Ingold Ed Encyclopedia of Anthropology Part 3

COMPANION ENCYCLOPEDIA
OF ANTHROPOLOGY
COMPANION
ENCYCLOPEDIA
OF
ANTHROPOLOGY
EDITED BY
TIM INGOLD
London and New York

First published in 1994
by Routledge
11 New Fetter Lane, London EC4P 4EE
This edition published in the Taylor & Francis e-Library, 2003.
Simultaneously published in the USA and Canada
by Routledge, Inc.
29 West 35th Street, New York, NY 10001
Structure and editorial matter© 1994 Tim Ingold
The chapters © 1994 Routledge
All rights reserved. No part of this book may be reprinted or reproduced or utilized in
any form or by any electronic, mechanical or other means, now known or hereafter
invented, including photocopying and recording, or in any information storage or
retrieval system, without permission in writing from the publishers.
British Library Cataloguing in Publication Data
A catalogue record for this book is available from the British Library.
Library of Congress Cataloging-in-Publication Data
A catalog record for this book is available on request.

ISBN 0-203-03632-8 Master e-book ISBN
CONTENTS
Preface IX
General introduction
Tim Ingold xm
The contributors xxm
PART 1: HUMANITY 1
1. Introduction to humanity
Tim Ingold 3
2. Humanity and animality
Tim Ingold 14
3. The evolution of early hominids
t tp . o tas
4. Human evolution: the last one million years
Clive Gamble 79
5. The origins and evolution oflanguage
Philip Lieberman 108
6. Tools and tool behaviour
Thomas ftYnn 133
7. Niche construction, evolution and culture
F.J. Odling-Smee 162
8. Modes of subsistence: hunting and gathering to agriculture
and pastoralism
Roy Ellen 197
9. The diet and nutrition of human populations
Igor de Garine 226
10. Demographic expansion: causes and consequences
MarkNCohen 265
11. Disease and the destruction of indigenous populations
Step hen J.Kunitz 297
PART 11: CULTURE 327

12. Introduction to culture
Tim Ingold 329
V
CONTENTS
13. Why animals have neither culture nor history

David Premack and Ann James Premack 350
14. Symbolism: the foundation of culture
Mary LeCron Foster 366
15. Artefacts and the meaning of things
Daniel Miller 396
16. Technology
Fran{ois Sigaut 420
17. Spatial organization and the built environment
Amos Rapoport 460
18. Perceptions of time
Barbara A dam 503
19. Aspects of literacy
Brian V.Street and Niko Besnier 527
20. Magic, religion and the rationality of belief
Gilbert Lewis 563
21. Myth and metaphor
James F. Weiner 591
22. Ritual and performance
Richard Schechner 613
23. The anthropology of art
Howard Morphy 648
Music an ance
Anthony Seeger 686
25. The politics of culture: ethnicity and nationalism
Anthony D.Smith 706
PART Ill: SOCIAL LIFE 735

26. Introduction to social life
Tim Ingold 737
27. Sociality among humans and non-human animals
R.I.M.Dunbar 756
28. Rules and prohibitions: the form and content of human
kinship
Alan Barnard 783
29. Understanding sex and gender
Henrietta L.Moore 813
30. Socialization, enculturation and the development of personal
identity
Fitz John Porter Poole 831
31. Social aspects of language use
Jean DeBernardi 861
32. Work, the division of labour and co-operation
Sutti Ortiz 891
Vl
CONTENTS
33. Exchange and reciprocity

C.A. Gregory 911
34. Political domination and social evolution
Timothy Earle 940
35. Law and dispute processes
Simon Roberts 962
36. Collective violence and common security
Robert A.Rubinstein 983
37. Inequality and equality
Andri Biteille 1010
38. The nation state, colonial expansion and the contemporary
world order
Peter Worsley 1040
Index 1067
Vll
PREFACE
This volume started life on the initiative of Jonathan Price, at that time
Reference Books Editor at Croom Helm. His idea was for an Encyclopedia of
Human Society whose subject would span the disciplines of anthropology,
sociology and archaeology. We first met to discuss the project in August 1986,
and it was then that he charmed me into agreeing to become the volume's
editor. It has been a big job, to put it mildly. In hindsight, it seems to me that I
must have been mad to take it on at all, let alone single-handed. No doubt my
motives were in part honourable, since I was strongly committed to the idea of
anthropology as a bridging discipline, capable of spanning the many divisions
of the human sciences. I wanted to prove that the possibility of synthesis
existed not just as an ideal, but as something that could be realized in practice.
o ou t, too, was motivate y a certam vamty: 1 a synt es1s was to e m t,
I wanted to be the one to build it, and to reap the credit! Seven years on, I am
both older and perhaps a little wiser-no less committed to the ideal of
synthesis, but a great deal more aware of the complexities involved, and rather
less confident about my own abilities to bring it about.
Following my initial meeting withJonathan Price, over a year passed before I
was able to begin serious work on the project, which we had decided to call
Humanity, Culture and Social Life. In October 1987 I drew up a prospectus for
the entire volume, which included a complete list of forty articles, divided
between the three parts spelled out in the title, and a rough breakdown of the
contents for each. Then, during the first half of 1988, I set about recruiting
authors for each of the articles. Meanwhile, Croom Helm had been subsumed
under Routledge, from whose offices J onathan continued to oversee the project.
My original schedule had been for authors to write their first drafts during
1989, allowing a further nine months for consultation and editorial comment,
with a deadline for final versions of September 1990 and a projected
publication date of April 1992. As always, things did not go entirely according
to schedule, and I soon found that I was receiving final drafts of some articles
while a pile of first drafts of others were awaiting editorial attention, and while
for yet others I was still trying to fill the gaps in my list of contributors. To my
great embarrassment, I found that I was quite unable to keep to my own
deadlines. The inexorable growth of other commitments meant that drafts,
IX
PREFACE
dutifully submitted by their authors at the appointed time, languished for many
months-and in some cases for more than a year-before I could get to work
on them. During the academic year 1990-1, pressures of teaching and
administration, coupled with my assumption of the Editorship of the journal
Man, grew so heavy that progress on the project more or less ground to a halt,
and my deadline for submitting the whole volume to the publishers-set for the
end of April, 1991-passed quietly by with most of the articles still at the first
draft stage.
The project was rescued by my good fortune in securing one whole year and
two subsequent terms of research leave from the University of Manchester.
The first year (1991-2) was made possible in part by a grant from the
University of Manchester Research Support Fund, for which I acknowledge
my profound thanks. The two following terms were taken as sabbatical leave,
and I should like to thank all my colleagues in the Manchester Department of
Social Anthropology for covering my teaching and administrative duties in my
absence. Shortly before his departure from Routledge to join the staff at
Edinburgh University Press, the ever-patient Jonathan Price was finally
rewarded for his forbearance. At noon on 14 October 1992, he arrived in my
office to collect the entire, edited manuscript, and to carry it off to London. I
had completed work on the manuscript only two hours before! But the editorial
introductions had still to be written, and it was not until well into the following
spring that they were eventually finished. Meanwhile, Mark Hendy was hard at
work on the Hercu ean task of su -e 1tmg the who e vo ume, whtch he
completed by the beginning of May. I owe him a debt of gratitude for his
efforts. Since Jonathan left for Edinburgh, responsibility for guiding the
volume through the press passed to Michelle Darraugh, who has been
wonderfully supportive, efficient and understanding. Most of all, however, this
book belongs to Jonathan, without whom it would never have been conceived in
the first place, and whose unflagging enthusiasm kept the project on the rails
even during the most difficult of times.
Looking back, I am surprised how closely the book, in its final form,
resembles the original plan drawn up so many years ago. Only four of the
projected articles have been lost, and the titles and ordering of the majority
have been changed little, if at all. There have been a few changes in the list of
contributors along the way: in particular, I should like to put on record the sad
loss of John Blacking, who died before he could begin work on his projected
article, 'Music and dance'; and I should also like to thank Anthony Seeger for
stepping into the breach at very short notice. There have also been some
changes in the volume's title. All along, I wanted it to be a book to be read, and
not merely consulted as a work of reference, and for that reason I was inclined
to relegate the phrase An Encyclopedia of Anthropology to the subtitle. In many
ways, the book is more akin to what might conventionally be called a handbook
or a reader, rather than an encyclopedia. Be that as it may, after much
discussion it was eventually decided to call it a Companion Encyclopedia, a
X
PREFACE
phrase which nicely combines the notion of encompassing a whole field of

knowledge with that of guiding and accompanying the reader in his or her
journey through it. The original working title, Humanity, Culture and Social
Life, accordingly became the volume's subtitle.
I would like to take this opportunity to extend my personal thanks to all the
many contributors to this book. They have put up patiently with endless delays,
and responded graciously to my many and sometimes inordinate editorial
demands. I have, moreover, learned a tremendous amount from working
through their articles. But for maintaining my sanity over all these years, my
greatest debt of gratitude is to my wife, Anna, and my children, Christopher,
Nicholas and Jonathan. Their support has been magnificent, and it is not
something that I shall ever be able to repay.
Tim Ingold
Manchester
September 1993
Xl
GENERAL INTRODUCTION
THE SCOPE OF ANTHROPOLOGY

Anthropologists study people. They do not study stars, rocks, plants or the
weather. But whilst we may have little difficulty in separating out the field of
anthropological inquiry from those of astronomy, geology, botany or
meteorology, it is not so obvious how-if at all-anthropology may be
distinguished from the many other branches of the human sciences, all of
which could claim to be studying people in one way or another. Medicine is
concerned with the workings of the human body, psychology with those of the
mind; history studies people's activities in the past, sociology their institutional
arrangements in the present, and so on. The list could be extended almost
m e mte y.
people?
Part of the difficulty we have in answering this question is attributable to the
fact that there is not one way of doing anthropology, but many. There are two
facets to this diversity, the first having to do with the circumstances of the
discipline's historical development, the second lying in its contemporary
subdisciplinary divisions. I begin with a few words about anthropology's
history.
In a sense, of course, anthropology can be traced to the earliest antiquity,
when human beings first began to speculate about their own nature, origins and
diversity. But as an explicitly defined field of academic inquiry, it is a creature
of the last two centuries of thought in that region of the world conventionally
known as 'the West'. Western thought, however, is not a monolithic edifice but a
complex interweaving of often opposing currents, and this is no less true of the
career of anthropology. Moreover, these currents did not flow in an historical
vacuum, but at every moment responded to dominant moral, political and
economic concerns of the time. Thus British anthropology developed alongside
the growth of empire; its preoccupations were fuelled by the need of the
colonial administration to take the measure of its presumed superiority over
administered nations, and to turn a knowledge of their social organizations and
cultural traditions to the service of indirect rule. In many countries of
Continental Europe, by contrast, the growth of anthropology (more commonly
xm
known as 'ethnology') was linked to emergent nationalist movements of the late

nineteenth and early twentieth centuries, and to the efforts, on the part of
adherents of each movement, to discover a national heritage in the traditions of
local folk or peasant culture. In North America the situation was different
again: the United States and Canada had their indigenous Indian populations,
and the first priority of many American anthropologists was to record as much
as possible about the physical features, material artefacts, languages and
cultures of extant Amerindian groups before it was too late. This was a kind of
salvage anthropology.
The second facet to the diversity of anthropological approaches lies in the
fact that anthropology, as it exists today, is not a single field, but is rather a
somewhat contingent and unstable amalgam of subfields, each encumbered
with its own history, theoretical agenda and methodological preoccupations. In
the American tradition of scholarship, it has long been customary to
distinguish four such subfields of anthropology, namely physical,
archaeological, cultural and linguistic. In the British tradition, by contrast,
there are only three subfields, of physical anthropology, archaeology and social
(rather than cultural) anthropology. The exclusion of linguistics from British
anthropology is a curious and somewhat scandalous anomaly to which I return
below. The more immediate question is: why these fields in particular? What
brought the study of physical types, ancient artefacts and supposedly
'primitive' ways of life under the umbrella of a single discipline of
anthropo ogy.
Most academic disciplines and their boundaries are, in fact, the fossilized
shells of burnt-out theories, and in this, anthropology is no exception. The
theory which, more than any other, established anthropology as a
comprehensive science of humankind held that people the world over are
undergoing a gradual, evolutionary ascent from primitive origins to advanced
civilization, and that the differences between societies can be explained in terms
of the stages they have reached in this progression. Anthropology, then,
emerged as the study of human evolution-conceived in this progressive
sense-through the reconstruction of its earlier stages. Physical anthropology
studied the evolution of human anatomy, archaeology studied the evolution of
material artefacts, and social and cultural anthropology studied the evolution of
beliefs and practices-on the assumption that the ways of life of contemporary
'primitives' afford a window on the former condition of the more 'civilized'
nations.
In short, it was progressive evolutionism that unified the study of human
anatomy, artefacts and traditions as subfields of a single discipline. Yet this kind
of evolutionary theory belongs essentially to the formative period of
anthropology in the nineteenth century and is, today, almost universally
discredited. So what, if anything, still holds the sublfields together? To the
extent that contemporary anthropologists concern themselves with this
question, their opinions differ greatly. Some argue that their continued
XIV
combination, for example within University Departments, is an anachronism

for which there is no longer any rational justification. Thus many cultural
anthropologists, concerned as they are with the manifold ways in which the
peoples among whom they have worked make sense of the world around them,
find more common ground with students of philosophy, language, literature
and the arts than with their colleagues in other fields of anthropology. Social
anthropologists, who would regard their project as a comparative study of the
generation, patterning and transformation of relationships among persons and
groups, profess a close affinity-amounting almost to identity-to sociologists
and historians, but have little time for archaeology (despite the obvious links
between archaeology and history). For their part, physical anthropologists (or
'biological anthropologists', as many now prefer to be known) remain
committed to the project of understanding human evolution, but their
evolutionary theory is of a modern, neo-Darwinian variety, quite at odds with
the progressive evolutionism of the nineteenth century. Having vigorously
repudiated the racist doctrines of the turn of the century, which cast such a
shadow over the early history of the discipline, anthropologists of all
complexions now recognize that social and cultural variation is quite
independent of biogenetic constraint. Thus physical anthropology, cut loose
from the study of society and culture, has virtually become a subfield of
evolutionary biology, devoted specifically to the evolution of our own kind.
Yet despite these tendencies towards the fragmentation of anthropology,
a ong the mes of the heav1 y mstitutwna tze tvtswn of aca emtc a our
between the humanities and social sciences on the one hand, and the natural
sciences on the other, many anthropologists remain convinced that there is
more to their discipline than the sum of its parts. What is distinctive about the
anthropological perspective, they argue, is a commitment to holism, to the idea
that it should be possible-at least in principle-to establish the
interconnections between the biological, social, historical and cultural
dimensions of human life that are otherwise parcelled up among different
disciplines for separate study. It was, of course, just such a synthesis that the
nineteenth century founders of anthropology claimed to have achieved with
their theory of evolution. But the fact that the theory is now judged, in
hindsight, to have been wrong does not mean that the project that gave rise to it
was entirely misconceived (although aspects of it-such as its assumption of
Euro-American superiority and its racist undertones-undoubtedly were). My
own view, which also furnishes the rationale for the present volume, is that a
synthesis of our knowledge of the conditions of human life in the world, in all
its aspects, is something worth striving for, and that working towards such a
synthesis is the essence of doing anthropology.
The obstacles, however, are formidable. Biological and cultural
anthropologists, for example, are divided not simply by their attention to
different kinds of facts, but by a more fundamental difference in their
respective understandings of the relations between fact and theory. True, the
XV
data of observation in every branch of anthropology have one thing in common:

they are not derived by experiment, but are gathered through the conduct of
fieldwork. But ethnographic fieldwork, as it is carried out by social and cultural
anthropologists in the settings of everyday life, is very different from the kind
of fieldwork that might be conducted by an archaeologist or physical
anthropologist in searching, say, for the fossilized remains of early hominids or
for evidence, in the form of preserved artefacts, of their activities. Fossils and
artefacts can be treated, to all intents and purposes, as inert objects of
investigation: they may be examined for every ounce of information they will
yield, but they are not themselves party to its interpretation. Living people,
however, cannot be treated as objects in this sense. In the field, ethnographers
engage in a continuous dialogue with their informants, who provide instruction
in the skills and knowledge that are entailed in their particular form of life. It
has been said, with some justification, that ethnographers do not so much study
people, as go to study among or with people, and the results of such study
emerge as the products of this mutual, dialogic encounter. Indeed much so-
called 'ethnographic data' is in fact autobiographical, describing the ways in
which the fieldworker experienced those events in which he or she participated.
Under these circumstances, a clear distinction between observation and
interpretation, between the collection of data in the field and their placement
within a theoretical framework, cannot readily be sustained. This did not,
however, prevent the first generation of British social anthropologists-
pwneers of the km of ong-term, mtensiVe 1e stu y that IS now cons1 ere
indispensable to competent ethnographic work-from pretending that it could,
apparently in an effort to secure recognition for their discipline as a true science
of society. This goes some way to explaining the curious neglect, by social
anthropologists of this generation, of language and its uses. Knowledge of the
native language was considered a prerequisite for ethnographic inquiry; as
such, however, it was regarded as a tool of the anthropologist's trade rather
than something to be investigated in its own right. One was to use language to
probe the details of culture and social organization much as a botanist uses a
microscope to examine the fine structure of plants. Only subsequently, as
anthropologists became more reflexive, more sensitive to the epistemological
conditions of their own inquiry, did language use re-emerge as a key focus of
attention. Even in North America, where linguistic anthropology has always
occupied its place among the four subfields of the discipline, its practitioners
have long been in the minority, often drawn into the anthropological camp
through their reaction against the excessive formalism of mainstream
linguistics, and its insensitivity to the social and cultural contexts in which
language is put to work.
But the challenge posed by ethnographic study among people whose
backgrounds and sensibilities are situated in environments very different from
those of the 'West' goes far beyond showing how the seemingly strange or
irrational 'makes sense' when placed in its proper context. For the knowledge
XVI
and wisdom that these people impart to the fieldworker, sharpened as it is by

their practical experience of everyday involvement in the world, strikes at the
heart of some of the most basic presuppositions of Western thought itself. To
take this knowledge seriously, and to be the wiser for it, means bringing it to
bear in a critical engagement with these presuppositions. In this engagement,
every single one of the key concepts of Western civilization-concepts like
society, culture, nature, language, technology, individuality and personhood,
equality and inequality, even humanity itself-becomes essentially contestable.
Theoretical work, in social and cultural anthropology, is largely a matter of
opening up these concepts for inspection and unpacking their contents,
thereby revealing the often hidden baggage that we carry with us into our
encounters with unfamiliar realities. If we are ever to reach a level of
understanding that breaks the barriers between Western and non-Western
worlds of life and thought, such work is indispensable. Yet it also leaves
anthropology perilously poised on a knife-edge. For how can a discipline whose
project is rooted in the intellectual history of the Western world meet the
challenge presented by non-Western understandings of humanity, culture and
social life without undercutting its own epistemological foundations?
Perhaps uniquely among academic disciplines, anthropology thrives on the
art of its own perpetual deconstruction. Caught at the intersection of two
cross-cutting tensions, between the humanities and natural sciences on the one
hand, and between theoretical speculation and lived experience on the other, it
eaves Itt e room for mte ectua comp acency. Like phi osophy, the remit of
anthropology is not confined to a delimited segment within a wider division of
academic labour; rather it exists to subvert any such tidy division, rendering
problematic the very foundations on which it rests. The best anthropological
writing is distinguished by its receptiveness to ideas springing from work in
subjects far beyond its conventional boundaries, and by its ability to connect
these ideas in ways that would not have occurred to their originators, who may
be more enclosed by their particular disciplinary frameworks. But to this
connecting enterprise it brings something more, namely the attempt to engage
our abstract ideas about what human beings might be like with an empirically
grounded knowledge of (certain) human beings as they really are, and of what
for them everyday life is all about. This engagement not only provides the
primary motivation-apart from that of sheer curiosity-for ethnographic
inquiry, but also carries anthropology beyond the closeted realms of speculative
philosophy. Anthropology, if you will, is philosophy with the people in.
No more today than in the past, however, is anthropological work conducted
in an historical vacuum. Just as much as the people they study, anthropologists
are participants in the one world which we all inhabit, and therefore carry their
share of the responsibility for what goes on in it. In many parts of the world,
people currently face appalling deprivations, whether due to poverty, famine,
disease, war, or some combination of these. There is no doubt that
anthropological knowledge, tempered as it is by an awareness of the practical
xvn
realities of life 'on the ground' in real human communities, has a vital
contribution to make in the alleviation of human suffering. Moreover, to an
increasing extent, anthropologists have involved themselves as advocates on
behalf of the peoples among whom they have worked-for example in the
struggle for recognition of indigenous rights to land-or as advisers or
consultants in various projects of development. In view of such involvements, it
has sometimes been suggested that a field of 'applied anthropology' should be
recognized, alongside those branches of the discipline that are already well
established.
If this suggestion has not met with wholehearted approval, the reason does
not lie in any desire to keep anthropology 'pure', nor does it indicate that
anthropologists prefer to wash their hands of the moral and political
entailments of their involvement with local communities. It is rather that in the
conduct of anthropological work it is practically impossible to separate the
acquisition of knowledge from its application. The distinction between pure
and applied science rests on a premiss of detachment, the assumption that
scientists can know the world without having to involve themselves in it. But
anthropology rests on exactly the opposite premiss, that it is only by immersing
ourselves in the life-world of our fellow human beings that we shall ever
understand what it means to them-and to us. Thus whatever else it may be,
anthropology is a science of engagement. Indeed it may be said that in
anthropology we study ourselves, precisely because it requires us to change our
conceptiOn of who we are, from an exc usiVe, Western we to an me usiVe,
global one. To adopt an anthropological attitude is to drop the pretence of our
belonging to a select association of Westerners, uniquely privileged to look in
upon the inhabitants of 'other cultures', and to recognize that along with the
others whose company we share (albeit temporarily), we are all fellow travellers
in the same world. By comparing experience-'sharing notes'-we can reach a
better understanding of what such journeying entails, where we have come
from, and where we are going.
HUMANITY, CULTURE AND SOCIAL LIFE

This is an encyclopedia of anthropology, it is not an encyclopedia about
anthropology. The distinction is critical, and underwrites both the content of
the articles that follow and the structure of the volume as a whole. There is a
tendency, common to many branches of scholarship, for specialists to become
so absorbed in debates internal to the discipline that they lose sight of their
original purpose, namely to extend the scope of our knowledge of the world.
The debates become an object of study in themselves. Though there must be a
place in every discipline for a consideration of its history and its methods, I
believe it is important to resist the inclination to detach such consideration
from the primary objective of enlarging human understanding. In the case of
anthropology, this means that however much we may tangle with the details of
xvm
particular arguments, we should never forget that the pursuit of

anthropological knowledge is for the benefit of people, and not the other way
round. The tapestry of human life, in other words, has not been woven for the
purpose of providing research opportunities for anthropologists; however,
anthropological research can help us to unravel the strands and to reveal the
subtleties of their patterning. This volume, then, is about human life in all its
aspects, and each article, focusing on some specific aspect, sets out what current
studies in anthropology (and in several cases, in contingent disciplines) have to
say about it.
The same principle informs the division of the volume into its three parts,
respectively entitled 'humanity', 'culture' and 'social life'. The emphasis, in the
first, is on human beings as members of a species, on how that species differs
from others, on how it has evolved, and on how human populations have
adapted to-and in turn transformed-their environments. The second part
focuses on the origination, structure, transmission and material expression of
the symbolically constituted forms of human culture, and on the role of culture
in action, perception and cognition. The third part examines the various
facets-familial, economic, political, and so on-of the relationships and
processes that are carried on by persons and groups, through the medium of
cultural forms, in the historical process of social life. Each part begins with an
introductory article that sets out the substantive areas to be covered in greater
depth, and places the articles that follow in their wider anthropological context.
f course any tvtswn of the entire 1e of human tfe IS oun to e
artificial, and there are perhaps as many common themes linking articles in
different parts as within each part of the volume. The point I wish to stress,
however, is that the division is not based on, nor does it correspond with, any of
the conventional divisions of the field of anthropology. It is true that the work
of archaeologists and physical (or biological) anthropologists figures relatively
prominently in the first part, and that work in cultural and social anthropology
predominates in the second and third parts. But if there is one thing that the
volume establishes, beyond any reasonable doubt, it is that the issues of our
common humanity, of cultural variation and of social process can be adequately
tackled only through the collaboration of scholars working in all the
conventional subfields of anthropology-biological, archaeological, cultural,
social, linguistic-and of others besides, whose backgrounds lie in fields as
diverse as medicine, ecology, psychology, cognitive science, history, sociology,
comparative religion, political science, law, philosophy, architecture, drama,
folklore and ethnomusicology. Indeed, practitioners of several of these latter
fields number among the contributors to this book.
To attempt to compress all of human life within two covers may seem a
hopelessly ambitious undertaking. For every topic included in the contents, a
thousand others could have been selected; for every discussion of a given topic,
a thousand others could have been presented, each drawing on different
material and with a different orientation. Though the overall conception of the
XIX
volume-including the definition of issues to be covered by individual articles,

their ordering and arrangement into parts-is my own, contributors have been
given a free hand to develop their ideas along whatever lines they find most
productive and congenial. The result is something of a pot-pourri of
approaches which, whilst they may accurately reflect the diversity of voices
currently to be heard within the discipline, hardly add up to any consistent
direction. So what possible justification can there be for collecting them all
together under the grandiose and all-encompassing rubric of an Encyclopedia?
To my mind, there are three good reasons for doing so.
The first, and most important, is to counteract the dangers of
overspecialization. One of the more worrying consequences of the exponential
growth in the volume of research and publication during the latter part of this
century is that we know more and more about less and less. It is hard enough for
any scholar to keep abreast of developments within a relatively narrow field, let
alone to follow what is going on in even closely related specialisms. What is lost, in
this process, is an awareness of the interconnectedness of phenomena, of their
positioning within wider fields of relationships. Knowledge is fragmented, its
objects treated in isolation from the contexts in which they occur. Yet it is only
thanks to our ability to connect that knowledge is rendered significant. Thus,
paradoxically, does the growth of knowledge breed ignorance, for the more we
know, the less we understand of what that knowledge means. Despite its holistic
aspirations, anthropology has suffered its own fragmentation, which some indeed
have we come as testimony to the rapt a vance of anthropo ogtca scho arshtp
in recent years, on a wide range of fronts. Gone are the days, it is said, when
anthropologists could read and contribute-as did the founders of the
discipline-across the entire spectrum of its concerns. I do not personally believe
this is the case, and if it is, I certainly do not welcome it. But there is no doubt that
the proliferation of interests and approaches threatens the coherence of
anthropology as a discipline, and that the need for integration and synthesis is
urgent. This volume exists to meet that need.
The second reason for an encyclopedic compilation of this kind is that it
serves to establish a baseline of anthropological knowledge upon which
subsequent generations can build. This is not merely to embark on a
stocktaking exercise, a survey of achievements to date in the various areas
covered. Indeed, little is to be gained from attempts to recapitulate or
paraphrase all that has been written on this or that topic: to do so leads at best
to the sterile rehearsal of obsolete arguments, at worst to the contrivance of
artificial 'schools of thought', each of which then becomes the subject of a
separate story. Contributors to this volume were asked not to write articles of
this sort, but were rather challenged to break new ground, not only by
presenting their own versions of the 'state of play' in their respective fields of
study, but by charting out new directions of inquiry hitherto unexplored. They
have, without exception, risen to the challenge, and the result is a volume that
takes anthropology beyond existing frontiers and that points unequivocally and
XX
sometimes provocatively towards the future. Many contributors, moreover, deal

with issues that lie on the evolving interface between anthropology and other
disciplines in the human sciences, from biology and psychology to linguistics,
history and sociology, and herein lies the third raison d'etre for the volume. For
besides bringing out the connections within the discipline of anthropology, the
articles collected here amply demonstrate the relevance of anthropological
insights to work in a host of related fields, and the capacity of the discipline to
build bridges across the frontiers between otherwise divided and mutually
impenetrable intellectual territories.
Let me conclude with a few words about what this Encyclopedia is not. I
have already pointed out that its subject is not anthropology but human life,
and that its orientation is to the future rather than the past. For this reason,
there are no articles dealing specifically with the history of anthropology. This
is not to say that no space is devoted to historical themes. However it has been
left to the discretion of individual contributors to dwell on the history of
approaches to the topical issues that concern them, in so far as it is conducive to
the elucidation of these issues themselves. The emphasis, in other words, is on
learning .from the history of the discipline rather than on learning about it. The
same goes for questions of anthropological research method. With the
reformulation of such questions as problems of 'methodology', they have
tended to become objects of investigation in their own right, rather than
questions whose resolution is but a means to the greater goal of enlarging
human un erstan m g. In this vo ume, matters of metho are not ma e m to the
subjects of separate articles, but are rather introduced where they belong, in the
context of inquiries into substantive anthropological topics. Finally, this is an
encyclopedia of anthropology, not of ethnography. It does not aim to catalogue
the range of human cultural variation, or to review the findings of
anthropological research in particular regions of the world. Each article has a
thematic rather than a regional focus, and authors have been free to draw on
illustrative material from whatever region or period best suits the purposes of
their exposition.
Though the volume qualifies as an encyclopedia, in that it encompasses the
full circle of current anthropological knowledge, it is also a book that is
designed not just to be consulted but to be read. While conceived as a work of
reference, its aims go far beyond that: namely to lay the foundations for an
integrated and synoptic perspective on the conditions of human life that is
appropriate to the challenges of the next century. For an encyclopedia, the
number of articles is relatively small, but by the same token, authors have had
the opportunity to develop their ideas and arguments at some length. Each
article, indeed, stands as a major contribution, an innovative synthesis at the
cutting edge of the discipline. Moreover, the ordering of articles is not
arbitrary, but has been carefully designed to bring out to best advantage the
connections between them, and to weld the volume into a coherent whole. The
resulting combination of breadth of coverage and depth in the treatment of
XXI
individual topics is, I believe, unparalleled in contemporary anthropological

literature.
I expect this book to be read primarily by students, teachers and academics
working in fields of anthropology or related disciplines, who need to turn to a
significant overview of current thinking to supplement their existing specialist
knowledge. But I hope it will also offer a source of ideas and inspiration to the
enthusiastic and informed 'general reader' who, once having encountered
anthropology, wishes to find out more about various aspects of the subject. To
cater for this wide readership, the articles are written so as to be both
authoritative and yet readily comprehensible to professionals, students and lay
persons alike. Each article is followed by a comprehensive list of references
detailing works cited in the text, and by a selected list of 'further reading'
recommended for those who wish to pursue the themes of the article in greater
depth. Naturally, there is often a good deal of overlap between items included
under 'further reading' and those listed in the references; however the costs of
duplication were felt to be outweighed by the advantages of presenting the
'further reading' as a single, integral list.
What lies ahead is a journey through some of the most exciting and
challenging domains of contemporary scholarship. I wish the reader bon voyage
while, with the merciful relief of a marathon completed, I lay down my own pen.
xxn
THE CONTRIBUTORS
BARBARA ADAM received her Ph.D. from the University of Wales, and is
currently Lecturer in Social Theory at the University of Wales, Cardiff. She is
the founder editor of the journal Time and Society. She has written extensively
on the subject of social time, and her book, Time and Social Theory (1990), won
the Philip Ab rams Memorial Prize in 1991, awarded by the British Sociological
Association for the best first book.
ALAN BARNARD is Senior Lecturer in Social Anthropology at the
University of Edinburgh. He was educated at the George Washington
University, McMaster University and the University of London, receiving his
Ph.D. in 1976. Before moving to Edinburgh in 1978, he taught at the
8). He has carried out fieldwork with the Nharo and other Khoisan peoples of
Botswana and Namibia and his research interests include kinship theory,
hunter-gatherer studies, and regional comparison. He is the author, with
Anthony Good, of Research Practices in the Study of Kinship (1984), and of
Hunters and Herders in Southern Africa (1992). He is co-editor of Kinship and
Cosmology (1989), and was editor of the journal Edinburgh Anthropology for
1988. Alan Barnard has published numerous articles on kinship, hunter-
gatherers and the Khoisan peoples of southern Africa, as well as a wordlist and
grammar of the Nharo language (1985) and a children's book on the Bushmen
(1993). His recent interests include the early history of social anthropology and
the relation between anthropology and popular literature.
NIKO BESNIER gained his Ph.D. from the University of Southern California
in 1986, and is now Associate Professor in the Department of Anthropology at
Yale University. He has conducted research in various locations in Western
Polynesia and Melanesia, principally on Nukulaelae Atoll, Tuvalu. His
published works deal with literacy, emotional life and the cultural construction
of the person, political rhetoric, gossip, and the relationship between verbal
accounts and social action.
ANDRE BETEILLE is Professor of Sociology at the University of Delhi
where he has taught since 1959. He was Simon Fellow at the University of
xxm
THE CONTRIBUTORS
Manchester (1965-6), Commonwealth Visiting Professor at the University of

Cambridge (1978-9), held the Tinbergen Chair at Erasmus University,
Rotterdam (1984), was Visiting Professor at the London School of Economics
(1986), Visiting Scholar in Residence at the University of California, Santa
Barbara (1988), Fulbright Distinguished Lecturer (1989) and a Fellow of the
Institute of Advanced Study, Berlin (1989-90). He has delivered the Auguste
Comte Memorial Lecture at the London School of Economics ( 1979), the
Kingsley Martin Memorial Lecture (1979) and the Commonwealth Lectures
(1985) at the University of Cambridge, and the Ambedkar Lectures at the
University of Bombay (1980). His research interests include stratification and
social class, equality and social justice, and race, caste and ethnicity. In addition
to many papers in scholarly journals, Beteille is the author of Caste, Class and
Power (1965), Castes: Old and New (1969), Studies in Agrarian Social Structure
(1974), Inequality among Men (1977), Ideologies and Intellectuals (1980), The
Idea of Natural Inequality and Other Essays (1983), Essays in Comparative
Sociology (1987), Society and Politics in India (1991) and The Backward Classes
in Contemporary India (1992). He is the editor of Social Inequality (1969) and
Equality and Inequality ( 1983).
MARK COHEN graduated from Harvard College and Columbia University
with degrees in anthropology. He has carried out archaeological fieldwork in
North, South and Central America, in southern Europe and in East Africa. He
Civilization ( 1989), and senior editor of Paleopathology at the Origins of

Agriculture (1984). He has been a Fellow of the Center for Advanced Study in
the Behavioral Sciences, Stanford University (1978-9), a Fellow of the John
Simon Guggenheim Memorial Foundation and Visiting Scholar at Cambridge
University (1985-6), and a Fulbright Lecturer at the Hebrew University,
Jerusalem (1989-90). Mark Cohen is currently Distinguished Teaching
Professor of Anthropology at the State University of New York College at
Plattsburgh, where he is working to reconstruct patterns of health in a
sixteenth century Christian May a population.
JEAN DEBERNARDI was educated at Stanford University and at the
University of Oxford, and received her Ph.D. from the University of Chicago
in 1986. She has carried out fieldwork in Penang, Malaysia ( 1979-81 ), and in
Taiwan and Fujian Province of the People's Republic of China (1987).
DeBernardi has taught at Beloit College, the University of Michigan and Bryn
Mawr College, and is presently Assistant Professor in the Department of
Anthropology, University of Alberta at Edmonton, Canada. Her research
interests include Chinese popular religious culture, the use of anti-languages in
Chinese secret societies, and linguistic nationalism in Taiwan. She is currently
working on a book entitled Empire over Imagination: Chinese Popular Religious
Culture in Colonial and Post-Colonial Malaysia.
XXIV
THE CONTRIBUTORS
ROBIN DUNBAR was educated at Oxford University and at the University of

Bristol, where he received his Ph.D. in 1974. He has subsequently held research
posts at Cambridge, Stockholm and Liverpool Universities. He is now
Professor of Biological Anthropology at University College London. His
research has been concerned mainly with the evolution of mammalian social
systems and has involved field studies of primates and ungulates in Africa and
Scotland. He is the author of Reproductive Decisions: An Economic Analysis of
Gelada Baboon Social Strategies (1984) and Primate Social Systems (1988), and
has published numerous articles in books and journals on themes in
primatology, sociobiology and human evolution.
TIMOTHY K.EARLE is Professor of Anthropology at the University of
California, Los Angeles. He received his Ph.D. in Anthropology from the
University of Michigan in 1973. His research interests include the evolution of
pre-industrial complex societies, institutional finance and prehistoric
economies. He has carried out field research on the Hawaiian Islands and in
Andean South America, and is presently involved in a long-term investigation
of Danish Neolithic and Bronze Age chiefdoms. Earle is the editor or co-editor
of a number of volumes, including Exchange Systems in Prehistory (1977),
Modeling Change in Prehistoric Subsistence Economies (1980), Contexts for
Prehistoric Exchange (1982) and Specialization, Exchange and Complex Society
(1987). He is the author of Economic and Social Organization of a Complex
The Evolution ofHuman Society (1987).

ROY ELLEN was educated at the London School of Economics, where he
received his B.Sc. in 1968, and his Ph.D. in 1972. He was Lecturer in Social
Anthropology at the LSE in 1972-3, and thereafter Lecturer, Senior Lecturer
and Reader at the University of Kent at Canterbury, England. Since 1988 he
has been Professor of Anthropology and Human Ecology at the University of
Kent. He has carried out several periods of social anthropological fieldwork
among the Nuaulu of Seram, Eastern Indonesia, as well as in Sulawesi and
Java, and in the Gorom archipelago. His interests include ecological
anthropology (especially of rain forest environments), regional organization of
trade, ethnobiology and classification, and anthropological research methods.
He is the author of Nuaulu Settlement and Ecology (1978), Environment,
Subsistence and System (1982) and The Cultural Relations of Classification (1993).
He has also edited or co-edited a number of collections, including Social and
Ecological Systems (1979), Classifications in their Social Context (1979),
Ethnographic Research (1984) and Malinowski Between Two Worlds (1988). He
was the Royal Anthropological Institute's Curl Lecturer for 1987.
CLIVE GAMBLE was born in 1951 and educated at Cambridge (MA and
Ph.D.). He is now Reader in Archaeology at the University of Southampton.
He has carried out extensive fieldwork in Palaeolithic Archaeology both in
XXV
THE CONTRIBUTORS
Europe and in Australia. He is the author of The Palaeolithic Settlement of

Europe (1986), Timewalkers: the Prehistory of Global Colonization (1993) and
(with C.B.Stringer) In Search of the Neanderthals (1993), as well as numerous
articles. He is editor, with O.Soffer, of The World at 18,000 B.P. (1990), and,
with WA.Boismier, of Ethnoarchaeological Approaches to Mobile Campsites
(1991). He has broadcast on the fate of the Neanderthals and lectured in many
countries on themes in Palaeolithic Archaeology.
IGOR DE GARINE was educated at the Sorbonne, Paris, and received his
Doctorate in Ethnology in 1962 for a thesis on the economic and social life of
the Massa of Cameroon. Since 1953 he has carried out fieldwork in numerous
countries of Africa (Cameroon, Chad, Senegal, Niger), South America (Brazil,
Argentina, Chile, Paraguay) and in South and South-east Asia (India, Nepal,
the Philippines). He has made eleven films and authored over a hundred
publications. Most recently, he was co-editor of Coping with Uncertainty in Food
Supply (1988) and Food and Nutrition in the African Rain Forest (1990). He
currently directs a 20-member research team of the Centre National de la
Recherche Scientifique, which is investigating the influence of seasonality on
food and nutrition in a number of African societies and in Nepal. In addition,
de Garine is Regional Commissioner for Europe of the International
Committee on the Anthropology of Food (ICAF), and Head of the Group on
the Anthropology of Food of the Maison Des Sciences de l'Homme, Paris. His
. . .
North Cameroon and Chad, and he has books in preparation on the social and
religious organizations of these peoples, as well as on theoretical and
methodological aspects of the anthropology of food.
CHRIS GREGORY studied Economics at the University of New South Wales
and the Australian National University. After spending three years lecturing in
Economics at the University of Papua New Guinea, where he developed an
interest in economic anthropology, he went on to obtain his doctorate in Social
Anthropology at the University of Cambridge in 1980. He subsequently
carried out anthropological fieldwork on rural marketing in a tribal area of
central India in 1982-3; his current ethnographic work in this region has
focused on material culture and mythology. Chris Gregory is currently Senior
Lecturer in the Department of Archaeology and Anthropology, Australian
National University. He is the author of Gifts and Commodities (1982) and (with
J.C.Altman) Observing the Economy (1989), as well as of numerous articles and
reviews.
TIM INGOLD is Professor of Social Anthropology at the University of
Manchester, England. After completing his first degree in Social Anthropology
at the University of Cambridge (1970), he carried out fieldwork among Saami
people in north-eastern Finland, leading to his Ph.D. and his first book, The
Skolt Lapps Today (1976). As well as pursuing this ethnographic interest, with
XXVI
THE CONTRIBUTORS
further fieldwork among northern Finnish farmers, he has written extensively

on comparative questions of hunting and pastoralism in the circumpolar North
(Hunters, Pastoralists and Ranchers, 1980), on evolutionary theory in
anthropology, biology and history (Evolution and Social Life, 1986), and on
human ecology (The Appropriation of Nature, 1986). He is a co-editor of the
two-volume work, Hunters and Gatherers ( 1988), and editor of What is an
Animal? (1988). His current research interests are in the anthropology of
technology (he is editor, with Kathleen Gibson, of Tools, Language and
Cognition in Human Evolution, 1993), and in issues of environmental
perception. Tim lngold was the Malinowski Memorial Lecturer for 1982, and
the Royal Anthropological Institute's Curl Lecturer for 1989. From 1990-2, he
was editor of Man (the Journal of the Royal Anthropological Institute). In 198 9
he was awarded the Rivers Memorial Medal.
STEPHEN KUNITZ is a physician and holds a Ph.D. in Sociology. He is
professor in the Department of Community and Preventitive Medicine at the
University of Rochester Medical Center, Rochester, NY, USA, as well as a
Visiting Fellow (part-time, 1990--4) at the National Centre for Epidemiology
and Population Health, Australian National University, Canberra. He has
carried out most of his fieldwork among Indian peoples of the American South-
west. Stephen Kunitz is author of Disease Change and the Role of Medicine: the
Navajo Experience (1983) and Diversity and Disease: the Impact of European
(1994). With J.E.Levy, he has also published Indian Drinking: Navajo Practices
andAnglo-American Theories (1974), Navajo Aging (1991) and Navajo Drinking
Careers: a Twenty-five Year Follow-up (1994 ), as well as several articles on
population and health among the Navajo and Hopi Indians.
MARY LECRON FOSTER is a full-time researcher at the Department of
Anthropology, University of California, Berkeley. She received her Ph.D. in
linguistics from that University in 1965, and has initiated and taught
programmes in social and cultural linguistics and in symbolic anthropology at
California State University, Hayward. She has carried out extensive research in
Mexico on indigenous languages and cultures, and has published grammars of
two Mexican languages and many articles on language evolution and aspects of
cultural symbolism. She is editor (with Stanley H.Brandes) of Symbol as Sense
(1980), and (with Lucy J.Botscharow) of The Life of Symbols (1990).
GILBERT LEWIS initially studied Medicine at the Universities of Oxford
and London, and held hospital medical posts in London between 1962 and
1965. He then went on to study Social Anthropology, and was Research Officer
at the London School of Economics from 1967-71. Following anthropological
fieldwork in West Sepik Province, Papua New Guinea, in 1968-70, Lewis
gained a Ph.D. in 1972. He is presently Lecturer in Social Anthropology at the
University of Cambridge. His books include Knowledge of Illness in a Sepik
xxvn
THE CONTRIBUTORS
Society (1975) and Day of Shining Red (1980). Lewis is currently engaged in
medical anthropological research in West Africa (Guinea-Bissau).
PHILIP LIEBERMAN received his initial training in Electrical Engineering,
at the Massachusetts Institute ofTechnology, but his Ph. D., completed in 1966,
was in Linguistics. He was Associate Professor in Linguistics and Electrical
Engineering, and subsequently Professor of Linguistics, at the University of
Connecticut, from 1967-74. In 1974, he joined the Faculty at Brown
University, where he is now the George Hazard Crooker University Professor
and Professor of Cognitive and Linguistic Sciences. He has carried out
research on the perception and production of speech, and on the innate
foundations and evolution of human linguistic competence. He is the author of
Intonation, Speech and Language (1967), The Speech of Primates (1972), On the
Origins of Language (1975), Speech Acoustics and Perception (1976), The Biology
and Evolution of Language (1984) and Uniquely Human: The Evolution of
Speech, Thought and Selfless Behavior (1991), as well as of numerous articles and
reviews.
DANIEL MILLER studied archaeology at the University of Cambridge, and
received his Ph.D. in 1983. Since 1981 he has worked in the Department of
Anthropology, University College London, where he is now Reader in
Anthropology. His principal research interests are in material culture and mass
consumption, and he has carried out fieldwork in Indonesia, the Solomon
Is an s, In 1a, Lon on an , most recent y, Tnm a . He IS the author of
Artefocts as Categories ( 1985), Material Culture and Mass Consumption ( 1987)
and Modernity: an Ethnographic Aproach (1994). He is editor, with C.Tilley, of
Ideology and Power in Prehistory ( 1984 ), and with M.Rowlands and C. Tilley, of
Domination and Resistance (1989). He is also sole editor of Unwrapping
Christmas (1993).
HENRIETTA MOORE was educated at the University of Cambridge,
receiving her Ph.D. in 1983. She has taught at the Universities of Kent and
Cambridge, and is currently Reader in Anthropology at the London School of
Economics. She has carried out fieldwork in Kenya, Burkino Faso, Sierra
Leone and Zambia, and her major research interests include economic
anthropology and development, gender and feminist studies, and
contemporary issues in anthropological theory. She is the author of Space, Text
and Gender: An Anthropological Analysis of the Marakwet of Kenya (1986) and
Feminism and Anthropology ( 1988), as well as of numerous articles in books and
journals.
HOWARD MORPHY studied Anthropology at the University of London and
at the Australian National University, receiving his Ph.D. in 1977. He was
Lecturer and subsequently Senior Lecturer in the Department of Prehistory
and Anthropology, Australian National University, from 1978-86. He is
currently Curator of Anthropology at the Pitt-Rivers Museum and Lecturer in
xxvm
THE CONTRIBUTORS
Ethnology at the University of Oxford. He has carried out anthropological

fieldwork over many years among the Yolngu, an Aboriginal people of Arnhem
Land, northern Australia, and has collaborated with the film-maker lan
Dunlop on several films dealing with art and religion in Arnhem Land,
including the Narritjin Series and a film ethnography entitled Djungguwan. He
is the author of Journey to the Crocodile's Nest ( 1984) and Ancestral Connections
(1991), and editor of Animals into Art (1989). He has also published extensively
on themes in the anthropology of art and material culture, visual anthropology,
and Australian Aboriginal ethnography. Howard Morphy was awarded the
Stanner Prize for Aboriginal Studies in 1985 and 1992, and the J.B. Donne
Prize for the Anthropology of Art in 1988. He was Malinowski Memorial
Lecturer for 1993.
JOHN ODLING-SMEE holds a joint lectureship in the Departments of
Biology and Biochemistry and of Human Sciences at Brunei University, West
London. After taking his first degree in Psychology, and a subsequent period
spent studying animal learning at the Medical Research Council's Neural
Mechanisms Unit, both at University College London, Odling-Smee received
his Ph.D. in 1973. He has worked extensively on animal behaviour and
evolutionary theory, and has published numerous articles, many of them in
collaboration with Henry C.Plotkin, in Behavioral and Brain Sciences, Advances
in the Study of Behaviour, and in the volumes (edited by H.C.Plotkin) Learning,
In 1993 he was appointed to a five-year Research Fellowship at the Institute of

Biological Anthropology, University of Oxford.
SUTTI ORTIZ studied at the University of London and received her Ph.D.
from the London School of Economics in 1963. She is currently Associate
Professor of Anthropology at Boston University. She has carried out several
periods of fieldwork in Colombia, and has recently completed a study of labour
markets in Colombian coffee production. She is the author of Uncertainties in
Peasant Farming (1973), and of numerous articles on themes in economic
anthropology and development, agrarian change, peasant sociopolitical
organization and decision making, and rural labour markets. Sutti Ortiz is
former president of the Society for Economic Anthropology, and editor of the
proceedings of its first conference, Economic Anthropology: Topics and Theories
(1983). With Susan Lees, she has also edited the proceedings of the Society's
decennial conference, Economy as Process (1992).
FITZ JOHN PORTER POOLE is Associate Professor of Anthropology at the
University of California, San Diego. His undergraduate studies were
undertaken at several institutions in the United States and Europe, and were
completed at New York University with a BA in Anthropology, Biology and
Philosophy. He received his doctorate in Anthropology and Social Psychology
from Cornell University in 1976, and went on to hold a position in the
XXIX
THE CONTRIBUTORS
Department of Anthropology at the University of Rochester. His theoretical

interests have focused primarily on the confluence of socialization and
enculturation, the formation of identity, gender and ritual experience. His
ethnographic research among the Bimin-Kuskusmin of Papua New Guinea is
represented in numerous articles and in a forthcoming book, The Rites of
Childhood.
ANN JAMES PREMACK graduated from the University of Minnesota, and
is now a distinguished science writer whose work has appeared in many
magazines and journals, including Scientific American, La Recherche, La Debat,
National Geographic and Geo. Her book, Why Chimps Can Read (1975), has
appeared in four foreign-language translations. With David Premack, she has
published The Mind of an Ape (1983). She is also editor, with David Premack
and Dan Sperber, of Causal Understanding in Cognition and Culture, to be
published in 1994.
DAVID PREMACK gained his Ph.D. from the University of Minnesota in
1955. He taught at the University of Missouri, Columbia, from 1956 to 1964
and at the University of California, Santa Barbara, from 1965 to 1975. Since
1975 he has been Professor of Psychology at the University of Pennsylvania,
and is currently affiliated to the Laboratoire de Psycho-Biologie de l'Enfant in
Paris. He has also held visiting posts at Harvard University, the Van Leer
Jerusalem Institute and the Institute for Advanced Study, Berlin, and is a
recipient of the lnternatwna Pnze of the Fyssen Foun atwn. Davi Premack
was one of the pioneers of experimental research on the linguistic capabilities
of great apes, and has written extensively on comparative questions of ape and
human language, learning and intelligence. His more recent work has been
concerned with acquisition of 'theories of mind' and related problems of
cognition. He is the author of Intelligence in Ape and Man (1976) and Gavagai!
Or the Future History of the Animal Language Controversy (1986), as well as
numerous articles in books and journals.
AMOS RAPOPORT has taught at the University of Melbourne, the
University of California at Berkeley, University College London and the
University of Sydney. Since 1972 he has been at the University of Wisconsin at
Milwaukee, first jointly in Architecture and Anthropology and subsequently,
from 1979, as Distinguished Professor of Architecture in the School of
Architecture and Urban Planning. He is one of the founders of Environment-
Behaviour Studies, and is the author of over two hundred articles in journals
and edited collections. Amos Rapoport's books include House Form and Culture
(1969), Human Aspects of Urban Form (1977), The Meaning of the Built
Environment (1982, revised edition 1990) and History and Precedent in
Environmental Design ( 1990). He has also edited or co-edited a number of
collections, including The Mutual Interaction of People and their Built
Environment (1976), Environment and Culture (1980) and Human and Energy
XXX
THE CONTRIBUTORS
Factors in Urban Planning (1982). His work has been translated into many
languages, including French, Spanish, Greek, Chinese, Korean and Japanese.
SIMON ROBERTS studied at the London School of Economics, and first
taught at the Law School, Institute of Public Administration, near Blantyre,
Malawi. He subsequently carried out two years of field research in Botswana.
He is currently Professor of Law at the London School of Economics, and
General Editor of The Modern Law Review. He is the author or co-author of
Order and Dispute ( 1979), Rules and Processes ( 1981) and Understanding Property
Law (1987). He is presently engaged in a field study of a London divorce
counselling and mediation agency.
ROBERT A.RUBINSTEIN received his Ph. D. in Anthropology from the State
University of New York at Binghamton. He also holds a Master's degree in
public health from the University of Illinois at Chicago. He has carried out
research in Egypt, Belize, Mexico, and in the United States. Robert Rubinstein
is currently Associate Research Medical Anthropologist with the Francis
!.Proctor Foundation at the University of California, San Francisco. He is eo-
chair of the Commission on the Study of Peace of the International Union of
Anthropological and Ethnological Sciences. He is the editor or co-editor of
Epistemology and Process (1984), Peace and War (1986), The Social Dynamics of
Peace and Conflict (1988) and Fieldwork: The Correspondence of Robert Redfield
and Sol Tax (1991). With C.D.Laughlin and J.McManus, he is a co-author of
Science as ognitive Process 984 .
RICHARD SCHECHNER holds a Chair at New York University and also
teaches in the Performance Studies Department at the Tisch School of the
Arts. He is editor of TDR: A Journal of Performance Studies. In 1967,
Schechner founded The Performance Group with whom he has directed many
plays, including Dionysus in 69, Mother Courage and her Children, Oedipus, The
Tooth of Crime and The Balcony. Since 1980, Schechner has staged works and
conducted performance workshops in several countries of Europe, Asia, Africa
and the Americas. In 1992 he founded and became artistic director of the East
Coast Artists, for which he directed Faust!Gastronome. Schechner's many
books include Environmental Theater (1973), Between Theater and Anthropology
(1985), Performance Theory (1988) and The Future of Ritual (1993). Schechner
is also editor of various books including (with M.Schuman) Ritual, Play and
Performance (1976) and (with WAppel) By Means of Performance (1990). He is
General Editor of the book series, Worlds of Performance. He has been awarded
numerous Fellowships and Prizes, and has lectured and taught at Princeton
University, Florida State University, Ball State University, and the School of
Art Institute of Chicago.
ANTHONY SEEGER received his BA in Social Relations from Harvard
University and his MA and Ph.D. in Anthropology from the University of
Chicago. His research has focused on the social organization, cosmology and
XXXI
THE CONTRIBUTORS
music of Brazilian Indians as well as on issues of land and cultural equity. He is

the author of Nature and Society in Central Brazil (1981), Why Suyd Sing
(1987), two other books and many articles and reviews. He has taught at the
Museu Nacional in Rio de Janeiro (1975-82), and at Indiana University where
he also served as Director of the Archives of Traditional Music (1982-8).
Anthony Seeger is currently Curator of the Folkways Collection and Director
of the Smithsonian Institution in Washington, DC, in connection with which
he has directed the production of over fifty audio recordings.
FRAN<;OIS SIGAUT was first trained in agronomy and received his degree
from the lnstitut National Agronomique, Paris, in 1964. After a few years of
professional work, he turned to the social sciences and received his Ph.D. in
Ethnology in 197 5 for a dissertation on the relation between agriculture and
fire. His main field of research remains in the technology of pre-industrial
European agriculture. Since 1985, his interests have widened to include the
basic concepts and methods of technology, understood as the study of technics
considered as an integral part of social life. He joined the Ecole des Hautes
Etudes en Sciences Sociales, Paris, in 1978, where he is now Professor. He has
co-edited four volumes on Les Techniques de Conservation des Grains (1979-85),
has acted as guest editor for two issues of the journal Techniques et Culture, and
is the author of numerous publications on technological, ethnological and
historical topics.
ANTH NY D.SMITH receive his rst egree m assics an Phi osophy
from the University of Oxford, going on to take a Master's Degree and Ph.D.
in Sociology at the London School of Economics. He has taught at the
Universities ofYork and Reading, and is currently Professor of Sociology at the
London School of Economics. His many books include Theories of Nationalism
(1971, 1983), The Concept of Social Change (1973), Social Change (1976),
Nationalism in the Twentieth Century (1979), The Ethnic Revival (1981), State
and Nation in the Third World (1983), The Ethnic Origin of Nations (1986) and
National Identity (1991). He is the editor of Nationalist Movements (1976). His
main interests are in ethnic identity and nationalism, but he also holds a
Doctorate in the History of Art and maintains a further interest in the sociology
of development. He is presently working on a study of the revival of ethno-
nationalism in a global era.
BRIAN STREET is Senior Lecturer in Social Anthropology at the University
of Sussex, England. After undertaking anthropological fieldwork in Iran during
the 1970s, he has written and lectured extensively on literacy practices from
both a theoretical and an applied perspective. He is best known for Literacy in
Theory and Practice (1984); more recent works include an edited volume on
Cross-Cultural Approaches to Literacy ( 1993) and a forthcoming collection of
articles entitled Social Literacies. He has been active in applying theory to
practice, both in relation to adult literacy programmes in the UK and in
xxxn
THE CONTRIBUTORS
connection with development programmes. His latest research involves the

investigation of everyday writing practices in the UK, focusing on self-selected
adults who have written responses to questions on social and personal issues for
the Mass-Observation Archive at the University of Sussex. Other research
interests have included popular and literary representations of 'other' peoples.
His study, The Savage in the Literature, appeared in 1975.
PHILLIP TO BIAS has recently retired from the headship of the Department
of Anatomy and Human Biology at the University of the Witwatersrand,
Johannesburg, South Africa, having held the position for thirty-two years. He
has been elected to two Honorary Professorships at the same University, in
Palaeoanthropology and in Zoology. His researches have focused mainly on the
present and past inhabitants of Africa and he has written or edited nineteen
books and published approximately 850 articles. Since 1959 he has been
responsible for the description and evaluation of the fossil hominid remains
from Olduvai Gorge and other sites in Tanzania and Kenya. With L.
S.B.Leakey and J.R.Napier, he recognized and named a new hominid species,
Homo habilis, in 1964; the appraisal of Homo habilis is the subject of his 2-
volume work on the species, published in 1991. He has been in charge of
excavations at two famous South African sites of early hominid discoveries,
Sterkfontein (since 1966) and Taung (since the 1980s), and at the former site
his team has brought to light some 550 new specimens of Australopithecus
Professor Tobias, and in 1987 he was awarded the Balzan International Prize
(the first time this prize had been given for achievements in Physical
Anthropology).
JAMES WEINER studied at the University of Chicago and at the Australian
National University, and gained his Ph.D. from the ANU in 1984. He held a
Lectureship in the Department of Prehistory and Anthropology, Australian
National University, from 1986-9, and from 1990 has been Lecturer in Social
Anthropology at the University of Manchester, England. Weiner has carried
out a total of two and a half years of fieldwork among the Foi people of Papua
New Guinea, on the basis of which he has published many articles and two
books, The Heart of the Pearlshell (1988) and The Empty Place (1991). He is also
the editor of Mountain Papuans (1988). He is currently working on a book on
Australian and Papuan myth, and is carrying out further work on Foi
Christianity.
PETER WORSLEY obtained his first degree in Anthropology and
Archaeology at the University of Cambridge in 194 7, and went on to read for
an MA in Social Anthropology at the University of Manchester. From 1956-64
he taught in the Department of Sociology at the University of Hull, and from
1964 until his retirement in 1982 he was Professor of Sociology at the
University of Manchester. He is now Professor Emeritus at that University. In
xxxm
THE CONTRIBUTORS
1955 Worsley received the Curl Bequest Essay Prize of the Royal
Anthropological Institute, and from 1971-4 he was President of the British
Sociological Association. He is the author of many books, including The
Trumpet Shall Sound (1957), The Third World (1964), Introducing Sociology
(1970, 1977), Marx and Marxism (1982) and The Three Worlds: Culture and
World Development (1984).
THOMAS WYNN received his Ph.D. in anthropology at the University of
Illinois, Urbana, in 1977. His research has centred on the application of theory
in developmental psychology, especially that of Jean Pia get, to the
archaeological record of prehistoric artefacts. He has recently published The
Evolution of Spatial Competence ( 1989), and has written numerous articles for
professional journals, including Journal of Human Evolution, World
Archaeology, Man, Cambridge Archaeological Journal, Journal ofAnthropological
Archaeology and American Anthropologist. He is currently Professor of
Anthropology, University of Colorado at Colorado Springs, USA.
XXXIV
3 SOCIAL LIFE
26
INTRODUCTION TO SOCIAL LIFE

Tim lngold
QUESTIONS OF SOCIALITY
Wherever people live, as they generally do, in the company of others, and act
with those others in mind, their mode of life may be called social. Questions
about social life have therefore to do with elucidating the dynamic properties of
human relationships, properties conveyed by such stock-in-trade
anthropological notions as kinship, exchange, power and domination. We may
ask how these features of human sociality are generated, maintained and
manage ; ow t ey are tmp tcate m t e 1 e- tstory o t e m IVI ua rom
childhood to old age; how they are represented and communicated in discourse;
how they structure-and are in turn structured by-the production and
consumption of material goods, and how they underwrite (or subvert) diverse
forms of moral or political order. These are the kinds of questions addressed in
the chapters making up the third part of this volume. They can, of course, be
posed on any number of different levels, from the minutiae of everyday life in
familiar contexts of face-to-face interaction to the trials and endeavours of
whole populations on a world-historical stage. Likewise the temporal scale on
which social processes are viewed may range from within a lifetime to the entire
span of human history. It is important to remember, however, that it is the
perspective of the observer that 'selects' the scale of the social phenomena
observed. When anthropologists claim that they generally study small-scale
societies rather than large-scale ones, this is not because the world of humanity
is objectively partitioned into social units of diverse size, of which the smaller
lend themselves more readily to anthropological investigation, but because
from a locally centred perspective, the horizons of the social field appear
relatively circumscribed.
The concept of society, moreover, is by no means neutral, but trails in its
wake a long history of controversy among Western philosophers, reformers and
statesmen about the proper exercise of human rights and responsibilities. In
this controversy, the meaning of 'society' has varied according to the contexts
737
SOCIAL LIFE
of its opposition, alternately, to such notions as individual, community and

state. Unlike the 'individual', society connotes a domain of external
regulation-identified either with the state itself or, in polities lacking
centralized administration, with comparable regulative institutions-serving
to curb the spontaneous expression of individual interests on behalf of higher
ideals of collective justice and harmony. In other contexts, however, society
comes to represent the power of the people-as a real or imagined community
bound by shared history, language and sentiment-against the impersonal and
bureaucratic forces of the state. And in yet other contexts, society stands opposed
to 'community', connoting the mode of association of rational beings bound by
contracts of mutual self-interest, as epitomized by the market, rather than by
particularistic ties of the kind epitomized by kinship relations. What is clear
from this diversity of usage is that the term 'society', far from forming part of a
value-free language of description, in fact belongs to a language of argument.
To use it is, inevitably, to make a claim about the world.
One further opposition, which has proved especially troublesome for
anthropology, is between society and 'culture'. So long as society could be
regarded as an association of individuals, and culture as the sum of their
knowledge, acquired by traditional transmission and imported into contexts of
interaction, this distinction seemed straightforward enough. Indeed it served
for much of the twentieth century as the rationale for a division between two
largely autonomous branches of anthropological inquiry, 'social' and 'cultural',
w ose mte ectua omes ay respective y m Bntam an Nort menca. In
recent years, however, this division has come to be seen as increasingly artificial.
The reasons are various, but at the most fundamental level, they are bound up
with a general rejection of what is known as an essentialist viewpoint -that is,
one that would treat societies and cultures as real entities 'out there', presenting
themselves to anthropological observation as objects to be described, compared
and classified. Contemporary anthropology veers more to a process-oriented
view, according to which cultural form does not come ready-made, like a suit of
clothing to cover the nakedness of the 'biological' individual, but is perpetually
under construction within the contexts of people's practical engagements with
one another. All culture, then, is social, in that its constituent meanings are
drawn from the relational contexts of such mutual involvement; conversely all
social life is cultural, since people's relationships with one another are
informed by meaning. In short, culture and social life appear to be caught in an
ongoing dialectic in which each, in a sense, 'constitutes' the other, through the
mediation of human agency.
As the emphasis has shifted from the study of societies as things to the study
of social life as process, anthropologists have begun to pose their questions in a
rather different way. Instead of asking 'Why do different societies take the
forms they do?', they are presently more inclined to ask 'What is it about a form
oflife that makes it social?'. They have moved, in other words, from questions
about society to questions about sociality. What, they ask, is necessary for there
738
to be social life at all? Recent discussions have thrown up three kinds of

answers. The first seeks the roots of sociality in some innate biological
endowment, an evolved predisposition to associate that will naturally be
expressed in varying manner and degree, depending upon the prevailing
conditions of the environment. For proponents of this view, sociality is by no
means confined to human beings, or even to the order of primates, but extends
right across the animal kingdom. The second answer is to identify sociality with
moral accountability; that is with the explicit recognition of rules and standards
by which people judge their own and others' actions. Insofar as the articulation
of these rules and standards depends upon a capacity for language, sociality in
this sense is generally attributed uniquely to human beings. The third kind of
answer locates the essence of sociality neither in individually held dispositions
nor in collective rules, but in the relationships that bind people together as fellow
participants in a life-process. To grasp the significance of this answer, however,
it is necessary to examine the notion of 'relationship' a little more closely. In
particular, we need to reconsider the dichotomy, which keeps cropping up
especially in discussions of kinship and gender, between social and biological
domains of relationship. This emerges as a central theme of Articles 27, 28
and 29.
SOCIAL AND BIOLOGICAL RELATIONSHIPS

Dun ar, m rttc e 7, wntes as a stu ent o amma e avwur wtt an mttmate
knowledge of the elaborate social intrigues typical of everyday life in
populations of non-human primates. He leaves us in no doubt that monkeys
and apes, like human beings, are caught up in complex networks of relationship
with others of their kind. Any one individual may indeed be simultaneously
involved in several different networks. Yet such networks, for all their
complexity, and for all the intensity of their constituent relationships, do not
imply the existence of large or stable groupings. The company an animal keeps
may be highly selective, and may vary from one moment to the next according
to a host of situationally specific factors. Thus to be bound by relationships is
not at all the same thing as to live in a group. In many animal species, including
the so-called 'social' insects, birds and mammals, individuals cluster into
aggregates of a size and permanence without parallel in the primate order, prior
to a relatively late phase in the evolution of human society, yet within these
aggregates there may be no relationships to speak of at all. Thus, explanations
of why animals live in groups, for example in terms of the facilitation of co-
operative foraging, defence of food resources or protection against predators,
do not in themselves account for the presence and quality of relationships
among their members. How then, are they to be explained? What is the source
from which relationships spring?
To these questions, as Dunbar remarks, biological and social anthropologists
are inclined to come up with rather different answers. One reason for this lies
739
SOCIAL LIFE
in a certain disparity between their respective views of what a relationship 15,

and of how it is to be recognized. Social anthropologists discover relationships
in the commitments and promises that people make towards one another, in
their agreements and obligations, which they do not always live up to in actual
practice. For biological anthropologists, on the other hand, the existence of a
relationship is a matter of empirical observation, whenever it is found that the
same individuals interact on numerous successive occasions, evidently with a
knowledge, based on past experience, of each other's identity and character.
Consequently, whereas the generative source of relationships lies, from a social
anthropological point of view, in the dynamics of total social systems, from the
viewpoint of biological anthropology it lies in the behavioural predispositions
of individuals. Though these viewpoints need not be mutually exclusive, the
former looks from the 'top down', the latter from the 'bottom up'. And this
difference in perspective may be linked directly to the fact that each is situated
within a different framework of interpretation. In the first, the challenge is to
'make sense' of people's behaviour by placing it in its social context of shared
cultural understandings. The second, by contrast, seeks an underlying
rationale for social behaviour in terms of its consequences for the survival and
reproduction of the individuals concerned, regardless of what construction-if
any-they may place on it.
This difference of approach is well illustrated by what Barnard, in Article
28, and Dun bar, in Article 27, have to say about the nature and significance of
re atwns o ms Ip. Barnar s positiOn IS m c ose accor Wit
mainstream social anthropology. Kinship, he argues, is not a fact of nature but
is rather constituted within a specifically human discourse on social
relationships. This discourse typically includes ideas about the sharing of
bodily substance, as conceived within the indigenous theory of procreation. In
Western societies, the substance of kinship has commonly been identified with
blood (as in the notion of consanguinity), though nowadays this is giving way to
a pseudo-scientific conception of genetic material. When people in these
societies say that they are of one blood, or that they have inherited the same
genes, their statements should be understood not literally, as having reference
to a given, 'biological' reality, but rather metaphorically, as ways of talking
about an experienced, social reality. In other words, kinship is 'biological' only
insofar as 'biology' enters into the vernacular discourse on social relations. And
it is in this light, too, that a social anthropologist would interpret the kinds of
comments that people make, probably in all societies, about the appearance of
children. A child may be said to resemble this parent or that, or to possess
features that it has 'received' from certain more distant relatives. The purpose
of such commentary, however, is not to discover evidence of actual genetic
connection but to place the child, and confer upon it a specific identity, within a
nexus of social relationships.
Yet Dunbar, resting his argument on the premisses of evolutionary biology,
reaches precisely the opposite conclusion! His objective is to show how
740
particular patterns and modes of relationship may have evolved through a

process of variation under natural selection. To achieve this it is necessary to
suppose, first, that manifest social behaviour is the output of a programme that
every individual brings into its encounters with conspecifics, and second, that
the constituent elements of this programme-commonly identified with
genes-are replicable across generations. The mechanism of genetic replication
is assumed to be sexual reproduction, though as Dunbar recognizes, the
relation between 'genes', as units of heredity conceived in the abstract, and the
actual genetic material in the chromosomes, is far from clear. According to the
logic of natural selection, any behaviour that has the effect of increasing the
representation, in future generations, of those genes of which it is the outward
expression, will tend to become established. To demonstrate that a behaviour
has this effect, by conferring a reproductive advantage on those who engage in
it, is sufficient to account for its evolution. But granted that an animal may
derive some benefit from its association with conspecifics, why should it choose
to establish relationships only with some particular individuals, while others are
avoided?
Sociobiological theory explains this selectivity, in part, on the grounds that,
depending on their genealogical proximity, individuals will have a greater or
lesser proportion of their genes in common. Hence the consequences of
associating with a close relative, in terms of genetic replication, will differ from
those of associating with a distant relative or a non-relative. However, if an
m IVI ua IS to associate pre erentia y Wit re atiVes, t en It must ave some
mechanism for their identification. The perception of physical resemblances,
according to Dunbar, furnishes just such a mechanism, and in the comments
that people habitually make about children's likenesses to their elders ('Doesn't
he have grandma's nose!'), we see it in operation. If this argument is correct,
then comments of this kind, quite contrary to the social anthropological
interpretation offered by Barnard, do have a forensic purpose: they are meant to
draw attention to actual genetic connections. Such connections, virtually by
definition, exist independently of, and prior to, any relationships at all. Thus
whatever the people themselves may claim, they are predisposed to attend to
physical resemblances not for what they reveal about the relationships within
which children come into being as members of society, but for guidance on
where to place their investments in succeeding generations. In other words, the
configuration of social relations follows from-rather than provides a context
for-the recognition of physical affinity.
What, then, makes a relationship a kinship relationship? Is it merely a matter
of the choice of idiom in which people talk about it, or does kinship have an
independent foundation in genetic relatedness? For the biological
anthropologist, actual genetic connections are critical, since the evolutionary
rationale for kin-based altruism rests on the presumption that individuals
identified as potential beneficiaries do in fact share a substantial proportion of
the altruist's genes. To the objection, commonly levelled by social
741
SOCIAL LIFE
anthropologists, that cultural designations of kinship are arbitrarily

superimposed upon genetic realities, the biologists' response is that their
theory requires no more than a statistical correlation. So long as there is
sufficient overlap between culturally perceived kinship and true genetic
kinship, the theory should work. Social anthropologists, for their part, while
not denying the facts of genetic connection, exclude them from their field of
inquiry. Their concern, they say, is with the ways in which certain relationships
come to be 'culturally constructed' as relations of kinship by virtue of their
grounding in an indigenous biology of shared substance. If genetics figures at
all in their discussions, it is as one particular instance of such a biology, namely
that of the Western biomedical establishment. To what extent this biology is
scientifically more 'correct' than others is not for them to judge.
Ask a social anthropologist to describe a relation of kinship, for some
particular society, and the answer-as Barnard shows in Article 28-is likely to
come in two parts. Consider, for example, the relationship between a father and
his son. First, there is a set of expectations surrounding the proper
performance of fatherhood, as there is attached to being a good son; in this
sense 'father' and 'son' are roles to be enacted, and the relationship between
them is inscribed within the framework of normative orientations of the society
in question. This relationship is said (by the anthropological analyst) to be one
of 'social' kinship. Second, members of the society claim that father and son are
linked by a bond of substance, by virtue of the father's material contribution to
t e ormatwn o t e o y o IS c 1 . T IS s anng o su stance IS sat agam
by the analyst) to be constitutive of a relation of 'biological' kinship. Armed
with this distinction between the social and biological components of paternity,
social anthropologists have gone on to draw attention to cases where one
component can occur without the other: where a man extends fatherhood
towards children who are not thought to share in his bodily substance; or
denies it to children who are. None of this, however, has anything to do with
actual genetic connection. Whether the individual who is socially identified as
the donor of paternal substance is or is not the same as the true genetic father is
irrelevant for social anthropological analysis. For the aim of such analysis is to
understand the concordance between social and biological kinship as culturally
perceived.
There is a remarkable parallel between the history of the biological/social
distinction in the study of kinship relations, and that of the sex/ gender
distinction in the study of relations between men and women. In both cases, the
distinction was drawn initially in order to emphasize the independence of
socially defined role relationships from underlying biogenetic constraint. Just
as the roles of 'father' and 'son' were held to have nothing to do with the genetic
connection-if any-between their incumbents, so the expectations defining
what it is to be a 'man' or a 'woman' were shown to vary widely from one society
to another, lending support to the view that they are independent of innate
predisposition. For any society, the distinction of gender-between men and
742
women-was said to belong to a normative order, superimposed upon a given

biological substrate of male/ female sex differences. While sex as biological fact
was thus eliminated from the field of gender relations, it reappeared as a focus
of social anthropological attention in quite another guise, as part of a discourse
for talking about them. The need therefore arose to introduce a further analytic
distinction, between sex as a fact of nature-a prerequisite for physiological
reproduction-and 'Sex' as constituted within the cultural discourse on
gender relations. The result, as Moore shows in Article 29, is the separation of
sex, 'Sex' and gender; a trichotomy that has its precise counterpart in the field
of kinship studies in the separation of biology as actual genetic connection,
'biology' as indigenous cultural discourse, and the structure of role relations on
which this discourse comments.
This solution, though neat, is not entirely satisfactory. The problem lies in
the very notion that the orders of gender and kinship are socially or culturally
constructed upon the foundation of biological reality. Critics have pointed out
that the recognition of sex differences as pre-existing in nature, independently
of the constructions placed on them, belongs to a specifically Western ontology.
In other words the distinction between sex and gender, as constituted
respectively within the domains of nature and society, is itself the product of a
particular set of discursive practices. By incorporating the distinction into its
own theoretical apparatus, social anthropology has taken for granted what it
should be seeking to explain. By and large, people in non-Western societies do
not ma e t IS m o tstmctwn. Far rom regar mg sexua 1 entities as rea y-
made, as though every new-born child came into the world completely and
unalterably male or female, they hold that these identities are fashioned within
life-cycle processes through the exchange and ingestion of male and female
substances. Likewise, they would not accept the distinction, built into the
framework of the anthropological analysis of kinship, between its biological and
social components. Contrary to Western genetics, they would argue that the
contributions of paternal and maternal substance that go to make up the body
of a child are themselves delivered within the context of an ongoing set of
nurturing relationships. In short, as they proceed through life, human beings
are thought to incorporate into their very biological constitution the matrix of
relationships that, at the same time, furnishes their identities as social persons.
It is tempting, as many anthropologists have done, to accommodate these
non-Western views by regarding them as so many cultural constructions of
reality, alternative to the Western one. This, however, will not do, for the simple
reason that the Western ontology-which yields the distinctions both between
sex and gender and between biological and social kinship-also underwrites
the logic of cultural construction itself. Applying this logic, what are
constituted within social processes are 'meanings' and 'understandings', that
are added on to bodies that have been biogenetically pre-programmed in
advance of their entry into the social arena. Through the reduction of 'biology'
to genetics, human relationships are withdrawn from the real world in which
743
SOCIAL LIFE
people dwell, as a preliminary to their reinscription on the level of its cultural

representation. Thus individuals are perceived to exist in hermetic isolation,
while relationships exist in the discourse. In this division between the
discursive worlds of culture and naturally given, biogenetic reality, no
conceptual space remains for the domain in which human beings live their lives
through an active engagement with those around them. The relationships
constitutive of this domain are indeed social, but they are no less 'real' or
'biological' for that. For in truth, no more than other animals do human beings
come biologically ready-made, to be 'topped up' by culture. They rather
emerge with their peculiar capacities, dispositions and intentions in the course
of a process which, in the literature of biology and psychology, goes by the
name of development. In this process, the contributions of other persons in the
social environment are critical. And as Poole shows in Article 30, the rather
belated recognition that human development is invariably embedded in
contexts of social relations requires us to take a fresh look at the time-worn
concepts of socialization and enculturation.
BECOMING A PERSON
Traditionally, the project of social anthropology has been distinguished from
that of psychology in terms of a distinction between the individual and the
person. In this division of intellectual labour, the nature of individual self-
awareness, postte as a uman umversa , was to e stu te y psyc o ogtsts,
while anthropologists focused on the person as a social being, formed within
the normative framework of society and its relationships. Having thereby
excluded the self as an aspect of human nature from their field of inquiry,
anthropologists were able to turn their attention instead to issues of indigenous
psychology. In a move strikingly similar to the developments reviewed above in
the study of kinship and gender, they could claim that their concern was with
the diverse ways in which notions of the self can be brought to bear in the
cultural construction of personhood, rather than with the 'actual'
psychological foundations of the self as a centre of individual experience. This
move, however, leads to precisely the same impasse, in that the opposition
between the individual (the psychological self) and the person (the social
being), on which the logic of cultural construction depends, is itself constituted
within a specifically Western discourse on nature and society. And again, people
in non-Western societies seem to be telling us something quite different:
namely that as agentive centres of awareness and experience, selves become, and
that they do so within a matrix of evolving relationships with others.
Personhood, in other words, is seen not as the imprint of society upon the pre-
social self, but as the emergent form of the self as it develops within a context of
social relations.
In Article 30, Poole advocates an approach to understanding the
development of personal identity that would take this view as its starting point.
744
To characterize it as 'non-Western' is perhaps misleading, since several

prominent social theorists, writing in the Western tradition of scholarship, have
gone out of their way to stress the relational aspects of selfhood. Nevertheless,
they have written against the grain of the doctrine of individualism which, for
many, is the hallmark of the political philosophy with which 'the West' is
popularly identified. Conscious of the dangers of importing assumptions based
in this doctrine into their studies of non-Western societies, anthropologists
have been inclined to portray these societies as holistic or socio-centric in
orientation, and to deny the applicability of such notions as 'the individual' and
'individuality' for fear that they may be tainted by their association with
individualism. This fear, as Poole points out, is misplaced. It is indeed essential
to distinguish between an analytic notion of individuality and the various and
shifting ideologies of individualism to be found in the history of Western
discourse. For the former has to do with the ways in which people, in all
societies, build unique identities for themselves and for one another out of their
own particular experiences and life histories, and the histories of their mutual
relationships. The individual of real life, equipped with such an identity, bears
only a distant and problematic relation to the abstract, atomic individual as
posited within the doctrine of individualism.
The recognition that individual selves are social in their very constitution
not only dissolves the conventional dichotomy between social anthropology
and psychology, but also has radical implications for our understanding of the
process o socta tzatwn. It as een usua , m t e past, to regar t IS process as
one in which the child, initially without a social identity or presence of any
kind, is gradually provided with the lineaments of personhood, in the shape of
schemata for categorizing and positioning others in the social environment, and
guidelines for appropriate action towards them. The acquisition of these
schemata and guidelines has been taken to be a precondition for meaningful
engagement with other persons, and hence for full participation in social life.
This implies, however, that the process of learning that prepares the child for
personhood can form no part of that life, and that to study this process is to
investigate not the dynamics of social life itself but rather the psychodynamics
of acquisition, by the immature individual, of the schemata that enable him or
her to enter into it. Herein lies the principal explanation for the unfortunate
separation between the psychological study of child development and the
anthropological study of culture and social life. The failure, until recently, of
developmental psychology to take account of the social context of learning has
been matched, by and large, by an equal failure on the part of social
anthropology to pay any attention to children and how they learn. Indeed it
would be fair to say that in the majority of anthropological accounts, children
are conspicuous by their absence.
Poole offers an approach to the rectification of this deficiency, though given
the dearth of anthropological studies of child development, it is necessarily
somewhat programmatic. Children, he argues, are not to be regarded as passive
745
SOCIAL LIFE
rectptents of social knowledge which 'descends' upon them from an

authoritative source in society. On the contrary, they participate actively and
creatively in the learning process. They do so by playing their own part in
shaping the contexts in which learning occurs and knowledge is generated.
Such interaction begins in earliest infancy, growing in complexity and
sensitivity as the child's communicative competence increases, above all with
the mastery of language. Learning, then, is not preliminary to involvement in
the social world, for it is above all through such involvement-in the 'hands-
on' experience of engaging with others in practical situations of everyday life-
that the child learns. Far from providing a prelude, in the career of each
individual, for his or her entry upon the social stage, these situations of
interactive learning are the very sites from which social life unfolds. In clear
contrast to the scenario of classical socialization theory, according to which
children start from a position outside or on the margins of society, whence they
must progressively work their way in, the view advanced here holds that
children are launched at birth into the very centre of a social world. They learn,
not in order to gain entry to this world, but to be able to make their way in it.
Nor does learning end with childhood. To be sure, childhood experience
may have a formative quality, underwriting all that occurs in later life. But adult
experience too, especially the experience of tutoring children, can be
transformative. Indeed as Poole stresses, socialization is a process that carries
on over an individual's entire lifetime. There is no point in the life-cycle at
w tc socta tzatwn cou e sat to e comp ete , mar mg o t e peno o
preparation from the attainment of full personhood. In a sense, then,
socialization and social life are two sides of the same coin: on the one side, the
enfolding of social relations in the experience and sensibility of the self; on the
other, the unfolding of the self in social action.
Perhaps no aspect of socialization is more crucial than the acquisition of
language, and in Article 31 DeBernardi presents a comprehensive review of the
large literature concerning the social dimensions of language acquisition and
use. Here a rather similar shift in perspective is evident to that described above,
from a view of language as an abstract structure or code that is 'taken on', more
or less unconsciously, by each new generation, to a view that gives a much
greater weight to the processual and performative aspects of language use. In
this latter view, language is regarded not as a pre-formed totality but rather as
perpetually under construction within the dialogic contexts of everyday
interaction, including interactions involving infants and young children.
Though the effectiveness oflinguistic communication depends on the existence
of community-wide verbal conventions, such conventions do not come ready
made but have continually to be worked at. Current conventions are the
sedimented outcomes of the struggles of past generations to make themselves
understood: thus words, as DeBernardi observes, condense a community's
recollections of its past. This kind of approach requires that much greater
attention be paid to the diversity of individual voices. Variations in usage which,
746
from a structuralist perspective, would be dismissed as merely idiosyncratic or

as defects of performance, reappear as instances of language in the making. It is
the tension between individual circumstance and common code, DeBernardi
argues, that provides language with its historical dynamic, keeping it forever 'in
play'. For language changes, even as we speak.
POLITICAL ECONOMY
Besides learning to speak, people in all societies must also learn to work, and
this generally entails the acquisition of a specific set of practical skills, along
with an understanding of the appropriate contexts for their deployment.
Learning to work is thus one aspect of socialization, which Ortiz describes in
Article 32. Her principal thesis is summed up in the statement that work is as
much a social as it is a technological process. This point needs to be argued only
because people in Western industrial societies-including many economists
and social theorists-are inclined to believe that work is somehow excluded
from the domain of social life. The reasons for this, as Ortiz shows, lie partly in
the experience of industrialization itself, and partly in the way in which the
meaning of work has been framed within the modern science of economics,
whose concern is exclusively with the dynamics of commodity production. We
may note the following points: first, that under conditions of industrial
capitalism, workers labour not for themselves or their families but for
emp oyers w o comman ot t eu capacities to a our an t e mstruments
and raw materials needed for these capacities to be realized; second, that with
the automation of production, manual skills tend to be replaced (albeit never
completely) by the operation of machines; third, that in the mass production of
commodities, the objects produced cease to be identified in any way with their
producers; and finally, that with the separation of the 'workplace' from the
'home', the latter comes to be seen as a place of consumption rather than
production. This, in turn, leads to the perception of 'housework' as an
anomalous category.
Clearly in non-industrial societies, where these conditions do not obtain, the
significance of work will be very different. For one thing, people retain control
over their own capacity to work and over other productive means, and their
activities are carried on in the context of their relationships with kin and
community. Indeed their work may have the strengthening or regeneration of
these relationships as its principal objective. For another thing, work calls for
the exercise of specific skills which identify their possessors as belonging to the
communities in which they were acquired. But it is not only by their skills that
persons in non-industrial societies are identified; they are also known for what
they produce. Through making things, people define themselves. Moreover
there is no obvious criterion for distinguishing work from non-work. Many
non-industrial societies lack any general term whose meaning would overlap
with that of 'work' in the Western industrial context (and even in that context,
747
SOCIAL LIFE
the term has manifold, often contradictory meanings). Instead, a host of more
specific terms are used to denote the various life-sustaining tasks that people
are called upon to perform. Thus work, in these societies, is embedded in social
life to the extent of being virtually indistinguishable from it. Our modern
tendency to see work as opposed to life, or to regard it as technological rather
than social, is the product of a particular history in the Western world.
This history has also given rise to the notion of the 'economy' as a domain of
activity separate from that of 'society', and operating exclusively on the basis of
market or market-like principles. The sub-discipline of economic anthropology
has emerged largely out of the attempt to show that where these principles do
not operate, the activities not only of production and consumption but also of
exchange, far from being external to society, are embedded in a social relational
matrix. However as Gregory shows in Article 33, neither of the two major
paradigms of Western economic thought-the 'commodity' paradigm of
nineteenth century political economy and the 'goods' paradigm of twentieth
century marginalism-was capable of addressing the questions raised by
anthropological work in societies where wealth is evaluated and transacted
according to principles other than those of the market. Classical political
economy distinguished between values in use and in exchange: the former
consist in the capacities of objects to fulfil human needs, the latter in the
amounts of labour that went into their production. But the distinction was
made simply in order to clear the way for an exclusive concern with exchange
va ue, as It IS revea e m contexts w erem o Jects are exc ange as mar eta e
commodities. The neoclassical economists, for their part, did away with both
these notions of value, replacing them with a single notion of utility, based not
in the objective properties of the wealth items themselves but in the subjective
preferences of individuals. This approach, apparently applicable to virtually
any kind of exchange, offered the prospects of building a deductive theory of
great generality and predictive power. Given a knowledge of individual
preferences, and of the means available for fulfilling them, one could predict
rational courses of action and their aggregate effects.
This theory, however, is quite indifferent to the particulars of social and
historical circumstance, and seemed to offer little to anthropologists more
interested in developing generalizations by induction, from a comparative
analysis of the ways in which wealth is evaluated and distributed in different
societies and periods. For them the commodity paradigm has always been more
attractive, and the gradual accumulation of ethnographic data from fieldwork-
based studies put them in the position of being able to address the questions
that it had left unanswered, particularly about the nature of non-commodity
exchange-ironically at a time when mainstream economists were abandoning
the commodity paradigm in favour of the abstract formalism of the theory of
goods. What made this possible, Gregory argues, was the development of a
positive theory of non-commoditized wealth as consisting in gifts. Gifts have
two crucial properties by which they may be distinguished from commodities:
748
first, they are evaluated in terms of a qualitative rather than a quantitative

standard (in other words, the principle of their ranking is ordinal rather than
cardinal); second, their exchange does not entail any severance of the bond of
identification with the persons exchanging them. Thus whereas the exchange
of commodities establishes a quantitative relation between the things
exchanged, that of gifts establishes a qualitative relation between the exchange
partners. Indeed, every exchange of this latter kind is a moment in the
constitution of a social relationship. Equipped with a theory of the gift,
anthropologists were able to show how the character and significance of
transactions, and the evaluation of the materials transacted therein, depend
upon the kinds of relationships in which they are embedded.
A further feature of the classical approach to political economy that
commended it to anthropological attention lay in its recognition of the ways in
which exchange is underwritten by relations of power. The imbalances in
exchange set up by these relations enable dominant elites to cream off the
surpluses needed to finance their activities and legitimating institutions. In
Article 34, Earle reviews the different sources of political power-social,
military, ideological and economic-and considers their respective strengths
and weaknesses. Social power, based on the ability to draw support from close
kin, is necessarily limited in scope, since the strength of support naturally
wanes with increasing genealogical distance. Military power, based on threat
and intimidation, or on the direct use of physical force, is perhaps more
e ectiVe ut a so more 1 tcu t to contro , smce a mt ttary ea re can a too
easily turn against a leader who has grown too dependent on it. Ideological
power is established through the promulgation of belief in the natural right to
rule of an elite, backed by religious sanction. But this, too, has its limitations,
for power of this kind can be used as much to resist domination as to support it.
The final source of power, namely economic, lies in controlling access to the
means of production and distribution of necessary goods, whether staple
foodstuffs or prestige-conferring valuables.
In different societies, or among competing factions within the same society,
elites have based their dominance on different sources of power. However these
sources, Ear le argues, are neither independent of one another nor equivalent in
the control they afford. Of all the four sources, the economic is most
fundamental, since it alone can provide a stable basis for the construction and
expansion of complex, politically centralized societies. This is not only because
of the ease with which economic processes can be controlled (especially with
the intensification of the regime of subsistence production), but also because
the products of the economy can be reinvested in order to secure control over
other sources of power. Social power can be extended by financing strategic
marriages, military power by supporting and arming the cadre, and ideological
power by financing religious institutions and ceremonies which uphold the
legitimacy of elite authority. It is by examining the historical interconnections
between the sources of power, Ear le contends, that we can best understand the
749
SOCIAL LIFE
evolution of complexity in human society, from the small, intimate bands of

hunter-gatherers, through tribal polities and chiefdoms, to large and highly
stratified urban states.
DISPUTE, NEGOTIATION AND SOCIAL ORDER

Political anthropology has its roots in the problem of order. Western
philosophers have looked to the institutions of the state as providing a
regulative framework within which individuals could peaceably pursue their
own legitimate interests. It was assumed that in the absence of such a
framework, social life would dissolve into chaos, a war of all against all. Such,
indeed, was supposed by many to have been the lot of humankind in its
primordial condition of savagery. Anthropological studies among peoples
without a state organization, or anything equivalent to it, revealed however that
they did not lead disordered lives. Nor did they experience levels of conflict
significantly greater than those encountered in state-organized societies. The
problem was thus to explain how order in these societies is established and
maintained. If behaviour is guided by rules or norms, in what authority are
they vested and how are they enforced? And how are disputes handled when
they do arise? In Western societies, the answers to these questions come under a
single rubric, that of 'law'. The term is used to refer both to a set of codified
rules and regulations-backed by the authority of the judiciary-for people to
IVe y, to an mstitutwna apparatus t roug w IC government can exercise Its
role in steering human affairs, and to a set of procedures for the adjudication of
disputes. Is there, then, anything equivalent to 'law' in non-Western, and
especially in stateless, societies? In Article 35, Roberts reviews the history of
attempts by both anthropologists and legal specialists to address this question.
He divides this history into five phases. The first is characterized by the
attempts of late nineteenth century scholars to discover the primitive
antecedents of what they perceived as an evolved state of modernity. In the
second phase, evolutionary questions were replaced by functional ones, as the
first generation of anthropologists to have conducted intensive fieldwork asked
how the institutions of the societies they studied worked to maintain order and
stability. The third phase was marked by a shift of focus from the maintenance
of order to processes of dispute, and led to a number of advances in the
understanding, for example, of how settlements may be reached without
involving third-party adjudication, of the role of litigation in struggles for
political power, of the relations between rules and outcomes, and of the
differences between legal disputation and physical fighting as mechanisms of
conflict resolution. In the fourth phase, anthropologists and historians
embarked on a highly critical reappraisal of earlier ethnographic depictions of
the so-called 'customary law' of native peoples. This law, it was argued, bore
only a tenuous relation to precolonial arrangements, and was largely invented
by the colonial authorities, with some assistance from their anthropological
750
advisers, as an instrument of domination by indirect rule. In the fifth and final

phase, this radical critique has given way to a more measured view which
recognizes the co-existence of a plurality of loosely bounded normative orders,
situated at a number of 'legal levels' from metropolitan centres to local
communities. The problem, in this 'legal pluralist' approach, is to understand
the linkages between what goes on at these different levels. Yet the applicability
of the concept of 'law' to normative orders which lack the attributes of state law,
that is where the order is not tied to a command structure, remains problematic.
In the last analysis, Roberts suggests, what is 'legal' about the anthropology of
law may have less to do with its subject matter than with the fact that the
majority of its practitioners are now lawyers rather than anthropologists!
While people may sort out their individual differences by verbal
contestation or by actually fighting, the same applies at the collective level as
well. Collective violence-whatever its causes, which are clearly multiple-
typically takes the form of warfare; and the threat or reality of war brings its
own countermeasures by way of attempts to promote common security through
negotiated settlement. In Article 36, Rubinstein considers the relevance of
anthropological understanding in tackling the many large-scale conflicts,
affecting the lives of millions of people, that are endemic in the contemporary
world. This entails a significant change of perspective, in that diplomats,
analysts and politicians who are professionally charged with negotiation at this
level are not so much the subjects as the consumers of social scientific research.
Ru mstem s ows t at t e approac a opte y t ese pro esswna s, a ove a m
the Western defence establishment, rests on a set of highly artificial
assumptions about how people behave. Ironically known as 'political realism',
this approach envisages a world in which the actors are nation states, and in
which actions follow the predictions of formal econometric or game-theoretic
models, calculated on the basis of a knowledge of objective social scientific facts
to maximize economic or military pay-offs. Thus no account is taken oflocal or
indigenous interests below the level of the state, or of the influences of social
and cultural experience on people's perceptions and actions, or of sources of
power other than the military and economic (for example what Rubinstein calls
'normative power'-roughly equivalent to Earle's concept of 'ideological
power'-which can furnish a significant means of resistance against economic
and military might). Where political realism does take cultural considerations
into account, these are typically based in crude stereotypes of how other people
behave, insensitive to both history and context. In their application, such
stereotypes generate chronic misunderstanding. Good negotiating practice,
Rubinstein argues, should be informed by a proper appreciation of the
cognitive and emotional force of symbolic forms. This is no minor plea: the
future security of entire populations depends on it.
Beteille's discussion, in Article 37, of the premisses of inequality in human
societies takes us back to the problem of order. A central question in the
anthropology of law, as we have seen, is whether any framework of norms and
751
SOCIAL LIFE
standards can be maintained without a system of imposed regulation: that is,

whether there can be 'order' without 'command'. Another way of posing the
same question is to ask whether there can be 'society' without 'inequality'. Is an
egalitarian society possible, even in theory? The answer, of course, depends on
what is meant by society, and by equality. On the face of it the trajectory of
social evolution, as laid out by Ear le in Article 34, from hunter-gatherer bands
through agrarian civilizations to modern nation states, seems to involve a steady
increase in inequality until a relatively recent point was reached, marked by the
transition to modernity (politically speaking, from aristocracy to democracy),
whereupon the trend went into reverse, and orders that were rigidly
hierarchical in both ideal and fact gave way to societies premissed upon a
formal commitment to equality. Yet it would be absurd to suggest that the
equality of the hunter-gatherer band remotely resembles that of the modern
industrial state.
Indeed these inequalities rest on entirely different principles: the first on the
lack of enduring commitments and dependencies among persons who are
nevertheless well-known to one another; the second on an individualistic
conception of the person as the singular yet anonymous embodiment of a
universal humanity. Those who hold that egalitarian society is an impossibility
are inclined to dismiss the evidence from hunter-gatherer societies on the
grounds that in the absence of any framework of normative obligation, or of
anything that might be recognized as 'social structure', they can scarcely be
recogmze as societies at a . T e very existence o society, t ey argue, epen s
on the presence of rules, and since it is in the nature of rules that they are
sometimes violated, giving rise to disputes that require adjudication, situations
are bound to arise which call for the exercise of power by some individuals over
others. By this argument, inequality is a necessary condition for ordered social
life. The argument, however, can also be turned on its head, such that systems
of rules, far from calling for the exercise of power, function as part of an
apparatus of domination through which power achieves its effects. The more,
then, that power is concentrated in the hands of a dominant elite, the more
elaborate the framework of rules and regulations imposed upon the subject
population.
When it comes to modern nation states, it is invariably the case that
ideological commitments to equality are combined with stark practical
inequalities. Citizens may, according to the democratic ideal, be equal before the
law, but they are very far from equal after it. However as Beteille shows, the
manner of their inequality will depend upon whether the public commitment is
to an ideal of competitive equality (judged by the balance of opportunity) or
distributive equality (judged by the balance of income or result). In the first
case, everyone is supposed to have the same chances to compete in the 'market
place' of civil society, but due to inherent variations of individual ability, some
are said to do better for themselves than others. The successful rise to the top,
the unsuccessful sink to the bottom, whilst the majority settle somewhere in
752
between, leading in aggregate to the observed pattern of social stratification. In

the second case, the state intervenes to ensure equality of results through an
enforced redistribution from the 'haves' to the 'have-nots'. Yet it can do so only
because of a concentration of power at the centre. Thus the very promotion of
distributive equality sets up further inequalities, experienced as constraints on
civil liberties.
Two further aspects of Beteille's argument warrant special emphasis. The
first is that while people in society may differ in all kinds of ways-such as in
gender, physical appearance, occupation and lifestyle-not all differences
count as inequalities. What converts difference into inequality is a scale of
evaluation, and such scales may vary within as well as between societies. In
stressing the possible existence of multiple and partially contradictory scales
within the same society, Beteille echoes a point also made by Moore (Article 29)
with regard to the evaluation of gender differences, and by Roberts (Article 35)
with regard to different fields of law. Second, whether or not a society appears
egalitarian will depend to a certain extent on the scale at which it is defined.
Large, highly stratified societies may encompass communities that are,
internally, markedly egalitarian; conversely small-scale, egalitarian societies
may be encompassed within wider social systems structured by relations of
pronounced inequality.
This point applies with equal force at a global level. To the extent that
Western societies have achieved a basic level of affluence for all, they have done
so at t e cost o t e ot er societies aroun t e wor t at t ey ave su JUgate
and exploited. How they did so is the subject of Worsley's account, in Article
38, of the history of colonial expansion, an expansion that laid the foundations
for the contemporary world order. This account effectively demolishes any
naive, evolutionist model of development that would portray the history of
non-European peoples as one which began with first European contact, and
which has gradually raised them from an original, primitive or traditional
condition to a position where they can take their fair share of the benefits of
modernity. For one thing, the societies first encountered by Europeans varied
from small stateless polities to great empires of a scale, wealth and
sophistication far exceeding anything that had been achieved in Europe itself.
In many parts of the world, European supremacy was by no means a foregone
conclusion, and was achieved only after long and bloody conflicts which often
set native peoples at war with one another. For another thing, far from
encouraging the development of local industries in the territories they
controlled, the colonial powers blocked such development, in order to
guarantee for Western industry its supply of raw materials and markets for its
goods. Today, in a post-colonial world, key resources remain concentrated in
the 'developed' nations of Europe, North America and of course the Far East,
while war and starvation stalk the continent of Africa, and indigenous people
are being wiped out in the name of progress in parts of South America and
South-east Asia. At the same time, the West is becoming painfully aware of the
753
SOCIAL LIFE
disastrous environmental consequences of its own expansion. Notwithstanding

rumours of a new world order, humankind has still a long way to go before it
reaches the far side of history, if indeed it ever will.
FURTHER READING
Barnard, A. and Good, A. (1984) Research Practices in the Study of Kinship, London:
Academic Press.
Beteille, A. (1987) The Idea of Natural Inequality and Other Essays, Delhi: Oxford
University Press.
Carrithers, M., Collins, S. and Lukes, S. (eds) (1985) The Category of the Person:
Anthropology, Philosophy, History, Cambridge: Cambridge University Press.
Collier, J. and Yanagisako, S. (eds) (1987) Gender and Kinship: Essays towards a Unified
Analysis, Stanford: Stanford University Press.
Comaroff, J.L. and Roberts, S.A. (1981) Rules and Processes: the Cultural Logic of
Dispute in an African Context, Chicago: University of Chicago Press.
Dumont, L. (1986) Essays on Individualism: Modern Ideology in Anthropological
Perspective, Chicago: University of Chicago Press.
Earle, T. (ed.) (1991) Chiefdoms: Power, Economy, and Ideology, Cambridge: Cambridge
University Press.
Fortes, M. (1983) Rules and the Emergence of Society (RAI Occasional Paper, No. 39),
London: Royal Anthropological Institute.
Foster, M.L. and Rubinstein, R. (eds) (1986) Peace and War: Cross-Cultural
Gregory, C.A. (1982) Gifts and Commodities, London: Academic Press.

Hinde, R.A. (1987) Individuals, Relationships and Culture, Cambridge: Cambridge
University Press.
Humphrey, C. and Hugh-Jones, S. (eds) (1992) Barter, Exchange and Value: an
Anthropological Approach, Cambridge: Cambridge University Press.
lngold, T. (1986) Evolution and Social Lift, Cambridge: Cambridge University Press.
Johnson, A. and Earle, T. (1987) The Evolution ofHuman Societies: from Foraging Group
to Agrarian State, Stanford: Stanford University Press.
Joyce, P. (ed.) (1987) The Historical Meanings of Work, Cambridge: Cambridge
University Press.
Mann, M. (1986) The Sources of Social Power, vol. 1, Cambridge: Cambridge
University Press.
Mauss, M. (1990) The Gift: the Form and Reason for Exchange in Archaic Societies (trans.
W.D.Halls), London: Routledge.
Moore, H.L. (1988) Feminism and Anthropology, Cambridge: Polity Press.
Moore, S.F (1978) Law as Process, London: Routledge & Kegan Paul.
Morris, B. (1991) Western Conceptions of the Individual, Oxford: Berg.
Riches, D. (ed.) (1986) The Anthropology of Violence, Oxford: Blackwell.
Roberts, S. (1979) Order and Dispute, Harmondsworth: Penguin.
Rubinstein, R.A. and Foster, M.L. (eds) (1988) The Social Dynamics of Peace and
Conflict: Culture in International Security, Boulder: Westview Press.
Sahlins, M. D. (1972) Stone Age Economics, London: Tavistock.
754
Schieffelin, B.B. and Ochs, E. (eds) (1986) Language Socialization across Cultures,
Cambridge: Cambridge University Press.
Schneider, D.M. (1984) A Critique of the Study of Kinship, Ann Arbor: University of
Michigan Press.
Strathern, A.M. ( 1988) The Gender of the Gift, Berkeley: University of California Press.
Taylor, C. (1989) Sources of the Self, Cambridge, Mass.: Harvard University Press.
Wallman, S. (ed.) (1979) Social Anthropology of Work, London: Academic Press.
Whiting, B.B. and Edwards, C.P. (1988) Children of Different Worlds, Cambridge,
Mass.: Harvard University Press.
Wilson, E.O. (1980) Sociobiology (abridged edition), Cambridge, Mass.: Harvard
University (Belknap) Press.
Wolf, E.R. (1982) Europe and the People without History, Berkeley: University of
California Press.
Worsley, P. (1984) The Three Worlds: Culture and World Development, London:
Weidenfeld & Nicolson.
755
27
SOCIALITY AMONG HUMANS AND

NON-HUMAN ANIMALS
R.I.M.Dunbar
The tendency to impute human motivations and forms of social organization to

animals has a long and distinguished history. As long ago as the fourth century
BC, Aristotle noted that animals differ from each other in disposition, some
being stubborn, others mean and scheming, yet others mischievous and
wicked, bashful or jealous. On the other hand, his description of bee society is
extraordinarily accurate (it was not correctly described again until the mid-
eighteenth century), though he persistently refers to the queen in each hive as
t e ma e) ea er .
During the three decades since 1960, there has been an unparalleled growth
in our knowledge and understanding of animal societies. In part, this has been a
consequence of the amount of detailed descriptive fieldwork that has been
undertaken. In addition, however, the second half of this period has coincided
with major new developments in theory that have revolutionized our
understanding of how and why animals behave in the ways they do. This
theoretical advance (known variously as 'sociobiology', 'behavioural ecology' or
'evolutionary ecology') has both revealed animal societies in a new light and
stimulated a vast amount of empirical research aimed at testing specific
hypotheses.
Attempts to interpret human social behaviour in the light of animal
behaviour are, of course, a far from recent feature in the history of modern
science. Undoubtedly the two most controversial attempts to do so during this
century were those associated with the behaviourists in the 1920s and 1930s
(e.g. Watson 1919, Skinner 1938) and the ethologists in the 1960s (e.g. Lorenz
1960, Morris 1967). By and large, these proceeded by analogy: if greylag geese
behave in a certain way, then human beings will do so too. With the benefit of
hindsight from three decades of fieldwork, such reasoning can now be seen as
naive: primates, for example, are so flexible in their behaviour that we cannot
always infer the behaviour of one population from that of another population
even when they belong to the same species (see Smuts et al. 1987, Dunbar
756
SOCIALITY AMONG HUMANS AND NON-HUMAN ANIMALS
1988). If we cannot do this reliably within a species, then the possibility that we
can do it between species (and the more so between those as distantly related as
geese and ourselves) becomes even more remote. One of sociobiology's
beneficial influences in this respect has been to shift the emphasis away from
analogical reasoning of this kind to give greater prominence to the specific
contexts in which particular animals or groups of animals find themselves. This
is not to say that there are no general principles that apply universally, but
rather to emphasize the fact that these principles are at a deep level: the same
universal principle may express itself in quite contrary ways in different
ecological or demographic contexts.
In this article, I summarize our current understanding of animal sociality
and ask what this can tell us about human sociality. I begin with the more
general problem of defining sociality, and then elaborate the theoretical
perspective that underlies all contemporary studies of animal behaviour. I go
on to consider the reasons why animals are social and conclude by attempting
to reassess the differences and similarities between animal and human societies.
THE PROBLEM OF SOCIALITY

Animals vary in their social arrangements from the wholly asocial solitary life
of some insects and lower organisms (as well as of primates like galagos and the
orang-utan) to the highly complex societies of many birds and mammals
parttcu ar y t e tg er pnmates me u mg, most conspicuous y, ourse ves).
Defining a scale of social complexity that would enable us to make quantitative
comparisons between species has always been a major difficulty. Biologists have
tended to simplify the situation by drawing a qualitative distinction between
the essentially asocial existence of many solitary and semi-solitary species and
the obviously social groups of many other species. This has led to a tendency
for sociality to be equated with group-living, and for social complexity to be
judged in terms of group size (see, for example, Wilson 1975). Thus, gorillas
would appear to be social because they live in permanent groups, whereas
orang-utans would appear to be asocial because they live solitarily. Such a
criterion creates anomalies that beg questions about what we mean by sociality.
Chimpanzees, for example, might seem asocial because they tend to forage in
'groups' that average somewhat less than two individuals in size, at least in
some populations. Yet, as the detailed studies of de Waal ( 1982) and Goodall
(1986) clearly show, chimpanzee social life is in fact extraordinarily rich.
Likewise, were we to judge complexity by group size alone, we would be
obliged to consider antelope that are found in herds of several hundred
thousand animals (e.g. wildebeest) as socially more complex than, say, baboons
that typically live in groups of 40-50 animals. Yet we know that relationships
among primates are of a very different social order from those of antelope,
whose groups are often passive aggregations of individuals brought together by
particularly good feeding conditions.
757
SOCIAL LIFE
This tendency to view sociality in terms of the disposition of animals in

groups was in part a consequence of the fact that, until the 1960s, most field
studies were of short duration (often no more than censuses): one of the few
things that can easily be quantified during a short field trip is the way the
animals are distributed around the habitat at any given moment. In addition,
the bulk of the fieldwork conducted up to 1960 was carried out on birds, and
this fostered a tendency to focus on a small number of functionally relevant
social units (notably those involved in mating and rearing) that have special
prominence in bird biology.
How, then, should we view animal sociality? The past two decades of field
research on primate behaviour have revealed that the complexity of primate
societies lies in the relationships among individuals that bind the group of
animals together. A shift in emphasis towards a focus on relationships began as
early as the 1970s, when Hinde ( 1976) suggested that a social system could
profitably be interpreted in terms of the patterning and quality of relationships
among a set of individuals, each of these relationships, in turn, being
interpretable in terms of the patterning and quality of the interactions between
the two individuals involved. In other words, what we, as observers, describe as
a relationship between two individuals is something we abstract from the set of
interactions that we observe between them, with the social system as a whole
being similarly derived from the set of such relationships between all the
individuals in a given region.
ne Imp Icatwn o t IS view IS t at amma s mig t ave re atwns Ips an
hence a form of social life), even though they do not spend all of their time
together in the same group. This has important implications in the case of
chimpanzees, for example. The earliest field studies suggested that
chimpanzees did not have social groups of any kind, the only form of stable
relationships in evidence being the associations of mothers and their offspring
(Goodall 1965). However, subsequent fieldwork at Gombe and in the Mahale
Mountains of Tanzania during the 1970s revealed that the loose parties in
which chimpanzees were most often found did not form at random. Rather,
their members are drawn from a pool of individuals (later known as the
'community') that share a common ranging area (see Wrangham and Smuts
1980). These individuals readily form foraging parties with each other and are
antagonistic towards members of other, neighbouring communities.
Subsequently, data from a number of primate species began to suggest that
multi-layered societies in which individuals were members of ever more
inclusive, hierarchically organized, social groupings might in fact be typical of
most, if not all, species (Dunbar 1988, 1989a). African vervet monkeys, for
example, live in archetypal primate groups of some ten to twenty individuals.
But far from being the socially amorphous entities presupposed in most of the
earlier literature, these groups turn out on closer inspection to be highly
structured: individuals living within the same group do not interact with one
another at random. Similarly, while neighbouring groups often do have
758
mutually antagonistic relationships and do defend their territories against

incursions, this is not necessarily true of all groups. Some groups turn out to
have positive relationships with each other, in that they regularly exchange
members (Cheney and Seyfarth 1983). From this observation emerged the
concept of a local community in which certain groups within the population
are bound together by historical and genetic ties.
The gelada baboon provides a more complex example. Gelada live in
reproductive units that consist of a single breeding male and up to ten
reproductive females and their dependent offspring. These units are
themselves organized into highly discrete coalitionary subgroupings, mostly
based on close female kin relationships (mother-daughter-sister). These
reproductive units, in turn, are grouped into higher-order clusters (bands)
based on their occupancy of a common core ranging area, while the bands
themselves are organized into still higher-level groupings (communities) which
more readily form combined foraging parties (Kawai et al. 1983).
Because the relationships between the reproductive units of a band are
relatively loose, these units may all be together in one large group on one day,
but then dispersed over a wide area on the next; on yet other occasions, some
units of one band may join up with units from another band in areas of range
overlap. The resulting groupings (termed herds) are temporary aggregations
that undergo continuous flux as units come and go, but they are none the less
an important feature of the gelada social system: they provide the animals with
t eu pnmary protectiOn rom pre ators see Kawat et a . ). In a Itwn,
the relationships they entail cut horizontally across the vertically structured
groupings that characterize the species's social system. This example serves to
remind us that animals may be simultaneously involved in several different
networks of social relationships.
Multi-tiered social systems of similar complexity have also been noted in
other Old World monkeys (e.g. hamadryas baboons: Kummer 1968, 1984;
pigtailed macaques: Robertson 1986). Hierarchically organized groupings
have, in addition, been documented in birds (e.g. bee-eaters: Hegner et al.
1982), rodents (e.g. prairie dogs: King 1955) and elephants (Moss and Poole
1983), suggesting, perhaps, that social complexity of this kind may be typical of
most higher vertebrates (birds and mammals).
What is important here is not so much the particular form of social
organization as the facts that an animal has relationships with a variety of
individuals, and that these relationships reflect differing degrees of familiarity
and intensity. They may also be expressed behaviourally in quite different ways.
Thus, when two individuals of the same elephant family meet after a temporary
separation, they commonly engage in an intense, excited, often noisy greeting
ceremony, whereas when members of different families (but the same clan)
meet they simply place the tips of their trunks in each other's mouths (Moss
and Poole 1983).
In many respects, these groupings resemble the social networks discussed by
759
SOCIAL LIFE
sociologists (see for example Milardo 1988), and they do in fact function in
rather similar ways. An individual animal may belong to a number of different
networks in just the same way as a man or woman may belong to a set of
partially overlapping networks (e.g. networks of relations among work
colleagues, friends, kin, members of a political party or church, etc.).
Despite this emphasis on relationships, the grouping patterns of animals
remain none the less a central concern for biologists. It is not just the nature of
the relationships that one animal has with another that is interesting, but also
their number. Why should an animal have intense relationships with a dozen
other individuals and not just with one? Why do some animals prefer to live
alone, even though they have relationships with those living nearby? Groups
themselves are, after all, simply a reflection of the way certain kinds of
relationships cluster in space-time. Understanding why some groups are
manifested in this sense, but others not, is an important endeavour. In addition,
the size of the group in which an animal lives inevitably places constraints on
the number and frequency of other kinds of relationships that it can have. For
example, if ecological conditions limit groups to a maximum of two individuals,
then the animals concerned will not be able to form coalitionary relationships
with third parties vis-a-vis one another when disputes arise. Nor will it be
possible for one individual to play another off against a third. Thus,
understanding the factors that foster the development of groups of a certain
size and type remains fundamental.
T IS raises an Important Issue o e Imtwn. W en 10 ogists re er to
relationships, it is clear that they mean something rather informal and low-
level, little more than a consistent patterning in the interactions of a pair of
animals. When social anthropologists refer to relationships, they often mean
something closer to a rule-bound contractual arrangement between consenting
parties. A similar difference exists in the use of the term 'group'. To a biologist,
a group is simply a set of animals bound together in some way and occupying a
discrete segment of space-time: groups may be dispersed (when their members
do not physically live together) or they may be spatially concentrated.
Moreover, a single individual on its own may, for some purposes, be said to
constitute a 'group' of size one. There is nothing particularly odd about this
usage, but it will probably strike social anthropologists as perplexing because,
in general, they tend to regard groups as being the product of contractual
arrangements among a set of individuals: by definition, then, it takes two or
more to make a group.
These contrasting usages arise from the different explanatory interests of
biologists and social anthropologists. That humans organize their relationships
on the basis of verbally negotiated contractual arrangements is neither here nor
there for the biologist (even though it may be recognized as an intrinsically
interesting property of human behaviour). The evolutionary biologist's
concern is with the functional consequences of those contractual
arrangements, not with their immediate causes. At this level, the mechanisms
760
involved in their creation are immaterial. I draw attention to this point now
because it is important to appreciate that the same term may be used to mean
quite different things in the two disciplines of social anthropology and biology.
I have more to say about the implications of this in the following section.
SOME BACKGROUND THEORY

However we define societies, merely to describe them is not enough. The very
fact of their existence, and a fortiori the existence of differences between species
in their forms of social life, beg questions about why a given society takes the
form it does. Biologists have long recognized that even such an apparently
innocuous question as asking why something is the case can (and indeed
should) be answered at a number of logically different levels (Huxley 1942,
Tinbergen 1963). Thus, in asking why a given society takes the form it does,
answers might be given in terms of ( 1) the mechanisms (or processes) that bind
it together (explanations of proximate causation), (2) the function (or purpose)
that a particular form of society serves for its members (functional
explanations), (3) the developmental processes (specified by some combination
of genetic, environmental and cultural factors) that give rise to the particular
forms of behaviour that underwrite it (ontogenetic explanations) and (4) the
historical sequence by which that particular form of society arose from some
ancestral form (explanations of phylogeny or evolutionary history).
Known nowa ays as Tm ergen s Four W y s , t ese questiOns are ogtca y
distinct. Our answer to any one of them in no way predisposes us to a particular
form of answer to any of the other three. This distinction between different
levels of explanation is fundamental; most disputes over the nature of biological
explanations have arisen as a result of a failure to distinguish either between
proximate causation and function or between ontogenetic and functional
explanations.
Despite the equal importance attached to all four types of explanation by
biologists, functional explanations have a particularly important role to play in
biology: for it is only by taking function into account that we can understand
how evolutionary change can be other than random and chaotic. As has been
repeatedly noted, there has been insufficient time since the first appearance of
life on earth for the enormous diversity of life-forms that have ever lived to
have evolved simply by random mutation, even when assisted by genetic drift.
Only Darwinian natural selection is a powerful enough process to direct
evolution sufficiently to account for what we observe. The logic of natural
selection is, in essence, as follows: in any self-replicating system, operating in a
finite environment, in which there is reasonable fidelity of copying between
successive generations, any character that serves to enhance the rate of
replication of the heritable component will inevitably be selected for.
Differential rates of replication will tend to establish characters that are
relatively more effective in facilitating future replication.
761
SOCIAL LIFE
In a conventional biological context, the material that is passed on from one

generation to the next is, of course, genetic, consisting of molecules of DNA.
Since only genetic material is transmitted from parent to offspring,
evolutionary change can occur only through the inheritance of genes (see,
however, Odling-Smee in this volume, Article 7). The rate with which these
elements are replicated will be a simple function of the impact they exert on the
carrier's ability to reproduce. The gene's rate of replication over time (relative
to that of alternative forms) is defined as its fitness, and it is this that provides
the criterion whereby functional explanations are judged. The structure of
Darwinian arguments has been shown to be logically self-contained and to
follow from a few simple axioms (see Williams 1970, Dunbar 1982a).
This insistence on the primacy of genes has sometimes been expressed
metaphorically by characterizing the gene as 'selfish' (Dawkins 1976). This, of
course, has nothing to do with either behaviour or morality; rather, it is
intended as a reminder that, since in a Darwinian world it is genes that are
passed on from one generation to the next, we must always adopt a 'gene's eye
view' when asking questions about the evolutionary function of a given
phenomenon. To take any other perspective (for instance, to claim that
evolution occurs for the benefit of the individual or the species) is to invite
error, because neither individuals nor species have any temporal stability on an
evolutionary timescale. This will have important implications when we come to
consider the evolution of altruism in a later section.
It IS Important to e aware t at t e term gene as use m t IS context IS t e
Mendelian gene, not the more familiar segment of DNA that lies at the heart of
molecular biology. Strictly speaking, the Mendelian gene is a trait or character,
not a segment of DNA, even though there must be some definable relationship
between these two components of the system. More importantly, they refer to
quite different constituencies in the nexus of biological explanation. DNA is
relevant to ontogenetic explanations, whereas Mendelian genes are appropriate
to functional explanations; as such, Mendelian genes are emergent properties
of the underlying DNA and not necessarily identical with it in any simple
sense. Indeed, Dawkins ( 1982) has correctly pointed out that the heritable
elements that make the theory of natural selection work need not even be
segments of DNA; only fidelity of copying between parent and offspring
generations is required. It just so happens that, in many of the contexts with
which we are most familiar in biology, the replicating entity is DNA. But it
need not be: anything that copies itself with reasonable fidelity is a replicating
entity and therefore subject to processes of natural selection. Conventional
trial-and-error learning is one example; cultural inheritance of behaviour is
another. For this reason it has been possible to develop models of cultural
transmission based on Darwinian principles that in no way invoke genetic
determinism (see for example Cavalli-Sforza and Feldman 1980, Boyd and
Richerson 1986).
Despite persistent claims to the contrary, sociobiological explanations are
762
wholly and only about evolutionary function. They are concerned with the
question: Why does a given trait persist in a population? As such, they do not
necessarily presuppose any genetic basis for a given behaviour. Indeed,
explanations about evolutionary function can, of themselves, tell us nothing at
all about the ontogenetic foundations of the trait in question: the mode of
inheritance might involve 'memes' (Dawkins's term for cultural replicators),
but it equally might involve genes.
This leads naturally to the consideration of a key feature of the social
behaviour of higher organisms (especially primates), namely its extraordinary
flexibility. The variety of social forms exhibited by different populations of
baboons, for example, makes nonsense of the assumptions about species-
specific behaviour patterns that characterized the classical ethology of the
1940s and 1950s. Even within a given social group, we may find individuals
pursuing radically different social strategies (see, for example, Dunbar 1982b,
Caro and Bateson 1986). These kinds of alternative strategies are inherent in
any biological system (see, for example, Maynard Smith and Price 1973,
Maynard Smith 1982). Indeed, evolution cannot occur within a biological
system unless there is variation in a character among the constituent organisms
that make up the population. This is not to suggest that species-typical
behaviour patterns do not exist: far from it-the smile, after all, is a human
universal. But such behaviours are, in themselves, rather uninteresting from an
evolutionary point of view: what is of interest is how these behaviours are used
m a strategic sense to ac teve unctwna y-re ate goa s.
Given this perspective, one obvious interpretation of social systems is that
they function as reproductive strategies. In other words, animals evolve the
social systems they do because these enable them to survive and reproduce
more effectively in the particular environments in which they live. In effect, a
group is a co-operative solution to one or more problems of mutual concern.
We can pursue this argument a little further by suggesting that the multi-level
societies discussed in the preceding section represent a series of such solutions,
each concerned with a different functional problem (Dunbar 1988, 1989a).
Finally, two caveats are in order.
First, it is important in this context to appreciate that evolutionary
functionalism has little in common with the functionalism that dominated
sociology and anthropology during the 1930s and 1940s. Structural-
functionalism in the social sciences concerned itself with the self-regulating
properties of whole societies and viewed the individual's place in the social
system as subservient to the perpetuation of the monolithic structures of the
system itsel£ In contrast to this top-down view, evolutionary biology adopts a
strictly bottom-up view: society as we perceive it is simply the outcome of the
series of decisions made by a set of individuals to associate with each other.
Even though the structural components of the system may impose constraints
on just what those individuals can do (they can, after all, only live in social
groups if they are prepared to compromise on their ideal strategies), the system
763
SOCIAL LIFE
as a whole is intrinsically dynamic and can be expected to change continuously

through time as the individual and collective interests of its members change.
Second, care needs to be taken to avoid misinterpreting the
anthropomorphisms commonly used by evolutionary biologists. Genes, as well
as individuals, are commonly spoken of as though they had goals and made
decisions about how to behave for the best, all couched in the language of
strategy and gamesmanship. It is, of course, easy to interpret these terms
literally. In evolutionary biology, however, they function as shorthand for what
otherwise would become impossibly convoluted expressions. This use of
metaphor works because evolutionary processes are teleonomic (goal-directed,
without being intentionally so). Indeed, in this context 'Tinbergen's Four
Why's' are clearly applicable: questions about intentionality or consciousness
are concerned with proximate mechanisms, and not with function. The same
functional requirements can be met by any number of different proximate
mechanisms, one of which entails fully conscious decision-making, another the
operation of a pre-programmed automaton.
WHY LIVE IN GROUPS?

