The Origins of Human Behaviour...development of social hierarchy and political centralization J.Gledhill et al. (eds) The walking larder: patterns of domestication, pastoralism, and

THE ORIGINS OF HUMAN BEHAVIOUR

TITLES OF RELATED INTEREST

Animals into artHMorphy (ed)Archaeological approaches to cultural identitySJShennan (ed)Archaeological heritage management in themodern worldHFCleere (ed)Centre and periphery comparative studies inarchaeologyTCChampion (ed)Chalcolithic Bronze and Iron Age cultures inSouth AsiaMLal (ed)Conflict in the archaeology of living traditionsRLayton (ed)Domination and resistanceDMiller et al (eds)The excluded past archaeology in educationPStone amp RMackenzie (eds)Food metals and towns in African historyAfrican adaptations in subsistence andtechnologyTShaw et al (eds)Foraging and farming the evolution of plantexploitationDHarris amp GHillman (eds)From the Baltic to the Black Sea studies inmedieval archaeologyDAustin amp LAlcock (eds)Hunters of the recent pastLDavis amp BOKReeves (eds)

The meanings of things materialculture and symbolic expressionIHodder (ed)The politics of the pastPGathercole amp DLowenthal (eds)Signifying animals human meaning inthe natural worldRGWillis (ed)State and society the emergence anddevelopment of social hierarchy andpolitical centralizationJGledhill et al (eds)The walking larder patterns ofdomestication pastoralism andpredationJClutton-Brock (ed)What is an animalTIngold (ed)Whatrsquos new A closer look at theprocess of innovationSEvan der Leeuw amp RTorrence(eds)Who needs the past Indigenous valuesand archaeologyRLayton (ed)The world at 18 000 BP highlatitudesOSoffer amp CGamble (eds)The world at 18 000 BP low latitudesCGamble amp OSoffer (eds)

THE ORIGINS OFHUMAN BEHAVIOUR

Edited by RAFoleyDepartment of Biological Anthropology University of Cambridge

London and New York

copy RAFoley and contributors 1991This book is copyright under the Berne Convention No reproductionwithout permission All rights reservedPublished by the Academic Division ofUnwin Hyman Ltd1517 Broadwick Street London W1V 1FP UK

Unwin Hyman Inc955 Massachusetts Avenue Cambridge MA 02139 USA

Allen amp Unwin (Australia) Ltd8 Napier Street North Sydney NSW 2060 Australia

Allen amp Unwin (New Zealand) Ltdin association with the Port Nicholson Press LtdCompusales Building 75 Ghuznee Street Wellington 1 New Zealand First published in 1991 Routledge is an imprint of the Taylor amp Francis Group This edition published in the Taylor amp Francis e-Library 2004

British Library Cataloguing in Publication Data

The origins of human behaviourmdash(One world archaeology)1 Man Behaviour OriginsI Foley Robert II Series1557

ISBN 0-203-16875-5 Master e-book ISBN ISBN 0-203-26400-2 (Adobe eReader Format)ISBN 0-04-445015-X (Print Edition) Library of Congress Cataloging in Publication DataThe Origins of human behaviouredited by RAFoleyp cmmdash(One world archaeology 19)Includes bibliographical references and indexISBN 0-04-445015-X $5500 (US)1 Human evolution 2 Human behaviourmdashOrigin3 Behavior evolution I Foley Robert II SeriesGN2814075 19915732ndashdc20 90ndash12920 CIP

List of contributors

RAFoley Department of Biological Anthropology University ofCambridge UKPhillip JHabgood Department of Anthropology University of SydneyNSW AustraliaSheldon Klein Computer Sciences Department University ofWisconsin-Madison USAWCMcGrew Department of Psychology University of Stirling UKFrancis BMusonda Livingstone Museum Livingstone ZambiaThomas Wynn Department of Anthropology University of Colorado atColorado Springs USAEzra BWZubrow Department of Anthropology University of New York atBuffalo USA

Foreword

This book is one of a major series of more than 20 volumes resulting from theWorld Archaeological Congress held in Southampton England in September1986 The series reflects the enormous academic impact of the Congress whichwas attended by 850 people from more than 70 countries and attracted manyadditional contributions from others who were unable to attend in person

The One World Archaeology series is the result of a determined and highlysuccessful attempt to bring together for the first time not only archaeologistsand anthropologists from many different parts of the world as well as academicsfrom a host of contingent disciplines but also non-academics from a wide rangeof cultural backgrounds who could lend their own expertise to the discussionsat the Congress Many of the latter accustomed to being treated as the lsquosubjectsrsquoof archaeological and anthropological observation had never before beenadmitted as equal participants in the discussion of their own (cultural) past orpresent with their own particularly vital contribution to make towards globalcross-cultural understanding

The Congress therefore really addressed world archaeology in its widestsense Central to a world archaeological approach is the investigation not onlyof how people lived in the past but also of how and why changes took placeresulting in the forms of society and culture which exist today Contrary topopular belief and the archaeology of some 20 years ago world archaeology ismuch more than the mere recording of specific historic events embracing as itdoes the study of social and cultural change in its entirety All the books in theOne World Archaeology series are the result of meetings and discussions whichtook place within a context that encouraged a feeling of self-criticism andhumility in the participants about their own interpretations and concepts of thepast Many participants experienced a new self-awareness as well as a degree ofawe about past and present human endeavours all of which are reflected in thisunique series

The Congress was organized around major themes Several of these themeswere based on the discussion of full-length papers which had been circulatedsome months previously to all who had indicated a special interest in themOther sessions including some dealing with areas of specialization defined byperiod or geographical region were based on oral addresses or a combinationof precirculated papers and lectures In all cases the entire sessions wererecorded on cassette and all contributors were presented with the recordings ofthe discussion of their papers A major part of the thinking behind the Congresswas that such a meeting of many hundreds of participants that did not leavebehind a published record of its academic discussions would be little more thanan exercise in tourism

x FOREWORD

Thus from the very beginning of the detailed planning for the WorldArchaeological Congress in 1982 the intention was to produce post-Congressbooks containing a selection only of the contributions revised in the light ofdiscussions during the sessions themselves as well as during subsequentconsultations with the academic editors appointed for each book Particularlyin the case of sessions based on precirculated papers all contributors were awareof the subsequent publication production schedulesmdashif their papers wereselected for publication they would have only a few months to revise themaccording to editorial specifications and they would become authors in animportant academic volume scheduled to appear within a reasonable periodfollowing the Southampton meeting

The origins of human behaviour results from the four and a half days at theCongress of discussions of precirculated papers and verbal contributionspresented under the overall title lsquoThe Pleistocene Perspectiversquo organized byProfessor Michael Day Dr Robert Foley and Mr Arthur ApSimon Morespecifically it derives from the Congress subtheme on lsquoHominid Behaviour andEcologyrsquo which was organized by the editor of this book Other contributionsoriginally linked to this overall theme of the Congress have been published inHunters of the recent past (edited by LBDa vis amp BOKReeves) The world at 18000 BP low latitudes (edited by CGamble amp OSoffer) and The world at 18 000BP high latitudes (edited by OSoffer amp CGamble)

Unfortunately the publication of this book has been much delayed and isthe last of those in this series to concentrate on the Pleistocene (those onlsquoHuman Evolutionrsquo lsquoHominid Dispersal Patternsrsquo and lsquoAdaptations at aroundthe PleistoceneHolocene Boundaryrsquo having very reluctantly been laidaside) The publication of The origins of human behaviour reflects therefore thecommitment by the World Archaeological Congress to palaeoanthropologyand palaeolithic archaeology despite the very complex and troubled history ofthe organization of sessions on lsquoThe Pleistocene Perspectiversquo at the 1986Congressmdashreviewed in detail elsewhere (Ucko 1987 pp 24 34 39ndash40 6164 82 125 129 131ndash2 144 151 214 222 224 235 247 249 266ndash9)mdashwhich inevitably led to a lack of thematic coherence and a somewhat variablequality in the papers presented at that time Since then numerous attemptshave been made to create several thematic books on the Pleistocene from themany 1986 contributions but finally it has been decided to concentrateexclusively on the present book

The origins of human behaviour is much more than simply a symbolicpublication Its concerns are central to many of the overall perceptions ofthe One World Archaeology series focusing attention on such very basicquestions as what constitutes lsquoculturersquo whether such a term is merelydescriptive or whether it may be used as an analytical tool and how it canmdashor cannotmdashserve to qualify the lsquohuman conditionrsquo As such The origins ofhuman behaviour is implicitly and necessarily concerned with concepts suchas innovation discovery and diffusionmdashmaking it an interesting companionto Whatrsquos new A closer look at the process of innovation (edited by SEvan derLeeuw amp RTorrence)mdashand it challenges the reader to determine whether

xiFOREWORD

there is any reason against assuming their equal importance to the non-human sphere

Likewise this book also has much of importance to say about the subjectmatter of three other One World Archaeology books Animals into art (edited byHMorphy) Archaeological approaches to cultural identity (edited by SJ Shennan)and What is an animal (edited by TIngold) With regard to the first the readeris urged to consider in detail the implications for the evolution of humanorganization and human thought of assumptions about any regular use of lsquorulesrsquoin the schemata of Upper Palaeolithic artworks With regard to the second theproblems of using archaeological evidence to establish whether or not therereally were two distinct forms of humans living contemporaneously apart ortogether should caution against too easy assumptions about ethnic visibility inthe archaeological record In the last casemdashif culture is not an exclusivelylsquohumanrsquo phenomenonmdashwhat really are the distinguishing points (if any)between human and animal

The nature of these kinds of enquiries are both wide-ranging andcomparative far removed from the climate of much prevalent opinion aboutlater archaeological material which currently stresses the necessity of inward-looking intracultural analysis and interpretation It is exciting therefore to findrenewed interest in other generalizing comparative approaches to humanactivity such as the possible relevance of Piagetian constructs to archaeologicaldata and their interpretation

This book therefore foreshadows other attempts (eg Ucko 1990) toreinvestigate the possible relevance to current archaeological theory of grandschemesmdashmainly constructed in the 19th century by psychologists as well asanthropologists and archaeologistsmdashabout the human past all being based onwhat has often been referred to as lsquothe comparative methodrsquo It is fascinating todiscover from this book how the need to investigate the mechanisms of non-genetic generational transmission of behavioural characteristics remains ofcentral concern to current interpretations of human behaviour in thePleistocene Classic lsquorecapitulationrsquo theories may be long outdated in biologicalterms but curiosity about an assumed similarity between the development froman infant to an adult and the growth of human lsquoculturalrsquo complexity is still inevidence

The origins of human behaviour is a book which challenges the archaeologist toreconsider the appropriateness of many interpretative models in use in thisdiscipline for periods much later than the Pleistocene It also however serves asa challenge to its own practitioners in publicly stressing the common humanityof its data base and therefore claiming an almost Olympian detachment fromthe political and social debates surrounding contemporary human rightsPleistocene studies have remained very much the exclusive domain of specialistsfrom or influenced by a restricted Western background If wide-rangingcomparative methodology is to be its forte such comparative schemata cannotafford to be based on assumptions deriving exclusively either from ethology orfrom presumed linear progressions of social development The challenge is torefrain from sheltering behind the chronological remoteness of the material

xii FOREWORD

under study andmdashas some (Cann et al 1987) have begunmdashto bring it into thepublic domain demonstrating its relevance to those living in the present

PJUckoSouthampton

References

Cann RL Stoneking M amp Wilson AC 1987 Mitochondial DNA andhuman evolution Nature 325 31ndash6

Ucko PJ 1987 Academic freedom and apartheid the story of the World ArchaeologicalCongress London Duckworth

Ucko PJ 1990 Whose culture is it anyway Frazer Lecture University of Glasgow

Contents

List of contributors viiForeword PJUcko ixPreface RAFoley xvIntroduction investigating the origins of human behaviour RAFoley 1

1 Chimpanzee material culture what are its limits and whyWCMcGrew 13

Introduction 13Culture and symbols 15Environment and adaptation 16Diet 17Home bases 19Technology 19Conclusions 21

2 How useful is the culture concept in early hominid studiesRAFoley 25

Introduction 25Definitions of culture 26The use of culture in palaeoanthropology 26The inadequacy of culture 27Those damned chimpanzees 29The evolution of complex and flexible behaviour 30Conclusions 35

3 The significance of modern hunter-gatherers in the study of earlyhominid behaviour Francis BMusonda 39

Introduction 39Environmental setting 39Settlement patterns 41Subsistence activities 42Social organization 45Conclusions 47

4 Archaeological evidence for modern intelligence Thomas Wynn 52

Introduction 52

xiv CONTENTS

Archaeology and intelligence 53A Piagetian approach to prehistoric intelligence 54Concrete operations 56Archaeological evidence for concrete operations 57Formal operations 58Archaeological evidence for formal operations 59Critique of formal operations 63Conclusions 65

5 The invention of computationally plausible knowledge systems in theUpper Palaeolithic Sheldon Klein 67

The problem of computing human behaviour by rules 67The basic structure of the invention 68ATOs language and culture 76ATOs and the ontogeny of shamanism 77The evidence of Leacutevi-Strauss 78Testing the ATO model in historical time 78Conclusions 79

6 An interactive growth model applied to the expansion of Upper Palaeolithicpopulations Ezra BWZubrow 82

The background 82The model 83Results from the model 86Conclusions 95

7 Aboriginal fossil hominids evolution and migrationsPhillip JHabgood 97

Index 115

Preface

The World Archaeological Congress meetings in Southampton in September1986 included a series of sessions on the problems of Pleistocene archaeologyThe chapters in this book derive from some of those discussions While theoriginal meetings were extremely diverse this volume focuses on the problemsthat face prehistorians and palaeoanthropologists trying to understand the long-term evolution of human behaviour and the patterns observable in the fossil andarchaeological record of a period of time stretching over several million years Itis not the intention to present a comprehensive analysis of the origins and evolutionof modern human behaviour but rather to illustrate the diversity of approachesand concepts that are required if we are to unravel what must surely be the mostcomplex of problems facing archaeologists and evolutionary biologists

Much gratitude is owed to the many people who contributed to the originaldiscussions and made the meeting a memorable one as well as to the organizersespecially Peter Ucko without whose drive little would have been achieved Iwould also like to express my appreciation to those who chaired sessions duringthe Congress Paul Callow Michael Day Clive Gamble Wu Rukang ChrisStringer and Jiri Svoboda Thanks must also go to Marta Lahr and Harriet Eeleyfor their helpful comments on many of the ideas discussed in this book

RAFoleyCambridge

Introduction investigatingthe origins of human behaviourRAFOLEY

It is hard to find a branch of anthropology and archaeology as disparate as thestudy of the origins and evolution of human behaviour which may be referredto as behavioural palaeoanthropology

At the outset there is the question of scale The closest living relatives ofhominids are the African apes and most probably the chimpanzee (Pantroglodytes and Pan paniscus) Current evidence suggests that the split from Panoccurred between 5 and 8 million years ago (Holmes et al 1989) Thedifferences between these two groups of hominoids therefore developed over aperiod of several million years a timescale of unique length withinanthropology (although relatively short on a general palaeontological andevolutionary scale) Such a timescale makes it very difficult to conceptualizeprocesses over long periods to explain long-term events and to document thetiming and sequence of the major developments As a result the question ofwhether the shift from archaic to modern hominids was of evolutionarysignificance in itself or whether the principal evolutionary changes occurredwith the appearance of the genus Homo some 2 million years earlier (Foley1989) for example remains problematic

Scale is not simply a chronological consideration Variability is central to allscience and anthropology is no exception However most of anthropology(including prehistoric archaeology) is concerned only with the development ofmodels to descr ibe and theories to explain intraspecific var iabilityPalaeoanthropology though must treat not just intraspecific variability but alsovariations across species genera and even higher taxonomic levels while at thesame time employing where necessary techniques concepts and assumptionsdeveloped both for later periods and for cultural rather than biologicalvariability in behaviour

Next comes the difficulty of lines of evidence As McGrew points out (Ch1) behaviour does not fossilize but must be inferred indirectly from either fossilhominid morphology or archaeological remains Both the fossil andarchaeological records are notoriously incomplete This is true of both thefrequency with which prehistoric hominids and their activities are preservedand discovered and the range of such activities that are subject to fossilizationprocesses We have access to only a minuscule proportion of the hominids thathave ever lived and to only a limited range of the results of their activities thesebeing confined principally to subsistence and technology

2 INTRODUCTION

This has led to a recognition that supporting evidence must come fromother indirect sources In practice this means some sort of analogical reasoningbased on living groups of humans or animals But here again difficulties ariseWhat are appropriate analogues for events and processes in early hominidevolution Do modern humans provide suitable material for drawing parallelsTraditionally living hunter-gatherers have been used to lsquoflesh out the fossilrecordrsquo but it is clear as Musonda discusses (Ch 3) that there are as manydifferences between these groups and earlier hominids as there are similaritiesthey are members of a different species they have undergone their ownsubsequent evolutionary and historical developments and they are the productof specific historical cultural and environmental contexts (McGrew 1990)Furthermore their technology is significantly different from that of prehistoricnonmodern human populations Simplistic hunter-gatherer models have beenheavily criticized for being based on outmoded notions of cultural evolutionthat place living hunter-gatherers in a primitive ancestral state

The alternative source of analogue models is the nonhuman primates butmany difficulties arise here as well While we share a close phylogeneticrelationship and evolutionary history with other primates they too haveevolved in response to their own unique evolutionary context Apes andmonkeys are not just hominids manqueacutes they are species as well adapted to theirniches as the hominids were to theirs Analogies may be based on eitherphylogenetic or environmental similarity or both Historically chimpanzees andbaboons respectively have been taken as models for the two roles but againthere are difficulties with this approach because evolution is the result ofinteraction between phylogeny and environment and for every species thisproduces a unique set of circumstances

Even if such analogies are useful the next question that arises is what exactlyis it in hominid evolution that ones wishes to model and seek an explanation forA wide variety of distinctive human features have been identified and proposed ascrucial to the evolution of the hominids from a more primitive ape precursorThese include bipedalism enlarged brains and intelligence tool-making food-sharing language consciousness enlarged social networks hunting territorialityand so on Histor ically there has been a shifting emphasis amongpalaeoanthropologistsmdashfor example from brains and technology to bipedalismand subsistencemdashthat reflects a variety of factors both internal and empirical andexternal and political As a result a wide range of social behavioural andecological traits have been selected as the key to the change from animal tohuman behaviour and work on the origins of human behaviour has necessarilybeen spread very thinly with little consensus about where the main issues lie orperhaps more importantly how the behavioural variables are related to each otherFor example is speech significant in the processes of encephalization thatoccurred during hominid evolution and how does it relate chronologically andfunctionally to other attributes such as conscious thought hunting behavioursocial cooperation etc While McGrew shows that the range complexity andvariation of chimpanzee behaviour are greater than previously thought (Ch 1)nonetheless analogical approaches remain problematic

3INTRODUCTION

Even taking into account these essentially empirical methodological andtechnical questions there remains yet one more black box for thepalaeoanthropologist to try to peer into that of explanation The study ofhuman evolution especially its behavioural aspects is on the cusp of the socialand natural sciences On one side lie essentially Darwinian explanations andtheories couched in terms of natural selection reproductive advantage costsand benefits and evolutionarily stable strategiesmdashthe terms and concepts ofbehavioural ecology and sociobiology On the other side lies social and culturalanthropology with its armoury of functionalism and structuralism culturalecology and cultural relativism social theory and societal norms In between lieapproaches that have attempted to integrate evolutionary and cultural theories(Shennan 1989) The problem is reminiscent of a large and complex landscapedivided by a deep and meandering river It is that river that somewhere duringthe course of their evolution hominids are thought to have crossed But whenand how Until these questions are answered what sort of explanation shouldbe sought for hominid behavioural evolutionmdashDarwinian or culturalmdashwillremain unclear (In fact at present there is not even any consensus that hominidsever crossed the river at all some feel they are still firmly placed on thesociobiological banks)

The study of the origins of human behaviour then is a difficult subjectbeset by problems of temporal and taxonomic scale imperfect and incompleteevidence an uncertain comparative framework and competing explanatorysystems Given all that behavioural palaeoanthropology may be likened toSamuel Johnsonrsquos comments on bipedal dogs lsquoIt is not done well but you aresurprised to find it done at allrsquo

Great advances though have been made in turning a subject that ispotentially an exercise in hindsight into an empirically testable project with itsown procedures (Kinzey 1987) Three factors are most probably responsibleFirst there has been a considerable improvement in the fossil record of hominidevolution This statement is not simply a reiteration of the oft-heard claim thatthe jigsaw of prehistory is gradually being filled in but is based on one aspect ofrecent developments in palaeoanthropology This is that as more fossils havecome to light it is clear that hominid evolution is not a simple unilinear processfrom the primitive to the advanced rather it is a complex process involvingcladogenetic speciation extinction and coexistence of species As a result of thissingle discovery it is no longer possible to be vague about phases of hominidevolution As McGrew points out (Ch 1) if we are to use chimpanzees as amodel for earlier hominids we should know for which hominid they areappropriatemdashcommon ancestor australopithecine early Homo Behaviouralpalaeoanthropology must explain variability in terms other than thechronological

Second field studies of primates have provided a vast amount of informationabout the evolutionary and behavioural ecology of complex and highly socialmammals This has been vital in opening up discussion about early hominidsbeyond such vague issues as male dominance and large groups Of particularimportance has been the move away from descriptions of stereotypic species

4 INTRODUCTION

behaviour towards the recognition that behaviour is variable and flexible withinspecies according to such factors as age sex rank and ecological conditions(Dunbar 1988 Standen amp Foley 1989) This means that application of theseresults to the early hominids can go beyond single species models (lsquothe baboonanalogyrsquo) to deal with issues of individual reproductive strategies and life historycorrelates

The third and related factor is the development of evolutionary theory itselfOf particular importance has been the increasing recognition of two facts thatbehaviour evolves through the mechanism of natural selection and thatbehavioural evolution can play a significant part in determining overall patternsof evolution (Bateson 1988 Shennan 1989) The integrating of behaviouralespecially social (Humphrey 1976) evolution within evolutionary biology as awhole has taken human evolution with its special behavioural and socialconsiderations away from the margins of the subject and into the mainstream

What is perhaps most interesting is that these critical developments havecome not from archaeology theoretically the subject most directly concernedwith the behaviour of early hominids but from the adjacent field ofevolutionary biology Certainly there have been critical developments both inmethodology and in empirical results within archaeology (and nowhere has thisbeen more significant than in studies of early hominid meat-eating behaviour)but the key development has been the recognition that evolutionary biology hasthe power to explain extremely complex behavioural phenomena withoutrecourse to the lsquospecial casersquo of humans that has dogged research into ourorigins

It is important to understand the background to the problems currentlyfacing those researching into the origins and evolution of human behaviourHowever it is perhaps even more vital to determine the way forward Theproblem appears overwhelmingly insoluble Questions of lsquooriginsrsquo by definitionrelate to events in the past which can never be directly observed and thereforerest uneasily on the boundaries of practical scientific investigation If science isthe art of the soluble as Peter Medawar has claimed (1967) it may be that alldiscussions of human origins must remain speculative resting ultimately onphilosophical preference rather than empirical validation The origins of theuniverse of life and of humans it may be argued are beyond formal scientificinvestigation If this is so then the way forward may be no different from theway backmdasha procession of untestable theories

However the link between the origins of human behaviour and othercosmological problems may perhaps show a means of escaping this ratherdespairing scenario (Foley amp Dunbar 1989) Physicists and mathematiciansstudying the origins of the universe are faced with a very similar problem tothat of palaeoanthropologistsmdashthe investigation of events that took placemillions of years ago occurring under conditions very different from those oftoday These events cannot be directly observed but must be inferred from theobservation of remnant effects The universe is in a sense a fossil of the big bangand subsequent developments Indeed cosmologists study events even moreremote than human origins and conditions that differ greatly from those found

5INTRODUCTION

today And yet physicists have managed to reconstruct what happened and toreveal the principles underlying events at the beginning of the universe Theapproach they have used to achieve this may provide a useful pointer to the wayforward for palaeoanthropologists

The answer seems to lie in the use and nature of theory and in itsarticulation with empirical study Relativity and quantum mechanicsmdashpurelytheoretical formulations based on mathematical principlesmdashhave allowed afairly narrow set of models to be constructed These have enabled certainpossibilities to be ruled out They have also permitted precise predictions to bemade about the outcomes arising from certain models and hence thespecification of empirically observable entities and relationships that will occuronly if the conditions outlined in the models are fulfilled In other wordsmoving from simple description and classification to both prediction andexplanation is dependent upon the development of theory and the discipliningof that theory to the constraints of empirical investigation

It is obvious that the complexities of the biological world pose a vastlydifferent set of problems from the certainties of physics Biologists especiallythose working at the lsquosofter endrsquo of ecology and behaviour cannot even start toclaim that their theories are as powerful as those of the cosmologists Indeed itmay be argued that biology dealing as it must with events strongly affected byhistorical contingency can never hope to rival the description of the physicalworld However physical scientists have two important lessons for us The first isthat considerable progress can be made not by seeking a single answer in oneleap but by eliminating certain possibilities It may not at this stage be possibleto determine whether early hominids were monogamous or polygamous but itmay be possible to exclude other alternatives such as polyandry or asociality ontheoretical grounds alone (Foley amp Lee 1989 Lee 19889) The second is thelink between theories models and empirical observation Theories expose theprinciples we think underlie the events we are studying These may betransformed into models which are essentially conditional statements that ariseout of the theoretical axioms Their utility is based on the extent to which theycan specify outcomes to be expected if certain conditions are fulfilled If thoseoutcomes have empirical manifestations then formal testing becomes apractical possibility

As discussed above the expansion of evolutionary theory and especially thedevelopment of evolutionary and behavioural ecology now makes it possiblefor biologists to construct formal theoriesmdashsets of expectationsmdashrelating to thebehaviour and adaptations of complex organisms While these are not completeand may be particularly problematic when applied to modern humans theynonetheless provide a series of expectations Only with these expectations canwe hope to measure the extent to which humans may or may not conform tothe general principles underlying the behaviour of biological organismsFurthermore for those studying living organisms they can be formulated interms of empirically testable (and quantifiable) hypotheses (Dunbar 1989) Theextension of these to events and situations in the past is of course a formidabletechnical problem but it remains the only option That this is not a simple

6 INTRODUCTION

attempt to reduce all behaviour to a system of energetics is perhaps evidencedby the way in which behavioural ecology is itself while not abandoning itsbasic principles moving towards incorporating cognitive and other morecomplex parameters into its framework (Byrne amp Whiten 1986) The sameshould be expected for behavioural palaeoanthropology Klein gives an exampleof this (Ch 5) when he attempts to build a model of the structure of humanthought which is consistent with both evolutionary principles and Leacutevi-Straussian structuralism

In elaborating these methodological principles two further points should bemade The first is a question of discipline Major advances in science are seldommade by asking questions in a very general form Crick and Watson did notsolve the problem of the structure of genes by asking metaphysical questionsabout the nature of heredity but by recognizing that certain key properties mustbe involved this in turn led them to focus on the exact molecular structure ofchemicals in the nucleus of the cell (Crick 1989) Similarly we should not beasking what the origins of human behaviour are but should be looking insteadfor the precise properties of humans that we seek to understand These may turnout to be energetic cognitive or even thermoregulatorymdashat present we simplydo not know

Second the key process is that of inference The task at hand involves notdirect observation of what we are trying to understand but indirect inferencefrom other observations The key to inference lies in the fact that we are usingposited theoretical principles to move from what is observable to what is notAny study of events in the past must proceed in this fashion That the advancesmade in cosmology are greater than those made in palaeoanthropology can beexplained by the fact that inference in physics is more certain as the underlyingprinciples of physics are simpler and more verifiable than those in biologyespecially behavioural biology However it is very important to recognize thatinference as the path to knowledge is not confined to the study of the past Allknowledge is based on inference from observations this is as true ofexperimental sciences (for example inferring chemical structure from X-raydiffraction patterns) and the social sciences (for example inferring socialstructure from discussions with informants) as it is of sciences investigating pastevents Differences between lsquoneorsquo and lsquopalaeorsquo sciences derive from the solidityand testability of the links of an inferential chain not from any inherentdifferences in the way in which knowledge is constructed

The way forward then lies in developing a judicious mixture of theory-building and empirical observation Is it possible to be more precise about thenature of these As already stated the theoretical framework now available forbehavioural palaeoanthropologists is considerably more sophisticated than itwas in the past and has been elaborated in considerable detail elsewhere (Foley1987a Kinzey 1987 Mithen 1989) The construction of models from thistheoretical framework is also an area that has been strengthened in particular bythe application of computer simulations and other quantitative techniques(Dunbar 1989) Zubrow (Ch 6) provides an elegant example showing bothhow it is possible to incorporate individual and population variations which

7INTRODUCTION

must be the key to past events into models and how small changes in thesevariables can have major effectsmdashin this case the replacements of Neanderthalsby anatomically modern humans These models in turn have enabled us to gaina better idea of what sort of data may be critical to the further investigation ofthe origins of human behaviour

One example may be br iefly mentioned here to illustrate theinterrelationship between theory model-building and empirical observationand hence the way in which knowledge about early hominid behaviour maybe advanced Body size has become a central focus of evolutionary ecology Itis now clear that a large number of energetic physiological and behaviouralparameters greatly affect or are affected by body size (Peters 1983 Martin1983) These include for primates and many other organisms brain sizemetabolic rate longevity reproductive rates patterns of growth sexualdimorphism diet and patterns of locomotion as well as many others (Harveyet al 1987) In many cases the nature of these relationships has beenquantified

What is particularly relevant is the fact that body size is a trait that may beinferred from observations of fossils (McHenry 1989) This means that access toa wide range of behavioural and ecological characteristics of an extinctorganism is possible Changes in body size during the course of hominidevolution or differences in body size across extinct taxa can therefore be usedto infer other attributes that may be of great evolutionary significancemdashdegreeof encephalization energetic requirements and costs patterns of rangingbehaviour and so on Apart from body size similar relationships betweenempirically observable entities and other evolutionarily significant parametersare currently being investigatedmdashfor example growth and development(Bromage amp Dean 1985 Smith 1989) longevity and demography (Trinkaus ampThompson 1987) and species diversity (Foley in press)

The selection of these predominantly palaeontological examplesundoubtedly reflects my own concerns and biological orientation Otherparameters such as technology should be equally to the fore in constructingsuch methodologies Indeed Wynn (Ch 4) provides an explicit example ofhow such inferential links can be made using technological evidenceHowever it is probably the case that the more biologically based aspects ofpalaeoanthropology have developed quantitative modelling rather morefruitfully until now than has been the case in palaeolithic archaeology Tosome extent this reflects the more formally constructed nature of theory inevolutionary ecology compared with archaeology (contrary to the positionoutlined in Clark 1989) More important though it highlights the need toincorporate the material archaeological record by far the most abundant formof direct evidence of the past into behavioural palaeoanthropology to agreater extent than has hitherto been the case (but see Wynn Ch 4)Palaeolithic archaeology needs to become much more integrated withevolutionary biology

The methodological framework outlined here assumes that the questionsof interest to behavioural palaeoanthropologists are relatively clear-cut

8 INTRODUCTION

However this is far from true To return to a point made earlier in order toadvance our understanding of the or igins and evolution of humanbehaviour we need to specify tightly formulated questions These do derivewithout doubt from a general understanding of what lsquobeing humanrsquomeansmdashthat is an intelligent behaviourally flexible technologicallydependent highly communicative and conscious species that lives inextensive and complex social contexts in very large numbers This broaddefinition sets the parameters for the specific questions to which we need toknow the answers

What is the pattern of development in human intelligence and moreparticularly what are the types of mental skills that constitute our mind

To what extent were hominids other than modern humans flexible intheir behaviour Can any differences be quantified and do they varyaccording to different types of behaviour (for example social versus foragingbehaviour)

What is the pattern of technological development during the course ofhuman evolution and especially to what extent does technology in prehumanhominids differ in its pattern of generation and use from that in modernhumans

When did the communicative skills of humans develop and how is suchcommunication related to the capacity for conscious and self-reflective thought

What are the patterns of social organization of hominids This is partly aquestion about group size but more interestingly concerns relationshipsbetween and within sexes and across age categories and the degree to whichthey varied within and between populations

While these questions approach a greater degree of specificity they are stillextremely general The ways in which they may be made even more precise canillustrate some important issues in behavioural palaeoanthropology For exampleunderlying each one of the above questions are a number of issues such as

Under what ecological conditions would for example certain mental skillsbe useful

Which hominid taxa exhibit particular types of mental skillsWhat were the consequences of their evolution for other attributes of

hominid lifestyleHow were they developed in an ontological sense

Foley (Ch 2) while focusing on analytical problems associated with cultureattempts to sketch out a model that links human behavioural and cognitiveevolution to ecological conditions

It is worth highlighting some aspects of our current understanding ofhominid evolution as they provide useful insights into what may havehappened in hominid evolution First it is abundantly clear that humanevolution is not a ladder-like progression from an ape ancestor to modern

9REFERENCES

humans Rather it is a bush of radiating populations and species each ofwhich may well have had characteristics unique to itself To force thevariation observable in the archaeological and palaeontological record into alinear framework is likely to be extremely misleading Perhaps the moststriking evidence for this lies in the fact that anatomically modern humansmay in fact have predated the classic manifestations of the Neanderthals (seeMellars amp Stringer 1989) Second human evolution may not have been apattern of constant change key behavioural changes may have beenspecifically located at particular points in time and space For many anexample of this would be the origins of modern humans who may haverepresented a radical departure from other hominids rather than acontinuation of existing trends (Foley 1989) Third there may be considerabledifferences in the way in which similar behaviours were generated in differenthominid taxa For example the actual process of manufacturing a bifacialhandaxe may lie within the technical compass of a variety of species ofhominidmdashand hominoid (McGrew Ch 1)mdashbut the way in which each ofthese taxa executed the technology may have been very different And finallyhominid morphology shows considerable regional continuity for someperiods with distinctive trajectories of change occurring and a similar claimmay be made for the archaeological record (Foley 1987b) Here it is possibleto see novel selective pressures which may have been the same in differentregions interacting with unique phylogenetic factors to produce divergentpatterns of evolution The implication is that there may be considerablehistorical contingency in the pattern of hominid evolution As Habgoodshows (Ch 7) even the morphological evidence requires dextrous integrationof historical genetic demographic and ecological factors

