individual involved for the time of the ceremony or permanently to mark his or her new social status. Tunisian women wear during their marriage ceremony a number of traditional outfits, one of them completely covered with gold, which are rented for the occasion (Demmerseman, 1998). Christian marriage-rings, by contrast, are a good example of personal ornaments symbolising a new social status that are kept by the individuals after a ceremony. Offerings Offering of beads to the gods, spirits and ancestors seems a fairly common practice. Its goal is to attract the goodwill of the worshipped entity or to acknowledge it for a received favour. Pendants in the form of body parts are suspended on the walls of Catholic churches and chapels, the Santa Rosalia sanctuary near Palermo in Sicily being one of the more famous examples. In Tibet, beads are ‘sowed’ by Buddhist priests as offerings to secure good harvests (Dubin, 1987).

530

Speaking with beads

Amulets and talismans Amulets (Budge & Wallis, 1968; Pavitt & Pavitt, 1970; Kunz, 1971; Evans, 1976) are meant to protect their wearer from particular misfortunes (illness, death, miscarriage, loss of beloved, etc.). Several amulets, each protecting against a different calamity, may be worn together. In Ancient Egypt amulets were granted great power. Complete necklaces composed of different amulets were supposed to protect owners during and after their earthly life (Dubin, 1987). The use of beads representing an eye to protect oneself from the ‘evil eye’ and keep away all sorts of misfortunes is well known in the Mediterranean region (Dundes, 1981). Unlike most personal ornaments, which are created for visual display, amulets may be worn hidden under clothes or disguised within ornaments having different functions. Talismans are items that secure prosperity. The Inuits made small ivory carvings of fishes, seals and whales to facilitate capture of these animals. Fertility talismans are common in African societies (Dubin, 1987). In southern Africa, little dolls adorned with beads are given to KwaZulu girls when they first menstruate to increase their reproductive capacity (Preston-Whyte, 1994). In the old harbour of Marseille, South of France, fishermen sell opercula of Turbo shells, called Lucy’s eyes, which are believed to bring luck to the wearer. Prophylactics When amulets fail to protect their wearers from illness, they may turn to prophylactic ornaments to recover (Dubin, 1987). In Nepal, each item of a healing necklace has its peculiar curative properties. In India, prophylactic properties are attributed to gold, symbol of the sun and the Ganges River, and amber is said to heal jaundice. The Lapidarium, a literary genre concerned with the symbolic virtues of prophylactic stones and minerals, was popular all over the Middle Ages as testified by the treaties of Isidorus from Sevillia, Marbodius, bishop of Rennes, and Albertus Magnus from Cologne. Even today the ‘power of gemstones’ has remained in vogue. Exchange media Beads are, with textiles, the elements of material culture that travel the longest distances (Curtin, 1984; Appadurai, 1986; Leroi-Gourhan, 1964). Due to their relative lightness, robustness and reduced size they have been excellent exchange media in all times. Ornaments can circulate in a gifts exchange network to reinforce social ties (Mauss, 1923–1924; Wiessner, 1990; Godelier, 1996; Marshall, 1998) or, when made of exotic material, may be pursued as prestige symbols (Ingold et al., 1988). Among the !Kung of southern Africa, the exchange of beadworks serves to maintain networks of mutual aid, essential to group survival (Wiessner, 1990; Marshall, 1998).

531

From Tools to Symbols

In a number of North American (Boas, 1897) and Pacific societies (Mauss, 1923–1924; Godelier, 1996) gifts are exchanged to acquire or maintain positions of power. Such gift exchange systems (potlach, big men or kula) have in common mechanisms of redistribution and/or destruction of large quantities of goods, including beadworks. Beads may also be used as currency to buy goods, acquire a wife, or to settle judicial affairs such as payment of compensation for robbery or murder (Testart, 2001). Colonial trade (Erikson, 1969; Ceci, 1982; Greaber, 1996; Sciama, 1998) utilised large quantities of ornaments – particularly cowrie shells and glass beads – to buy slaves and a variety of goods (gold, furs, ivory, spices). In Western societies, beads are still used as money in special contexts such as holiday centres where, as is the case of the Club Mediterranée, beads of different colour correspond to different values. Famous is the case of the New Orleans Mardi Gras during which necklaces are given to convince members of the same or opposite sex to display hidden parts of their body (Wilkie, 1998). Inalienable possessions Family or tribal inalienable possessions may include beadworks (Weiner, 1992; Sciama, 1998; Godelier, 1996). These treasures often have a sacred character as they symbolise the link with the ancestors and the group identity. Beadworks included in these possessions may never have been worn or are only displayed on particular occasions by designated members of the group. Erikson (1969) mentions cases in which beads are so old that it becomes impossible to trace back their origin. Among the Kelabit of Eastern Malaysia, beads as old as 2 000 years have been identified by Janowski (1998). Price (1991) and Francis (1992) note that whenever bead types become rare, they are removed from exchange networks to integrate inalienable possessions. Such unique items only leave family treasures to pay blood money in case of a murder. Communication systems Studies on writing systems reveal that beads have been used to create means by which to store and transmit information (Gelb, 1952). Zulu girls make bead necklaces for the boys they are in love with (Twala, 1951; Schoeman, 1983; Preston-Whyte, 1994). By varying motifs and colours, type of beads and quality of manufacture, these necklaces communicate complex messages that allow boys to understand the feelings of their potential partners. Wampum – necklaces or belts made of little white and purple perforated disks or tubes of Mercenaria mercenaria shells – were used in North America as mnemotechnical devices carried by messengers. Each belt concealed the story of a political or diplomatic event (discussion, meeting, treaty between tribes). Colours and motifs gave a general indication of their content, black for example meaning war, but their real message was learned and remembered by the messenger.

532

Speaking with beads

Counting devices Beads are part of counting devices. The Incas wore complex storing devices made of threads and knots, called quipu (Ifrah, 1994). Items are represented in this device by multicolour strings, quantities by the location and type of knots. Samoan men of New Guinea (Weyler, 1959) wear headbands with little hollow sticks, each representing one of their properties (woman, pig, shell, handaxe, …). Rosaries are used by believers of monotheistic religions (Catholicism, Greek Orthodox, Islam, Buddhism) to recite accurately from memory the correct number of prayers and incantations required by their faith (Dubin, 1987; Dransart, 1998).

From ethnographic to prehistoric beads How to infer the function of prehistoric beadworks from the personal ornaments recovered in the archaeological record? This might be possible if ethnographically documented functions were mutually exclusive and resulted in distinct long-lasting features that one could identify on archaeological specimens. Unfortunately, it is not that simple. A single beadwork may have multiple functions and apparently similar beadworks may have different functions according to the society in which they are used. However, cross-cultural analysis of bead use shows that when multiple functions are documented, one largely dominates over the others. Also, visual or material similarity between beadworks used for different functions by different human cultures fades out once information on production, context of use and disposal is taken into account. Contrary to what is generally believed, a substantial amount of this information is not lost in prehistoric beads and may be recovered through detailed analysis of archaeological specimens and thorough examination of their archaeological context. Taphonomic, zooarchaeological, technological and use-wear analyses of prehistoric beads, in combination with comparative analyses conducted on ethnographic objects, may create informed frameworks of inference which may help in identifying the more likely function of prehistoric beads (d’Errico & Vanhaeren, 2002; Vanhaeren & d’Errico, 2001, 2003a–c; Vanhaeren, 2002). Before turning to the earliest jewellery, I will briefly illustrate the potential of this approach presenting the results of the ongoing analysis of San ostrich egg-shell beads (OESBs) from the Fourie collection. The San ornaments that we have analysed were collected in the 1920s by Louis Fourie, a medical officer with a strong interest in social anthropology (Wanless, 1999). His passion led him to take 600 photographs of Khoisan people and collect more than 3 000 San objects, now kept at Museum Africa, Johannesburg. A set of photographs from his collection illustrates the different stages of the manufacture of OESBs (Fig. 1), i.e. a woman biting with her teeth into eggshell fragments to produce disk-shaped preforms, which she then drills with two borers,

533

From Tools to Symbols

Figure 1 Photos taken by Louis Fourie in the 1920s of a Khoisan woman making ostrich eggshell beads (top and middle), of the tools she used in the manufacturing process (bottom left) and a sample (bottom right) of ostrich eggshell, bead preforms and finished beads (Museum Africa, Johannesburg).

534

Speaking with beads

strings on a leather thong, regularises with a horn hammer, and eventually smoothes with a grinding stone. In addition, Fourie collected the tools involved and a woman’s bag with eggshell beads at different stages of manufacture (Fig. 2). The bag contains 555 ostrich eggshell items, more than half of which consist of preforms with traces of drilling on one aspect, a third of drilled unfinished beads, the remainder being composed of irregular blanks, disk-shaped preforms with no traces of drilling, and only two finished beads. Microscopic analysis of these two beads reveals use-wear suggesting they come from a broken beadwork. The bag also contains tiny fragments of ochre, responsible for the red patches observed on some of the beads, plant remains and two glass beads. Detailed technological and morphometric analysis of this and similar collections is particularly useful to define stages of manufacture, characterise personal and group variability, understand how beadwork manufacture is integrated in the individual’s life, and interpret archaeological accumulations of beads and preforms. Such data, combined with ethnographic information on the function of beadwork in San societies (Wiessner, 1984; Marshall 1998), are crucial when it comes, for example, to exploring the significance of morphologically similar beads found in Middle Stone Age (MSA) layers.

The earliest personal ornaments Europe The invention of personal ornaments has long been considered synonymous with the colonisation of Europe by anatomically modern populations bearing the Aurignacian culture (Klein, 1989, 1992, 1995; Davidson & Noble, 1992; White, 1993a, 1995; Stringer & Gamble, 1994; Noble & Davidson, 1996). Aurignacian beads dated between 37 Ky and 30 Ky BP are certainly numerous and diverse. A recent survey counted 153 bead types made of ivory, antler, bone, stone, teeth and marine shells from 93 sites (Vanhaeren, 2002; Vanhaeren & d’Errico, 2004). Evidence for bead use by Mousterian Neanderthals is scant and it has been shown that in a number of instances purported beads are the result of natural phenomena (d’Errico & Villa, 1997). It is now generally acknowledged, however, that the Aurignacian is not the first European culture that has produced personal ornaments. Personal ornaments are also reported from European sites attributed to other Early Upper Palaeolithic technocomplexes, stratigraphically underlying the Aurignacian (Châtelperronian, Uluzzian, Bachokirian, Szeletian, Ahmarian, Stzreletian). Human remains associated with the Châtelperronian suggest that Neanderthals were the authors of this and possibly other pre-Aurignacian technocomplexes.

535

From Tools to Symbols

Figure 2 A Khoisan woman’s bag with eggshell beads (top) at different stages of manufacture (middle) and tools used in ostrich eggshell bead manufacture (bottom) collected by Louis Fourie in the 1920s and curated in the Museum Africa, Johannesburg.

536

Speaking with beads

Among these (Fig. 3), the oldest bead evidence comes from the 43 Ky levels of the Bacho Kiro site, Bulgaria, where a perforated wolf canine and bear incisor were found associated with a lithic industry called Bachokirian and interpreted by the excavators as a possible precursor of the Aurignacian (Kozlowski, 1982, 2000). At Kostienki 14 (Markina gora), a Mediterranean shell with two holes has been recovered from a level yielding a Streletskian lithic assemblage and radiocarbon dates ranging between 32,6 and 36,5 Ky BP (Sinitsyn, 2003). A fish-tail ivory pendant has been found in the Szeletian layers of the Bryndzeny site, Moldavia (Ketraru, 1989; Kozlowski, 2000). Dentalium sp. shells come from the Uluzzian layers of Klisoura cave, Greece (Koumouzelis et al., 2001). The same shell species, as well as Natica sp., Trochus sp. and Glycymeris sp. shells, were recovered from the contemporaneous site of Grotta del Cavallo, South of Italy (Palma di Cesnola, 1993). The Uluzzian layers of Castelcivita, in the south of Italy, yielded a Pecten sp. shell (Palma di Cesnola, 1993). In France, a varied collection of perforated or gouged beads is reported from the Châtelperronian layers of Grotte du Renne, in the Yonne region. It comprises eight fox canines, four bovid incisors, three reindeer incisors, two bear incisors, two marmot incisors, one red deer canine, five bone pendants, three ivory beads and two fossil belemnites (Leroi-Gourhan & Leroi-Gourhan, 1965; d’Errico et al., 1999; White, 2000). Perforated wolf, fox and red deer canines were also found in the Châtelperronian layers of Quinçay Cave (Granger & Lévêque, 1998), a perforated fox canine at the eponymous site of Châtelperron (White, 2000). Bovid incisors and an ivory ring come from the contemporary layers at Roche au Loup (White, 2000), a bear incisor and a Pecten sp. shell from the Trilobite Cave (Taborin, 1993), a Turitella sp. shell from Cauna de Belvis Cave (Taborin, 1993). Dentalium sp. shells were apparently found at Saint-Césaire (Lévêque, in d’Errico et al., 1998), and a carnivore canine, identified as a lynx canine, at Roc de Combe (Sonneville-Bordes, 2002). Near East Personal ornaments may have appeared earlier in the Levant than in Europe (Fig. 3). Four Glycymeris sp. shells with perforations on the umbo (Taborin, 2003) come from Mousterian layers of Qafzeh Cave, Israel, dated by TL between c. 90 and 100 Ky (Valladas et al., 1998). Brought to the site by the inhabitants of the cave, identified as modern humans on the basis of the burials found in the same layers, these shells may represent the oldest known beads. However, no anthropic traces were detected on the perforations, which we know may occur naturally on the umbo of bivalves (d’Errico et al., 1993). Also, the large size of the shells and the presence of pigment on one specimen indicate they may have been used as ochre containers rather than as ornaments (Taborin, 2003).

537

From Tools to Symbols

Figure 3 Earliest personal ornaments from Europe: (a) Bacho Kiro; (b) Kostienki 14 (Markina gora); (c) Bryndzeny; (d) Grotte du Renne (Arcy-sur-Cure); (e) Quincay; (f) Roc de Combe, the Near East; (g) Qafzeh; (h) Uçagizli; (i) Ksar Akil; (j) Blombos Cave; (k) Zombepata; (l) Border Cave; (m) Enkapune ya Muto; (n) Loyangalani.