Having, I hope, cleared the debris of past misunderstandings out of the way, I
can now concentrate on the functional significance of sociality in animals. For
purely logistic reasons, it is convenient to partition any discussion of sociality
mto a num er o separate questiOns. Domg so a ows us to concentrate m turn
on each of the key problems that arise without confounding issues that are
logically distinct. I have already raised the general question of why animals
might be social (in the sense of maintaining affiliative or friendly relationships
with each other). Once an animal has 'decided' to form relationships of a
particular kind, two logically quite separate questions arise. One of these
concerns the size of the group it should live in. It might decide to live in a
group of size 1 (i.e solitarily) or it might decide to live in a group of size 101.
The issue here is the simple question of what determines group size. The
second question concerns whom it should form these groups with. I shall deal
with the first question in this section; the second question will be the subject of
the penultimate section.
One of the central tenets of sociobiology is that whichever sex has the most
at stake with respect to reproduction will always be able to exert more influence
over the form of the species's mating system. Two points are worthy of
comment here. First, it is a matter of accident as to which (genetic) sex incurs
the greater cost in reproduction, although we then habitually refer to this sex as
'female' and its gametes as 'eggs' or 'ova'. (In birds, for example, it is the XY
sex that lays the eggs, whereas in mammals it is, of course, the XX sex.)
Secondly, the imbalances between the sexes in the initial investment in
reproduction are very much greater in mammals than in any other taxonomic
group (in part because of internal gestation but, more importantly, because of
764
lactation), and they are proportionately more acute in primates than in other
mammalian taxa because of the prolonged periods of infant dependency that
characterize this group.
In primates, then, females may be expected to exert the most influence and
will thus distribute themselves around the habitat in the ways that are most
conducive to their successful reproduction, forming groups only when and if
these are advantageous. Males will then map themselves onto the distribution
pattern of the females in such a way as to maximize their own reproductive
success. In primates, at least, there is direct evidence to support this claim (e.g.
from releases of animals into new habitats: Charles-Dominique 1977).
Given that the animals in a population distribute themselves around the
habitat in groups of a particular size, what factors influence this size? In
biological systems, most questions of this kind turn out to have rather complex
answers that rest on a balance between the costs and benefits of a given strategy.
In principle, the benefits that would accrue to the individual from living in a
group, taken on their own, would be expected to result in 'runaway' selection in
favour of ever larger groups. However, the fact that organisms are systemic
entities means that evolutionary change along one dimension inevitably creates
costs along one or more of the system's other dimensions and these act to
counter-balance the evolutionary forces driving the system towards any one
extreme. Increased group size may have advantages in terms of territorial
defence, for example, but large groups impose greater costs on their members
ecause t e stze o terntory t at as to e e en e to provt et e group wtt
the resources it needs increases faster than the area the group can patrol
effectively. In most cases, the solutions that animals adopt turn out to be
compromises between the conflicting demands stemming from a number of
different considerations.
Biologists have suggested four main selective advantages to explain why
animals live in groups (see Wrangham 1983, 1987, Dun bar 1988). These are:
(1) improved care of young; (2) co-operative hunting; (3) defence of food
resources, and (4) protection against predators. Table 1 lists a selection of
species for which each of these explanations can plausibly be invoked.
In general, the advantages that accrue in terms of parental care are likely to
be restricted to species that are monogamous (i.e. those in which mating occurs
only between one male and one female who normally live alone as a pair).
Sharing of parental duties between both the male and the female is common
only in such species. Indeed, monogamy is exclusive to those species in which
male parental care is possible (Kleiman 1977). Of all bird species, for example,
90 per cent are monogamous, whereas monogamy is found in only about 5 per
cent of mammalian species. Male birds are just as competent at incubating eggs
and feeding the young as are female birds, but internal gestation and lactation
make it difficult for male mammals to contribute directly to the business of
reanng young.
765
SOCIAL LIFE
Table 1 Main benefits gained from group-living, with some

examples that seem to constitute instances of these
Hypothesis Examples
1 Parental care Jackals, foxes, most birds

2 Co-operative hunting Lions, wolves, hyaenas
3 Defence of food resources Primates
4 Protection from predators Antelope, buffalo, primates
However, even within the mammals there are marked differences between
taxonomic groups. Monogamy is ubiquitous, for example, among the canids
(the dog-wolf-jackal family), and in most of these species biparental care is the
norm. Lactation in these species is relatively brief and the male is able to feed
the female in the den by bringing meat back to her; once the pups are weaned,
both male and female are able to bring food back to the den for the pups to eat.
Monogamy is also typical of about 15 per cent of all primate species, though
here the distribution is very uneven. Most small New World monkeys (e.g.
marmosets, tamarins, titis) and all gibbons are monogamous, with monogamy
associated with male parental care in the marmosets and tamarins but not in the
titis or gibbons. By contrast, monogamy is extremely rare in all other groups of
pnmates. However, It seems un 1 e y t at parenta care can e a actor
promoting the evolution of monogamy in primates. Instead, it seems more
likely that the opportunity for paternal care is a by-product of monogamy
rather than its cause, with the evolution of monogamous mating systems having
more to do with the risk that females run of infanticide by other males (see van
Schaik and Dunbar 1990).
Carnivores may often gain considerable advantages from hunting in groups:
there is ample evidence to show, for example, that the size of prey caught
increases with group size in species that hunt co-operatively (e.g. lion: Schaller
1972; hyaena: Kruuk 1982). While the advantages of co-operative hunting
might be seen as relevant to the later hominids, they are unlikely to be relevant
to any non-human primates or to have been so to our early hominid ancestors
(e.g. the australopithecines) whose hunting skills appear to have been minimal.
There seems to be general agreement that primates live in groups either as a
defence against predators or in order to defend food sources against other
members of their own species (van Schaik 1983, Wrangham 1987, Dun bar
1988). It should be noted that defence against predators does not necessarily
imply that they are actively driven away, either by members of the group as a
whole or by one class of individuals from the group (e.g. adult males). Although
examples of male baboons driving leopards away have been documented, the
only unequivocal evidence for active deterrence of predators is that provided by
Busse (1977) for red colobus (where the presence of adult males does seem to
766
deter chimpanzees from preying on the smaller members of the group). There
is, however, an extensive literature, both theoretical and experimental, on the
advantages that prey species gain from living in groups: clumping of prey may
make it harder for predators to locate a group, more likely that a neighbour will
be taken rather than yourself, more likely that a predator will be detected before
it can approach close enough to the group to launch an attack and more likely
that a predator may become confused by prey fleeing in all directions (for a
general review, see Bertram 1978).
The hypothesis that primates group to defend food resources in the face of
intra-specific competition rests largely on the evidence that primate
populations, like those of many other organisms, are ultimately food-limited:
the amount of food available in the habitat sets the upper limit on a
population's capacity to increase its size. Whether this hypothesis specifies an
important determinant of group-living thus ultimately turns on the issue of
whether or not animals like primates are ever at such high densities that their
populations are close to the maximum that their habitats can sustain.
The primate literature is more or less evenly divided on the question of
which of these last two hypotheses-reduction of predation risk or defence of
food resources-is correct as a general explanation for group-living in
primates. Wrangham ( 1980, 1987), Dittus ( 1986) and Cheney and Seyfarth
(1987) favour the latter; van Schaik ( 1983), de Ruiter ( 1986) and Dun bar ( 1988)
favour the former. Where direct tests between the two hypotheses have been
posst e, owever, t ese ave ten e to come own m avour o t e re uctwn o
predation risk (van Schaik 1983, Dunbar 1988).
In Darwinian terms, no benefit can be viewed in isolation from the
corresponding costs. These costs increase with increasing group size and come
in two main forms: (1) direct costs in terms of increased competition over
specific items of food and other social stresses that ultimately influence an
individual's survival and/ or reproduction directly, and (2) indirect costs in the
form of disrupted time budgets and longer day journeys (because the group
needs to search an area each day that is proportional to its size). These effects
are well documented for primates (see, for example, Wrangham 1977, van
Schaik et al. 1983, Watts 1985, Stacey 1986), and in some cases the
physiological mechanisms that mediate them are well understood (e.g. Abbott
1984, French et al. 1984 ).
The sizes of the social groups in which animals live in a given habitat will
thus depend on a balance between the benefits and the costs. If there are few
predators in the habitat, little advantage will be gained from living in large
groups; the costs of doing so will therefore push group size towards a minimum
value. In habitats with many predators, the benefits of large groups will
outweigh the costs, and group size will tend to increase despite the
disadvantages incurred by the animals.
767
SOCIAL LIFE
THE STRUCTURE OF SOCIAL GROUPS
Primate groups differ from those of other animals in two important respects.
One is the intensity of their social bonding; the other is in the extent to which
alliances are used to minimize the costs of group-living.
Grooming is the main way of cementing social relationships in primates, and
monkeys and apes may devote up to 20 per cent of their total waking time to
grooming with each other. The actual amount of time spent on social activities
seems to be directly related to the size of the group: even within species,
animals that live in larger groups devote a higher proportion of their day to
social interaction than those that live in smaller groups. Although grooming has
an obvious hygienic function, individuals of the more social species groom one
another far more than is required to keep the fur clean and free of parasites.
Quite how grooming serves to maintain relationships is far from clear,
though it is now known that grooming increases the production of endogenous
opiates (the brain's own painkillers) (Keverne et al. 1989). Grooming may also
provide an excuse for animals to spend time in close proximity and thus to get
to know each other better. Familiarity is an important requirement for coalition
partners because the value of an ally in a conflict depends on his or her
reliability, and knowledge of another individual's reliability can only come
through repeated interaction.
Although coalitions are known to occur in other animals besides primates
e.g. wns: Pac er an Pusey ; swans: cott , most o t ese are
straightforward relationships based on immediate mutual advantage (for a
review, see Harcourt 1989). The coalitions of higher primates (monkeys and
apes) seem to differ from these in three key respects.
First, coalitionary relationships are established long before they are needed.
Whether grooming partners are more likely to become allies because they
spend so much of their time together or whether monkeys and apes deliberately
groom with those who might be the most profitable allies at some future time
remains uncertain. However, there is at least some evidence to suggest that even
juvenile baboons are aware of who the best allies are in that they actively seek to
establish grooming relationships with just those individuals (Cheney 1977).
Second, coalitionary relationships among primates are often directed
towards minimizing the costs of group-living rather than-as is usual among
other mammals-simply enabling animals to gain immediate access to a
resource. Among gelada baboons, for example, females form coalitions whose
primary function is to reduce the levels of harassment that they inevitably
suffer while living in groups (Dun bar 1980, 1989b). Coalitions make it possible
for the females to minimize these costs and thereby to remain together and gain
the primary advantage that groups are intended to serve (probably the
reduction of predation risk: Dunbar 1986).
Third, the coalitions of primates often involve the exploitation of third-
party power relationships, whereas those of non-primates tend to be based on
768
mutual interest in a common resource (Harcourt 1989). In other words,

monkeys and apes compete for allies, not just for resources (see for example
Seyfarth 1977, 1983). Moreover, these allies are recruited to provide assistance
not just in the heat of the moment (as is the case in other mammals and birds),
but for use at some uncertain and unspecified future time: primates anticipate
the need for coalitionary support (Dunbar 1988). Higher primates exhibit
many other behavioural strategies that suggest that they can evaluate the power
differentials between other individuals (see Kummer 1982, Byrne and Whiten
1988) as well as recognize higher-order aspects of relationships of a more
general nature (e.g. recognizing that A's relationship to B is similar to C's
relationship to me: see Cheney and Seyfarth 1986, 1990).
Given the relative sophistication of the coalitionary relationships
characteristic of primates, it should not be too surprising to find that the more
social primates (i.e. baboons, macaques and chimpanzees) also have
mechanisms that allow them to keep alliance relationships going even after they
have been destabilized. Many studies have demonstrated that an animal's
willingness to go to an ally's aid depends critically on its perception of the risks
involved. Macaques and baboons, for example, will not support their allies in a
fight if the opponent can call on more powerful allies or if the likelihood of
losing the fight anyway is high (Datta 1983, Chapais 1983, Netto and van Hooff
1986). De Waal (1989, de Waal and van Roosmalen 1979) has pointed out that
reneging on an alliance in this way weakens a coalition and makes it less likely
t at t e eserte a y wt support Its partner w en t at m IVI ua n s ttse m
a similar situation on some future occasion. De Waal has suggested two
behavioural processes (termed 'reconciliation' and 'consolation') which
macaques and chimpanzees use to restore the equilibrium in a destabilized
relationship. Both processes involve behaviours such as approaching and
putting an arm around the deserted ally. Reconciliation occurs when the two
members of an alliance have themselves been involved in a fight; consolation
occurs when one member has failed to support its ally when the latter was
involved in a fight with a third party (and especially if the ally lost the
encounter).
Such observations have led to the suggestion that primates owe their
unusually large brains (and thus greater intellectual abilities) to the need to
manage complex social relationships (see Jolly 1966, Humphrey 1976). This,
the so-called 'social intellect' or 'Machiavellian intelligence' hypothesis, has
gained considerable ground in recent years at the expense of its main rival, the
more traditional view that primates' high intelligence is mainly ecological in
function. Indeed, comparisons across species reveal a simple linear relationship
between relative neocortex size and group size in primates as a whole
(Sawaguchi and Kudo 1990, Dunbar 1992).
One interpretation of these results is that there is a cognitive limitation on
primates' abilities to hold social groups together that is directly dependent on
aspects of brain size (and in particular on neocortex volume). The constraint
769
SOCIAL LIFE
appears to lie in the brain's ability to process information (in this case,
specifically information about social phenomena). This in turn suggests that,
among primates, the evolution of larger group sizes under environmental or
other selection pressures was dependent on the evolution of the larger brains
necessary to service the relationships involved. (Note that the nature of the
relationships is crucial in this respect: antelope can form very much larger
groups, but these are unstable aggregations that are very different from the
highly structured congregations of the higher primates.)
If we use the regression equation derived from monkeys and apes to predict
group sizes in anatomically modern humans, we obtain a figure of around 150.
Evidence from hunter-gatherer societies does in fact suggest a level of grouping
of about this size, corresponding among sedentary peoples to the typical village
and among nomadic peoples to the regional band (Dunbar 1993). In addition,
there is considerable sociological evidence to suggest that, even in modern
Western societies, the number of people with whom an individual interacts on a
regular basis is such as to give rise to groups (or extended networks) of about
this size.
These observations have interesting implications for the evolution of
language. As we have seen, relationships in primate groups are cemented by
social grooming, and the larger the group, the greater the amount of time that is
spent grooming (Dunbar 1990). It is not entirely clear why this should be so,
and several different mechanisms have been suggested. What is clear, however,
IS t at t e e ort t at as to e mveste m socta groommg mcreases wtt group
size. If we use this relationship to determine the grooming time that would be
required for modern humans to service relationships within groups of the size
expected on the grounds of their neocortex volume we find that they would
have to spend more than a third of their waking hours in social grooming. For
organisms that also have to make a living in the world, this is not a feasible
proposition. Language is an ideal solution to the resulting problem of time-
budgeting as it allows an individual to engage in extensive time-sharing in ways
that are not possible with grooming. One cannot walk and groom at the same
time, nor can one groom more than one individual at a time, but it is possible to
walk and talk and to hold several interlocutors in conversation simultaneously.
Thus it looks as though the capacity for language might have evolved to solve a
problem of social bonding in the large groups in which our ancestors were
obliged to live by some (as yet undetermined) aspect of their ecology.
The question of why our ancestors should have been obliged to live in such
large groups remains unclear. Our current understanding is that primates live
in groups either to defend food resources or for protection against predators,
with the latter certainly being the more significant factor in the case of
terrestrial species inhabiting relatively open environments. However, no
primate species has a mean group size of more than about a hundred
individuals (much lower than that predicted for anatomically modern humans).
Since many of these live in habitats where the risk of predation is high, it is
770
difficult to see how primates as large as the later hominids could conceivably
have been so much more at risk from predators as to require groups that are
more than twice the size of most other open-country primates.
Whatever factor lay behind the evolution of very large groups in our
ancestors, it must clearly have been some feature of their ecology that is not
shared by other primates. Defence against other human groups under
conditions of rapidly rising population density has been suggested as one
possibility (e.g. Alexander 1989); another lies in the evolution oflarge-scale co-
operative hunting; a third possibility might have been the need to share access
to key resources (such as waterholes or dry-season foraging areas) when
individual foraging parties were otherwise obliged to disperse over very wide
areas. Of these, the hunting hypothesis can almost certainly be ruled out
because large-scale hunting is not observed in the archaeological record until
well after the evolution oflarge brain size (see also Wynn 1988).
What is certain, however, is that this increase in group size must have been a
relatively late development. Brain size for the australopithecines lies well
within the range for extant great apes, suggesting that australopithecines
probably did not have group sizes significantly larger than those observed in
modern chimpanzees (i.e. 50 to 100 individuals). Although there is an increase
in relative brain size with the appearance of the first members of the genus
Homo (i.e. Homo habilis and H. erectus), the real jump in brain size does not
come about until the appearance of our own species (Homo sapiens)
SOME SPECIAL ORGANIZING PRINCIPLES

Sociobiology may be said to owe its origins to the attempt to solve the problem
of altruism. In this context, altruism is defined in a rather specific way as
behaviour that increases another individual's fitness (i.e. its relative
contribution to the species' future gene pool) at the expense of the altruist's
fitness. If the altruist does not incur such a genetic cost, the behaviour does not
count as biological altruism. Paradoxically, perhaps, giving the price of a meal
to a beggar is unlikely to count as altruism (unless the altruist is another
beggar), but committing suicide would do if it resulted in an improvement in
the beneficiary's ability to reproduce (e.g. by reducing the competition for
scarce resources or realizing a large insurance claim). That altruistic behaviour
clearly exists (and thus must have evolved) is puzzling from a Darwinian point
of view: on the face of it, any gene (or meme) for altruism would be heavily
selected against and would inevitably be eradicated from the population each
time it appeared as a new mutation.
The solution arrived at by Hamilton ( 1964) depended on the recognition
that an individual can contribute copies of its genes to the next generation
either by reproducing itself or by enhancing the reproduction of its relatives.
Because relatives share a proportion of their genes (by virtue of their descent
771
SOCIAL LIFE
from a common ancestor), any copy of a shared gene that is passed on by a

relative is just as good (from the gene's point of view) as any copy passed on by
the altruist. The number of copies of a given gene contributed to the next
generation by both routes is referred to as the gene's 'inclusive fitness' and acts
as the accounting basis for assessing the evolutionary value of alternative
strategies. (Note that individuals are also commonly said to have an inclusive
fitness; though strictly speaking incorrect, this is a convenient shorthand used
by most biologists.) A gene for altruism can survive and prosper in a Darwinian
world even if the altruistic action results in the death of the altruist, providing
the act of altruism results in more copies of that gene reaching the next
generation via the assisted relative than the altruist would have been able to
contribute by its own reproduction had it lived. The mathematical conditions
under which this can occur (known as 'Hamilton's rule' for the spread of
altruism) are rather precise; together with the associated definition of inclusive
fitness and related concepts, they constitute the theory of 'kin selection'.
Kin selection has come to be seen in some quarters as the central tenet of
sociobiology. One common assumption, for example, is that kin selection
obliges animals (including humans) to behave altruistically towards, or choose
to live in groups only with, their kin. Unfortunately, given the definitions of
inclusive fitness and Hamilton's rule, this cannot be the case. An organism
always has at least two options on how to behave, and it is the balance in their
net pay-offs that determines which option is best. The catch is that the net pay-
o ttse epen s on t e a ance etween t e trect ene Its con erre vta t e
actor's own reproduction and the indirect ones conferred via its relatives. With
the exception of a few cases that have odd genetics (e.g. bees), an individual is
always more closely related to its own offspring than to a relative's offspring;
hence, there will always be a predisposition towards personal reproduction at
the expense of that of relatives (see Dunbar 1983). For this reason, mutualism
(i.e. cases in which all parties benefit from co-operating: see Wrangham 1982) is
probably more important as a driving force behind the evolution of social
groups in animals (and certainly in primates) than is kin selection.
In point of fact, kin selection is not the only Darwinian explanation for the
evolution of altruistic behaviour: there are at least two others, namely reciprocal
altruism (Trivers 1971) and mutualism. Reciprocal altruism allows altruistic
behaviour to evolve even when the parties concerned are unrelated to each
other genetically (indeed, they may even belong to different species) because
the (genetic) 'debt' that the altruist incurs by its behaviour is recouped within
its lifetime. In fact, reciprocal altruism and kin selection differ only in the
length of time over which the 'debt' is recouped: under kin selection, the debt
can be reclaimed in the next generation. Mutualism shortens the time-scale still
further to the point where the return to the altruist is immediate (because both
parties are simultaneously altruist and beneficiary).
What kin selection may do, however, is bias the choice of fellow group-
members in favour of relatives, given that an animal has 'decided' to live in a
772
group in the first place. Even so, kin selection is just one of several evolutionary
forces that contribute to an individual's inclusive fitness. An individual will
often benefit more in terms of inclusive fitness by forming a coalition or living
in a group with an unrelated individual than by doing so with a relative. Thus,
male lions will sometimes form coalitions with non-relatives in order to gain
control over a pride of females (Packer and Pusey 1982).
In fact, the evidence suggests that, rather than responding automatically to
the call of kinship, higher primates (at least) weigh up the relative advantages of
kin and non-kin in a given social context. Cheney (1983), for example, has
shown that high-ranking vervet monkeys tend to form alliances with close
relatives; but low-ranking monkeys tend to prefer alliances with unrelated
dominant individuals to alliances with relatives. For an animal that is already
high-ranking, additional support from other high-ranking allies may be of
marginal benefit, and more may be gained through kin selection by supporting
relatives. For a low-ranking animal, by contrast, little is gained by an alliance
with a relative, since relatives are also likely to be low-ranking; much more is to
be gained by forming an alliance with a high-ranking individual who is likely to
have a significant impact on the ally's dominance rank within the group (and
hence directly on its own ability to reproduce).
The fact that animals do discriminate between relatives and non-relatives
naturally raises important questions about the mechanisms of kin recognition.
Even though animals may be unable to recognize relatives in any direct genetic
sense, m uect cues 1 e amt tanty wt usua y su ce to a ow m se ectwn to
work: evolutionary processes are statistical rather than deterministic and
simply require the balance of probabilities to work in favour of a particular
effect. Given that mammals have to spend time with their mothers and,
perhaps, siblings, it is not hard to see that a simple rule of thumb such as 'Be
altruistic towards more familiar individuals' will often have the same genetic
effect (and therefore be selected for) as the more direct rule 'Be altruistic
towards genetically more closely related individuals'.
There is, however, growing evidence that animals can sometimes recognize
genetic relatives independently of their familiarity with them. Species as
different as Japanese quail and rhesus monkeys have been shown to be able to
discriminate relatives from non-relatives, even though they were separated
from them at birth (Bateson 1983, Wu et al. 1980). Much of the interest in this
context has focused on smell and on the genes of the major histocompatibility
(MHC) complex that provide the basis for our immune system (Yamazaki et al.
1976, 1978). Even in humans, emphasis may be placed on identifying features
that might suggest genetic relatedness. In some cultures, relatives' comments
about newborn babies often emphasize inherited features ('Doesn't he have
Grandma's nose!' (Daly and Wilson 1982)). That newborn babies are, of
course, all but indistinguishable as far as most such features are concerned
makes the fact that such references invariably favour paternal relatives rather
than maternal ones all the more significant: among mammals, only the female
773
SOCIAL LIFE
knows for sure that she is related to the infant. In cases where the father
normally makes a significant contribution to parental care, the need to convince
the father of his genetic relatedness to the offspring becomes important (and
becomes proportionately all the more important when there is a possibility of
doubt). Obviously, if the male does not invest heavily in his wife's offspring,
then the problem does not arise and we would not expect to find comments of
this kind that stress paternity. This, of course, is just what we see in those
societies where promiscuity is high, paternity certainty (the male's knowledge
of his biological fatherhood) low and the avunculate a common practice.
The question of whether, in humans, biological kinship is related to the ways
in which kin are culturally classified has long been a bone of contention.
Evolutionary biologists have been inclined to insist that such a relationship
does exist, mainly on the grounds that cultural kinship classifications are never
entirely arbitrary with respect to biological kinship. Given the statistical nature
of all biological effects, the fact that humans sometimes make some kinship
assignations that have no basis in biological kinship does not, in itself,
invalidate this claim. The question at issue is whether there is so much
misidentification of biological relatives that the evolutionary process would be
undermined.
By contrast, social and cultural anthropologists have usually insisted that
kinship terminology bears little or no relationship to any underlying biological
'reality' (e.g. Sahlins 1976, Bryant 1981; see also Barnard's discussion in the
o owmg arttc e . However, w ere spect tc cases ave een put orwar m
support of this claim, detailed investigation has invariably revealed that
biological kinship does, in fact, underwrite people's behaviour (e.g. Silk 1980,
Hughes 1988).
Hughes's (1988) analyses of many examples of human kinship-naming
patterns are particularly important in this context because he draws attention to
a misconception underlying many interpretations of sociobiological arguments
about kinship and kin selection-one of which even biologists have been guilty.
The key issue from an evolutionary viewpoint is not whom you are most closely
related to, but rather who is most likely to produce offspring that are most
closely related to you. Hence, we need to look at how coefficients of relationship
within a group of individuals map onto those individuals' own future
reproductive prospects. Hughes's mathematical analyses demonstrate that
genetic fitness is maximized not by allying with relatives in proportion to their
degree of relatedness, but by allying with those relatives who will produce the
largest number of most closely related descendants. These will not always be
the individuals who are most closely related to you in absolute terms; indeed,
they will seldom be the adults in the population. Rather, they will tend most
often to be the older members of the offspring generation (i.e. those
approaching or just past puberty). In an analysis of Bryant's (1981) own data
for a rural Tennessee community, for example, Hughes was able to show that
declared family allegiances (which genuinely bear little relationship to direct
774
genetic relatedness, as Bryant rightly noted) in fact fit rather closely to a pattern
of relatedness concentrating on twelve focal groups of siblings in the current
offspring generation.
Even though kinship classifications may be underpinned by genetic
relatedness in this way, focusing one's kinship allegiances on the offspring
generation creates a serious problem: the offspring generation is never the same
in two consecutive time periods, because offspring continuously age and join
the adult cohort. How, then, is one to establish kin group stability over time?
Hughes points out that the obvious solution is, in fact, to refer the kinship
group backwards to some ancestor, since the ancestor's status will always
remain constant through time, thereby providing a firm point on which to
anchor the pedigree. What is particularly interesting in this context is that it
makes very little difference whether the members of the kinship group are
themselves directly related to that 'ancestor' or not, since beyond about four
generations removed in time the coefficients of relationship between two
individuals are so low that they are, to all intents and purposes, unrelated.
Indeed, it makes little difference whether that ancestor actually existed or not:
the sun, the moon and Mother Earth are as functional in this context as one's
great-great-great-grandfather.
ANIMAL VERSUS HUMAN SOCIETIES

evera Important essons ave een earne y 10 ogtsts over t e past two
decades. One is that animal societies are not all of a piece. Species that belong to
different groups of organisms may differ radically both in the way their
societies are organized and in the biological bases that underlie them. While
invertebrates, for example, may reasonably be considered to be genetically
determined automata, mammals (and a fortiori primates) cannot. None the less,
it may still be the case that the general evolutionary principles that underlie the
one also underlie the other. Sociobiologists are willing to switch from bees to
humans within the same sentence, not because they believe that the behaviour
of both bees and humans is determined by the same set of genetic molecules,
but because the same functional considerations apply universally.
At the same time, biologists have learned to be very cautious of attempts to
argue analogically from one species to another. That geese, for example, should
behave in a particular way does not mean to say that humans should do so too.
Indeed, there is little justification for supposing that generalizations of this
kind can be made even across closely related species. This is most clearly the
case in the primates where behavioural flexibility is so great that adjacent
groups within the same population may be organized on quite different
principles. This highlights an important distinction between 'deep structure'
and 'surface structure' in the context of social organization. The particular
pattern of relationships that the observer notes in any given group is a
consequence of the patterning of interaction among the animals, and this in
775
SOCIAL LIFE
turn is founded on sets of rules that the animals use in order to identify the
most profitable social partners. It is these deep structural rules that turn out to
be universally true for all species, but the particular social pattern that a given
deep structural rule produces depends crucially on the context in which the
animals have to apply it. In different contexts, the application of the same rule
may yield completely contradictory expectations about the optimal behavioural
strategy. Thus, vervet monkeys appear to operate with the rule 'Form those
alliances that will allow you to maximize your chances of contributing to the
species' gene pool'. Which particular alliance partners best allow one to achieve
that goal is, however, very different for high-ranking and low-ranking
individuals (Cheney 1983). Similarly, gelada baboon females appear to operate
with a similar rule, but the choice of preferred alliance partner depends upon
who is available (Dun bar 1984). In part, this is a consequence of the fact that
what animals actually do is almost always a compromise between what they
would really like to do (in an ideal world) and what the demographic and
ecological context allows them to do.
Studies of animals can tell us a great deal about the underlying processes in
human societies. But, as biologists have long been aware, we cannot learn
anything useful by analogical reasoning: studies of non-human animals can tell
us little about the fine details of human social behaviour. Much of that may well
be culturally determined and owe its origins as much to cultural drift
(analogous to genetic drift, itself a perfectly respectable concept in evolutionary
10 ogy as to se ectwn an a aptatwn. However, e avwur ecomes
sociobiologically interesting as soon as it has some influence on the rate at
which the units of selection (either genes or their cultural analogues, 'memes',
in most real world contexts) replicate themselves. From an evolutionary point
of view selection on memes in the m emetic universe is no less interesting than
selection at the genetic level. Indeed, it is quite likely that many cultural
institutions actually comprise a number of facets that are subject to quite
different evolutionary processes. An example might be the need for a religious
system with an omnipotent deity. Belief in a deity of this kind may well help
individuals to survive and reproduce more effectively because it gives
coherence to an apparently chaotic world. Such a belief might be selected for at
the genetic level. But the particular choice of deity to fill this role may have
little or no genetic import: Allah, God, Zeus and the Great Spirit in the Sky
might all be equally good candidates. If they genuinely do not differ in their
consequences for the genetic fitnesses of those who hold them, choice of deity
will then be influenced at the memetic level by a process analogous to genetic
drift. Alternatively, they may be under memetic (but not genetic) selection for
their 'fit' in relation to other aspects of the cultural system: the meme 'Allah'
may not mesh well with key elements of the set of cultural institutions of which
it tries to become a part, and thus be selected against in favour of a more
compatible alternative.
776
The crucial lesson here is that we have to approach evolutionary issues in a

very different (and very much more sophisticated) way from that which has
often been the case hitherto. This is especially so with respect to the kinds of
evolutionary explanations that have been prevalent in the social sciences. As
Ingold (1986) points out, most of these owe their origins to Spencerian and not
to Darwinian views of evolution. What makes human sociobiology particularly
interesting is the sheer complexity of the biological system once cultural
processes are introduced. Its analysis requires a much deeper understanding of
the many intrinsic and extrinsic factors that influence an organism's ability to
reproduce successfully.
Equally, it is important to remember that questions of the kind that
commonly interest social and cultural anthropologists are very different from
those that interest evolutionary biologists. As 'Tinbergen's Four Why's' should
remind us, this does not mean that one set of interests must be right and the
other wrong. Questions about origins or function, for example, cannot be
brushed under the carpet merely because they are inconvenient or difficult to
answer. In most cases, it is clear that issues that interest one group of scholars
are simply not relevant to the interests of the other group. Nevertheless, we will
not be said to have achieved a satisfactory understanding of our world until
both sets of questions have been answered.
Abbott, D.H. (1984) 'Behavioural and physiological suppression of fertility in

subordinate marmoset monkeys', American Journal of Primatology 6:169-86.
Aiello, L.C. and Dunbar, R.I.M. (1993) 'Neocortex size, group size and the evolution
of language', Current Anthropology 34:184--93.
Alexander, R.D. (1989) 'Evolution of the human psyche', in P.Mellars and C. Stringer
(eds) The Human Revolution, Edinburgh: Edinburgh University Press.
Bateson, P.P.G. (1983) 'Optimal out breeding', in P.Bateson (ed.) Mate Choice,
Bertram, B.C.R. (1978) 'Living in groups: predators and prey', in ].Krebs and N.
Davies (eds) Behavioural Ecology, Oxford: Blackwell.
Boyd, R. and Richerson, P. (1986) Culture and the Evolutionary Process, Chicago:
Chicago University Press.
Bryant, F.C. (1981) We're All Kin: A Cultural Study of a Mountain Neighbourhood,
Knoxville: University of Tennessee Press.
Busse, C. (1977) 'Chimpanzee predation as a possible factor in the evolution of red
colobus monkey social organisation', Evolution 31:907-11.
Byrne, R. and Whiten, A. (eds) (1988) Machiavellian Intelligence, Oxford: Oxford
University Press.
Caro, T. and Bateson, P. (1986) 'Organisation and ontogeny of alternative tactics',
Animal Behaviour 34:1483-99.
Cavalli-Sforza, L. and Feldman, M. ( 1980) Cultural Transmission and Evolution,
Princeton: Princeton University Press.
777
SOCIAL LIFE
Chapais, B. (1983) 'Dominance, relatedness and the structure offemale relationships in

rhesus monkeys', in R.Hinde (ed.) Primate Social Relationships, Oxford: Blackwell.
Charles-Dominique, P. (1977). Ecology and Behaviour of Nocturnal Primates, New York:
Columbia University Press.
Cheney, D.L. (1977) 'The acquisition of rank and the development of reciprocal
alliances among free-ranging immature baboons', Behavioural Ecology and
Sociobiology 2:303-18.
--(1983) 'Extrafamilial alliances among vervet monkeys', in R.Hinde (ed.) Primate
Social Relationships, Oxford: Blackwell.
Cheney, D.L. and Seyfarth, R.M. (1983) 'Nonrandom dispersal in free-ranging vervet
monkeys: social and genetic consequences', American Naturalist 122: 392-412.
--,--(1986) 'The recognition of social alliances by vervet monkeys', Animal
Behaviour 34:1722-31.
--,--(1987) 'Influence of intergroup competition on survival and reproduction of
female vervet monkeys', Behavioural Ecology and Sociobiology 21:375-86.
--,-(1990) How Monkeys See the World, Chicago: Chicago University Press.
Daly, M. and Wilson, M. (1982) 'Whom are newborn babies said to resemble?' Ethology
and Sociobiology 1:301-9.
Datta, S. (1983) 'Relative power and the maintenance of rank', in R.Hinde (ed.) Primate
Dawkins, R. (1976) The Selfish Gene, Oxford: Oxford University Press.
--(1982) 'Universal Darwinism', in D.Bendall (ed.) Evolution from Molecules to Men,
Dittus, W.P.J. (1986) 'Sex differences in fitness following a group take-over among
toque macaques: testmg mo e s o soCia evo utwn,
Sociobiology 19:257-66.
Dunbar, R.I.M. (1980) 'Determinants and evolutionary consequences of dominance
among female gelada baboons', Behavioural Ecology and Sociobiology 7:253-65.
--(1982a) 'Adaptation, fitness and the evolutionary tautology', in King's College
Sociobiology Group (eds) Current Problems in Sociobiology, Cambridge: Cambridge
University Press.
--(1982b) 'lntraspecific variations in mating strategy', in P.Klopfer and P.Bateson
(eds) Perspectives in Ethology vol. 5, New York: Plenum Press.
--(1983) 'Lifehistory tactics and alternative strategies of reproduction', in P. Bateson
(ed.) Mate Choice, Cambridge: Cambridge University Press.
--(1984) Reproductive Decisions, Princeton: Princeton University Press.
--(1986) 'The social ecology of gelada baboons', in D.Rubenstein and R. Wrangham
(eds) Ecological Aspects of Social Evolution, Princeton: Princeton University Press.
--(1988) Primate Social Systems, London: Chapman & Hall.
--(1989a) 'Social systems as optimal strategy sets: the costs and benefits of sociality',
in V.Standen and R.Foley (eds) Comparative Socioecology, Oxford: Blackwell.
--(1989b) 'Reproductive strategies of female gelada baboons', in A.Rasa, C.Vogel
and E.Voland (eds) The Sociobiology of Sexual and Reproductive Strategies, London:
Chapman & Hall.
--(1990) 'The functional significance of social grooming in primates', Folia
Primatologica 57:121-31.
--(1992) 'Neocortex size as a constraint on group size in primates', Journal of
Human Evolution 20:469-93.
778
--(1993) 'Co-evolution of neocortex size, group size and language in the human
lineage', Behavioral and Brain Sciences (in press).
French, J.A., Abbott, D.H. and Snowdon, C.T. (1984) 'The effects of social
environment on oestrogen excretion, scent marking and sociosexual behaviour in
tamarins (Saguinus oedipus)', American Journal of Primatology 6, 155-67.
Goodall, J. (1965) 'Chimpanzees of the Gombe Stream Reserve', in I.DeVore (ed.)
Primate Behavior, New York: Holt Rinehart and Winston.
--(1986) The Chimpanzees ofGombe, Cambridge: Mass.: Harvard University Press.
Hamilton, WD. (1964) 'The genetical evolution of social behaviour, I, 11', Journal of
Theoretical Biology 7:1-52.
Harcourt, A.H. (1989) 'Social influences on competitive ability: alliances and their
consequences', in V.Standen and R.Foley (eds) Comparative Socioecology, Oxford:
Blackwell.
Hegner, R.E., Emlen, S.T. and Demong, N.J. (1982) 'Spatial organisation of white-
fronted bee-eaters', Nature, London 298:264-6.
Hinde, R.A. (1976) 'Interactions, relationships and social structure', Man (N.S.) 11: 1-
17.
Hughes, A. (1988) Evolution and Human Kinship, Oxford: Oxford University Press.
Humphrey, N.K. (1976) 'The social function of intellect', in P.Bateson and R.Hinde
(eds) Growing Points in Ethology, Cambridge: Cambridge University Press.
Huxley, J.S. (1942) Evolution: The Modern Synthesis, London: Alien & Unwin.
lngold, T. (1986) Evolution and Social Lift, Cambridge: Cambridge University Press.
Jolly, A. (1966) 'Lemur social behaviour and intelligence', Science 153:501-6.
Kawai, M., Dunbar, R., Ohsawa, H. and Mori, U. (1983) 'Social organisation of gelada
a oons: soCia umts an e Imtwns , rtmates
Keverne, E.B., Martenz, N.D. and Tuite, B. (1989) 'Beta-endorphin concentrations in
cerebrospinal fluid of monkeys are influenced by grooming relationships',
Psychoneuroendocrinology 14:155-61.
Kleiman, D. (1977) 'Monogamy in mammals', Quarterly Review of Biology 52:39-69.
King, J.A. (1955) 'Social behaviour, social organisation and population dynamics in a
blacktailed prairiedog town in the Black Hills of South Dakota', Contributions of the
Laboratory of Vertebrate Biology, University of Michigan, Ann Arbor 67:1-123.
Kruuk, H. (1982) The Spotted Hyaena, Chicago: Chicago University Press.
Kummer, H. (1968) Social Organisation of Hamadryas Baboons, Easel: Karger.
--(1982) 'Social knowledge in free-ranging primates', in D.R.Griffin (ed.) Animal
Mind-Human Mind, Berlin: Springer.
--(1984) 'From laboratory to desert and back: a social system of hamadryas
baboons', Animal Behaviour 32:965-71.
Lorenz, K. (1960) On Aggression, New York: Harcourt Brace and World.
Maynard Smith, J. (1982) Evolution and the Theory of Games, Cambridge: Cambridge
University Press.
Maynard Smith, J. and Price, G.R. (1973) 'The logic of animal conflict', Nature,
London 246:15-18.
Milardo, R.M. (1988). 'Families and social networks: an overview of theory and
methodology', in R.Milardo (ed.) Families and Social Networks, Newbury Park,
Calif.: Sage.
Morris, D. (1967) The Naked Ape, London: Constable.
779
SOCIAL LIFE
Moss, C. and Poole, J.H. (1983) 'Relationships and social structure of African
elephants', in R.Hinde (ed.) Primate Social Relationships, Oxford: Blackwell.
Netto, W.J. and van Hooff, J. (1986) 'Conflict interference and the development of
dominance relationships in immature Macaca foscicularis', in ].Else and P.C.Lee
(eds) Primate Ontogeny, Cognition and Social Behaviour, Cambridge: Cambridge
University Press.
Packer, C. and Pusey, A. (1982) 'Co-operation and competition within coalitions of
male lions: kin selection or game theory?', Nature, London 296:740-2.
Robertson, J.M. Y ( 1986) 'On the evolution of pigtail macaque societies', PhD. thesis,
University of Cambridge,
de Ruiter, J. (1986) 'The influence of group size on predator scanning and foraging
behaviour of wedgecapped capuchin monkeys (Cebus olivaceus )', Behaviour 98: 240-
58.
Sahlins, M. (1976) The Use and Abuse of Biology, Ann Arbor: University of Michigan
Press.
Sawaguchi, T. and Kudo, H. (1990) 'Neocortical development and social structure in
primates', Primates 31:283-90.
van Schaik, C.P. (1983) 'Why are diurnal primates living in groups?', Behaviour 87:
120-44.
van Schaik, C.P. and Dunbar, R.I.M. (1990) 'The evolution of monogamy in large
primates: a new hypothesis and some crucial tests', Behaviour 119:30-62.
van Schaik, C.P., van Noordwijk, M.A., Wasone, M.A. and Sitriono, E. (1983) 'Party
size and early detection in Sumatran forest primates', Primates 24:211-21.
Schaller, G. (1972) The Serengeti Lion, Chicago: Chicago University Press.
Seyfarth, R.M. (1977) 'A model of social grooming among adult female monkeys',
Journal of Theoretical Biology 65:671-98.
--(1983) 'Grooming and social competition in primates', in R.Hinde (ed.) Primate
Silk, J.B. ( 1980) 'Adoption and kinship in Oceania', American Anthropologist 82: 799-
820.
Skinner, B.F. (1938) The Behaviour of Organisms: An Experimental Analysis, New York:
Appleton-Century-Crofts.
Smuts, B.B., Cheney, D.L., Seyfarth, R.M., Wrangham, R.W. and Struhsaker, T. T.
(eds) (1987) Primate Societies, Chicago: Chicago University Press.
Stacey, P.B. (1986) 'Group size and foraging efficiency in yellow baboons', Behavioural
Ecology and Sociobiology 18:175-87.
Tinbergen, N. (1963) 'On the aims and methods of ethology', Zeitschrift fiir
Tierpsychologie 20:410-33.
Trivers, R.L. (1971) 'The evolution of reciprocal altruism', Quarterly Review of Biology
46:35-57.
de Waal, F. (1982) Chimpanzee Politics, London: Jonathan Cape.
--(1989) Peacemaking Among Primates, Cambridge, Mass.: Harvard University
Press.
de Waal, F. and van Roosmalen, A. (1979) 'Reconciliation and consolation among
chimpanzees', Behavioural Ecology and Sociobiology 5:55-66.
780
Watson, J.B. (1919) Psychology from the Standpoint of a Behaviourist, Philadelphia:

Lippincott.
Watts, D. (1985) 'Relations between group size and composition and feeding
competition in mountain gorilla groups', Animal Behaviour 32:72-85.
Williams, M.B. (1970) 'Deducing the consequences of evolution: a mathematical
model', Journal of Theoretical Biology 29:343-85.
Wilson, E.O. (1975) Sociobiology: The New Synthesis, Cambridge, Mass.: Harvard
University Press.
Wrangham, R.W. (1977) 'Behaviour of feeding chimpanzees in the Gombe National
Park, Tanzania', in T.Clutton-Brock (ed.) Primate Ecology, London: Academic
Press.
--(1980) 'An ecological model of female-bonded primate groups', Behaviour 75:
262-300.
--(1982) 'Mutualism, kinship and social evolution', in King's College Sociobiology
Group (eds) Current Problems in Sociobiology, Cambridge: Cambridge University
Press.
--(1983) 'Ultimate factors determining social structure', in R.Hinde (ed.) Primate
--(1987) 'Evolution of social structure', in B.B.Smuts, D.L.Cheney, R.M. Seyfarth,
R.W.Wrangham and T.T.Struhsaker (eds) Primate Societies, Chicago: Chicago
University Press.
Wrangham, R.W. and Smuts, B. B. (1980) 'Sex differences in the behavioural ecology of
chimpanzees in the Gombe National Park, Tanzania', Journal of Reproduction and
Fertility, Supplement 28: 13-31.
u, ., o mes, . ., e 1na, . . an ac ett,
Macaca nemestrina', Nature, London 285:225-7.
Wynn, T. (1988) 'Tools and the evolution of human intelligence', in R.Byrne and A.
Whiten (eds) Machiavellian Intelligence, Oxford: Oxford University Press.
Yamazaki, K., Boyse, A.E., Mike, V., Thaler, H.T., Mathieson, B.J., Abbott,J., Boyse,J.,
Zayas, Z.A. and Thomas, L. (1976) 'Control of mating preferences in mice by genes
in the major histocompatibility complex', Journal of Experimental Medicine
144:1324-35.
Yamazaki, K., Yamaguchi, M., Andrews, P., Peake, B. and Boyse, E. (1978) 'Mating
preference of F2 segregants of crosses between MHC-congenic mouse strains',
Immunogenetics 6:253-9.
FURTHER READING
Alexander, R. (1979) Darwinism and Human Affoirs, Seattle, University of Washington
Press.
Chagnon, N.A. and Irons, W. (eds) (1979) Evolutionary Biology and Human Social
Behavior, North Scituate, Mass.: Duxbury Press.
Daly, M. and Wilson, M. (1982) Sex, Evolution and Behavior, North Scituate, Mass.:
Duxbury Press.
Dunbar, R.I.M. (1988) Primate Social Systems, London: Chapman & Hall.
Dawkins, R. (1976) The Selfish Gene, Oxford: Oxford University Press.
781
SOCIAL LIFE
Hinde, R.A. (1987) Individuals, Relationships and Culture, Cambridge: Cambridge

University Press.
Quiatt, D. and Reynolds, V. (1993) Primate Behaviour: Information, Social Knowledge
and the Evolution of Culture, Cambridge: Cambridge University Press.
Rasa, A., Vogel, C. and Voland, E. (eds) (1989) The Sociobiology of Sexual and
Reproductive Strategies, London: Chapman & Hall.
Rubenstein, D.I. and Wrangham, R. (1986) Ecological Aspects of Social Evolution,
Princeton: Princeton University Press.
Smuts, B.B., Cheney, D., Seyfarth, R., Wrangham, R. and Struhsaker, T. (eds) (1987)
Primate Societies, Chicago: University of Chicago Press.
Standen, V. and Foley, R. (eds) (1989) Comparative Socioecology Oxford: Blackwell.
Wilson, E.O. (1980). Sociobiology, abridged edn, Cambridge, Mass.: Harvard
University (Belknap) Press.
782
28
RULES AND PROHIBITIONS: THE

FORM AND CONTENT OF HUMAN
KINSHIP
Alan Barnard
There was a time not long ago when kinship firmly commanded the highest
position among the theoretical realms of anthropology. This is probably no
longer true, but nor is it true that kinship is an idea with a past and no future.
Kms Ip remams as Important as ever as an e ement o uman society, an new
perspectives within the social and biological sciences offer opportunities to
reconsider some old arguments in a new light and to look forward to new
debates and new ideas.
The anthropological study of kinship has traditionally been divided into
three broad areas: group structure (including descent and residence), alliance
(relations through marriage), and the classification of relatives. Rules and
prohibitions are marked out within each of these areas, and the very existence
of such rules engenders an overlap between them, an overlap which is itself,
arguably, the very essence of 'social structure'. The purposes of this article are
first to highlight the significance of such rules and prohibitions in the
foundation of human society, and second to look at recent developments and
reconsider some common preconceptions about kinship, of which some but
certainly not all were inherent in the old debates. My focus is on topics which I
think have the greatest relevance for the future study of humankind, and not
necessarily on those most significant in the history of kinship studies. Thus, for
example, the transformational analysis of relationship terminologies is not
treated at all (for a review, see Borland 1979), 'prescription' and 'elementary
structures' will be treated only in passing, while the debate concerning the
meaning and applicability of biological notions of 'kinship' receives greater
emphasis.
783
SOCIAL LIFE
WHAT'S SO SPECIAL ABOUT KINSHIP ANYWAY?
Although the significance of kinship in anthropological discourse has been on

the wane during the last twenty years, there is little doubt that it has been
treated as the single most important aspect of society, as far as anthropological
theory is concerned, throughout the history of the discipline. There are several
reasons for this.
First, whether in respect of the incest taboo or in respect of nurturing and
socializing children and thereby establishing social groupings, kinship
simultaneously marks a boundary and a bridge between non-human and
human social orders. The incest taboo is pre-eminently human, although
selective mating is found among non-human primates too. Extensive
socialization and the formation of social groups through ties of relationship is
an attribute of most primate societies, but the ideological recognition of such
ties is commonly thought to define the essence of human social organization.
Second, kinship has long been conceived as somehow logically distinct from
other aspects of society. Beattie ( 1964: 102), for example, describes kinship as
'the idiom through which certain kinds of political, jural, economic, etc.,
relations are talked about and thought about'. He portrays kinship almost as a
contentless form which humans employ to create social relations. Although
politics, economics, etc. may function in a similar way (and in my estimation,
frequently do), kinship has come to be perceived as different, or at least as of
pnme stgm tcance, m t IS regar .
Third, in the history of anthropology, it was through kinship that the variety
of human conceptual systems and the internal logic of diverse social structures
came to be recognized. This was as true of evolutionists such as M organ ( 1871)
as it was of functionalists such as Radcliffe-Brown ( 1941 ). There has indeed
been a loosely relativistic streak in most brands of social anthropology ever
since the acceptance of monogenism in the late nineteenth century. By
definition, monogenism entails an acceptance of humankind's common origin,
and one consequence was that anthropologists had to learn to explain
differences in kinship structure as variations on a common theme. In the
nineteenth century it was supposed that the differences represented stages in a
single sequence of evolutionary development, through which all humankind
was destined to pass; but in the twentieth century different systems came to be
seen more as alternative cultural imprints on a 'blank slate' of universal human
biology, or as variants of a limited set of logically possible arrangements for
organizing human relationships. I vividly remember my own, somewhat naive,
conversion to relativistic anthropology, which was stimulated more than
anything else by the revelation that the Iroquois have an 'Iroquois' relationship
terminology structure just as logical as, but profoundly different from, my own
English-language one, which in turn is formally of the 'Eskimo' type. Whether
the logic of such systems is best understood within evolutionist, interpretivist,
or structuralist paradigms is one of the great questions of anthropology.
784
RULES AND PROHIBITIONS: HUMAN KINSHIP
Fourth, on a not unrelated point, the reason why anthropologists have been
especially prone to 'discovering' structural parallels and contrasts within the
realm of kinship is that kinship is the most transparently structured of all
realms of human life. It is not merely that kinship experts have devised complex
notation systems and other technical devices beyond the ken of, say, specialists
in religion or politics. The logical primacy of the genealogical grid in kinship
studies gives specialists a tool for cross-cultural comparison of a kind that is not
available in other fields. Not since the Romans first recognized the equivalence
of their gods and goddesses to those in the Greek pantheon have Western
minds come up with such a clear-cut datum point for structural comparison-
or, if one prefers, cultural 'translation'-as the notion of 'genealogical
relationship'.
Fifth, and to turn full circle, kinship studies have promoted a quasi-fallacy
that kinship is built on models that are more 'real' than those of religion, of
economics, of politics, or of law. In truth, kinship is no more real than these
other notions; it is rather that kinship structures are more apparent cross-
culturally than the structures identified and studied by specialists in religion,
economics and politics. The apparent cross-cultural 'reality' of kinship stems
from an erroneous equation of 'kinship' with 'biology'. Kinship, for virtually
all human societies, is built upon a putative biological foundation.
Nevertheless, this is a cultural phenomenon and not per se a biological one. The
incest taboo, the family, and the genealogical grid are substantive universals of
KINSHIP AND BIOLOGY
Humans and animals

Is kinship a distinctively human attribute? Dunbar, in the previous article,
stressed the scope of comparison between human and non-human sociality,
including what biologists often refer to as (non-human) 'kinship'. In another
key paper, Fox ( 197 5) considered the specific relationship between human and
non-human primate 'kinship'. Fox's argument is in essence one of continuity:
human kinship is not, in his biologically oriented view, merely a cultural
construct, but is rooted in primate behaviour. Thus, according to Fox non-
human primates possess all the important rudiments of human kinship. What
makes them different is that some non-human primate societies have 'alliance'
without 'descent' (namely those which live in single-male groups), while others
have 'descent' without 'alliance' (namely those which live in multi-male
groups). Human kinship systems invariably have both alliance and descent.
Such arguments may seem cogent, but they hold only if we accede to the
premiss that human society is to be comprehended primarily in behavioural
rather than cultural terms. Many social or cultural anthropologists would
785
SOCIAL LIFE
regard this premiss with some anathema, as a capitulation to 'biological

determinism'. The emphasis on culture as sui generis, pure and untainted by
nature (or 'biology'), marks a common spirit within traditions as diverse as
late-nineteenth-century monogenic evolutionism, early-twentieth-century
Boasian relativism, and late-twentieth-century structuralism and symbolic
interpretivism. On this score, I must declare my sympathies with the
mainstream and against Fox, not because he is wrong and they are right, but
because, to me, kinship concerns 'biology' in quite a different sense. Fox glosses
over those aspects of kinship which are founded in the capacity for language
and symbolic thought, of which the notion of 'biology' is itself a product. It is
this capacity that both distinguishes human kinship from the so-called 'kinship
systems' of non-human animals, and underwrites the diversity of forms of
kinship to be found in human societies. This diversity, in turn, provides the
point of departure-in most if not all anthropological traditions-for
structural and evolutionary comparisons between societies.
While it is reasonable to speak of other animals as having 'kinship', at least in
a metaphorical sense, human kinship is fundamentally different from that of
other species in that it is characterized by culturally articulated sets of rules
which may operate to a great extent independently of observable behaviour.
Studies of non-human sociality, among primates for example, are based on the
data of observation and, as such, are unavoidably behaviourist in perspective.
Studies of human sociality, on the other hand, depend on an understanding of
t e re atwn etween e avwur an ru es. Most stu tes o uman socta tty m
the sphere of kinship have emphasized the pre-eminence of such rules (see, for
example, Sahlins 1976, Fortes 1983). 'Kinship', as defined in the human
context, depends on the existence of these rules, which in turn are understood
by ordinary human beings in relation to culturally specific sets of linguistic and
extralinguistic categories. The notion that kinship has a biological foundation is
really dependent on the cultural definition of 'biology'. Even in Western
societies, 'biological' kinship is often as much a metaphor for social relations as
a statement of relevant biological fact. To me as a social anthropologist, this is
exemplified in expressions like 'She has her mother's eyes', or 'He's my own
flesh and blood', though sociobiologists may disagree. (Dun bar, in the
preceding article, uses a similar example as evidence for precisely the opposite
point of view!)
The great debate

Among humans, then, kinship is everywhere a cultural and social construction,
whatever facts of reproduction may lie behind the variety of kinship systems to
be found. This seemingly simple observation masks an intractable problem of
definition, one which was first brought to light in the late 19 50s but still casts
its shadow on present-day thinking: the problem of 'the concept of kinship'.
The key protagonists in the great debate on this problem have included
786
Gellner, Needham and Barnes. The debate was played out in a series of five
articles, originally published in the journal Philosophy of Science between the
years 1957 and 1963. I shall cite here from the reprints of Gellner's three
articles, which appeared, along with other relevant essays, in his book The
Concept of Kinship (1973:154-203), as well as from Needham's (1960) reply to
the first of Gellner's articles and Barnes's (1961) reply to the second. The
minutiae of attack and counter-attack need not concern us. Much more
important are the implications of the arguments, which hinge on the relation
between 'biological' and 'social' kinship.
Gellner's position is that social kinship is axiomatically bound to a
'biological' foundation (1973 [1957]:154-62). For Gellner, and indeed for the
other protagonists, the terms 'biological' and 'physical' were taken as essentially
synonymous and applied to the facts of reproduction. However, his purpose in
his original paper was less to explain kinship per se than to use 'kinship
structure' as a device with which to illustrate the operation of an 'ideal
language', as conceived by Wittgenstein and other early-twentieth-century
philosophers. The specific aspect of kinship structure which Gellner employed
was the relation between generations as constructed in a hypothetical 'naming'
system, in which children would bear the 'names' of their ancestors in a certain
logical order. For example, 'if Joan has three sons and Joan's name is J, their
names would beJlX,J2X, andJ3X where X conveys the necessary information
about their respective fathers or father and in turn their ancestry' (Gellner
7 : e ner was apparent y assummg t at mem ers o IS ypot ettca
society had much the same theory of procreation as members of his own society,
for a biological physical relationship is assumed to be recognized as a
preliminary to the naming of children.
Needham's attack centred on Gellner's apparent confusion of biological and
social relationships. For Needham, only the latter are of any significance at all
for the anthropologist, although he admits a degree of 'concordance' between
'biology' (here being defined as actual genetic parentage) and 'descent' (the
socially defined rules for stipulating the relations between members of different
generations). Gellner retorted that this latter admission on Needham's part
rules out any attempt to separate biology from descent. They are inextricably
bound together. What is important is that 'physical relationships' are used 'for
social purposes' (Gellner 1973: 170), to be enunciated or ignored by social
actors according to social customs and perhaps individual aspirations.
Barnes's position-which Gellner regarded as a 'refinement' of
Needham's-is that the social anthropologist should be concerned with
'physical relationships' but only in the form in which these are defined by
members of the society under consideration: the Trobriand Islanders, the
Nuer, the English, etc. Thus there are three possible levels of analysis: of true
genetic relationships between individuals, of 'biological' relationships (as
defined by people of the society in question), and of social relationships. For
Barnes, the first is irrelevant for social anthropology, while the latter two, and
787
SOCIAL LIFE
the relation between them, fall firmly within the subdiscipline's rightful
domain of enquiry. Gellner (1973: 198-200) counters this suggestion by
pointing out that social anthropologists themselves, in their ethnographies, do
indeed take for granted 'physical reality' as defined within their own
anthropological culture.
It seems to me that despite the apparent plausibility of Gellner's counter-
attacks, Needham and more particularly Barnes have got it more or less right.
Needham re-articulates the view, generally accepted both before and after
Gellner, that anthropology should not concern itself with the truth or lack of
truth in other people's belief systems (on this point, see Lewis in this volume,
article 20). Thanks to his recognition of a flaw in Gellner's argument-the
failure to distinguish between (1) true biological knowledge, or more
specifically the knowledge of genetic relationships and the facts of
reproduction, and (2) socially constructed 'biological' knowledge-Barnes has
given us a useful analytical insight. He has defined precisely what had earlier
lain implicit in anthropological understanding: the existence of three rather
than two levels of analysis. The interplay between the biological and the social
has a middle ground (socially constructed 'biological' knowledge), and the
terms of this middle ground are not universal but culturally specific.
Nevertheless, none of the contributors to the original debate tackled the final
problem alluded to by Gellner: the fact that there is something, which we call
'kinship', that is understood cross-culturally and is described by
ant ropo ogtsts m a way w tc presupposes certam umversa s.
Needham (1971) and Barnard and Good (1984:187-9) have commented on
the need for a polythetic definition of 'kinship'. Kinship is understood cross
culturally not because it has a single defining feature in all societies, but because
similar sets of features are found in every society, without any single feature
being necessary as the defining one. There are universals in kinship, but these
universals are the constructs of anthropologists rather than of informants.
Gellner's initial premiss was that human kinship systems are based on biology,
and his conclusion, seven years on in the debate, was that this must be true
because anthropologists themselves share a knowledge of certain biological
facts and use this biological idiom in their ethnographic descriptions. But does
the same biological idiom form part of the knowledge of all human societies? I
think not. There is no reason to suppose that Australian Aborigines, Bedouin
nomads, or Chinese peasants have the same notions of procreation that we have
in the West. Even Western scientists have only relatively recently-in the
nineteenth and twentieth centuries-come to understand anything of the
mechanisms of ovulation and fertilization which educated Europeans now take
for granted (see Barnes 1973:65-6). The fact that scientific knowledge is itself
defined, not in nature, but according to the culture of science, is a further
complication.
Gellner's 'cultural universal' is neither true biological knowledge nor a
shared cultural knowledge of biology. It is the genealogical grid, a device
788
defined within anthropological culture, but one which is presumed by

anthropologists to be of utility for the explanation of any kinship system, no
matter what the indigenous theory of biology or of social relationship might be.
The genealogical grid is an extremely useful methodological tool, but beyond
that its existential status is difficult to establish.
'Fatherhood'
The problems of defining biological fatherhood have been couched in terms of
the theoretical issues outlined above. True biology being irrelevant, the notion
of 'father' is supposed to encompass two basic elements which, in relation to
any particular child, may or may not specify the same individual. These
elements are ( 1) the indigenous recognition of having contributed something
by way of material substance to the child, and (2) the recognized conferral on
the child of a specific social identity with its attendant rights and obligations.
Drawing on a pair of Latin terms, the first element is said in conventional
anthropological accounts to specify the child's genitor, whereas the second is
said to specify its pater, or 'social father'.
However, the ethnographic situation is often more complicated than this
simple distinction implies. Consider the beliefs of the Trobriand Islanders.
Malinowski (1932 [1929]:140-78) reports that, when a person dies, that
person's spirit is believed first to go to the Island of the Dead and later to
return to eart to Impregnate a woman o Its own su c an. ImpregnatiOn IS sat
to be either through the head or through the vagina, but there is no suggestion
in Malinowski's account that the woman's husband is believed to be involved in
the contribution of genetic substance to the child. He simply 'opens up' the
woman for childbirth. Children are supposed to resemble their mothers'
husbands because of the close physical relationship between husband and wife,
not because of the implantation of semen.
In his account, Malinowski took these beliefs at face value, as a reflection of
the Trobrianders' alleged ignorance of the male role in conception. In their
rather different ways, Spiro (1968) and Leach (1968) argued against this
position, by appealing to the contrast between a public doctrine of the denial of
physiological paternity and a more matter-of-fact (according to Leach) or
repressed (according to Spiro) knowledge of the 'true' process of procreation
(see Barnard and Good 1984: 170-4). Yet whatever the Trobriand equivalent of
'genitor' might be, there is little doubt that the Trobrianders have a concept of
'pater' as both genealogical (mother's husband) and social father, at least
comparable to the Latin or English notion, though different in some respects.
The Trobriand word is tama, which is applied rather more widely than its Latin
or English equivalents (it denotes not only the father but also the father's
brother, father's sister's son, etc.). Some evidence that tama really is
genealogically, if not biologically, similar in definition to the notion of 'father'
in other languages, may be found in the fact that its reciprocal is also the
789
SOCIAL LIFE
reciprocal of ina (mother): both 'parents' apply the word latu (child) to their
'children'. Further evidence is that the wife of any (classificatory) tama is called
one's ina, and the husband of any ina is called one's tama. Socially, we can
speak of those called tama as 'fathers' because of these genealogical
equivalences, but it is always worth remembering that the notion of 'fatherly'
attitudes and behaviour in one society may be quite different from that in
another. For the Trobrianders, the father is an indulgent figure more akin to a
favourite uncle in Western societies, whereas the Trobriand mother's brother is
just the opposite, a Freudian father-figure par excellence.
'Motherhood'
The concept of 'motherhood' is even more interesting. It actually entails three
distinct elements, each of which has potential for social recognition. We can
distinguish: ( 1) the culturally defined genetic mother, (2) the bearing or
carrying mother, and (3) the social mother. Following ancient Roman (Latin)
usage, modern anthropologists have generally conflated the first two, often
under the term genetrix, and distinguished the last by the term mater. Yet the
conflation of the former cannot be sustained, either on logical or on biological
grounds. Indeed the distinction, which ancient Romans and anthropologists
alike have failed to make, is not new within Western thought.
The culturally defined genetic mother is the female recognized by society as
avmg gtven matena su stance to t e c 1 . T IS su stance, o course, nee
not correspond to that which modern biological science deems to be definitive
of 'true' genetic motherhood. For example, the common belief, found in many
parts of the world, that the child's 'flesh' comes from its mother's 'blood', while
its 'bone' comes from its father's 'semen', reflects a notion of genetic
motherhood (and fatherhood) which differs from that of Western science. As
noted above, the true facts of genetics discovered by nineteenth- and twentieth-
century scientists are, as such, largely irrelevant for the anthropological study
of kinship; what are important are the indigenous theories. The peculiarity in
the study of Western kinship is that Western science itself, or aspects of it, form
part of our own folk knowledge.
The bearing or carrying mother is the person who gives birth to the child.
Of course, in the overwhelming majority of cases, this is the same person as the
culturally defined genetic mother, but the concept is nevertheless distinct.
Consider three cases: the Orthodox Christian doctrine of the Virgin Mary as
Theotokos, the Aranda belief in 'conception clans', and the modern medical
notion of 'test-tube babies'. In each of these cases, the definition of the
'mother' is closely bound up with the relationship between this 'mother' and a
culturally designated 'father'.
Greek-speaking Christians recognized the distinction between the genetic
mother and the bearing mother as early as AD 4 31, when the Council of
Ephesus formally proclaimed the doctrine of the Virgin Mary as Theotokos
790
(literally 'God-bearer', though generally translated as 'Mother of God').

Nestorius, Bishop of Constantinople, had precipitated a debate on the matter
when he denied that such a term should be used, since (he argued) Mary could
only be the 'mother' of the human, not the divine, aspect of Christ's person. St
Cyril of Alexandria, supported by the Council, disagreed, and it was his view
that was accepted. Cyril argued from scriptural evidence that Mary did indeed
bear the simultaneously fully divine and fully human Christ, and that to assume
otherwise would be tantamount to splitting Christ into two separate persons.
Similarly today, when modern Orthodox Christians (or indeed Roman
Catholics) use the term 'Mother of God', they are not asserting anything about
genes or chromosomes. On the contrary, to the Orthodox the doctrine of the
Theotokos 'safeguards the unity of Christ's person' (Ware 1984:33).
The idea of the 'conception clan' among the Aranda, an Aboriginal people of
central Australia, may be rather different, but a similar distinction between
aspects of 'motherhood' is implied in their belief system. The Aranda possess
three types of clan: matrilineal, patrilineal and 'conception'. The matrilineal
and patrilineal clans are clear enough; a person belongs to the matriclan of his
or her mother and to the patrician of his or her father. The conception clans are
of a different order. Unlike the matrilineal and patrilineal clans, the conception
clans are not exogamous. A man, his wife, and one or more of their children,
may all belong to the same or to different conception clans. Membership in a
conception clan is acquired through the belief that part of the 'genetic'
su stance o a c I IS contn ute y a spint, representmg a totemic emg,
whose power is vested in a sacred site. If a woman should happen to pass by
such a site, she may be fertilized by the totemic spirit. While virtually all
Australian Aboriginal peoples recognize some form of spiritual fertilization
(often by a spirit of the father's clan), those such as the Aranda who recognize
conception filiation are unusual in that they posit a special relationship between
the mother and the spirit itself, without reference to the father (see, for
example, Maddock 1972:30-2, Strehlow 1947:86-96).
Our third example is the idea of the 'test-tube baby'. As an idea, this is
hardly new. It was heralded several decades ago by prophetic if little known
science-fiction writers like Sam Fuller (1936). Nor is it new to anthropologists.
Nearly all existing high-tech procedures of procreation have their equivalents,
albeit mystical rather than technical, envisaged in the indigenous ideologies of
non-Western peoples (Heritier-Auge 1985). What is different about this notion
is that the idea was eventually realized in technical practice. In vitro fertilization
is now practised in several technologically advanced countries. Modern
medicine has also made it possible to implant an ovum (either fertilized or
unfertilized) from one woman into another. While 'test-tube babies' may be
genetically related to both their prospective social parents, surrogate
motherhood is a very real possibility.
All this has important and interesting implications for kinship theory. The
distinctions between the various forms of biological and social parenthood and
791
SOCIAL LIFE
the potential kinship ideologies and social relationships that may emerge from
putting them into practice are intricate and enigmatic. While some artificial
methods are both possible and realized (artificial insemination in utero,
fertilization in vitro, and ovum transfer), others are theoretically possible if not
yet practical or practised. Consider, for example, the possibility of a person
other than the genetic mother receiving a fertilized ovum or embryo, and where
neither this surrogate mother nor the genetic mother becomes the social
mother of the child, and where the sperm donor, too, is a different person from
the child's intended social father. This would give the child no less than two
'fathers' and three 'mothers'! This is only one, if the most complex, of a great
number of vexing possibilities (see Laborie et al. 1985: 14-16). The existence of
customs such as wet-nursing among the European upper classes of historical
times raises a further consideration, as in such cases the nurturing mother is yet
another category, and one with a biological as well as a social role.
Of course, knowledge flows both ways. Just as 'test-tube babies' have
important implications for kinship theory, so indigenous ideologies (such as
those of the Australian Aborigines) have much to add to Western
understandings of such medical practices and their social consequences. Yet, as
Riviere ( 1985) has pointed out, little note has been taken of the contribution
that anthropology can make in unravelling the cultural perceptions behind
biological 'facts'. Politicians, lawyers, doctors and moral philosophers, all new
to the problem, have had at least as much trouble explaining the relationship
etween genetics an parentage as ave t e urc Fat ers, t e ran a, or t e
anthropologists who have considered such issues. Peculiarly, the Committee of
Inquiry into Human Fertilization and Embryology, which reported to the
British government in 1984 (Warnock 1985), included no anthropologists. For
the time being at least, the philosophers have been left holding the baby.
Social parenthood
Social parenthood is best defined as a culturally recognized relationship which
involves one or more of the following roles: nurturing and socialization (these,
of course, are not necessarily exclusive of biological input), obligations of
guardianship, and equivalent rights as a guardian (either in rem or in personam).
In particular cases, social parenthood may or may not coincide with any specific
kind of biological parenthood, but within a given society as a whole it is
generally expected that those designated as 'parents' will normally have
biological (in the sense of 'shared substance') or pseudo-biological ties, as well
as jural ties, to their 'children'.
This definition is clearly imprecise, but it is exactly this imprecision which
makes it more or less universally applicable to the great diversity of human
societies (cf. Barnard and Good 1984: 187-9). The nature of 'parenting', of
course, is extremely variable. As Malinowski (1966 [1927]:14-19) found, the
free and indulgent father-son relationship among the Trobriand Islanders
792
manifested a very different notion of fatherhood from that of his native Central
European experience.
Social parenthood is an outgrowth of the nuclear family, another commonly
cited human universal. The simple two-parent nuclear family is the statistical
norm in many Western and industrialized societies and serves as a basis for
family organization elsewhere. Yet two qualifying factors (among other
possibilities) deserve special mention: (1) the existence of alternative and more
complex domestic arrangements, and (2) the practice of acquiring children
from outside the nuclear family. The former is represented by such practices as
eo-residence of parents and adult children or of siblings, or polygamy and
concubinage, and the latter is exemplified by fostering and adoption. Let us
examine each in turn.
The classical formulation of the idea of the nuclear family as a universal may
be attributed to Murdock (1949:1-40), who argued that other human family
types can be identified as variations on the nuclear family theme, like atoms
'aggregated, as it were, into molecules' (1949:23). In such societies as have
them, these alternative forms often entail 'parental' obligations on the part of
other senior members of the domestic unit. Such family types include the
compound family (defined as a polygamous household, e.g. a man, his wives
and children), the joint family (involving a formalized collectivity of relatives,
e.g. a group of brothers, their wives and children), and the extended family
when defined as a domestic unit (usually understood as a less formalized
eo ecttvtty o re attves s anng t e same we mg pace . Event e one-parent
family is, arguably, a form of nuclear family-simply one that involves one
rather than the typical two parents.
The practice of acquiring children from outside the nuclear family is not
uncommon. Two ways in which children are brought into the family for
nurturing and socialization are 'fostering' and 'adoption'. The distinction
between them is commonplace in modern legal systems, especially in the West.
Fostering, or fosterage, involves nurturing and socialization without full social
parenthood (often as an initial step towards adoption), while adoption does
involve full social parenthood. However, this distinction is not always as clear-
cut in other societies as it may seem in ours, and finer distinctions are
sometimes called for. In West Africa, for example, there exists a complex of
fostering practices based upon notions of legal obligation within and between
kin groups, as well as upon the economic and political considerations of
particular families. Domestic help on the part of fostered girls, apprenticeship
and inherited clientship on the part of boys, even the 'pawning' of children
between creditors, have all formed part of West African fosterage in recent
times (E.N.Goody 1984).
A further word of caution: while in modern Western societies, 'adoption' is
most commonly thought of as a method of incorporating and legitimating
parentless children into a nuclear family, this is far from its original
significance. Nor does the term convey much information about the culturally
793
SOCIAL LIFE
varied practices which it is called upon to designate. The ancient Roman notion
of adoptio ('adoption' of a legal dependent), like the related concept of adrogatio
('adoption' of an independent), had much more in common with establishing
political alliance through marriage than it did with the upbringing of hapless
waifs. It had to do with the potential inheritance of wealth by one's chosen
'son'. The adopted 'son', more often than not an adult, would maintain filial
affection towards his original parents, while acquiring the legal status of'son' to
someone else (E.N.Goody 1971:340-2; cf. Rawson 1986:173-86.)
GROUP STRUCTURE: DESCENT AND RESIDENCE

No one would deny the importance to human groups of close family ties, but
'kinship' in anthropological discourse generally connotes ties beyond even the
extended family, and group structures beyond the domestic unit. Fortes
(1969:63-6, 100) distinguished close kin ties (in what he called the 'domestic'
or 'familial domain') from more distant ones (representing the 'politico-jural
domain'). Although these terms may seem to modern anthropologists both
unwieldy and overly functionalist, the distinction drawn is nevertheless useful.
The politico-jural domain is our concern in this section.
In describing the elements of that domain, it is also worthwhile to
distinguish what may be termed the formal properties of rules of descent and
residence, or even of descent systems, from those which might be called
su stantive. T e orma properties o a ru e are t ose w tc are enve trect y,
with mathematical or logical precision, from the rule itself. By contrast, the
substantive implications of descent and residence rules will differ according to
culture and social context.
Rules of descent
In formal terms, there are six logical possibilities for the transmission of group
membership (or of other rights, as with the inheritance of property or
succession to office) from one generation to the next (N eedham 1971: 10).
These correspond to the traditional notions of patrilineal, matrilineal, double,
bilateral, parallel, and cross-descent. Patrilineal and matrilineal descent are
virtually self-explanatory, whereas the distinction between double and bilateral
descent tends to be more troublesome. The final two forms are in fact
opposites.
Double (or duolineal) descent comprises simultaneous patrilineal and
matrilineal descent: a child belongs to the patrilineal group of its father and the
matrilineal group of its mother, and patrilineal and matrilineal groups belong
to different sets. To take an imaginary example, the patrilineal groups may be
localized clans, say 'Alsace', 'Burgundy' and 'Bourdeau', while the matrilineal
groups are non-localized moieties, say the 'Whites' and the 'Reds'. If the
groups are exogamous, as ethnographically they often are, then a 'White Alsace'
794
must marry a 'Red Burgundy' or a 'Red Bourdeau' (not, say, a 'White

Bourdeau'). Bilateral (or cognatic) descent, by contrast, comprises a
recognition of descent from both sides of the family in the absence of any
specified lines of descent. In a system based on cognatic descent there are no
unilineal descent groups: no localized clans, no 'White' or 'Red' moieties.
Whereas double descent is fairly rare, being found in a few societies in West and
Southern Africa and in Australia, bilateral descent is very common, especially
at the two ends of the evolutionary scale: in small-scale societies of hunter-
gatherers and in modern industrialized societies.
Parallel and cross-descent are very rare forms. Parallel descent involves the
transmission of sex-specific group or category membership from father to son
and from mother to daughter. Cross (or alternating) descent involves
transmission from father to daughter and mother to son (or mother's father to
daughter's son and father's mother to son's daughter). These are typically
found as secondary modes in conjunction with simultaneous patrilineal or
matrilineal descent. For example, the Nama of Namibia formerly possessed
both patrilineal clans, which were localized, and exogamous cross-descent
name lines, which were not (Hoernle 1925:9, 16, Barnard 1975:9-11).
Rules of residence
There are some seven possible rules of postmarital residence, several of which,
w en coup e wtt ru es o escent, ave orma tmp tcatwns or group
structure. These possible rules include virilocal (in the natal locale of the
husband), uxorilocal (in the natal locale of the wife), avunculocal (with the
husband's mother's brother), amitalocal (with the wife's father's sister),
duo local (the separate residence of husband and wife), ambilocal (in either the
natal locale of the husband or that of the wife), and neolocal (in a new locale)
residence. Amitalocal residence-formally the inverse of avunculocal
residence-is ethnographically unattested, though it might be anticipated in a
strongly patrilineal, strongly female-dominated society, if such were ever
found.
Virilocal and amitalocal residence, when coupled with patrilineal descent,
have the logical propensity to foster the recruitment of de focto patrilineal
groups. Virilocal residence would keep the men of the group together, while
amitalocal residence would keep the women together. Uxorilocal residence and
avunculocal residence, when coupled with matrilineal descent, would similarly
foster the recruitment to matrilineal groups. Uxorilocal residence would keep
the women of the group together, while avunculocal residence would keep the
men together. Duolocal residence preserves the stability of existing residential
groupings but is generally unstable and, where it is found (as among the
Ashanti of West Africa), usually occurs only as an initial stage in married life
(see Fortes 1949). In contrast to all of these, ambilocal residence and neolocal
residence bear no formal relationship to potential unilineal groups. A
795
SOCIAL LIFE
composite type, uxori-virilocal residence, is common in societies in which

bride-service is practised, since it permits the husband to engage in activities,
such as hunting, on behalf of his affines in the early phases of marriage.
Residence and descent: cause and effect?

Over the years, many anthropologists have speculated on the relationship
between residence and descent, on the general supposition that rules of
postmarital residence are prerequisite to the formation of certain types of
descent grouping. Most notable among those who have looked into the problem
are Lowie, Murdock, and Ember and Ember.
Lowie (1947 [1920]:157-62) was among the first to suggest a causative
relationship between residence and descent. Drawing on a small number of
geographically scattered cases, he argued that unilineal descent is derived from
a combination of rules of residence and rules for the transmission of property.
Murdock (1949:201-18) took Lowie's hypothesis a bit further. He suggested
that residence and descent are simply the middle part of a chain of causation
from environment to social structure. In Murdock's view, environmental
conditions and changes in subsistence practices affect patterns of residence. As
in Lowie's model, residence rules affect descent group organization. The type
of descent group organization, in turn, affects other aspects of social structure,
such as relationship terminologies. For example, a system of production in
w tc women per orm t e most Important agncu tura tas s mtg t avour t e
development of uxorilocal residence, perhaps de focto at first and de jure later
on. The recognition of such a rule of residence might result in the formation of
matrilineal descent groups. The development of strong matrilineal groups, in
turn, could favour the use of relationship terms that give recognition of this
fact. Individuals might develop the habit of applying a single term to all
relatives of a particular lineage, thus creating a so-called 'Crow' relationship
terminology. Murdock set out to test such hypotheses statistically by using a
carefully chosen sample of the world's societies to see what broad patterns
emerged. Causation was inferred largely through intuition.
A more sophisticated attempt along the same lines was made by Ember and
Ember. Using an updated version of Murdock's method of cross-cultural
comparison, the Embers (1983) demonstrated a number of interesting
correlations and cast doubt on some common assumptions. For example, they
virtually disproved the traditional notion that the division of labour is a major
determinant of the rule of residence. Instead, their findings indicate that
virilocal residence is favoured by societies engaged in internal warfare, while
uxorilocal residence is favoured by societies engaged in external warfare where
women are involved in subsistence work (1983:151-97). Avunculocal residence
is seen to be a result of the conjunction of uxorilocal residence, matrilineal
descent, a high male mortality rate, and a change towards internal (as opposed
to external) conflict (1983:249-59). With regard to descent, Ember and Ember
796
noted (1983:359-97) that warfare, in the presence of unilocal residence

patterns, may act as a catalyst for the formation of unilineal descent groups.
They argued that 'putative descent groups' (that is, groups commonly defined
as 'clans') develop earlier than what they term 'demonstrated descent groups'
(that is, 'lineages', where the genealogy is claimed to be known).
Kinship and property

A significant effect of descent group formation is the establishment of a
mechanism for the transmission of property from one generation to the next.
This fact was crucial to nineteenth-century evolutionists such as Maine ( 1861)
and Morgan (1877), who regarded changes in the rules which govern the
transmission of property as fundamental to the development of 'higher' forms
of human society (see Kuper 1988). The theme has continued to inform more
recent work, especially that of Goody (e.g. 1969, 1976, 1983), who has paid
particular attention to the significance of bridewealth and dowry.
Dowry, which is often found in settled agricultural societies, is a payment by
the family of the bride to the bride herself or to her husband. In many societies
it is conceived of as payment in lieu of inheritance, in recognition of the fact
that a woman has, upon marriage, left her natal kin group, whereas a man
remains a member of his. Bridewealth, found commonly in patrilineal societies,
and especially in pastoral ones, is a payment by a man or his kin to the kinsmen
o t e n e. Its usua unctiOn IS t e egtttmatwn o c 1 ren. I It IS not pat ,
children may belong to their mother's rather than their father's patrilineal
group (see also Goody and Tambiah 1973). Bride-service, found commonly in
hunting and gathering and small-scale horticultural societies, is similar but
involves the exchange of labour rather than of wealth. Bride-service may
however have different implications for relations between men and women, in
that it can give male labour (hunting) pre-eminence over female activities
(Collier and Rosaldo 1982). Ultimately, all these exchanges are as crucial to
alliance as they are to descent, for they help to determine relations between as
well as within kin groups.
MARRIAGE AND MARITAL ALLIANCE
Problems in the definition of marriage

There has been much debate concerning the definition of 'marriage'. Riviere
(1971:57) has suggested that 'marriage as an isolable phenomenon of study is a
misleading illusion'. Needham (1971:7-8) concluded that the concept is 'worse
than misleading in comparison and of no real use at all in analysis'. Both these
authors, in their negative approaches to the problem, were essentially following
the more positive, but nevertheless polythetic usage of Leach. Leach (1955:183)
explicitly defined the institution as 'a bundle of rights'. In any specific society
797
SOCIAL LIFE
these rights may include: legal fatherhood, legal motherhood, a monopoly of

sexual access between married partners, rights to domestic services and other
forms oflabour, rights over property accruing to one's spouse, rights to a joint
fund of property for the benefit of the children of the marriage, and recognized
relations of affinity such as that between brothers-in-law. According to Leach,
there is no single right which defines the institution of marriage, because in
each society different sets of these rights will apply.
The most famous 'problem-case' for the universal definition of marriage is
that of the N ayar, a high-status South Indian caste group. According to Fuller
(1976:99-122), the problem lies in the fact that traditional Nayar marriage
entails two separate male roles which elsewhere in India are combined. Earlier
ethnographers sought to define Nayar marriage only according to one or other
of these two roles: the sambandham partner and the tali-tier. In ordinary, non-
Nayar Hindu marriage, the bridegroom ties a tali (the gold emblem that
symbolizes the union) around the neck of the bride. In Nayar marriage, a high-
caste person, often a Brahman, ties the tali. In an Indian context, the ceremony
clearly indicates the first stage of a Hindu marriage. Yet in a world-wide
context, the ceremony would seem to resemble more a puberty rite than a
marriage, in that it grants the girl full womanhood and enables her to take
lovers. The Nayar girl does not sleep with her tali-tier; instead, she takes a
series oflovers, called 'sambandham partners', and they become the genitors of
her children. The sambandham partners have little to do with the children they
at er. 1 ren owe a egtance nett er to t e man w o tie t etr mot er s ta i
nor to their genitors. Rather, since descent is reckoned matrilineally, they owe
allegiance to their mother's brothers (c£ Gough 1959).
In an overwhelming majority of human societies, marriage is the mechanism
which provides for the legitimation of children and defines their status in
relation to the conjugal family and the wider kin group. Thus marriage is often
distinguished from concubinage, which may serve similar social functions to
marriage but denies full legitimacy to the union. However, in some societies
marriage is considered as a process rather than an event, and is not easily
distinguished from concubinage. This is the case among the Nharo Bushmen
of Botswana (Barnard 1980: 120-2). Those Nharo who form liaisons with
members of the neighbouring Kgalagari communities are regarded by the
Nharo as married to them. Yet the Kgalagari, although they also recognize a
processual aspect to marriage, do distinguish concubinage as a separate
institution. Accordingly, they frequently regard their Nharo mates simply as
concubines (c£ Kuper 1970).
In most societies, marriage may be ended by either divorce or death, though
death need not always be the end of the union. Sometimes a spouse is claimed
(with or without that person's right of refusal) by another member of the family
of the deceased. Thus the custom of the sororate involves the marriage of a man
to the sister of his deceased wife (a woman to the husband of her deceased
sister), and the custom of the levirate involves the marriage of a man to the wife
798
of his deceased brother (a woman to the brother of her deceased husband). The
latter term may also include cases, as among the Nuer of southern Sudan,
where the woman is taken in by the brother of her husband, but remains legally
married to her dead husband and may bear children in his name.
The Nuer are also anthropologically famous as exemplars of other, more
unusual forms of marriage (Evans-Pritchard 1951 :29-123). Apart from
leviratic marriage and 'normal marriage', in which bridewealth of cattle is paid
from the bridegroom to the family of the bride, there are two intriguing
varieties known as 'ghost marriage' and 'woman marriage'. Ghost marriage is
somewhat similar to leviratic marriage, except that it occurs when a man dies
childless, especially if the death is a result of fighting. The ghost of the man is
married to a woman and bridewealth is paid in his name. Children are fathered
on his behalf by one of his close kinsmen. Woman marriage occurs usually
when a woman is thought to be barren. She becomes socially 'male' and the
'husband' of another woman. The male kin of this female 'husband', or other
males, beget children on her behalf. The males are recognized as genitors of her
children, while she is defined as the pater.
Woman marriage is in fact found in other parts of Africa as well, perhaps most
notably among the Lovedu of South Africa, where historically it has had a
powerful impact on the political system. Since around 1800 the Lovedu have
been ruled by a line of biologically female, but socially male women, the remote
and mysterious 'Rain-Queens'. Ideally, each queen since that time has been
po ygamous y marne to ot er ema es m a pattern m w tc an actua not
classificatory) brother's daughter of each of the wives follows her father's sister to
the queen's harem. They may remain there to be impregnated by male members
of the royal house, or they may be redistributed to the queen's relatives or other
subjects elsewhere. This pattern maintains alliances between the royal house and
the people of scattered localities, each distant group proudly assenting to the
power of the queen, their kinsman (Krige 1975:249-52).
Incest and exogamy

'Incest' is defined technically as a sexual act between individuals prohibited
from engaging in such acts because of their relationship. The relationship may
be specified according to affinity or fictive kinship, as well as consanguinity. It
may be one of close kinship or, as often as not, of distant relationship of a
particular, culture-specific kind. Whereas proximity of relationship is
important in Western society, other societies more frequently define as
incestuous sexual acts between individuals of the same moiety, phratry or clan,
between those defined as members of the mutual classificatory 'sibling'
category or of the same household, or even between people related in a ritual or
fictive way, such as those who share the same godparents.
The origin of the universal prohibition of incest has been the subject of
debate since the late nineteenth century. Some early theorists concentrated on
799
SOCIAL LIFE
the incest taboo proper. Among these, Westermarck ( 1891) contended that
humans have a natural abhorrence of mating with close kin, while Freud (1960
[1913]) suggested just the opposite. Other writers emphasized the relation
between incest regulation and marriage prohibitions. Chief among these was
McLennan who, in his famous Primitive Marriage ( 1970 [ 1865]), argued that
the basis of human society was exogamy, or 'marrying out', a term which he
invented. In McLennan's view, the prohibition of sex and consequently
marriage within the tribe was directly related to the development of the capture
of women, first as concubines, and later as wives. Indeed the symbols of
capture, of yielding, or of giving away the bride are still common throughout
the world in societies at all levels of social evolution. Problems like the relative
importance of nature and nurture, the relation between descent and alliance,
and the genesis of human society itself are also fundamentally related to the
origin of this supreme taboo (see Fox 1980, Arens 1986). While the precise
definition of an incestuous relationship is specific to each culture, the
prohibition of incest is virtually universal. For this reason, Levi-Strauss, in his
most profound book The Elementary Structures of Kinship ( 1969 [1949]: 12-25),
equated the incest taboo with the origin of culture itself.
More recently, exogamy has returned as a focus of serious interest. Knight, a
contemporary anthropologist in the nineteenth-century mould, has been
developing a theory of the origin of culture which hinges on the exchange of sex
(from dominant women) for meat (from male hunters). The intricacies of this
comp ex t eory are we eyon t e scope o t IS arttc e, ut t e mterestmg
point about it is that Knight has placed incest in the context of food exchange.
Central to the theory is the so-called 'own kill rule', according to which
individual hunters will not eat the animals they kill but will instead exchange
them (Knight 1986, see also Knight 1991 ). Among many hunting and gathering
and small-scale cultivating peoples, eating one's own kill is likened to incest,
and the exchange of food is likened to the exchange of women.
Of course, exogamy is only one side of the marital coin. Endogamy also
needs to be explained. 'Societies', 'tribes', and 'traditional communities' the
world over are largely endogamous, almost by definition, but frequently
marriage takes place within smaller units than these. While major reasons for
societal, tribal or community endogamy may be geographical proximity and
familiarity, the principal reasons for marriage to close kin may be more subtle.
One reason is the preservation of the kin group or kin line itself, perhaps
expressed symbolically by the notion of the 'purity of the blood'. A classic case
of this is brother-sister marriage in Ancient Egypt (Hopkins 1980). Another
explanation often encountered is that close kin marriage keeps property in the
family or acts to preserve its unity. The best-known case here is preferential
marriage to the father's brother's daughter, common in North Africa and the
Near and Middle East. In this latter case, the explanation may be made explicit
by informants, or, perhaps more commonly, may lie deep in the indigenous
kinship consciousness-in what Bourdieu (1977:30-71) calls the habitus, an
800
environment of dispositions available to given individuals, which may be

simultaneously strategic and unpremeditated. Indeed Bourdieu exemplifies
this notion in a discussion of the reasons for father's brother's daughter
marriage among the Kabyle of Algeria.
Finally in this connection, it is worth remembering that the rules of incest
and those of exogamy need not coincide. As Fox (1967:54) has put it, 'this is
really only the difference between sex and marriage, and while every teenager
knows these are different, many anthropologists get them confused.' Nor is it
only anthropologists who sometimes make this confusion. The term 'incest' as
used in the popular press more often than not refers to the incestuous rape of
minors, a very specific form of incest indeed. In common language the term
generally has a much wider meaning, though one which is often equated with
sex between 'blood' relatives, and this does cause confusion when one considers
incest as a legal category, which of course it is in both Western and non-Western
societies. It is interesting to note that in Scotland, for example, the laws
affecting incest and marriage coincide, and these cover both consanguines and
affines. Yet in England and Wales it is perfectly legal to engage in sexual
intercourse with certain classes of relatives who are forbidden in marriage. This
is because English civil law forbids marriage to close affines and step-relatives
while the criminal law of incest applies only to consanguines (Seear et al.
1984:12-13).
Marital exchange
It would offend many in the field of kinship studies if I were to pass on without
a word about alliance theory and prescriptive marriage. As Maybury-Lewis
(1965:228) has written: 'To paraphrase Dr Johnson, a man who is tired of issues
such as these is tired of social anthropology'. Be that as it may, the theories and
debates surrounding this subject are too numerous and too complex for me to
review them in detail here (but see Barnard and Good 1984:95-118). Instead I
shall draw attention to just a few of the ideas which have sprung from the
copious writings on the subject.
Alliance theory is usually traced to Levi-Strauss's The Elementary Structures
of Kinship, published in French in 1949, though not translated into English
until 1969 from the French second edition (see also Levi-Strauss 1963 [1945]:
46-9). In contrast to descent theory, alliance theory concentrates not on the
formation of groups, but on the relations established between them through
marriage. Levi-Strauss distinguished 'elementary' from 'complex' structures,
the former characterizing societies which have a positive marriage rule (e.g. one
must marry someone of the category cross-cousin), the latter characterizing
societies which have a negative marriage rule (e.g. as in Western society, one
may marry anyone who is not close kin). Elementary structures are further
divided into those where there is direct or restricted exchange (normally,
marriage to the category of the first or second cross-cousin), those where there
801
SOCIAL LIFE
is generalized exchange (normally, marriage of men to the category of their

matrilateral cross-cousins), and those which practise delayed direct exchange.
The last would be a product of the repeated marriage of men to their
patrilateral cross-cousins. For formal reasons, matrilateral marriage creates a
pattern in which men take women in marriage from the same group as did their
fathers, while patrilateral marriage creates a delayed pattern, with women going
in one direction in one generation and back again in the next. For this and other
reasons, true patrilateral cross-cousin marriage with delayed direct exchange is
unworkable and found nowhere as a stable form. By contrast, generalized
exchange is common, especially in Asia, and direct exchange is common in
Amazonia and Australia (for further discussion of elementary structures in
relation to the properties of exchange, see Gregory in this volume, Article 33).
French writers such as Levi-Strauss, Dumont (e.g. 1975) and Heritier (e.g.
1981) have developed a high degree of sophistication in the comparative and
theoretical analysis of structures of kinship alliance, but their approaches differ
in subtle ways from their British counterparts, notably Leach (e.g. 1951) and
Needham (e.g. 1962). To put it crudely, French theorists have long been
expanding the applicability of alliance theory to cover a wider variety of
societies, while British theorists (and to some extent also Dumont) have tended
to narrow their concerns to the so-called 'prescriptive' systems, originally
conceived as those with a precise rule of marriage to members of one particular
class of kin. Yet, as Needham (1973) later revealed (if not in so many words), all
st es a mtsta en y assume t at t ey were ea mg wtt two sets o
phenomena when really there were three. The choice was not between
'prescription' and 'preference', however defined, but between 'prescription',
'preference', and 'practice'. In his new formulation, 'prescription' was equated
with formal relations between categories in an idealized terminology structure,
'preference' was applied to the jural rules which are recognized in any given
society, and 'practice' described the actual social behaviour of individuals, who
may or may not follow such rules, regardless of the type of kinship system they
possess. Contrary to earlier usage, these sets of rules and categories may be
applied to the study of any kind of society, not just those elusive dwellers of
isolated islands and impenetrable jungles who may or may not possess the
perfect prescriptive system (cf. Schneider 1965). (For a summary of the current
debate on this issue, see Barnard and Good 1984:95-106.)
THE CLASSIFICATION OF RELATIVES
Morgan's discovery
M organ's discovery of the 'classificatory system of relationship' is, in my view,
the single most significant ethnographic breakthrough of all time. The idea that
different societies classified relatives differently had been noted before M organ,
notably by the eighteenth-century missionary among the Iroquois, Joseph
802
Lafitau (1724). Moreover, there is no reason to doubt that North American

Indians were aware of the differences between the various indigenous systems
of classification before Europeans were. What makes Morgan the pioneer is
that he was the first to posit historical and sociological reasons for such
differences in kinship classification, and to express these within a context of
anthropological theory (cf. Rivers 1968 [1914]:41-45, Service 1985:13-34,
Trautmann 1987).
Morgan had lectured and published on Iroquois kinship from the early
1850s, but his truly definitive statement on the subject was Systems of
Consanguinity and Affinity of the Human Family (Morgan 1871). In the final
chapter of this great work, Morgan advances a theory of the evolution of
human society, a theory which hinges on the notion that relationship
terminologies are slow to change and thus retain clues to ancient customs no
longer practised. For example, if one calls father and father's brother by a
single term, this might suggest an earlier rule or practice of marriage of a group
of brothers all to the same woman. If one also calls one's mother and mother's
sister by a single term, this might further suggest a rule or practice of 'group
marriage', where a group of brothers would collectively be married to a group
of sisters.
Very few anthropologists today subscribe to Morgan's specific evolutionary
theory. Yet ever since Morgan, kinship studies have attempted to tease from
relationship terminologies a multiplicity of sociological ideas, both grand and
sma . T e exceptiOns are ew. T e most promment exceptiOn m Morgan s own
time was McLennan (1876:329-407), arguably the founder of alliance theory.
Although his own, earlier formulation of the stages of human social evolution
was largely in agreement with Morgan's, he considered Morgan's classificatory
system 'a system of mutual salutations merely' (1876:366). Even some of those
who have denied altogether the sociological importance of relationship
terminologies, notably Kroeber ( 1909), have nevertheless recognized in them a
deep psycho-structural significance. Levi-Strauss, the doyen of both psycho-
structuralism and alliance theory, dedicated his Elementary Structures of
Kinship ( 1969 [ 1949]) neither to Kroeber nor to McLennan, but to M organ.
Relationship terminologies
Morgan's own taxonomy of relationship terminologies included only two
types: 'descriptive' and 'classificatory'. Descriptive terminologies were defined
as those which distinguish direct relatives (direct ancestors and descendants of
ego, plus ego's siblings) from collaterals (all other consanguineal relatives).
Classificatory terminologies were defined as those which, at least from some
genealogical points of reference, fail to make such a distinction. For example,
the relationship terminology of the !Kung Bushmen would be regarded as
descriptive, because, as in English, father and father's brother are
terminologically distinguished (respectively, ba and tsu in !Kung; 'father' and
803
SOCIAL LIFE
'uncle' in English). The relationship terminology of the Nharo Bushmen

would be regarded as classificatory, because, as in Iroquois, father and father's
brother are called by the same term (auba, sauba or g I I oba in Nharo, depending
on linguistic context; ha-naih in Seneca Iroquois). In both Nharo and Iroquois,
the mother's brother is called by a different term (tsxoba or mamaba in Nharo;
hoc-no-sih in Seneca Iroquois).
After Morgan, typologists came to realize the inadequacy of such a simple,
dual classification. It is not merely the presence or absence of one particular
distinction (direct versus collateral) which is important. Other distinctions are
made in the relationship terminologies of the world, and these, too, require
anthropological understanding and explanation. Kroeber (1909), perhaps over-
ambitiously, posited eight such distinctions, and of these one emerged as being
of special ethnographic and theoretical importance. This is the distinction
between parallel relatives (direct relatives and those collaterals who are related
through a same-sex sibling link, e.g. mother's sister, mother's sister's son,
mother's sister's daughter) and cross-relatives (those who are related through
an opposite-sex sibling link, e.g. mother's brother, mother's brother's son,
mother's brother's daughter).
On the basis of these two distinctions, applied in the generation of ego's
parents, Lowie (1928, 1929) and Kirchhof (1931) recognized four ideal types.
Take males of this generation, for example (the structure for female relatives is
the mirror image of that for males). (1) One might call father, father's brother,
an mot er s rot er a y a smg e term as m t e Hawanan anguage).
Alternatively, (2) one might classify father and father's brother by one term and
mother's brother by a different one (as in Iroquois or Nharo). Another
possibility (3) is to classify father by one term and father's brother and
mother's brother by another (as in English or !Kung). Finally, (4) one might
classify each genealogical position by a distinct term (as, say, in Nuer, Armenian
or Gaelic). Lowie called these four ideal types, respectively, 'generational',
'bifurcate merging', 'lineal', and 'bifurcate collateral'; the fifth logical
possibility, which would be to equate the father and mother's brother while
distinguishing the father's brother, is unattested. Generational terminologies
make neither direct-collateral nor parallel-cross distinctions. Bifurcate merging
terminologies make parallel-cross distinctions only. Lineal terminologies make
direct-collateral distinctions only. Bifurcate collateral terminologies make both
kinds of distinction.
While Kirchhof and Lowie emphasized the classification of relatives in the
first ascending generation, Murdock ( 1949) emphasized their classification in
ego's own generation. His typology consists of six classes. 'Hawaiian'
terminologies are those which make no distinction between siblings and
cousins, except of sex. 'Iroquois' terminologies distinguish cross-cousins from
parallel cousins (and often classify siblings by the same term as that for parallel
cousins). 'Eskimo' terminologies do not make parallel-cross distinctions, but
instead distinguish cousins (collaterals) from siblings (direct relatives).
804
'Sudanese' terminologies, like Lowie's 'bifurcate collateral' ones, make both

kinds of distinction. Typically, they lack any general word equivalent to English
'cousin' and call all cousins by strings of possessives. For example, mac brathair
mdthar (literally 'son of the brother of the mother') is the traditional Scots
Gaelic term for a person in that relationship, though it is nowadays also
possible to employ a generic, namely co-ogha 'eo-grandchild'. (It is of course
only since the Norman Conquest that English has been transformed from a
Sudanese into an Eskimo terminology, thanks to its acquisition of the French
word cousin.)
Murdock's other two types, 'Crow' and 'Omaha', like Iroquois, distinguish
parallel from cross-relatives. The difference is that they treat generational
differences in a peculiar way, that is, peculiar for those who do not grow up in
such a culture. The defining feature of Murdock's Crow type (also known as
'Choctaw') is that it classifies father's sister and father's sister's daughter by the
same term. The defining feature for Omaha is that it classifies mother's brother
and mother's brother's son by the same term. Very often, such terminologies
make further equations across generational lines. For example, Crow
terminologies generally classify father and father's sister's son by the same
term, and Omaha terminologies generally classify mother and mother's
brother's daughter by the same term. The logic here is simple. In societies
which possess Crow terminologies, more often than not matrilineal groups are
present, and ego simply assimilates all members of his or her father's
matn mea group an ea s t em y two terms, one or ma es an one or
females. The Trobriand Islanders are a well-known example. Likewise, Omaha
terminologies are frequently found in strongly patrilineal societies, where
similar equations are made in reference to ego's mother's patrilineal group.
This type is common among North American Indians and throughout much of
Africa and South-east Asia. Further equations following the same logic may
occur. For example, in many Omaha systems one's mother's mother's brother,
mother's mother's brother's son, and other male members of one's mother's
mother's patrilineage are all called by the same term.
Crow-Omaha terminology and alliance structures

Over the years, the notions of Crow and Omaha systems have taken on a new
significance in the light of Levi-Strauss's theory of elementary and complex
structures. Levi-Strauss himself (1966: 18-20, 1969:xxxv-xlii) stressed the
importance of 'Crow-Omaha' structures, not simply as terminology types, but
as mechanisms which make possible the extension of marriage prohibitions to
such a degree that the resulting complex structures come to resemble
elementary ones.
Take, for example, the Samo of Burkina Faso, described by Heritier
(1981:81-7, 105-26). They trace descent patrilineally, and have a terminology
structure of the Omaha type. They prohibit marriage to anyone born into ego's
805
SOCIAL LIFE
father's, ego's mother's, ego's father's mother's, or ego's mother's mother's

patrilineal groups. Members of each of these are classified by terms which
indicate their status as 'kin' forbidden in marriage, while the system is further
complicated by the practice of polygyny and consequent restrictions on the
choice of second and subsequent spouses. The prohibitions are so numerous
that specific alliances come to be made between groups by direct exchange.
Crow-Omaha structures are, in effect, ideologically complex, because they
rely upon negative marriage rules, but empirically elementary, because they
narrow the range of choice of spouse to specified descent groups (Barnard
1978:73-4). Although in the Samo case direct exchange is reported, other cases
more closely resemble generalized exchange, which Crow-Omaha structures
simultaneously mimic and reverse as if through a distorted mirror. Unlike the
classic idea of generalized exchange in which sons or daughters repeat the
marriages of their parents, the patterns of exchange in Crow-Omaha structures
can only be egocentric. In 'true', sociocentric generalized exchange (including
classic cases such as the Purum of north-eastern India or the Kachin of Burma)
the bonds and the hierarchical relations between groups are strengthened with
each marriage. In Crow-Omaha structures, by contrast, the constellation of
permitted and prohibited potential spouses changes with each marriage
contracted, and repetition is not allowed. In Levi-Strauss's terms ( 1966: 19),
the one turns 'kinsmen into affines', while the other turns 'affines into
kinsmen'. In other words, while to the individual seeking a spouse there may be
THE FUTURE OF KINSHIP STUDIES

Levi-Strauss's classic essay on 'The future of kinship studies' (1966) picked
out the Crow-Omaha or 'semi-complex' systems as the loci of interest for the
coming generation of scholarship. With Heritier's work, as well as that of a host
of others in diverse traditions, this prediction has, to some degree, been borne
out (see Welter 1988). Nevertheless, there is little doubt that systems of
prescriptive alliance have captured more attention than have Crow-Omaha
ones. A recent attempt to put Needham's (1973) three-level distinction
(between categories, rules, and practices) into effect is Good's (1981) insightful
study of the Kondaiyankottai Maravar of South India. But can we go any
further in the study of 'prescription'? The major contributors to the theory of
prescriptive systems have all gone on to other things. There seems to be little if
any room for further refinements in that sphere of interest.
One area which has long been increasing in popularity is mathematical
modelling, though this is nothing new. Malinowski ( 1930: 19), for whom
kinship was 'a matter of flesh and blood, the result of sexual passion and
maternal affection, of long intimate daily life, and a host of personal intimate
interests', berated Rivers and others for their exclusive attention to the formal
806
properties of kinship systems. Indeed Rivers looked forward to the day when
'many parts of the description of the social systems of savage tribes will
resemble a work on mathematics in which the results will be expressed by
symbols, in some cases even in the form of equations' (1914, I: 10). There is a
sense in which Rivers's prediction has come true. The form, if not the content,
of human kinship may share properties of algebraic systems, just as algebraic
systems in turn share properties with geometric ones. To some, alliance
structures are but patterns which can be created through actual or imagined
marriages between people belonging to actual or imagined groups or categories.
In recent years Australian Aboriginal kinship has been likened to the double
helix (Denham et al. 1979); twisted cylinders (Tjon Sie Fat 1983);
kaleidoscopic reflections, frieze patterns and wallpaper designs (Lucich
1987:1-150); and barn dances (Alien 1982). But building mathematical models
for the sake of it cannot be the answer.
On firmer ground, Levi-Strauss (1968:351) once suggested that in
presenting Aboriginal kinship in formal terms we might, unknowingly, be
'trying to get back to the older [indigenous] theory at the origin of the facts we
are trying to explain'. He had in mind the idea of a primeval 'Plato' or
'Einstein', who used kinship as a means of abstract expression. This idea was
ridiculed by Hiatt ( 1968), and most anthropologists today would probably
prefer Hiatt's sceptical empiricism to Levi-Strauss's conjectures. Von
Brandenstein ( 1970:49), in turn, has likened Aboriginal thought to medieval
a c emy an t e our- umour system o a en, w 1 e Turner o ers a
brilliant reanalysis of the Book of Genesis through Aboriginal spectacles. The
truth is that while wild ideas and grudging squabbles like these give
anthropology its purchase on the broader issues of Western philosophy and
social theory, at the same time they do as much to obscure as to reveal the
cultural systems whose properties and workings we set out to understand.
I believe that the future of kinship studies lies in two directions: first, in the
greater awareness of the importance of such studies in understanding our own
preconceptions, as they touch on such issues as in vitro fertilization, surrogacy,
and the law of incest; and second, in the continuing rediscovery that kinship is
good to think with. Travel in these two directions can be simultaneous, and by
doing so we shall not necessarily end up in two separate places.
REFERENCES
Alien, N.J. (1982) 'A dance of relatives', Journal of the Anthropological Society of Oxford
13:139-46.
Arens, W. (1986) The Original Sin: Incest and its Meaning, New York: Oxford
University Press.
Barnard, A. (1975) 'Australian models in the South West African highlands', African
Studies 34:9-18.
--(1978) 'Universal systems of kin categorization', African Studies 37:69-81.
807
SOCIAL LIFE
--(1980) 'Sex roles among the Nharo Bushmen of Botswana', Africa 50:115-24.
Academic Press.
Barnes, J.A. (1961) 'Physical and social kinship', Philosophy of Science 28:296-99.
--(1973) 'Genetrix : genitor :: nature : culture?', in ].Goody (ed.) The Character of
Kinship, Cambridge: Cambridge University Press.
Beattie, J.H.M. (1964) 'Kinship and social anthropology', Man 64:101-3.
Borland, C. H. (1979) 'Kinship term grammar: a review', Anthropos 74:326-52.
Bourdieu, P. (1977) Outline of a Theory of Practice, Cambridge: Cambridge University
Press.
Brandenstein, C.G.von (1970) 'The meaning of section and section names', Oceania
41:39-49.
Collier, J.F. and Rosaldo, M.Z. (1982) 'Politics and gender in simple societies', in
S.B.Ortner and H.Whitehead (eds) Sexual Meanings: the Cultural Construction of
Gender and Sexuality, Cambridge: Cambridge University Press.
Denham, W. W., McDaniel, C.K. and Atkins, JR. ( 1979) 'Aranda and Alyawara kinship:
a quantitative argument for a double helix model', American Ethnologist 6: 1-24.
Dumont, L. (1975) Dravidien et Kariera: /'alliance de marriage dans l'inde du sud et en
australie, Paris/The Hague: Mouton.
Ember, M. and Ember, C. (1983) Marriage, Family, and Kinship: Comparative Studies of
Social Organization, New Haven: Human Relations Area Files Press.
Evans-Pritchard, E.E. (1951) Kinship and Marriage Among the Nuer, Oxford:
Clarendon Press.
Fortes, M. (1949) 'Time and social structure: an Ashanti case study', in M.Fortes (ed.)
octa tructure:
University Press.
--(1969) Kinship and the Social Order: the Legacy of Lewis Henry Morgan, London:
Routledge & Kegan Paul.
--(1983) Rules and the Emergence of Society, London: Royal Anthropological
Institute Occasional Paper no. 39.
Fox, R. (1967) Kinship and Marriage: an Anthropological Perspective, Harmondsworth:
Penguin.
--(1975) 'Primate kin and human kinship', in R.Fox (ed.) Biosocial Anthropology,
A.S.A. Studies no. 1, London: Malaby Press.
--(1980) The Red Lamp of Incest: an Inquiry into the Origins of Mind and Society,
Notre Dame, lnd.: University of Notre Dame Press.
Freud, S. (1960 [1913]) Totem and Taboo, London: Routledge & Kegan Paul.
Fuller, C.J. (1976) The Nayars Today, Cambridge: Cambridge University Press.
Fuller, S. (1936) Test Tube Baby, New York: God win.
Gellner, E. (1973) The Concept of Kinship and Other Essays, Oxford: Blackwell.
Good, A. (1981) 'Prescription, preference and practice: marriage patterns among the
Kondaiyankottai Maravar of South India', Man (N.S.) 16:108-29.
Goody, E.N. (1971) 'Forms of pro-parenthood: the sharing and substitution of parental
roles', inJ.Goody (ed.) Kinship: Selected Readings, Harmondsworth: Penguin.
--(1984) 'Parental strategies: calculation or sentiment? Fostering practices among
West Africans', in H.Medick and D.W.Sabean (eds) Interest and Emotion: Essays on
the Study of Family and Kinship, Cambridge and Paris: Cambridge University Press
and Editions de la Maison des Sciences de L'Homme.
808
Goody, JR. (1969) Comparative Studies in Kinship, London: Routledge & Kegan Paul.
--(1976) Production and Reproduction: a Comparative Study of the Domestic Domain,
--(1983) The Development of the Family and Marriage in Europe, Cambridge:
Cambridge University Press.
Goody, JR. and Tambiah, S.J. (1973) Bridewealth and Dowry, Cambridge: Cambridge
University Press.
Gough, E.K. (1959) 'The Nayars and the definition of marriage', Journal of the Royal
Anthropological Institute 89:23-34.
Heritier, F. (1981) L'Exercice de la parente, Paris: Gallimard.
Heritier-Auge, F. (1985) 'La cuisse de Jupiter. Reflexions sur les nouveaux modes de
procreation', L'Homme 94(xxv[2]):5-22.
Hiatt, L.R. (1968) 'Gidjingali marriage arrangements', in R.B.Lee and I.DeVore (eds)
Man the Hunter, Chicago: Aldine.
Hoernle, A.W. (1925) 'The social organization of the Nama Hottentots of Southwest
Africa', American Anthropologist 27:1-24.
Hopkins, K. (1980) 'Brother-sister marriage in Roman Egypt', Comparative Studies in
Society and History 22:303-54.
Kirchhof, P. ( 19 31) 'Die Verwand tschaftsorganisation der U rwaldstiimme
Siidamerikas', Zeitschrifi for Ethnologie 63:85-193.
Knight, C. D. (1986) 'The hunter's own kill rule', paper presented at the Fourth
International Conference on Hunting and Gathering Societies, London, 8-13
September 1986.
--(1991) Blood Relations: Menstruation and the Origins of Culture, New Haven and
on on: a e mvers1ty ress.
Krige, E.J. (1975) 'Asymmetric matrilateral cross-cousin marriage: the Lovedu case',
African Studies 34:231-57.
Kroeber, A.L. (1909) 'Classificatory systems of relationship', Journal of the Royal
Kuper, A. (1970) 'The Kgalagari and the jural consequences of marriage', Man (N.S.)
5:355-81.
--(1988) The Invention of Primitive Society: Transformations of an Illusion, London:
Routledge.
Laborie, F., Marcus-Steiff, J and Moutet, J (1985) 'Procreations et filiations. Logiques
des conceptions et des nominations', L'Homme 95(xxv[3]):5-38.
Lafitau, J (1724) Moeurs des sauvages ameriquaines, comparie aux moeurs des premiers
temps, 2 vols, Paris: Saugrain L'aine.
Leach, E.R. (1951) 'The structural implications of matrilateral cross-cousin marriage',
Journal of the Royal Anthropological Institute 81:54-104.
--(1955) 'Polyandry, inheritance and the definition of marriage', Man 55:105-13.
--(1968) 'Virgin birth' [letter]. Man (N.S.) 3:655-66.
Levi-Strauss, C. (1963) Structural Anthropology, New York: Basic Books.
--(1966) 'The future of kinship studies', Proceedings of the Royal Anthropological
Institute for 1965:13-22.
--(1968) 'The concept of primitiveness', in R.B.Lee and I.DeVore (eds) Man the
Hunter, Chicago: Aldine.
--(1969 [1949]) The Elementary Structures of Kinship (second edition), London:
Eyre & Spottiswoode.
809
SOCIAL LIFE
Lowie, R.H. (1928) 'A note on relationship terminologies', American Anthropologist 30:
263-68.
--(1929) 'Relationship terms', Encyclopaedia Britannica, 14th edn, 19:84-6.
--(1947 [1920]) Primitive Society, 2nd edn, New York: Liveright.
Lucich, P. (1987) Genealogical Symmetry: Rational Foundations of Australian Kinship,
Armidale: Light Stone Publications.
McLennan, J.F. (1970 [1865]) Primitive Marriage. Chicago: University of Chicago
Press.
--(1876) Studies in Ancient History, Comprising a Reprint of Primitive Marriage,
London: Bernard Quaritch.
Maddock, K. (1972) The Australian Aborigines: a Portrait of Their Society, London:
Alien Lane.
Maine, H. (1861) Ancient Law, London: John Murray.
Malinowski, B. (1930) 'Kinship', Man 30:19-29.
--(1966 [1927]) The Father in Primitive Psychology, New York: WWNorton.
--(1932 [1929]) The Sexual Lift of Savages in North-Western Melanesia, London:
Routledge & Sons.
Maybury-Lewis, D.H.P. (1965) 'Prescriptive marriage systems', Southwestern Journal
ofAnthropology 21:207-30.
Morgan, L.H. (1871) Systems of Consanguinity and Affinity of the Human Family,
Washington: Smithsonian Institution.
--(1877) Ancient Society: Researches in the Lines of Human Progress from Savagery
through Barbarism to Civilization, New York: Holt.
Murdock, G.P. (1949) Social Structure, New York: Macmillan.
--(1962) Structure and Sentiment, Chicago: University of Chicago Press.