These very brief examples indicate an important trend in the study of theorigins and evolution of human behaviour that should lead to future researchmdashthat is the need to be increasingly precise As mentioned earlier it is no longerpossible to refer generally to hominids or early hominids behavioural modelsmust be specific to time place and taxon Furthermore any model mustincorporate both proximate (for example developmental) and ultimate orselective parameters And finally any model must tackle the problem of theinteraction of immediate functional causation with problems of historicalcontingency In this way behavioural palaeoanthropology may move towards amore precise understanding of the way in which the unique features of themodern human species evolved

References

Bateson PPG 1988 The active role of behaviour in evolution In Evolutionary processesand metaphors MWHo amp SWFox (eds) 119ndash207 New York John Wiley amp Sons

Bromage TG amp MCDean 1985 Re-evaluation of the age of death of immature fossilhominids Nature 317 525ndash7

Byrne R amp AWhiten 1986 Machiavellian intelligence Oxford Clarendon Press

10 INTRODUCTION

Clark GA 1989 Alternative models of Pleistocene biocultural evolution a response toFoley Antiquity 63 153ndash62

Crick F 1989 What mad pursuit Harmondsworth PenguinDunbar RIM 1988 Primate social systems London Croom-HelmDunbar RIM 1989 Ecological modelling in an evolutionary context Folia

primatologica 53 235ndash46Foley RA 1987a Another unique species patterns in human evolutionary ecology Harlow

LongmanFoley RA 1987b Hominid species and stone tool assemblages how are they related

Antiquity 61 380ndash91Foley RA 1989 The ecological conditions of speciation a comparative approach to the

origins of anatomically modern humans In The human revolution behavioural andbiological perspectives on the origins of modern humans PAMellars amp CBStringer (eds)298ndash320 Edinburgh Edinburgh University Press

Foley RA in press How many hominid species should there be Journal of HumanEvolution

Foley RA amp RIMDunbar 1989 Beyond the bones of contention New Scientist 124(1686) 37ndash41

Foley RA amp PCLee 1989 Finite social space evolutionary pathways andreconstructing hominid behavior Science 243 901ndash6

Harvey PH RD Martin amp THClutton-Brock 1987 Life histories in comparativeperspective In Primate societies BBSmuts et al (eds) 181ndash96 Chicago University ofChicago Press

Holmes EC GPresole amp CSaccone 1989 Stochastic models of molecular evolutionand the estimation of phylogeny and rates of nucleotide substitution in the hominoidprimates Journal of Human Evolution 18 775ndash94

Humphrey NK 1976 The social function of intellect In Growing points in ethologyPPGBateson amp RAHinde (eds) 303ndash17 Cambridge Cambridge University Press

Kinzey WG (ed) 1987 The evolution of human behavior primate models AlbanyStateUniversity of New York Press

Lee PC 19889 Comparative ethological approaches in modelling hominid behaviourOssa 14 113ndash26

Martin RD 1983 Human brain evolution in ecological context James Arthur Lectureon the Evolution of the Human Brain American Museum of Natural History NewYork

Medawar PB 1967 The art of the soluble London MethuenMellars PA amp CBStringer (eds) 1989 The human revolution Edinburgh Edinburgh

University PressMithen S 1989 Evolutionary theory and post-processual archaeology Antiquity 63

483ndash94McGrew WC 1990 Chimpanzee material culture implications for human evolution

Unpublished PhD thesis University of StirlingMcHenry HM 1989 New estimates of body weight in early hominids and their

significance to encephalization and megadontia in robust australopithecines InEvolutionary history of the robust australopithecines FEGrine (ed) 133ndash48 New YorkAldine

Peters RH 1983 The ecological implications of body size Cambridge CambridgeUniversity Press

Shennan S 1989 Cultural transmission and cultural change In Whatrsquos new A closer lookat the process of innovation SEvan der Leeuw amp RTorrence (eds) 330ndash46 LondonUnwin Hyman

11REFERENCES

Smith BH 1989 Dental development as a measure of life history in primates Evolution43 683ndash7

Standen V amp RAFoley (eds) 1989 Comparative socioecology the behavioural ecology ofhumans and other mammals Oxford Black well Scientific

Trinkaus E amp DDThompson 1987 Femoral diaphyseal histomorphometric agedeterminations for the Shanidar 3 4 5 and 6 Neanderthals and Neanderthallongevity American Journal of Physical Anthropology 72 123ndash9

1 Chimpanzee material culturewhat are its limits and whyWCMcGREW

Introduction

It is a truism to say that behaviour and ideas do not fossilize Hencereconstruction of the origins of culture depends on artefacts and other remnantsthe use and meaning of which are then inferred Such inference may seemsimple in principle but it is difficult in practice for several reasons first culturalobjects are not always distinguishable from natural ones second only a subsetof enduring objects remains while perishable ones have been lost and thirddeposition and sometimes recovery is nonrandom and so what remains in thearchaeological record is biased The upshot of this can be summed up in twoaphorisms lsquoAbsence of evidence is not evidence of absencersquo and lsquoPresenceproves only possibility not probabilityrsquo In the case of the former one couldnot infer that early hominids did not use digging sticks as these tools wouldinevitably be lost to us In the case of the latter concentrations of fragments offired clay need not imply human agency as natural sources may be equallylikely

These difficulties may be eased by looking for the closest livingapproximation to the extinct hominoid forerunner Such referential models(Tooby amp DeVore 1987) will not be identical obviously but the closer the fitthe better The big advantage thus gained is that directly observable behaviouraldata will be available to supplement the artefacts At the very least this showsthe minimal capacity of an early hominid It is necessary but not sufficientevidence

Consider an idealized example An ape is seen to make and use a stone toolwhich is indistinguishable from a similar object thought to have been made byan early hominid This may mean nothing more than limited but certainknowledge of one way in which that artefact could have come about Howeverit is a tremendous advance because one now has available for study thebehaviour and mind of the user For an empiricist this is worth all thespeculation however fascinating in the world

So how to choose the best model Even a glance at thepalaeoanthropological literature shows no consensus Some models rely solelyon living human beings and exclude other species Such approaches rightlyfavour tropical open-country-living hunting-and-gathering people especiallyin Africa (Musonda Ch 3 this volume) Other models follow either homologyand make use of nonhuman primates or analogy and make use of socialcarnivores My aim here is to focus on the common chimpanzee Pan troglodytesin both ways

14 CHIMPANZEE MATERIAL CULTURE

The chimpanzee is apt for several reasons It is well studied both in the wildand in captivity Anatomically and genetically it is our closest livingevolutionary relation Most important it is a culture-bearing creature in its ownright (This last point is debatable and may be a curse as well as a blessing seeFoley Ch 2 this volume McGrew 1990)

Of course as others have used the chimpanzee model before the readermight well ask that more can usefully be said The answer is that if the mostcomplete and thoughtful earlier work was done by Tanner (1981 1987 1988)then new findings have already rendered it out of date The new knowledgecomes from several sources In captivity unprecendentedly rich and creativeexperiments and observations have been done with chimpanzees In natureseveral new field sites and studies have been developed as well as older onesexpanded In palaeoanthropology new techniques for analysis of artefacts andfossils yield data never before available and in some cases not even imaginedArchaeological evidence is much improved in quality as well as in quantityespecially in terms of careful systematic collection Perhaps most importantanalysis is tighter and more rigorous relying on explicit step-by-step argumentand stated falsifiable hypotheses rather than seductive but slippery scenariosThus my aim restated is to answer the following question given recentknowledge is the chimpanzee a better or worse model for human evolutionthan before In tackling this problem the catchier question posed in thischapterrsquos title should be dealt with too

A cautionary note culture is not a concrete entity but a mental construct Itis a set of concepts and as such cannot truly evolve However capacities forculture can evolve in organisms and manifestations of culture such as artefactscan evolve in the sense of showing changes in design features etc Beingtangible material culture is the easiest point at which to start retrospectiveanalysis but it is not enough to stop at the material To be used meaningfully theterm culture implies associated symbol-use by the culture-bearer Thusreconstructing cultural evolution may start with objects for convenience butmust carry on to assess the symbolic significance that is implicit in the artefacts

On another front there is disagreement about the various extinct forms forwhich the chimpanzee has been proposed as a model At least four have beenput forward in print first a Miocene stem-form of ancestral ape second a Mio-Pliocene ancestral hominoid which cannot be assigned confidently to eitherPongidae or Hominidae third the first recognizable Pliocene hominid andfourth a later Plio-Pleistocene hominid the first indisputable member of thegenus Homo (McGrew 1989) The first of these may correspond to Proconsal thesecond is unknown the third an australopithecine or H habilis the last Herectus

For reasons elaborated upon elsewhere (McGrew 1990) it seems likely thatonly the middle two need be taken seriously The first is too conservative in thatit was probably a precultural form more like a present-day gibbon The last istoo advanced in that living chimpanzees arguably show neither the cultural northe bodily similarities needed to draw direct comparisons with large-brainedHomo Of the middle two the ancestral hominoid is to be preferred on grounds

of caution but it must be an underestimate if the living chimpanzee is a culture-bearer (unless one assumes cultural devolution) It is hardly likely that pongidculture has stood still over the last 6 million years On the other hand if theliving chimpanzee most resembles an australopithecine then we must be facedwith cultural convergence or parallelism since phylogenetically the pongid andhominid lines had already diverged or to be more precise at least one radiationhad occurred

Culture and symbols

As might be expected culture is as hard for anthropologists to define operationally(as opposed to theoretically) as intelligence is for psychologists or language forlinguists It is not enough to say that culture includes all things human for aconcept that explains everything explains nothing it is not heuristic Thus a workingdefinition is crucial if one is to tackle the evolutionary transition from a preculturalto a cultural state If the change had intermediate protocultural states then precisionand explicitness are all the more important For the sake of this chapter culturewill be thought of as being made up of observable actions and inferrable thoughtsThe former has been covered in an earlier work using a set of eight criteriainnovation dissemination standardization durability diffusion traditionnonsubsistence and natural adaptiveness (McGrew amp Tutin 1978) Chimpanzeesshow all of these but it is the last which needs attention here

Cultural organisms must have the mental abilities to create and use symbolsotherwise one would have pseudoculture which is essentially mindless sociallearning In other words self-aware symbol-use is a necessary prerequisite ofculture (Contrary to what is sometimes saidmdashfor example by Washburn andBenedict (1979)mdashit is symbol-use and not language which is critical The twoare not synonymous as language is only a subset of symbol-use This is clearfrom studies of nonverbal humans such as autistic children) So do chimpanzeesshow self-awareness and use symbols

The answer on both points seems to be yes Galluprsquos (1970) elegantexperiment on chimpanzees recognizing themselves in a mirror has beenmuch repeated and elaborated upon Woodruff and Premack (1979) haveshown that chimpanzees seek to deceive humans both by omission and bycommission Savage-Rumbaugh et alrsquos (1978 1980) series of careful studieshas shown chimpanzees using simple symbols to label sort and ask for objectssuch as food or tools They do so among themselves in the absence of humansAll of these demonstrations come from laboratories and so remain to beconfirmed in nature but the capacities are clearly present and are used insocial life (de Waal 1982)

CULTURE AND SYMBOLS

Environment and adaptation

Despite evidence to the contrary prehistorians continue to think dichotomouslyabout the correlation between environment and hominoids Living chimpanzeesand their pongid ancestors are classed as forest-dwellers while hominization islinked with adaptation to the savanna Given this neat division chimpanzeeswould seem to be precluded as evolutionary models for any hominid on basicecological grounds In fact recent findings show that neither distinction holdsWhile the picture of an African Plio-Pleistocene vegetational mosaic of scrubdeciduous open woodland grassland and scant gallery forest remains typical thiswas punctuated by periods of expansion of humid rainforest (Williamson 1985)

More to the point wild chimpanzees live in hot dry and open environmentstoday both in East and West Africa Long-term studies have been carried out atMount Assirik Senegal where less than 3 per cent of the surface area is forested(McGrew et al 1981) Climatologically hydrologically floristically andfaunistically this is a savanna At the least the studies show that chimpanzees asa species are much more adaptable than usually credited and so must be eligibleas possible models for either forest-living ancestral hominoids or savanna-livinghominids

More pertinent to the origins of culture is the extent to which thechimpanzee toolkit (and hence by analogy the ancestral formrsquos material culture)varies with environment While some aspects of chimpanzee material life suchas nest-building are largely constant (see p 19) variation in other aspects occursacross populations Tools used to obtain termites for food differ over threewidespread sites (McGrew et al 1979) Moreover such variation also occursbetween communities in the same population (McGrew amp Collins 1985) Inboth cases some differences reflect contrasting features of habitat (for exampleavailability of prey) while others appear to reflect contrasting social customs(for example preferences for raw materials) Finally pan-African comparison oftechniques by which a specific food item the oil palm nut (Elaeis guineensis) isprocessed and eaten by chimpanzees shows a range of cultural complexity(McGrew 1985) Some populations ignore the nuts others eat only the outerenergy-rich husk and other also use stone tools to extract the protein-richkernel

All of this sounds familiar when one recalls early hominid cultural traditionsdescribed on the basis of different lithic industries (Leakey 1975) Recentreassessments (Toth 1985b) are more cautious in their interpretations andinferences and are even closer to the chimpanzee model For example earlyhominids at Koobi Fora may not have depended on stone tools and may haveused them only in certain habitats In some cases prehistor ians andprimatologists working independently with different data have come to verysimilar conclusions for example that design of tools is demonstrably a functionof the size shape and mechanical properties of the raw materials (Jones 1981McBeath amp McGrew 1982 Boesch amp Boesch 1983)

In summary recent ecological studies of wild chimpanzees and their materialculture strengthen rather than weaken the case for their being used as models of

human cultural evolution Also they underline the importance of environmentalvar iables as influences on culture However they do not yet allowpalaeoanthropologists to match the chimpanzee model with a particularancestral form

Overall diets of wild chimpanzees and ancestral hominoids and hominids lookmore and more similar Both are omnivores or more exactly mainly frugivoreswith a taste for opportunistic faunivory

All closely studied populations of chimpanzees are known to eat animal aswell as plant matter Short- and long-term studies of both tamed and untamedwild apes in a variety of habitats show them to eat social insects and small-sizedmammals such as monkeys and young ungulates (Teleki 1973 McGrew 1983Takahata et al 1983 Boesch amp Boesch 1989) Also cannibalism onceconsidered aberrant is now recognized and explicable (Goodall 1977 Nishidaamp Kawanaka 1985) Thus meat-eating is species-typical

However there are dietary differences between ape populations in thespecies of prey chosen and these seem to be understandable only in terms ofcultural differences (McGrew 1983) Techniques for getting meat vary tooChimpanzees stalk as well as stumble upon prey and after seizing the victimmay kill it in several ways (Teleki 1973 Boesch amp Boesch 1989) Sometimesextractive foraging of hidden prey occurs for example chimpanzees atMount Assirik winkle out bushbabies (Galago senegalensis) from theirsleeping holes If in the company of other apes intense scrounging andsharing usually follows Even eating showrsquos special features bites of meat areeaten with mouthfuls of leaves a habit not yet understood Chimpanzeespirate prey from other predators for example freshly filled bush-buck fawns(Tragelaphus scriptus) are taken from baboons (Papio anubis) (Morris andGoodall 1977) Significantly true scavenging also occurs when the carcassof an ungulate not seen or heard to be killed is eaten when found later(Hasegawa et al 1983)

Cross-population differences also exist for plant foods eaten (Nishida et al1983) but fruits remain the staple of chimpanzee diet from the wettest to thedriest habitats (McGrew et al 1988 Nishida amp Uehara 1983) Every placehowever dry at which chimpanzees have been studied for a long time (that isover several annual cycles) has yielded fleshy fruits in the gallery forest at leastat some times of year (Baldwin 1979)

What wild chimpanzees do not eat is equally important especially asnegative evidence is now strong from a few very long studies For animal foodsthey avoid fast-moving solitary prey which are in effect more trouble thanthey are worth such as nonsolitary insects reptiles amphibians and most smallmammals Moreover mammals weighing more than about 15 kg are not preyedupon This is true even of species whose young are taken such as bushpigs(Potatnochoerus porcus) This pointed omission is in all likelihood due to the

adultsrsquo abilities to avoid or repel predatory apes which kill only by grabbingtheir prey

For plants what is ignored is even more telling for example undergroundstorage organs such as tubers and rhizomes are not eaten even when present andexploited by sympatric primates such as Papio papio at Mount Assirik Thisomission is notable given the proposed prominence of roots in hominidadaptation to the savanna (Hatley amp Kappelman 1980) Similarly chimpanzeesfeed only rarely on the seeds of grasses This is true even on savannas whereedible grasses abound and are heavily exploited by baboons (McGrew et al1988 Sharman 1981) It seems likely in both cases that the frugivorous ape isoutdone by more dentally specialized competitors

Recent palaeoanthropological evidence of diet is impressive butinconclusive Happily much data has replaced earlier speculation Butchery andpresumed carnivory by tool-using Plio-Pleistocene hominids is now knownfrom cut marks on the bones of large herbivores At least some of these havebeen found close by hominid fossils or artefacts (Shipman amp Rose 1983 Potts1984b) However other data such as patterns of damage to bones in marrow-extraction are equivocal (Bunn 1981) As yet little can be said about feeding oninvertebrate prey at any time or about feeding on vertebrates before stone toolsemerged in hominid evolution In principle micro wear on the teeth of thehominoid could yield signs of for example bone-crushing or scanning electronmicroscopy of the bones of vertebrates might yield distinctively human toothmarks but these experiments remain to be done or are inconclusive

Evidence of nonfossilizing plant foods in the diet is even more tenuous Thusit is not surprising that results and conclusions conflict On the basis ofmicroscopic tooth-wear Walker (1981) concluded that robust australopithecineswere fruit-eaters and not grass- leaf- or bone-eaters On the basis of dentalanatomy and biomechanical analysis of chewing Lucas amp Corlett (1985)concluded that the same creature was a specialized eater of grass seeds legumesand roots Thus the chimpanzee seems to be either the most or the least aptmodel for this early hominid More detailed comparative data are available on aMiocene hominoid Sivapithecus indicus According to Teaford amp Walker (1984)its pattern of dental microwear is indistinguishable from that of Pan but differsfrom other living primates which are more specialized for hard-object- andleaf-eating In contrast on the basis of their thick molar enamel Kay (1981)decided that the ramapithecines (including S indicus) ate hard fruits seeds andnuts There is agreement that Miocene hominoids were not grass-eaters (see alsoCovert amp Kay 1981)

Overall the patchy evidence now available suggests that the diet ofchimpanzees may resemble that of an ancestral hominoid more than anintermediate hominid Only further systematic and experimental studies (Peters1982) will clarify the picture More data are needed not more speculationWhatever the uncertainty about the fossil evidence though the chimpanzeelooks markedly better than any other living primate as a dietary analogue

Home bases

Early field studies of chimpanzees stressed their nomadism and lack of fixed oreven revisited living or sleeping sites This is easy to understand in woodland andforest where food and trees are many and scattered More recent studies of theseapes in open high-seasonal habitats show a different picture At Mount Assirikby the end of the seven-month dry season chimpanzees were sleeping only inthe narrow strips of gallery forest (Baldwin et al 1982) Moreover within thisforest sleeping sites were concentrated around the few remaining sources ofclean water

Similarly early reports concluded that sleeping platforms (lsquonestsrsquo) built byapes were largely stereotypes in their making and resulting form However ifone compares nests point by point across populations differences emerge evenin such seemingly arbitrary features as the proportion which were open to thesky instead of sheltered by overhanging foliage (Baldwin et al 1981)

Finally chimpanzees in their daily activities do not wander constantly oraimlessly Travel (as opposed to feeding on the move) follows paths much of thetime and favoured resting spots are used again and again At some times of yearcertain resources such as termite mounds or hammers and anvils may be visitedseveral times a week (Boesch amp Boesch 1984 McGrew and Collins 1985)

Palaeoanthropologists seeking to interpret evidence of home bases in thearchaeological record focused on safety and food-sharing as key reasons for basecamps (Isaac 1978) More recent reinterpretations of such concentrations ofbones and tools argue for caution as other agents such as flowing water naturalclusters of dead animals or scavenging carnivores may have been responsible(Potts 1984a) Even accumulations of stone tools may have been occasionallyused as caches rather than longer-term occupation sites Thus there is no reasonto assume that either early Homo or earlier ancestral hominoids were morehome-based than living chimpanzees are

Technology

Chimpanzees are skilful makers and users of tools They make a variety of toolsfrom a variety of raw materials to serve a variety of purposes Differentcommunities have different toolkits both within and across populations Thebasic picture is well known (Goodall 1964) but recent findings refine it Forexample diffusion of a tool-use pattern has been seen for example termite-fishing females have migrated from one group to another in the Mahale MountainsTanzania (Takahata 1982) Telling cases of tool-use in hunting have been seensuch as an adult male throwing a rock to break up a stand of bushpigs allowingpiglets to be grabbed and eaten (Plooij 1978) Spontaneous appearance andrapid dissemination of hammer-stone-use has been followed in a group of 16chimpanzees (Hannah amp McGrew 1987) Chimpanzees in the Tai Forest IvoryCoast use hammer-stones of differing raw materials and dimensions to crackopen nuts of different species and show sex differences in doing so (Boesch amp

TECHNOLOGY

Boesch 1983 1984) Finally chimpanzees will transport raw materials tools anditems to be processed for hundreds of metres before use even if the resource orplace of use is out of sight (Boesch amp Boesch 1985 Hannah amp McGrew 1987)Far from being exhausted the technological capacities of chimpanzees continueto be revealed

However there are certain things that the apes have not yet been seen todo In most cases this reflects the limits of their naturally endowed mostlydental features They do not make flaked stone tools presumably because theyalready possess cutting edges on their canine teeth But they do use hammer-stones to open nuts presumably because their molar teeth are not robustenough to break the hardest-shelled species without risk of damage Theyhave enough strength to dismember small prey such as monkeys by handwithout butchering tools but resort to flimsy probes when the prey is anunderground termite which requires delicate extraction There are lsquogapsrsquo toosuch as the lack of digging sticks to get roots as noted above Neither do theymake or use shelters or containers in nature though in captivity they readilytake to both They do not use missiles or ladders to bring down or gain accessto for example out-of-reach fruit in the wild but will do so if taught byhumans

Attr ibuting function to the finished product the artefact in thearchaeological record is notoriously difficult Reconstructing the making ofsuch a tool is even more difficult as this so far has depended on experimentalreplication by knappers (Jones 1981) Occasionally ingenious approaches mayyield new knowledge from old items for example microwear on tool edges canbe an indicator of what material was worked (Keeley 1977) the sequence offlake removal can reveal whether the worker was right-handed or left-handed(Toth 1985a) However such studies have so far tended to concentrate onartefacts from later in cultural evolution typically those of H erectus rather thanon those from earlier forms

Finally the most challenging type of inference is that of the mind behind theact which produces the artefact This twice-removed operation is fraught withuncertainty Gowlettrsquos (1984) lsquoprocedural templatesrsquo (though really flow chartsof action rather than thought) at least make explicit the possible sequences ofmanufacture from start to finish

Perhaps the most ambitious attempt to tackle the problem of extinctintelligence is that of Wynn (1979 1981 Ch 4 this volume) He uses thegenetic epistemology of Jean Piaget the Swiss polymath to re-create the mindsof the makers of tools from the toolsrsquo topological attributes For Oldowan toolshe concludes that no more intelligence than that of a living chimpanzee wouldbe needed For Acheulean tools he goes to the other extreme and claims thattheir making required a level of intellect no less than that of adult H sapiensHowever a closer look at the four mental operations positedmdashwholendashpartrelations qualitative displacement spatiotemporal substitution and bilateralsymmetrymdashshows all of them to be involved in the chimpanzeesrsquo making ofprobes for termite-fishing (McGrew 1990)

21REFERENCES

Conclusions

New knowledge from the last decade makes the chimpanzees a better model ofthe origins of human culture than ever before To be more exact the relativenumber of points of similarity has increased and those of dissimilarity havedeclined For example specificity of raw materials for tools is independently andunexpectedly confirmed in both ape and ancestor hence a similarity existsScavenging of carcasses of unknown origin has now been seen in chimpanzeeshence an hypothesized dissimilarity has disappeared Also the relative degrees ofsimilarity have more often edged closer together than moved apart Diffusion ofa tool-use skill long known in captivity has now been confirmed in nature Putanother way more and more hypothetical differences between a living pongidand an extinct form ancestral to the living hominid are now seen to be quantitativerather than qualitative

Ironically however the new knowledge carries with it complications Ifchimpanzees are culture-bearers then they too have a heritage of culturalevolution If this goes all the way back to a common ancestral hominoid thenAfrican ape and hominid cultural evolution may have gone on in parallel oreven intertwined for millions of years Oldowan tools could have been made byapes not humans (Wynn amp McGrew 1989)

More likely it means analogy is just as important as homology incomparisons between chimpanzees and possible extinct counterparts Apparentlimits on chimpanzee achievements may be cultural not organic As such theyare not fixed If a previously isolated human society which has no written formof language is found the assumption is made that this is a cultural absence notan organic one Now if one finds a chimpanzee population without stone tool-use the same sort of interpretation must follow This means that use of achimpanzee model for help in reconstructing human evolution can no longerbe species-typical Instead attention must now be paid to ethnographic details inanother species Having learned more about chimpanzee nature we now knowless about its limits

References

Baldwin PJ 1979 The natural history of the chimpanzee (Pan troglodytes verus) at MtAssirik Senegal PhD thesis University of Sterling

Baldwin PJ WCMcGrew amp CEGTutin 1982 Wide-ranging chimpanzees at MtAssirik Senegal International Journal of Primatology 3 367ndash85

Baldwin PJ JSabater Pi WCMcGrew amp CEGTutin 1981 Comparisons of nestsmade by different populations of chimpanzees (Pan troglodytes) Primates 22 474ndash86

Boesch C amp HBoesch 1983 Optimization of nut-cracking with natural hammers bywild chimpanzees Behaviour 83 265ndash86

Boesch C amp HBoesch 1984 Possible causes of sex differences in the use of naturalhammers by wild chimpanzees Journal of Human Evolution 13 415ndash40

Boesch C amp HBoesch 1989 Hunting behavior of wild chimpanzees in the TaiNational Park American Journal of Physical Anthropology 78 547ndash73

Bunn HT 1981 Archaeological evidence for meat-eating by Plio-Pleistocenehominids from Koobi Fora and Olduvai Gorge Nature 291 574ndash7

Covert HH amp RFKay 1981 Dental microwear and diet implications for determiningthe feeding behaviors of extinct primates with a comment on the dietary pattern ofSivapithecus American Journal of Physical Anthropology 55 331ndash6

de Waal F 1982 Chimpanzee politics London Jonathan CapeFoley RA 1991 How useful is the culture concept in early hominid studies In The

origins of human behaviour RAFoley (ed) Ch 2 London Unwin HymanGallup GG 1970 Chimpanzees self-recognition Science 167 86ndash7Goodall J 1964 Tool-using and aimed throwing in a community of free-living

chimpanzees Nature 201 1264ndash6Goodall J 1977 Infant killing and cannibalism in free-living chimpanzees Folia

primatologica 28 259ndash82Gowlett JAJ 1984 Mental abilities of early man a look at some hard evidence In

Hominid evolution and community ecology RFoley (ed) 167ndash92 LondonAcademicPress

Hannah AC amp WCMcGrew 1987 Chimpanzees using stones to crack open oil palmnuts in Liberia Primates 28 31ndash46

Hasegawa T MHiraiwa TNishida amp HTakasaki 1983 New evidence on scavengingbehaviour in wild chimpanzees Current Anthropology 24 231ndash2

Hatley T amp JKappelman 1980 Bears pigs and Plio-Pleistocene hominids a case for theexploitation of belowground food resources Human Ecology 8 371ndash87

Isaac GLi 1978 The food-sharing behavior of protohuman hominids Scientific American238 (4) 90ndash108

Jones PR 1981 Experimental implement manufacture and use a case study fromOlduvai Gorge Tanzania Philosophical Transactions of the Royal Society London B292189ndash95

Kay RF 1981 The nut-crackersmdasha new theory of the adaptations of theRamapithecinae American Journal of Physical Anthropology 55 141ndash51

Keeley LH 1977 The function of Palaeolithic stone tools Scientific American 237 108ndash26

Leakey MD 1975 Cultural patterns in the Olduvai sequence In After theaustralopithecines KWButzer amp GLi Isaac (eds) 477ndash93 The Hague Mouton

Lucas PW amp RTCorlett 1985 Plio-Pleistocene hominid diets an approach combiningmasticatory and ecological analysis Journal of Human Evolution 14 187ndash202

McBeath NM amp WCMcGrew 1982 Tools used by wild chimpanzees to obtaintermites at Mt Assirik Senegal Journal of Human Evolution 11 65ndash72

McGrew WC 1983 Animal foods in the diets of wild chimpanzees (Pan troglodytes) whycross-cultural variation Journal of Ethology 1 46ndash61

McGrew WC 1985 The chimpanzee and the oil palm patterns of culture Social Biologyand Human Affairs 50 7ndash23

McGrew WC 1989 Why is ape tool use so confusing In Comparative Socioecology VStanden amp RA Foley (eds) 457ndash72 Oxford Blackwell Scientific

McGrew WC 1990 Chimpanzee material culture implications for human evolutionUnpublished PhD thesis University of Stirling

McGrew WC amp DACollins 1985 Tool-use by wild chimpanzees (Pan troglodytes) toobtain termites (Macrotermes herus) in the Mahale Mountains Tanzania AmericanJournal of Primatology 9 47ndash62

McGrew WC amp CEGTutin 1978 Evidence for a social custom in wild chimpanzeesMan 13 234ndash51

McGrew WC PJBaldwin amp CEGTutin 1981 Chimpanzees in a hot dry and openhabitat Mt Assirik Senegal West Africa Journal of Human Evolution 10 227ndash44

23REFERENCES

McGrew WC PJBaldwin amp CEGTutin 1988 Diet of wild chimpanzees (Pantroglodytes verus) at Mt Assirik Senegal I Composition American Journal of Primatology16 213ndash26

McGrew WC CEGTutin amp PJBaldwin 1979 Chimpanzees tools and termitescross-cultural comparisons of Senegal Tanzania and Rio Muni Man 14 185ndash214

Morris K amp JGoodall 1977 Competition for meat between chimpanzees and baboonsof the Gombe National Park Folia primatologica 28 109ndash21

Musonda FB 1991 The significance of modern hunter-gatherers in the study of earlyhominid behaviour In The origins of human behaviour RAFoley (ed) Ch 3 LondonUnwin Hyman

Nishida T amp KKawanaka 1985 Within-group cannibalism by adult male chimpanzeesPrimates 26 274ndash84

Nishida T amp SUehara 1983 Natural diet of chimpanzees (Pan troglodytes schweinfurthuuml)long-term record from the Mahale Mountains Tanzania African Study Monographs 3109ndash30

Nishida T RWWrangham JGoodall amp SUehara 1983 Local differences in plant-feeding habits of chimpanzees between the Mahale Mountains and Gombe NationalPark Tanzania Journal of Human Evolution 12 467ndash80

Peters CR 1982 Electron-optical microscopic study of incipient dental microdamagefrom experimental seed and bone crushing American Journal of Physical Anthropology57 283ndash301

Plooij FX 1978 Tool-use during chimpanzeesrsquo bushpig hunt Carnivore 1 (2) 103ndash6Potts R 1984a Home bases and early hominids American Scientist 72 338ndash47Potts R 1984b Hominid hunters Problems of identifying the earliest huntergatherers

In Hominid evolution and community ecology RFoley (ed) 129ndash66 London AcademicPress

Savage-Rumbaugh ES DMRumbaugh amp SBoysen 1978 Symbolic communicationbetween two chimpanzees (Pan troglodytes) Science 201 641ndash4

Savage-Rumbaugh ES DMRumbaugh STSmith amp JLawson 1980 Reference thelinguistic essential Science 210 922ndash5

Sharman MJ 1981 Feeding ranging and social organisation of the Guinea baboonPhD thesis University of St Andrews

Shipman P amp JRose 1983 Early hominid hunting butchering and carcass-processingbehaviors approaches to the fossil record Journal of Anthropological Archaeology 2 57ndash98

Takahata Y 1982 Termite-fishing observed in the M Group chimpanzees MahaleMountains Chimpanzee Research Project Ecological Report No 18

Takahata Y THasegawa amp TNishida 1983 Chimpanzee predation in the MahaleMountains from August 1979 to May 1982 International Journal of Primatology 5213ndash33

Tanner NM 1981 On becoming human Cambridge Cambridge University PressTanner NM 1987 The chimpanzee model revisited and the gathering hypothesis In

The evolution of human behavior primate models WGKinzey (ed) 3ndash27 Albany StateUniversity of New York Press

Tanner NM 1988 Becoming human our links with our past In What is an animalTIngold (ed) 127ndash40 London Unwin Hyman

Teaford MF amp AWalker 1984 Quantitative differences in dental microwear betweenprimate species with different diets and a comment on the presumed diet ofSivapithecus American Journal of Physical Anthropology 64 191ndash200

Teleki G 1973 The omnivorous chimpanzee Scientific American 228 (1) 33ndash42Tooby J amp IDeVore 1987 The reconstruction of hominid behavioral evolution through

strategic modeling In The evolution of human behavior primate models

WGKinzey (ed) 183ndash237 Albany State University of New York PressToth N 1985a Archaeological evidence for preferential right-handedness in the Lower

and Middle Pleistocene and its possible implications Journal of Human Evolution 14Toth N 1985b The Oldowan reassessed a close look at early stone artifacts Journal of

Archaeological Science 12 101ndash20Walker A 1981 Diet and teeth-dietary hypotheses and human evolution Philosophical

Transactions of the Royal Society B292 57ndash64Washburn SL amp BBenedict 1979 Non-human primate culture Man 14 163ndash4Williamson PG 1985 Evidence for an early Plio-Pleistocene rainforest expansion in