538

Speaking with beads

Forty-three perforated marine shells, most of which belong to the species Nassarius gibbosula, were found at Uçagizli, South of Turkey (Kuhn et al., 2001) in layers dated to 41 400 ± 1 100 BP (AA37625). They are associated with a lithic assemblage attributed to the Ahmarian, an Upper Palaeolithic technocomplex spread in the East of the Mediterranean that stratigraphically underlies the Aurignacian. In Lebanon, at Ksar’Akil (Mellars & Tixier, 1989), 243 shell beads (146 Nassarius gibbosulus, 22 Columbella rustica and 26 other marine gastropods as well as 48 Glycymeris sp. and one other marine bivalve) are reported from layers that have yielded lithic assemblages similar to those found at Uçagizli (Kuhn et al., 2001) and stratigraphically lying between layers dated to 43 750 ± 1 500 BP and 32 Ky BP. As in Europe, the author of this transitional industry is uncertain. The cast of a lost infant skull from the Ahmarian layers of the Ksar’Akil site, Lebanon, bears modern features. The dating and archaeological context of these remains, however, is uncertain (Bergman & Stringer, 1989). Five Levantine Aurignacian sites, contemporaneous with those from Europe, have also yielded personal ornaments. These consist of perforated animal teeth – mostly fox, wolf and red deer canines – and perforated shells. The latter belong to the same species as those used as beads by the Ahmarian inhabitants of the region (Vanhaeren & d’Errico, 2002). Africa The African evidence for early use of personal ornaments has long been underestimated. The uncertain chronology of a number of sites that have yielded potentially old ornaments is probably the main reason for this. We now know that the oldest securely dated beads come from Blombos Cave, South Africa, where 41 Nassarius kraussianus shell beads bearing human-made perforations are found associated with a Still Bay assemblage dated by OSL and TL at c. 75 Ky BP (Henshilwood et al., 2004; d’Errico et al., 2005). Personal ornaments are reported from seven other South and East Africa MSA sites. A complete and an unfinished ostrich eggshell bead (OESB) were found in the OLP member of Boomplaas, South Africa (Deacon, 1995). Three 14C dates, one U/Th, and one AAR are available for this member. Two 14C dates on charcoal give an age of > 40 Ky BP (UW 305) (Fairhall et al., 1976) and 37 400 ± 1 370 (Pta-1811) (Vogel, 2000) respectively; the third 14C date, obtained from a speleothem, provides an age of 31 680 ± 550 (Pta-2302) (Vogel, 2000). The U/Th dating, also on speleothem, gives 35 200 ± 2 600 (U-366) (Vogel, 2000). Dating by amino acid racemisation on an ostrich eggshell fragment gives an age of 44 000 ± 4 000 (Miller et al., 1999). A fragment of a stone ring made of micaceous schist comes from an MSA layer of the Zombepata

539

From Tools to Symbols

cave, Zimbabwe (Cooke, 1971). The weight (28 grams) and small size of the fragment, which constitutes a third of the original object, contradict its use as a weight on a digging stick and rather suggest it was worn as an ornament. The 14C date for this layer, 40 720 ± 1 620 BP (SR 190) (Cooke, 1971), should probably be considered as a minimum age. It is noteworthy that a second fragment from the same site was found in an undated deeper layer. OESBs and OESB preforms come from undated MSA contexts at Bushman Rock Shelter (Plug, 1982), and Cave of Hearths (Mason, 1962, 1993; Mason et al., 1988). OESBs were also found at Border Cave in ‘Early LSA’ layers (Beaumont et al., 1978). The twenty-four 14C dates from these layers range between 33 Ky and 39,8 Ky. These dates are consistent with those obtained for the same layers by ESR (Grün & Beaumont, 2001). At the Kenyan site of Enkapune Ya Muto, thirteen complete ostrich eggshell beads, twelve perforated OESB preforms, and 593 ostrich eggshell fragments were found in a stratigraphical unit (DBL1) containing an early Later Stone Age (LSA) lithic assemblage (Ambrose, 1998). The 14C date obtained from one fragment, 39 900 ± 1 600 BP (Pta4889 F2), consistent with those from the enclosing archaeological layers, convincingly supports the view of an East African bead-working tradition dating back to at least 40 Ky BP. Two other sites in East Africa, Kisese II (Inskeep, 1962) and Mumba (Mehlman, 1989, 1991; Brooks & Robertshaw, 1990), have yielded OESBs associated with transitional MSA/LSA lithic industries. Radiocarbon dates on an ostrich eggshell from the former site has given an age of 31 480 BP (Deacon, 1995). Direct AMS 14C dates on beads from the latter site range between 29 Ky and 33 Ky (Conard, this volume); 14C determinations on bone and snails from the same layer provide dates ranging between 29 570 ± 1 400 and > 37 Ky BP (Mehlman, 1989, 1991). An older age for this layer (46,6–65,6 Ky) is suggested by U/Th dating of bone (Mehlman, 1989, 1991) and AAR (52 Ky) on OESB (Hare et al., 1993). Two OESBs have also recently been reported from the site of Loiyangalani (Thompson et al., 2004) in Tanzania. The associated MSA industry is tentatively attributed by the authors to a period ranging between 50 and 120 Kya. In North Africa beads are reported from four Aterian sites, a bone pendant from Grotte Zouhra, Morocco (Debénath, 1994), four perforated quartzite flakes from Seggédim, Nigeria (Tillet, 1978; Debénath, 1994), and a perforated Nassarius gibbosula from Oued Djebanna in Algeria (Morel, 1974). The anthropogenic nature of perforations on these objects, however, remains to be verified.

Conclusion The earliest evidence for bead use dates back to 90–75 Kya, is found in Africa and the Near East, and consists in all cases of perforated marine shells. However, while

540

Speaking with beads

the anthropogenic origin of Blombos beads is demonstrated by detailed taphonomic, technological and use-wear analysis of the archaeological material, information on the possible shell beads from other sites is still too scant to accept this evidence as confirmed. Anatomically modern humans seem to be the authors of these cultural remains at Blombos and Qafzeh. Dating the ostrich eggshell beads and stone rings found at a number of southern and eastern Africa MSA and MSA/early LSA sites is problematic. Most sites lack radiometric determinations or were dated long ago with conventional 14C and provided infinite dates or dates close to the limit of the method. Furthermore, when dates obtained with other methods are available, they often differ from the 14C dates and from each other. At present, the most parsimonious interpretation of this evidence is that ostrich eggshell beads were produced in sub-Saharan Africa, certainly by modern humans, since at least 40 Ky and possibly 10–20 Ky earlier. Evidence for the use of personal ornament in Europe before 40 Ky is scant or absent. After this date ornaments appear associated with techno-complexes attributed to Neanderthals, to modern humans and of unknown authorship. Contrary to African and Near Eastern early instances, Early Upper Palaeolithic beadworks consist, among both Neanderthals and moderns, of a variety of types, produced with different raw materials and techniques. Discovery of carnivore teeth with unfinished perforations in the Châtelperronian levels of Roc de Combe (Sonneville-Bordes, 2002) and byproducts of ivory rings and decorated bone tubes in the same cultural layers of Grotte du Renne (d’Errico et al., 1998), suggests Châtelperronian ornaments were produced by Neanderthals rather than obtained from modern humans. This is confirmed by the fact that some Châtelperrronian bead types are not found in the Aurignacian. One site in Western Australia yielded twenty-two Conus sp. shell beads dated to c. 32 000 BP (Morse, 1993), but in many areas of the globe (West Africa, India, China, Far East Asia) ornaments are not reported from sites older than 20 Ky BP and may be relatively rare or absent until 10 Ky BP. Future research will establish whether this absence is due to lack of data or of long-lasting bead-making traditions. Concerning the function of personal ornaments in early symbolic societies, a remarkable difference appears between the African and the European record. Small disk-shaped ostrich eggshell beads are virtually the only bead type found at late MSA and early LSA sites over an area covering a large part of sub-Saharan Africa. Their production persists apparently unchanged in this area until historical times. In contrast, dozens if not hundreds of different bead types characterise the Early Upper Palaeolithic of Europe and the Near East, from its very beginning. Ongoing analysis of the geographical distribution of 153 Aurignacian bead types found at 93 archaeological sites identifies a clear regional trend in the bead type used, as well as evidence for trade

541

From Tools to Symbols

of shell beads over relatively long distances (Vanhaeren & d’Errico, 2004; d’Errico & Vanhaeren, 2004). Cross-cultural survey of personal ornament use in traditional societies suggests the main, though certainly not exclusive, function of MSA beads was, as for known San societies, that of exchange media used in gift-giving systems, the role of which is to strengthen networks of social and economic relationship (Wiessner, 1982, 1986; Ambrose, 1998). Three properties of MSA OESBs support this view. They are highly standardised, which warrants their exchange value. The raw material used was arguably available all over the area occupied by these groups. This prevents creation of unbalanced exchange networks typical of cases in which exotic prized items are involved. Manufacture requires a significant investment of time and labour without requiring specialised craftsmanship, which makes their production feasible for everybody. The Early Upper Palaeolithic ornaments best fit instead an interpretation as integrated markers of ethnic, social and personal identity. Their ethnic dimension is suggested by regional patterns in bead type association not explained by raw material availability. Use as social markers may be indicated by recurrent occurrence, within a region, of different types found in significantly different proportions. Presence of unique bead types may reflect use of ornaments by an individual to highlight his or her particular social role. Thus, the reasons that have stimulated the creation of beadwork traditions in the two areas, and probably in other areas of the globe, seem to have been different. In Africa beads had the function of reinforcing reciprocity networks to ensure the survival of hunter-gatherer groups in times of stress. Beads were used in Europe to strengthen affiliation to a group and to manifest social and individual roles within the group.

Acknowledgements I thank the organisers of the conference, Francesco d’Errico and Lucinda Backwell, for having provided such a wonderful, rich, stimulating and cordial environment for discussion and for their useful comments, help with finding references, and efficient editing of the present and earlier drafts of this paper. I am especially grateful to Francesco d’Errico for the many discussions on the earliest beads and help with correcting and clarifying the ideas presented here. My gratitude also goes to Ann Wanless for giving access to the Fourie collection and interesting discussions. This research would not have been possible without a postdoctoral grant given by the French CNRS and two travel grants given by the French Ministry of Research (Aires Culturelles) and the University of Bordeaux 1. Research conducted in this paper has been funded by the Origine de l’Homme, du Language et des Langues program of the French CNRS, the Origin of Man, Language and Languages programme of the European Science Foundation and the ACI Espaces et Territoires programme of the French Ministry of Research.

542

Speaking with beads

References Alvarez Fernandez, E. (1999). Las perlas de Madera fosil del terciaro y los objetos de adornocolgantes sobre dientes de zorro y ciervo del magdaleniense de Gönnersdorf y de AndernachMartinsberg – 2 (Neuwied, Rheinland Pfalz, Alemania). Zephyrus 52, 79–106. Alvarez Fernandez, E. (2001). L’axe Rhin-Rhône au Paléolithique supérieur récent: l’exemple des mollusques utilisés comme objets de parure. L’Anthropologie 105, 547–564. Ambrose, S.H. (1998). Chronology of the Later Stone Age and Food Production in East Africa. Journal of Archaeological Science 25, 377–92. Appadurai, A. (Ed.) (1986). The Social Life of Things: Commodities in Cultural Perspective. Cambridge: Cambridge University Press. Bahn, P. (1982). Inter-site and inter-regional links during the Upper Palaeolithic: the Pyrenean evidence. Oxford Journal of Archaeology 1, 247–268. Bar-Yosef, O. (1992). The role of Western Asia in modern human origins. Philosophical Transactions of the Royal Society B 337, 193–200. Bar-Yosef, O. (2001). The Middle and Early Upper Paleolithic in southwest Asia and neighboring regions. In (O. Bar-Yosef & D. Pilbeam, Eds) The Geography of Neandertals and Modern Humans in Europe and the Greater Mediterranean, pp. 107–156. Peabody Museum Bulletin: Peabody Museum of Archaeology and Ethnology. Massachusetts: Harvard University. Bar-Yosef Mayer, D.E. (1991). Changes in the selection of marine shells during the transition from the Natufian to the Neolithic. In (F. Valla & O. Bar-Yosef, Eds) The Natufian Culture in the Levant, pp.629–636. International Monographs in Prehistory. Ann Arbor. Bar-Yosef Mayer, D.E. (1989). Late Paleolithic and Neolithic Marine shells in the southern Levant as cultural markers. In (C.F. Hayes & L. Ceci, Eds) Proceedings of the 1986 Shell Bead Conference, pp. 169–174. Rochester Museum and Science Center Research Records 20. Beaumont, P.B., de Villiers, H. & Vogel, J.C. (1978). Modern man in Sub-Saharan Africa prior to 49,000 years B.P.: a review and evaluation with particular reference to Border Cave. South African Journal of Science 74, 409–419. Beckwith, C. & Fisher, A. (1999). African Ceremonies. New York: Harry N. Abrams Inc. Publishers. Bednarik, R.G. (1995). Towards a better understanding of the origins of body decoration. L’Anthropologie 33, 201–212. Bednarik, R.G. (1997). The role of Pleistocene beads in documenting hominid cognition. Rock Art Research 14, 27–41. Bergman, C.A. & Stringer, C.B. (1989). Fifty years after: Egbert an Early Upper Palaeolithic juvenile from Ksar Akil, Lebanon. Paléorient 15(2), 99–111. Boas, F. (1897/1966). Kwakiutl Ethnography. Chicago: Chicago University Press. . Borgia, G. (1987). Sexual selection in bowerbirds. Scientific American 254(6), 92–100. Bosinski, G. & Hahn, J. (1973). Der Magdalenian Fundplatz, Martinsberg. Rheinische Ausgrabungen 11, 81–266.

543

From Tools to Symbols Brain, R. (1979). The Decorated Body. London: Hutchinson. Brooks, A.S. & Robertshaw, P.T. (1990). The Glacial Maximum in tropical Africa: 22,000 to 12,000 BP. In (C. Gamble & O. Soffer, Eds) The World at 18,000 BP, Vol. 2: Low Latitudes, pp. 121–169. London: Unwin Hyman. Budge, E. & Wallis, A. (1968). Amulets and Talismans. New York: Collier Books. Burch, E.S. (1980). Traditional Eskimo societies in Northwest Alaska. In (Y. Kotani & W.B. Workman, Eds) Alaska Native Culture and History, pp. 253–304. Senri Ethnological Studies 4. Carey, M. (1986). Beads and Beadwork of East and South Africa. Aylesbury: Shrire. Carey, M. (1991). Beads and Beadwork of West and Central Africa. Aylesbury: Shrire. Ceci, L. (1982). The value of wampum among the New York Iroquois: a case study in artefact analysis. Journal of Anthropological Research 38(1), 97–107. Chase, P. & Dibble, H. (1987). Middle Paleolithic symbolism: a review of current evidence and interpretations. Journal of Anthropological Archaeology 6, 263–296. Chase, P. (1991). Symbols and Paleolithic artefacts: style, standardization, and the importance of arbitrary form. Journal of Anthropological Archaeology 10, 193–214. Cole, H.M. (1975). Artistic and communicative values of beads in Kenya and Ghana. The Bead Journal 1(3), 29–37. Coles, J. & Budwig, R. (1998). Le monde des perles. Paris: Flammarion. Conard, N.J. & Bolus, M. (2003). Radiocarbon dating and the appearance of modern humans and timing of cultural innovations in Europe: new results and new challenges. Journal of Human Evolution 44, 332–371. Cooke, C.K. (1971). Excavations in Zombepata Cave, Sipolilo District, Mashonaland, Rhodesia. South African Archaeological Bulletin 26, 104–126. Cordier, G. (1956). Les coquilles des faluns de Touraine ont-elles été colportées en Dordogne à l’Age du Renne? Bulletin de la Société d’Etudes et de Recherches Préhistoriques des Eyzies 6, 39–55. Corwell, J. & Schwarz, R.A. (Eds.) (1979). The Fabrics of Culture: The Anthropology of Clothing and Adornment. Paris: Mouton. Curtin, P.D. (1984). Cross-Cultural Trade in World History. Cambridge: Cambridge University Press. d’Errico, F. (2003). The invisible frontier: a multiple species model for the origin of behavioural modernity. Evolutionary Anthropology 12(4), 188–202. d’Errico, F. & Vanhaeren, M. (2000). Mes morts et les morts de mes voisins. Le mobilier funéraire de l’Aven des Iboussières et l’identification de marqueurs culturels à l’Epipaléolithique. In Epipaléolithique et Mésolithique: Les derniers chasseurs-cueilleurs d’Europe occidentale, pp. 325-342. Actes du colloque international de Besançon, October 1998. Besançon: Presses Universitaires Franc-Comtoises.