--(1971) 'Remarks on the analysis of kinship and marriage', in R.Needham (ed.)
Rethinking Kinship and Marriage, A.S.A. Monographs no. 11, London: Tavistock.
--(1973) 'Prescription', Oceania 42:166-81.
Radcliffe-Brown, A.R. (1941) 'The study of kinship systems', Journal of the Royal
Rawson, B. (1986) 'Children in the Romanfomilia', in B.Rawson (ed.) The Family in
Ancient Rome: New Perspectives, London: Croom Helm.
Rivers, WH.R. (1914) The History of Melanesian Society, 2 vols, Cambridge:
--(1968 [1914]) Kinship and Social Organization, London: Athlone Press.
Riviere, P.G. (1971) 'Marriage: a reassessment', in R.Needham (ed.), Rethinking
Kinship and Marriage, A.S.A. Monographs no. 11, London: Tavistock.
--(1985) 'Unscrambling parenthood: the Warnock Report, Anthropology Today
1(4):2-7.
Sahlins, M. (1976) The Use and Abuse of Biology: an Anthropological Critique of
Sociobiology, London: Tavistock Publications.
Schneider, D. M. (1965) 'Some muddles in the models: or, how the system really works',
in M.Banton (ed.) The Relevance of Models for Social Anthropology, A.S.A.
Monographs no. 1, London: Tavistock.
Seear, Baroness et al. (1984) No Just Cause. The Law of Affinity in England and Ttales:
Some Suggestions for Change, London: CIO Publishing.
810
Service, E.R. (1985) A Century of Controversy: Ethnological Issues from 1860 to 1960,
Orlando: Academic Press.
Spiro, M. E. (1968) 'Virgin birth, parthenogenesis and physiological paternity: an essay
in cultural interpretation', Man (N.S.) 3:242-61.
Strehlow, T.G.H. (1947) Aranda Traditions, Melbourne: Melbourne University Press.
Tjon Sie Fat, F.E. (1983) 'Age metrics and twisted cylinders: predictions from a
structural model', American Ethnologist 10:585-604.
Trautmann, T.R. (1987) Lewis Henry Morgan and the Invention of Kinship, Berkeley:
University of California Press.
Turner, D. H. (1985) Life Before Genesis: a Conclusion, New York: Peter Lang.
Ware, T. (1984) The Orthodox Church, 2nd edn, Harmondsworth: Penguin.
Warnock, M. (1985) A Question of Life, Oxford: Blackwell.
Welter, V. (1988) Verwandtschaftsterminologie und Sozialorganisation: Einige
ethnosoziologische Interpretationen der Crow/Omaha-Systeme, Emsdetten: Andreas
Gehling.
Westermarck, E. (1891) The History of Human Marriage (3 volumes), London:
Macmillan.
FURTHER READING
Academic Press.
Bohannan, P. and Middleton, J. (eds) (1968) Kinship and Social Organization, Garden
Cit : Natural Histor Press.
Dumont, L. (1971) Introduction d deux theories d'anthropologie sociale, Paris and The
Hague: Mouton.
Fox, R. (1967) Kinship and Marriage: an Anthropological Perspective, Harmondsworth:
Penguin.
Goody, JR. (1969) Comparative Studies in Kinship, London: Routledge & Kegan Paul.
--(ed.) (1971) Kinship, Harmondsworth: Penguin.
--(ed.) (1973) The Character of Kinship, Cambridge: Cambridge University Press.
Graburn, N. (ed.) (1971) Readings in Kinship and Social Structure, New York: Harper &
Row.
Heritier, F. (1981) L'Exercice de la parente, Paris: Gallimard.
Leach, E.R. (1961) Rethinking Anthropology, London: Athlone Press.
Levi-Strauss, C. (1966) 'The future of kinship studies', Proceedings of the Royal
Anthropological Institute for 1965:13-22.
--(1969 [1949]) The Elementary Structures of Kinship, 2nd edn, London: Eyre &
Spottiswoode.
Morgan, L.H. (1871) Systems of Consanguinity and Affinity of the Human Family,
Washington: Smithsonian Institution.
Needham, R. (ed.) (1971) Rethinking Kinship and Marriage, A.S.A. Monographs no. 11,
London: Tavistock.
--(1973) 'Prescription', Oceania 42:166-81.
--(1986) 'Alliance', Oceania 56:165-80.
Schneider, D.M. (1980 [1968]) American Kinship: a Cultural Account, 2nd edn,
Englewood Cliffs, NJ: Prentice-Hall.
811
SOCIAL LIFE
--(1984) A Critique of the Study of Kinship, Ann Arbor: University of Michigan

Press.
Trautmann, T.R. (1987) Lewis Henry Morgan and the Invention of Kinship, Berkeley:
Wolfram, S. (1987) In-laws and Outlaws: Kinship and Marriage in England, London:
Croom Helm.
812
29
UNDERSTANDING SEX AND

GENDER
Henrietta L.Moore
BIOLOGY AND CULTURE

In the discussion of sex and gender in human social life, one term emerges as
particularly problematic, and that term is 'natural'. In public debates
concerning the origins of so-called sex differences and the nature of relations
between women and men-debates which are conducted in the media, in day-
to- ay mteractwns an m aca emtc tscourses-a senes o assertiOns are ma e
which utilize the word 'natural' in ways which are fundamentally misleading.
These assertions are of several kinds, but a common feature of many is that they
describe the differences established in social life between women and men as
originating in biology. This apparently quite straightforward proposition has
been strongly contested by work in the social sciences over the last two decades.
The labour of contestation and refutation has been complicated by a particular
view of biology itself: a view which has been shared by many commentators
both academic and non-academic.
It is often, as Fausto-Sterling (1985) points out, extremely difficult to unravel
arguments about the way in which biology is supposed to determine human
behaviour, because of the large number of unconnected or very tenuously
associated phenomena which are thrown together under that rubric. One
prominent example is the relationship which is supposed to exist between male
hormones and aggression. Under a variety of stimuli, these hormones are
claimed to provide the biological basis for warfare (understood as organized
group aggression), for the political and economic dominance of men, for juvenile
delinquency rates in young males, for violent crime in general and for reckless
driving (Fausto-Sterling 1985: 125). Fausto-Sterling examines these arguments,
and other familiar ones about the biological basis of sex differences, and shows
them to be unfounded for a variety of reasons. However, she emphasizes a
particular difficulty with arguments of this kind, which is that they imply that the
813
SOCIAL LIFE
relationship between biology and social behaviour can be understood as a

straightforward one of cause and effect. Contemporary research in biology
explicitly rejects this view, arguing instead that biology is a dynamic component
of our existence and not a one-way determinant. As Fausto-Sterling observes, it is
now possible to argue for a more complex analysis, in which
an individual's capacities emerge from a web of interactions between the biological

being and the social environment. ... Biology may in some manner condition
behaviour, but behaviour in turn can alter one's physiology. Furthermore, any
particular behaviour can have many different causes. This new vision challenges the
hunt for fundamental biological causes at its very heart, stating unequivocally that
the search itself is based on a false understanding of biology.
(1985:8)
This 'new vision' of the relationship between biology and behaviour, and the
revised view of biology on which it is based, have been relatively slow to
influence thinking in the social sciences because of the way in which social
scientists have been, and continue to be, haunted by the shadow of biological
determinism, especially in its most recent guise as sociobiology. It was, in part,
to try and combat biologically deterministic arguments that feminist
anthropologists in the 1970s emphasized the importance of distinguishing
biological sex from gender. The idea that the terms 'woman' and 'man' denote
Margaret Mead who, in Sex and Temperament (1935), had argued that
considerable cultural variability exists in definitions of femaleness and
maleness. This approach was extended and developed in the 1970s, and much
new ethnographic evidence for variability in what the categories 'woman' and
'man' mean in different cultural contexts demonstrates clearly that biological
differences between the sexes cannot provide a universal basis for social
definitions. In other words, biological differences cannot be said to determine
gender constructs, and, as a result, there can be no unitary or essential meaning
attributable to the category 'woman' or to the category 'man' (Moore 1988:7).
The distinction between biological sex and gender has proved absolutely
crucial for the development of feminist analysis in the social sciences, because it
has enabled scholars to demonstrate that the relations between women and
men, and the symbolic meanings associated with the categories 'woman' and
'man', are socially constructed and cannot be assumed to be natural, fixed or
predetermined. Cross-cultural data have been particularly useful in this regard,
providing the empirical evidence to show that gender differences and gender
relations are culturally and historically variable.
In spite of this work, however, the actual relationship between biological sex
and the cultural construction of gender has remained largely unexamined, since
this relationship has been assumed to be relatively unproblematic. Thus, while it
is acknowledged that gender constructs are not determined by biological sex
814
UNDERSTANDING SEX AND GENDER
differences, there has been a tendency, in much social science writing, to assume
that gender categories and gender meanings are cultural devices designed to
comprehend and manage the obvious fact of binary sex differences. These sex
differences, in turn, are taken to be clearly visible in the physical attributes of the
human body, and are recognized as crucial for the biological reproduction of
human populations. In short, there has been an implicit assumption that binary
biological sex differences underlie, even if they do not determine, gender
categories and gender relations (Yanagisako and Collier 1987: 15).
However, this point needs further clarification in the light of the arguments
of a number of anthropologists to the effect that some cultures do not
emphasize the biological-by which they appear to mean 'physiological'-
differences between women and men. In other words, differences between
women and men are said to exist in certain domains of social life, for example
with regard to spiritual potency, ritual efficacy or moral worth, but are not
thought to be derived from biological differences. In such instances, women
and men are often conceived to be essentially similar in their physical makeup.
This has led some writers to argue that biology does not, in fact, even underlie
gender constructs, let alone determine them:
Natural features of gender, and natural processes of sex and reproduction, furnish
only a suggestive and ambiguous backdrop to the cultural organization of gender
and sexuality. What gender is, what men and women are, what sorts of relations do
. . .
elaborate upon biological 'givens', but are largely products of social and cultural
processes. The very emphasis on the biological factor within different cultural
traditions is variable; some cultures claim that male-female differences are almost
entirely biologically grounded, whereas others give biological differences, or
supposed biological differences, very little emphasis.
(Ortner and Whitehead 1981:1)
There are two important points to be made about arguments of this kind. First,
they do still posit a radical distinction between (biological) sex and (culturally
constructed) gender. In fact, the distinction they suggest is even more radical
than in those arguments which assume that gender systems are cultural
mechanisms for managing sex differences and the problems of social and
biological reproduction. It is clear that such a radical distinction effectively rules
out altogether any possibility for the social sciences to address the relationship
between biology and culture. The primary difficulty here, as Errington ( 1990)
has pointed out, lies in how to understand human bodies. The meanings given to
bodies, and the practices in which they are engaged, are culturally and historically
highly variable. However, the experience of embodiment-whereby these
meanings and practices are incorporated as the enduring dispositions and
competences of real human agents (Bourdieu 1977:87-95)-is something which
could be said to be universal. Although the exact nature of that experience differs,
unless social scientists are prepared to consider the relationship between
815
SOCIAL LIFE
biological sex and gender-that is, between biological entities and social
categories-they will make no progress in understanding the manifold ways in
which culture interacts with biology to produce that most distinctive of human
artefacts: the human body (Errington 1990: 11-15).
It seems very probable that in coming years much new work will be done on
the question of embodiment and on the relationship between biology and
culture, but this depends not only on the willingness of social scientists to
rethink the radical distinction between sex and gender, but also on the
willingness of certain biologists to give up their outmoded ideas about
biological determinism.
'SEX', SEX AND GENDER

The second issue raised by arguments about whether or not biological
differences underlie gender constructs has been that addressed by Yanagisako
and Collier (1987) in their recent discussion of the relationship between gender
and kinship.
Yanagisako and Collier argue that both gender studies and kinship studies in
anthropology are premised on a Western folk model of human reproduction.
This folk model assumes that the difference between women and men is natural,
given in biology and thus pre-social, and although social constructions are built
upon this difference, the difference itself is not viewed as a social construction
7: . T e act t at t IS Western o mo e assumes t at gen er IS
everywhere rooted in a binary, biologically based sex difference means that
anthropological analysis effectively takes for granted a dichotomy which it should
in fact be aiming to explain (1987:15). Thus, the overall argument advanced by
Yanagisako and Collier is that gender and kinship studies have failed to free
themselves from a set of assumptions about natural differences between people,
in spite of a commitment to a social constructionist perspective.
Their critique is a very powerful one, and is of particular interest since they
anticipate that the questioning of conventional assumptions that it enjoins 'will
eventually lead to the rejection of any dichotomy between sex and gender as
biological and cultural facts' (1987:42). This expectation rests on their view
that both sex and gender (rather than gender alone) are socially constructed,
each in relation to the other. Bodies, physiological processes and body parts
have no meaning outside of socially constructed understandings of them.
Sexual intercourse and human reproduction are not just physiological
processes, they are also social activities. The notion of sex, like the concept of
gender, is constructed within a set of social meanings and practices: it therefore
cannot be a pre-social fact (1987:31). The conclusion to which Yanagisako and
Collier move is that if we recognize that the Western concept of sex is socially
constructed, then we cannot argue that this particular model of 'biological' sex
everywhere provides the 'raw material' for gender constructs, nor can we argue
that it everywhere supplies the basis for people's understanding of the
816
processes of human reproduction. In making this argument Yanagisako and

Collier seek to draw attention to the Western cultural assumptions embedded
in our analytic categories (1987:34).
The socially constructed nature of sex and of biological differences is
further elaborated by Errington ( 1990), who develops a notion of the body as a
system of signs. She demonstrates that within the Western model, genitals are
signs of other differences which are interior to the body and which are
themselves indexical signs of an individual's sexual identity. Errington is here
referring to 'internal' features of sexual difference such as those based on
chromosomes and hormones. As she points out, the contradictions in this
model were all too apparent in the cases of the Olympic athletes who classified
themselves as women, but were reclassified as men when they turned out not to
have perfect chromosomes ( 1990: 19-20). This example serves to emphasize the
point that even the supposed natural or biological facts of sex are subject to
interpretation and reinterpretation in the context of a specific discourse of sex
and sexual identity.
The arguments advanced by Yanagisako and Collier, and by Errington, are
clearly Foucauldian in nature, if not necessarily in inspiration. In the first
volume of his History of Sexuality, Foucault argues that 'sex' is an effect rather
than an origin, and that far from being a given and essential unity, it is, as a
category, the product of specific discursive practices.
anatomical elements, biological functions, conducts, sensations, and pleasures, and

it enabled one to make use of this fictitious entity as a causal principle, an
omnipresent meaning, a secret to be discovered everywhere: sex was thus able to
function as a unique signifier and as a universal signified. Further, by presenting
itself in a unitary fashion, as anatomy and lack, as function and latency, as instinct
and meaning, it was able to mark the line of contact between a knowledge of human
sexuality and the biological sciences of reproduction; thus, without really borrowing
anything from these sciences, excepting a few doubtful analogies, the knowledge of
sexuality gained through proximity a guarantee of quasi-scientificity; but by virtue
of this same proximity, some of the contents of biology and physiology were able to
serve as a principle of normality for human sexuality.
(Foucault 1984: 154-5)
Foucault's point about the mutually constitutive nature of Western discourses
of sexuality and biology underscores the argument, made by Yanagisako and
Collier, about the mutually constitutive nature of the concepts of sex, gender
and kinship in the discourse of anthropology. The realization that sex as a
unitary category is established in and through Western discursive practices
clearly entails that the Western concept of sex cannot be said to underlie gender
constructs around the world. Yanagisako and Collier are therefore correct to
argue that as a concept of analysis, gender should be freed from assumptions
about the biological 'given-ness' of sex, as a foundation from which to
deconstruct the Western model of sex and gender relations on which
817
SOCIAL LIFE
anthropological work in these areas has rested for far too long. Such an aim is,
in any case, firmly in line with feminist theorizing in anthropology which, from
the earliest days of its inception, has sought to deliver an internal critique by
unpacking the Western assumptions underlying many of the analytical
constructs most central to the discipline.
Errington's more radical critique of the Western concept of sex mirrors
other aspects ofFoucault's argument in The History of Sexuality, namely those
about the constructed nature of binary and exclusive sex categories.
Genitals ... along with invisible body fluids and substances of which they are believed
to be signs, are classed in this [Western] culture as part of the 'natural', 'objective'
realm, and humans are assumed to be 'naturally' divided into two categories without
the help of cultural ideas or social institutions-and the main raison d' etre of those two
categories is generally believed, whether by religious persons or by secular
evolutionary biologists, to be reproduction. I will call this the taxonomy of Sex, with a
capital 'S' ... By 'Sex' I mean to include the whole complex of beliefs about genitals as
signs of deeper substances and fluids and about the functions and appropriate uses of
genitals; the assignment of the body into the category of the 'natural' (itself a
culturally constructed category); and the cultural division of all human bodies into
two mutually exclusive and exhaustive Sex categories.
(Errington 1990:21)
While recognizing that 'Sex' is socially constructed, Errington nevertheless
attempts to distinguish between (in her terms) 'Sex', sex and gender. By 'Sex',
s e means a parttcu ar construct o uman o tes preva ent m Euro- menca,
while sex means the physical nature of human bodies, and gender refers to what
it is that different cultures make of sex. She criticizes Yanagisako and Collier for
collapsing the distinction between 'Sex' and sex, on the grounds that while we
can recognize that the Western understanding of 'Sex' is socially constructed, it
is also important to recognize that humans have bodies with distinguishing
genitals and that there is therefore a material reality-i.e. sex-which must be
taken into account when discussing the meanings which cultures give to bodies
and embodied practices-i.e. gender (Errington 1990:27-8).
The point Errington makes is an important one, but there are still further
confusions to be sorted out in this discussion about the relationship between
'Sex', sex and gender. While both Errington and Yanagisako and Collier
acknowledge that 'Sex' is an effect of a particular Western discourse for
comprehending and categorizing the apparent differences between women and
men-a discourse which underlies the analytic categories of anthropological
theorizing-they do not seem to acknowledge the point that sex is everywhere
'Sex'; in other words, that although the particular constitution, configuration
and effects of 'Sex' clearly vary between cultures, there is, in each case, no way
of knowing sex except through 'Sex'. The specific Western discourse of 'Sex'
may have influenced anthropological theorizing, but there are many other
discourses of 'Sex', and these discourses need to be specified through
anthropological analysis.
818
The fact that all cultures have ways of making sense of, or giving meaning to,
bodies and embodied practices, including physiological processes and bodily
fluids and substances, means that all cultures have a discourse of 'Sex'. In each
case, this discourse stands in a relationship of partial dependency and partial
autonomy with other discourses, including, very often, what anthropologists have
referred to as the discourse of gender. Gender discourses themselves are refracted
in many other discursive domains of culture, giving rise in some instances to
discourses of power, potency, cosmology, fertility and death which also appear
highly gendered. One example from Western societies of such a gendered
discourse is that of nature and culture. Conversely, and by virtue of that relation
of mutual constitution so well described by Foucault, the discourse of gender is
itself shot through with ideas about what is natural and what is cultural.
There is, in short, no way in any culture to approach sex except through the
discourse of 'Sex', and this must surely be particularly true of cultures which
have lacked, either now or in the past, the technological means for revealing the
true nature of the underlying physiological processes and substances, and thus
for distinguishing between sex and 'Sex'. What Errington and Yanagisako and
Collier do not seem to realize is that the notion of sex, of a biological property
or set of biological processes, existing independently of any social matrix, is
itself the product of the biomedical discourse of Western culture. There is a
fundamental sense in which, outside the parameters and spheres of influence of
this biomedical discourse, sex does not exist. In other words, in most cultures in
t e wor , w ere m tgenous or oca now e ge retgns supreme, t ere IS no sex,
only 'Sex'.
Anthropology has been very slow to grasp this point, partly because the
question is obscured, paradoxically, by an approach which posits a radical
separation of sex from gender, and, by extension, of biology from culture. The
question we have to ask ourselves for the future is whether it makes sense,
except where spurious biological arguments are being used to justify
discriminatory social practices, to insist upon separating sex from gender, when
the real issue is not sex, but 'Sex'. A further question, however, needs to be
posed: is it appropriate to separate 'Sex' from gender, when 'Sex' is understood
as the culturally specific discursive practices which make sense of body parts
and their relation, indexical or otherwise, to physiological processes and
substances, including those associated with human reproduction? This
question is the more difficult one, and involves a consideration of the problem
of binary sex categorization.
The determination, as I have already pointed out, of two mutually exclusive
and fixed categories of sex, the female and the male, is an effect of the Western
cultural discourse of 'Sex'. This discourse stands in a mutually constitutive
relation with biomedical discourse, such that the former becomes scientized,
while the latter is constructed according to a set of understandings about the
meaning of sex differences and about the relations to be established between what
is cultural and what is natural (Hubbard 1990). The difficulty with the Western
819
SOCIAL LIFE
discourse on 'Sex', as Yanagisako and Collier point out, is that the 'naturalness' of
binary sex categorization is apparently reinforced by the fact that biological
females and biological males are required for human sexual reproduction.
However, we do not have to suppose that there are peoples around the world who
are unable to recognize the differences between female and male genitalia, or who
are unaware of the different roles which women and men play in sexual
reproduction, in order to question the assumption that biological differences
between women and men provide a universal basis for cultural categorizations
that assign every individual to one or other of two, discrete and fixed categories,
'female' and 'male', in the manner of Western discourse. There is ample
ethnographic evidence to show that this kind of binary categorization is culturally
specific, and that it does not arise automatically from the recognition of
differences in roles and in physical appearance.
This point is most apparent when we turn to consider theories about the
physical constitution of persons. In many societies it is believed that persons are
made up of female and male parts or substances. Levi-Strauss (1969) identified
what he called the flesh-bone complex for South Asian societies, in which bones
are inherited from the father and flesh from the mother. Marilyn Strathern
( 1988) has recently discussed the multiply gendered and partible nature of bodies
as they are conceived by the people of the Mount Hagen region in the New
Guinea Highlands. Hageners see gender as a process rather than as a category:
how one becomes rather than what one is. Likewise, according to Meigs (1990),
t e Hua-anot er Htg an s peop e-mcorporate m t etr vtew o gen er t e
idea that persons can become more female or more male depending on how much
they have been in contact with and have ingested certain bodily substances
thought to be female (e.g. menstrual blood, parturitional fluids and vaginal
secretions). Hua men take in these substances as a result of eating food prepared
by reproductively active women, as a consequence of sex, and through daily
casual contact (Meigs 1990: 109). This means of conceptualizing gender is a
processual and multiple one, and it exists in parallel with a mode of categorization
based on external genitalia. It is evident from recent ethnographies that many
societies have more than one way of conceptualizing and classifying gender, and
that this fact has been obscured by the reliance of the social sciences on a model of
gender which stresses the fixed and binary nature of sexual difference.
One difficulty which ethnographic data of this kind raise is how to establish,
and indeed whether it is possible to establish, the distinction between sex and
gender at all. If sexual difference is thought to exist within bodies as well as
between them, should we view this as a matter of sex or of gender (Moore
1993)? This question becomes particularly crucial in the light of the previous
argument that sex as well as gender must be understood as socially constructed.
The result is that the analytic distinction between sex and gender seems very
blurred. At least in the context of cross-cultural analysis, it seems that the
attempt to uphold a radical distinction between sex and gender will not
necessarily help us to gain an improved theoretical perspective.
820
INEQUALITY AND SUBORDINATION
However, one area in which a distinction between sex and gender has proved
very useful to the social sciences is the analysis of gender inequalities. The
immediate question raised by cross-cultural analysis has been how to account
for the enormous variability in local understandings of gender and gender
relations, in the context of what appears to be the universal subordination of
women to men (Moore 1988: eh. 2; see also Beteille in this volume, Article 37).
It seemed that there must be some cultural or sociological regularities which
would account for male dominance (Rosaldo 1974, Reiter 1975). The value of
enquiry into this question was that it drew the social sciences away from a
debate about the biological basis for gender inequality towards a discussion of
its cultural and sociological determinants. However, there are a number of
difficulties with the theoretical solutions which have been proposed to the
question of universal female subordination.
Sherry Ortner (1974) proposed that the universal devaluation of women is
connected to their symbolic association with the realm of nature, which is itself
viewed as subordinate to the realm of culture associated with men (Ortner 1974).
At the same time, Michelle Rosaldo (1974) suggested that it is women's
association with the domestic sphere, in contrast to men's dominance in the
encompassing public sphere of social life, which accounts for the universal
tendency for women to be subordinated to men. Ortner's account thus stressed
cu tura an sym o tc actors, w 1 e Rosa o s emp astze socw ogtca
considerations. Both these explanations have been widely criticized (Moore 1988:
eh. 2). A number of scholars have pointed out that the distinction between
women and men is not necessarily associated with the division between nature
and culture, and that concepts or notions of nature and culture vary greatly from
one society to another, if indeed they exist at all (MacCormack and Strathern
1980). The problem once again lies in the imposition of an analytic dichotomy
derived from Western thought in situations where it is not always appropriate. In
the case of the domestic-public distinction, Rosaldo made it clear that women's
identification with the domestic domain is a consequence of their role as mothers
(Rosaldo 1974:24). This view was strongly criticized by a number of writers, who
argued that a search for the universal causes of gender inequality inevitably ends
up implying some form of biological determination, even if the theory proposed
appears to offer a social or cultural explanation (e.g. Leacock 1978, Sacks 1979).
Rosaldo later modified her view and argued that the domestic-public distinction
could not provide a universal explanation for women's subordination because,
both analytically and sociologically, it is the product of historical developments in
Western society (Rosaldo 1980).
A variant of the domestic-public dichotomy emerged in Marxist feminist
writing on the position of women and the sexual division of labour. The
starting point for much of this debate was the distinction, first made by En gels,
between production and reproduction. Engels regarded the subordination of
821
SOCIAL LIFE
women as being linked to their exclusion from the sphere of production (Moore
1988:46-49). This debate was initially very ethnocentric in formulation and
tended to draw much of its theoretical inspiration from arguments about the
role of women's reproductive labour under conditions of capitalist production,
where such labour is situated inside the household-as opposed to productive
labour which is situated outside it. This is obviously an inappropriate model for
understanding production systems in which women are engaged in both
productive and reproductive labour within the household (Moore 1988: eh. 3).
There was also a tendency to assume that the nature of women's reproductive
labour does not change over time. Some writers conflated reproduction with
biological reproduction (Meillassoux, 1981 ), while yet others equated it with
women's domestic labour (Boserup, 1970). In spite of attempts to clarify the
notion of 'reproduction', and to include not only biological reproduction and
domestic work, but also social reproduction in the wider sense (Harris and
Young 1981), there has been a persistent tendency to link women's
subordination to their role in reproduction, and thus to their position in the
sexual division of labour.
However, gender relations cannot be understood as a simple reflection of the
sexual division of labour. Cultural representations of gender rarely mirror
accurately women's and men's activities, their contributions to society, or their
relations with each other (Ortner and Whitehead 1981:10). A number of
feminist scholars have suggested that women's position in society is
etermme y t e extent to w tc t ey contro t eu own a our an t e
products of their labour. But ethnographic analysis has revealed that even this
proposition is too straightforward. The difficulty with investigating gender
inequality is that one has to analyse not only the political and economic
contexts in which gender relations are operative, but also the cultural and
symbolic meanings accorded to gender differences.
Ortner and Whitehead have suggested a method for combining the symbolic
and the sociological approaches by focusing on what they call 'prestige
structures'. Prestige structures are understood as those lines of social
evaluation, positions and roles through which a given set of social statuses and
cultural values are reproduced (Ortner and Whitehead 1981: 13). As Yanagisako
and Collier point out (1987:27), it is not easy to grasp exactly what is meant by
this. Ortner and Whitehead, however, suggest that gender systems-that is
gender meanings and gender relations-are themselves prestige structures and
that they are correlated in many societies with other axes of social evaluation,
such as rational versus emotional or strong versus weak (Ortner and Whitehead
1981:16-17). This argument is compelling, in one sense, because it does help us
to understand why evaluative statements are so often rendered in gendered
terms, when what is actually being referred to are relations between people of
the same sex or between people of different classes, rather than relations
between women and men. Nonetheless, Ortner and Whitehead assume, rather
than demonstrate, that prestige structures are rooted in the male-dominated,
822
public sphere of social activity, and that they encompass or dominate the
domestic sphere of social life (Ortner and Whitehead 1981:19). As Yanagisako
and Collier note, this means that the notion of prestige structures does little
more than replicate the problems inherent in the domestic-public distinction,
and that as such it simply assumes a priori what it should be seeking to
investigate (Yanagisako and Collier 1987:28).
The notion of a prestige structure is nevertheless useful because it directs
attention to the social evaluations of women's and men's behaviour, and to the
meanings given to the differences in their activities and roles. Marilyn
Strathern (1981) has used the idea of social evaluation to demonstrate that in
Hagen, pursuing socially valued goals ('acting like a man') and pursuing
individual interests ('acting like a woman') are types of behaviour open to both
women and men, and that although gender idioms are used to describe moral
qualities and socially valued behaviour, this does not determine how the actual
behaviour of individual women and men would be evaluated in any particular
context. The disparity between the cultural representations of gender and the
activities of individual women and men raises once more the questions of how
the status of women is to be evaluated in any given context, and of what kind of
information is necessary to be able to determine the nature and extent of
women's subordination to men.
For example, Keeler has pointed out for Java that although gender
differences can be used to make distinctions among individuals, differences
ase on sty e an status can a so e use to o t e same t mg Kee er
1990: 128-9). Keeler notes that while gender distinctions are relevant in
domestic and public life, they do not prevent women from exercizing control
within the household, and from managing their own money as well as their
husband's income. Women apparently participate fully in discussions about
children's education, business plans and marriage arrangements. Men do some
child care, and both women and men farm, and take part in business activities.
Many Javanese women enjoy positions of prestige and respect in public life, as
government officials and heads of schools, although the numbers holding high
office are not great (Keeler 1990: 129-30). Nonetheless, despite women's
activities and achievements, they tend to be described as lacking socially and
morally valued characteristics such as self-control, patience, spiritual potency,
sensibility and insight. Keeler links this overtly negative discourse to the fact
that women are believed to lack potency, which is related in Javanese thinking
to both prestige and status. However, as he points out, it would be a mistake to
assume that because women lack culturally defined prestige they are
automatically considered to be inferior in social life.
MULTIPLE MODELS AND MULTIPLE DISCOURSES

What the example from Java demonstrates is the difficulty of combining
cultural representations of gender relations and local views of women and men
823
SOCIAL LIFE
as persons, together with the actual activities and roles of women and men, to
produce a single model of gender relations. It is similarly difficult to combine
these different types of data to arrive at some formulation of a single position
which women could be said to hold in society (Strathern 1987). For one thing,
gender is not the only axis of social differentiation within a society, and there
may be manifest differences between women due to class, race, religion or
ethnicity. This gives rise to a situation in which not all women are subordinate
to all men. For example, in the Javanese case, high-ranking women have low-
ranking men as subordinates, and in many contemporary societies, class and
race are significant axes of social differentiation which organize access to
resources, including education, employment and public office, in ways which
often cross-cut gender distinctions. However, recent work in anthropology has
emphasized that it is a mistake to imagine that societies have only one model, or
only one discourse, of gender and gender relations. The recognition of the
existence of multiple models and discourses, and the investigation of how those
models and discourses intersect in any given context, is providing a new
direction for the analysis of gender in anthropology.
Anna Meigs points out that until recently, anthropological work on gender has
been based on three assumptions: first, that there are two clear-cut and
monolithic categories, female and male; second, that female status is singular and
unitary, and third, that each society has a single gender model (Meigs 1990: 102).
The cross-cultural variability of the categories female and male, and the manifold
nature o ema e status, ave a rea y een tscusse . However, wtt re erence
again to the Hua people of Highland New Guinea, Meigs details three gender
models or discourses which exist simultaneously in their society. The first of
these emphasizes that female bodies are disgusting and dangerous to men, and
that women lack knowledge and insight. This model is enshrined in many rituals
and social institutions, especially male initiation. The second is quite contrary to
the first and concerns the Hua belief that the female body is superior to the male.
Hua men imitate menstruation and believe that they can become pregnant. This
second model is embodied in myth, local belief and ritual practices, including the
bloodletting which is an imitation of menstruation. The third model is egalitarian
and emphasizes that although female and male bodies are different, neither one is
more desirable than the other. This model emphasizes interdependency and
complementarity, and is connected to the respect which women and men show
each other in daily life (Meigs 1990: 102-3).
Meigs's work is useful because she emphasizes not only that societies will
probably have more than one gender discourse or model, but also that many of
these different ideas about gender, and about the nature of women and men,
will likely conflict with and contradict each other. She also points out that from
among these multiple models or discourses of gender, certain of them are more
appropriate to particular contexts, or to particular stages in the life-cycles of
individuals, than others. Thus, young Hua males at initiation are taught that
female bodies are dangerous, and they are forbidden to look at women or eat
824
food from women's gardens or consume any of the foods which resemble the
female reproductive system. However, as they get older, these rules fall away,
and the ideal of sexual avoidance is replaced by one of relative egalitarianism
and co-operation (Meigs 1990: 103-4).
The recognition of the existence of multiple gender models in anthropology
has been stimulated by a transformation, over the last fifteen years or so, in the
anthropological understanding and definition of culture. Whereas culture was
once defined as an overarching set of beliefs and customs that were equally
shared by all members of a society, recent work in the social sciences has
emphasized the contested and contingent nature of culture. One result of this
has been a shift away from imagining culture and social life as based on rules
and rule-following to a view which emphasizes that they are constituted
through performance and practice (Ortner 1984). The current emphasis in the
cross-cultural study of gender on the way in which women's and men's
activities are informed by a multiplicity of discourses of gender and gender
relations, which are themselves produced and reproduced through those same
activities, is a consequence of this shift in the understanding of culture.
CHANGING DISCOURSES OF GENDER AND

GENDER RELATIONS
The sexual division of labour is constantly being transformed as social and
economic c ange ta es p ace. In t e process, oca 1 eas a out women an men,
and about the nature of gender relations, change too. The analysis of what
determines these changes and of how they take place is complex, and factors
which have to be taken into account include the nature of kinship systems,
existing political and state structures and the level of development of the
economy (Moore 1988: eh. 4). Many writers have noted that colonialism and
missionary activity, in their attempts to extract male labour and to construct a
public domain in contrast to a domestic life, have often had deleterious effects
on local systems of gender and gender relations. Here, economic, political and
socio-ideological forces have worked together to transform both concepts of
gender and the sexual division of labour. However, it is a mistake to
oversimplify this picture. The forces of change have been uneven in their
spread and impact, and have not had a uniform effect on gender and gender
relations. Until recently, work in this area tended to imply that local systems
were simply passive and unable to resist the imposition of exogenous socio-
economic and political structures (Moore 1988:74). In the following
paragraphs I briefly review some new studies which challenge this assumption.
Cristina Blanc-Szanton (1990) has described the effect of Spanish
colonization and Catholicism, and the later impact of North Americans and the
modern state, on concepts of gender among the Ilonggo of the Philippines. She
notes that despite three centuries of pressure on aspects of sexual morality and
gender symbolism, including attempts to impose a particular view of sexuality
825
SOCIAL LIFE
and to restructure the nature of gender relations, the result has not been
capitulation to external gender models. Even in the context of unequal and
coercive power relations, the Ilonggo have selectively absorbed and adapted
new notions of gender and, in the process, have creatively reworked many of
these new ideas. For example, Blanc-Szanton argues that in the 1970s, the
Ilonggo manifested a 'remarkable mixture of symbolic references to machismo
and virginity of both Spanish and Judeo-Christian origin, an awareness of the
new sexuality, but also an emphasis on the sameness and comparability of the
sexes' (Blanc-Szanton 1990:378). In her conclusion, she suggests that the
reason for the syncretic vitality of the Ilonggo gender system lies in the fact that
while the Spanish, the Americans and the modern state have each sought to
transform this system, they have never been able to undermine the
reproductive basis of the system itself. Ilonggo gender is based on a notion of
non-hierarchical comparability and equivalence which is enshrined in features
of social organization, including bilateral kinship, and, unless this principle is
undermined, the system will continue to respond syncretically and adaptively
(Blanc-Szanton 1990:381-2).
Aihwa Ong (1987, 1990) has examined young women's participation in
industrial production in Malaysia, and has investigated the ways in which
sexual symbolism and gender constructs are reinterpreted and transformed in
situations of rapid social change and power conflict. She argues that conflicts
over class and national identity are often constructed as conflicts over gender
an gen er meanmgs, t us trans ormatwns m gen er re atwns are t e ey to
understanding processes of social change. Ong also emphasizes the way in
which contradictory and competing discourses on gender can be produced as a
result of the different interests and struggles of social groups.
Old concepts of gender and gender relations can acquire new meanings and
serve new purposes under changed circumstances (Ong 1990:387). One notable
example of this last point is the way in which factory managers guarantee the
safety of young women workers, maintain control over their movements
between home and factory, and impress upon parents their concern for the
moral reputations of the young women. This system of surveillance accords
well with parental values and wishes, and while it has the added advantage of
ensuring an adequate supply of labour to the factory, it also provides the
management with a socially legitimate method for controlling women within
the factory, thereby contributing to the formation of a disciplined and docile
workforce. As Ong points out, the traditional moral authority of men in
domestic affairs has been transformed into a system for the industrial
exploitation of Malay women (Ong 1990:402-3).
On the factory floor, the regime is paternalistic, with male foremen
subjecting female workers to control, questioning and surveillance. This
situation is made worse by the public image of factory workers as morally loose
women who pay insufficient attention to family values and Muslim standards of
behaviour. Women workers daily try to resist the factory regime through such
826
methods as making excuses to leave the factory floor for religious reasons or
'female problems' (Ong 1990:417). A new phenomenon, however, has been the
increase in episodes of spirit possession, which often result in women shouting
at and resisting male supervisors. The resistance to male control on the factory
floor is paralleled by a resistance to male control in the domestic sphere.
Working daughters often demonstrate their resistance to their fathers and to
the ethic of communal consumption by protesting against undesired marriages,
using their savings to plan an alternative career, engaging in premarital sex and
refusing money to parents who remarry. These forms of resistance, in the home
and the workplace, have to be understood in the context of the increased
control over young women's lives exercised through the discipline and
surveillance of the factory regime and through the increased vigilance of
Islamic state institutions. The result is a situation in which young women
actively resist, while simultaneously maintaining family loyalty, Islamic
asceticism and male authority as central values (Ong 1990:420). The struggle
over gender and gender relations is also the struggle over family, religious and
national identity in the context of political and economic domination.
The confusing nature of resistance and its ambivalent relation to
emancipation are well demonstrated by Abu-Lughod's (1986, 1990) work on
Bedouin women. Bedouin women have traditionally evaded male control
through the institution of the sexually segregated women's world, in which
they are able to avoid male surveillance and enjoy a degree of self-
etermmatwn. T ey ave a so exercise power an a certam amount o contro
through their resistance to undesirable marriages. Lyric poetry and other
subversive discourses provide a further medium through which dominant
discourses on gender and gender relations can be reinterpreted and resisted
(Abu-Lughod 1986). However, Abu-Lughod notes that these traditional forms
of resistance are being eroded. The poetry is becoming increasingly associated
with young men, who sing the songs, make money out of locally produced
cassettes, and use these poems to resist the power of older kinsmen (Abu-
Lughod 1990:325).
Sedentarization has led to a situation in which women's movements are
more closely controlled, and in which women spend more time veiled and less
time in the relative freedom of the desert camps. However, this has been
paralleled by the increasing consumer orientation of young women, as shown in
their purchase of items like face creams and lingerie, which has put them at
odds with their mothers and female kin. Young women are now less interested
in resisting marriage than in trying to secure the kind of marriage which will
provide them with access to consumer goods and fulfil their fantasies of a
romantic match with an educated and progressive man. Young women aspire to
be housewives in a way which their mothers would never have done, because
their own security and standard of living is dependent on the favour of
husbands in a situation in which everything costs money, but in which women
have no independent access to cash (Abu-Lughod 1990:326-7). Men's power
827
SOCIAL LIFE
over women now includes the power to buy things and to give or withhold these
things. Along with the desire for consumer goods goes a desire for Egyptian
music and soap opera, within which many of the new ideas about gender and
gender relations are encoded. Older Bedouin women and men try to resist these
forms of Egyptianization. As Abu-Lughod points out, although young women
are resisting their elders in taking up new patterns of consumption, they are
simultaneously becoming caught up in the new forms of subjection which these
patterns entail (Abu-Lughod 1990, 328).
All forms of social change involve the reworking of gender relations to
greater or lesser degrees. This is because changes in production systems involve
changes in the sexual division of labour; political conflicts involve the
reconfiguration of power relations within the domestic domain and beyond;
and gender as a powerful form of cultural representation is caught up in
emerging struggles over meaning and in attempts to redefine who and what
people are. This has nowhere been clearer than in the transformations in
gender relations which have been sought in many (formerly) socialist and
communist countries (Moore 1988: 136-49). That the policies pursued by these
countries have enjoyed only partial success demonstrates that politicians, like
social scientists, have yet to comprehend how and why gender relations might
be transformed in present-day societies and in those of the future.
Abu-Lughod, L. (1986) Veiled Sentiments: Honor and Poetry in a Bedouin Society,

Berkeley: University of California Press.
--(1990) 'The romance of resistance: tracing transformations of power through
Bedouin women', in P.Sanday and R.Goodenough (eds) Beyond the Second Sex:
New Directions in the Anthropology of Gender, Philadelphia: University of
Pennsylvania Press.
Blanc-Szanton, C. (1990) 'Collision of cultures: historical reformulations of gender in the
lowland Visayas, Philippines', in J.Atkinson and S.Errington (eds) Power and
Diffirence: Gender in Island Southeast Asia, Stanford, Cal.: Stanford University Press.
Boserup, E. (1970) Women's Role in Economic Development, London: George Alien &
Unwin.
Bourdieu, P. (1977) Outline of a Theory of Practice, trans. R.Nice, Cambridge:
Errington, S. (1990) 'Recasting sex, gender and power: a theoretical and regional
overview', in J.Atkinson and S.Errington (eds) Power and Difference: Gender in
Island Southeast Asia, Stanford, Cal.: Stanford University Press.
Fausto-Sterling, A. (1985) Myths about Gender: Biological Theories about Women and
Men, New York: Basic Books.
Foucault, M. (1984) The History of Sexuality, vol I, Harmondsworth: Penguin.
Harris, 0. and Young K. (1981) 'Engendered structures: some problems in the analysis
of reproduction', in ].Kahn and J.Llobera (eds) The Anthropology of Pre-Capitalist
Societies, London: Macmillan.
828
Hubbard, R. (1990) The Politics of Women's Biology, New Brunswick, NJ: Rutgers
University Press.
Keeler, W. (1990) 'Speaking of gender in Java', in J.Atkinson and S.Errington (eds)
Power and Difference: Gender in Island Southeast Asia, Stanford, Cal.: Stanford
University Press.
Leacock, E. (1978) 'Women's status in egalitarian society: implications for social
evolution', Current Anthropology 19(2):247-5.
Levi-Strauss, C. (1969) The Elementary Structures of Kinship, Boston: Beacon Press.
MacCormack, C. and Strathern, M. (eds) (1980) Nature, Culture and Gender,
Mead, M. (1935) Sex and Temperament in Three Primitive Societies, New York: William
Morrow & Co.
Meigs, A. (1990) 'Multiple gender ideologies and statuses', in P.Sanday and R.
Goodenough (eds) Beyond the Second Sex: New Directions in the Anthropology of
Gender, Philadelphia: University of Pennsylvania Press.
Meillassoux, C. (1981) Maidens, Meal and Money, Cambridge: Cambridge University Press.
Moore, H.L. (1988) Feminism and Anthropology, Cambridge: Polity Press.
--(1993) 'The differences within and the differences between', in Tdel Valle (ed.)
Gender and Social Life, London: Routledge.
Ong, A. (1987) Spirits of Resistance and Capitalist Discipline: Factory Women in
Malaysia, Albany: State University of New York Press.
--(1990) 'Japanese factories, Malay workers: class and sexual metaphors in West
Malaysia', in J.Atkinson and S.Errington (eds) Power and Difference: Gender in
Island Southeast Asia, Stanford, Calif.: Stanford University Press.
rtner, . eory m ant ropo ogy smce t e SIXties , omparattve tu tes tn
Society and History 26(1): 126--66.
--(1974) 'Is female to male as nature is to culture?', in M.Rosaldo and L.Lamphere
(eds) Woman, Culture and Society, Stanford, Calif.: Stanford University Press.
Ortner, S. and Whitehead, H. ( 1981) 'Introduction: accounting for sexual meanings', in
S.Ortner and H.Whitehead (eds) Sexual Meanings: The Cultural Construction of
Gender and Sexuality, Cambridge: Cambridge University Press.
Reiter, R. (1975) Toward an Anthropology of Women, New York: Monthly Review Press.
Rosaldo, M. (1974) 'Woman, culture and society: a theoretical overview', in M.Rosaldo
and L.Lamphere (eds) Woman, Culture, and Society, Stanford, Cal.: Stanford
University Press.
--(1980) 'The use and abuse of anthropology: reflections on feminism and cross-
cultural understanding', Signs 5(3):389-417.
Sacks, K. (1979) Sisters and Wives: The Past and Future of Sexual Equality, Westport,
Conn.: Greenwood Press.
Strathern, M. (1981) 'Self-interest and the social good: some implications of Hagen
gender imagery', in S.Ortner and H. Whitehead (eds) Sexual Meanings: The Cultural
Construction of Gender and Sexuality, Cambridge: Cambridge University Press.
--(1987) 'Introduction', in M.Strathern (ed.) Dealing with Inequality: Analysing
Gender Relations in Melanesia and Beyond, Cambridge: Cambridge University Press.
--(1988) The Gender of the Gift, Berkeley: University of California Press.
Yanagisako, S. and Collier, J. (1987) 'Toward a unified analysis of gender and kinship',
in ].Collier and S.Yanagisako (eds) Gender and Kinship: Essays Toward a Unified
Analysis, Stanford, Cal.: Stanford University Press.
829
SOCIAL LIFE
FURTHER READING
Atkinson, J. and Errington, S. (eds) (1990) Gender and Power in Island Southeast Asia,
Stanford, Calif.: Stanford University Press.
Bledsoe, C. (1980) Women and Marriage in Kpelle Society. Stanford, Cal.: Stanford
University Press.
Caplan, P. (1985) Class and Gender in India: Women and their Organisations in a South
Indian City, London: Tavistock.
Collier, J. and Yanagisako, S. (1987) Gender and Kinship: Essays Towards a Unified
Analysis, Stanford, Calif.: Stanford University Press.
Herdt, G. (1987) Guardians of the Flutes: Idioms of Masculinity, New York: Columbia
University Press.
Kendall, L. (1985) Shamans, Housewives and Other Restless Spirits: Women in Korean
Ritual Life, Honolulu: University of Hawaii Press.
Kondo, D. (1990) Crafting Selves: Power, Gender and Discourses of Identity in a Japanese
Workplace, Chicago: University of Chicago Press.
Lamphere, L. (1987) From Working Daughters to Working Mothers, Ithaca: Cornell
University Press.
di Leonardo, M. ( ed.) (1991) Gender at the Crossroads of Knowledge, Berkeley:
Meigs, A. (1984) Food, Sex and Pollution, New Brunswick, NJ: Rutgers University
Press.
Moore, H.L. (1986) Space, Text and Gender: An Anthropological Study of the Marakwet
of Kenya, Cambridge: Cambridge University Press.
Ong, A. (1987) Spirits of Resistance and Capitalist Discipline, Albany: State University of
New York Press.
Ortner, S. and Whitehead, H. (eds) (1981) Sexual Meanings, Cambridge: Cambridge
University Press.
Redclift, N. and Sinclair, T. (eds) (1990) Working Women: International Perspectives on
Labour and Gender Ideology, London: Routledge.
Standing, H. (1990) Dependence and Autonomy: Women's Employment and the Family in
Calcutta, London: Routledge.
Strathern, M. (1988) The Gender of the Gift, Berkeley: University of California Press.
830
30
SOCIALIZATION, ENCULTURATI ON
AND THE DEVELOPMENT OF
PERSONAL IDENTITY
Fitz John Porter Poole
This article explores a particular view of the anthropological study of processes

of socialization and enculturation focused on the acquisition of interwoven
social, cultural, and personal dimensions of identity, and of their meanings in
the context of socio-cultural understanding and action. The article proceeds
rom presentmg a genera portrait o t e c aracter o socta tzatwn an
enculturation, to mapping analytically the mosaic of personhood, selfhood, and
individuality as variable and variably interconnected aspects of identity, to
considering the ways in which these aspects of identity emerge and are
elaborated in the course of early and middle childhood. The exercise is more a
sketch of an agenda than a comparative assessment of ethnographic studies. It
is suggested that this focus on the confluence of processes of socialization and
enculturation, on the one hand, and of schemata of person, self, and
individuality, on the other, nevertheless holds some promise for making sense
developmentally of Sapir's ( 1949: 515) prescient claim that the 'true locus of
culture is in the interactions of specific individuals and, on the subjective side,
in the world of meaning which each one of these individuals may ... abstract for
himself from his participation in these interactions'.
SOCIALIZATION AND ENCULTURATION

The processes of socialization and enculturation, though interconnected, are
analytically distinguishable in emphasis (cf. Herskovits 1948:38, Mead 1963,
Schwartz 1976). Mead (1963: 185) suggests that socialization has to do with 'the
set of species-wide requirements and exactions made on human beings by
human societies', whereas enculturation refers to 'the process of learning a
culture in all its uniqueness and particularity'. From the perspective of this
831
SOCIAL LIFE
article, however, my focus is on how, when, where, why, with whom, under what
circumstances, and with what individual significance and psychological force,
personal configurations of learning of the social and of the cultural occur, and
on how social and cultural forms and forces are intertwined in the process of
such learning. Thus, the distinction between socialization and enculturation is
predicated, in part, upon a further analytic distinction, between society as
consisting in pragmatically constituted, negotiated, co-ordinated, and
replicated patterns of interaction, and culture as consisting in socially
distributed and more or less shared knowledge (including knowledge of social
interaction) manifested in those perceptions, understandings, feelings,
evaluations, intentions, and other orientations that inform and shape the
imagination and pragmatics of social life from the imperfectly shared
perspectives of social actors. Somewhat controversially, this perspective instals
the individual as the focal locus of culture and as the significant agent in social
interaction, and thus calls for a 'person-centred enthnography' (Le Vine 1982).
Socialization
Socialization implicates those interactive processes-their structures, contents,
contexts, and actors-in and through which one learns to be an actor, to engage
in interaction, to occupy statuses, to enact roles, and to forge social
relationships in community life, as well as acquiring the competence, skills,
sensitivities, an IspositiOns appropnate to sue socia participatiOn. It IS
concerned with the character and condition of learning processes, entailed in
the learner's participation in social practices appropriate to particular
relationships, by which he or she becomes adapted to, integrated in, and
competent at those interactions involved in becoming or being an actor in
society. It is bound up with the social apparatuses, institutional arrangements,
or socio-ecological contexts, and with significant categories of persons, that
together define and exemplify the ranges of socially appropriate behaviours for
people having certain social identities and occupying particular statuses in the
varied situations of community life (Bronfenbrenner 1979, Le Vine 1969, 1977).
If, as Giddens (1979:251) maintains, some negotiation of mutual knowledge is a
necessary precondition for social interaction, then socialization is tied in with
the interactive processes that promote and facilitate such negotiation through
particular forms of 'interactional display of the socio-cultural environment'
(Wentworth 1980:68).
Studies of socialization not only focus on the character of the situations and
events in which learning occurs, but also attend to the organization of the
interactive processes that promote and facilitate learning of different kinds and
significance. It is generally agreed that social interaction both enfolds and
shapes the ways in which socio-cultural phenomena are encountered and
learned (Bruner and Bornstein 1989), although the manner and circumstances
in which this occurs, and its consequences, are highly variable. One of the most
832
SOCIALIZATION, ENCULTURATION AND IDENTITY
important contexts of interactive learning has been identified by Vygotsky

(1978) as the 'zone of proximal development'. Learning in this zone involves
interactions in which less and more knowledgeable and skilled actors actively
contribute to an intersubjective collaboration through engaging in joint
activities whose form, focus and direction is intricately and sensitively adjusted
to the demands of 'scaffolding' or 'guided participation' (Rogoff 1990). By
subtly setting goals and shaping activities, the more competent participant in
the interaction attends to the less adept partner's kinds and degrees of
knowledge, skill, involvement, reaction, and elicitation, as the learner is given
the opportunity to experiment with and discover new understandings. It is also
in such contexts of interaction, however, that often tacit messages about
personhood, selfhood, and individuality are communicated, apprehended,
interpreted, negotiated, and sometimes internalized (Cooley 1902, Lee and
Hickman 1983, Lee et al. 1983, Mead 1934).
Within anthropology, probably the most influential model of socialization
derives from the studies ofWhiting and Whiting (1975), which focused on the
social processes of transmission of cultural knowledge to children, who were
nevertheless portrayed as more or less passive recipients of such knowledge in
the context of 'learning environments'. In the service of comparison,
generalization, and explanation, this model attends to supposedly objective
features of the milieu of child-rearing and of children's behaviour, but it
explores little of how children construct their own experiences of the socio-
cu tura wor m w tc t ey IVe an o ow t ey put t ose expenentta
understandings to varied personal and social use. In the most recent
formulation of Whiting and Edwards (1988:2), the effort is directed toward
'observations collected in naturally occurring settings of children', with
particular attention to children's routine activities in space and time and to the
significant social others with whom they interact in these 'learning
environments'. Despite the importance of understanding the salient
characteristics of the social settings, partners, and interactions that influence
children's behaviour, this approach is insufficient on its own, and needs to be
complemented by an exploration of how the child-actively and creatively-
apprehends, experiments with, represents, communicates, and internalizes
aspects of cultural knowledge, and brings them to bear in making sense of and
in navigating through the social world of the community. From the perspective
of this article, an understanding of the salience of 'learning environments' must
take into account the nuances of children's own understandings of them.
Enculturation
Turning from socialization to enculturation, the focus is on those processes by
which one acquires understanding, orientation, and competence in the
ideational realm that constitutes a culture-schemata, scripts, models, frames,
and other images of the organization and contextualization of knowledge that
833
SOCIAL LIFE
are culturally constituted, socially distributed, and personally construed.

Enculturation concerns the acquisition of those rules, understandings and
orientations that provide, among other things, contoured maps of the landscape
of community life and heuristic guides for effective participation. Through the
lens of a 'person-centred ethnography', the study of enculturation attends to
how individuals come to develop more or less adaptive (or maladaptive)
interpretations, representations, expectations, evaluations, feelings, intentions,
and so on, concerning their socio-cultural milieu and their positions within it
from perspectives that are both socio-centric and personal.
Research on enculturation holds much promise for enhancing our
understanding of the acquisition of cultural schemata (D' Andrade 1981, 1984,
Schwartz 1981 ). More or less widely shared, such schemata (variously
conceived as frames, maps, models and scripts) are fundamentally involved in
perception, recognition, interpretation, assessment, and other modes of
processing information that enable actors to represent the contexts in which
they find themselves and to guide their action within them. A schema generally
consists of a number of conceptual elements connected to one another within a
semantic network. The conceptual elements of a schema exhibit a range of
values and connective potentials, can be variously interlinked within a focal
schema or among other schemata, and can be variably bound to different
aspects of the environment in different instantiations of the schema.
As D'Andrade (1992:29) suggests, 'a schema is an interpretation which is
requent, we orgamze , memora e, w tc can e ma e rom mmtma cues,
contains one or more prototypic instantiations, is resistant to change, etc.'
Cultural schemata facilitate the construction of hypotheses about the identity
or conceptual properties of objects, events, situations, actions, or persons, and
allow interpretation to proceed beyond the information given in guiding
inferences about unobserved, tacit, or ambiguous aspects of such phenomena.
Yet each conceptual element of a given schema may itself be a complex schema,
for more general, overarching schemata encompass more specific, context-
bound subschemata in a hierarchical arrangement. Such subschemata may be
activated in various ways and at various hierarchical levels, on the basis of
minimal information, as the construction of an interpretation proceeds.
As D' Andrade ( 1981, 1984) further implies, a distinctive characteristic of
most cultural schemata, as marked simplifications of reality, is that they are not
entirely explicit or well specified in ordinary discourse. Indeed, certain aspects
of such schemata remain tacit, ambiguous, or even opaque, yet are often
'transparent' (Hutchins, 1980: 14) in the sense that they are taken for reality
rather than recognized as interpretations of reality. In consequence, any
particular schema may imply other schemata through often tacit criteria of
relevance and inference, criteria that appear so obvious and natural as not to
require explicit notation, thus permitting the formation of multiple and various
linkages. Indeed, the interwoven networks of concepts that constitute schemata
may exhibit strong or weak, complex or simple, dense or diffuse, bounded or
834
unbounded (in Strauss's (1992) sense), and other variable qualities of

connectivity, or even gaps in articulation. That is, cultural schemata may vary
in the kind and degree of their schematicity, although they ordinarily involve
distinct, well-integrated conceptual structures which can be activated rapidly
on the basis of a minimal set of learned cues.
Certain cultural schemata, or congeries of schemata, may have a particular
centrality, being positioned at the highest hierarchical levels of organization or
generality, potentially interconnecting a wide range of other schemata. Such
schemata are implicated and instantiated in a large array of objects, events,
situations, actions, or perceptions, and come to be bound up with the ways such
phenomena are personally construed and evaluated. Indeed, as Strauss
(1992:211-21) suggests, any individual's repertoire of schemata implicates a
mosaic of bounded and unbounded semantic networks, both cultural and
personal, which are interconnected in complex ways through individual
interaction and experience in a socio-cultural milieu. Thus, an appreciation of
how schemata provide personally compelling interpretations for any particular
individual requires an exploration of the life-history of that individual with
respect to his or her self-understandings, marked memories of salient and self-
defining experiences, prototypical instantiations of schemata having personal
significance, and the kinds of schemata that the individual has experienced and
reconstituted for himself or herself in the processes of socialization and
enculturation.
W y IS It t at certam earne cu tura se emata, eyon provi mg
intellectual interpretations that may be put to occasional and superficial-
albeit appropriate-social use, come to be invested with cognitive, emotional,
evaluative or motivational force for the individual, and even to instigate
personal action? Not all cultural schemata are found to be so compelling in this
sense: that a schema should be implicated in the processes of socialization and
enculturation does not in itself tell us whether or why it should carry such
personal significance or psychological force. In other words, it does not account
for the schema's internalization. Concerning the acquisition of cultural
knowledge or belief, Spiro (1982) distinguishes situations where such
knowledge is both familiar and well understood but does not command assent,
situations where it is seen to make defensible claims but nevertheless does not
carry sufficient personal significance to instigate behaviour, situations where it
amounts to a personal belief system serving to guide and shape action, and
situations where it exerts a compelling force over the entire range of a person's
perception, understanding, affectivity, evaluation, and motivation. Only in the
last two situations, in which cultural schemata may be said to have been
personally appropriated, can we claim that they have been internalized.
Following D' Andrade's (1992:30) hypothesis that 'a person's most
important interpretations of what is going on will function as important goals
for that person', it may be suggested that among the most pervasive and
compelling interpretations are likely to be those having to do with personhood,
835
SOCIAL LIFE
selfuood, and individuality, and that these will furnish a significant connection
between cultural and personal semantic networks (D' Andrade 1990, Quinn
1992 ). It is probable, then, that those schemata will be internalized that are
most closely bound up with schemata of personhood, selfuood, and
individuality. They may, too, be associated with identifications of significant
other persons who are involved with the learner in the acquisition, through
interaction, of these latter schemata.
Rethinking socialization and enculturation

As Schwartz (1976:ix) notes, processes of socialization and enculturation
enable one to embody in one's own experiences, to inscribe in one's sense of
identity, and to enact in one's learned behaviour, a part of the culture of the
community in which one lives one's life (Schwartz 1981). Yet, since different
individuals encounter, learn, and internalize different aspects of their socio-
cultural environment, their knowledge of, perspectives on, and investments in
varying aspects of community life will differ. Even in relatively simple, small-
scale, and homogeneous communities, bound by face-to-face relationships, no
individual could conceivably take on the entire body of cultural knowledge.
This raises the considerable problems of the ways in which, and of the degree
to which, cultural knowledge is shared (Roberts 1951, Wallace 1970, Schwartz
1978, Swartz 1991). The acquisition of more or less shared understanding is
eve opmenta y comp ex Ne son 8 . ommona 1t1es an 1 erences m
status-role configurations, socio-cultural experiences, family milieux,
circumstances of socialization and enculturation, and so on, result in what
Schwartz (1978:428-30) calls 'idioverses' as the distributive loci and personal
organizations of culture. An important aspect of what a child must learn to
negotiate, therefore, are the recognized entailments, consequences, and
adjustments of its known and intuited idioverse for becoming an effective
participant in socio-cultural life. It may be suggested that a child's idioverse is
significantly structured, in part, with respect to its senses of personhood,
selfuood, and individuality as organizing schemata of great generality. Beyond
these considerations, and in view of traditionally prevailing images of the
nature of socialization and enculturation, several caveats, some overlapping in
concern, are however in order.
First, the phylogenetic heritage underwriting the process of neurobiological
maturation undoubtedly provides fundamental developmental resources and
constraints encompassing how development can be effected and take shape in
any socio-cultural milieu. This evolutionary legacy presumably also sets certain
parameters on the ways in which cultures and societies can be designed,
realized, rendered adaptively viable, and, thus, shape processes of socialization
and enculturation. From a developmental perspective, however, it is essential to
recognize that these neurobiological constraints afford not only wide yet
ultimately limited variation in the character of the socio-cultural milieu in
836
which socialization and enculturation occur, but also considerable if

constrained plasticity in developmental potential, especially during the earlier
years of the life-cycle (Brauth et al. 1991, Gollin 1981, Lerner 1984, Toulmin
1981).
Second, at the confluence of neurobiological maturation and the shaping of
development through socialization and enculturation, there may well be certain
sensitive or critical periods during which the nexus between the capacities for
gaining significant experience and the incidence of certain structured, marked,
elaborated, valued, repeated, and sanctioned experiential opportunities is
particularly close (Bornstein 1987). What those formative developmental
periods and their conditions are, however, remains problematic and may vary
for different abilities, individuals, and socio-cultural environments, although
childhood does appear to have at least certain broad, stage-setting functions in
shaping some dimensions of the course of development. Yet studies of
socialization and enculturation, which are processes that unfold over a lifetime,
cannot simply assume that early to middle childhood establishes the essential
grounds and configurations of all that emerges thereafter. We must, therefore,
direct serious attention to the various ways in which certain periods of
development may be formative, and to the socio-cultural conditions under
which such periods and their formative potentials arise.
Third, even the youngest children are active, creative participants and also
agents in the processes of socialization and enculturation that embed their
mteractiVe earnmg expenences m commumty 1 e. Furt er exammatwn IS
needed of how children variously shape the course of their own development by
eliciting, instigating, and otherwise affecting the character of the interactions in
which they are engaged. Indeed, more theoretical and ethnographic attention
to the entailments and consequences of various genres of interaction in
socialization and enculturation is essential (Bruner and Bornstein 1989,
Whiting and Edwards 1988). Above all, to grasp the complexities of
socialization and enculturation from a person-centred perspective, we need
detailed case studies of individual children in diverse cultures and societies,
however much such studies may hamper our attempts to reach even local
generalizations (Briggs 1991, Poole 1987).
Fourth, processes of socialization and enculturation are both intentional and
unintentional, explicit and implicit, marked and unmarked. More
ethnographic attention should be paid to the ranges of socio-cultural contexts
that function as 'learning environments' for children, and to which they have
variable kinds and degrees of access, both direct and indirect. Little is to be
gained from supposing that 'learning environments' are either limited to those
situations explicitly designed to be so or, at the other extreme, that they
embrace the whole of social life. Ethnographic maps of greater subtlety are
required to understand how significant 'learning environments' are socio-
culturally constituted, without narrowing them down to the intentional,
explicit, and marked designs inscribed in local ethnopsychology, or expanding
837
SOCIAL LIFE
them to embrace the totality of socio-cultural experience. What constitutes an

effective 'learning environment' must be understood in terms of a complex
interaction of cultural and social factors as these are subjectively and
intersubjectively construed and negotiated.
Fifth, socialization and enculturation have potentially transformative
significance throughout the life-cycle; for humans retain a capacity for change
from early infancy through to old age. Altered circumstances wrought by the
individual's movement through the life-cycle and by historical changes in the
community may necessitate or provoke various kinds and degrees of personal
accommodation. Further understanding of the character and conditions of
developmental stability, mutability, and change in the course of the life-cycle
requires not only a greater appreciation of the enduring effects of childhood
experiences, but also an exploration of the potentially distinctive and
transformative nature of adult socialization, enculturation, and development.
Sixth, socialization and enculturation involve not only various kinds of
learning, but also varied ways of'learning to learn' (in Bateson's (1972b) sense).
Of deutero-learning, Bateson ( 1972b: 170) suggests that 'individuals who have
complex emotional patterns of relationships with other individuals ... will be
led to acquire or reject apperceptive habits by the very complex phenomena of
personal example, tone of voice, hostility, love, etc .... The event stream is
mediated to them through ... cultural media which are structured at every point
by tramlines of apperceptive habit.' Up to now, anthropological studies of
socta tzatwn an encu turatwn ave ma e 1tt e ea way m graspmg ow
children apprehend the cues, signs, or other features that mark and frame what
is important and to be learned, what is a 'learning environment', who are the
persons with and from whom one should learn, and how is effective learning to
proceed. Children do not simply learn what there is to be learned; they also
learn how variously to go about the processes of learning and what they must
attend to and take into account to do so. Significant dimensions of the strategies
and tactics of learning are themselves learned through socialization and
enculturation.
Seventh, learning through socialization and enculturation is geared not only
to understanding the cultural concepts, categories, schemata, frames, and
scripts governing engagements in social interaction, but also to comprehending
the character of the contexts and their focal events, the relevant background
knowledge, and the situational cues in which social interaction is embedded
(Goodwin and Duranti 1992). Understanding any social activity requires often
tacit background knowledge implicated by tacit frames of relevance. Various
senses of context must be learned with some subtlety, nevertheless, if
individuals are to be able to apprehend and utilize schemata as context-
dependent interpretive devices, and thereby to make sense of and deal with the
complexities of the social life surrounding them. A child must learn to
recognize, attend to, construct, and manipulate myriad aspects of the
interactive contexts of those activities in which he or she participates.
838
Eighth, studies of socialization and enculturation must take account of the

interwoven effects of several kinds of change-in the individual's
circumstances, positions, and experiences in the life-cycle, in the social forms
enveloping the individual (as regards social relationships within family,
household, domestic group and beyond), and in the historical character of a
particular culture and society. As the individual moves through the life-cycle,
his or her positions in and, in part, perspectives on the social world are altered
by virtue not only of social maturation and its entailments in terms of
generational position, but also of varying kinds and degrees of historical
transformation of the society or culture itself. Perhaps in microcosm, such
changes are reflected in transformations in the experiences, circumstances,
identities, and social relationships of the developing child, for, as Giddens
(1979:130) suggests, the 'unfolding of childhood is not ... just for the child'. As
a child is born and socially matures, its presence and development take place in
a qualitatively ever-changing network of persons and social relationships
centred on the family, and gradually extending to wider and more complex
domains of community life.
Ninth, the socially interactive transmission of cultural knowledge, however
it may be personally received, interpreted, and reorganized, is not only
transgenerational, but also intra-generational (Tuzin 1990:82-5, Weisner and
Gallimore 1977). In early childhood, the rudiments of a child's self-
constructed or interactively constituted understandings of the socio-cultural
rea m are ne y portraye m preten p ay or ot er antasy Images, w tc may
be solitary or collaborative in their production. Beginning usually in middle
childhood, however, children actively, energetically, and creatively construct
elaborate socio-cultural worlds of their own that are by no means simply
impoverished replicas of adult worlds (Poole 1987, Tuzin 1990). Thus, any
exploration of socialization and enculturation must attend to the two-fold
milieux of childhood-the child-centred and the adult-centred-and to how
each informs the other. The part played in socialization and enculturation by
peers-friends and kin of varying ages and different genders-and by the
'societies' and 'cultures' of children are only beginning to be understood
(Asher and Gottman 1981, Hartup 1983, Rogoff 1981, Youniss 1980).
A theoretical approach to human development

Any study of socialization and enculturation is predicated on some theoretical
perspective on human development, which usually privileges childhood. As
Schwartz ( 1981:4) observes, however, anthropology has tended to ignore
children in culture and society, whereas developmental psychology has tended
to ignore the socio-cultural in children. Whatever theoretical approach is
adopted, it must rest on some set of foundational assumptions about the
unfolding of certain aspects of human nature, the causes or influences
governing or shaping their emergence and their stability or change, and the
839
SOCIAL LIFE
pathways of their enhancement, diminution, stabilization or transformation

(Bruner 1986, Feldman and Bruner 1987, Kagan 1986, Shweder 1991, White
1983 ). The key questions of contemporary developmental theory may be
phrased in the following terms: What characteristics are common to all
children everywhere at particular moments of development, and what
characteristics are peculiar to particular socio-cultural milieux? How does
development unfold, and why does it take the course or courses that it does?
Which developmentally constituted capacities are preserved and elaborated,
which are not, and what conditions determine or shape the stability or
mutability of acquired abilities? These questions are founded upon certain
basic oppositions: between the developmental roles of biology and experience
(Kagan 1981 b, Konner 1981, Le Vine 1969), between continuity and
discontinuity (Em de and Harm on 1984, Kagan 1980, 1983, 1984 ), between
qualitative and quantitative change, between subjective and objective frames of
reference, between constructs of a general and of a specific nature, between
constancy and mutability across domains or contexts, between stages or
transitions and continua, and between the individual and society. It is now
generally recognized, however, that developmental trajectories are best
understood not in terms of hard-and-fast dichotomies, but as the result of a
complex interplay of mutual influences over time.
Theoretical models of socialization and enculturation have as their objective
to explain the significance of the cultural and the social in development-and
espeCia y t ose aspects o eve opment t at ma e posst e, act ttate, promote,
channel, focus, and constrain the orientations, sensitivities, understandings,
competences, and agential powers necessary to participate as an actor in society.
Central to the development of the potential to become an actor and to its
pragmatic realization, however, is the emergence of a sense of identity-a sense
of having a place in the community as a more or less complete, normal human
being capable of apprehending socio-cultural forms and forces, of
communicating and otherwise acting, of authoring, monitoring, and
representing one's actions, and of being a recognized person, a reflexive self,
both distinguished through one's own actions and 'reflected back' in and
through the actions of others (Cooley 1902, Mead 1934 ).
In consequence, it is essential to explore the unfolding of three intertwined
aspects of identity-those of personhood, selfhood, and individuality-
through processes of socialization and enculturation in contexts of
participation as an actor in a community. Although, as I have already remarked,
identity is forever subject to change in at least some respects throughout the
life-cycle, my emphasis here will be primarily on early through middle
childhood (from about two to twelve years)-a presumably formative segment
of the life-cycle with which most anthropological studies of socialization and
enculturation have been primarily concerned. The sense in which any aspect of
early or later childhood is formative in kind, degree, or force is, nevertheless,
theoretically and empirically problematic (Clarke and Clarke 1976, Shweder
840
1991). Various aspects of identity may always be subject to transformation

under certain circumstances. Thus, the rationale for the emphasis on early to
middle childhood in this article is the assumption that cultural understanding
and social competence, and the relevant senses of personal identity, however
they may be subsequently transformed, first begin to be acquired, organized,
made personally significant, and put to social use in the early interactions in
which children engage and in the 'societies' and 'cultures' that they construct
for themselves both seriously and playfully-and always imaginatively.
Although some aspects of what is acquired in childhood socialization and
enculturation affect, mutatis mutandis, the shape of certain dimensions of
subsequent development, these early (re)constructions of socio-cultural
experience provide a privileged base for perceiving how an individual comes
into being in a 'culturally constituted behavioural environment' (Hallowell
1955:57).
PERSONHOOD, SELFHOOD AND INDIVIDUALITY

As Fogelson (1982), La Fontaine (1985), Poole (1991), and Whittaker (1992)
variously maintain, ethnographic explorations of person, self, and individuality
are still at an early and theoretically confused stage of mapping a complex
terrain. Analytically, notions of person and self are often conflated,
confounded, or seen as synonymous (Geertz 1973, 1984), and ideas about the
m IVI ua are common y constgne to a rea m o pecu tar y Western cu tura
and intellectual biases. My present perspective, however, proposes a tentative
set of analytic distinctions-not prematurely formal definitions, but
orientations of emphasis-among personhood, selfhood, and individuality as
different dimensions of identity. These distinctions may help us to show how,
in diverse socio-cultural contexts, various discriminations and linkages may be
made among 'human beings as .. .living entities among many such entities in
the universe, ... [as] centres of being or experience, or ... [as] members of society'
(Harris 1989: 599).
However such aspects of identity may or may not be recognized, constituted,
discriminated, interconnected, emphasized, or valued, in whole or in part, in
any socio-cultural milieu, there is some heuristic utility in separating them out
analytically. Such distinctions may provide a framework for conceptually
mapping any particular local configuration onto a comparative landscape, and
for exploring what contribution, if any, local concepts make to the shape and
texture of personal, cultural, and social senses of identity in any community
(Poole 1991). It must be noted, nevertheless, that our analytic notions of
person, self, and individual will inevitably bear a problematic and at most
approximate relationship to local concepts of identity. Moreover, cultural
schemata of identity in any society will variously draw upon a broad range of
other culturally embedded ideas, both explicit and implicit.
841
SOCIAL LIFE
Personhood
Following Mauss (1938), Fortes (1973), and Harris (1989), the notion of
personhood refers to those culturally constituted and socially conferred
attributes, capacities, and signs that mark a moral career and its jural
entitlements in a particular society. An analytic concern with personhood
focuses on those cultural forms and social forces that together confer on the
individual a public presence, in the sense of a human nature which is socially
encompassed, and place him or her in an array of social positions that establish
the contexts, entitlements and emblems for the enactment and achievement of
particular kinds and degrees of social agency. Indeed, social personhood
endows the culturally recognized individual as a social being with those powers
or capacities upon which agency depends, makes possible and constrains his or
her proper actions, casts him or her as possessed of understanding and
judgement and, thus, of responsibility, and renders him or her accountable as
an actor in a socio-moral order.
Although the capacities of personhood may be anchored to the powers and
limitations of the human body, they consist fundamentally of the cognitive,
emotional, motivational, evaluative and behavioural abilities that are entailed in
becoming an actor in community life. Thus, the person is essentially a social
being with a certain moral status, is a legitimate bearer of rights and
obligations, and is endowed with those characteristics of agency that make
posst e socta actiOn. onverse y, certam cu tura y mar e a norma 1t1es or
deficiencies in capacity or in action, whatever their locally understood
foundations, may lead to the social denial or withdrawal of particular kinds and
degrees of personhood.
Yet a person also has, by cultural implication, a sense of self and of
individuality, and a notion of past and future. He or she can hold values,
perceive goals, experience motivations, recognize resources, acknowledge
constraints, make choices and, thus, adopt plans of action. Moreover, these
plans are attributable to him or her as a social being with the conscious,
reflective capacity to frame culturally appropriate representations of
phenomena and to have purposes, desires, and aversions that require
judgement and demand accountability. To be a person, or a moral agent, is to be
sensitive to certain standards of the socio-moral order of the community and to
suffer a sense of shame (or guilt) when a breach of this order may be attributed
to one's personal judgement and responsibility. Thus, personhood consists
generally in a conceptual adjustment of a culturally constituted sense of human
nature to a socially constituted jural and moral order, with entailments
concerning both selfhood and individuality.
Selfhood
In the tradition of Hallowell ( 1955, 1960), selfhood implicates a set of
842
orientations towards the 'culturally constituted behavioural environment',

which, in turn, significantly enfolds, informs, and shapes the self. The self
refers to the perceiving and experiencing ego as it is known to himself or
herself. Thus, it implies an understanding of the human being as a locus of
experience, encompassing the experience of being a more or less distinctive
'someone' beyond one's identity as a person. It is that conceptualized self
which is the referent of such notions as 'I' (the subject and author of
experiential states and processes and, thus, of thoughts, feelings, evaluations,
motivations, and behaviours), and 'me' (the cognized and recognized object of
my own and others' attention), and which is distinguished from a contrasting
set of phenomena experienced and represented as 'other-than-self.
It appears that the sense of self is bound up, developmentally and otherwise,
with the perceptions and interpretations of other people. Understanding the
emergence of self-awareness therefore requires some consideration of the ways
in which one's actions are interpreted by others. But that the self emerges vis-
a-vis others also implies a sense of separateness and distinctiveness predicated
on some developmental process of individuation, and a sense of those
phenomena that confer individuality and mark personal difference.
Perhaps the constitution of the self is best seen as a process of establishing
problematic linkages between relatively inward-facing, private self-
understandings and outward-facing, public presentations of self, and between
images of past, present, and future selves. An enduring question, therefore,
concerns w et er se oo s ou e regar e as smgu ar or mu ttp e,
transcendent or context-bound. As Ewing (1990:274) observes, it is necessary
to attend to 'how multiple self-representations are organized, contextualized,
and negotiated'. Self-understandings are often experienced, anchored and
remembered in particular frames of reference that refer to personally salient
and culturally shaped social encounters. Aspects of such understandings may
emerge in the course of a process of self-construction that continues
throughout the life-cycle. Ever shifting across contexts, they are always being
reconfigured by experiences of marked personal significance. On the other
hand, cultural and personal representations of selfhood often invoke a more or
less cohesive, stable and enduring sense of self. Notions of the unity, stability,
and boundedness of certain aspects of selfhood must be placed, nevertheless,
against the background of the multiplicity, openness, and fluidity of selfhood as
it emerges in contexts of experience and action throughout the life-cycle.
Individuality
Whereas notions of person and self figure frequently in anthropological maps
of cultural landscapes, the notion of the individual has come to be seen as
problematic. Indeed, the question of how to take account of the individual in a
principled way in relation to the character, dynamics, and reproduction of
socio-cultural phenomena has long been recognized as a profound problem in
843
SOCIAL LIFE
anthropology (Emmet 1960, Evens 1977). Although people in all societies seem
to recognize the individual as an empirical agent, the ways in which any sense of
individuality is culturally inscribed in schemata of person, self, and the
activities of social life are enormously varied. With regard to the 'individualism'
of the West, Geertz notes that:
The Western conception of the person as a bounded, unique, more or less integrated
motivational and cognitive universe, a dynamic center of awareness, emotion,
judgment, and action organized into a distinctive whole and set contrastively both
against other such wholes and against its social and natural background, is, however
incorrigible it may seem to us, a rather peculiar idea within the context of the
world's cultures.
(1984: 126)
Characterized along these lines, Western individualism is now commonly

assumed in anthropology to be the biased source of any putatively analytic
notions of the individual, individuation, or individuality. Beyond the West, a
relational, socio-centric, holistic, or socially embedded understanding of self is
said to prevail (Shweder and Bourne 1991, Shweder and Miller 1991), and it is
supposed that this latter sense of self pervades non-Western cultural, social,
and personal self-understandings. Both logically and empirically, however, this
assumption seems unwarranted, on the following grounds.
First, there is ample evidence that, even in the West, concepts of the
m IVI ua ave un ergone const era e permutatiOns an trans ormatwns, not
only on a broad historical scale but also locally (Lukes 1973, Macfarlane 1979,
Taylor 1989). Much of the scholarly representation of Western individualism
has been drawn from metaphysical or ideological images, variously embedded
in diverse economic, moral, political, and other social philosophies. Such
images bear a highly problematic relationship to notions of individuality
adduced in local cultural contexts, and to the ways in which individuals
experience themselves or are seen by others in their socio-cultural
environments. To suppose that representations of personal difference or some
sense of individuality can exist only in conjunction with a highly elaborated,
metaphysically complex, ideologically prominent, and institutionally
embedded concept of the individual would be unduly restrictive.
Second, images of identity are unlikely to be only singular, altogether
coherent, and pervasively uniform, and, as Geertz (1973:406) notes, 'patterns
counteractive to the primary ones exist as subdominant but nonetheless
important themes in ... any culture'. Within any society, a variety of senses of
person, self, and individuality are manifested in diverse contexts. Whatever
their cultural elaboration and social prominence, culturally recognized aspects
of personal difference are variously marked in a number of non-Western
societies (Briggs 1970, 1991, Fajans 1985, Kirkpatrick 1985, McHugh 1989,
Poole 1987, 1991). Moreover, a longstanding tradition of thought on the
development of the self among Westerners, from a social interactionist
844
perspective, has emphasized the relational character of the 'looking-glass self

(Cooley 1902, Mead 1934 ); and social psychology has begun to consider the
more general applicability of a social-constructionist perspective on identity to
any socio-cultural milieu (Gergen 1985a, b). It is probable that in any society,
the full range of explicit and implicit cultural ideas about identity, as they are
variously interconnected, contextualized, socially embedded, and personally
construed, exhibits a complex web of collectivistic and individualistic themes,
woven together in different ways and for differing purposes.
Third, if notions of the self, however culturally constituted and socially
realized, are also personally constructed, they represent individuality in at least
three senses: they are (re)constructed by individuals; they are assembled from
and embedded in life experiences of personal salience; and they mark, in part,
certain kinds and degrees of personal distinctiveness. Self-understandings
seem always to encompass various senses of personal difference, and
personhood seems inevitably to entail some recognition of certain kinds and
degrees of individuality. Thus, it is in those places and on those occasions
where such facets of individuality are most readily imagined to be at issue that
account must be taken of the dimensions of recognized personal difference.
Finally, to imagine that whole cultures or societies may be classified in terms
of mutually exclusive, monolithic categories, as either individualistic or socio-
centric, is simplistic and misreads the ethnographic and historical record. Such
a view also inhibits cross-cultural comparison, blunts the subtlety of
et nograp tc ana ysts, tstorts et nopsyc o ogtca un erstan mg, an once
again privileges the West as having a uniqueness different in kind from that of
other socio-cultural traditions, thus placing it beyond comparison.
Anthropology needs to reconsider how to take account not only of the
ethnographer's field experience of individuals who cannot readily be
assimilated to a view of the generalized 'other', but also oflocal understandings
of the conditions under which certain recognized personal differences become
socio-culturally significant.
All societies must come to terms with the problem of individuality, of the
presence of the individual in society, or of the differentiation of each person from
each and every other, and must recognize the entailments and consequences of
such personal differences for maintaining and perpetuating a socio-moral order.
Yet the kinds and degrees of cultural emphasis on individuality, its social force, its
expressive possibilities and necessary constraints, and its psychological salience,
will vary both within and between societies and cultures. In different socio-
cultural situations, certain personal differences may be seen, for example, as
valued resources, as behavioural propensities requiring constraint, as natural
inevitabilities, as pathological signs, or as insignificant dispositions. Whether as
figure or ground, individuality is variously articulated within, and qualifies,
broader patterns of personal and socio-cultural identity.
An analytic notion of individuality makes no assumption of analogy with the
various and shifting historical notions of individualism attributed to certain
845
SOCIAL LIFE
genres of socio-cultural discourse in the West (cf. Dumont 1970, 1977, 1986;
see Beteille in this volume, Article 37). Referring to personally construed,
culturally recognized, and socially expressed personal differences, individuality
may involve various kinds and degrees of unity, separateness, exclusiveness,
boundedness, privacy, interiority, autonomy, naturalness, value, power, control,
agency, and other distinctive qualities. Nor does a notion of individuality
necessarily connote absolute qualities of uniqueness or idiosyncrasy; it may also
refer to relatively distinctive qualities that could, in principle, be plotted on an
ethnopsychological or ethnosociological map.
All of the intertwined senses of identity discussed above are essential to
social life because they enable members of a community to identify a particular
embodied being as human, normal, cognizant, sentient, intentional, in
possession of agential capacities, having a certain history and place in a system
of social relations, and bearing a certain responsibility and accountability for
action. Such interwoven senses of identity, as they are personally constituted,
in turn significantly affect an individual's understandings of, and participation
in, the activities of social life. As Le Vine and White (1986:38) note,
There are concepts of the person and the self in all cultures. Self-awareness and a
sense of one's continuity over time are universal in human experience, and all
human adults distinguish between actions of the self as opposed to those of another.
THE EMERGENCE OF IDENTITY IN CHILDHOOD

The developmental emergence, elaboration, and perpetuation or
transformation of schemata of personhood, selfhood, and individuality
through processes of socialization and enculturation among non-Western
children has rarely been examined in anthropology. Indeed, in view of the
proliferation of studies of culturally constituted constructs of person and self
and, to a lesser extent, of individuality in recent anthropological literature,
remarkably little attention has been paid not only to how such constructs are
socially distributed and embedded, are acquired in socio-cultural contexts, and
are personally construed and deployed, but also to local ethnopsychological
portrayals of children or childhood (Poole 1985, 1987, Reisman 1992). How
various aspects of identity are developmentally formed, consolidated,
contextualized, perpetuated, or transformed over the early life-cycle with
respect to socio-cultural forms and forces remains largely unknown terrain,
waiting to be ethnographically mapped. Such understanding, however, is
essential to the development of theories of socialization and enculturation, as
well as to the theories of society and culture upon which they are founded. As I
have already noted, cultural schemata of person, self, and individuality,
particularly to the extent that they are personally acquired and internalized, are
central to how an individual finds his or her orientation, place, significance, and
purpose in the community. They are possibly among the most complex of
846
acquired schemata, of perhaps the greatest centrality and generality, and of the
greatest personal significance and psychological force. Consequently, it might
be expected that the processes of continually both perpetuating and
(re)constructing such senses of identity, and of interconnecting them with
other salient schemata, are channelled through socialization and enculturation
in critically important ways.
Infancy and early childhood

Explorations of the enculturation and socialization of infants from a cross-
cultural perspective remain few and far between (Field et al. 1981, Leiderman
et al. 1977, Super 1981). Most ethnographic attention has been directed toward
adults' values, understandings, expectations, and actions vis-a-vis infants, for
prelinguistic children are not considered to be readily accessible to ordinary
modes of ethnographic inquiry. In developmental psychology, however, it is
now recognized that well before the end of the first year of life, infants exhibit
an incipient capacity to form images and concepts, to express emotions and
perhaps intentions, and to engage in reciprocities of interaction (Mandler
1990). Indeed, not only does a family-centred social world envelop the infant
interactively, but the infant also soon begins to participate in those interactions
in a more or less patterned way (Kaye 1982, Stern 1985). In the course of such
early development, the rudiments of what might be termed a 'proto-selr begin
to emerge.
From infancy onwards, beliefs held by parents and others about the nature
and capacity of the child significantly shape the character of their interactions
with the child (Kaye 1991, Sigel 1985). There is considerable cross-cultural
variation in such beliefs (Le Vine 1974, Poole 1985, Whiting 1974). The cultural
construction of images of the developing child, and their consequences for
socialization and enculturation, are complex (Gergen et al. 1990). In the first
year of life, however, infants begin to respond to and to accommodate the
actions directed towards them and gradually to transform such actions into
increasingly intricate interactions (Poole 1985). They also begin to construct
what will become more and more contoured maps of their 'developmental
niche' (Super and Harkness 1986), which they play an increasing part in
constituting.
After a few months, the mutual accommodation and co-ordination of certain
caretaking activities begins to take on a more interactive character, and
modifications of routines in regard to the everyday patterns of family,
household, and community seem to become integrated into infants'
expectations. At about two and a half months, infant development comes to be
marked by signs of anticipation and aversion, retention of learning between
interactions, increased attention span and focus, more visual acuity and
scanning of the environment, smiling as a means of social elicitation or
response, reduction of diffuse or generalized signals of distress, and, thus, a
847
SOCIAL LIFE
new quality of contact with caretakers. Increased complexity in reaching,

grasping, and later locomotive capacities, allows the infant to explore both its
own body and nearby objects, persons, and situations, as well as to attempt to
repeat and extend motoric skills that are often encompassed and shaped by
early socialization and enculturation. Such advances set the stage for certain
rudimentary sensitivities to, and recognitions of, facets of what will become a
more integrated sense of selfhood.
By seven to nine months, however, attachment behaviours to mothers and
significant other persons begin to be manifested in various forms of ensuring
contact, protesting against separation, and expressing wariness of unfamiliar
persons and strange situations (Ainsworth 1967, Lamb 1982). Although the
character, onset, foci, strength, security, and duration of varying kinds of
attachment seem to differ in some respects, depending on how they are shaped
by socialization and enculturation and by the qualities of relations with
multiple caretakers, they all involve social relationships with significant others,
are invested with personal significance and psychological force, and imply
incipient abilities to categorize and remember persons. Attachment behaviours
are bound up with expanding communicative abilities involving not only bodily
expression and social referencing but also instrumental vocalizations and
understandings of simple expressions.
After the end of the first year, naming and more complex categorization
begin to emerge and to be directed, in part, towards an early mapping of
persons, p aces, t mgs, an acttvlttes m commumty 1 e. spects o t e
environment are manipulated in elementary puzzle-solving, and play facilitates
more fluid forms of experimenting with, and imitating, environmental
knowledge. In various contexts, infants appear to acquire implicit knowledge of
the scripts of routine events, as we can infer from their signs of anticipation or
recognition of such events, and from their reactions of surprise and protest
when ordinary expectations are confounded and habitual routines altered. The
ability to represent everyday scripts explicitly, and outside their ordinary and
immediate contexts, emerges in rudimentary form at the end of the first year
(Bates et al. 1979). Thereafter, there often appears to be a growing ability to
construct increasingly complex pretend representations of everyday persons,
roles, and actions-representations that entail decentering, individuation, and
recognition that the world consists of objects and agents other than the self.
Collaborative play, in turn, requires not only the metacommunicative framing
of action (in Bateson's (1972a) sense) among children, but also some sharing of
scripts of the social activities enacted.
In infancy and early childhood, the role of the family and of a relatively
intimate and stable sphere of significant caretakers, however this may be
bounded, is particularly significant in processes of socialization and
enculturation. Dunn (1988), Hess and Handel (1985), and Reiss (1981)
variously argue that the psycho-social interior of the family serves to mediate
and to interpret for the child the latter's encounters with, and understandings
848
of, socio-cultural phenomena beyond its realm. Indeed, the family often
provides a gateway for contact between a child and the wider community.
Although much attention has been focused on the importance of mother-child
attachments and interactions within the family, there is now greater realization
of the significance of the roles of fathers (Hewlett 1992, Katz and Konner
1981 ), of siblings (Dunn 1988, Mendelson 1990), and of other important
persons in early socialization and enculturation within the domestic-familial
realm of social life.
As infancy merges into early childhood, there is often a marked decline in
expressions of distress at separation, an expansion of exploration, and the
emergence of a more elaborated sense of a distinctive self in a world of other
persons. The child exhibits a growing sensitivity to social standards and a
greater capacity to judge actions, to influence others, and to control or modify
its own behaviour with respect to them, and such perceived standards become
inscribed in the scripts of play activities. Utterances reveal an incipient sense of
agency in self-descriptions, which are also increasingly elaborated in other
respects. The qualities of social interaction become not only more complex and
differentiated, but also more co-ordinated and sensitive to person, activity, and
context. Folk models of child development in many societies indicate a broad
recognition that the child is now on a threshold of new understandings and
competences in its expanding realm, and processes of socialization and
enculturation are often adjusted in consequence of such recognition.
In ear y c 1 oo rom a out two to our years , owever, mgmsttc
competence advances rapidly and, through its developing conversational
abilities, the child is increasingly able to encounter and to understand what is of
significance in the socio-cultural environment and to bring such
understandings to bear on its expanding realms of interaction. Yet linguistic
ability does not merely imply greater and more sophisticated access to cultural
knowledge. As Schieffelin ( 1990) demonstrates, acquiring a language is
profoundly affected by the process of becoming a competent social actor, which
in turn is significantly realized through language (see also DeBernardi's
discussion of this point in the next article). Conversational capacity, requiring
some recognition of distinctions between self and other, categories of other
persons, interactive contexts, and socio-linguistic patterns of interaction, is
established as early as the age of two.
This advance is marked in two especially significant ways in regard to the
child's understanding of personhood, selfhood, and individuality. First, as
Bates (1990) notes, a child's acquisition of the linguistic features of person-
marking and pronominal referencing illuminates some aspects of an emerging
self-concept in the transition from infancy to early childhood. By the age of
three to four years, the child appears to have learned to map certain
understandings of self onto a complex set of lexical and grammatical forms,
and to use these forms conversationally in socially appropriate ways
(Miihlhausler and Harre 1990). Second, as the mastery of speech proceeds, the
849
SOCIAL LIFE
child develops the ability to construct narratives of increasing complexity, and,

as Bruner (1990), Kerby (1991), and Schafer (1992) propose, much that can be
glimpsed of self-understandings is cast in narrative constructions about the
self-stories about self and others both in everyday events and contexts and in
salient life experiences. Such self-narratives include not only autobiographical
portraits, but also a variety of less complex but none the less revealing requests,
refusals, excuses, and other comments, interpretations, and explanations that
implicate the child as subject or object in social encounters and experiences.
Early childhood also marks the appearance of a more elaborated map of
persons in an experientially expanding sense of community. Self-identification
of and with other persons through observation, differentiation, imitation, and
affiliation involves a complex array of processes of perception, categorization,
appraisal, and comparison. Gender comes to figure prominently in
understandings of person and self, and to inflect the character of social
interaction. Of particular importance to the development of perceptions and
understandings of persons other than the self, and of the child's senses of its
own personhood, selfhood, and individuality, however, is its acquisition of a
'theory of mind' (Astington et al. 1988, Frye and Moore 1991, Wellman 1990).
Drawing on ethnopsychological schemata of 'mind' encountered in the
interactive contexts of socialization and enculturation, a child begins to develop
senses of sensitivity (Light 1979) and empathy (Eisenberg and Strayer 1987) in
regard to other persons, and to acquire a subtler sense of the
et nopsyc o ogtca contours o mteractiVe contexts mvo vmg se an ot er.
Understandings and representations of other persons and of the self become at
once more richly connected and distinguished in diverse respects (Shields and
Duveen 1986).
Beginning in the second year, as Kagan (198la) notes, self-awareness and
self-understanding, with implications for how the child conceives its
personhood and individuality, become increasingly marked in self-descriptions
which reflect more elaborated senses of social standards, agency, mastery,
control, evaluation, and so on, coupled to the child's perceptions of its own
thoughts, feelings, expectations, intentions, competence, and actions. Various
phenomena come to possess meanings which are significant in socio-cultural
contexts of morally shaped interaction, and which are thus of relevance to other
persons besides oneself. In the context of an expanding awareness of its ability
to initiate, cease, and reflexively monitor its own actions, the child acquires
greater sensitivity to its own capacity to generate goal-directed activity, and to
the resources and constraints affecting the attainment of goals. As the child's
own awareness of its qualities, capacities, competences, agential powers, and
abilities to affect other persons in contexts of social interaction proceeds, its
experience of its environment and of its domain of community life becomes
significantly organized in terms of its senses of being a person, a self, and an
individual in a socio-cultural world.
850
Middle childhood
It is at the point in early to middle childhood when self-understanding is

consolidated, deployed, and elaborated that folk models of childhood in most
societies also establish the beginnings of social personhood. The bestowal of
personhood, coupled to a recognition of a child's advancing cognitive,
emotional, motivational, and behavioural capacities in various arenas of social
interaction, implies an attribution of judgemental ability appropriate to the
assumption of responsibility and, thus, of social accountability as an incipient
actor. In consequence, as social horizons expand and interactions begin to
pervade newly experienced domains of community life, children come to know
much about the statuses they occupy, the roles they are expected to enact, how
to act accordingly, how to control expressions of socially inappropriate feelings,
and how to respect the rights of other persons. They develop richer
understandings of the socio-cultural perspectives of other persons--especially
in contexts of friendship-and of the implications of such perspectives for the
course of social interaction and in the management of co-operation and
conflict. Their recognition of the social significance of self-control, and of
strategies and tactics for controlling others in interaction, becomes more
elaborate. Their play, now more and more in the contexts of peer interaction, is
infused with understandings of socio-moral rules and conventions, of the
consequences of their breach, and of the cultural values of various genres of eo-
operatiOn an con tct.
On the threshold of middle childhood, a child's notions of its own
personhood, selfhood, and individuality become increasingly infused with
experiential anchorages in interactive contexts and with personal investments
of significant psychological force. Various emotions and motivations come to be
personally bound up with understandings of both the outward-facing and the
inward-facing aspects of these senses of identity. In this context, the role of
shame (and guilt) may be particularly important (Epstein 1992:198-247, Lewis
1992 ), and an understanding of shame seems to be acquired by the time of the
transition from early to middle childhood. Although shame, in its varying
guises, is a cultural construct that operates as a sanction in social control, it is
manifested not only in public arenas of social interaction, but also in a more
intimate way in experiences of one's own personhood, selfhood, and
individuality. The experiential sense of shame, sometimes verging on guilt, is
less visible socially but may perhaps affect more profoundly one's senses of
identity, to the extent that such senses are forged in contexts of social
interaction. Indeed, the effectiveness of shame as a mechanism of social control
depends upon the way in which the cultural schemata of shame are linked, in
the course of their acquisition and internalization, to schemata of person, self,
and individuality, thereby taking on personal significance and psychological
force. However, the ways in which shame and other emotional complexes come
to be shaped by and connected to the schemata of person, self, and
851
SOCIAL LIFE
individuality, through socialization and enculturation, are largely unknown for

any culture and society.
The period of middle childhood is culturally marked at its inception (about
five to seven years) in most societies by the recognition of a dramatic advance in
social competence. As middle childhood begins to unfold, children are seen in
almost every society to possess a different cognitive, emotional, evaluative,
motivational, and agential capacity and, thus, a new order of independence,
action, responsibility, and accountability. This transformation is recognized not
only in local ethnopsychological portrayals of personhood, selfhood, and
individuality at the inception of middle childhood, but also in developmental
psychological characterizations of intellectual (Piaget 19 54), moral (Kohlberg
1981, 1984) and other social advances in understanding. Processes of
socialization and enculturation often involve taking significant account of such
changes in articulating expectations, demands, and sanctions in regard to an
expanding sphere of social opportunities, responsibilities, and interactions.
As middle childhood advances towards the brink of adolescence, senses of
personhood, selfhood, and individuality are increasingly manifested in the
interactions of a more or less distinctive world of peers, which is often
segregated by gender. It is in this context that the distinctive 'societies' and
'cultures' of children take significant shape (Poole 1987, Tuzin 1990), while
relationships with adults-both within the family and beyond-are
transformed from emphasizing the dependence, management, and control of
c 1 ren to a greater stress on t e mutua negotiatiOn o m epen ence an
interdependence. Yet however segregated the child-centred and adult-centred
milieux may be in social space, what is learned and experienced in each is
brought to bear on the other. Earlier fantasy play tends to give way to games
involving more complex understandings of moral rules, social conventions,
reciprocal responsibilities, resource distributions, and co-ordinations of
personal orientations in the regulation of social interaction. Relations of
friendship come to involve increasing emphasis on the sharing of interests, the
construction of mutual understandings, and the creation of senses of mutual
sensitivity, responsibility, and trust. As social horizons widen, socio-cultural
knowledge expands across increasingly diverse contexts, and social
relationships and interaction become more extensive and complicated.
CONCLUSION
Although these brief sketches of aspects of early through middle childhood
only partially portray the myriad complexities of this time of life in any culture
or society, they do indicate some of the largely untapped potential for an
ethnographic exploration of the formation of identity in processes of
socialization and enculturation. An approach to the anthropological study of
socialization and enculturation focused on the character of personal construals
of cultural schemata of personhood, selfhood, and individuality, as they are
852
variously experienced, apprehended, and rendered salient in and through the

social interactions shaping a child's emerging sense of everyday community
life, and of its locus in its socio-cultural milieu, holds considerable promise for
theoretical understanding on several fronts.
First, it adopts a perspective that considers, in an integrated and principled
manner, the developmental interconnections of the cultural, the social, and the
individual over time. This perspective is centred on how the personal
apprehension and appropriation of socio-cultural understandings, acquired in
various ways and with variable psychological significance and force, are bound up
with interactive processes oflearning in the varied contexts of community life.
Second, the approach is not predicated, as are more orthodox accounts, on a
view of culture as a largely seamless and shared web of significance somehow
descending upon and altogether encompassing the individual, who often
appears to vanish in the notion of the cultural constitution of phenomena.
From the present perspective, culture is rather seen to be organized and
experienced in various contextually embedded ways that exhibit
inconsistencies, contradictions, ambiguities, lacunae and loose
interconnections, to be socially distributed and problematically shared, and to
be personally construed or constituted.
Third, the approach is also predicated on a view of society as consisting in
pragmatically organized and contextualized events of patterned interaction
involving the negotiation of mutual understandings about, and co-ordinations
o , sue mteractwns y m IVI ua actors w ose now e ge o w at IS gomg on
and what is at stake is incompletely shared. For the individual actor's
understandings are not automatically congruent with those of others, nor are
the interactions that flow from them necessarily co-ordinated. Cultural models
of the proper patterning of social interaction do not translate
unproblematically, by way of individual understandings of them, into actual
patterns of interaction.
Fourth, the approach recognizes the individual as the proper locus of
culture, as the actor in social interaction, and as the bearer of senses of identity
that inform the apprehension and appropriation of culture and the engagement
in social interaction. In so doing, it denies that an analytic construct of the
individual is merely, or must necessarily be, an artefact of some Western
tradition of individualism. To the contrary, it posits an analytic sense of
individuality by attending not only to the individual as an empirical agent, but
also to how that agent is at once culturally constituted, interactively realized,
and personally constructed in the ebb and flow of the varied situations of
community life. Thus, it attends to how personal differences, constituted in
various ways, do or do not become culturally, socially, and personally significant
in different ways in different communities.
Fifth, the approach makes possible a developmental, person-centred,
context-embedded appreciation of person and self, agency and emotion, as well
as of local ideologies, philosophies, and worldviews, in a manner that explores
853
SOCIAL LIFE
how what is culturally constituted and socially negotiated can become

personally significant and endowed with psychological force. For it is not
enough simply to assert that this or that phenomenon is culturally constituted
or socially negotiated. We need to attend also to the theoretical and
epistemological implications of our notions of 'constitution' and 'negotiation',
and to probe the interactive contexts wherein the social and the cultural are
woven, through socialization and enculturation, into the life-histories of
individuals.
Finally, the approach suggests that certain interlinked schemata of
identity-of personhood, selfhood, and individuality-are centrally involved
in the processes whereby personally apprehended realms of socio-cultural
knowledge are variously acquired, (re)constructed, comprehended, rendered
significant, and sometimes internalized and imbued with psychological force as
they affect the thoughts, feelings, orientations, evaluations, intentions, plans,
and actions of individuals in the course of their continual socialization and
enculturation in the contexts of community life. Anthropological theorizing
about socialization and enculturation still remains in its infancy. As Benthall
( 1992) suggests, however, ethnographic study of the socialization and
enculturation of children may hold the promise of recasting anthropological
perspectives not only on the distinctiveness of children's socio-cultural worlds
and of children's images of the adult-centred socio-cultural realm, but also on
how socio-cultural worlds in general are learned by individuals and constituted
REFERENCES
Ainsworth, M.D.S. (1967) lnfoncy in Uganda, Baltimore: Johns Hopkins University
Press.
Asher, S.R. and Gottman, J.M. (eds) (1981) The Development of Children's Friendships,
Astington, J.W., Harris, P.L. and Olson, D.R. (eds) (1988) Developing Theories ofMind,
Bates, E. (1990) 'Language about me and you: pronominal reference and the emerging
concept of self, in D. Cicchetti and M.Beeghly (eds) The Self in Transition, Chicago:
University of Chicago Press.
Bates, E., Benigni, L., Bretherton, 1., Camaioni, L. and Volterra, V. (1979) The
Emergence of Symbols, New York: Academic Press.
Bateson, G. (1972a) 'A theory of play and fantasy', in G.Bateson Steps to an Ecology of
Mind, New York: Ballantine Books.
--(1972b) 'Social planning and the concept of deutero-learning', in G.Bateson Steps
to an Ecology of Mind, New York: Ballantine Books.
Benthall, J. (1992) 'A late developer? the ethnography of children', Anthropology Today
8(2): 1.
Bornstein, M.H. (1987) 'Sensitive periods in development: interdisciplinary
854
perspectives', in M.H.Bornstein (ed.) Sensitive Periods in Development, Hillsdale,

NJ: Lawrence Erlbaum.
Brauth, S.E., Hall, WS. and Dooling, R.J. (1991) 'Epilogue', in S.E.Brauth, WS.Hall
and R.J.Dooling (eds) Plasticity of Development, Cambridge, Mass.: MIT Press.
Briggs, J.L. (1970) Never in Anger, Cambridge, Mass.: Harvard University Press.
-(1991) 'Mazes of meaning: the exploration of individuality in culture and of culture
through individual constructs', The Psychoanalytic Study ofSociety 16: 111-53.
Bronfenbrenner, U. (1979) The Ecology of Human Development, Cambridge, Mass.:
Harvard University Press.
Bruner, J.S. (1986) 'Value presuppositions of developmental theory', in L.Cirillo and
S.Wapner (eds) Value Presuppositions in Theories of Human Development, Hillsdale,
NJ: Lawrence Erlbaum.
--(1990) 'Autobiography and self, in J.S.Bruner, Acts of meaning, Cambridge,
Bruner, J.S. and Bornstein, M.H. (1989) 'On interaction', in M.H.Bornstein and
J.S.Bruner (eds) Interaction in Human Development, Hillsdale, NJ: Lawrence
Erlbaum.
Clarke, A. M. and Clarke, A.D.B. (1976) Early Experience, New York: Free Press.
Cooley, C.H. (1902) Human Nature and the Social Order, New York: Charles Scribner's
Sons.
D' Andrade, R.G. (1981) 'The cultural part of cognition', Cognitive Science 5:179-95.
--(1984) 'Cultural meaning systems', in R.A.Shweder and R.A.LeVine (eds) Culture
Theory, Cambridge: Cambridge University Press.
--(1990) 'Culture and personality: a false dichotomy', in D.K.Jordan and M.J.
wartz e s ersona tty an t e onstructton o
University of Alabama Press.
--(1992) 'Schemas and motivation', in R.G.D'Andrade and C.Strauss (eds) Human
Motives and Cultural Models, Cambridge: Cambridge University Press.
Dumont, L. (1970) Homo Hierarchicus, Chicago: University of Chicago Press.
--(1977) From Mandeville to Marx, Chicago: University of Chicago Press.
--(1986) Essays on Individualism, Chicago: University of Chicago Press.
Dunn, J. (1988) The Beginnings of Social Understanding, Cambridge, Mass.: Harvard
University Press.
Eisenberg, N. and Strayer, J. (1987) Empathy and Its Development, Cambridge:
Emde, R.N. and Harmon, R.J. (1984) 'Entering a new era in the search for
developmental continuities', in R.N.Emde and R.J.Harmon (eds) Continuities and
Discontinuities in Development, New York: Plenum Press.
Emmet, D. (1960) 'How far can structural studies take account of individuals?', Journal
of the Royal Anthropological Institute 90:191-200.
Epstein, A.L. (1992) In the Midst of Life, Berkeley: University of California Press.
Evens, T.M. (1977) 'The predication of the individual in anthropological
interactionism', American Anthropologist 79:579-97.
Ewing, K.P. (1990) 'The illusion of wholeness: culture and the experience of
inconsistency', Ethos 18:251-78.
Fajans, J. (1985) 'The person in social context: the social character of Raining
"psychology"', in G.M.White and J.Kirkpatrick (eds) Person, Self, and Experience,
855
SOCIAL LIFE
Feldman, C. and Bruner J.S. (1987) 'Varieties of perspective: an overview', in J.Russell

(ed.) Philosophical Perspectives on Developmental Psychology, Oxford: Blackwell.
Field, T.M., Sostek, A.M., Vietze, P. and Leiderman, P.H. (eds) (1981) Culture and
Early Interactions, Hillsdale, NJ: Lawrence Erlbaum.
Fogelson, R. (1982) 'Person, self and identity: some anthropological retrospects,
circumspects and prospects', in B.Lee (ed.) Psychosocial Theories of the Self, New
York: Plenum Press.
Fortes, M. (1973) 'On the concept of the person among the Tallensi', in G.Dieterlen
(ed.) La Notion de personne en afrique noire, Paris: Centre National de la Recherche
Scientifique.
Frye, D. and Moore, C. (eds) (1991) Children's Theories of Mind, Hillsdale, NJ:
Lawrence Erlbaum.
Geertz, C. (1973) 'Person, time, and conduct in Bali', in The Interpretation of Cultures,
New York: Basic Books.
--(1984) "'From the native's point of view": on the nature of anthropological
understanding', in R.A.Shweder and R.A.LeVine (eds) Culture Theory, Cambridge:
Gergen, K.J. (1985a) 'Social constructionist inquiry: context and implications', in
K.J.Gergen and K.E.Davis (eds) The Social Construction of the Person, New York:
Springer.
--(1985b) 'Theory of the self: impasse and evolution', in L.Berkowitz (ed.) Advances
in Experimental Social Psychology, vol. 17, New York: Academic Press.
Gergen, K.J., Gloger-Tippelt, G. and Berkowitz, P. ( 1990) 'The cultural construction
of the developing child', in G.R.Semin and K.J.Gergen (eds) Everyday
n erstan mg, on on: age.
Giddens, A. (1979) Central Problems in Social Theory, Berkeley: University of
California Press.
Gollin, E.S. (1981) 'Development and plasticity', in E.S.Gollin (ed.) Developmental
Plasticity, New York: Academic Press.
Goodwin, C. and Duranti, A. (1992) 'Rethinking context: an introduction', in A.
Duranti and C.Goodwin (eds) Rethinking Context, Cambridge: Cambridge
University Press.
Hallowell, A.I. (1955) 'The self and its behavioral environment', in A.I.Hallowell
Culture and Experience, Philadelphia: University of Pennsylvania Press.
--(1960) 'Self, society and culture in phylogenetic perspective', in S.Tax (ed.)
Evolution After Darwin, vol. 2, Chicago: University of Chicago Press.
Harris, G.G. (1989) 'Concepts of individual, self, and person in description and
analysis', American Anthropologist 91:599-612.
Hartup, W.W. (1983) 'Peer relations', in E.M.Hetherington (ed.) Handbook of Child
Psychology, vol. 4, New York: John Wiley.
Herskovits, M.J. (1948) Man and His Works, New York: Alfred A.Knopf.
Hess, R.D. and Handel, G. (1983) 'The family as a psychosocial organization', in G.
Handel (ed.) The Psychosocial Interior of the Family, 3rd edn, New York: Aldine de
Gruyter.
Hewlett, B.S. (ed.) (1992) Father-Child Relations, New York: Aldine de Gruyter.
Hutchins, E. (1980) Culture and Inference, Cambridge, Mass.: Harvard University
Press.
856
Kagan, J. (1980) 'Perspectives on continuity', in O.G.Brim Jr and J.Kagan (eds) Constancy

and Change in Human Development, Cambridge: Harvard University Press.
--(198la) The Second Year, Cambridge, Mass.: Harvard University Press.
--(1981b) 'Universals in human development', in R.H.Munroe, R.L.Munroe and
B.B.Whiting (eds) Handbook of Cross-Cultural Human Development, New York:
Garland STPM Press.
--(1983) 'Developmental categories and the premise of connectivity', in R.M.
Lerner (ed.) Developmental Psychology, Hillsdale, NJ: Lawrence Erlbaum.
--(1984) 'Continuity and change in the opening years of life', in R.N.Emde and
R.J.Harmon (eds) Continuities and Discontinuities in Development, New York: Plenum
Press.
--(1986) 'Presuppositions in developmental inquiry', in L.Cirillo and S.Wapner
(eds) Value Presuppositions in Theories of Human Development, Hillsdale, NJ:
Lawrence Erlbaum.
Katz, M.M. and Konner, M.J. (1981) 'The role of the father: an anthropological
perspective', in M.E.Lamb (ed.) The Role of the Father in Child Development, 2nd
edn, New York: John Wiley.
Kaye, K. (1982) The Mental and Social Lift of Babies, Chicago: University of Chicago
Press.
--(1991) 'The parental frame', in M.Woodhead, R.Carr and P.Light (eds) Becoming
a Person, London: Routledge.
Kerby, P. (1991) Narrative and the Self, Bloomington: Indiana University Press.
Kirkpatrick, J. (1985) 'How personal differences can make a difference', in K.J. Gergen
and K.E.Davis (eds) The Social Construction of the Person, New York: Springer.
o erg, ssays on ora eve opment, vo . , ew or : arper & ow.
--(1984) Essays on Moral Development, vol. 2, New York: Harper & Row.
Konner, M.J. (1981) 'Evolution of human behavior development', in R.H.Munroe,
R.L.Munroe and B.B.Whiting (eds) Handbook of Cross-Cultural Human
Development, New York: Garland STPM Press.
La Fontaine, J.S. (1985) 'Person and individual: some anthropological reflections', in
M.Carrithers, S.Collins and S.Lukes (eds) The Category of the Person, Cambridge:
Lamb, M.E. (1982) 'Parent-infant interaction, attachment, and socio-emotional
development', in R.N.Emde and R.J.Harmon (eds) The Development of Attachment
and Affiliative Systems, New York: Plenum Press.
Lee, B. and Hickman, M. (1983) 'Language, thought and self in Vygotsky's
developmental theory', in B.Lee and G.G.Noam (eds) Developmental Approaches to
the Self, New York: Plenum Press.
Lee, B., Wertsch, J.V. and Stone, A. (1983) 'Towards a Vygotskian theory of the self, in
B.Lee and G. G.Noam (eds) Developmental Approaches to the Self, New York: Plenum
Press.
Leiderman, P.H., Tulkin, S.R. and Rosenfeld, A. (eds) (1977) Culture and lnfoncy, New
York: Academic Press.
Lerner, R.M. (1984) On the Nature of Human Plasticity, Cambridge: Cambridge
University Press.
LeVine, R.A. (1969) 'Culture, personality and socialization: an evolutionary view', in
D. Goslin (ed.) Handbook of Socialization Theory and Research, Chicago: Rand
McNally.
857
SOCIAL LIFE
--(1974) 'Parental goals: a cross-cultural view', Teachers College Record 76:52-65.