East Africa Nature 315 487ndash9Woodruff G amp DPremack 1979 Intentional communication in the chimpanzee the

development of deception Cognition 7 333ndash62Wynn T 1979 The intelligence of later Acheulean hominids Man 14 371ndash91Wynn T 1981 The intelligence of Oldowan hominids Journal of Human Evolution 10

529ndash41Wynn T amp WCMcGrew 1989 An apersquos view of the Oldowan Man 24 383ndash98

2 How useful is the cultureconcept in early hominid studiesRAFOLEY

Introduction

Culture is a central concept in anthropology An understanding of themechanisms of processes of cultural formation cohesion maintenance andchange forms a central focus of anthropological studies The notion of culturehas been extensively developed as a unique area of study within the disciplinedistinguishing much of anthropology from other branches of the social andbehavioural sciences Furthermore the concept has acquired a connotationof what is uniquely humanmdashthat which distinguishes humans from otheranimals

In this context the culture concept has accumulated a significance inpalaeoanthropological studies and in particular in models of the evolution ofhuman behaviour For example many attempts have been made to identifythe origins of truly cultural behaviour in the human evolutionary record andvar ious markers of these or igins have been suggested meat-eatingcooperative behaviour food-sharing home bases language symbolic thoughttool-making The occurrence of evidence for these in the fossil orarchaeological record has been used to argue that at this stagemdashusually placedeither in the Early or Middle Pleistocene and associated variously with theappearance of the genus Homo or the species Homo erectusmdashhominids hadacquired a new behavioural grade culture-bearing animals The existence ofthis new grade has been employed to suggest that non-Darwinian models ofevolutionary change are more appropriate to studies of human evolution forexample the coevolutionary models of genes and culture developed byLumsden amp Wilson (1981) and Cavalli Sforza amp Feldman (1981) amongothers In particular it is often thought that the role of natural selectionbecomes minimal once culture is established or that its presence will preventthe operation of certain evolutionary processes (for example speciation)giving human evolution a distinctive pattern of unilineal and rapid change(Wolpoff 1971)

This chapter lies at one end of a spectrum of views about the value andmeaning of the concept of culture At this extreme is the reductionist approachof evolutionary ecology which attempts to accommodate new behaviouralphenomena within an existing explanatory framework without recourse tonovel entities and processes At the other is the framework derived from socialanthropology which accepts much more readily hierarchical and emergententities in developing analytical procedures The critical question here is which

26 THE CULTURE CONCEPT IN EARLY HOMINID BEHAVIOUR

view will be most useful for developing an understanding of the patterns andprocesses of human behavioural evolution

Evolutionary ecology attempts to explain phenomena in terms of neo-Darwinian principles of natural selection principles that place emphasis onindividuals rather than larger-scale units such as societies or cultures Theevolution of culture should therefore be explicable in terms of the advantages itbrings to the individual in increased relative reproductive success In thiscontext the concept of culture is of little analytical value This chapter attemptsto justify this contention

Definitions of culture

Definitions of culture are almost as numerous as are anthropologists Kroeber ampKluckhoim (1952) have shown the vast range of definitions and usage that existand these have probably increased still further in the intervening 38 years Inherentin a common-sense understanding of culture are such characteristics as learningnongenetic transmission of information between and among generations highlevels of intra- and especially interpopulation behavioural variability tool-useand manufacture and the use of symbolic systems of communication Beyondthis there is some confusion as to whether culture is these observable phenomenaor whether it lies in the structure of the mind that makes cultural activitiespossible As the purpose of this chapter is to question the utility of the cultureconcept and to suggest that the complexity of human behaviour is comprehensiblewithout it it is not appropriate to develop a new definition However thedefinition Lumsden amp Wilson give (1981 p 3) may be quoted for illustrativepurposes lsquoculturehellip[is] the sum total of mental constructs and behavioursincluding the construction and employment of artefacts transmitted from onegeneration to the next by social learningrsquo

The use of culture in palaeoanthropology

Two examples of the use of the culture concept in palaeoanthropology aredescribed here These should be taken simply as illustrations of the ways inwhich culture has been employed both in explanatory terms and throughmodelling terms in human evolutionary studies The first is Wolpoffrsquos (1971) useof the competitive exclusion principle to argue for a single-species model ofhuman evolution Wolpoff held that culture was the means by which humans(and hominids) adapted to their environment and therefore that their niche wasdefined by the ecological space filled by culture As culture permits hominids tofill virtually all available ecological space it could therefore be argued that notwo culture-bearing hominid species could exist since these would overlap intheir requirements and so according to the Gauss competitive exclusion principleone would come to outcompete the other Wolpoffrsquos model uses culture definedas lsquostructured learningrsquo (although it is difficult to see how learning could not be

structured) in an adaptive context to argue that certain evolutionary options areremoved once culture existsmdashthat is speciation In his model culture is used ina dichotomous presence or absence manner and its presence inferred fromattributes that might correlate with the increased capacity for structured learnedbehaviour reduced canine dimensions tool-making and delayed maturation Inthis example culture is used to explain the pattern supposedly unilinear ofhuman evolution Indeed the model is analytically very powerful allowingpredictions about the nature of the fossil record on the basis of explicit ecologicaland evolutionary theory

The second way in which culture has been used in palaeoanthropologicalstudies is for the development of coevolutionary models These attempt toestablish a coevolutionary relationship between genes evolving through naturalselection and culture evolving in parallel through some alternative mechanism(Lumsden amp Wilson 1981 Cavalli Sforza amp Feldman 1981) Essentially thesemodels show that with the establishment of culture however definedevolutionary patterns will change in such a way that Darwinian selection can nolonger account for patterns or modes of change These changes can be seen asthe actual changes occurring in behaviour (culture) which do not refer back tothe genome and also as the impact of these cultural changes on geneticevolutionmdashhence the coevolutionary model the interaction between culturalevolution and Darwinian biological evolution The critical point in thesemodels is that they all start with the assumption that culture constitutes a singleentity often equivalent to the category of lsquobiologyrsquo Given this assumption thentwin evolutionary processes can occur with varying levels of interactionbetween the biology and the culture What is problematic about this approach isthat it asserts that the development of culture is the creation of a new entity inthe evolutionary process Rather than consisting of a series of epiphenomenalcomponents that have a major impact on the evolutionary process involving amassive increase in the complexity and variability of the selective environmentoperating on an individual culture here is an emergent property and a newevolutionary process

The inadequacy of culture

Is culture though a particularly useful concept in the study of human originsThis is a question relating to the practical utility of culture in analysing aparticular evolutionary event Culture is a composite term bringing togethera whole series of attributes that are important in the way in which humans livetoday However in studying the origins of these attributes it may not beparticularly useful to link them together We do not knowmdashindeed this is thevery thing we are trying to find outmdashwhen any of them first occurred withinthe hominid lineage Each of themmdashthe enhanced capacity for learning speechtool-making etcmdashmay have evolved separately subject to independent selectiveforces Thus to treat them collectively as lsquoculturersquo is to remove the possibilitythat hominids may in the past have possessed only part of their present

INADEQUACY OF CULTURE

behavioural repertoire or that repertoire combined in different ways There islittle advantage in using a term that bestows the advantages of a descriptiveshorthand (which the term culture certainly does) if it begs the very questionwe are asking thus buying descriptive ease at the expense of analytical precisionor evolutionary process

When looked at separately many of the features that collectively constitutehuman culture can be found in at least rudimentary form in nonhuman animalsChimpanzees are known to be tool-users and tool-makers (Goodall 1970Boesch amp Boesch 1983 McGrew Ch 1 this volume) and tool-making as wellas the extensive modification of the materials found in the environment (such asoccurs in nest-building) are found in other members of the animal kingdom aswell as chimpanzees Although controversial experiments with chimpanzeeshave shown them to be capable of systematic use of gestural language (such asAmerican Sign Language) in ways that suggest a grasp of symbols andgrammatical structure (Gardner amp Gardner 1969 Savage-Rumbaugh et al1983) Furthermore in their natural state primates employ a wide variety ofcommunicatory systems Cheney amp Seyfarth (1980) for example have shownthat vervet monkeys (Cercopithecus aethiops) use vocalizations in precise ways thatare close to what we understand as lsquowordsrsquomdashthat is specific sounds that haveparticular meanings Learned behaviour is also of course extremely widespreadamong animals from blue tits learning to open milk bottles (Hinde amp Fisher1951) to birds learning songs from their neighbours (Thorpe 1961) to Japanesemacaques acquiring the skill of cleaning the sand off their food (Itani 1958) Ineach of these instances not only is learning involved but there is also rapidtransmission of information and the development of lsquotraditionsrsquo withinpopulations (McGrew amp Tutin 1978)

When treated independently therefore most of the features that go to makeup the composite world of culture occur elsewhere in the animal kingdom Useof the term culture in palaeoanthropology obscures this continuity That theycan occur independently or are combined in ways different from that found inmodern humans or are developed to different degrees suggests that areductionist approach is more productive in investigating the origins of modernhuman behaviour employing minimalist categories of behaviour Inpalaeoanthropology the culture concept makes too many unwarranted andtautological assumptions

This is not to say that once such complex characteristics as tool-makingand communicatory skills have developed they do not result in novel patternsof evolution It is important however to distinguish between causes andconsequences in behavioural evolution As consequences not causes ofbehavioural evolution they would undoubtedly alter the nature of theselective pressures operating on hominids but not the mechanisms by whichselection acts To assume cultural evolution or coevolution at the outset of ananalysis of human evolution is to predetermine that natural selection isinadequate It is the purpose of an evolutionary ecological analysis todetermine whether this is the case not to prejudge the issue Behavioural notcultural evolution is an adequate term making the fewest assumptions and

providing the greatest flexibility This does not remove from consideration thecharacteristics that make humans uniquemdasha vast capacity for learninginnovation and imitation complex communication and extreme plasticity ofbehaviourmdashbut deals with them in ways that make comparisons betweenspecies feasible

Those damned chimpanzees

Culture then has three basic flaws as a concept in palaeoanthropology first it istoo high a level of abstraction to be of much empirical value second it assumesa permanent interrelationship of the components that constitute this high levelof abstraction an assumption that is invalid in evolutionary terms and third asthe basic intention is to define something that is uniquely human it is constantlyredefined in the context of studies of nonhuman primates that show continuitiesbetween humans and nonhumans In the light of these flaws palaeoanthropologistshave the choice of either abandoning the term altogether or else incorporatingchimpanzees and possibly other primates as well within the realm of lsquoculture-bearing animalsrsquo (see Bonner 1980)

McGrew amp Tutin (1978) have opted for the second of these strategiesThey have criticized the way in which culture has been defined so as to bevirtually synonymous with lsquobeing humanrsquo and have instead attempted toset up empirical criteria for defining the existence of culture that allowthe behaviour of other animals to be tested for the extent to which itfulfils those cr iter ia McGrew amp Tutin derived their cr iter ia from achallenge put out by the most culture-bound of social anthropologistsKroeber (1928) as to what he would accept as cultural behaviour in anape These criteria are

innovation diffusiondissemination traditionstandardization nonsubsistencedurability natural adaptiveness

According to these criteria chimpanzees do display cultural behaviour andwould have to be considered as culture-bearing animals Other primates mightalso fall within this category Chimpanzees therefore can be said to havedemonstrated yet further the continuities between humans and the rest of theanimal kingdom

As McGrew argues (Ch 1 this volume) this makes chimpanzees an excellentmodel for studying the development of complex human behaviour But atanother level labelling chimpanzees as culture-bearing merely extends theproblem beyond humans to chimpanzees The central problem is not whetherother animals have the capacity for culture but whether culture is ananalyticallymdashas opposed to descriptivelymdashuseful concept In other wordsrather than seeking empirical criteria for discerning culture in humans and

THOSE DAMNED CHIMPANZEES

other animals we should instead be trying to establish what is actually involvedin the evolution of complex behaviour and how it can be selected for

The evolution of complex and flexible behaviour

The key characteristic that links humans chimpanzees and to a lesser extentother primates is the complexity and flexibility of their behaviour It has beenargued here that the evolution of this type of behaviour cannot be understoodadequately through composite terms such as lsquoculturersquo but through consideringthe components that contribute towards increased behavioural complexity andflexibility The remainder of this chapter will attempt to construct a model thatcan account for the evolution of this type of behaviour However prior to doingso it is important to establish the basic framework something that previousmodels have not always done In particular I wish to emphasize the distinctionbetween (a) conditions (b) the phenotypic characters that are selected and (c)the epiphenomenal characteristics that arise as a consequence Conditionsconstitute the context or selective pressure that prompts or requires complexbehaviour Phenotypic characters are the actual behavioural characteristics thatare selected for and come to be incorporated in the behavioural repertoireresting in the individual and providing it with a reproductive advantageEpiphenomenal characteristics which are not themselves selected for arise as aconsequence of the behavioural changes and become part of the selectiveenvironment (Table 21)

The second important point to establish at the outset is the central role thatthe brain plays in the evolution of complex behaviour In strict evolutionaryterms behaviours we observe do not necessarily evolve what does evolve is theneurological capacity for these behaviours Primate and human evolution hasincorporated a large amount of encephalization and so a good starting point isto establish the function of the brain in the development of human Table 21 Overall structure of a model to account for the evolution of complex and

flexible behaviour patterns

complexity As far as humans are concerned it may be argued that the keycharacteristics of the brain are its propensity for conscious thought and for theuse of symbols

Conscious thought is an analogue for the real world What goes on inside ourheads is the constant construction of models that have some relationship tooutside events be they plans memories fantasies calculations intentions etc Inthat sense they are very similar to computer simulation programmes whichsimulate particular events and processes I would suggest as others have donethat encephalization in human evolution is essentially the evolution of a largecomputer filled with simulation programmes What has been selected for is theability to simulate the real world inside our heads which then results in suchcomplex behaviour patterns This argument has been developed most fully byHumphrey (1976) in the context of social interactions (predicting modellingand empathizing with the behaviour of other individuals in a social context)and it has been central to recent discussions of manipulative or lsquoMachiavellianrsquoprimate behaviours (Byrne amp Whiten 1986)

This perspective prompts two questions that are essentially separate but oftenconflated in many discussions what is the advantage of the ability to simulatethe external world (that is the advantage to the individual that possesses thisability) and what are the conditions likely to promote its evolution (that is theenvironmental context in which that advantage occurs)

The principal advantage of a simulation programme is that it answers lsquowhatifrsquo questions very rapidly and at very low cost and risk For example economistscan look at the effects of say altering tax structures without actually having tocarry out the reforms Simulation programmes enable individuals andinstitutions to look ahead at the consequences of their decisions and to weighup alternative courses of action If the human brain is a computer runningsimulation programmes then its advantage will be that an organism canexamine inside its head the possible options for responding to the ecologicaland social problems it faces in its environment Obviously the advantages thataccrue are only as good as the alternatives considered and the viability of themodel used (as many economists have found to their and our cost) but moreappropriate models can be developed with experience

To return to the original framework then when we talk about complex andflexible behaviour in terms of the phenotypic characters involved we mean theevolution of the brain into an extremely efficient computer for simulations Asstated at the outset these attributes reside in the individual and are the focus forselection The conditions on the other hand that promote them are a complexsocial and ecological environment in which an organism needs to make rapidand flexible responses to the problems it faces (Table 22)

From this basic relationship between conditions and selection forphenotypes and this focus on the evolution of the brain as the geneticallycontrolled generator of behaviour (Table 23) arise other characteristics someof which represent other phenotypic and genetic changes others of which aremore epiphenomenal Among the former are symbol-use conscious thought andsensory perception Symbol-use arises as a consequence of the evolution of the

EVOLUTION OF COMPLEX AND FLEXIBLE BEHAVIOUR

Table 22 Environmental conditions necessary to promote complex and flexible socialbehaviour

Table 23 Processes of selection the brain is the focus for selection for complexbehavioural strategies based on the ability to simulate the external world

human brain as a computer simulator All simulation programmes must be writtenin a language and languages are forms of symbols In terms of the computeranalogy pursued here simulations are likely to have been lsquowrittenrsquo in machinecodemdashthat is fundamental neurobiological entitiesmdashlinked to simple probablyiconic languages With continued selection though the ability of the brain tohandle more and more languages (symbolic systems) and for those languages tobecome more user friendly (conscious) would increase These lsquolanguagesrsquo maybe both internal (thought) or external (for example speech) and an advantage ofthis approach means that they can be treated independently The tremendouscapacity for humans to use symbols both in language and in other systems ofinternal and external communication is a consequence of the sophistication ofthe human brain (Table 24)

Another consequence of the evolution of the human brain as a simulator isthe evolution of the senses Any simulation programme is only as good as theinformation on which it is based The effectiveness of the simulator will dependtherefore not just upon the internal efficiency of the brain but also upon theefficiency with which the brain can receive and process information In parallel

Table 24 Processes of selection that lead to the development of symbolic systems ofthought and language as a consequence of the evolution of the brain

to the evolution of the brain will come an increasing sensitivity in certainabilities the ability to observe what is going on and to perceive and monitorevents the ability to retain events in the memory the ability to imitate othersand the ability to put out and receive signs and signalsmdashthat is to communicateAs these are also like the brain phenotypic characters they will become foci forselection The reproductive success of individuals with these abilities would behigher in the conditions specified earlier (Table 25)

Other consequences are more epiphenomenalmdashthat is they no longer restwith the individual and therefore cannot be the focus for selection but arise asa consequence of the behavioural capacity of individualsmdashand feed back to theindividual and its selective environment (Bateson 1988) These include theelaboration of symbolic systems the degree of intra- and interpopulationvariability the rate of behavioural change the rate of dissemination ofbehaviours and the degree of standardization These are pr incipallyconsequences of the ability to innovate behaviours (a response to bettersimulations) and the improved sensory perceptions (to monitor and adapt towhat is going on around an organism) These epiphenomena are what aregenerally considered to form the basis for cultural evolution (Table 26)

Table 25 Processes of selection the development of sensory perception in theevolution of human behaviour

Conclusions

In summary then the evolution of the capacity of complex and flexible behaviouris accountable in reductionist and Darwinian terms if attention is paid to thedistinction between conditions for selection phenotypic characters andepiphenomenal characteristics arising as a result It is stressed that all thecomponents usually considered to constitute culturemdashfor example those ofMcGrew amp Tutin (1978)mdashare incorporated in this model (Table 27) and thereis no need for recourse to higher entities or emergent properties to account forthem Instead it is argued that analysis should concentrate on the relatively simplecomponent parts and their empirically observable manifestations As far aspalaeoanthropological studies go culture is a redundant concept except as alinguistic shorthand

The particular model proposed here is tentative only but does have severaladvantages and implications Principal among these is that by concentrating onthe function of the brain it is possible to recognize the essentially mentalisticview of cultural capacity that is central to modern anthropological thinkingrather than its material manifestations These manifestations though are alsoincorporated as are also the outcomes of certain selective pressures andphenotypic properties Another advantage is that the central place given to thedistinction between the conditions necessary to promote complex and flexiblebehaviour and the properties themselves opens up the possibility of investigatingthe ecological and social context in which these properties evolved (Foley

Table 26 Epiphenomenal consequences of the evolution of complex and flexiblebehaviour

CONCLUSIONS

1989) This is particularly important given the current debate concerning thesignificance of the evolution of anatomically modern humans (Mellars amp Stringer1989) And finally by focusing on the role of the brain as a means of simulatinginternally the external world it may be argued that the development of thought(the internal act of simulation) is independentmdashand many would argue prior to(Lieberman 1986)mdashthe development of language the communication of theresults of such simulations to others

It must be stressed that the thrust of this chapter is not to suggest that thereare no differences between humans and other animals There clearly are and theexplanation of these in Darwinian terms remains one of the central problems inevolutionary biology and palaeoanthropology The intractability of this problemlies in developing models that do not minimize the differences between humansand other species and yet still use a truly comparative framework Themethodological reductionism of evolutionary biology provides the best scopefor this task (Foley 1987) As a summary term culture perhaps aptly encapsulatesmany aspects of human uniqueness However as generally understood culture isa synthetic concept not an analytical one and as such can have little role to playin the actual investigation of the differences between humans and other formsof life What is proposed here is not an exact model for the pattern of human

Table 27 Characteristics accounted for by the model prest(compare McGrew amp Tutin 1978)

37REFERENCES

behaviour but a framework for developing such a model and hence movingtowards a truly comparative and evolutionary explanation of humanbehavioural uniqueness

References

Bonner JT 1980 The evolution of culture in animals Princeton Princeton UniversityPress

Byrne R amp AWhiten 1986 Machiavellian intelligence Oxford Clarendon PressCavalli Sforza L amp MFeldman 1981 Cultural transmission and evolution Princeton

Princeton University PressCheyney DL amp RMSeyfarth 1980 Vocal recognition in free-ranging vervets Animal

Behaviour 28 362ndash7Foley R 1987 Another unique species patterns in human evolutionary ecology Harlow

LongmanFoley R 1989 The ecological conditions of speciation comparative perspectives on the

origins of modern humans In The human revolution behavioural and biologicalperspectives on the origins of modern humans PAMellars amp CBStringer (eds) 298ndash320Edinburgh Edinburgh University Press

Gardner RA amp BTGardner 1969 Teaching sign language to a chimpanzee Science165 664ndash72

Goodall J 1970 Tool-using in primates and other vertebrates In Advances in the study ofbehaviour 3 DLehrman RHinde amp EShaw (eds) New York Academic Press

Hinde RA ampJFisher 1951 Further observations of the opening of milk bottles bybirds British Birds 44 393ndash6

Itani J 1958 On the acquisition and propagation of a new food habit in the troop ofJapanese monkeys at Takasakiyama Primates 1 131ndash48

Kroeber AL 1928 Sub-human cultural beginnings Quarterly Review of Biology 3 325ndash42

Kroeber AL amp CKluckholm 1952 Culture a critical review of concepts anddefinitions Papers of the Peabody Museum of American Archaeology and Ethnology 47

Lieberman P 1986 The evolution of language Cambridge Ma Harvard University PressLumsden CJ amp EOWilson 1981 Genes mind and culture the coevolutionary process

Cambridge Ma Harvard University PressMcGrew WC 1991 Chimpanzee material culture what are its limits and why In The

origins of human behaviour RAFoley (ed) Ch 1 London Unwin HymanMcGrew WC amp CEGTutin 1978 Evidence for a social custom in wild chimpanzees

Man 13 234ndash51Mellars PA amp CBStringer (eds) 1989 The human revolution behavioural and biological

perspectives on the origins of modern humans Edinburgh Edinburgh University Press

Savage-Rumbaugh ES JLPate JLawson STSmith amp SRosenbaum 1983 Can achimpanzee make a statement Journal of Experimental Psychology 112 457ndash92

Thorpe WH 1961 Bird Song Cambridge Cambridge University PressWolpoff MH 1971 Competitive exclusion among Lower Pleistocene hominids the

single species hypothesis Man 6 601ndash14

3 The significance of modernhunter-gatherers in the studyof early hominid behaviourFRANCIS BMUSONDA

Introduction

The majority of modern hunter-gatherers live mainly in marginal areasmdashtheKalahari desert tropical rainforests and the tundramdashwhich are in most casesunsuitable for pastoralism and agriculture Through specialized adaptationsthey have been able to survive in these harsh environments Their subsistencedepends to a large extent on hunting and gathering but because of differencesin habitat characteristics particular activities vary in importance from one regionto another

Certain aspects of our knowledge of modern hunter-gatherers are relevantto the interpretation of the subsistence and sociological behaviour patterns ofour early hominid ancestors In this chapter I focus on four of theseenvironmental setting settlement pattern subsistence activities and socialorganization These are areas which may help to elucidate the cultural dynamicsof human evolutionary development

The fossil evidence itself has so far failed to answer questions relating tothe social life of early hominids adaptive mechanisms that led to bipedallocomotion processes of tool-use and tool-making and subsistencepatterns The last two decades have however witnessed a proliferation ofethnographic research on hunter-gatherer and nonhuman pr imatesubsistence patterns and social organization (Lee amp DeVore 1968 1976Bicchieri 1972 Coon 1971 Silberbauer 1981 Lee 1979 Nelson 1973) andthese studies provide an insight into food-gathering strategies and socialbehaviours that are relevant to the study of early hominids Much of thedata used in this chapter is drawn from studies of hunter-gatherers ofeastern and southern Afr ica where intensive field work has beenundertaken

Environmental setting

Hunter-gatherer communities on the African continent today are foundmainly in three areas northeastern Zaire where the Pygmies live northernTanzania around Lake Eyasi where the Hadza are found and the Kalaharidesert in southern Africa the home of the San These three areas illustrate

40 MODERN HUNTER-GATHERERS AND EARLY HOMINID BEHAVIOUR

the diversity of environmental setting to which these communities have hadto adapt

The region presently occupied by the Hadza is dry rocky savannadominated by thorn scrub and acacia trees (Woodburn 1968 p 50) Althoughgame is plentiful vegetable foods constitute about 80 per cent of the intake Incontrast the Pygmies live in a vast expanse of dense and damp forest whichreceives abundant rainfall throughout the year (Turnbull 1968 p 132) Theysubsist largely on plant foods and on small and medium-sized game (Harako1981 p 552) Closely adapted to the forest environment they continue topractise a hunting and gathering way of life this adaptation is expressed in theirtechnology and subsistence and is deeply rooted in their ideology (Turnbull1968 p 133)

The Kalahari desert home of the Kung San the Gwi San and otherhunter-gatherer groups experiences low rainfall which varies between 230 and600 mm a year (Yellen amp Lee 1976 Fig 11) As a result these foragers have tocope with extreme scarcity of water throughout the year Within the Kalaharidesert there exist regional ecological differences which have brought aboutimportant shifts in adaptation and cultural and social organization The centralKalahari which forms the habitat of the Gwi San is much drier than the Dobearea where the Kung are found Despite being a drier area than the Dobe thecentral Kalahari provides a wider spectrum of plant foods which the Gwi arebetter able to exploit than their Kung neighbours for whom the mangetti nutsare a staple Thus the subsistence pattern of the Kalahari hunter-gatherers islargely dictated by the availability of rainfall which in turn results inconsiderable local annual variations in the plant and animal life upon whichthese hunter-gatherers depend

Although the environmental settings in which the Hadza Pygmies and theSan live today are diverse they are indeed very restricted compared to those ofthe Plio-Pleistocene hominids of between 3 and 15 million years ago TheAfrican hominids during this period inhabited grassland and woodland areasswamps and river valleys rather than arid areas or homogeneous expanses offorest Examples include lake basins and valley floors of major rivers such as theLower Omo in Ethiopia East Lake Turkana and Afar (Leakey 1971 Isaac 1977Coppens et al 1976) sites which are lowlands around 430 m above sea levelThese were apparently favoured as habitation areas in eastern Africa by Plio-Pleistocene hominids The only exception is Laetolil in Tanzania wherefootprints and hominid fossils have been found preserved on an upland plain(Leakey et al 1976) around 1300 m above sea level

In southern Africa important Plio-Pleistocene hominid localities have beenstudied They are all highland sites ranging from 1161 m above sea level atTaung (Peabody 1964 p 674) to 1478 m at Sterkfontein to over 1829 m atMakapansgat (Sampson 1974 p 18) providing hominid fossil evidence fromfissures and caverns Vrbarsquos (1975) analyses of bovid fossil remains from thesehominid sites indicate the presence of open plains and a grassland environment

Determination of the physical environment in which our early ancestorslived is an essential part of the study of the palaeoecology of early hominids

However a comparison of modern hunter-gatherer and early hominidenvironments reveals that these two groups would have required differentadaptations because the environments are different The modern situation is notclose enough to that existing in the Plio-Pleistocene period to enable it to playa key part in any reconstruction of the exploitation patterns of our ancestors

Settlement patterns

The hunter-gatherers of southern and eastern Africa like others elsewhere in theworld possess tools for their day-to-day economic activities These largely consistof hunting and gathering tools such as bows and arrows digging sticks and anarray of domestic tools oriented toward food procurement There are alsoornamental tools and those used for personal hygiene Simplicity of personalpossessions is an advantage to these communities as they constantly have toabandon camps and set up new ones in places where plant foods and animals areto be found Since edible vegetable foods vary with the season and the movementof animals depends to a large extent on the availability of plant foods and waterthe distribution of hunter-gatherer camps is tailored to meet their need formobility

The Kung and Gwi San live in small widely scattered camps during therainy season and aggregate in large camps during the dry season (Yellen 1976)moving frequently within overlapping territories Territorial boundaries are notwell defined or defended Several groups may move in an area where foodresources are available and exploit them together Usually these groups are smalltheir numbers varying according to season and they have been observedmoving campsites from twice to ten times annually (Lee 1976 p 74) Locationof camps is to a large extent determined by the availability of water and foodresources This is especially so with the Gwi San who move their camps onlywhen they are in search of these resources (Barnard 1979 Hitchcock amp Ebert1984) Thus the shifting of camps is determined by changes in food-procurement strategies and preference and availability of new food resources ornew knowledge about the location of wide-ranging and constantly movinggame (Yellen 1976 p 56)

Similarly Kung San settlements tend to be located at least frac12 km from awater source in order not to frighten away the game that utilizes it (Yellen1976) As is true of a Gwi San camp a Kung settlementrsquos size and location andthe length of time it is occupied depend on the food resources available tosupport the group The placement spacing and utilization of work areas andstructures in a Kung camp are influenced by group structure socialorganization and division of activities among different units These camps arecharacterized by structural features such as hearths pits and tool-manufactureareas which serve as semipermanent reminders of human habitation Debrisscatter is generally confined to areas surrounding hearths and may consist ofvegetable remains animal bones and wooden implements Although most ofthe usable tools tend to be carried away to new locations when a camp is

SETTLEMENT PATTERNS

abandoned a careful study of debris scatter can make it possible to differentiatebetween dry- and wet-season camps based on food residue size of camp (wet-season camps are small have few huts and occupants and are briefly inhabited(Yellen 1977 p 78)) and amount of debris (less in wet- than in dry-seasoncamps) Food resources and length of occupation can be ascertained from theby-products of manufacturing wooden bone and stone tools A similar situationin settlement patterning has been observed among the Aborigines of Australiawho live in large camps when water is plentiful and disperse into smaller groupsduring drier periods of the year (Gould 1969)

The activity patterning in modern hunter-gatherer camps and the settlementpattern evidence in general are important to our understanding of the nature ofearly hominid sites Although the ancient eastern and southern Africanenvironments were indeed dissimilar from those currently occupied by the SanHadza and Pygmies and although the spatial organization of present-dayhunter-gatherers has doubtless been affected by movements of other people andpolitical changes there are significant archaeological implications to be derivedfrom such studies for the elucidation of early hominid settlements patternsSchick (1984) has undertaken experiments to study site formation processesand the effects they have upon Palaeolithic archaeological materials in stratifiedwater-laid deposits especially alluvial sediments The results have been appliedto the study of tool-manufacturing activities at Lower Pleistocene sites at KoobiFora Kenya and have added substantially to our understanding of thebehavioural processes involved in the formation of sites and stone tool-manufacture (Toth amp Schick 1986)

Subsistence activities

Studies involving modern hunter-gatherers of eastern and southern Africa haveshown that these peoplersquos survival is largely dependent on their intimateknowledge of the plant and animal communities that they exploit (Lee amp DeVore1968 Marshall 1976 Bicchieri 1972 Tanaka 1976) For example it has beenshown that the Kung San depend for their survival on the knowledge of placeswhere edible fruits seeds roots bulbs and other plant foods are to be found andthe conditions under which they grow as well as the feeding habits movementsand ecological requirements of the game animals upon which they dependResources are not uniformly distributed in the Kalahari desert environment butthe San tend to possess extensive knowledge of the environment and this enablesthem to be self-sufficient in plant and animal foods During times of plenty plantfoods that have a sour taste are not generally attractive or are inferior in nutrientsare not exploited although these are eaten during periods of food shortagesHowever the Dobe area in the Kalahari desert is rich in various vegetable foodsthroughout most of the year so the hunter-gatherers here can afford to exerciseselectivity in their food quest

Unfortunately there are as yet no known Plio-Pleistocene sites which haveyielded plant remains suggestive of early hominid diet However we do knowthat the majority of early hominid sites are located close to permanent watersources (Butzer 1978 p 209 Harris 1980 p 32 Isaac 1977 Leakey 1971)which may have attracted hominids because of concentration of food resourcesA recent study of vegetation transects across east African riparian andnonriparian habitats has found that both diversity and abundance of potentiallyedible high-quality plant foods were greatest in riparian habitats withabundance peaking in the wet season (Sept 1984 1985) To gain a betterunderstanding of the nature of diet and procurement strategies in the pastarchaeological models have to be formulated based on both contemporaryhunter-gatherer subsistence patterns and the results of ecological studies such asthe one undertaken by Sept

Division of labour among hunter-gatherers is based on gender and plays animportant part in food acquisition and sharing Women remain primarilyresponsible for procuring and preparing vegetable foods whereas hunting gameis largely the responsibility of adult males although young males and able-bodied females may take part when the need arises Despite the existence ofdivision of labour among hunter-gatherers food acquisition remains a collectiveresponsibility Tasks performed by a hunter-gatherer group become increasinglydifferentiated with age with young boys and girls taught to do different kinds ofthings at an early age (Draper 1975)

An understanding of social aspects relating to division of labour and food-sharing practices may be helpful in elucidating the development of permanentmalendashfemale relationships among the early hominids Based on the fact thatboth males and females in a modern hunter-gatherer society collect differentkinds of foodstuffs which they then transport back to campsites to share withinthe social group an early hominid couple could have paired for the matingseason on the basis of food-procurement arrangements Gradually a matingsystem among members of the group would emerge as a result of division oflabour and food-sharing practices (see also Lovejoy 1981)