544

Speaking with beads d’Errico, F. & Vanhaeren, M. (2002). Criteria for identifying red deer (Cervus elaphus) age and sex from upper canines. Application to the study of Upper Palaeolithic and Mesolithic ornaments. Journal of Archaeological Science 29, 211–232. d’Errico, F. & Vanhaeren, M. (2004). Did they talk to each other? Ethnolinguistic diversity of Aurignacian populations and their contacts with Neanderthals revealed by personal ornaments. Présentation au colloque ‘Modern Human and Neanderthals Meet’, Blaubeuren-Tübingen 7–10 July 2004. d’Errico, F. & Villa, P. (1997). Holes and grooves: the contribution of microscopy and taphonomy to the problem of art origins. Journal of Human Evolution 33, 1–31. d’Errico, F., Jardon-Giner, P. & Soler Major, B. (1993). Etude des perforations naturelles et artificielles sur coquillages. In Traces et fonction: les Gestes Retrouvés, pp. 243–254. Liège: ERAUL. d’Errico, F., Zilhao, J., Baffier, D., Julien, M. & Pelegrin, J. (1998). Neandertal acculturation in Western Europe? A critical review of the evidence and its interpretation. Current Anthropology 39, 1–44. d’Errico, F., Henshilwood, C., Lawson, G., Vanhaeren, M., Tillier, A.-M., Soressi, M., Bresson, F., Maureille, B., Nowell, A., Lakara, J., Backwell, L. & Julien, M. (2003). The search for the origin of language and symbolism. The contribution of an ongoing multidisciplinary research project. Journal of World Prehistory 17(1), 1–70. d’Errico, F., Henshilwood, C., Vanhaeren, M., van Niekerk, K. & Jacobs, Z. (2005). Nassarius kraussianus shell beads from Blombos Cave: evidence for symbolic behaviour in the Middle Stone Age. Journal of Human Evolution. Davidson, I. & Noble, W. (1992). Why the first colonisation of the Australian region is the earliest evidence of modern human behaviour. Archaeology in Oceania 27, 135–142. de Waal, F.B.M. (1996). Good Natured: The Origins of Right and Wrong in Humans and Other Animals. Cambridge, MA: Harvard University Press. Deacon, H.J. (1995). Two Late Pleistocene-Holocene archaeological depositories from the Southern Cape, South Africa. South African Archaeological Bulletin 50, 121–131. Debénath, A. (1994). L’Atérien du nord de l’Afrique du Sahara, Sahara 6, 21–30. Demmerseman, A. (1998). Aspects de la société tunisienne d’après Ibn Abî I Dhiyâf. Tunis: Publications de l’IBLA. Dransart, P. (1998). A short history of rosaries in the Andes. In (L.D. Sciama & J.B. Eicher, Eds) Beads and Beadmakers: Gender, Material Culture and Meaning, pp. 129–146. Oxford: Berg. Dubin, L.S. (1987). The History of Beads. From 30,000 B.C. to the Present. London: Thames & Hudson. Dundes, A. (1981). The Evil Eye. A Folklore Casebook. New York: Garland. Eicher, J.B. (1998). Beaded and bedecked Kalabari of Nigeria. In (L.D. Sciama & J.B. Eicher, Eds) Beads and Beadmakers: Gender, Material Culture and Meaning, pp. 95–116. Oxford: Berg.

545

From Tools to Symbols Erikson, J.M. (1969). The Universal Bead. New York: W.W. Norton. Evans, J. (1976). Magical Jewels of the Middle Ages and the Renaissance. New York: Dover. Fairhall, A.W., Young, A.W. & Erickson, L.J. (1976). University of Washington dates IV. Radiocarbon 18, 221–239. Faris, J.C. (1972). Nuba Personal Art. London: G. Duckworth. Fischer, H. (1896). Note sur les coquilles récoltées par M.E. Piette dans la grotte du Mas d’Azil. L’Anthropologie 6, 633–652. Fischer, P. (1876). Sur les coquilles récentes et fossiles trouvées dans les cavernes du midi de la France et de la Ligurie. Bulletin de la Société géologique de France (3e série) IV, 329–342. Fisher, A. (1984). Africa Adorned. New York: Harry N. Abrams. Floss, H. (2000). Le couloir Rhin-Saône-Rhône: axe de communication au Tardiglaciaire? In Les derniers chasseurs-cueilleurs d’Europe occidentale, pp. 313-321. Actes du colloque international de Besançon, October 1998. Besançon: Presses Universitaires Franc-Comptoises. Francis, P. (1992). Heirlooms of the Hills: Southeast Asia. Lake Placid, New York: Center for Bead Research. Fritz, C. & Simonnet, R. (1996). Du geste à l’objet: les contours découpés de Labastide : résultats préliminaires. Techne 3, 63–77. Geary, C.M. (1994). Beaded Splendor. Washington DC: National Museum of African Art, Smithsonian Institution. Gelb, I.J. (1952). A Study of Writing. Chicago: University of Chicago Press. Godelier, M. (1996). L’Enigme du Don. Paris: Fayard. Graeber, D. (1996). Beads and money: notes towards theory of wealth and power. American Ethnologist 23(1), 4–24. Gramble, G. (1982). Interaction and alliance in Palaeolithic Society. Man 1, 92–107. Granger, J.-M. & Lévêque, F. (1997). Parure castelperronienne et aurignacienne: étude de trois séries inédites de dents percées et comparaisons. Comptes Rendu de l’Académie des Sciences de Paris (Préhistoire) 325, 537–543. Grün, R. & Beaumont, P.B. (2001). Border Cave revisited: a revised ESR chronology. Journal of Human Evolution 40, 467–482. Hahn, J. (1992). Eiszeitschmuck auf der Schwabischen Alb. Ulm: Suddeutsche Verlagsgesellschaft. Hare, P.E., Goodfriend, G.A., Brooks, A.S., Kokis, J.E. & von Endt, D.W. (1993). Chemical clocks and thermometers: diagenetic reactions of amino acids in fossils. Carnegie Institution Washington Yearbook 92, 80–85. Harrisson, B. (1961). Orang-Utan – der ‘Waldmensch’ in Gefahr. Umschau 61: 745–748. Hayden B. (1993). The cultural capacities of Neandertals: a review and re-evaluation. Journal of Human Evolution 24, 113–146. Heizer, R.F. (Ed.) (1978). Handbook of North American Indians, Volume 8. California. Washington D.C.: Smithsonian Institution.

546

Speaking with beads Henshilwood, C., d’Errico, F., Vanhaeren, M., van Niekerk, K. & Jacobs, Z. (2004). Middle Stone Age shell beads from South Africa. Science 304, 404. Hodder, I. (1977). The distribution of material culture items in the Baringo District, Western Keya. Man 12, 239–269. Hodder, I. (1991). The Meanings of Things. Material Culture and Symbolic Expression. London: Harper Collins. Ichon, A. (1973). Acculturation et abandon du costume traditionnel. In: L’Homme, hier et aujourd’hui : recueil d’études en hommage à André Leroi-Gourhan, pp. 695–704. Paris: Cujas. Ifrah, G. (1994). Histoire universelle des chiffres. L’intelligence des homes racontée par les nombres et le calcul. Paris: Ed. Robert Laffont. Ingold, T., Riches, D. & Woodburn, J. (1988). Hunters and Gatherers. Property, Power and Ideology. Oxford: Berg. Inskeep, R. (1962). The age of the Kondoa rock paintings in light of recent excavations at Kisese II rock shelter. In (G. Mortelmans & J. Nenquin, Eds) Actes du IVe Congres Panafricain de Préhistoire de l’étude du Quaternaire, pp. 249–256. Tervuren: Sciences Humaines 40. Jackson, J.W. (1917). Shells as Evidence of the Migration of Early Cultures. Manchester: Manchester University Press. James, W. (1988). The Listening Ebony. Moral Knowledge, Religion and Power among the Uduk of Soudan. Oxford: Clarendon Press. James, W. (1996). Uduk Resettlement. Dreams and Realities. In (T. Allen, Ed.) In Search of Cool Ground: War, Flight and Homecoming in Northeast Africa. London: Currey. Janowski, M. (1998). Beads, prestige and life among the Kelabit of Sarawak, East Malaysia. In (L.D. Sciama & J.B. Eicher, Eds) Beads and Beadmakers: Gender, Material Culture and Meaning, pp. 213–246. Oxford: Berg. Jantschke, F. (1972). Orang-Utans in Zoologischen Gärter. München: Piper. Jones, S. (1997). The Archaeology of Ethnicity. Constructing Identities in the Past and Present. London: Routledge. Ketraru, N.A. (1989). Une amulette du site du Paléolithique supérieur dans la grotte de Brynzeny I. Monuments de l’art le plus ancien sur le territoire de Moldavie: Kichinev, pp. 5–10. Kinietz, W.V. (1972). The Indians of the Western Great Lakes 1615–1760. Ann Arbor: University of Michigan Press. Klein, R. (1989). The Human Career. Chicago: University of Chicago Press. Klein, R. (1992). The archaeology of modern human origins. Evolutionary Anthropology 1, 5–14. Klein, R. (1995). Anatomy, behaviour, and modern human origins. Journal of World Prehistory 9, 167–198. Klump, D. & Kratz, C. (1993). Aesthetics, expertise and ethnicity: Okiek and Masai perspectives on personal ornament. In (T. Spear & R. Waller, Eds) Being Masai: Ethnicity and Identity in East Africa. London: Currey.

547

From Tools to Symbols Köhler W. (1921). Intelligenzprüfungen an Menschenaffen. Berlin: Springer. Koumouzelis, M., Kozlowski, B.G., Kozlowski, J., Pawlikowski, M., Bar-Yosef, O., Albert, R.A., Litynska-Zajac, M., Stworzewicz, E., Wojtal, P., Lipecki, G., Tomek, T. & Bochenski, Z.M. (2001). The Early Upper Palaeolithic in Greece: The Excavations in Klisoura Cave. Journal of Archaeological Science 28, 515–539. Kozlowki, J.K. (2000). L’art de la préhistoire en Europe orientale. CNRS Editions: Paris. Kozlowski, J.K. (Ed.) (1982). Excavation in the Bacho Kiro Cave (Bulgaria): Final Report. Warsaw: Polish Scientific Publishers. Kuhn, S.L., Stiner, M.C., Reese, D.S. & Güleç, E. (2001). Ornaments of the earliest Upper Palaeolithic: new insights from the Levant. Proceedings of the National Academy of Science 98(13), 7641–7646. Kunz, G.F. (1971). The Curious Lore of Precious Stones. New York: Dover. Kuper, H. (1973). Costume and identity. Comparative Studies in Society and History 15, 348–367. Leroi-Gourhan, A. (1964). Le geste et la parole. La mémoire et le rythme. Paris: Albin Michel. Leroi-Gourhan, A. & Leroi-Gourhan, A. (1965). Chronologie des grottes d’Arcy-sur-Cure. Gallia Préhistoire 7, 1–64. Lethmate, J. & Dücker, G. (1973). Untersuchungen zum Selbsterkennen im Spiegel bei OrangUtans und einigen anderen Affenarten. Zeitschrift fur Tierpsychologie 33, 248–269. Lindley, J.M. & Clark, G.A. (1990). Symbolism and modern human origins. Current Anthropology 31, 233–240. Marshall, L. (1998). Sharing, talking, and giving: relief of social tensions among the !Kung. In (R.B. Lee & I. DeVore, Eds) Kalahari Hunter-Gatherers: Studies of the !Kung San and Their Neighbors, pp. 349–371. Cambridge: Harvard University Press. Mason, R.J. (1962). Prehistory of Gauteng. Johannesburg: Witswatersrand University Press. Mason, R.J. (1993). La Grotte de Hearths, Vallée de Makapansgat, Transvaal, Afrique du Sud, 1937–1988. L’Anthropologie 97, 85–96. Mason, R.J., Brain, C.K., Brown, A.J.V., Cooke, H.B.S., Deacon, H.J., Ewer, R.F., de Graaff, B., Hofmeyr, I., Malan, B.D., Tobias, P.V., du Toit, A.P., Turton, K. & Wells, L.H. (1988). Cave of Hearths, Makapansgat, Transvaal. Johannesburg: University of the Witswatersrand Archaeological Research Unit, Occasional Papers 21. Masson, A. (1982). Circulations paléolithiques: une question de longueur. Bulletin de la Société préhistorique française 79(7), 197. Mauss, M. (1923-1924). Essai sur le don: forme et raison de l’échange dans les sociétés archaïques. L’Année Sociologique I, 30–186. McBrearty, S. & Brooks, A.S. (2000). The revolution that wasn’t: a new interpretation of the origin of modern human behaviour. Journal of Human Evolution 39, 453–563. McKinnon, J. (1971). The orang-utan in Sabak today. A study of a wild population in the Ulu Segama Reserve. Oryx 11, 141–191.

548

Speaking with beads Mehlman, M.J. (1989). Late Quaternary Archaeological Sequences in Northern Tanzania. PhD Thesis. University of Illinois, Urbana. Mehlman, M.J. (1991). Context for the emergence of modern man in eastern Africa: some new Tanzanian evidence. In (J.D. Clark, Ed.) Cultural Beginnings, pp. 177–196. Bonn: Dr. Rudolf Habelt GMBH. Meisch, L.A. (1998). Why do they like red? Beads, ethnicity and gender in Ecuador. In (L.D. Sciama & J.B. Eicher, Eds) Beads and Beadmakers: Gender, Material Culture and Meaning, pp. 147–177. Oxford: Berg. Mellars, P. & Tixier, P.J. (1989). Radiocarbon-accelerator dating of Ksar’Akil (Lebanon) and the chronology of the Upper Palaeolithic sequence of the Near East. Antiquity 63, 761–768. Mellars, P. (1989). Major issues in the emergence of modern humans. Current Anthropology 30, 349–385. Mellars, P. (1996). The Neanderthal Legacy. Princeton: Princeton University Press. Miller, G.F. (2001). Aesthetic fitness: how sexual selection shaped artistic virtuosity as a fitness indicator and aesthetic preferences as mate choice criteria. Bulletin of Psychology and the Arts 2(1), 20–25. Miller, G.H., Beaumont, P.B., Deacon, H.J., Brooks, A.S., Hare, P.E. & Jull, A.J.T. (1999). Earliest modern humans in southern Africa dated by isoleucine epimerization in ostrich eggshell. Quaternary Science Review 18, 1537–1548. Mithen, S.J. (1996). The Prehistory of the Mind: A Search for the Origins of Art, Religion and Science. London: Thames & Hudson. Morel, J. (1974). La station éponyme de l’Oued Djebbana à Bir-el-Ater (Est algérien), contribution à la connaissance de son industrie et de sa faune, L’Anthropologie 78, 53–80. Morse, K. (1993). Shell beads from Mandu Mandu Creek rock-shelter, Cape Range peninsula, Western Australia, dated before 30,000 b.p., Antiquity 67, 877–883. Newell, R.R., Kielman, D., Constandse-Westermann, T.S., Van der Sanden, W.A.B. & Van Gijn, A. (1990). An Inquiry into the Ethnic Resolution of Mesolithic Regional Groups. The Study of Their Decorative Ornaments in Time and Space. Leiden: Brill. Noble, W. & Davidson, I. (1996). Human Evolution and Mind: A Psychological and Archaeological Inquiry. Cambridge: Cambridge University Press. O’Hear, A. (1998). Lantana beads: gender issues in their production and use. In (L.D. Sciama & J.B. Eicher, Eds) Beads and Beadmakers: Gender, Material Culture and Meaning, pp. 117–128. Oxford: Berg. O’Shea, J. (1984). Mortuary Variability: An Archaeological Investigation. New York: Academic Press. Palma di Cesnola, A. (1993). Il Paleolitico Superiore in Italia. Firenze: Garlatti e Razzai. Pavitt, W.T. & Pavitt, K. (1970). The Book of Talismans, Amulets & Zodiacal Gems. Hollywood: Wilshire Book Co.