--(1977) 'Child rearing as cultural adaptation', in P.H.Leiderman, S.E.Tulkin and
R.Rosenfeld (eds) Culture and Infoncy, New York: Academic Press.
--(1982) Culture, Behavior, and Personality, 2nd edn, New York: Aldine.
LeVine, R.A. and White, M. (1986) Human Conditions, London: Routledge & Kegan
Paul.
Lewis, M. (1992) Shame, New York: Free Press.
Light, P. (1979) The Development of Social Sensitivity, Cambridge: Cambridge
University Press.
Lukes, S. (1973) Individualism, Oxford: Blackwell.
Macfarlane, A. (1979) The Origins of English Individualism, Cambridge: Cambridge
University Press.
Mandler, J.M. (1990) 'A new perspective on cognitive development in infancy',
American Scientist 79:236-43.
Mauss, M. (1938) 'Une categorie de l'esprit humain: la notion de personne, celle de
"moi'", Journal of the Royal Anthropological Institute 68:263-282.
McHugh, E.L. (1989) 'Concepts of the person among the Gurungs of Nepal',
American Ethnologist 16:75-86.
Mead, G. H. (1934) Mind, Self and Society, Chicago: University of Chicago Press.
Mead, M. (1963) 'Socialization and enculturation', Current Anthropology 4:184--8.
Mendelson, M.J. (1990) Becoming a Brother, Cambridge: MIT Press.
Miihlhiiusler, P. and Harre, R. (1990) Pronouns and People, Oxford: Blackwell.
Nelson, K. (1985) Making Sense, Orlando: Academic Press.
Piaget, J. (1954) The Construction of Reality in the Child, New York: Basic Books.
oo e, . . . ommg mto soCia emg: cu tura Images o m ants m Imm-
Kuskusmin folk psychology', in G.M.White and J.Kirkpatrick (eds) Person, Self,
and Experience, Berkeley: University of California Press.
--(1987) 'Personal experience and cultural representation in children's "personal
symbols" among Bimin-Kuskusmin', Ethos 15:104--35.
--(1991) 'Cultural schemas and experiences of the self among the Bimin-
Kuskusmin ofPapua New Guinea', Psychoanalytic Study of Society 16:55-85.
Quinn, N. (1992) 'The motivational force of self-understanding: evidence from wives'
inner conflicts', in R.D'Andrade and C.Strauss (eds) Human motives and cultural
models, Cambridge: Cambridge University Press.
Reisman, P. (1992) First Find your Child a Good Mother, New Brunswick, NJ: Rutgers
University Press.
Reiss, D. (1981) The Family's Construction of Reality, Cambridge, Mass.: Harvard
University Press.
Roberts, J.M. (1951) Three Navaho Households: a Comparative Study of Small Group
Culture, Peabody Museum of America Archaeology and Anthropology Papers vol.
40(3).
Rogoff, B. (1981) 'Adults and peers as agents of socialization: a highland Guatemala
profile', Ethos 9:18-36.
--(1990) Apprenticeship in Thinking, New York: Oxford University Press.
Sapir, E. (1949) Selected writings of Edward Sapir in Language, Culture and Personality,
ed. D.G.Mandelbaum, Berkeley: University of California Press.
Schafer, R. (1992) 'Narratives of the self, in R.Schafer, Retelling a Life, New York:
Basic Books.
858
Schieffelin, B.B. (1990) The Give and Take of Everyday Life, Cambridge: Cambridge
University Press.
Schwartz, T. (1976) 'Introduction', in T.Schwartz (ed.) Socialization as Cultural
Communication, Berkeley: University of California Press.
--(1978) 'Where is culture? personality as the distributive locus of culture', in
G.D.Spindler (ed.) The Making of Psychological Anthropology, Berkeley: University
of California Press.
--(1981) 'The acquisition of culture', Ethos 9:4--17.
Shields, M.M. and Duveen, G. M. (1986) 'The young child's image of the person and
the social world: some aspects of the child's representation of persons', in J. Cook-
Gumperz, W.A.Corsaro and J.Streeck (eds) Children's Worlds and Children's
Language, New York: Mouton de Gruyter.
Shweder, R.A. (1991) 'Rethinking culture and personality theory', in R.A.Shweder,
Thinking Through Cultures, Cambridge, Mass.: Harvard University Press.
Shweder, R.A. and Bourne, E.J. (1991) 'Does the concept of the person vary cross-
culturally?', in R.A.Shweder, Thinking Through Cultures, Cambridge, Mass.:
Shweder, R.A. and Miller, J.G. (1991) 'The social construction of the person: how is it
possible?', in R.A.Shweder, Thinking Through Cultures, Cambridge, Mass.: Harvard
University Press.
Sigel, I.E. (1985) 'Introduction', in I.E.Sigel, Parental Belief Systems, Hillsdale, NJ:
Lawrence Erlbaum.
Spiro, M.E. (1982) 'Collective representations and mental representations in religious
symbol systems', inJ.Maquet (ed.) On Symbols in Anthropology, Malibu: Udena.
tern, e nterpersona or o t e n ant, ew or : as1c oo s.
Strauss, C. (1992) 'What makes Tony run? Schemas as motives reconsidered', in R.
D' Andrade and C.Strauss (eds) Human Motives and Cultural Models, Cambridge:
Super, C.M. (1981) 'Behavioral development in infancy', in R.H.Munroe, R.L.
Munroe and B. B. Whiting (eds) Handbook of Cross-Cultural Human Development,
New York: Garland STPM Press.
Super, C.M. and Harkness, S. (1986) 'The developmental niche: a conceptualization at
the interface of the child and culture', International Journal of Behavioral
Development 9:545-69.
Swartz, M.J. (1991) The Way the World Is, Berkeley: University of California Press.
Taylor, C. (1989) Sources of the Self, Cambridge, Mass.: Harvard University Press.
Toulmin, S. (1981) 'Epistemology and developmental psychology', in E.S.Gollin (ed.)
Developmental Plasticity, New York: Academic Press.
Tuzin, D.F. (1990) 'Of the resemblance of fathers to their children: the roots of
primitivism in middle-childhood enculturation', The Psychoanalytic Study of Society
15:69-103.
Vygotsky, L.S. (1978) Mind in Society, Cambridge, Mass.: Harvard University Press.
Wallace, A.F.C. (1970) Culture and Personality, 2nd edn, New York: Random House.
Weisner, T.S. and Gallimore, R. (1977) 'My brother's keeper: child and sibling
caretaking', Current Anthropology 18:169-91.
Wellman, H.M. (1990) The Child's Theory of Mind, Cambridge, Mass.: MIT Press.
Wentworth, W.M. (1980) Context and Understanding, New York: Elsevier North
Holland.
859
SOCIAL LIFE
White, S.H. (1983) 'The idea of development in developmental psychology', in R.M.

Lerner (ed.) Developmental Psychology, Hillsdale, NJ: Lawrence Erlbaum.
Whiting, B.B. (1974) 'Folk wisdom and child rearing', Merrill-Palmer Quarterly of
Behavior and Development 20:9-19.
Whiting, B.B. and Whiting, J.WM. (1975) Children of Six Cultures, Cambridge, Mass.:
Whittaker, E. ( 1992) 'The birth of the anthropological self and its career', Ethos 20:
191-219.
Youniss, J. (1980) Parents and Peers in Social Development, Chicago: University of
Chicago Press.
FURTHER READING
Ainsworth, M.D.S. (1967) Infancy in Uganda, Baltimore, Md.: Johns Hopkins
University Press.
Briggs, J.L. (1970) Never in Anger, Cambridge: Harvard University Press.
Chisholm, J.S. (1983) Navajo Infoncy, New York: Aldine.
Dunn, J. (1988) The Beginnings of Social Understanding, Cambridge, Mass.: Harvard
University Press.
Erchak, G.M. (1992) The Anthropology of Self and Behavior, New Brunswick, NJ:
Rutgers University Press.
Field T.M. Sostek A.M. Vietze P. and Leiderman P.H. eds
Early Interactions, Hillsdale, NJ: Lawrence Erlbaum.
Hewlett, B.S. (1991) Intimate Fathers, Ann Arbor: University of Michigan Press.
Kelly-Byrne, D. (1989) A Child's Play Lift, New York: Teachers College Press.
Leiderman, P.H., Tulkin, S.R. and Rosenfeld, A. (eds) (1977) Culture and Infoncy, New
York: Academic Press.
LeVine, R.A. (1982) Culture, Behavior and Personality, New York: Aldine Publishing
Company.
Munroe, R.H., Munroe, R.L. and Whiting, B.B. (eds) (1981) Handbook of Cross-
Cultural Human Development, New York: Garland STPM Press.
Reisman, P. (1992) First Find Your Child a Good Mother, New Brunswick, NJ: Rutgers
University Press.
Rogoff, B. (1990) Apprenticeship in Thinking, New York: Oxford University Press.
860
31
SOCIAL ASPECTS OF LANGUAGE

USE
Jean DeBernardi
Language use is fundamental to the creation and expression of social identity

and difference, and the translation of cultures has always depended on
understanding the complexities of language use in other social worlds. Such
understanding is crucial even in the work of anthropologists who would not
describe themselves as 'linguists': the analysis of kinship systems, for example,
epen s on a sop tsttcate un erstan mg o t e way t at terms o re erence
and address both classify social relationships and pattern social interaction.
Even Radcliffe-Brown's work (1965 [1952]) on joking relationships, though
defined as a study in social structure, is in fact a concise statement of the social
meaning of certain norms of linguistic interaction.
Contemporary ethnographic linguists are driven by functional questions
regarding the role of linguistic interaction in expressing social identity and
shaping value. Research into the pragmatics of language use suggests that
people not only speak about the world 'out there'; they also create a good deal
of their social reality in the very act of speaking (Silverstein 1979: 194). Thus
the acquisition of a language is not only the internalization of a linguistic code,
but also entails the learning of status and role, of appropriate social affect, and
(ultimately) of a worldview. Language provides both the foundation of a shared
cultural identity and the means for the reproduction of social difference.
Early cross-cultural research on language use emphasized the shared
dimensions of language, attending to the role of language in constraining
thought, revealing worldview, and determining social action, and these
theoretical premisses have had a deep influence on the work of both linguists
and anthropologists. Malinowski (himself highly sensitive to the 'verbal
contour of native thought' ( 1922:23)) was among the first to shift theoretical
attention to questions about social aspects of language use (1978 [1935]) in
work that influenced the development of the field of functional linguistics (see
861
SOCIAL LIFE
for example J.R.Firth ( 19 57) and Halliday ( 1978) ). The emergence of the
approach known as the 'ethnography of speaking' and its application to the
cross-cultural study oflanguage use was an important step in the establishment
of an interdisciplinary sociolinguistics, and gave impetus to the growth of both
ethnopoetics and 'dialogic anthropology'. Interdisciplinary interest in the
relationship between language, ideology, and power has also had a significant
impact in the field, and anthropologists have made important contributions to
research on the politics of language use.
While much recent work places theoretical emphasis on performance and
choice rather than system and code (Luong 1990), in practice the cross-cultural
analysis of language in society must explore the interaction between the
particular and the general. Thus Abu-Lughod (1986) interprets a Bedouin
poetic performance in the light both of the personal history of the singer and of
a shared cultural code governing the social expression of affect; Scotton (1988)
explains an instance of code-switching in Kenya in terms both of the strategic
aims of the speaker and of the shared social meaning of two linguistic varieties.
With this dialectic in mind, between individual circumstance and common
code, let me begin with a discussion of the social nature of the linguistic sign.
THE LINGUISTIC SIGN

Languages in their variability and diversity are profoundly social. Sapir
emp astze t IS act w en e e ne anguage as a pure y uman an non-
instinctive method of communicating ideas, emotions, and desires by means of
a system of voluntarily produced symbols' (1949a [1921]:8), and as 'the most
massive and inclusive art we know, a mountainous and anonymous work of
unconscious generations' ( 1949a:220). Since, ultimately, each language is both
an abstraction from and a classification of experience, each gives
'predetermined form' to the symbolic expression of its speakers.
Consider, then, the nature of the linguistic sign. De Saussure made the
arbitrariness of the association between signifier and signified fundamental to
his theory of language (1959 [1915]:67-9). On the foundation of the arbitrary
association between sounds and meanings, a diverse range of linguistic codes
may be constructed. Within those codes, by contrast, arbitrariness is relative
(de Saussure 19 59: 131-4 ), limited for example by the use of metaphor to
construct chains of association. To illustrate the relativity of arbitrariness,
examine the word 'tree,' which clearly has no necessary relationship with the
branched plants that we so label. But 'tree-diagrams', 'shoe trees', and 'family
trees' all bear an iconic relationship to a prototypical tree, and are non-
arbitrarily named (Benveniste 1971, Friedrich 1979).
Many scholars would assent to the proposition that 'all language is almost
totally tropological' (Friedrich 1991:24). Nietzsche noted that 'every concept
originates through our equating what is unequal', and ironically characterized
'truth' as a 'mobile army of metaphors, metonyms, anthropomorphisms'
862
SOCIAL ASPECTS OF LANGUAGE USE
(1954:46-7). In a development of this insight, the linguists Lakoff and Johnson

assert that 'our ordinary conceptual system, in terms of which we both think
and act, is fundamentally metaphorical in nature' (1980:3). In their study of
American English they explore conceptual metaphors such as 'time is money'
by detailing the expressions that build on it: 'You're wasting my time', 'You're
running out of time', 'You must budget your time' (Lakoff and Johnson
1980:7).
Cross-cultural comparison foregrounds the intimate relationship between
society, language, and concept. Time is money for us because of the way that
work is accomplished and rewarded (Thompson 1967). In contrast, Nuer
pastoralists of the Southern Sudan are said to live by a 'cattle clock', with
activities co-ordinated by events rather than by an abstract system (Evans-
Pritchard 1940: 103). The Nuer have no concept of time comparable to our own,
and Evans-Pritchard (whom Ardener (197l:lix) credits with securing a place
for language in British social anthropology) described the difference in these
terms:
The Nuer have no expression equivalent to 'time' in our language, and they cannot,
therefore, as we can, speak of time as though it were something actual, which passes,
can be wasted, can be saved, and so forth. I do not think that they ever experience
the same feeling of fighting against time or of having to co-ordinate activities with
an abstract passage of time, because their points of reference are mainly the
. .
logical order, but they are not controlled by an abstract system.
(1940:103)
Indeed, as Evans-Pritchard wistfully concluded, 'Nuer are fortunate'. (See,

however, Adam in this volume, Article 18, for a critical commentary on Evans-
Pritchard' s analysis of N uer time.)
Metaphoric and conceptual elaboration are thus keys to areas of socio-
cultural importance (and difference). While the Nuer may not elaborate
temporal metaphors, they do have a highly developed vocabulary with which to
discuss another key concern-their cattle. Cattle are not only a source of food
but also a medium of exchange in bridewealth payments and bloodfeud
settlements. The vocabulary that describes and names them is rich and
detailed, and songs that praise them also praise their owners-young men who
take their names from their cattle. Finally, the tribe itself is metaphorically
construed as a cattle camp (Evans-Pritchard 1940:16-50).
LANGUAGE AND WORLD VIEW

The first generation of American ethnographic linguists asserted strong links
between language and worldview. They suggested not only that language
channels perception, but also that it contains the 'genius' of the people who use
863
SOCIAL LIFE
it as their means of verbal expression. The view that language was essential to
the continuation of the unique identity and destiny of a group was fundamental
to the German Romanticism of the late eighteenth and early nineteenth
centuries (Edwards 1985:23-7). Divorced from its evolutionist and nationalist
matrix, this perspective influenced the formation of Boasian anthropology,
with its emphasis on the mastery of American Indian languages (ironically, in a
period of widespread language extinction). In particular, the argument that a
language shapes its speakers more than its speakers shape language (that
'language speaks man' in Heidegger's felicitous expression) is one that recurs
repeatedly in studies of language and worldview. The emphasis given to
language by Boas and his students led to the establishment of 'linguistic
anthropology' in North American universities as one of the four basic subfields
of the discipline, together with cultural anthropology, archaeology, and
physical anthropology.
Sapir (1949a [1921], 1949b) and Whorf (1964) were among those who
developed a linguistic relativism based on the premiss that language shaped
worldview, while rejecting the assumption that the languages used by members
of technologically less advanced and non-literate societies were inferior vehicles
for conception. Sapir, for example, argued that:
Both simple and complex types of language of an indefinite number of varieties may
be found spoken at any desired level of cultural advance. When it comes to linguistic
. . . . .
hunting savage of Assam.
(1949a [1921]:219)
In a relativistic inversion of earlier evolutionist arguments, Boas (1966), Sapir

( 1949a [ 1921], 1949b ), Rei chard ( 19 51), and Whorf ( 1964) promoted an
appreciation of the formal elegance of non-Western languages as vehicles for
thought, and Whorf reversed evolutionist schemes when he praised Hopi
representations of time as truer analyses of temporal experience than the
objectifications of 'Standard Average European' (1964:151-5; see also Lucy
1985).
Basic to studies of language and worldview is the attention they give to the
detail of linguistic structure, particularly in so far as that structure provides a
classification of experience. As Sapir observed, 'the world of our experiences
must be enormously simplified and generalized before it is possible to make a
symbolic inventory of all our experiences of things and relations and this
inventory is imperative before we can convey ideas' (1949a [1921]:12). A
community of speakers must agree tacitly to a classification of experience if
they are to communicate, and this classification forms a foundation for their
worldview.
Not surprisingly, anthropologists often analyse society and culture through
the prism oflanguage. Much work in the field of symbolic anthropology in fact
864
entails analysis of the linguistic metaphors that inform classification, ritual

practice, and concepts of the person. Metaphor and metonymy, for example,
figured as key tropes in Levi-Strauss's (1966) characterization of primitive
thought (la pensie sauvage), and interpretive anthropologists explored 'key' or
'root' metaphors (also described as 'key cultural ideas' or 'key symbols') in
order to gain insight into cultural values (Turner 1974, Ortner 1973). The
translation of culture cannot proceed without exploration of the associational
base that natural languages provide to their speakers.
To take a single example, Mary Black (1984) analysed Hopi corn metaphors
using a model drawn from the work of Lakoff and Johnson (1980). She
explored a small number of 'conceptual metaphors' summarized as 'people are
corn' and 'maidens are corn.' The implications of these analogies are part of
ordinary language, in which the same terms apply to both people and corn, and
the analogy receives further elaboration in ritual speeches and song. This
example richly illustrates the relation between language and poetry: the
associations implicit in the Hopi language linking the life-cycle of humans and
the life-cycle of corn are developed in poetic form in the public performances
of ritual. Language gives shape to the individual imagination, and poetic
performance realizes the implicit in an aesthetic form.
The classificatory implications of linguistic structure continue to be a source
of insight into the relationship between language and culture. Witherspoon, for
example, in his studies of Navaho 'language in culture and culture in language',
giVes care u const eratwn to t e trans atwn o Nava o wor s sue as oz o
('beauty', but also 'goodness'), with their penumbra of associations, as he seeks to
elucidate essential cultural values (1977:23-46). Moreover, he delves into the
classificatory implications of syntactic structure, and associates these with aspects
of worldview. For example, he explores the use of 'subject-object' inversion in
Navaho, and demonstrates that restrictions on the use of this syntactic form are
enmeshed with assumptions about who may act upon whom in a society in which
beings are implicitly ranked in terms of their ability to shape outcomes by their
will (Witherspoon 1980:8). Cultural ideas constrain linguistic form, and
linguistic form attests to the depth and tenacity of cultural ideas.
FROM STRUCTURE TO PROCESS

In his Course in General Linguistics, de Saussure emphasized the code-like
properties of language, and provided the founding statement for a semiotics of
society. He suggested that language was but one of many human sign systems,
and proposed a new science of semiology to study the 'life of signs within
society' (1959 [1915]:16). His proposal for the synchronic study oflanguage as
langue, organized as a system of differences, inspired the development of
structuralist anthropology. Building on this insight, and inspired by Jakobson's
breakthroughs in the understanding of phonological codes, Levi-Strauss
inferred semiotic codes (or structures) from the data of kinship systems and
865
SOCIAL LIFE
mythologies ( 1963, 1966, 1969a, b; see also Leach 1970). He claimed that
through these cross-cultural studies of structure, he could elucidate aspects of
the organization of the human mind. Ultimately, the theoretical base of his
model was cognitive, and it gave linguistic theory a new prominence in
anthropological analysis. However, with its focus on the analysis of abstract
structures, the theory did not address questions dealing with social action and
language use in context.
By contrast, recent studies of language in society emphasize social process
rather than structure, and performance rather than code. The development of
the interdisciplinary field of sociolinguistics was an important step in the move
from structure to process, and major theoretical and methodological
contributions include the works of Fishman (1986), Gumperz (1982),
Gumperz and Hymes (1986 [1972]), Hymes (1974), and Labov (1972a, b). The
'ethnography of speaking' (see below) laid stress on the cross-cultural study of
language use, and has contributed an important comparative dimension to the
formation of a socially constituted linguistics. The turn towards the study of
language use in social process has been given further impetus by an
interdisciplinary interest in 'discourse analysis', which seeks to interpret the
diversity of discursive practices in the light of unequal socio-economic
conditions.
The study of linguistic performance in relation to social process has important

roots in the 'ethnography of speaking', an approach developed by American
linguistic anthropologists and folklorists. The approach was given early
theoretical definition by Hymes ( 1974) and Gumperz and Hymes ( 1986
[1972]), and has been developed by Basso (1979, 1984, 1990), Bauman (1977,
1983), Bauman and Sherzer (1974), and others. Articulated here is a forceful
challenge to linguistic models that emphasize formal structure to the exclusion
of practice. H ymes ( 197 4), for example, opposed the Chomskyan emphasis on
'competence' (often equated with de Saussure's concept of langue) and the
concomitant exclusion of 'performance' (corresponding to de Saussure's
parole), arguing that a fundamental aspect of linguistic competence is
communicative competence, i.e. the ability to produce utterances that are
appropriate to the occasion. Linguistic skills entail more than mastery of a
linguistic code that allows the speaker to produce grammatical sentences; they
also involve knowing how to speak appropriately in different social settings.
Important theoretical sources for the ethnography of speaking were the
works of the Prague School, including (most importantly) Jakobson (see Caton
1987) and Mukarovsky (1977). The Prague School theorists developed a model
for understanding the structure of the communicative act that has been
foundational for study of the social use of language. A key statement of this
theoretical programme is Roman Jakobson's ( 1960) 'Closing statement:
866
linguistics and poetics', in which he outlines a model of the major components

of a communicative situation and their associated functions. In this model, the
linguistic code is but one of six components or 'aspects' into which the speech
event may be analysed. Other aspects include the addresser, the addressee, the
channel, the message, and the context. The functions associated with these
aspects of the speech event include the expressive (focus on the speaker), the
conative (focus on the audience), the phatic (focus on establishing social contact
rather than communicating ideas), the poetic (elaboration of the message form
using linguistic parallelism), and the referential (focus on the context)
(1960:353-7). Any one or more of these functions might be emphasized in a
given speech event.
This model provides the starting point for more recent discussion and
analysis of situated language use. With the aim of analysing 'linguistic
competence', Hymes (1974:54-62) developed a model rooted in Jakobson's
work on the speech event, which he summed up with the mnemonic
'SPEAKING' (setting and scene; participants; ends; acts; key;
instrumentalities; norms of interaction and interpretation; genres). These
terms are used in the description of speech events, and Hymes sets himself the
analytic goal of creating a taxonomy to classify the range of cross-cultural
variation in language use in context (1974:33-5).
Basso's studies of Western Apache language and culture illustrate Hymes's
claims regarding cross-cultural variability in language use. Basso observes that
Western pac e norms o mteractwn are un amenta y 1 erent rom ng o-
American' norms. For example, silence is enjoined in a number of social
situations: when strangers meet, when two young people court, when children
come home (from college, say), and when someone is drunk or in deep mourning
(and thus likely to be emotionally volatile). He concludes that silence is a response
'to uncertainty and unpredictability in social relations' (1972:83). Unfortunately,
the Apache response easily leads to cross-cultural misunderstanding in situations
such as job interviews, where reticence might be interpreted as sullen
defensiveness. For their part, the Western Apache caricature Anglo-American
assertiveness in joking performances that are themselves a mocking inventory of
contrasting communicative norms (Basso 1979).
A development of this insight, and a convergence of interest with folklore
studies, appears in the approach to performance of Bauman and others
(Bauman 1977, Bauman and Sherzer 1974 ). The approach entails a critique of
genre studies of the kind which elicit folklore items in the artificial setting of
the interview and, in analysis, abstract these items from their social context. By
contrast, ethnographers of speaking emphasize the importance of studying
language and the genres of verbal art that interest the folklorist in the social
context of performance. At the same time, Bauman and Briggs ( 1990:7 4)
observe that an important feature of works of verbal art is that they are often
created as texts that may be detached from the context of their creation without
significant loss of meaning (see also Hanks 1989).
867
SOCIAL LIFE
Abraham's (1977) analysis of tea meetings in the West Indies provides a

good illustration of the importance of a focus on the social context of
performance. Members of the community concerned recognize two contrasting
styles of speaking: the 'rude' (associated with young men, and street corner
society), and the 'behaved' (associated with women, home, and church). The
two styles confront one another in the 'tea meeting', a speech contest in which
the speechmaker, who is the master of an elevated style of speaking that
combines Latin with erudite English, is challenged to continue despite rude
(and comic) heckling from the audience (1977:117-19). Attention to genre
conventions of the speech alone would fail to convey the social meaning and
humour of the event.
The ethnographic approach to verbal art persuasively demonstrates that
meaning is often only completed in the context of the speech event. The
contexts in which people live their lives are themselves endowed with meaning,
and a full understanding depends on our comprehension of meanings
associated with physical settings and community history. In the Western
Apache practice of 'shooting with stories,' reprimands or didactic messages are
conveyed indirectly by moralistic stories that are linked to physical locations
where memorable past events took place. Retelling the story, or mentioning the
name of the story ('It happened at "men stand above here and there'"), invites
the hearers to search for analogies between their own behaviour and the
disapproved behaviour of the anti-hero of the tale. Moreover, the place itself is
a constant remm
Basso puts it:
After stories and storytellers have served this beneficial purpose [to make the listener
think about his life], features of the physical landscape take over and perpetuate it.
Mountains and arroyos step in symbolically for grandmothers and uncles.
(1984:43)
Words condense a community's recollections of its experiences within a unique

environment, and the environment in turn evokes a community's memories of
its past (see also DeBernardi 1993).
One of the more fruitful developments in the field of the ethnography of
speaking has been cross-cultural work on language socialization (Ochs and
Schieffelin 1983, Schieffelin and Ochs 1986). Following Vygotsky ( 1978, see
also Wertsch 1985), Ochs suggests that the genesis of thought and language
must be placed within the context of social interaction, a point that is also
central to Poole's discussion of socialization and enculturation in the previous
article. Invoking Sapir and Whorf, she further proposes that 'children acquire a
worldview as they acquire a language' (1986:3). In her view, children acquire
performance competence through participation in social interactions that are in
part constructed as an 'interactional display (covert or overt) to a novice of
expected ways of thinking, feeling, and acting' (1986:2).
868
In a formulation that recalls Hymes's emphasis on competence in

performance, Ochs emphasizes the fact that the child learning a language is also
acquiring 'social competence', which entails 'the ability to recognize/interpret
what social activity I event is taking place and to speak and act in ways sensitive
to the context' (1986:3). Types of knowledge acquired in language socialization
include the ability to express status and role through language use, and to
recognize and express feelings in context. For example, in Kaluli society,
teasing and shaming are assertive interactive strategies that are important
modes of social control in adult life. Schieffelin demonstrates that both are a
systematic component of interactions with children, and are used extensively
'to teach children how to be part of Kaluli society' ( 1986: 179).
Recent anthropological research challenges the premiss that emotions are
psychobiologically universal, and relativizes different kinds of affect as
sociocultural constructs that are, like kinship systems or concepts of time,
cross-culturally variable (Besnier 1990, Lutz and Abu-Lughod 1990, Lutz and
White 1986). The key to this research has proven to be the study of discourse,
defined as 'the situated social practices of people speaking, singing, orating, or
writing to and about each other' (Abu-Lughod and Lutz 1990: 10). In the view
of Abu-Lughod and Lutz, discourse does not merely carry emotion; rather
emotional discourse is viewed as a learned 'form of social action that creates
effects in the world' (1990:12). Emotion is both content and effect; hence the
expressive and conative functions of language (to use Jakobson's terms) are
uppermost ere.
Abu-Lughod thus determined to study sentiment through the observation
of emotion talk (Abu-Lughod and Lutz 1990: 15). In her study of the Bedouin
use of poetic form in social life, she observes that there are two basic codes
governing the expression of emotion in Bedouin life, a 'code of honour' that
precludes expressions of vulnerability and weakness, and a 'code of modesty'
that women realize in their conduct through self-restraint and effacement-or
what Abu-Lughod calls 'the honour of the weak' (1986: 108). Bedouin women
contravene both codes in singing poetic couplets that express love, attachment,
and loss. These set poetic forms express standard sentiments; but at the same
time, they are interpreted in the light of the life histories of their singers, whose
inner emotions are inferred from their songs. For example, a woman responded
with anger when her husband took a second wife, but communicated sadness
and grief through song ( 1986:189-94). As with linguistic etiquette (discussed
below), poetic discourse is a strategy of indirectness, and Abu-Lughod
concludes that 'individuals are shielded from the consequences of making
statements and expressing sentiments that contravene the moral system if they
do so in poetry' (1986:248).
869
SOCIAL LIFE
ETHNOGRAPHIC POETICS
Linguistic anthropologists have recently established the new subfield of

'ethnopoetics', focused on the cross-cultural study of poetic language and
performance (Friedrich 1991, Hymes 1981, Sherzer and Woodbury 1987,
Tedlock 1983). This emphasis on the poetic is in part justified by the
philosophical view that language is, in its essence, tropological-a 'work of
art'-and studies often emphasize the creative and emergent qualities of poetic
performance (see Weiner in this volume, Article 21). For example, in a recent
statement that Sapir would surely have endorsed, Friedrich has forcefully
argued that 'language, whether at the individual, sociocultural, or some
universal level, is inherently, pervasively, and powerfully poetic' ( 1986: 17).
Language (and culture) influence the imagination, but the imagination of the
individual speaker, particularly the more poetic person, remains relatively
unconstrained, and is thus able to innovate and reorder cultural and linguistic
materials (1986:17).
The poetic also assumes prominence in the work of scholars who advocate
the study of performance as a means of gaining insight into social life (Abu-
Lughod 1986, Bauman 1977, Caton 1990, Sherzer 1983). Frequently the poetic
elaboration of language is an element of socio-cultural performance, from
political oratory to magical invocation. While the aesthetic qualities of these
performances may be difficult to convey in translation, the social use of the
poetic IS a topic o contmumg re evance an Importance.
Intensified poetic forms are often displayed in formal and public contexts
(see for example Fox 1977, Sherzer 1983), a fact that led Bloch to examine their
different social functions. In his consideration of Merina oratory, he concluded
that such poetic ways of speaking were in fact designed to further the goals of
power-holders, since the formal elegance of the words and the contexts in
which they were spoken made them inaccessible to debate and challenge (1975,
1989). Bloch's interpretation has however been questioned by others who,
finding his perspective over-reductionistic, have in turn refocused on the
aesthetic qualities of verbal art forms. Nonetheless, the debate has drawn
attention to the cross-cultural importance and centrality of poetic language in
public (and often political) contexts.
Caton (1990) explores the poetry of Yemeni men as cultural practice,
enmeshed in the history and public life of a tribal society. For Khawlani
tribesmen, the art of composing poetry and the rhetoric of public persuasion
are closely connected, making politics and poetics inseparable ( 1990: 155). The
poet has 'the power to enter into a discourse in which honour is created or
defended ... and persuasion is exercised' (1990:178), at weddings for example
(pp. 65-71), or in the mediation of disputes (pp. 71-4). These rhetorical
outcomes are only possible if the poet can exhibit skill in the manipulation of
poetic form, and Caton concludes that verbal skill gives the poet power in the
constitution of social reality ( 1990:268). As Gal describes the force oflanguage
870
in politics, 'power is more than an authoritative voice in decision-making; its

strongest form may well be the ability to define social reality, to impose visions
of the world. Such visions are inscribed in language and enacted in interaction'
(Gall989:26).
While ethnopoetics is a relatively new subfield, it has deep roots in the
discipline. Many anthropologists would agree with Victor Turner, who summed
up a fundamental premiss of his anthropology when he stated:
Experience always seeks its 'best', i.e. most aesthetic expression in performance ....
Cultures, I hold, are better compared through their rituals, theatres, tales, ballads,
epics, operas than through their habits.
(Turner and Bruner 1986:13)
In this view, aesthetic expression is inseparable from social use, and art
becomes a window on the most fundamental values in a society (in Article 22 of
this volume, Schechner develops this idea at greater length).
DIALOGIC ANTHROPOLOGY
The turn towards an analytical focus on speaking and dialogue coincides with a
critique of totalizing concepts of culture (or ideology) that leave no place for
variation or the individual voice. Ideally, in a dialogic study many voices, both
capture the system at the expense of a nuanced understanding both of creative

and emergent cultural or linguistic forms (metaphor, for example), and of the
development of personal or subcultural symbols. The writings of the Russian
literary critic Mikhael Bakhtin ( 1981, 1984) on the 'dialogic imagination' in
literature have provided a point of departure in the formulation of this dialogic
critique. From the Bakhtinian perspective, culture is:
an open-ended, creative dialogue of subcultures, of insiders and outsiders, of

diverse factions. A 'language' is the interplay and struggle of regional dialects,
professional jargons, generic commonplaces, the speech of different age groups,
individuals, and so forth.
(Clifford 1988:46)
This reworked concept of culture shifts the analytic focus away from system
and towards the study of the diversity of human practice. In addition, it takes
discourse and dialogue as keys to understanding human experience (see Bruner
1984).
The dialogic approach, with roots in Bakhtinian literary criticism and the
ethnography of speaking, has also had a pronounced impact on the writing of
ethnography (Marcus and Fischer 1986:67-73). The discourse of fieldwork has
itself become the object of critical analysis (Briggs 1986, Moerman 1988), and the
recognition that ethnographic knowledge is the product of a dialogue between
871
SOCIAL LIFE
researcher and informants has led to criticism of standard styles of representation

in ethnographic writing. Anthropologists are seeking alternatives to an
'ethnographic authority' constructed through third-person objective reporting
(Clifford 1988), and have explored ways of using the first-person voice and of
presenting data in the form of a dialogue between ethnographer and informant
(Tedlock 1979, 1987, Tedlock and Mannheim 1994 ).
THE POLITICS OF LANGUAGE USE

Linguistic varieties or ways of speaking may be conscious or unconscious
markers of personal and social identity, and through these markers language
finds an important function. Linguistic usage expresses and creates social
difference, and language realizes the power structure of society in so far as it
expresses, symbolizes, and maintains the social order (Halliday 1978: 172). The
social prestige or stigma attached to linguistic varieties often supports and
expresses the value attached to social identities.
That a linguistic item or variety may function as a badge of identity is
beyond dispute. Linguistic varieties are associated with national identities (see
Smith in this volume, Article 25), class differences, ethnic differences,
subcultural differences, gender and generational differences. Distinctive
aspects of language use range from pronunciation to syntax, from the use of
slang to norms of interaction: speakers may use virtually any aspect of the
mgmsttc co e as a ve tc e or 1 enttty Hu son 8 , Tru gt 7 ). LePage
and Tabouret-Keller (1985:14) argue provocatively that linguistic behaviour 'is
a series of acts of identity in which speakers reveal both their personal identity
and their search for social roles.'
While the relativistic anthropologist might argue that all languages are equal
in so far as they are adequate to the communicative needs of their speakers, the
social fact is that linguistic varieties are stratified (Grillo 1989a, b, Wolfson and
Manes 1985). The question then arises as to how certain linguistic varieties
come to be socially ranked as more prestigious than others. The response most
frequently encountered is that the most prestigious form will be that of the
most powerful group in society, because it is this group that controls such
channels of influence as educational institutions and the media.
Often (though not always) the prestige form is the standard language.
Standardized linguistic varieties are typically the creations of modern nation
states that seek to use language to unify populations and promote literacy (see
Article 19). As Haugen observes, a national language has the two-fold potential to
create internal cohesion and to foster external distinction-a duality that forms a
powerful base for national identity (1972:245). While political goals may have led
to the establishment of national languages, economic factors often motivate their
acceptance. Frequently, material advancement depends on mastery of the
national language, and language shift (and sometimes the 'death' of minority
languages) is not an uncommon outcome (Edwards 1985:91-6, Dorian 1989).
872
The abandonment of a minority language in favour of the dominant

language may be the pragmatic choice from an economic perspective. Minority
languages often persist, however, despite the lack of institutional support for
their transmission. These languages may serve to mark off ethnic differences
within multi-ethnic societies, but loyalty to or revival of marginalized languages
may also express the political aspirations of their speakers.
The case of modern Chinese (Mandarin) well illustrates the political aspects
of language. Prior to the twentieth century, China was united by a shared
written language, but divided into eight distinct major 'dialect' areas whose
speakers in fact spoke topolects as different from one another as French is from
Italian. Each topolect had its 'reading pronunciation' for characters, which
meant that shared literacy did not confer a shared spoken language. High-
ranking scholar-literati who left their regions for national service compensated
for this diversity by learning guan hua, literally 'official language', a lingua
franca based in Northern Chinese.
In the early part of the twentieth century, the republican leaders of China
recognized the potential political and economic value of a shared national
language, and the Ministry of Education called a Conference on Unification of
Pronunciation. In a spirit of compromise, the new national language, based in
Mandarin, incorporated elements of southern dialects (and indeed was called
'Blue-green Mandarin' because of its mixed quality (DeFrancis 1972:66)).
'Blue-green Mandarin' failed, however, to be adopted in use, since the new
natwna anguage a no pre-extstmg commumty o spea ers Ramsey 87: .
In 1932, standard pronunciations were normalized to reflect those in use in
Beijing, in an important step that some feared would 'force the south to follow
the north' (Ramsey 1987:11). Significant progress in the adoption of the
national language was not made until after the Second World War, when the
communists actively promoted the use of Mandarin as well as the simplification
of characters as a means towards widespread literacy. Because they followed the
Soviet model of toleration of topolects, China shifted from being a diglossic
nation to become a nation of bilingual speakers (DeFrancis 1972, 1984,
Norman 1988:249-53).
The pragmatic usefulness of the shared national language does not always
guarantee that it should enjoy an exclusively prestigious status (Norman
1988:245-9). Regional languages continue to be spoken in China and Taiwan in
informal contexts, and in Taiwan, 'Southern Min' (or 'Taiwanese', as it is
known) is now promoted in certain quarters as a symbol of the desire for
independent nationhood. The promotion of Taiwanese involves reclaiming the
reading pronunciation of characters, as well as devising original ways to create a
written standard for the language, since it is a topolect with much vocabulary
that is non-cognate with Northern Chinese (DeBernardi 1991).
Economic power as well as political aspiration may allow the promotion of
regional languages at the expense of national ones. Barcelona, for example, is a
community that is economically powerful but politically peripheral to the
873
SOCIAL LIFE
Spanish state. Castilian is the language of the nation, but it has been
marginalized in Barcelona owing to the social configuration of that urban
community, in which Castilian is the language of an emigrant working
population and Catalan the language associated with elite social standing
(Woolard 1989).
The restoration of a language in decline may also be undertaken as an aspect
of 'ethnic revival'. In Ireland, for example, English policies of settlement and
education (many of them begun as early as the mid-sixteenth century) led to
the decline of Irish. As language shift occurred, English became the language of
social prestige, while Irish became a language maintained primarily among the
poor. In the mid-nineteenth century an Irish nationalist movement emerged
that sought to encourage and revive the original language. This effort at
language restoration was part of a larger national movement, which culminated
in the founding of the Irish Free State (Edwards 1985:53-5).
With the founding of that state Irish became its first official language, and
support for its revival now exists in the form of compulsory education in the
language, the standard use of Irish in bilingual government publications, and
the establishment of a government board to promote its use (Edwards 1985:56-
9). Edwards observes that 'in daily Irish life there are places for the language
but almost all are either ceremonial or trivial, or exist only in tandem with
English' (1985:59-60). The effort to revive Irish as a spoken language has
failed, and Edwards concludes that the original language that romantic
natwna tsts so ar ent y e en e IS not, a ter a , essentta to t e mamtenance
of a strong Irish identity (1985:64).
For multi-lingual speakers, linguistic varieties may index aspects of identity,
and a number of close-grained studies of code-switching have demonstrated
that language choice is both systematic and socially meaningful (Breitborde
1983, Gall979, 1988, Gumperz 1982, Gumperz and Hymes 1986, Helier 1988,
Hill and Hill 1986, Urciuoli 1991 ). Code-switching often involves the use of
both the state-supported language (with associations of power and prestige)
and ones used by minority groups (perhaps stigmatized). The choice of code is
often strategic, and as Gal notes, it is used in conversation 'to establish, cross or
destroy group boundaries; to create, evoke or change interpersonal relations
with their accompanying rights and obligations' (1988:247).
In Kenya, for example, use of the official language indexes membership in a
multi-ethnic elite (Scotton 1988: 162). There are contexts, however, in which
minority languages have greater social value, and Scotton (1988: 169) describes
an interaction in Kenya in which a young woman switches from Swahili, an
ethnically neutral lingua franca, to a shared tribal language, in an attempt to
smooth over a minor conflict with the gatekeeper at her club. In this context,
the minority language is used to establish ethnic eo-identity and to negotiate a
different (more solidary) relationship. In multi-ethnic Kenya, minority
languages may also function to ensure privacy, excluding outsiders from
comprehension (Scotton 1988: 174-5).
874
ACTS OF IDENTITY: LINGUISTIC STYLES
Even within a national or regional language variation exists, which may range
from regional or local dialects to registers (varieties according to use), from
subcultural or social class or ethnic styles to gender differences. For many years,
of course, scholars have noted that any 'language' will vary in use, and what a
transformational grammarian might dismiss as idiosyncratic aspects of
performance, or 'free variation', is of central concern to those who study the
linguistic expression of social identity. As a technique of the body, language use in
this instance is an aspect of habitus, defined by Bourdieu as 'the system of
structured, structuring dispositions ... which is constituted in practice and is
always oriented towards practical functions' (1990:52). One practical function of
the linguistic habitus is the communication of identity through linguistic style. As
a form of 'linguistic capital', style may express and confirm the speaker's position
in society, and mastery of what Bourdieu terms the 'authorized language' may
yield a profit in terms of authority or distinction, since language 'represents,
manifests, and symbolizes authority' (1982: 103-5; see also Irvine 1985).
In the United States, for example, consistent use of the double negative ('He
don't know nothing') is characteristic of a linguistic variety termed 'black
English vernacular' (BEV), and associated with African American speakers. In
the 1960s, this linguistic form came to be stigmatized by educators as 'bad
grammar', or as an expression of illogicality, a judgement made from the
stan pomt o stan ar Eng IS . Wt tarn La ov, owever, as argue or t e
grammatical integrity of BEV, observing with characteristic acuteness that
although the double negative (termed by him 'negative concord') is employed
in Russian, Spanish, French, and Hungarian, those languages are not
stigmatized as 'illogical' (1972a:226). Bolinger sums up the situation thus:
Attitudes toward a form of speech are hardly other than attitudes towards the
speakers. Inferior people speak in inferior ways. Naturally. And the differences that
mark their speech tend to be stigmatized.
(1980:45)
Or as Halliday puts the matter: 'the conscious motif of "I don't like their
vowels" symbolizes an underlying motif of "I don't like their values'"
(1978:179).
There may be an objective basis for the social usefulness and prestige of a
language such as Standard English, which is a highly 'developed' language with
a wide range of functions (Halliday 1978: 194). However, advocates of the
standard language (and in America, proponents of an 'English-only' policy)
have often been ideologically driven. Silverstein observes (1987:3) that 'the
culture of Standard is aggressively hegemonic, dominating .. .linguistic
situations with an understanding of other linguistic usages as locatable only in
terms of Standard.'
875
SOCIAL LIFE
Deep differences between Standard and non-Standard speakers have been

inferred on the basis of patterns of language use. In an influential formulation,
Bernstein ( 1972) suggested that class differences in the use of language in social
interaction could explain differences in academic achievement in Britain. He
proposed that the middle class had achieved mastery of a form of speech that he
termed an 'elaborated code', verbally explicit and relatively independent of
context, a way of speaking that 'maintained social distance, demanded
individuated responses, and made no assumptions about the hearer's intent'
(Halliday 1978:87). The speech of the English working class, by contrast, was
limited to a 'restricted code', which Bernstein described as socially and
situationally bounded, and particularistic rather than universalistic. The
restricted code was a more verbally implicit and context-dependent, but also a
more socially intimate form of speech in which meaning was 'tied to a local
relationship and to a local social structure' (Bern stein 1972:164 ). The
educational system demanded an elaborated code, and children who-due to
the manner of their socialization-did not wholly master this code were put at a
disadvantage.
In the United States, Bernstein's model was (mis)applied by researchers in
the field of education to explain the poor academic performances oflower-class
black children. Labov has singled out for critique the work of Bereiter, who
described the speech of pre-school African American children as a series of
emotional cries amounting to 'a nonlogical mode of expressive behaviour'
Beretter, ctte m La ov 72a:2 . La ov questwne t e researc met o s
that produced these conclusions, pointing out that the structure of the testing
situation itself had an inhibiting effect on the verbal performance of the
children tested (1972a:205-13). He also reviewed the criteria used to define
'elaborated' and 'restricted' codes, and observed that the 'elaborated code' used
by educated speakers was often merely an elaborated style, and was no more
logically incisive or conceptually universalistic than the 'restricted code' of
BEV speakers (1972a:216-20).
'Women's language', as it is termed, also demonstrates the relationship
between linguistic style, identity, and prestige. Gender identity and gender
attitudes find expression in both linguistic form and style, and in studies of
gender and language, the interaction of system and use, representation and
choice, is perhaps more fully explored than in other areas of linguistic
anthropology (McConnell-Ginet 1988). Lakoff long ago pointed out that
'language uses us as much as we use language' (1975:3), and interdisciplinary
feminist scholarship has explored the implications of representations of male
and female implicit in the very structure oflanguage. Use of'man' as a generic
term for 'men and women' has been criticized as 'sexist' for example, as has use
of 'he' as the pronoun of indefinite reference. In the history of the English
language, gender-related vocabulary has undergone a process in which terms
associated with men tend to retain connotations of high social standing
('master', 'lord', 'king'), while those associated with women have acquired
876
derogatory, negative (and often sexual) connotations ('mistress', 'lady', 'queen')

(Graddol and Swann 1989: 112-18 ). Like linguistic relativists, feminist scholars
argue that language shapes worldview, and conclude that linguistic practice
must change if society is to change.
Gender differences in language use have also been documented, to the extent
that Tannen can confidently assert that 'male-female conversation is cross-
cultural communication' (1986:133). Lakoff observed that American women are
taught 'women's language', which she describes as consisting of polite and
deferential ways of speaking which ultimately subordinate women in society
(1975:6-8). Supporting Lakoffs contention that women are more deferential
than men in conversation, research on middle-class Americans has shown that
men take longer conversational turns, and interrupt more frequently in order to
take the floor from the current speaker (West and Zimmerman 1983). Women, by
contrast, do a disproportionate amount of 'maintenance work', by providing
encouraging responses, asking questions, and listening (Fishman 1983 ). At the
same time, however, it has been questioned whether women's language really
exists in English (by comparison with Japanese, where women use special terms
of self-reference and address; Jorden 1974, see also Shibamoto 1985, 1987), or
whether the variety of speech described by Lakoff is not rather a 'powerless
script' used by both men and women who are in socially subordinate positions
(O'Barr and Atkins 1980). Cross-cultural research also suggests that the
association of women with greater politeness is not universal: in Malagasy women
are seen as uect an a rupt, w 1 e men are seen as spea mg wtt care an
indirectness (Keenan 1974).
Words, too, have value. As Bakhtin stated, words have 'owners', and 'each
word tastes of the context and contexts in which it has lived its socially charged
life; all words and forms are populated by intentions' (1981:293). Variability in
language creates and embodies different social lives and intentions, which is
one reason why dislike of a person's vowels, syntax, or choice of words may in
fact, to recapitulate Halliday's point ( 1978: 179), be symbolic of dislike or
disregard of their values.
Slang, for example, is a linguistic style that is associated with young
speakers, and has low value when compared with the formal language of
academic or legal discourse. In a study ofZuni, Newman (1964) observed that
the sacred language used in the ceremonial house (kiva) carried high social
prestige, whereas slang language (while metaphorical and witty) carried low
prestige. The two varieties of Zuni were used by different speakers in distinct
contexts: community elders used sacred language in the kiva, while young
people used slang in informal settings. The use of slang terms like 'cotton-
wood' for 'Anglo-American' was prohibited in the kiva, and the Zuni
compared the use of slang in that setting to bringing a radio into the sacred
precincts. Newman observes that 'status differentiation is applied to age
groups, and the speech peculiar to young people is low-valued, while that
associated with old people is prestigeful.' He concluded that 'words acquire
877
SOCIAL LIFE
connotative gradations in accordance with the cultural values assigned to ideas,

status groups, and situations' ( 1964:402).
The elaboration of linguistic style in the defence of identity is often a
characteristic of subcultural groups. Durkheim classically suggested that the
social contract was inscribed in linguistic categories (1965 [1915]:482-7), and
those who stand outside the norms of the social contract frequently reshape
language to express their values. Such cases are illuminating, since they
illustrate vividly the claim that conversation creates and sustains inter-
subjective reality. Attention to the creative dimensions of language use also
serves as a reminder that 'it is people who retain the power to name, entitle, and
objectify others, who determine the terms of discourse' (Parkin 1984:359).
Halliday coined the term 'anti-language' to describe the argots of socially
marginal persons (primarily thieves) who refashion the language's lexicon in
order to have a secret code in which to speak of such things as criminal acts, the
tools of their trade, their opponents (the police), their victims, and the penalties
for their crimes. Halliday observed that such persons need anti-languages in
order to 'act out a distinctive social structure' wherein is inscribed an
'alternative social reality' ( 1978:165, see also DeBernardi 1987).
David Maurer's classic study of the 'Whiz Mob' provides a rich example of
an anti-language. In the technical argot of pickpockets, victims of their trade
are 'beats', 'clips', 'clouts', and 'nails' ( 1964:49). By contrast, pickpockets are
'class cannons', 'careful tools', or 'bangup operators' ( 1964:9 5). As Halliday
o serves 78: 7 , t ose o w om society at arge tsapproves can respon to
the negativity of their disapprobation by themselves redefining positive and
negative values. The argot of pickpockets also contains a vocabulary of
disapprobation: a pickpocket who is too careful is a 'centre fielder' or 'sneeze
shy'; one who is small-time is a 'doormat thief, and a 'forty-second street thief
is one who will not leave New York (Maurer 1964:96-7). In Halliday's terms,
'an anti-language is the means of realization of a subjective reality: not merely
expressing it, but actively creating and maintaining it' (1978:172).
IDENTITY IN INTERACTION
Social distinctions are also constructed and expressed in interaction through
the use of 'indexical' linguistic items. Indexicals are items that mark features of
the speaker's and/ or the hearer's identity, and they include pronouns, kinship
terms and titles, and the differential use of speech levels. In social use,
indexicals often create and sustain a relational social identity, and thus have a
performative function. As Silverstein notes, 'social indexes such as deference
vocabularies and constructions ... are examples of maximally creative or
performative devices, which, by their very use, make the social parameters of
speaker and hearer explicit' ( 1976:34 ).
The model developed by Brown and Levinson (1987) to discuss linguistic
etiquette is a useful starting-point. They approach politeness from the
878
perspective of a 'model person' who desires to secure the co-operation of others

while avoiding the appearance of imposition. 'Politeness', therefore, involves
linguistic strategies of indirectness that are understandable in the light of
human 'face wants'. They define 'face' as:
the public self-image that every member wants to claim for himself, consisting of
two related aspects:
(a) negative face: the basic claim to territories, personal preserves, rights to non-
distraction-i.e. to freedom of action and freedom from imposition
(b) positive face: the positive consistent self-image or 'personality' (crucially
including the desire that this self-image be appreciated and approved of)
claimed by interactants.
(1987:61)
Brown and Levinson assume that all competent adult members of society have
both 'face', and the rationality to devise means (including strategies of
politeness) to achieve their ends (1987:61 ).
While in the short run direct communication would be a more efficient way
to accomplish interactional work, it would threaten the 'face needs' of others by
imposing demands on them. Forms of politeness, however, communicate
through strategies of indirectness that invite the addressee to draw implications
from non-explicit speech. Brown and Levinson suggest that all languages
emp oy wo pnmary s ra eg1es o po 1 eness, w 1c mvo ve nega 1ve po 1 eness
(or respect and distance), and 'positive politeness' (or familiarity) (1987:2).
These strategies correspond to the axes of power and solidarity often
employed in analyses of pronoun use in European languages. In a classic
study, Brown and Gilman ( 1960) detailed the development and contemporary
use of pronouns in European languages that employ two forms of the second-
person pronoun 'you' (e.g., French vous and tu; Spanish, usted and tu;
German Sie and du). In these pairs of pronouns, the first (which is in fact the
plural 'you') is formal and distant, while the second is informal and familiar.
Use of the familiar form expresses solidarity with a peer, but will express
condescension when used to a social inferior, while use of the plural 'you'
marks respect to the addressee or social distance. When the forms are used
non-reciprocally, an asymmetric power semantic is set up in which the
higher-status speaker condescends to the lower-status hearer with the
familiar form, while the lower-status speaker shows respect to the higher-
status hearer with use of the plural form.
For Brown and Levinson, honorific forms, broadly defined to include such
linguistic phenomena as pronoun use, are a case of 'frozen conversational
implicature' (1987:23). In the case of the second-person pronoun, reciprocal
use of the familiar form marks an assumption of 'positive politeness', while
reciprocal use of the formal form implies the social distance of 'negative
politeness'. The power semantic established through non-reciprocal pronoun
879
SOCIAL LIFE
use implies a social difference: to borrow a Wolof expression, 'one person has
shame, the other has glory' (Irvine 1974:17 5).
The social structural assumptions embedded in such linguistic forms
surface when they are contested, as occurred when seventeenth-century
Quakers refashioned their linguistic practices to express radical social ideals.
The Quakers sought to replace the socio-religious dominance of Anglican
ministers with the authority of persons 'speaking in the light' of divine
revelation. In pursuit of this goal they eliminated mannered and polite ways of
speaking, and substituted 'plain speech', which for them was literal speech.
Use of the informal 'thee' replaced the asymmetric pronominal semantic of
'you' and 'thee', and they reasoned that use of the plural 'you' to address a
single person was untruthful. Against the backdrop of conventional norms, the
Quakers' use of 'thee' to address their social superiors was however interpreted
as insolent behaviour, and was considered deeply offensive (Bauman 1983). A
contemporary observer wrote in 1655 that:
We maintain that Thou from superiors to inferiors is proper, as a sign of command;

from equals to equals is passable, as a note of familiarity, but from inferiors to
superiors, if proceeding from ignorance, hath a smack of clownishness; if from
affectation, a tang of contempt .... Such who now quarrel at the honour will
hereafter question the wealth of others.
(cited in Hill 1972:247)
Clearly the 'frozen implicature' of pronoun use functions as part of a linguistic

habitus that is both structured and structuring: consequently, a change in that
habitus is a threat to structural continuity. The observer whose words are cited
above did not err when he interpreted the Quaker challenge to linguistic norms
as a challenge to the social structure itself.
Occasionally social and political elites have attempted to transform social
structure through language planning. During the French revolution, for
example, the Committee for the Public Safety banned use of the aristocratic
vous as a feudal remnant, and promoted adoption of mutual tu and the
egalitarian Citoyen (Brown and Gilman 1960:266). More recently, the Chinese
equivalent of Citoyen, Tongzhi (literally, 'with a common will'), was promoted
to replace inegalitarian titles with the goal of remaking Chinese society through
a reworking oflinguistic norms (Scotton and Zhu 1983).
As in France, in China linguistic asymmetry persisted, and Tongzhi
('Comrade') came to be used primarily in situations where the status title of the
addressee was unknown. In circumstances of status inequality, use of
'Comrade' took on marked meanings. When a social inferior addressed a social
superior as 'Comrade', the meaning invoked was one of solidarity, and most
often the term was used when making a request. When a social superior
addressed a social inferior as 'Comrade', then most often the superior implied a
rebuke (Scotton and Zhu 1983:483-4).
880
While social hierarchy persists in China, the social system has nevertheless
changed dramatically, and terms of address such as 'Comrade' are one index of
this change. Halliday has observed that:
semantic style is a function of social relationships and situation types generated by

the social structure. If it changes, this is not so much because of what people are now
speaking about as because of who they are speaking to, in what circumstances,
through what media, and so on. A shift in the fashion of speaking will be better
understood by reference to changing patterns of social interaction and social
relationships than by the search for a direct link between the language and the
material culture.
(1978:77)
Linguistic style may indeed be a function of social structure, but social

structure is realized only in social activity, including, conspicuously, the
activity of discourse.
Asian languages provide a good illustration of the realization of social
structure in discourse. A number of East and South-east Asian languages make
extensive use of speech levels and honorific vocabulary, and the relative statuses
of speaker and hearer find linguistic expression in these forms. Like pronoun
use in European languages, the use of these forms of politeness is often
expressive of identity within a social hierarchy.
For s eakers of a anese choice from a variet of s eech levels indicatin
formality and politeness is basic to all communication (Shibamoto 1987:269).
The choice of plain, polite, or deferential style depends on the speaker's
attitude towards the person(s) addressed. Within each of these styles, Japanese
speakers also have the choice of humble, neutral, or exalted expressions,
depending on their attitude towards the subject of the expression (Martin
1964:408-9). Age difference, gender difference, social position, and group
membership are all factors in the choice of speech levels and the degree of
politeness. A study of Japanese concluded that 'politeness of usage seems to be
in inverse proportion to feeling that one has the upper hand in a situation'
(Martin 1964:411 ).
Traditionally, strict codes of etiquette ruled the Thai court and the Malay
sultanates, and the use of polite language was one aspect of the enactment of
social hierarchy. Politeness entailed the correct use of formal modes of
addressing royalty with linguistic terms that exalted royalty and humbled those
oflower status. These terms often gave vivid expression to social difference. For
example, one Sanskrit-derived term of address for Malay royalty is paduka,
literally 'shoe', defined by Coope as 'a royal title derived from the fact that the
subject addresses the raja's feet, being unworthy to address the prince himself.
Common people were pacal, meaning 'slave of a slave, the humblest of the
humble' (Coope 1976: 197). This highly self-deprecatory term was used as a
first-person pronoun when non-royalty addressed royalty, and ministers also
881
SOCIAL LIFE
referred to their family members as 'slaves of slaves' when they addressed their
raja (Ghazali 1977:275).
In Thailand a palace language also existed, which was primarily derived
from Sanksrit (with some vocabulary from Khmer). As Wales describes the
Thai court, 'it was taboo to use words of the common language, or common
modes of address, when speaking to or about the King and princes' (1931:39).
The first-person pronoun used when addressing the King meant 'I, the slave of
the Lord Buddha'; the second-person pronoun meant 'the dust beneath the
sole of your august feet', meaning that the speaker did not dare to address the
king directly, and directed his comments instead to the dirt on the floor
(1931:40). A range of lexical items also had a court form, including terms for
body parts, articles used by royal persons, food and drink, kinship
relationships, verbs of bodily action, and names of certain animals, fish, fruit,
and flowers (1931:39-40).
In contemporary Javanese, the norms governing linguistic etiquette are basic
to correct language use, and Javanese provides a complex illustration of the use
of language in the construction of identity in interaction (Keeler 1984). Geertz
noted as follows:
In Javanese it is nearly impossible to say anything without indicating the social

relationship between the speaker and the listener in terms of status and familiarity.
Status is determined by many things-wealth, descent, education, occupation, age,
. . . . . . . .
linguistic forms as well as speech style is in every case partly determined by the
relative status (or familiarity) of the conversers.
(1960:248)
Speech levels, however, are but one complex and nuanced aspect of Javanese
etiquette, an etiquette that governs not only speaking but also 'sitting, standing,
pointing, composing one's countenance, and so on' (Errington 1988:11 )-in
short, what Bourdieu would term the habitus of Javanese society.
In a study of the speech levels of the Javanese elite (priyayi), Errington cites
an elderly Javanese who instructed him thus:
Whenever two people meet they should ask themselves: 'Who is this person? Who
am I? What is this person to me?' (Here he held out his hands, palms up, as if they
were pans of a scale.) That's 'relative value' (unggah-ungguh).
(1988: 11)
Speakers of Javanese choose among lexical variants, and 'the system is based on
sets of precisely ranked, or style-coded morphemes that are semantically
equivalent but stylistically contrastive' (Wolfowitz 1991: 121). Most basic to the
system is the distinction between ordinary and polite vocabulary, but speakers
draw on these to create a continuum of stylistic mixes (Wolfowitz 1991: 123).
Also important is an honorific vocabulary referring to the 'possessions,
882
attributes, states, and actions of persons', a vocabulary that includes honorific

kin terms (since kin may be regarded as attributes of persons (Errington
1988: 139)).
CONCLUSION
Anthropologists involved in the cross-cultural study of language use have
contributed to the formation of a socially grounded linguistics that has great
relevance for socio-cultural analysis. The structure of language and norms for
language use are basic to the matrix of social life, and a wide range of
anthropological questions cannot be addressed without taking account of the
data provided by linguistic form and function.
The analysis of 'discourse' holds a central place in contemporary
scholarship, comparable to that once held by structural analysis or
hermeneutics, and anthropologists have also made essential contributions to
interdisciplinary dialogues on this topic. For many scholars, the study of
discourse is closely linked to questions of social power and ideological control,
and those who study the 'political economy' of language use argue that
discourse is a fundamental means through which hegemonic ideas are imposed
or contested, and social differences reproduced (Eagleton 1991: 193-220). The
cross-cultural study of language use provides discourse analysis with essential
comparative insight into the conceptual and practical ordering of social life,
an un erscores t e Importance o t e unconsciOus 1 eo ogtca tmenswns o
language use. At the same time, however, the close focus on the relationship
between language and power should not be allowed to overshadow the other
manifold uses to which speakers put language, including (most notably) the
aesthetic.
Socio-cultural anthropologists of various theoretical persuasions have
explored the role of language in the construction of social thought and practice.
As demonstrated above, language use shapes the formation of the conceptual
systems shared by speakers of a language, and at the same time constitutes
diverse social identities in interaction. In speaking, these two aspects of
language use converge, as when metaphorically derived polite terms of address
simultaneously image hierarchy and index identity: the Thai noble who
addresses his comments to the dirt beneath his king's shoe is invoking a
cultural image of 'low' status, but he is also indexing relative identity in the
social interaction of discourse. Language is profoundly social, and language use
both constitutes shared worlds and realizes social diversity in practice.
REFERENCES
Abrahams, R.D. (1977) 'The training of the man of words in talking sweet', in
R.Bauman (ed.) Verbal Art as Performance, Prospect Heights, Ill.: Waveland Press.
883
SOCIAL LIFE
Abu-Lughod, L. (1986) Veiled Sentiments: Honor and Poetry in a Bedouin Society,

Abu-Lughod, L. and Lutz, C. A. (1990) 'Introduction', in C.A.Lutz and L.Abu-
Lughod (eds) Language and the Politics of Emotion, Cambridge: Cambridge
University Press.
Ardener, E. (ed.) (1971) Social Anthropology and Language, A.S.A. Monographs no. 10,
London: Tavistock.
Bakhtin, M. (1981) The Dialogic Imagination, ed. M.Holquist, trans. C.Emerson and
M.Holquist, Austin: University of Texas Press.
--(1984) Problems of Dostoevsky's Poetics, ed. and trans. C.Emerson, Minneapolis:
University of Minnesota Press.
Basso, K. (1972) "'To give up on words": silence in Western Apache culture', in
P.P.Giglioli (ed.) Language and Social Context, New York: Penguin (reprinted in
Basso 1990).
--(1979) Portraits of the 'Whiteman' : Linguistic Play and Cultural Symbols Among the
Western Apache, Cambridge: Cambridge University Press.
--(1984) "'Stalking with stories": names, places, and moral narratives among the
Western Apache', in E.M.Bruner (ed.) Text, Play, and Story: The Construction and
Reconstruction of Self and Society, Prospect Heights, Ill.: Waveland Press (reprinted
in Basso 1990).
--(1990) Western Apache Language and Culture: Essays in Linguistic Anthropology,
Tucson: University of Arizona Press.
Bauman, R. (ed.) (1977) Verbal Art as Performance, Rowley: Newbury House.
--(1983) Let Your Words be Few, Cambridge: Cambridge University Press.
auman, . an nggs, oet1cs an per ormance as cnt1ca perspectives
on language and social life', Annual Review ofAnthropology 19:59-88.
Bauman, R. and Sherzer, J. (eds) (1974) Explorations in the Ethnography of Speaking,
Benveniste, E. (1971 [1966]) 'The nature of the linguistic sign', in Problems in General
Linguistics, trans. M.E.Meek, Coral Gables, Fla.: University of Miami Press.
Bernstein, B. (1972) 'Social class, language and socialization', in P.P.Giglioli (ed.)
Language and Social Context, New York: Penguin.
Besnier, N. (1990) 'Language and affect', Annual Review ofAnthropology 19:(419-51 ).
Black, M.E. (1984) 'Maidens and mothers: an analysis of Hopi corn metaphors',
Ethnology 23(4):279-88.
Bloch, M. (1975) 'Introduction', in M.Bloch (ed.) Political Language and Oratory in
Traditional Society, London: Academic Press.
--(1989) 'Symbols, song, dance and features of articulation: is religion an extreme
form of traditional authority?', in M.Bloch, Ritual, History and Power: Selected
Papers in Anthropology, London: Athlone Press.
Boas, F. (1966 [1911]) Introduction to the Handbook of American Indian Languages,
Washington, D.C.: Georgetown University Press Reprint.
Bolinger, D. (1980) Language, The Loaded Weapon. The Use and Abuse of Language
Today, London: Longman.
Bourdieu, P. (1982) Ce que Par/er veut dire: l'economie des echanges linguistiques, Paris:
Fayard.
--(1990 [1980]) The Logic of Practice, Stanford, Cal.: Stanford University Press.
884
Breitborde, L.B. (1983) 'Levels of analysis in sociolinguistic explanation: bilingual code

switching, social relations, and domain theory', International Journal ofthe Sociology
of Language 39:5-43.
Briggs, C.L. (1986) Learning How to Ask: A Sociolinguistic Appraisal of the Role of the
Interview in Social Science Research, Cambridge: Cambridge University Press.
Brown, P. and Levinson, S. (1987) Politeness: Some Universals in Language Usage,
Brown, R. and Gilman, A. (1960) 'The pronouns of power and solidarity', in T.A.
Sebeok (ed.) Style in Language, Cambridge, Mass.: MIT Press.
Bruner, E.M. (ed.) (1984) Text, Play, and Story: The Construction and Reconstruction of
Self and Society, Prospect Heights, Illinois: Waveland Press.
Caton, S.C. (1987) 'Contributions of Roman Jakobson', Annual Review ofAnthropology
16:223-60.
--(1990) 'Peaks of Yemen I Summon': Poetry as Cultural Practice in a North Yemeni
Tribe, Berkeley: University of California Press.
Clifford, J. (1988) 'On ethnographic authority', in The Predicament of Culture:
Twentieth-Century Ethnography, Literature, and Art, Harvard: Harvard University
Press.
Coope, A.E. (1976) A Malay-English Dictionary, Kuala Lumpur: Macmillan Malaysia.
DeBernardi, J. (1987) 'The god of war and the vagabond Buddha', Modern China
13(3):310-32.
--(1991) 'Linguistic nationalism: the case of southern Min', in V.Mair (ed.) Sino-
Platonic Papers, University of Pennsylvania.
--(1993) 'Historical allusion and the defense of identity: Malaysian Chinese popular
re Igwn, m . ar acre, . eyes an
Honolulu: University of Hawaii Press.
DeFrancis, J. (1972 [1950]). Nationalism and Language Reform in China, New York:
Octagon Books.
--(1984) The Chinese Language: Fact and Fantasy, Honolulu: University of Hawaii
Press.
Dorian, N. (ed.) (1989) Investigating Obsolescence: Studies in Language Contraction and
Death, Cambridge: Cambridge University Press.
Durkheim, E. (1965 [1915]) The Elementary Forms of the Religious Life, New York: Free
Press.
Eagleton, T. (1991) Ideology: An Introduction, London: Verso.
Edwards, J. (1985) Language, Society, and Identity, Oxford: Blackwell.
Errington, J.J. (1988) Structure and Style in Javanese: A Semiotic View of Linguistic
Etiquette, Philadelphia: University of Pennsylvania Press.
Evans-Pritchard, E.E. (1940) The Nuer, Oxford: Oxford University Press.
Firth, J.R. (1957) Papers in Linguistics 1934-1951, Oxford: Oxford University Press.
Fishman, J.A. (1986 [1972]) 'Domains and the relationship between micro- and
macrosociolinguistics', in J.J. Gumperz and D.Hymes ( eds) Directions in
Sociolinguistics, Oxford: Blackwell.
Fishman, P.M. (1983) 'Interaction: the work women do', in B.Thorne, C.Kramarae and
N.Henley (eds) Language, Gender, and Society, Rowley, Mass.: Newbury House.
Fox, J.J. (1977) 'Roman Jakobson and the comparative study of parallelism', in C.H.van
Schooneveld and D.Armstrong (eds) Roman Jakobson: Echoes of his Scholarship,
Lisse, The Netherlands: Peter de Ridder Press.
885
SOCIAL LIFE
Friedrich, P. (1979) 'The symbol and its relative non-arbitrariness', in Language.

Context, and the Imagination, Stanford, Cal.: Stanford University Press.
--(1986) The Language Parallax: Linguistic Relativism and Poetic Indeterminacy,
Austin: University of Texas Press.
--(1991) 'Polytropy' inJ.WFernandez (ed.) Beyond Metaphor: The Theory ofTropes
in Anthropology, Stanford, Cal.: Stanford University Press.
Gal, S. (1979) Language Shift: Social Determinants of Linguistic Change in Bilingual
Austria, New York: Academic Press.
--(1988) 'The political economy of code choice', in M.Heller (ed.) Codeswitching:
Anthropological and Socio linguistic Perspectives, Berlin: Mouton de Gruyter.
--(1989) 'Between speech and silence: the problematics of research on language and
gender' IPrA Papers in Pragmatics 3(1):1-38.
Geertz, C. (1960) The Religion ofJava, Chicago: University of Chicago Press.
Ghazali, D. M. (1977 [1933]) 'Court language and the etiquette of the Malays',
Malaysian Branch of the Royal Asiatic Society Reprint no. 4: A Centenary Volume,
Singapore: Times Printers.
Graddoll, D. and Swann,J. (1989) Gender Voices, Oxford: Blackwell.
Grillo, R. (ed.) (1989a) Social Anthropology and the Politics of Language, London:
Routledge.
--(1989b) Dominant Languages: Language and Hierarchy in Britain and France,
Gumperz, J.J. (1982) Discourse Strategies, Cambridge: Cambridge University Press.
Gumperz, J.J. and Hymes, D. (eds) (1986 [1972]) Directions in Sociolinguistics: The
Ethnography of Communication, Oxford: Blackwell.
a 1 ay, anguage as octa emtottc: e octa nterpretatton o
Language and Meaning, Baltimore, Md.: University Park Press.
Hanks, WF. (1989) 'Text and textuality', Annual Review ofAnthropology 18:95-127.
Haugen, E. (1972) 'Dialect, language, nation', inA.S.Dil (ed.) The Ecology ofLanguage,
Stanford, Calif.: Stanford University Press.
Helier, M. (ed.) (1988) Codeswitching: Anthropological and Sociolinguistic Perspectives,
Berlin: Mouton de Gruyter.
Hill, J.C. (1972) The World Turned Upside Down: Radical Ideas During the English
Revolution, New York: Viking Press.
Hill, J.H. and Hill, K.C. (1986) Speaking Mexicano: Dynamics of Syncretic Language in
Central Mexico, Tucson: University of Arizona Press.
Hudson, R.A. (1980) Sociolinguistics, Cambridge: Cambridge University Press.
Hymes, D. (1974) Foundations in Sociolinguistics: An Ethnographic Approach,
Philadelphia: University of Pennsylvania Press.
--(1981) 'In Vain I Tried to Tell You': Essays in Native American Ethnopoetics,
Irvine, J.T. (1974) 'Strategies of status manipulation in the Wolof greeting', in
R.Bauman and J.Sherzer (eds) Explorations in the Ethnography of Speaking,
--(1985) 'Status and style in language', Annual Review ofAnthropology 14:557-81.
Jakobson, R. (1960) 'Closing statement: linguistics and poetics', in T.A.Sebeok (ed.)
Style in Language, Cambridge, Mass.: MIT Press.
Jorden, E.H. (1974) 'Female speech: persisting myth and persisting reality', in Report
886
ofthe Second US. Japan Joint Sociolinguistics Conference, Tokyo: Japanese Society for
the Promotion of Social Science.
Keeler, W. (1984) Javanese: A Cultural Approach, Athens: Ohio University
Monographs in International Studies.
Keenan, E. (1974) 'Norm-makers, norm-breakers: uses of speech by men and women
in a Malagasy community', in R.Bauman and J.Sherzer (eds) Explorations in the
Ethnography of Speaking, Cambridge: Cambridge University Press.
Labov, W (1972a) Language in the Inner City, Philadelphia: University of Pennsylvania
Press.
--(1972b) Sociolinguistic Patterns, Philadelphia: University of Pennsylvania Press.
Lakoff, G. and Johnson, M. (1980) Metaphors We Live By, Chicago: University of
Chicago Press.
Lakoff, R. (1975) Language and Woman's Place, New York: Harper & Row.
Leach, E.R. (1970) Claude Levi-Strauss, Harmondsworth: Penguin.
LePage, R. and Tabouret-Keller, A. (1985) Acts of Identity: Creole-based Approaches to
Language and Identity, Cambridge: Cambridge University Press.
Levi-Strauss, C. (1963 [1958]) Structural Anthropology, trans. C.Jacobson and
B. Grundfest Schoepf, New York: Basic Books.
--(1966 [1962]) The Savage Mind, Chicago: The University of Chicago Press.
--(1969a [1949]) The Elementary Structures of Kinship, trans. J.H.Bell, J.R.von
Sturmer, ed. R.Needham, Boston: Beacon Press.
--(1969b [1964]) The Raw and the Cooked: Introduction to a Science of Mythology,
vol. 1, trans. J. and D.Weightman, Chicago: The University of Chicago Press.
Lucy, J.A. (1985) 'Whorfs view of the linguistic mediation of thought' in E.Mertz and
.. armentier e s emtottc e tatton:
Perspectives, New York: Academic Press.
Luong, H.V. (1990) Discursive Practices and Linguistic Meanings, Amsterdam and
Philadelphia: John Benjamins.
Lutz, C. and Abu-Lughod, L. (eds) (1990) Language and the Politics of Emotion,
Lutz, C. and White, G.M. (1986) 'The anthropology of emotions', Annual Review of
Anthropology 15:405-36.
McConnell-Ginet, S. (1988) 'Language and gender' in F.J.Newmeyer, (ed.) Linguistics:
The Cambridge Survey, vol. 4, Language: the Sociocultural Context, Cambridge:
Malinowski, B. (1922) Argonauts of the Western Pacific, London: Routledge & Kegan
Paul.
--(1978 [1935]) Coral Gardens and Their Magic, vol. 2: The Language of Magic and
Gardening, New York: Dover.
Marcus, G.E. and Fischer, M.M.J. (1986) Anthropology as Cultural Critique: An
Experimental Moment in the Human Sciences, Chicago: University of Chicago Press.
Martin, S.E. (1964) 'Speech levels in Japan and Korea', in D.Hymes (ed.) Language in
Culture and Society: A Reader in Linguistics and Anthropology, New York: Harper &
Row.
Maurer, D.W (1964 [1955]) Whiz Mob: A Correlation of the Technical Argot of
Pickpockets with Their Behavioral Pattern, New Haven: College and University
Press.
887
SOCIAL LIFE
Moerman, M. (1988) Talking Culture: Ethnography and Conversational Analysis,

Mukarovsky, J. (1977) The Word and Verbal Art. Selected Essays by ]an Mukarovsky,
trans. and ed. J.Burbank and P.Steiner, New Haven: Yale University Press.
Newman, S. (1964) 'Zuni sacred and slang usage', in D.Hymes (ed.) Language in Culture
and Society: A Reader in Linguistics and Anthropology, New York: Harper & Row.
Nietzsche, F. (1954) 'On truth and lie in an extramoral sense', in W.Kaufmann (ed.)
The Portable Nietzsche, New York: Viking.
Norman, J. (1988) Chinese, Cambridge: Cambridge University Press.
O'Barr, W. and Atkins, B.K. (1980) "'Women's language" or "powerless language"?', in
S.McConnell-Ginet, R.A.Borker and N.Furman (eds) Women and Language in
Literature and Society, New York: Praeger.
Ochs, E. (1986) 'Introduction', in B.B.Schieffelin and E.Ochs ( eds) Language
Socialization Across Cultures, Cambridge: Cambridge University Press.
Ochs, E. and Schieffelin, B.B. (eds) (1983) Acquiring Conversational Competence,
London: Routledge & Kegan Paul.
Ortner, S.B. (1973) 'On key symbols', American Anthropologist 75:1338-46.
Parkin, D. (1984) 'Political language', in Annual Review ofAnthropology 13:345-65.
Radcliffe-Brown, A.R. (1965 [1952]) 'On joking relationships', in Structure and
Function in Primitive Society: Essays and Addresses, New York: Free Press.
Ramsey, S.R. (1987) The Languages of China, Princeton: Princeton University Press.
Rei chard, G. (1951) Navaho Grammar, New York: J.J.Augustin.
Sapir, E. (1949a [1921]) Language: An Introduction to the Study of Speech, New York:
Harcourt Brace Jovanovich.
e ecte rtttngs o
Sapir, J.D. and Crocker, C. (eds) (1977) The Social Use of Metaphor, Philadelphia:
University of Pennsylvania Press.
Saussure, F. de (1959 [1915]) Course in General Linguistics, ed. C.Bally and A.Sechehaye,
trans. WBaskin, New York: McGraw-Hill.
Schieffelin, B.B. (1986) 'Teasing and shaming in Kaluli children's interactions', in
B.B.Schieffelin and Ochs, E. (eds) Language Socialization Across Cultures,
Schieffelin, B.B. and Ochs, E. (eds) (1986) Language Socialization Across Cultures,
Scotton, C. M. (1988) 'Code switching as indexical of social negotiations', in M.Heller
(ed.) Codeswitching: Anthropological and Sociolinguistic Perspectives, Berlin: Mouton
de Gruyter.
Scotton, C.M. and Zhu, W (1983) 'Tongzhi in China', Language in Society 12:477-94.
Sherzer, J. (1983) Kuna Ways of Speaking: An Ethnographic Perspective, Austin:
University of Texas Press.
Sherzer, J. and Woodbury, A.C. (eds) (1987) Native American Discourse: Poetics and
Rhetoric, Cambridge: Cambridge University Press.
Shibamoto, J.S. (1985) Japanese Women's Language, New York: Academic Press.
--(1987) 'Japanese sociolinguistics', Annual Review ofAnthropology 16:261-78.
Silverstein, M. (1976) 'Shifters, linguistic categories, and cultural description', in
K.H.Basso and H.A.Selby (eds) Meaning in Anthropology, Albuquerque: University
of New Mexico Press.
888
--(1979) 'Language structure and linguistic ideology', in P.R.Clyne, W.F.Hanks and

C.L.Hofbauer (eds) The Elements: A Parasession on Linguistic Units and Levels,
Chicago: Chicago Linguistic Society.
--(1987) Monoglot 'Standard' in America, Working Papers and Proceedings of the
Center for Psychosocial Studies no. 13, eds R.J.Parmentier and G. Urban, Chicago:
Center for Psychosocial Studies.
Tannen, D. (1986) That's Not What I Meant! How Conversational Style Makes or Breaks
Your Relationships, New York: William Morrow.
Tedlock, D. (1979) 'The analogical tradition and the emergence of a dialogical
anthropology', Journal ofAnthropological Research 35(4): 387-400.
--(1983) The Spoken Word and the Work of Interpretation, Philadelphia: University of
Pennsylvania Press.
--(1987) 'Questions concerning dialogical anthropology', Journal ofAnthropological
Research 43(4): 325-44.
Tedlock, D. and Mannheim, B. (eds) (1994) The Dialogic Emergence of Culture, Urbana:
University of Illinois Press.
Thompson, E.P. (1967) 'Time, work-discipline, and industrial capitalism', Past and
Present 38:56-97.
Trudgill, P. (1974) Sociolinguistics: An Introduction, Harmondsworth: Penguin.
Turner, V. (1974) Dramas, Fields, and Metaphors, Ithaca: Cornell University Press.
Turner, V.W. and Bruner, E.M. (eds) (1986) The Anthropology of Experience, Urbana:
University of Illinois Press.
Urciuoli, B. (1991) 'The political topography of Spanish and English: the view from a
New York Puerto Rican neighborhood', American Ethnologist 18(2):295-310.
ygots y, tn tn octety. e eve opment o tg er syc o ogtca rocesses,
eds M.Cole, V.John-Steiner, S.Scribner and E.Souberman. Cambridge, Mass.:
Wales, H.G.Q (1931) Siamese State Ceremonies: Their History and Function, London:
Bernard Quaritch.
Wertsch, J.V. (1985) Vygotsky and the Social Formation of Mind. Cambridge, Mass.:
West, C. and Zimmerman, D. (1983) 'Small insults: a study of interruptions in cross-
sex conversations between unacquainted persons', in B.Thorne, C.Kramarae, and
N.Henley (eds) Language, Gender, and Society, Rowley, Mass.: Newbury House.
Whorf, B.L. (1964) Language, Thought, and Reality, ed. J.Carroll, Cambridge, Mass.:
MIT Press.
Witherspoon, G. (1977) Language and Art in the Navaho Universe. Ann Arbor:
University of Michigan Press.
--(1980) 'Language in culture and culture in language', International Journal of
American Linguistics 46:1-13.
Wolfowitz, C. (1991) Language Style and Social Space: Stylistic Choice in Suriname
Javanese, Urbana: University of Illinois Press.
Wolfson, N. and Manes, J. (eds) (1985) Languages of Inequality, Berlin: Waiter de
Gruyter.
Woolard, K. (1989) Doubletalk: Bilingualism and the Politics of Ethnicity in Catalonia,
Stanford: Stanford University Press.
889
SOCIAL LIFE
FURTHER READING
Basso, K. (1990) Western Apache Language and Culture: Essays in Linguistic
Anthropology, Tucson: University of Arizona Press.
Bauman, R. and Sherzer, J. (eds) (1974) Explorations in the Ethnography of Speaking,
Bourdieu, P. (1982) Ce que Par/er veut dire: l'economie des echanges linguistiques, Paris:
Fayard.
Brown, P. and Levinson, S. (1987) Politeness: Some Universals in Language Usage,
Brown, R. and Gilman, A. (1960) 'The pronouns of power and solidarity', in T.A.
Sebeok (ed.) Style in Language, Cambridge, Mass.: MIT Press.
Edwards, J. (1985) Language, Society, and Identity, Oxford: Blackwell.
Fernandez, J.W. (1991) Beyond Metaphor: The Theory of Tropes in Anthropology,
Stanford, Calif: Stanford University Press.
Friedrich, P. (1986) The Language Parallax: Linguistic Relativism and Poetic
Indeterminacy, Austin: University of Texas Press.
Grillo, R. (ed.) (1989) Social Anthropology and the Politics of Language, London:
Routledge.
Helier, M. (ed.) (1988) Codeswitching: Anthropological and Sociolinguistic Perspectives,
Berlin: Mouton de Gruyter.
Hudson, R.A. (1980) Sociolinguistics, Cambridge: Cambridge University Press.
Hymes, D. (1974) Foundations in Sociolinguistics: an Ethnographic Approach.
Psychological Perspectives, New York: Academic Press.

Newmeyer, F.J. (ed.) (1988) Linguistics: The Cambridge Survey, vol. 4, Language: The
Socio-cultural Context, Cambridge: Cambridge University Press.
Sapir, E. (1949 [1921]) Language: An Introduction to the Study of Speech, New York:
Harcourt Brace Jovanovich.
Schieffelin, B.B. and Ochs, E. (eds) (1986) Language Socialization Across Cultures,
890
32
WORK, THE DIVISION OF LABOUR

AND CO-OPERATION
Sutti Ortiz
WORK, LABOUR AND LEISURE

Farmers, factory workers, housewives, hunters, secretaries and children spend
part of their days working, studying, training, and then enjoying themselves.
We use different terms to describe these various activities, even though what is
done in leisure time and work time may be very similar. A weekend gardener
ten mg IS tomatoes m IS ac yar IS per ormmg an activity compara e to
what a coffee farmer does when he plants tomatoes among the immature coffee
trees. A young person spending his day in a trade school toils in the same way as
a young apprentice learning a trade. Yet, we think of home gardeners as
enjoying themselves while farmers work; of students as applying themselves to
acquire knowledge, while apprentices work to receive instruction in a trade.
Why do we use different words? What is the significance attached to the
term 'work' that renders it inapplicable to the student and weekend gardener?
Adam Smith pointed to a crucial distinction in the various uses of the term in
his book of 1776, The Wealth of Nations (Smith 1982 [1776]). He would have
answered our question by pointing out that the tomatoes of the farmers, in
being exchanged in the market, would have become commodities that
generated capital, whereas the tomatoes of the gardener would have been likely
only to generate pleasure. The significant difference was not in the quality of
the effort but in how the product of the effort was used. Since the effort of the
farmer yielded wealth and capital it could be called 'productive' effort or
'work'. The gardener, on the other hand, was engaged in an 'unproductive'
effort since the products were consumed. Likewise, the work of servants
'seldom leaves any trace of value behind' (Smith 1982:430) since nothing
exchangeable is produced by their effort.
Economists since Adam Smith have narrowed the meanings of the concept
of work, focusing on what they have called 'productive labour'-a labour that
891
SOCIAL LIFE
produces commodities with exchange value-and disregarding efforts that they

regard as unproductive or activities that they consider to represent leisure.
They argue that the growth of an economy depends, in part, on the willingness
of the suppliers of labour to give up leisure time, and on the capacity of the
economy to absorb the labour supplied. By plotting the flows of labour, the
income it can offer to those who provide it, and the capital it can generate to
those who purchase it, economists are indeed able to uncover some of the
dynamics of economic systems. Since productivity is the corner-stone of the
definition of labour, its productive potential must be one of the determinants of
the wage rate. The capability of skilled or educated labourers to produce either
more items within a unit of time or items with higher market value-that is,
their ability to generate more capital-is remunerated with higher wages. A
tractor driver is paid more than a field hand; piece rate payments ensure that
the more experience harvester receives the higher remuneration. A more
complex reformulation of this argument, incorporating conditions of supply
and demand, has permitted economists to construct elaborate models to
describe and predict the movement of wages, trends in the economy, and the
behaviour of firms and farmers. But to do so they have had not only to rely on
Adam Smith's narrower meaning of the term 'productive labour', but also to
quantify it in terms of its market value and the units of time required.
By concentrating on the narrower analytical definition oflabour, economists
have excluded from their analysis, until recently, the impact of other work
e orts t at ma e t e avat a 1 tty o a gtven quantity o pro uctiVe a our
possible. For example, the time spent by the wives and daughters of the wage
labourers preparing their meals and washing their clothes-effort that allows
the labourer to spent more time working for wages and that ensures the
reproduction of the labour force-is often ignored. Likewise, the
'unproductive' labour of the service sector has remained on the sidelines of
most economic models.
There are other, less obvious aspects of work effort that should not be
ignored, as they reveal important social meanings affecting the performance
and the form of remuneration. For example, a wife who spends much of her
time doing 'housework' is said to be a 'housewife'. Her work is considered a
duty, not a remunerable service. It is a duty assumed at marriage and
reciprocated with consideration and financial support. Her work not only
allows for the reproduction of the domestic unit, but also helps to strengthen
the marital bond. She helps her man. The time she spends on housework is not
simply motivated by the need to feed and care for the family, but rests also on
the nature of the marital bond and the prestige associated with a particular style
of demonstrating one's dedication to the family. Not surprisingly, when a wife
of some means pays someone to come and help her discharge these duties, she
is said to contract 'household help' rather than household workers.
Work can help both to strengthen existing social relationships, and to
generate new ones. The farmer who helps his brother with the harvest through
892
WORK, THE DIVISION OF LABOUR AND CO-OPERATION
an exchange-of-hands agreement is reasserting a moral obligation, while at the

same time solving a harvest-labour problem and overcoming a cash constraint.
When he helps a neighbour under a similar arrangement he gives new meaning
to that neighbourly relationship. Simmel argued, in 1907, that a gift always
brings forth a counter-exchange which is constrained by morality (Simmel
1971:43-69). Mauss (1967), using examples from Melanesia and Polynesia,
suggested that it was because a gift contains some part of the spiritual essence
of the giver that the recipient is moved to make a return; a new relationship is
sealed and gifts and counter-gifts flow endlessly.
When a Colombian Paez Indian decides to clear a large field to plant his food
crops, he calls a minga or labour party, inviting his close kinsmen and intimate
friends. The work day ends in celebration, not of what has been accomplished,
but of what was ascertained: the communal bond of the participants. The feast
serves as an initial and only partial retribution for help received. It has to be
followed by the collaboration of the host in the future mingas called by his
guests; only then will the obligation be fully reciprocated and the host's social
debt cancelled. This cancellation, in turn, makes the initial guest indebted; he
is now the one who must reciprocate with a feast and return help. Prestations
and counter-prestations follow one another as a set of social debt and credit
relations, permanently linking friends and kinsmen to each other.
The central theme of the exchange is the significance of the obligation
shared by related families, rather than the quantity of food that is produced in
t IS or t at te an t e cost o pro ucmg It. T us, t ere IS no pomt m eepmg
track of how much work was done by each participant, nor of the cost oflabour
in beer, meat and tubers. Instead, the Paez farmer remembers who comes, and
who are the close friends and kinsmen who consistently, and without reason,
shirk invitations to mingas and other requests for help. In fact, all Paez agree
that a minga working party is neither the most efficient nor the cheapest way to
go about clearing a large field. However, if there is enough cane to prepare beer
and enough money to buy meat, the Paez farmer prefers to call a minga in order
to reassert his social commitments, while clearing a large enough field to assure
the survival of his family (Ortiz 1979). After all, if the crops fail to grow and not
enough food is produced to feed the family, the farmer will have no alternative
but to approach his kinsmen. It is significant that work parties are not used to
clear fields for commercially grown crops, for in the case of such crops, costs
and productivity are the key concerns-indeed it is only in these terms that
their cultivation can be justified.
Work is also an activity that can bring an individual closer to supernatural
deities. In the Polynesian island of Tikopia, the sago palm is not just a food-
producing tree but represents one of their gods. When a chief is about to cut
down a tree, a number of activities requiring some effort must be performed.
Because they are endowed with religious symbolism, we are inclined to regard
them as ritual activities. Yet the effort is similar to that involved in productive
activities. In fact, it is expected to yield two outcomes: the collaboration of the
893
SOCIAL LIFE
deity, and the successful production of starch. The intertwining of productive

pursuits and religious activities in Tikopia was so intense that Firth (1967a)
adopted the phrase 'work of the gods' to describe the religious cycle he
observed in that island in 1928, and again in 1952. The rites are obligatory, and
involve the production and presentation of food to gods and spirits, food that is
eventually put into circulation and consumed.
Cutting the trees, scraping the pith and processing the sago not only bring
Tikopian men and gods closer to one another but also, as with the Paez, bring a
work party of kinsmen together (Firth 1967b:269-83). This joint co-operative
effort is regarded as a social obligation: 'Anything done in the house of a
brother-in-law, I rise and go after it' (Firth 1967a: 147). The kinsman who
responds in such a manner is also mindful of the complexities of the social
bond and will be careful to acknowledge symbolically the important differences
of status and role. If the chief is the one who has invited others to work for him,
he must offer a more elaborate feast than a host of lower rank. If a senior
individual helps in the production of starch, he must receive a bowl of starch as
well as the customary amount of flour.
In some cases, the offering and receipt of help can be used to formalize
status differences. In England, large staffs of cooks, gardeners and cleaners
allowed the gentry to keep well-manicured gardens, entertain with great
formality and assume a style of life becoming to their social status. However,
just as significant was the demeanour adopted by the managing butler when
Isc argmg IS uties. He a to act m a Igm e manner, never revea mg IS
emotion or anxieties, and had to speak eloquently with a good accent on
appropriate topics. This professionalism was attributed to the squire, who was
thereby freed to act in a more relaxed manner. In other words, the work effort
of the butler allowed upper-class gentlemen to mark their class difference by a
particular style, without having overly to constrain their behaviour. The
butler's ability to enhance the status of his master put him in a class apart from
the manservants, housekeepers and domestic help hired by middle-class
housewives.
Relations forged while labouring are both complex and delicate. In the case
of the butler, he must distance himself from the squire whose status he
enhances, and from other members of the staff, in order not to demean the
symbolism of his pose. All social aspects of the work relation are brought to
bear on the forging of rules of conduct among eo-workers. In turn, the patterns
of conduct, as well as pervasive social conflicts, class antagonisms, or more
ramifying loyalties, are likely to affect performance. This is most striking in
rural areas and in small-scale societies, where there is little separation between
domestic tasks, farming tasks at home and work on the farms of neighbours and
kinsmen. Class differences are said to have a negative effect on the work
performance of field hands in England. Farm owners, concerned with
productivity, try to offset the impact of class differences by relating to their
workers as informally and as supportively as possible. They drink together in
894
pubs and work along with the labourers as often as they can. In this way, farm
owners foster the loyalty, commitment and deference of their workmen, and
attempt to transform the contractual tie into a socially meaningful work
relation (Newby 1977:303).
The social significance of work extends beyond the conveyance of symbolic
information. Work involves the transformation of nature into objects which
then become identified with the people who produced them. As Marx long ago
observed, it is through labour, through the transformation of nature into
commodities, that human beings define themselves (see also Miller in this
volume, Article 15). The exercise of care and ability, as demonstrated by the
quality of workmanship, further enhances their stature. Well-tended fields,
smooth and delicious cubes of uncentrifuged sugar and tightly woven cloth all
bring prestige to a Paez Indian and many of them are willing to give up leisure
time to gain the recognition attained through such products. Some of these
carefully manufactured craft objects become so identified with the producer
that it is not easy to convince them to part with them (Ortiz 1979). Even in
societies where crafts are readily sold in the market the identification of the
object with the person who made it or designed it is retained (Annis 1987).
Much has been written about the existential significance of work, how and
when it serves to integrate a person, to give him or her a sense of satisfaction
and identity, or alternatively leads to a sense of insignificance and alienation.
Chinoy's classical study of automobile workers (Chinoy 1965) has set some
By transforming nature through their work, individuals can establish a

claim over what they produce. In most pre-capitalist societies work serves to
establish rights of use over fields, trees or crops. Territories may belong to
certain clans, but individual members of a clan can carve out sections for
themselves and their families by clearing them and planting them with crops.
Although the recognition that work generates rights of use or disposition is
most common in pre-capitalist societies, it is not unknown in market
economies. In Colombia, individuals can claim rights over unused land by
clearing it and using two thirds of the claimed territory. The disastrous fate of
the forest of Amazonia is in part rooted in that fundamental notion, which has
received formal legal recognition in most Latin American countries.
The recognition that labour is a force that serves to link commodities to their
producers is probably what induced Ricardo (1911:5-18) and Marx (1965:48-
54) to think of labour as a value-endowing process. For example, they supposed
that in non-market economies, the value of the hunt could be measured by the
time spent tracking animals, plus the time spent making the spear used to kill
them (in proportion to the durability of the tool). Marx introduced the
qualification that this was likely to be the case only in economies in which
producers controlled the means of production, and in which capital
requirements were similar among industries. Morishima ( 1973) has added
895
SOCIAL LIFE
another condition: that the mobility of labour should not be impeded

sociologically or geographically.
Some anthropologists have adopted Marx's proposition that the exchange
value of a commodity rests on the labour expended in producing it, in order to
explain the rates at which commodities were exchanged, as well as the rigidity
of such rates, in pre-capitalist economies. However, in an analysis of the
exchange of salt blocks for bark capes among the Baruya of Highland New
Guinea, Godelier ( 1971) found that although there was a certain
correspondence between the rate of exchange and the proportional input of
labour required, on average, to manufacture these two kinds of goods, the
Baruya themselves justify the terms of this exchange with different arguments.
The request for a specific number of bark capes in exchange for a given
quantity of salt is initially backed by arguments of need. Only when the trading
partner remains unmoved does the Baruya salt producer bring up the issue of
the work required for the item offered in exchange. Godelier concludes that the
rate of exchange is determined not by the relative amounts of labour
'congealed' in the items exchanged, but by the relative social need of the
trading partners for these items ( Godelier, 1971:66-7; see also Gregory's
discussion of the Baruya case in Article 33). Cook (1976) has attempted a rather
similar type of analysis to determine whether the exchange value of metates
(grinding stones) produced by Zapotec peasants in Mexico is pegged-
allowing for seasonal fluctuations in supply and demand-around a value fixed
y t e a our cost o t etr pro uctwn. He cone u es t at t e metate pro ucers
are indeed concerned to ensure that they receive an appropriate compensation
for their labour. In other words, in this peasant economy, labour time is
adduced as one factor in assigning value to commodities, but only after
excluding certain costs (such as the time spent at the quarry to extract the
stone, and the time and money required to maintain and replace the required
tools).
Thus, the meanings of work and labour are multiple. Work is an activity that
can generate or strengthen social relations. As a social process it reflects and
symbolizes all other conditions: rank differences, role characteristics, kinship
obligations. On ritual occasions, work acquires a religious connotation. It is
because work evolves in this social context that it can serve to link the
individual to the produce of his or her labour, but only if that labour has not
been sold as a commodity to another party (the employer). In the latter
situation, for example in modern factory production, the relation that
individuals retain with the commodities they produce is extremely tenuous.
The meaning of their work then shifts from the items produced to the status of
their job and to social relations within the factory.
More importantly, work in industry and commercial agriculture generates
capital for the owner of the means of production or for the buyer of the labour.
This is the meaning of work to which economists draw most attention. It is of
course an important meaning. Yet, since the production of capital is affected by
896
the social context of labour, it is wrong to disregard the other meanings of work
alluded to in this section. To do so would render our explanations incomplete,
and our arguments only partial.
None of these meanings is intrinsic to work. Rather, the significance of work
is drawn from, and reflects, the social contexts of production and exchange. An
individual who sells his labour-power enters into a contract. It is the contract,
and the socio-economic relations implied by it, that gives to work its particular
tenor. Moreover, all of these diverse meanings are learned: workers must be
socialized, for example, into the experience of factory employment. Once
socialization is achieved, the experience becomes generalized and is often
reified as a code of work ethics and of job characterizations.
MEASURING LABOUR AND WORK

These different meanings and significances are sometimes conveyed by using
separate terms. Domestic helpers may be referred to as domestics, servants,
cleaning women, household helpers, or housewives, rather than-say-home
labourers. Servants are said to produce 'services' instead of capital. Kinsmen in
a working party are known as 'guests' rather than 'workers'.
A Paez farmer distinguishes linguistically between the various social and
economic meanings of his work effort. He describes his participation in a minga
as 'accepting an invitation', and his several hours' walk in taking salt to his
catt e-raise as an msurance agamst starvatiOn-as a VISit' to t e amma .
When a Paez sells his energy for money, however, he 'works'. He also 'works',
as does his wife and children, when clearing his field and planting his crops.
Work is seen as what generates capital and sustenance. Mingas, of course, also
generate sustenance, but more importantly, they generate sociability and future
work resources (Ortiz 1979).
The context of the work experience affects the way that experience is
recalled. A farmer who has helped a kinsman does not keep track of the number
of hours spent at each task; when he returns home, he thinks about his
relationship to his kinsman. Likewise, his collaboration is remembered not with
a figure but as an affective gesture. To keep an exact account of time
expenditure may signal a disinterest in the social obligation. Loathing to convey
such a message, the farmer may undervalue or even overvalue the extent of his
contribution. When however the work is provided under contract and a wage is
paid, farmers are more likely to keep track of time spent at a job, and to evaluate
financially the wage received.
I noticed that Paez farmers did not include, in their estimates of labour
time expenditure, the contribution of guests at a minga, or the time spent
visiting their cattle. They can give some very rough estimates of work
requirements for a field planted with subsistence crops, since they are
labouring at this task. But this work effort is clouded by the social
significance of subsistence crops, and a strict accounting of time spent cannot
897
SOCIAL LIFE
convey what the effort is all about: the ability to care for a family. This ability,
in the view of the farmers themselves, is more clearly demonstrated by
bountiful dinners and feasts, than by individual expenditure of effort. Hence,
time accounting seems to them to be somewhat irrelevant. This finding
accords with those cited earlier, concerning the relative inaccuracy, in terms
of labour costs, of the evaluation of metates by Zapotec stone workers and of
salt blocks by Baruya. Time spent working becomes more crucial when the
effort is vested in planting cash crops, because the question then present in
the mind of a Paez farmer is whether the alternative of wage work may not be
a better solution to the problem of meeting his cash needs.
The introduction of time accounting (see Adam in this volume, Article 18)
does not automatically follow from the commercialization of labour, nor from
industrialization. 'Rather, time in relation to work has been continuously
shaped, defined and contested by workers and employers in the context of
changing structural pressures contained within the spheres of production and
social reproduction' (Whipp, 1987:211 ). Even as late as 1940, accounts of
labour costs in pottery production were not kept by most factories in Britain.
Instead of using time to regulate labour costs, the factory owner divided the
production process into tasks and paid by task completed. Whipp quotes a
moulder as saying: 'We have no set time for stopping and starting here, that is
in regard to moulded work, should any job be given out, a piece-work rate is at
once fixed on. So the Boss troubles no more about one's comings and goings'
W tpp 987:22 . In t IS way pottery emp oyees cou mtx omesttc wor
with productive work. The first was called play, the second work.
These uneven patterns of labour-time accounting in farming and industry
undermine the economists' attempts to evaluate the relative significance of
different factors of production. They can represent labour inputs by payments
made for labour, but only when the payments correspond to the quantity and
quality of effort vested in the production process.
Anthropologists should be even more careful when gathering estimates of
labour input in non-market economies. Data cannot be obtained through recall
of past events, but must instead be carefully adduced from daily observations
and time budgets. The field observer should not ask 'what work did the
informant carry out on that day?' but 'what did she or he do on that day?' In this
way one can avoid the risk of failing to attend to tasks that may be classed,
whether in the conventions of the observer or in those of the local people, as
other than work Oohnson 1975).
DIVISION OF LABOUR
In all societies there are some tasks that require training and experience. But
most other tasks could, in principle, be performed by just about anyone. Yet in
practice this is rarely the case. Tasks tend to be categorized as appropriate for
certain sets of individuals rather than others. Young men are often the warriors,
898
women the ones to prepare food, and older men are the political managers. To
some extent one can account for specialization and the division of labour as
responses to technical requirements and time constraints.
For example, some activities are very time-consuming and need exclusive
attention in locations distant from where other activities must be performed.
This and other conflicting demands can be resolved by differential allocation
of responsibilities among individuals who recognize their social
interdependence and who share the proceeds of work. Thus, it is most often
within households composed of close kinsmen that one can note task
specialization and interdependence. It is understandable, within this setting,
that women who must nurse children, and who are thus more limited in their
mobility, should take on tasks that can be performed within or near the
household. Tending the fire, processing staples, caring for domestic animals,
preparing the food and tilling nearby gardens are jobs often assigned to
women. These are tasks that can readily be combined with the responsibilities
of child care. Men can more easily assume responsibilities that take them
further away from their homes: going to war, hunting large animals that must
be followed for days, herding and trading in faraway territories. In fact,
women are likely to assume a major responsibility for agriculture in societies
where men are called away by war or must tend distant herds of livestock
(Ember 1983:297-99). This pattern is replicated when wage labour is
introduced and men must go away to work (Burton and White 1984:580). The
mtermtttent yet mtermma e at y tas o
left to children who have little else to do.
Long dry seasons, followed by short periods of rain, skew work rhythms in
ways that are difficult to manage. Agricultural tasks have to be completed
within short periods of time, while working very intensively. Men may be in a
better position to handle such work cycles (Burton and White 1984:579).
Alternatively, some of the agricultural activities that have to be completed
within a short period of time can be subdivided into tasks that are each
judiciously allocated to either men or women. By so doing, and by also taking
into account the domestic constraints on women, time pressures can be at
least partially resolved. When the Bemba have to clear a field before the rains
come, the men are assigned to pollarding the trees and cutting the
underbrush. The women pile the branches that men will later burn (Richards
1939:289-94).
The problems posed by uneven seasonal demands can also be resolved
through co-operation rather than specialization. The production of salt from
grasses, as practised by the Baruya of New Guinea, requires a number of
operations, some of which cannot be carried out by a single individual within
the short dry periods. The grass must be cut, transported, dried for two weeks,
and burned before the rains come. Baruya men with rights over grassland solve
their problem by enlisting the help of other kinsmen who lack such rights, or by
working co-operatively with other salt producers. Once the grass is burned, the
899
SOCIAL LIFE
ash residues have to be soaked and filtered, and the sediment extracted by heat-
induced evaporation of the water. This second stage requires much time but
only limited attention; it can be carried out by a single individual with free time
to spare-time he can make available to other salt producers. Altogether, the
production of one bar of salt requires 21 person-days, but this labour
requirement can be selectively distributed among a group of co-operating men
(Godelier 1971:55-8).
Work rhythms, however, are not simply dictated by the demands of growing
crops. Before cocoa farming was introduced into West Africa, important Beti
men were able to mobilize enough labour to clear large tracts ofland. They took
advantage of this opportunity to instigate not only an intensive period of land
preparation, but also a subsequent season of intensive planting. Women, clients
and junior men who lacked access to a large labour force had to content
themselves with cultivating land previously used; they also had to avoid
practices requiring peak labour periods (Guyer 1988:256). Even today, Beti
women are limited to growing crops that can be combined in such a way that
periods requiring intensive inputs oflabour are avoided (Guyer 1984:381).
Thus technological arguments, though relevant, cannot fully explain
seasonal work rhythms, nor can they explain why certain tasks, in specific
societies, are performed by women. If it is more rational and efficient for
women to concentrate on jobs that can be performed near their dwellings,
why is it that in New Guinea their gardens are often so far from the village?
ccor mg to a IS ury 9 2: 9- 2 , t e men go to t eu te s every t u
day and spend about four hours a week travelling, while women go every day
and spend twelve hours a week on their return trips. Technological
arguments also fail to explain why, among the Baruya of New Guinea, women
are allowed to help with the cutting and piling of the grass, yet they are
excluded from helping in the collective task of wrapping the salt bars. The
explanation, in this case, is cultural: women can pollute salt bars (Godelier
1971:56). Carlstein (1982:339) has depicted diagrammatically the daily
routine of members of a Gusii agropastoralist household in East Africa,
illustrating the imbalances in the time invested by husband, wives, daughters
and sons in their respective tasks, and the locations of such tasks (Figure 1).
Men spend more time around their dwellings than women do. Johnson
(197 5: 639) estimated that among the Machiguenga-slash-and-burn
horticulturalists in Peru-men spend hours daily in agriculture and foraging
and about 15 to 20 minutes helping with childcare and cooking. When other
productive tasks are included, men work a total of 6 hours per day. Women
spend more time on food processing and childcare, and less time in
agricultural and gathering activities, adding up to a total of 7.3 hours of work
per day.
To understand how labour is allocated among members of a household, and
how these allocations correspond to social categories (males, females, parents,
adolescents and younger children), one must consider the social as well as the
900
Time
(hours)
24-
--,
I
I
22- I
1 -Attending
beer part)
Sleeping- 1
I
20- ____ j
r
Eating- I
I
18- - L- -------y - - --Dusk
I
I -Local
.......................................... I litigation
16-
I
-grazing r ____ J
14-
cattle
I.-'·'-------------'
1- Yisiting
neighbours
············· ..L:
Grinding corn - [ Watering ~~ttl~ ·.:: ~ I
at power mill _I
12- .......! ·rl -Eating
:····························~·:..:_··.:.:.·:.:·:..:.; ·:· -Milking co" s
-grazing I -work in
10- cattle I coffee
I garden .
Cultivation- ............................1................. •..
I ::
8- I
L ___ l
I ::
6- I ; : -Eating
---- I::_- --Dawn
Fetching water- -.---=-=J ~ j
Elimination- ~;;
4- 1 ..
Sleeping - 1 ..
I ::
I
2- I"
Man - ;.......... Daughter
1
Woman--t ~ Son j--
1 ..
_ _ _ _ 1::
0-
Work Service Market Yard or Own Place of Others School
place place place garden dwelling meeting d11 elling
Figure 1 A reconstruction of the daily round of men, women and children among the
Gusii of Kenya (from Carlstein 1982:339).
SOCIAL LIFE
technical aspects of work. Since work can generate power and prestige,
establish relationships and symbolize status, the allocation of tasks is likely to
reflect both the prevailing social organization and the politico-economic
context within which the labour process unfolds.
If only some people are allowed, or are in a position, to produce an item
which is basic for survival, others must become dependent on those people.
When the goods produced can be used to gain other assets, the control so
conferred can yield significant political and economic power. Among the Lele
of West Africa, the elders reserve the right to produce the decorated bark cloth
that circulates as part of bridewealth payments (Douglas, 1963:54-61 ). When
new cash crops are introduced, it is often the men who reserve to themselves
the right to plant them, and by so doing they retain control over capital assets,
thus ensuring their own economic ascendancy.
However, the planting of staples can only bring power to producers when
they control the means of production. For example, in New Guinea, it is the
women who produce the yams. Their specialization in this activity has not,
however, enhanced their position because, while controlling the production of a
basic staple, they remain dependent on their husbands for the land on which to
plant the yams and for the clearing of their fields (Salisbury 1962:46).
For power to be gained through control over the production process, the
producer must not only control the means of production, but also be in
command of his or her own efforts. A Paez woman in Colombia may gain
const era e prestige y pro ucmg a we - estgne an ttg t y woven ruana or
poncho; in theory she could also make non-weavers dependent on her effort.
But before a wife sets a loom she must ask for her husband's permission. He
has a prior claim over her labour and may prefer that she works in the family
field, rather than remain at home weaving a blanket. More importantly, and
because he has a claim over her labour, the cloth she produces, even when she
uses the wool of her own sheep, has to be committed to a family member. Her
bargaining powers are, thus, limited.
The power and status that may potentially be achieved through
specialization can also be contained by controlling the distribution of
commodities. In Guatemala, women are often the weavers of blankets, but
seldom do they sell what they themselves produce. The explanation given is
that the markets, and tourist craft buyers, are too far away (Bossen 1984 ). The
question remains as to whether men are not also trying to link themselves to a
work activity that is profitable and that might, if left totally in the hands of
women, allow them to build considerable bargaining power within the domestic
circle. Thus power can only be gained by specialists when they have free access
to the means of production or to capital assets, total claim to their own labour
and the freedom to distribute what they produce.
Since producers are often identified with the fruits of their labours,
specialists can gain stature, as well as power, when their products are significant
luxuries or prestigious staples. In India, activities are ranked according to a
902
scale of religious purity, with professions at the top of the list and defiling
activities, like those of street sweepers, at the bottom-restricted to the lowest
untouchable castes. Any activity connected with dead things or bodily
emissions pollutes, and is reserved for the lowest castes and subcastes. But
other, less polluting occupations are also evaluated according to how degrading
it is to perform them and hence whether they should be avoided by higher-
ranked castes. Occupations thus serve as indices of the social rank of castes.
A more subtle illustration of occupational stratification is to be found with
the emergence of guilds and the nineteenth-century artisan societies. Elaborate
technologies do of course account for the institution of apprenticeship, through
which novices were enabled to acquire the required skills. However, the
training was often overly protracted in order both to limit entry and to enhance
the prestige of the trade. The Amalgamated Society of Engineers insisted that
for a man to be recognized as a member of the trade, he had to serve a five-year
apprenticeship. The knowledge thus acquired became the artisan's personal
capital, marking him as an adult, a special individual in society and a member
of a prestigious social group (McClelland 1987:191 ). Prestige rested not only
on the importance of the trade, but also on a well established code of behaviour.
A shipwright who came to work on Mondays unshaven and wearing a dirty
shirt was likely to be fined (McClelland 1987:193). A labourer who worked
alongside a mason was often physically chastised by being struck hard with the
mason's leather apron, as a way of symbolically humiliating him (Rule
987: 09 . In act, most tra es were ran e m terms o onoura e' or
'dishonourable' they were.
Because, as we have seen, work is an activity that links individuals to
commodities, services, communities and gods, and because it can create
relations of debt and power, it is a social as much as a technological process.
Hence, the division of labour must always be closely connected with political
structures and must lie at the heart of any process of stratification. Indeed,
Godelier argues that the division of labour emerges alongside social hierarchies
rather than being the prior cause of them (Godelier 1980).
With this in mind, we should re-examine Durkheim's visionary portrayal of
the evolution of societies from those of a simple and undifferentiated kind, held
together by shared values that were routinely obeyed, ritually reinforced and
communally sanctioned, to larger and more secular societies where
specialization prevailed, fostering social interdependence. He characterized the
former as integrated through mechanical solidarity, and the latter as integrated
through organic solidarity (Durkheim, 1933: 129-32). In Tikopia, for example,
the ritual cycle brings each clan together and links it to the three other clans.
Likewise, the feasts that end the labour parties of the Colombian Indians help
to reassert the significance of social bonds. In these and many other small-scale
societies, solidarity rests in the routinized performance of communal acts.
Once specialization makes interdependence a more constant and pervasive
reality, the ritual re-enactment of solidarity becomes less crucial. Societies
903
SOCIAL LIFE
become secularized, new codes of conduct are forged, and new sanctions are
established to ensure the smooth operation of society. The division of labour,
Durkheim argued, allows individuals to strive for equal opportunity, to have
greater autonomy and to be constrained not so much by tradition as by
contractual obligation (Durkheim 1933: 147-56).
Inter-caste exchanges illustrate how specialization can bring
interdependence and solidarity, much as Durkheim suggested. However, they
also highlight another point made by Durkheim: that specialization can lead to
hierarchical and exploitative social arrangements, and that religion can buttress
some of these unbalanced relationship. The exchanges of rice and services
between carpenters and landlord, for example, are fixed by tradition. At times,
these exchanges may be balanced, but the fact is that the carpenter cannot hold
land and that the landlord, who is a member of the dominant caste, controls the
distribution of rice and services. This is a right that is sanctioned by religion
and renders the landlord more powerful, despite his dependence on a number
of specialists (Dumont 1972: 138-48). It is important to note that, in this
example, it is not so much individuals who become interdependent as social
groups, which remain in such a position generation after generation.
It becomes clear that the emerging division of labour can be accompanied by,
and can consolidate, class stratification. Durkheim himself drew attention to
this in his admission that neither America nor Western Europe had achieved
true organic solidarity. He characterized these as transitional capitalist
societies, m w IC wor ers w o wou ot erwise e unemp oye are o ten
forced to enter into contractual agreements that benefit only industrialists, and
in which a state of anomie prevails. Only in societies that offer equal
opportunities to all, where entry to certain occupations is determined by ability
and capacity rather than by birth, can a division of labour generate organic
solidarity. In these (ideal) societies no hidden power structures determine how
people use their time and what occupation they can practise. The state is
supposed to protect individual freedoms and to ensure access to education and
other important resources for the acquisition of skills.
The division of labour has not yet given us the Utopian society that
Durkheim wished for. If anything, the consequences of this social process have
been more accurately described by Marx. Specialization, when one class retains
control over the means of production and circulation, often leads to
domination. Short of remaining unemployed, the factory worker has no choice
but to abide by the contractual terms imposed by the capitalist employer. His
occupation is specified by the technical engineer who designs the machinery
that he will operate, and is manipulated by the foreman who allocates tasks and
schedules times. Concerns other than those of mechanical efficiency are also
present in the minds of these designers and managers. These are concerns
about how to hold down the costs of production, either by controlling wage
payments, or by increasing the intensity of the labour process, or by
minimizing the possibilities of industrial action. Reorganizations of the
904
division of labour often reflect management concerns as well as technical

problems; together they deeply affect the structure of wage labour markets.
THE DIVISION OF LABOUR AND THE STRUCTURE

OF LABOUR MARKETS
Durkheim was correct in pointing out that the division of labour became more
pervasive with the expansions of capitalism. For it is this that locks a large
sector of the population into an exchange of labour for wages and of money for
subsistence needs. A new set of social classes emerged against the background
of already existing social hierarchies, and within each a shared consciousness
began to be formulated. The division of labour took on a strongly political
dimension.
It also became more spatially defined. Work was removed from the domestic
circle, creating, in factories and plantations, new social relations of production.
The differentiation between labour and leisure began to be acknowledged more
categorically. Clock time started to regulate labour, taking away from the
individual the management of his or her own work effort. Labour, more often
than not, began to be sold by units of clock time rather than by tasks completed
(Thompson 1967, Adam, this volume, Article 18). Legal contracts emerged as
the mode for negotiating the sale and exchange of labour. These
transformations in the organization of production thus began to create new
tvtswns o a our: etween JO s m t e amt y t at are unremunerate , wage
jobs that may vary according to what is being produced, manual jobs, and
administrative jobs.
Analysts of Marxist persuasion point to another parallel development with
industrialization: de-skilling, the degradation of work and the resulting
homogenization of the labour force (Braverman 1974 ). This transformation of
the production process became possible, it is argued, with technification. It
represents a counter-move against the division of labour that had been fostered
by the system of craft guilds. Artisans hoping to safeguard their interests had
drawn sharp lines around their professions, tightening entry conditions and
requiring ever lengthier apprenticeship. Capitalists, who had to rely on skilled
workers for a large part of the production process, were at the mercy of the
regulations that limited the supply of competent labourers and augmented the
cost of their labour. De-skilling has to be considered as part of the
rationalization process employed by capitalists to reduce the costs of
production. Once achieved, capitalist producers could tap much cheaper
sources of labour: the unskilled, women and children. Alternatively, they could
bypass the skills of the artisans by simply allowing the production of cruder
produce (Rule 1987:101).
It was partly through de-skilling that capitalists were able to gain control
over the labour process and open up the labour market to a vast sector of the
population. Gordon et al. (1982) have rephrased the same historical trajectory,
905
SOCIAL LIFE
adding that only certain industries had enough available capital to achieve
almost complete de-skilling. Eventually, these industries were the only ones
that could advance technologically and thereby increase productivity and
returns so as to be able to offer better wages. Other enterprises, constrained by
access to capital, were unable to achieve high productivity through
mechanization, and had to retain a labour force with lower wages. They had to
rely on a supply of workers who were socially rejected by the core industries:
above all, women and members of ethnic minorities. This process of
differentiation differs from the division of labour in that it does not categorize
prestige, quality of work and payment by the type of task, but by the type of
firm and the social type oflabourer it prefers to recruit. It has been given a new
label: the segmentation oflabour markets.
Neoclassical economists countered that there is no real segmentation of the
labour market. In free-market economies, specialization is a rational response
to both technical problems and management issues. Jobs are open to anyone
who is qualified, and remuneration is based on the cost of acquiring the
required skills, productivity and the supply of labour. The division of labour
within the market sector, they argued, is based solely on an open network for
achievement and promotion (Cain 1975).
While this may be the case for some sectors of the labour market, it is hard to
believe that social processes of differentiation, empowerment and stratification,
which had operated until the emergence of industrial society, would have
su en y cease to m uence t e orgamzatwn o t e economy. T e neoc asstca
argument also fails to explain certain historical patterns of recruitment and
wage rates. When the Dutch engine loom was introduced by large
manufacturers in 1815, only men were hired to operate them, relegating women
weavers to the more unskilled tasks of operating the single-hand looms. Neither
technology, nor mechanization, nor skill could explain this segmentation of the
labour market along gender lines. Women had worked in the weaving industry
and would have probably been perfectly able to work with the new looms, had
they been allowed to operate them. What probably kept women from these
looms was a combination of cultural notions about the kinds of work that are
appropriate to different categories of people (in this case, men and women), and
the wish to prevent the empowerment of women through their admission to
relatively prestigious and financially rewarding occupations. In fact women, off
and on, had alternately participated in, and then again been barred from, the
weaving trade in Spitalfield (Berg 1987). When they did participate in trades
along with men, as in nail manufacturing, they received much lower wages.
When women and ethnic minorities consistently concentrate on more marginal
industries, we may expect to find causes for this other than the requirements of
training, competence and scheduling. Given what has been said about the social
significance of work, it seems much more plausible that this segmentation is
akin to a gender division of labour.
More conscious subsequent resegmentations of the market, targeting
906
particular categories of individuals for newly defined skills, have been

associated with strategic attempts by farmers and industrialists to subvert
protest, to control the labour force, and to lower wages. Redefining certain tasks
as requiring experience and skill that are gender-specific, contractors can
control the recruitment process to avoid labour conflicts, and to influence pay
scales. By praising the dexterity of women as lettuce wrappers, lettuce growers
in California, fearful of the spread of unions, targeted a population of
supposedly less troublesome workers as the suppliers of labour for their crews.
The task of cutting, leafing, wrapping and crating the lettuce in the field used
to be done by an undifferentiated crew of individuals. When the process
became partly mechanized, men, women and machines had to move in unison,
but since the job was redefined as consisting of three different tasks, a different
type of individual could be hired for each. By assigning different modes and
levels of remuneration to each task, the lettuce-growers were able to lower the
costs of production. While migrant male cutters and packers received a piece-
rate payment, the women who followed the cutters were paid by the hour.
Thomas (1985) calculates that the wages thus earned by women were much
lower than those earned by men, though the skill required did not differ.
By ranking occupations in terms of prestige, growers also affect the sources
of labour. Whether intentionally or not, farmers often downgrade certain tasks,
allowing entrepreneurial contractors to set a cap on wages and to seek out
sources of supply willing to provide it: usually the most vulnerable sectors of
t e popu atwn. F on a cane growers mstst t at mencans are not use to ar
work and would not apply for cane-cutting jobs. With this argument in hand
they petition for special immigration licences to bring in foreign labourers for a
season at a time (Martin 1988). In fact, crops seem to be ranked by a scale of
prestige, and workers are loath to move from the 'ladder crops' to the less
esteemed jobs in cotton, grapes and potatoes, or in the still lower status crops
that require stooping. With industrialization in manufacture and agriculture we
find a new redivision of tasks that combines with other processes of gender,
ethnic and social differentiation, once again to generate closed occupational
groups. This is a response on the part of management to the escalating costs of
production, the demands from workers who are unionized, and the difficulty in
asserting a work ethic that is profitable to capital.
With the spread of wage work there is also a shift in the way that the labour
relationship is defined, from the exchange of labour as a gift, or as part of a
social obligation, to exchange with remuneration. The shift was gradual as it
entailed the sanctioning of contracts, and the freedom of both parties to initiate
or terminate them. In England, labourers did not gain such freedom until late
in the nineteenth century. Labour relations then became bureaucratized, the
traditional base of pre-existing relations between labourers and employers was
weakened and the force of the supply-demand tension began to be felt in full.
Only then did the new forms of redivision of labour, alluded to above, begin to
take shape.
907
SOCIAL LIFE
Although in principle this shift has now been completed, a large part of the
agricultural produce consumed, or of manufactured goods purchased, in
modern societies is nevertheless not produced by individuals who work in
plantations, workshops or factories, and who enter freely into clearly defined
contractual agreements. Much food is still produced by self-employed farmers,
or sharecroppers, or labourers who offer their services either to repay a loan or
in exchange for access to land. Although some of these arrangements are not
necessarily exploitative, they are not always explicitly negotiated. Robertson
(1987), in a survey of sharecropping contracts in Africa, points out that the
terms of the contracts are often implicit, and are negotiated over a number of
occasions. Many informants, upon hearing Robertson's summary of his
findings, were struck by the unfairness of some of the contracts into which they
themselves had entered, and were inspired to drive a harder bargain on future
occasions. In the highly commercialized sector of Colombian coffee agriculture,
many labourers are not at present certain of what pay they are going to receive
at the end of the week. When negotiations are not open, wages cannot reflect
market conditions. They are more likely to reflect other capital costs of the
enterprise and the landlord's or manufacturer's perception of what is an
appropriate standard of living for the labourer. Such perceptions are coloured
by class and ethnic differences and are likely to perpetuate them. When
contractual agreements are unclear and negotiations not bureaucratized, a
labour market may be more readily segmented along social lines than along
tas -spect c mes. T us t e nature o t
nature of the wage negotiating process.
REFERENCES
Annis, S. (1987) God and Production in a Guatemalan Town, Austin, Tex.: University of
Texas Press.
Bossen, L. (1984) The Redivision of Labour: Women and Economic Choice in Four
Guatemalan Communities, Albany: State University of New York Press.
Berg, M. (1987) 'Women's work, mechanisation and the early phases of
industrialisation in England', in P.Joyce (ed.) The Historical Meanings of Work,
Braverman, H. (1974) Labour and Monopoly Capital, New York: Monthly Review Press.
Burton, M. and White, D.R. (1984) 'Sexual division oflabor in agriculture', American
Anthropologist 86:568-83.
Cain, G.G. (1975) 'The challenge of dual and radical theories of the labor market to
orthodox theory', American Economic Review 65:16--22.
Carlstein, T. (1982) Time Resources, Society and Economy, vol. 1, Pre-industrial Societies,
London: Alien & Unwin.
Chinoy, E. (1965) The Automobile Worker and the American Dream, Garden City, N.Y.:
Doubleday.
Cook, S. (1976) 'Value, price and simple commodity production: Zapotec stone
workers', Journal of Peasant Studies 3:395-428.
908
Douglas, M. (1963) The Le le of Kasai, London: Oxford University Press.