The ethnographic literature on food sharing among hunter-gatherers is notsufficiently detailed to describe how different food items are shared or in thecase of meat how specific anatomical parts are distributed What is commonlyportrayed is a situation in which every group member receives a share of theavailable food irrespective of its size quantity or nutritional value Howeverethnographic literature provides some insights into the nature of food-procurement strategies and consumption It has been noted that huntergathererseat some of the food collected almost immediately but also carry some back tothe campsites to share with those who stayed behind (Musonda 1986) Foodsharing is a characteristic of hunter-gatherers that is deeply entrenched in theireating behaviour (Marshall 1976) However this behaviour mainly applies tobig animals rather than small ones such as tortoises lizards and duikers Huntingparties go out to hunt big animals and meat is shared more or less evenlyAccording to Marshall (1976 p 358) when a kill is made hunters eat the liverand other perishable parts on the spot as well as other body parts until they are

SUBSISTENCE ACTIVITIES

full The animal is then dismembered and carried back to the campsite If it istoo large to be carried the bones are left behind after chopping off the meatButchering of small animals is carried out at the camp and not necessarily at thekill site Meat is then shared among members of the group

This conclusion has important implications for hunter-gatherer responsesdeveloped to cope with problems associated with small total calorie andprotein intakes Speth and Spielmann (1983 pp 18ndash21) and Speth (1987 p17) have discussed some of those responses in relation to modern temperateand northern-latitude hunter-gatherers When both total calories and proteinare in short supply sharing of meat has to include fats and carbohydrateswhich are nutritionally significant to prevent hunter-gatherers from losingbody weight

Wilmsen (19781982) Truswell (1977) and Truswell amp Hansen (1968) havestudied the diet and nutrition of the Kalahari desert hunter-gatherers andhave demonstrated that these people undergo significant loss of body weighteach year during the late dry season and early rainy season (that is late springand early summer) Wilmsen has attributed the weight loss to food shortagesin the late dry season and early rainy season This phenomenon may beapplicable to early hominid procurement strategies because they too mayhave faced levels of seasonal food stress more or less comparable to the levelsfaced by contemporary San (Speth 1987 p 21) According to Speth thestrategies of early hominids towards the procurement of animal proteinshould be highly dependent on the nutritional status of both hominids andprey and that nutritional status in turn varies in a systematic fashion withseason Thus the current debate about whether early hominids obtained meateither largely or entirely by hunting or by scavenging (Bunn et al 1980Binford 1981 Isaac amp Crader 1981 Isaac 1983 1984 Bunn 1983 Potts 1983Shipman 1983) may have to look critically at the arguments presented bySpeth (1987) concerning the procurement of animal protein during thedifferent seasons of the year

Subsistence-related behaviour is also reflected in the possession ofvarious kinds of equipment essential to the hunter-gatherer food questStudies of toolkits employed by hunter-gatherers are important in ourunderstanding of the economic and social behaviour of early hominids Thefact that almost all the food-acquisition activities of modern hunter-gatherers are accompanied by the use of tools leads us to speculate that asimilar kind of behaviour prevailed among early hominids Ebertrsquos (1979)research among the San of the Kalahari has suggested that certain aspects oftool-use and tool-discard or loss are probably similar to those in thearchaeological record although he admits that the metal knives and axesused by hunter-gatherers today differ in their economic value effectivenessand longevity and in the cultural or symbolic value placed upon them fromthe stone implements of earlier huntergatherers in the same region Thisbehaviour relating to tool-use and discard is important to an understandingof past technological remains (Ebert 1979 p 63)

Studies of modern hunter-gatherers have shown that more than 70 per centof their food intake consists of plant foods contrary to the previous emphasisplaced on meat-eating and hunting (Ardrey 1961 1976) Therefore the primacyof hunting and meat-eating in hominid evolution is not supported byethnographic studies Studies of tooth-wear patterns of early hominids suggest adiet that was not dominated by meat (Wallace 1972 Wolpoff 1973) aconclusion that points to the fact that meat-eating was probably not central tohominid origins

Lower Pleistocene sites in eastern and southern Africa have yielded evidencethat points to dependence on a wide range of animal foods by early hominidsBecause of preservation problems no evidence of plant-food gathering has beenfound at these sites Todayrsquos hunter-gatherers display a broad dietary rangeinvolving a wide spectrum of plant and animal foods and their intake of thesefoods ranges from deeply buried tubers to fruits high on trees and from smallcrawling animals to large mammals The acquisition of most of these foods isgreatly facilitated by the use of tools wooden spears bored stones diggingsticks and bows and arrows

Modern hunter-gatherers transport meat to campsites in more or less thesame fashion that early hominids did as reflected in the archaeological recordfrom East Lake Turkana (KBS) Kenya and Olduvai Gorge (FLKN Level 6)Tanzania (Isaac 1976 p 561 Leakey 1971 p 252) On the basis of evidencerelating to meat-eating Isaac (1980 p 226) has argued that the course of humanevolution was characterized by a broadening of the range of foods which wereimportant to protohuman ancestral populations Isaacrsquos argument offers analternative interpretation to earlier views on human evolution advanced fromthe 1950s (Dart 1953 Ardrey 1961 1976 Morris 1967 Jolly 1970) that huntinginfluenced human evolution and was responsible for the division of labourbetween the sexes

Social organization

Models formulated to understand the social behaviour and anatomy of earlyhominids have largely been based on studies of chimpanzee behaviour andanatomy These primates are strikingly similar to humans in social behaviourThey prepare and use tools for a variety of purposes prey upon small animalsoccasionally walk bipedally for short distances share certain foods andcommunicate social and environmental information (Goodall 1968 Teleki 1975)

The Pygmy chimpanzee provides an even better fit because this primate isless sexually dimorphic than other apes and is less specialized in habitat dietand social behaviour (Zihlman 1979) Zihlman amp Tanner (1978 p 168) haveargued that the similarities in behaviour anatomy and genes between humansand chimpanzees are so extensive that it is most unlikely that these shared traitsare due to convergent evolution

However despite the varied activities that chimpanzees are able toperform their relatively small brain limits their ability to develop highly

SOCIAL ORGANIZATION

skilled tool-use and tool-making Cooperative hunting and food-sharing(plant foods are rarely shared) are evident but division of labour is not aselaborate as it is among humans These complex behaviours include carryingfood back to a base camp for preparation and sharing common amongcontemporary hunter-gatherers

Studies of modern hunter-gatherers have shown that these communitiesare characterized by very fluid population distribution over a geographicalarea Group structure like campsite location is oriented to food and waterresources (Yellen amp Harpending 1972) Owing to scattered and variableresources the San constitute a loose confederation of small bands organizedthrough kin and marriage relationships (Silberbauer 1972 p 273 1981) andare mobile independent of others in order to achieve a close fit betweenresource and population density (Yellen amp Harpending 1972) The carryingcapacity of a territory sets a limit on the size of the band while the availabilityof food plants and water is the principal determinant of the bandrsquos socialorganization

Observations of San bands (Lee 1968 1976) show that they constitutenoncorporate bilaterally organized groups that live in a single settlement andmove together for at least part of the year Group structure is very variableindeed perhaps because of changes in rainfall levels and the sparse distributionof standing water in the northern Kalahari

The social organization of the Kung is very similar to that of other hunter-gatherersDamas (1969) has shown that central Eskimos concentrate in large groups inwinter when there is good seal-hunting Also the Aborigines of Australia followa concentration-dispersion pattern determined by seasonal differences in wateravailability This pattern has also bee n observed among the Pygmies of theCongo Forest in northeastern Zaire where the huntergatherer movementpattern is based on the seasonal exploitation of key resources and social factors(Turnbull 1965 1968) According to Lee (1976 p 91) the existence of thispattern in different kinds of environment suggests that it is basic to the huntingand gathering adaptation There are indeed several advantages to this kind ofpattern first a high population density is a distinctive possibility second thereis a likelihood of responding favourably to the local imbalance in foodresources and third there is a good chance of keeping the threat of violence toa minimum (Lee 1976 p 91)

Explanatory models for the social life of early hominids have been drawnfrom the interpretation of tool-making processes and the way tools weretransported These models are important as they help to define human patternsof behaviour Evidence from Olduvai Gorge and East Lake Turkana sites hasbeen used to explain how early hominids made stone tools which were carriedaround and how hunting the butchering of animals and the sharing of meatwere important aspects of social organization (Leakey 1971 Isaac 1978)Language important for the exchange of information about various aspects oflife and a regulating factor of social relations among modern hunter-gatherersmust have been instrumental in the success of an early hominid band With thedevelopment of a mating system and division of labour between sexes language

must have enabled early hominids to develop an alternative lsquoinheritancersquocapable of changing faster than genetic systems

Conclusions

The foregoing is a brief summary of some of the important aspects of modernhunter-gatherer behaviour that palaeoanthropologists are emphasizing in thereconstruction of the cultural history of Plio-Pleistocene hominids (Clark 1968p 276) However opinion remains divided on the question of whether modernhunter-gatherers can be used as exact models for early hominids especially inview of the formerrsquos association with marginal environments One school ofthought argues that a judicious use of ethnographic data may provide a uniqueopportunity for the reconstruction of the way of life of past populations (Clark1968 p 280) A more cautious approach in the use of ethnographic data isadvocated by Clark Howell in his contribution at the symposium Man theHunter (Lee amp DeVore 1968 p 287) He suggests that reconstruction of earlyhominid life based on the present should be discouraged or very severely curtailedexcept for very recent time periods However later researchers have revealedthat some behavioural elements of sociocultural systems have material correlatesand can be incorporated in the archaeological interpretation helping in themaking of inferences about early hominid behaviour (Kramer 1979 p 1)

Yellenrsquos (1977) research among the San has revealed that modern hunter-gatherersocieties do provide very significant data for formulating models that are useful in theinterpretation of archaeological material Studies involving the subsistencebehaviour of hunter-gatherers point to dependence on both gathering huntingand division of labour between sexes behaviours which were certainlycharacteristic of early hominids Hunting for instance has been overemphasizedas a factor responsible for speeding up human evolution (Washburn amp Lancaster1968 Pfeiffer 1972) whereas vegetable foods have until recently received littleattention in discussions related to human evolution Studies of hunter-gatherersshow that meat is a minor component in their diet (between 20 and 50 per cent)so in the light of this information hunting cannot be regarded as a factor responsiblefor human development

Although modern hunter-gatherer studies have made it possible forarchaeologists to speculate on the size of early hominid social groups the lengthof time involved in refuse accumulation subsistence and settlement patternsserious misgivings must remain about developing models based on present-dayhunter-gatherer activities Modern groups are far removed in time from theearly hominids To use them to postulate past activities is to suggest that thesubsistence base and technology have not changed since the Plio-PleistoceneWhile such studies are undeniably useful it is important to realize that modernhunter-gatherers inhabiting marginal areas may differ from prehistoric peoplesinhabiting different environments Moreover in the course of time thesehunter-gatherers may have undergone considerable change (Tanaka 1976)requiring different adaptations Thus the use of ethnographic analogy in the

CONCLUSIONS

interpretation of archaeological data that are greatly removed in time and spaceis risky to say the least (Binford 1968 Isaac 1972) and may have only limitedapplication It should not be assumed that the observed differences betweenagricultural and pastoral societies on the one hand and hunter-gatherers on theother are an indication of the closeness of the latter to the Plio-Pleistocenehominids However as long as contemporary hunter-gatherers are not viewed aslsquoliving fossilsrsquo surviving from more or less remote periods (Isaac 1968 p 253)prehistoric studies can use the insights they provide to devise research in thearchaeological context (Isaac 1968 1972 p 172 Clark 1968) Such an approachoffers unique opportunities for the reconstruction of early hominid activities inthe distant past

Acknowledgement

I am greatly indebted to Florence Nchimunya of the Livingstone Museum whotyped the draft of this chapter

References

Ardrey R 1961 African genesis New York CollinsArdrey R 1976 The hunting hypothesis New York AtheneumBarnard A 1979 Kalahari bushmen settlement patterns In Social and ecological systemsPBurnham amp RFEllen (eds) 131ndash44 London Academic PressBicchieri M (ed) 1972 Hunters and gatherers today New York Holt Rinehart amp

WinstonBinford LR 1968 Post-Pleistocene adaptations In New perspectives in archaeology

SRBinford amp LRBinford (eds) 313ndash41 Chicago AldineBinford LR 1981 Bones ancient men and modern myths New York Academic PressBunn HT 1983 Evidence on the diet and subsistence patterns of Plio-Pleistocene

hominids at Koobi Fora Kenya and Olduvai Gorge Tanzania In Animals andarchaeology 1 Hunters and their prey JClutton-Brock amp CGrigson (eds) 21ndash30 BARInternational Series 163 Oxford British Archaeological Reports

Bunn HT JWKHarris GIsaac ZKaufulu EKroll KSchick NToth ampAKBehrensmeyer 1980 FxJjSO an early Pleistocene site in northern Kenya WorldArchaeology 12 109ndash36

Butzer KW 1978 Geological perspectives on early hominid evolution In Earlyhominids of Africa CJJolly (ed) 191ndash217 New York St Martinrsquos Press

Clark JD 1968 Studies of hunter-gatherers as an aid to the interpretation of prehistoricsocieties In Man the hunter RBLee amp IDeVore (eds) 276ndash80 Chicago Aldine

Coon CS 1971 The hunting peoples Boston Little BrownCoppens Y FCHowell GIsaac amp REFLeakey (eds) 1976 Earliest man and

environments in the Lake Rudolf Basin Chicago University of Chicago PressDamas D 1969 Characteristics of central Eskimo band structure In Contributions to

anthropology band societies DDamas (ed) 116ndash38 National Museum of CanadaBulletin 228

Dart RA 1953 The predatory transition from ape to man International Anthropologicaland Linguistic Review 1 (4) 201ndash19

49REFERENCES

Draper P 1975 Kung women contrasts in sexual egalitarianism in the foraging andsedentary contexts In Toward an anthropology of women RReiter (ed) New YorkMonthly Review Press

Ebert JI 1979 An ethnoarcheological approach to reassessing the meaning ofvariability in stone tool assemblages In Ethnoarcheology implications of ethnography forarcheology CKramer (ed) 59ndash74 New York Columbia University Press

Goodall J 1968 The behaviour of free-living chimpanzees in the Gombe StreamReserve Animal Behaviour Monographs 1 165ndash311

Gould RA 1969 Subsistence behaviour among the Western Desert Aborigines ofAustralia Oceania 39 253ndash74

Harako R 1981 The cultural ecology of hunting behaviour among Mbuti Pygmies inthe Ituri Forest Zaire In Omnivorous primates gathering and hunting in human evolutionRSOHarding amp GTeleki (eds) 499ndash555 New York Columbia University Press

Harris DR 1980 Commentary human occupation and exploitation of savannaenvironments In Human ecology in savanna environments DRHarris (ed) 31ndash39London Academic Press

Hitchcock RK amp JIEbert 1984 Foraging and food production among Kalaharihunter-gatherers In From hunters to farmers the causes and consequences of food productionJDClark amp SABrandt (eds) 328ndash48 Berkeley University of California Press

Isaac GL 1968 Traces of Pleistocene hunters an East African example In Man thehunter RBLee amp IDeVore (eds) 253ndash61 Chicago Aldine

Isaac GL 1972 Early phases of human behaviour models in Lower Palaeolithicarchaeology In Models in archaeology DLClarke (ed) 167ndash99 London Methuen

Isaac GL 1976 The activities of early African hominids a review of archaeologicalevidence from the time span two and a half to one million years ago In Humanorigins Louis Leakey and the East African evidence GLIsaac amp TMcCown (eds) 462ndash514 Menlo Park California WABenjamin Inc

Isaac GL 1977 Olorgesailie archaeological studies of a Middle Pleistocene lake basin in KenyaChicago University of Chicago Press

Isaac GL 1978 The foodsharing behaviour of protohuman hominids Scientific American238 (4) 110ndash23

Isaac GL 1980 Casting the net wide a review of archaeological evidence for earlyhominid land use and ecological relations In Current argument on early man LKonigsson (ed) 226ndash51 Oxford Pergamon Press

Isaac GL 1983 Bones in contention competing explanations for the juxtaposition ofearly Pleistocene artefacts and faunal remains In Animals and archaeology 1 Huntersand their prey JClutton-Brock amp CGrigson (eds) 3ndash19 BAR International Series163 Oxford British Archaeological Reports

Isaac GL 1984 The archaeology of human origins studies of the Lower Pleistocene inEast Africa 1971ndash1981 Advances in World Archaeology 3 1ndash87

Isaac GL amp DCCrader 1981 To what extent were early hominids carnivorous InOmnivorous primates RSOHarding amp GTeleki (eds) 37ndash103 New York ColumbiaUniversity Press

Jolly C 1970 The seed-eaters a new model of hominid differentiation based on ababoon analogy Man 5 (1) 5ndash26

Kramer C (ed) 1979 Ethnoarcheology implications of ethnography for archeology New YorkColumbia University Press

Leakey MD 1971 Olduvai Gorge Volume 3 Cambridge Cambridge University PressLeakey MD RLHay GHCurds REDrake MKJackes amp TDWhite 1976 Fossil

hominids from the Laetolil beds Nature 262 460ndash66Lee RB 1968 What hunters do for a living or how to make out on scarce resources

In Man the hunter RBLee amp IDeVore (eds) 30ndash48 Chicago Aldine

Lee RB 1976 Kung spatial organization an ecological and historical perspective InKalahari hunter-gatherers studies of the Kung San and their neighbors RBLee amp IDeVore (eds) 74ndash97 Cambridge Ma Harvard University Press

Lee RB 1979 The Kung San men women and work in a foraging society CambridgeCambridge University Press

Lee RB amp IDeVore (eds) 1968 Man the hunter Chicago AldineLee RB amp IDeVore 1976 Kalahari hunter-gatherers studies of the Kung San and their

neighbors Cambridge Ma Harvard University PressLovejoy CO 1981 The origin of man Science 211 341ndash50 Marshall L 1976 Sharing

talking and giving relief of social tensions among the Kung In Kalahari hunter-gatherers studies of the Kung San and their neighbors RB Lee amp IDeVore (eds) 349ndash71 Cambridge Ma Harvard University Press

Morris D 1976 The naked ape London Jonathan CapeMusonda FB 1986 Plant food in the diet of the prehistoric inhabitants of the

Lunsemfwa drainage basin Zambia during the last 20 000 years Zambia GeographicalJournal 36 17ndash27

Nelson RK 1973 Hunters of the nor them for est Chicago University of Chicago PressPeabody FE 1964 Travertines and cave deposits of the Kaap escarpment of South

Africa and the type locality of Australopithecus africanus Dart 1924 Bulletin of theGeological Society of America 65 671ndash706

Pfeiffer J 1972 The emergence of man New York HarperPotts R 1983 Foraging for faunal resources by early hominids at Olduvai Gorge

Tanzania In Animals and archaeology 1 Hunters and their prey JClutton-Brock amp CGrigson (eds) 51ndash62 BAR International Series 163 Oxford British ArchaeologicalReports

Sampson CG 1974 The Stone Age archaeology of southern Africa New York AcademicPress

Schick KD 1984 Processes of Palaeolithic site formation an experimental study UnpublishedPhD thesis University of California Berkeley

Sept JM 1984 Plants and early hominids in east Africa a study of vegetation in situationscomparable to early archeological site locations Unpublished PhD thesis Department ofAnthropology University of California Berkeley

Sept JM 1985 Edenrsquos forbidden fruit Plant food foraging opportunities in east Africanhabitats Paper presented at the 50th Annual Meeting of the Society for AmericanArchaeology Denver Col (4 May 1985)

Shipman P 1983 Early hominid lifestyle hunting and gathering or foraging andscavenging In Animals and archaeology 1 Hunters and their prey JClutton-Brock ampCGrigson (eds) 31ndash50 BAR International Ser ies 163 Oxford Br itishArchaeological Reports

Silberbauer GB 1972 The Gwi Bushmen In Hunters and gatherers today M Bicchieri(ed) 271ndash325 New York Holt Rinehart amp Winston

Silberbauer GB 1981 Hunter and habitat in the central Kalahari desert CambridgeCambridge University Press

Speth JD 1987 Early hominid subsistence strategies in seasonal habitats Journal ofArchaeological Science 14 13ndash29

Speth JD amp KSpielmann 1983 Energy source protein metabolism and huntergatherersubsistence strategies Journal of Anthropological Archaeology 2 1ndash31

Tanaka J 1976 Subsistence ecology of central Kalahari San In Kalahari huntergatherersstudies of the Kung San and their neighbors RBLee amp IDeVore (eds) 98ndash119Cambridge Ma Harvard University Press

Teleki G 1975 Primate subsistence patterns collector-predator and gatherer-hunterJournal of Human Evolution 4 125ndash84

51REFERENCES

Toth N amp KDSchick 1986 The first million years the archeology of protohumanculture In Advances in archeological method and theory Volume 9 MBSchiffer (ed) 1ndash96 New York Academic Press

Truswell AS 1977 Diet and nutrition of hunter-gatherers In Health and disease in tribalsocieties 213ndash26 Ciba Foundation Symposium 49 Amsterdam Elsevier

Truswell AS amp JDLHansen 1968 Medical and nutritional studies of Kung bushmenin northwest Botswana a preliminary report South African Medical Journal 42 1338ndash9

Turnbull C 1965 Wayward servants the two worlds of the African Pygmies Garden CityNatural History Press

Turnbull C 1968 The importance of flux in two hunting societies In Man the hunterRBLee amp IDeVore (eds) 132ndash37 Chicago Aldine

Vrba E 1975 Some evidence of chronology and palaeoecology of SterkfonteinSwartkrans and Kromdraai from the fossil Bovidae Nature 254 301ndash4

Wallace JA 1972 Tooth chipping in the australopithecines Nature 244 117ndash18Washburn SL amp CSLancaster 1968 The evolution of hunting In Man the hunter

RBLee amp IDeVore (eds) 293ndash303 Chicago AldineWilmsen EN 1978 Seasonal effects of dietary intake on Kalahari San Federation

Proceedings 37 65ndash72Wilmsen EN 1982 Studies in diet nutrition and fertility among a group of Kalahari

bushmen in Botswana Social Science Information (Sage London and Beverly Hills) 21(1) 95ndash125

Wolpoff MH 1973 Posterior tooth size body size and diet in South African gracileaustralopithecines American Journal of Physical Anthropology 39 375ndash94

Woodburn J 1968 An introduction to Hadza ecology In Man the hunter RBLeeampIDeVore (eds) 49ndash55 Chicago Aldine

Yellen JE 1976 Settlement patterns of the Kung an archeological perspective InKalahari hunter-gatherers studies of the Kung San and their neighbors RBLee amp IDeVore (eds) 47ndash72 Cambridge Ma Harvard University Press

Yellen JE 1977 Archeological approaches to the present Model for reconstructing the past NewYork Academic Press

Yellen JE amp Harpending 1972 Hunter-gatherer populations and archaeologicalinference World Archaeology 4 (2) 244ndash53

Yellen JE amp RBLee 1976 The Dobe-Duda environment background to a huntingand gathering way of life In Kalahari hunter-gatherers studies of the Kung San and theirneighbors RBLee amp IDeVore (eds) 27ndash46 Cambridge Ma Harvard UniversityPress

Zihlman A 1979 Pygmy chimpanzee morphology and the interpretation of earlyhominids South African Journal of Science 75 163ndash5

Zihlman A amp NTanner 1978 Gathering and the hominid adaptation In Femalehierarchies LTiger amp HFowler (eds) Chicago Fowler AVC Inc

4 Archaeological evidence formodern intelligenceTHOMAS WYNN

Introduction

Many prehistorians assume that the evolution of anatomically modern humanscoincided with the appearance of certain attributes of behavioural complexityin the archaeological record parietal art exchange systems and curated toolsto name just a few Some argue that this complexity reflects a more powerfulintelligence and that Homo sapiens sapiens was blessed with a cleverness thatgave him a marked advantage over his archaic predecessors According toRedman there was lsquoa change in adaptive strategies and organizational abilitiesat the beginning of the Upper Palaeolithic This transition signifies the rapidlyincreasing ability of human beings to recognize environmental potentialsthat existed [and] to communicate these potentials to othersrsquo (1978 pp 51ndash2) In a discussion of one aspect of complexity storage Binford makes thefollowing contention lsquoIt is my impression that the ability to anticipate eventsand conditions not yet experienced was not one of the strengths of our ancestorsprior to the appearance of clear evidence for symboling eg personalornaments graphics in the form of painting lsquoartrsquo and notation (1982 p 178emphasis in original) In other words prior to the Upper Palaeolithic Homowas incapable of planning very far ahead Both the supposed foresight of Hsapiens sapiens and his increased organizational ability if true must be aspectsof a more powerful intelligence

The question of intelligence is not a matter of sophistry If the behaviouralcomplexity we see in the archaeological record was tied to intelligence then wemust incorporate a factor of biological evolution into our interpretation ofculture change (see Foley Ch 2 this volume) Intelligence at least as commonlyconceived has a physiological component that must have evolved If on theother hand this complexity was not tied to intelligence then we must interpretculture change rather differently These two alternatives constitute very differentunderstandings of the nature of later human evolution

In this chapter I address the question of the appearance of modernintelligence using the theory of Jean Piaget perhaps the most influentialdevelopmental psychologist of the 20th century For evidence I use thearchaeological record Most of my examples come from the European UpperPalaeolithic not because it is somehow more typical but because the transitionto increased cultural complexity was relatively abrupt In particular I discuss thesignificance of the following behaviours

1 technology especially curated tools and facilities2 subsistence especially seasonal hunting and fishing3 exchange systems4 ritual systems especially Magdalenian parietal art

Archaeology and intelligence

In any study of intelligence we immediately run into some methodologicalproblems The first is definition In introductory psychology courses one is taughtthat intelligence is something measured by IQ tests In other words it isperformance on a standardized test This definition clearly has very littleevolutionary potential we cannot give Neanderthals the Stanford-BinetFurthermore saying that Johnny has a higher IQ than Tommy is not quite thesame as saying that elephants are more intelligent than monitor lizards Intelligenceis too general and fuzzy a concept to use without narrowing it down a bit Thesecond problem is one of evidence What do we look for Archaeology mustwith few exceptions rely on the analysis of the end products of behaviour Someof these may have required more intelligence than others but if so it is notobvious which How do we select our attributes We can solve both of thesemethodological difficulties if we turn to established theories of intelligence

Unfortunately archaeologists more often than not turn to common sense Weconsider ourselves to be intelligent people and by self-reflection we decidewhat it is in prehistory that should require a high intelligence Self-reflection isa notoriously faulty source for scientific concepts This problem is not restrictedto intelligence Binford (various see for example 1983) has shown that mostfaunal analysis has been based on common-sense ideas that are simply wrongHe has attempted to replace the common sense with experimentally basedlsquomiddle-range theoryrsquo as he terms it One problem with common-sense ideasof intelligence is their tendency to confuse intelligence with complexity thereare more tool types in the Upper Palaeolithic therefore people must have beensmarter Pursuing such reasoning one would in turn have to argue that 20th-century Europeans are more intelligent than 19th-century Europeans There aremore insidious dangers Based largely on self-reflection and common sense19th-century scientists assumed that men were smarter than women andnorthern Europeans were smarter than southern Europeans After all it seemedobvious Worse still this bias led them to find (or make) measures corroboratingtheir ideas (Gould 1981) Common sense is simply insufficient Luckily for thestudy of intelligence well-developed and well-tested theories exist and unlikeBinford we need not construct our own experimental base

For a theory of intelligence to be useful for the archaeologist it must do twothings First it must define intelligence in such a way that it encompasses thebehaviours of nonhumans It must see intelligence as an entity that varies fromtaxon to taxon and which can evolve within a single lineage It must be able tocompare elephants to monitor lizards apes to humans and then to measure thedifferences in some way The IQ definition fails here Second the theory must

ARCHAEOLOGY AND INTELLIGENCE

54 ARCHAEOLOGICAL EVIDENCE FOR MODERN INTELLIGENCE

be able to assess the end products of behaviour Many theories are based on theassessment of sequences of behaviour or on verbal accounts by subjects We haveneither in the archaeological record A third characteristic is also important Thetheory must be persuasive The categories of intelligence defined by the theoryneed to have been confirmed again and again in comparative studies and cross-cultural studies The truth and reliability of the theory must be established oncontemporary data Only then can the theory be applied to prehistory Thearchaeological record does not have the resolution to generate and test theoriesof intelligence on its own

One methodological caveat cross-cuts all theories of intelligence used byarchaeologists the problem of minimum necessary competence We cannotassume that the behaviours we see represented in archaeological evidencerequired the highest abilities of the prehistoric people They may well have usedvery sophisticated thinking in domains that are archaeologically invisiblemdashsocial structure or cosmogony for example But when we assess intelligencearchaeologically we can reach conclusions only about the minimumcompetence necessary for the behaviour that we see It is therefore possible tounderestimate intelligence especially because archaeological evidence consistsmostly of mundane day-to-day behaviours that may not have taxed prehistoricintelligence just as they do not tax ours On the other hand it would be verydifficult to overestimate intelligence since we must assess minimum abilities

A Piagetian approach to prehistoric intelligence

Piagetrsquos genetic epistemology is well known as a theory of child development atheory that describes a sequence of stages through which all children pass frominfancy to adolescence But Piaget considered the theory to be much moregeneral and indeed intended that it should describe the development of allforms of knowing from the evolution of intelligence to the history of scientificthinking (Piaget 1970 1972) He studied human children because they presenta readily available sequence of development not because he was interested ineducation or in child-rearing Piagetrsquos theory has been extensively applied instudies of development including cross-cultural and interspecific studies It isprobably the most widely applied and closely examined theory of intelligenceyet devised

Piagetian theory is a structural theory that defines intelligence asorganizational ability This encompasses the way an organism places itself inmoves about in and manipulates its surroundings The theory makes veryspecific predictions about the form a child will use to solve a particular kind ofproblem and while the theory was not designed to assess results many of theforms of organization it describes can be used to evaluate the products ofbehaviour The theory then meets the criteria already outlined it can be used tocompare nonhumans and to assess products of behaviour

The theory is also a stage theory The stages were described on the basis ofobservations of childrenrsquos approaches to tasks ways of solving problems and soon Piagetrsquos scheme includes four major stagesmdashsensorimotor preoperationalconcrete operational and formal operationalmdashand each of the stages includessubstages The sequence is invariant Every child passes through the stages andsubstages in the same order though the age at which each stage is achievedvaries from child to child The defining criteria are qualitative and not based onstatistical trends An important part of most Piagetian experiments is a dialoguebetween the experimenter and the child for a childrsquos reasons for behaving in aparticular way are as enlightening as the products themselves The dialogueaspects cannot of course be applied in prehistory but the typical products ofcertain kinds of organization can be used though the precision of the analysis isreduced One kind of behaviour used often by Piaget is spatial ability arrangingobjects drawing figures reconstructing scenes mapping and so on It is thisemphasis on spatial ability that allows us to use the scheme as a yardstick inprehistory

As important as the stages are to Piagetian theory they are not its core Thislies in Piagetrsquos view of the nature of intelligence and the process ofdevelopment It is a structural theory but does not view structure as innateRather structure is constructed by means of an interaction between individualsand their environment Individuals apply their internal organization to theexternal context of their surroundings and if their organization is inadequatemodify the internal structure based on this experience It is an activeconstruction of a new organization not a behaviourist kind of passive learningThe new organization is in turn applied until it is inadequate then modifiedand so on The result is a sequence of more and more powerful organizationsthat are expressed behaviourally as the stages The theory is not innatist likeChomskian structuralism nor is it behaviourist like Skinnerian psychologyHowever it is cognitive in the sense that the brain actively constructs theorganizations Piaget is not mystical about the structures but sees them as beingthe manifestations of a brain organized by its own action

While Piagetrsquos idea of constructivism is most easily understood in thecontext of ontogeny he intended it to apply to all development includingphylogenetic development Piagetrsquos first publications were in biology at a timewhen Haeckelrsquos idea of recapitulation was still influential and while Piagetnever argued for a strict lsquoterminal addition with accelerationrsquo (Gould 1977) healways maintained that the ontogenetic sequence informed us about thephylogenetic sequence The constructivist nature of development accounts forthe parallel The structure typical of one stage is a logically necessaryprerequisite for the next in that the succeeding stage builds on and out of theorganization of its antecedent This logical necessity must be true of anysequence including both ontogeny and phylogeny Piaget himself did notpursue prehistory although he occasionally mentioned it and was content tostudy his invariant sequence in its most accessible form Even if we do not sharehis certainty about the parallel the scheme still provides a powerful hypothesisfor the phylogenetic sequence This is the approach I will take

A PIAGETIAN APPROACH TO

Because I am concerned here with the appearance of anatomically modernhumans I will deal only with Piagetrsquos final stages of operational intelligenceElsewhere (Wynn 1981 1985 1989) I have considered preoperational stagesThere are two organizational features that are central to operational thinkingbut which do not appear in earlier stages reversibility and conservation Asimple example of reversibility is in arithmetic where every operation has animplied inverse for example addition is the inverse of subtractionConservation is one of Piagetrsquos most famous concepts In transitivity when A=Band B=C it must follow that A=C Something has been conserved across therelationship A preoperational child does not see any logical necessity intransitivity and insists that A and C must be directly compared before theanswer can be known Reversibility and conservation provide thinking withsome very useful organizational features One is precorrection of errors lsquoWhatthis means is that an operational system is one which excludes errors beforethey are made because every operation has its inverse in the systemhelliprsquo (Piaget1970 p 15) With operational thinking an individual can make detailedcontingency plans by in a sense returning to a starting point in thought(reversibility) after anticipating possible difficulties The preoperational thinkercan proceed only by trial and error because reversibility in his or her planningis lacking Reversibility and conservation also allow classification Classificationrequires reversibility (subclass A+subclass Arsquo=B class B-subclass Arsquo=subclass A)and the conservation of some definitional variable across disparate items orgroups Preoperational thought can group accordingly to similarity but cannotcreate logically consistent classifications or reclassifications From this baredescription I hope it is clear that operational thinking is indispensable to manykinds of human behaviourmdashcomplex kinship systems and interplanetaryexploration to name two It must have evolved but when