549

From Tools to Symbols Plug, I. (1982). Bone tools and shell, bone and ostrich eggshell beads from Bushman Rock Shelter (BRS), Eastern Transvaal. South African Archaeological Bulletin 37, 57–62. Powell, I. (1995). Ndebele, A People and Their Art. New York: Cross River Press. Preston-Whyte, E. (1994). Speaking with Beads. London: Thames & Hudson. Price, S. (1991). Parure. In (P. Bonte, & M. Izard, Eds) Dictionnaire de l’ethnologie et de l’anthropologie, pp. 560–561. Paris: PUF. Ray, D.J. (1975). The Eskimos of Bering Strait 1650–1898. Seattle: University of Washington Press. Reid, P. & Wilson, E. (1981). An Analysis of Trade Mechanisms in European Prehistory. London: University of Ann Arbor. (Microfilm IX) Reynolds, B. (1978). Beothuk. In (B.G. Trigger, Ed.) Handbook of North American Indians. Vol. 15. Northeast, pp. 101–108. Washington D.C.: Smithsonian Institution. Reynolds, V. & Reynolds, F. (1965). Chimpanzees of the Budongo Forest. In (I. DeVore, Ed.) Primate Behaviour. New York: Holt, Rinehart & Winston. Rivière, E. (1887). De l’Antiquité de l’homme dans les Alpes Maritimes. Paris: J.-B. Baillière. Rivière, E. (1904a). Bracelets, parures, monnaies d’échange, fétiches. Bulletin de la Société préhistorique française 1, 83–89. Rivière, E. (1904b). Les parures en coquillage. Bulletin mensuel de la Société d’Anthropologie de Paris, 4 March 1904, 199–201. Roach, M.E. & Eicher, J.B. (Eds.) (1965). Dress, Adornment and Social Order. New York: J. Wiley. Sacchi, D. (1986). Le Paléolithique supérieur du Languedoc occidental et du Roussillon. (XXIe suppléments à Gallia Préhistoire) Paris: CNRS. Schaller, G.B. (1965). The behaviour of the Mountain Gorilla. In (I. DeVore, Ed.) Primate Behaviour. New York: Holt, Rinehart & Winston. Schenkel, R. (1964). Zur Ontogenese des Verhaltens bei Gorilla und Mensch. Zeitschrift fur Morphologie und Anthropologie 54, 233–259. Schoeman, S. (1983). Eloquent beads, the semantics of a Zulu art form. Africa Insight 13(2): 5–25. Sciama, L.D. & Eicher, J.B. (Eds) (1998). Beads and Beadmakers: Gender, Material Culture and Meaning. Oxford: Berg. Sciama, L.D. (1998). Gender in the making, trading and uses of beads: an introductory essay. In (L.D. Sciama & J.B. Eicher, Eds) Beads and Beadmakers: Gender, Material Culture and Meaning, pp. 1–46. Oxford: Berg. Seligman, C.G. & Seligman, B.Z. (1932). Pagan Tribes of the Nilotic Sudan. London: Routledge. Shea, J.J. (2003). Neandertals, competition and the origin of modern human behavior in the Levant. Evolutionary Anthropology 12(4), 173–187.

550

Speaking with beads Shennan, S. (1989). Introduction: archaeological approaches to cultural identity. In (S. Shennan, Ed.). Archaeological Approaches to Cultural Identity, pp. 1–32. London: Routledge, Unwin Hyman. Shennan, S. (2001). Demography and cultural innovations: a model and its implications for the emergence of modern human culture. Cambridge Archaeological Journal 11, 5–16. Sinitsyn, A.A. (2003). The most ancient sites of Kostienki in the context of the Initial Upper Paleolithic of northern Eurasia. In (J. Zilhao & d’Errico F., Eds) The Chronology of the Aurignacian and of the Transitional Technocomplexes. Dating, Stratigraphies, Cultural Implications, pp. 89–107. IPA (Trabalhos de Arqueologia 33): Lisboa. Sonneville-Bordes, D. de (2002). Roc de Combe. Préhistoire du Sud-Ouest 13, 37–58. Strathern, A. & Strathern, M. (1971). Self-decoration in Mount Hagen. London: G. Duckworth. Stringer, C. (2002). Modern human origins: progress and prospects. Philosophical Transactions of the Royal Society London B 357, 653–579. Stringer, C. & Gamble, C. (1994). In Search of the Neanderthals: Solving the Puzzle of Human Origins. London: Thames & Hudson. Taborin, Y. (1993). La parure en coquillage au Paléolithique. Paris: CNRS. (XXIXe suppléments à Gallia Préhistoire). Taborin, Y. (1996). La parure paléolithique et la notion de territoire. In (A. Beltran & A. Vigliardi, Eds) L’Art au Paléolithique et au Mésolithique, pp. 143–148. XIII Congress International de l’UISPP, Forli (Italie), 8)14 September 1996. Forli: Abaco. Taborin, Y. (2003). La mer et les premiers hommes modernes. In (B. Vandermeersch, Ed.) Echanges et diffusion dans la préhistoire méditerranéenne, pp. 113–122. CTHS: Paris. Tainter, J.R. (1978). Mortuary practices and the study of prehistoric social systems. Archaeological Method and Theory 1, 105–141. Testart, A. (2001). Deux Polithiques funéraires. Texte de Conférence tenue à l’Institut Français de Porto en Janvier 2001. Thackeray, A.I. (1992). The Middle Stone Age south of the Limpopo River. Journal of World Prehistory 6, 385–440. Thompson, J. (1972). Preliminary Study of Traditional Kutchin Clothing in Museums. Ottawa: National Museums of Canada. Tillet, T. (1978). Présence de pendeloques en milieu atérien au Niger oriental. Bulletin de la Société Préhistorique française 75, 273–275. Thompson, J.C., Bower, J.R.F., Fisher, E.C., Mabulla, A.Z.P., Marean, C.W., Stewart, K. & Vondra, C.F. (2004). Loiyangalani: behavioral and taphonomic aspects of a Middle Stone Age site in the Serengeti Plain, Tanzania. Paper presented at the Paleoanthropology Society Meeting, Montreal, 30–31 March 2004. Torti, C. (1983). Circulations paléolithiques : questions de longueur…et de prudence. Bulletin de la Société préhistorique française 80(2), 44–45.

551

From Tools to Symbols Trigger, B.G. (Ed.) (1978). Handbook of North American Indians. Volume 15. Northeast, pp. 101– 108. Washington D.C.: Smithsonian Institution. Turner, T. (1980). The social skin. In (J. Cherfas, Ed.) Not Work Alone: Cross-cultural Accounts of Activities Superfluous to Survival. London: Temple Smith. Twala, R. (1951). Beadwork in the Zulu cultural tradition. African Studies 10(3), 113–123. Valladas, H., Mercier, N., Joron, J.-L. & Reyss, J.-L. (1998). Gif Laboratory dates for Middle Paleolithic Levant. In (T. Akazawa, K. Aoki & O. Bar-Yosef, Eds) Neanderthals and modern humans: discussing the transition: central and Eastern Europe from 50,000 – 30,000 BP, pp. 92–111. Mettmann, Germany, Neanderthal Museum. Vanhaeren, M. (2002). Les fonctions de la parure au Paléolithique supérieur: de l’individu à l’unité culturelle. IPGQ: Talence. Thèse de Doctorat de l’Université Bordeaux I. Vanhaeren, M. & d’Errico, F. (2001). La parure de l’enfant de La Madeleine (fouilles Peyrony). Un nouveau regard sur l’enfance au Paléolithique supérieur. Paléo 13, 201–237. Vanhaeren, M. & d’Errico, F. (2003). The body ornaments associated with the burial. In (J. Zilhao & E. Trinkaus, Eds) Portrait of the Artist as a Child. The Gravettian Human Skeleton from the Abrigo do Lagar Velho and its Archaeological Context, pp. 154–186. Trabalhos de Arqueologia, 22. Instituto Portugues de Archeologia: Lisbonne. Vanhaeren, M. & d’Errico, F. (2003). Childhood in the Epipaleolithic. What do personal ornaments associated to burials tell us? In (L. Larsson, H. Kindgren, K. Knutsson, D. Leoffler & A. Akerlund (Eds) Mesolithic on the Move, pp. 494-505. Papers presented at the Sixth International Conference on the Mesolithic in Europe, Stockholm 2000. Oxford : Oxbow Monographs. Vanhaeren, M. & d’Errico, F. (2003). Le mobilier funéraire de la Dame de Saint-Germain-laRivière et l’origine paléolithique des inégalités sociales. Paléo 15, 195–238. Vanhaeren, M. & d’Errico, F. (2004). Speaking with Beads: The Ethnolinguistic Geography of Early Upper Paleolithic Europeans. Présentation à la Paleoanthropology Society, Montreal, 30–31 March 2004. Vanhaeren, M., d’Errico, F., Billy, I. & Grousset, F. (2004). Tracing the source of Upper Palaeolithic shell beads by strontium isotope dating. Journal of Archaeological Science 31, 1481–1488. Vitebsky, P. (1995). Shamanism. University of Oklahoma Press. Vogel, J.C. (2000). Radiometric dates for the Middle Stone Age in South Africa. In (P.V. Tobias, M.A. Raath, J. Moggi-Cecchi & G.A. Doyle, Eds) Humanity from African Naissance to Coming Millennia. Florence University Press and Witwatersrand University Press. Wadley, L. (2001). What is cultural modernity? A general view and a South African perspective from Rose Cottage Cave. Cambridge Archaeological Journal 11(2), 201–221. Wanless, A. (1999). The shadow and the background: Louis Fourie in South West Africa in the 1920’s. Paper presented at the Encounters with Photography conference, Cape Town, 16 July 1999.

552

Speaking with beads Weiner, A. (1992). Inalienable Possessions: The Paradox of Keepingwhile-Giving. Berkeley: University of California Press. Weyler, E. (1959). Peuples primitifs d’aujourd’hui. Paris: Horizons de France. White, R. (1993). Technological and social dimensions of Aurignacian age: body ornaments across Europe. In (H. Knecht, A. Pike-Tay & R. White, Eds) Before Lascaux: The Complex Record of the Early Upper Palaeolithic, pp. 277-300. Boca Raton: CRC Press. White, R. (1995). Ivory personal ornaments of Aurignacian age: technological, social and symbolic perspectives. In (M. Menu, P. Walter & F. Widemann, Eds) Le travail et l’usage de l’ivoire au Paléolithique supérieur. Ravello: Centre Universitaire européen pour les Biens Culturels. White, R. (2000). Personal ornaments from the Grotte du Renne at Arcy-sur-Cure, Athena Review 2(4), 41–46. Wiessner, P. (1982). Risk, reciprocity and social influences on !Kung San economics. In (E. Leacock & R. Lee, Eds) Politics and History in Band Societies, pp. 61–84. Cambridge: Cambridge University Press. Wiessner, P. (1984). Reconsidering the behavioral basis for style: a case study among the Kalahari San. Journal of Anthropological Archaeology 3, 190–234. Wiessner, P. (1986). !Kung San networks in a generational perspective. In (M. Biesele, R. Gordon & R. Lee, Eds) The Past and the Future of !Kung San Ethnography, pp. 103–136. Quellen zur Khoisan-Forshung, Band 4. Hamburg: Helmut Buske Verlag. Wiessner, P. (1990). Is there a unity to style? In (M. Conkey & C. Hastorf, Eds) The Uses of Style in Archaeology, pp. 105–112. Cambridge: Cambridge University Press. Wilkie, L.A. (1998). Beads and breasts: the negotiation of gender roles and power at New Orleans Mardi Gras. In (L.D. Sciama & J.B. Eicher, Eds) Beads and Beadmakers: Gender, Material Culture and Meaning, pp. 177–192. Oxford: Berg. Wrangham, R.W., McGrew, W.C., de Waal, F.B.M. & Heltne, P. (1994). Chimpanzee Cultures. Cambridge, MA: Harvard University Press. Zilhao, J. & d’Errico, F. (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.

553

Personal names index This index lists authors cited in references, as well as the names of persons referred to in the text but not necessarily cited. Names mentioned in Acknowledgements are not listed here, neither are authors cited in figure captions. A lower-case ‘n’ preceding a number indicates that the name is mentioned in an endnote; the number after the ‘n’ refers to the page number on which the reference to the endnote occurs. Compiled by Marthina Mössmer

A

B

Abbot 233 Abitbol 186 Adachi 123 Adamiec 465, 469 Adcock 284 Adler 308 Aguirre 136, 243 Ahern 221 Aielllo 40 Aitken 461, 468, 501 Albrecht 304, 319 Alexeev 199, 215 Allen 313 Alvarez Fernandez 527 Ambrose 312, 391, 499, 500, 526, 540, 542 Anderson-Gerfaud 397, 400, 407 Andrews 96, 105, 108, 136, 138 Appadurai 531 Archibold 345 Archie 207 Armstrong 89 Arnason 123, 125, 126, 130, 137 Ascenzi 304 Assefa 305, 306, 336, 339, 340, 343, 349, 361, 362 Aston n28 Audouze 422 Avery 111, 249, 497, 499, 500, 504 Ayala 96 Azoury 312

Baas 109 Backwell 5, 68, 71, 244, 251, 253 Bacon 139, 142 Baffier 313, 320 Bahn 527 Bailey 123 Balout 31 Bamford 3, 109, 111, 112 Barham 309, 391, 504 Barral 243 Bartram 242, 264, 265, 340 Bar-Yosef 2, 311, 362, 391, 526 Bar-Yosef Mayer 527 Bauchot 90 Beaumont 243, 494, 495, 499, 540 Beaune 73 Beck 298 Beckwith 530 Bednarik 313, 526 Begun 124, 126, 127, 129, 146 Behrensmeyer 259 Benade 186, 191 Benefit 136 Ben-Itzhak 350, 351 Berge 186 Berger 5, 111, 153, 154, 156, 350 Bergman 539 Besancon 299 Beynon 234 Beyriès 54, 68 Biddittu 243 Biegert 144

Bielicki 91 Biesele 382 Binford 242, 305, 306, 307, 336, 337, 338, 339, 340, 343, 349, 363, 393 Bishop 105, 123, 137 Bleek 22, 382 Blumenschine 95, 96, 199, 201, 338, 347 Blurton Jones 306 Boas 532 Boaz 112 Bocherens 305, 306, 348, 353, 358 Boëda 302, 307, 393, 422, 423, 424, 437 Boesch 56, 57, 59, 60, 84 Boesch-Achermann 56, 57, 60 Bolus 296, 297, 298, 315, 318, 320, 322, 344, 362, 409, 526 Bon 296 Bonifay 243 Bonnefille 106, 107, 108, 109, 114, 115 Bonnichsen 242 Bordes 29, 299, 310, 392, 406, 408 Bordy 166 Borgia 528 Boschian 240, 243, 261 Bosinski 299, 300, 527 Bøtter-Jensen 465 Boule 18 Bourgeois 15 Bourguignon 396