Dumont, L. (1972) Homo Hierarchies, London: Palladin.
Durkheim, E. (1933) The Division of Labour in Society, New York: Free Press.
Ember, C. (1983) 'The relative decline in women's contribution to agriculture with
intensification', American Anthropologist 86:285-304.
Firth, R. (1967a) The Work of the Gods in Tikopia, London School of Economics,
Monographs in Social Anthropology, vols 1 and 2, London: Athlone Press.
--(1967b) Tikopia Ritual and Belief, London: Alien & Unwin.
Godelier, M. (1971) 'Salt currency and the circulation of commodities among the
Baruya of New Guinea', in G.Dalton (ed.) Studies in Economic Anthropology,
Anthropological Studies no. 7, Washington DC: American Anthropological
Association.
--(1980) 'Work and its representation: a research proposal', History Workshop 10:
164--74.
Gordon, D., Edwards, R. and Reich, M. (1982) Segmented Work, Divided Workers, The
Historical Transformation of Labor in the United States, Cambridge: Cambridge
University Press.
Guyer, J. (1984) 'Naturalism in models of African production', Man (N.S.) 19: 371-88.
--(1988) 'The multiplication of labour: historical methods in the study of gender
and agricultural change in modern Africa', Current Anthropology 29:247-72.
Johnson, A. (1975) 'Time allocation in a Machiguenga community', Ethnology 14: 301-
10.
Martin, P.L. (1988) Harvest of Confusion, Migrant Workers in US. Agriculture, Boulder,
Colo.: Westview Press.
arx, ew
York: Random House.
Mauss, M. (1967 [1925]) The Gift, Forms and Functions of Exchange in Archaic Societies,
New York: W.WNorton.
McClelland, K. (1987) 'Time to work and time to live: some aspects of work and the re-
formation of class in Britain, 1850-1880', in P.Joyce (ed.) The Historical Meanings of
Work, Cambridge: Cambridge University Press.
Morishima, M. (1973) Marx's Economics: a Dual Theory of Value and Growth,
Newby, H. (1977) The Deferential Worker, Harmondsworth: Penguin.
Ortiz, S. (1979) 'The estimation of work: labour and value amongst Paez farmers', in
S.Wallman (ed.) Social Anthropology of Work, London: Academic Press.
Ricardo, D. (1911) The Principles of Political Economy and Taxation, Dutton, NY:
Everyman's Library.
Richards, A. (1939) Land, Labour and Diet in Northern Rhodesia, Oxford: Oxford
University Press.
Robertson, A.F. (1987) The Dynamics of Productive Relationships: African Share
Contracts in Comparative Perspective, Cambridge: Cambridge University Press.
Rule, J. (1987) 'The property of skill in the period of manufacture', in P.Joyce (ed.) The
Historical Meanings of Work, Cambridge: Cambridge University Press.
Salisbury, R. (1962) From Stone to Steel; Economic Consequences of Technological Change
in New Guinea, Cambridge: Cambridge University Press.
Simmel, G. (1971) On Individuality and Social Forms: Selected Writings, ed. D.N.
Levine, Chicago: University of Chicago Press.
909
SOCIAL LIFE
Smith, A. (1982 [1776]) The Wealth of Nations, Harmondsworth: Penguin.

Thomas, R.J. (1985) Citizenship, Gender and Work: Social Organization of Industrial
Agriculture, Berkeley: University of California Press.
Thompson, E.P. (1967) 'Time, work discipline, and industrial capitalism', Past and
Present 38:56-97.
Whipp, R. (1987) "'A time to every purpose": an essay on time and work', in P.Joyce
(ed.) The Historical Meanings of Work, Cambridge: Cambridge University Press.
FURTHER READING
Bossen, L. (1984) The Redivision of Labour: Women and Economic Choice in Four
Guatemalan Communities, Albany: State University of New York Press.
Braverman, H. (1974) Labour and Monopoly Capital: The Degradation of Work in the
Twentieth Century, New York: Monthly Review Press.
Burawoy, M. (1979) Manufocturing Consent: Changes in the Labour Process Under
Monopoly Capitalism, Chicago: University of Chicago Press.
Burton, M.L. and White, D. (1984) 'Sexual division of labor in agriculture', American
Anthropologist 86:568-83.
Gill, L. (1990) 'Painted faces: conflict and ambiguity in domestic servant-employer
relations in La Paz, 1930-1988', Latin American Research Review 25:119-36.
Guyer, J. (1984) 'Naturalism in models of African production', Man (N.S.) 19: 371-88.
Joyce, P. (ed.) (1987) The Historical Meanings of Work, Cambridge: Cambridge
University Press.
Kahn .S. 1981 'Mercantilism and the emer ence of servile labour in colonial
Indonesia', in J.S.Kahn and R.Llobera (eds) The Anthropology of Pre-Capitalist
Societies, London: Macmillan.
Newby, H. (1977) The Deferential Worker, Harmondsworth: Penguin.
Ortiz, S. (1992) 'Market, power and culture as agencies in the transformation of labor
contracts in agriculture', in S.Ortiz and S.Lees (eds) Economy as Process, Lanham,
Md.: University Press of America.
Pahl, R.E. (1984) Divisions of Labour, Oxford: Blackwell.
Richards, A. (1939) Land, Labour and Diet in Northern Rhodesia, London: Oxford
University Press.
Robertson, A.F. (1987) The Dynamics ofProductive Relations, African Share Contracts in
Comparative Perspective, Cambridge: Cambridge University Press.
Rueschemeyer, D. (1986) Power and the Divisions of Labour, Stanford, Calif.: Standford
University Press.
Scott, J.W. and Tilly, L.A. (1975) 'Women's work and the family in nineteenth century
Europe', Comparative Studies in Society and History 17:36-64.
Thomas, R.J. (1985) Citizenship, Gender and Work: Social Organization of Industrial
Agriculture, Berkeley: University of California Press.
Wallman, S. (1979) Social Anthropology of Work, London: Academic Press.
910
33
EXCHANGE AND RECIPROCITY

C.A. Gregory
The concepts of 'exchange' and 'reciprocity' are closely related. This much is
clear from the Oxford English Dictionary, which defines exchange as 'the action,
or act of, reciprocal giving and receiving', and reciprocity as 'mutual action,
influence, giving and taking'. Indeed the words are often used as synonyms.
However, in the anthropological literature over the past century the term
'reciprocity' has acquired a special meaning, and a distinction between
exchange and reciprocity of great theoretical importance has arisen. The
tstmctwn turns on ne 1 erences m meanmg etween t e wor s mutua tty',
'giving', 'receiving' and 'taking', and to understand these nuances it is
necessary to situate the anthropological theory of exchange in the broader
historical and theoretical context from which it has emerged.
The general theory of exchange is concerned with analysing acts of
exchanging things, people, blows, words, etc. Exchange is a 'total social
phenomenon', to use Mauss's (1990 [1925]:3) famous expression, and, as such,
its study involves the fields not only of economics but also of law, linguistics,
kinship and politics, among others. Most anthropological theorizing about
exchange, however, has been restricted to exchanges of wealth. But what is
'wealth'? The answers to this question fall into three broad categories. For
economists of the nineteenth century wealth consisted in commodities, whereas
for those of the twentieth, it consists in goods. Either way, it is stuff which is
valued by the market. For anthropologists, on the other hand, wealth consists
above all in gifts, products that are valued according to the non-market
principle of reciprocity. The notion of reciprocity, then, is at the heart of
theoretical debates concerning the distinction between market and non-market
forms of valuation. But ethnographers have also found the principle of
reciprocity operating in tribal trading systems and peasant markets, and these
findings have led to a revision of the theory of commodity exchange itself
To understand the anthropological concept of reciprocity, it is necessary to
compare and contrast economic and anthropological theories of the exchange
911
SOCIAL LIFE
of wealth. I propose to do this under the following five headings. Under the
first, 'Commodities as wealth', I briefly discuss the nineteenth-century political
economy approach to exchange. This is followed, under the heading, 'Goods as
wealth', by a discussion of twentieth-century economic theories of exchange. In
the third section, 'Gifts as wealth', I provide an overview of the anthropological
notion of reciprocity. This is followed by a discussion of 'Barter and other
forms of counter-trade', in which I introduce some anthropological revisions to
the theory of the commodity. In the final section, on 'Market-place trade', I
show how anthropological work also requires certain revisions to the theory of
market exchange.
COMMODITIES AS WEALTH
'The wealth of those societies in which the capitalist mode of production
prevails', Marx (1954 [1867]:43) declares in the first sentence of Capital,
'presents itself as an immense accumulation of commodities.' This notion of
wealth was part of the conventional orthodoxy of eighteenth- and early
nineteenth-century European thought, and all the leading theorists of the
time-Quesnay, Smith, Ricardo-developed their particular conceptions of
the principles regulating the market within this general paradigm. The truly
radical break came in the 1870s with the development of the theory of goods
and, with it, a new set of answers to the fundamental questions of market
exc ange: W at IS pro t. W at etermmes re attve pnces. W at etermmes
the level of wages? In this section I present, in very general terms, the answers
developed within the commodity-theory paradigm; in the next section these
will be set in contrast to the answers given by the goods-theory paradigm.
The notion of a commodity has its origins in the Aristotelian idea that a
product has two distinct values: a value in use and a value in exchange. Shoes,
for example, are useful because they protect one's feet when walking over rough
ground. This value is called 'use-value' and is quite distinct from the value
shown on the price-tag of a pair of shoes displayed in a shop window. The
price-tag value is called 'exchange-value', and it was value in this sense that
pre-twentieth-century commodity theorists sought to explain; the study of the
useful properties of objects, and that of the manner in which they satisfy
human wants, were regarded as falling outside the scope of political economy
(Marx 1954 [1867]:43). Within the overall commodity-theory paradigm many
different theories of exchange-value were formulated. Quesnay, the eighteenth-
century French physiocrat, found the answer in the natural productivity of
land; Adam Smith, the so-called father of economics, opposed this theory and
developed a labour theory of value in its place; this theory was developed, in
turn, by Ricardo and Marx, among others.
Marx's contribution to the theory of the commodity, though it hinged upon
the labour theory of value, gave it a new twist. Marx (1954 [1867]:53) claimed that
he was the first to point out that labour, too, possesses a use-value and an
912
exchange-value. To make shoes, for example, requires a certain quality of effort, a

particular kind of skilled technical practice. The use-value of the shoemaker's
labour lies in these qualitative aspects of his performance. As such, it differs from
the use-value of the labour of the farmer in growing wheat. If, however, we
'abstract out' the particular, technical qualities of the shoemaker's and the
farmer's labour, reducing both to their lowest common denominator, then they
may both be regarded as representing certain quantities of labour, which may be
measured and compared in terms of identical units of (chronological) time. As a
commodity measured in these terms, and exchangeable for other things, labour
has exchange-value. Now suppose that the shoemaker exchanges one pair of shoes
for, say, ten kilograms of wheat. In quality and hence in their respective use-
values, shoes and wheat are quite different. How come, then, that they are
equated in exchange? Marx's answer was that they are equated because an
identical amount of 'abstract human labour' was expended in the production,
respectively, of the pair of shoes and the ten kilos of wheat. Or more generally, the
equation in exchange of heterogeneous commodities comes about by virtue of the
equality in the amounts of homogeneous, abstract labour which they embody.
Marx distinguished between a number of different historical forms of the
commodity. Following Aristotle, he speculated that commodities emerged on
the boundaries of tribal communities where people would enter into
transactions with strangers and exchange products for which they had no use in
return for those which they desired. Following its birth in these marginal
regwns t e commo tty orm, Marx argue , egan to grow 1 e a ce ,
developing ever more complex manifestations as it divided and multiplied.
Initially, on the boundaries of the community, exchange took the form of the
barter of one commodity (C) directly for another (C-C). With the
development of peasant markets, and the need for a generalized medium of
exchange-namely money (M )-to facilitate trade, barter gave way to selling
(C-M) in order to buy (M-C), and the tribal community disappeared under
the corrosive influence of the new commodity form. The subsequent
development of mercantile capitalism-buying (M-C) in order to sell at a
profit (C-M')-and of moneylending at interest (M-M'), further eroded the
agrarian pre-capitalist society. Following a series of bloody struggles involving
the emerging capitalist class and the pre-capitalist peasants and landlords, a
new class of propertyless wage-labourers was born and the commodity
assumed its most generalized form, C=c+v+s, where c is constant capital, v
variable capital, and s surplus value, these values corresponding to the
quantities of abstract labour embodied in raw materials, wages and profit
respectively.
The historical prerequisite for the emergence of capitalism, and of the
C=c+v+s form of the commodity, was, according to Marx, the emergence of a
proletariat. With this, labour-power became a commodity with an exchange-
value like any other. But this historical development gave rise to a new problem:
What determines the exchange-value of labour-power?
913
SOCIAL LIFE
Marx's answer is long and involved but is based on a simple idea. He

observed that under feudalism the rate of exploitation, i.e. the ratio of surplus
labour to necessary labour, is expressed in a simple and direct form. Suppose,
for example, that a serf farms three days per week for his own benefit and that
he gives his landlord three days of labour per week as rent for the land he uses.
The former is called 'necessary labour' and the latter 'surplus labour'. This
relation, Marx argues, did not disappear with the development of capitalism; it
merely changed its form as rent was transformed, firstly, into a share of the
physical output (e.g. half the harvest) and then into money (e.g. half the money
profits or a fixed rent per acre of land used). With the emergence of labour-
power as a commodity, necessary labour assumed the abstract form of variable
capital (v), corresponding to the amount of money paid as wages, while the
profit over and above this amount-corresponding to the product of surplus
labour-was shared between the capitalist and the landlord.
Marx's immediate concern was to analyse the principles governing the
production, consumption, distribution and exchange of this most generalized
form of the commodity; pre-capitalist forms of commodity exchange were of
interest to him only to the extent that they illuminated the social preconditions
of the capitalist form. In any case, his understanding of the pre-capitalist and
non-commodity forms of exchange was severely constrained by the absence, at
the time, of reliable ethnographic and historical data.
Apart from the distinction between use-value and exchange-value, two other
e mng c aractensttcs o t e commo tty-t eory para tgm eserve mentiOn.
One of these concerns the prominence given to the notion of reproduction.
Exchange was not seen as an isolated act but as a phase in a reproductive cycle
consisting of successive acts of production, exchange and distribution. This
conception of the economy was first developed in 17 59 by Quesnay, in his
Tableau Economique (1962 [1759]), and it has provided the conceptual
framework for all discussions of commodity-value theory ever since. The
notion has been refined over the years, the most recent and logically
sophisticated being the version found in Sraffa's (1960) Production of
Commodities by Means of Commodities. As the title of this book suggests, a
commodity is both an output and an input, and Sraffa argues that these input-
output ratios, combined with a given distribution of income, determine the
exchange-values of commodities.
The other defining characteristic of the commodity-theory paradigm is the
focus on class relations. Even though some of the early commodity theorists
espoused naturalistic theories of wealth, they all addressed the problem of the
principles governing the distribution of surplus among competing classes.
Quesnay's theory, for example, distinguished three classes-the landlords, the
peasant farmers and the artisans-a division that captured the essence of the
social organization of the eighteenth-century French countryside; Ricardo
based his theory of value on the opposition between landlords and capitalists, a
key social conflict in the England of his time; and Marx, as is well known, based
914
his theory on the opposition between wage-labour and capital. These class
relations of production were seen to be crucial because they gave exchange
relations their particular form and content. In other words, the classical
political economists conceptualized exchange as an expression of underlying
power relations.
GOODS AS WEALTH
Following the marginalist revolution of the 1870s, which saw the fall of the
theory of commodities and the rise to dominance of the theory of goods,
exchange came to be seen as an expression of the subjective preferences of
individuals rather than of underlying power relations. This change in thinking
was reflected in a new concept of wealth. The marginalists, or 'neoclassicals', as
they are sometimes called, no longer saw the unit of wealth as a commodity but
as a good whose magnitude was measured by its subjectively attributed 'utility'.
In other words, the concept of a commodity, with its distinction between use-
value and exchange-value, was replaced by the concept of a good with an
undifferentiated utility value.
This new concept of wealth quite literally affected the way people viewed
the world. Emphasis was placed on consumption and scarcity rather than on
reproduction, and on choice and subjective preferences rather than on
objective class relations of production. The new paradigm also provided a novel
conceptua ramewor or posmg, an answenng, t e o questiOns concernmg
wages, prices and profit. This can be seen by examining, in a little more detail,
the notion of a good.
Despite appearances to the contrary, the word 'utility' does not mean the
same as use-value. Use-values refer to the objective properties of things and are
a function of the technological and scientific knowledge available to a society at
a given point in its history. For example, the discovery of photography in the
nineteenth century meant that silver acquired a new use-value to complement
its other uses as a store of value, as jewellery, as cutlery, and so on. Utility, on
the other hand, refers back to the subjective preferences of an individual
consumer. A cup of tea, for example, has positive utility to a thirsty person, but
that utility will be less for each additional cup consumed. Thus the marginal
utility of the tea declines, until the point is reached when the consumer's thirst
is quenched and she desires no more; at this point the tea ceases to be a 'good'
and, logically speaking, becomes a 'bad'-although this term is rarely used.
The utility of a good, then, derives from its ability to yield subjective
satisfaction. It refers to individual psychological feelings about scarce objects
and not to the objective properties of different things. As Robbins (1932:47) has
noted, 'Wealth is not wealth because of its substantial qualities. It is wealth
because it is scarce.' This conception of wealth is obviously very different from
its classical precursor. Among other things it contains the paradoxical
915
SOCIAL LIFE
implication that wealth, because it is the subjective sum of enjoyments, might

increase as material abundance declines (Heilbroner 1987:882).
This new theory of wealth opened up a new angle on ancient problems.
Consider the water-diamond paradox which was concerned with the following
question: Why does water have a high use-value and negligible exchange-value
while diamonds have a high exchange-value and negligible use-value? The
classical economists regarded this problem as peripheral and answered it in
terms of a highly technical theory of rent. The neoclassical theorists, on the
other hand, regarded it as central and made it the basis for an alternative theory
of price. Their answer, in a word, was scarcity. The marginal utility of water is
low because of its great abundance, the marginal utility of diamonds is high
because of their scarcity, and the ratio of the marginal utility of water to that of
diamonds determines their relative prices.
This proposition applies to the prices of all goods, according to an argument
that runs as follows. Suppose water became scarce because of a drought. In this
case its marginal utility would rise relative to the marginal utility of diamonds
and a price for water would emerge; if the drought was particularly severe the
price of water would rise even further and it may even become more valuable
than diamonds. Thus all exchanges, and hence prices, are expressions of
relative scarcity as manifested in the ratios of marginal utility.
The great appeal of this theory of exchange and price lies in its generality.
Indeed, it is difficult to imagine a form of exchange to which it cannot be
app te . Ta e, or examp e, t e we - nown exc ange o u a va ua es-
armbands and necklaces of polished shell-that are traded by the people of the
Trobriand Islands off the eastern tip of Papua New Guinea (described by
Malinowski 1922). Traders obviously prefer the scarce, high-ranking shells to
the relatively abundant, low-ranking shells. Thus it is possible to conceptualize
kula exchange in terms of marginal utilities. Furthermore, it could be argued
that the paradoxical conception of wealth contained in the theory of goods is
just what is needed to explain the notorious destruction of wealth that occurs in
the potlatch ceremonies of the Kwakiutl and other indigenous peoples of the
Northwest Coast of North America (Codere 1950). In the face of a super-
abundance of material goods, their worth is effectively eroded.
It is not without some justification, then, that Jevons, one of the founding
fathers of marginalism, could claim in 1871 that the science of economics is in
some degree peculiar, owing to the fact 'that its ultimate laws are known to us
immediately by intuition', and that from the notion of utility it is possible to
reason deductively to theories of value and exchange Oevons 1970 [1871]: 18).
The leading figures in the theory-of-goods paradigm today-Nobel
prizewinners such as Samuelson (1947), Debreu (1959) and Friedman (1962)
-have all developed highly complicated theories of value by reasoning
deductively in this way.
Like the theory of commodities, the theory of goods has many internal
divisions. However, these are mere dialectal variations of the one language. The
916
language of the theory of goods shares a common grammar and lexicon which
have nothing in common with those of the language of the theory of
commodities. The terms 'good' and 'commodity' epitomize this difference, the
etymology of the former suggesting a subjective approach to value, the latter an
objective approach.
The new language in which economists began to talk around the 1870s can
be likened to Esperanto, and the old language of the commodity theorists to,
say, German. This is not to say that one is better than another. Indeed, there is
no meta-theory by which they can be compared and evaluated. The two
paradigms have different consequences for understanding human life which
can only be evaluated in specific contexts. This implies some notion of
adequacy in relation to practical aims. To pursue the language analogy, do we
try to overcome the communication problems of the world by teaching people
Esperanto or do we try to learn some particular languages in order to develop
our general ideas from a comparative analysis of them? Needless to say,
anthropologists have tended to find the latter path more attractive, and few
have embraced the Esperanto of the theory of goods.
In this sense, then, anthropologists took up the implicit questions left
unanswered by the commodity theorists: What, in positive terms, does non-
commodity exchange mean, and by what principles is it governed? What
principles govern the circulation of commodities on the periphery of tribal
communities? Is commodity exchange the end of an evolutionary sequence?
Does It ave a corrosive m uence on ot er orms o exc ange.
the right questions to be asking anyway?
The fieldwork tradition pioneered by Malinowski, Boas and others has
provided us with the means to answer these questions. It is ironic that at the
very time that these means were becoming available-in the era of European
capitalist imperialism (1870-1914)-economists ceased to be interested in the
concrete problems posed by the theory of commodities and turned instead to
the abstract and formal problems posed by the new paradigm of goods. Many
of the theories formulated within the latter paradigm contained ill-considered
assumptions about the workings of tribal economies, and these provided
ethnographers such as Malinowski (1922) with easy targets to criticize in the
course of developing their own ideas. But fieldwork anthropologists, for the
most part, remained ignorant of the classical tradition of economic thought and
of the challenging questions that lay waiting to be answered. (It was not until
the 1970s, when neo-Marxist anthropology flourished, that anthropologists
took the theory of commodities seriously.) But it is now possible, with the
benefit of hindsight, to see that anthropologists have provided implicit answers
to the questions posed by the commodity theorists and, in the process, have laid
the foundations for a whole new approach to the theory of exchange of wealth
by posing many new questions.
917
SOCIAL LIFE
GIFTS AS WEALTH
Though the terms 'wealth' and 'valuables' are often used in anthropological
literature in the common dictionary sense of 'riches' and 'abundance', they also
take a more precise and anthropologically specific meaning. The word 'gift'
captures this meaning in a very general way and, like the words 'commodity'
and 'goods', it signifies a distinct paradigm (see Belshaw 1965, Sahlins 1972,
Gregory 1982, Strathern 1988 and Weiner 1992 for analytical overviews of the
literature).
The notion of gifts as wealth has assumed a variety of concrete forms, most
of which are now very familiar: the celebrated coppers and blankets of the
Kwakiutl, the armshells and necklaces of the Trobriand Islanders, the brass
rods and cowrie shells of the Tiv of Nigeria, the pigs and pearlshells of the
peoples of the Papua New Guinea Highlands, and so on. Many of these objects
are nowadays valuable in conventional money terms. In the Trobriand Islands,
for example, a vigorous trade in real and counterfeit (plastic) necklaces goes on
outside the tourist centres. However, the most highly prized shells are quite
literally priceless, and have remained in circulation in the kula ring throughout
the colonial period despite the attempts of outsiders to buy them (Campbell
1983). The reasons for this are complex, but it would seem that they have as
much to do with the intricacies of local-level politics as with subjective
preferences. What is clear, though, is that these objects, when exchanged as
gt ts, are va ue y transactors accor mg to a stan ar t at as qua tty rat er
than quantity as its basis.
Consider Campbell's (1983) discussion of the ranking criteria used for
Trobriand armshells (mwari). Five named categories are distinguished and
these are ordered according to their personal history, personal name, colour,
and size. Shells of the top category, mwarikau, have personal names and
histories; they have red striations and are the largest of all. Shells of the lowest
category, gibwagibwa, have neither names nor personal histories; they are white,
unpolished and small in size. Necklaces (vaiguwa) are ranked in a similar way.
The ranking system, then, is ordinal rather than cardinal.
Ordinal ranking systems are ethnographically widespread and their
character is commonly captured in anthropological literature by the expression
'spheres of exchange'. Bohannan (1959), for example, uses this expression to
describe the ranking system of the Nigerian Tiv. Among the Tiv objects are
classified as belonging to one of three spheres. The first, and lowest sphere, is
what the Tiv call yiagh. This includes locally produced foodstuffs, some tools,
and raw materials which are traded at the markets. The second sphere
comprises items of a kind that carry prestige (shagba) and whose transaction is
independent of the markets. Slaves, cattle, horses, white (tugudu) cloth, and
brass rods circulate within this sphere. The third sphere, considered 'supreme',
contains a single item: rights in human beings, especially women and children.
The theoretical significance of this distinction between quality and quantity
918
has, by and large, escaped the notice of twentieth-century economists. J.M.

Keynes (1982) was the prominent exception. Unlike many of his
contemporaries, he had an interest in comparative economy and, during the
1920s, studied the monetary system of the ancient Greeks. In a remarkable
essay, not published until 1982, he noted that cows, corn, iron and bronze,
which were the principal materials of exchange, had 'a conventional order of
value and even a conventional relation of value for customary purposes'
( 1982:256, original emphasis). He argued that 'three types of monetary or
quasi-monetary practice' coexisted: (1) a cow-sheep standard for purposes of
ostentation, religion, reward and punishment; (2) a corn standard for
agricultural rents, wages and loans; and (3) an iron or bronze standard for
market purposes.
It is not surprising that Keynes and Bohannan independently developed the
notion of spheres of exchange, since the ethnographic and historical evidence
repeatedly posed a question for which this theory was the answer. This
question took many forms, one of which was to explicate the puzzling notion of
'equivalence' which Malinowski (1922:355), among others, used to describe the
relationship between objects of gift exchange. This notion is broadly similar to,
but nevertheless subtly different from, the notion of 'equality' which holds in a
price relation. For example if a loaf of bread costs two dollars, then this relation
can be expressed as an equation of the type $2= 1 loaf of bread. But if a kula
bagiriku necklace is held to be equivalent to a mwarikau armshell, then this
re atwn cannot e put m t e orm o an equatiOn ecause to o so wou e to
reduce a qualitative relation to one of quantity.
An analogy with playing cards helps to clarify this point. The four aces
constitute an equivalent set superior in rank to the set of four kings; the set of
kings, in turn, constitutes an equivalent set superior in rank to the four queens,
and so on. This relation can be expressed using the 'greater than' sign (>) as
follows: aces> kings>queens>jacks>tens>nines>etc. These relationships are
ordinal and cannot be expressed in equations of the type '1 ten=2 fives'. Within
equivalent sets ordinal relations also hold. An ace of hearts, for example, is of a
higher rank than an ace of clubs. Objects of gift exchange are similarly ranked,
though the analogy should not be pushed too far. The ranking of gifts is a
serious matter of politics rather than a mere game, and the ordering is often
disputed, especially at the lower end of the scale.
The existence of these qualitative standards poses new questions concerning
the principles governing the exchange and distribution of wealth items. It was
the great achievement of Mauss, in his classic essay on The Gift, first published
in 1925, to pose these questions in a precise way. What, he asked, is the basis of
the obligations to give, to receive, and to return gifts? His answer-implied in
one of the ways he phrased the question: 'What power resides in the object
given that causes its recipient to pay it back?' (1990 [1925]:3)-can be criticized
for its implicit objectification of power relations, by which a property of the
relations between persons is made to appear as though residing in things.
919
SOCIAL LIFE
Nevertheless, his comparative analysis of the ethnographic evidence, which

demonstrated the widespread importance of inalienable bonds between
persons and things, was an extremely valuable contribution to the theory of
exchange. Among other things, it replaced the vacuity of Marx's theory of the
'non-commodity' with a positive theory of the gift. Mauss makes no reference
to Marx in his essay, but his ghost haunts its every page as a kind of invisible
antithesis. Mauss's method was, like Marx's, dialectical, evolutionary,
comparative and political. Dialectics enabled Mauss to see that even though
gifts appear voluntary they are, in reality, repaid under obligation; his
evolutionary approach led him to suggest the primacy of gift exchange over
barter; and his comparative method enabled him to see the significance of the
distinctions between stranger and relative and between the alienable and the
inalienable in terms that were the mirror-image of those employed by Marx.
Mauss's essay on the gift is also very much a political tract. Indeed, it could
be argued that the essay is primarily about early-twentieth-century France. In
the last chapter he discusses the implications of his survey of 'archaic'
economies for the France of his day. Mauss, it must be remembered, was a
socialist but not a communist revolutionary, and his conclusion offers a gift
theory of capitalism to counter Marx's surplus-value theory. He likens the
wage-labour contract to a gift exchange, notes that the worker is giving his time
and life, and that he wishes to be rewarded for this gift (1990 [1925]:77). He
favours a form of welfare capitalism because, as he put it (1990 [1925]:69),
ver-generostty, or commumsm, wou e as arm u to tmse [t e wor er]
and to society as the egoism of our contemporaries and the individualism of our
laws.'
Mauss's theory of the gift owed much to the ethnographic work of
Malinowski and a limited number of other scholars. Since Malinowski's time
the number of high-quality ethnographic reports on exchange has been
increasing apace. These detailed first-hand accounts have been synthesized and
generalized by, among others, Polanyi (1944), Levi-Strauss (1969 [1949]) and
Sahlins (1972). The significant logical and conceptual developments which
anthropologists have made to the theory of exchange can be identified by
comparing the approaches of these three authors.
Polanyi, an economic historian, first became interested in the theory of gift
exchange in order to understand the 'extraordinary assumptions' underlying
the market economy of Europe, the principal subject of his magnum opus, The
Great Transformation ( 1944). For him the market economy was a system of self-
regulating markets in which the prices of commodities organized the whole of
economic life. The basis of this system was seen to lie in the profit motive and
in the existence of commodities in the form of land, labour and money. The
non-market economy, he noted, is the very opposite of this: the motive of gain
is absent, there is no wage-labour, and no distinctively economic institutions.
How then, he asked, is production, exchange and distribution organized? His
argument rests on the identification of three principles of economic
920
Principle of economic behaviour Form of social organization Institutional pattern
Reciprocity Kinship Symmetry

Redistribution Polity Centricity
Householding Household Autarky
Table 1 Relations between principles of economic behaviour, forms of social organization

and institutional arrangements, according to Polanyi (1944).
behaviour-'reciprocity', 'redistribution', and 'householding'-which are 'a

mere function of social organization' (1944:49) and which, in turn, are
associated with distinct institutional patterns. Polanyi's argument is
summarized in Table 1.
Reciprocity has family and kinship as its basis. The reciprocal obligations
that parents and children, brothers and sisters, and husbands and wives have
towards each other 'help safeguard production and family sustenance'
(1944:48). Exchange between groups of kin is facilitated by the symmetry
inherent in the principle of duality upon which many tribal societies are based.
The subdivision of a tribe into moieties, the pairing of villages in different
ecological niches, alliances of individuals from different communities, and
other expressions of the duality principle lend themselves to the creation of
exc ange partners tps w tc persona tze t e re atwn o reciproCity an ma e
long-term exchanges possible.
Redistribution refers to the process by which a substantial part of the annual
produce of a society is delivered to a central figure of authority, who keeps it in
storage for subsequent disposal on special occasions such as annual feasts, the
ceremonial visit of neighbouring tribes, and so on. The social basis of this form
of exchange is a political organization headed by village elders, a chief, king or
despot. It was practised, says Polanyi, in ancient China, the empire of the In cas,
the kingdoms of India, by Hammurabi of Babylonia, in the feudal society of
Europe and in the stratified societies of Africa and the Pacific.
The third principle, householding or production-for-use, is based on the
closed, self-sufficient and territorial household group. The internal
organization and size of the group is a matter of indifference-Polanyi lists the
European peasant farming household and the Carolingian magnates as
examples-because the principle is always the same, namely, 'that of producing
and storing for the satisfaction of the wants of the members of the group'
(1944:53).
For Polanyi, then, the comparative economic history of humanity is
characterized by a great divide: on one side is the self-regulating market; on the
other, economies based on the principles of reciprocity, redistribution and
householding (or some combination of these three principles). This is just
another way of saying that the capitalist economy that emerged at the end of the
921
SOCIAL LIFE
eighteenth century ushered in a radically new form of economic organization,

which was unique in world history.
This is a bold generalization, but it is still a great advance on the mistaken
idea that there was no divide at all-that all economies were commodity
economies. This fallacy, which lay at the base of the writings of Adam Smith,
was uncritically accepted by many twentieth-century economists, and
Polanyi-like Malinowski-was concerned to challenge it. Like Marx, Polanyi
started with Aristotle's distinction between householding and money-
making-'probably the most prophetic pointer ever made in the realm of the
social sciences' (1944:53)-but was able to develop the distinction much
further. Marx, we have seen, developed the category of production-for-
exchange by calling it 'commodity exchange' and distinguishing between its
various forms: barter (C-C); selling-in-order-to-buy in peasant markets (C-
M-C); buying-in-order-to-sell for mercantile profit (M-C-M'), usurious
money lending (M-M'), and industrial capitalism (M-C-M') where labour-
power is the principal commodity. Polanyi's 'great divide' is not between the
presence and absence of commodities but between industrial capitalist
exchange and all other forms. Thus Polanyi was not claiming, any more than
was Marx, that commodity exchange did not exist prior to the emergence of
capitalism. His claim was rather that prior to capitalism, commodity exchange
was subordinate to the principles of reciprocity and redistribution; in other
words that it was socially embedded and hence regulated rather than self-
regu atmg.
Perhaps the most enduring legacy of Polanyi's work was the equation of
'reciprocity' with 'gift exchange'. However, this usage is something of a coded
shorthand because the adjective 'positive' is elided. Thus it is positive
reciprocity which is being equated with gift exchange. The logical corollary of
this formulation was that negative reciprocity came to be synonymous with
commodity exchange. This much is clear from Sahlins's well-known essay 'On
the sociology of primitive exchange' (first published in 1965 and reprinted in
Sahlins 1972: eh. 5), which revises Polanyi's arguments in the light of new
ethnographic data.
Sahlins does not use the term 'positive reciprocity', but it is implicit in his
notion of a kind of reciprocity that he called 'generalized' (as opposed to
'negative'). Generalized reciprocity and negative reciprocity are defined,
respectively, as the 'solidary' and 'unsociable' extremes in a spectrum of
reciprocities. Negative reciprocity is 'the attempt to get something for nothing
with impunity' (Sahlins 1972: 195): haggling, barter, gambling, chicanery, theft,
and other varieties of seizure are examples. Generalized reciprocity 'refers to
transactions that are putatively altruistic, transactions on the line of assistance
given and, if possible and necessary, assistance returned' ( 1972:194 ): examples
include food-sharing, the suckling of children, help and generosity.
The distinction between positive (generalized) and negative reciprocity that
Sahlins proposes here is really an application of Aristotelian logic to the
922
Remittance Admittance
Positive reciprocity Giving +--- Converse -+ Receiving

i ~ ~ i
Contrary Contradictory Contrary
~ ~ ~ ~
Negative reciprocity Losing +--- Converse -+ Taking
Table 2
oppositions giving-receiving and losing-taking. Giving, for example, is the

converse of receiving and the contradictory of taking; losing, the contrary of
giving, is the converse of taking and the contradictory of receiving. This logic
defines two modes of mutuality between the transactors, positive reciprocity
(giving-receiving) and negative reciprocity (losing-taking), and two positions in
relation to the transmission of objects, remittance (giving-losing) and admittance
(receiving-taking). These logical relations are summarized in Table 2.
This table clarifies, at least in a formal sense, the distinction between
reciprocity and exchange. Exchange is the transmission of wealth from one
transactor to another, whereas reciprocity refers to the specific quality of the
relationship between the transactors. This relationship is characterized by
the other (negative reciprocity). Thus specification of the qualitative form of

reciprocity enables particular forms of exchange to be distinguished from
exchange in general. Where wealth is defined as either commodities or gifts this
specification of the quality of the relationship necessarily involves a concrete
investigation into the spatio-temporal forms of social and political organization
of the economy in question. This much is common to the approaches of Smith,
Ricardo, Marx, Mauss and Polanyi, notwithstanding the great differences
between them.
Sahlins places positive and negative reciprocity at two ends of a continuum
whose mid-point specifies a third type which he calls 'balanced' reciprocity.
This form of reciprocity is 'less personal' than generalized reciprocity and
'more economic' (in the Western sense of the term). It expresses the need to
transcend hostility in favour of mutuality, to strike a balance in a relationship.
Examples include formal friendship or kinship involving compacts of solidarity
and pledges of brotherhood, and the affirmation of corporate alliances in the
form of feasts, peace-making ceremonies and marital exchanges.
Sahlins summarizes his argument in terms of a diagram (Figure 1) in which
kinship distance is correlated with reciprocity. Kinship distance, he argues, is
defined by the intersection of consanguinity and territoriality. This defines a
set of ever-widening spheres of eo-membership-household, lineage, village,
etc.-such that as one moves out through these spheres positive reciprocity is
923
SOCIAL LIFE
Negative
reciprocity
Figure 1 Reciprocity and kinship residential sectors. (After Sahlins 1972: 199)
gra ua y counteracte y a negative c arge. ac sector Is, t ere ore,

characterized by the dominance of a certain type of exchange, with the purest
form of gift exchange occurring in the closest sphere and the purest form of
commodity exchange in the most distant-i.e. in the inter-tribal sector.
This model of exchange has been modified and developed by Ingold ( 1986 ),
who has noted that the negativity of reciprocity is independent of kinship
distance. Ethnographic reports on contemporary hunter-gatherer economies
show that negative reciprocity, in the form of 'demand sharing' (Peterson, in
press), exists at the very core of the system; it also exists on the outermost
periphery in the form of theft and burglary. Likewise, positive reciprocity exists
not only at the core, as sharing in which the donor takes the initiative, but also
on the periphery, in the form of haggling and barter. Thus, as kinship distance
increases it is not that positive reciprocity gradually becomes negative but
rather that one form of positive (or negative) reciprocity is transformed into
another form of positive (or negative) reciprocity. Figure 2 illustrates Ingold's
argument; intermediate cases have been omitted for ease of exposition.
Sahlins's neat model-and Ingold's variation-is of course complicated by
the presence of 'other factors'. The most important of these is political rank,
which can be thought of as a vertical axis that intersects, and interacts with, the
horizontal axis of kinship distance. This vertical axis is associated with what
Sahlins calls a system of reciprocities. Under this system products are pooled in
a many-to-one and a one-to-many pattern of exchange.
924
Aliens
(+barter)
(-theft)
Figure 2 Modified model of reciprocity and residential sectors. (After Ingold 1986:232)
Sahlins's conceptual framework, which has Mauss's theory as its basis

a ms 972: c . ), provt es an answer to Mauss's questiOn a out t e
obligation to return gifts. As Figure 1 illustrates, this answer is given in terms
of the social organization of kinship and rank typical of tribal societies. Like
Polanyi, Sahlins was more concerned to examine the implications of this
proposition than to investigate its philosophical basis. Furthermore, there is a
sense in which he considered such a task to lie beyond the scope of his analysis,
for it had already been undertaken by Levi-Strauss in his great work, The
Elementary Structures of Kinship (1969 [1949]).
One of the innovations of Levi-Strauss's book was to conceptualize
marriage as the gift exchange of sisters (daughters) by brothers (fathers). This
notion, as Sahlins (1972: 181) observed, provoked a reaction from British and
American anthropologists who 'recoiled at once from the idea, refusing for
their part to "treat women as commodities"' (emphasis added). Such a reaction,
as Sahlins correctly noted, betrayed a misunderstanding of the comparative
theory of exchange and an inability to distinguish gifts from commodities.
Levi-Strauss's conceptualization of marriage as gift exchange enabled him
to develop a definition of reciprocity of great generality and rigour; this, in
turn, enabled him to synthesize a vast amount of data from Oceania and Asia
and to find patterns of exchange where others had found none. His primary
distinction is between elementary and complex structures of kinship; the
former, the focus of his analytical attention, are further subdivided into
structures of restricted and generalized exchange.
925
SOCIAL LIFE
Restricted exchange, his theoretical starting point, takes the following

general dyadic form:
A +---+ B
Levi-Strauss shows that alliance relations based on the bilateral marriage rule,
that a man should marry a woman in the combined kinship category of
mother's brother's daughter and father's sister's daughter, take this form. The
notion of 'sister exchange', which is often used to describe this form of
reciprocity, precisely captures the essence of restricted exchange. The
perspective is, of course, from the male point of view (compare Strathern 1988
and Weiner 1992), but this is as much the indigenous male's point of view as it
is the ethnographer's. In other words, Levi-Strauss's perspective on exchange
is that of the powerful men who do the exchanging, rather than that of the
women whose place of residence is usually changed as a result of marriage. He
draws illustrative examples from ethnographic studies of the Australian
Aborigines, among whom dual organization is widespread. This conception of
restricted exchange corresponds exactly with Polanyi's correlation of
reciprocity with a symmetrical kinship structure, the difference being only that
Polanyi was mainly concerned with the exchange of objects rather than the
exchange of persons (i.e. sisters or daughters).
By contrast to restricted exchange, generalized exchange takes the following
form:
This form of exchange
establishes a system of operations conducted on credit. A surrenders a daughter or a

sister to B, who surrenders one to C, who, in turn, will surrender one to A. This is
its simplest formula. Consequently, generalized exchange always contains an
element of trust .... There must be the confidence that the cycle will close again, and
that after a period of time a woman will eventually be received in compensation for
the woman initially surrendered.
(Levi-Strauss 1969 [1949]:265)
Generalized exchange is another way of expressing the matrilateral marriage

rule that a man should marry a woman in the kinship category of mother's
brother's daughter, and it is associated with a long cycle of reciprocity.
These two systems of exchange are conceived of as extremes between which
lies a third form of exchange, delayed exchange, which establishes a short cycle
of reciprocity of the following form:
926
Marriage with the father's sister's daughter, the patrilateral rule, is consistent
with exchanges of this type, in which the direction of exchange is reversed
rather than repeated in each successive generation.
These three forms of exchange are all of the 'elementary' type. Levi-Strauss
sees in such elementary structures of kinship the basis for the gift exchange of
things. A bridewealth exchange, for example, is 'a process whereby the woman
provided as a counterpart is replaced by a symbolical equivalent' (1969 [1949]:
470). A transformation such as this can only occur, however, if the marriage is
of the generalized or delayed kind. All three elementary forms, in Levi-
Strauss's scheme, are then opposed to complex structures which leave 'the
determination of the spouse to other mechanisms, economic or psychological'
(Levi-Strauss 1969 [1949]:xxiii). Levi-Strauss's 'great divide', between
elementary and complex structures of kinship, can be mapped onto Polanyi's
between non-market and market exchange. The fit is by no means perfect, but
the degree of correlation is high.
These different forms of exchange, argues Levi-Strauss, are an expression
of the incest taboo, the 'supreme rule of the gift' (1969 [1949]:22). As he put it,
t e pro 1 1t10n o mcest IS ess a ru e pro 1 ttmg marnage wtt t e mot er,
daughter or sister, than a rule obliging the mother, sister, or daughter to be
given to others' (1969 [1949]:22). This is not only Levi-Strauss's answer to
Mauss's question about the basis of the obligation to give, it also underwrites
his theory of cultural evolution. Levi-Strauss argues (1969 [1949]: eh. 28) that
it was man's desire to maximize the kinship distance between himself and his
wife that saw society progress through different evolutionary stages of
development.
There is, in sum, a sense in which the theories of Mauss, Levi-Strauss,
Polanyi and Sahlins, taken together, provide a conceptual framework which is
the mirror image ofMarx's. Whereas the gift theorists begin their analyses with
the direct gift exchange of people and then progress through various mediating
forms to the generalized gift exchange of things, commodity theorists like Marx
begin with the direct commodity exchange of things and progress to the
generalized commodity exchange of labour (Gregory 1982:68). This method of
analysis is 'evolutionary' to the extent that it is making claims about actual
historical processes, but it can also be seen as a 'logical historical' method
(Meek 1967), a mode of reasoning employed in the process of developing an
abstract conceptual framework. This distinction is important because the
logical historical method makes no claims about actual historical processes.
Thus, an evolutionary theory can be rejected without affecting the legitimacy
of the logical historical method. The importance of this distinction should
927
SOCIAL LIFE
become clear in the course of the following discussion of barter and other forms
of counter-trade.
BARTER AND OTHER FORMS OF COUNTER-TRADE

Counter-trade is the general form of non-monetized commodity exchange.
Barter, the simultaneous exchange of commodities (C-C), is the best known
example of counter-trade, but there are many other non-simultaneous forms
(e.g. delayed barter exchange). The latter necessarily involve a time element
and, in consequence, some notion of credit. In formal terms they are analogous
to delayed exchanges of gifts, and in practice it is often impossible to
distinguish between gift and commodity components of a counter-trade
transaction.
The phenomenon of counter-trade poses two questions: What is the
evolutionary status of barter? And what determines the rate of exchange?
For both classical and neoclassical economists barter is the origin of all
exchange. They believe that the original economy was a 'natural' one based on
an elementary division of labour and lacking any form of money. This
inefficient system gave way to money-based exchanges with the progressive
division of labour and the development of markets. Thus the invention of
money was the answer to the 'problem' of barter. This 'origin myth', as Hart
( 1987) has aptly called it, is based on a priori logical reasoning about an
tmagme past rat er t an on contemporary et nograp tc evt ence. T e myt
was repeatedly attacked by early anthropologists as ethnographic evidence on
actual barter exchanges began to accumulate. The evidence shows that different
forms of exchange co-exist rather than following one another in a temporal
sequence. An object can participate in many different forms of exchange in the
course of a day. For example, a pig may begin the day by being sold in a market
for cash, then be bartered for another commodity, later resold at a profit, then
given away as a gift, and finally consumed as a good.
The first reliable evidence to point along these lines came from Malinowski's
( 1922) classic study of the tribal economics of the Trobriand Islanders. From a
comprehensive list of gifts, payments and commercial transactions he
distinguished seven types of exchange, and showed how they were interrelated
in the concrete ecological and social context of the Trobriand Islands in the
early part of this century. His study showed that much geographically based
barter trade took place within the framework of the annual kula gift-exchange
ritual. Recent studies from the Milne Bay area show that these gift exchanges
continue to take place today under the umbrella of the world market economy.
This complexity poses few problems for indigenous transactors in the region,
who know exactly what type of transaction they are entering into, but it has
posed many theoretical problems for anthropologists who have tried to
comprehend what is going on.
Mauss was one of the early synthesizers. He recognized the implications of
928
Malinowski's data for the economists' theory of barter and developed the
alternative thesis that gift exchange preceded commodity exchange. He
proposed a three-stage theory: first came the restricted exchange of gifts within
a tribe, next came generalized gift exchange, and finally the money economy
originated when the ancient Semitic societies 'invented the means of
detaching ... precious things from groups' (1954 [1925]:94).
It is interesting to note, in passing, that barter exchange is re-emerging in
the heartland of international financial capitalism as the hegemony of the
United States wanes and with it the value of the dollar. Barter has long been a
major component of international trade between East and West (i.e. on the
boundaries of United States power), but now many multi-national companies
are resorting to it to safeguard losses from deals involving a declining dollar;
within the United States the rise of computerized exchange, where debts can be
cancelled without the aid of money, has begun to worry the Internal Revenue
Service (Hart 1987: 197).
The ethnographic and historical evidence, then, does not support any
simplistic theory of the evolution of economic forms. This is not to say that the
logical historical method, which organizes concepts in a sequence from simple
to complex, is invalid. To the contrary, as the above discussion has shown, it has
underlain all the significant conceptual developments in the theory of exchange
over the past two centuries.
Let me now turn to consider the question of exchange-rate determination.
asstca po 1t1ca economy, as we ave seen, propose t e a our t eory o
value as the key to understanding the exchange-rate of commodities. It was
argued that two heterogeneous commodities can be equated in value because of
the equality of the labour time contained therein. Neoclassical economists, on
the other hand, proposed that scarcity and utility determine the prices of
goods. What contribution have anthropologists been able to make to this
debate?
The controversy has been uppermost in the minds of many ethnographers
as they observed and collected quantitative data on tribal systems of barter and
counter-trade. Godelier ( 1977), for example, explicitly addressed the debate on
the theory of value in his article on '"Salt money" and the circulation of
commodities among the Baruya of New Guinea'.
The Baruya are a people of the Eastern Highlands ofPapua New Guinea for
whom the production of sweet potatoes is the principal economic activity. They
are also specialists in the production of vegetable salt which is redistributed
among relatives within the tribe and bartered for various products and services
beyond its borders. The latter exchanges were conducted in pre-colonial times
by daring individuals who made contact with hostile neighbours and managed
to establish 'trade and protection' pacts with certain members of the host
groups. Trading partners would feed and protect their guests and do their best
to find the merchandise which the latter desired. Salt was a highly desired
prestige item which was stored above the hearth to be used on ceremonial
929
SOCIAL LIFE
occasions involving the exchange and consumption of gifts. Baruya traders

bartered their salt for a range of commodities, one of which was bark cloth
obtained from the Youndouye, long-time friendly neighbours. Godelier noted
that the exchange rate was one bar of salt (average weight 2 kg) for 6 bark cloths,
and he calculated the labour time required to make the two products. A single
bar of Baruya salt entailed, on average, days of labour, whereas 6 bark cloths
entailed 4 days of labour. In other words, the exchange rate was imbalanced in
labour terms, with the Baruya receiving the equivalent of almost three times
more labour than they gave.
Godelier denies, however, that the Baruya exploit other people's labour.
'What counts', he argues
is the reciprocal satisfaction of their need and not a well-kept balance of their labour
expenditure. For this reason, the inequality of exchange expresses the comparative
social utility of exchanged products, their unequal importance in the scale of social
needs and the diverse monopolist positions of exchange groups.
(1977:150)
Godelier's conclusion is based on an interpretation of the statement of one of

his informants, who declared that 'If we receive enough, then work belongs to
the past, it is forgotten.'
This looks like a victory for neoclassical theory. However, it could also be
labour because the skills required to produce salt are much more highly
specialized than those involved in making bark cloth. The labour theory of
value requires that differences of quality be reduced to those of quantity, and if
the reduction factor was such that three hours ofBaruya labour is equivalent to
one hour of Youndouye labour, then it could be argued that the exchange is
indeed equal.
The conclusion that both theories are valid is, perhaps, to be preferred,
because evidence such as this cannot resolve the fundamental problems of the
theory of value. What is at issue are different methods of apprehending the
world. If we conceptualize Baruya salt as a commodity certain implications
follow; if we conceptualize it as a good different implications follow. They are
incommensurable paradigms and, as such, no way of comparing them exists.
Furthermore, accurate accounting of utility value and labour value is
impossible, even in the Baruya's own terms. How does one reduce skilled
labour time to unskilled? How does one measure marginal utility and make
interpersonal comparisons? There are no satisfactory answers to these hotly
debated questions.
The labour-value and utility-value paradigms do not exhaust the universe of
possibilities, and there is room for other theories of value. Sahlins's essay,
'Exchange value and the diplomacy of primitive trade' (in Sahlins 1972), can be
seen as an attempt to develop an alternative. He begins by noting that the
930
'characteristic fact of primitive exchange is indeterminacy of the rates'

(1972:278). By this he means that similar commodities move against each other
in different proportions in different transactions. He addresses the usual
explanations, finds them wanting, and argues that, in partnership trade, the
rates are set by social tact, 'by the diplomacy of good measure appropriate to a
confrontation between comparative strangers' (1972:302). Sahlins maintains
that in times of scarcity when, according to neoclassical theory, prices are
supposed to rise, the partnership absorbs the pressure and the exchange rate
remains undisturbed. In other words, the 'flexibility of the system depends on
the social structure of the trade relation' (1972:313).
This theory moves the focus of attention from the economic value of objects
to the political value of the trade partnership. In this regard, it invites
comparison with Marx's (1954 [1867]:76) theory of the 'fetishism' of
commodities, which argues that in a world of generalized commodity
circulation, relations between people assume the fantastic form of relations
between things. Sahlins, by way of contrast, argues that in the highly
particularized world of counter-trade, relations between people always appear
as such.
MARKET PLACE TRADE

In the discussion of market exchange, it is essential to distinguish between
mar et princip es an mar et paces. T e ormer ave to o wtt t e a stract
principles that determine the formation of wages, prices and profit. Market
places, by contrast, are the loci where concrete exchanges take place. Marketing
systems are organized frameworks for the purchase and sale of commodities;
their features include customary market centres and a calendar of market days
so that buyers and sellers can meet in regular and predictable ways.
Economists have shown little interest in market places, confining themselves
almost exclusively to the analysis of abstract principles. Most research on
market places has been carried out by geographers and anthropologists, the
former concentrating on their spatial aspects and the latter on their social and
cultural aspects (as Bromley's (1979) comprehensive bibliography illustrates).
Anthropological studies of markets mainly take the form of ethnographic
accounts of particular local regions. General and comparative studies of the
kind that Mauss, Polanyi, Sahlins and others made of gift exchange are rare.
However, while all the classic ethnographic studies-Skinner ( 1964-5) on
China, Mintz (1959, 1961) on the Caribbean, Dewey (1962a) on Indonesia,
Malinowski and Fuente (1982 [1957]) on Mexico, Geertz (1979) on Morocco-
are concerned with the analysis of particular situations, they do contain many
important general analytical points. Some of these have been drawn out by
Bohannan and Dalton (1962) in their introduction to an important collection of
essays on African rural markets.
931
SOCIAL LIFE
Markets can be classified by the types of commodities transacted, by the

forms of trade, by the roles of traders, and by the mode of spatio-temporal
organization.
The commodities traded in markets can be divided into those of the
'vertical' and those of the 'horizontal' type (Mintz 1959). Vertical commodities
are either 'upwardly mobile' or 'downwardly mobile' (Skinner 1964-5).
Upwardly mobile commodities are those which are produced in a rural area and
exported from it by wholesalers. They usually consist of agricultural products,
but may also include products of artisanship such as ceramic pots, basketware,
iron tools, folk art, and so on. These products will ascend through a hierarchy
of wholesalers and eventually become the downwardly mobile commodities of
another area, usually urban and possibly overseas, where they will be
consumed. For a rural area, downwardly mobile commodities usually consist of
manufactured commodities of urban provenance such as clothing or jewellery.
They are brought into a marketing area by wholesalers who resell them to
retailers who, in turn, offer them for sale at the market place. The image of a
vertical commodity, then, captures the nested hierarchy of markets that
characterizes many peasant marketing systems, and locates them in a system
that incorporates local economies into the regional, national and international
economy. By contrast with vertical commodities, horizontal commodities move
across a limited local space. They are usually sold direct to the final consumer
by the producer at the market place, without the mediation of wholesalers.
T IS tstmctwn, t en, provt es t e rst means o c asst ymg mar ets: some
will be characterized by a predominance of vertical commodities (e.g. China),
others by a predominance of horizontal commodities (e.g. West Africa-see
Hill 1966:298).
A second method of classifying markets is by the form of trade. All the forms
of commodity trade discussed above-barter C-C, selling in order to buy C-M-
C, buying in order to sell M-C-M', and moneylending M-M'-are found in
rural peasant markets. Barter trade is extremely rare. I observed barter
transactions in the markets I studied in Central India, but they accounted for a
negligible proportion of total commerce. Reports exist of Andean markets
operating almost exclusively by barter (Mintz 1959:29), but the vast majority of
transactions are of the C-M-C and M-C-M' variety. In Central India selling in
order to buy is the basis of the system from the farmer's perspective. Farmers,
or rather the female members of farming households, bring small loads of
agricultural produce to sell at weekly markets in order to purchase kerosene,
cloth, ornaments or other items. Selling is obviously more intense at the end of
the harvest, but so too is buying. I was struck by the difference in the trading
patterns of markets in the more prosperous areas of northern India. Here
markets are solely of the M-C-M' variety, as farmers only go to the markets to
buy. Their produce, which is grown using more capital-intensive techniques, is
not brought to the market but sold through other channels in a manner similar
to that found in the rich capitalist countries.
932
Classification by the roles of traders is yet a third way to characterize

different market place systems. Traders fall into two main categories: the
mobile trader and the settled shopkeeper. Mobile traders include pedlars, who
wander from place to place hawking commodities in an unsystematic,
opportunistic way. They may be artisans who provide a service or add value to a
product, or they may be pure merchants who buy in order to sell at a profit.
Pedlars of this type usually possess a very small trading capital and travel by
foot, but there are also relatively wealthy individuals who possess motorized
transport. Thus pedlars can be distinguished by the type of trade they do, by
the size of their capital and by their means of transport.
Pedlars are only one kind of mobile trader, and they should be distinguished
from periodic market place traders. These traders have a set round of places,
which are visited on specified weekly market days. A cloth trader I interviewed
in India, for example, would set up his ten thousand rupees'-worth of stock at
the big market in his home town ofKondagaon on Sunday, travel20 km by jeep
to Sampur on Monday, 50 km to Makdi on Tuesday, 70 km to Randha on
Wednesday, rest on Thursday, travel 80 km to Bare Dongar on Friday, and 50
km to Mardapal on Saturday-and so on in this way for 52 weeks of the year.
Like the pedlars, traders in this category can be divided into artisans and
merchants and ranked in terms of their capital (which for many is often less
than a hundred rupees); they can also be divided into wholesalers and retailers.
In addition, traders who attend periodic markets can be distinguished by the
parttcu ar ocatwns m t e mar et w ere t ey set up s op. Rtc tra ers usua y
have a fixed establishment (e.g. a thatched-roof stall covering a small piece of
cleared ground), poor traders will crouch in the dust under an umbrella, while
the pedlar will wander around the market place hawking his or her
commodities.
Mobile traders of all kinds are to be distinguished from shopkeepers. Again,
this category can be subdivided along a variety of axes: wholesaler-retailer,
rich-poor, and so on, all of which are salient in rural areas. They usually
surround the central market place. In Western European countries the relative
importance of the periodic market trader has declined significantly over the
past two centuries as shopkeeping has emerged as the dominant form of
exchange. Nowadays the large department stores reign supreme as the central
loci of exchange. In many non-European countries, however, periodic market
traders are still the key merchants in most rural areas. They capture almost all
of the trade and customers only go to stores for emergency purchases or to buy
insignificant items such as a toothbrush, a pencil, or a packet of biscuits.
It is obvious that the distinct ways of characterizing markets outlined above,
according to types of commodities, forms of trade and roles of traders, are
interrelated. But it is also obvious that they allow for very diverse combinations
of features, and hence for a great variety of possible market place systems. The
task of the observer, looking at such systems in their empirical manifestations,
is to identify the principal tendencies and to account for them. But before we
933
SOCIAL LIFE
examine some of the factors which have been used to explain the observed
patterns (such as the predominance of periodic market traders in the rural areas
of poor countries, of hierarchical markets in China, of C-M-C markets in
Central India, and so on), it is necessary to consider some of the many different
types of spatio-temporal organization found in periodic markets.
Periodicity concentrates the demand for a product to a certain place on a
specified day between set hours. A trader can, by repositioning him or herself at
regular intervals, tap the demand of a market area and obtain an income from
commerce that is adequate for survival. From the point of view of farming
households the periodicity of markets reduces the distance they must travel in
order to sell their produce and to buy goods for consumption. In effect,
periodicity disperses the central market-town throughout the countryside and
converts sleepy backwaters into thriving commercial centres for a few hours
each week. This pattern of dispersal is a function of the availability of
transport: for rich market-town traders there is a limit to how far they are
prepared to drive each day, and for poor farmers there is a limit to how far they
are prepared to walk.
The distribution of periodic markets over time and space poses a problem
that can be expressed in mathematical terms. Christaller's (1966 [1933]) classic
application of central place theory is one such expression that has proved very
influential with geographers and with some anthropologists (e.g. Skinner
1964-5). However, rather than elaborating on formal models of this kind, it is
more appropnate ere to gtve some m tcatwn o t e actua vanatwns oun m
the spatio-temporal organization of marketing systems.
In China market schedules are usually based on a ten- or twelve-day week.
This structure allows for the development of cyclical systems of great
complexity. For example, the 12-day cycle yields three regular cycles of 12-day,
6-day and 3-day market weeks; within these cycles many further possibilities
for scheduling are found. Six different schedules make up the 6-day week for
example: the first consists of the 1st and 7th day of the cycle, the second of the
2nd and 8th day, the third of the 3rd and 9th day, and so on. If town A chooses
the first schedule, town B the second, town C the third and so on, then it can be
seen that a farming household living equidistant from these three towns has 3
markets close by on 6 of the 12 days of the market week; towns D, E, F, G, etc.
will provide the household with a range of more distant markets to choose from
on the other days of the week.
In Central India the system is comparatively simple. The market week is a 7-
day one. The major market is held on Sundays at the central market town;
intermediate level centres hold their markets on Fridays, Saturdays and
Mondays; and small centres hold their markets on the remaining days. In West
Africa there is a standard market week of 3, 4, 5, 6, 7 or 8 days in length, such
that all markets in a given locality are based on the same cycle.
In areas where vertical commodities predominate, space becomes ordered in
a hierarchical way with market centres of various sizes constituting the nodal
934
points. Skinner ( 1964-5) proposes a multi-level typology of central places,

ranging from the local minor marketing area, based on a central village, to the
major regional trading area based on a regional city. He finds that the
arrangement of minor market areas in China approximates to a honeycomb
pattern and that this hexagonal spatial grid is reproduced at the higher levels.
Abstract models like this are useful heuristic devices for understanding the
empirical complexity of market systems and have stimulated much
geographical research (Smith 1978), but it is the social organization of markets
that has been the anthropologists' prime concern.
A persistent theme of anthropological literature is the ubiquity of economic
and social differentiation in periodic market places. Mintz (1959), for example,
notes that in Haiti horizontal exchange occurs among class equals, while
vertical exchange occurs between class unequals. In other words, as we move up
the hierarchy of vertical commodities we also move up through a class
hierarchy. Dewey (1962b) found a similar situation in Java. Here Javanese
farmers dominate the small-scale trade, Chinese, Arabs and Indians dominate
the large-scale inter-market trade, and Europeans retain control over the really
large-scale economic enterprises. This pattern, she claims, is found throughout
South-east Asia. In India the marketing system is entirely in the hands of
Indians, but the general correlation of class and ethnicity is still to be found.
The elite traders found in almost all market areas of India are the Marwaris.
They are migrants from Marwar in Rajasthan and, as a group, control a
tsproportwnate s are o t e m ustna an mercantt e wea t o t e country.
Explaining this social differentiation has been a central concern for
anthropologists. How is it, they ask, that markets that are the closest known
approximation to the economists' ideal of free competition are nevertheless
characterized by such gross social and economic inequalities? Many
explanations have been put forward. Investigations have focused, among other
things, on culture, ecology, population pressure, the labour-intensive
technology of poor farmers, and systems of land tenure.
One paradox that anthropologists have identified is that the competitive
market system is backed by a 'strong personalistic element which affects the
nature of internal marketing activity' (Mintz 1959:25). In Haiti this personal
relationship is called pratik. It means that buyer and seller emphasize 'the
reciprocal nature of relationships' (1961:55, my emphasis). Women who buy
and sell on these terms call each other bel me ('stepmother') or matelot
('concubine of the same man'). Reciprocity of this kind between buyer and
seller is called 'goodwill' in European countries, where it has been converted
into a commodity: shopkeepers and sellers of professional services (e.g. doctors,
lawyers, dentists) pay huge sums for it.
Another important kind of reciprocity in market systems is that obtaining
between sellers of a given type of vertical commodity. As we have seen, these
merchants tend to belong to families or ethnic groups whose members identify
with each other in opposition to the world at large. These groups, Dewey ( 1962b)
935
SOCIAL LIFE
notes, develop a social structure that enables them to bring informal sanctions to
bear on their members. In these localized power systems coercion and
collaboration create solidary relations which bring benefits to the in-group and
problems for the out-group. One of the greatest benefits to the in-group is access
to credit. This provides members of the in-group with initial capital and the
ability to accumulate more. Whereas debt can enchain a consumer, for merchants
it is their lifeblood, for without it they cannot expand their capital. It is obvious
that credit will not be extended where there is neither trust nor sanction and, in
periodic market systems, this marks the boundary between the in-group and the
out-group. Thus we find that credit for merchant capital expansion flows upon
the foundations laid by consanguinity and territoriality. Here, then, is an
important factor behind the observed hierarchies found in market places and,
when considered in the light of the particular history of a merchant class, it goes
some way towards explaining the wealth of some and the poverty of others.
Dewey's argument, for which a wide range of supporting evidence can be
marshalled, amounts to the claim that positive reciprocity asserts itself in unique
ways in the heartland of negative reciprocity, the market place.
This argument seems to contradict Sahlins's theory of positive and negative
reciprocity. However a distinction must be maintained between the analysis of
abstract principles of exchange and the analysis of exchange in concrete
situations. The theories developed by scholars such as Smith, Ricardo, Marx,
Malinowski, Mauss, Polanyi and Sahlins are abstractions which must be
recogmze as sue an app te wtt cautiOn to t e ana ysts o concrete rea tty.
The message of Dewey's argument-and of the growing body of literature
concerned with applying the theory of the gift to European history (White
1988), literature (Hyde 1984), economy (Zelizer 1989) and culture (Agnew
1986)-is that concrete reality is riddled with contradictions. This means that
any attempt, say, to characterize the European economy as a commodity
economy and the Melanesian economy as a gift economy, is bound to fail
because positive and negative reciprocity is at work in both economies. The
notion of reciprocity, then, can be defined in the abstract but its real meaning
will always depend on the concrete political context.
REFERENCES
Agnew, J-C. (1986) Worlds Apart: The Market and the Theater in Anglo-American
Thought, 1550-1750, Cambridge: Cambridge University Press.
Belshaw, C.S. (1965) Traditional and Modern Markets, Englewood Cliffs, NJ: Prentice-
Hall.
Bohannan, P. (1959) 'The impact of money on an African subsistence economy', The
Journal of Economic History 19:491-503.
Bohannan, P. and Dalton, G. (eds) (1962) Markets in Africa, Evanston, Ill.:
Northwestern University Press.
Bromley, R. (1979) Periodic Markets, Daily Markets, and Fairs: A Bibliography
936
Supplement to 1979, Centre for Development Studies Monographs no. 5, University

of Swansea.
Camp bell, S. (1983) 'Attaining rank: a classification of shell valuables', in ].Leach and
E.R.Leach (eds) The Kula: New Perspectives on Massim Exchange, Cambridge:
Christaller, W. (1966 [1933]) Central Places in Southern Germany, Englewood Cliffs,
N.J.: Prentice-Hall.
Codere, H. (1950) Fighting with Property, New York: J.J.Augustin.
Debreu, G. (1959) Theory of Value: An Axiomatic Analysis of Economic Equilibrium,
New Haven: Yale University Press.
Dewey, A. (1962a) Peasanting Marketing in Java, New York: Free Press.
--(1962b) 'Trade and social control in Java', Journal of the Royal Anthropological
Institute 92:177-90.
Friedman, M. (1962) Price Theory: A Provisional Text, Chicago: Aldine.
Geertz, C. (1979) 'Suq: the bazaar economy in Sefrou', in C.Geertz, H.Geertz and
L.Rosen, Meaning and Order in Moroccan Society, Cambridge: Cambridge
University Press.
Godelier, M. (1977) Perspectives in Marxist Anthropology, Cambridge: Cambridge
University Press.
Hart, K. (1987) 'Barter', in J.Eatwell, M.Milgate and P.Newman (eds) The New
Palgrave: A Dictionary of Economics, New York: Macmillan.
Heilbroner, R.L. (1987) 'Wealth', in J.Eatwell, M.Milgate and P.Newman (eds) The
New Palgrave: A Dictionary of Economics, New York: Macmillan.
1, . otes on tra 1t10na mar et aut onty an mar et peno !City m est
Africa', Journal ofAfrican History 7:295-311.
Hyde, L. (1984) The Gift: Imagination and the Erotic Life of Property, London: Vintage.
lngold, T. (1986) The Appropriation of Nature: Essays on Human Ecology and Social
Relations, Manchester: Manchester University Press.
Jevons, W. (1970 [1871]) The Theory of Political Economy, Harmondsworth: Penguin.
Keynes, J.M. (1982) 'Ancient currencies', in D.Moggridge (ed.) Collected Writings,
New York: Macmillan.
Levi-Strauss, C. (1969 [1949]) The Elementary Structures of Kinship, London: Eyre &
Spottiswoode.
Malinowski, B. (1922) Argonauts of the Western Pacific, London: Routledge & Kegan
Paul.
Malinowski, B. and de la Fuente, J. (1982 [1957]) Malinowski in Mexico: The Economics of
a Mexican Market System, ed. S.Drucker-Brown, London: Routledge & Kegan Paul.
Marx, K. (1954 [1867]) Capital, vol. I: A Critical Analysis of Capitalist Production,
Moscow: Progress.
Mauss, M. (1954 [1925]) The Gift: Forms and Functions ofExchange in Archaic Societies,
trans. I.Cunnison, London: Routledge & Kegan Paul.
--(1990 [1925]) The Gift: The Form and Reason for Exchange in Archaic Societies,
trans. W.D.Halls, London: Routledge.
Meek, R.L. (1967) Economics and Ideology and Other Essays, New York: Chapman.
Mintz, S. W. ( 19 59) 'Internal market systems as mechanisms of social articulation',
Proceedings of the 1959 Annual Spring Meeting of the American Ethnological Society,
ed. V.F.Ray, Seattle: Washington University Press.
937
SOCIAL LIFE
--(1961) 'Pratik: Haitian personal economic relationships', Proceedings of the 1961

Annual Spring Meeting of the American Ethnological Society, ed. V.E.Garfield,
Seattle: Washington University Press.
Peterson, N. (in press) 'Demand sharing: reciprocity and the pressure for generosity
among foragers', American Anthropologist.
Polanyi, K. (1944) The Great Transformation, New York: Rinehart.
Quesnay, F. (1962 [1759]) 'The tableau economique', in R.Meek (ed.) The Economics of
Physiocracy, London: George Allen.
Robbins, L. (1932) An Essay on the Nature and Significance of Economic Science,
London: Macmillan.
Sahlins, M. (1972) Stone Age Economics, London: Tavistock.
Samuelson, P. (1947) Foundations of Economic Analysis, Cambridge, Mass.: Harvard
University Press.
Skinner, G.W (1964-5) 'Marketing and social structure in rural China', Journal of
Asian Studies 24:3-43, 195-228, 363-99.
Smith, R.H.T. (ed.) (1978) Market-Place Trade: Periodic Markets, Hawkers, and Traders
in Africa, Asia, and Latin America, Vancouver: University of British Columbia Press.
Sraffa, P. (1960) Production of Commodities by Means of Commodities: Prelude to a
Critique of Economic Theory, Cambridge: Cambridge University Press.
Weiner, A.B. (1992) Inalienable Possessions: The Paradox of Keeping-While-Giving,
White, S.D. (1988) Custom, Kinship, and Gifts to Saints: The Laudatio Parentum in
Western France, 1050-11 SO, University of North Carolina Press: Chapel Hill.
FURTHER READING
Belshaw, C.S. (1965) Traditional and Modern Markets, Englewood Cliffs, NJ: Prentice-
Hall.
Bohannan, P. and Dalton, G. (eds) (1962) Markets in Africa, Evanston, Ill.:
Northwestern University Press.
Dalton, G. (ed.) (1967) Tribal and Peasant Economies, Austin: University of Texas Press.
Geertz, C. (1963) Peddlars and Princes: Social Change and Economic Modernization in
Two Indonesian Towns, Chicago: University of Chicago Press.
University Press.
Gouldner, A. (1960) 'The norm of reciprocity: a preliminary statement', American
Sociological Review 25:161-78.
Gregory, C. A. and Altman, J. C. ( 1990) Observing the Economy ( ASA Research Methods
in Social Anthropology, 3), London: Routledge.
Humphrey, C. and Hugh-Jones, S. (eds) (1992) Barter, Exchange and Value: an
Anthropological Approach, Cambridge: Cambridge University Press.
Leach, J. and Leach, E.R. (eds) (1983) The Kula: New Perspectives on Massim Exchange,
938
Levi-Strauss, C. (1969) The Elementary Structures of Kinship, London: Eyre &

Spottiswoode.
Malinowski, B. (1921) 'The primitive economics oftheTrobriand Islanders', Economic
Journal3l:l-16.
Mauss, M. (1990) The Gift: the Form and Reason for Exchange in Archaic Societies, trans.
WD.Halls, London: Routledge.
Parry, JP. and Bloch, M. (eds) (1989) Money and the Morality of Exchange, Cambridge:
Polanyi, K. (1944) The Great Transformation, New York: Rinehart.
Polanyi, K., Arensberg, C. and Pearson, H. (eds) (1957) Trade and Markets in the Early
Empires, New York: The Free Press.
Sahlins, M. (1972) Stone Age Economics, London: Tavistock.
Asian Studies 24:3-43, 195-228, 363-99.
939
34
POLITICAL DOMINATION AND

SOCIAL EVOLUTION
Timothy Earle
The evolution of societies from small-scale intimate groups to large and

complex urban states is a fact of human history. To explain the processes
responsible for this evolution has been a challenge for social philosophers,
sociologists, and anthropologists. 1 In this article, I argue that the evolution of
social complexity needs to be understood first and foremost as a political
process. Fun amenta IS t e 1ssue o contro : ow emergmg ea ers 1p
establishes and extends control over the labour of a non-kin support group
(Earle 1989, Webster 1990).
Control is essential to mobilize the resources needed to finance emerging
institutions. It is grounded in different sources of power: military, ideological
and economic; but these different sources of power are not of equal use to an
emerging elite. To construct the hierarchical power structure that is the
backbone of a complex society, access to power must itself be controllable, and
the different sources of power vary considerably in their ability to be thus
controlled. As I argue here, it is the grounding of each source of power in
controllable economic systems that becomes the critical factor for the evolution
of socially stratified and politically centralized societies.
In this article, I proceed in two steps. First, I summarize briefly the
evolutionary typologies for human society that have proliferated in the last
century or so. My goal is to clarify the key variables in these typologies as they
relate to different mechanisms of change. Second, I propose a synthetic model
for the evolution of complex societies that draws extensively on existing
formulations, but focuses on the political processes tied to different sources of
power.
940
POLITICAL DOMINATION AND SOCIAL EVOLUTION
TYPOLOGIES OF CULTURAL EVOLUTION
The proliferation of evolutionary typologies in the literature has been greeted

by scholars with some dismay (Feinman and Neitzel 1984, Kristiansen 1991).
In the sections that follow, I review the most important typological schemes
that have been proposed, their diagnostic variables, and their respective engines
for change. Table 1 summarizes the main stages in the various schemes and
indicates how they intersect typologically.
Although the authors of these various schemes have been at great pains to
recognize the complicated and multi-causal nature of societal change, three
rather distinct schools of thought emerge. One emphasizes technology, another
stresses the scale of integration, and the third singles out social structure as the
most significant dimension of variability.
Technology and social evolution

According to the 'technological' theories of social evolution, human beings
solve problems of living by developing material culture, and in the course of
time they gradually accumulate knowledge of how to adapt better to their
environments. Each successful, novel solution increases the effectiveness of
subsistence practices, allows for population growth, and has many social
consequences both foreseen and unforeseen.
ommant m uence m t IS se oo o evo utwnary t oug t as een t at o
Marx. Based on a common, nineteenth-century belief in technological progress
(see Morgan 1877), Marx (1904) and Engels (1972 [1884]) constructed their
seminal theory that human history has been propelled by complicated
interplays between developing forces of production and the social relations
within which these forces have been worked out. During the twentieth century,
Table 1 Some common anthropological typologies of social evolution
Service (1962) Sahlins ( 1963)

Childe ( 1936) Johnson and Earle (1987) Earle (1978) Fried (1967)
Hunter- Band
gatherers (family level) Head man Egalitarian society
Farmers Tribe
(local group) Big man
- Ranked society
Simple
Chiefdom ~--·~""' ""
Civilization Complex Stratified society
State State State
941
SOCIAL LIFE
an eclectic mix of Marxist, Weberian, and functionalist theoreticians have

emphasized the role of technology in social evolution (Childe 1936, White
1959, Lenski 1966, Glassman 1986).
The evolution of social complexity, according to this view, is tied to the
development of effective technologies that permit human populations to grow,
to establish permanent settlements and eventually to generate a surplus that
supports administrators, merchants, priests, artists, and craftsmen. The major
technological stages are adduced from Morgan's (1877) tripartite division into
hunter-gatherer societies (savagery), agricultural societies (barbarism), and
complex specialized societies (civilization).
Such a tripartite division is a common feature of many evolutionary
schemes. Perhaps most influential have been the writings of the archaeologist
V.Gordon Childe (1936, 1951). Childe conceived of two major revolutions in
human history-the Neolithic Revolution and the Urban Revolution. In
Childe's scheme, hunter-gatherers were wanderers, eking out an existence from
foods available in nature. Following the domestication of plants and animals,
formers were able to settle down and create a village life with permanent houses,
a richer material culture, and a more reliable subsistence. Then, further
technological advances, especially the beginnings of metallurgy, brought into
being a specialized and highly efficient economy from which the urban life of
civilization could flourish.
The primary engine for change, according to these theories, is the dynamic
re atwns tp etween tee no ogy an uman popu atwn. In or er to survtve
and prosper, human groups must solve critical problems of making a living;
technology is the means to solve these problems of subsistence. This 'vulgar'
form of Marxism (c£ Friedman 1974, Johnson and Earle 1987:9) emphasizes
how technology's most basic function is to extract energy and materials used to
support human populations. The dynamic relationship between technology
and population has been conceived in both a positive and a negative light.
Following nineteenth-century optimism, technological development has
been seen as the means to liberate human societies from the bondage of nature,
allowing a growing and settled human world. This was Childe's position in
Man Makes Himself(l936), in which he emphasized the successful revolutions
in human technology. Leslie White (1943) elaborated and systematized this
view of technological development as increasing the efficiency of energy
capture and thus permitting the evolution of more complex societies. By
contrast, a more pessimistic view, which goes back to Malthus ( 1798) and his
followers, turns the equation around. Human populations, like those of any
other animal species, have the biological potential for sustained growth, and
will expand until that growth is curtailed by disease, starvation and war.
Unique to human beings is the capability to increase by technological means
the environment's productivity, and thus to permit the population to expand.
Boserup (1966) thus shows how agricultural technology developed in response to
the population-driven need to intensify subsistence production. Following this
942
logic, Cohen ( 1977) argues that the origins of agriculture in different regions of
the globe resulted from sustained population growth that led to the peopling of
the world and pressed against the availability of wild resources in a hunter-
gatherer economy (see also Cohen's contribution to this volume, Article 10).
This is a classic chicken-and-egg debate: on the one hand, an inherent
growth in human technology is said to have permitted population growth
(Childe and White); on the other hand, inherent growth in human population
is said to have caused hardships that either limited growth (Malthus), or caused
technological and social innovations that permitted further growth (Boserup).
Most probably the two suggested prime movers-demographic expansion and
progressive technological innovation-are inexorably bound together Oohnson
and Earle 1987). The reproductive potential of human populations and the
cultural capability to enhance the productivity of environmental resources
together generate a growth-oriented system of a kind hitherto unknown. But a
further implication of this interpretation is that technological change did not
result in a 'better world' with higher per capita consumption; rather, the result
was simply a greater population.
The positive feedback between population growth and technological
innovation is linked to social modes of production which provide the material
basis for social differentiation. Marxist analyses of capitalism, for example,
describe how the ownership of industrial technology conferred control over the
productive process and exclusive rights to the profits derived therefrom.
evera recent evo utwnary se ernes, Marxist m conceptiOn, emp astze ow
the characteristics of the new technologies and economies affect the political
and social character of life. Major syntheses include those of Lenski (1966),
who distinguishes between hunting and gathering societies, simple
horticultural societies, advanced horticultural societies, agricultural societies,
and industrial societies; and Glassman (1986), who separates the stages of
democracy in hunting-gathering band society, democracy in hunting-
horticultural or herding-hunting tribal society, and despotism in horticultural
village society or nomadic herding society. On the basis of this separation into
stages, Glassman attempts to show how gender and political relations derive
from the way in which persons stand, vis-a-vis each other, with regard to their
respective labour roles in subsistence production.
Clearly, what is needed is a systematic way to link the processes of
technological elaboration and population growth with the development of more
complex societies. The next school of social evolutionary thought to be
considered deals with this linkage explicitly.
Scale of integration and social evolution

The scale of integration in human society expands through the creation of
overarching levels of organization that embed pre-existing structures. The
school of thought that emphasizes this aspect of social evolution holds that
943
SOCIAL LIFE
institutional mechanisms develop to integrate the larger and more complex

sociopolitical groups needed to solve economic problems. Such institutions
include central leadership, social hierarchies, and related features, such as
writing, state religions and bureaucracies, that are characteristic of complex
societies. Why new levels of integration are created is still a matter of debate
between those who emphasize, respectively, managerial and political causes.
Evolutionary schemes based on levels of integration have a long pedigree.
The original conquest theories of the origins of the state, for example,
explained the creation of large-scale societies as resulting from conquest and
incorporation. In the nineteenth century, Spencer ( 1967) conceived of social
evolution as the political process by which stronger societies expanded to
dominate politically weaker ones. In recent anthropology, Steward (1955)
regarded the development of new levels of social integration as a solution to
problems of adaptation. This perspective from cultural ecology was
popularized in the influential evolutionary typology of Elman Service (1962),
who distinguished the successive levels of band, tribe, chiefdom, and state.
This scheme, based on four levels and mechanisms of integration, has been
recast in the recent synthesis by Johnson and Ear le ( 1987), as follows.
Family level society (the band)

Describing them as the 'most rudimentary form of social organization', Service
9 2: 07 envtswns an s as exemp ars o t e pnmor ta uman soCia orm.
The band was originally thought to consist of a small, patrilineal, exogamous
group organized for effective hunting. Subsequent studies, notably those
included in the symposium volume Man the Hunter (Lee and De Vore 1968; c£
Williams 1974), forced a significant modification of the band concept.
Gathering is now recognized as a primary source of food in many hunter-
gatherer societies, and residential groups are seen as small and flexible in
composition, allowing people to respond to changing opportunities in their
environments. By renaming this level of integration 'the family level', Johnson
and Earle (1987) emphasize the informal and intimate character of the small
social groups that are based on close kinship relationships (compare Steward
1977).
The local group (the tribal level)

The local group forms by organizing and embedding several extended families
into a village-sized group of a few hundred. Originally, Service (1962: 113)
emphasized the importance of pan-tribal sodalities, such as clans and warrior
societies, that create a regional, decentralized organization integrating many
villages. Criticizing this concept, Fried (1967: 154) argued that tribes as regional
organizations were largely constructs of colonial governments, designed for
administrative purposes. Sahlins (1968) then reconceived the tribal level of
944
integration as inherently fragmented into local villages. Lacking regional

political institutions that could resolve conflict, villages in tribal societies
constantly fight with each other. Johnson and Earle (1987) describe these
regionally fragmented social systems, organized politically at the village level
and extended regionally only through shifting political alliances and ritual
cycles.
Leadership within a local group varies considerably in terms of the degree
and concentration of political power Oohnson and Earle 1987). It is not,
however, formalized in political offices (see Sahlins 1963; c£ Earle 1987).
The chiefdom: simple and complex

The primary distinguishing characteristic of the chiefdom is the extension of
the polity to incorporate and integrate multiple communities within a region
(Carneiro 1981). For Service (1962), chiefdoms are redistributive societies in
which a formalized central agency, personified as the chief, emerges to co-
ordinate the distribution of specialized goods within the regional polity.
Johnson and Earle (1987) emphasize the inherent social inequality and control
that emerge with chiefdoms. They differentiate between simple chiefdoms-
polities of a few thousand people with modest social differentiation, and
complex chiefdoms-polities of tens of thousands with marked social
stratification. Integration is provided by a highly generalized and
The state
State societies are complex and internally differentiated by class, economic
specialization, and ethnicity. The expansion in scale, in comparison to
chiefdoms, is connected to various institutions of integration: military,
religious and bureaucratic. As Wright (1984) emphasizes, the centralization of
leadership is based on the formalization of decision-making hierarchies. In
contrast to chiefdoms, in which a leadership stratum remains generalized, state
bureaucracies are internally specialized with a differentiation of decision-
making and control functions. Special characteristics of states may include
writing systems, which are important for record-keeping, and elaborate
systems of transportation and communication. The economies of states are
based either on a system of pooling and redistribution or on market exchange,
or on a combination of the two.
In the schemes presented above, which classify societies by their levels of

integration, the main engine for change must be one that leads human societies
to increase in scale. When it comes to specifying this engine for change, two
competing approaches are evident (Service 1978). The first stresses issues of
adaptation, arguing that central leadership is a social technology developed to
945
SOCIAL LIFE
solve problems of survival. The second approach stresses the political

dimension, arguing that central leadership is an outcome of expanding
domination.
The adaptationist approach of Steward (1955) and Service (1962) envisaged
a simple dynamic for the evolution of human society. Human institutions are
organizational solutions to critical problems of adaptation. Directly analogous
to technological advance, the evolution of central leadership was conceived as
the development of administrative forms designed to manage the economy for
the benefit of the collectivity. More recently scholars have emphasized the role
of population growth in creating new problems, which in turn call forth novel
social solutions (see Johnson and Ear le 1987).
Examples in which the evolution of social forms is seen to have a managerial
basis abound in the literature. The flexible organization of family-level society
is seen as an adaptation to the fluctuating resource base of a gathering economy
(Yellen and Harpending 1972). The decentralized organization at the local-
group level could be similarly appropriate to the management of intergroup
relations, including those of exchange, insurance against risk, military co-
ordination, and exogamy (see Dalton 1977, Braun and Plog 1982). Service
(1962) envisioned chiefdoms as having evolved through the adaptation of
expanding populations to a settled life in diverse environments requiring local
specialization. The possibility this opened up for a more productive and secure
agriculture based on irrigation encouraged groups to accept despotic leaders
responst e or eo-or matmg t e constructiOn an mamtenance o ungatwn
works. These leaders then created the bureaucratic state (Wittfogell957). In all
of these cases, the logic is quite similar: the 'need' to adopt a particular solution
for collective survival, or the advantage it confers, encourages the group to
accept a leader's central direction.
The alternative approach emphasizes the political basis for the evolution of
central authority and corresponding social stratification. In a view derived
loosely from nineteenth-century Marxism, the social stratification of chiefdoms
and states is seen to emerge from the material conditions of control. In his oft-
quoted preface to A Contribution to the Critique of Political Economy, Marx
(1904:12) argued that:
The sum total of these relations of production constitutes the economic structure of
society-the real foundation, on which rise legal and political superstructures and
to which correspond definite forms of social consciousness. The mode of
production in material life determines the general character of the social, political
and spiritual processes oflife.
Anthropologists have returned to this basic materialism, emphasizing the

importance of economic control, rather than management, in the evolution of
stratification and related social complexity. Carneiro (1970), attributing his
approach to Spencer, has argued that complexity derives from conquest. Where
946
geographical or 'social' (i.e. demographic) circumscription exists, a subjugated

population has no escape from its conquerors. Similarly, irrigation and
intensive agriculture act to circumscribe a population by tethering it to
improved lands; ownership of these lands by elites confers the control on which
stratification can emerge (Ear le 1978, Gilman 1981 ). The redistributive and
market systems of chiefdoms and states are viewed as mechanisms of
institutional finance, rather than of ecological adaptation (Brumfiell976, 1980,
Earle 1977). In an elegant argument, Haas ( 1982) shows that systems of
resource extraction rely on a balance whereby the cost to a commoner of his
compliance with an elite's demand for labour and resources must remain less
than the cost of refusal. The commoner is caught in an asymmetrical power
relationship; the greater the asymmetry in the relationship the more likely is
the commoner to comply.
A peasant will give labour and resources to his lord as long as his options for
refusal are limited. But what limits the options available to the commoner? The
answer to this question should solve the dilemma of how complex systems
evolve. One obvious limitation is structural, as I shall now show.
Social structure and social evolution

The basic premise of evolutionary schemes emphasizing structural change is
that a historical transformation in the nature of social relationships and
resource owners tp un er tes t e evo utwn o uman society. For examp e,
what Polanyi ( 1944) called the 'great transformation' refers essentially to a
change in how people are structurally related to one another. In nineteenth-
century writings, Maine (1861) distinguished between the structuring of
societies by status and by contract, and Engels (1972 [1884]) distinguished
between societies based on kinship and those based on territory. From an
economic standpoint, Durkheim (1933 [1893]) contrasted divisions of labour
based respectively on mechanical and organic solidarity, while Mauss (1954
[1925]) saw an evolution from societies characterized by gift exchange to
societies in which the exchange of commodities prevailed. Moreover, many
anthropologists have drawn a structural division, albeit non-evolutionary in
conception, between stateless and state societies (Evans-Pritchard and Fortes
1940, Mair 1962). All such formulations envisage a fundamental contrast in the
structural basis of society: on the one hand are traditional societies integrated
by social relations, and on the other are modern societies organized by a
combination of codified laws and the economic relations of the market.
The most clearly elaborated evolutionary synthesis deriving from this
intellectual tradition is to be found in the work of Fried ( 1967). His four stages
are those of egalitarian, ranked, stratified and state societies. In many respects
these social stages may readily be identified with those posited by Service, but
Fried emphasizes the structural transformations involved.
947
SOCIAL LIFE
Egalitarian society
Equated with family-level integration and with local groups with headmen,
this has 'as many positions of prestige in any given age-sex grade as there are
people capable of filling them' (Fried 1967:33). These societies are structured
traditionally by kin relationships and by the universal criteria of age and sex.
Ranked society
Equated with Big Man polities and simple chiefdoms, this has a limited
number of positions of valued status such that 'not all those of sufficient talent
to occupy such statuses actually achieve them' (Fried 1967: 109). Local polities
have ritual and political leaders who acquire their positions on the basis of
traditional principles that rank individuals with respect to each other. Fried
believed that such positions were not based on economic power or privilege.
Leadership carried traditional rights of obedience and obligations to manage
economic projects such as the construction of irrigation systems and the
redistribution of specialized goods (compare Service 1962).
Stratified society
Equated loosely with complex chiefdoms, this 'is one in which members of the
resources that sustain life' (Fried 1967: 186 ). This stage is poorly defined
ethnographically, and its separation from that of ranked society seems to be
based only on the structural transformation from communal to private
property.
My impression, though disputed by some (Kristiansen 1991 ), is that ranked
and stratified societies are not qualitatively different. Rather, in both, leaders
attempt to maximize their political advantage, but their ability to do this varies
according to the available systems of control and finance; the outcome is
quantitative variation in the strength and extent of political centrality and in
the resulting developmental dynamics (Sanders and Webster 1978, Ear le 1989).
State society
This is identified as having 'specialized institutions and agencies ... that
maintain an order of stratification' (Fried 1967:235). Thus states simply
represent the expansion and institutionalization of the structural changes
underwriting the emergence of stratified society, and Fried explains the rarity
of the proto-typical stratified society on the grounds that once the critical
structural transformation giving rise to stratification has occurred, the state
will necessarily develop quickly to solidify it.
Friedman and Rowlands (1977) significantly expand on Fried's formulation
948
by looking at the inherent developmental characteristics of non-stratified

'tribal' society, including both local groups and chiefdoms. Leaders seek to
enhance their positions of eminence by manipulating prestige goods, exchange
ties and linked social relationships involving political marriages and alliances.
Asiatic states, based on the replacement of kinship by territorial principles,
develop as chiefs control systems of redistribution that are grounded in
ownership of productive resources (i.e. agricultural lands). Perhaps the most
important element of Friedman and Rowlands's model is that the society is
seen as having its own, growth-oriented dynamic based both on its internal
structure (compare Earle 1978) and on its regional and long-distance
articulation with world economic systems. Thus development can be
understood in terms of broad political interactions that link together the
internal changes taking place within individual societies.
The important point to note is that social systems have internal dynamics
responsible for change. This idea is, of course, a direct intellectual descendent
of Marx's view of social evolution as a working out of the internal
contradictions in historically specific modes of production. Thus the inherent
conflicts between feudal lord and merchant underlay the developments that
took European societies out of feudalism. But what causes this transformation?
So-called structural Marxists, such as Godelier (1977) and Friedman (1974,
1975, Friedman and Rowlands 1977), take the weight off Marx's original
insistence that social relationships derive ultimately from economic
re atwns tps, an o t e correspon mg c aracter o power. T e structura
Marxists' view of the dynamic character of human social relations must,
however, be extended to consider what is practical under varying material
conditions.
TOWARDS A SYNTHETIC THEORY OF SOCIAL EVOLUTION

In what follows, I argue that the three schools of thought outlined above each
have important contributions to make to a synthetic theory of social evolution
(see Johnson and Ear le 1987). These contributions and some of the unanswered
questions to which they give rise, are summarized below:
1 The 'technological' schemes identify a dynamic interrelationship between human

population expansion and technological development. But why should this lead to
the development of more complex societies?
Za The 'integrationist' schemes that invoke processes of adaptation identify the ways
in which human organizations function to solve critical problems of survival. But
for any particular problem, multiple organizational solutions seem practicable.
The difficulty with these arguments is not that central leadership, for example,
might not meet the needs of local populations, but that there is no a priori reason
why these needs should be fundamentally different from those of people in
decentralized societies. Adaptationist theories of evolutionary integration have by
949
SOCIAL LIFE
now been extensively criticized, and it has been shown that, by and large, the
adaptive advantage of central management, by contrast to other social solutions, is
unclear (see Earle 1987).
Zb The 'integrationist' schemes that invoke political processes emphasize that control
over populations is the basis for the development of more integrated systems. But
why would a political system expand in the first place, and why should different
political systems, all operated by equally sophisticated strategists, vary in their
capacity to expand and dominate?
3 The 'social structuralist' schemes emphasize that different social systems have
fundamentally different organizational structures and thus have contrasting
developmental dynamics and trajectories. But whence come these different
structures? Are they simply the outcome of historical differences?
A synthetic theory can draw on the strengths of each theoretical scheme, which
complement each other in many respects. To construct such a theory, we should
recognize that every human society depends on a conjunction of subsistence
economy and political economy (Earle 1978, Johnson and Earle 1987:11-15).
The differences emphasized in the 'social structural' schemes do not, I argue,
represent qualitative differences in society, but rather represent different
properties of the two economic spheres of subsistence and politics.
The subsistence economy meets the direct survival needs of a population. Its
character depends on the scale of these needs (reflecting largely the
population's size) and on the availability of resources in its environment which
may e trans orme y t e uman pro uctiVe process. T e ynamtc
relationship between technology and population growth underlies the gradual
expansion in the subsistence economy, subject however to environmental
constraints on technological intensification. Problems of survival may require
the establishment of social networks and group leaders backed by traditional
reciprocal rules of aid and support.
The political economy provides the finance to support emerging elites and
their related institutions. As I have argued for the Hawaiian case (Earle 1978),
unlike the subsistence economy, the political economy is inherently growth-
oriented. In essence, competition for positions ofleadership puts a premium on
the mobilization of resources used to support contending factions. Growth in a
political economy can be very rapid, constrained only by the ability to mobilize
resources from a commoner population. But what confers and limits this
ability? Local elites will actively seek to develop means of control, but control
can also be seen as derived fundamentally from the character of the subsistence
economy.
Long-term intensification of the subsistence economy, of a kind that might
require local management, also creates conditions of control in the political
economy (Johnson and Earle 1987). As Lenski (1966) argues, the
institutionalization of privilege that underlies social complexity derives from a
balance of power and need. Although state coercion never lies far below the
surface, stable systems of domination depend upon the ruling elite's ability to
950
provide (or deny) essential products or services. To show how this works we
need to examine the varying sources of power in chiefdoms and states, and how
these sources offer potentially conflicting and complementary bases for control.
Sources of power
In this section, I argue that four sources of power exist: social, military,
ideological, and economic (compare Mann 1986, Earle 1987). A person's
political position depends on using one or more of these power sources. The
political process involves the selective application of power to control access to
these power sources and thus to weaken potential opposition.
Power relationships, either overt or thinly disguised, underlie the dynamics
of all societies; however, an ability to control access to power can be realized in
the mobilization of resources to finance those institutions of rule on which
complex societies rest Oohnson and Earle 1987). I shall first summarize briefly
the literature on the alternative sources of power and then go on to argue that
the economy is primary, in that it alone permits control over all other sources of
power.
Social power
Characterizing all societies, social power derives from the ability to draw
po 1t1ca support an resources rom c ose m, an most pro a y ta es Its
strength from the intimacy of such kinship bonds. In Yanomamo axe fights, for
example, when interpersonal confrontations arise, factions of close kin form
such that brothers and cousins side with each respective combatant (Chagnon
1983). Similarly, in the political machinations of Big Men, critical to a man's
initial success is the size of his immediate kindred on which he can rely for
support (Oliver 1955). In their model of tribal society, Friedman and Row lands
( 1977) emphasize its kin-based character and the chiefs political strategy to
extend his power through marriage. Each marriage unites a leader with an
affinal group from which he can draw politically. Among the Trobriand
Islanders, the wives that a chief takes enable him to collect affinal gifts which
become a rudimentary form of tribute Oohnson and Earle 1987:216-23; see
Malinowski 1935). Webster (1990) offers a recent review of power relations in
the chiefdoms of Africa and prehistoric Europe, which also emphasizes their
personal, kin-based nature.
Military power
Military power is based on might and intimidation. Gilman ( 1987)
characterizes this as a protection racket in which commoners must give to the
elites what they demand, or face reprisal. Certainly military might both
maintains and extends political control. In the complex Hawaiian chiefdoms, a
951
SOCIAL LIFE
military cadre was supported by the paramount chief, who used it to conquer
new lands and peoples and to retain control internally (Earle 1978). The
warriors of the Polynesian chiefdoms (Sahlins 1958), and of European
medieval society, were the most direct instruments of oppression. Control over
the manufacture and ownership of the weapons of war could form a basis for
political domination (Childe 1951 ). Traditional African chiefdoms and states,
for example, show how control over the technology of war can translate into a
monopoly offorce (Goody 1971).
Ideological power
Another important source of power is a society's ideology. Chiefs and kings
maintain domination through perpetuating the belief that their superiority is
part of the natural order, sanctioned by superhuman powers and authority.
This is done by hosting ceremonies that present the legitimate ascendancy of
the leaders, frequently grounding that ascendent position in history and
genealogy. For example, in the Merina state of Madagascar (Bloch 1989) and
the Mapuche chiefdoms of Chile (Dillehay 1990), leaders use ceremonial
occasions at burial monuments to proclaim the legitimate genealogical basis for
their rule. One is drawn to the parallels in Neolithic and Bronze Age Europe,
where the monumental burial grounds of chiefs dominated both the physical
and the political landscape. In the complex Hawaiian chiefdoms, rulers were
not stmp y ea ers; t ey were go s w o ru e y re tgwus y sanctwne
authority (Earle 1978, 1991). Through dress and ceremony, chiefs identified
their status with that of the gods. Most impressively, in chiefdoms and states,
ceremonial and political spaces were organized according to a cosmic order that
created a celestial stage on which leaders acted out their sanctified roles (see
Krupp 1983). This theme is exemplified in Geertz's (1980) notion of the
nineteenth-century Balinese state as a theatre, and by Fritz's (1986) analysis of
spatial organization in a medieval Indian capital. Individuals thus ruled not by
might but by their sanctified place in a universal order.
Economic power
Economic power derives from control over the production and distribution of
necessary goods. These goods may be either staple supplies or valuables.
Staples, such as food and clothing, are goods required by all for subsistence;
by contrast, a society's valuables, such as the items used in marriage
payments or political displays, are necessary for establishing personal (and
group) standing.
The actual sources of economic power are quite variable, depending on how
control is exercised by a ruling elite. In the establishment of institutions
supporting political integration, leaders must assemble the goods needed to
compensate political supporters and others who work for them. Essentially this
952
is a problem of institutional finance for which goods must be mobilized and

distributed strategically. In traditional societies, I have distinguished two forms
of mobilization-staple finance and wealth finance (D' Altroy and Ear le 1985,
Brumfiel and Earle 1987). Staple finance involves the mobilization of foods,
typically by virtue of ownership of the land, followed by their disbursement to
provide subsistence for supporters. Wealth finance involves procuring valuables
through exchange or sponsored manufacture, which are then distributed as a
political currency.
Staple and wealth finance are linked to strategies of economic manipulation
involving, respectively, feasting and resource ownership, and long-distance
exchange. Each strategy has characteristics of stability and centrality that affect
the dynamics of political evolution.
Feasting is, to different degrees, important in societies of highly variable
social complexity. Friedman and Rowlands (1977) show how leaders in 'tribal'
societies amass foods to host feasts. The success of an individual leader and his
supporters is measured by the scale of the feast and its associated gift
exchanges. Thus among the Mae Enga of New Guinea, a Big Man builds his
political position by hosting feasts in the regional te exchange cycle; success in
these feasts establishes an individual's prestige and translates directly into his
ability to attract additional political supporters, marriage partners, and allies
Oohnson and Earle 1987:183-6, Meggitt 1974).
What is described as the redistributional economy of chiefdoms
c aractensttca y mvo ves t e mo 1 tzatwn an tstn utwn o oo an wea t
in annual ceremonies (see, for example, Sahlins 1958 on Polynesian chiefdoms).
Even in complex states, like the Inca empire, major annual ceremonies were
among the most prominent and economically significant of the events staged at
political centres.
These feasts are a rudimentary and often composite form of institutional
finance. Subsistence goods are mobilized through various means such as
personal ties to Big Men, first-fruit obligations to chiefs, and more directly
controlled systems of staple production. Wealth items are most commonly
obtained through intergroup exchange dominated by leaders and channelled
through ceremonial exchange. For example, a Trobriand chief collects goods
through extended kin relationships and political patronage, and then invests
these goods by hosting competitive feasts that reflect directly on both his
prestige and the renown of his political supporters Oohnson and Ear le 1987).
Payment to supporters is thus direct, in the form of ceremonial food and wealth
distributions, and indirect, in the form of increased personal status deriving
from association with a successful leader.
One way to stabilize economic control is through the assertion of ownership
of the means of subsistence production. In simpler chiefdoms this is often
manifested as an overarching system of land tenure in which the chief is
considered to be ultimately the owner of all lands by virtue of his religiously
sanctioned position as the focal point for the polity. This is the case, for
953
SOCIAL LIFE
example, among the Bemba of Central Africa (Richards 1939). Social

anthropologists frequently emphasize that chiefs control labour by calling on
kin relationships. However, although the kinship structure is certainly a basis
for the recruitment of support, the ability to centralize kinship responsibilities
around the chief (as opposed to anyone else in the society) is based on reputed
ownership and rights to allocate productive resources (contra Sahlins 1972). For
example, the ceremonial and burial monuments of the European Neolithic and
Early Bronze Ages can be interpreted as chiefly assertions of ownership over
political territories (Ear le 1991 ). By funding the construction of monuments,
chiefs transformed the landscape into spaces associated with particular
genealogical lines and with the performance of ceremonies that they themselves
financed. Thus, chiefs ground their structural power (derived from a
hierarchical kinship system represented by the burials) and their ideological
power (derived from imputed connections to superhuman forces) in a
constructed and owned landscape.
Direct ownership by a ruling elite, however, becomes most explicit in
situations where the landscape is developed and divided, as for example in the
case of irrigated lands. In the complex protohistoric chiefdoms of the Hawaiian
Islands, chiefs developed tracts of irrigated land and then bestowed use rights
in specific plots upon commoners in return for the latter's labour on land
producing directly for the chief (Earle 1978). In this example of a system of
staple finance, ownership is used to mobilize subsistence goods that were then
use to support t e c te s, t etr speCia tsts, an easts.
Staple finance, based on ownership of irrigated and other developed field
systems, probably supports many complex societies. Gilman (1981, 1987)
suggests that the evolution of chiefdoms in prehistoric Europe was based on
ownership of developed lands, especially in south-eastern Spain, where
irrigation was practised. Where social complexity has already emerged, the
subsequent abandonment of ceremonial monuments and the development of
elaborately laid-out field systems may indicate the establishment of direct land
ownership by a ruling elite (Ear le 1991 ). By establishing direct control over
subsistence production, elites trade off access to land for corvee labour or a
proportion of commoner production.
After feasting and resource ownership, a third source of economic power lies
in control over exchange. Exchange in wealth objects (social valuables) is
facilitated by the fact that the objects are easily moved over great distances. This
is because, in comparison to staples, they have a very high value for their bulk
and weight. Therefore, the production and movement of wealth can be
centralized and controlled over a broader region (D' Altroy and Ear le 1985). In
addition to their use in marriage payments and similar social transactions,
wealth objects-if they come from afar-may bespeak exotic and esoteric
knowledge not available to local commoner populations (Helms 1979). Thus
foreign wealth is associated with special ritual knowledge that legitimizes the
elevated status of chiefs and links them with the gods.
954
Chiefs and kings control exchange in wealth by virtue of their foreign

contacts (Friedman and Rowlands 1977). In their role as foreign diplomats,
chiefs can maintain the trade partnerships through which their wealth is
obtained. Through ownership of the means of transportation, chiefs can also
monopolize participation in foreign exchanges. For example, the Trobriand
chiefs were able to dominate the exchange of kula objects by virtue of their
ownership of sea-going canoes. Burton (1975) argues that the unusual
development of social complexity in the Trobriand Islands, by comparison with
neighbouring Melanesian societies, is an outcome of the Islands' marginal
position in the kula exchange system. This system connects many islands off
the eastern tip of New Guinea; most of these islands are relatively close to each
other such that inter-island trade is comparatively easy for anyone to engage in,
and thus difficult to control. The Trobriand Islands, by contrast, are more
isolated than others in the system, and large sea-going canoes are needed to
participate in the competitive external exchanges. By owning these canoes,
chiefs can control the exchange.
A synthetic model of power and control

Are the four sources of power alternative foundations for political
development? To some degree, they are. From my review of the emergence of
complexity in chiefdoms, it is evident that different societies have maintained
compara e eve s o po 1t1ca comp extty on t e asts o qmte 1 erent sources
of power (Earle 1987, 1989). In his archaeological comparison of the
development of chiefdoms in Colombia, Panama, highland Mexico, and coastal
Veracruz in Mexico, Drennan (1991) emphasizes the individual character of
each sequence; power is manifested quite differently in each case, and these
differences are reflected in the developmental trajectories involved. Socio-
political change can, in at least some situations, be seen as an outcome of
factional competition, in which each faction draws its power from a different
source (Bradley 1991 ).
The different sources of power are, however, not equivalent, and they cannot
be conceived to be independent of one another. They are not equivalent
because emerging elites cannot control them equally. Thus it is not power per se,
but the ability to control potential power, that is the crucial factor in
understanding the evolution of social complexity.
Social relationships constitute a web in which any individual is in the centre
of his or her personal network. To try to extend your social support group is
thus to draw on relations of increasing genealogical distance, in which the
kinship bond is progressively weaker. Any polity based on kinship will harbour
a constant tendency towards fission (Sahlins 19 58). This was the case, for
example, with Polynesian chiefdoms, which were structured on the basis of
kinship ranking. When these chiefdoms expanded, they split into independent
polities. One would hardly expect a junior line to have retained its allegiance to
955
SOCIAL LIFE
a senior line, when the expansion of the latter structurally undermined the
former's rank and political position. Consequently, it was common for a junior
line to break away and to form its own polity (Sahlins 19 58, Ear le 1978). It is
difficult to see, then, how kinship could form the bedrock of a stable
arrangement of political power.
Military power, in its application, is manifested as naked force. Although no
complex society can exist without it, military force is inherently difficult to
control. As proverbial wisdom has it, 'He who lives by the sword dies by the
sword.' The military cadre on which a leader depends is often his greatest
threat because its force can be quickly turned against him. What gives a
monopoly of force and allows that force to be controlled?
Ideological power derives from an accepted notion of order,
characteristically backed by religious sanction. But what limits access to
esoteric knowledge and religious sanctity? Cannot anyone-a new shaman,
priest, prophet, or a man on a soapbox-claim to have direct communication
with the gods and create a new religious order? Tradition may constrain what
can be done and said, but in this respect it can be used as much against
centralizing power as to support it.
Power can be an equalizing force. It is used not only to dominate, but also to
resist domination. Complex societies are especially complicated because the
competing sources of power are continually dissolving centralization (Mann
1986). Modern state society may actually 'devolve' as the multiplicity of the
sources o power ma es po 1t1ca centra tzatwn tmpracttca .
The evolution of complex social systems, while certainly encompassing
complicated and conflicting power relationships, is fundamentally based on
control over material conditions, which in turn permits control over the other
sources of power. Economic power alone provides the stability that allows for
the creation and extension of politically centralized societies. It does this
because of the ease with which economic processes can be controlled and used
to control the other sources of power.
Economic forces can be controlled by restricting access to the means of
production and distribution. In evolutionary development, the intensification
of production increases the ease with which control can be established, by
gradually replacing labour with technology as the critical limiting factor. For
example, with a shift to irrigated agriculture, improved lands become centrally
important, and access to these improvements can be regulated by an emerging
elite (see Earle 1978). Economic power becomes increasingly centralized as
income from owned facilities is used to finance further economic development
with the construction of agricultural facilities, the attachment of specialists,
and the development of transport technology.
The products of the economic system can also be reinvested materially in
control over the other sources of power. A chiefs kinship network is extended
by polygamous marriages secured through rich gift exchanges (Friedman and
Rowlands 1977). Military forces are controlled by providing material support
956
to the cadre and by control over the manufacture and importation of their
weaponry (Goody 1971 ). Ideological power is controlled by the substantial
capital required to finance religious institutions and the spectacular ceremonies
of legitimation.
The primary dynamic in the evolution of complex society lies in an intensely
competitive political arena (Earle 1978, Johnson and Earle 1987). Survival in
that arena depends on astute strategies on the part of individual leaders in
manipulating their investments in the alternative sources of power and in
mechanisms for establishing control. Thus within the political arena there is a
social process of leadership selection; at times a leader's success centralizes his
polity, but miscalculation can as quickly lead to collapse.
Chiefdoms are characteristically cyclical. For example, the prehistoric
Mississippian chiefdoms of the South-eastern United States were never stable;
they expanded and declined rapidly as different localities rose and fell from
political dominance (Anderson 1990). The different bases of political power
were continually tested and the ability to maintain and extend domination
formed the foundation for political development. With the emergence of states,
the frequency of cycling may be reduced by increasingly centralized and
institutionalized control; nevertheless the rise of states anticipates their
eventual fall (see, for example, Khazanov 1984 ).
Elites must continuously seek out mechanisms of domination. These may
include the establishment of a police force and of religious institutions. The
economy may e systemattca y mampu ate to mcrease t e epen ency o t e
peasantry. However, stability in control may equally be the outcome of long-
term changes in the subsistence economy that make commoners dependent on
the ruling elite for necessary goods and services that cannot be obtained
independently.
Successful systems of domination are characterized by the intertwining of
the sources of power and control. Income from a growth-oriented political
economy is invested in economic expansion, political alliances, military
support, and religious extravaganzas. Thus economic dependence, social
relationships, naked force, and sacred legitimacy are continually bound up with
one another. The binding thread is the economic flow of resources. Material
wealth begets both more wealth and political control.
NOTE
1 Major syntheses abound, including Morgan (1977), Marx (1904), Engels (1972 [1884]),
Spencer (1967), Childe (1936, 1951), Steward (1955), Service (1962), Wittfogel (1957),
Lenski (1966), Cameiro (1970), Fried (1967), Harris (1977), Glassman (1986), Mann
(1986), and Johnson and Earle (1987). Some excellent histories of social evolutionary
theory are also available, among them Lenski (1966), Harris (1968), Service (1975,
1978).
957
SOCIAL LIFE
REFERENCES
Anderson, D. (1990) 'Political change in chiefdom societies: cycling in the late
prehistoric Southeastern United States'. PhD dissertation, Department of
Anthropology, University of Michigan.
Bloch, M. (1989) Ritual, History and Power, London: Athlone Press.
Boserup, E. (1966) The Conditions of Agricultural Growth: the Economics of Agrarian
Change Under Population Pressure, Chicago: Aldine.
Bradley, R. (1991) 'Ideology and economy in the prehistory of Southern England', in
T.Earle (ed.) Chiefdoms: Power, Economy, and Ideology, Cambridge: Cambridge
University Press.
Braun, D. and Plog, S. (1982) 'Evolution of "tribal" social networks theory and
prehistoric North American evidence', American Antiquity 43:504-25.
Brumfiel, E. (1976) 'Regional growth in the eastern Valley of Mexico: a test of the
"population pressure" hypothesis', in K.Flannery (ed.) The Early Mesoamerican
Village, New York: Academic Press.
--(1980) 'Specialization, exchange and the Aztec state', Current Anthropology 21:
459-78.
Brumfiel, E. and Earle, T. (1987) 'Introduction', in E.M.Brumfiel and T.K.Earle (eds)
Specialization, Exchange, and Complex Societies, Cambridge: Cambridge University
Press.
Burton, R. (1975) 'Why do the Trobriands have chiefs?', Man (N.S.) 10:544--58.
Carneiro, R.L. (1970) 'A theory of the origin of the state', Science 169:733-8.
--(1981) 'The chiefdom as precursor of the state', in G.Jones and R.Kautz (eds) The
Chagnon, N. (1983) Yanomamij; the Fierce People, New York: Holt, Rinehart &
Winston.
Childe, V. G. (1936) Man Makes Himself, London: Watts.
--(1951) Social Evolution, London: Watts.
Cohen, M.N. (1977) The Food Crisis in Prehistory: Overpopulation and the Origins of
Agriculture, New Haven: Yale University Press.
D' Altroy, T. and Earle, T. (1985) 'Staple finance, wealth finance, and storage in the Inca
political economy', Current Anthropology 26:187-206.
Dalton, G. ( 1977) 'Aboriginal economies in stateless societies', in T. Earle and J. Ericson
(eds) Exchange Systems in Prehistory, New York: Academic Press.
Dillehay, T. (1990) 'Mapuche ceremonial landscape, social recruitment and resource
rights', World Archaeology 22:223-41.
Drennan, R. (1991) 'Prehispanic chiefdom trajectories in Mesoamerica, Central
America, and Northern South America', in T.Earle (ed.) Chiefdoms: Power,
Economy, and Ideology, Cambridge: Cambridge University Press.
Durkheim, E. (1933 [1893]) The Division of Labor in Society, New York: Free Press.
Earle, T. (1977) 'A reappraisal of redistribution: complex Hawaiian chiefdoms', in T.
Earle and J.Ericson (eds) Exchange Systems in Prehistory, New York: Academic Press.
--(1978) Economic and Social Organization of a Complex Chiefdom: the Hale/ea
District, Kaua'i, Hawaii, University of Michigan, Anthropological Papers, vol. 63,
Ann Arbor: University of Michigan.
--(1987) 'Chiefdoms in archaeological and ethnohistorical perspective', Annual
Review ofAnthropology 16:279-308.
958
--(1989) 'The evolution of chiefdoms', Current Anthropology 30:84-8.