Concrete operations

Operational structures do not emerge overnight in ontogeny Piagetrsquos scheme isoften caricatured as if this were the case but he never suggested that the transitionfrom one kind of thinking to another occurred in a single flash of insight Ratheroperational thinking is first applied in a narrow range of domains and thenapplied to more and more situations For example children can conserve quantitybefore they can conserve weight However the scheme does entail a developmentwithin operational thinking This is the distinction between concrete operationsand formal operations Piaget considers that these styles of thinking constitutetwo separate stages Formal operations are the final achievement of adultintelligence and contain abstract features not found in concrete operations

Concrete operations are characterized by all of the organizational features ofoperations reversibility conservation precorrection of errors and so on Theyare the first operations to appear and are used to organize tangible things likeobjects and people and simple concepts like numbersmdashhence the termconcrete Hypothetical entities or abstract concepts are not the stuff of concrete

operations Using concrete operations one can classify objects according tocolour and reclassify them according to shape but cannot then hypothesizeabout the class of all classes One accepts the necessity of division as the inverseof multiplication but sees no necessity in the square root of minus oneConcrete operations are nevertheless a powerful organizational tool indeedthey are the principal organizational tool for day-to-day living Tasks toolskinship politics and religion are all organized in this manner Concreteoperations have been documented again and again in cross-cultural contexts(Dasen 1977 Dasen amp Herron 1981) The cross-cultural use of Piagetrsquos theoryis fraught with methodological problems (see p 63) but it does appear that thestage sequence is the same for all groups and that concrete operations areachieved by modern adults everywhere

We can now turn to the archaeological record

Archaeological evidence for concrete operations

There is good evidence for the use of concrete operations by 300 000 years agoI have presented this argument in detail elsewhere (Wynn 1979 1989) but aprecis is appropriate here

One of the advantages of Piagetian theory for prehistory is its emphasis onspatial relations Archaeologists have stone tools in abundance and many of thesepresent patterns that can be used to infer the minimum spatial competence ofthe stone knapper By the end of the Acheulean and perhaps a bit earlier therewere stone tools of considerable spatial sophistication The one familiar to mostprehistorians is the fine handaxe with true bilateral symmetry and lenticularcross-section True symmetry is a Euclidean relation that is achieved only in theconcrete operational stage (Piaget amp Inhelder 1967) The mirroring of a shapeacross a midline requires reversibility because the shape must be inverted inthought It cannot have been achieved by trial-and-error copying because thestone could not be folded to compare one side to the other (as is done in thesymmetry of paper dolls) I must emphasize that I am referring only to thosehandaxes that demonstrate an almost perfect symmetry and one that was theresult of extensive trimming These demand that the knapper have a concept ofsymmetry Most handaxes indeed probably all of the early ones are onlyroughly symmetrical and there are ways to achieve this without a symmetryconcept (Wynn 1985) The lenticular cross-sections of fine handaxes are evenmore demanding These are symmetrical figures but more importantly theycannot be directly perceived by the knapper They must be constructed inthought Especially fine handaxes have a virtual infinity of symmetrical cross-sections all of which the knapper must have considered while trimming thepiece Such a feat is beyond the ability of preoperational trial-and-error planswhich can consider only one variable at a time By the end of the Acheuleanthere are also minimally trimmed handaxes which achieve a remarkablesymmetry with very little trimming These suggest a sophisticated idea of therelation of whole to parts in this case the relation of short trimming segments

CONCRETE OPERATIONS

to the conceived final product Again such a concept requires reversibility andprecorrection of errors and is beyond the scope of trial-and-error plans

The spatial evidence from stone-tool geometry may seem meagre but it isactually quite compelling At least three different spatial relations that were usedrequired operational structures These spatial patterns cannot be produced bypreoperational organizations We must therefore conclude that the minimumcompetence of these later Acheulean stone knappers was concrete operationalintelligence

Formal operations

The structures of formal operational thinking are more generally appliedthan those of concrete operations No longer is the logic applied only toobjects or to real data sets it is used to establish generalities about all possiblesi tuations This development also includes the capacity forhypotheticodeductive reasoning the use of propositional logic and the abilityto disassociate form from content In other words formal operations arecharacteristic of the most sophisticated kind of reasoning we know It is thefinal stage of Piagetrsquos scheme and also the most controversial I will hereinvestigate the possibility that formal operations were associated with theappearance of anatomically modern humans (Homo sapiens sapiens) and thatthis development supplied them with some advantage

In addition to the general claims for hypothetical reasoning and so on Piaget(Inhelder amp Piaget 1958) argues for a very specific change in the logic of formaloperations While concrete operations employ reversibility formal operationscoordinate two kinds of reversibility inversion and reciprocity In inversion atransformation is combined with an inverse that negates the transformation forexample +A-A=0 This is the kind of reversibility used in the classificationexample above (A+Arsquo=B B-Arsquo=A) and also in the whole-part relationsdemonstrated by minimally trimmed handaxes where the addition orelimination of potential trimming segments in thought is a matter of inversion(see Wynn 1979 for detailed argument) Reciprocity is simply a reversal of order(Piaget 1970 p 22) A transformation combined with its reciprocal yields anequivalence (as opposed to negation) for example AB combined with itsreciprocal BA results in B=A This kind of reversibility is beautifullyexemplified by the fine bilateral symmetry of the later handaxes where theshape is mirrored by its reciprocal While the handaxes demonstrate both kindsof reversibility we cannot argue that they were coordinated into a formalsystem Such a coordinated system has some interesting properties and anexample from Piagetrsquos work is in order

Understanding the relation of weight to distance on a balance scale requiresan understanding of proportion which here requires coordinating inversionand reciprocity An individual using formal operations knows that a balance canbe achieved by adding and subtracting weight (inversion) moving the weightsin or out on the arms (reciprocity) or by adding weight to one arm and moving

a smaller weight further out on the other (a coordination of the two) After onlya brief experimentation the formal operational thinker can generalize theproportions to all possible situations Individuals using concrete operations canbalance by adding weight or by moving weight but do not construct a systemof proportions that they see as being always and everywhere true It is not thatconcrete operational individuals cannot balance the weights only that they haveno foolproof system

Archaeological evidence for formal operations

Ideally the arachaeologist would look for evidence of the system of coordinatedreversibilities since this is the most specific difference between formal and concreteoperations Unfortunately such evidence is not easy to find and in the absenceof texts may well be impossible in prehistory The difference between concreteand formal operations is not so much in the end product as in the way thesolution is achieved After all concrete operations can balance the scale butarchaeologists would find only the balanced scale not the thought process behindit Formal operations generally considered are not used on tangible things buton hypotheses generalizations and contentless forms Unlike concrete operationsthey will not be directly preserved in patterns whose minimum necessarycompetence is formal operations Patterns of objects demand at most concreteoperations As a consequence the archaeologist must take one step beyond thephysical evidence and assess competence based on interpretations of prehistoricbehaviour

In the following analysis I will focus on selected examples from fourdomains of behaviour technology subsistence social organization and ritualand art

Although technology alone was sufficient to document concrete operationalthinking at 300 000 years ago it is of little help in documenting formaloperations Stone tools in particular are uninformative even though they are themost abundant source of palaeolithic evidence Much has been written aboutthe sophistication of such techniques as the Levallois and prismatic blade corebut they are no more conceptually difficult than the fine bifaces of the laterAcheulean They may require more skill and practice (though this is debatable)but the minimum conceptual requirement is reversibility in monitoring therelation of core and flakes (wholendashpart relations) and precorrection of errors(Wynn 1985) Nothing in stone knapping requires the coordinated systems offormal operational intelligence Prismatic cores may have made more efficientuse of raw material but efficiency is not necessarily a mark of intelligence

Two post-Acheulean developments in technology are provocative curatedtools and the use of facilities Again both curation and the manner in whichfacilities were used is a matter of interpretation not a pattern directlyobserved Curated tools are not manufactured for a specific task but have ageneral function and are used again and again carried from place to placeElements of Upper Palaeolithic technology were almost certainly curated

FORMAL OPERATIONS

Magdalenian bone points and Solutrean points to name just two Commonsense tells us that curated tools are elements of a longer-range technology thannoncurated tools and therefore more intellectually taxing But how far in thefuture one plans is not in itself relevant to the organizational complexity of thetask Short-range plans can be more complex than long-range plans It is therelation of the elements of the plan not the length of the forecast that iscrucial Unless we know more about the long-term strategy of curated-tool-use their minimum competence remains concrete operational intelligenceThe same is true of facilities stationary technologies like pit falls and fish trapsdesigned to capture without direct human participation (Oswalt 1973) Againnothing about the construction or geometry requires more than concreteoperational intelligence so we must consider the strategies of usemdashsomethingnot directly observable

The strategies of use fall under the rubric of subsistence Here again theUpper Palaeolithic appears more complex than earlier subsistence systemsBut does this reflect as Binford (1982) maintains a more powerfulreasoning ability Upper Palaeolithic subsistence contrasts with earlierperiods in at least two respects first some groups (though by no means all)appear to have specialized on gregarious herd animals and second towardsthe end of the Upper Palaeolithic at least there is reliance on fishingBinford (1982) argues that specialization on gregarious mammals likereindeer is linked to a periodic aggregation of the species At such timesthey can be exploited more easily but such mass hunting almost requiressome form of storage Bahn (1977) in a similar vein argues for selectivekilling of males in autumn Such a system must be based on a year-roundstrategy rather than the short-term hunting or scavenging episodes ofprevious times Binford further argues that the emergence of curatedtechnologies in the Upper Palaeolithic corroborates the use of long-rangestrategies Fish are difficult to exploit (Dennell 1983) except in spawningseasons with the use of facilities This is again a matter of long-range plansThese appear to be fair interpretations but the new strategy is not in factmore intellectually demanding than hunting episodes of a few daysrsquoduration Piaget studied the development of concepts of time (Piaget 1969a)and concluded that the relations used in constructing a concept of time (itis not perceived) are the same as those used in constructing space Ofparticular importance are spatiotemporal operations like substitutionsimilar to those used in conceiving the cross-sections of bifaces (see above)Concrete operations are perfectly capable of constructing temporal framesof years cycles of seasons and cycles of game availability In other wordswhile common sense may suggest that long-range planning is especiallydifficult formal theory argues that the minimum competence is concreteoperational intelligence

The two most easily documented domains of prehistoric behaviourtechnology and subsistence have provided no evidence for formaloperations I will now turn to the more elusive domains of socialorganization and ritual Here I will rely entirely upon interpretations many

of which are themselves controversial As in the domains of technology andsubsistence one difference between Upper Palaeolithic social organizationand earlier systems is a difference in magnitude in this case not of time butof space Bahn (1977) observed that European Upper Palaeolithic hunterscarried or traded shells and other raw material hundreds of kilometres Thegeographic and presumably social range of groups was apparently largerthan those of earlier periods Gamble (1982) argues that the subsistencesystem of Europe during the Upper Palaeolithic required the exchange ofinformation about far-flung resources and conditions and that suchinformation could come only from distant kin real or fictive He seesevidence of these regional information-exchange patterns in thedistribution of distinctive artefact styles which may have operated as indicesof social affiliation (Wynn in press) Piaget himself rarely commented onthe cognitive prerequisites of social organization Nevertheless if we look atthe organizational requirements of exchange systems we must conclude thatthe minimum competence was again concrete operationsmdashsimplereversibility in planning and the organization of real as opposed tohypothetical information It is unnecessary for there to have been a generaltheory of information or style for such a system to work Simplecontingency plans would suffice

It is only in the realm of r itual that we find the glimmerings oforganizations beyond the scope of concrete operations Of course evensimple interpretations of palaeolithic ritual behaviour are controversial andLeroi-Gourhanrsquos (1967) the one I choose to examine is not simple I am notso much interested in Leroi-Gourhanrsquos conclusions about dualistic systemsand malendashfemale symbols as I am in his documentation of associations andrepetitions in parietal art These suggest something interesting about theprehistoric classification system For example 91 per cent of the painted bisonare found in central portions of caves 64 per cent of the bison are associatedwith horses lsquowide signsrsquo dominate the central panels (Leroi-Gourhan 1967pp 112ndash37) I am aware of the problems of cave topography and Leroi-Gourhanrsquos occasionally odd method of counting (Ucko amp Rosenfeld 1967)but he does make a good case for certain redundancies in composition a casemade stronger by its independent discovery by Laming-Emperaire (Leroi-Gourhan 1967 p 110) If bison horses and lsquowide signsrsquo do represent acoherent association of symbolic value as Leroi-Gourhan maintains then theyrepresent a rather sophisticated form of classification Concrete operationalclassification groups members on the basis of tangible similarities nothypothetical commonalities lsquohellipeven in a zoological classificationhellipyoucannot extract two noncontiguous classes like oysters and camels and makethem into a new ldquonaturalrdquo classrsquo (Piaget 1969b quoted Gruber amp Voneche1977 p 398) One could argue that horses and bison are lsquonaturalrsquo in this sensebut this does not appear to be what the Magdalenian painters had in mindThe animals the signs and the positions were grouped according to someabstract common feature (whether or not this is lsquofemalenessrsquo is irrelevant) andnot a tangible similarity This requires formal operations at least as Piaget

FORMAL OPERATIONS

generally defined them Unfortunately even if this assessment were true wehave documented formal operational intelligence only for the Magdalenianperhaps 16000 years ago and this is so close to the present as to beunremarkable

There is evidence of nonutilitarian behaviour prior to the MagdalenianFrom the German site of Hohlenstein-Stadel there is a lion-headedanthropomorph carved in ivory that dates to before 30 000 years ago (Marshack1989) tempting one to posit an iconographie system as subtle as that inMagdalenian art Marshack makes the persuasive argument that such figuresrepresent lsquohellip the end product of a long developmentrsquo Engraved bones appearin European sites as early as 190 000 years ago (Gamble 1980) Marshack (1972)has made a case for the notational function of such engraved bones and leavingaside his interpretation of lunar calendars it appears that the makers werekeeping account of something Other provocative bits of evidence argue forsymbolic behaviour during the Middle Palaeolithic but interpretations arecontroversial (Chase amp Dible 1987 Marshack 1989) However none of thesedemonstrates associational patterns as complex as those of Magdalenian parietalart If the engraved tablets were notational tallies of some sort this requires atmost a concept of number Piaget (1952) has shown that a concept of numberis based on concepts of class inclusion and seriation both of which areconstructed with the simple reversibilities of concrete operations Given thetool geometries at 300 000 years ago the possible appearance of notation at 190000 is not a surprise Even lion-headed anthropomorphs are within the abilitiesof concrete operations (lions and humans share the tangible feature of lsquoanimatebeingrsquo though I suspect that this is again not quite what the Aurignacian carverhad in mind) Such evidence suggests a subtle symbolic system perhaps but noleap in intelligence

To summarize the archaeological evidence for formal operations is mostlynegative None of the technological or subsistence developments that appearedafter 300000 years ago requires more than concrete operational organizations Iinclude here both developments associated with archaic H sapiens andanatomically modern humans At most we can conclude that technology andsubsistence developed a larger temporal scope This is not organizationalhowever and all remain within the competence of concrete operations Thesame is true of the greater geographic scope of certain Upper Palaeolithic socialcontacts The only possible evidence for the most abstract organizations offormal operations comes from Magdalenian parietal art This is based on Leroi-Gourhanrsquos controversial theory and even if it is sound it places formaloperations so late in prehistory as to be unremarkable It cannot be used toargue for the intellectual supremacy of the anatomically modern humans overarchaic H sapiens

Because archaeology can document only the minimum necessarycompetence it is still possible that anatomically modern humans did in factemploy formal operations but that the relevant behaviours have simply left noclues This is a weak argument in its absence of evidence but it is at least

possible A closer examination of formal operations especially the cross-culturalevidence will I think weaken the possibility still further

Critique of formal operations

Piagetrsquos stage of formal operations is elusive in prehistory as I have just shownIt is also elusive in the modern world lsquoThe very few cross-cultural studies thathave included tasks of the formal operational stage have found very littleevidence of formal operational performancersquo (Dasen amp Herron 1981 p 332)Concretre operations on the other hand appear to be universal in adultsUnfortunately cross-cultural studies are fraught with methodological problemsespecially when using Piagetian tasks As Piaget observes few anthropologistsare well enough trained to administer the tests and few psychologists are familiarenough with a native people to create a comfortable testing situation (Bringuier1980) Also most cross-cultural applications of Piaget are based on a small setof tasks occasionally only one and this is insufficient for a reliable assessment(Cole amp Scribner 1977) Nevertheless the absence of formal operations isthought-provoking especially given the complex behaviours produced by so-called primitive peoples Micronesian sailors can travel hundreds of milesbetween tiny atolls using an elaborate system of sidereal navigation oceancurrents birds and so on and yet they do not perform well on Piagetian tasks(Gladwin 1970) Much of the discrepancy appears to be in the domains testedBalancing a scale is familiar and relevant to Swiss high school students butunfamiliar to a Micronesian sailor Indeed not all Western adults perform atthe formal operational level (Dasen amp Herron 1981) Formal operations mayin fact be a kind of thinking used by literate educated adults Results suggestthat some kind of schooling may be necessary for formal operationalperformance (Rogoff 1981) and more specifically that experience in textsmay be of crucial significance (Scinto 1984) The nature of texts is self-reflectiveand forces the writer to transcend content and deal in form This may in turnlead the individual to use this new style of organization in other domainsFormal operations then may be a rather artificial style of thinking one learnsin school This makes it no less useful but does seriously challenge its status asa stage of intellectual development

The problem we have just encountered is that of disentangling intelligence assome inherent competence from culture as a learned set of solutions and itbears heavily on our understanding of recent human evolution Piaget himselfwas aware of the tangle and indeed incorporated it into his constructivist viewof ontogeny lsquoMoreover the history of formal structures [formal operations] islinked to the evolution of culture and collective representations as well as [to]their ontogenetic historyrsquo (Inhelder amp Piaget 1958 as quoted in Gruber ampVonech 1977 p 436) Because new cognitive structures are constructed througha dialectic (assimilation and accommodation) between existing structures andthe external milieu some aspect of the milieu must force the disequilibrium inthought that leads to the reorganization of concrete operations into formal

CRITIQUE OF FORMAL OPERATIONS

operations Piaget sees the social environment as being crucial in this transitionan environment that has its own history In Western thought this historyincludes reliance on texts formal proofs deductive logic and so on But thehistory of Micronesian sailors did not include these things and as aconsequence their final operational structures appear rather different fromthose of Western adults This is not to say that the cultural milieu determinesintelligence only that it presents rather different problems for intelligence tosolve

So formal operations may not be universal They are probably an artefact ofoperational relationsmdashreversibility conservation etcmdashapplied in a Westerneducated milieu that has its own history As a consequence it may be ratherfoolish to look for them in prehistory Nevertheless the enigma of formaloperations does supply some insights into the relationship between intelligenceand culture a relationship that has probably held for some 300 000 years

Conclusions

Before entering into a more speculative discussion I would first like to reviewthe conclusions of this Piagetian analysis

The geometry of 300 000-year-old stone tools requires the operationalorganizations supplied by reversibility and conservation Preoperationalorganizations are incapable of conceiving or executing the fine bilateralsymmetries and the multiple symmetrical cross-sections of later Acheuleanbifaces The minimum necessary competence was that of concrete operationalintelligence I am not arguing that later Acheulean culture was indistinguishablefrom modern culture only that the cognitive organizations behind it wereequivalent to those of most modern culture

None of the Stone-Age developments after 300 000 years ago require anintelligence more sophisticated than concrete operations I include heretechnological developments such as prismatic cores curated tools and facilitiessubsistence developments such as specialization and fishing both of whichrequire long-term plans and social developments such as exchange networksNone of these requires more than the mental reversibility and conservation ofconcrete operations The only Stone-Age patterns that would have required themore abstract organizations of formal operations are those of Magdalenianparietal art assuming of course that Leroi-Gourhanrsquos scheme is correct In otherwords from a Piagetian perspective Middle Palaeolithic and Upper Palaeolithiccultures are indistinguishable

As a consequence we have no rigorous base from which to argue thatanatomically modern humans had some innate capacity for culture that wasmore powerful than that of their archaic antecedents Piaget does see a relationbetween formal operations and physiology lsquoIt seems clear that the developmentof formal structures in adolescence is linked to the maturation of cerebralstructures [But] the exact form of linkage is far from simple since theorganization of formal structures must depend on the social milieu as wellrsquo

65REFERENCES

(Inhelder amp Piaget 1958 quoted in Gruber and Voneche 1977 p 435)However the cross-cultural evidence suggests that the social milieu may in factbe the only relevant variable in the acquisition of formal operations Formaloperations are rarely achieved outside of Western educated adults and are noteven universally true for these It appears to be a style of operational thinkingrather than a stage The basic organizational principles of operational thoughtappear to be the final stage generally attained by modern humans Thedifferences between Micronesian sailors and Swiss high school students are amatter of social and cultural milieu not some inherent difference in the cerebralphysiology of the populations Why should it have been any different in the finalperiods of the Stone Age

If physiological evolution cannot be used to account for the documentedincrease in complexity what can Certainly Upper Palaeolithic culture is morecomplex than Middle Palaeolithic culture in terms of the number of itselements the temporal range of its subsistence and the geographic range of itssocial contacts But so is Western culture compared with that of highland NewGuinea and so is 20th-century technology compared with that of the 19th It isculture itself that has become more complex and odd as it sounds moreintelligent The complex social and technological fabric of which we are a partallows us to solve complex organizational problems lsquothe most generalized formsof thought those that can be dissociated from their content are by that veryfact forms of cognitive exchange or of interindividual regulationrsquo (Piaget 1971pp 360ndash1) This lsquointerindividual regulationrsquo has I maintain been the crucialcomponent of human behaviour for the last 300 000 years It has increased inscope and organizational power but this is not a matter of brain evolution Wecannot disentangle the evolution of intelligence from the evolution of culturebecause from 300 000 years ago they are one and the same thing UpperPalaeolithic culture may have been more finely adapted than that of the MiddlePalaeolithic but this was not because the participants were more intelligentTheir culture was simply different and it appears in the long run moresuccessful

References

Bahn PG 1977 Seasonal migration in southwest France during the late glacial periodJournal of Archaeological Science 4 245ndash57

Binford LR 1982 Comment on lsquoRethinking the Middle Upper Palaeolithictransitionrsquo by RWhite Current Anthropology 23 177ndash81

Binford LR 1983 In pursuit of the past London Thames amp HudsonBringuier J 1980 Conversations with Jean Piaget Chicago University of Chicago PressChase P amp HDibble 1987 Middle Palaeolithic symbolism a review of current evidence

and interpretations Journal of Anthropological Archaeology 6 263ndash96Cole M amp SScribner 1977 Developmental theories applied to cross-cultural research

New York Academy of Sciences Annals 285 366ndash73Dasen P 1977 Piagetian psychology cross-cultural contributions New York Garden Press

Dasen P amp AHerron 1981 Cross-cultural tests of Piagetrsquos theory In Handbook of cross-cultural psychology vol 4 developmental psychology HTriandis amp AHerron (eds) 295ndash341 Boston Allyn amp Bacon

Dennell R 1983 European economic prehistory London Academic PressFoley RA 1991 How useful is the culture concept in early hominid studies In The

origins of human behaviour RAFoley (ed) Ch 2 London Unwin HymanGamble C 1980 Information exchange in the palaeolithic Nature 283 522ndash3Gamble C 1982 Interaction and alliance in palaeolithic society Man 17 92ndash107Gladwin T 1970 East is a big bird Cambridge Ma Harvard University PressGould S 1977 Ontogeny and phylogeny Cambridge Ma Harvard University PressGould S 1981 The mismeasure of man New York NortonGruber H amp JVoneche 1977 The essential Piaget New York Basic BooksInhelder B ampJPiaget 1958 The growth of logical thinking from childhood to adolescence

(Trans AParsons amp SMilgram) New York Basic BooksLeroi-Gourhan A 1967 Treasures of prehistoric art (Trans NGuterman) New York

AbramsMarschack A 1982 Upper Palaeolithic notation and symbol Science 178 817ndash28Marshack A 1989 Evolution of the human capacity the symbolic evidence Yearbook of

Physical Anthropology 32 1ndash34Oswalt W 1973 Habitat and technology New York Holt Rinehart amp WinstonPiaget J 1952 The childrsquos conception of number (Trans CGattegno amp FHodgson)

London Routledge amp Kegan PaulPiaget P 1969a The childrsquos conception of time (Trans APomerans) London Routledge amp

Kegan PaulPiaget J 1969b The psychology of the child (Trans HWeaver) London Routledge amp

Kegan PaulPiaget J 1970 Structuralism (Trans CMaschler) New York HarperPiaget J 1971 Biology and knowledge Chicago University of Chicago PressPiaget J 1972 The principles of genet ic episte mology (Trans WMays) London Kegan PaulPiaget J amp BInhelder 1967 The childrsquos conception of space (Trans FLanglon amp JLunzer)

New York NortonRedman C 1978 The rise of civilization San Francisco FreemanRogoff B 1981 Schooling and the development of cognitive skills In Handbook of cross-

cultural psychology vol 4 developmental psychology HTriandis amp AHerron (eds) 233ndash94 Boston Allyn amp Bacon

Scinto L 1984 The architectonics of texts produced by children and the developmentof higher cognitive functions Discourse Processes 7 371ndash418

Ucko P amp ARosenfeld 1967 Palaeolithic cave art New York McGraw HillWynn T 1979 The intelligence of later Acheulean hominids Man 14 371ndash91Wynn T 1981 The intelligence of Oldowan hominids Journal of Human Evolution 10

529ndash41Wynn T 1985 Piaget stone tools and the evolution of human intelligence World

Archaeology 17 31ndash43Wynn T 1989 The evolution of spatial competence Urbana University of Illinois PressWynn T In press The evolution of tools and symbolic behaviour In The evolution of

human symbolic behaviour ALocke amp CPeters (eds) Oxford Oxford University Press

5 The invention ofcomputationally plausibleknowledge systems in theUpper PalaeolithicSHELDON KLEIN

In recent years with the expansion of computer science researchers in thecognitive sciences have been attracted towards the use of computational modelsfor understanding the structure of human thought Such work as has beendone has proved extremely powerful for tackling contemporary situations(Haugeland 1985) and so it is appropriate to ask whether such approacheshave the potential for explaining the evolutionary development of systems ofhuman knowledge In this chapter I shall examine how such knowledge systemsmay be structured and whether there is evidence for their origins in humanprehistory

The problem of computing human behaviour by rules

Contemporary artificial-intelligence researchers find the problem ofcomputing human behaviour by rules intractable for large-scale knowledgesystems While excellent results have been obtained for small-scale knowledgedomains the time it takes to make such computations can increaseexponentially or even combinatorially with the size and heterogeneity of theknowledge system If the human brain like a computer is a finite-stateautomaton then the problem of generating and parsing behaviour must presentthe same computational difficulty for the human mind1 The problem ofmaking such computations at a pace fast enough for ordinary social interactioncan be solved if appropriate constraints apply to the structure of the rulesThere seems to be evidence that systems of such constraints were invented inthe Upper Palaeolithic and were of such power as to guarantee that the timenecessary for computation of behaviour would increase only linearly withthe size and heterogeneity of the world knowledge systems The evidencecan be found in the material and symbolic artefacts of a variety of culturesand the major sources are classification schemes divination systemsiconographie systems language structures and shamanistic mythic or religoussystems

The purpose of this chapter is to establish a model by which thecomplexities of human behaviour can be generated using a system of rules that

68 KNOWLEDGE SYSTEMS IN THE UPPER PALAEOLITHIC

is consistent with how human thought operates is parsimonious allowing forthe processing and manipulation of knowledge to occur rapidly is internallyconsistent and permits knowledge to be accumulated In other words this is anattempt to construct a model of the mind that is capable both of being practicalin computational terms and of accounting for the heterogeneity in humanknowledge systems The key attribute for this lies in the use of rules governingthe association and transformation of items of knowledge This in turn rests onthe use of formal logic for treating the classification of knowledge andconsequently such formal logic provides the methodological framework forwhat follows

The basic structure of the invention

Fundamentally there was one computational invention capable of unifying thefull range of human sensory domains and consisting of an analogical reasoningmethod used in combination with global classification schemes The structure ofthe human brain may be a factor in the history of this invention but its utilityexists independently of such a connection Every culture seems to have a globalclassification scheme in the history of its knowledge structures and usually suchschemes can be linked to myth systems The use of this invention to computehuman behaviour is explained fully elsewhere (Klein 1983 1988) The strongequivalence operator of logic is shown to define ATOs (appositionaltransformation operators) that relate the input and output states of behaviouralrules by analogical transformations It is argued that a given culture has a relativelysmall set of such ATOs and that they apply to diverse domains of human behaviourwith a processing time that increases only linearly with the number of elementsrelevant to those rules The global classification scheme makes it possible toselect and apply the appropriate ATOs in a variety of domains by specifyingequivalence classes of elements that may serve as substitution sets for the extensionof each ATO The result can be compared to a set of canonical analogies forwhich the extension and application are determined by equivalent analogues inthe global classification scheme The classification scheme for Chinese culture(Table 51) is a typical example (Klein 1983 p 159)

Each semantic domain is seen to have its equivalent in another domain Forexample lsquoEastrsquo is the direction counterpart of the element lsquowoodrsquo and its seasoncompanion is lsquospringrsquo Each of these terms is itself a metonym representinganother class of items The Chinese scheme is also linked to the I Chingdivination system which may be viewed as a knowledge-based query systembased on analogical principles The divination system is associated with a set ofcanonical texts containing specific terms of reference that function asmetonyms for higher-level classes Each text may be viewed as a formulaicbehaviour pattern awaiting the substitution of appropriate values for its variableterms by the user of the divination system The computationally difficultproblem is the selection of a culturally consistent set of elements for

the terms in the text For a computer program operating with rules formulatedin propositional logic this could involve a combinatoric computation processThe Chinese global classification scheme reduces the process to looking up thecorresponding elements in a table However the classification scheme used in agiven divination is actually a transformation of the basic one shown in Table 51The divination process yields an ATO which generates an analogical realignmentof the original table in correspondence with the situation of the moment asdetermined by the divination process A widespread African divination systemoperates on the same principles and they can be seen to work also in the visualand verbal iconography of Navaho curative ceremonies Tibetan and esotericJapanese Buddhist iconography functions as an ATO system which is visualencoding of ATOs applicable to specific world domains in conjunction with amyth system and a global classification scheme (Klein 1983)

At this point let me offer some intuitive examples of how ATOs work inverbal and visual analogical reasoning problems and also examples of analogicalcomputation of behaviour using situation descriptions linked by ATOs (Klein1983 pp 152ndash4)

ATOs relate situation descriptions in the form of arrays of features A two-valued version can be defined by the strong-equivalence operator of logicwhich can be used to compute ATOs

The lsquorsquo means that a result is to be computed using the above truth table ATmeans lsquotruersquo a lsquo0rsquo means lsquofalsersquo and lsquorsquo means lsquodoes not applyrsquo ATOs may alsobe computed with the rules for binary addition (mod-2 arithmetic) if theinterpretations of 1 and 0 are reversed

and has a mathematical group property Consider for example the ATO relatingtwo hypothetical feature arrays A and B Each feature value in A is matched withits positional counterpart in B to compute its component in the ATO AB

Some simple analogies will illustrate how ATOs work (Klein 1983 pp152ndash4)

A feature array referencing lsquomalersquo lsquofemalersquo lsquoyoungrsquo lsquoadultrsquo lsquoloversquo lsquohatersquolsquolightrsquo and lsquodarkrsquo is sufficient to formulate the following analogy

The same method can be applied to visual analogies For example if a set ofvisual features is used to create a pictorial analogy (Fig 51) the answer can becalculated using ATOs (Fig 52) If we give natural-language interpretations tothese visual features we can obtain the results shown in Figure 53

BASIC STRUCTURE

Another example

Figure 51 A pictorial anologyKey M=male F=female Y=young A=adult L=love H=hate Lt=light D=dark

Figure 52 Calculation of a pictorial analogy

Figure 53 The pictorial analogy with a natural-language interpretation

Complex analogies may also be computed as in the following abstractexample

A concrete illustration of this abstract example is as follows

BASIC STRUCTURE

Where La means lsquoloves Arsquo etc $ means lsquohas moneyrsquo and Ma means lsquomarried toArsquo etc the X and Y states may be represented as follows

If we depict lsquolovesrsquo as a nose pointing at the beloved (in between if twoloves) if a noseless state means lsquoloves no onersquo if holding hands depicts lsquomarriedtorsquo and if a lsquo$rsquo indicates lsquohas moneyrsquo we obtain the visual interpretation ofFigure 54

Figure 54 A visual interpretation of XrarrY where X is lsquoA loves B has no $ and isunmarried B loves A has no $ and is unmarried C loves no one has $ and isunmarriedrsquo and Y is lsquoA loves B has no $ and is married to B B loves A has no $ andis married to A C loves no one has $ and is unmarriedrsquo

Continuing with this complex example

This yields the visual interpretation of Figure 55

Figure 55 A visual interpretation of ZrarrW where Z is lsquoA loves no one has no $ and ismarried to B B loves A has no $ and is married to A C loves A has $ and is unmarriedrsquoand W is lsquoA loves no one has $ and is married to C B loves no one has no $ and isunmarried C loves A has $ and is married to Arsquo

If we then postulate a situation P

we can compute its successor state by analogy with the combined results of XmdashV and ZmdashW by solving

where which can be represented as follows

A loves B and C has no $ and is married to BB loves C has $ and is married to A and C Cloves A and B has $ and is married to B

lsquosurrealisticrsquo interpretation

Figure 56 A visual interpretation of the lsquosurrealisticinterpretationrsquo (XY) (ZW) lsquoA loves B and C has no$ and is married to B B loves C has $ and is married toA and C C loves A and B has $ and is married to Brsquo

A loves B and C has no $ and isunmarried B loves A has no $ and isunmarried C loves A has $ and isunmarried

A loves B and C has $ and is marriedto C B loves no one has no $ and isunmarried C loves A has $ and ismarried to A