555

Authors and personal names index Bousfield 177 Bowcock 286 Braga 4, 221, 232, 233 Brain 3, 5, 41, 42, 44, 68, 186, 230, 242, 243, 244, 247, 256, 529 Braumann 207 Bray 468 Brett 249 Breuil 11, n11, 16, n18, 18, 19, n28, 29, 30, 31, 243 Brézillon 64 Broca 95 Broglio 315, 316 Bromage 259 Brooks 2, 85, 175, 179, 240, 243, 296, 298, 299, 300, 304, 309, 362, 378, 390, 391, 405, 420, 454, 494, 500, 504, 526, 540 Broom 41, 110, 186, 188, 230 Bross 528 Brown 283 Brunet 125, 130, 143, 144 Bruni 243 Budge 531 Budwig 527 Buisson 319 Bunn 68, 242, 359 Burch 351, 529 Burke 306 Burkitt n11, 16, 18, 22 Butzer 497 Byrne 56

C Cadman 115 Cahen 243 Campbell 88 Cann 5, 282, 284 Capparos 206 Carbonell 396 Carey 529 Carlquist 109 Carrion 115, 500 Cartwright 483 Cassoli 243 Catuneanu 166

556

Ceci 532 Cerling 105 Chacón 307 Chamberlain 199, 201, 220 Chance 64 Chaplin 352 Chapman 137 Chase 242, 340, 343, 344, 390, 408, 526 Chazan 422, 425 Chen 287 Chiarelli 136 Churchill 318 Clark 112, 177, 179, 253, 299, 300, 375, 526 Clarke 94, 110, 112, 159, 186, 221, 222, 247, 266 Clottes 317, n373 Coates Palgrave 112 Coetzee 109 Cole 528 Coles 527 Collard 199, 222 Conard 6, 296, 297, 298, 299, 300, 302, 306, 307, 308, 311, 312, 313, 315, 316, 318, 320, 321, 322, 344, 362, 409, 526, 540 Connan 302 Conroy 136 Cooke 177, 540 Coolidge 393, 401 Cooper 284 Cope 207 Coppens 143, 146, 185, 188 Cordier 527 Corruccini 138 Corwell 529 Costamango 484, 490 Coupland 345 Courty 520, 521 Cowling 483 Coye n14, n17 Creutzfeldt 96 Cruz-Uribe 340, 483 Curnoe 256

Curtin 531 Curtis 108 Cuvier 88

D Darroch 200 Dart 22, 95, 242, 243 David 407 Davidson 240, 391, 420, 526, 535 Davis 515 Day 88 de Bonis 124, 136, 146 de Heinzelin 114 de Morgan 17 de Mortillet 11, 13, 15, 17 De Vos 345 de Waal 528 de Wit 166 Deacon n13, 96, 111, 179, 243, 249, 298, 345, 420, 422, 454, 477, 490, 494, 497, 499, 500, 539, 540 Dean 221, 234 Debenath 243, 540 Dechamps 112, 114 Defleur 309 Deino 123 Delagnes 396, 401 Delattre 186 Delpech 340, 343, 344, 358 Delporte 296, 406, 407 Delson 232 Demars 401 Demmerseman 530 Dennell 240 d’Errico 2, 5, 68, 71, 242, 243, 244, 251, 253, 259, 262, 296, 298, 303, 304, 309, 310, 311, 313, 318, 320, 335, 344, 361, 390, 391, 408, 409, 446, 452, 454, 526, 527, 533, 535, 537, 539, 541, 542 deRuiter 256 Diamond 288, 391

Authors and personal names index Dibble 390, 405, 408, 526 Dillehay 513, 520 Disotell 124 Dissanayake 373 Dixon 513, 514 Dobosi 243 Dobzhansky 96, 97 Donald 454 Doran 59 Dowson 379 Drake 108 Dransart 533 Dubin 527, 528, 529, 530, 531, 533 Dubow 20 Dücker 528 Duller 298, 465, 469 Dundes 531 Dunn 20 Dupont 15 Duport 243 Duren 123

E Eccles 96, 97 Eckhardt 94, 139 Edelman 376, 377 Edgar 296 Eicher 529, 530 Enard 289, 290 Endo 48 Erikson 528, 532 Evans 456, 503, 531

F Fairhall 539 Falk 93, 95, 96, 97 Faris 529 Farris 207 Feathers 422, 479, 485, 500, 501 Feblot-Augustins 308, 405 Feibel 139 Ferguson 139, 199 Fernandez-Carriba 241 Fernández-Jalvo 105 Field 230, 253 Fischer 299, 300

Fisher 288, 527, 528, 530 Fleagle 223 Floss 307, 308, 527 Foley 299, 393 Forge 378, 380 Foth 520 Fourie 533 Francis 532 Franks 15 Freeman 243 Frey 340 Frisancho 335, 350, 352 Fritz 527 Fullagar 513, 518

G Gagneux 123, 125, 130 Galbraith 465, 466, 470 Galef 57 Gamble n373, 526, 535 Gardner 436 Gargett 309, 375 Gastaldo 106 Gatti 304 Gaudzinski 243, 303, 304, 306 Gautier 242 Geary 529 Gelb 532 Geleijnse 420, 490, 497 Geneste 307 Genet-Varcin 89, 136 Gentry 105, 345 Gibbon 167 Gibson 64, 73, 94 Gifford-Gonzalez 242 Gilbert 206 Gille 59 Giocobini 242 Godelier 531, 532 Goede 499 Goldstein 286 Gombrich 373, 386 Gommery 4, 142, 143, 185, 186, 187, 188, 191, 192 Goodall 56, 84 Goodman 94, 287

Goodwin n11, 11, n13, 13, 14, 15, n17, 17, 18, 19, 22, 23, 24, 26, 27, 31, 32, 419, 446 Gordon Childe 21, 24 Goren-Inbar 45, 313 Gosselain 60 Gould 48, 89, 408 Gramble 527 Granger 408, 537 Graves 401 Grayson n14, 340, 343, 344, 358 Greaber 532 Grimaud-Hervé 93 Grine 220, 223, 233, 234 Grootes 318 Grove 345 Groves 199, 220 Grun 179 Grün 422, 499, 501, 540 Grüsser 89, 91 Guillaume 56 Guille-Escuret 59 Güleç 124, 312 Gunness-Hey 351

H Haddon 32 Hahn 304, 307, 315, 316, 318, 526, 527 Haidle 310 Haile-Selassie 143, 146 Halkett 490 Hall 178 Ham 56 Hamilton 345 Hammer 286 Harding 286, 287 Hardy 513, 517, 519 Hare 540 Harris 105, 287 Harrison 108, 123, 528 Hart 490 Harvey 89 Hasegawa 123 Hauman 111 Häusler 186 Hawkes 306

557

Authors and personal names index Hay 106, 108 Hayden 361, 526 Haynes 242 Hays 397, 400 Heese 23 Heiple 89 Heizer 529 Heizmann 124, 126 Henderson 112 Hendley 497 Henri-Martin 242 Henshilwood 2, 7, 243, 303, 310, 311, 362, 391, 442, 446, 447, 448, 449, 450, 451, 452, 454, 456, 460, 461, 468, 494, 503, 504, 539 Herman 47 Hey 287 Hill 137, 138, 139 Hodder 529 Hoering 44 Hofman 89 Holliday 2, 350 Holloway 47, 67, 89, 90 Holt 349 Honeyman Heath 83 Höss 284 Hovers 309, 310, 454 Howell 199, 243 Howells 94, 139 Hublin 320, 321, 344, n373, n390, 401 Huffman 56 Hufford 521 Hughen 298 Hughes 4, 111, 186, 221, 222 Hunt 48, 49, 59, 63 Hürzeler 124

I Ichon 529 Ifrah 533 Imbrie 352 Ingman 282, 283 Ingold 57, 531 Inizan 23, 423, 425, 426, 437 Inoue 58

558

Inskeep 540 Isaac 64, 67, 68, 71, 94, 95 Ishida 125, 130, 137 Itani 56

J Jablonski 352 Jackson 527 Jacobs 7, 175, 298, 444, 447, 450, 454, 465, 468, 469, 470, 503, 504 Jaeger 105 James 529, 530 Janke 123, 130 Janowski 532 Jansen 23, 24 Jantschke 528 Jarvik 379 Jaubert 393 Jelinek 336 Jenkins 5 Jerison 89, 91 Johanson 185, 191 Johnson 20, 192, 193 Jolly 456 Jones 504, 529 Joris n396 Jöris 300 Joron 501 Joulian 3, 54, 56, 57, 59, 60, 61, 64, 71, 74, 84, 85, 241 Julien 29, n31 Jungers 89, 139, 207 Justus 243

K Kaessmann 287 Kandel 168, 487 Kannemayer 22 Kappelman 105 Kaszycka 231 Kattwinkel 108 Kawai 56 Keeley 68 Kelley 123, 136 Kelly 345, 360, 361, 422 Kempson 167, 168

Kennedy 351 Kent 106 Kenyon 84 Ketraru 537 Keyser 247, 263 Khroustov 84 Kibunjia 266 Kim 305, 306, 340, 349, 353 Kimbel 220, 221 Kinietz 529 Kitching 243 Klein 2, 85, 179, 240, 242, 249, 296, 297, 298, 306, 340, 343, 344, 361, 363, 363, 390, 391, 420, 450, 451, 460, 477, 483, 490, 494, 497, 500, 526, 535 Kluge 207 Klump 530 Klüver 379 Knight 391 Knuttson 425 Kohler 56 Köhler 84, 528 Kölbl 311 Kolen 307 Kooij 84 Korisettar 266 Kortlandt 84 Koufos 146 Koumouzelis 537 Kozlowski 537 Kramer 207, 220 Krantz 89 Kratz 530 Krebs 345 Krings 283 Kroll 68 Kuhn 308, 312, 313, 362, 526, 539 Kuman 5, 112, 159, 164, 166, 177, 179, 221, 222, 230, 247, 266 Kunz 531 Kuper 529

L Lacruz 115, 156, 159 Lacy 207

Authors and personal names index Lague 139 Lahr 299 Lai 288 Laitman 96 Lancaster 84, 345 Lande 89 Langejans 178 Lartet 13 Lashley 91 Laslett 465 Laughlin 350 Lazenby 350, 352 Le Baron 166, 168, 170, 175, 178 Le Gros Clark 142 Leakey 67, 94, 95, 96, 105, 106, 108, 129, 130, 136, 137, 139, 140, 142, 199, 200, 214, 220, 230, 256, 258, 259, 261, 265 Leatherwood 47 Lebeltzer 20 Lee 64, 345 Lemonnier 59, 60, 242 Lenoble 476, 484 Leonard 350 Léonard 109 Leroi-Gourhan 31, 54, 63, 64, 65, 71, 73, 242, 313, 317, 400, 531, 537 Lethmate 528 Lévêque 408, 537 Lewis-Williams 6, 297, 318, 322, n373, 375, 376, 379, 383, 385 Lieberman 94, 96, 199, 201, 217, 220, 222, 223 Likens 345 Lindley 526 Lockwood 207, 223 Loeches 241 Lombard 513 Lorblanchet 311 Lorenz 499, 500 Lovejoy 89, 186 Loy 513, 514, 520 Lubbock 13 Lupo 349 Lyman 242

M Mace 89 Madrigal 349 Maes 112, 114 Maguire 242, 247 Malan 456, 466 Malerba 242 Mania 243, 302, 310 Mann 313, 319 Marais 85 Marchant 58 Marean 2, 6, 305, 306, 336, 339, 340, 343, 344, 345, 347, 349, 351, 353, 359, 361, 362, 391, 456 Marks 137, 423 Marshall 531, 535 Martin 89, 234 Martinaud 476 Martindale 375 Martinson 496, 501, 504 Marzke 63, 72 Mason 540 Masson 527 Mather 92 Matsuzawa 58 Maureille 392 Mauss 27, 54, 55, 56, 57, 59, 64, 531, 532 Mayr 91 Mazak 199 McBrearty 2, 85, 167, 175, 177, 178, 179, 240, 296, 298, 299, 300, 309, 362, 378, 390, 391, 405, 420, 454, 494, 500, 504, 526 McBurney 299 McCrossin 136 McDougall 139 McGrew 56, 58, 59, 60, 66, 84, 241 McHenry 89, 138, 201, 222, 223 McKee 232 McKinnon 528 McPherron 405 Mehlman 540 Meignen 307, 396, 430 Mein 126, 127 Meisch 529

Melentis 136 Mellars 335, 340, 343, 344, 348, 351, 353, 358, 359, 360, n373, 390, 391, 392, 401, 526, 539 Meller 302 Menter 247 Menzel 56 Mercier 7 Merrick 64 Meyerson 56 Miller 442, 500, 504, 528, 539 Milo 344 Minichillo 362 Mithen 420, 526 Miyamoto 287 Moerdyk 164 Molefe 186 Monahan 359 Moran 350 Morel 540 Morris n390 Morse 313, 541 Mosimann 200, 207 Motsumi 186 Mourre 393 Muhesen 302 Müller 31, 32 Münzel 318 Murray n11, 465, 469 Mussi 340

N Napier 63, 95, 199 Nathan 472 Neugebauer-Maresch 317 Neville Jones 24 Newell 527 Newman 41, 263 Nigro 158, 159 Nilssen 456 Nilsson 40 Nishida 56 Nishimura 56 Noble 240, 391, 420, 526, 535 Nowell 313

559

Authors and personal names index

O Oakley 83, 302 Obermaier 16 O’Connell 306, 313, 347 O’Hear 529 Ohman 142 Olsen 242 O’Shea 529 Ovchinnikov 283 Owen 305

P Pain 48, 49 Pallary 17 Palma di Cesnola 537 Palmer 109, 111, 112 Parker 73, 94 Parkington 7, 296, 297, 306, 310, 311, 322, 359, 454, 456, 476, 478, 483, 487, 490, 497, 499, 503, 504 Partridge 87, 110, 111, 164, 166 Patterson 139, 140 Pavitt 531 Pearson 335, 350 Peers 456 Pei 242 Pelegrin n31, 404, 408, 425, 426 Peltz 313 Pepperberg 48 Peresani 393 Péréz-Lézaun 286 Péringuey 20 Perpère n28 Petraglia 266 Peyrony 407 Pfeiffer 352 Phillips 242 Phillips-Conroy 256 Pianka 345 Pickford 4, 123, 125, 126, 127, 130, 136, 137, 138, 141, 142, 144, 146, 192 Pilbeam 89, 123, 124, 125, 126, 130, 136, 137 Pitman 109, 111, 112

560

Pitts 242 Plug 540 Plummer 105 Poggenpoel 476 Pollarolo 172 Post 89, 90 Potts 68 Povinelli 60 Powell 529 Prat 4, 199, 200, 201, 206, 207, 214, 216, 220, 221, 222, 223 Premack 57 Preston-Whyte 529, 531, 532 Price 527, 528, 532 Prindiville 306

R Radmilli 240, 243, 261 Rak 220, 221 Rando 307 Rappaport 385 Ray 529 Rayner 115 Reck 108 Reed 105, 111 Reeves 47 Reichel-Dolmatoff 382 Reid 234, 527 Reinach 15 Renfrew 408 Rensink 307 Révillion 299, 396, 401, 423 Reygasse 17 Reynolds 528, 529 Richards 348, 353, 358, 359 Richter 298, 318, 396 Ricklan 193 Riek 304 Riel-Salvatore 375 Rigaud 499 Rightmire 199, 200, 220 Riollet 106, 107 Rivière 527 Roach 529 Roberts 242, 468, 469 Robertshaw 540