--(1991) 'Property rights and the evolution of chiefdoms', in T.Earle (ed.)
Chiefdoms: Power, Economy, and Ideology, Cambridge: Cambridge University Press.
Engels, F. (1972 [1884]) 'The origin of the family, private property and the state', in
K.Marx and F.Engels (eds) Selected Works in One Volume, New York: International
Publishers.
Evans-Pritchard, E.E. and Fortes, M. (eds) (1940) African Political Systems, London:
Oxford University Press.
Feinman, G.M. and Neitzel, J. (1984) 'Too many types: an overview of sedentary
prestate societies in the Americas', in M.Schiffer (ed.) Advances in Archaeological
Method and Theory, vol. 7, New York: Academic Press.
Fried, M.H. (1967) The Evolution of Political Society: an Essay in Political Economy,
New York: Random House.
Friedman, J. (1974) 'Marxism, structuralism and vulgar materialism', Man (N.S.) 9:
444--69.
--(1975) 'Tribes, states and transformations', in M.Bloch (ed.) Marxist Analyses and
Social Anthropology, New York: Malaby Press.
Friedman, J. and Rowlands, M.J. (1977) 'Notes towards an epigenetic model of the
evolution of "civilization"', in ].Friedman and M.Rowlands (eds) The Evolution of
Social Systems, London: Duckworth.
Fritz, J. (1986) 'Vijayanagara: authority and meaning of a South Indian imperial
capital', American Anthropologist 88:44--55.
Geertz, C. (1980) Negara: the Theater State in Nineteenth-Century Bali, Princeton:
Princeton University Press.
1 man, e eve opment o soCia strati catiOn m ronze ge urope ,
Current Anthropology 22:1-24.
--(1987) 'Unequal development in Copper Age Iberia', in E.M.Brumfiel and
T.K.Earle (eds) Specialization, Exchange, and Complex Societies, Cambridge:
Glassman, R. (1986) Democracy and Despotism in Primitive Societies: a Neo-Weberian
Approach to Political Theory, New York: Associated Faculty Press.
University Press.
Goody, J. (1971) Technology, Tradition and the State in Africa, Oxford: Oxford
University Press.
Haas,J. (1982) The Evolution ofthe Prehistoric State, New York: Columbia University Press.
Harris, M. (1968) The Rise of Anthropological Theory: a History of Theories of Culture,
--(1977) Cannibals and Kings: the Origins of Cultures, New York: Vintage Books.
Helms, M.W. (1979) Ancient Panama: Chiefs in Search of Power, Austin: University of
Texas Press.
Johnson, A. and Earle, T. (1987) The Evolution of Human Societies: From Foraging
Group to Agrarian State, Stanford: Stanford University Press.
Khazanov, A.M. (1984) Nomads and the Outside World, trans. J.Crookenden,
Kristiansen, K. (1991) 'Chiefdoms, states and systems of social evolution in Northern
Europe', in T.Earle (ed.) Chiefdoms: Power, Economy, and Ideology, Cambridge:
959
SOCIAL LIFE
Krupp, E.C. (1983) Echoes ofthe Ancient Skies: the Astronomy of Lost Civilizations, New
York: Harper & Row.
Lee, R.B. and DeVore, I. (eds) (1968) Man the Hunter, Chicago: Aldine.
Lenski, G.E. (1966) Power and Privilege: a Theory of Social Stratification, New York:
McGraw-Hill.
Maine, H.S. (1861) Ancient Law, London: John Murray.
Mair, L. (1962) Primitive Government, Harmondsworth: Penguin.
Malinowski, B. (1935) Coral Gardens and their Magic: a Study ofthe Methods of Tilling the
Soil and ofAgricultural Rites in the Trobriand Islands, New York: American Book Co.
Malthus, T. (1798) An Essay on the Principle of Population, London: Johnson.
Mann, M. (1986) The Sources of Social Power, vol 1: A History of Power from the
Beginning to A.D. 1760, Cambridge: Cambridge University Press.
Marx, K. (1904) A Contribution to the Critique of Political Economy, trans. N.I.Stone,
Chicago: Charles H.Kerr.
Mauss, M. (1954) The Gift: Forms and Functions of Exchange in Archaic Societies,
London: Cohen & West.
Meggitt, M. (1974) 'Pigs are our hearts!: the te exchange cycle among the Mae Enga of
New Guinea', Oceania 44:165-203.
Morgan, L.H. (1877) Ancient Society, Chicago: Charles H.Kerr.
Oliver, D.L. (1955) A Solomon Island Society: Kinship and Leadership Among the Siuai of
Bougainville, Cambridge, Mass.: Harvard University Press.
Polanyi, K. (1944) The Great Transformation, New York: Farrar & Rinehart.
Richards, A.I. (1939) Land, Labour and Diet in Northern Rhodesia, London: Oxford
University Press.
a Ins,
Press.
--(1963) 'Poor man, rich man, big man, chief: political types in Melanesia and
Polynesia', Comparative Studies in Society and History 5:285-303.
--(1968) Tribesmen, Englewood Cliffs, NJ: Prentice-Hall.
--(1972) Stone Age Economics, London: Tavistock.
Sanders, W.T. and Webster, D. (1978) 'Unilinealism, multilinealism, and the evolution
of complex societies', in C.L.Redman, M.J.Berman, E.V.Curtin, WT.Langhorne,
Jr, N.M.Versaggi and J.C.Wanser (eds) Social Archaeology: Beyond Subsistence and
Dating, New York: Academic Press.
Service, E. (1962) Primitive Social Organization: an Evolutionary Perspective, New York:
Random House.
--(1975) Origins of the State and Civilization: the Process of Cultural Evolution, New
York: Norton.
--(1978) 'Classical and modern theories of the origins of government', in R.Cohen
and E. Service (eds) Origins of the State: the Anthropology of Political Evolution,
Philadelphia: Institute for the Study of Human Issues.
Spencer, H. (1967) The Evolution of Society: Selections from Herbert Spencer's Principles
of Sociology, ed. R.L.Carneiro, Chicago: University of Chicago Press.
Steward, J. (1955) Theory of Culture Change, Urbana: University of Illinois Press.
--(1977) 'The foundations of Basin-Plateau Shohonean society', in ].Steward and
R.Murphy (eds) Evolution and Ecology, Urbana: University of Illinois Press.
Webster, G. (1990) 'Labor control and emergent stratification in prehistoric Europe',
960
White, L. (1943) Energy and the evolution of culture, American Anthropologist 45: 335-56.
--(1959) The Evolution of Culture, New York: McGraw-Hill.
Williams, B.J. (1974) 'A model of band society', Society for American Archaeology
Memoirs no. 29, Washington, DC: Society for American Archaeology.
Wittfogel, K. (1957) Oriental Despotism, New Haven: Yale University Press.
Wright, H. (1984) 'Prestate political formations', in T.Earle (ed.) On the Evolution of
Complex Societies, Malibu: Undena Publications.
Yellen, J. and Harpending, H. (1972) 'Hunter-gatherer populations and archaeological
inference', World Archaeology 4:244--53.
FURTHER READING
Carneiro, R.L. ( 1981) 'The chiefdom as precursor of the state', in G.Jones and R.
Kautz (eds) The Transition to Statehood in the New World, Cambridge: Cambridge
University Press.
Cohen, R. and Service, E. (eds) (1978) Origins of the State: the Anthropology of Political
Evolution, Philadelphia: Institute for the Study of Human Issues.
Earle, T. (ed.) (1991) Chiefdoms: Power, Economy, and Ideology, Cambridge: Cambridge
University Press.
Fried, M.H. (1967) The Evolution of Political Society: an Essay in Political Economy,
Friedman, J. and Rowlands, M. (eds) (1977) The Evolution of Social Systems, London:
Duckworth.
Glassman, R. (1986) Democracy and Despotism in Primitive Societies: a Neo-Weberian
Approach to Political Theory, New York: Associated Faculty Press.
Goody, J. (1971) Technology, Tradition and the State in Africa, Oxford: Oxford
University Press.
Johnson, A. and Earle, T. (1987) The Evolution ofHuman Societies: from Foraging Group
to Agrarian State, Stanford, Cal.: Stanford University Press.
Kirch, P.V. (1984) The Evolution of Polynesian Chiefdoms, Cambridge: Cambridge
University Press.
Lenski, G.E. (1966) Power and Privilege: a Theory of Social Stratification, New York:
McGraw-Hill.
Mann, M. (1986) The Sources of Social Power, vol 1: A History of Power from the
Beginning to A.D. 1760, vol. 1, Cambridge: Cambridge University Press.
Sahlins, M. D. (1972) Stone Age Economics, London: Tavistock.
Service, E. (1962) Primitive Social Organization: an Evolutionary Perspective, New York:
Random House.
--(1975) Origins of the State and Civilization: the Process of Cultural Evolution, New
York: W.W.Norton.
Steward, J.H. (1955) Theory of Culture Change, Urbana: University of Illinois Press.
Upham, S. (ed.) (1990) The Evolution of Political Systems: Sociopolitics in Small-Scale
Sedentary Societies, Cambridge: Cambridge University Press.
Webster, G. ( 1990) 'Labor control and emergent stratification in prehistoric Europe',
White, L.A. (1959) The Evolution of Culture, New York: McGraw-Hill.
Wright, H. (1984) 'Prestate political formations', in T.Earle (ed.) On the Evolution of
Complex Societies, Malibu: Undena Publications.
961
35
LAW AND DISPUTE PROCESSES

Simon Roberts
INTRODUCTION
A sociology of those specialized, differentiated arrangements which we would
unambiguously label 'law' in the contemporary West is in itself problematic.
Law lays claim to a dual character: it furnishes the normative 'map' informing
the life-world of a society's members as they experience it; and it provides one
of the central means through which government exercises a steering role.
Hence a sociology of law must be concerned with commonly accepted
stan ar s an wtt Impose regu atwn, wtt t e omams o or er' an o
'domination'. Thus the ambition must be to keep these domains analytically
distinct, without losing sight of the strands which undoubtedly connect them.
Whatever these difficulties, law's robustly self-defined character at least
provides the 'folk' categories upon which a sociological analysis of 'norms' and
of 'government' can be brought to bear. But this quality at once poses a
problem when we try to imagine what an anthropology of law might be. The
very concept of 'law', with its claimed separation of the cognitive and
normative domains, its identification with a discrete sphere of the 'ought', may
not always find counterparts in the small-scale and technologically simple
societies which anthropologists have traditionally studied. The institutional
arrangements which we associate with law in the West-the differentiation of
legal norms; a specialized judiciary within a compartmentalized, self-conscious
governmental structure; the emergence of a legal profession-are all specific to
a particular socio-political context. Even in functional terms, law's almost
inextricable identification with 'government', the exercise of a steering role,
raises problems as soon as we move beyond the bounds of the sovereign state.
These concerns, which are surely of a different order from those associated
with marking out such broad, general categories as kinship, politics, economics
and religion, have not inhibited the growth of legal anthropology. Despite an
important shift in perspective, the interest of nineteenth-century scholars in
'primitive law' survived the transition into modern anthropology through
962
Malinowski's early monograph, Crime and Custom in Savage Society (1926). At

the same time, Radcliffe-Brown confidently identified 'law' as one of the
principal compartments into which anthropological studies should be divided,
and in his important essay 'On social structure' ( 1940) law appears as a separate
and privileged element in the proposed 'social physiology'. Subsequently, some
of the leading Anglo-American anthropologists of the next generation made
their names with books about law; and today, the writings of legal
anthropologists provide one inspiration for a new jurisprudence in the West,
enlarging the realm of legal studies to embrace formerly 'suppressed
discourses' of 'non-state law' (Cotterrell 1983, Fitzpatrick 1992, Teubner
1992).
While 'law' has thus provided a durable label, the appearance of continuity
in anthropological interest is deceptive. Looking back, what we see is an
unbroken succession of quite different 'anthropologies of law'. In the mid- and
later nineteenth century 'primitive law' featured prominently in efforts to
characterize, and provide an ancestry for, 'modernity'. Under Malinowski and
Radcliffe-Brown, these evolutionary studies were replaced by an anthropology
of 'order'. After the Second World War, legal anthropology became the study of
dispute processes. This focus gave way in turn to a new legal anthropology
which examined the part played by law in the imposition of colonial
domination, and which has now itself been transformed into a 'legal pluralism'
which cuts across boundaries between the anthropologies and sociologies of
aw. vwus y, t e a ove compartments ave not een waterttg t, an t e
sediments of these successive anthropologies of law may suggest the elements
of what legal anthropology might become. In what follows, however, I attend to
them individually, in the order of their appearance.
PRIMITIVE LAW AND THE CHARACTERIZATION OF

MODERNITY
An enduring source of interest in 'primitive law' lay in the ambition of classical
and recent social theorists to characterize the condition of modernity. In the
course of this project, 'tradition' was invoked both as a means of highlighting
modernity through contrast and as an aid in reconstructing the route along
which we (in the West) have travelled to the present. The widespread
invocation of an opposition between 'tradition' and 'modernity' may also
conceal shifting levels of focus: upon differences between traditional and
modern persons; upon the diverse ways in which traditional and modern
societies 'hold together'; upon contrasting features of traditional and modern
authority. Thus the opposition may be located at the levels of action, order or
domination and these levels may be mutually entangled. At the level of
domination, a major focus of interest has been on the origin and development
of the state. In these studies, 'law' is deployed in varied ways; sometimes it is
itself the focus of attention, as it was for Maine in Ancient Law (1861),
963
SOCIAL LIFE
sometimes it is invoked as a means of understanding society, as in Durkheim's

De la Division du travail social (1893).
In Ancient Law, and in his later Dissertations on Early Law and Custom
(1883), Maine mapped out a broad transition from small, kin-based groups to
larger, territorial units. While the famous 'status to contract' formulation
suggests a concern with different foundations of social solidarity, his history is
one of 'government' rather than 'order'. For Maine, the story of society is a
story of decision-making. The very origin of social life is identified in the
steering role exercised within a group of kin by the senior male agnate. These
old patriarchs made decisions on an ad hoc basis; no consistent rules
underpinned the decisions they took, yet government was supposed to be by
adjudication by the senior male, before whom all disputes were brought.
In the form of society that followed, collections of these small groups of
agnates became clustered together under chiefs, but the (sometimes fictional)
assumption of shared kinship remained the basic organizing principle. Then
came the territorial stage, in which members identified themselves through
their common occupation of a defined tract of land, rather than through
kinship. Around the end of the second stage and the beginning of the third,
'law' developed as rulers began to pronounce the same judgments in similar
situations, providing their decision-making with an underlying set of rules.
Later in the development of territorially based societies, the settlement of
disputes fell into the hands of a specialized elite, who alone had access to the
prmctp es to e o owe m t eu reso utwn. s Mame wrote: W at t e
juristical oligarchy now claims is to monopolize the knowledge of the laws, to
have the exclusive possession of the principles by which quarrels are decided'
(1861:7). There followed the 'era of codes', and so on, but we can leave the
developmental process at this stage.
Several important features are clear from this summary. First, for Maine
there were no structural changes in the process of dispute settlement over the
three fundamental stages of societal development. From the senior male agnate
onwards, disputes were resolved by decision, handed down by a third party;
there was no suggestion of negotiatory modes of settlement giving way to
processes of third-party adjudication. Secondly, the presence of a normative
basis for decision-making was the key attribute of law for Maine, and the
emergence of this feature heralded the transition from the pre-legal to the legal
world. Thirdly, there was the later development of specialization as legal rules
became separated off from other rules operating in society. Thus for Maine,
social life is the product of 'government', law develops in the course of that
process, and the fundamental way in which kingly power is revealed is through
adjudication.
In examining the foundations of social order the classical sociologists,
writing towards the close of the nineteenth century, continued to make use of
explicit oppositions between 'tradition' and 'modernity'. On one level the
writings of Durkheim and Weber can be seen as a bridge between scholars like
964
Maine and Morgan on the one hand, and modern social anthropology on the
other. They posed a 'problem of order' in terms which are recognizably the
same as those in which it was addressed by Radcliffe-Brown and Malinowski.
But they remain remote in that their central interest, like that of their
predecessors, was in understanding modernity; the past, along with
contemporary examples of the primitive, was still invoked in the project of
getting to grips with the present. They are also remote in their partly concealed
presupposition, reinforced by the poor quality of the ethnography then
available, that governmental action is an inevitable concomitant of life in a
social world.
In opposing 'mechanical' and 'organic' solidarity in De la Division du travail
social (1893), Durkheim purports to elucidate the different ways in which
traditional and modern societies hold together rather than to examine the
nature of governmental action. But the use he makes of law in this discussion
reveals a conflation of the problems of order and of domination. In arguing that
the predominance of 'repressive' sanctions can provide us with a criterion for
identifying societies characterized by mechanical solidarity, and by similarly
linking 'restitutive' sanctions with organic solidarity, primitive societies were
credited with regimes of criminal law, and hence with mechanisms of
adjudication and coercive governmental action.
In his Economy and Society (1978 [1917]) Weber invokes an opposition
between tradition and modernity primarily at the level of government, rather
t an at t e eve o society. T IS oppositiOn IS use to e uci ate t e I erent
kinds of legitimacy claims made by traditional and modern (rational-legal)
authorities, and in an examination of the underpinnings of traditional and
modern forms of adjudication. For Weber, 'law' was a creature of the modern
world, linked to the application of general rules, and served to differentiate
bureaucratic government and specialized, rule-based adjudication from their
'traditional' forerunners.
Although an assumption that developed law is an achievement of the
modern world is implicit in a great deal of English legal anthropological
writing, explicit interest in legal evolution had fallen away by the 1920s. In
North America, on the other hand, this interest was sustained in such works as
Hoebel's Law of Primitive Man (1954), Redfield's influential essay on
'Primitive Law' (1967) and Newman's Law and Economic Organisation (1983).
All three works search through the ethnographic record for the pre-legal and
the proto-legal, mapping out with anthropological findings the path along
which law has evolved.
While processes of state formation have now become a source of renewed
interest among social theorists (see Giddens 1986, Mann 1986), 'law' has not
yet found a prominent place in these discussions. It is perhaps surprising that
no one has pursued in detail Maine's tantalizing aside, in his Dissertations on
Early Law and Custom, to the effect that the origins of adjudication are
intimately linked to those of kingship (1883:160). But Bloch's recent account of
965
SOCIAL LIFE
the formation and expansion of the Merina kingdom in Madagascar is entirely

consistent with the idea that 'law', as a differentiated corpus of regulations, is
best seen as a by-product of the business of rule (Bloch 1971 ). Initially brought
into being as 'custom', in the sense of an undifferentiated repertoire of
communal understandings, it is eo-opted by rulers seeking to establish and
consolidate their ascendancy. Bloch presents a scenario of would-be kings
seeking to associate themselves with the traditional norms of the acephalous
rice-growing Merina communities of the valleys, and ultimately presenting
themselves as the source of these norms, at the same time seeking to play an
adjudicatory role in local-level dispute processes.
THE PROBLEM OF ORDER

From the 1920s, a second anthropology of law began to develop. It appeared as
part of the general changes which were noticeable in anthropology at that time,
with a shift in attention away from invoking 'the primitive' as a means of
characterizing modernity, and away from interest in larger questions having to
do with change and historical development. There was also an explicit reaction
against attempts to understand particular features of the culture under
observation in terms of survivals from some earlier 'stage'; every institution was
rather to be understood in terms of its contemporary 'function'.
Once sustained attention came to be directed, at close quarters, to those
sma -sea e, re atiVe y simp e societies w IC were oun ng t across t e
colonial world, some new questions arose. It quickly became clear that many of
these societies had no obvious centralized authority, let alone the differentiated
institutional arrangements associated with government in the West. For
observers coming from cultures where 'order' had become linked to the
accomplishments of kingship or some other form of self-conscious
administration, this seemed problematic. Here were societies without 'kings,
courts and constables', as Malinowski put it (1934:lxii); and yet they were not
the savage anarchies which Hobbes had postulated as the inevitable alternative
to the presence of a sovereign. How were the evident coherence and regularity
of these groups to be explained? What held them together? The absence of
explicit governmental arrangements, and of anything looking like a legal
system, placed the 'problem of order' at the top of the agenda. It also raised a
difficulty for observers interested in law. Did these societies have 'law'? If so,
what form did it take and what were its central attributes?
Malinowski's response to the problem of order, at least so far as the
Trobriand Islanders whom he had studied were concerned, was that
compliance with socially approved norms was ensured through the complex of
reciprocal economic obligations which bound members of the society to each
other. Among these relationships the simplest bound together the group of
fishermen who shared a boat on the lagoon. Each of these men carried out a
particular task in manning the boat and the net, and through its performance
966
acquired a right to a share in the catch. Repeated failure to accompany the

fishing expeditions of 'his' boat would deprive a man of his share of the fish.
Another relationship bound the fisherman on the lagoon with an inland
partner, the yam grower. (Both fish and yams were staples of the Trobriand
diet.) The fisherman supplied the inland farmer with fish, and the farmer
supplied the fisherman with yams. If either party persistently failed to honour
his side of the arrangement, he would soon find himself without an essential
element in his overall subsistence budget: no fish, no yams. Malinowski
suggested that while a breakdown of this kind could possibly be endured for a
while, it would over time have such a destructive effect on other relationships
that in the end the recalcitrant partner would be forced back into line or obliged
to live elsewhere. One of the other relationships which could be directly
affected was that between husband and wife. In Trobriand society, instead of
being responsible for feeding his own household-himself, his wife and
children-a man's efforts are directed towards providing for his sister and her
husband and their children, while his own needs in this respect are met by his
wife's brother. It is not hard to see how the breakdown of any one of these
relationships will immediately place the remaining strands in jeopardy. Under
such circumstances the mechanism of enforcement lies within the complex of
relationships itself, and no external sanction is necessary.
Malinowski presented this account of the forces securing the coherence of
the Trobriand social world in the form of an explicit attack on what he saw as
t e conventwna vtew, as represente m t e wor o Dur etm. In De a
Division du travail, Durkheim had claimed that in societies characterized by
'mechanical solidarity' order is secured primarily through a shared repertoire
of common understandings which are comprehensively internalized by the
societies' members-a position which Durkheim partially reiterated in his last
book, Les Formes ilimentaires de la vie religieuse ( 1912), where he described the
inhabitants of the primitive world as more embedded in society than their
modern counterparts. Malinowski ridicules Durkheim for envisaging people in
primitive societies as virtual automata, blindly and unthinkingly complying
with long-standing customs. His vigorous polemic appears to set up an exciting
argument, but the issue is never really joined since Durkheim's discussion is
located at the level of rules and structure, whereas Malinowski's eye is on the
actions and motivations of persons.
Malinowski's contemporary, Radcliffe-Brown, adopted a position much
closer to that of Durkheim. His early The Andaman Islanders ( 1922) had a 'rule-
centred' quality; and although he stressed the need for 'sanctions' to ensure
compliance with rule, implying attention to motivation, his theoretical work
(see especially Radcliffe-Brown 1952) was largely focused at the level of
structure. At first glance, the argument here seems polemical and capable of
ready solution, but while subsequent ethnographies have struggled to achieve a
multi-dimensional quality, they have on the whole revealed the clear imprint of
either a Radcliffe-Brownian or a Malinowskian approach. Overall, it has proved
967
SOCIAL LIFE
difficult to achieve a satisfactory balance of 'rule' and 'practice', to articulate

the level of 'order' with the level of 'action' (Comaroff and Roberts 1981 ).
In retrospect, a striking quality of Crime and Custom in Savage Society is that
Malinowski seems to escape effortlessly from longstanding presuppositions in
social, political and legal theory about the need for certain actors to occupy
positions of command if stability is to be sustained in social life. But these
assumptions were not easily abandoned in legal anthropology. In Kapauku
Papuans and their Law ( 19 58), for example, Pospisil reasserted the idea that the
presence of authorities playing command roles is an essential feature of human
association, and he has subsequently sought to reinterpret the ethnography
which appears to cast doubt upon this proposition. While it would be absurd to
underplay the importance which 'government'-in the form of self-conscious
steering mechanisms-assumes in contemporary polities, questions of 'order',
in the sense of the reproduction of pattern in the social world, remain all too
easily conflated with those of command and domination.
While Crime and Custom inaugurated an anthropology of law which
embraced broad questions of order and social control, and so by-passed
potentially troubling questions about the nature of law and its institutional
location in stateless societies, arguments about the definition of law and the
conditions under which it is to be found have continued. On the whole, the
working definitions of law offered by anthropologists have been influenced
strongly by the predominantly imperative and positivist orientation of Anglo-
mencan ega t eory. In IS entry on aw' m t e e 1t10n o t e
Encyclopedia of the Social Sciences, Radcliffe-Brown explicitly followed Roscoe
Pound, the American jurist, in identifying law as 'social control through the
systematic application of force in politically organized society' (1933:202).
Malinowski immediately responded, in his Introduction to Hogbin's Law and
Order in Polynesia (1934), by reasserting the position he had adopted in Crime
and Custom. Subsequently a divide has remained between those who have held
to institutional definitions derived from Western legal and political theory,
those who have followed Malinowski in adopting a conception of law which
does not distinguish it from social control in general, and those who have
declined to talk about 'law' at all outside the context of the modern state.
These disagreements surfaced in a different form in the 19 50s, in the
context of a celebrated debate between Max Gluckman and Paul Bohannan
(see Nader 1969:337-418). In The Judicial Process among the Barotse (1955),
Gluckman, following Schapera (1938), had made deliberate use of the
linguistic, conceptual and institutional categories of Western law. Bohannan,
for his part, in his study of Justice and Judgment among the Tiv (1957),
claimed that such Western categories are inappropriate for understanding the
legal concepts, procedures and rules of a non-Western culture. Ultimately,
however, the argument raised worries as to the extent to which it is proper to
talk about 'law' at all in a cross-cultural context (Bohannan 19 57:4-6, Moore
1978: 135-48).
968
AN ANTHROPOLOGY OF LAW AS THE STUDY OF

DISPUTE PROCESSES
The 19 50s saw the appearance of a number of major ethnographies built

around case histories of'dispute', and for a couple of decades this work became
a central concern of leading anthropologists in Britain and North America.
Bailey (1960), Bohannan (1957), Fallers (1969), Gluckman (1955), Gulliver
(1963, 1971), Pospisil (1958) and Turner (1957), major figures in post-war
social anthropology, all made or consolidated their reputations with such
ethnographies.
At first glance it would appear that this shift from an anthropology of law
focused on the problem of order to an anthropology of law as the study of
dispute process took place rather abruptly. Yet there was an undercurrent of
continuity. The 'processual' emphasis in most of these studies builds directly
on strands present in Malinowski's work. 'Dispute' also became an obvious
focus in the context of sustained participant observation; the very accessibility
of public quarrels made it likely that they would become the centrepieces of the
research. Disputes were also topical at this moment for another reason. In
Africa, at least, dispute management was one of the main tasks of the
traditional and neotraditional authorities established under 'indirect rule'.
Whether or not the agency under observation was directly drawn into colonial
local government, 'disputes' were flagged for special attention by that system.
For a maJonty o t e aut ors mentwne a ove, a consciOus ec1s10n to ocus
on disputes was also influenced by the example of an earlier work, The
Cheyenne Way (1941 ). This book, on a North American Indian group, was the
product of a collaboration between Karl Llewellyn, a law professor, and
Adamson Hoe bel, a social anthropologist. Their focus on 'trouble cases' flowed
from Llewellyn's commercial law teaching and research at Columbia Law
School. A member of what is now labelled the Realist School, Llewellyn
thought that law was best approached and understood from the study of
superior court litigation: the 'cases' which were the product of such processes
were the central materials with which law teachers and their students worked.
Confronted with Hoebel's desire to study Cheyenne law, it seemed natural to
suggest that he should look for the equivalents of these cases in the Cheyenne
context. In the end, the book they wrote together was constructed around
remembered case histories from the Cheyenne past, recalled for Hoebel by
elderly Cheyenne informants. According to Llewellyn and Hoebel, it initially
proved difficult to get their informants to understand what it was they wanted
from them; but in the end a series of 'trouble cases' was assembled, from which,
in true lawyerly manner, the two researchers managed to extract what they saw
as some fundamental principles of Cheyenne law. This work is vividly and
confidently written, and it is easy to recognize the possibilities which
anthropologists immediately saw in it. Malinowski ( 1942) himself reviewed it
favourably just before he died.
969
SOCIAL LIFE
Looking back, it does seem that The Cheyenne Way is vulnerable to the
criticism that Llewellyn's lawyerly preoccupation with superior court litigation
obtruded too strongly into this attempt to understand conflict in another culture.
There must be some question as to just how much of Cheyenne culture survived
accommodation within the format of an American law school text. But that kind
of criticism cannot be made of the rich and wide-ranging studies which followed.
These made advances in at least four important directions: freedom from a rigid
adjudicatory model of decision-making; escape from a narrow view of conflict as
necessarily pathological and linked to rule breach; progress towards a more
sophisticated understanding of the relationship between rules and outcomes; and
rejection of an inflexible 'law-war' dichotomy.
In Ancient Law and in subsequent writings, Maine had treated third-party
decision-making as the basic means of resolving disputes across all known
societies. From the old patriarchs who stood at the head of the earliest social
groups to Victorian High Court judges, he saw the mode of resolution as one of
imposed decision; it was just that as different stages of civilization were
reached, different kinds of people made the decisions and new criteria
underpinned their judgments. Today, now that we recognize the possibility of
'order' without 'command', and are thus no longer constrained to invoke the
necessity of the king and the judge (although in the West still expecting to find
them somewhere in the picture), it becomes possible to characterize the range
of dispute-processing institutions in a far less restricted way.
n t e asts o t e et nograp tes o tspute w tc appeare m t e Os
and 1960s a number of tentative typologies of dispute institutions were put
forward (Gulliver 1963, Abel 1974, Koch 1974). These emphasized various
features, such as the presence or absence of third-party intervention, or the
form which such intervention might take in those cases where it was to be
found. A measure of agreement also began to emerge as regards the essential
range of variation which empirical studies disclosed. At the heart of these
variations appear to lie three basic forms which settlement-directed discourse
may take: the disputants may feel their way towards a settlement through
bilateral negotiation; they may try to resolve the matter with the help of a
neutral mediator; or they may submit the quarrel to an umpire for decision. My
discussion of these alternatives in the following paragraphs conforms closely to
Roberts (1979:69-71).
Bilateral negotiation represents the least complex form of settlement
process. Here the rival disputants approach each other without the intervention
of third parties and try to bring the dispute to an end through discussion. No
intermediaries or supporters are involved; the achievement of communication
and the subsequent process of settlement lie in the hands of the two parties
alone. A variation of this mode of settlement occurs when partisans align
themselves in support of one or other of the disputants; but while the 'strength'
of the respective sides may be altered by this procedure, the structural form of
the encounter remains unchanged.
970
In each of the remaining modes of settlement this bilateral element is

removed by the intervention of third parties in some intermediate position.
Where this role is mediatory, the third party helps the disputants towards their
own solution rather than imposing a solution upon them. The most limited
form of mediation arises where the third party acts as a 'go-between'. His role
is passive in the sense that while he operates as a bridge or a conduit between
the two disputants, he does no more than carry messages backwards and
forwards between them. Through this means of communication the disputants
themselves reach some kind of settlement. The go-between has not actively
contributed by tendering advice or urging particular avenues of conduct; but he
has enabled the disputants to communicate with each other. This form of
mediation may be contrasted with a more active one, in which the third party
takes a positive part in promoting a settlement. His intervention may take the
form of advice, suggested solutions, reasoned pleas, or even impassioned
cajoling, threats and bullying. Unlike the go-between, he actively pursues a
settlement, while remaining ostensibly neutral and without seeking to impose
an outcome.
Under the third mode of settlement the neutral party seeks to resolve the
dispute by making a decision, rather than by assisting the disputants towards
their own solution. Within this broad category we can distinguish two types of
umpire, whom I shall call the arbitrator and the adjudicator. The arbitrator
derives his authority to decide the dispute from the invitation of the disputants
t emse ves, w o ave vo untan y su mltte to IS ec1s10n. T e a JU tcator, y
contrast, derives his authority from some office in the community, and
intervenes to impose a decision by virtue of that office rather than by the
invitation of the disputants. In some respects this last distinction is of limited
importance, as both kinds of umpire possess the authority to resolve a dispute
in the face of competing claims by imposing a decision. Nevertheless, the
distinct sources from which this authority is derived may (as we shall see later
in this section) be of critical importance.
This typology underlines some of the important variables which affect these
different processes: the achievement of a solution by negotiated agreement or
imposed decision; the presence or absence of third-party involvement; the
nature of the intervener as either partisan or neutral; and the derivation of
authority in decision-making. It also points to what may be considered the
crucial feature of any dispute process, namely the location of the power to
decide the outcome.
The lawyer's 'folk' view of conflict as pathological, as arising out of rule
breach and requiring remedial intervention, is central to the analysis in The
Cheyenne Way. Underlying this view is a determination to keep the 'legal' and
the 'political' apart, to treat disputes associated with departures from
commonly accepted understandings as somehow different from those
associated with competitive processes in which there is a struggle for a scarce
resource which one may win and another lose without either departing from
971
SOCIAL LIFE
mutually accepted standards of conduct. Although this distinction is deeply

rooted in Anglo-American legal ideology, it does not provide a safe point of
departure for the study of dispute processes in other cultures. One of the
strengths of Turner's discussion, in his Schism and Continuity in an African
Society (1957), of the cyclical processes through which the headship of
N dembu villages devolves, lies in his demonstration of the way in which claims
about 'wrong doing' are closely interwoven with struggles for political
ascendancy and competition for resources.
Conventional accounts of judicial decision-making in Anglo-American
courts postulate a clear-cut relationship between rule and decision: the facts
adduced by the parties identify for the judge the relevant rule which is then
invoked as determining the outcome. This seemingly mechanical process is
little more than a caricature of judicial decision-making, even in the superior
courts; but while it may provide an ideal model of how things should be done in
a particular legal culture, it is not a model which provides a safe starting point
in understanding the dispute processes of other societies. The legal
ethnographies produced since the 19 50s reveal wide differences from one
society to another in the nature of the normative repertoire and in the manner
in which the repertoire is invoked in the context of dispute. Norms may be
vague and general, seldom explicitly invoked, but none the less implicitly
shaping the contours of claim and argument; or they may be clear-cut and
detailed, exposed to explicit discussion and scrutiny in the context of dispute,
an seen as etermmatiVe o t e outcome. r a ternatiVe y, w 1 e regar e as
important, norms may be treated as but one kind among many of the resources
that can be invoked.
Another durable strand in legal ideology is the time-honoured idea that
fighting and talking are opposed-in the sense that talking tends to be rule-
governed whereas fighting does not. Introducing a volume of papers on Law and
Watfore, Bohannan (1967) promotes this idea in his comment that there are
'basically two forms of conflict resolution: administered rules and fighting, Law
and War' (1967:xiii). A number of carefully observed accounts of intercommunal
fighting in New Guinea indicate that this formulation should be treated with
caution. They reveal that such fighting almost invariably takes on an
institutionalized form, in some cases constituting elaborate set-piece encounters
(e.g. Rappaport 1967). Elsewhere conflict is taken even further into the sphere of
ritual, as in the Eskimo nith-songs, where the participants 'fight' with words
(Weyer 1932). Here, fighting is talking, rather than being opposed to it.
One ground on which anthropological studies of dispute processes have
been fairly criticized is their tendency to present disputes in very much the way
that lawyers do, as typically involving the clash of two relatively evenly matched
individuals. It is argued that the implications of stratification and the presence
of control from the centre have frequently been ignored (Cain and Kulcsar
1981 ). This neglect can perhaps be partially explained as a consequence of the
kind of society which anthropologists have typically studied; but as soon as we
972
move away from small-scale, relatively egalitarian cultures, at least three broad
categories of dispute have to be distinguished:
1 Disputes between parties in relationships of relative equality.

2 Disputes which cross lines of stratification (e.g. confrontations between
lord and villein; between employer and employee).
3 Disputes which arise directly out of a ruler's efforts to govern and in which
the ruler himself or his agents will be directly involved.
Dispute processes within each category may be expected to take a different

shape; and variations in institutional structure may be observable, as also in the
criteria invoked by the disputants and by those attempting to achieve an
outcome.
AN ANTHROPOLOGY OF LAW AS THE STUDY OF

LAW IN THE COLONIAL CONTEXT
The anthropologies of law as the study of order and of dispute processes
tended to focus upon small, local communities, cut away from the larger
colonial context within which nearly all of them had become encapsulated by
the time the studies discussed above were carried out. This excision was
typically a conscious choice, taken because it was the uniqueness of the society
m questiOn t at was o centra mterest. But rom t e 7 s-an m a ew cases
from much earlier-this fiction tended to be dropped: the implications of
'contact' were foregrounded and the points of contact of small local
communities with the larger encapsulating colonial order became the explicit
focus of interest. Thus a historical dimension and an interest in change were
restored to legal anthropology. The important questions became: What was the
link between the governmental arrangements and normative understandings of
the pre-colonial world, and those prevailing in the same localities in the post-
colonial present? What was 'customary law', and what was its relationship to
the colonial project? What was the relationship between the colonial legal order
and life in the localities?
In jural terms what happened is largely uncontested. Across Africa, Asia and
the Pacific, overarching, territorially based legal orders were imposed, founded
on the metropolitan law of whichever happened to be the colonial power
concerned-Britain, France, Germany, Holland, or Portugal. Subject to that
dominant legal order, the pre-existing normative orders of local encapsulated
groups enjoyed qualified, parallel survival. At the same time, in many
territories, 'traditional authorities' became ever more caught up in the project
of colonial rule. In the case of Britain, this was first a matter of necessity-
there were simply not enough expatriates to go round; later, a virtue was made
out of necessity, as native intermediaries were employed under the policies of
'indirect rule'. The provenance of these authorities was rather varied: in some
973
SOCIAL LIFE
instances they had occupied apical positions in pre-colonial polities; in others

they were virtually the creations of the colonial power. This led, in formal
terms, to the development of a 'dual system' of government, of which the
'native' (later 'local' or 'customary') courts became part. The national law
provided the regime of norms in the superior courts and the magistrates'
courts; and 'native' or 'customary' law provided it in the local courts. In general
terms, these arrangements survived to the end of the colonial period; and in
some countries they still survive today.
The way in which this process has been regarded by scholars has undergone
considerable modification, in line with changing views concerning both the
nature of the pre-colonial world onto which the colonial legal order was
superimposed, and the character and effects of the colonial project itself.
In general, older accounts offer a picture of order and continuity, later ones
of abrupt transformation. The earlier view depicted an imposition of colonial
rule upon a stable egalitarian consensus. Life in most encapsulated
communities was said to have altered little: at first, because the colonial power
lacked the resources to bring about rapid, ameliorating change; later, because
the survival and continuity of 'traditional' life was deliberately fostered under
the policies of 'indirect rule'.
Later accounts (e.g. Chanock 1985, Ranger 1983, Snyder 1981, Woodman
1983) tend to contradict this picture rather sharply. They tell a story of
discontinuity and abrupt transition which left members of encapsulated
commumttes expose to t e ar ttrary ru e o neotra 1t10na aut orlttes, an
drawn to their disadvantage into new forms of economic relations. Colonial
local government is now presented as having had few links with the past:
authorities had to be 'found' and placed in charge of formerly acephalous
groups, or, at best, holders of existing offices were made to perform roles quite
different from their accustomed ones. The 'customary law' which was
recognized in colonial legislation, and developed and 'applied' in the newly
established 'native' courts, was a tendentious montage with only a superficial
connection with the past, supportive of the project of colonial rule, and
entrenching the position of elders over juniors, men over women. Some have
even called it an 'invented tradition' (Ranger 1983).
Overall, this revision is a valuable one, a necessary antidote to earlier
accounts which had postulated a deceptively harmonious and egalitarian pre-
colonial context, and which had overemphasized the extent to which long-
standing indigenous institutions had been there in the first place and then
survived. There is no doubt, either, of the coercive nature of 'indirect rule', or
about the disruption to the lives of colonized peoples resulting from their
association, often involuntary, with European economic operations.
Nevertheless, the new picture is arguably still an incomplete one, and care
must now be taken to avoid distortions of an opposite kind to those present in
the earlier accounts. There are real difficulties with seeing 'customary law'
solely in terms of domination. Similarly there are problems in regarding it as of
974
entirely recent manufacture. Lastly, there must be doubt as to how far 'colonial
customary law' was successfully transmitted into, and assimilated within, the
life-worlds of most colonized peoples.
First, while it was important, as a counter to the consensus implied in earlier
writings, to reveal the extent and nature of colonial domination, this is by no
means the whole story. Even if we freely concede the coercive nature of local
government in the colonial period, and the ideological quality of what passed
for 'customary law', an exclusively one-way, top-down view of the colonial
encounter would be misleading. There is no need to repeat here the now well-
articulated and generally accepted concerns about placing too literal a reliance
upon a conception of 'sovereign' power (Foucault 1984 ). 'Power' resides at
different levels, takes on diverse forms, and runs in all directions (Giddens
1985). Thus while 'customary law', in the sense of the repertoire of rules
applied in the colonial courts, did provide an instrument of rule, it also offered
avenues of escape and resistance for the ruled. Similarly, 'customary law' in the
different sense of the meanings and commitments which furnished the life-
worlds of indigenous peoples, while subject to covert penetration and eo-option
(de Sousa Santos 1980), also provided the means to achieve qualified autonomy.
The insistence of scholars like Chanock, Snyder and Ranger that 'customary
law' is of recent manufacture, a creature of the colonial period rather than the
pre-colonial past, is helpful in a number of ways. It is essential to recognize that
the relationship between contemporary and past forms is, at the very least,
pro emattc. Moreover t e associatiOn etween custom' an a suppose y
egalitarian context must be questioned. Further, the specific idea of 'invention'
restores and gives prominence to a conception of agency, the essential notion
that custom is linked to the affairs of living men and women-that it is both at
the root of action and the product of it. But there are difficulties in pressing
this view of customary law too far. First, it risks conflating two separate, if
interlinked spheres: the 'customary law' of the colonial and post-colonial
courts, and that which furnishes the everyday life-world of local people.
Second, the connotation of novelty, of a clean break, which 'invention' carries,
draws attention away from crucial aspects of what was happening. The very
strength of customary law, the source of its supposedly coercive power, lay in
the links it could claim with a past, established and approved state of affairs.
Foreign novelties do not lay claim through existing commitments; yet that-if
anything-is what custom does. Thus we should be looking not for novelty but
for the exploitation of an existing repertoire, or the artificial sustaining of
ancient forms, with detrimental, constraining effects upon the ruled.
The idea of an 'invented' tradition seems also to imply an impoverished and
grossly simplistic understanding of the operation of ideology. It calls up a
vision of the manufacture, transmission and assimilation, intact, of some new
world view, and the corresponding destruction of the pre-existing cognitive and
normative foundations of the life-world. Much more persuasive is an account
of ideology as working with what is already to hand, covertly upon and within
975
SOCIAL LIFE
an existing life-world, transforming without eradicating. Such an account

seems to me to be essentially that suggested by Althusser ( 1977), and in the
specific context of customary law, by de Sousa Santos (1980). But even here we
must not neglect the very significant extent to which some cultures are resistant
to transformation through eo-option. Empirical observations reveal wide and
interesting variations in the response of encapsulated groups to the experience
of colonial rule.
The problem of transmission and assimilation raises a final concern over
recent revisions. How far, in fact, were colonial subjects affected in their
everyday lives by colonial 'customary law'? It would be foolish to underestimate
the consequences of economic changes during and following the colonial
period, and the operation of colonial customary law was undoubtedly in some
respects supportive of those changes. But we should nevertheless keep an open
mind as to the extent to which the worldviews of those in power came to the
attention of the ruled. Even where they did, there is a question over how far
they ever came to be shared. As Weber indicated in Economy and Society, the
importance of ideology may lie more in supporting the agents of those in power
than in engendering commitments among the ruled. Recent writings, including
those with a direct focus upon the colonial period, offer confirmation of this
view (see Hobsbawm and Ranger 1983).
PLURALISM
The shift to a focus on the operation of law in the colonial context brought with
it a number of important gains. First, it reinforced the recognition that law had
a political dimension, in the sense of its implication in processes of domination.
Thus, 'power' remained in the centre of the picture, to which it had already
been drawn in the more sophisticated discussions of dispute processes (Starr
and Collier 1989:1-25). As a result, attention inevitably moved to the role
which 'the rule of law' and the process of adjudication might play in the
legitimation of particular forms of government. Accordingly, the ideological
aspects of law achieved a new prominence, as did the nature of legal ritual.
Secondly, the focus upon the operation of law in the colonial context forced
scholars to give much more careful thought to the nature of indigenous
governmental arrangements and normative understandings in the pre-colonial
world, and to the transformation which these subsequently underwent (for an
important example, see von Benda-Beckman 1979). In this respect, the process
of incorporation of the 'traditional authorities' into regimes of colonial and
post-colonial government was of central interest (Mann and Roberts 1991).
All of this posed some troubling questions, which ultimately resolved
themselves into a single problem: with the imposition of a national, formally
dominant legal order upon the diversity of pre-colonial indigenous
communities, how can we best conceptualize the relationship between what was
976
going on at the centre, in 'the secretariat'; and what was happening on the
periphery, in the localities? One way of looking at this is in terms of what
Kidder (1979) has vividly called 'the static hypodermic model'. This involved a
vertical, top-down, command view of the operation of law in the colonial
world. Rules enacted by government at the centre were transmitted to the
localities, where they produced direct, matching changes in behaviour,
resulting in 'development' or progress towards modernity. 'Law', made at the
centre, superseded existing 'customary' regimes.
Pospisil, in his early work, Kapauku Papuans and their Law (1958), rejected
this extreme positivism. 'Law' should be located at different points in the social
world, wherever 'authorities' could be found imposing normatively based
decisions. Accordingly, whether you looked at the developed West, or at the
territories then undergoing colonial encapsulation, 'law' should be seen as
residing at a number of hierarchically ranged, more or less discrete, 'legal
levels'. In so far as these levels were connected, the linkage was still seen to be
vertical, with change being transmitted down from the top. Most important,
perhaps, was Pospisil's rejection of an exclusive focus upon state law, allowing
as much attention to be given to other normative fields. Why, he asked, should
national law be privileged: should we not treat as 'law' the normative
understandings prevailing within local groups at any level? For Pospisil, norms
operative at the village level were just as much 'law' as those enacted at the
centre.
more ex1 e approac was propose y Moore m er semma essay, Law
and change: the semi-autonomous social field as an appropriate area of study'
(1973). Here, Moore substitutes the concept of 'social field' for that of 'legal
level'. Normative orders, including that presented by the national legal system,
are best seen as partially discrete, but nevertheless overlapping and
interpenetrating social fields, within which meaning is communicated on a two-
way, interactive basis. The social field is identified in terms of its 'semi-
autonomy', by 'the fact that it can generate rules and customs and symbols
internally, but ... is also vulnerable to rules and decisions and other forces
emanating from the larger world by which it is surrounded' (Moore 1978:55).
Moore was not talking exclusively about 'law', but rather about 'normative
fields' in general; nevertheless her approach proved immediately congenial to
legal anthropologists. She depicted change as a fluid, interactive process, full of
imponderables and unintended consequences.
Pospisil's insistence that in examining the 'legal' we should not focus on the
level of national law alone, and Moore's lead in turning attention to the
relationship between coexisting normative fields, together constituted the
principal agenda and approach for the anthropology of law during the latter part
of the 1970s and the 1980s. Under the label of'legal pluralism' the anthropology
oflaw virtually became the study of how several normative regimes may coexist in
the same social field. Legal anthropologists formed themselves into a professional
association under the grandiose title of the 'Commission on Folk Law and Legal
977
SOCIAL LIFE
Pluralism'; the journal African Law Studies re-emerged as the Journal of Legal
Pluralism and Unofficial Law, a conference was held at Bellagio in 1981 to
inaugurate this movement, and a large literature emerged which sought to re-
present the anthropology of law as legal pluralism and to delimit this new field
(see specially, Griffiths 1986, Merry 1988, Allott and Woodman 1985). Following
Moore's lead, societies of the West became as much a focus of attention as did
those of the post-colonial world.
In retrospect, this new anthropology of law brought important insights.
First, the move away from 'legal centralism' (Griffiths 1986), from according
privileged attention to national law, and from treating it as unproblematically
determinative of social forms, represented something of a release for legal
scholars. Equally significant was the way in which the relationship between
adjacent, semi-autonomous fields came to be perceived-as fluid, interactive
and imponderable. The very focus of lawyers' attention on a wider slice of the
social world, which legal pluralism implied, was in itself welcome; as was a new
openness to social and anthropological theory.
But there are also costs entailed in 'melting it all together as "law"' (Moore
1978:81 ). As Merry notes (1988:878), to extend the term law to forms of
ordering that are not state law may lead to a loss of analytic rigour. Depending
upon the focus of analysis, while 'recognizing the existence of and common
character of binding rules at all levels, it may be of importance to distinguish
the sources of the rules and the sources of effective inducement and coercion'
Moore 7 : ). T e tstmctiVe c aracter o state aw m t e West enves rom
its implication in the growth of a particular form of government; this
provenance accounts for crucial differences between it and other normative
orders. Correspondingly, in labelling other normative orders as 'law', it is
important to avoid the trap of investing them with the attributes of state law.
This seems to be exactly the trap into which Pospisil himself had fallen.
Insisting that 'Kapauku law' takes the form of norms derived from legal
decisions, which have to enjoy the attributes of 'authority' and 'intention of
universal application' in order to have a legal quality, he imputes an
adjudicative, command character to Kapauku processes which seriously
distorts their nature. While most advocates of the approach of legal pluralism
are entirely conscious of the hazards of distorting non-state processes through
investing them with a framework derived from Western law, the designation of
the approach as one of legal pluralism should perhaps sound a warning note.
It is significant that the field of the anthropology of law has become almost
exclusively occupied by lawyers rather than anthropologists. Until the 1960s it
was occupied almost entirely by anthropologists, with lawyers showing
relatively little interest; but since then the position has been entirely reversed.
And because it has been colonized by lawyers, it has inevitably been treated as
an area of 'legal' scholarship. Overall, it is difficult to avoid the impression that
the invocation of 'legal pluralism' has more to do with the entrenchment of an
academic discipline than with the struggle to understand the social world.
978
CONCLUSION
The potential gains in understanding to be achieved with the rejection of 'legal

centralism' were made clear in the writings of Malinowski more than sixty
years ago; it is good that these should now be recognized by lawyers working
within a pluralist paradigm. These gains can be consolidated by giving equal
attention to different normative orders, and by sensitivity to the ways in which
these orders intersect and are interwoven.
Inevitably this extension of the legal gaze brings with it renewed, agonized
attempts to delineate the 'legal'. Teubner's proposal that this boundary should
be marked by the use of 'the binary code of legal communication' ( 1992:1451) is
the latest in a heroic line; but it is hard to forecast a wider consensus for this
effort than for any of its predecessors. More worrying is the fact that
sociological understanding is immediately imperilled once we impose an
imprint of 'law' across plural normative fields. Merry (1988:878) senses this in
reflecting upon the boundary problems and renewed struggles for definition
which a 'legal' pluralism involves:
Why is it so difficult to find a word for nonstate law? It is clearly difficult to define
and circumscribe these forms of ordering. Where do we stop speaking of law and
find ourselves simply describing social life? Is it useful to call these forms of
ordering law? In writing about legal pluralism, I find that once legal centralism has
. . .
confounds the analysis. The literature in this field has not yet clearly demarcated a
boundary between normative orders that can and cannot be called law.
These difficulties are self-imposed, inherent in the project of legal pluralism. In

the context of such a project they are inevitable, given the extent to which our
ideas about law are bound up with Judaeo-Christian beliefs on the one hand,
and the development of secular government in Europe on the other. The
specific, situated character of those roots should be enough to warn us against
the enlargement of the realm of law which legal pluralism demands. Looking
forward, the anthropology of law should be content to resolve itself into the
respective anthropologies of norms and of government, attentive to the distinct
but nevertheless related problems of 'order' and 'domination'. Within the
ambit of these general enquiries, 'law' is best viewed as an interesting folk
category, encountered under specific and limited conditions.
REFERENCES
Abel, R.L. (1974) 'A comparative theory of dispute institutions in society', Law and
Society Review 8(2):218-347.
Allott, A.N. and Woodman, G. (eds) (1985) People's Law and State Law, Dordrecht:
Foris.
979
SOCIAL LIFE
Althusser, L. (1977) 'Ideology and the ideological state apparatus', in Lenin and
Philosophy and Other Essays, London.
Bailey, F. G. (1960) Tribe, Caste and Nation, Manchester: Manchester University Press.
von Benda-Beckmann, F. (1979) Property in Social Continuity: Continuity and Change in
the Maintenance of Property Relationships in Minangkabau, West Sumatra. The
Hague: Nijhoff.
Bloch, M. (1971) 'Decision-making in councils among the Merina of Madagascar', in
A.Richards and A.Kuper ( eds) Councils in Action, Cambridge: Cambridge
University Press.
Bohannan, P. (1957) Justice and Judgment among the Tiv, London: Oxford University
Press for the International African Institute.
--(1967) 'Introduction', in P.Bohannan (ed.) Law and Warfore, New York: Natural
History Press.
Cain, M. and Kulcsar, K. (1981) 'Thinking disputes: an essay on the origins of the
dispute industry', Law and Society Review 16:375-402.
Chanock, M. (1985) Law, Custom and Social Order, Cambridge: Cambridge University
Press.
Comaroff, J.L. and Roberts, S.A. (1981) Rules and Processes: the Cultural Logic of
Dispute in an African Context, Chicago: University of Chicago Press.
Cotterrell, R.M.B. ( 1983) 'The sociological concept oflaw', Journal of Law and Society
10:241-55.
de Sousa Santos, B. (1980) 'Law and community: the changing nature of state power in
late capitalism', International Journal of the Sociology of Law 8:379-97.
Durkheim, E. (1893) De la Division du travail social, Paris: Alcan.
es ormes e ementatres e a vte re tgteuse, ans: can.
Fallers, L. (1969) Law Without Precedent, Chicago: University of Chicago Press.
Fitzpatrick, P. (1992) The Mythology of Law, London: Routledge.
Foucault, M. (1984) The History of Sexuality, vol. 1, Harmondsworth: Penguin.
Giddens, A. (1985) The Nation-State and Violence, Cambridge: Polity Press.
--(1986) The Constitution of Society, Cambridge: Polity Press.
Gluckman, M. (1955) The Judicial Process Among the Barotse of Northern Rhodesia,
Manchester: Manchester University Press.
Griffiths, J. (1986) 'What is legal pluralism?', Journal of Legal Pluralism 24:1.
Gulliver, P.H. (1963) Social Control in an African Society, London: Routledge & Kegan
Paul.
--(1971) Neighbours and Networks, Berkeley: University of California Press.
Hobsbawm, E.J. and Ranger, T.O. (eds) (1983) The Invention of Tradition, Cambridge:
Hoebel, E.A. (1954) The Law ofPrimitive Man, Cambridge, Mass.: Harvard University
Press.
Kidder, R.L. (1979) 'Toward an integrated theory of imposed law', in S.B.Burman and
B.Harrell-Bond (eds) The Imposition of Law, New York: Academic Press.
Koch, K.F. (1974) War and Peace in Jalemo, Cambridge, Mass.: Harvard University
Press.
Llewellyn, K.N. and Hoebel, E.A. (1941) The Cheyenne Way, Norman: University of
Oklahoma Press.
Maine, H.S. (1861) Ancient Law, London: John Murray.
--(1883) Dissertations on Early Law and Custom, London: John Murray.
980
Malinowski, B. (1926) Crime and Custom in Savage Society, London: Kegan Paul,
Trench & Trubner.
--(1934) 'Introduction', in H.I.Hogbin, Law and Order in Polynesia, New York:
Harcourt Brace.
--(1942) 'A new instrument for the study of law-especially primitive', Yale Law
Journal5l:l237-54.
Mann, K. and Roberts, R. (eds) (1991) Law in Colonial Africa, Portsmouth, NH:
Heinemann.
Mann, M. (1986) The Sources of Social Power, vol. 1, Cambridge: Cambridge
University Press.
Merry, S.E. (1988) 'Legal pluralism', Law and Society Review 22:869-96.
Moore, S.F. (1973) 'Law and change: the semi-autonomous social field as an
appropriate area of study', Law and Society Review 7:719-46.
--(1978) Law as Process, London: Routledge & Kegan Paul.
Nader, L. (1969) Law in Culture and Society, Chicago: Aldine.
Newman, K. (1983) Law and Economic Organisation, Cambridge: Cambridge
University Press.
Pospisil, L. (1958) Kapauku Papuans and their Law, New Haven: Yale University
Publications in Anthropology no. 54.
Radcliffe-Brown, A.R. (1922) The Andaman Islanders, Cambridge: Cambridge
University Press.
--(1933) 'Primitive law', in Encyclopedia of the Social Sciences 9:202-6. New York.
--(1940) 'On social structure' ,Journal of the Royal Anthropological Institute 70: 1-12.
--(1952) Structure and Function in Primitive Society, London: Cohen & West.
anger, . . e mventwn o tra ItiOn m co oma nca , m . o s awm an
T.O.Ranger (eds) The Invention of Tradition, Cambridge: Cambridge University
Press.
Rappaport, R.A. (1967) Pigs for the Ancestors, New Haven: Yale University Press.
Redfield, R. (1967) 'Primitive Law', in P.Bohannan (ed.) Law and Warfore, New York:
Natural History Press.
Schapera, I. (1938) A Handbook of Tswana Law and Custom, London: Oxford
University Press.
Snyder, F. G. (1981) 'Colonialism and legal form', Journal of Legal Pluralism 19: 49-90.
Starr, J. and Collier, J.F. (eds) (1989) History and Power in the Study of Law, Ithaca, NY:
Cornell University Press.
Teubner, G. (1992) 'The two faces oflegal pluralism', Cardozo Law Review 13: 1443-
62.
Turner, V. (1957) Schism and Continuity in an African Society, Manchester: Manchester
University Press.
Weber, M. (1978 [1917]) Economy and Society, trans. and ed. G.Roth and C.Wittich,
Weyer, E.M. (1932) The Eskimos, New Haven: Yale University Press.
Woodman, G.R. (1983) 'How state courts create customary law in Ghana and Nigeria',
in H.W.Finkler (compiler) Papers of the Symposia on Folk Law and Legal Pluralism,
XIth International Congress of Anthropological and Ethnological Sciences, Vancouver,
Canada, 19-23 August 1983.
981
SOCIAL LIFE
FURTHER READING
Abel, R.L. (1974) 'A comparative theory of dispute institutions in society', Law and
Society Review 8(2):218-347.
Bohannan, P. (1957) Justice and Judgment among the Tiv, London: Oxford University
Press for the International African Institute.
Cain, M. and Kulcsar, K. (1981) 'Thinking disputes: an essay on the origins of the
dispute industry', Law and Society Review 16:375-402.
Fallers, L. (1969) Law without Precedent, Chicago: University of Chicago Press.
Gluckman, M. (1955) The Judicial Process among the Barotse of Northern Rhodesia,
Manchester: Manchester University Press.
Griffiths, J. (1986) 'What is legal pluralism?', Journal of Legal Pluralism 24:1.
Gulliver, P.H. (1963) Social Control in an African Society, London: Routledge & Kegan
Paul.
Hamnett, I. (ed.) (1977) Social Anthropology and Law, New York: Academic Press.
Hoebel, E.A. (1954) The Law ofPrimitive Man, Cambridge, Mass.: Harvard University
Press.
Koch, K.F. (1974) War and Peace in Jalemo, Cambridge, Mass.: Harvard University
Press.
Llewellyn, K.N. and Hoebel, E.A. (1941) The Cheyenne Way, Norman: University of
Oklahoma Press.
Malinowski, B. (1926) Crime and Custom in Savage Society, London: Kegan Paul,
Trench & Trubner.
Merry, S.E. (1988) 'Legal pluralism', Law and Society Review 22:869-96.
Nader, L. (1969) Law in Culture and Society, Chicago: Aldine.

Newman, K. (1983) Law and Economic Organisation, Cambridge: Cambridge
University Press.
Pospisil, L. (1958) Kapauku Papuans and their Law, New Haven: Yale University
Publications in Anthropology no. 54.
Turner, V. (1957) Schism and Continuity in an African Society, Manchester: Manchester
University Press.
982
36
COLLECTIVE VIOLENCE AND

COMMON SECURITY
Robert A.Rubinstein
Anthropologists have taken a long and varied interest in studying armed

conflict and aggression. Acts of collective violence, however, have relatively
rarely been a principal topic of anthropological concern. Instead, descriptions
of collective violence have been embedded in ethnographies or in theoretical
discussions focused primarily on such other topics as social organization, legal
systems, or political evolution (see the previous two articles in this volume).
Moreover, t IS attentiOn as most o ten een pat wtt m t e context o stu tes
of small, well-bounded 'preliterate' or 'primitive' societies-the traditional
subjects of anthropological study.
Although collective violence is usually considered only incidentally, and in
the context of small-scale societies, it has occasionally formed the main subject
of anthropological study. This shift of focus has often occurred at times of
major international crisis-like the Second World War, the Vietnam War, or the
threat of nuclear war-when anthropologists have sought to bring their
knowledge to bear on contemporary circumstances (see, for example,
Malinowski 1941, Swanton 1943, Fried et al. 1968, Worsley and Hadjor 1987).
As a result of the direct and indirect anthropological study of collective
violence, many approaches to the topic have been elaborated and the relevant
literature is large. It would be impossible to review it all. In this article,
therefore, I have selected for review those anthropological materials relevant to
the understanding of collective violence and security in the modern world. I
focus primarily on how anthropological data and theory can contribute to
contemporary discussions of collective violence and security, as these are
carried out by the various professional communities--of diplomats, analysts
and politicians-which are charged with deciding related policy issues. I argue
that during the last forty years these communities have been dominated by
methods and topics of analysis which produce too narrow an understanding of
the social and cultural phenomena involved.
983
SOCIAL LIFE
The article is divided into four general sections. First I present a brief
overview of some of the concerns that anthropologists have traditionally
brought to the study of violence and security. This first section is highly
schematic, intended simply to indicate the range of approaches that
anthropologists have taken, and to direct the interested reader to the relevant
literature. The second section describes the tenor of contemporary discussions
of collective violence and security. The assumptions underlying the dominant
forms of analysis are presented and some examples are given to illustrate the
results of applying these assumptions. The third section focuses on how the
introduction of anthropological materials forces us to enlarge our
understanding of two key concepts: 'power' and 'collective violence'. The
fourth discusses how anthropology can directly contribute to avoiding,
managing, and resolving collective violence by attending to cultural aspects of
negotiations.
SOME THEMES IN THE ANTHROPOLOGICAL

STUDY OF VIOLENCE
Several broad themes characterize the anthropological literature on violence
and aggression. Perhaps the most frequent is the discussion of the biological
basis of human aggression. At various times during this century theorists have
asserted that individual and collective violence is a necessary result of the
circumstances o uman 10 ogica evo utwn. Ear y c aims to t IS e ect were
grounded in a relatively crude biological determinism, while later claims have
been based in more sophisticated elaborations of biological theory.
For instance, basing his work on ethology and a general understanding of
the evolution of aggressive behaviour in non-human species, Lorenz (1963, also
Ardrey 1966, Morris 1967) argued that humans have a heritage of intraspecific
aggression. This heritage, he further argued, is especially troublesome because
it is linked to the rapid development of weapons and yet is unconditioned by
biological mechanisms of restraint, as are aggressive drives in other species.
Thus he writes (1963:42):
It is more than probable that the destructive intensity of the aggression drive, still a
hereditary evil of mankind, is the consequence of a process of intraspecific selection
which worked on our forefathers for roughly forty thousand years, that is,
throughout the Early Stone Age. When man reached the stage of having weapons,
clothing, and social organization, and so overcoming the dangers of starvation,
freezing and being eaten by wild animals, and these dangers ceased to be the
essential factors influencing selection, an evil intraspecific selection must have set
in. The factor influencing selection was now the wars waged between hostile
neighbouring tribes.
This view of the biological basis of human aggression has been widely
criticized as based on faulty inference, and especially on inappropriate and
984
COLLECTIVE VIOLENCE AND COMMON SECURITY
oversimplified analogies between the behaviour of modern humans and that of

non-human animals (see Dunbar in this volume, Article 27). More recently, the
growth of sociobiology has raised anew the issue of the biological basis of
human aggression (Wilson 1975, 1978). Chagnon (1988:985), for instance,
recently reported that among the Yanomamo Indians of Brazil, 'men who have
killed have more wives and offspring than men who have not killed', and he
went on to argue, on these grounds, that reproductive variables were critical to
understanding tribal violence.
Such sociobiological attempts to account for human aggression in genetic
and evolutionary terms have, however, been as vulnerable as their ethological
precursors to the charge of depending upon faulty inference. Moreover they
fail to specify a mechanism of action (Sahlins 1976), and are ultimately
unhelpful in accounting for specific forms of violence.
The general consensus currently is that attempts to understand violent
human action in terms of some hereditary load are misconceived, especially
when an innate tendency towards aggression is invoked in order to attribute
collective violent action to an essential human nature (Hinde 1988). Indeed, as
Koch has observed:
It really does not matter whether or not one assumes an innate drive toward
aggression. History and comparative anthropology show that people fight not
because they need to satisfy some instinct, but because their interests clash with
culturally defined.
(Koch 1974:52-5)
A second approach widely adopted in the anthropological analysis of human

collective violence views it in ecological terms, as serving to preserve a viable
relationship between a population and the environmental resources available to
it (Vayda 1968, 1974, Tefft 1974). For example, in a study of the Maring of
Highland New Guinea, Rappaport (1967) attempts to show that population
pressure leads to conflicts whose effect is to redistribute human population
over available land.
Although such ecological accounts have been proposed independently of the
biologically deterministic views of human aggression reviewed above, the two
approaches are often contrasted. This contrast is particularly evident in
anthropological discussions of warfare in Amazonia. Anthropologists dispute
among themselves whether Amazonian warfare is best accounted for by
reference to protein scarcity, reproductive fitness, or something else altogether
(see Gross 197 5). Those who take an ecological view argue that game animals
are relatively scarce in the area and that protein shortage is therefore a limiting
aspect of the environment (e.g. Ross 1980:38-39, Ross and Ross 1980). Their
opponents, however, argue that rather than being due to limitations of the
area's carrying capacity, Amazonian warfare results either from pressures of the
985
SOCIAL LIFE
socio-political environment (Chagnon 1967), or from the reproductive benefit

it confers (Chagnon 1988; see Chagnon and Hames 1979). Although attempts
have been made to reconcile these positions (Ferguson n.d.), warfare in
Amazonia remains to be adequately accounted for.
Anthropological concern with collective violence, and especially with
understanding and defining war, has in part been motivated by the objective of
interpreting the growth and evolution of human societies. Many analysts view
warfare as having been particularly important for the growth of states as
centralized political systems (Sahlins 1968, Cohen 1983, 1986, Ember and
Ember 1988, Vincent 1990:90-2). Political evolution is treated in detail by
Ear le, in Article 34 of this volume. Here I merely note that whatever positive
role it may have played in the development of 'early' states, collective violence
no longer supports the stability of states in the contemporary world (Foster and
Rubinstein 1986, Beeman 1989).
More generally, as anthropologists have gained more experience in the
analysis of collective violence, it has become clear that human violence and
aggression cannot adequately be accounted for in terms of relatively simple
models, and that it is essential to appreciate their complex and multi-causal
nature (Foster and Rubinstein 1986, Rubinstein and Foster 1988). In this light,
Vayda (n.d.) has reconsidered his earlier attempts at an ecological explanation
of primitive war, and finds them inadequate. He argues that his previous
accounts placed too much of an explanatory burden on the annexation of
terntory. By contrast, e now cone u es t at we nee to e more aware o t e
context-relatedness of human behavior and of how answers to "why-questions"
differ depending upon differences in assumptions'.
Finally, the anthropological literature reflects a concern with the definition
of war. Anthropologists have sought to distinguish warfare as a unique form of
collective violence. Yet because collective violence is a complex social
phenomenon, a single definition of warfare necessarily proves inadequate. As
Koch (1974:52-3) put it, 'linguistic distinctions between raids, feuds, and war
tend to obscure rather than elucidate the problem of explaining why people
resort to violent methods of confrontation in pursuit of their interests'.
It is principally because collective violence is resorted to as a means by which
groups pursue their own perceived interests (Wright 1968, Herran 1988, Foster
1989), which are patterned by socially and culturally based symbolic forms,
that attempts at all-encompassing definitions and explanations of war and
collective violence necessarily fail. In place of global explanations of war, a more
particularistic approach, one that adequately deals with the multiple levels-
from small scale to large scale-on which collective violence occurs, yields
greater understanding. In addition to the economic, ecological, and material
concerns that are routinely included in attempts to define and understand
collective violence, it is also necessary to consider culturally specific symbolic
and organizational systems. Anthropological enquiry is particularly well suited
to the identification of such systems (Foster and Rubinstein 1986, Rubinstein
986
and Foster 1988, Turner and Pitt 1989), and I focus on these in the remainder
of this article.
ETHNOCENTRISM, VIOLENCE AND SECURITY IN

INTERNATIONAL AFFAIRS
Recent critiques of anthropological writing have argued that despite the claim
of the discipline to present a view of other societies and cultures 'from the
inside', the images of the 'other' which it presents are distorted by their passage
through the warped lenses of Western, logocentric discourse. It is therefore
somewhat out of fashion to offer anthropological descriptions of other societies
and cultures as possible correctives to ethnocentrism. But while no description
of another culture can be perfect, some conceptions of culture, and the
descriptions that derive from them, are for some purposes more adequate than
others (Rubinstein 1992). This is especially the case when we come to consider
approaches to violence and collective security.
Despite the lively intellectual debate surrounding this topic, discussion has
been dominated during the past forty years by a single and widely held
approach, often called 'political realism'. The vigour of the debate tends to
obscure the fact that this approach continues to furnish the context within
which issues of collective violence and security are presented and evaluated.
Furthermore, it forms the symbolic matrix that shapes discourse about these
Issues m contemporary circumstances o n 7, Bras set ). Po Itica
realism places a premium on the production of information that is
characterized as 'objective', 'rational' (in a logical sense), amenable to formal
modelling, and derived from 'correct scientific methods' (Beeman 1986,
Rubinstein and Foster 1988:3-7). In an important sense, the role accorded to
such information in the analysis of social and cultural life derives from, and
perpetuates, a pervasive ethnocentrism.
A few examples can illustrate how an ethnocentric hegemony is reinforced in
discussions of violence and collective security. Although 'culture' has become a
category of some concern in diplomacy, attempts to understand its role in
negotiations tend to rely on caricatures of national negotiating and decision-
making styles. These attempts seek to specify how the national culture affects
negotiations in order that diplomats may be advised about what to expect in
their dealings with representatives of different countries. In contrast to
anthropological descriptions of the dynamic and symbolic nature of social and
cultural life, the resulting accounts treat culture as homogeneous and stable.
They discover 'cultural patterns' by collecting the impressions of diplomatic
and military personnel of 'what it was like to deal with them', or by gathering
impressions from the personnel of a third country. Because of the elements of
self-presentation or national self-interest that generally permeate these
accounts, the descriptions they provide are highly unreliable (see, for example,
Fahmy 1983). Nevertheless, they form the basis of the caricatures of culture
987
SOCIAL LIFE
that inform most discussions of world affairs. Thus, for example, interviews
with Polish personnel have been used to reveal the cultural basis of Soviet
negotiation strategies (Checinski 1981 ), and Middle Eastern negotiation styles
are portrayed as deriving from the haggling behaviour sometimes observed in
bazaars (Binnendijk 1987).
In discussions of collective violence and security in the Third World, the
local-level concerns that motivate less powerful nations and local groups tend to
fall from view. Instead, a privileged position is accorded to the interests and
interpretations of the superpowers, and diplomatic and military initiatives are
treated from the perspective of ideological, political, and economic superpower
contests. A recent study of constraints on United States policy in relation to
Third World conflicts (Hosmer 1985; see also Record 1985) reflects this
excessively narrow-minded view of global affairs. This report considers United
States involvement in the Third World almost entirely from the perspective of
military concerns. It treats that involvement primarily in relation to the Soviet
Union, for the most part ignoring the specific interests and concerns of Third
World countries and groups. This preoccupation with East-West relations, to
the exclusion of numerous regional concerns around the world, is revealed in
the fact that in his study, Hosmer makes explicit reference to the Soviet Union
on no fewer than 90 of the monograph's 130 pages, and on those pages where
he does not do so, it is only because he dwells instead on Chinese communist
interests or actions.
T IS m o et nocentnsm contmues to o sway esptte t e recent
superpower detente, and the collapse of the Soviet Union. An otherwise
instructive, recent three-volume analysis of the Lessons of Modern War
examines wars fought mainly in and by Third World actors principally from
the perspective of relations with great powers outside the Third World, and
by emphasizing military technology. Local-level political, social and cultural
factors are neglected, being considered 'only to the extent necessary to
understand military events' (Cordesman and Wagner 1990:xv). This neglect
is remarkable because, especially during the last decade, many
anthropological studies have appeared which show how analyses that ignore
cultural and symbolic factors are bound to fail (Foster and Rubinstein 1986,
Worsley and Hadjor 1987, Rubinstein and Foster 1988, Turner and Pitt
1989). To the extent that attention has been paid to the human arrangements
underlying the formulation and implementation of policy, it has largely been
by resort to formal, econometric or game-theoretic models of behaviour,
decision-making, and negotiation (Brams 1985, Ball and Richelson 1986).
And for the most part, the socio-cultural processes which qualify the
application of these models have not been considered (Rubinstein 1988a:23-
31 ). Worsley's ( 1982, 1986, 1987) discussions of the Third World, and of the
consequences of excluding cultural considerations from analysis, provide a
more general perspective on this issue.
988
POWER AND COLLECTIVE VIOLENCE IN

ANTHROPOLOGICAL PERSPECTIVE
The perspective of political realism, embodied both in the aforementioned

examples and in most contemporary discussions of violence and security,
depends, as anthropologists especially have pointed out, on a number of
unwarranted assumptions (see, for example, Foster and Rubinstein 1986,
Rubinstein and Foster 1988, Kim 1983, Beeman 1986, 1989, Myrdall969). Here
I wish to draw attention to three assumptions, in particular, that anthropological
work has seriously questioned: that actions in relation to collective violence and
security are based on objective social scientific knowledge; that they are rational
to the extent that they conform to formal models of econometric analysis or game
theory; and that the proper unit of concern for understanding such action in the
contemporary world is the nation state.
As regards the first point, it is rarely acknowledged by advocates of the
realist approach that it is the approach itself that determines what counts as
analytically relevant information. Thus the local-level meanings and symbolic
significance entailed in conflict situations are dismissed out of hand (Kim
1983:9). Yet facts, of course, are never 'just facts'. They depend upon value
judgements that can be consciously presented and explored or, for whatever
reasons, hidden. As Myrdal (1969:51-2) has observed:
1ases m sona science cannot e erase s1mp y y eepmg to t e acts an re mmg

the methods of dealing with statistical data. Indeed data and the handling of data are
often more susceptible to tendencies toward bias than 'pure thought' .... Biases are
thus not confined to the practical and political conclusions drawn from research.
They are more deeply seated than that. They are the unfortunate results of
concealed valuations that insinuate themselves into research at all stages, from its
planning to its final presentation. As a result of their concealment, they are not
properly sorted out and can thus be kept undefined and vague.
When it comes to rationality, the realist approach assumes that both decision-
making and action are mechanical processes: once a group has the 'objective
facts' at its disposal, it (through its leaders) will act rationally, according to the
predictions of formal models. For a typical example, we could cite the philosophy
and methods used by the RAND Corporation Strategy Assessment Center. The
work of the Center is based on automated war games in which rule-guided
decision models for managing behaviour and for co-ordinating responses are
substituted for human decision-makers. RAND representatives argue that
the power of the approach is due in large part to its emphasis on realism (relative to
more standard approaches) and to the use of artificial intelligence and force
modelling techniques that make behaviour rules and other key variables transparent
and interactively variable.
(Davis and Winnefeld 1983:vii)
989
SOCIAL LIFE
The rationality implied here is of a purely 'technical' kind, which excludes any
consideration of substantive cultural and social influences (Simon 1983). It
might perhaps be more appropriately described as logical rather than rational.
A corollary of the realist approach is that decisions and actions are attributed
to corporate groups, and especially in the last twenty-five years or so these have
generally been taken to be nation states. As a result, local or indigenous views of
intergroup relations are simply disregarded (Rubinstein and Foster 1988).
Anthropological analyses, to the contrary, are most often concerned with
interactions below the level of the nation state. In one such analysis, Beeman
(1986, 1989) demonstrates how United States foreign policy decisions
regarding the Middle East operate on the assumption that the world consists
only of nation states, and he shows that this leads analysts to ignore crucially
relevant information.
In general, political realism presents us with a telling example of what can
happen when the original reasons for adopting particular approaches, forms of
evaluation, or indices of measurement are ignored or forgotten, such that these
techniques become ends in themselves, regardless of their applicability in
actual contexts of human affairs. When this takes place in any field of inquiry,
the result is to narrow the perspective to the point at which it must ultimately
fail to yield a convincing account (Rubinstein et al. 1984 ). Yet it is just this kind
of process that has characterized discussions of power, violence and security.
Power
Discussions of the relationships among political groupings often focus on
disparities in access to advanced military technologies. For the most part, power
is taken in these discussions to refer to the range of measurable military,
technological or other such outcomes that can be effected by one group in its
relations with other groups (cf. Thibault and Kelly 1959, Cordesman and
Wagner 1990). In this sense, power is the ability to coerce other individuals or
groups to change their behaviour in some intended direction (Dahl 1969,
Zartman 197 4). The result of this kind of reasoning is that power has come to
be measured in terms of such indicators as concession rates, economic or
military pay-offs, and the like.
When policies are developed on the basis of such realist assumptions,
groups that control the disposition of material resources tend to be regarded as
powerful. Groups that do not control these resources are taken to be powerless.
Only physical and material resources are included in calculations of relative
power. Kim (1983:9) notes that 'the concept of "power" in mainstream realism
is excessively narrow and limited. This realism respects only material and
physical power and is contemptuous of "normative power" .... It denies the
existence of the world normative system.' By taking power as resting only on
material strengths, the domain of activities that are considered legitimately to
represent power is artificially restricted. This narrow view is indeed thoroughly
990
unrealistic, for it ignores the entire range of traditional and non-Western

conceptions of power. These alternative conceptions are much like what Kim
(1983:44) calls normative power, which 'is the ability to define, control and
transform the agenda'.
By explicating how this normative dimension of power works through
studies of actions below the level of the nation state, anthropologists have
focused on areas of experience that, at first glance, might appear to have little to
do with collective violence and security. Of course, they have not ignored the
material and technological aspects of power, for to do so would be naive (see
Otterbein 1973). However, anthropological work recognizes that normative
power can work even in the face of apparently superior power, as measured in
material terms. According to Wahrhaftig and Lukens-Wahrhaftig ( 1977:231 ),
the Cherokee conceive of this power as
sacred, not secular. It is an aspect of permanence granted each people at creation.

Autonomy and self-government are inseparable attributes of primordial power;
these are in the created nature of peoples, for each of the many distinct peoples set
forth at creation, of which Cherokees are one, was created self-governing. In
Cherokee myth, even animals and plants meet in council to determine their own
course of action-often with greater wisdom than humans. Such power is; it cannot
be gained.
More generally, this power can be said to accrue to a person through the
expenence o ea mg a mora y goo 1 e, w tc IS mar e y ea mg wtt ot er
people through social relations that are considerate and mutually respectful. It
is the process of living according to principle, not material force, that produces
power. 'To live according to one's laws is to be powerful' (Wahrhaftig and
Lukens-Wahrhaftig 1977:231).
The hardships experienced by Native Americans in the United States as a
result of military defeat, disease, external political control, and other kinds of
disasters and deprivations have been accompanied by material powerlessness.
In the face of such material hardships the focus of Indian groups on how things
are done rather than on what is done has allowed them to retain a sense of the
continuity of their ways of life and thus to retain their normative power. As the
case of the Cherokee demonstrates in particular, this normative power has
consequences in the political arena. It is their concept of, and respect for,
normative power that has enabled the Cherokee to build autonomous social,
political, and economic institutions, despite the repeated exercise of secular,
material power by whites (Gearing 1958, Wahrhaftig and Lukens-Wahrhaftig
1977). Indeed, normative power is rarely the inconsequential factor that it is
sometimes made out to be. To the contrary, normative power is an important
force which must be understood and counted in any reckoning of the 'balance
of power'.
Anthropological descriptions of normative power and its consequences show
that actions based in such conceptions of power can successfully challenge
991
SOCIAL LIFE
materially more powerful groups. For example, the Dene (Kehoe 1988) have
successfully opposed uranium mining and other nuclear-related actions, and
the Cherokee have successfully resisted their economic and cultural extinction
(Rubinstein and Tax 1985). The Palestinian intifoda is also an example of the
force of such non-materially based power (Schiff and Ya'ari 1990). The intifada,
like the actions of the Dene, the Cherokee, and of other indigenous peoples, or
of people in Iran (Bateson 1988; see also Beeman 1986) and China (Potter
1988), shows how power grounded in non-material, symbolic, normative
aspects of social and cultural life can achieve very real effects which, although
they cannot be neatly estimated by some quantitative index, make a significant
difference in the political arena.
Collective violence
Just as conceptions of power are culturally patterned, so conceptions of
collective violence have been channelled by cultural understandings, both lay
and professional. For the most part, discussions of collective violence have
focused on observable acts of violence launched by one group against another,
on the size and relations of military forces and on technological aspects of
fighting capabilities. Collective violence is described in terms of its intensity as
this is defined by battlefield deaths or the military technology used in a dispute.
Thus, for example, in deciding what to consider as 'war', only those conflicts in
w tc some cnttca num er o eat s uect y resu ts rom corn at are me u e
(Cohen 1986). The 'Correlates ofWar' Project undertaken at the University of
Michigan, for instance, defines a conflict as war only if it involved at least 1,000
battlefield deaths. And a 'conflict spectrum' (Sarkesian 1986: 116) has been
defined in terms of the destructive capabilities of the armaments employed or
in terms of the number of deaths directly resulting from combat.
It is obvious, however, that collective violence extends well beyond the range
of military aggression. War and violence, as contemporary political realities, are
nowadays very different from the conventional wars of other eras of human
history. Combat between opposing armies is now infrequent. In its place, 'war
is focused on the Third World, and pits guerrilla insurgencies against state
governments and states against indigenous nations' (Nietschmann 1987:1 ). The
direct killing and maiming of combatants is the unfortunate goal of war. But
civilians also die: in the Middle East, for example, since the Second World War
1.1 million deaths have resulted directly from wars and civil conflicts in the
regwn.
A less obvious effect is of the loss of this human power for society-the loss
of teachers, engineers, and manual workers to carry on the daily business of
keeping a society going. In the aftermath of war the society must support and
care for disabled veterans, and suffers the effects of angry men in its midst who
have been trained to kill. Some researchers have suggested that people
maturing in a society at war may suffer a form of moral and social retardation.
992
These people, and others, may develop long-term personality difficulties

resulting from the abnormal conditions in which they have grown up.
Moreover, war can profoundly affect civilian health. The civilians need not be
members of an enemy group; war may provide an excuse for the genocide of a
national minority population. Examples of such genocide are numerous, from
the German Holocaust of European Jews to the 'ethnic cleansing' of Bosnian
Muslims and the Guatemalan extermination of the indigenous Indian peasants
(Carmack 1988). Direct health effects on civilian enemies are also numerous,
but these are often ignored since the people killed are typically women,
children, and elders.
Less obvious effects of war on civilian health are the disruption of food
distribution and the breakdown of health care. In Sudan, the largest country in
Africa, the brutal civil war between the Moslem North and non-Moslem South
has undone much of the progress achieved by past development efforts, and
has reduced the prospects for development in the future. For example, for the
South the war has meant the near cessation, since 1985, of the drilling of
boreholes for fresh water, an exceptionally high infant mortality rate of 180 per
thousand, prevalent malnutrition among children of twelve years and below,
and the decimation of the infrastructure for primary and secondary health care
in the region.
There are no reliable estimates of how many children may have died in the
Sudanese war from 1983 to 1989. However, donor officials estimate that in 1988
a one a tota o 0,000 out erners te rom starvatiOn an re ate 1 nesses,
when both the SPLA and government troops bombed and threatened to
destroy food convoys. Roughly half of the Sudanese population is below 16
years of age, but reports indicate that many more than half of the 250,000 dead
were children. Reports prepared by Medecins Sans Frontieres, who were
working in the southern town of El Meiram, indicate that about two times as
many children died as adults. Extrapolating from this proportion, some
165,000 children may have died in one year of the war. In some areas the rate of
child mortality may have been even higher. A United States congressional fact-
finding committee reported that in the Abiye refugee camp in Southern Sudan,
every child under the age of two years and six months died. One factor
contributing to the higher death rate in children was a measles epidemic that
struck the malnourished and unimmunized child population in the summer of
1988. A UNICEF report on children in nine Southern African countries found
that war contributed directly to 20 per cent of the mortality of children under
the age of five.
In Zimbabwe from 1978 to 1980 the military carried out Operation Turkey,
destroying crops, livestock, and food supplies in order to starve the guerrillas.
The unfortunate consequence of this strategy was widespread malnutrition of
rural children and increased infant and childhood mortality. In Nicaragua the
Contra forces explicitly targeted health workers and health institutions. From
1981 to 1985, 38 health workers were killed and 28 kidnapped while they were
993
SOCIAL LIFE
performing medical duties; 61 health units were destroyed and 37 others forced
to close due to Contra activity. Because of the reduced availability of health
services, immunization, sanitation, nutrition and other health programmes
have been curtailed, and the health, especially of the rural peasants, has
suffered.
That the devotion of a disproportionate share of a nation's economy to
maintaining a military effort has negative effects on human services and on
social supports in that nation has been well documented (Melman 1965, 1986,
Pinxten 1986). Furthermore, the devotion of resources to the procurement of
arms has worldwide effects, causing distortions both within and between
national economies. Indeed, much of the inability of Third World countries,
and even some industrialized countries, to provide a basic level of food,
housing, and health care to their peoples can be traced directly to the distorting
effects of military expenditures.
Wars and civil conflicts over issues of ethnicity, self-determination, access to
resources and equity directly involve massive civilian populations. Violent
disruptions in a society disproportionately affect the most vulnerable: the poor,
women, and children. Like most pathogenic conditions, for every mortality
there are many more who are injured or suffer permanent disability. War affects
people, perhaps especially children, directly through death, disabling injury,
and psychological stress; indirect effects are disruption of health services and
education, impeded food distribution, family disruption and displacement,
estructwn o ousmg, water an samtatwn act 1t1es, an
funds for military needs (Zwi and Uglade 1989).
CULTURE AND NEGOTIATION

Negotiations are those communicative processes through which individuals or
groups try to resolve the disagreements that exist among them. Nearly every
human communicative interaction involves negotiation. Sometimes the
negotiation process is explicit. At other times it is taken for granted, and takes
place without the participants recognizing that they are involved in negotiation
at all. Whether explicit or not, negotiation is a shared process that occurs
within a social and cultural matrix that shapes both how problems are defined
and what solutions are conceivable. In general, negotiators seek to resolve
disagreements, which may involve eliminating the source of controversy.
Resolution may also result from reframing items under discussion, so that there
is no longer disagreement, or so that whatever disagreement persists is no
longer considered meaningful by those involved. In any event, negotiators work
within the boundaries of their cultural expectations and symbolic frameworks
to judge the outcomes of their efforts.
When negotiators come from a common background many fundamental
aspects of the negotiation process are part of their shared tacit knowledge-like
whether a proposal should be made with blunt straightforwardness or instead
994
with artful indirection. When negotiation involves actors with sensibilities,

understandings, and expectations grounded in different cultural backgrounds,
additional complexities are involved. In such instances, structures of
understanding and patterns of behaviour and communication that might
otherwise be effective, and thus taken for granted, may produce paradoxical
results-such as unintended insult or confusion where clarity was intended.
In this section I sketch some of the ways in which culture provides the
context for negotiation and the control of collective violence. For illustration I
show how matters of culture and communication impinge on negotiations
between Arabs and Israelis. (Because discussions among residents of the region
are sometimes referred to as taking place between Arabs and Israelis, it is easy
to suppose that all Arabs share a single culture and set of metacommunicative
rules. It is important to recognize that just as differences exist between and
within the Palestinian and Israeli communities, there are also cultural and
metacommunicative differences among and between Palestinians and other
Arabs.) The literature concerning negotiation and conflict resolution is large. It
is not my intention to survey that literature here. Rather, I merely wish to
illustrate how symbolic repertoires and cultural traditions shape, and are in
turn shaped by, processes of negotiation.
Studying negotiation
ommumcatmg wtt ot ers m or er to arnve at a reso utwn o 1 erences IS
the essence of negotiation. It 'is a basic means of getting what you want from
others. It is a back-and-forth communication designed to reach an agreement
when you and the other side have some interests that are shared and others that
are opposed' (Fisher and Ury 1981 :xi). Like many other processes that are
ubiquitous in social life, negotiation ranges from the mundane and taken-for-
granted to the elaborately formal and institutionalized.
The process and patterns of various kinds of negotiations have been studied
in some depth. In general, such studies have had two very different emphases.
The first is most evident in analyses of institutionalized forms of negotiation,
like bargaining in the context of labour relations or in arms control talks. These
analyses have tended to study negotiation through one or more of three general
strategies: ( 1) through laboratory experiments, (2) in terms of abstract
mathematical decision and game theoretic models, or (3) through qualitative
analysis of the recollections of participants in particularly important
negotiations, like the Cuban Missile Crisis Oanis 1983) or the Camp David
negotiations (Raiffa 1982). Especially when laboratory analyses and
mathematical modelling have been used, this approach to the study of
institutionalized negotiations has sought to describe their formal
characteristics.
The second and less commonly adopted approach examines the implicit
negotiations in daily life. The aim of these studies is to understand how
995
SOCIAL LIFE
agreement is reached through the interaction of individuals. Anselm Strauss

(1978), for instance, argues that all social orders are to some degree negotiated
orders. To understand the forms that negotiation takes, researchers focus on the
effects of the larger social context on the ways in which people in particular
societies actually resolve their differences. Much of this study of mundane
negotiation is to be found embedded in anthropological accounts of conflict
resolution. Greenhouse (1986:54-8), for example, describes how the resolution
of significant differences among the inhabitants of a small American town
depends upon a calm, negative attitude towards conflict. By contrast, in
Egyptian popular culture one form of negotiation involves a ritual
pantomime-a dowsha-in which sham gestures of violence are used to focus
attention on a dispute, and to attract and justify the intervention of third
parties with a view to resolving the dispute and re-establishing harmony (Rugh
1982:xvi). By explicating episodes of mundane negotiation these studies help to
reveal the cultural and symbolic components that contribute to successful
negotiations in particular societies.
Lessons from formal negotiations

Studies of institutionalized negotiation have been of central concern to
students of international affairs (e.g. Ikle 1964, Schellenberg 1982), and have
provided considerable insight into its formal aspects. By examining how groups
an m IVI ua s reso ve tsputes m contro e settmgs, t ese stu tes ave
explicated the structural stages in the process of achieving agreement and the
formal properties of decision-making in bargaining situations. The resulting
literature mainly develops two lines of thinking. The first often describes both
actual and possible negotiations (i.e. those simulated in laboratory settings) as
instances of'n-player games', and analyses how decisions conform to models of
rational decision-making (e.g. Raiffa 1982). The second, related approach is to
consider how prospects for negotiations can be improved, for example by
creating situations in which both sides can win or by developing a variety of
confidence-building mechanisms. This line of work has resulted in several
guidebooks which describe how to negotiate successfully and fairly (e.g. Fisher
and Ury 1981 ). Both approaches provide useful starting points for thinking
about negotiation and peacemaking.
Studies of this general kind began in the 1940s with the analysis of two-
player, single-choice games. The field rapidly developed, however, into one in
which sophisticated analyses are made of ongoing, multiple-player
negotiations. According to these analyses a number of structural features are
critical to the success of negotiation, including: (I) the number of negotiating
parties involved; (2) the degree of consensus existing within each negotiating
group; (3) whether the negotiation is ongoing or discrete; (4) the number of
issues being considered and the connections between them; ( 5) the linkage of
the negotiations to other issues; (6) whether discussions are held in public or in
996
private; and (7) how agreements reached through the negotiation will be
enforced (Raiffa 1982: 11-19).
For instance in the context of negotiations between Israelis and Palestinians,
neither of the parties can be taken to be single actors. Each represents a diverse
constituency and any negotiating team contains internal divisions.
Furthermore, there are variations among Palestinians in their perceptions of
the land of Palestine and the possibilities for satisfactory settlement, and these
run along the lines of regional and religious-ideological identity (Lesch and
Tessler 1989, Grossman 1988). Between Palestinians living in the Occupied
Territories of the West Bank and Gaza, Arabs living within Israel and
Palestinians living in Jordan, there exist differences in the perception of the
nature of the 'problem' and possible solutions to it. Shehadeh ( 1982), who
chose to remain in the West Bank as a samid-one who resists Israeli
occupation by leading a life of principled non-co-operation and non-
acquiesence in Israeli authority-describes how his perceptions of political
action and of his attachments to the land came to differ markedly from those of
his cousin residing in Jordan, and how he felt almost alienated from Arabs in
Acre (Shehadeh 1982:7-11, 20-3). In addition, there are ideological loyalties
that cross-cut and confuse this variation: the scorn of the freedom fighter and
political prisoner for the samidin is keenly felt, as is the frustration felt by the
samidin in response to the romanticization of the conflict by Palestinians living
abroad (Shehadeh 1982:23-6, 56-8).
T e Israe 1 commumty IS s1m1 ar y IVI e m opmwn an perceptiOn,
depending upon religious-ideological and regional factors. Views on the nature
and possible resolution of the 'problem' of the Occupied Territories are shaped
by political affiliations, religious commitments, and personal experience,
among other factors. Benvenisti ( 1989) describes the range of these variations,
and Shavit (1991) describes the variety of reactions to military service in a Gaza
Strip internment camp. The divisions internal to Israeli society are evident in
the diversity of political parties, both religious and ideologically based, and of
social movements like Peace Now and Gush Emunim (the latter of which seeks to
develop Israeli settlements in the West Bank).
Under such circumstances, presenting a united front in negotiations is an
extremely difficult task for each party. Privately and in public, both must
negotiate among themselves in order to arrive at bargaining positions that can
be put forward, and considerable intra-group negotiation is needed in order to
arrive at responses to proposals made by their interlocutors. These intra-group
negotiations, moreover, may themselves be explicit or tacit, conducted in public
or in private. In addition, negotiators must continually touch base with their
constituencies. All of these tasks are difficult, and failure in either group's
internal negotiations may place in jeopardy the possibilities and potentials for
intergroup negotiations (Fahmy 1983, Maksoud 1985, Eban 1985, Grossman
1988, Friedman 1989, Schiff and Ya'ari 1990, PASSIA 1991, Alternative
Information Centre n.d.).
997
SOCIAL LIFE
In addition to identifying the structural characteristics of negotiations, some

analysts have attempted to explicate the role of different techniques within
negotiation settings. Ikle (1968:117-18), for instance, describes the techniques
of threat and commitment. In the former, one of the negotiating parties asserts
its intention to cause the other party the loss of some valued asset should the
other party not comply. Threat, of course, may be credible or bluff.
Commitment, on the other hand, imposes constraints on the party making it.
By making a commitment a negotiating party makes it difficult for itself to
renege on a position it has advanced. Because such limitation is self-imposed,
the act of commitment is a move to convince the other negotiating parties of the
sincerity of the position advanced.
Other researchers have sought to transform the information derived from
analytic studies of negotiation into practical and straightforward advice for
improving negotiation practice (Karrass 1970, Coffin 1976, Fisher and Ury
1981 ). Some of the books in this genre offer useful suggestions about how to
conduct negotiations. Fisher and Ury (1981:11), for instance, develop a method
they call 'principled negotiation' or 'negotiation on the merits', which is really
concerned with meta-negotiation. Each move is to be made with the awareness
that it 'helps structure the rules of the game you are playing' (Fisher and Ury
1981: 10 ). They contrast this to the more usual kind of account which regards
negotiation as a process of 'positional bargaining' in which negotiators define
and defend their respective positions. The parties, however, are inclined to
a opt t ese positiOns as t etr own raison etre, an t IS can east y cause t etr
underlying interests to be overshadowed.
The method of principled negotiation depends upon four general strategies
for ensuring good negotiations, which Fisher and Ury ( 1981:11) sum up as
follows:
Separate the people from the problem. Focus on interests not positions. Generate a
variety of possibilities before deciding what to do. Insist that results be based on
some objective standard.
The method is intended to have very practical results; to produce wise

agreements, to do so efficiently, and to allow the parties to separate on amicable
terms. Fisher and Ury (1981:14) claim that
in contrast to positional bargaining, the principled negotiation method of focusing

on basic interests, mutually satisfying options, and fair standards typically results in
a wise agreement. The method permits you to reach a gradual consensus with a joint
decision efficiently without all the transactional costs of digging into positions only
to have to dig yourself out of them. And separating the people from the problem
allows you to deal directly and empathetically with the other negotiator as a human
being, thus making possible an amicable agreement.
The method of principled negotiation has been put to very good use. Its
998
directives are admirable and productive, especially in situations where

negotiators share tacit understandings of the general nature and purpose of
negotiation. Like other methods of understanding negotiation, however, its
application encounters a unique set of obstacles when applied in cross-cultural
contexts.
Culture and negotiation

Once a method of understanding or action is developed which is said to be
universally applicable, it is easy to become over-optimistic about the
possibilities for using it to solve previously intractable problems. This is
especially prone to happen in the context of international and intercommunal
disputes (Rubinstein 1988a, Rubinstein and Foster 1988).
An example of the complexities involved in real international disputes can
be drawn from the Camp David negotiations between Egypt and Israel. After a
long and difficult process of negotiation, Egypt and Israel signed the Camp
David peace accords in September 1978. It is widely acknowledged, however,
that the successful conclusion of this accord did not result in an equally
successful peace. Israeli and Egyptian accounts and interpretations of the
course of their post-accord relations vary widely, and each side has found that
its expectations have not been met (Lesch and Tessler 1989, Fahmy 1983,
Cohen 1990). For these reasons, the peace between the two countries has been
escn e as a co peace.'
Fisher and Ury's (1981:4) view is that a good negotiation method 'should
produce a wise agreement if agreement is possible. It should be efficient. And it
should improve or at least not damage the relationship between the parties.' By
these criteria negotiations between Egypt and Israel over the Camp David
accords and subsequently must be judged as wanting: resolutions have been
achieved, but with each subsequent negotiation the relationship between the
two countries appears to have deteriorated (see Cohen 1990). However,
negotiators whose tacit cultural knowledge leads them to see efficiency and the
improvement of interpersonal relations as mutually exclusive may not view
these criteria so positively. More obviously, the search for objective standards
for use in resolving disputes may produce greatly varying responses: what one
person takes to be neutral objectivity is not infrequently taken by another to be
biased in the extreme (Rubinstein 1989:52-6). In the Egyptian-Israeli case, for
instance, the record seems to indicate that each side would view its conduct in
negotiations as principled. Yet, each views the other as having dealt with it in
bad faith (Fahmy 1983, Cohen 1990).
Dealing with longstanding problems in cross-cultural negotiations reveals a
variety of pitfalls that guides to negotiation technique and formal models of the
negotiating process are unable to overcome. In order to deal successfully with
the problems presented by cross-cultural negotiation, it is necessary to have an
understanding of culture as a dynamic, symbolically based system through
999
SOCIAL LIFE
which people construct and enact meaning (Kertzer 1988). One of the most
salient symbolically based aspects of the Israeli-Palestinian issue is the way in
which the devotion to the land of Israel/Palestine has become invested with
multiple meanings and emotions. Both Palestinian and Israeli interlocutors
bring to their discussions a symbolic understanding which frames their
discourse. The Palestinian concept of 'the preserving' (samid), and the Israeli
conception of a special homeland (moledet) exert powerful emotional and
cognitive influences on those who hold them (Shehadeh 1982, Benvenisti
1989).
Successful cross-cultural negotiation depends, therefore, upon integrating
the results of formal studies of negotiation with contextual information about
the role of culture in mundane negotiation processes. The following section of
this article considers the importance of intra-cultural variability and the role of
symbols in political discourse.
Culture and internal variability

In part because negotiating cross-culturally introduces new difficulties, interest
in the formal aspects of negotiations has been supplemented by attempts to
characterize national negotiating and decision-making styles. It is, however,
misguided to rely on stereotyped characterizations of cultural negotiating
styles, since this is to assume that cultures are homogeneous and stable, and
t at once escn e t e patterns stay mtact. T e cu tura c aractenzatwn o
patterns of behaviour, belief, and interaction is not in itself fallacious. Such
characterizations can be useful if they are clearly anchored in specific
circumstances. But it is always misguided and unhelpful to treat them as
though they had a permanent existence, outside time and history. To do so is to
commit what I call the 'fallacy of detached cultural descriptions'.
Anthropological work shows that cultural norms and preferences, such as
for social harmony or directness, do indeed exist, but that not all individuals
from a particular society will hold or behave according to a single set of norms.
And, of course, such norms are constantly affected by social, political,
economic, and other processes and contradictions within the society. Thus,
cultural styles are not stable, even if they may be clearly discerned in relation to
a particular problem or situation. This is because societies always contain
within themselves a variety of styles, some of which will be in direct tension
with each other.
Ismail Fahmy, former Egyptian Minister of Foreign Affairs and Deputy
Premier, recalls (1983:124) that,
It takes time to learn to deal with the Soviets and understand their tactics. For
example, the Russian negotiator never answers 'da' (yes) at the outset. The answer is
always 'niet'. Often the first 'niet' means 'da', but at other times 'niet' is 'niet.' The
problem is to learn to tell the difference. Once I learned, I enjoyed tremendously
1000
negotiating with the Soviets. It was always tough, but they could be outmanoeuvred
once their tactics were understood.
Yet during the period in which Minister Fahmy was dealing with the Soviets,
their interests in the region shifted many times, as did the constraints on their
actions. As even the record of missed opportunities and misunderstandings
reported in his own memoirs shows, Fahmy's view that once understood, Soviet
negotiators could henceforth be handled with aplomb, was in fact a chimera.
Beeman ( 1986, 1989) and Bate son ( 1988), for example, describe how the
assumptions of United States negotiators about Iranian political styles proved
inaccurate, precisely because they failed to be aware of cultural heterogeneity.
Bateson and her colleagues (see Bateson 1988) isolated two distinct forms of
political discourse in Iran-the opportunistic and the absolute. At the time of
the Iranian revolution public rhetoric and public policy changed in ways that
baffled United States analysts. Yet, Beeman and Bateson argue, when it is
recognized that contrasting themes generally coexist in any culture, these
events are more readily understandable. As Bateson (1988:39) puts it,
Iranian public policy and public rhetoric, both domestically and internationally,
went through an apparent radical change at the time of the revolution into a style
that appeared totally different and therefore unpredictable, but we would argue that
the two styles-and more significantly the tendency to think of them as alternatives
. . . . .
Understanding that opposing styles exist in any society, and being aware of
which styles are ascendant in a particular situation, requires that the analyst be
aware of the different contexts in which negotiators frame their work, and
further requires them to understand how the give-and-take of social process in
these situations keeps the cultural matrix in which actions are situated in a
constant state of flux. Indeed, 'the truth of the matter is that people have mixed
feelings and confused opinions, and are subject to contradictory expectations
and outcomes, in every sphere of experience' (Levine 1985:8-9).
In sum, it is as misleading to attend exclusively to autobiographical
recollections of formal negotiations as it is to rely on laboratory simulations or
on the mathematical modelling of decision-making processes. Studies that rest
on such analyses direct our attention towards a limited number of
characteristics of negotiations, and away from other less easily explained or
measured, but nevertheless equally critical, aspects of the negotiation process
(Rubinstein 1989).
Culture, symbols, and negotiation

The elements of negotiating competence in one culture may ensure failure in
negotiation in another. This is because metacommunicative rules of negotiation
1001
SOCIAL LIFE
are culturally specific. Egyptian communicative competence places a high value

on maintaining face agreement and a smooth and harmonious social order. As a
result negotiations are often structured in a way that is cyclical in form,
incorporating within them a large amount of repetition. Once a point is put
forward, in a relatively indirect way, it is discussed until a sense of closure
appears imminent. At this stage, the discussion might return to consider the
point anew. Again closure is approached, and again discussion is reopened.
This next episode of discussion may be briefer than its predecessor; this
process continuing until all parties have had a chance to speak fully to the point
and consensus is presumed. Each of these episodes of discussion may be quite
animated, and important information may be conveyed in an indirect fashion.
All of this might well strike a Western observer as wasteful of both time and
energy. It is true that this pattern of negotiation is not efficient in reaching a
conclusion-but it is efficient for maintaining social harmony. (Descriptions of
Egyptian communicative styles derive from my own work and from that of
Cohen (1990). The analysis of Israeli negotiating styles presented in this
section is based primarily on Cohen.)
The rules of communicative competence characterizing Israeli negotiations
are very different. There, according to Cohen (1990), little care is taken to
sugarcoat positions that may be unpalatable to an interlocutor. Rather the
emphasis is on direct, forthright, 'clear' communication. Thus, negotiating
positions tend to be put forward directly, and little attention is paid to the
uman st e o t e socta transactiOn. n t e ot er an , w en every wor IS
listened to, analysed, and taken seriously, as it is by Israeli negotiators, the use
of artful ambiguity and hyperbole, often employed by Egyptian negotiators,
rankles and insults just as deeply as does blunt disregard for social niceties.
Communication, of course, is more than just the content of a message.
Language, like all symbols, is essentially ambiguous. There is nothing novel in
the observation that the same words, spoken in different ways or in different
contexts, may convey a range of different meanings (on this, see DeBernardi in
this volume, Article 31 ). Indeed, Cohen (1990) shows that Israeli and Egyptian
interlocutors repeatedly misunderstand one another, and take insult from their
interaction, precisely because their metacommunicative expectations are not
mutually consistent.
Among the many examples that Cohen offers, his description of the first
meeting between Boutros Boutros-Ghali and Moshe Dayan, who at the time
were acting as foreign ministers of Egypt and Israel respectively, is instructive.
Cohen (1990:57-58) observes:
Within hours of President Sadat's historic arrival in Israel, on the evening of 19

November 1977, with nerves at a high pitch of anticipation, Israeli diplomacy made
its first tactless and maladroit overture .... Without trying to soften the blow in any
way, Dayan brusquely informed Boutros-Ghali, with astonishing insensitivity, that
since there was no chance of Jordan or the Palestinians' joining in the negotiations-
1002
as Sadat hoped at that point, anxious to avoid isolation in the Arab world-Egypt
had to be ready to sign a peace treaty with us [Israel] even if she were not joined by
others.
Boutros-Ghali was profoundly shocked by Dayan's ill-timed proposal of a
separate peace, as was Sadat when it was reported to him. At issue was not the idea
itself, which was based on an objective analysis of the situation .... It was the
unsubtle directness of the approach that was utterly repellent to the Egyptian
minister. This first conversation with an Israeli leader rankled in Boutros-Ghali's
mind for years afterward.
The value placed on directness is not the only communicative expectation over
which Egyptians (and other Arabs) and their Israeli counterparts diverge.
Israeli negotiators often appear to be immediately concerned with working out
the details of an agreement. By contrast, Arab diplomats have tended to seek
frameworks for solution, leaving aside the details. For the Israeli actor attention
to the precise wording of an agreement is considered an expression of good
faith, whereas for the Egyptian negotiator good faith is displayed by agreement
to a broad conceptual framework; the details are left to be worked out at a
future time (see Carter 1982:342, Fahmy 1983:285-308).
Raymond Cohen (1990) traces these and other obstacles to negotiations
between the Israelis and the Egyptians, and other Arabs. Such obstacles all
belong outside the structural character of formal negotiations. Indeed, both the
Israeli and the E tian ne otiators understand and seek to adhere to the
structural features of negotiations, as these are understood by the international
diplomatic community. The stumbling blocks that remain are the result of
conflicting metacommunicative expectations.
CONCLUSION
Expectations about what is proper and good are cultural, and they are encoded
in a society's symbolic forms. Most importantly, symbols are ambiguous in that
they may have several meanings-being often imprecisely defined-and they
may invoke emotional responses. As Abner Cohen (1979:89; see also Kertzer
1988) notes, cultural symbols have great political impact because they allow
political relationships to be 'objectified, developed, maintained, expressed, or
camouflaged by means of symbolic forms and patterns of symbolic action'.
Such symbolic forms include, among other things, the repetitive, ritual
organization of negotiations (Rubinstein 1988b ), the public rhetoric of political
leaders (Cohen 1990:45-8), and the literature of resistance (Lesch and Tessler
1989: 125-39). Because symbolic forms have both an ambiguous cognitive
component and a strong emotional load they are powerful factors in structuring
political perceptions.
Such cultural factors affect the patterning of collective violence both direct
and indirect, and of conceptions of power and security. Moreover, the cultural
1003
SOCIAL LIFE
factors that affect negotiation, such as metacommunicative expectations, are

encoded in symbols. These cultural factors structure the way that negotiators
respond to their interlocutors, they affect the perception of what is fair and
objective, and of how to begin and end discussions. Especially when
disagreement is emotionally laden and rich in symbolic elements, it is all the
more necessary to appreciate the role of culture in the dynamics of negotiation,
if we are to gain a better understanding of collective violence, an understanding
that is vital to present and future security.
ACKNOWLEDGEMENT
Preparation of this article was supported in part by a grant from the
Ploughshares Fund, which I gratefully acknowledge. I thank Mary LeCron
Foster and Sandra D.Lane for comments on an earlier draft. Many colleagues
responded to my circular letter requesting references to pertinent literature. I
am grateful for their helpful replies. Much of the substance of this article is
drawn from my earlier published papers, especially Rubinstein 1988a, 1989,
and 1992.
REFERENCES
Alternative Information Centre (n.d.) Three Years of Intifoda. News from Within from
ecem er to t e u rtsts, erusa em: erna ive n orma ion en re.
Ardrey, R. (1966) African Genesis, New York: Doubleday.
Ball, D. and Jeffrey R. (eds) (1986) Strategic Nuclear Thinking, Ithaca, NY: Cornell
University Press.
Bateson, M.C. (1988) 'Compromise and the rhetoric of good and evil', in
R.A.Rubinstein and M.L.Foster (eds) The Social Dynamics of Peace and Conflict.
Culture in International Affoirs, Boulder, Colo.: Westview Press.
Beeman, W.O. (1986) 'Conflict and belief in American foreign policy', in M.L.Foster
and R.A.Rubinstein (eds) Peace and War: Cross-Cultural Perspectives, New
Brunswick, NJ: Transaction Books.
--(1989) 'Anthropology and the myths of American foreign policy', in P.R.Turner
and D.Pitt (eds) The Anthropology of War and Peace. Perspectives on the Nuclear Age,
Granby, Mass.: Bergin and Garvey.
Benvenisti, M. (1989) Conflicts and Contradictions, New York: Eshel Books.
Binnendijk, H. (ed.) (1987) National Negotiating Styles, Washington, DC: United
States Department of State.
Brams, S.J. (1985) Superpower Games: Applying Game Theory to Superpower Conflict,
New Haven: Yale University Press.
Brass et, D. ( 1988) 'Values and the exercise of power: military elites', in R.A.Rubinstein
and M.L.Foster (eds) The Social Dynamics of Peace and Conflict: Culture in
International Security, Boulder, Colo.: Westview Press.
Carmack, R.M. (ed.) (1988) Harvest of Violence. The May a Indians and the Guatemala
Crisis, Norman: University of Oklahoma Press.
Carter, J. (1982) Keeping Faith: Memoirs of a President, New York: Bantam.
1004
Chagnon, N. (1967) The Fierce People, New York: Prentice-Hall.