This yields Figure 57

ATOS AND SHAMANISM

The visual interpretation obtained is that in Figure 56

ATOs language and culture

I wish to argue that the invention of computational knowledge consisting ofthe idea of a global classification scheme in combination with behaviour rulesrelated by a limited set of analogical transformation operators was responsiblefor the elaboration of language and culture structures in a process of coevolutionPhrase-structure grammar operates on ATO principles this can be verified bycreating a categorial grammar in which grammar codes consist of appropriatelychosen binary integers If one adds information indicating right- or left-combining properties and also adds semantic-feature vectors it is possible touse ATO logic for decoding both syntax and semantics in the same notationAn implication is that world knowledge systems and language systems havecoevolved If this is so then 1 The Sapir-Whorf hypothesis that the structure of grammar determines world

view may remain true synchronically diachronically however the two systemsare in an intimate relationship of mutual influence and modification

2 While the ATO model does not lsquorefutersquo Chomskyrsquos view that there is aninnate genetic basis for language structure it makes that assumption unnecessaryto account for human linguistic behaviour The structure of the human brainmay be a passive factor in the invention of structures that are computationallyefficient in a given lsquohardwarersquo environment

The extension and elaboration of culture content can be interpreted as theextension of the global classification scheme to new elements and as theapplication of existing ATO patterns to new behavioural situations The result isa formally definable explanation of the process of creating new patterns ofbehaviour by analogy with patterns in other domains If this process is part ofthe growth of a culture and its social institutions then its symbolic behaviouraland material artefacts will contain many homologies It is this aspect that gives aculture its coherency and enables its members to know what culture elementsare appropriate

Figure 57 A visualinterpretation ofPrarrP((XY) (ZW))where P is lsquoA loves B and Chas no $ and is unmarriedB loves A has no $ and isunmarried C loves A has $and is unmarr iedrsquo andP((XY) (ZW)) is lsquoA lovesB and C has $ and is marriedto C B loves no one has no

ATOs and the ontogeny of shamanism

Religious systems can be interpreted as the symbolic medium in which ATOsystems are encoded The hierarchy of ATOs that govern the structure of aculture are inevitably encoded surrealistically in verbal and plastic domainsincluding myth systems and representations of spirits and deities

Consider the following aspects of the computation of behaviour with ATOs(Klein 1983 p 154)

If a sequence of events A B C D occurs then

If we wish to obtain a state E instead of D without changing any of the ATOswe derive by analogy a sequence leading to E by replacing A B C respectivelywith A(DE) B(DE) C(DE) If we wish to make a plan that specifiesmore than one goal state in the event sequence we must alter some ATOs

The meaning of lsquoculturally defined behaviourrsquo is that members of a societyplan in a way that minimizes the level and number of ATOs affected It followsthat deviant behaviour may be interpreted as behaviour that violates acceptablelevels and numbers of ATOs ATO patterns are part of the knowledge acquiredby children They are encoded in multiple media of expression both materialand symbolic and are the source of metaphor It is this encoding that gives formto a culture and it is the widely distributed presence of ATOs in theenvironment that makes calculation of social behaviour computationallyfeasible for the human mind

The emergence of a canonical hierarchy of ATOs applicable to multipledomains of social reality through the mediation of a global classification schemewould be a natural consequence of organizing social life on the basis of ATOlogic If we make the assumption that the human mind encodes ATOs in iconicimagery we may also suggest that such imagery is given metaphysicalinterpretation A hierarchic ATO system may be interpreted by the humanmind as hierarchy of spiritual beings and the spirit journey of a shaman seekingto resolve problems in a spirit realm can be interpreted as precisely the kind ofATO manipulation described above Magic spells and rituals would appear asdevices for inserting desirable ATOs in given situations and it might be possibleto predict their form and general content from the global classification schemeThe implication of this model is that shamanism is a consequence of theadoption of computationally plausible knowledge systems Several theoreticalpossibilities are implied

ATOS AND SHAMANISM

(1) The ATO system concept was invented once and spread by diffusion(2) Computation with ATO logic may be a part of the functioning of the

human brain(3) ATO systems may have been invented independently in conjunction with

elaboration of social life(4) If (2) and (3) are true then the concept of lsquothe shamanistic traditionrsquo may

reflect phenomena which are of independent origin (Eliade 1964 Artscanada19734)

The evidence of Leacutevi-Strauss

The ATO logic I have described in more detail elsewhere (Klein 1983) is amodel of the structuralism of Claude Leacutevi-Strauss It was originally formulatedin 1976ndash7 in an attempt to replicate the reasoning processes that Leacutevi-Straussused in Mythologiques (Leacutevi-Strauss 1964ndash71) Given his semantic units thearguments linking myth structures can be verified and replicated by ATOcomputation (Klein 1977) My 1983 paper was intended as a validation of theATO concept with independent data La penseacutee sauvage (Leacutevi-Strauss 1962) is anexplication of human reasoning with ATO systems the four volumes ofMythologiques represent an overwhelming body of empirical evidence that ATOsystems exist The work is an analogue of historial reconstruction linguisticsWhile he does not reconstruct a protosystem Leacutevi-Strauss has demonstrated thatproto-ATO systems must have existed at least as early as the Upper Palaeolithicand that they have contemporary descendants Given this perspective muchwork seemingly critical of the structuralism of Leacutevi-Strauss can be reinterpretedas supportive (Hodder 1982 Miller 1982 Tilley 1982 Wylie 1982)2

Testing the ATO model in historical time

My discussion elsewhere of ways one might obtain empirical validation of theATO concept (Klein 1983 p 178) includes the following observations

(4) the ATO model can be used as a heuristic device to suggest culturalcorrelations that can be verified by other methods This approach might evenextend to predictions about the location of buildings with specific functionsin archaeological sites Analysis of symbolic artefacts by ATO logic mighthelp to decode or unlock large systems of correlations hellip

(5) One might examine the possibility that ATOs can be sources of socialand cultural change A large-scale classification system can imply a structureduniverse which no participant in a culture can contemplate as a whole If aclassification system incorporates 50 features it can imply a conceptual universewith 2n=250 elements ATOs that function in a subset of the implied universecan be used as an exploratory tool to extend knowledge by analogy A suddenexternally caused change in iconography (or mythology) would imply a new

system of correlations and would offer the potential for new analogies aboutthe structure of the world that might imply new patterns of behaviour A testof such a possibility would require an adequately documented historicalsituation

Major testing of the theory requires a detailed analytic perusal of broad streamsof history in a number of cultures I would cautiously cite Toynbee (1934ndash61)and Spengler (1926ndash8) whose general theoretical analyses can be interpreted inan ATO framework I do not endorse any particular details of their analyses butrather note that in their surveys of massive amounts of data they found relationsand structures which are compatible with the theory of ATO systems Theprinciple that I value in Toynbee is his relation of religious systems to socioculturalsystems (after disassociating his ideas from his personal religious bias) In the caseof Spengler I value the perception of the analogical relationships among theartefacts of a culture (Spengler 1926 p 47)

From this moment on relations and connexionsmdashpreviously oftensuspected sometimes touched on but never comprehendedmdashpresentedthemselves in ever-increasing volume The forms of the arts linkedthemselves to the forms of war and state-policy Deep relations were revealedbetween political and mathematical aspects of the same culture betweenreligious and technical conceptions between mathematics music andsculpture between economics and cognition-forms Clearly andunmistakably there appeared the fundamental dependence of the mostmodern physical and chemical theories on the mythological concepts of ourGermanic ancestors the style-congruence of tragedy and power-technicsand up-to-date finance and the fact (bizarre at first but soon self-evident)that oil-painting perspective printing the credit system long-rangeweapons and contrapuntal music in one case and the nude statue the city-state and coin-currency (discovered by the Greeks) in another were identicalexpressions of one and the same spiritual principle

Conclusions

The criterion that a model of human cognition must account for the ability ofhumans to compute social behaviour in real time has to my knowledge notbeen addressed before The thesis that ATO systems were invented in the UpperPalaeolithic and are responsible for the growth of sociocultural structures providesa mechanism for a variety of seemingly disparate theories It makes structuralismand systems anthropology appear as different aspects of the same phenomenonand if ATO logic proves to be hardwired in the human brain it will be particularlycompatible with sociobiology3

CONCLUSIONS

1 An assumption that the brain is a massively parallel computer does not mitigate theproblem The addition of n parallel processors can reduce the computation time by afactor of n but the problem domain involves a processing time that can increasecombinatorially with the size of the data base If an additional computer processor isadded for each new item in the data base the processing time may increase at a rate ofnn=(n-1) A connectionist brain model presents an analogous difficulty the needfor combinatorially increasing processing time is replaced by a need for combinatoriallyincreasing connectivity

2 The seemingly supportive evidence of Leroi-Gourhan (1965) is not supportive becauseit is not substantiated by knowledge of the global classification scheme of the culturethat produced the Lascaux paintings A recent discussion of the evidence is containedin Marshack (1985 pp 538ndash9)

3 A very recent analysis suggests that a developmental sequence in lithic technologydating to the MiddleUpper Palaeolithic transition in the Negev reflects the groupconcept and ATOs in the cognitive processes of the concerned tool-makers (Klein1990)

References

Artscanada 19734 Stones bones amp skin ritual and shamanic art 184ndash7 30th anniversaryissue

Blofeld J 1978 Taoism the road to immortality Boulder ShambhalaEliade M 1964 Shamanism archaic techniques of ecstasy Princeton Princeton University

PressHaugeland J 1985 Artificial intelligence the very idea Cambridge Ma MIT PressHodder I 1982 Theoretical archaeology a reactionary view In IHodder (ed) Symbolic

and structural archaeology 1ndash16 Cambridge Cambridge University PressKlein S 1977 Whorf transforms and a computer model for prepositional appositional

reasoning Paper presented at the Applied Mathematics Colloquium University ofBielefeld at the Computer Science Colloquium University of Paris-Orsay and at ajoint colloquium of the Anthropology and Computer Science DepartmentUniversity of California Irvine

Klein S 1983 Analogy and mysticism and the structure of culture Current Anthropology24 151ndash80

Klein S 1988 Reply to SDSiemensrsquo critique of SKleinrsquos lsquoAnalogy and mysticism andthe structure of culturersquo Current Anthropology 29 478ndash83

Klein S 1990 Human cognitive changes at the MiddleUpper Palaeolithic transitionthe evidence of Boker Tachtit In The emergence of modern humans the archaeologicalperspective PAMellars (ed) 499ndash516 Edinburgh Edinburgh University Press

Leacutevi-Strauss C 1962 La peacutensee sauvage Paris PlonLeacutevi-Strauss C 1964ndash71 Mythologiques 4 vols Paris PlonLegeza L 1975 Tao magic the Chinese art of the occult New York Pantheon BooksLegge J (trans) 1964 (1899) The Yi King 2nd edn New Hyde Park New York

University BooksLeroi-Gourhan A 1965 Preacutehistoire de lrsquoart occidental Paris MazenodMarshack A 1985 More on serpents in the mind Current Anthropology 26 537ndash9Miller D 1982 Artefacts as products of human categorization processes In Symbolic and

structural archaeology IHodder (ed) 17ndash25 Cambridge Cambridge University Press

81REFERENCES

Spengler O 1926ndash8 (1918ndash22) The decline of the West Vol 1 1926(1918) Form andactuality Vol 2 1928(1922) Perspectives of world-history New York Alfred A Knopf

Tilley C 1982 Social formation social structures and social change In Symbolic andstructural archaeology IHodder (ed) 26ndash38 Cambridge Cambridge University Press

Toynbee AJ 1934ndash61 A study of history 12 volumes London Oxford University PressWylie MA 1982 Epistemological issues raised by a structuralist archaeology In

Symbolic and structural archaeology IHodder (ed) 39ndash46 Cambridge CambridgeUniversity Press

Yu-lan Fung 1953 (1934) A history of Chinese philosophy Vol 2 (Trans Derk Bodde)Princeton Princeton University Press

6 An interactive growth modelapplied to the expansion ofUpper Palaeolithic populationsEZRA BW ZUBROW

Outlined against a blue-gray October sky the Four Horsemen rode againhellip Indramatic lore they are known as Famine Pestilence Destruction and Death These areonly aliases

Grantland Rice

And power was given unto them over the fourth part of the earth to kill withsword and with hunger and with death and with the beasts of the earth

Revelation 68 There has been considerable speculation on the relationship of the twosubspecies Homo sapiens sapiens and Homo sapiens neanderthalis during the periodjust prior to the Neanderthalsrsquo extinction There is little fact This chapter usesa simulation model to create possible scenarios for the interaction of the twospecies at different locations in Europe at about 30 000 BC The models indicatethat there is a very small window which existed in the growth and interactionrates of the two species which would have allowed the Neanderthals to continueFurthermore this chapter suggests that the Neanderthal demise was more likelythe result of small numbers and chance in a competitive situation than lack ofadaptive characteristics Finally it suggests that one advantage that H sapienssapiens had was its more rapid rate of attaining demographic and geographicstability As is the case in all simulations reality is modelled it is not re-createdTherefore this chapter admittedly contributes to the realm of speculation ratherthan that of fact

The background

Since 1856 scholars have been aware of the unusual skeleton found at a quarryin the Neander valley near Duumlsseldorf Now with more than 100 sites analyseda broadly drawn picture of Neanderthal adaptation has been developed throughthe efforts of numerous archaeologists and physical anthropologists A briefsketch would note that they were hunter-gatherers with the emphasis probablyon gathering lived in small family bands made stone tools with Mousterian

techniques and were sufficiently sophisticated to bury their dead It was asuccessful adaptation surviving major changes in climate Neanderthals appearedin Europe about 125 000 years ago and became extinct approximately 30ndash35000 years ago During their 100 000-year existence there was a sufficientgeographical radiation for them to have been found in Europe the MiddleEast and Asia

Early H sapiens sapiens such as Cro-Magnon generally correspond to theUpper Palaeolithic in Europe From approximately 35 000 to 10 000 years BPcultural variation increased as indicated by the diversity of the PerigordianAurignacian Magdalenian and Solutrean cultures as well as by the increasinglyfunctionally specific types of sites As hunters and gatherers they were able toadapt to both the climate of the last glaciation in Europe and the warmingwhich followed It has been assumed that they lived in small bands of about 75to 100 Ethnographic analogy has suggested that labour use was relativelyefficient and that their existence was not limited to a Malthusian minimum Tosome extent this viewpoint is substantiated by the great art of the period atplaces such as Lascaux and by the rapid adaptive and eventually culturaldiversity of our species

Considerable interest has been expressed in the transitional examples ofhominids It has long been suggested that the Neanderthals found at Tabun andAmud in Israel were aberrant They are more similar to H sapiens sapiens thanare many other skeletons Similarly the sites of Skhul and Qafzeh contain classicNeanderthal Mousterian tools but modern hominids Alternatively at St Ceacutesaireone finds Neanderthal skeletons associated not with Mousterian but withChatelperronean stone tools

The stage is thus set to enter the realm of speculation and consider what therelationships between the two populations may have been There are severalpossibilities which include 1 H sapiens sapiens and Neanderthals are two distinct populations with the former

deriving from the ancestral latter2 H sapiens sapiens and Neanderthals are two distinct and partially

contemporaneous populations in which the latter became extinct due tocompetitive pressures from the former

3 H sapiens sapiens and Neanderthals are one ancestral population and the sapienssapiens characteristics survived due to adaptive or competitive advantage

The rest of this chapter will be concerned with examining how the simulationmodel addresses this transition and these three possibilities

The model

Imagine two bands of hominids moving through a Pleistocene landscape followingtheir respective game animals The sun rises and falls on their respective campsAs the seasons pass each traverses a route through their territory These routes are

THE MODEL

84 A GROWTH MODEL APPLIED TO THE UPPER PALAEOLITHIC

established by the schedule of harvesting wild plants game routes predators andthe location of water They are also determined by a variety of imponderablesvolition religion idiosyncratic personality and simple chance

These populations are not static They grow and decline they break up andreaggregate This depends on many factors the local environment the skill ofthe subsistence gatherers disease and the fertility of the child-bearers

These populations inhabit areas of very low density so low that it is almostinconceivable to the modern urban dweller The idea of walking for two weeksand never seeing another individual is true solitude If you did see someone itwould be a member of your immediate household or local band However eventhese small populations are not completely isolated Occasionally one of thesepopulations meets another (Fig 61) When this happens a complex set ofinteractions takes place There may be immediate withdrawal competition forresources warfare or trade and exchange This study will be concerned with amodel which addresses all but the first alternative Each alternative is a type ofinteraction and thus I call my model a model of interactive growth

Figure 61 The modelrsquos scenariopopulations of modern humansand Neander thals pursuingforag ing strategies withinoverlapping ter r itor ies KeyHSS=Homo sapiens sapiensHSN=Homo sapiens neanderthalensis

My simulation model has several features First there are four major groupsof parameters Each is an input entered prior to running the simulation Theyare the initial sizes of the populations the initial growth rates the competitionor replacement rates and the probability that the two populations come intocontact (Fig 62) The model positions both populations according to theirrespective initial sizes and growth rates These values will change interactivelyas each population grows and declines Second the growth functions may beapplied to as many populations as the simulator is interested in studying In mycase I will limit this study to two populations Each will be considered as anexample of how the growth of many small populations might take place Thislimit of two populations creates a highly simplified world which brings outthe similarities and the differences in the populations Third the model allowsthe dependency of the two populations to vary Within the confines of themodel it is possible for the two populations to be totally independent of eachother On the other hand it is also possible that one population is dependentupon the resources of the other to whatever degree or that both are dependentupon the resources of each other Finally the model allows one to census thetwo populations at any time

The growth functions of the populations are standard growth equationswhich operate on the entire population The model is modular and it is possibleto use age and sex-specific growth rates as well as stable population equationsHowever these are separate topics and are discussed in detail elsewhere(Zubrow 1989) In addition to the size and growth rates of the two populationsthe initial inputs included the probability that members of one population willmeet members of the other population It is assumed that the interactionbetween the two populations is direct By this I mean that the members of thetwo populations meet or they are in direct competition If not there is nointeraction and the populations grow independently For competition to occurthe populations do not actually have to meet They may compete serially for thesame resources or require the same land or water For example one populationmay enter an area harvest the game then leave When the second population

Figure 62 The initial parameters forthe interactive growth model

THE MODEL

arrives in the same area their harvest has been diminished What is not allowedfor is indirect competition One population may reduce the resources in theirimmediate area this in turn may lower the resources of an adjacent area Thisreduction affects the size of the second population In short competition whichoperates through adjacency is not modelled

The initial rate of replacement is another parameter What happens when themembers of the two populations meet is modelled Possibilities include thecomplete or partial replacement of the members of one population in theresource area by members of the second population Alternatively thepopulations may meet compete and remain in a position of status quo The sizeand rate of the replacement function allows one to simulate the full range ofreplacement

Results from the model

As of the time of writing I have simulated more than 300 variations of theseparameters A good indicator of the demographic viability of a population is thenumber of generations to extinction If a population does not become extinct inthe first 100 generations I consider it successful In these first runs I set themaximum limit for the number of generations to be simulated at no more than200 I also set the size of the initial populations as very small usually betweentwo and 200 In almost 60 per cent of the cases simulated one of the twopopulations survived for more than 100 generations In only three simulationswere the Neanderthal populations able to survive for over 100 generations Incomparison in over 20 simulations the H sapiens sapiens populations were ableto survive over 100 generations

I have rerun the first 50 simulations and added 250 more variations raisingthe total number of variations to over 300 Additionally I have increased thenumber of generations to 500 I have also expanded the range of initialpopulations growth rates replacement rates and meeting rates The results bothconfirm and elaborate the original conclusions and so I will emphasize thesenew results The simulations were run varying one parameter and holding theothers constant Each initial population was allowed to range from one to 6400individuals The population growth rates were allowed to vary far beyond realityThey could and did take on any value from 0000 to 0050 To interpret thesevalues so that they are not just sterile figures one should remember that 0020would be a 2 per cent annual increase If this rate was applied constantly thepopulation would double in 35 years increase four times in 70 and be slightlygreater than eight times the original population in slightly more than a centuryThe meeting rate varied also from 0001 to 0500 This means that of all possibleoccasions when interactions could occur the populations actually met andinteracted from one out of 1000 times to every other time The replacementrates were also run between 0001 and 05 At 0001 in every 100 interactions areplacement took place Similarly 0500 means in one out of every twointeractions a replacement took place A member of one population replaced a

member of the other population This replacement occurs in the context ofcompetition within the localized resource system

The window for successful Neanderthal survival is very small As we will seein the following discussion Neanderthal extinction almost always occursbetween 100 and 250 generations that is between 2500 years and 7500 years Itcan be as short as 30 generations and as long as 350 generations

However before disclosing all of the conclusions I wish to discuss the resultsin a systematic manner What I propose to do is to examine the results ofvarying one parameter at a time Then I will discuss the simulations which useparameters based upon ethnographic analogy and epipalaeolithic values

In the following figures the parameters for the simulations were set tostandard settings Then each parameter was varied while the others were heldconstant These settings were initial population of H sapiens sapiens equals teninitial population of H sapiens neanderthalensis equals 100 initial growth rate ofH sapiens equals 0010 initial growth rate of Neanderthals equals 0010 initialmeeting rate equals 0010 and initial replacement rate equals 0010

Figure 63 Varying theinitial population sizesof modern sapiens Theparameters are set atinitial Neanderthalpopulation equals 100initial modern sapiensequals 20 30 40 50100 400 initialNeanderthal growthrate equals 001 initialmodern sapiens growthrate equals 001 themeeting rate equals001 and thereplacement rate equals001

RESULTS

Figure 63 shows the growth of the interacting Neanderthal and modern Hsapiens populations when one increases the initial size of the contacting

population of modern H sapiens The upper graph shows the growth ofNeanderthals the lower graph shows the growth of modern H sapiens Eachfamily of curves represents the change resulting from varying the initial H sapienspopulation from 20 to 30 40 50 100 and 400 There are several obviousdescriptive generalities which should be noted In all cases the Neanderthalsbecome extinct in less than 200 generations while the H sapiens sapiens grow ina more or less logarithmic function The rapidity of H sapiens growth is directlyrelated to the size of the initial H sapiens contacting population The swiftness ofNeanderthal extinction is inversely related to the size of the initial H sapienspopulation There is a threshold between 100 and 150 for the initial H sapienscontacting population If this population is above the threshold the Neanderthalpopulation simply decreases and becomes extinct This occurs between 70 and150 generations If on the other hand the number of contacting H sapiens is lessthan this threshold both populations grow for a period after contact and it isonly later that the Neanderthal populations begin to decline as the growth ofthe modern H sapiens overtakes them I call this the lsquocontact thresholdrsquo

Figure 64 Varying theinitial population sizes ofNeanderthals Theparameters are set atinitial Neanderthalpopulation equals 100200 400 800 and 1600initial modern sapiensequals 10 initialNeanderthal growth rateequals 001 initialmodern sapiens growthrate equals 001 themeeting rate equals 001and the replacement rateequals 001

Conversely Figure 64 shows the growth of the interacting populationswhen I vary the size of the initial Neanderthal population from 100 to 1600The larger the initial Neanderthal population the shorter the time toextinction Thus the Neanderthal population of 1600 becomes extinct in 50generations while the Neanderthal population of 100 becomes extinct in about200 generations The reason this occurs is that the larger Neanderthalpopulation creates the potential for a much larger number of contacts Thus onecould suggest that at time of contact it actually would have been maladaptive ifthe Neanderthals were in larger groups The modern H sapiens populationgrows again more or less logarithmically and with a rate which is directlyrelated to the size of the initial Neanderthal population Changing the size ofthe initial H sapiens sapiens population actually causes somewhat less growthmarginally than does changing the size of the Neanderthal population Forexample a change from 100 to 400 initial H sapiens results in a change fromapproximately 6500 to 16 000 H sapiens at the 350th generation A 400 per cent

Figure 65 Varyingthe initial growth rateof modern sapiens Theparameters are set atinitial Neanderthalpopulation equals 100initial modern sapiensequals 10 initialNeanderthal growthrate equals 001 initialmodern sapiensgrowth rate equals0001 0002 00030005 0020 0050the meeting rateequals 001 and thereplacement rateequals 001

RESULTS

increase in the initial population results in a 250 per cent increase by the finalgeneration On the other hand changing the initial Neanderthal populationfrom 100 to 400 results in an increase in H sapiens at the 350th generation from3500 to 13 000 or a resultant 370 per cent increase

Figure 65 depicts the interaction when I vary the growth rate of thecontacting modern H sapiens The growth rate ranges from 0001 to 0050 inthis graph As one expects increasing the growth rate of the modern H sapiensis inversely related to the rapidity of Neanderthal extinction as well as directlyrelated to modern H sapiens growth Neanderthal extinction may be as rapid as100 generations and as slow as 300 After the initial contact both populationsgrow For the Neanderthals they continue to grow for approximately half oftheir postcontact existence Thus when the contacting population growsrapidly at 0050 the growth period of the Neanderthals is 50 generations Thedecline from approximately 120 individuals to extinction takes approximatelythe same number of generations At lower growth rates the Neanderthalpopulation grows for longer periods and declines for a larger number ofgenerations There is an important threshold in modern H sapiens growth It

Figure 66 Varying theinitial growth rate ofNeanderthals Theparameters are set at initialNeanderthal populationequals 100 initial modernsapiens equals 10 initialNeanderthal growth rateequals 0001 0005 00200040 initial modernsapiens growth rate equals001 the meeting rateequals 001 and thereplacement rate equals001

occurs at about 0010 If one examines the lower graph the curves from 0001to 0005 show a logistic form of growth By this I mean the growth is relativelyslow for the first 100 generations becomes more rapid for the second 100generations and slows down again for the third 100 generations Thiscorresponds to a model which would suggest a period of successful adaptationthen rapid adaptive radiation and finally another successful adaptation At thehigher growth rates of 0020 or 0050 the modern H sapiens just take off in analmost logarithmic growth pattern I call this threshold the sapiens growththreshold

Figure 66 is the corresponding variation of the Neanderthal growth ratesThe growth rates vary from 0001 to 0040 in this illustration The Neanderthalsfollow the same patterns as we have noted before Extinction takes placebetween between 100 and 250 generations or less than 10 000 yearsNeanderthal growth rates are inversely correlated to Neanderthal survival Themaladaption of rapid growth is clear A growth rate of 0040 results in apopulation of almost 1000 in 70 generations Extinction however occurs 40generations later The results are only slightly less dramatic with rates of 0020Once more there is a threshold Its character is only sketched in this graphHowever if the growth of the Neanderthals is less than 0005 they do not growafter contact Contact by the modern H sapiens populations simply rings thedeath knell of the Neanderthals It is however a long concert taking more than150 generations The growth rate of the Neanderthals is directly related to thegrowth of the H sapiens The greater the Neanderthal growth the greater theresultant H sapiens population The growth is relatively slow It is not until morethan 50 generations have occurred that one can begin to pick out significantdifferences in the numbers of modern H sapiens This is partly a result of thescale but not entirely Previously at similar scales differences in initialpopulations and growth rates could be determined One should also note inpassing that if the Neanderthal growth rate is high enough one has a logisticcurve in the growth of modern H sapiens After the second plateau or moreaccurately the quasi-plateau caused by a decreasing growth rate the growthpicks up significantly and then continues to grow logarithmically

Briefly an increase in Neanderthal growth rates from 0001 through 0005 to0020 results in an eightfold increase in modern H sapiens while the sameincrease in H sapiens results in first a doubling and then an additional fivefoldincrease

Figures 67 and 68 are very similar Each represents the changes caused bydecreasing their respective parameters that is the meeting rate and replacementrate respectively The demise of the Neanderthals takes place in approximately250 generations in each of these cases As the meeting rate and the replacementrate increase the time to extinction becomes shorter For the meeting rate thereis a threshold between 0003 and 0005 If the value is less than 0003 theNeanderthal population grows before becoming extinct If more then thepopulation rapidly becomes extinct without any growth The threshold for thereplacement rate is approximately 0010 If the competition rate is greater thanthis value extinction takes place without any preliminary increase in the

RESULTS

Neanderthal population Although structurally similar in that the shapes of theresultant graphs are the same there is a quantitative difference between themeeting and the replacement rates It takes a smaller change in the meeting ratethan the replacement rate to create the same decrease in the time for extinction

By now the reader must be crying lsquoenoughrsquo We have a good idea of how theparameters cause changes in the prehistoric populations However there is alimit to speculation without relating it to what is actually known about specificethnographic and prehistoric populations In order to replace the readerrsquos feetfirmly on the terra firma of anthropological reality I ran a series of simulationsusing ethnographic and prehistoric rates There are of course a considerablenumber of ethnographic examples which could be used There are also a greatnumber of assumptions and stretches of imagination that are necessary to usesuch data For this chapter I will report on only one set of three cases I culledthe following data from Kung bushmen ethnographies and demographicstudies (Howell 1979 Lee 1972a 1972b) I set the number of members in thecontacting population of H sapiens sapiens as 20 50 and 500 This correspondsto the range of the ethnographic extended household of 20ndash50 and to the full

Figure 67 Varying themeeting rate for thepopulations Theparameters are set atinitial Neanderthalpopulation equals 100initial modern sapiensequals 10 initialNeanderthal growthrate equals 001 initialmodern sapiens growthrate equals 001 themeeting rate equals0003 0005 0020 andthe replacement rateequals 001

band size of approximately 500 I set the Neanderthal population to 500 or theband size The growth rates for both populations were set to the ethnographicvalues of 00026 The meeting and replacement rates were set at 0010

Figure 69 illustrates the resulting population curves for these ethnographically grounded populations In all three cases extinction occurs prior to 150generations Indeed when one population meets another coming down a foragingpathmdashthat is band meets band rather than household meets bandmdashthe extinctiontakes place remarkably quickly in only 50 generations The growth curves forthe replacing population the modern H sapiens sapiens have no surprises for usIn all three cases there is rapid growth for the first 30 to 50 generations then thegrowth continues but it is a slower indeed almost constant rate

If one tries to estimate real prehistoric population growth rates one isentering a very difficult and speculative area There are not a lot of data to relyupon and what there are have been beset by problems These problems are notcreated by the analyst Rather they are the result of limited samples poorpreservation and the difficulty of the task Acsadi amp Nemeskeri (1970) havepresented one series of data They are not the only ones nor are they necessarilythe best but they are well known and appear to be reasonable If one takes theepipalaeolithic the best of their earlier sequences the parameters are set so thatthe initial populations are both 185 (which was the size of their skeletal

Figure 68 Varying thereplacement rate for thepopulations Theparameters are set atinitial Neanderthalpopulation equals 100initial modern sapiensequals 10 initialNeanderthal growthrate equals 001 initialmodern sapiens growthrate equals 0001 00020003 0005 00200050 the initialmeeting rate equals001 and the initialreplacement rate equals0005 0020 00400080

RESULTS

populations) The growth rates for both populations are 000013 The meetingand replacement rates are 001 These latter rates were chosen because they werereasonable and convenient They were not based upon specific anthropologicaldata Figure 610 shows these results They conform with the general pattern wehave seen in the previous ethnographic cases Extinction takes place for theNeanderthal population in the first 150 generations Growth for the modern Hsapiens is rapid and then slows down to a linear form

If you review all of the above there are several generalities worthemphasizing First no matter which parameter one varies or relaxes theNeanderthal population goes extinct Usually the time to extinction is in theneighbourhood of 150 generations Second the modern H sapienspopulations are particularly hardy and the question is far more frequently howrapid is their growth rather than how long to time of extinction Third thereare two demographic regimes which can be separated In one of these regimesthe Neanderthal populations continue to grow after initial contact for severalgenerations Then it appears that the processes of competition replacementand the increasing numbers of modern forms overwhelm the Neanderthalgrowth In the other regime there is simply decline and extinction aftercontact It may be slow and then occurring at an increasing rate or it may be

Figure 69 Simulationsbased on ethnographicanalogy The parametersare set at initialNeanderthal populationequals 500 initialmodern sapienspopulation equals 20 50500 initial Neanderthalsand modern sapiensgrowth rates equal00026 initial meetingand replacement rateequals 0010

95REFERENCES

fast But the Neanderthals never seem to be able to maintain any growth aftercontact These two regimes are separated by a series of thresholds or thresholdvalues for the parameters One could say that all other things being equal if thecontacting modern H sapiens are a population greater than 150 or have a growthrate greater than 0010 or if the competition rate is above 0003 or if themeeting rate is 0010 then the Neanderthal population enters this second regimeand simply declines

Conclusions

In this chapter I have briefly surveyed some of my ongoing research on simulatinginteracting prehistoric populations I have developed a simulation model basedon complex interactive growth It shows that under many different demographicand interactive variations Neanderthal survival was impossible The demographicwindow which could have made it possible was quite improbablemdashit requiresunreasonably low sizes and growth rates for the populations of H sapiens sapiensEven this was insufficient for Neanderthal survival Survival would haveadditionally required very low replacement and interaction rates In generalNeanderthal continuation was more prolonged in competitive situations whereboth populations were small It would appear that one advantage the modernforms had was ability to reach more rapidly a form of lsquostablersquo growth Needlessto say far more may be accomplished with these models as they become moresophisticated and more simulation runs are completed

References

Acsadi Gy amp JNemeskeri 1970 History of human life span and mortality BudapestAkamemiai Kiado

Howell N 1979 The demography of the Dobe Kung New York Academic Press

Figure 610 Simulationsbased on epipalaeolithicrates The parameters are setat initial Neanderthal andmodern sapiens populationsequal 185 initialNeanderthal and modernsapiens growth rates equal000013 and the initialmeeting and replacementrate equals 001

Lee RB 1972a Population growth and the beginnings of sedentary life among theKung bushmen In Population growth anthropological implications BSpooner (ed)329ndash42 Cambridge Ma MIT Press

Lee RB 1972b Kung spatial organization An ecological and historical perspectiveHuman Ecology 1 125ndash47

Zubrow EBW 1989 The demographic modelling of Neanderthal extinction In Thehuman revolution behavioural and biological perspectives in the origins of modern humansPAMellars amp CStringer (eds) 212ndash31 Edinburgh Edinburgh University Press

7 Aboriginal fossil hominidsevolution and migrationsPHILLIP JHABGOOD

The earliest evidence for human occupation of Sahul the combined landmass ofAustralia New Guinea and Tasmania is at least 40 000 BP (Groube et al 1986Jones 1989 Nanson et al 1987 Pearce amp Barbetti 1981 White amp Habgood1985 White amp OrsquoConnell 1982) The sites in question (Huon Upper SwanLake Mungo Keilor and Cranebrook Terrace) are located in the northeasternsouthwestern and southeastern parts of the continent