Robinson 41, 89, 186, 188, 191, 233, 243 Roche 64, 67, 266 Roebroeks 306, 307 Rose 242, 256 Rosenzweig 345 Roubet n17, 17 Ruff 350 Rumbaugh 47 Rust 299 Ruther 488 Rutherford 111

S Sabater-Pi 66 Sacchi 527 Sacher 94 Sackett n14, 29 Sampson 177, 179 Sanday 361 Sarich 287 Sawada 141, 192 Schäfer 311 Schaller 528 Schenkel 528 Schepers 97, 230 Schick 68 Schirmer 456 Schlanger 2, n11, 14, 17, 20, 22, 27, n28, n31, 32, 55, 59, 242 Schmid 143, 185, 186, 315 Schmidtgen 306 Schmitt 349 Schnapp n11 Schoeman 532 Schrire n13 Schwarz 529 Sciama 527, 529, 532 Scott 107, 112, 115 Sealy 243, 359, 450, 451, 454, 503 Seeberger 318 Segre 243, 304 Seielstad 286, 287 Semaw 64, 266

Authors and personal names index Senut 4, 123, 126, 127, 128, 130, 136, 137, 138, 139, 140, 141, 143, 144, 146, 186, 188, 192, 353 Sept 68, 346 Shackelton 497, 503 Shahack-Gross 521 Shapiro 190 Shea 299, 302, 526 Shennan 297, 385, 526 Shepherd n13 Shipman 68, 105, 242, 243, 244, 256, 258, 259, 261, 264 Shrewsbury 192, 193 Siegel 379 Sigmon 93 Sillen 44, 359 Simmonet 527 Simons 136 Simpson 287 Singer 243, 299, 300, 303, 309, 419, 420, 422, 484, 490, 494 Sinitsyn 537 Skelton 201, 222, 223 Smith 23, 178, 180, 318, 320 Smuts n11, 13, 14, n18, 20, 21, 22, 58 Sokal 207 Solecki 309 Sollas 16 Sonneville-Bordes 537, 541 Soodyall 5 Sorensen 350 Soressi 6, 309, 310, 392, 396, 397, 400, 401, 404, 406, 407, 446 Sorg 242 Spencer 348 Speth 348, 349 Spicer 106 Spielmann 348, 349 Spoor 221, 350 Stauffer 123, 130 Steguweit 310 Stent 95 Stepanchuk 243 Stephan 90 Stern 191

Steudel 89 Stewart 124 Stiner 305, 312, 313, 336, 339, 340, 343, 344, 363 Stocking n14 Stokes 468 Stoneking 282, 286 Strahler 345 Strait 201, 217, 220, 222, 223 Strathern 529 Stringer 96, 199, 201, 220, n373, 391, 526, 535, 539 Suddendorf 60 Sugiyama 56, 58, 66, 73 Susman 143, 191, 192, 193, 231 Sutcliffe 242 Sutton 175, 176, 178 Suwa 233 Swofford 206 Szalay 94

T Taborin 454, 526, 527, 537 Taieb 185 Tainter 529 Takeshita 63 Tanaka 346 Tardieu 188 Testart 532 Texier 396, 484, 499 Thackeray 4, 186, 221, 223, 230, 231, 232, 233, 235, 420, 422, 433, 494, 497, 499, 500, 503, 504, 526 Therin 520 Thiele 206 Thieme 302, 305 Thierry 57, 59 Thoma 350 Thompson 351, 454, 456, 529, 540 Tiercelin 64 Tillet 540 Tishkoff 286 Tixier 31, 539

Tobias 3, 4, 46, 67, 84, 88, 91, 93, 95, 96, 97, 110, 156, 188, 199, 201, 221, 222, 230, 233, 234 Toepfer 302 Tomasello 57 Tomlinson 513 Tononi 376, 377 Torigoe 63 Torke 304 Torti 527 Toth 68 Tribolo 7, 480, 501, 503, 504 Trigger n11, 529 Trinkaus 309, 350 Tuffreau 396, 401 Turk 319, 520 Turner 529 Tuross 513, 520 Turq 396 Tutin 56, 59 Tuttle 63, 93, 94 Twala 532 Tyson 164 Tzedakis 344

U Underhill 286

V Valladas 7, 501, 503, 537 Van Andel 344, 358, 499 Van der Merwe 359 Van der Ryst 249 Van Doornum 178 Van Hoepen 20, 23 Van Peer 423, 424 Van Riet Lowe 11, n11, 13, n13, 14, 15, 18, 19, 20, 21, 22, 24, 26, 27, n28, 28, 29, 419, 446 Van Schaik 84 Vandermeersch 310, 311 Vanhaeren 7, 311, 527, 533, 535, 539, 542 Vaquero 307, 396 Vargha-Khadem 290 Vauclair 63

561

Authors and personal names index Vaufray n28 Vaughn 517 Veil 302, 397 Velo 517 Vértes 310 Vignaud 144 Villa 242, 264, 265 Vincens 114 Vincent 243, 346 Vitebsky 530 Voelker 298 Vogel 105, 422, 495, 500, 503, 539 Vogelsang 303, 310, 311, 313, 488 Volkman 423 Volman 175, 177, 420, 497 Von Leuvan-Smith 235 Vrba 232, 235

W Wadley 249, 300, 307, 391, 420, 494, 513, 514, 517, 519, 522, 526 Wagner 304, 306 Wakefield 96 Walker 63 Wallis 531 Walraven 63 Walter 186 Wandsnider 349 Wanless 533 Ward 123, 138, 139, 142, 190 Watson 41, 84, 263 Watts 63, 309, 310, 391, 454 Weber 243 Weiner 532 Weiss 89, 91, 312 Werger 109 Westfall 111 Weyler 533 Wheeler 40, 109 White n28, 67, 130, 139, 140, 221, 309, 454, 526, 535, 537 Whiten 1, 47, 56, 60, 241 Whittaker 345 Wiessner 311, 456, 529, 531, 535, 542

562

Wildman 287 Wilkens 96 Wilkie 532 Williamson 7, 513, 514, 515, 519 Wilson 282, 287, 527 Wintle 298, 465, 469 WoldeGabriel 140, 143, 146 Wolpoff 89, 94, 123, 125, 144, 145 Wood 199, 201, 220, 221, 222, 233, 234 Woodborne 478, 504 Woodbourne 167, 175 Woodburn 361 Wrangham 56, 123, 125, 126, 130, 137, 528 Wu 124 Wullstein 63, 72 Wurz 6, 391, 420, 422, 427, 428, 429, 430, 433, 435, 454, 477, 479, 494, 499, 500 Wymer 243, 299, 300, 303, 309, 419, 420, 422, 484, 490, 494 Wynn 310, 393, 401

Y Yamakoshi 66, 73 Yates 478 Yellen 243, 362, 454 Yerkes 56

Z Zavada 115 Zeitoun 201, 206 Zilhão 243, 296, 321, 335, 344, 390, 409, 526

Subject Index This index lists terms and subjects, while a separate index on page 555 lists cited authors and other personal names mentioned in the text. The terms hominid, hominin, and hominoid are indexed as they are used by individual authors in the text, as are scientific names for taxa, for example Australopithecus vs Paranthropus. A lower-case ‘n’ preceding a number indicates that the subject is mentioned in an endnote; the number after the ‘n’ refers to the page number on which the reference to the endnote occurs. Compiled by Marthina Mössmer

A aardvark 263 aardwolf 263 Aboriginal mtDNA 284 Abri Blanchard 317 Abri Castanet 317 Abri Cellier 317 accelerator mass spectrometry (AMS) 297 acculturation 404 Acheulean 13, 45, 153, 154, 156, 167, 242, 261, 266, 267 adaptedness vs adaptability 92 admixture 461 aDNA 283–284 damage to 284 adzes 477, 514 African terminology 11, 13, 14, 18, 20 afropithecines 123 afterlife 309 agate 24 Ahmarian 535, 539 alcelaphines 235 Allia Bay 140 altered states of consciousness 376, 377, 379, 383, 384, 385 amber 531 anatomical modernity vs cultural modernity 321, 390

ancestors, hominid 145 animal tissues 513, 515, 522 antbear 263 antelopes 105 ant-fishing 241 anthropoid apes. See great apes, chimpanzee, gorilla, orangutan antler 240, 243, 303, 535 anvils bone 258, 261, 262 stone 258 ape-human divergence. See dichotomy apes fossil 136 Miocene 123 See also great apes, chimpanzee, gorilla, orangutan Apollo 11 303, 309, 311, 313, 321, 488, 500, 505 Aramis 139 palaeoecology 140 arboreal adaptations 143, 188, 192 archaeology history of 2, 11 Palaeolithic 13 architecture 307 Arcy-sur-Cure 243

Ardipithecus 123, 128, 130, 145 Ardipithecus kadabba 143 Ardipithecus ramidus 88, 139, 140 arrow points, bone 249 art 313–318, 373, 375, 378–379, 380, 386, 489 as indicator of behavioural modernity 318 art mobilier 375 artefacts. See bone tools, pottery, stone tools, tools ash 442, 444, 480, 484 assemblage composition 27 Aterian 17, 362, 540 Aurignacian 13, 17, 304, 313, 318, 319, 362, 373, 374, 375, 404, 408, 535, 537, 539, 541 aurochs, images of 379 Australia 313 australopithecines 41, 44, 96, 106, 136, 142 brain size of 86–87, 90, 91 brain surface 93 humerus 143 postcranial morphology 185–191 teeth 125–126, 128 thumb 192–193 vertebrae 186–191 tool use 263

563

From Tools to Symbols Australopithecus 199, 220–223, 230 as monophyletic group 216 See also Paranthropus Australopithecus afarensis 114, 138, 139, 185 brain size 86–87, 88, 90 Australopithecus africanus 95, 110, 111, 139, 186, 221–222, 234 brain size 86–87, 88, 90, 91 language 97 thumb 193 tool-making 193 Australopithecus anamensis 88, 140 Australopithecus antiquus 127, 186 Australopithecus boisei 216, 221, 230, 234 brain size 86–87, 88 tool-making 266 Australopithecus garhi 88 Australopithecus ramidus. See Ardipithecus Australopithecus robustus 108, 216, 230 brain size 86–87, 90 autosomal DNA markers 286–287 Awash Basin 114 awls 241 bone 243, 249, 446

B baboons 41, 58 Bacho Kiro 537 Bachokirian 535, 537 backed stone tools 420, 429, 488, 489, 494 Balerno Main Shelter 178 banded ironstone 166 beads 527–542 antler 535 bone 535 glass 535 ivory 313, 535

564

oldest known 537 ostrich eggshell 312, 449, 529 shell 313, 446, 452–454, 535 stone 535 teeth 535 bears figurines of 316 teeth 537 behavioural modernity 2, 85, 240, 267, 295–297, 377–378, 390, 393, 420, 454–456, 460, 494, 506, 513, 526 criteria for 308, 318, 319–322 evidence for 297–322 evolution of 297, 321, 390–391 origins of 296, 298 vs anatomical modernity 321, 407–409 See also cultural modernity belemnites 537 Berekhat Ram 313 beta counting 297 betaglobin 286–287 biconical 154 bifaces 167, 168, 169, 172, 174, 177, 242, 266, 362, 405, 406, 446 cordiform 392 production of 396–401, 405 bifacial points 460, 494, 503, 514 Bilzingsleben 310 bipedalism 141, 142, 143, 144, 145, 146, 184–193, 188, 191, 193 birds figurines of 316 images of 379 tool use by 48–49, 83 birth canal 46 bison figurines of 316 images of 379 blades 179, 299, 362, 374, 420, 437, 479, 480

Blombos Cave 243, 303, 310, 311, 321, 442–456, 460–472, 494, 503, 504, 505, 506, 539, 541 blood 513, 514, 515 boats 313 body height 89 size 88 techniques 54–62 weight 88, 89 Bois Roche 264, 265 bone 448, 478, 480, 520 bone tools 158, 238–267, 240, 362, 449, 450–451, 454 and behavioural modernity 240, 241, 303–304 and Homo erectus 267 as indicators of behavioural modernity 267 formal 241, 446, 460 grinding 250–251, 264, 266 harpoons 304 Olduvai 256–260, 266, 267 Sterkfontein 243–256 Swartkrans 243–256, 267 techniques 240 bone as raw material 240 beads 535 burnt 42–44 cut marks on 45, 158, 235 engraved 446, 454 figurines 315 flutes 318 pendants 537, 540 preservation of 442 residue on tools 515 tubes 541 bones elephant 242–243 hippopotamus 484 in cave walls 382–383 incised 310 postcranial 88–89, 185–191 sheep 477 with puncture marks 258

Subject Index bonobo 201, 217, 287 brain size 86–87 traditions 241 vertebrae 187 Boomplaas 500, 505, 539 Border Cave 496, 497, 499, 500, 505 dating 232 bores 241, 392 bovids 106, 111, 158, 235, 256, 306, 340, 484 horn cores 243 teeth 537 bowerbirds 528–529 brain size 39–41, 46, 86–92, 144 absolute 88, 90 advantages of larger 91–92 reasons for increased 91 relative 88–91 brain evolution 85 size. See brain size surface 92–93, 93 brain-to-body weight ratio 40, 48, 88–89 breccia 110, 115, 153, 154, 159 Broca’s area 93, 94, 95, 97 Bryndzeny 537 buffalo 306, 340, 484 burials 309, 310, 378, 490, 537 Neanderthal 309, 375 burins 480, 489 Burkitt affair 17 Bushman Rock Shelter 540 Bushman. See San

C Candir 126 canids 111, 154, 158, 352 Capsian 17, 20 capuchin monkeys 84 carbon isotope analysis 105–106 carnivores 40, 45, 48, 156, 158, 235 marks on bone made by 265 carvings 375, 379

Castelcivita 537 Cauna de Belvis Cave 537 Cave of Hearths 540 caves 380, 382, 383–385 cellulose 513, 514, 515 cephalopods 47 cercopithecoids 111 cerebral cortex 92, 93 cerebrum 92 cetaceans 92 Chad 143 chaîne opératoire 31, 214–242 defined 422 chalcedony 166 character-age dependence 207, 217, 223 character-sex dependence 207, 217, 223 Charaman 177, 178, 179 charcoal 447, 461, 477, 478, 480, 483, 484 Châtelperronian 313, 373, 375, 401, 404, 408, 535, 537, 541 Chauvet Cave 316–317 Cheboit 129, 130 Chemeron 199 chert 24, 154, 166 chimpanzee 1, 47, 56, 123, 201, 205, 217 and language 289 as species of Homo 287 brain size 86–87, 90, 91 culture 57 gestures 58 grooming 59 humerus 143 laterality 58 nut-cracking 60–62, 84 phylogeny 123, 125, 126, 130 postcranial morphology 185 self-decoration 528 teaching 60–61 teeth 125–126, 127, 129, 137, 142, 144 termite-fishing 241, 263 thumb 192 tool use 84, 184

tools 54 traditions 59, 241 vertebrae 187 See also primates, great apes chopper-cores 177 Clacton-on-Sea 302 cladistic analysis 199–223 ingroup 205 outgroup 205 cleavers 167, 168, 170, 172, 174, 179 climate 105, 109, 164, 484, 497 and Neanderthal evolution 344–345 change 167, 180, 344–345, 499, 506 climbing adaptations 143, 188, 192 collagen 513, 514, 515 Columbella rustica shells 539 Combe Grenal 337, 340 common ancestor ape-human 127, 129–130, 137, 145–146 chimpanzee-gorilla 125 chimpanzee-human 126, 138, 287, 289, 290 Homo sapiens 282, 284, 286, 287 most recent 282, 284, 286, 287 composite tools 300 conjoined tools 168 consciousness 375–379 higher-order 377, 378, 379 primary 377 contamination 297 Conus shells 541 Coopers D 153–156 coprolites 42, 158, 353 core-axes 177 cores 23, 28, 30, 167, 171, 172, 175, 362, 401, 422, 428–429, 437 platform 299 polyhedral 235 prepared 177, 179