--(1988) 'Life histories, blood revenge, and warfare in a tribal population', Science
239:985-92.
Chagnon, N. and Hames, R. (1979) 'Protein deficiency and tribal warfare in Amazonia:
New data', Science 203:910-13.
Checinski, M. (1981) A Comparison of the Polish and Soviet Armaments Decisionmaking
Systems, Santa Monica: RAND Corporation.
Coffin, R.A. (1976) Negotiator: A Manualfor Winners, New York: Everyday Handbook
Service.
Cohen, A. (1979) 'Political symbolism', Annual Review ofAnthropology 8: 87-113.
Cohen, Raymond (1990) Culture and Conflict in Egyptian-Israeli Relations,
Bloomington: Indiana University Press.
Cohen, Ronald (1983) 'War, war machines, and state formation in pre- and post-
industrial states', paper presented at the International Congress of Anthropological
and Ethnological Sciences, Vancouver, BC, Canada, August 1983.
--(1986) 'War and war proneness in pre- and postindustrial states', in M.L.Foster
and R.A.Rubinstein (eds) Peace and War: Cross-Cultural Perspectives, New
Brunswick: Transaction Books.
Cohn, C. (1987) 'Sex and death in the rational world of defense intellectuals', Signs
12(4):687-718.
Cordesman, A. and Wagner, A. (1990) The Lessons of Modern War, vol. 1, Boulder,
Colo.: Westview Press.
Dahl, R. (1969) 'The concept of power', in R.Bell, D.Edwards and R.Wagner (eds)
Political Power, New York: The Free Press.
avis, . . an Wmne e , . . e trategy ssessment enter: n
Overview and Interim Conclusions about Utility and Development Options, Santa
Monica, Cal.: RAND Corporation.
Eban, A. (1985) 'Multilateral diplomacy in the Arab-Israeli conflict', in A.Lall (ed.)
Multilateral Negotiation and Mediation. Instruments and Methods, New York:
Pergamon Press.
Ember, M. and Ember, C. (1988) 'Fear of disasters as an engine of history: resource
crises, warfare, and interpersonal aggression', paper prepared for the conference
'What is the Engine of History?', Texas A&M University, 27-29 October 1988.
Fahmy, I. (1983) Negotiating for Peace in the Middle East, London: Croom Helm.
Ferguson, R.B. (n.d.) 'Game wars? Ecology and conflict in Amazonia', manuscript,
Department of Anthropology, Rutgers University, Newark, NJ.
Fisher, R. and Ury, W. (1981) Getting to Yes, Boston: Houghton Mifflin.
Foster, M.L. (1989) 'The causes of conflict', paper prepared for the World Affairs
Council of Northern California, 14 July.
Foster, M.L. and Rubinstein, R.A. (eds) (1986) Peace and War: Cross-Cultural
Perspectives, New Brunswick, NJ: Transaction Books.
Fried, M., Harris, M. and Murphy, R. (eds) (1968) War: The Anthropology of Armed
Conflict and Aggression, Garden City, NY: Natural History Press.
Friedman, T. (1989) From Beirut to Jerusalem, New York: Farrar Straus Giroux.
Gearing, F. (1958) 'The structural poses of 18th century Cherokee villages', American
Anthropologist 60: 1148-57.
Greenhouse, C. (1986) 'Fighting for peace', in M.L.Foster and R.A.Rubinstein (eds)
Peace and War: Cross-Cultural Perspectives, New Brunswick: Transaction Books.
1005
SOCIAL LIFE
Gross, D. ( 1975) 'Protein capture and cultural development in the Amazon Basin',
American Anthropologist 77:526-49.
Grossman, D. (1988) The Yellow Wind, London: Jonathan Cape.
Herran, F. ( 1988) 'Conclusions of a transdisciplinary investigation of the causes of war
and the conditions of peace', manuscript, Universidad Nacional de Salta, Argentina.
Hinde, R.A. (1988) Aggression: Integrating Ethology and the Social Sciences. Medicine
and War, New York: John Wiley.
Hosmer, S. (1985) Constraints on US. Strategy in Third World Conflict, Santa Monica,
Calif.: RAND Corporation.
Ikle, F. (1964) How Nations Negotiate, New York: Harper.
--(1968) 'Negotiation: A definition', in D.Sills (ed) International Encyclopedia of the
Social Sciences, New York: Free Press, Macmillan.
Janis, I. (1983) Groupthink. Psychological Studies of Policy Decisions and Fiascoes, 2nd
edn, Boston: Houghton Mifflin.
Karrass, C.L. (1970) The Negotiating Game: How to Get What you Want, New York:
Crowell.
Kehoe, A.B. (1988) 'Fourth world responses to external threats: the Dene', in
R.A.Rubinstein and M.L.Foster (eds) The Social Dynamics of Peace and Conflict:
Culture in International Security, Boulder: Westview Press.
Kertzer, D. (1988) Ritual, Politics, and Power, New Haven: Yale University Press.
Kim, S. (1983) The Quest for a Just World Order, Boulder, Colo.: Westview Press.
Koch, K.-F. (1974) The Anthropology of Warfare, Addison-Wesley Modules in
Anthropology no. 52, Reading, Mass.: Addison-Wesley.
Lesch, A. and Tessler, M. (1989) Israel, Egypt, and the Palestinians. From Camp David to
t e ntt a a, oommgton: n 1ana mvers1ty ress.
Levine, D.N. (1985) The Flight from Ambiguity: Essays in Social and Cultural Theory,
Chicago: University of Chicago Press.
Lorenz, K. (1963) On Aggression, New York: Harcourt, Brace and World.
Maksoud, C. (1985) 'Arab League negotiations', in A.Lall (ed.) Multilateral Negotiation
and Mediation. Instruments and Methods, New York: Pergamon Press.
Malinowski, B. (1941) 'An anthropological analysis of war', American Journal of
Sociology 46(4):521-50.
Melman, S. (1965) The Permanent War Economy, New York: Simon & Schuster.
--(1986) 'The war-making institutions', in M.L.Foster and R.A.Rubinstein (eds)
Peace and War: Cross-Cultural Perspectives, New Brunswick, NJ: Transaction Books.
Morris, D. (1967) The Naked Ape, New York: Doubleday.
Myrdal, G. (1969) Objectivity in Social Research, Middletown, Conn.: Wesleyan
University Press.
Nietschmann, B. (1987) 'Militarization and indigenous people', Cultural Survival
Quarterly 11(3):1-16 ..
Otterbein, K. (1973) 'The anthropology of war', in J.J.Honigmann (ed.) Handbook of
Social and Cultural Anthropology, Chicago: Rand McNally.
PASSIA (Palestinian Academic Society for the Study of International Affairs) (1991)
Palestinian Assessments of the Gulf War and its Aftermath, East Jerusalem: PASS lA
Publications.
Pinxten, R. (1986) 'The developmental dynamics of peace', in M.L.Foster and
R.A.Rubinstein (eds) Peace and War: Cross-Cultural Perspectives, New Brunswick
NJ: Transaction Books.
1006
Potter, J. ( 1988) 'The communist ethic and the spirit of China's party cadres', in R.
A.Rubinstein and M.L.Foster (eds) The Social Dynamics of Peace and Conflict:
Culture in International Security, Boulder, Colo.: Westview Press.
Raiffa, H. (1982) The Art and Science of Negotiation, Cambridge, Mass.: Harvard
University Press.
Rappaport, R. (1967) Pigs for the Ancestors, New Haven, Conn.: Yale University Press.
Record, J. (1985) 'Third world conflicts: implications for US security and force
structure', in K.A.Dunn and W.O.Staudenmaier (eds) Alternative Military
Strategies for the Future, Boulder, Colo.: Westview Press.
Ross, J. (1980) 'Ecology and the problem of tribe: A critique of the Hobbesian model of
preindustrial warfare', in E.Ross (ed.) Beyond the Myths of Culture: Essays in
Cultural Materialism, New York: Academic Press.
Ross, E. and Ross,J. (1980) 'Amazon warfare', Science 207:590-1.
Rubinstein, R.A. (1988a) 'Cultural analysis and international security', Alternatives
13(4):529-42.
--(1988b) 'Ritual process and images of the other in arms control negotiations',
Human Peace 6(2):3-7.
--(1989) 'Culture, international affairs and peacekeeping: confusing process and
pattern', Cultural Dynamics 2(1):41-61.
--(1992) 'Culture and negotiation', in E.Fernea and Mary Hocking (eds) The
Struggle for Peace: Israelis and Palestinians, Austin: University of Texas Press.
Conflict: Culture in International Security, Boulder, Colo.: Westview Press.
Rubinstein, R.A. and Tax, S. (1985) 'Power, powerlessness, and the failure of political
rea Ism , m ens roste et a . e s attve ower, ergen: mvers1tets or aget.
Rubinstein, R.A., Laughlin, C. and McManus, J. (1984) Science as Cognitive Process,
Rugh, A. (1982) 'Foreword', in Khul-Khaal. Five Egyptian Women Tell Their Stories,
Nayra Atiya, Syracuse: Syracuse University Press.
Sahlins, M. (1968) Tribesmen, Englewood Cliffs, NJ: Prentice-Hall.
--(1976) The Use and Abuse of Biology, Ann Arbor: University of Michigan Press.
Sarkesian, S. (1986) 'The nature of war and the American military profession', in M.
L.Foster and R.A.Rubinstein (eds) Peace and war: Cross-Cultural Perspectives, New
Brunswick, NJ: Transaction Books.
Schellenberg, J. (1982) The Science of Conflict, Oxford: Oxford University Press.
Schiff, Z. and Ya'ari, E. (1990) Intifoda: The Palestinian Uprising-Israel's Third Front,
New York: Simon & Schuster.
Shavit, A. (1991) On Gaza Beach, The New York Review of Books 38(13):3-5.
Shehadeh, R. (1982) The Third Way: A Journal of Life in the West Bank, London:
Quartet Books.
Simon, H.A. (1983) Reason in Human Affoirs, Stanford, Cal.: Stanford University
Press.
Strauss, A. (1978) Negotiations, San Francisco: Jossey-Bass.
Swanton, JR. (1943) Are wars Inevitable?, Washington, DC: Smithsonian Institution.
Tefft, S.K. (1974) 'Warfare regulation: a cross-cultural test of hypotheses', in M.
Nettleship, R.D.Givens and A.Nettleship (eds) War: Its Causes and Correlates, The
Hague: Mouton.
1007
SOCIAL LIFE
Thibault, J. and Kelly, H. (1959) The Social Psychology of Groups, New York: John
Wiley.
Turner, P. and Pitt, D. (eds) (1989) The Anthropology of War and Peace. Perspectives on
the Nuclear Age, Granby, Mass.: Bergin and Garvey.
Vayda, A.P. (n.d.) 'Explaining why Marings fought', manuscript, Department of
Anthropology, Rutgers University, New Brunswick, NJ.
--(1974) 'Warfare in ecological perspective', Annual Review of Ecology and
Systematic 5:183-93.
--(1968) 'Hypotheses about the functions of war', in M.Fried, M.Harris and
R.Murphy (eds) War: The Anthropology of Armed Conflict and Aggression, Garden
City, NY: Natural History Press.
Vincent, J. (1990) Anthropology and Politics: Visions, Traditions, and Trends, Tucson:
University of Arizona Press.
Wahrhaftig, A. and Lukens-Wahrhaftig, J. (1977) 'The thrice powerless: Cherokee
Indians in Oklahoma', in R.Fogelson and R.Adams (eds) The Anthropology ofPower:
Ethnographic Studies from Asia Oceania and the New World, New York: Academic
Press.
Wilson, E.O. (1975) Sociobiology, Cambridge, Mass.: Harvard University Press.
--(1978) 'What is sociobiology?', in M.Gregory, A.Silvers and D.Sutch (eds)
Sociobiology and Human Nature, San Francisco: Jossey-Bass.
Worsley, P. (1982) 'Non-western medical systems', Annual Review of Anthropology 11:
315-48.
--(1986) 'The superpowers and the tribes', in M.L.Foster and R.A.Rubinstein (eds)
Peace and War: Cross-Cultural Perspectives, New Brunswick, NJ: Transaction Books.
ntro uctwn , m . ors ey an . . a JOr e s n t e rtn :
Proliftration and the Third World, London: Third World Books.
Worsley, P. and Hadjor, K.B. (eds) (1987) On the Brink: Nuclear Proliftration and the
Third World, London: Third World Books.
Wright, Q (1968) 'War', in D.Sills (ed.) International Encyclopedia of the Social
Sciences, vol. 16, New York: Free Press, Macmillan.
Zartman, W (1974) 'The political analysis of negotiations', World Politics 26: 385-99.
Zwi, A. and Uglade, A. (1989) 'Towards an epidemiology of political violence in the
Third World', Social Science and Medicine 28(7):633-42.
FURTHER READING
Beer, F. (1981) Peace Against War: The Ecology of International Violence, San Francisco:
WH.Freeman.
Bramson, L. and Goethals, G. (eds) (1964) War: Studies from Psychology, Sociology,
Anthropology, New York: Basic Books.
Cohen, R. (1990) Culture and Conflict in Egyptian-Israeli Relations: A Dialogue of the
Deaf, Bloomington: Indiana University Press.
Coser, L. (1956) The Functions of Social Conflict, New York: Free Press.
Eibl-Eibesfeldt, I. (1979) The Biology of Peace and War: Men, Animals, and Aggression,
New York: Viking Press.
Falk, R. and Kim, S. (eds) (1980) The War System: An Interdisciplinary Approach,
Boulder, Colo.: Westview Press.
1008
Ferguson, R.B. (ed) (1984) Warfore, Culture and Environment, New York: Academic
Press.
Foster, M.L. and Rubinstein, R. ( eds) (1986) Peace and War: Cross-Cultural
Perspectives, New Brunswick: Transaction Books.
Fried, M., Harris, M. and Murphy, R. (eds) (1967) War: The Anthropology of Armed
Conflict and Aggression, Garden City: Natural History Press.
Kertzer, D.I. (1988) Ritual, Politics and Power, New Haven: Yale University Press.
Koch, K.-F. (1974) The Anthropology of Warfare, Addison-Wesley Modules in
Anthropology no. 52, Reading, Mass.: Addison-Wesley.
Nettleship, M., Givens, R.D. and Nettleship, A. (eds) (1974) War: Its Causes and
Correlates, The Hague: Mouton.
Nordstrom, C. and Martin, J.-A. (eds) (1992) The Paths to Domination, Resistance and
Terror, Berkeley: University of California Press.
Otterbein, K. (1970) The Evolution of War, New Haven: HRAF Press.
Riches, D. (ed.) (1986) The Anthropology of Violence, Oxford: Blackwell.
Conflict: Culture in International Security, Boulder, Colo.: Westview Press.
Turner, P. and Pitt, D.(eds) (1989) The Anthropology of War and Peace: Perspectives on
the Nuclear Age, Granby, Mass.: Bergin & Harvey.
Turney-High, H. (1971) Primitive War: Its Practice and Concepts, Columbia: University
of South Carolina Press.
Viiyrynen, R. (ed.) (1987) The Quest for Peace, London: Sage.
Worsley, P. and Hadjor, K.B. (eds) (1987) On the Brink: Nuclear Proliferation and the
Third World, London: Third World Books.
1009
37
INEQUALITY AND EQUALITY

Andri Biteille
EVOLUTIONARY PERSPECTIVES AND THE

COMPARATIVE METHOD
A striking feature of the modern world is the deep and pervasive disjunction
between the ideal of equality and the reality of inequality. The ideal of equality
is widely endorsed and, as Isaiah Berlin has put it, 'The assumption is that
equality needs no reasons, only inequality does so' (1978:84 ). At the same time,
there is extensive and sometimes extreme inequality in the distribution of
matena an ot er resources, an m t e re atwns etween m IVI ua s, groups
and categories of every conceivable kind.
In studying inequality systematically we have to keep in mind the fact that
inequalities differ not only in degree but also in kind. Inequalities in the
distribution of income or of wealth are difficult to compare directly with
inequalities in the distribution of power, or with inequalities of status, prestige
or esteem. Moreover, the idea of equality is not a simple or a homogeneous one,
so that when people say they value equality, they may not all mean the same
thing. There are striking differences of orientation and perception between
those who emphasize competitive equality or equality of opportunity, and those
who stress distributive equality or equality of results. For these reasons it may
be misleading to argue about the nature and forms of inequality without
keeping in mind the various meanings of equality which, in our age, is both an
ideal and a value.
While social theorists are agreed that the societies in which they live are
marked by many forms of inequality, there is disagreement about whether
inequality is inevitable. Perhaps the majority believe that inequality is inherent
in the very nature of collective life, and some would go even further and argue
not only that inequality or stratification is inevitable but also that it has a
definite social function (Davis and Moore 1945; see Bendix and Lipset 1966).
Others maintain that inequality or stratification is not inevitable, and that an
egalitarian society is possible as a reality and not merely as an ideal. Most of the
1010
latter would probably concede that it is possible to attain only what Tawney
(1964 [1931]) described as 'practical equality' rather than absolute or perfect
equality.
Those who argue that in spite of the wide prevalence of inequality,
egalitarian societies are in fact possible, have sought to demonstrate either that
such societies have existed in the past or that they can be constructed in the
future, or both. Characteristically, the faith in the possibility of constructing
such a society in the future has been sustained by the belief that equality and
not inequality was the original condition of human life.
Among modern social and political philosophers, Rousseau was one of the
first to argue that equality or near-equality was the original or natural condition
of humanity, although Hobbes and Locke had put forward similar arguments
before him (Beteille 1980). Rousseau did not deny the existence of natural or
physical inequalities, but he believed these to be slight or insignificant. The
inequalities that really mattered were political or moral inequalities which,
being based on a kind of convention, could in principle be abolished or at least
diminished by a different convention. Rousseau's views were considered
radical in his time and they left a lasting impact on succeeding generations,
both in Europe and elsewhere. 1
The writings of Marx and Engels gave rise to the doctrine that the first stage
of social evolution was one of 'primitive communism' and that the final stage
would also be one of communism, both stages being marked, despite many
1 erences, y t e a sence o c asses. However, t ere was a 1 erence m
approach and method between Rousseau and the nineteenth-century
proponents of the theory of primitive communism. Rousseau constructed his
model from first principles, observing, 'Let us begin then by laying facts aside,
as they do not affect the question' (1938 [1762]:175). Marx, and more
particularly Engels (1948 [1884]), on the other hand, turned to the available
evidence from primitive societies to demonstrate that classless societies existed
in reality.
The second half of the nineteenth century saw the emergence of the new
science of ethnography, based largely on accounts of primitive societies by
explorers, missionaries, traders and administrators. A whole new world was
opened up for systematic enquiry. The early ethnographers were enthusiastic
advocates of the comparative method, by which contemporary primitive
societies were likened to those that were supposed to have existed at earlier
stages in the development of more advanced civilizations, and they used it to
construct ambitious evolutionary schemes. Perhaps the most famous among
these, and one which had a lasting influence in the Soviet Union, was
formulated in 1877 by Lewis Henry Morgan (1964). According to Morgan, the
first stage of evolution, designated as 'savagery' and represented by a number of
surviving primitive societies, was marked by an absence of inequality and class.
The theory of primitive communism aroused great interest in the late
nineteenth and early twentieth centuries. Inevitably, the discussion turned
1011
SOCIAL LIFE
around the presence or absence of individual property in the early stages of

evolution. The predominant view was that the concept of property-and
indeed of the individual-was absent in primitive societies; and it was tacitly
assumed that where there was no individual property, there could be no classes,
no strata, and no significant inequality. This view was challenged in a landmark
study published in 1921 by the American anthropologist Robert Lowie (1960).
Through a meticulous examination of the ethnographic record, he showed that
primitive societies were far more varied and far more differentiated than had
been allowed for in the theories of his predecessors. It is fair to say that most
anthropologists are nowadays sceptical about the existence of a universal stage
of primitive communism.
This scepticism does not of course mean that anthropologists altogether
reject the view that primitive societies, or at least some primitive societies, may
be genuinely egalitarian in their constitution. Recently the characterization
'egalitarian' has been applied to a number of societies in which 'equalities of
power, equalities of wealth and equalities of prestige or rank are not merely
sought but are, with certain limited exceptions, genuinely realized' (Woodburn
1982). The use of the term 'egalitarian' in the case of these societies is justified
on the grounds 'that the "equality" that is present is not neutral, the mere
absence of inequality or hierarchy, but is asserted' (1982:431-2). Examples of
such societies include the Mbuti Pygmies of Zaire, the !Kung Bushmen of
Botswana and Namibia, the Pandaram and Paliyan of South India, the Batek
Negntos o Ma aysta an t e Ha za o Tanzama.
The egalitarian societies referred to above are all based on a foraging or
hunting-and-gathering economy. Indeed according to Woodburn, not all
hunter-gatherer societies are egalitarian, but only those characterized by
'immediate-return' as against 'delayed-return' systems of production
(1982:431 ). An immediate-return system is one in which there is no time-lag, or
only a small one, between the investment of labour in production and the
realization of the product, so that no complex chain of rights and obligations is
entailed in production, whereas such a chain is a necessary part of delayed-
return systems. Hunter-gatherers with immediate-return systems live and
move about in very small groups which have no fixed membership and only a
very rudimentary division of labour, a condition that comes close to the outer
limit of organized social life. It is very difficult to draw any significant
conclusion from their study for the future of equality in more organized
societies.
Although evolutionary theories are no longer as popular as they were in the
past, those engaged in the comparative study of equality and inequality often
adopt an evolutionary perspective, either implicitly or explicitly. A
characteristic expression of the evolutionary perspective on the subject is to be
found in a recent essay by Gellner. Commenting on the work of a well-known
American author, Gerhard Lenski (1966), he observed that:
1012
The pattern of human history, when plotted against the axis of equality, displays a
steady progression towards increasing inequality, up to a certain mysterious point in
time, at which the trend goes into reverse, and we then witness that equalisation of
conditions which preoccupied Tocqueville.
(Gellner 1979:27)
This view of the course of human history is very widely held, and it merits a
brief discussion.
Implicit in the evolutionary scheme outlined above is a classification of
societies into three broad types: (1) primitive societies, (2) agrarian
civilizations, and (3) industrial states. Primitive societies, including bands,
segmentary tribes as well as tribal chiefdoms, are small in scale and relatively
undifferentiated; though few of them are egalitarian in every sense, they are
generally not divided into distinct classes or strata. Agrarian civilizations of the
kind that prevailed in Europe, India or China are or were hierarchical both by
design and in fact; their characteristic divisions were into castes or estates
whose boundaries were relatively clear and acknowledged by custom and law.
Industrial states, whether of the capitalist or the socialist type, have a formal
commitment to equality rather than hierarchy; their characteristic divisions are
classes and strata 2 which must accommodate themselves to the ideals of
democratic citizenship and equality of opportunity. It is not that inequalities
are unknown or even uncommon in industrial societies, but rather that they
The ' distinction between 'aristocratic'

' and 'democratic' societies, and the
historical passage from the former to the latter, were described in memorable
prose by Alexis de Tocqueville in 1835. He wrote:
In running over the pages of our history, we shall scarcely find a single great event of
the last seven hundred years that has not promoted equality of condition.
And again,
The gradual development of the principle of equality is, therefore, a providential

fact. It has all the chief characteristics of such a fact: it is universal, it is lasting, it
constantly eludes all human interference, and all events as well as all men contribute
to its progress.
(1956, 1:5-6)
De Tocqueville set out to demonstrate the progress of equality in every sphere

oflife: in the material conditions of human beings, in the pattern of their social
relations, and in their ideas, beliefs and values.
It must be remembered that de Tocqueville's argument about 'aristocratic'
and 'democratic' societies was an historical one, intended to bring out the
continuity as well as the contrast between the two. The contrast has been
extensively applied, both to different historical phases in the life of the same
1013
SOCIAL LIFE
society and to different societies independently of historical connections. The

second kind of contrast does not have any necessary link with the evolutionary
perspective, and might in fact be accompanied by an explicit rejection of such a
perspective.
Western scholars have long been fascinated by the Indian caste system,
which has often been represented as the prototype of rigid hierarchy. Some of
them have seen in it an extreme form of tendencies present in their own society,
while others have viewed it as a qualitatively different, if not an altogether
unique, system. The French anthropologist Louis Dumont ( 1966, 1977) has
developed a body of work in which the contrast between traditional Indian
society and the modern West is presented in the sharpest possible terms,
epitomized in the respective notions of Homo hierarchicus and Homo aequalis.
The contrast, as he draws it, is confined largely to the plane of values, to what
people believe or say they believe rather than to what they do or practise. When
Dumont talks about Homo aequalis, what he means is that modern societies
have an egalitarian ideology-that they are egalitarian in intention-not that
they have attained or are likely to attain equality in the distribution of material
resources. I discuss some of these issues more fully below (p. 1028f£), and note
here only that it may be misleading to characterise whole societies as either
'egalitarian' or 'hierarchical'.
How tangled the question is can easily be seen by returning briefly to de
Tocqueville. When he spoke of the providential advance of equality, de
Tocquevt e c ear y e teve t at equa tty was, m IS own 1 ettme, a vancmg
simultaneously on all fronts. But that, plainly, was an illusion. We have no
reason to believe that equality of condition, or equality in the distribution of
material resources, always advances simultaneously with equality as a moral or
philosophical value.
An important aspect of inequality in all modern societies is inequality in the
distribution of income. Now it is a well-established truth that there was an
increase rather than a decrease in inequality of income in the early stages of
economic growth in most, if not all Western countries (Kuznets 1955). In other
words, inequality in one significant sense was increasing during precisely that
period when the modern egalitarian ideology was spreading rapidly in the
West. Not all societies have had, or can be expected to have, the same historical
experience in every respect. But it is obvious that 'legal equality' and 'economic
equality' do not have the same rhythms of change and might, arguably, change
in opposite directions.
An additional difficulty arises from the fact that different concrete forms of
inequality may coexist in the same society: for instance, an open class system
and a rigid system of racial stratification. This was noted by de Tocqueville for
the United States (1956 [1835, 1840], I: eh. 18). Lack of internal consistency
makes comparison difficult, and the difficulty is compounded when the units
being compared differ vastly in scale-for instance, a small foraging band and a
large nation state. A society on a large scale with a complex pattern of
1014
stratification may contain within it component units which have an appearance

of remarkable homogeneity and equality, as indeed was the case in traditional
India; and a small-scale egalitarian community, enjoying a degree of isolation
and autonomy, may depend for its survival on its articulation with a large and
complex system of stratification.
The inadequacy of treating the nation state as an irreducible unit in the study
of equality and inequality has become increasingly apparent. Societies are at all
levels in continuous interaction with each other, and modern anthropologists no
longer regard them as isolated or self-sufficient units with fixed and rigid
boundaries. Recent studies have shown how the rise of European societies from
the seventeenth century onwards was often at the cost of smaller or less developed
or less powerful societies in Asia, Africa and Latin America which they oppressed
and exploited (Wolf 1982). One must not be too quick to characterize the former
as egalitarian societies by looking only at the ideals they set for themselves while
looking away from their actual treatment of others.
APPROACHES TO THE STUDY OF INEQUALITY

It is clear that when we compare different societies, we are dealing with
inequalities that differ not only in degree but also in kind. There is no
universally accepted criterion which enables us to conclude that a given society
corresponds more closely than another to some general standard of equality;
an common sense IS not a ways a very re ta e gm e. Economists o ten smg e
out a specific aspect of the problem, namely inequality in the distribution of
income, on the grounds that it lends itself most easily to quantitative treatment.
But even here they find it difficult to judge unequivocally whether a given
distribution shows more or less inequality than another (Sen 1973). And
inequality of income has to be viewed alongside other aspects of inequality
which differ significantly among themselves.
The conclusion we reach from a comparison of different patterns of
inequality will depend in part on our method and approach. Of the several
approaches to the study of inequality, two are of particular importance. The
point of departure for the first approach lies in the inequalities inherent in the
distribution of abilities among the individual members of a society; for the
second, it lies in the inequalities inherent in their arrangement into an
organized whole. The first approach stresses that individuals are unequal to
begin with, and that their unequal abilities will be bound to show up no matter
how or where they are initially placed; the second maintains that since
individuals are unequally placed from the start, they develop and display
unequal abilities.
Individual variations are a matter of common observation, and they are to be
found in every society. No two individuals are exactly alike, and identical twins
are the exception that proves the rule. However, we must be careful to
distinguish between difference and inequality-an obvious distinction that is
1015
SOCIAL LIFE
easily overlooked by proponents of the theory of natural inequality (Beteille

1980). Two individuals may be quite different from each other without being in
any meaningful sense unequal.
Do individuals differ to the same extent in all societies? It is difficult to give
an unequivocal answer to this question. One might like to distinguish between
variations in purely physical or biological traits and those in mental or 'moral'
characteristics, or between 'natural differences of kind' and 'social
differentiation of positions' (Dahrendorf 1968); but the distinction is by no
means easy to sustain. Comparison of the degrees of individual variation
becomes difficult where societies differ very greatly in scale. Moreover,
variations among individuals of one's own kind always appear greater than
among individuals of a different kind. Explorers, missionaries and colonial
administrators systematically underestimated individual variations, even in
physical characteristics, among the natives whom they observed and described.
Some anthropologists take the view that the stress on individual variation, if
not the very fact of it, is unique to modern societies, being undeveloped or
weakly developed in primitive or traditional societies. Emile Durkheim, whose
work has left a lasting impression on the French school of sociology, put
forward this view in his very first book, The Division of Labour in Society (1982
[1893]), in which he argued that primitive societies (conceived in a very broad
way) were held together by mechanical solidarity which was based on 'likeness'
as against 'complementary difference'. He believed that people in these
societies ac e m IVI ua Ity to sue an extent t at even t e I erences
between men and women, including their physical differences, were weakly
expressed or rudimentary in them. This is an extreme position to which few
would assent today.
While individual differences are present in all societies, they may be
culturally restrained in some cases and encouraged in others. They tend to be
encouraged to such an extent in modern societies that individualism has come
to be regarded as the dominant ideology of these societies (Dumont 1977,
1983). De Tocqueville believed that there was a close connection between
individualism and equality (1956, 2:98-100). But individualism has more than
one implication, just as equality has more than one meaning (Beteille 1986). To
the extent that individualism stresses the autonomy and the dignity of the
individual, it places itself against all forms of ascribed inequality. But to the
extent that it stresses competition and achievement, it justifies and promotes
inequality in other forms.
The preoccupation with individual achievement (and with individual
quality on which it is presumably based) has given a distinctive character to
contemporary debates on equality and inequality. The logic of capitalism is that
opportunities are in principle equally available to all individuals who,
nevertheless, do not all benefit from them to the same extent because they differ
in their endowments and fortunes. For many, this difference in individual
endowments and individual fortunes lies at the heart of the problem of
1016
inequality in modern societies (Hayek 1960: eh. 6, E ysenck 1973 ). Such a view
reveals a bias in favour of methodological individualism, seen most commonly
in writings on inequality by economists and psychologists.
Methodological individualism, or the procedure which treats the individual
as the basic and irreducible unit in social analysis, faces many difficulties in the
study of variation and change in patterns of inequality. It can perhaps account
for the ranks assigned to individuals on a given scale, but it cannot as easily
account for the scale itself. An issue that all students of social inequality must
face is what may be called the passage from difference to inequality. It is a
truism that not all differences count as inequalities. Why, then, do only some
differences count as inequalities, and not others? Do the same differences count
as inequalities in all places, at all times? What is actually involved when a set of
differences is transformed into a system of inequalities? These questions
cannot be addressed without considering some of the constitutive features of
human society and culture.
The majority of sociologists and anthropologists take as their point of
departure not the individual agent, but the framework of collective life within
which he acts (Bendix and Lipset 1966, Helier 1969, Beteille 1969). Every
individual acts within a framework of society and culture which both provides
him with facilities and, at the same time, imposes constraints. The language he
speaks, the technology he uses, the division of labour within which he works,
all exist to some extent independently of his exertions. The regularities
governmg anguage, tee no ogy an tvtswn o a our are o a 1 erent m
from those governing individual action.
Language provides us with a convenient example of the place of collective
representations in human life. Without language, human life as we know it
would be impossible, and human language, in its turn, would not exist in the
absence of collective life. But collective representations include much more
than language. They consist of the full range of beliefs and values shared by
individuals as members of society. At this point it will be enough to say that
collective representations include both cognitive and evaluative elements-
which are, moreover, closely intertwined-so that the individual members of a
society share not only common modes of thought but also common standards
of evaluation. Indeed, it is difficult to see how collective life would be possible
in the complete absence of shared beliefs and shared values.
Durkheim stressed the contrast between the fullness and variety of the
collective representations of a society and what it is possible for any individual
mind to create or comprehend on its own. Subsequent investigations by
anthropologists in the field have fully confirmed the truth of Durkheim's
insight. People with a simple Neolithic technology, such as the Bororo or the
Nambikwara Indians of the Amazon basin, show a richness and complexity in
their collective representations that seem to surpass what even Durkheim
might have expected. The luxuriance of expressive life commonly encountered
in the primitive world at the level of cosmology and taxonomy can scarcely be
1017
SOCIAL LIFE
explained by the practical requirements of material existence (Levi-Strauss

1966).
A seminal paper published by Durkheim and Mauss in 1903 opened up a
new field of cultural anthropology devoted to the study of systems of
classification (Durkheim and Mauss 1963). We now know that such systems,
which are sometimes extraordinarily elaborate, are present in all societies, even
those of the smallest scale. They not only arrange the vast multitude of
culturally recognized items into broad classes but also order them according to
principles that may be implicit or explicit. This means that as well as there
being socially preferred items of food, dress, adornment, and so on, there are
also recognized preferences in regard to colours and other attributes of nature.
These preferences reveal the aesthetic and moral categories of a society. Once
again, it is difficult to see how a human society could exist in their absence.
Now, it would be strange in a culture to have standards of evaluation that
apply to food, dress, adornment, plants and animals, but none that apply to
human beings and their activities. In other words, where people are able to
discriminate between good and bad food, they will also discriminate between
good and bad cooks; where they judge some gardens to be superior to others,
they will also judge some gardeners to be superior to others; where there are
preferences as between artefacts, there are likely to be preferences also as
between artisans. I am of course talking now of culturally prescribed, or at least
culturally recognized, preferences, and not the personal preferences of
parttcu ar m IVI ua s.
Every culture, no matter how rudimentary, has its own bias, not only for
certain types of human performance but also for certain types of human
quality. Quality and performance are closely related in the minds of people, but
they may be given different priorities in different cultures. Men and women
may be believed to have different qualities, and where these qualities are
themselves ranked, as they often are, men and women will also be ranked (for
further discussion of this point, see Article 29). Even where qualities are
assigned priority, there is always room to take performances into account. For
instance, women may be considered to excel in gardening and men in hunting,
but then hunting may rate higher than gardening, in which case men will be
ranked higher than women. The stress on quality tends to be associated with
the segregation of distinct sections of society into separate fields of activity, so
that their members do not compete with each other on a common ground.
Where the stress is on performance, men and women-or, to vary the example,
whites and blacks-may be allowed to compete for the same prizes and then
ranked according to their performance, irrespective of gender (or race). But
here again, success or failure will be attributed, at least in part, to the presence
or absence of some quality such as intelligence. Moreover, the fact that whites
and blacks (or men and women) may in certain spheres compete on equal
terms, and be judged on merit, does not mean that they will not in other
spheres be treated differently or even unequally.
1018
Thus, it is clear that what transforms differences into inequalities are scales
of evaluation. A scale of evaluation is not a gift of nature; to speak in the
language of Rousseau, 'it depends on a kind of convention, and is established,
or at least authorized by the consent of men' (1938:174). Even while invoking
the name ofRousseau, however, it is important to guard against the dangers of a
constructivist argument. The conventions by which human beings rank each
other-their qualities and their performances-are rarely the outcome of
conscious design. Most people use these scales as they use language, without a
clear awareness of their structure.
Once we realize that scales of evaluation are not usually the products of
conscious design and are not always clearly recognized for what they are, we
have to turn to consider the coexistence of a multiplicity of scales and the
problem of their mutual consistency. It is a common experience that where A
ranks higher than B in scholastic ability, B may rank higher than A in athletic
ability, leaving open the question of the overall rank of A in relation to B. Some
occupations are more remunerative, others permit greater freedom of
individual action; how are they to be ranked in relation to each other? How
complicated the general problem is may be seen from a glance at the
voluminous literature that has grown around so specific a topic as the social
grading of occupations (see Goldthorpe and Hope 1974).
To assign a central place to evaluation in the explanation of inequality is not
to deny that different values coexist in the same society. One can go further and
argue t at 1 erent va ues ten to pre ommate m 1 erent sectors o t e same
society. Manual workers and professionals may not rank occupations in the
same way; blacks and whites may not assign the same significance to colour in
social ranking; and men and women may show different kinds of bias in the
personal qualities they value. While this is true, it should not lead to the
conclusion that there can be as many scales of evaluation as there are individual
members of society, for no society can endure without some coherence in the
domain of values.
Advocates of the so-called 'structural-functional' approach in social theory
tend to stress the integration of values in the societies about which they write
(Parsons 1954). One form of the functionalist argument is that, although there
may be different scales of evaluation in the same society, these scales themselves
can be arranged in a hierarchy, since every society has a 'paramount value'
which determines the alignment of all its other values (Dumont 1980, 1987).
This is a tendentious argument which should not be allowed to divert attention
from the empirical investigation of the actual extent to which different values
reinforce or subvert each other in concrete historical situations.
Where there are competing or conflicting values in a society, each associated
with a particular section of it, they do not always rest in a state of stable
equilibrium. Of course, the discordance may be reduced through reflection,
argument and self-correction, and accommodation may be achieved on the
plane of beliefs and values itself. But this is not the only or even the most
1019
SOCIAL LIFE
typical way in which the problem of value conflict is resolved. Differences that
cannot be resolved on the plane of values are typically resolved on the plane of
power. Or, to put it plainly, 'Between equal rights force decides' (Marx 1954
[1867]: 225).
The resolution of conflict (including the disagreement over values) through
the exercise of power brings to our attention a second important source of
inequality in collective life. The importance of force (as against common values)
in maintaining order and stability in society has been noted by many, and there
are some who would say that it is not only important but decisive (Dahrendorf
1968). This is particularly true of those who deal with the place of the state in
human affairs. As Hobbes wrote in his Leviathan of 1651, 'And Covenants,
without the Sword, are but Words, and of no strength to secure a man at all'
(1973:87).
The state provides the most striking example of inequalities in the distribution
of power, but by no means the only one. Such inequalities are commonly found in
many domains, including the domestic domain, that are a part of society but not,
strictly speaking, of the state. No doubt it can be argued that where the state exists
it provides sustenance to inequalities of power in every domain and that with the
collapse of the state, those inequalities should also collapse. This has been a
familiar argument among Marxists who have found support for it in a work
published by Engels a century ago (Engels 1948 [1884]). At that time it was hoped
that the argument would be confirmed by the imminent collapse of the bourgeois
state. T e ourgems state, owever, as co apse many times over, ut t e en o
the inequality of power is nowhere in sight.
There is, besides, plenty of evidence for inequality of power in what are
commonly described as 'stateless societies' (Tapper 1983; see also this volume,
Article 34). There are, firstly, the chiefdoms, varying greatly in size and degree
of organization, with tribal or clan chiefs who might exercise considerable,
though intermittent authority in organizing people for collective activities.
Much depends on the scale and importance of the collective activities that have
to be organized. Pastoral tribes have leaders whose voice carries considerable
authority in matters concerning the movement of people and animals, and in
conducting and coping with raids.
There are then the segmentary systems proper-segmentary tribes as
against tribal chiefdoms, to follow the terminology of Sahlins (1968)-which
do not have chiefs in the accepted sense of the term. Here the system works not
so much through a hierarchical distribution of power as through the balance of
power between groups at different levels of segmentation (see Evans-Pritchard
1940 for a classic account). Two kinds of groups are especially significant in
such societies: descent groups and local groups. Where descent groups are
corporations-whether among the patrilineal Tallensi (Fortes 1945, 1949) or
the matrilineal Truk (Goodenough 1951 )-the senior male members have a
decisive say in the disposal of the productive and reproductive resources of the
corporation, mainly land, livestock and women. This is particularly true at the
1020
lower levels of segmentation where the descent group is functionally most

effective as a corporation.
It is on the level of the local group rather than the descent group, however,
that the crucial evidence for the kind of argument that I am trying to make will
have to be found. The evidence seems to me to be clear, though perhaps not
decisive. Evans-Pritchard, whose book on the Nuer of southern Sudan (1940)
was a turning point in the study of tribal political systems, deliberately
excluded the internal organization of the village from his consideration of N uer
political structure. We can nevertheless say something about the exercise of
power in maintaining the stability of such groups, even while conceding that
this stability is itself a matter of degree.
The problem is of the following kind. Every stable group has a division of
labour, no matter how rudimentary, which is regulated by rules regarding the
rights and obligations of its individual members. It is in the nature of human
life that these rules do not operate mechanically, with clock-like regularity and
precision. They are occasionally, if not frequently, violated, if only because
individuals have different perceptions of the rules themselves, as well as
divergent interests. These divergences, which are found in even the simplest
local groups, may appear trivial in scale by comparison with those that occur in
industrial societies, but they are nevertheless important in their own context.
Disputes have to be settled, decisions that are binding on all have to be made,
and this provides the basis for the exercise of power by some individuals over
ot ers. To e sure, matters may e sett e rom one sttuatwn to anot er y a
the members of the group acting together so that no individual accumulates
more power or authority than any other. But that would be the limiting case and
not the typical one.
We may recall at this point the egalitarian society based on an 'immediate-
return' economy of hunting and gathering. It will be a little more clear now why
I regard it as a limiting case. It stands at one extreme, the other extreme being
represented by the monolithic and authoritarian industrial state with its
massive apparatus of coercion and manipulation which reached perfection, or
near perfection, in the Soviet Union under Stalin, and, more briefly, in
Germany under Hitler. We can learn a great deal about equality and inequality
from both social types, although it is my judgement, which I cannot
substantiate here, that they are both highly unstable.
SOME COMMON HISTORICAL FORMS OF INEQUALITY

In an important essay on the origin of inequality, Ralf Dahrendorf (1968)
distinguished, on the one hand, between natural differences of kind and
natural differences of rank, and on the other, between the latter and social
stratification. We shall set aside for the moment the significance of 'natural
differences of rank', or what is more commonly called natural inequality
(Beteille 1980). The relation between natural difference and social inequality is
1021
SOCIAL LIFE
a very important one, although it is by no means as simple as might at first sight

appear. Natural differences do not present themselves to us directly, but are
perceived in a highly selective manner, through the lenses of socially
established systems of classification (Levi-Strauss 1966). What needs to be
stressed is that not merely the evaluation of differences, but to some extent their
very recognition, is a social process.
Differences that are assigned cardinal significance in one society may be
ignored or overlooked in another. The differences between men and women are,
however, taken into account in all human societies, and it is difficult to see how
it could be otherwise. This does not of course mean that they are taken into
account in the same way or to the same extent in every society. Where men and
women are given distinct social roles, they develop differences, and sometimes
marked differences, in temperament and ability; these differences in
temperament and ability are then taken-by women as well as by men-to be
the reason for their being given different roles. It is clear that much of this rests
on convention which varies from one society to another (see the classic but
controversial account in Mead 1963 [1935]). What is not clear is whether,
outside of procreation and parturition, there have been or can be conventions
for the social division of labour that ignore altogether the differences of gender.
Leaving aside the question of what is possible, we have to consider how far the
differences of gender are in fact treated as inequalities. This is a vexed question
where the facts are confusing and are open to conflicting interpretations. There is
a vast Iterature on t e positiOn o women m pnmitive societies w IC It IS
impossible to summarize here. In a lecture delivered on the subject in 1955 and
first published in 1965, Evans-Pritchard, then Britain's foremost anthropologist,
observed that the acrimonious debates on the subject belonged to the past and
that it could at last be discussed with scholarly detachment (Evans-Pritchard
1965: eh. 2). That has turned out to be a monumental error of judgement, for no
field of anthropology is more deeply embattled today than the one that deals with
gender and inequality (Ardener 1975, MacCormack and Strathern 1980; see also
Moore in this volume, Article 29).
The historical record of the development of the subject is roughly as follows.
Early anthropologists commonly subscribed to the theory of the primitive
matriarchate or the view that the first stage of social evolution was marked
uniformly by the prevalence of matriarchy or mother-right. This view
gradually became obsolete, particularly after Lowie's critique of it in Primitive
Society (1960 [1921 ]). At about the same time, Rivers (1924: eh. 5) also pointed
out that power lay generally in the hands of men, irrespective of the form of
descent, and that there was no uniform relationship between the position of
women on the one hand and forms of descent, inheritance and succession on
the other. The considerable body of empirical material that was available when
Evans-Pritchard wrote his lecture seems to have borne out Rivers's basic point
that women were in general subordinated to men in public life, and that parity
between men and women was unusual if not unknown.
1022
The whole subject has now been thrown open once again, mainly through
the recent spate of feminist studies (see, for example, Leacock 1978). New
dimensions have been brought to light which were not perceived by even the
most acute minds among the earlier anthropologists. These studies have
implications, only now beginning to be explored, for understanding not just the
disparity between the sexes but inequality in general. I here merely touch upon
two such issues, one relating to power and the other to values.
Those who have stressed the subordination of women to men have tended to
dwell mainly upon the politico-jural domain rather than the domestic domain.
Clearly, in even the most strongly 'patriarchal' societies, women sometimes play
an important, not to say a crucial, role in domestic affairs. They may play the
major part in everything concerned with food, health and nurture, and exercise
independent initiative in all these regards. As against the 'jural' inferiority of
the wife to the husband or the sister to the brother, there might be a
'psychological' dominance of the son by the mother. A contemporary Indian
psychologist has indeed argued with regard to his own society, which is to all
appearances strongly patrilineal, that 'the Indian lives in his inner world less
with a feared father than with a powerful, aggressive and unreliable mother'
(Nandy 1980: 107; see also Kakar 1978). All this, however, would require a
reconsideration of the concepts of power and dominance as conventionally
used in the social sciences to an extent that would take us far beyond the scope
of the present article.
Just as It may e unreasona e to assume t e existence o a smg e
homogeneous domain in which some individuals invariably exercise power over
others, it may also be unrealistic to assume the existence of a homogeneous
conceptual or moral universe whose categories of classification and evaluation
are accepted in the same way by all. The important contribution of women's
studies has been to draw attention to the existence of alternative beliefs and
values whose implications for the social ranking of persons have yet to be fully
explored.
Distinctions of race, though also marked by physical or biological traits,
differ significantly from those of gender. They are less clear and less fixed, and
are not universally present. Only some societies have or recognize them while
others do not. Within a given society racial differences exist and are
perpetuated because they have cultural significance. If people simply ignored
those distinctions in their social interactions and married without any regard
for them, the distinctions themselves would cease to exist or become
substantially different (Beteille 1977, eh. 5). The same can hardly be said about
gender.
There is a very wide range of variation of physical features in the human
species, much wider than in most other animal species. However, variation by
itself does not give us distinct races; the variation has to be clustered in a
particular way for races to become visibly apparent. That can happen in either
of two ways: when populations are territorially dispersed to an extent which
1023
SOCIAL LIFE
practically rules out interbreeding; or when, though sharing the same territory,
they are prevented or discouraged from interbreeding by law, custom and
convention. The continued presence of distinct races in a society and their
social segregation are, in a sense, two sides of the same coin.
Racial discrimination in its characteristic modern form is a feature of
societies that owe their origin to historical circumstances of a particular kind.
These are circumstances of sudden and violent encounter between populations
differing sharply in physical appearance, language and material culture,
associated with the European conquest of Africa and the New World (and to a
much lesser extent of Asia). This is not to say that the violent penetration by
people of one physical type into the territories of another never took place in
the past. But the European penetration of Africa and the New World in the
seventeenth and eighteenth centuries was unique in its global character, in its
swiftness and violence, and in the scale on which it led to the dislocation of
populations (Wolf 1982).
We find today two distinct patterns of racial inequality, both involving
whites and blacks, one in the United States and the other in South Africa
(Beteille 1977). In the United States racial inequality survives under a liberal
democratic regime which has shown some commitment to affirmative action; in
South Africa it holds its own under a minority racist regime committed to a
policy of apartheid (i.e. 'apartness'). 3 Apart from differences in constitutional
history and background, there is an important demographic difference
etween t e two countnes. In t e Umte tates t e w ttes are not on y
politically dominant, they are also in a majority, having overwhelmed other
races on account of their superior firepower, the devastating impacts of
introduced diseases on indigenous populations (see Article 11 ), and sheer
strength of numbers. In South Africa the whites are politically dominant but
numerically in a minority, being surrounded, moreover, by states which are
totally hostile to white-minority rule. What is notable in the United States is
the ambivalence of the blacks, whereas what is striking in South Africa is the
anxiety of the whites.
Even where two distinct races are initially brought together by the use of
force, and are then kept at least partially segregated also by the use or the threat
of force, their coexistence over successive generations can lead them to share
certain common values. To be sure, these 'common' values are largely the
values of the dominant race, but the point is that they tend to be internalized, at
least to some degree, also by the subordinate race. A striking example of this
may be found in the extent to which upwardly mobile blacks in the United
States have internalized white values and standards in regard to personal
beauty, elegant dress and refined speech (Frazier 1957). Where, on the other
hand, the subordinate race fails or refuses to internalize the 'common' values of
the dominant race, we have an unstable and a potentially explosive situation, as
exists in South Africa.
We have seen that the inequality of races is, in the typical case, established by
1024
the exercise of power and maintained by the hold of a common culture which
assigns higher values to the traits characteristic of one race as against those
characteristic of another. There is nothing 'natural' about either of these
processes. Indeed, if the present population of either the United States or
South Africa were allowed to revert to its 'natural' state, all distinctions of race,
or at least those distinctions now considered significant, would disappear with
the passage of time. This is quite apart from the fact that no matter what we
might think of 'domination', evaluation cannot in any meaningful sense be
regarded as a natural phenomenon.
Caste and race are sometimes considered together as they are both regarded
as extreme forms of rigid social stratification maintained by strict rules of
endogamy. Both Lloyd Warner, who pioneered the empirical study of social
stratification in the United States (Warner 1941 ), and Gunnar Myrdal, who
conducted a monumental study of the blacks in the same country (Myrdal
1944, 1: eh. 31 ), found it convenient to use the concept-and not merely the
metaphor-of caste in analysing stratification by race. They both pointed out
that neither the blacks nor the whites were a race in the scientific sense, that the
whole system rested on social conventions, and that, therefore, to represent it in
a biological idiom was misleading. They also felt that the barriers separating
blacks and whites were qualitatively different from those between classes within
each of these populations. Thus, the choice of the term 'caste' was to some
extent dictated by negative considerations, since neither 'race' nor 'class'
seeme appropnate.
But other anthropologists, too, have pointed to certain fundamental
similarities between the Indian caste system and the colour-caste system of the
United States (Berreman 1960, 1966). One of these similarities relates to
attitudes towards women. Both white males in the United States and upper-
caste males in India have shown an obsessive concern with the 'purity' of their
own women while engaging freely in the sexual exploitation of black or
untouchable women. All of this can be related to ideas about bodily substance
and the conditions appropriate for its exchange. The general importance of
these ideas in American culture has been stressed by Schneider (1968), and in
the Hindu caste system by Marriott and Inden (1974). In other words,
inequalities of caste are illuminated in the same way as those of race by a
consideration of gender (Beteille 1990).
There are of course differences between caste and race, and the tendency
among contemporary anthropologists is to stress the differences more than the
similarities (Dumont 1961, de Reuck and Knight 1968). At any rate, the Hindu
caste system is a sufficiently important historical example of inequality to
deserve attention in its own right. Recent writers on caste, notably Dumont
(1966), have seen in it the most complete example of a hierarchical society, one
which in its traditional form was hierarchical not only in fact but also by
design, and in which the hierarchical principle animated every sphere of life.
Viewed in this light, the Hindu caste system had its analogue in the European
1025
SOCIAL LIFE
system of estates which was also governed by the 'hierarchic conception of

society' (Huizinga 1924: eh. 3).
The caste system may be viewed at two levels, those of varna and jati, for
both of which the same English word 'caste' has been commonly used (Srinivas
1962). Varna represents the formal order of caste, the 'thought-out' rather than
the 'lived-in' system, and the traditional discourse on caste has been typically
in the idiom of varna. All humankind and, indeed, all created beings were in
principle divided into four varnas which were both exclusive and exhaustive.
The Manusmriti declares that Brahman, Kshatriya, Vaishya and Shudra are the
four varnas and that there is no fifth. The same four varnas, in the same order
of precedence, were acknowledged by Hindus throughout India for more than
two millennia until disowned by the new constitutional and legal order.
The varna order is expounded in detail in the classical socio-legalliterature
known as the Dharmashastra, particularly in the Manusmriti or the
Manavadharmashastra, which dates back roughly two thousand years (Kane
1974 ). Anyone who reads this literature will be struck by the elaborate and
comprehensive manner in which human beings-their qualities and actions-
and all things around them are classified and ordered. To take a well-known
example, it is decreed that the sons of a male Brahman shall inherit property in
the following proportions: the son of the Brahman mother, four parts; the son
of the Kshatriya mother, three parts; the son of the Vaishya mother, two parts;
and the son of the Shudra mother, one part only. To be sure, the classification
an t e or enng are tg y se emattc, an present us wtt mo e s rat er t an
descriptions.
The invariance and fixity characteristic of the varna model are reduced to
some extent when we move down to the plane of jatis. Jati is a regional rather
than a national system, and the number of jatis, as well as their names, vary
from one part of the country to another. Moreover, there is reason to believe
that oldjatis have disappeared and that new ones have come into being with the
passage of time in each and every region. Although Hindu theory states that the
whole of humankind is embraced by the varna order, jatis have in fact freely
existed outside of that order, among Muslims, among Christians and, to some
extent, also among so-called 'tribals' (Bose 1975 [1949]). The problem of the
correspondence between varna andjati is a difficult one (Srinivas 1962, Lingat
1973), although the assumption of such a correspondence was a part of Hindu
beliefs about caste.
Whereas the varnas are only four in number, the jatis in each region are very
many; exactly how many is difficult to say, because they are frequently
segmented in a manner that has baffled census takers over the distinction
between caste and subcaste (Beteille 1964). Suffice it to say that there may be in
a single village as many as 30 to 35 subcastes (Beteille 1965). The jatis in a
region are not merely differentiated from each other; they are also mutually
ranked. This ranking manifests itself in a variety of social contexts through
transactions of different kinds (Marriott 1959, 1968). Traditionally, a very large
1026
social distance had to be maintained between the Brahmans at one extreme and
the Harijans or Untouchables at the other.
The ranking of jatis differs, and has always differed, from the ranking of
varnas in a number of important ways. There is no clear linear order of jatis as
there is of varnas. It is no doubt true that the Brahmans are at the top and the
Harijans at the bottom, but each of these two categories is made up of a number
of distinct jatis, which themselves cannot be readily placed in a linear order.
This ambiguity has always left some room for mobility among castes and
subcastes (Srinivas 1968). An upwardly mobile jati not uncommonly phrased
its claim to superior status in the idiom of varna.
While there is general agreement that the ranking of jatis is very elaborate
and, compared with other systems of social ranking, also very rigid, there is
considerable disagreement about the sources of caste rank. The actual ranks
enjoyed by the different castes arise from a variety of factors, although the
idiom in which caste ranking is phrased is typically a ritual one, more
specifically the idiom of purity and pollution. This had led some observers to
exaggerate the importance of ritual factors, giving the system an appearance of
mechanical rigidity without any room for freedom of action.
Despite the impressive stability and continuity of the caste structure, Hindu
ideas behind the ranking of persons are fluid and complex, and perhaps
heterogeneous. Varna, which may loosely be rendered as 'order' or 'kind',
provides an overall framework, but it does not stand by itself Besides the four
varnas etat e m t e D armas astras, t ere are t e t ree gunas or qua 1t1es
discussed elsewhere, particularly in the Samkhya texts (Rege 1984, 1988,
Larson and Bhattacharya 1987). The three gunas are: sattva (signifying light,
purity, intellect), rajas (energy, valour), and tamas (darkness, inertness). The
gunas enter as constituents into the make-up of different persons. In addition to
guna, there is also karma, which refers to action or works: what a person does
rather than what he or she is.
Guna and karma are commonly discussed in relation to persons rather than
groups, although they may also be linked more or less explicitly to the four
varnas. In the Bhagavadgita, Lord Krishna declares, 'caturvarnyam mayam
sristam, guna-karma-vibhagasah' ('the four varnas did I create, dividing (or
distributing) the gunas and the karmas')(Zaehner 1969:4/13). Some modern
interpreters of the Gita, including the great nationalist leaders B. G. Tilak and
M.K.Gandhi, have tried to argue that it represents an activist philosophy;
however, it cannot be too strongly emphasized that throughout the long course
of Indian history individual action has been severely constrained by the social
framework of caste.
Some contemporary anthropologists (e.g. Dumont 1964) have overstressed
the hierarchical completeness of Hindu society in order to bring out the
distinctive features of their own. Modern societies do indeed have a number of
distinctive features, both in their organizational structures and in their value
patterns. These features stand out when we contrast the modern West not only
1027
SOCIAL LIFE
with traditional India but also with its own medieval past (Beteille 1986,
Dumont 1987).
In the context of our present theme, perhaps the most striking feature of
modern societies is the notion of equality before the law. As an explicit principle
governing the relations between persons, it has found its fullest expression only
in modern times. It developed first in the West, in England, France and the
United States, and came to be widely adopted in the present century so that
there are very few parts of the world today where it is not acknowledged. The
far-reaching implications of this should not be overlooked, for equality before
the law requires equality not only between the rich and the poor or the high-
and the low-born, but also between blacks and whites and between men and
women. Medieval European society and, to an even greater extent, traditional
Indian society, was a society of privileges and disabilities; by contrast we now
have a society of citizens entitled to, if not actually enjoying, the equal
protection of laws.
The acceptance in principle of equality before the law or of equality of
opportunity does not mean, of course, that inequalities of status and power
have ceased to exist. There is a vast body of sociological literature showing
beyond a shadow of doubt that such inequalities do exist in all modern
industrial societies (Bendix and Lipset 1966, Helier 1969, Beteille 1969). There
is, as one would expect, a polemical side to this. Socialist writers from the
Soviet Union and from East European countries have argued that since such
mequa 1t1es enve pnman y rom t e pnvate owners tp o property, t ey are to
be found in their most extreme form in capitalist countries, notably the United
States. Liberal writers from the West, on the other hand, have asserted that the
truly oppressive forms of inequality are those arising from the monolithic
concentration of power in the apparatus of state and party, as exemplified in
countries like the Soviet Union. 4
We might begin on neutral ground with a consideration of the occupational
structure of modern societies. It would be difficult to exaggerate the importance
of that structure in industrial societies, whether of the capitalist or the socialist
type. Occupations have become highly specialized, and the occupational system
has become more elaborate, more complex and more autonomous than in any
society previously known to history. Industrialization is accompanied not only by
a new attitude to work but also by a new organization of work (see this volume,
article 32). Much of a person's adult life is spent in his or her occupational role,
and early life is largely a preparation for it.
The hundreds of named occupations present in an industrial society are
classified and ranked. The principles of occupational ranking have been
discussed even more exhaustively by sociologists than have those of caste
ranking by anthropologists (Goldthorpe and Hope 1974). Studies in the
United States have shown that, although new occupations displace old ones
with great rapidity, the structure of occupational ranking shows a high degree of
stability. Moreover, comparative studies of occupational ranking in different
1028
industrial societies, of the capitalist as well as the socialist types, have shown
that this structure is not only remarkably stable but also relatively invariant
(Hodge et al. 1966a, b).
In general, non-manual occupations rank higher than manual ones, not only
in the United States, but also in the Soviet Union in spite of official theory
which assigns pride of place to manual work in the creation of value in the form
of material products. Doctors rank higher than typists, not only in the United
States where they are independent professionals, but also in the Soviet Union
where, like typists, they are state employees. Soviet attempts to level out
differences of income between occupations had limited success, despite strong
pressures from the state. They eventually had to be abandoned, and were later
condemned by Stalin (Lane 1971 ).
The question of why some occupations are consistently ranked higher than
others is in some ways as difficult to answer as the question of why some castes
are always ranked higher than others. It no more suffices to say that space
scientists rank higher than plumbers because they receive higher earnings, than
it does to say that Brahmans rank higher than Oilpressers because they have
greater purity. One might just as well ask why the space scientist should earn
more than the plumber. Various kinds of explanations, none of them very
satisfactory, have been offered, in terms of 'scarcity', 'function' and so on
(Bendix and Lipset 1966). It is quite clear, as Parsons (1954) consistently
stressed, that occupational ranking is governed by the value system of a society,
an t e more u y a gtven occupatiOn em o tes or expresses Its core va ues, t e
more highly it is likely to be ranked. There are only two qualifications to be
added: first, occupations alone do not express the core values of a society; and
secondly, their ranking is also governed, at least in part, by considerations of
power which are different from those of esteem.
Although occupational ranking may be as elaborate as caste ranking, the
nature of occupational status differs from that of caste status. Caste status is
ascribed whereas occupational status is, at least in principle, achieved. There is
no guarantee that an individual will have the same occupation, or even the same
occupational level, as his father, and the same individual may in fact move
considerably from one occupational level to another in his own lifetime.
Therefore, sociologists who study occupational structure and occupational
ranking also study occupational mobility. Indeed, the enormous literature on
the social grading of occupations has grown largely in response to the problems
of describing, analysing and measuring occupational mobility (Goldthorpe
1980: eh. 1).
The literature on occupational mobility in industrial societies is not only
very large but in parts highly technical (Blau and Duncan 1968), so that casual
inferences drawn from it are likely to be misleading. But some of the studies
have come to conclusions that at first sight appear surprising. In a pioneering
study made in the 1950s, Lipset and Bendix emphasized at the outset that 'the
overall pattern of social mobility appears to be much the same in the industrial
1029
SOCIAL LIFE
societies of various Western countries' (Lipset and Bendix 1967: 13). They
found their own conclusions 'startling' in view of the universal assumption that
the United States had much higher rates of mobility than European countries
like Britain and France. The earlier studies operated with such broad
differences of level as between 'manual' and 'non-manual' workers; more
refined analyses have naturally revealed variations in rates of mobility within
the same overall pattern.
An important issue in the study of social mobility relates to its implications
for the formation and stability of classes (Goldthorpe 1980). Sociologists who
deal with this question tend to approach it from two different points of view.
There are those who maintain that the multiplicity of occupational levels
together with high rates of individual mobility renders the formation of
distinct and stable social classes difficult if not impossible in advanced
industrial societies. Blau and Duncan (1968) argued in an influential book that
high rates of mobility make most individual positions impermanent to such an
extent that few individuals are likely to develop a lifelong commitment to any
particular class. 'Class' then becomes a statistical construct rather than a
socially significant category.
The second approach is a Marxian one. Marxists have traditionally held an
ambivalent attitude towards individual mobility. On the one hand, they have
questioned whether capitalist societies have high or even rising rates of
mobility. On the other hand, they have maintained that rates of mobility have
1tt e, 1 anyt mg, to o wtt t e po anzatwn o c asses-w tc t ey see as an
historical tendency generated by contradictions within capitalism (Poulantzas
1976). A reasonable position would seem to be that, while rising rates of
individual mobility do alter the context of class conflict, they do not abolish
class identity as such, certainly not the identity of the working class
(Goldthorpe 1980).
Marxists, as is well known, contrast class with occupation (Dahrendorf
1959: pt. I), and assign far more importance to the former than to the latter, at
least in the analysis of capitalist societies. The importance that we assign to
class in industrial societies in general, as against the capitalist variant alone, will
depend on what we mean by class. In the Marxian scheme, the inequality of
classes is much less a matter of status and esteem than of unequal power in the
economic domain. The inequality of power is itself seen to be rooted in the
particular historical institution of private property. Thus, in this scheme,
although inequalities of power are crucial and quite large in capitalist societies,
they can, at least in principle, be greatly reduced, if not eliminated, by the
abolition of private property.
Others argue that property is only one of the bases of power, and that power
has other bases that would survive the abolition of property and might even be
strengthened in consequence. These writers also tend to subordinate esteem or
status to power in their analytical schemes, but in a way that is different from
that of the Marxists (Dahrendorf 1968). In their view power is a universal and
1030
inescapable source of inequality which permeates all forms of human life,

particularly in societies organized on a large scale. To be sure, there are
variations in the patterns of its distribution, and the resulting inequalities can
be controlled or regulated to some extent, but never eliminated altogether.
Some have taken the further step of trying to redefine class by substituting
power (or more narrowly, authority) for property. Thus, for Dahrendorf
(1959:204), 'the term "class" signifies conflict groups that are generated by the
differential distribution of authority in imperatively co-ordinated associations'.
The presumption behind this definition (like the one behind the definition it
seeks to supersede) is that inequality of power (like inequality of property)
generates conflict. Whether it does so or not, and under what conditions, to
what extent and in what forms, are important questions on which there is much
disagreement and some confusion among sociologists. Some regard 'class' as an
aspect of 'stratification', being primarily an expression of the economic ranking
of groups. Others regard 'stratification' and 'class' as fundamentally different;
according to them, 'stratification' relates to the ranking of groups, whereas
'class' relates to the conflict of interests between them (Dahrendorf 1959).
EQUALITY AS CONCEPT AND IDEAL

We are now in a position to return to a consideration of equality as a concept
and an ideal. There is no doubt that inequalities of status and power exist
everyw ere m t e mo ern wor , ut t ey now extst m a great y a tere
and moral environment.
'Hierarchy', wrote Marx and Engels, 'is the ideal form of feudalism'
(1968: 190), and it was also the ideal form of other past civilizations. This is not
the case today. People live with inequality, they may seek to explain or even to
justify it, but they no longer idealize it. This is true not only of England, France
and the United States, where the modern ideal of equality first took shape, but
also of countries like India to which it later spread.
But the ideal of equality is no less confusing a subject than the reality of
inequality, and the confusion is compounded when we seek to consider it in a
comparative perspective. Two questions may be asked at this point: first,
whether the ideal of equality is indeed unique to modern ideology and, if so, in
what sense; and, second, whether the ideal, or rather the concept behind it, is a
coherent one.
Some scholars believe that egalitarian values have not only originated in the
West but are, moreover, somehow incompatible with non-Western societies and
cultures. Others maintain that they are neither uniquely Western nor uniquely
modern. I have already alluded to Wood burn's ( 1982) argument about the
'egalitarian societies' of certain hunting and gathering peoples. References to
egalitarian values are not uncommon in the comparative anthropological
literature on tribal societies including those of pastoralists and agriculturalists.
On a larger scale, Islamic civilization was in many respects more markedly
1031
SOCIAL LIFE
egalitarian than medieval Christianity, not to speak of medieval Hinduism

(Gellner 1981 ). However, Islamic egalitarianism lacked the universality
characteristic of modern egalitarianism: it denied equality, even in the formal
sense, to women as well as to adherents of other faiths. In the case of most tribal
societies as well, this lack of universality also qualifies such commitment to
equality as there is.
Although modern societies have universalized the idea of equality and have
elaborated it in moral, legal and political discourse to an unparalleled degree,
they have not come anywhere near to the equality of condition said to be
common in many tribal communities, including those adhering to Islam. The
modern idea of equality arose under specific historical conditions, in response
to a society where hierarchy was deeply and firmly entrenched. It was under
such conditions that 'equality of opportunity'-or, in Napoleon's famous
phrase, 'careers open to talent'-became a powerful slogan. Equality of
opportunity could hardly be a forceful idea in a tribal society where equality of
condition, or near-equality of condition, is an established datum of experience.
The idea of equality of opportunity, which was a new one in Napoleon's
time, had already lost its shine a century later. R.H.Tawney, one of the
strongest advocates of equality in the inter-war years, saw clearly that in a
society marked by acquisitiveness and untempered competition, equality of
opportunity by itself could do little to reduce the gap between the rich and the
poor, and might in fact increase it (Tawney 1964). Thus he contrasted equality
o opportumty wtt w at e ea e practtca equa tty , an soug t to ma e t e
latter the central focus of social policy.
It is through considerations of social policy rather than abstract speculation
that the ambiguities in the concept of equality have become manifest. We know
today that legal equality, equality of opportunity and even rising rates of
mobility can coexist with increasing inequality in the distribution of income. As
we have seen, equality of opportunity can be of significance only in a society
based on the competition of individuals. But this means that there can be
equality only before the competition, and not after it. From this it may be
argued that the commitment to equality requires not only that the competition
itself should be free, but also that the rewards of success should not be too
lavish nor the penalties of failure too severe.
Thus, equality may signify equality of opportunity, or it may signify
equality in the distribution of things (Beteille 1985). If it is true that modern
ideology sets a high value on equality, it is also true that it is deeply divided
between these two conceptions of what equality is. Several positions may be
taken on this. One may argue that there is no real contradiction between the
two, that the contradiction is only apparent. If we take equality of opportunity
to mean an equality that is 'fair' and not merely 'formal', then we can more
easily reconcile it with equality, or at least equalization, in the distribution of
things (Rawls 1972:83-9, 298-303).
Others would maintain that the idea of a 'fair' equality of opportunity is
1032
subjective and arbitrary, and that the ideal of equality cannot be tested against
any preconceived model of distribution. This being so, substantive equality is a
kind of mirage whose pursuit is bound to be self-defeating. More importantly,
it can subvert the ideal of formal equality, or equality before the law, which in
this view is where the essence of equality lies (Hayek 1960: eh. 6, Joseph and
Sumption 1979).
If we now look back on the transition from the 'aristocratic' to the
'democratic' type of society, or from the 'hierarchical' to the 'egalitarian' type,
we realize how complex the issues are. When we look at that transition in
Europe, and also elsewhere, we cannot but be struck by the crucial part played
in it by the forces of the 'self-regulating market'. These forces broke down old
barriers and created new cleavages. In Europe the old distinctions of estate,
guild and parish yielded before the expanding forces of the market to the extent
that the latter took less account of social origin than of individual ability.
However, the market did not dissolve all the old distinctions, some of which
survived, although in altered forms, and accommodated themselves to it. First
of all, there are countries like India where market forces have not penetrated far
enough and where so-called 'semi-feudal' arrangements, based on caste and
patronage, are still well entrenched. It can of course be argued that what
survives from the past will inevitably decay as and when the market takes full
command. But this argument loses much of its force when we see that
distinctions of race and ethnicity, and sometimes marked disparities based on
t em, ouns even m sue a mature captta 1st society as t e Umte tates.
The market also sharpens old distinctions, and creates new ones, the most
important being the distinction between capital and labour. The widening gap
between capital and labour, and the simultaneous enrichment of the few and
impoverishment of the many in mid-nineteenth-century England, were noted
not only by Marx and Engels but also by many others who witnessed the
expansion of market forces at first hand. It is true that the worst excesses of this
phase of capitalism have to some extent been corrected, at least in the advanced
capitalist societies, but it is not true that they have all been corrected solely by
the 'self-regulating market'. Few of those who are witnessing the expansion of
market forces and the accompanying rise in economic disparity in India and
other Third World countries can seriously believe that they should wait for the
market itself to correct these disparities in the long run.
The belief that the inequalities inherited from the past and those being
generated at present can and should be corrected by some form of social
intervention is widely, if not universally, held in countries like India, and is also
held by varying and fluctuating sections of society in countries like Britain and
the United States. Of course, such intervention can be of many different kinds,
and opinion is naturally divided on who should intervene, to what extent and in
which areas of social life. A certain consensus on these issues, however fragile
and momentary, was embodied in the institutions of the welfare state created in
a number of West European countries in the wake of the Second World War.
1033
SOCIAL LIFE
Given the full range of historical possibility and experience, the welfare state
of post-war Western Europe appears as a relatively mild instrument for the
containment of inequality. Far more powerful apparatuses of state and
government have been devised in the Soviet Union and elsewhere, at least in
part with the objective of reducing inequality. Nor should we underestimate
their achievements. There were notable successes in controlling
unemployment, in giving workers a better deal and in reducing income
differentials between 'mental' and 'manual' workers. Some advances were also
made since the Bolshevik Revolution in reducing disparities between the
different ethnic groups and nationalities, but many disparities still remain, as is
becoming evident in the rising tide of ethnic conflicts.
The notable gains in equality mentioned above were achieved at some cost,
which, by any reasonable account, was at times exceedingly high. A
consideration of this cost at once reveals one of the paradoxes of equality. The
very attempt to regulate and reduce inequality through direct intervention in
social and economic processes led, some would say inevitably, to a tremendous
concentration of power in the apparatuses of state and party. In other words,
the instruments for the suppression of inequality are not neutral, but generate
their own inequalities. One could then ask whether, in moving from the
inequalities of estate prevalent until the eighteenth century to the inequalities
of class about which Marx wrote, and from those again to the inequalities of
power of the twentieth century, any real or demonstrable gain was made in the
ac tevement o equa tty.
A monolithic structure of power imposes constraints on the realization not
only of equality but also of other social values, notably liberty. It may be
possible in principle to envisage an ideal world where liberty and equality
would complement rather than contradict each other; but such an ideal world is
not yet within reach, and perhaps for most, not even within sight. Libertarians
do not question the principle of equality before the law, or even of equality of
opportunity to the extent that it is consistent with the former. But they do
question the 'legitimacy of altering social institutions to achieve greater
equality of material condition' (Nozick 1980:232), whether in the name of
distributive equality or of 'fair equality of opportunity'.
The stress on distributive equality may be viewed as a threat not only to
liberty but also to efficiency. Few people would place efficiency on the same
plane as equality and liberty in their hierarchy of values. It is nevertheless true
that efficiency has a central place in the economic ideology that dominates
much of modern life. Some of the most crucial debates in the realm of social
and economic policy relate to the comparative advantages of market and plan as
two alternative forms of rationality (Dahrendorf 1968). A major test of these
advantages, even for those who believe that the two alternatives cannot be
mutually exclusive, is the degree of efficiency attainable under each, either
singly or in combination with some elements of the other.
1034
Modern egalitarians have always argued that an order that tolerates extremes
of inequality is not only socially unjust but also economically wasteful and
inefficient. But the considerable experience now available of centrally-regulated
economies has shown up the other side of the coin. In the socialist countries,
the market was for decades held responsible for both generating and sustaining
economic inequality, and one of the main objectives of centralized planning was
precisely to restrict that role. If the market is viewed with less suspicion in these
countries today, it is not because its role in sustaining inequality has been
completely lost to sight, but rather because people are now a little better
prepared to accept some economic inequality as a price to pay for the efficiency
guaranteed by a measure of competition.
Thus, although equality is undoubtedly an important value in modern
societies, there is a considerable distance between a minimal definition of it as
equality before the law and a definition that also tries to take into account the
distribution of income, wealth and various social services, such as health and
education. One must always keep in mind that there are not only strong
advocates of equality in these societies but also critics of it (Letwin 1983).
These critics point not only to the high political and economic costs of realizing
equality, but also to the conceptual ambiguity inherent in the very idea of
equality. 'The central argument for Equality', a contemporary political
philosopher has written, 'is a muddle' (Lucas 1965:299). And even of the more
specific ideal of 'equality of opportunity', a distinguished American
e ucatwmst as wntten, seemmg y m espair, t at It IS a a se I ea
1973:135).
Perhaps equality is not so much a false ideal as one which cannot be
meaningfully conceived in an historical vacuum. It can only make sense in the
context of, and in response to, the specific challenge that a given society
presents to its reflective members. Sometimes the challenge comes from an
order established by age-old religious tradition, such as that of caste;
sometimes it comes from a recklessly competitive economic system such as that
of free-enterprise capitalism; or again, it may come from a monolithic political
apparatus itself designed to solve the problem of inequality once and for all.
Equality is today too powerful an idea to be set aside simply because it cannot
be precisely defined. It is like the djinn which, once released from the bottle,
cannot be put back into it again.
NOTES
1 To take an example from outside the West, Rousseau influenced the great nineteenth-
century Bengali writer Bankimchandra Chattopadhyay, who published a tract on
equality entitled Samya in 1879; for an English translation, see Haldar (1977); see also
Ganguli (1975).
2 Soviet writers have generally preferred the term 'strata' to 'classes' to describe the
characteristic divisions of their own society which, according to them, was marked by an
absence of 'contradiction' or, at least, of 'antagonistic contradiction'.
1035
SOCIAL LIFE
3 This article was completed in 1989. No account is taken of political developments in

SouthMrica since that time.
4 The collapse of the Soviet Union, which took place after this article was written, has
rendered much of this polemic at least temporarily obsolete. In what follows, references
to the Soviet Union specify conditions predating the collapse.
REFERENCES
Ardener, S. (ed.) (1975) Perceiving Women, London: Dent.
Bendix, R. and Lipset S.M. (eds) (1966) Class, Status and Power, New York: Free Press.
Berlin, I. (1978) Concepts and Categories, London: Hogarth Press.
Berreman, G. D. (1960) 'Caste in India and the United States', American Journal of
Sociology 66:120-7.
--(1966) 'Caste in cross-cultural perspective', in G.De Vos and H.Wagatsuma (eds)
Japan's Invisible Race, Berkeley: University of California Press.
Beteille, A. (1964) 'A note on the referents of caste', European Journal of Sociology, 5:
130-4.
--(1965) Caste, Class and Power, Berkeley: University of California Press.
--(1977) Inequality Among Men, Oxford: Blackwell.
--(1980) The Idea of Natural Inequality, London: London School of Economics.
--(1985) Equality of Opportunity and the Equal Distribution of Benefits, Pune:
Gokhale Institute.
--(1986) 'Individualism and equality', Current Anthropology 27(2): 121-54.
--(1990) 'Race, caste and gender', Man (N.S.) 25:489-504.
Blau, P.M. and Duncan, O.D. (1968) The American Occupational Structure, New York:
Free Press.
Bose, N.K. (197 5 [1949]) The Structure of Hindu Society, Delhi: Orient Longman.
Coleman, J.S. (1973) 'Equality of opportunity and equality of results', Harvard
Educational Review 43(1):129-37.
Dahrendorf, R. (1959) Class and Class Conflict in Industrial Society, London: Routledge
& Kegan Paul.
--(1968) Essays in the Theory of Society, Stanford, Cal.: Stanford University Press.
Davis, K. and Moore, W.E. (1945) 'Some principles of stratification', American
Sociological Review 10(2):242-9.
de Reuck, A. and Knight, J. (eds) (1968) Caste and Race. London: J. & A. Churchill,
de Tocqueville, A. (1956 [1835, 1840]). Democracy in America, 2 vols, New York: Alfred
Knopf.
Dumont, L. (1961) 'Caste, racism and "stratification"', Contributions to Indian
Sociology 5:20-43.
--(1964) La Civilization indienne et nous, Paris: Armand Colin.
--(1966) Homo hierarchies, Paris: Gallimard.
--(1977) Homo aequalis, Paris: Gallimard.
--(1980) On Value, London: British Academy.
--(1983) Essais sur l'Individualisme, Paris: Seuil.
--(1987) 'On individualism and equality', Current Anthropology, 28(5):669-77.
Durkheim, E. (1982 [1893]) The Division of Labour in Society, London: Macmillan.
1036
Durkheim, E. and Mauss, M. (1963 [1903]) Primitive Classification, London: Cohen &
West.
Engels, F. (1948 [1884]) The Origin of the Family, Private Property and the State,
Moscow: Progress Publishers.
Evans-Pritchard, E.E. (1940) The Nuer, Oxford: Clarendon Press.
--(1965) The Position of Women in Primitive Societies, and Other Essays in Social
Anthropology, London: Faber & Faber.
Eysenck, H.J. (1973) The Inequality of Man, London: Temple Smith.
Fortes, M. (1945) The Dynamics of Clanship Among the Tallensi, London: Oxford
University Press.
--(1949) The Web of Kinship Among the Tallensi, London: Oxford University Press.
Frazier, E.F. (1957) Black Bourgeoisie, New York: Free Press.
Ganguli, B. N. (197 5) Concept of Equality, Simla: Indian Institute of Advanced Study.
Gellner, E. (1979) 'The social roots of egalitarianism', Dialectics and Humanism 4: 27-
43.
--(1981) Muslim Society, Cambridge: Cambridge University Press.
Goldthorpe, J.H. (1980) Social Mobility and Class Structure in Modern Britain, Oxford:
Clarendon Press.
Goldthorpe, J.H. and Hope, K. (1974) The Social Grading of Occupations, Oxford:
Clarendon Press.
Goodenough, WH. (1951) Property, Kin and Community on Truk, New Haven: Yale
University Press.
Haldar, M.K. (1977) Renaissance and Reaction in Nineteenth Century Bengal, Calcutta:
Minerva Associates.
aye , . . e onstttutton o t erty, on on: out e ge & egan
Helier, C.S. (ed.) (1969) Structured Social Inequality, London: Macmillan.
Hobbes, T. (1973 [1651]) Leviathan, London: Dent.
Hodge, R.W, Siegel, P.M. and Rossi, P.H. (1966a) 'Occupational prestige in the United
States: 1925-63', in R.Bendix and S.M.Lipset (eds) Class, Status and Power, New
York: Free Press.
Hodge, R.W, Trieman, D.H. and Rossi, P.H. (1966b) 'A comparative study of
occupational prestige', in R.Bendix and S.M.Lipset (eds) Class, Status and Power,
New York: Free Press.
Huizinga, J. (1924) The Waning of the Middle Ages, London: Edward Arnold.
Joseph, K. and Sumption, J. (1979) Equality, London: John Murray.
Kakar, S. (1978) The Inner World, Delhi: Oxford University Press.
Kane, P.V. (1974) History of Dharmasastra, vol. 2, pt. 1, Poona: Bhandarkar Oriental
Research Institute.
Kuznets, S. (1955) 'Economic growth and income inequality', American Economic
Review 45(1).
Lane, D. (1971) The End of Inequality? Harmondsworth: Penguin.
Larson, G.J. and Bhattacharya, R.S. (eds) (1987) Encyclopedia of Indian Philosophy:
Samkhya, Princeton: Princeton University Press.
Leacock, E. (1978) 'Women's status in egalitarian societies', Current Anthropology
19(2):247-75.
Lenski, G. (1966) Power and Privilege, New York: McGraw-Hill.
Letwin, W (ed.) (1983)Against Equality, London: Macmillan.
Levi-Strauss, C. (1966) The Savage Mind, London: Weidenfeld & Nicolson.
1037
SOCIAL LIFE
Lingat, R. (1973) The Classical Law of India, Berkeley: University of California Press.
Lipset, S.M. and Bendix, R. (1967) Social Mobility in Industrial Society, Berkeley:
Lowie, R. (1960 [1921]) Primitive Society, London: Routledge & Kegan Paul.
Lucas, JR. (1965) 'Against Equality', Philosophy 40(154):296-307.
MacCormack, C. and Strathern, M. (eds) (1980) Nature, Culture and Gender,
Marriott, M. (1959) 'lnteractional and attributional theories of caste ranking', Man in
India 39(2):92-107.
--(1968) 'Caste ranking and food transaction', in M.Singer and B.S.Cohn (eds)
Structure and Change in Indian Society, Chicago: Aldine.
Marriott, M. and lnden, R.B. (1974) 'Caste systems', Encyclopaedia Britannic a
(Macropaedia), vol. 3.
Marx, K. (1954 [1867]) Capital, vol. 1, Moscow: Progress Publishers.
Marx, K. and Engels, F. (1968) The German Ideology, Moscow: Progress Publishers.
Mead, M. (1963 [1935]) Sex and Temperament in Three Primitive Societies, New York:
Morrow.
Morgan, L.H. (1964 [1877]) Ancient Society, Cambridge, Mass.: Harvard University
(Belknap) Press.
Myrdal, G. (1944) An American Dilemma, New York: Harper.
Nandy, A. (1980) At the Edge of Psychology, Delhi: Oxford University Press.
Nozick, R. (1980) Anarchy, State and Utopia, Oxford: Blackwell.
Parsons, T. (1954) Essays in Sociological Theory, New York: Free Press.
Poulantzas, N. (1976) Les Classes sociales dans le capitalisme aujourd'hui, Paris: Seuil.
aw s, . eory o usttce, on on: x or mvers1ty ress.
Rege, M.P. (1984) Concepts of Justice and Equality in the Indian Tradition, Pune:
Gokhale Institute.
--(1988) 'Dharma: man, society and polity', New Quest 69:133-40.
Rivers, WH.R. (1924) Social Organization, London: Kegan Paul.
Rousseau, J-J. (1938 [1782]) The Social Contract and Discourses, London: J.M.Dent.
Sahlins, M.D. (1968) Tribesmen, Englewood Cliffs, NJ: Prentice-Hall.
Schneider, D. M. (1968) American Kinship, Englewood Cliffs, NJ: Prentice-Hall.
Sen, A. (1973) On Economic Inequality, Oxford: Blackwell.
Srinivas, M. N. (1962) Caste in Modern India and Other Essays, Bombay: Asia Publishing
House.
Srinivas, M.N. (1968) 'Mobility in the caste system', in M.Singer and B.S.Cohn (eds)
Structure and Change in Indian Society, Chicago: Aldine.
Tapper, R. (ed.) (1983) The Conflict of Tribe and State in Iran and Afghanistan, London:
Croom Helm.
Tawney, R.H. (1964 [1931]) Equality, London: Unwin.
Warner, W.L. (1941) 'Introduction', inA.Davis, B.B.Gardner and M.R.Gardner, Deep
South, Chicago: University of Chicago Press.
Wolf, E.R. (1982) Europe and the People Without History, Berkeley: University of
California Press.
Wood burn, J. (1982) 'Egalitarian societies', Man (N.S.) 17:431-51.
Zaehner, R.C. (ed.) (1969) The Bhagvad-Gita, London: Oxford University Press.
1038
FURTHER READING
Bendix, R. and Lipset, S.M. (eds) (1966) Class, Status and Power, New York: Free Press.
Beteille, A. (1977) Inequality Among Men, Oxford: Blackwell.
--(1987) The Idea of Natural Inequality and Other Essays, Delhi: Oxford University
Press.
Bourdieu, P. (1985) Distinction, London: Routledge & Kegan Paul.
Duby, G. (1980) The Three Orders, Chicago: University of Chicago Press.
Dumont, L. (1980) Homo hierarchicus, Chicago: University of Chicago Press.
Eysenck, H.J. (1973) The Inequality of Man, London: Temple Smith.
Franklin, J.H. (ed.) (1968) Color and Race, Boston: Houghton Mifflin.
Goldthorpe, J.H. (1980) Social Mobility and Class Structure in Modern Britain, Oxford:
Clarendon Press.
Jencks, C. (1973) Inequality, Harmondsworth: Penguin.
Leach, E.R. (1970) Political Systems of Highland Burma, London: Athlone Press.
Letwin, W. (ed.) (1983) Against Equality, London: Macmillan.
Marshall, T.H. (1977) Class, Citizenship and Social Development, Chicago: University of
Chicago Press.
Ossowski, S. (1963) Class Structure in the Social Consciousness, London: Routledge &
Kegan Paul.
Sen, A. (1973) On Economic Inequality, Oxford: Blackwell.
Strathern, M. (ed.) (1987) Dealing with Inequality, Cambridge: Cambridge University
Press.
Tawney, R.H. (1964) Equality, London: Unwin Books.
Academic Press.
Tumin, M.M. (1985) Social Stratification, Englewood Cliffs, NJ: Prentice-Hall.
Wesolowski, W. (1979) Classes, Strata and Power, London: Routledge & Kegan Paul.
1039
38
THE NATION STATE, COLONIAL

EXPANSION AND THE
CONTEMPORARY WORLD ORDER
Peter Worsley
THE RISE OF THE NATION STATE

The rise of the nation state is a modern phenomenon. Its origins, in Europe,
date back only two centuries. The earlier rise of the centralized state entailed
t ree mterre ate processes: t e concentratiOn o po 1t1ca power, economic
centralization, and cultural hegemony.
The establishment of monarchical supremacy over hitherto vigorously self-
assertive aristocracies-especially the great feudal magnates whose vast estates
provided them with strong regional bases of power-was a long-drawn-out
struggle. The new absolutist monarchs also had to construct their states out of
feudal polities in which the consent of the traditional estates had to be gained
for major taxation (Anderson 1973: chs 1 and 2). The creation of a system of
centralized taxation through which money was directly available to the
monarch made it possible to raise military forces that came immediately under
the sovereign's command. 'A prince', Machiavelli wrote, 'should ... have no
other thought or aim than war.' The new armies and navies were used, not just
to bring the magnates to heel, but to expand the economic system by protecting
the domestic market and stepping-up overseas trade.
Culturally, the consolidation of the absolutist monarchy led to the
domination of the culture of the victorious heartland over provincial cultures:
for example, the transformation of the dialect of the Isle de France around Paris
into a 'national' French, and of the East Midlands dialect into 'Standard
English'. But thorough-going cultural standardization was not achieved by any
Absolutist state. 'The ideological conceptions of "nationalism"', Anderson has
remarked, 'were foreign to the inborn nature of absolutism' (Anderson
1973:38). It was the bourgeoisies that inherited the centralized polities created
1040
COLONIALISM AND THE WORLD ORDER
by the absolutist monarchs who instituted national systems of education to

meet the requirements of industrial society and to integrate the citizen with the
state (Gellner 1983 )-a model that was subsequently exported to the rest of the
world (see this volume, Article 25).
NON-EUROPEAN POLITIES AT THE TIME OF

EUROPEAN COLONIAL EXPANSION
Empires and states

The societies encountered by Europeans during the expansion of the West
varied enormously, from the 'stateless' societies Columbus found in the
Caribbean to the great empires ofTurkey and China. The latter were far greater
in size and wealth than any European state. To the Ottomans, rulers of fifty
million people at a time when Queen Elizabeth inherited a state with only five
million, the inferiority of Europe was self-evident: in 1666, the Grand Vizier of
the Ottoman court addressed the French Ambassador as a 'Giaour [unbeliever],
a hogge, a dogge, a turde eater'. Fifty years later, Oliver Cromwell's grandson,
Governor of Fort William (Calcutta), was expected to make obeisance to the
Moghul Emperor 'with the reverence due from a slave' (Stavrianos 1981: 157).
Technologically, Europe had no great superiority over Asian economies,
w ose m ustnes range rom arge-sca e s tp- m mg to sop tsttcate
textiles. Superior military equipment had proved decisive in the conquest by
Europeans of the Aztec and Inca Empires, and their superiority in sea power
and in naval tactics enabled them to establish small coastal trading-posts. But
for the two and a half centuries after Da Gama, they were effectively excluded
from the Indian subcontinent. Even at their height, the Portuguese in Asia as a
whole were merely middlemen in a purely intra-Asian trade in which European
goods were unimportant (Stavrianos 1981:158,230, van Leur 1955:281).
Asia was also the centre from which most of the great 'world-religions'
(Weber 1956) reached the adjoining regions: Hinduism spread into South India
and South-east Asia (Fuller 1984); Buddhism spread from North India
southwards, reaching present-day Sri Lanka several centuries before the birth
of Christ (Geiger 1986, Gunawardana 1979), as well as northwards into the
Himalayan zone and China, and into countries south and east of the
subcontinent. The spread of Chinese influence over Korea, Japan and ludo-
China was as much cultural as political. In these countries, where religious
'Great Traditions' (Tambiah 1970) flourished, resistance to European culture
was correspondingly stronger.
Yet not all the societies encountered by Europeans were large empires. Many
states did not conform to Weber's 'rational-legal' ideal type, with monopoly of
the legitimate use of physical force within a given territory, a centralized
bureaucracy, and an effective system of state economic organization (Weber
1041
SOCIAL LIFE
1961:249-58). China, the 'world's largest enduring state' (Elvin 1973:15),

certainly possessed, in the mandarinate, a highly organized administrative
apparatus. Yet the political articulation and integration of the provinces varied
with the vicissitudes of power at the centre: at times of weak central control, the
empire would divide into lower-level regions. Similarly, the range of economic
articulation and integration of the different levels of marketing system
fluctuated over the centuries (Chi 1936, Skinner 1964-5).
But many states were endemically weak. The authority of the ruler was often
little more than the acceptance of the loose suzerainty, often largely ritual, of
one political grouping-an aristocratic house or tribal group-over others
equally noble or powerful. Hence succession was often determined less by clear
rules than by civil war between followers of rival royal or noble houses. In lieu
of a 'rational-legal' system of administration, noble representatives of tribal
houses or conquered tribes were made to serve at court. Random levies,
patronage, or campaigns to secure booty abroad took the place of a 'rational'
system of securing income for the state.
Weak states of this kind were therefore often segmentary in structure rather
than centralized (Southalll956). The prime focus of individual loyalty was not
the state but local authority-figures and communities; the first allegiance was to
one's lord, or to clan and tribe, or to religious communities-the 'little
traditions' of earth and ancestor-cults.
These variations of state structure and civil society were of major
consequence w en t e Europeans arnve . Even t e great emptres contame
serious structural weaknesses, visible to Europeans, which Sir Waiter Raleigh
summed up as a 'void oflibertie' and a 'want of Nobilitie'-the absence of any
checks on the sovereign's power, especially on the part of an independent land-
owning class; a state of affairs which Montesquieu was later to designate as
'Asiatic Despotism' (Anderson 1973:462ff.). Weak state structures were
susceptible to division and manipulation by determined invaders. Thus Cortes,
with only 600 Spaniards under his command, was able to conquer Mexico
because he was assisted by tens of thousands of traditional enemies of the
Aztecs from Tlaxcala and Texcoco; Pizarro, in Peru, was able to exploit
divisions resulting from a very recent succession war. Clive's victory at Plassey
in 1757 turned, in the end, on the defection of one of the enemy's generals.
Stateless societies
By no means all of the world penetrated by Europeans was inhabited by
populations living in states, let alone empires. Large parts of Amazonian South
America and virtually all of North America, as well as Australia and many other
parts of the globe, were occupied by societies without a state apparatus. The
social and political institutions of 'stateless' societies, however, were very
varied. There were societies with chiefs and hereditary aristocracies, even with
slaves, as well as societies where age and sex were the primary bases of status,
1042
rank and authority (see Article 34). Using such political criteria, societies of
this kind have often been called 'tribes' or 'bands', and on the grounds of
techno-economic criteria, these have been associated respectively with
'agricultural' and 'hunting and gathering' economies. Whichever criteria are
adopted, whether political or techno-economic, such designations ignore the
profound differences between peoples around the world whose cultures are as
dissimilar as the languages they speak.
Terms such as 'stateless', or 'acephalous', are in any case only negative,
residual categories; they tell us what these societies are not. In an attempt to
provide a positive designation of their common attributes, Wolf has called them
societies based on a 'kin-ordered' mode of production (Wolf 1982:88-100).
Kinship systems, he accepts, may be of many kinds; moreover kinship is neither
equally salient in all stateless societies, nor does it fulfil the same functions. But
in so far as it is used to regulate descent and marriage, it does affect the
deployment of economic and political power.
However exiguous the material equipment of such peoples, their systems of
religious belief are rich and complex, and the empirical knowledge they possess of
their environments, in particular of the plants and animals on which they
depend, is both wide-ranging and intellectually highly organized (Waddy 1988).
In such societies, it is people and their knowledge, rather than things or capital,
that are the crucial social resource: their labour-power, their skills and, in the case
of women, their capacity to produce more people. They are not, as nineteenth-
century et no ogtsts t oug t, pnmtttve commumsts : t ere IS mstttutwna tze
differentiation, particularly of sex and age, which recurs generation after
generation-inequalities, for instance, as between the original settlers of the land
and newcomers, or between senior and junior lines of descent (see Article 37).
Even stateless polities, lacking kings or chiefs and specialized military forces,
were capable of co-ordinated and steady campaigns of resistance or aggression
against neighbouring peoples. For example, the segmentary lineage
organization of such tribal peoples as the Nuer of the southern Sudan was
preadapted to a process of what Sahlins ( 1961) has called 'predatory
expansion'. Such polities were also capable of radical political innovation. The
arrival of British colonial forces in Nuerland, for instance, resulted in the rise
of religious prophets who were able to mobilize very large numbers of people
(Evans-Pritchard 1937). Similarly, in Melanesia, individuals and communities
who believed in the other-worldly source of material commodities, and in a
future apocalypse, followed prophets who foretold the imminent end of the
world-one in which the whites would be defeated and their goods would fall
into the hands of the natives (Worsley 1957).
The establishment of European rule was not necessarily accomplished
suddenly, as in South America. In North America, the struggle between Britain
and France for control of the fur trade and for political domination of the
region sucked different Amerindian peoples into a succession of wars. In the
process, institutions which had brought separate groups together, often for
1043
SOCIAL LIFE
ritual purposes, were transformed. The Iroquois Confederacy, for example,

established initially for the peaceable settlement of disputes between the five
'nations', and for co-ordinating defence against outsiders, became a mechanism
for organizing war against their neighbours and was increasingly wracked by
violent internal battles for hegemony. The alliance of the Iroquois with the
losing side, the British, during the American War of Independence, proved to
be their final undoing. Subsequent movements among the Iroquois, notably the
Handsome Lake prophetic movement around the beginning of the nineteenth
century, focused on personal spiritual revival and revelation and were staunchly
opposed to war (Kehoe 1981 :244--50).
THE BEGINNINGS OF EUROPEAN EXPANSION

In pre-capitalist Europe, state power was agrarian power, consisting in control
over land and over the labour which produced wealth from the land.
In classical antiquity, trade had been both predominantly seaborne, due to
the geographical location of the Mediterranean countries, and largely confined
to luxury commodities, notably spices, silks and fine cottons imported from the
Orient. But by the late medieval period, the centre of European trade had
moved to the north of the continent. This was a trade not in luxuries but in
necessities, notably the export, on a massive scale, of timber and grain from
Eastern Europe to Western Europe. The states of Western Europe could only
so ve t e resu tmg negative a ance o payments y exportmg go an st ver.
Beginning in the fifteenth century, the trading relations of the largest
Eastern European state, Russia, shifted eastwards. In 1584 the Cossacks began
crossing the Urals, and by 1637 the Russians had reached Okhotsk on the
Pacific, having traversed a distance half as far again as that between the Atlantic
and Pacific coasts of the United States, and at a time when the English colonists
in America had not yet crossed the Alleghany mountains (Stavrianos 1981 :69).
This orientation to the Asian hinterland deflected Russia from colonial
adventure outside the Euro-Asian landmass, while the preservation of a social
structure founded on serfdom at the bottom and autocracy at the top became
the principal preoccupation of the Tsars. The nobility exchanged power over
the state for power over their serfs, while trade passed into the hands of foreign
middlemen. These developments also cut the Russian Empire off from the
technological and economic advances that were taking place in the West.
Despite periodic attempts of autocrats like Peter the Great and Catherine to
imitate the West, Russia and the other states of Eastern Europe increasingly
became an underdeveloped agrarian region.
In Western Europe, by contrast, the rise of the absolutist monarchies led to a
quite different organization and geographical orientation of trade, based on co-
operation between monarchical rulers and the rising mercantile bourgeoisies
which undertook overseas ventures in which the state invested funds and
provided military and political backing.
1044
International trade within Western Europe, especially in textiles, was

growing steadily and was based in towns where merchants had acquired
immunities from arbitrary state action. Increasingly, they involved themselves
in trade outside Europe, above all in the spice trade.
THE GROWTH OF THE SPICE TRADE

Eastern spices were needed in medieval times to preserve meat or to disguise
the taste. But spices such as nutmeg and cloves were also valued as
medicaments, even as aphrodisiacs, or simply for their flavours or scents. Their
importance was reflected in the immense social value placed upon them: in the
conspicuous consumption of a year's supply of cinnamon which Nero burned
at the funeral of his wife, or in the payment of 1,200 kilograms of pepper to the
Gothic king, Alaric, in return for his undertaking (later broken) not to sack
Rome. As early as the first century AD, the Romans had sailed as far as the
Malabar Coast of South India in their search for pepper, and to present-day Sri
Lanka for cinnamon, deliberately spreading economic 'disinformation' as to
where the spices were actually produced and who produced them.
The trade with Europe was to become the most fateful for the world. It had
been pioneered by Indonesian traders who had taken spices over 6,500
kilometres of ocean to Madagascar, whence Arabs or perhaps Phoenicians had
shipped them via the Red Sea, or overland up the Nile Valley, to the
Me tterranean. But wtt t e esta IS ment o ra ommatwn over t e
Eastern Mediterranean, control over the spice trade gradually became a
stranglehold. The Arab conquest of Cairo (Abu-Lughod 1971) signalled the
beginning of Muslim control over the land routes to China and the sea routes to
the East, which even the later capture of Constantinople by the Crusaders in
1204 could not break. Though the spice trade within Europe became a Venetian
monopoly, Venice was dependent on Arab compliance for access to the sources
of the spices. And even at the height ofVenetian power gold continued to flow
eastwards to pay for the spices. When the Ottomans took Constantinople in
1453, control of trade between the Mediterranean and the Orient seemed to
have fallen irrevocably into Muslim hands.
But within only a few years, Da Gama had rounded the Cape and Magellan
had circumnavigated the globe. European explorer-traders now poured into the
Indian and Pacific Oceans.
The possibility of defeating the Saracens now seemed on the cards. After the
Crusades, Christians in Europe had realized that they were only a minority in
the world, and could not hope to defeat Islam on their own (Southern 1962:27-
31 ). The idea of establishing alliances against the Turk with non-Muslim
empires in the East was canvassed as early as the thirteenth century. Some of
these projects-like the idea of making contact with the legendary Prester John
in Abyssinia, or with the Nestorian Christians in China-were fantasies;
others, especially the attempt to build an alliance with the Mongols, were more
1045
SOCIAL LIFE
realistic. Between 1245 and 1253, no less than four missions were sent to the
Mongol Khan by the Papacy alone, visits which were reciprocated by Mongol
embassies to Rome (Southern 1962:39-65).
But the dream of breaking the power of Islam only began to seem realizable
after the Arabs had been driven out of the Iberian peninsula and following the
conquests in America. Spanish confidence now knew no bounds. Muslim
resistance, some thought, could be broken by diverting the Nile to the Red Sea
or by raiding Mecca and seizing the Prophet's body; five thousand Spaniards, it
was even suggested, could take China. Thus inspired, Spanish and Portuguese
'discoverers' set out on voyages that were to end with the unification of the
entire globe. The central purpose of these expeditions was unambiguous:
Magellan's first round-the-world voyage, westwards, was an expedition to
reach the Spice Islands; the eastwards route, round the Cape, was aimed at
securing the sources of pepper on the Mala bar Coast.
THE CONQUEST OF SOUTH AMERICA: GOLD,

SILVER, SUGAR AND SLAVERY
In order to avoid head-on confrontation, Spain and Portugal accepted the Papal
division of the New World under which all lands west of a line near Cape Verde
were allocated to Spain, while those to the east went to Portugal. But an attempt
to define similar spheres of influence in the Spice Islands themselves, where the
pamar s a esta IS e t emse ves m Tt ore an t e Portuguese m Ternate,
was only resolved by bitter warfare between the colonizers and their respective
local allies (Spate 1979:99-100).
The absence of spices in the Americas was a bitter disappointment to
Columbus, but the gold ornaments of the Carib Indians suggested another
source of profit. Since there was not much gold to be had locally, the Spaniards
were encouraged to invade the mainland. Gold rapidly became the main
driving-force of the Spanish Conquest: 'We Spaniards', Cortt!S wrote, 'suffer
from an affliction of the heart which can only be cured by gold'. To a priest who
criticized his lack of concern with saving Indian souls, Pizzaro replied: 'I have
not come for any such reasons. I have come to take away from them their gold'.
The initial looting of Mexico-what Weber termed 'booty capitalism' (Bendix
1960:306)-soon exhausted the gold available. The conquerors were forced into
organizing the production of gold themselves on a massively expanded scale. By
the end of the sixteenth century, bullion, mainly silver, came to constitute more
than 95 per cent of all colonial exports, tripling the supply of silver in Europe, a
flow so gigantic that when Drake returned from 34 months of piracy in the
Spanish Main, Queen Elizabeth was able to pay off the whole of England's
foreign debt and finance the Levant Company, the predecessor of the East India
Company (Spate 1979:263). The line between legitimate trade (always
accompanied by ruthless violence) and buccaneering-whether private
enterprise, directly state-financed, or financed indirectly by the state through the
1046
grant of 'letters of marque'-was often hard to perceive. Such was the continuing
wealth of the Americas, though, that in the mid-seventeenth century, the
prostitute-filled silver centre ofPotosi could boast 14 dance-halls, 80 churches, 36
gaming-houses and seven or eight hundred professional gamblers. On his round-
the-world voyage of 1770-4, Anson captured only one large Spanish silver vessel.
But 32 wagon-loads of Spanish treasure were conveyed in triumphant procession
to the Tower of London (Spate 1983:256-65).
In the process, the indigenous population was decimated: partly worked to
death, but in the main succumbing to disease (see this volume, Article 11 ). The
population of Mexico declined from some 25 million in 1519 to 5.3 million in
1548 and 1.05 million by 1605; in Peru, from possibly 7 million to 1.8 million
by 1580.
The imperial connection also proved fatal for the Spanish economy,
intensifying the relative economic backwardness and the social ossification of
that country vis-a-vis its more dynamic northern neighbours. The abundance
of bullion inhibited investment in manufacturing industry and encouraged
costly wars of expansion. Eventually, the Spanish empire became chronically
bankrupt, the Spanish imperial system little more than a mechanism for
transferring the wealth of America to pay for the manufactured goods it had to
buy from northern Europe. By the end of the sixteenth century, only 3.8 per
cent of the goods carried to the New World in Spanish ships were products of
Spain (Spate 1983:335). Portugal became a client state of England.
W ere go an st ver were a sent or ecame wor e out, sugar ecame t e
major source of colonial profit. During the Crusades, the Christians had
become acquainted with sugar and with the technology which the Arabs had
developed to produce it. Arab production, and the industry which the
Portuguese and the Spaniards implanted on their new Atlantic island
possessions off the West African coast, had been based on a mixture of free
labour, indentured labour and slave labour. Slavery had not been the dominant
form of labour, nor-as the world 'slave', derived from the name 'Sclavus'
('Slav'), indicates-had slavery been confined to Africans. But in the New
World, production came to be organized entirely on the basis of plantations
worked by African slave labour (Mintz 1986:28-32). The sugar plantation was
an agro-industry, in which centralized discipline and a concern to achieve
maximum economy in the use of time constituted in many ways the prototype
for the factories of the subsequent Industrial Revolution (Patterson 1982).
England and France now came into head-on conflict for control of the sugar
trade in the Caribbean. By the end of the century, William Pitt the Younger
estimated that four-fifths of British overseas income derived from the West
Indies, while two-thirds of French external commerce was with one island,
Saint-Domingue (Haiti). Holland exchanged New York for the far more
important sugar-fields of Surinam, while France let Britain have Canada rather
than lose Guadeloupe.
1047
SOCIAL LIFE
Whichever power ran the plantations, new and larger supplies of slave
labour were needed. A whole continent, Africa, was converted into the major
source of supply. Thirty-six million people died without reaching the
Americas; perhaps twelve millions got there: together, nearly fifty million
human beings were transported. In the process, the indigenous economies of
Africa were destroyed; powerful kingdoms were broken or converted into
machines for capturing slaves, and new slave-raiding and slave-trading states
were brought into being.
FROM COLONIALISM TO IMPERIALISM

The decline of Spain and Portugal opened up a struggle between England,
France and Holland for global mastery. Once independent of Spain, the Dutch
turned their attention to supplanting their former imperial masters and the
Portuguese in colonial trade.
The location of the Spice Islands and the routes to them had been subjects
of the tightest security. The Portuguese had probed southwards along the cost
of West Africa and rounded the Cape of Good Hope in 1487. To protect the
slave trade, Manoel I, King of Kongo, had forbidden the inclusion in maps of
the route southwards to the Cape. But neither this nor subsequent Portuguese
attempts to keep these secrets to themselves succeeded. Da Gama was able to
reach Calicut, in India, thanks to an Arab pilot. (Drake's method of acquiring
navtgatwna mte tgence a een stmp e an uect: 1 nap a oca pt ot
(Spate 1983:298 n. 1, and eh. 9).) In 1595, van Linschoten, a Dutchman who
had lived in Goa, published the sailing instructions for the Cape route in his
Itinerario. Immediately, a Dutch fleet set out for the Indies. By the middle of
the seventeenth century, they had conquered Ceylon, captured Malacca, and
finally seized the ultimate prize: the Spice Islands themselves. They were also
implanted in Recife in Brazil and were raiding the west coast of South America.
Military conquest led to the replacement of trading-posts, or 'factories', by
colonies. Direct political power now allowed the Dutch to dominate the
production process itself For more than a century they enforced the most
rigorous control over the production of spices. The island of Banda, the only
source of nutmegs, was depopulated and the entire council of headmen
butchered in order to break indigenous resistance. In the Moluccas, where
cloves grew widely, native people were exterminated on every island except
Amboyna, where the guns of the colonial forts ensured that no one grew,
transported or possessed a single seedling of cloves (Greenberg and Ortiz
1983:17,20, 61).
Other European colonizing states did the same. The historic pattern of
mercantile competition now gave way to global struggle between these
European states for the direct and permanent conquest of colonies. The power
of the historic trading companies was replaced, step by step, with that of
representatives of the metropolitan state. By 1778, a new principle of bourgeois
1048
political economy was introduced even in the Spanish Empire: free trade
within and between the Spanish colonies, and with the metropolis. In 1790, the
Casa de Contrataci6n, in Seville, which had controlled trade with the Americas
since the Conquest, was abolished. The consequent increase in both the
production and the trade of the colonies resulted in a new realization that the
economic interests of the colonies were not necessarily identical with those of
Spain. It led, in other words, to the strengthening of a sense of nationalism.
In 1830, the Dutch replaced the system under which the Dutch East India
Company had managed trade with the Indies for over two centuries with a new
'Culture [Cultivation] System'. This established incentives designed to
stimulate peasant production for the market: those who produced export crops
on a fifth of their land had their taxes remitted. Java was soon transformed into
a 'mammoth state plantation' (Geertz 1963:53) for the production of coffee and
sugar; 'a whole people ... converted into a nation of... estate coolies, with their
own natural aristocracy reduced to the position of foremen and
superintendents' (Panikkar 1959:88).
Despite the measures taken by the Dutch to preserve their monopoly of
spices, they failed. In any case, spices and sugar were fast becoming less
important as the major sources of colonial wealth. The monopoly over the
plants themselves was broken by British 'botanic imperialism', as seedlings of
cocoa, tea and rubber plants, and of cinchona (for the production of quinine),
were smuggled, often by agents of the British, including diplomats, from the
East In tes, Brazt an Peru, an tssemmate rom Kew ar ens to new
colonial Botanic Gardens in Kingston Oamaica), Peredeniya (Ceylon) and
Raffles Gardens (Singapore), where they became the bases of new and
immensely profitable tropical agro-industries (Brockway 1979).
The struggle between Britain and France for control of North America and
India had left India as the jewel in Britain's crown. One major consequence of
the subsequent desperate attempt to throw off the British yoke, the 'Mutiny' of
1857, was the final abolition of the (by then) weakened British East India
Company, and its replacement by a regime of direct control of both the polity
and the economy by the colonial state.
Agriculture and industrial revolution in the West now led to a new pattern of
economic relations between the metropoles and the colonies. In India,
traditional industries, notably shipbuilding and textiles, were destroyed. In
their place, a new division of international labour arose: Indian agriculture
supplied the raw material for Lancashire's new cotton mills, whose products
were then exported back to India. The wealth extracted from the colonies thus
went to fuel the British agricultural and industrial revolutions. Liverpool, the
world's leading slave port, survived the ending of the trade by converting itself
into a centre of international commerce and industry.
There was also a revolution in consumption: a near-doubling of wages in
Western countries after the middle of the nineteenth century stimulated a mass
demand for tropical commodities like sugar and fruits which had once been
1049
SOCIAL LIFE
supreme luxuries, so valuable that sculptures in sugar were conspicuously

displayed on royal banqueting tables in the Middle Ages (Mintz 1986: eh. 3). In
1815--44, the average Briton still consumed less than 20 pounds of sugar per year;
by the 1890s, this had risen to between 80 and 90 pounds (Hobsbawm 1969:74).
THE CONSOLIDATION OF GLOBAL IMPERIALISM

India became the springboard for Britain's onslaught upon the most populous
country on earth, China. When Lord Macartney had proposed the opening of
China to foreign trade and to Christian missionizing in 1793, the response from
the Chinese Emperor had been one of polite incredulity. His respectful spirit of
submission was appreciated, Macartney was told, but China had no need of the
manufactures of 'barbarians'. Plainly, however, they needed Chinese products,
so would be allowed to establish small trading-posts on the coast, under strict
supervision. The notion that they might proselytize Christianity, however, was
dismissed as 'utterly unreasonable'.
One commodity introduced by the Europeans from India soon outstripped
all others: opium. Within a few decades, millions of Chinese had been turned
into drug addicts and in two Opium Wars all barriers to trade were destroyed.
A joint British-French force took advantage of the Taiping Rebellion (1848-64)
to impose its will on the enfeebled imperial government, burning the Summer
Palace and opening the country's trade to foreigners. By the end of the century,
ma a een IVI e mto Bntts , Frenc , Japanese an
influence'.
The final act in the establishment of modern imperialist control over
virtually the entire world came with the Berlin African Conference of 1884--5,
when Africa was divided between a handful of industrialized European powers.
The establishment of virtually global European rule depended not just on
technological superiority in general, but on one specific kind of production:
armaments. The technological edge in the Spanish conquest of South
America-armour, swords, muskets, horses and dogs-had not been very
great. But by the nineteenth century, European industry provided its armed
forces with new weapons of terrible destructive power. Nevertheless, people
continued to resist. In early clashes in New South Wales, between 2,000 and
2,500 settlers were killed by Aborigines armed only with spears; however,
settlers and the military wiped out upwards of 20,000 Aborigines (Hughes
1988:277). In well-organized empires and states with large armies, resistance
was more effective: the Ashanti wiped out a British army in 1824. So did the
Mahdi in the Sudan and the Zulu in 1879, while Abyssinian forces destroyed an
Italian army as late as 1896, at Adowa.
THE COLONIAL POLITY

European conquest also depended, as in the Americas, on using local forces. In
1050
West Africa, for instance, an army of 1,200 men, most of whom were Africans,
had defeated 30,000 of their enemies at Sokoto. Colonial troops from countries
outside Africa were also used.
The consolidation of military victory entailed the construction of new states
that were entirely subordinate to the mother state back in Europe.
Administration was designed to cost as little as possible. Sir Harry Johnston
governed Nyasaland with his own salary plus £10,000 a year, one British officer
and 75 Indian soldiers. Lord Lugard had an annual budget of just over
£100,000, five European administrators and one African regiment to govern
ten million people. Hence administration necessarily depended on eo-opting
indigenous political authorities and dividing any possible indigenous
opposition. The Dutch in the East Indies, likewise, governed with only a small
European administrative staff.
'Divide and rule' involved more than the elimination of any potentially
threatening physical force that might have remained in the hands of others. In
India, the British organized their army recruitment on the basis of obsessional
divisions of the population not only into castes and subcastes, but even into
sub-subcastes, in their racist search for uncontaminated 'martial' stock (Mason
1974:350-61).
In post-Mutiny India, a cultural offensive was launched to persuade the
conquered that their future lay in joining the British in building a new imperial
order. The Queen now became monarch of both Britain and India, and in 1877
s e was resty e Empress o In ta. In tan pnnces an nota es were won over
not just by showering them with material rewards, but by the award of honours
and an elaborate series of durbars in which an act of incorporation was the
central ritual.
New 'traditions' were invented to incorporate and divide India's old
aristocrats and new civil servants. Competition and division between the
princes was instilled by creating fine distinctions according to their new
positions in the imperial hierarchy: distinctions of title; of clothing and
uniform; in the numbers of retainers and soldiers that princes were allowed,
and so on. A whole array of new orders, escutcheons, armorial bearings, robes,
banners, etc. was created-a bizarre iconic mix of 'Victorian feudal', Mughal,
Hindu, imperial Roman, Sikh and Rajput elements (Cohn 1983:165-209).
'Indirect rule' was much older than Lugard's subsequent formulation of the
idea; it had been used for centuries in territories where populations were
numbered in tens, even hundreds of millions, and it continued to the end of the
colonial epoch. On the eve of the Second World War the Dutch East Indies
were divided into directly administered areas and areas of indirect rule with
269 'native states'; India's constitution was similarly heavily weighted in favour
of the princely states.
The principal task of the colonial authorities in India was the collection of
taxes to pay for the costs of administration, as the title of the Indian
administrative official-the 'Collector'-indicated. In the sphere of
1051
SOCIAL LIFE
production, the promotion of capitalism was the major economic priority. The
pioneer transformation of colonial land-holding and taxation systems was the
Permanent Settlement of 1793 in Bengal, whereby 3,000 zamindars and
jaghirdars who, until then, had possessed rights over labour and the products of
that labour on lands granted to them by the Moghuls, were made absolute
owners of the land-which they had never been before. Their loyalty was
assured by allocating to them one-tenth of the taxes collected.
The same basic principles informed policy a century later, in a quite
different kind of colony. In Kenya, a white settler colony, Africans were forced
to become wage-labourers on lands allocated to Europeans. Even so, in 1914
more than 70 per cent of exports were still coming from African peasant
smallholdings. European farmers were now given a vast range of government
services-railways, roads, schools, hospitals, extension services, etc.-together
with subsidies built into the customs tariffs. Africans were forced to pay head
and hut taxes, in cash, and each individual had to carry a kipande which
recorded their tax payments and labour-history, and which had to be presented
on demand by employers and officials. A Masters and Servants Ordinance
bound the African to serve out a contract on pain of imprisonment. For those
Africans who stayed on lands allotted to white settlers, the Resident Labourers
Ordinance permitted them a small subsistence plot, on condition that they put
in 180 days of work a year for the settler-owner. By 1920, more than half of the
men of the Kikuyu and Luo, the largest agricultural tribes, were working for
Europeans. T ese economic measures were rem orce y a co our- ar
excluding Africans from legislative and other public bodies and prohibiting
African trade unions, together with a whole social apartheid of separate
schools, separate residential areas, exclusive access for whites to hotels and
recreational facilities, and 'whites only' seats in buses and public places (Leys
1975:30-4).
CULTURAL HEGEMONY AND CULTURAL