Figure 71 Map of Sahul showing maximum low sea level and major latePleistocene sites

98 ABORIGINAL FOSSIL HOMINIDS

(Fig 71) So as to allow for the movement from the most probable entry pointinto the continentmdashthe northwest (Birdsell 1977)mdashto these dispersed sites themost widely quoted date for the initial migration(s) to Sahul is around 52 000BP when there was a major glacio-eustatic lowering of sea levels (Chappell1976 1982) However the initial entry into Sahul could have been significantlyearlier (Jones 1989)

A major debate still rages as to who these colonists were Three majorexplanations have been postulated Two are based on the premise of a number ofmigrations by morphologically different groups which subsequently interbredwhile the other contends that the Australian Aborigines migrated from a singlebiological homeland (see Kirk amp Thorne 1976 for references)

Birdsellrsquos (1979) trihybrid theory based on studies of contemporaryAborigines postulates three waves of colonists into Australia The first wavecomprised the Oceanic Negritos whose remnants he saw in Tasmania and therainforests of northeastern Australia No geographic homeland for the OceanicNegritos has been specified by Birdsell but he does see them as being present inother areas such as the highlands of New Guinea the Andaman Islands parts ofthe Malay Peninsula and on some of the Philippine islands (Birdsell 1977) Thesecond wave comprised the Murrayians who are linked with the Ainu and whodisplaced the Oceanic Negritos from most of the continent Birdsellrsquos final andmost recent wave of colonists comprised the Carpentarians who possibly camefrom India The morphologically variable Australian Aborigines were seen byBirdsell as a hybrid of all three groups

This trihybrid explanation has some major problems Studies of prehistoricTasmanian crania (Pardoe 1984 Thorne 1971b) suggest that they are lsquohellipvariants of a southern Australian population based on a morphology existingabout the time of Tasmaniarsquos connection and subsequent separation from themainlandrsquo (Thorne 1971b p 319) A study of Queensland crania failed todistinguish between those from the northwestern rainforests and those from therest of Queensland (Larnach amp Macintosh 1970) Also the Carpentarians seemto be the result of recent contact between Aborigines along the northernAustralian coast and Macassan trepang fishermen from Indonesia and Papuantraders (Larnach amp Macintosh 1970 Thorne 1971b) An Ainu link for theMurrayians has also been challenged (Yamaguchi 1967)

A dual-source explanation based on the differentiation of the latePleistocene and earlier Holocene Australian skeletal material into two distinctmorphological types one lsquorobustrsquo and the other lsquogracilersquo has been proposed byThorne (1971a 1976 1977 Thorne amp Wilson 1977 see also Freedman ampLofgren 1979) The lsquorobustrsquo type as typified by crania from Kow SwampCohuna Coobool Creek Talgai Mossgiel and Cossack is relatively low andrugged with flat receding frontal bones marked postorbital constriction largesupraorbital tori and occipital tori moderate gabling of the thick cranial vaultbroad prognathic faces and large palates mandibles and teeth The lsquogracilersquotype as exemplified by material from Lake Mungo Lake Tanou Lake Nitchieand Keilor has high rounded and in general more modern-looking craniawith thin vault bones expanded frontal and temporal squama

slight brow-ridge development lightly constructed nonprognathic facial regionsand relatively small palates mandibles and teeth At present the earliest date forthe lsquorobustrsquo type is approximately 14 000 BP at Kow Swamp whereas thelsquogracilersquo Lake Mungo 3 skeleton has a date of at least 30 000 BP Thorne(1977) explained these two types as being the result of two morphologicallydistinct and chronologically separated groups entering Australia The lsquorobustrsquotype which display lsquothe mark of ancient Javarsquo came from Indonesia while thelsquogracilersquo type which have lsquothe stamp of ancient Chinarsquo came from East Asia(Thorne 1977 see also Freedman amp Lofgren 1979) Interbreeding betweenthese two groups is thought to have eventually led to the modern AustralianAboriginal morphology

As with Birdsellrsquos trihybrid explanation Thornersquos dual-source hypothesishas some major problems Recent multivar iate analyses of cranialmeasurements have shown that the late Pleistocene and early HoloceneAustralian crania are more similar to each other than they are to either East orSoutheast Asian crania (Habgood 1985 1986b) Also placing of individualcrania into one or other of the two morphological extremes is more difficultthan Thornersquos hypothesis would suggest Thorne places the Lake Nitchiecranium in his lsquorobustrsquo group (Thorne 1977) based solely one assumes on itslarge size yet examination of the cranial measurements and midsagittalcontour suggests a strong similarity with the Keilor cranium one of Thornersquoslsquogracilersquo types (Fig 72 Freedman amp Lofgren 1979) Most workers wouldplace Lake Nitchie in Thornersquos lsquogracilersquo group (Macintosh 1971 Howells1973 Freedman 1985 Freedman amp Lofgren 1979 Habgood 1985 1986b)What the individual crania display is a range of morphological forms (Fig72) not two extremes as Thorne suggests This range is well illustrated at thetwo sites with large samples Kow Swamp and Coobool Creek which

Figure 72 Midsagittal cranial contours of Australian Aboriginal crania orientated onthe Frankfurt plane (after Freedman amp Lofgren 1979) Key A=Cohana B=Keilor

C=Lake Nitchie D=Kow Swamp 1

ABORIGINAL FOSSIL HOMINIDS

lsquoinclude individuals at the opposite ends of the morphological spectrum andindeed most of the specimens at these sites represent the intermediatesbetween these extremesrsquo (Wolpoff et al 1984 p 445) For example Wolpoff(1980) identified a similar ity in the development of the frontal andsupraorbital regions between Lake Mungo 3 and some of the more gracileKow Swamp males such as KS 14 and KS 15 and thought the Lake Mungo 1cranium resembled the more complete female crania from Kow Swamp suchas KS 4 and KS 16 Webb (1989) however does not feel that the degree ofgracility of Willandra Lakes hominids such as Lake Mungo 1 (his WLH 1) ismatched by females from Kow Swamp

Finally some of the variation within the late Pleistocene and earlyHolocene cranial material appears to be due to cranial deformation Brown(1981) has demonstrated that features including a flat frontal bone and aprebregmatic eminence which are said to be typical of the lsquorobustrsquo type(Thorne 1976) are likely to be the result of deformation caused by cranialpressing This form of cranial deformation would allow a great deal ofvariation in the amount of deformation and associated effects on the crania(Brown 1981) What has been produced is a gradation from skulls such asKow Swamp 5 and 7 Cohuna and Coobool Creek 1 49 and 65 whichdisplay marked deformation to those such as Kow Swamp 1 8 and 9 andMossgiel which display little or no deformation

After a recent study of the hominid sample from the Willandra Lakes (theser ies numbers from WLH 1ndash135) Webb (1989) concluded that themorphological range was too great to be encompassed within a singlemorphological population Like Thorne he argued that there was a lsquorobustrsquogroup and a lsquogracilersquo group that were the result of separate migrationsInstead of arguing for separate geographical homelands for the two types heproposed that they both came from the same area Indonesia but wereseparated by a considerable period of time during which gracilizationoccurred That is the lsquorobustrsquo type entered Australia first and was laterfollowed by the lsquogracilersquo type that had subsequently developed in Sunda Bythe late Pleistocene the Australian population reflected the wide range ofmorphological variation produced by the intermixing of the two types Hefound it difficult to decide whether the lsquogracilersquo type constituted a secondpopulation or just a link in the chain of human migrations to Australia andeven postulated that the lsquogracilersquo type may be an indication of theemergence and spread throughout parts of Sunda and Sahul of precursorpopulations that eventually gave rise to the smaller human phenotypes suchas the modern lsquonegritorsquo stocks

It is hard to argue against his propositions because of the fragmentarycondition of most of the Willandra Lakes hominid material and the lack ofchronological control for the sample (it may represent a very long period oftime) Also he does not explain in any detail the mechanisms involved in thegracilization of the Sunda population especially when one considers the robustnature of late Pleistoceneearly Holocene material from other regions such asNorth Africa (Anderson 1968 Greene amp Armelagos 1972) He argued that the

Willandra Lakes hominid sample was made up of two types (lsquogracilersquo andlsquorobustrsquo) but does not explain how the two types could live in the same regionat the same time without interbreeding and the range of variation decreasingFinally he does not explain why his widespread lsquonegritogracilersquo type wasgenetically swamped by the lsquorobustrsquo type when the two groups interbred toproduce the late Holocene and modern Australian Aborigines This point isespecially important if the two types coexisted in the Willandra Lakes over along period of time as he infers

Brown (1987 1989) has also argued that although variation is present at anindividual level there is a consistent Australian Pleistocene morphology not twoseparate morphologies

The third explanation the homogeneity hypothesis suggests that theAustralian Aborigines are the result of migration from a single biologicalhomeland Proponents of this explanation have studied both living Aborigines(Abbie 19631968) and skeletal material (Macintosh 1971 Howells 1973 1976Macintosh amp Larnach 1976 Habgood 1985 1986b Brown 1987 1989) andhave concluded that the data suggest a homogeneous founding population forAustralia It should be noted that both Abbie and Howells regarded theTasmanians as Melanesian and so different to the mainland Aborigines but aswe have seen prehistoric Tasmanian Aboriginal crania are mainland Australianin their affinities (Pardoe 1984 Thorne 1971b)

What we have in Australia is a corpus of late Pleistocene and early Holoceneskeletal material that displays a continuum of cranial forms across a large rangeof morphological variation (see Habgood 1985 1986a 1986b) As Macintosh ampLarnach stated the various fossil crania are lsquoequal representatives ranged towards(because they fall inside the extremes of the range) either end of a continuum ofa single populationrsquo (1976 p 114)

The homogeneity explanation proposes that the morphological variationevident in the late Pleistocene and early Holocene Australian skeletal materialwas caused by genetic processes and not due to subsequent migrations bymorphologically distinct groups from different geographical homelands andbiological sources This does not necessarily mean that more than onemigration from the original source area could not have occurred Theadditional colonists could have come from the same biological stock as thosethat preceded them or have come in sufficiently small numbers so as not toadd substantially to the genetic-morphological make-up of the continentalpopulation

At present this explanation fits the available morphological data better thaneither of the other two hypotheses It is also compatible with the archaeologicaldata in that the corpus of late Pleistocene Australian stone tools is so similareven when they are made from different materials or utilized in differentenvironments that they are grouped into the pan-continental Australian lsquocore-tool and scraper traditionrsquo Similarly the first distinct corpus of rock art inAustralia the Panaramitee style is found throughout the continent includingTasmania and forms a relatively homogeneous stylistic entity (Franklin 1990Maynard 1979 White amp Habgood 1985 White amp OrsquoConnell 1979 1982) The

late Pleistocene archaeological record therefore also suggests that the continentof Australia was colonized by a group (or groups) with a homogeneous culturalbackground and by inference from a single source area and not from amultitude of cultural sources

If we assume that Sahul was colonized by people from one source area as thecurrent evidence suggests where was this area The most likely region isIndonesia which incorporates the southern part of Sunda Land and theWallacean islands (Fig 71) Whether the colonists had lived in Sahul for a longperiod or had recently migrated into the area is open to much debate (seeHabgood in press Smith et al 1979 Stringer amp Andrews 1988 for discussions ofthis problem) At present the only human fossils from Indonesia that couldpossibly be the ancestors of the earliest inhabitants of Sahul are the Ngandonghominids which are generally classified as Homo erectus (Coon 1962 Santa Luca1980 Stringer 1984 Groves 1989 Habgood 1989 in press)

The date of the Ngandong hominid sample remains difficult to ascertain(see Bartstra et al 1988 Habgood in press) The hominids were recoveredfrom the upper or 20-m terrace of the Solo river near Ngandong central Java(Santa Luca 1980 Weidenreich 1951) They were not localized in anyparticular spot or within a single layer but were irregularly distributedthroughout the entire site and so while they can be regarded as a sample theydo not necessarily represent a single biological population This terrace of theSolo river which is referred to as the Notopuro formation containedabundant predominantly extant mammalian fauna (the Ngandong fauna)and so is usually considered to be of Upper Pleistocene age (Santa Luca 1980)This has meant that some scholars such as Coon (1962) have given theNgandong hominids an Upper Pleistocene date

Santa Luca (1980) however has provided taphonomic evidence to suggestthat the Ngandong fauna is a mixed assemblage with the nonhominid faunarepresenting a death-assemblage buried after minimal exposure while thehominid remains appear to have been redeposited into younger levels Thissuggestion is consistent with the arrangement of the hominid remains withinthe terrace The nonhominid fauna is also well preserved with complete andarticulated vertebral columns and crania together with associated mandibleswhereas the hominid material is fragmented and consists predominantly ofcalvaria The hominid sample is composed of only the most durable structureswith the calvaria displaying evidence of surface damage and lacking the facialskeleton which could be the result of rolling and transportation of the crania bywater (Boaz amp Behrensmeyer 1976) This pattern is consistent with thesuggestion that the hominid sample had been redeposited It would appear thatthe Ngandong hominid sample has had a complex depositional history whichwill make the dating of it very difficult A middle or late Middle Pleistocene agefor the Ngandong hominids would seem a reasonable estimate at present (butsee Bartstra et al 1988 for an earlier dating)

If the concept of regional morphological continuity within Australasia iscorrect an idea that is much debated (see Groves 1989 Stringer 1984 Stringeramp Andrews 1988 Stringer et al 1984 Brauer 1984 1989 Habgood 1986b

1989 in press) and there is an evolutionary line leading from the earlyIndonesian H erectus type through to the Australian Aborigines via theNgandong form (Weidenreich 1943 Coon 1962 Macintosh 1965 Thorne ampWolpoff 1981 Wolpoff et al 1984) then the earliest inhabitants of Sahul shouldbe advanced Ngandong type H sapiens (assuming a date around 52000 BP forthe initial occupation of the continent)

The discovery of Willandra Lakes hominid 50 (WLH 50) may support thisassumption (Flood 1983 Thorne 1984 Webb 1989) A calvaria some portionsof the facial skeleton and fragmentary postcranial material were found on thesurface near Lake Garnpung north of Lake Mungo This essentially unpublishedhominid is opalized with all the normal phosphate in the bone being replacedby silicates (Flood 1983) The large and robust calvaria is fully sapient in overallconfiguration and Australian Aborigine in affinity (Webb 1989) WLH 50 hasvery thick vault bones Over 70 per cent of this thickness is made up of diploeicbone (Delson 1985 Flood 1983 Thorne 1984 Webb 1989) The very angularcalvaria has a flat and receding frontal marked temporal crests a protrudingoccipital with a well-developed transverse torus marked angulation betweenthe occiput and the nuchal plane prominent brow-ridges (especially the medialsegment) and maximum parietal breadth located towards the parietal mastoidangle The fragmentary postcranial elements are also quite large and robust(Flood 1983)

The date of WLH 50 has not been conclusively established The skeletalmaterial was a surface find and so it is difficult to ascertain to which geologicallayer it should be equated Quoting unpublished papers written by AG ThorneFlood (1983 p 67) recorded that lsquoRadiocarbon and trace element analysisindicate a minimum age of 25 000 to 30 000 BP but the remains are probablymuch olderrsquo while Delson (1985 p 298) stated that lsquoUsing an experimentalelectron spin resonance approach [ESR] the oldest specimen WLH 50 is mucholder than 30 000 perhaps something like 60 000 years oldrsquo These dates arepreliminary estimates and have not been substantiated in print Caddie et al(1987) calculated an ESR date of 29 000 plusmn 5000 years for WLH 50 based onthe assumption that the natural dose rate at the site of WLH 50 was the same asthat for other sites studied in the area The crust of calcium carbonate aroundWLH 50 has provided a minimum radiocarbon date of 15 000 BP(AGThorne pers comm) Webb (1989) contends that WLH 50 along withthe rest of the Willandra Lakes hominid sample predates the end of lunetteformation at the Willandra Lakes around 15000 BP It is however unlikely thatWLH 50 is older than 45 000 BP as this is the estimated date for the bottom ofthe Mungo Unit and as yet no archaeological remains have been recoveredfrom the underlying Golgol Unit (Bowler 1976 White amp OrsquoConnell 1982)

Thorne (1984 p 227) declared that WLH 50 was lsquomuch more robust andarchaic than any Australian hominid found previouslyrsquo He and Wolpoff(1985) have proposed resemblances between early Indonesian crania andWLH 50 especially in the form of the frontal squama and the skull shapeviewed in norma occipitalis with Thorne (pers comm) suggesting that thetransverse arc of WLH 50 was close to the mean value for this measurement in

the Ngandong sample Elsewhere I have argued that the morphologicalfeatures that have been suggested to demonstrate a morphological linkbetween Australian crania and earlier Indonesian material are generallycharacteristic of H erectus and archaic H sapiens throughout the Old World(Habgood 1989 in press) However a combination of a number of fronto-facial features (a long and sagittally flat frontal bone with a posterior positionof minimum frontal breadth malars with everted lower margins andprominent zygomaxillary tuberosities and possibly very prognathic faces) didseem to indicate some degree of regional morphological continuity inAustralasia (Habgood 1989a in press) WLH 50 has this combination offeatures (Habgood 1989a) and so may be evidence for regional continuity inAustralasia especially if it is older than the Lake Mungo material which doesnot support morphological continuity with earlier Indonesian hominids(Habgood 1989a in press) WLH 50 may therefore resemble the originalcolonizers of Sahul

However one cannot discount the possibility of some pathologicalchanges to the vault of WLH 50 caused most probably by some form ofhaemoglobinopathy (Brown 1987 Webb 1989 GEKennedy pers comm)Although the heavy mineralization of the calvar ia interfered withradiographie visualization X-rays of WLH 50 reveal a somewhat lineal(hair-on-end) arrangement of the diploeic trabeculae around the bregmawhich is consistent with a diagnosis of severe anaemia Even if pathologicalmodifications did occur WLH 50 would have been a very robust craniumand would have presented the combination of morphological features listedabove

A continued biological link with or the continued influx of genes fromIndonesia is suggested by the morphological similarity between the undatedbut probably mid-Holocene Wajak 1 cranium and the 12 900-year-old craniumfrom Keilor in southeastern Australia (Coon 1962 Habgood 1985 1986bWeidenreich 1943 Wolpoff et al 1984) It is also probable that the dingo asemidomesticated dog which appears in Australia about 4000 BP was broughtinto the continent by people

We can assume that the initial population of Sahul was relatively small Atthis time the north west coast of the continent would have been coolerhaving an average mean annual temperature drop of up to 5degC and drierwith a decrease in rainfall of possibly 30 per cent as well as a change in theseasonality of the rainfall and a diminished effect from tropical cyclones(Bowler et al 1976 Deacon 1985 Webster amp Streten 1978 Chappell ampGrindrod 1983) It was most probably covered by open woodland andsavanna associated with generally low nutrient soils on laterites while thecoasts may have offered little permanence of tenure due to an unstableenvironment caused by frequent fluctuations in sea levels on the gentlysloping continental shelf (Chappell amp Thorn 1977 Chappell amp Grindrod1983 Deacon 1985 Hope amp Hope 1976) These environmental conditionscould have reduced carrying capacities and kept extinction rates high andpopulation numbers low (Deacon 1985 Hay den 1972) Population growth

might have been slow due to a low survivorship while inhabitants adaptedto the new environment and its resources and because of physiologicalcontrols such as infanticide prolonged lactation abortion and abstinence(Hayden 1972)

To be successful a colonizing group needed to have the genetic capacity tocontinue the population McArthur (1976 Me Arthur et al 1976) conducted anumber of simulation models of the chances of long-term survivorship of smallfounding populations It was found that the larger the initial group the greaterits chance of survival and that the populations founded by younger adults hadbetter chances of survival Groups that did not practise monogamy would alsohave had a substantially better chance of survival Differential extinction ofpopulations that may have survived for some time would have helped keeppopulation numbers low

A low continental population during the late Pleistocene fits with thedearth of sites that are dated to this period even allowing for poorarchaeological visibility and a high degree of site destruction (Jones 1989White amp Habgood 1985 White amp OrsquoConnell 1982) The early sites are alsonot rich in archaeological remains which suggests a very dispersed low-density population (Deacon 1985) Some areas such as Tasmania and thesoutheastern and southwestern corners of Australia appear to have beenrefug ia dur ing this per iod Substantial population growth is notarchaeologically visible in Australia until the mid-Holocene (Beaton 1983Lourandos 1983 Veth 1989)

As groups colonized the Sahul continent they moved into a diversity ofterrains and climates By at least 20000 BP most parts of Sahul showevidence of habitation or resource exploitation for example KoonaldaCave Lindner and Puritjarra in arid regions Kutikina Cave BeginnerrsquosLuck Cave and Kosipe in cold upland regions Matenkupkum and WallenWallen Creek in coastal environments and Lake Mungo Keilor and UpperSwan in riverine lake environments (Jones 1989 Veth 1989) If populationnumbers were relatively small as I have suggested it is probable that thesecolonizing groups would have become isolated from each other due to theenormous size of the continent (see White amp Habgood 1985 White ampOrsquoConnell 1982 for an evaluation of the various theories regarding thecolonization of Australia) There are few significant topographical barriersin Australia except the expanse of ocean that now separates Tasmania fromthe mainland and the three major sand-ridge deserts of central Australia(Veth 1989) The whole of the arid core of Australia may also have acted asa temporary barrier to human occupation (and possibly movement) duringthe last glacial maximum (Veth 1989)

If small groups were isolated one could expect to find unique localartefactual specializations which could be adaptations to local conditions Thetula adze found in the arid regions of northern Australia ground stone hatchetsfrom sites in Arnhem Land northern Australia and small tools recovered fromearly sites in southwestern Western Australia (White amp OrsquoConnell 1982) may beexamples of this form of local adaptation Also within the Panaramitee style

there are differing regional emphases on motif types although the overall rangeis similar and the technique of rendering the motifs varies from peckingabrading and pounding through to painting (Franklin 1990) These differencesmay be the result of and evidence for isolation during the late Pleistocene

Geographical isolation would have been accentuated by the last glacialmaximum centred around 18000 BP when one can expect populations tohave decreased Deacon amp Thackeray (1984) proposed a model for southernAfrica which suggested marked depopulation as a consequence of loweredusable productivity of the environment due to climatic changes at this timeThey assumed changes in both population distribution and density withlocal population extinctions Gamble (1983) has documented a similardepopulation of large parts of central Europe at various times during theUpper Pleistocene Based on a biogeographic model that divides Australiaup into refuges corridors and barriers Veth (1989) identified a lack ofevidence for the occupation of large tracts of Australia (the corridors andbarriers) during the glacial maximum Also Hiscock (1984) hypothesizesthat increases in discard rates at Colless Creek Cave between 13 500 BP and17 000 BP a period of increased aridity in northwest Queensland couldhave been caused by a reduction in the territory the occupants utilized anda more intensive utilization of the resources of the well-watered Lawn HillGorge complex Smith (1989) has argued for a general model of reducedforaging territory during the glacial maximum for the arid region ofAustralia

Geographical isolation is a reversible phenomenon which in itself does notaffect the separated groups (gene pools) but allows other processes toaccumulate genetic differences It is probable that small colonizing groupswould not be a representative cross-section of the parent population and sodue to founder effect would be genetically different This effect can beaccentuated if the group is composed of members of the same family lineage(lineal effect) Utilization of vastly different environments would meandifferent kinds of selection would have been acting on the small populationsThis may have caused genetic frequency variation between groups becausehuman biological variation is determined by the interaction between theenvironment and genetic systems Random genetic changes or mutationswhich are the source of new genetic variation within a gene pool stand amuch greater chance of becoming fixed in small populations and causing arandom genetic differentiation between the isolated colonizing groupsGenetic drift or chance fluctuations in gene frequencies may also causegenetic differentiation For example in isolated and possibly polygynousgroups a few males may father the majority of the children having a profoundeffect on the gene fixation of each generation Lourandos (1983) hassuggested that the late Pleistocene may have been typified by restrictedmarriage systems which would have accentuated isolation by furtherrestricting gene flow

Under this model what we would have had in the late Pleistocene inAustralia is small isolated groups scattered throughout the continent

developing genetic variation Genetic variation does not necessarily mean ahigh degree of morphological variation but we can assume a certaincorrelation between the two Each isolated group therefore could havedeveloped unique genetic and morphological combinations which becamefixed These groups could have been relatively homogeneous internally butwould have differed substantially from other such groups An example of thismay be the differences between the approximately 12 000- to 13 000-year-old Keilor and Talgai crania During the late Pleistocene and earlier Holocenethe continental population of Greater Australia would under the abovescenario have been genetically and morphologically heterogeneous whilebeing made up of many small and relatively isolated morphologicallyhomogeneous groups However as Brown (1982 1989) has demonstratedalthough there is variation there was a consistent Australian Pleistocenemorphology

Genetic differentiation and the development of unique geneticcombinations could have been restrained by gene flow but isolation would havebeen limiting its potential effect It is significant that in areas such assoutheastern Australia where isolation would have been greatly reduced andgene flow high the sites of Kow Swamp and Coobool Creek have skeletalsamples that display a large range of morphological variation and not fixedunique morphologies (Brown 1982 1987 1989 Thorne 1976 Wolpoff 1980Lourandos 1983 takes a different position see discussion in Pardoe 1988) Thismay also have been the case throughout the Willandra Lakes system (Webb1989) During the late Holocene and at the time of European contactsoutheastern Australia especially along the Murray river corridor is suggestedto have had a higher population density than many other regions of Australia(Butlin 1983 Webb 1984) This situation may also have typified earlier periodsif the appearance of burial grounds can be taken to imply population increaseandor higher population densities (Pardoe 1988) The snow-fed Murray-Murrumbidgee river system may also have experienced less extreme conditionsduring the last glacial maximum thus preventing major depopulation of theregion during this period

Genetic variation caused by isolation and small population numbers alongwith cranial deformation (Brown 1981) can account for the morphologicalpattern evident in the late Pleistocene and earlier Holocene Australian cranialmaterial

After the last glacial maximum population numbers would have slowlyincreased In its simplest terms this population increase (repopulation may bemore accurate) would mean more people or groups inhabiting the landscapeIsolation of groups would therefore have gradually lessened This increase isespecially evident (archaeologically visible) during the Holocene (Beaton 1983Lourandos 1983 Veth 1989)

As population numbers and densities increased gene flow would have beenhigher introducing new genetic material andor changing gene frequencies inthe previously isolated groups The trend would have been for graduallyincreasing differentiation within the individual groups which previously had

unique morphologies until all groups displayed similar types and ranges ofvariation Unique morphologies would have disappeared and the chance ofnew ones becoming fixed would have been greatly lessened in possibly larger-sized groups that were no longer isolated

The genetic and morphological variation within the smaller groups wouldhave increased whereas the continent-wide range of variation may havedecreased in that there would have been a reduction in the occurrence ofgroups displaying unique morphologies In this way the overall continentalrange of genetic and morphological variation may have been reduced fromthat of the earlier period whereas the range of variation of individual groupscould have increased with the introduction of new variation due to increasedgene flow Lourandos (1983) contends that during the Holocene increasinglywidening marriage systems developed further increasing the gene flowUnder this scenario the continental population which would have seemedmore homogeneous would have been made up of morphologically morevariable groups Areas that may have remained isolated longer than the moreoptimal regions could still have produced unique morphologies well into theHolocene An example of this could be the Cossack cranium from WesternAustralia

There may however have been a change in the pattern described aboveprior to European contact Groups living in resource-r ich areas thatsupported high population densities appear to have had rigid territorialboundaries and short marriage distances therefore restricting gene flowwhereas in arid regions with low population densities and unpredictableresources groups maintained more fluid boundaries and extensive socialnetworks which would have promoted gene flow (Pardoe 1988 Peterson1976 1986 White 1979) These patterns most probably came into existencewith repopulation after the last glacial maximum That is when groups movedback into and permanently occupied arid regions they took with themextensive social networks and as population densities reached critical levels inregions such as the Murray river corridor (Webb 1984) more rigid territorialboundaries became necessary so as to maintain control of resources Thischange may have influenced the morphological range of more recentAboriginal crania but would not affect the pattern of the late Pleistocene andearlier Holocene material

From the preceding discussion one can see that to account for the largerange of morphological variation of the late Pleistocene and earlier HoloceneAustralian Aboriginal crania it is unnecessary to resort to explanationsinvolving independent migrations by different groups who remainedbiologically separated for over 30000 years before interbreeding to producethe modern Australian Aborigines A more pedestrian and parsimoniousexplanation proposes migrations by small groups from a single source(geographic and biological) which due to isolation during continentalcolonization and demographic variation such as marked depopulation duringthe glacial maximum were acted upon by genetic processes includingfounder effect selection mutation genetic drift and varying amounts of gene

109REFERENCES

flow causing the development of a large range of morphological and geneticvariation Cranial deformation was also a contributing factor During the latePleistocene there would have been small and relatively homogeneous groupsmaking up a heterogeneous continental population With population increaseduring the mid-Holocene isolation would have decreased while gene flowincreased causing a reduction in genetic and morphological variation Therewould therefore have been possibly larger and more heterogeneous groupsmaking up a relatively homogeneous continental population during the mid-Holocene

Acknowledgements

Financial support for the research upon which this chapter is based and to attendthe 1986 World Archaeological Congress in Southampton where an earlier draftwas presented was provided by the Carlyle Greenwell Research Fund theUniversity of Sydney and the Australian Institute of Aboriginal Studies I wouldalso like to thank Natalie Franklin for typing the final draft of the chapter

References

Abbie AA 1963 Physical characteristics of Australian Aborigines In Australian aboriginalstudies HShiels (ed) 89ndash107 Oxford Oxford University Press

Abbie AA 1968 The homogeneity of Australian Aborigines Archaeology and PhysicalAnthropology in Oceania 3 221ndash31

Anderson JE 1968 Later Paleolithic skeletal remains from Nubia In The prehistory ofNubia FWendorf (ed) 996ndash1040 Dallas Southern Methodist University Press

Bartstra G-J SSoeghondho amp Avan der Wijh 1988 Ngandong man age and artefactsJournal of Human Evolution 17 325ndash7

Beaton JM 1983 Does intensification account for changes in the Australian Holocenearchaeological record Archaeology in Oceania 18 94ndash7

Birdsell JH 1967 Preliminary data on the trihybrid origin of the Australian AboriginesArchaeology and Physical Anthropology in Oceania 2 100ndash55

Birdsell JH 1977 The recalibration of a paradigm for the first peopling of GreaterAustralia In Sunda and Sahul prehistoric studies in Southeast Asia Melanesia andAustralia JAllen JGolson amp RJones (eds) 111ndash67 London Academic Press

Birdsell JH 1979 A reassessment of the age sex and population affinities of the Niahcranium American Journal of Physical Anthropology 50 419

Boaz NT amp AKBehrensmeyer 1976 Hominid taphonomy transport of humanskeletal parts in an artifical fluviatile environment American Journal of PhysicalAnthropology 45 53ndash60

Bowler J M 1976 Recent developments in reconstructing late Quaternaryenvironments in Australia In The origin of the Australians RLKirk amp AGThorne(eds) 55ndash77 Canberra Australian Institute of Aboriginal Studies

Bowler JM amp AGThorne 1976 Human remains from Lake Mungo discovery andexcavation of Lake Mungo III In The origin of the Australians RLKirk ampAGThorne (eds) 127ndash38 Canberra Australian Institute of Aboriginal Studies

Bowler JM RJones HAllen amp AGThorne 1970 Pleistocene human remains fromAustralia a living site and human cremation from Lake Mungo western New SouthWales World Archaeology 1 39ndash60

Bowler JM GSHope JNJennings GSingh amp DWalker 1976 Late Quaternaryclimates of Australia and New Guinea Quaternary Research 6 359ndash94

Brauer G 1984 The Afro-European sapiens-hypothesis and hominid evolution in Asiaduring the later Middle and Upper Pleistocene Courier Forschungsinstitut Senckenberg69 145ndash66

Brauer G 1989 The evolution of modern humans a comparison of the African andnon-African evidence In The human revolution behavioural and biological perspectives onthe origins of modern humans PAMellars amp CBStringer (eds) 123ndash54 EdinburghEdinburgh University Press

Brown P 1981 Artificial cranial deformation a component in the variation inPleistocene Australian Aboriginal crania Archaeology in Oceania 16 156ndash67

Brown P 1987 Pleistocene homogeneity and Holocene size reduction the Australianhuman skeletal evidence Archaeology in Oceania 22 41ndash7

Brown P 1989 Coobool Creek Terra Australis 13 Department of Prehistory ResearchSchool of Pacific Studies Australian National University Canberra

Butlin N 1983 Our original aggression Canberra Australian National University PressCaddie DS DSHunter PJPomery amp HJHall 1987 The ageing chemistmdashcan

electron spin resonance (ESR) help In Archaeometry further Australasian studiesWRAmbrose amp JMJMummery (eds) 167ndash76 Canberra Australian National

University PressChappell J 1976 Aspects of later Quaternary palaeogeography of the AustralianmdashEast

Indonesian region In The origin of the Australians RLKirk amp AGThorne (eds) 11ndash22 Canberra Australian Institute of Aboriginal Studies

Chappell J 1982 Sea levels and sediments some features of the context of coastalarchaeological sites in the tropics Archaeology in Oceania 17 69ndash78

Chappell J amp AGrindrod 1983 CLIMANZ Proceedings of the First Climanz 1981Canberra Department of Biogeography and Geomorphology Australian NationalUniversity

Chappell J amp BGThorn 1977 Sea levels and coasts In Sunda and Sahul prehistoricstudies in Southeast Asia Melanesia and Australia JAllen JGolson amp RJones (eds)275ndash91 London Academic Press

Coon CS 1962 The origin of races New York Alfred A KnopfDeacon HJ 1985 How did past climates affect prehistoric people in Australia and

South Africa The Digging Stick 2(2) 5ndash6Deacon HJ amp JFThackeray 1984 Late Pleistocene environmental changes arid

implications for the archaeological record in southern Africa In Late Cainozoicpalaeoclimates of the southern hemisphere JCVogel (ed) 375ndash90 Rotterdam Balkema