565

From Tools to Symbols core-scrapers 177 cosmos, tiered 381, 386 coup-de-poing 13, 23, 24 courtship 528 cracks 380 craniofacial analysis 200 crows 48–49 cul-de-sac 9, 10, 19, 20, 22, 27 cultural modernity 296, 306, 318, 321 vs anatomical modernity 321 cultural identity 529 tradition 153 transmission 84–85 culture 1, 2, 19, 57, 84, 92, 93 defined 85 primate 56 culture-historical approach 16, 19, 27 cut marks 45, 158, 235, 343 cuttlefish 47

D daggers, bone 243 Dale Rose Parlour 456, 503 Darwin’s finches 83 dassie 484 dating methods 297–298, 498–503 contamination and 297 debitage 158, 401, 514 decolonisation 20 decoration 310–311, 374–375, 489 deflation 174 dendroclimatology 109 Dentalium shells 537 denticulates 167, 168, 171, 172, 173, 177, 406 dentition. See teeth DHA (docosahexaenoic acid) 91 dichotomy ape-hominid 127, 129–130, 136–146, 145–146

566

chimpanzee-gorilla 125 chimpanzee-human 126 Homo-Australopithecus 215–216 Die Kelders 249, 343 Diepkloof 310, 311, 321, 476–490, 494, 497, 503 diet australopithecine 263 fat in 348–349, 351, 352, 353 Homo 263 in tropical environments 345–347 mammals in 306 meat in 347, 522 Neanderthal 305, 358–359 plants in 363, 521 plants in hominid 305 plants in hunter-gatherer 360–361 protein in 263, 347, 348, 358 diffusion 373–374, 401 diffusionism 16, 19–20, 21, 29 digging tools 258 dimorphism, sexual 207 Dinofelis 42, 235 diorite 23 discoid technology 393 divergence 126, 137, 146, 287, 290 See also dichotomy, common ancestor division of labour 360–361 Divje Babe 319 DNA ancient 283–284 ancient, damage to 284 Australian 284 autosomal markers 286–287 mitochondrial (mtDNA) 279, 281–284 Neanderthal 283–284 nuclear 280–281 X-chromosome 287 Y-chromosome 279, 281–282, 284–286

docosahexaenoic acid (DHA) 91 dolerite 23, 166 Dolní Vestonice 309 dolphins 47, 92 dreams 376, 377 Drimolen 191, 199, 232, 247, 263, 265 dryopithecines 124, 139 Dryopithecus 123, 124, 126, 127, 128, 129, 130, 144 dyspraxia 288–290

E Earlier Stone Age 164, 302 final 168, 176, 177 late 167 earliest hominid ancestor 139 East Rift 114 éboulis secs 497, 499 Egyptian geese 83 eland 343, 484 Elands Bay 359, 476, 483, 490 elephant bones 242–243, 256, 258, 260, 261 hunting 305 tusks 242–243 elephantids 111 enamel, tooth 126, 128, 130, 137, 139, 234 encephalisation 89, 90, 91, 95 end products 424–426, 429–430, 437 endocranial capacity. See brain size endocranial casts 86, 92–93, 93, 94, 95, 96, 97 Engelswies 126 Enkapune Ya Muto 312, 540 Enlène 382 entrapment 156 epigenetic transmission 84–85 epistemology 2, 11 equids 111, 243, 256, 305, 484 ethology 56–57 Eurocentrism 85 European terminology 13

Subject Index evolutionism 14, 16, 27, 419 exchange networks 527, 542 experimental tools. See replication

Fumane 315 fungi 521

F

G

factory sites 28, 29, 405 fat in diet 348–349, 351, 352, 353 rendering 349, 352, 353 fault lines 158 fauna 154, 158 faunal analyses 105 faunal exploitation. See hunting, scavenging Fauresmith 24, 179 feldspar 500 felids 111 images of 379 felines 42 figurative art 313–318, 319–320, 375, 378–379 figurines 313, 315–316, 317, 375, 379 fire 42–45, 485 management of 45 fish 40, 503 fish gorges 249 fissures 380 flakes 28, 30, 154, 167, 171, 172, 173, 175, 362, 392, 406, 477, 514, 517 bone 260 flint 21, 23, 24, 27, 299, 374, 405 properties of 24 Florisbad 112, 116, 179 flutes 318–319 flying 380 food sharing 305 storage 351, 357, 360 formal tools. See tools fox teeth 537, 539 FOXP2 gene 288–290 French-South African collaboration 10–32

Geißenklösterle 313, 315, 316, 318 genome, human 280, 282 Geographic Information Systems 159 Gesher Benot Yaaqov 45 giraffe bones 256, 258, 261 giraffids 106 GIS 159 Giza 299 Gladysvale 115, 153, 156 glass beads 532 tools 22 Glycymeris shells 537, 539 Goergap 249 gold 530, 531 Gondolin 247 gorilla 47, 123, 201, 205, 217 brain size 86–87 divergence from human lineage 287 fossil 129 language 289 phylogeny 125, 130 sexual dimorphism 207, 214 teeth 125, 127, 128, 129, 137, 144 Graecopithecus 124 grave goods 309, 375, 378 great apes 85 ancestor 137 and language 290 bipedalism 185 brain size 88 taxonomy 287 trunk 190 See also bonobo, chimpanzee, gorilla, orang-utan, primates Great handaxe culture 28 Great Leap Forward 288

grid-based excavations 159–161 Grimaldi Caves 309 grinding of beads 535 of bone tools 250–251, 264, 266 of pigments 310 Griphopithecus 126, 127, 129 Große Grotte 304 Grotta del Cavallo 537 Grotte Chauvet 316–317 Grotte du Renne 313, 537, 541 Grotte Vaufrey 337, 340, 343 Grotte XVI 344, 358, 392, 405 Grotte Zouhra 540

H Hackthorne 164–168, 174, 176, 177, 178, 179 Hadar 86, 91, 114, 139, 186, 199 hafting 179, 300–302, 406, 490, 514, 517, 519, 522 and ochre 519 hair 513 halites 484 hallucinations 376, 379, 380–385, 384 hallucinogens 376, 382, 383 hammer stones 154, 352 hammers 425–426, 428, 430, 437, 480, 535 bone 240, 261, 263, 266 hand axes 28, 169, 177, 179, 261, 299, 310, 489 bone 260–261, 266, 304 handedness 49 handprints 375 hares 484 harpoons, bone 304 hearths 307, 349, 442, 444, 446, 452, 484, 490 Hemmer’s Constant of Cephalisation 90 herders 477 hide-working 258, 262, 517

567

From Tools to Symbols higher-order consciousness 377, 378, 379 hippopotamus bones 484 teeth 256 Hoedjiespunt 157 Hohle Fels 313, 315 Hohlenstein-Stadel 313, 315 Hollow Rock Shelter 309, 456, 503 Holocene Age 442, 490 hominids 2, 40, 106, 230 ancestors 145 cladistic analysis 199–223 earliest 136 evolution of 105 thumb 231 hominin evolution 334–335 hominoids 123, 126, 136 Homo, early 127, 207, 217, 220–223, 230, 231, 232, 234, 235 taxonomy of 200 Homo anatomically modern 320, 321 as monophyletic group 216 chimpanzees as species of 287 subgenera 287 Homo erectus 39, 96, 108, 266, 350 and bone tools 267 brain size 87, 90 Homo ergaster 39, 96 Homo habilis 39, 84, 89, 93–94, 95, 96, 97, 108, 199, 206, 207, 214–215, 216, 220–223, 230, 234 brain size 87–88, 90, 91 brain surface 93 language 96 thumb 231 tool-making 266 Homo heidelbergensis 179 Homo helmei 112 Homo neanderthalensis. See Neanderthals

568

Homo rhodesiensis 179 Homo rudolfensis 39, 88, 199, 214–215, 220–223 Homo sapiens 40, 157, 373 archaic 179 brain size 87, 90 postcranial morphology 185–191 Homo sapiens sapiens 320, 321 horn 243, 247, 250, 252 hornfels 166 horses 305 figurines 316 images of 379 hot-rock technology 349, 351, 352 Howieson’s Poort 175, 300, 310, 311, 420, 422, 477, 478, 479, 480, 484, 488, 490, 494–503, 505–506 dating 498–503 Humallian 299 humans anatomically modern 179 evolution of modern 345–347 human-style intelligence 49 hunter-gatherers 21, 170, 299, 306, 307, 308, 345–347, 348, 349, 351, 357, 360, 363, 374, 382, 527, 542 hunting 305, 306, 336–344, 354–357, 360, 379, 490 as men’s domain 360–361 dogs 42 effectiveness 337 efficiency 337 for attaining status 347 goals of 347 in cold climates 347 in tropical environments 347 Inuit 350 killing ability 337 hyaenas 42, 349, 352, 487 marks on bone made by 264–265 hyaenids 111 hyracoids 111, 484

I Ibero-Maurusian 17 image-making 373, 375, 378–379, 379–380, 386 implements. See bone tools, stone tools, tools incised objects 310–311, 313, 446, 454, 479, 494, 506 industries 14 defined 13 discussion of 178, 179 ingroups 205 innovation 393, 404 insectivory 263 intelligence 49 internalisation 57 Inuit as control group 349–351 osteoporosis 352–353 Iron Age 153, 154, 170 bone tools 249 ironstone 166 isotope analyses 105–106, 305, 358–359 Isturitz 319 ivory 243, 303, 313, 535 beads 313, 537 figurines 315, 531 flutes 318 pendants 537 rings 541

J Jacovec Cavern 87 Jerison’s Encephalisation Quotient 89–90

K Kalambo Falls 506 Kanapoi 139, 140 Kapcheberek 143 Kapsomin 127, 128, 129, 130, 142, 144, 192 molar 144 Kasteelberg 251

Subject Index Katanda 304 Keilmesser 396 Kenyanthropus 200, 220 Kenyanthropus platyops 205, 216 kenyapithecines 123 Kenyapithecus 126, 129, 136, 137 Kenyapithecus africanus 142 Keratic Koppie 168, 174, 176, 179 Khoisan 533 mtDNA 283 killing ability 337, 343 kinship systems 377 Kisese II 540 Klasies River Mouth 157, 303, 309, 321, 337, 340, 343, 419–438, 446, 484, 490, 494, 496, 497, 499, 500–503, 504, 505, 506 dating 232 knapping 24, 27, 28, 29, 240, 253, 260, 265, 266, 299, 405 knives 398, 514 backed 406, 479 knuckle-walking 185 Kobeh 353 Königsaue 302 Koobi Fora 86, 95, 96, 105, 114, 139, 232, 234 Kostienki 537 Kromdraai 115, 153, 154, 194, 230, 231, 232, 235 dating of deposits 232 toolmaking 231 Ksar Akil 312, 539 Kudu Koppie 170–175, 176, 179

L La Ferassie 317 La Roche-Cotard 311 La Rochette 392, 401, 405, 407 Laetoli 105, 106, 106–108, 116 climate 107 flora 107–108 lagomorphs 106, 484 Lake Mungo 284 Lake Turkana 86, 96, 114, 139

lanceolates 177 language 46, 49, 93, 95–96, 311, 318, 335, 377, 386, 391 centres 94 disorders 288–290 evolution of 96, 288–290, 361 FOXP2 gene 288–290 spoken 95, 96, 97, 288 Lascaux 383 laser theodolites 159 Late Pleistocene Age 296, 299, 304, 305, 306, 320 Later Stone Age 153, 167, 170, 172, 176, 178, 249, 297, 299, 300, 302, 306, 340, 341, 442, 444, 446, 447, 448, 449, 450, 451, 452, 454, 460–472, 476, 485, 490 early 309, 312, 540 Le Moustier 392, 405, 407 leatherworking 258, 262, 310, 517 tanning 310 Lehringen 302 lenses 484 leopard 41, 42 Les Trois Frères 382 Levallois 21, 23, 24, 28, 299, 362, 393, 396, 422, 429, 437 Libytherium bones 256 limb bones 88 lions 42, 235 figurines of 316 lithic technology. See stone tools Little Foot 87, 186 living floors 94 Loiyangalani 540 Lothagam 139 Lower Palaeolithic 240, 243, 302, 304, 305, 307, 310, 321 Lucy 143, 185, 186, 191, 192 lufengpithecines 123 Lufengpithecus 124 Lukeino 129, 130, 138, 141, 192 luminescence dating 298, 461–466 Lupemban 177, 178, 362

lydianite 23, 24 lynx teeth 537

M Mabaget 138, 139 macaques 56 and language 289 Magdalenian 13 Makapansgat 86, 115, 247 dating 232 Makgadikgadi Pans 177 mammoths figurines of 316 images of 379 in Neanderthal diet 348, 351, 358–359 ivory 310, 313, 315, 318 Mandu Mandu Creek Rock Shelter 313 manganese 513, 521 Manis 263 mapping 159 Mapungubwe 164, 249 marine resources 306 Markina gora 537 marmot teeth 537 marrow extraction 241, 265, 349, 352 masks 310 mastic 302 Mbere Shelter 178 medicinal plants 514 Megantereon 42 Micoquian 396, 397, 401 microlithic tools 170, 172 microwear analysis 517–518 middens 490 Middle Palaeolithic 240, 243, 300, 302, 304, 305, 306, 307, 309, 310, 319, 320, 321, 336, 343, 349, 361, 362, 396, 454, 455 Middle Pleistocene Age 242, 243, 299, 300, 305, 306, 313 Middle Stone Age 153, 154, 157, 158, 167, 176, 178, 243,

569

From Tools to Symbols 267, 300, 302, 303, 304, 305, 306, 309, 310, 311, 320, 321, 336, 340, 361, 362, 420, 422, 423, 435, 442, 444, 446, 447, 448, 449, 450, 451, 452, 453, 454, 455, 456, 460–472, 476, 477, 480, 484, 485, 490, 495, 499, 500, 503, 535, 539, 540, 542 defined 179, 419 early 167, 168, 172, 175, 177, 179 late 168, 307, 312, 313 Middle to Upper Palaeolithic transition 85, 320, 344, 373, 375, 386 mitochondrial DNA. See mtDNA mobility 308 modern behaviour. See behavioural modernity molecular clock 123, 127, 137, 282 monkeys 92 mormyrid fish 40 most recent common ancestor 284, 286, 287 Motopithecus 137 Mousterian 13, 358, 373, 374, 391–397, 535, 537 Mousterian-Aurignacian boundary 344 MRCA. See most recent common ancestor Mrs Ples 110, 186 mtDNA 279, 281–284 Aboriginal 284 advantages of using 282 mud-wasps 46–47 multi-disciplinary research 1, 2, 8, 104–105, 242 Mumba Cave 312, 506, 540 musical instruments 318–319 mycohyphae 521