Delson E 1985 Late Pleistocene human fossils and evolutionary relationships InAncestors the hard evidence EDelson (ed) 296ndash300 New York Liss

Flood J 1983 Archaeology of the dreamtime Sydney CollinsFranklin NR 1990 Explorations of variability in Australian prehistoric rock

engravings Unpublished PhD dissertation Department of Archaeology La TrobeUniversity

Freedman L 1985 Human skeletal remains from Mossgiel NSW Archaeology in Oceania20 21ndash31

Freedman L amp MLofgren 1979 Human skeletal remains from Cossack WesternAustralia Journal of Human Evolution 8 283ndash99

Freedman L amp MLofgren 1983 Human skeletal remains from Lake Tandou NSWArchaeologyy in Oceania 18 98ndash105

111REFERENCES

Gamble C 1983 Culture and society in the Upper Palaeolithic of Europe In Hunter-gatherer economy in prehistory GNBailey (ed) 201ndash11 Cambridge CambridgeUniversity Press

Greene DL amp GArmelagos 1972 The Wadi Halfa Mesolithic population AmherstResearch Report 11 Amherst Department of Anthropology University ofMassachusetts

Groube L JChappell JMuke amp DPrice 1986 A 40000-year-old human occupationsite at Huon Peninsula Papua New Guinea Nature 324 453ndash5

Groves C 1989 A regional approach to the problem of the origin of modern humansin Australasia In The human revolution behavioural and biological perspectives on the originsof modern humans PAMellars amp CBStringer (eds) 274ndash85 Edinburgh EdinburghUniversity Press

Habgood PJ 1985 The origin of the Australian Aborigines an alternative approach andview In Hominid evolution past present and future PVTobias (ed) 367ndash80 New YorkLiss

Habgood PJ 1986a A late Pleistocene prehistory of Australia the skeletal materialPhysical Anthropology News 5 1ndash5

Habgood PJ 1986b The origin of the Australians a multivariate approach Archaeology inOceania 21 130ndash7

Habgood PJ 1989 The evolution of modern humans evidence from Australasia seen ina global context In The human revolution behavioural and biological perspectives on theorigins of modern humans PAMellars amp CBStringer (eds) 245ndash73 EdinburghEdinburgh University Press

Habgood PJ In press A morphometric investigation into the origin of anatomically modernhumans British Archaeological Reports

Hayden B 1972 Population control among huntergatherers World Archaeology 4 205ndash21

Hiscock P 1984 Preliminary report on the stone artefacts from Colless Creek CaveNorthwest Queensland Queensland Archaeological Research 1 120ndash51

Hope JH amp GSHope 1976 Palaeoenvironments for man in New Guinea In The originof the Australians RLKirk amp AGThorne (eds) 29ndash55 Canberra Australian Instituteof Aboriginal Studies

Howells WW 1973 The Pacific islanders London Weidenfeld amp NicolsonHowells WW 1976 Metrical analysis in the problem of Australian origins In The origin

of the Australians RLKirk amp AGThorne (eds) 141ndash60 Canberrra AustralianInstitute of Aboriginal Studies

Jones R 1989 East of Wallacersquos line issues and problems in the colonization of theAustralian continent In The human revolution behavioural and biological perspectives onthe origins of modern humans PAMellars amp CBStringer (eds) 743ndash82 EdinburghEdinburgh University Press

Kirk RL amp AGThorne 1976 In The origin of the Australians RLKirk amp AG Thorne(eds) 1ndash8 Canberra Australian Institute of Aboriginal Studies

Larnach SL amp NWGMacintosh 1970 The craniology of the Aborigines of QueenslandOceania Monograph 15

Lourandos H 1983 Intensification A late PleistocenendashHolocene archaeologicalsequence from southwestern Victoria Archaeology in Oceania 18 81ndash94

McArthur N 1976 Computer simulations of small populations Australian Archaeology 453ndash7

McArthur N IWSaunders amp RLTweedie 1976 Small population isolates a micro-simulation study Journal of the Polynesian Society 85 307ndash26

Macintosh NWG 1965 The physical aspects of man in Australia In Aboriginal man inAustralia RMBerndt amp CHBerndt (eds) 29ndash70 Sydney Angus and Robertson

Macintosh NWG 1971 Analysis of an Aboriginal skeleton and a pierced toothnecklace from Lake Nitchie Australia Anthropologie 9 49ndash62

Macintosh NWG amp SLLarnach 1976 Aboriginal affinities looked at in worldcontext In The origin of the Australians RLKirk amp AGThorne (eds) 115ndash26Canberra Australian Institute of Aboriginal Studies

Mayer E 1959 Isolation as an evolutionary factor Proceedings of the American PhilosophicalSociety 103 221ndash9

Maynard L 1979 The archaeology of Australian Aboriginal art In Exploring the visual artof Oceania SMMead (ed) 83ndash110 Honolulu University Press of Hawaii

Nanson GC RWYoung amp EDStockton 1987 Chronology and palaeoenvironmentof the Cranebrook Terrace (near Sydney) containing artefacts more than 40000 yearsold Archaeology in Oceania 22 72ndash8

Pardoe C 1984 Prehistoric human morphological variation in Australia UnpublishedPhD dissertation Department of Prehistory Australian National UniversityCanberra

Pardoe C 1988 The cemetery as symbol The distribution of prehistoric Aboriginalburial grounds in southeastern Australia Archaeology in Oceania 23 1ndash16

Pearce RH amp MBarbetti 1981 A 38000-year-old archaeological site at Upper SwanWestern Australia Archaeology in Oceania 16 173ndash8

Peterson N (ed) 1976 Tribes and boundaries in Australia Canberra Australian Institute ofAboriginal Studies

Peterson N (in collaboration with Jeremy Long) 1986 Australian territorial organisation aband perspective Oceania Monograph 30

Santa Luca AP 1980 The Ngandong fossil hominids a comparative study of a Far EasternHomo erectus group Yale University Publications in Anthropology 78

Smith MA 1989 The case for a resident human population in the Central Australianranges during full glacial aridity Archaeology in Oceania 24 93ndash105

Smith FH JFSimek amp MSHarrill 1979 Geographical variation in supraorbital[torus] reduction In The human revolution behavioural and biological perspectives on theorigins of modern humans PMellars amp CStringer (eds) 172ndash93 Edinburgh EdinburghUniversity Press

Stringer CB 1984 The definition of Homo erectus and the existence of the species inAfrica and Europe Courier Forschungsinstitut Senckenberg 69 131ndash44

Stringer CB amp PAndrews 1988 Genetic and fossil evidence for the origin of modernhumans Science 239 1263ndash8

Stringer CB JJHublin amp BVandermeersch 1984 The origins of anatomicallymodern humans in western Europe In The origin of modern humans FHSmith amp FSpencer (eds) 51ndash135 New York Liss

Thorne AG 1971a Mungo and Kow Swamp morphological variation in PleistoceneAustralia Mankind 8 85ndash9

Thorne AG 1971b The racial affinities and origins of the Australian Aborigines InAboriginal man and environment in Australia DJMulvaney amp JGolson (eds) 316ndash25Canberra Australian National University Press

Thorne AG 1975 Kow Swamp and Lake Mungo Unpublished PhD dissertationDepartment of Anthropology University of Sydney

Thorne AG 1976 Morphological contrasts in Pleistocene Australians In The Origin ofthe Australians RLKirk amp AGThorne (eds) 95ndash112 Canberra Australian Instituteof Aboriginal Studies

Thorne AG 1977 Separation or reconciliation Biological clues to the development ofAustralian society In Sunda and Sahul prehistoric studies in Southeast Asia Melanesia andAustralia JAllen JGolson amp RJones (eds) 187ndash204 London Academic Press

113REFERENCES

Thorne AG 1984 Australiarsquos human originsmdashhow many sources American Journal ofPhysical Anthropology 63 227 (abstract)

Thorne AG amp PGMacumber 1972 Discoveries of late Pleistocene man at KowSwamp Australia Nature 238 316ndash19

Thorne AG amp SRWilson 1977 Pleistocene and recent Australians a multivariatecomparison Journal of Human Evolution 6 393ndash402

Thorne AG amp MHWolpoff 1981 Regional continuity in Australasian Pleistocenehominid evolution American Journal of Physical Anthropology 55 337ndash41

Veth P 1989 Islands in the interior a model for the colonization of Australiarsquos aridzone Archaeology in Oceania 24 81ndash92

Webb SG 1984 Intensification population and social change in southeastern Australiathe skeletal evidence Aboriginal History 8 154ndash72

Webb SG 1989 The Willandra Lakes hominids Canberra Department of PrehistoryResearch School of Pacific Studies Australian National University

Webster PJ amp NAStreten 1978 Late Quaternary Ice Age climates of tropical Australiainterpretations and reconstructions Quaternary Research 10 279ndash309

Weidenreich F 1943 The skull of Sinanthropus pekinensis a comparative study of aprimitive hominid skull Palaeontologia Sinica D 10

Weidenreich F 1947 Facts and speculations concerning the origin of Homo sapiensAmerican Anthropologist 49 187ndash203

Weidenreich F 1951 Morphology of Solo man Anthropology Papers of the AmericanMuseum of Natural History 43 205ndash90

White JP amp PJHabgood 1985 La prehistoire de lrsquoAustralie La Recherche 167 730ndash7White JP amp JFOrsquoConnell 1979 Australian prehistory new aspects of antiquity Science

203 21ndash8White JP amp JFOrsquoConnell 1982 A prehistory of Australia New Guinea and Sahul Sydney

Academic PressWhite NG 1979 The use of digital dermatoglyphics in assessing population

relationships in Aboriginal Australia Birth Defects Original Article Series 15 437ndash54Wolpoff MH 1980 Paleoanthropology New York Alfred A KnopfWolpoff MH 1985 Human evolution at the peripheries the pattern at the eastern

edge In Human evolution past present and future PVTobias (ed) 355ndash65 New YorkLiss

Wolpoff MH Wu Xin Zhi amp AGThorne 1984 Modern Homo sapiens origins ageneral theory of hominid evolution involving the fossil evidence from East Asia InThe origins of modern humans a world survey of the fossil evidence FHSmith amp F Spencer(eds) 411ndash83 New York Liss

Yamaguchi B 1967 A comparative osteological study of the Ainu and the Australian AboriginesAustralian Institute of Aboriginal Studies Occasional Papers 10 Human BiologySeries 2

Afar 40age specificity 4 8 43 85agriculture 39 48Ainu 98Amud Israel 83analysis presenceabsence 13 27Andaman Islands 98animal foods

role in hunter-gatherer diet 47significance in hominid evolution 45

anthropology 3 25 29 36 79mentalism in 36

ape Miocene 14archaeological record visibility of 54archaeology palaeolithic and evolutionary

biology 7Arnhem Land Australia 105art parietal Upper Palaeolithic xi 52ndash3

61ndash2 64artefact

nature of 13specialisation as evidence of cultural

isolation 105ndash6 variability indevelopment of specific form

13 35Aurignacian culture 83Australia populating of

dual-source hypothesis 98ndash9 108homogeneity hypothesis 101ndash2 108trihybrid theory 98ndash9 108

Australian Aborigines 42 46 98 103 108cranial morphology 98ndash9 108

australopithicines 3 14ndash15 18

baboons 2 4 17ndash18Beginnerrsquos Luck Tasmania 97 105behaviour

concepts stereotypic vs variable 4environmental mediation of 4 8intraspecies variability 4 8 26 29 35ndash6models single species 4non-genetic transmission xi

behavioural analysisanalogy inference in 6 13 68 70 andempirical evidence 3ndash7 36 47 78conditions vs properties 36

behavioural evolutioncauses vs consequences 28 phenotypicvs epiphenomenal aspects 30

behaviourism 55bias in analysis xi 13 53Binford L 53bipedal locomotion 2 39 45Birdsell JH 98ndash9body size factors affecting 7bones notations on 62boundaries territorial effect on gene flow

108brain as a simulator 31ndash4

computer analogies 31ndash4 67 80development 2 7 30ndash3 36role in behavioural evolution 30 size 732ndash3 45ndash6

burial grounds and population density 107burials 83bushbaby 17bush-buck 17bushpig 18ndash19butchery 18 44 46 cannibalism in chimps 17carnivory hominids 18Carpentarians 98carrying capacity 46 104Cercopithecus aethiops see monkey vervetchewing biomechanical analysis 18chimpanzee

as model for human evolution 2 14ndash1729ndash30 45ndash6

communication information exchange21 45

cooperation in 45ndash6 culture in 21 2945ndash6 developmental constraints 2145ndash6 food procurement processing16ndash20 45ndash6 habitat shelterpreferences 16 19 28 language 28Pygmy 45 selectivity in 21 self-awareness in 15 situationaladaptability of 16 technology tool-making use 1619ndash20 28 45ndash6

China 99classification 56ndash7 61 67ndash70coevolution coevolutionary models 25

27ndash8 70 76cognitive sciences 67Cohuna Australia 97ndash100Colless Creek Cave 106

116 INDEX

communication x 8 26 28ndash9 33ndash6 4561 78

in animals 21 28 45relation to conscious thought 8

competition interspecies intraspecies 2685ndash6 94

complexity relation to intelligence 52ndash3computational knowledge

definition 68metaphorical forms and devices 77

computer science 67consciousness 2 8 31constructivism 55 63Coobool Creek Australia crania 98 100 107cosmology 4ndash6Cossack Australia cranium 97ndash8 108Cranebrook Terrace 97cranial deformation 100 104 107ndash8cranial morphology hominids 9 97ndash109Cro-Magnon people 83cultural behaviour

nonhuman 21 29 45ndash6origins 25

cultural elaboration role of computationalknowledge in 76

cultureconcept of x 14ndash15 25ndash37 77in nonhuman primates xi 14ndash15 2128ndash9

depopulation genetic effects of 108depopulation and extinction 3 83ndash95 104

106 108development concepts linearity in xi 8ndash9diet early hominid 17ndash18 43digging sticks 20 41 45dingo 104diversity species 7duikers 43

East Lake Turkana Kenya 40 45ndash6ecological diversity modern vs Plio-

Pleistocene 40ecology behavioural 3 5 7electron spin resonance 103Eskimos central 46ethnographic analogy 1ndash2 5 13 39ndash48

54 83 87 92ndash5 98ethology xievolution

behavioural vs biological 4convergent 15 45Darwinian 3 27 30 32ndash5 37genetic influence of culture 27 human

use of nonhuman analogues 13ndash17neo-Darwinian 26non-Darwinian 25parallel 15 21 27relation to complexity 64ndash5single-species model 26ndash7unilineal 3 8ndash9 25 27

evolutionary change constancy of 9exchange systems 53 64ndash5 84

field studies primates 3ndash4 14 19fishing 53 60 64food

preferences procurement hunter-gatherer41 44ndash5

processing 16 19ndash20procurement early hominids 43sharing 2 19 25 43ndash6stress 44

fossil analysis 3 7 9 14 25 39ndash40 97ndash109

founder effect 108functionalism 3

Gwi San 40ndash1Galago senegalensis see bushbabiesgender specificity 4 8 43 85gene flow 107ndash8gene pools 106genetic drift 106 108genetic variation effect of isolation 106ndash7gibbon 14group size early hominids 47 Hadza society 39ndash40 42heredity 6Hohlenstein-Stadel Germany 62hominids

behaviour nonhumans as analogues2ndash5 14ndash17

diet 17ndash18ecological variation among 16Miocene Pliocene 14 18selectivity in 21social economic behaviour 44

hominization and savanna 16Homo erectus 14 20 25 102ndash4Homo habilis 14human concept of xi 2 4ndash6 8 21 25 29

37hunter-gatherers 13 39ndash48 as models for

early hominids 47ndash8hunting role in human evolution 45ndash7Huon New Guinea 97

117INDEX

I Ching 68ndash70iconography iconographie systems 67 70

79imitative learning 29 34 36India 98Indonesia 98ndash100 102ndash3infanticide 105information transmission diffusion

exchange x 26 28ndash9 35ndash6 61 78innovation x 29 32 36intelligence 2 8 15 20 52ndash65

and complexity 52ndash3 archaic Hsapiens vs modern 62 artificial 67concept of 53ndash4 cultural milieu and64ndash5nonhumans and 53

interbreeding 101invention independent 78IQ 53ndash4isolation geographical 105ndash8

Java 99 Kalahari desert 39ndash40 42 44 46

Dobe area 40 42Keilor Australia cranium 97ndash9 107Koobi Fora 16Koonalda Cave Australia 97 105Kosipe New Guinea 97 105Kow Swamp Australia crania 97ndash100Kroeber AL 29Kung San 40ndash1 46 92ndash5Kutikina Cave Tasmania 97 105

labour division of 43 45ndash7Laetolil Tanzania 40Lake Eyasi Tanzania 39Lake Garnpung 103Lake Mungo Australia crania 97ndash100 103ndash5Lake Nitchie Australia 97ndash9Lake Tandou Australia 97ndash8language 2 15 25 33 36 46ndash7 67 70

and computational knowledge 76relation to thought 36 role in culturalevolution 46ndash7

Lascaux 80 83Lawn Hill Gorge 106learning 26ndash9 77Leroi-Gourhan A 61ndash2 64 80Leacutevi-Strauss C 6 78ndash80Lindner Australia 97 105linguistics historical 78lithics in hominid classification 16Lower Omo river Ethiopia 40

macaques Japanese 28Macassan trepang fishermen 98Magdelanian culture 83Mahale Mountains Tanzania 19Makapansgat southern Africa 40Malay Peninsula 98mangetti nuts 40Matenkupkum 97 105material culture chimpanzee 13ndash21mating systems 5 46 105ndash6 108maturation rate 27meat procurement hominids 44meat-eating 4 25 40 42 44ndash5 47microwear analysis tools 20mind 8 20 68monkey vervet 28morphological variation and genetic

variation 101 106ndash7Mossgiel Australia 97ndash8 100Mount Assirik Senegal 16ndash19multivariate analysis cranial morphology

99Murray River corridor 107ndash8Murrayians 98mythology myth systems 67ndash8 70 77ndash9

natural selection 3ndash4 25ndash8 30 108focus on individuals 26ndash7

Navaho curative ceremonies 70navigation Micronesian 63ndash5Neander valley Germany 82Neanderthals

and H sapiens sapiens xi 7 9 53ndash6582ndash95

simulation of population dynamicsextinction 83ndash95

Negev desert 80negritoes modern 100ndash1New Guinea highlands 65 97ndash8Ngandong (Java) hominids fauna 102ndash4nonhumans as hominid analogues 1ndash5notation in Palaeolithic 52 62

Oceanic Negritoes 98Olduvai Gorge Tanzania 45ndash6origins problems in study of 4ndash6 9 Pan paniscus see chimpanzeePan troglodytes see chimpanzeePapio anubis see baboonPapio papio 18perception sensory 31 34 36Perigordian culture 83Philippine Islands 98

118 INDEX

physics 4ndash5Piaget Jean 20Piagetian theory xi 20 54ndash65

cross-cultural use 57 63planning long-term 52 60 64plants

foods 17ndash18 20 40ndash3 45remains at Plio-Pleistocene sites 43

Pleistocene studies Western bias in xipopulation

controls 105dynamics simulated Upper Palaeolithic

84ndash95effect of isolation 105ndash6

population growtheffect on morphological variation107ndash9 epipalaeolithic period 93ndash5Sahul 104ndash5 stability in 95

Potamochoerus porcus see bushpigProconsul 14psychology xi 15 55 63Puritjarra Australia 97 105Pygmies 39ndash40 42 46 Qafzeh 83quantum mechanics 5

radiocarbon analysis 103radiography 104ramapithecines 18recapitulation theories xi 55reductionism methodological 25ndash6 35

37relativism historical cultural 3 5 9 64relativity theory of 5religion ritual activity 53 57 59 61ndash2 67 77reproductive advantage 3 26resource competition 84rock art Australia Panaramitee style 101ndash2

105ndash6rules in behaviour xi 67 76

Sahul populating of 97ndash109San society 39ndash42 44 46ndash7Sapir-Whorf hypothesis 76scale in analysis 1 87scanning electron microscopy 18scavenging 17 19selectivity specificity 21 83settlement patterns early hominid 42sexual dimorphism 7shamanism 67 77ndash8simulations computer 6ndash7 31 67ndash95 105Sivapithecus indicus 18

Skhul 83social interaction 2 31ndash2 82ndash95social organisation 32 41 59 61sociobiology 3Solutrean culture 83spatial relations 55 57speciation 3 25 27speech 2 27 33Spengler O 79St Cesaire 83standardization 35Sterkfontein southern Africa 40stone tools see toolsstorage 60structuralism 3 6 55 76 78ndash80style and information exchange patterns

61subsistence

hominid 39nonhuman primate 39Upper Palaeolithic vs earlier 60

Sunda 100 102symbols symbolic use communication

14ndash15 25ndash6 31 33 35ndash6 52 61Middle Palaeolithic 62

Tabun Israel 83Tai Forest Ivory Coast 19Talgai Australia 97ndash8 107Tasmania 97 101 105Taung southern Africa 40technology

and formal thinking 59nonhuman 9 13ndash21stone tool early vs modern 2 8

teethmorphology 18 20 27study of tooth-wear patterns 18 45

termite probing by chimpanzees 16 19ndash20theory

and empirical validation 5middle range 53

Thorne AG 98ndash9thought

computational models of 67ndash80concrete vs formal (abstract) 56 63environmental mediation of 8operational vs preoperational 56relation to language 36 Western vsnon-Western 63ndash5

thought formal (abstract)and Magdalenian art 62relativity of 63ndash4

119INDEX

tool-making use 2 9 13ndash21 25ndash7 33 3941ndash2 44ndash5 53 57ndash60 62 64 83101 105

by nonhumans 9 16 19ndash20 28 45ndash6tool types technologies

Acheulean 20 57 59 64Chatelperronean 83 Levallois 59Mousterian 83 Oldowan 20ndash1Solutrean 60

toolscuration of 19 53 59ndash60 64design relation to raw material 16stone geometry of 57ndash60 64

tortoises 43Toynbee A 79trace element analysis 103trade 61 84tradition 28ndash9 35ndash6 78

as a concept 28ndash9Tragelaphus scriptus see bush-bucktransportation of material objects food

20 43 45ndash6 61

tula adze 105

ungulates 17Upper Palaeolithic behavioural change in

52 64ndash5 80 83Upper Swan Australia 97 105

Wajak 1 cranium 104Wallen Wallen Creek Australia 97 105warfare 84Willandra Lakes Australia hominids 100ndash1

103ndash4Wolpoff MH 26ndash7World Archaeological Congress I xworld view 76 X-ray diffraction 6 Zaire 39 46

Book Cover

Contents

List of contributors page

Foreword

Preface

Introduction investigating the origins of human behaviour

Chimpanzee material culture what are its limits and why

Culture and symbols

Home bases

Conclusions

How useful is the culture concept in early hominid studies

Conclusions

The significance of modern hunter-gatherers in the study of early hominid behaviour

Settlement patterns

Social organization

Conclusions

Archaeological evidence for modern intelligence

Concrete operations

Formal operations

Conclusions

The invention of computationally plausible knowledge systems in the Upper Palaeolithic

The evidence of Levi-Strauss

Conclusions

An interactive growth model applied to the expansion of Upper Palaeolithic populations

The model

Conclusions

Aboriginal fossil hominids evolution and migrations

London and New York

Foreword

x FOREWORD

xiFOREWORD

xii FOREWORD

PJUckoSouthampton

References

Contents

Introduction 52

xiv CONTENTS

Index 115

Preface

RAFoleyCambridge

2 INTRODUCTION

3INTRODUCTION

4 INTRODUCTION

5INTRODUCTION

6 INTRODUCTION

7INTRODUCTION

8 INTRODUCTION

9REFERENCES

References

10 INTRODUCTION

11REFERENCES

Introduction

Culture and symbols

CULTURE AND SYMBOLS

Home bases

Technology

TECHNOLOGY

21REFERENCES

Conclusions

References

23REFERENCES

Introduction

Conclusions

CONCLUSIONS

37REFERENCES

References

Introduction

Settlement patterns

SETTLEMENT PATTERNS

Social organization

SOCIAL ORGANIZATION

Conclusions

CONCLUSIONS

Acknowledgement

References

49REFERENCES

51REFERENCES

Introduction

Concrete operations

CONCRETE OPERATIONS

Formal operations

FORMAL OPERATIONS

Conclusions

65REFERENCES

References

BASIC STRUCTURE

Another example

BASIC STRUCTURE

ATOS AND SHAMANISM

Conclusions

CONCLUSIONS

References

81REFERENCES

Grantland Rice

The background

The model

THE MODEL

RESULTS

95REFERENCES

Conclusions

References

109REFERENCES

Acknowledgements

References

111REFERENCES

113REFERENCES

61ndash2 64artefact

116 INDEX

117INDEX

118 INDEX

61subsistence

119INDEX

20 43 45ndash6 61

tula adze 105

Book Cover

Contents

Foreword

Preface

Culture and symbols

Home bases

Conclusions

Settlement patterns

Social organization

Conclusions

Concrete operations

Formal operations

Conclusions

The model

Conclusions

London and New York

Foreword

x FOREWORD

xiFOREWORD

xii FOREWORD

PJUckoSouthampton

References

Contents

Introduction 52

xiv CONTENTS

Index 115

Preface

RAFoleyCambridge

2 INTRODUCTION

3INTRODUCTION

4 INTRODUCTION

5INTRODUCTION

6 INTRODUCTION

7INTRODUCTION

8 INTRODUCTION

9REFERENCES

References

10 INTRODUCTION

11REFERENCES

Introduction

Culture and symbols

CULTURE AND SYMBOLS

Home bases

Technology

TECHNOLOGY

21REFERENCES

Conclusions

References

23REFERENCES

Introduction

Conclusions

CONCLUSIONS

37REFERENCES

References

Introduction

Settlement patterns

SETTLEMENT PATTERNS

Social organization

SOCIAL ORGANIZATION

Conclusions

CONCLUSIONS

Acknowledgement

References

49REFERENCES

51REFERENCES

Introduction

Concrete operations

CONCRETE OPERATIONS

Formal operations

FORMAL OPERATIONS

Conclusions

65REFERENCES

References

BASIC STRUCTURE

Another example

BASIC STRUCTURE

ATOS AND SHAMANISM

Conclusions

CONCLUSIONS

References

81REFERENCES

Grantland Rice

The background

The model

THE MODEL

RESULTS

95REFERENCES

Conclusions

References

109REFERENCES

Acknowledgements

References

111REFERENCES

113REFERENCES

61ndash2 64artefact

116 INDEX

117INDEX

118 INDEX

61subsistence

119INDEX

20 43 45ndash6 61

tula adze 105

Book Cover

Contents

Foreword

Preface

Culture and symbols

Home bases

Conclusions

Settlement patterns

Social organization

Conclusions

Concrete operations

Formal operations

Conclusions

The model

Conclusions

London and New York

Foreword

x FOREWORD

xiFOREWORD

xii FOREWORD

PJUckoSouthampton

References

Contents

Introduction 52

xiv CONTENTS

Index 115

Preface

RAFoleyCambridge

2 INTRODUCTION

3INTRODUCTION

4 INTRODUCTION

5INTRODUCTION

6 INTRODUCTION

7INTRODUCTION

8 INTRODUCTION

9REFERENCES

References

10 INTRODUCTION

11REFERENCES

Introduction

Culture and symbols

CULTURE AND SYMBOLS

Home bases

Technology

TECHNOLOGY

21REFERENCES

Conclusions

References

23REFERENCES

Introduction

Conclusions

CONCLUSIONS

37REFERENCES

References

Introduction

Settlement patterns

SETTLEMENT PATTERNS

Social organization

SOCIAL ORGANIZATION

Conclusions

CONCLUSIONS

Acknowledgement

References

49REFERENCES

51REFERENCES

Introduction

Concrete operations

CONCRETE OPERATIONS

Formal operations

FORMAL OPERATIONS

Conclusions

65REFERENCES

References

BASIC STRUCTURE

Another example

BASIC STRUCTURE

ATOS AND SHAMANISM

Conclusions

CONCLUSIONS

References

81REFERENCES

Grantland Rice

The background

The model

THE MODEL

RESULTS

95REFERENCES

Conclusions

References

109REFERENCES

Acknowledgements

References

111REFERENCES

113REFERENCES

61ndash2 64artefact

116 INDEX

117INDEX

118 INDEX

61subsistence

119INDEX

20 43 45ndash6 61

tula adze 105

Book Cover

Contents

Foreword

Preface

Culture and symbols

Home bases

Conclusions

Settlement patterns

Social organization

Conclusions

Concrete operations

Formal operations

Conclusions

The model

Conclusions

Foreword

x FOREWORD

xiFOREWORD

xii FOREWORD

PJUckoSouthampton

References

Contents

Introduction 52

xiv CONTENTS

Index 115

Preface

RAFoleyCambridge

2 INTRODUCTION

3INTRODUCTION

4 INTRODUCTION

5INTRODUCTION

6 INTRODUCTION

7INTRODUCTION

8 INTRODUCTION

9REFERENCES

References

10 INTRODUCTION

11REFERENCES

Introduction

Culture and symbols

CULTURE AND SYMBOLS

Home bases

Technology

TECHNOLOGY

21REFERENCES

Conclusions

References

23REFERENCES

Introduction

Conclusions

CONCLUSIONS

37REFERENCES

References

Introduction

Settlement patterns

SETTLEMENT PATTERNS

Social organization

SOCIAL ORGANIZATION

Conclusions

CONCLUSIONS

Acknowledgement

References

49REFERENCES

51REFERENCES

Introduction

Concrete operations

CONCRETE OPERATIONS

Formal operations

FORMAL OPERATIONS

Conclusions

65REFERENCES

References

BASIC STRUCTURE

Another example

BASIC STRUCTURE

ATOS AND SHAMANISM

Conclusions

CONCLUSIONS

References

81REFERENCES

Grantland Rice

The background

The model

THE MODEL

RESULTS

95REFERENCES

Conclusions

References

109REFERENCES

Acknowledgements

References

111REFERENCES

113REFERENCES

61ndash2 64artefact

116 INDEX

117INDEX

118 INDEX

61subsistence

119INDEX

20 43 45ndash6 61

tula adze 105

Book Cover

Contents

Foreword

Preface

Culture and symbols

Home bases

Conclusions

Settlement patterns

Social organization

Conclusions

Concrete operations

Formal operations

Conclusions

The model

Conclusions

Foreword

x FOREWORD

xiFOREWORD

xii FOREWORD

PJUckoSouthampton

References

Contents

Introduction 52

xiv CONTENTS

Index 115

Preface

RAFoleyCambridge

2 INTRODUCTION

3INTRODUCTION

4 INTRODUCTION

5INTRODUCTION

6 INTRODUCTION

7INTRODUCTION

8 INTRODUCTION

9REFERENCES

References

10 INTRODUCTION

11REFERENCES

Introduction

Culture and symbols

CULTURE AND SYMBOLS

Home bases

Technology

TECHNOLOGY

21REFERENCES

Conclusions

References

23REFERENCES

Introduction

Conclusions

CONCLUSIONS

37REFERENCES

References

Introduction

Settlement patterns

SETTLEMENT PATTERNS

Social organization

SOCIAL ORGANIZATION

Conclusions

CONCLUSIONS

Acknowledgement

References

49REFERENCES

51REFERENCES

Introduction

Concrete operations

CONCRETE OPERATIONS

Formal operations

FORMAL OPERATIONS

Conclusions

65REFERENCES

References

BASIC STRUCTURE

Another example

BASIC STRUCTURE

ATOS AND SHAMANISM

Conclusions

CONCLUSIONS

References

81REFERENCES

Grantland Rice

The background

The model

THE MODEL

RESULTS

95REFERENCES

Conclusions

References

109REFERENCES

Acknowledgements

References

111REFERENCES

113REFERENCES

61ndash2 64artefact

116 INDEX

117INDEX

118 INDEX

61subsistence

119INDEX

20 43 45ndash6 61

tula adze 105

Book Cover

Contents

Foreword

Preface

Culture and symbols

Home bases

Conclusions

Settlement patterns

Social organization

Conclusions

Concrete operations

Formal operations

Conclusions

The model

Conclusions

Foreword

x FOREWORD

xiFOREWORD

xii FOREWORD

PJUckoSouthampton

References

Contents

Introduction 52

xiv CONTENTS

Index 115

Preface

RAFoleyCambridge

2 INTRODUCTION

3INTRODUCTION

4 INTRODUCTION

5INTRODUCTION

6 INTRODUCTION

7INTRODUCTION

8 INTRODUCTION

9REFERENCES

References

10 INTRODUCTION

11REFERENCES

Introduction

Culture and symbols

CULTURE AND SYMBOLS

Home bases

Technology

TECHNOLOGY

21REFERENCES

Conclusions

References

23REFERENCES

Introduction

Conclusions

CONCLUSIONS

37REFERENCES

References

Introduction

Settlement patterns

SETTLEMENT PATTERNS

Social organization

SOCIAL ORGANIZATION

Conclusions

CONCLUSIONS

Acknowledgement

References

49REFERENCES

51REFERENCES

Introduction

Concrete operations

CONCRETE OPERATIONS

Formal operations

FORMAL OPERATIONS

Conclusions

65REFERENCES

References

BASIC STRUCTURE

Another example

BASIC STRUCTURE

ATOS AND SHAMANISM

Conclusions

CONCLUSIONS

References

81REFERENCES

Grantland Rice

The background

The model

THE MODEL

RESULTS

95REFERENCES

Conclusions

References

109REFERENCES

Acknowledgements

References

111REFERENCES

113REFERENCES

61ndash2 64artefact

116 INDEX

117INDEX

118 INDEX

61subsistence

119INDEX

20 43 45ndash6 61

tula adze 105

Book Cover

Contents

Foreword

Preface

Culture and symbols

Home bases

Conclusions

Settlement patterns

Social organization

Conclusions

Concrete operations

Formal operations

Conclusions

The model

Conclusions

Documents

The Origins of Human Behaviour...development of social hierarchy and political centralization J.Gledhill et al. (eds) The walking larder: patterns of domestication, pastoralism, and