N Namurungule 137

570

Nassarius gibbosula shells 539, 540 Nassarius kraussianus shells 539 Natica shells 537 natural selection 91 Neanderthals 373 adaptations 335, 353, 354 anatomy 350, 351, 352 and Homo sapiens 385–386 as makers of art 317–318 body form 350 bone tools 304 burials 309, 375 coexistence with modern humans 320 compared to Inuit 349–351, 352 consciousness 378 diet 305, 348, 349, 351, 353, 358–359 DNA 283–284 evolution 344–345, 349–351 extinction 288, 320, 321, 334–336, 344, 357, 361, 363, 373 fat in diet 350–353 hunting strategies 335, 336– 340, 343–344, 354–357, 360, 363 hybrids 288 incidence of trauma 350 innovativeness 404 language 386 long-term planning by 405 modern behaviour of 296, 321, 335–336, 373, 374, 390, 391, 401, 407–409 mortality rates 350 muscle mass 350 ornaments 313, 320, 374–375, 535, 541 scavenging 336–340, 343–344 settlement 307 sites 243 skin colour 353, 354

social strategies 374–375, 378 stature 354 stone tools 299, 360, 374, 392, 401 Near East 299, 302 Nelson Bay Cave 249, 497 Neolithic, defined 13 nest-building 46–47 Neumark-Nord 302 Ngorora molar 123–127, 129, 130, 137, 144 non-Levallois blade production 299, 423, 429 non-shivering thermogenesis 350, 352 North Africa 16–17 nuclear DNA 280–281 nut-cracking, chimpanzee 60–62, 84, 241

O ochre 310, 311, 513, 517, 519, 521–522, 535, 537 and hafting 519 engraved 446, 454, 460, 494, 506 in beads 453 octopus 47 Oldowan 93, 94, 108, 153, 154, 230, 235, 266, 299 Olduvai 86, 88, 93, 94, 95, 96, 105, 106, 108–110, 116, 199, 230, 231, 232, 234 bone tools 256–260, 263–264, 265, 266, 267 climate 109 Event 232, 234 fauna 105 palynology 108 sedimentology 108 stone tools 108 vegetation 108–109 Olieboomspoort 249 Omo 86, 96, 114, 199

Subject Index operational chain. See chaîne opératoire optically stimulated luminescence dating 461–466 orang-utan 47, 201, 217 and language 289 brain size 86–87 divergence from human lineage 287 self-decoration 528 traditions 84–85 Oreopithecus 124 organic technology. See wooden tools, bone tools ornaments 311–313, 374–375, 526–542 bone 243 functions of 527–533, 541–542 Orrorin 145 teeth 128, 129, 130, 142 Orrorin tugenensis 88, 126, 127, 141–142, 144, 192 bipedalism 142–143 postcranial mophology 142–143 tool-making 193, 194 Orycteropus 263 Osteodontokeratic culture 242 osteoporosis 352–353, 354 ostrich eggshell 311 bead manufacture 533–535 beads 312, 449, 529, 539, 540, 541 incised 494 marked 476–490 Otavipithecus namibiensis 136 otters 48, 83–84 Oued Djebanna 540 Ouranopithecus 124 Out of Africa theory 279, 282–283, 284, 286, 287, 334, 361–363, 391, 454 outgroups 205

P paintings 313, 315, 316–317, 373, 476 palaeobotany 105–116, 483 palaeoecology 105–116, 140 Palaeolithic 299, 307, 526 defined 13 research 29 vs Stone Age 13 See also Lower, Middle and Upper Palaeolithic palaeomagnetic data 232 palynology 106, 107, 108, 115 See also pollen Pan paniscus. See bonobo Pan troglodytes. See chimpanzee panda 48 pangolin 263 Paranthropus 153 See also Australopithecus Paranthropus boisei 108, 233, 234 Paranthropus robustus 154, 186, 191, 230, 231, 233, 234, 235 dating 232 thumb 193 tool-making 231, 234–235, 266 Parma Farms 175–176 parrots 48 Pasalar 126 Pech-de-l’Azé 310, 392, 401, 405, 406, 407 Pecten shells 537 Peers Cave 309, 311, 456, 503 pelvis 186–187 pendants 313 percussion 260, 265, 349, 425–426, 428, 430 flaking 240 instruments 319 marks 343 perforations in beads 537, 539, 540 in bone 258 in shells 453 photosynthetic pathways 105–106 phylogeny 127

phytoliths 106 picks 167, 169, 172, 174, 177, 179 pigments 309–310 See also manganese, ochre plant residues 514, 518, 520, 521 platform cores 299, 425, 430, 437 Pleistocene Age 299, 393 Plovers Lake 153, 157–158 points 172, 173, 175, 177, 179, 241, 300, 302, 420, 437, 446, 480, 484, 487, 494, 503, 514, 517 bifacial 450 bone 243, 249 polishes 517–518 pollen 106, 107, 111–112, 114, 115, 520 See also palynology pollical distal phalanx. See thumb Pongo pygmaeus. See orang-utan postcranial bones 88–89, 185–191 pottery 170, 178 Praeanthropus 128, 188 Praeanthropus africanus 127, 140 predation 40, 41–42, 45, 46 predators 105 premature discoveries 95 primary consciousness 377 primates 1, 2, 41, 53, 54, 56, 90 brain-to-body weight ratio 40 culture 56 encephalisation 89 evolution 287 primatology 56, 57, 58 Proconsul africanus 139 Proconsul nyanzae 142 proconsulines 123 protein, in diet 263, 358 Proteles 263 pseudotools 242, 244, 247, 256, 258, 260, 264–265 punctures, in bones 258

571

From Tools to Symbols

Q Qafzeh 310, 311, 537, 541 quartz 154, 155, 158, 166, 167, 460, 479, 480, 500, 504, 515 quartzite 23, 157, 158, 166, 423, 428, 476, 479, 484, 500, 504, 540 Quina 396 Quincay 243, 537 quipu 533

R race 279–280 radiocarbon dating 297–298 rainfall 107 Ramapithecus 124, 126, 136 raw materials 23, 24–26, 27, 153, 166, 362, 407, 428, 479, 494 availability of 542 distribution of 24 transport of 307–308, 360, 405 See also flint, dolerite, quartz, quartzite, shale red deer teeth 537, 539 refitting 423, 424, 488 reindeer images of 379 teeth 537 religious beliefs 310, 386 rendering, fat 349, 352 replication 32, 243, 247, 250, 256, 258, 260, 264, 426 See also wear patterns representational art 313–318 residue analysis 513–522 resin 513, 514 retouched tools 158, 171, 172, 175, 177, 240, 242, 392, 396, 405, 420, 429, 436–437, 480, 514, 517, 519 rhinocerotids 106, 484 Rhodesian Proto-Stillbay 177 rhyolite 166 Rietvlei Dam 111 right-handedness 49

572

Riparo Mochi 312 ritual 383, 530 specialists 382 Roc de Combe 537, 541 Roche au Loup 537 rodents 92, 105, 111, 484 Rose Cottage 249, 307, 321, 503, 505

S sabre-toothed cats 42, 235 Sahelanthropus 123, 128, 130, 145 Sahelanthropus tchadensis 88, 125, 143–144 teeth 144 Sahul, colonisation of 313 Saint-Césaire 537 Salzgitter-Lebenstedt 304, 306 Samburu Hills 137 Samburupithecus 130 Samburupithecus kiptalami 125, 137, 138 San 21, 22, 27, 533, 542 arrow points 249–250 link shafts 250 mtDNA 283 See also hunter-gatherers sandstone 166, 168 Sangoan 167, 168, 172, 174, 175, 176, 177–178, 179 savoir-faire 61 scaling 89, 90 scavenging 305, 306, 336–340, 343–344, 347–348 arguments for 337–340 schist 539 Schöningen 302, 305 scrapers 167, 171, 172, 177, 179, 374, 392, 398, 406, 446, 514, 517 seasonal activities 255 sedimentology 108 seeds 108, 520 Seggédim 540 selective pressure 40 self-consciousness 377

Semliki 112 sensory deprivation 383 sequential processing 49 Serengeti 108 settlement as indicator of behavioural modernity 307–308 patterns of 307–308 sexual dimorphism 207, 220, 223 shale 23, 24, 155 shamans 382, 530 sheep bones 477 shell beads 452–454, 460, 535, 537, 538, 539, 540, 541, 542 shellfish 446, 490 shells 311, 312, 313, 442, 446, 448, 461, 497, 503, 520 Columbella rustica shells 539 Conus shells 541 cowrie 532 Dentalium shells 537 Glycymeris shells 537, 539 Mercenaria mercenaria 532 middens 490 Nassarius gibbosula shells 539, 540 Nassarius kraussianus shells 539 Natica shells 537 Pecten shells 537 Trochus shells 537 Turbo 531 Turbo sarmaticus shells 461 Turitella shells 537 shelters 307 Sibudu 321, 503, 513, 514–522 silcrete 480, 504 silica polish 518 single nucleotide polymorphisms 280 site functions 28 sivapithecines 123 Sivapithecus 124, 130 Skildergat 456 skin colour 352–353, 354 slate 157 Smithfield 24

Subject Index SNPs 280 social differentiation 376, 378, 383–385, 386, 526, 529–530, 542 sociological perspective 27 soil analysis 519–520, 522 Solutrean 494 South Africanisation 20 spatial organisation, modern 307 spears, wooden 302, 305 speech areas 93–95 speech. See language spheroids 154, 155 split, ape-hominid. See dichotomy spores 106 squids 47 squirrel monkeys 40 starch grains 513, 514, 519, 520–521, 522 stature 89 Sterkfontein 42, 86, 87, 89, 96, 110–112, 115, 116, 153, 157, 158, 186, 193, 199, 230, 232, 243, 244, 250, 265 Still Bay 310, 311, 362, 442, 444–446, 450, 456, 460, 494, 503–505, 539 Stone Age 14, 180 vs Palaeolithic 13 See also Earlier, Later and Middle Stone Age stone tools 13, 21, 22, 23, 45, 95, 108, 112, 153, 154, 155, 158, 166, 172, 174, 175, 194, 230, 234, 235, 240, 261, 307–308, 360, 361, 374, 419–438, 444–446, 456, 477, 478–480, 489 as indicators of behavioural modernity 298–302 backed 420, 429, 488, 489, 494 conjoined 168 formal 169 materials 23 microlithic 170, 172 Mousterian 392–407

residues on 513–522 retouched 158, 171, 172, 175, 177, 240, 242, 392, 396, 405, 420, 429, 436–437, 480, 514, 517, 519 Swartkrans 253, 254 stone figurines 315 rings 539–540 storage, food 351, 357, 360, 488–489 Stratzing 317 striations 517–518 striking platform 422 subsistence, patterns of 305–306, 345–347, 374–375, 460, 484, 488, 513 as indicators of behavioural modernity 306 suids 111, 154, 306, 340 teeth 256 Sungir 309 Swartkrans 40–45, 86, 115, 153, 154, 157, 186, 194, 199, 230, 231, 234, 243, 244 bone tools 243–256, 264, 267 dating 232 symbolic communication 310–311, 321, 408, 489, 491, 506, 522, 526 synapomorphies 216

T Tabarin mandible 139 Tachnegit 24, 31 Tan Tan 313 tanning leather 310 taphonomy 105, 106 Tata 310 Taung 86, 115, 230 teaching 60–61 technological competence 46–49 teeth australopithecine 125–126, 127, 233, 234

cave bear 382 chimpanzee 125–126, 127, 129, 142, 144 enamel 126, 128, 130, 137, 139, 234 gorilla 125, 127, 128, 129, 144 human 125–126, 127 Ngorora molar 123–127 Orrorin 128 used as beads 535 used as ornaments 313 terminology 2, 19, 362 African 11, 14, 18, 20 European 13 lithic 12, 24 termite mounds 250 excavating 247, 263, 266 termite-fishing 241 theodolites, laser 159 therianthropes 313, 316, 375 thermogenesis, non-shivering 350, 352 thumb anatomy 143, 185, 191–193, 231 chimpanzee 192 hominid 231 human 192 Tönchesberg 302, 306 tool use 83, 230, 522 in animals 48, 58, 83–84, 184 in birds 48–49, 83 tool-making 28, 83, 192, 193, 230, 234 Kromdraai 231 tools 54 bone. See bone tools broken 519, 521 composite 300 conjoined 168 digging 258 formal 169, 172, 241, 362, 429, 514, 522 glass 22 hide-working 258, 262

573

From Tools to Symbols identifying true 242 leatherworking 258, 262 microlithic 170, 172 pseudotools 242, 244, 247, 256, 258, 260, 264–265 retouched 158, 171, 172, 175, 177, 240, 242, 392, 396, 405, 420, 429, 436–437, 480, 514, 517, 519 stone. See stone tools wood. See wooden tools tooth enamel 126, 128, 130, 137, 139, 234 trade 348, 527, 531–532, 541–542 tradition 56, 57, 58 animal 59, 529 internalisation of 57 transition Earlier Stone Age-Middle Stone Age 178 Middle-Upper Palaeolithic 85, 320, 344, 373, 375, 386, 454, 479 Middle Stone Age-Later Stone Age 479, 494 transitional industries 179 transmission 95 tree shrews 92 Trilobite Cave 537 Trochus shells 537 tropical environments 345–347, 347 Tswaing 111 tubes, bone 243 tuff 106, 108, 115 Tugen Hills 141 Tunisia 17 tunnels 380 Turbo sarmaticus shells 461 Turitella shells 537 Turkana 139, 199 Twiggy 94 Twin Rivers 309, 506 typology 27, 28, 178, 179, 436–437, 479, 480

574

U Üçagizli 312, 539 Ugandapithecus 142 Uluzzian 535, 537 Umm-el-Tlel 302 unifaces 168, 172 unifacial points 484, 487, 514 Upper Palaeolithic 240, 241, 297, 299, 300, 302, 312, 317, 318, 319, 321, 343, 383, 393, 404, 422, 539, 541, 542 transition 85, 320, 344, 373, 375, 386, 454, 479 Upper Pleistocene Age 242, 300, 306, 491 Uraha 199 usage. See replication, tool use, wear patterns use-wear analysis. See wear patterns

whales 47, 92 wild dogs 42 wildebeest 235, 484 Wilton 494 wolf teeth 537, 539 women’s status 361 Wonderkrater 111 wood 106, 480, 484 fossil 108, 109, 111, 112, 115 wooden tools 112, 302–303, 305, 514 as indicators of behavioural modernity 302–303 woodworking 177 workshops 28, 29, 405 writing systems 532

X X-chromosome DNA 287

V vegetation 105, 106 vertebrae 88, 186 Victoria West 23, 24, 31 vitamin D deficiency 352–353, 354 Vogelherd 304, 315, 317, 379 volumetric conception 401–404, 422, 428, 437 vortex 380

W Wallertheim 306 wampum 532 waste products 27, 28, 29, 424, 514 water flasks 487, 488, 490 wear patterns 243, 244, 247, 250–251, 253, 256, 258, 259, 260, 264, 397–400, 406, 454, 517–519, 533, 535 experimental 264 See also replication Wernicke’s area 93, 94, 95 West Rift 112

Y Yabrudian 299 Y-chromosome DNA 279, 281–282, 284–286

Z Zagros 353 Zinjanthropus 230 Zinjanthropus boisei 108, 234 Zombepata Cave 539–540