RESISTANCE
Political and economic domination was reinforced by the dissemination of
values and institutions designed to promote the acceptance, by the colonized,
of their place in the colonial order. To the European colonizers, the superiority
of their culture was a total superiority: not just one of technology and
productive systems, but also of ideas and values. If material pre-eminence was
based on modern science, the spiritual superiority of European culture over all
forms of indigenous culture, including even the religions of Asia, was equally
unquestioned.
Missionaries of all denominations now flocked to the colonies. In inter-war
Papua, 15 per cent of Europeans were missionaries belonging to eleven
different missionary societies. The colonial state was so thin on the ground that
the missions often carried out functions which would elsewhere be the business
1052
of the state. Schools run by missions, not by the state, were the main vehicles
for the dissemination of European culture. But even where paganism was
stamped out, as in Latin America, and Christianity became the religion of the
people, it was still informed by indigenous ideas (Wachtel 1971 ). The
quintessence of European culture, it seemed, was religious rather than secular.
In a society where all positions of power and wealth were monopolized by
whites, the missions were often the only available avenues of social mobility
open to the more enterprising and ambitious individuals. Some-whom Asians
called 'rice Christians'-'converted' in order to learn to read and write, or
because the missionaries provided them with free health services or food in
times of scarcity. Innovators and entrepreneurs seeking to carve out a place in
commerce and market agriculture found the Protestant ethic as attractive as did
their European predecessors during the Reformation (Long 1968).
For those who resented white authority, interpretations of Christian
doctrine which emphasized fraternity, hope and charity, and the righteousness
of the meek and the humble as against the arrogance of the mighty, provided a
'critical' ideology of social dissent. The more radical found Biblical authority
for deviant, even apocalyptic ideas, or developed syncretic mixtures of
Christianity and indigenous belief, organizing their followers into new
churches independent of white missionaries. In South Africa, where blacks
were kept out of even Christian churches, the formation of their own, 'Zionist'
and 'Ethiopian' separatist churches was one of the principal outlets for the
mte tgent an t e am Itwus. T ese c urc es were a so extraor man y
fissiparous, since would-be leaders constantly broke away from the parent body
to found their own sects (Sundkler 1948).
A tiny minority, normally sons of the aristocracy, went on to higher levels of
European education designed to fit them for positions of responsibility in
systems of indirect rule, or, in the economic field, as supervisors on estates or as
managers in urban business. But for the vast majority without capital, there was
little hope of rising above the level of the small farm or the small shop, and even
these niches were often occupied by people from immigrant business cultures,
such as Ismaili Muslims in East Africa (Morris 1968).
An even tinier minority had access to literature which was critical of
European society or informed them about the values and institutions of their
own pre-coloniallegacies. Given their socially privileged backgrounds, most of
them were not disposed to respond to such ideas.
But eventually, secularism, liberalism and nationalism did filter through:
first the Enlightenment ideas of Rousseau, Locke and Voltaire, and later the
positivism of Comte and Saint-Simon and the liberalism of Mill and Spencer.
The classical cultures of Mediterranean antiquity had inspired the thinkers of
the Enlightenment in Europe; a century later, they still inspired pioneer
Egyptian nationalists like Ahmad Lutfi al-Sayyid. Others, like Ram Mohan
Roy in India, struggled to modernize their own cultural traditions by
combining elements of Hinduism and of Western thought. Early nationalists
1053
SOCIAL LIFE
were also naturally inspired not only by philosophers, but also by their
counterparts in Europe: liberal, positivist, radical, revolutionary, and Utopian
political activists from Mazzini and Cavour to Tolstoy, Kossuth and Parnell.
Today, at a time when it is uncritically assumed that what is labelled
'fundamentalism' (in fact, modern interpretations of Muslim belief) is the
authentic and immanent essence of Islam, it is worth remembering that
nationalists in Turkey, for instance, had been predominantly secularist since as
far back as the epoch of the Tanzimat (1839) right through into the period when
Ataturk abolished the Caliphate in 1924 (Zubaida 1989).
But a religious heritage going back thousands of years and deeply imbricated
in the institutions of everyday life, especially in rural areas, was not to be
overthrown by secularist modernizers, even less by a few European
missionaries. In particular, the great religions rendered believers impervious to
the message of Christianity. The priority of the colonizing state, however, was
not the saving of souls but the exploitation of the colonies in the interests of the
Motherland. Whatever the degree of cultural resistance and persistence,
therefore, the colonial impact could not be prevented from transforming
secular life.
ECONOMIC TRANSFORMATION
The colonial powers crushed early attempts by non-European states to develop
mo ern Western-sty e m ustnes, especta y armaments m ustnes. Peno tc
requests from Ethiopia, from 1520 through to the nineteenth century, for the
technology with which to manufacture European-style swords, muskets,
textiles and books, were refused. Other African projects-to import foreign
tailors, smiths and carpenters into Dahomey in the 1720s; to develop cotton
production among the Fante; to establish sugar refineries in Calabar-were all
blocked. In Egypt, Mohammed Ali was more successful in developing the
cultivation and processing of cotton, and he used the profits to set up state
factories for the manufacture of cotton, woollen, silk and linen textiles, as well
as sugar, paper, glass, leather, sulphuric acid, and guns and gunpowder.
Palmerston thereupon invaded Egypt and imposed a 'capitulation' treaty under
which Egypt's internal trade was opened to foreigners, the state monopolies
were abolished, and Egyptian finances were plundered (Stavrianos 1981:118-
19, 215-16).
Even after the abolition of the slave trade in the British Empire-as a result
not only of the growth of Abolitionist sentiment in Europe and the growing
costs of slavery, but also of the major armed slave rebellions in Saint Domingue
and Jamaica (Blackburn 1988)-slavery continued to expand in the USA,
Brazil and Cuba. Nor did the abolition of slavery mean the end of the
plantation system. Rather, it became the main method of organizing the
production of tea, coffee, sisal and rubber, involving the transporting of large
populations, often overseas, to plantations where the intensity of work and the
1054
methods of control over the workforce amounted to instances of what Goffman

has called 'total institutions' (Goffman 1968). The system of indentured labour
in Burma has been described as a 'new system of slavery' (Tinker 1974 ). In the
South Pacific, nineteenth-century 'blackbirders' recruited islanders-often by
force or deceit-for labour in the Queensland or Fiji plantations, whence as
many as 750 out of every thousand failed to return (Belshaw 1954:39-40).
Large-scale agro-industry was not the only mode of capitalist production,
however. The purchase of peasant produce by large trading companies was an
alternative way of securing the volume of raw material needed by modern
industry in the West. What was destined to become one of the largest of
modern multinational corporations, Unilever, built its fortunes, initially, upon
the basis of the West African peasant production of palm oil. In 1810, West
Africa exported a mere 1,000 tons of oil; by 1860, it was exporting up to 50,000
tons annually. The peasant was now intimately affected by the ups and downs of
world prices.
In the first major continent to be colonized, another productive system had
flourished for centuries: the large estate or hacienda, based upon the use of tied
Indian labour, often living in communities bordering on the large estates, or,
like the 'squatters' of Kenya, living off the produce of small plots of land
belonging to the hacienda in return for so many days' obligatory labour on the
estate. Alternatively, rent might be paid to landlords in cash or in kind. 'Free'
labourers, without any land on which to produce their families' subsistence,
cou on y se t etr a our-power. W atever t e a our-regtme, t e ups ot was
the same: dependence and debt which could last beyond a person's life-time.
Even in the twentieth century a child in the Andes could be born inheriting its
father's debts to the landlord (Redclift 1978, Zamosc 1986). Yet because
peasant ideology was based on norms of reciprocity which included the
legitimacy of obligations to the landlord, and which distinguished good
landlords from bad, and because the expectations of peasants rarely aspired
beyond ensuring a bare subsistence, they did not become revolutionaries (Scott
1976).
The new imperialism also transformed the city. In China, pre-capitalist
cities had often been large centres of craft production centuries before the
'urban revolution' of the eleventh century AD. By the thirteenth century,
Hang-chou, which Marco Polo described as 'without doubt the finest and most
splendid city in the world', probably had between five and seven million
inhabitants and was forty miles in circumference. Yet Chinese cities did not
play the same historic role as their smaller medieval counterparts in Europe.
The towns were places where rural produce was marketed, and merchants
lacked the privileges, and craftsmen the guilds, which provided their European
contemporaries with a crucial measure of protection from the whims of kings.
Imperial power remained unchallenged for millennia: there was no bourgeois
revolution. It was the market towns which grew in the eighteenth and
nineteenth centuries, not the cities. By 1900, only about 4 per cent of the
1055
SOCIAL LIFE
Chinese population lived in cities of 100,000 inhabitants or more; less than in

the thirteenth century (Elvin 1973: 175-8).
But once connected to the world market, cities, some old but most quite new,
began to expand rapidly, usually in coastal locations like Bombay and Calcutta,
Lima, Valparaiso or Buenos Aires, Canton and Shanghai. As they grew, they
sucked in huge supplies of labour from the rural hinterland or, if that supply
was lacking, from further afield. Regions with vast populations, like India and
China, supplied less-populated but growing countries around the Pacific rim-
Ceylon, Fiji, Malaya, Hawaii, California, Australia-and the West Indies with
labour. After the abolition of slavery in Brazil in 1885, immigrant labour
flooded in to replace the slaves on the coffee plantations. The growth of a city
like Buenos Aires, based on the export of wheat, wool and sugar, and later the
centre of a world trade in meat, generated the rapid expansion of construction,
service and other industries such as the railways and the docks which serviced
the export trade. Later, the city became an industrial centre supplying the
growing internal market, and also producing for export to adjoining regions. By
1914, Buenos Aires had one and a half million inhabitants, a fifth of the total
national population. With only a thinly-populated rural interior, the new
labour force was predominantly an immigrant one: three out of four
inhabitants of the city were born abroad (Roberts 1978:49-56).
The rise of industry in the West led to a vastly expanded demand for new
kinds of raw materials from the colonial world, such as petroleum. The expansion
o wor commerce, too, ea e or new supp tes o go on an unprece ente
scale. The whole of southern Africa now became a vast reservoir of labour
servicing the mines of the Rand and the Copperbelt. The African village,
denuded of its menfolk, was inhabited by children, the elderly, and by women
upon whom the responsibility now fell for working the land, bringing up the
children, and taking care of the old people. Consumption increasingly included
purchased imports of foreign manufacture, from kerosene lamps and bicycles to
medicines and clothing. To pay for these, cash crops had to be cultivated
alongside subsistence produce. But the major source of cash with which to satisfy
these new wants consisted in the remittances sent by men from the mining areas.
Rather then constituting a 'dual' economy, then, mine and village were by now
integral parts of a single economic network (van Onselen 1976).
Most of the wealth from the mines did not go to Africans at all: only a third
of the value of the output from the mines of Northern Rhodesia, for instance,
even stayed in the country; of this, European mineworkers, who monopolized
the skilled and supervisory jobs, received twice as much as a member of the
much larger African labour force (Epstein 1958). But though African miners
only earned an average of around £40 a year, this was far superior to wages on
European-owned farms or what could be earned by producing cash crops.
The internal market in a country of this size was limited by the earning
power and by the size of the population. In countries with much larger
populations, not only was the internal market of a far greater scale, but there
1056
were also indigenous classes in possession of wealth which could now be

invested not just in commerce or light industry but eventually, in the case of
India, in modern heavy industry and manufacturing. Thus the coal mines and
iron works of Jamshedpur in Orissa were built up, not by British capitalists, but
by the Tata dynasty. Modern industry therefore generated two new classes
which were to play important parts in ending British imperial rule: an
indigenous bourgeoisie and an urban working class. The main financial support
for the Indian Congress party was to come from the Tatas.
THE IDEA OF THE NATION

Modern conceptions of the nation, and the linking of the nation to the state,
were novel ideas, pioneered principally by Herder and Kant (see this volume,
Article 25). Before the rise of the absolutist state, individuals had identities
ascribed to them, horizontally, as belonging to a certain rank: in feudal Europe,
as members of an estate, as nobles or commoners; in Hindu India, as members
of a caste. Relationships to the state were mediated by vertical ties to superior
groupings in a hierarchy of dependence and authority; in medieval Europe,
through ties to one's lord (Worsley 1984:252 f£ ).
Absolutism meant the concentration of internal sovereignty in the hands of
the monarch by breaking the power of the magnates and by refusing to
recognize claims to universalistic Papal authority over secular monarchs.
In t e su sequent ourgems revo utwns o t e seventeent century, t e
rights of lower-level corporate groups, especially the propertied, were
entrenched. There had been too much sovereignty; now what was needed were
the checks and balances of constitutional government. By the time of the
French Revolution and the first successful revolt against European colonialism,
namely the American War of Independence, civic rights had been widened and
defined as the Rights of Man-not just the propertied. The unit of civil society
was the individual, not the group. Corporate interests intermediate between the
individual and the public weal were a constraint on freedom. The interests of
society were to be decided by the general will: the aggregate of the individual
wills of all citizens. But for these choices to be based on reason, education was
needed. Under Napoleon's corporatist version of equality of education, the
ideal was itatiste uniformity: every French school-child would turn over the
same page in the same authorized textbook at the same time on the same day of
the year.
Equality, the atomic relationship to the state, and the supersession of older
identities of ethnic group and nationality, contained no collective element,
however. The gap was filled by formulating a new, supplementary kind of
identity: not just that of citizen (of the state), but a national identity-that of
Frenchman.
But when these ideas were transferred to the colonies, national identity was
an attribute which set off the indigenous population from their foreign rulers.
1057
SOCIAL LIFE
Members of the indigenous upper classes who collaborated with the

colonialists, by serving as administrators, were dubbed traitors; in the
economic field, indigenous intermediaries in the export trade with the
metropoles were branded pejoratively, in China, as 'compradores'.
Initial resistance to European penetration drew upon both indigenous and
new foreign sources in developing analyses of imperialism and in devising
programmes for creating a more modern and improved society. The ideology of
the biggest revolution of the nineteenth century, the Taiping 'Rebellion' in
China, which lasted nearly two decades and in which between twenty and
thirty millions died, was not exclusively Chinese in inspiration: many of its
ideas and social ideals had been borrowed from Protestant missionaries in
Canton. But they were combined with radical social ideals and modes of
organization taken from the perennial secret societies (Chesneaux 1971, 1973),
and with beliefs that had flourished for centuries in unorthodox schools of
Mahayana Buddhism. The Rebellion had been a revolt, in the first place, not
against Western imperialism but against the rule of the foreign Manchus. But
the support given to the imperial counter-revolution by the Western powers,
which was decisive in crushing the revolutionary regime, ensured that the next
major revolt, the 'Boxer Rebellion', would be virulently anti-European.
In Latin America likewise, the first movements of resistance against colonial
rule drew heavily on indigenous sources. Despite enforced Christianization,
the Indian heritage was still strong. Messianic and millenarian movements
occurre t roug out t e co oma peno , rom t e n es to Yucatan,
culminating in the revolt, under Tupac Amaru in 1780-83, to recover Indian
land and Indian cultural identity. The 'American' nationalism that
subsequently developed, largely among mestizo strata, during the struggle for
independence, was a quite different phenomenon.
By 1848, liberal nationalism had been crushed in Europe, especially in
Austria-Hungary and Russia. Thereafter, new and much more radical creeds,
notably socialism and eventually Marxism, were to challenge liberalism and
give rise to new kinds of mass movements and organized parties. From then on,
the removal of foreign rule and the removal of their own indigenous ruling
classes were two sides of the same coin for social revolutionaries.
The first successful revolts against colonial rule occurred in the Americas in
the eighteenth century: the establishment of the 'First New Nation' in the
thirteen colonies (Lip set 1964 ), followed by the winning of independence by
Haiti despite savage British and French repression (Blackburn 1988). But a
revolution of black slaves was not to be tolerated, even if it could not be
undone. Haiti was therefore strangled economically and was totally isolated
politically by the outside world Games 1963). Whatever happened in Europe,
the principles of the French Revolution were not to be applied anywhere in the
French or Dutch colonies. Thus in South Africa, where two small towns,
Swellendam and Graff-Reinet, declared themselves independent, liberty,
equality and fraternity were to apply only to whites (De Kiewiet 1941 ).
1058
The compliance of the colonized could often be ensured without recourse to

direct intervention. Britain controlled South America mainly by economic and
financial pressures or by using surrogates to crush troublesome regimes. But
Britain's successor converted the Monroe Doctrine into a new wave of
interventionism in the Teddy Roosevelt epoch: revolt against Spain in the
Philippines was hijacked, and the much bloodier revolt against both Spain and
slavery in Cuba ended with de focto American domination of the new states.
Both were policed by a strong US military presence. Thenceforth, the United
States intervened regularly in the 'banana republics' of the Caribbean. The
Mexican Revolution of 191 0-in which one in three of the population of the
state of Morelos died-was not repeated in neighbouring countries; and even
here the edge of peasant discontent was blunted by the ejido land-reform, while
capitalism took off in the cities (Womack 1969).
THE GROWTH OF NATIONALISM

1885, the year in which the great European powers dismembered Africa, had
seemed to Lenin, in his study of imperialism (Lenin 1915), an appropriate date
from which to reckon the consolidation of the new world system. True, Latin
America had been politically independent, with the exception of one or two
small countries, for most of the nineteenth century, and a few countries had
never fallen into European hands at all: Siam, Afghanistan, China, Persia and
Tur ey, an yssmta untt t e 9 Os . T eu orma po 1t1ca autonomy,
however, deceived no one, least of all their indigenous populations, about their
true status: that of 'semi-colonies'. By 1914, China, for instance, had been
divided into 'spheres of influence' of no less than fourteen foreign powers,
based on 'concessions' in the modern cities on the coast.
At the time of the First World War, nationalist movements in Asia were too
weak to take advantage of the rivalries between the imperial powers. Rather it
was a new, Asian imperialism, that of Japan, which snapped up colonial
territories. Those that had belonged to Germany were everywhere simply
transferred to the victors.
But within three years of the outbreak of the First World War, a new threat
to world capitalism had emerged, with the victory of the Bolsheviks in Russia.
Nationalists had long made common cause both with their brothers in other
colonial countries and with those in the metropoles themselves-liberals and
socialists for the most part-who opposed the imperialist policies of their own
governments. But for those who saw the struggle for national independence as
entailing social revolution as well, the USSR presented a new and impressive
model. The prestige of that revolution, too, ensured that Lenin's theory of
imperialism quickly eclipsed both J.A.Hobson's liberal analysis and
Hilferding's earlier Austro-Marxist one, and became the dominant global
revolutionary theory. Yet the ideological inspiration provided by the infant
USSR was not matched by the provision of material assistance to the tiny
1059
SOCIAL LIFE
groups of communists in the colonies. Even in China, where the communists

had emerged as a significant armed force, Soviet assistance was limited to a few
military advisors. The Chinese communists were wiped out in the cities and
forced to retreat to a remote rural base in Yenan, where they developed their
revolution on the basis not of the proletariat, but of the peasantry.
The experience of China, however, was different from that of other major
Asian colonized countries. Between the wars, nationalist movements arose
practically everywhere in Asia. But in India, the largest country after China, the
rapid growth of Congress did not eventuate in armed struggle. Elsewhere, the
colonial powers remained entrenched for the entire inter-war period. In
Southern Africa, white power was in the saddle; in East and West Africa
nationalist movements only emerged on the eve of the Second World War.
When pressed, colonial governments displayed consummate skill in fobbing off
nationalist pressure by interminable discussions about gradual instalments of
self-government and the necessity of ensuring the slow internalization of the
key values-accountability, impartiality, and so on-of Western civic culture.
It was not until the whites were actually defeated in Asia, by Japan in the
Second World War, and the subsequent defeat of the Japanese themselves by
the Allies, that the opportunity arose for nationalists to seize power themselves
through armed revolution, notably in Indonesia.
The Western powers were able to crush communist-led revolution in
Malaya, to contain the communists in Indo-China, and to restore themselves in
Hong Kong. But m ma t ey met wtt a wor - tstoriC e eat rom w tc a
that could be rescued was the quarantining of the new revolution, and the
separation off ofTaiwan-into which country foreign capital, mainly Japanese,
flowed on such a scale that the island grew to be a far larger industrial power
than mainland China itself. Korea was divided between a capitalist South and a
North that achieved Soviet-style modernization of agriculture and industry,
but which was faced, by the 1990s, with the same problems that undermined
the economy of the Soviet Union. In India, the transfer of power took place
peacefully before the continent could explode. With the 'Green Revolution' in
agriculture, and the creation of a large industrial sector, India was to become a
'regional superpower' in Asia. In most of Asia, then, the survival of capitalism
was ensured. The second major defeat of the West-the triumph of a nation of
peasants, Vietnam, over the greatest military power in world history-was
followed by a period of stagnation, in which the organizational structures which
had won the war proved incapable of developing the peacetime economy.
NATION-BUILDING
In 1960 alone, seventeen new African countries appeared on the world scene.
Most of them were colonial constructs, entities such as 'Nigeria' or 'Kenya',
much less than a century old. Within (and sometimes across) their boundaries,
people commonly identified much more strongly with their ethnic group-an
1060
identification reinforced by colonial policies of using the 'tribe' as the favoured

unit of administration-than with their class or nation. The priorities of the
new rulers who were catapulted into power were therefore to divest their
society and their culture of European influence, and to build a new state based
upon a new national identity, 'Tribalism' was henceforth seen as a sin, as were
any other identities or groupings which threatened to divide the nation. Thus,
according to the new populist ideologies of 'African socialism', class struggle
was a foreign ideology which had no place in the new Africa. Classes were
colonial phenomena and had never been part of the authentic pre-colonial past
(Worsley 1964: eh. 4, Ionescu and Gellner 1969). Given the power of the state,
and despite attempts by outside powers to foment divisions, nation-building
proved effective enough in most countries-especially where there were a large
number of different ethnic groups-to avert the kinds of internal
confrontations which, in Nigeria, resulted in the attempted secession of Biafra
and civil war.
The movements which came to power with great popular support embarked
upon programmes of nationalization. Politically, nationalization meant placing
nationals in all the positions of state legislative and administrative power, both
central and local. In Europe, 'nationalization' had meant state-ownership, and
indeed 'parastatals' were also created in the new states, notably the marketing
boards which monopolized the purchase of peasant cash-crop produce and
then sold it on the world market, making a profit for the state. New oil
m ustnes were common y eve ope un er state contro , an m some
countries older mining industries too. But foreign companies operating outside
the field of mineral exploitation were normally left untouched, though they
were pressured to appoint African directors to their boards. The state also
provided generous assistance to would-be indigenous entrepreneurs. In the
process, two new classes were created: a political 'new class' which controlled
the state, including the parastatals, and a new national bourgeoisie in the
private sector, brought into being through the provision of capital (and
sometimes ofland) by the state and society.
The dominant parties proceeded, step by step, to concentrate power in their
hands, by destroying rival parties and bringing the whole of civil society under
their control. The logical end-product was the single-party state. Once that had
been brought into being, it was open to whosoever could mobilize enough
power to take over the state and society. Increasingly, that meant not those who
could mobilize votes, but those who wielded military power. Within two
decades, military coups had taken place in a majority of black African states.
Outside powers--often the former colonial rulers-were able to manipulate
this situation to restore de facto economic control, and even at times to
intervene militarily.
Whatever the nationalist rhetoric and intent, key economic resources often
remained in foreign hands. Massive borrowing by the new states also increased
their vulnerability to outside economic control. So long as prices for their
1061
SOCIAL LIFE
primary goods remained buoyant, as they did in the 1960s and 1970s, the
problem of repaying the debt was deferred. But when export prices fell,
increased output often proved insufficient to service interest payments, let
alone to pay off the capital. Even those countries which had aimed at self-
reliance found that they were at the mercy of large corporations, since the latter
could charge high prices for manufactured goods. On the other hand, demand
for Third World commodities was inelastic, or could be undercut by
competitors, or even eliminated by substituting man-made materials. More
importantly, the 'impersonal' power of the market was such that it was
controlled, collectively, by the giant corporations which produced what the
Third World needed and which purchased Third World commodities.
In 1974, it seemed that there was one major exception to all this, when the
OPEC oil cartel raised its prices to the outside world. This looked like a model
which could be applied across the board: all that was needed was for the
producers to act in concert, and the West could be held to ransom. Even with
such a strategic commodity as oil, however, the panic created resulted in the
rapid development of alternative modes of energy use in the West, and to a
consequent reduction in world demand for petroleum relative to its increased
production. The economies of non-oil states in the Third World itself also
suffered severely. And when attempts were made to create a banana producers'
cartel, for example, they failed: in part because the West did not need bananas
as much as it needed oil; in part because it was easily able to break the solidarity
o t e supp ters.
Before the 1950s, there had been no such entity as the 'Third World'. But
the common interests of the new ex-colonies increasingly brought them
together, firstly in a series of regional conferences in Asia and Africa, then in
Afro-Asian conferences, of which the Bandung Conference of 1955 was the
most important, and culminating in the establishment of the Non-Aligned
Movement. Initially, its major preoccupations were with political
decolonization, both domestically and in the remaining colonies, and-as the
name of the Movement indicates-with the attempt, in the epoch of the Cold
War, to create a global grouping which would not itself be a bloc, but would be
independent of both superpowers. But by the 1980s, the major problems had
turned out to be economic ones: unequal terms of trade on the world market,
and the rising tide of debt. It was Third World pressure which forced the
United Nations to establish the UNCTAD conferences on trade and
development.
The Third World also found that it had to deal not just with the political
power of foreign states, but also with the economic power of giant multinational
corporations, which were now organized on the basis of a global division of
labour (Henderson 1989). Within this 'new international division of labour',
even some Third World countries-notably the 'four little tigers' (South
Korea, Hong Kong, Taiwan and Singapore), Mexico, Brazil and other states-
1062
became 'newly industrializing countries'. By 1990, the majority of the world's

population was living in towns and cities.
Yet in the 1980s, black Africa had slipped backwards in terms of production
while its debts increased. Countries like India and Brazil still contain huge
rural sectors comprising tens of millions of people mired in agrarian poverty.
The evolutionist notion, based on the experience of the minority of 'newly
industrializing' countries, that the Third World is 'disappearing' (Harris 1987)
therefore seems premature; indeed, those countries where famine is rife appear
to be 'underdeveloping'. The older Marxist evolutionary notion, still strongly
held in some parts of the Third World, that communism is the 'wave of the
future', seems equally destined to disappear as the collapse of the Soviet bloc
begins to have its impact on communist regimes in the Third World.
Academic debate about the causes of underdevelopment, and the search for
international remedies, continues. But it would be naive to neglect popular
conceptions of the Third World; these include racist views that people in poor
countries are simply lacking in intellectual ability. Others attribute their
backwardness to cultural beliefs and institutions, such as fatalistic 'other-
worldly' religions; while yet others see 'traditional', 'pre-modern' social
structures as the problem: aid from the developed world is considered a waste
of resources, because corrupt rulers will simply pocket it or use it inefficiently.
Finally, the whole notion of 'development', along with the assumption that
Western production-systems and patterns of consumption are models to be
tmttate , as come mto mcreasmg questiOn m t
(Redclift 1984 ).
REFERENCES
Abu-Lughod, J.L. (1971) Cairo: 1001 Years of the City Victorious, Princeton, Princeton
University Press.
Anderson, P. (1973) Lineages of the Absolutist State, London: New Left Books.
Belshaw, C.S. (1954) Changing Melanesia, Melbourne: Oxford University Press.
Bendix, R. (1960) Max Weber: an Intellectual Portrait, New York: Doubleday.
Blackburn, R. (1988) The Overthrow of Colonial Slavery, 1776-1848, London: Verso.
Brockway, L.H. (1979) Science and Colonial Expansion: the Role of the British Royal
Botanic Gardens, New York: Academic Press.
Chesneaux, J. (1971) Secret Societies in China in the 19th and 20th Centuries, London:
Heinemann.
--(1973) Peasant Revolts in China 1840-1949, London: Thames & Hudson.
Chi Ch'ao-ting (1936) Key Economic Areas in Chinese History as Revealed in the
Development of Public Works for Water-Control, London: Alien & Unwin.
Cohn, B.S. (1983) 'Representing authority in Victorian India', in E.Hobsbawm and
T.Ranger (eds), The Invention of Tradition, Cambridge: Cambridge University
Press.
De Kiewiet, C.W. (1941) A History of South Africa-Social and Economic, Oxford:
Oxford University Press.
1063
SOCIAL LIFE
Elvin, M. (1973) The Pattern of the Chinese Past, London: Eyre Methuen.
Epstein, A.L. (1958) Politics in an Urban African Community, Manchester: Manchester
University Press.
Evans-Pritchard, E.E. (1937) The Nuer: a Description of the Modes of Livelihood and
Political Institutions of a Nilotic People, Oxford: Clarendon Press.
Fuller, C.J. (1984) Servants of the Goddess: the Priests of a South Indian Temple,
Geertz, C. (1963) Agricultural Involution: the Process of Ecological Change in Indonesia,
Berkeley, University of California Press.
Geiger, W (1986) The Mahava?sa or the Great Chronicles of Ceylon, New Delhi, Asian
Educational Services.
Gellner, E. (1983) Nations and Nationalism, Oxford: Blackwell.
Goffman, E. (1968) Asylums: Essays on the Social Situation of Mental Patients and Other
Inmates, Harmondsworth: Penguin.
Greenberg, S. and Ortiz, E.L. (1983) The Spice of Life, London: MichaelJoseph.
Gunawardana, R.A.L.H. (1979) Robe and Plough: Monasticism and Economic Interest in
Early Medieval Sri Lanka, Tucson, Arizona: Association for Asian Studies.
Harris, N. (1987) The End of the Third World: Newly Industrializing Countries and the
Decline of an Ideology, Harmondsworth: Penguin.
Henderson, J. (1989) The Globalization of High Technology Production: Society, Space
and Semi-conductors in the Restructuring of the Modern World, London: Routledge.
Hobsbawm, E.J. (1969) Industry and Empire, Harmondsworth: Penguin.
Hughes, R. (1988) The Fatal Shore: a History of the Transportation of Convicts to
Australia 1787-1868, London: Pan Books.
onescu, . an e ner, . e s opu tsm, ts
Characteristics, London: Weidenfeld & Nicolson.
James, C.L.R. (1963) The Black Jacobins: Toussaint L 'Ouverture and the San Domingo
Revolution, New York: Vintage Books.
Kehoe, A.B. (1981) North American Indians: a Comprehensive Account, Englewood
Cliffs, NJ: Prentice-Hall.
Lenin, V.I. (1915) Imperialism: the Highest Stage of Capitalism (many editions),
van Leur, J.C. (1955) Indonesian Trade and Society: Essays in Social and Economic
History, The Hague: van Hoeve.
Leys, C. (1975) Underdevelopment in Kenya: the Political Economy of Neo-colonialism,
1964-1971, London: Heinemann.
Lipset, S.M. (1964) The First New Nation: the United States in Historical and
Comparative Perspective, London: Heinemann.
Long, N. (1968) Social Change and the Individual: a Study of the Social and Religious
Responses to Innovation in a Zambian Rural Community, Manchester: Manchester
University Press.
Mason, P. (1974) A Matter of Honour: an Account of the Indian Army, its Officers and
Men, London: Cape.
Mintz, S.W (1986) Sweetness and Power: the Place of Sugar in Modern History,
Harmondsworth: Penguin.
Morris, H.S. (1968) The Indians in Uganda, London: Weidenfeld & Nicolson.
van Onselen, C. (1976) Chibaro: African Mine Labour in Southern Rhodesia, 1900-1933,
London: Pluto Press.
Panikkar, K.M. ( 19 59) Asia and Western Dominance, London: George All en & Unwin.
1064
Patterson, 0. (1982) Slavery and Social Death: a Comparative Study, Cambridge, Mass.:
Redclift, M. (1978) Agrarian Reform and Peasant Organization on the Ecuadorian Coast,
London: Athlone Press.
--(1984) Development and the Environmental Crisis: Red or Green Alternatives?,
London: Methuen.
Roberts, B. (1978) Cities of Peasants: the Political Economy of Urbanization in the Third
World, London: Edward Arnold.
Sahlins, M. (1961) 'The segmentary lineage: an organization of predatory expansion',
American Anthropologist 63(2):332-45.
Scott, J.C. (1976) The Moral Economy of the Peasant: Rebellion and Subsistence in
Southeast Asia, London: Yale University Press.
Asian Studies 34(1-3):3-43, 195-227, 363-99.
Southall, A. (1956) Alur Society: a Study in Processes and Types of Domination,
Cambridge: Heffer.
Southern, R.W (1962) Western Views of Islam in the Middle Ages, Cambridge: Mass.:
Spate, O.H.K. (1979) The Spanish Lake, London: Croom Helm.
--(1983) Monopolists and Freebooters, London: Croom Helm.
Stavrianos, L.S. (1981) Global Rift: the Third World Comes ofAge, New York: Morrow.
Sundkler, B. G. M. (1948) Bantu Prophets in South Africa, London: Lutterworth Press.
Tambiah, S.J. (1970) Buddhism and Spirit Cults in North-east Thailand, Cambridge:
m er, ew ystem o avery: t e xport o verseas,
1830-1920, Oxford University Press.
Wachtel, N. (1971) La Vision des vaincus: les Indiens du Perou devant la Conquhe
espagnole, Paris: Gallimard.
Waddy, J.A. (1988) Classification of Plants and Animals from a Groote Eylandt Aboriginal
Point of View, 2 vols, Darwin: Australian National University.
Weber, M. (1956) The Sociology of Religion, London: Methuen.
--(1961) General Economic History, New York: Collier.
Wolf, E.R. (1982) Europe and the People without History, Berkeley: University of
California Press.
Womack, J. (1969) Zapata and the Mexican Revolution, Harmondsworth: Penguin.
Worsley, P. (1957) The Trumpet Shall Sound: a Study of 'Cargo' Cults in Melanesia,
London: MacGibbon & Kee.
--(1964) The Third World: a Vital New Force in International Affoirs, London:
--(1984) The Three Worlds: Culture and World Development, London: Weidenfeld &
Nicolson.
Zamosc, L. (1986) The Agrarian Question and the Peasant Movement in Colombia:
Struggles of the National Peasant Association, 1967-1981, Cambridge: Cambridge
University Press.
Zubaida, S. (1989) Islam, the People and the State: Essays on Political Ideas and
Movements in the Middle East, London: Routledge.
1065
SOCIAL LIFE
FURTHER READING
Blackburn, R. (1988) The Overthrow of Colonial Slavery, 1776-1848, London: Verso.
Brockway, L.H. (1979) Science and Colonial Expansion: the Role of the British Royal
Botanic Gardens, New York: Academic Press.
Cohn, B.S. (1983) 'Representing authority in Victorian India', in E.Hobsbawm and
T.Ranger (eds) The Invention of Tradition, Cambridge: Cambridge University Press.
Geertz, C. (1963) Agricultural Involution: the Process of Ecological Change in Indonesia,
Berkeley: University of California.
Gellner, E. (1983) Nations and Nationalism, London: Blackwell.
Hughes, R. (1988) The Fatal Shore: a History of the Transportation of Convicts to
Australia, 1787-1868, London: Pan Books.
Kehoe, A.B. (1981) North American Indians: a Comprehensive Account, Englewood
Cliffs, NJ: Prentice-Hall.
Lenin, V.I. (1915) Imperialism: the Highest Stage of Capitalism (many editions).
van Leur, J.C. (1955) Indonesian Trade and Society: Essays in Social and Economic
History, The Hague: van Hoeve.
Mason, P. ( 1974) A Matter of Honour: an Account of the Indian Army, its Officers and
Men, London: Cape.
van Onselen, C. (1976) Chibaro: African Mine Labour in Southern Rhodesia, 1900-1933,
London: Pluto Press.
Pannikar, K.M. ( 19 59) Asia and Western Dominance, London: George Allen & Unwin.
Redclift, M. (1978) Agrarian Reform and Peasant Organization on the Ecuadorian Coast,
London: Athlone Press.
World, London: Edward Arnold.

Asian Studies 34(1-3):3--43, 195-227, 363-99.
Spate, O.H.K. (1979) The Spanish Lake, London: Croom Helm.
Stavrianos, L.S. (1981) Global Rift: the Third World Comes ofAge, New York: Morrow.
Wolf, E.R. (1982) Europe and the People Without History, Berkeley: University of
California Press.
Worsley, P. (1984) The Three Worlds: Culture and World Development, London:
Zubaida, S. (1989) Islam, the People and the State: Essays on Political Ideas and
Movements in the Middle East, London: Routledge.
1066
INDEX
Abelam of New Guinea structural linguistics 369

art analysis 660,665-6,670-1,673 templates 379
Aboriginal peoples of Australia time 515
art analysis 665-6, 674, 675-7 Upper Palaeolithic 391
belief systems 571, 582 Abu-Lughod, L. 827-8, 862, 869
ceremonial centres 618-19 academic achievement
colonization 1050 language 876
conceptual organization 461 access
cosmology 338 spatial organization 472, 49 5
diabetes 317 accidents
dogs 209-10 American Indian comparison 312-13
Dreamtime 410, 619 Acheulean
ua orgamzat10n . erectus ,
motherhood 791-2 tool-making 383
myth603 acquired immune deficiency syndrome
population 265, 284 (AIDS)
privacy mechanism 467 Mrica 301
religious ceremonies 697 acute infections see disease
small-scale husbandry 206 Adair, ]. 15 4
spatial organization 481,483,487,490 Adam, Barbara 338-9, 503-23
structural analysis 662 adaptation
subsistence systems 203 Broca's area 8
technology 421 cultural factors 67
toas 668-9 diseases 300
trade 220 hominid evolution 71
Abrahams, R.D. 868 innovation 442-5
absolutism nutrition 227-8
European expansion I 044 rituals 372
power structure 1057 social evolution 946
world order 1040--1 writing systems 531
abstract art adaptive niches
see also art tool-use 140, 143--4
analysis 665 adjudicators
abstraction law 971
belief systems 564 administration
cultural role 368 colonialism 1051
likeness concept 372 nation formation 718
prehistoric art 384 social evolution 94 2
1067
INDEX
adoption agro-literate societies

social parenthood 793-4 ethnicity 708
adult literacy AIDS see acquired immune deficiency
education 552-3 syndrome
aesthetic drama Aiello, L. 66
see also drama; Akrich, M. 449
performance Alcock,J. 138-9
interactive nature 342 alcoholism
performance 626--9, 63(}-1, 632 comparative studies 312-13, 412-13
aesthetics disease 12
anthropological study 343-6 food selection 238, 256
art analysis 655, 672-7 Aleuts
belief systems 583 literacy 541
vocal styles 693 Alexander, J. 236
Africa alienation
climacteric events 69 material culture 335, 415
colonialism 81, 1052, 1056--7 Alikosh (Iran)
endemic disease 30(}-1 agriculture 236
evolution debate 79-80 allegory
fossils 50-63, 54, 57 definition 604-5
genetic migration 92-3 detotalization 606
H. erectus 81-2 alliance
hominids animal behaviour 768-9
classification 58-61 marriage 798-803
migration 72 property 797
Late Stone Age 94, 96 theory 801-3
nation-buildin 1061-2
aggressiOn societal impact 529, 532
causes of 984--6 Althusser, L. 976
agrarian societies altruism
ethnicity 711 definition 771-2
order 752 reciprocal164, 191
social evolution 1013 Alzheimer's disease
agricultural revolution speech production 118
colonialism 1049 Amazonia
concept of 442 European ecological impact 306-9
agriculture myth602
see also swidden agriculture symbolism 375
definition 207 warfare 985
division of labour 899 ambilocal residence
food production debate 10, 11
rules 795-6
Green Revolution 1061
Ambrona (Spain)
intensification 218-19
hunting 85
labour productivity 277
America
life expectancy 283
see also New World;
Neolithic 267-8
United States ...
nutrition 236--7
indigenous peoples 305-6
population effect 214-18,270
Spanish/English comparison 302-6
social evolution 942-3,946--7
American War of Independence 1044,
subsistence mode transition 197-221
1057
temporality 515
Americanization
agro-industry
material culture 405
colonialism 1055
1068
INDEX
amitalocal residence anthropocentrism

rules 795 human uniqueness debate 15, 27, 29,
Amud cave (Israel) 30
stone technology 94 Anthropoidea
anaemia early hominid evolution 35, 38
sedentism 284--5 anthropology
analogic drive cultural657, 1018
symbolism 334, 366-7 dialogic 862, 871-2
analogies evolution debate 79
definition 373 human uniqueness debate 28-30
symbolism 381, 382 process orientation 738
anatomically modern humans study of 331
see also modern humans anthropomorphism
chronology 81 evolutionary biology 764
classification debate 5 anti-language
cultural transition 93 definition 878
group size 770 Apaches
hunting debate 86 language 867, 868
tool-use 145-6 literacy 540
variation 60 apes
anatomy see also chimpanzees
cultural transition 91 behaviour 768, 769
hominid comparison 89 crania 47
Andaman Islanders hand comparison 145
belief systems 564 hominid evolution 33-4, 37
Anderson, C.A. 548 human comparison 4, 18, 146
Anderson, R.L. 649, 650 social networks 739

Andrade, R.G.D. 834, 835-6 social transmission 355
animal domestication symbolism debate 382
herding 235-6 teeth 44-5
slavery debate 451 aphasia see Broca's aphasia
social evolution 942 Apollo XI shelter (Namibia)
animal husbandry rock art 98
classification 211-12 Appadurai, A. 417
development 209-14 apprenticeship
food production debate 10 division of labour 903
animality memory 147
human comparison 14--30 social parenthood 793
Linnaean classification 27 technology 446
animals tool-use 151-3, 157
culture debate 350-63 work 891
kinship 785-6 Aquili, Eugene d' 638-9
language 26, 108-10 Aquinas, St Thomas 563
nutritional resources 4 Arabs
population growth 269 ethnicity 710
population regulation 277-8 negotiations 997
prehistoric art 384 religion 716
sociality 739, 756-77 Aranda
spatial colonization 467-9 motherhood 790-1
subsistence 10 arbitrariness
tool-use 133 language 862
1069
INDEX
arbitrators cultural significance 101, 343-6

law 971 definitions 54-6, 648-53
archaeology figurative 665
cultural orientation 67, 659 language 99-100
evolution debate 79 representation systems 664--70
food production 10 semiological analysis 664-5
fossil records 50 style 670--2
H. erectus 83-4 temporality 515
H. sapiens 83-4 Upper Palaeolithic 96--9
hominid comparison 89 verbal867-8
human evolution 6 Western categorization 650--3
reconstruction debate 413 art objects
spatial organization 480--1 categorization 655
technology 269, 440 artefacts
temporality 514 aesthetics 344--6
archaic H. sapiens see Homo sapiens art analysis 660, 662, 664
architecture clocks 512-16
myth 601-2 cultural significance 334--40
spatial organization 480 environment 337
temporality 515 evolutionary classification 142
archives function 436--7
music 700 human evolution 5-6
Ardener, Edwin 378, 379, 390 materiality 406-8
aristocracies meaning 396-417
evolution 1013, 1033 prehistoric symbolism 383
world order 1040 specificity 408-9
Aristotle 563 613 687 756
commodities 913, 922, 923 temporality 409-15, 515
technology 424 artificial intelligence
Armenians cultural instruction 186
ethnicity 710, 713, 715 artisans
religion 716 division of labour 903, 905
vertical ethnie 714 artists
armies training 63 7
world order 1040 arts
arms technology 421
warfare994 Ascher, R. 64
arms control Asia
iconicity 385 evolution debate 79
arms race Assyria
environment-organism feedback
ethnicity 713, 714
cycles 172
nationalism 718
Armstrong, J. 709 religion 716
Armstrong, R.P. 623
asymmetric power semantics
art
language 879-80
Aboriginal 619
audio recorders
abstract 665
technological development 690, 692,
aesthetics 672-7
700
analysis, methodological perspective
Auge, M. 573, 574
662-4
Australia
anthropological theory 330, 648-78
colonization 81, 88, 93, 94
belief systems 582-3
rock art 98
cross-cultural studies 653-4, 656-62
1070
INDEX
Australopithecines see also exchange;

Broca's area 62 trading
chronology 81 counter-trade 928-31
classification 58-9 Earth, F. 709
cranial centre of gravity 47 Barthes, Roland 406
culture 68, 71 basal ganglia
diet 228 reptiles 128
endocranial casts 46 Basques
evolutionary divergences 50 ethnicity 715
extinction 72 Basso, K. 867, 868
fossil records 53, 55 Bates, E. 386
frontal lobes 63 Bateson, Gregory 256
hominid divergence 6 communication theory 594
hominid evolution 36, 37, 43 learning 83 8
stone tools 70, 71 metacommunication 623, 639
supralaryngeal vocal tracts 126 Bateson, Mary 1001
authority Baudrillard, J. 406
centralized 966 Bauman, R. 867
stateless organization 750 Beattie, J.H.M. 784
traditional 973 beauty
automobile accidents art analysis 666-7
American Indian comparison 312-13 Bechoefer, W. 487
autonomous model Beck, B.B. 139
literacy 339, 532-3, 546 Bedouin
autonomy language 862, 869
ethnie 715 sex 827-8
rules 79 5, 797 '

behaviour
Aztecs altruism 128, 771-2
food selection 23 7 American Indian comparison 312-13
survival rates 306 analogic drive 334
animal transmission 351
baboons consolation 769
behaviour 767, 768 cultural concept 329
social transmission 356 cultural inheritance 762-3
sociality 757, 759 cultural transition 91, 93
tool-use 135 division of labour 903
Babylon economic 278, 921, 923-5
ethnicity 710 ethnicity 707
bad faith feeding 229, 230-1, 232
negotiations 999 foraging 228
Bailey, F.G. 969 human evolution 5-9
Bakhtin, Mikhael871, 877 hunter-gatherers 275--6
bands linguistic 872
level of integration 944 natural selection 741
social evolution 1013 Palaeolithic 87, 97
Barasana phenotypes 163
myth 602-3 political realism 751
Barba, Eugenio 623, 641 ritual638 behaviour (continued)
Barnard,Alan 740,741,783-808 sex 813-14
Barnes, J.A. 504 symbolism 101, 381-92
barter transition process 382
1071
INDEX
warfare 988 Binford, S.R. 95

weaning 238 biological anthropology
behaviour settings relationship 739-44
definition 461-2 biological determinism
behavioural ecology kinship 786
definition 479-80 biological relationships
development 756 social life 739-54
behavioural geography biology
definition 480 see also evolutionary biology;
behavioural space molecular biology
definition 478-9 culture dichotomy 7, 332-3
behaviourism human food behaviour 232
kinship 785 kinship 785-92
technology 440 population 169
belief pre-Darwinian 142
cultural relationship 340--3, 361-3 sex 813-16
ideological power 749 bipedalism
indoctrination 582-7 hominid evolution 40, 41-2,43, 51
nature 578-80 human uniqueness debate 27
observation 575-8 speech 119, 121
pedagogy 333 birds
rationality 563-87 animal behaviour 766
religion 567-72 proto-cultural behaviour 167, 185
translation 57 5-8 reproduction 765
Bell, D. 707 social transmission 353
Bellah, R. 570 sociality 758, 759
Birdsell . 270
fossil records 55 bison
Bengal hunting debate 85
Bay of 15 Black, F.L. 308
Berlin, Brent 374, 390 Black, Mary 865
Berlin, Isaiah 1010 Blacking, John 688
Bernstein, B. 876 blacksmithing 149
Besnier, Niko 339-40, 527-54 blades
Beteille,A. 751-3,1010-35 Levallois technique 85
Bhola, H. 548 Blanc-Szanton, Cristina 826
big-game hunting Blau, P.M. 1030
see also hunting Blier, S.P. 664
efficiency studies 279-80 Bloch,M. 571,870,965-6
H. erectus 85 blood pressure
bilateral negotiations Pacific island societies 315-16
Bloom, D. 550, 551
law 970
Bloomfield, H. 536
Bilzingsleben (Germany)
Boas, Franz 550, 582,657,659,671
hunting 85
language 864
bimolecular surveys
music 687, 688
populations 5
myth 593
Binford, L.R.
non-commodity exchange 917
cultural transition 95
body
function analysis 471
anthropological study 332
hominid subsistence 203
decoration 666, 675
planning depth 87
myth 598-602
sedentism 272
symbolism 378
1072
INDEX
Boesch, C. 356 bride-service 797

Bohannan,Paul968,969,972 bridewealth 797
Boker Tachtit (Israel) Briggs, C.L. 867
cultural transition 91, 95 Brillat-Savarin, A. 246, 255
Bolinger, D. 87 5 Britton,James 552
bonding broad-spectrum revolution
human 770 diminishing returns 271
labour 893 division of labour 289
Bonner, John T. 380, 382 economic transition 11, 208, 278
Boone, S.A. 667,676 feeding behaviour 266--7
Bordes, F. 95-6 nutrition debate 280--1
Boserup, Ester 942-3 Broca's aphasia
agriculture 219, 276-7 hand movements 127
cultural models 270--1 speech production 116--18
palaeopathology 278 Broca's area
population growth 275 adaptation 8
botanic imperialism early hominid evolution 48, 62-3
world order 1049 H. habilis 99
Boule, P.M. 90 motor control127
boundaries Neanderthals 125
language 874 speech production 116, 117, 118-19
nation-building 1061 Broken Hill (Zambia)
spatial organization 468-70,472,477, Rhodesia Man 126
481-2,490,493 Bronze Age
Bourdieu, P. 254, 801, 875, 882 burial grounds 952, 954
bourgeoisie life expectancy 282
inequality 1020 Brown,' P. 879

nationalism 727 Brown, R. 879
revolutions 1057 Bruner,J.S. 849
world order 1040--1 Buddhism
Boutros-Ghali, Boutros 1002-3 ethnicity 712
Bown, L. 545 literacy 537
Boyd, R. 182 Mahayana 1058
Brace, C.L. 90, 92 Buffon, Count de 20-2, 33
brain Buhler, K. 64
animal behaviour 769-70 buildings
enlargement 44--5 settings 464
evolution 127-9 built environments
handedness 145-6 conceptualization 460--2, 463-4,
hominid evolution 44, 51, 61 465-7
modern humans 62, 116-19 cultural relationship 474-5
ritual638 meaning 473-4
speech evolution 119-21, 128-9 origins 467-70
Brazil purposes 470, 471, 472
development policy 307-9 spatial organization 460--97
music 693-701 writing 473-4
Bretons Buka (Solomon Islands)
ethnicity 715 fossil records 94
Breuil, Abbe 615 bureaucracy
Brewer, S. 137 ethnicity 708
bricoleur 149 literacy 532
1073
INDEX
nation formation 719 cash crops

social evolution 944, 946 European imperialism 1056
burials Castaneda, Carlos 581
Australia 94 caste systems
Neanderthals 91, 96-7 India 1014, 1025-8
Old World 84 <;:atal Huyuk (Turkey)
Upper Palaeolithic, Europe 96 agriculture 236
Burnet,James see Monboddo ... catastrophism
Burridge, K. 594 replacement/ continuity debate 90-1
Burton, R. 955 categorization
Bushmen see San ... anthropological442-3
Byrne, David 701 art objects 655
Byzantium prototype theory 374
religion 716 catharsis
musical697
Caesar,Julius 710 Catholicism
Cakchiquels sex 826
child mortality rates 304 Caton, S.C. 870
Calame-Griaule, G. 602 cats
calendars social transmission 355
temporality 338, 507, 509-10, 512 causation
calorie requirements kinship 796-7
human populations 232, 276 proximate 761
camp sites symbolism 372, 380, 391
planning depth 87 Cavalli-Sforza, L.L. 92-3
Upper Palaeolithic 92 caves
Canada ritual sites 614,617

indigenous peoples 305 centralized authority
cancers law 966
disease 12 cereals
canine teeth cooking 427-9, 430
evolutionary divergencies 51 ceremonial centres
cantometrics hunter-gatherers 616--17,618,619-20
sound analysis 688 ritual618
capitalism ceremonies
agro-industry 1055 see also rituals
booty 1046 art analysis 664
commodity exchange 922 ethnicity 712
division of labour 904--5 mobilization 698
ethnicity 708 potlatch 916
inequality 1016 Chaikin, Joseph 623
Marxism 943 Chalcolithic
material culture 335 copper axes 154
Third World 722 Chaloupka, G. 98
work 747 Chang, K.C. 245, 251
car accidents Chanock, M. 975
American Indian comparison 312- charters
13 myth 592-5
Carneiro, Robert L. 946 Chatwin, Bruce 619
carmvores Chemeron (Kenya)
animal behaviour 766 fossils 53-4, 57
1074
INDEX
Cheney, D.L. 773 folk medicine 57 5

Cherokees language 873, 880-1
literacy 540 Taiping rebellion 1058
normative power 991-2 temporality 514
Chesowanja (Kenya) writing system 529, 530
fossil site 57 Chinoy, Ely 895
Cheyenne Chomsky, Noam
law 969-70 generative grammars 401
subsistence modes 232 language 146, 153
chiefdoms linguistic theory 373
inequality 1020 choreometrics
social evolution 1013 movement analysis 688
society 945 Christianity
Childe, V.Gordon 101, 442, 941, 942 belief systems 569, 571
children egalitarianism 1032
development 745-6 ethnicity 712
division of labour 899-900, 901, Eucharist 635
902-5,906 literacy 537, 543
food selection 23 7 medieval mass 625
identity 846-52 missionaries 1053
marriage 798 ritual 64 3-4
socialization 847-50,851-2 chronic infections see disease
warfare 993, 994 chronogeography
world view 868-9 temporal organization 465-6
Chilkats chronology
art analysis 671,672 industrialization impact 338
behaviour 767, 769 definition 274

classification debate 39 cirrhosis
diet 228 alcoholic 312-13
gesturallanguage 63 cities
hierarchy 156 European imperialism 1056
history debate 350-1 Neolithic 267
hominid evolution 34, 42-3, 50, 51 origins of 469
human uniqueness debate 28 citizenship
likeness exploitation 391 nationalism 717-18
linguistic abilities 109, 112, 118-19 Civil Rights Movement 698
pedagogy 333, 357-9 civilization
prefrontal cortex comparison 119 definition 273-4
social transmission 353-4, 356--7, demography 267, 268-9, 273-4
360,362 diseases of 287, 299
sociality 757 literacy 527
speech perception 128 social evolution 942
supralaryngeal vocal tracts 121, 122 state-level organization 274
termiting techniques 138 technology 421
tool-use 133, 136, 137-8 clad ism
China classification 39
belief systems 584 cladogenesis
colonialism 1050, 1055-6 stone tools 70
cooking 245 Clanchy, M.T. 543
ethnicity 709 Clark, G.A. 91
female literacy 546 Clark, Legros, W.E. 35, 37, 38,40
1075
INDEX
class structure see also knowledge

commodities 914--15 animal behaviour 770
concept of 1031 animal skill transmission 333
division of labour 908 belief systems 565-6, 582
ethnicity 708, 713 human/animal comparison 380
inequality 1011 Levallois technique 85
labour 894--5 literacy 534-5, 536--7
language 876, 877 social transmission 353-4
lateral ethnie 714 spatial organization 483-4
occupations 1030 symbolism 389-90
racial stratification I 014 tool-making debate 71, 146--50
vertical ethnie 714 cognitivism
classical economics anthropological study 332
barter 928 spatial organization 337
value systems 748 cognized space
wealth 916, 929 definition 479
classification Cohen,Abner504, 1003
anthropological442-3 Cohen, Mark Nathan 11-12, 265-91,943
belief systems 574 Cohen, Raymond 1002-3
cultural anthropology 1018 Cohen, Y.E. 253
Hominidae 58-61 cold foods
Hominoidea 35-6 taboos 249-50
language 388 Cole, M. 534
museums400 collective representations
spatial organization 484 individualism 1017
symbolic role 367-8 collective violence
cultural transition 91-2 common security 983-1004

evolution debate 79-81 Collier, J. 816-18, 820, 822
language 100-1 Colombus, Christopher 1041, 1046
local variation 272 colonialism
Clive, Robert 1042 culture 1052-4
clocks economics 1054-7
temporality 338, 508-10, 512-16 imperialism 1048-50
clonal selection law 973-6
immunology 185 missionaries 1053
clothing nationalism 727
function 443-4 non-European polities 1041-4
Club of Rome 441 polity 1051-2
eo-evolution South America I 046-8
environment-organism relationship spice trade 1045-6
171-2, 176,177,178,179, 180--1 world order 1040--63
co-operation colonization
work 891-908 Australia 93
coalitions chronology 80, 81
animal behaviour 768-9, 773 disease impact 300
code-switching English/Spanish comparison 302-6
language 862 European4
multi-lingual speakers 874 H. erectus 81-2
codes human evolution 9
language 865-6 hunting debate 85, 86
cognition language 10 I
1076
INDEX
literacy 339-40 comparative method

planning depth 88 evolution 1010--15
sedentism 101-2 competition
social order 753 equality 752
technological role 94-5 componential analysis
colour paradigms 374
semantics of 390 Comte, Auguste 1054
colour-caste systems conception
USA 1025 belief systems 571-2, 576
Colson, E. 273 motherhood 790-1
commercial literacy Trobriand Islanders 789
see also literacy conceptualization
Iran 541 symbolic role 372
commitment concordant structure
negotiation technique 998 definition 379
commodities concubinage
colonialism 1050 marital ties 798-9
fetishism 931 social parenthood 793
horizontal 932 conditioning
Marxism 913 social transmission 352
political economy 748-9 conflict spectrum
temporality 507, 513 definition 992
vertical 932 Conroy, G.C. 53
wealth 912-15 consanguinity
work 747 gift exchange 923
communication incest 800
musical traditions 688-9 conservatism

negotiation 99 5-6 religion 716
non-verbal 64 consolation behaviour
organization 465-6 animals 769
social evolution 945 consonantal alphabets
spatial organization 484 literacy 529
temporality 508, 510 constellations of knowledge
commumsm apprenticeship 151, 153-4
gift exchange 920 definition 149-50
modern world 723 constitutions
primitive 1011-12 modern world 724
community consumer goods
demotic ethnicity 714 material culture 335, 397
ethnicity 711 consumption patterns
ethnie 709 subsistence systems 203
nation formation 721 wealth 915
political economy 747 context
social life 738 belief systems 577
socialization 841 music 686, 690
spatial organization 482 spatial organization 491-2
vertical ethnie 714 symbolism 369
community identity contingency planning
music 699-700 tool-use 150
comparative economy continuity hypothesis
wealth 919 art 99-100
1077
INDEX
cultural transition 95 crops

modern humans 88, 90, 91-2 see also plants
contraception agricultural intensification 219
hunter-gatherers 290 cash 1056
lactation 289 combinations 216
contracts domestication 209
division of labour 905, 908 cross-cultural studies
labour agreements 892 art 653-4, 656--62
control belief systems 567
law 973 built environment 463, 464
spatial organization 484, 485 emic categorizations 488
synthetic model955-7 infancy 847
temporality 513 language 861, 863, 868, 877
conventions literacy 537
myth 592-5 negotiations 994-5,999-1003
Cook, Scott 896 sex 814, 820, 821
cooking structure 866
cereals 427-9, 430 temporality 503
functional analysis 470--1 Crow system
nutrition 243-5 relationship terminologies 796, 805,
subsistence 10 806
Coon, Carleton 79, 80 crowding
Coote,J. 672 complex social organization 27 3
core-areas Crump, T. 531
territoriality 486, 487 Crusaders
Corner, George Washington 66 ethnicity 714
trade 1045-7
Corvinus, G. 70 culinary revolution
cosmology hominid evolution 91
art analysis 664 culinary triangle 24 3
environment 337 cultivation
Hopi 506, 516 domestication distinction 210
myth 591-2 Natufian practices 206
N eo lithic 1017 origins 207-9
sex 819 cultural ecology
social structure 583 Stewardian 197
cosmopolitanism cultural isolation
modern nations 722 politics 716
counter-trade cultural landscapes
barter 928-31 latent functions 471-2
Coursey, D. G. 236 cultural relativism
art analysis 648, 679
courts
belief systems 581-2
colonialism 974
cultural space
decision-making 972
definition 479
litigation 970
cultural tradition
craftsmanship
East Africa 83-4
division of labour 905
Neanderthals 84--5
labour 894
cultural transition
social evolution 942
behaviour 93
creation myths
BokerTachtit 91
ethnicity 712
evolution debate 80
Crelin, 65
planning depth 87
1078
INDEX
culture Czerniewska, P. 552

animal/human comparison 350-63
art analysis 343-6, 653-4 Dahrendorf, Ralf 1016, 1020 1021
belief 340-3 1031 ' '
biology dichotomy 7 Daisyworld model
built environment 474-5 environment-organism feedback
ceremonial centres 620 cycles 173
cognition 71 dance
collective violence 987 anthropological theory 331
components 475, 476--7 community identity 699-700
concept of95, 329-30 cultural significance 343-6
definitions 151, 368, 474--5 definition 694--6, 696
disease 12 innovation 700--1
early hominids 66--7,68,69-72 music 686-702
ethnicity 346--8 performance 698-9
evolution physiological effect 696--8
classical models 268-9 Suya 693-701
divergencies 51 technology 690-3
human species 4 Yaqui614,616
theories of 142, 927 Dart, R.A. 36
typologies 941-9 Darwin, Charles
food-processing 232-3, 243-6
altruistic behaviour 771-2
geneticlinks 164-5 animal behaviour 767
hegemony 1052-4
classification 142
human capacity 332-4
descent 162
inheritance systems 188
domestication 11
introduction 329-48
kinship 786-9
food selection 23 7
language 100, 128
natural selection 5, 181, 761-2
literacy 532-7
Origin ofSpecies 17, 33, 80, 125
local variation 272
rationality 563
material334-5, 398-9, 656
readaptation 127
materialist theories 141
species modification 210
nature dichotomy 22-3
Darwinism
negotiation 994--5
neural185
niche construction 162-92
phenotypes 163, 178 David, Jacques-Louis 725 726
Davidson, D.S. 444 '
politics 346--8, 706--28
process orientation 738 Davidson, I. 99-101
resistance to 1052-4 Davis, C.M. 237
schemata 834-6, 846-7 Davis, P.K. 989
sex 813-16, 821 Dawkins, R. 172, 175,762-3
skills 334--40 de-s killing
society dichotomy 738 division of labour 905-6
symbolism relationship 366--92 Dean, C. 66
tool-use 144, 151-7 death
culturgens 9 American Indians 305
Cuppy, Will 22 marital ties 799
customary law sex 819
colonialism 974-5 temporality 518-19
social order 750 DeBernardi,Jean 746--7,861-83
cyclicality Debreu, G. 916
temporality 511, 517-19 debt
1079
INDEX
Africa 1062-3 Descartes, Rene 563

health impact 301 descent
Dechend, H.von 390 inequality 1020--2
decision-making kinship 794-5
collective violence 989 niche construction 162
courts 972 despotism
law 964,970 social evolution 943
social evolution 945 determinism
warfare 988 biological 786
decoration niche construction 163
art analysis 665, 675 detotalization
deduction allegory 606
belief systems 57 5 myth600
defence development
sociality 771 literacy 547-50
Western establishment 751 phenotypes 163
Deforge, Y. 436 socialization 831-54
delayed exchange DeVore, I. 275
see also exchange Dewey, A. 931, 936
marriage 926--7 diabetes
demand disease 12
community size 273 Pacific island societies 314--15 317
318, 319 ' '
negative reciprocity 924
population pressure 271-2 diachrony
demarcation modern/traditional debate 504, 522
spatial organization 49 3-4 dialectics
ift exchan e 920
evolution 943, 1013, 1033 dialects
order 752 mating habits 127
demography politics 873
American Indians 301-2 dialogic anthropology
expansion 265-91 language 862, 871-2
food selection 238, 253-4 Dibble, H.L. 95
human species 4 Diderot, Denis 423
subsistence 10-13 diets
demoralization agriculture 236-7
colonization impact 308 broad-spectrum revolution 266
demotic ethnicity fishing 234--5
see also ethnicity food-processing 232-3
politics 714 herding 235-6
dentition human behavioural specificity 229,
230--1,232
cultural transition 91
human populations 226-56
hominids 61, 228
hunter-gatherers 233-4,275-6
stress rates 281
differentiation
deoxyribonucleic acid (DNA)
ethnicity 711
see also mitochondrial DNA
myth 603-4
fossil records 48-9
spatial organization 482
gene codes 17
totemism 596-8, 601
sociality 762
diffusionist hypothesis
deportation
art analysis 659
ethnicity 714
digital technology
Derrida, Jacques 408
music 700
1080
INDEX
diglossia animal husbandry 209-14

definition 544 cultivation distinction 210
discourse husbandry comparison 11
literacy 536 lactose tolerance 167
political100 1 Neolithic 267-8
study of869 plant cultivation 207-9
disease slavery 451
colonialism 1047 social evolution 942
demographic expansion 265 subsistence 10
episodic high mortality 286--8 dominant-submissive relationships
human species 4 captive chimpanzees 357
hunter-gatherers 276 domination
indigenous populations 297-320 law 963-4, 965
Pacific island societies 314--17, 318, political 940--57
319 social life 737
palaeopathology 281-2 Dorze
prehistoric populations 12 symbolism 372
sedentism 284 double-bind
theories of 361 food behaviour 256
displacement activities Douglas, Mary 243,246,247
definition 622 dowry
disputes kinship 797
law 962-79 Doxiadis, C.A. 491
social life 750--4 drama
distributed neural networks see neural interactive nature 342-3
networks performance 626--9, 630--1, 632
egalitarianism 1032-3 Australian Aborigines 410, 619

division of labour Drennan, Robert 955
co-operation 891-908 Druse
egalitarianism 1012 ethnicity 713
gender 450--1, 899-900,901,902-5 dryomorphs
hunter-gatherers 234 early hominid evolution 37
individualism 1017 dualism
inequality 1016 animality /humanity comparison 21
market structure 905-8 artefacts 415-17
residence 796--7 temporality 505, 511, 516-22
sexual 822, 825 Dubois, E. 82
subsistence 11 Duchamp,Marcel652
divorce Duchin, L.E. 66
marital ties 799 Dumont, Louis 1014, 1025, 1028
Dmanisi (Caucasus) Dunbar, R.I.M. 739-41,756-77,785
H. erectus 82 Duncan, O.D. 1030
DNA see deoxyribonucleic acid Dunne,J.S. 518-19
Dog on Dunnell, R. 670
myth602 duolocal residence
dogs rules 795-6
domestication 209-10 Durham, W.H. 167, 182
Dolgin, Janet 369 Durkheim, Emile
DolniVestonice (Moravia) Australian Aborigines 697
burials 96, 97 art analysis 660
domestication collective representations 1017-18
1081
INDEX
division of labour 903-5 economic power

food selection 23 8 sources of952-5
food sharing 253 economics
inequality 1016 Africa 1062
language 878 colonialism 1047, 1054-7
law 964--5, 967 comparative 919
nationalism 726 exchange 911-36
religion 569-70, 583-4, 586-7, 595 food behaviour 254
social evolution 94 7 individualism 1017
social relations 415, 417 inequality 1015
technics 432 national languages 872-4
technology 422 neoclassical 906
Dushanbe (former USSR) warfare994
H. erectus 82 wealth 911
dwellings work 747, 891-2
built environment 462, 463 economies of scale
settings 464 agriculture 277
dyads Edelman, G.M. 185
social transmission 351-2 Edison, Thomas 690
Dyan, Moshe 1002 Edmondson, M.E. 445
Dynastic Period education
Egypt, animal domestication 235 colonialism 1053
equality 1058
ethnicity 708
Earle, Timothy 749, 752, 940-57
literacy 550-4
Eastern nationalism
nationalism 717-18
definition 717
Third World 723
warfare994
ecological anthropology
world order 1041
spatial organization 481
Edwards, C.P. 833
ecological inheritance
Edwards,J. 874
alternative cycles 187, 189
efficacy
definition 9
definition 448-9
externall76
entertainment 622-4, 625, 626
natural selection 178
egalitarianism
sociobiology 165
concept of 1012
ecology
social order 752-3
Amazon 306-9 typology 948
cultural factors 67, 197 Egypt
definition 479 animal domestication 235
general systems theory 175 Camp David agreement 999
hominid evolution 50, 69,71 communicative style 1001-3
spatial organization 469-70 ethnicity 709
systems 198 nationalism 718, 1054
warfare 985-6 water clocks 514
econometric analysis Ehrenzweig, Anton 637
collective violence 989 Eisenberg, L. 22
economic anthropology Eisomon, T. 550
development 748 ekistics
economic behaviour definition 480
optimal foraging theory 278 scale 491
economic geography electronic space
definition 480 definition 479
1082
INDEX
elites slavery 450

demography 274 social evolution 941, 947
ethnicity 708, 713 England
evolution 947 civil law 80 I
Java 882 ethnic bureaucracy 719
language 874 New World impact 303
law 964 Enlightenment
power sources 749 colonialism 1054
social power 951-2,954 entertainment
Third World 722 concept of 343
Ellen, Roy 10-11, 197-221 efficacy 622-4, 625, 626
Elliot, D. G. 37 environment
Ellul, Jacques 140--1 agricultural modification 216-17
Ember, C. and M. 796-7 cultural factors 67
embodiment cultural transition 91-2
myth600 evolution debate 79-80
em1cs human impact 391-2
built environment 463 language 100-1
categorizations 488 modern world 754
emotions modifications 197
language socialization 868 music 696
music 697 orgamsm
socialization 851 eo-evolution 176, 177, 178, 179,
symbolic resonance 377 180-1
empires feedback cycles 168,169-71,172-6
non-European polities 1041-2 relationship 9, 167-8
belief systems 578 proto-cultural behaviour 167

employers selfhood 842-3 environment
political economy 747 (continued)
employment shared 164--5
Third World 723 socialization 832
emulation spatial organization 330, 337, 460-
analogical processing 383 97
encephalization symbolism 381
hominids 61 tool-use 133
selective 44--5 environment-behaviour studies
technology relationship 81 spatial organization 480
enculturation Eoanthropus
identity 831-54 early hominid evolution 37
endocasts epidemics see disease
hominid 61-2 episodic high mortality
endocranial capacities population debate 286-8
early hominid evolution 46, 47 epizootics see disease
endogamy equality
food taboos 247 concept of 1031-5
kinship 801 education 1058
Engels, F. equivalence
equality 1011 gift exchange 919
hierarchy 1031 Errington, J.J. 882
industrial revolution 442 Errington, S. 815, 817-18
inequality 1020 Espinas, A. 448
1083
INDEX
essentialism ethnomusicology
social life 738 anthropological theory 331
ethics comparative study 687
spatial organization 481 technological development 690, 692
ethnic categories ethnopoetics
nationalism 709 language 862, 870-1
ethnic election ethology
politics 712-13 animality /humanity comparison 22
ethnic minorities definition 480
division of labour 907 displacement activities 622
industrialization 906 primates 440
ethnicism etic models
politics 709-10 built environment 463, 481
ethnicity etiquette
cultural significance 346--8 language 879, 881-2
dissolution 711-13 euphoria
politics 706-28 music 696
survival 711-13 Europe
ethnie see also Old World
lateral/ vertical 713-17 colonial expansion 1041-2, I 044--5
modern nations 721-5 colonization 4, 297-9
politics 709-10 evolution debate 79-80
ethnoarchaeology Neanderthals 84--5
spatial organization 481 Upper Palaeolithic 92, 94
ethnocentrism evaluation
art analysis 655, 656 scales of 1019
human uniqueness debate 28, 29, 30 inequality 1020, 1021-3

violence 987-8 language 863
ethnochoreology 688 marriage 799
ethnogenesis pastoralism 213
politics 711 temporality 505-7, 509
ethnography witchcraft 572, 579, 581-2
anthropological role 370-1 evolution
dialogic anthropology 871-2 see also eo-evolution;
dispute processes 969 human evolution
inequality 1011-12 analogic drive 334
infancy 847 art analysis 659
law 965, 968, 972 behaviour 5-9
literacy 550 cognition 389-90
person-centred 832, 834 comparative method 1010-15
personhood 841 cultural 268-9, 941-9
poetics 870--1 human society 266--8
sex 822 hunting debate 86
socialization 837 knives 431-3
speaking 862, 866, 866-9 language 63-4, 100--1, 119-21, 127-9
verbal art 868 legal 965
Ethnological Society 142 monogenism 784
ethnology multiple-level181-6
dance 688, 690, 692 niche construction 162-92
symbolism 375 political 986
technology 446 recursive heuristic 185-6
1084
INDEX
revolution concept 442 exploitation

social 940--5 7, 1011-12 feudalism 914
subsistence systems 204 exploitative systems
symbolism 380 evolution 166, 168, 190
synthetic theory 162-5 extractive systems
theory 22 food production debate 10
tool-use 139 subsistence modes 204-6
evolutionary biology eye contact
definition 480 shared experience 360
kinship 774
neo-Darwinism 9 Fabian, J. 504
relationships 740--1 face
sociality 763-4 definition 879
theoretical debate 9, 30 facts
evolutionary ecology technical 424--33
development 756 Fahmy, Ismail 1000
evolutionary systematism Falk, D. 62-3, 146
classification 38-9 Fallers, L. 969
evolutionism fall owing
classification 442-3 subsistence modes 217-18
kinship 784 family
Ewing, K.P. 843 band society 944
exchange reciprocity 921
see also barter; socialization 836, 848-9
trading warfare 994
art analysis 664 work 747
food behaviour 253 episodic high mortality 287-8

generalized 926, 929 New World impact 302
marriage 801-3 farming
Marxism 913 see also agriculture
Neolithic 267 division of labour 907
non-commodity 917 labour 891-3, 895
power 915 social evolution 942
rate of 896 fasting
reciprocity 911-36 food taboos 248
restricted 926, 929 fatherhood
social life 737 see also paternity
social power 954--5 kinship 789-90
value 748 Fausto-Sterling, A. 813-14
exercise feasting
fertility 289 ethnicity 712
exogamy music 698
food taboos 246 social power 953
kinship 800-1 featural alphabets
social evolution 946 literacy 529
expenence Febvre, Lucien 436
belief systems 584 federalism
sharing of 359-61 ethnicity 713
experiential space 479 modern world 723
experiments Feld, S. 687, 697
belief systems 57 5 feminism
1085
INDEX
inequality 1023 colonization 102

language 876-7 food production debate 10
material culture 404 nutrition 234--5
sexual division oflabour 822 sedentism 205
Ferguson, C.A. 544 trade 234--5
fertility Fishman, A. 544
American Indian comparison 310-12 fitness
population growth 269, 289-91 genetic 176, 178
prehistoric populations 12 inclusive 164
sex 819 neo-Darwinism 162-3, 181
fetishism fixed feature elements
commodities 931 built environment 463, 491
feudalism Flannery, K. V. 208
exploitation 914 Fogelson, R. 841
hierarchy 1031 Foi of Papua New Guinea
lateral ethnie 713 myth 605-9
world order 1040 naming system 595-7, 598-601
fieldwork folk culture
art analysis 658 ethnicity 711
non-commodity exchange 917 folk medicine
sociality 756, 758 China 575
fighting folk taxonomies
law 972 hierarchical paradigms 374
figurative art folklore
see also art ethnography of speaking 866
analysis 665 literacy 533
cultural transition 87 food

Upper Palaeolithic, Europe 97 animal behaviour 765-7
film contemporary trends 253-6
technological development 690 cooking 243-5
fine art local variation 272
definition 648, 653 meals 250-3
Fingeret, A. 553 music 698
Finley, M.I. 450 processing 232
fire seasonality 239, 243
hominid diet 228 selection 237-8
natural selection 168 staples 239-42
subsistence 10 taboos 246--50, 249-50
tool-making debate 71, 72 traditional system 250
vegetation patterns 206 warfare 993-4, 994
First World War food avoidance see taboos
imperialism 1059 food production
Firth, Raymond definition 201
art analysis 659 human evolution 10
belief systems 57 5-6, 577, 582 food prohibition see taboos
food 226 foraging
labour 894 animal networks 739
language 862 broad spectrum 11-12
symbolic terminology 369 efficiency 226, 228
Fisher, Roger 998, 999 home-bases 205
fishing hunting debate 86
1086
INDEX
optimality theory 267, 277-83, 289 food taboos 246-7

prefrontal cortex 119 incest 800
sociality 7 59 magic 635-6
subsistence systems 200 Freyre, Gilberto 402
tool-use 136, 138 Fried, Morton H. 941, 944, 947
Ford, Gerald 627-8, 630, 632 Friedlander,J.S. 315
Forde, C.D. 442 Friedman, J. 949, 951, 953
forests Friedman, Milton 916
tropical 307-9 Friedrich, Paul 870
Forge,J.A.W. 659-60, 665,670-1,673 Frost, G.T. 145-6
form Fuller, C.J. 798
anthropological view 345 function
art analysis 657-8, 659, 662-4 art analysis 654, 662-4
cultural332 evolutionary classification 142, 155
definition 662 iconicity 386
evolutionary classification 142, 155 style 671
kinship 783-808 symbolism 369
music 695 technology 436
style 671 functional literacy
formalism India 546
anthropological role 370-1 functionalism
methodology 373-5 art analysis 655
symbolic role 372-3 evolutionary 763
Fortes, M. 794, 842 inequality 1019
Forty, A. 403 kinship 784
fossil records linguistics 862
human evolution 5-6 social evolution 942

Foster, Mary LeCron 333-4, 342, 366--92 technology 440
fostering temporality 504
social parenthood 793
Foucault, Michel399, 817-18,975 Gal, S. 870-1, 874
Fox, Lane see Piu-Rivers ... Galapagos Islands
Fox, R. 785-6 bird behaviour 167
France Galen 249
colonialism 1047-8 Gamble, Clive 6, 9, 79-102
ethnic bureaucracy 719 Gapun
fraternity literacy 539
doctrine 725, 726 Garine, lgor de
free occupancy alcoholism 12
spatial organization 482 human diet 226--56
free-market economies subsistence 10
specialization 906 Garusi (Tanzania)
freedom fossil records 56, 57
corporate interests 1057 Gatewood,J. 147, 151-3
material culture 414 gathering see hunter-gatherers
Freire, P. 548-9 Geertz, Clifford
French Revolution 880, 1059 aesthetics 698-9
frequency patterns agriculture 277
human speech 112 ethnicity 707
Freud, Sigmund ideological power 9 52
body 378 individuality 844
1087
INDEX
myth 594 organism-environment eo-evolution

small communities 693 176, 177, 178, 179, 180--1
symbolism 368, 370-1, 372, 375 population regulation 277-8
geladas replication 741
behaviour 768 genitalia
sociality 7 59 identity 817-18
Gellner, E. 542, 544 genocide
belief systems 581 colonialism 1048
ethnicity 708 ethnicity 714
inequality 1012-13 warfare 993
kinship 786-9 genotypes
gender human evolution 7, 8
biology 813-16 species variation 18
classification 816--21 geography
culture 813-16 innovation 445
division of labour 450-1, 899-900, geometry
901,902-5,906-7 built environment 461
human uniqueness 26 geopolitics
inequality 821-5, 1022-3 inter-state warfare 711-12
language 876-7 Germany
literacy 543, 545-7 Romantic Movement 724, 725, 864
material culture 404 gesturallanguage
music 699 hominid evolution 63-4
relationships 742-3 ghost marriages
sex 813-28 alliance 799
subordination 821-5 gibbons
kinship 785 early hominid evolution 37

general systems theory tool-use 135
ecology 175 Gibson, J.J. 471
generalization Giddens, A. 519, 839
analogical processing 382 gifts
belief systems 574 belief systems 576
symbolic categorization 367 cultural anthropology 416
generalized exchange division of labour 907-8
see also exchange food behaviour 235, 253
marriage 926, 929 labour 893
generalized reciprocity marriage 925-8
see also reciprocity political economy 748-9
gift exchange 922-3 wealth 915-28
genes Gilman, A. 879, 951, 954
altruism 772 Gilsenan, M. 400-1
culture 9, 164--5 Gioscia, V. 509
evolutionary biology 764 Girard, R. 633-4, 637
human evolution 6 glaciation cycles
selfish 762 Neanderthal migration 126
species 17-18, 25 Pleistocene 80-1
Genet,Jean 638 Gladysvale (South Africa)
genetics fossil records 56
epistemology 383 glasnost
inheritance 162-3 modern world 723
migration 92-3 Glassie, H. 152-3,401-2
1088
INDEX
Glassman, Ronald 943 Gregory, W.K. 37

global imperialism Griaule, M. 567,659
consolidation 1050--1 Grimaldi caves (Italy)
global template analysis Upper Palaeolithic 96
anthropological impact 390 groommg
globalization animal behaviour 768
cultural effect 347-8 Grotowski, Jerzy 623, 641
Gluckman, Max 968, 969 groups
Godelier, Maurice 896, 929-30, 949 inequality 1020--2
Goehr, L. 649 structure 794--7
Goffman, Erving 1055 guilt
gold socialization 851
colonialism 1046--8 Gulliver, P.H. 969
Gona River (Africa) Gusii, Kenya
fossil sites 70 division of labour 901
Gi:innersdorf (Germany) Guss, D. G. 664
tent remains 97 Guthe, E. 226
Good, A. 788
Goodall,J. 112, 118, 156, 351 Haas,Jonathan 947
Goodenough, W.H. 66 Habermas, J. 504
Goodman, N. 670 Hadar (Ethiopia)
goods fossil records 55, 57, 70
political economy 748-9 Hiigerstrand, T. 515
theory of916-17 Haiti
wealth 915-28 imperialism 1059
Goody, E.N. 797 Halliday, M.A.K. 862, 875, 877, 878, 881
' . ' ' '
Gordon, David 906 Hallpike, C.R. 564-5
gorillas hallucinogens
classification debate 39 music 696
fossil records 56 Halprin, Anna 641
hominid evolution 34, 41, 50, 51 Hamilton, W.D. 164, 191,772
sociality 7 57 Hammond, M. 91
tool-use 135 hand movements
Gould, Richard A. 619 cortical control127-9
government Neanderthal125
colonialism 973-4 hand axes
law 964,966 Acheulean 81
gradualism H. erectus 82
human uniqueness debate 26--8 handedness
Granet, M. 245 anatomical features 14 5-6
Graves, P.M. 102 Handelman, Don 639
Gray, John E. 36 hands
Great Divide model tool-use 145
literacy 532, 536 Hanuno6
Greece writing system 531-2, 543
ethnicity 710, 715, 716 Harpending, H. 286
slavery 450--1 Harries-Jones, P. 504
Green Revolution Harris, G.G. 842
agriculture 1061 Harris, J. W.K.
Greenhouse, Carol 996 stone tools 70
Gregory, C. 746, 911-36 Harris, M. 141,246,277
1089
INDEX
Harrison, Tom 411 human evolution dichotomy 7, 9

Hart, K. 928 literacy 533
Hart, Mic key 701 template semantics 378
Hassan, F. 290 temporality 514
Haua Fteah (Cyrenaica) Hittites
stone technology 94 lateral ethnie 713
Haudricourt, A. G. 420 Hobbes, Thomas 966, 1011, 1020
Haugen, E. 872 Hobson, J.S. 1060
health Hockett, C.L.F. 64
belief systems 57 3 Hoebel, E.Adamson 965,969-70
sedentism 282-9 Hogbin, H.J. 968
warfare 993-4 Holland
hearths colonialism 1048
Old World 83-4 Hollis, M. 582
Upper Palaeolithic, Europe 92 home-bases
Heath, S.B. 535, 551 foraging 205
Heidegger, Martin 511, 521, 864 spatial organization 469
Heider, K.G. 423 home-ranges
Hellman, M. 37 territoriality 486, 487
Hennig, W. 38 homeland see territory
Herder,Johann Gottfried 725 homicide
herding American Indian comparison 312-13
see also animal domestication hominids (Hominidae)
nutrition 23 5-6 ancestra13
Herskovitz, Melville 659, 687 biological features 61-3
Hertz, Robert 374 classification 58-61
Hewes, 'G.W. 63 cultural factors 67

hierarchy culture 66--7,68,69-72, 380-6
division of labour 904, 905 ecology 69
gifts 918 evolution 33-72, 37, 38,49-51, 52,
language 880-1 53
political domination 940 food 226, 227-8
social evolution 1013-14 fossil sites 49, 56
Higham, C. 206 fossils 48-50, 53-6, 54
Hill, C. 553 likeness concept 372, 391
Hinde, R.A. 758 mimicry 368
Hinduism morphological features 39-48
caste system 1025-8 spatial organization 468-9
colonial expansion 1041 stone tool-makers 69-72
cooking 245 subsistence systems 203
egalitarianism 1032 Homininae
food 237,247,412 evolution 37
literacy 537 hominization
Hippocrates 249 teeth 44-5
historical geography Hominoidea
definition 480 classification 35-6
historicism early hominid evolution 35, 38, 50,
nation formation 720 51
history Homo
animal comparison 350-63 classification 58-9
anthropological comparison 521 Homo erectus
1090
INDEX
archaeology 83-4 cosmology 506, 516

brain size 771 environmental adaptation 309-13
chronology 81 metaphors 865
colonization 81-2 time 864
cranial centre of gravity 47 Hoppius, C.E. 15, 16
endocranial casts 46 Horace 613
evolution 79 horizontal commodities
hominid evolution 36, 37 markets 932
hunting 85-6 hormones
language 99 violence 813
morphology 61, 62 Horsman, J. 546
vocal tracts 126 horticultural societies
Homo habilis evolution 943
brain size 771 horticulture
Broca's area 99 development 208, 215
endocranial casts 46, 62 Sahara 102
evolutionary divergencies 50, 53 Horton, R. 565, 586
spoken language 8, 63 Hosmer, Stephen 988
stone tool-making 69-71 hot foods
symbolism 372 taboos 249-50
technology 440 housework
tool-use 133 labour 892 housework (continued)
vocal tracts 66 political economy 747
Homo sapiens housing
archaeology 83-4 Third World 723
art 384 warfare 994
Broca's area 118, 119 ' Poort industry

Howiesons ' '
chronology 81 South Africa 94
cultural factors 66--7, 68, 69-72 Hoxne (England)
cultural transition 88-92 cultural traditions 84
culture debate 380 Hrdlicka, A.
demographic expansion 265-91 art 98
endocasts 62-3 colonization 101, 102
fossil records 49 evolution debate 80
Linnaean classification 35 Hubert, H. 575-6
morphological features 39-48 Huffman,~. 139-40
symbolism 380, 384 Hugh-Jones, Stephen 602-3, 604
tool-use 145-6 Hughes, A. 774--5
Homo sapiens neanderthalensis human ecology
see also N eanderthals definition 480
human species 3 human evolution
Homo sapiens sapiens see also evolution
Broca's area 8 subsistence 10-13
cranial centre of gravity 47 human geography
human evolution 3-5, 6 definition 480
species variability 17-18 human hands see hands
Honneth, A. 522 human populations
honorifics see also populations
definition 879 calorie requirements 232
Honorius 625 diet and nutrition 226--56
Hopi protein requirements 229
1091
INDEX
human sciences behaviour comparison 228

infrastructure 187-90 definition 199
humans H. erectus 85--6
anthropocentrism 15, 25-30 performance 620--2
classification debate 39 prefrontal cortex 119
concept of 23-4, 26 residence rules 796
condition definition 15, 19-25 Upper Palaeolithic 204, 209
disease 4 husbandry
evolution 3 see also animal husbandry
H. s. sapiens 3-5 domestication comparison 11
language 20, 63-4 Hutchins, E. 577
molecular biology 38 Huxley, Thomas 28
sociality 7 56-77 hygiene
species definition 15-19 social structure 403
spoken language 3 Hylobatinae
theoretical debate 14--30 early hominid evolution 37
tool-use 3 Hymes,I>. 866,867
humming hypertension
non-verbal communication debate 64 disease 12
Hunsgi (India) Pacific island societies 315
Acheulean 84
Hunt, George 593 . . .
ICOlllCSigns
hunter-gatherers tools 156
bands 944 iconography
behaviour debate 275--6 anthropological study 343--6
categorization 442 art analysis 655, 658, 665
619-20 definition 369, 389

colonization 102 linguistic motivation 390
definition 200-1 literacy 528
demand sharing 924 prehistoric art 384-5
disease 285, 298 symbolic evolution 391
efficiency studies 279 identity
egalitarianism 1012 ethnicity 709-10,711,714-15
famine 287-8 genitalia 817-18
fertility rates 289-91 indexicals 878
food production debate 10 interaction 878-83
food-collecting populations 200--4 language 861, 872
group size 770 linguistic styles 87 5-8
hunting debate 86 masks 663
incest 800 music 697,699,699-700
nutrition 23 3-4 national1058
Palaeolithic 7 personhood 744--5
population debate 284--6 physical resemblance 741
population growth 271 sexual743
population size 11 skill-producing groups 448
social evolution 942, 943 socialization 831-54
social order 752 temporality 411-13
spatial organization 469 ideo-logique
subsistence modes 197-221 belief systems 574
hunting ideology
animal behaviour 765, 766 art analysis 664
1092
INDEX
belief systems 583 India

dialogic anthropology 871 caste system 1014, 1025-8
distribution 1032 colonialism 1051
equality 1031-5 division of labour 903
individualism 745, 1016 European imperialism 1057
kinship 784, 785 Mutiny 1049
language 862 indigenous populations
law 976 see also populations
legal 972 disease 297-320
literacy 339, 533, 534 indirect rule
material culture 404-5 colonialism 973-4, 1051-2
nationalism 725, 727, 1040 individualism
power 749, 952 inequality 1015-17
spatial organization 489 Western model 844-6
imitation individuality
analogical processing 382 identity 843-6
culture 334 individ uals
ontogenesis 386 personhood 744 individuals
social transmission 352-4 (continued)
immune systems social life 738
genes 774 indoctrination
immunization belief systems 583
warfare 994 experience 582-7
immunology industrial revolution
clonal selection 185 colonialism 1049
phenotypes 163 industrial societies
Im erialism
botanic 1049 social evolution 1013
colonialism 1048-50 industrialization
global1050-1 division of labour 905
nationalism 727 European imperialism 1056
non-commodity exchange 917 labour 898
Third World 722 occupations 1028-30
impoverishment see poverty temporality 338, 515
imprinting work 747
phenotypes 163 inequality
improvisation common historical forms 1021-31
music 689 comparative 1010-15
neurology 343 gender 821-5
ritual613 market classification 935
Inca
sex 821-5
survival rates 306
study of 1015-21
incest
infancy
kinship 800-1
socialization 847-50
marriage 927
infant mortality rates
income
prehistoric populations 12
inequality 1010, 1014
infections see disease
indexes
infinity loop model
definition 369
drama types 629, 630
symbolism 380-1
inflation
indexicals
health impact 301
definition 878
influenza
status symbols 155-6
1093
INDEX
Amazonia 308 International Congress of Prehistoric

disease 12 Archaeology 142
New World impact 299 International Union of Anthropological
information and Ethnological Sciences 226
social transmission 351-4 interpretivist paradigms
lngold, Tim relationship terminology 784
hominid evolution 34 interventionism
humanity USA 1059
animality distinction 14-30 intifada
introduction 3-13 normative power 992
hunting definition 199 intonation
pastoralism definition 211-12 human speech 112-13
reciprocity model 925 lnuit
temporality 504 diet 229, 233
inheritance hunting debate 86
culturall88 snow 398
ecologicall65, 178, 179 subsistence systems 202
selfish gene theory 762 invention
initiation technology 421
belief systems 583 involution
hunting 622 agricultural intensification 219
innovation Iran
adaptation 442-5 ethnicity 714
music 700-1 literacy 541
instincts political discourse 1001
behaviour 563 Ireland
likeness ex loitation 391
institutionalization national languages 874
symbolism 367 Iroquois
instrumental meanings confederacy 1044
environment 337 kinship 803, 805
instrumentalism irrigation
ethnicity 346, 707 health impact 301
politics 706-9 social evolution 946
integrationism Isaac, Glyn 63, 83, 205
social evolution 941,943-7,949-50 Isernia (Italy)
intellectuals see intelligentsia H. erectus 82
intelligence hunting 85
human 360--1 Islam
intelligentsia belief systems 569
nation formation 718, 720, 722
egalitarianism 1032
intentionality
ethnicity 712, 716
non-human animals 381
literacy 537
social transmission 352, 355
prayer 694
socialization 837
spice trade 1045-6
symbolism 381
Israel
technical424
Camp David agreement 999
interest
communicative style 1002-3
Marxism 913
ethnicity 710
internalization
negotiations 997, 999-1000
socialization 836
international affairs
ethnocentrism 987-8 Jackson,Michael603-4,607-8
1094
INDEX
Jainism Kant, lmmanuel25, 505

food taboos 247 Kaplan, B. 312
Jakobson, Roman 243, 865, 866--7 Kardiner, A. 226, 243
Japan Karolta (Australia)
dance analysis 694 rock art 98
imperialism 1059-60 Katz, S.H. 237
language 881 Keali'inohomoku, J. 688
writing system 531 Kebara cave (Israel)
Jarmo (Iraq) fossil records 93
agriculture 236 Keeler, W. 823
jati Keesing, R.M. 227
caste system 1026--7 Keller, C. and J. 149, 151, 153
Java Kenya
colonialism 1049 language 862, 874
language 882 Kerby, P. 849
sex 823, 824 Keynes, John Maynard 919
Jebel ~zfeh (Israel) see ~zfeh ... Kgalagari
Jelliffe, D.B. 239 marriage 799
Jericho Oordan) Khazanov, A.M. 211
agriculture 236 Kidder, R.L. 977
animal domestication 235 Kim,Samuel990-1
Jevons, W. 916 kin selection
jewellery inclusive fitness 164, 191
cultural transition 87 theory of 772-3
Jews King, A. D. 476
ethnicity 710, 715 kings
vertical ethnie 714 Kingsbury, H. 688

Joas, H. 522 kinship
Johnson, Alien 941, 944--5 adjudication 965
Johnson, D. 312 biology 785-92
Johnson, G. 273 causation 796-7
Johnson,M. 863,865 conceptual debate 786--9
joking content 783-808
relationships 861 culture 786--9
judgement descent 794--5
pedagogy 355 exchange 801-3
jurisdiction exogamy 800-1
territoriality 486, 487 fatherhood 789-90
food taboos 247
Kafka, Franz 638 form 783-808
Kagan, J. 850 freedom debate 414
Kalaunas future study 807-8
myth 606--9 gift exchange 923
Kaluli group structure 794-7
catharsis 697 incest 800-1
Kanapoi (Kenya) language 861
fossils 55, 57 law 966
Kangaki (Kenya) marriage 798-803
fossil site 57 motherhood 790-2
Kangatukuseo (Kenya) organization 771-5
fossil site 57 prohibitions 783-808
1095
INDEX
property 797 Kurds

reciprocity 921, 924 ethnicity 715
recognition 773
relationships 741-3 La Chapelle-aux-Saints (France)
relatives 803-6 skull and mandible 123, 124
residence 794--5 La Cotte de St Brelade Oersey)
rules 783-808 hunting 86
sedentism 267-8 La Fontaine, J.S. 841
social life 7 37, 740 La Marc he (France)
social parenthood 792-4 art 97
social power 955-6 Labanotation
stateless societies 1043 dance 691, 693
structuralist approach 374 labour
subsistence systems 203 agricultural technology 270--1, 277
totemism 595-6 co-operation 891-7
Western folk mode1816 division of 1012
Kirchhoff, P. 804-5 exchange-value 913
Kirkpatrick, M. 179 market structure 905-8
Klasies River Mouth (South Africa) measurement systems 897-9
fossil records 94 slavery 1047
Knight, C.D. 800 temporality 513
knives
Labov, William 875, 876
concel?t of 430, 431-3, 434, 437 lactation
evolutiOn of 431-3
fertility 289
knowledge
lactose tolerance
see also cognition
natural selection 167
beliefs stems 565-6
pedagogy 333
acquisition device
technical438-42
Laetoli (Tanzania)
technography 336
fossil site 57
theories of 504-5
Lafitau, Joseph 803
Koch, Klaus-Frederick 985
Lafitte, J. 423-4
Kohn, Hans 717
Laitman,J.T. 121-2
Koobi Fora (Kenya)
Lake Mungo (Australia)
cultural traditions 83
burials 97
fossil site 57
hominid diet 227-8 fossil records 94
stone tools 70 Lake Turkana (Kenya)
Ki:iping, Nicolas 15-16 H. erectus 126
Koranic literacy Lakoff, R. 863, 865, 877
Liberia 534, 535 Lamarck, Jean Baptiste
Krebs, J.R. 172 acquired inheritance 184
Kroeber, A.L. 803-4 evolutionary change 142
Kromdraai (South Africa) genetics 163, 190
fossil records 37, 56, 57 land charters
Kubler, G. 666 ethnicity 712
Kuechler, S. 414 land tenure
kula social power 953-4
art analysis 664 Lande, R. 179
wealth 916, 918, 928 Landes, D.S. 441
Kunitz, StephenJ. 12, 297-320 landscape architecture
Kuranko definition 480
myth 603-4, 607-9
1096
INDEX
landscape ecology animal domestication 210

definition 480 Late Stone Age
matrices 486-7 Africa 94, 96
Langer, S. 407 latent functions
language spatial organization 470
see also spoken language lateral ethnie 713-15
acquisition device 153 Latin American Indians see American
acquisition support system 153 Indians
American English 863 law
analogic drive 334 ancient Greece 585
anthropological study 331 anthropology of969-73, 976-9law
arbitrariness 862 (continued)
art 99-100 colonialism 973-6
definition 862 dispute processes 962-79
ethnicity 707 equality before 1028
evolution 63-4, 100-1 modern 963-6
generalization 367 order 966-9
handedness 146-7 primitive 963-6
human uniqueness debate 20, 26--8 reciprocity 966--7
inequality 1017 state organization 750, 752
literacy 527, 535 Layton, R. 650
material culture 406 Leach, Edmund 19, 374, 571, 789, 798,
meaning 370 802
motor speech areas 63 leadership
music 345-6, 694 civilization 274
myth 591-5,601,603 control940
negotiations 1002 social evolution 944, 946

origins 108-29 Leakey, L.S.B. 59, 69
pedagogy 333, 360, 387 Leakey, Mary 63,69
politics 872-4 learning
prehistoric art 386 apprenticeship 151-2
primordial 389 child development 746
reconstruction 387-9 literacy 535
reference 377 memory 147-8
semiotics 369 ontogenesis 386-7
social aspects 861-83 phenotypes 163
social relationships 770 social transmission 351-4
socialization 746--7 socialization 833, 837-8
symbolist theory 373 strategy 371
totemism 341 tool-use 139
world view 863-5 trial-and-error 762
writing 408 Lee, Richard
langue cultural models 271
structuralism 865 diet 234
larynges population growth 275
human 111, 113 sedentism 272-3
Lascaux cave (France) Leenhardt, Maurice 594
art 384 legal pluralism
LASS see language anthropology of 97 6--9
acquisition support system normative regimes 977-8
Late Neolithic Leibnitz, G.W. 423
1097
INDEX
leisure Pacific island societies 314

labour 891-7 palaeopathology 283
Lemonnier, P. 154-5 post-war rates 301
Lenin, Vladimir Ilyich 1059, 1060 life-cycles
Lenski, Gerhard 943, 950, 1013 food taboos 248
LePage, R. 872 material culture 335
Lerner, D. 548 Limber, J. 66
Leroi-Gourhan, A. 420, 421, 430, 432 Lindly, J. 91
letter writing linearity
literacy 539 temporality 506, 517-19
Levallois technique Western time 509, 511
Middle Palaeolithic 94 linguistics
Neanderthals 85 functional 862
Levant identity 875-8
N atufian, cultivation practices 206 literacy 536
Levi-Bruhl, L. 572 post-structural analysis 406
Levi-Strauss, Claude relativism 864
alliance theory 802, 804 Linnaeus, Car! 563
art analysis 660 critical interpretation 4-5
bricoleur 149, 153 hominid evolution 33-4, 35, 37
cooking 243 human uniqueness debate 27
food 226-7 tailed men 15-16
food behaviour 253 Linton, R. 226
food selection 23 7 literacy
food taboos 246 adult 552-3
gift exchange 920, 925-8 aspects of 527-54
language 865' consequences 532-4

modern/traditional debate 504, 516-- development 547-50
17,519 education 550-4
myth 593-5 environment 337
myth reconstruction 390 ethnicity 713
ordering process 401 gender 545-7
sex 820 invented 540
structuralism 370--1, 372, 374, 375 nationalism 542-4
technology 422 practices 540--2
templates 378-9 resistance to 537-9
totemism 595-8, 600, 602, 604 social impact 339-40
Levine, R.A. 846 socio-cultural construct 532-7
Levinson, S. 879 sociopolitical processes 542-54
Lewis, Gilbert 340, 563-87 spread of 537-42
Lewis-Williams, J.D. 98 technology 528-32
Lewontin, R.C. 166, 177 types 53 4--6
Lex, Barbara 638-9 litigation
liberalism courts 970
colonialism 1053 Llewellyn, Karl969-70
Liberia Lloyd, G. 585
Koranic literacy 534-5 local subsistence
Lieberman,P 8,63,65,99-100,108-29 system comparison 220--1
life expectancy Lock, A. 391
American comparison 304 Locke,John 1011, 1054
hunter-gatherers 285 locomotion
1098
INDEX
hominid 43-4, 61 Freudian view 635

human speech 119,121 myth 605-10
logographic writing nature 578-80
literacy 529 Mahayana Buddhism
Lomax, Alan 688-9 European colonialism 1058
Lomekwi (Kenya) Maine,ll. 797,947,963,965,970
fossil site 57 Makapansgat (South Africa)
longevity fossil records 56, 57
artefacts 410-11 maktab literacy
longitudinal ethnie 713-15 Iran 541
Lorenz, Konrad 756, 984 malaria
Lothagam (Kenya) colonization policy 300, 302, 307,
fossil records 53, 55, 57 308
Lovedu natural selection 167
marital ties 799 Malaysia
Lovelock,J. 173 sex 826-7
Lower Palaeolithic Malinowski, Bronislaw
hunting debate 86-8 belief systems 576
Lowie, Robert 796, 804-5, 1012, 1022 economics 928-9
Lubbock, Sir John 20, 442 equivalence 919
Lukeino (Kenya) fatherhood 789, 793
fossil records 53, 55 food 226
Lukens-Wahrhaftig,J. 991 gift exchange 920
lunar cycles kula exchange 916
temporality 514 language 861
law 963, 965, 966-8, 969, 970
'
non-commodity exchange 917
fossil records 55
McAllister, David 687, 688 technology 445
McAllister, L.N. 302 malnutrition
macaques disease 12
animal behaviour 769 food selection 237-8
potato-washing behaviour 229 warfare 993, 994
social transmission 356 Malthus, Thomas 269,943
sociality 7 59 mammals
tool-use 135-6, 139 animal behaviour 766
Macaulay, D. 437 mimicry 368, 391
McGrew, W.C. 137 reproduction 765, 774
Machiavelli, Niccolo 1040 sociality 757
machine space symbolic behaviour 382-3
definition 478 mammoth
MacKenzie, M.A. 664 hunting debate 85, 86
McLennan, J.F. 800, 803 'Man the llunter' paradigm 203
McNeill, W.R 298 man-made disease
macro-sociology see also disease effects 12
ethnicity 706 mana
Madagascar concept of 57 5-6
law 966 Mantoux, P. 442
Magellan, Ferdinand 1046 Maori
magic belief systems 576
belief in 563-87 maps
Foi spell formula 605-6 spatial organization 468, 484
1099
INDEX
marginalism massacres
political economy 748 ethnicity 714
wealth 916 Matenkupkum cave (Pacific)
markets fossil records 94
classification 931-6 material culture
social order 752-3 aesthetics 673
trading 9 31-6 art analysis 656
weaving902 definition 334-5, 398-9
Marquesas freedom 414
double-headed club 662, 663 gender 404
marnage mass consumption 405-6
definition 798-803 material resources
gift exchange 925-8 inequality 1010
housework 892 normative power 990--2
lateral ethnie 714 materialism
prescriptive 801-3 subsistence patterns 197
templates 379 materiality
Marshack, A. 99 artefacts 406--8
Martin, Debra 283 maternity
Marx,Karl gender 743
belief systems 583 mating
division of labour 904 dialects 127
equality 1011 reproduction 764
fetishism 931 matriarchy
forces of production 141 social evolution 1022
hierarchy 1031 matrices
industrial revolution 442 landscape ecology 486-7
inequality 1020, 1030--1 temporality 508
instrumentalism 707 matrilineal descent
labour 895-6 belief systems 571-2
modes of production 204 inequality 1020--1
non-commodity exchange 920 residence rules 796--7
religion 632 rules 794--5
small-scale societies 416 Mauer (Germany)
social evolution 941, 942, 946 mandible 84
technology 143,441 Maurer, David 878
wealth 912-15,922,928 Mauss, Marcel
Marxism artefacts 416
structural 949 belief systems 57 5-6
Marzke,M. 145,440 collective representations 1018
masks exchange 911,927,929
art analysis 663, 666--7, 676 food behaviour 253, 256
structuralist study 40 I gifts 893, 919-20, 925
Mason, O.T. 423, 446 music 687
mass consumption social evolution 94 7
freedom 414,417 socialization 842
material culture 405-6 technology 422, 423, 424--5, 449
Mass Observation 411-12 Maxwell, R.J. 503
mass production Maya
political economy 747 population collapse 266
1100
INDEX
survival rates 306 folk 575

temporality 514 food taboos 249
Maybury-Lewis, D.H.P. 801 palaeopathology 278
Mbuti Meek, M. 552
ceremonial centres 618-19 megafauna
Mead, Margaret 550, 814 hunting debate 86
food 226 Meigs, Anna 824
inequality 1022 Meinig, D.W. 494
socialization 831 Melanesia
meals see also Papua New Guinea
definition 250 Melanesia (continued)
sharing rituals 250--3 disease impact 314--15
meanmg Mellars, P.A. 95
see also semiotics melogram
anthropological role 370 song recording 692, 693
art analysis 658, 659, 664, 667 memes
artefacts 396-17, 397 bird behaviour 185
built environment 473-4 inheritance 763
definition 368-9 modern synthesis 9
derivation of 369-70, 377-8 memory
early 389 folk culture 711
gender 743 literacy 533, 535
latent functions 470 tool-use 144, 147
levels of 473 men
music 686-7,689,695 division of labour 899-900, 901,
organization 465-6 902-5,906-7
prehistoric art 385 inequality 1022-3
spatial organization 469 relationships 742-3
symbolism 366-7, 376-7, 380 sex 814-15, 821, 824
measles Mende
Amazonia 308 art analysis 666--7, 67 5-7
disease 12 Mendel, Gregor 179
New World impact 298-9, 302 Mendelian genes
meat-eating molecular biology 762
early hominids 203 merchants
foraging 228 European expansion 104-5
herding 235 social evolution 942
nutrition debate 280 Merriam, Alan 687, 688
mechanical solidarity Merry, S.E. 978
division of labour 903 Mesoamerica
inequality 1016 literacy 528
law 965,967 Mesolithic
mechanics animal domestication 210
laws of 512 broad-spectrum revolution 278-9
mechanology population 284, 289
definition 424 seed-processing tools 266
Medes trade 220
lateral ethnie 714 Mesopotamia
mediation ethnicity 709
law 971 literacy 528
medicine metacommunication
1101
INDEX
definition 623 minority groups

negotiation 1001-2 language 872-3
metallurgy missionanes
social evolution 942 colonialism 1053
metaphor missionization
American English 863 literacy 537-8
belief systems 568, 577 mitochondrial DNA
concordant structure 379 migration 92-3
language 862, 865 Mivart, J. 35
music 697 mnemomcs
myth 591-610 spatial organization 462, 493
prehistoric art 386 mobile traders
role 341-2 market classification 933
symbolism 373, 376-7, 380, 390 mobility
Mexico see also movement
literacy 544 social1030
microcosms spatial organization 474, 49 5
symbolism 376-7 subsistence systems 204-5
Middle East mobilization
political realism 990 ceremonies 698
Middle Palaeolithic 9 5 mode of production see production .. .
cultural transition 93 mode of subsistence see subsistence .. .
France 95 modern humans
hunting debate 86--8 see also anatomically modern humans
stone technology 94 behaviour 93
tool manufacture 383 brains 62
Middle Stone Age classification 60
Africa 94 culture 329-48
migration demographic expansion 265-91
African labour force 301 fertility rates 289-91
American Indians 310 high mortality 284-8
language divergence 389 intelligence 360-1
Pacific island societies 314, 316--17 pedagogy 354--7
Third World 722-3 population growth factors 283-4
Mikulecky, B. 546 population pressure 271-4
military power replacement/ continuity debate 88,
politics 749 90,91-2
social evolution 956 social evolution 266--8
sources of951-2 symbolism 369-92
technology 990 training 359-60
Mill, John Stuart 1054 modern law
Miller, Daniel335-6, 337, 344, 396-417 primitive comparison 963-6
mimicry modern societies
analogic drive 334, 382 anthropological study 504
mammals 368, 372, 391 belief systems 564
social transmission 353 modern synthesis
mind cultural variation 9
theories of 359-61, 850 modern world
mmmg political realism 751
European imperialism 1057 modernism
non-infectious disease 314
1102
INDEX
modernity classification of relatives 803-4

definition 708 evolution 1011
law 964--5 kinship 784
molecular biology law 965
clonal selection 185 property 797
cultural transition 88 social evolution 942
hominid evolution 38 Morishima, M. 896
human evolution 6 morphological change
Mendelian genes 762 human evolution 6
palaeontology 56 Morphy, Howard 344, 648-78
molecular clocks mortality
fossil records 48-9 American Indian comparison 310-11
molecular dating population pressure 284-8
human evolution 6 post-war rates 301
monarchies prehistoric populations 12
European expansion 1044 mosaicism
world order 1040 evolution 61
Monboddo, Lord Games Burnet) hominization 53
humanity concept 23-4 motherhood
orang-utans 20 kinship 790-2
tailed men 15-17 motor accidents
money economy American Indian comparison 312-13
gift exchange 929 Mousterian
money-lending art 384
Marxism 913 cultural transition 95-6
Mongols movement
ethnicity 709, 714 see also mobility
herding 236 spatial organization 487
monkeys Muller, Johannes Ill
analogical processing 382 Miiller, Max 594
animal behaviour 766, 768, 769 multi-level eo-evolution
genetic relationships 773 ecological inheritance 187
pedagogy debate 359 supplementary processes 181-6
social networks 739 multinational corporations
social transmission 352, 353, 355-6 Third World 1063
sociality 758, 759 Mumford, Lewis 141,421
tool-use 135 mumps
monocroppmg Amazonia 308
colonization 300 Munn, N.D. 659-60, 664, 665, 677
monogamy Murdock, G.P. 793, 796, 805
animal behaviour 765-6 Muruyur (Kenya)
monogemsm fossil records 55
definition 784 mUSIC
monopolies anthropological theory 331
law 964 catalogues 700
monuments community identity 699-700
longevity 410 cultural significance 343-6
Moore, H.L. 813-28 dance 686-702
Moore, S.F. 977-8 definition 694-6
Morenhout, J.A. 246 industry 700--1
Morgan, Lewis Henry innovation 700--1
1103
INDEX
performance 698-9 modern world 721-5

physiological effect 696--8 rise of I 040--1
production 698 world order 1040--63
Suya 693-701 nationalism
technology 690-3 anthropological study 346-7
mutualism collective violence 987
altruistic behaviour 772-3 colonialism 1049, 1053
Myrdal, Gunnar 989, 1025 ideology 1040
myth liberal105 8
anthropological theory 330 literacy 542-4
charter 592-5 politics 706-28
convention 592-5 retrospective 708
cultural knowledge !53 role 725-8
cultural significance 340--3 world order 1059-61
ethnicity 710, 711 Natufian
form 598-602 Levant, cultivation practices 206
magic 605-10 natural
metaphor 591-610 definition 813
naming systems 595-8 natural history tradition
seasonality 243 tool-use 137-40, 143-4
spatial organization 469 natural selection
structural analysis 602-5 see also selection
temporality 514, 515 Darwinism 5
totemism 595-8 lactose tolerance 167
mythic transformation neo-Darwinism 162
structuralism 379 relationships 741
mythomoteur self-induced 165-8
politics 716-17 sociali ty 761
nature
Nadel, S. 578 belief systems 579
Nagandong Oava) gender 743
H. erectus 84 human 21
Nambiquaras magic 578-80
population estimate 309 ritual615
nammg state of 21
abstraction 369 Navaho
ontogenesis 386 apprenticeship 154
totemism 595-8 art analysis 664
Napier,J. 69 environmental adaptation 309-13
Nariokotome (Kenya) language 390, 865
fossil site 57 learning behaviour 147
narrative ontogenesis 387
hunting 621 navies
nation states world order 1040
see also state Nayar
building of 1061-3 marital alliance 798
collective violence 990 Neanderthals
formation 717-21 see also Homo sapiens neanderthalensis
ideology 1057-9 burials 96-7
inequality 1015 catastrophism 90--1
modern security 751,752 chronology 81
1104
INDEX
cultural traditions 84-5 technical426

cultural transition 88-92 Upper Palaeolithic 96--9
evolution debate 79 neural networks
human species 3 properties 109
language 99 speech production 128
Mousterian 9 5 syntax 108
vocal abilities 63, 65, 123, 124, 125-7 neurobiology
Needham, R. 373-4, 390, 786-9, 798, 802 socialization 837
Neel,J.V. 308 neurology
negative reciprocity improvisation 343
gift exchange 922, 923-4 performance 638-9, 640, 641
negotiations New Guinea see Papua New Guinea
bilateral 970 New World
culture 994--5 see also America
dynamics of 996--9 conquest of 1046--8
principled 998-9 inequality 1024
social life 750--4 Old World impact 299-302
study of995-1003 New Zealand
warfare 988 migration 317
neo-Darwinism Newman, Stanley 877-8
evolutionary biology 9 newspapers
fitness theory 162 literacy 540
subsistence systems 200 N galoba (East Africa)
neoclassical economics anatomically modern humans 100
barter 928 N gorora (Kenya)
division of labour 906 fossil records 56
wealth 748, 915, 916, 929-30 Nharo
Neolithic marriage 798-9
animal domestication 210 Nicaragua
burial grounds 952, 954 literacy 548, 549
contraception 290 niche construction
innovation 445 evolution 162-92
life expectancy 282 phenotypes 165-81
plant domestication 267 niche destruction
technology 1017 definition 178, 191-2
tool-use 136 Nietszche, F. 862
Neolithic revolution 101 Nishida, T. 355
concept of 442 Nixon, Richard M. 627
population debate 283-4 Noble, W. 99-101
social evolution 942 nobles
neolocal residence lateral ethnie 713
rules 795-6 nomadism
N ettl, Bruno 687, 688 definition 211
networks hunter-gatherer comparison 233-4
cultural transition 87 sedentary politics 213-14
early hominid trading practices 220 spatial organization 47 3-4
language 100-1 non-verbal communication
social 739 singing debate 63-4
sociology 760 Normans
subsistence 12 ethnicity 709
1105
INDEX
normative power Olduvai (Tanzania)

collective violence 990-1 built environment 460, 469
social order 751 fossil site 57, 70
normative regime H. erectus 81
legal pluralism 977-8 hand anatomy 145
norms hominid diet 228
cultural! 000 oligarchy
definition 373 law 964
food behaviour 253 Omaha system
North American Indians see American relationship terminologies 805, 806
Indians Omo (Ethiopia)
nuclear families anatomically modern humans 100
social parenthood 793 fossil site 57
nuclear war hominid diet 227
security 983 mitochondrial DNA 93
Nuer one-parent families
language 863 social parenthood 793
marital ties 799 Ong, Aihwa 826--7
political system 1021 ontogenesis
Nukulaelae symbolic behaviour 386--7
literacy 539 ontology
nut-cracking Western 743
social transmission 356-7 operations
Nutels, N. 308 analysis 425-6, 427-9, 430
nutrition concept of 434
chronological comparison 280--3 opmm
disease 12 colonialism 1050
human populations 226-56 opportunity
hunter-gatherers 275--6 equality of 1032
palaeopathology 278 individualism 1016--17
inequality 1028
objectification opposable thumb
artefacts 415-17 human uniqueness debate 27
observation optimal foraging theory
belief 575-8 demographic expansion 267,277-83
pedagogy 355 fertility 289
occupations subsistence populations 200
inequality 1028-30 orang-utans
Ochs, E. 868-9 classification debate 39
Odling-Smee, F.J. hominid evolution 34, 50
cultural factors 69 humanity concept 20, 24
human evolution 9 tool-use 135
niche construction 162-92 order
Oedipus complex see also categorization;
symbolism 378 classification;
Ojibwa Indians 24---5 taxonomy
Old World law 965, 966--9
see also Europe meanings 399-406
New World impact 299-302 social life 750--4
Oldowan culture organic solidarity
stone tools 69, 82
1106
INDEX
division of labour 903-4 palaeopathology

law 965 fertility 289
orgamsms optimal foraging theory 277-83
environment feedback cycles 167-8, prehistoric populations 12
169-71, 172-6 palaeosoils
environmental eo-evolution 176, 177, Dushanbe 82
178, 179, 180--1 Palestinians
environmental relationship 9 intifoda 992
species variability 17-18 negotiations 997, 999-1000
organization Pan 34, 50
space 483-6, 487, 488-9 pandemics see disease
Ortiz, Sutti 747, 891-908 Panslavism
Ortner, Sherry 821, 822-3 doctrine 725
Oswalt, Wendelll43 Papua New Guinea 594, 595-6
ownership apprenticeship 154
social power 953-4 art analysis 656, 665-6, 673, 675
state 1061 belief systems 583
territoriality 485 diet 229 Papua New Guinea
(continued)
P-structures disease 314
definition 378, 379 feasting 953
iconicity 386 labour 896
Pacific island societies learning behaviour 147
disease 314-17,318,319 literacy 538, 539
horticulture 102 metaphor 342
Paddayya, K. 84 music 697
pagamsm myth606
cultural isolation 716 ritual617
palaeoanatomy satin bower bird 135
fossil records 49-50 sex 820, 824
palaeoanthropology paradigms
classification debate 5 belief systems 57 3, 57 4
cultural factors 67 definition 373
palaeoculture hierarchical 374
conceptualization I 00 Paranthropus
language 10 I classification 58-9
palaeoecology parenthood
fossil records 49-50 naming 387
Palaeolithic pedagogy 351-5
behaviour 87 social 79 2-4
cultural transition 88-92 pariah castes
global template analysis 390 ethnicity 715
human evolution 7 Parkinson's disease
hunting debate 86--8 speech production 118
performance 614, 615-16 Parpall6 cave (Spain)
theatre 343 art 97
palaeontology Parry, K. 553
cultural factors 67 Parsons, T. 1029
fossil records 49 pastoralism
human evolution 6 colonization 102
molecular biology 38, 56 definition 211-12
1107
INDEX
food production debate 10, 11 Peninj (Tanzania)

herding 236 fossil site 57
seasonality 212-13 pepper
spatial organization 469 spice trade I 046
subsistence mode transition 197-221 perceived space
paternity definition 479
see also fatherhood perception
gender 743 anthropological study 331
reproduction 774 belief systems 57 3
social relationships 742 time 514-15
paths perennialism
operational analysis 426 ethnicity 708
technical427-9 performance
territoriality 486 cultural genres 640
patriarchy drama 626--9, 630-1, 632
social evolution 1023 efficacy 622-4, 625, 626
patrilineal descent entertainment 622-4, 625, 626
inequality 1020--1 ethnopoetics 870
rules 794--5 hunter-gatherers 616--17,618,619-
patterns 20
built environment 461 hunting ritual620--2
communication 466 language 862
settlements 482 music 689, 698-9
structure 378 neurology 63 8-9, 640, 641
Paul, R. 378 Palaeolithic 614, 615-16
peacemaking physiological effects 696-8
negotiations 996 ritual613-45
peasants sacrifice 632-8
ethnicity 708 training 64 1, 64 2, 64 3
lateral ethnie 713 violence 632-8
nationalism 727 periodicity
social evolution 94 7 market classification 934
pebble tools Perrault, Charles 452
H. erectus 82 Persians
Pacific 94 ethnicity 710, 712, 713, 716
pedagogy lateral ethnie 714
animal skills transmission 333 nationalism 718
chimpanzees 357-9 personal space
education 67 territoriality 486, 487
humans 354-7 personalization
language 360 individual 477
literacy 535 material culture 335
parents 387 personhood
social transmission 352 concept of 23-4
sports 440 identity 831-54
pedlars social life 744--7
market classification 933 Pfaffenberger, B. 143
Peirce, Charles Saunders 368-9 phenomenology
pendulums anthropological study 332
development of 513 art analysis 660
1108
INDEX
phenotypes planning
definition 162 tool-use 150
extended 17 5 planning depth
genetic inheritance 163-4 hunting debate 86--8
multiple-level evolution 181-2, 183, language 101
184-5 plant cultivation
organism-environment eo-evolution origins 207-9
176, 177, 178, 179, 180--1 plant domestication
selection procedure 187-8 Neolithic 267
self-induced natural selection 165- social evolution 942
6 plantation system
species variation 18 colonialism 1055
Philippines plants
sex 826 husbandry, food production debate 10
Philistines resources
ethnicity 714 exploitation 4
philosophy nutritional significance 4
state institutions 7 50 spatial colonization 467
technology 421-2 taxonomy 374
temporality 514 play
Phoenicians concept of 343
religion 716 hunting 621-2
phonation imitation 382
human speech 112, 114 performance 620--2,638-9, 640,641
phonemes Pleistocene
structuralism 374 climacteric events 72
phonemic alphabets cultural traditions 98
literacy 529 domestication 209
phonology glaciation cycles 80-1
structuralist approach 374 palaeosoils 82
photography pluralism
technological development 690 law 751
physicality legal 976--9
artefacts 407-8 modern world 724
physics poetics
temporality 520 ethnographic 870--1
physiology Polanyi, K. 920, 921,922-3, 925,927,947
dance, music 696--8 polio
social 963 Amazonia 308
Piaget,Jean 148, 150,383 politeness
pictographs language 879, 881-2
literacy 528-9 political economy
Pilbeam, D. 40 anthropological study 346
Piltdown social evolution 950
early hominid evolution 37 social life 747-50
Pithecanthropus political evolution
see also Homo erectus collective violence 986
early hominid evolution 36, 37 political geography
Trinil82 definition 480
Pitt-Rivers, A.Lane Fox 142, 143, 432, 435 political realism
1109
INDEX
behaviour 751 food-collecting 200--4

collective violence 987, 989 growth 4, 11-12, 80,208
politics indigenous, disease 297-320
belief systems 585 pressure 271-4
cultural isolation 716 social evolution 942-3,946--7
cultural significance 346--8 virgin soil 300, 308
culture 706--28 population biology
ethnic election 712-13 environment-organism feedback
ethnic survival 711-13 cycles 169-71, 172-6
ethnicism 709-10 population pressure
Green 1063 aggression 985
language 862, 872-4 Portugal
mythomoteur 716--17 colonialism 1046--8
nation formation 717-21 colonization policy 307
nationalism 706-9 positive reciprocity
primoridalism 706-9 gift exchange 922, 923-4
religion 712, 716-17 positivism
social drama 627 temporality 504
social evolution 940--57 Pospisil, L. 968,969,977,978
spatial organization 491 post-structuralism
territorialization 711 linguistics 406-7
warfare 711-12 social relations 416
pollution posture
disease 12 evolutionary divergencies 51
polyethnic nations hominids 61
modern world 723 potato-washing behaviour
polygamy macaques 229
social parenthood 793 potlatch ceremonies
Polynesia wealth 916
belief systems 57 5 Potts, R. 205
food economy 235 Pound, Roscoe 968
food taboos 246 poverty
Pongidae health impact 301
classification debate 39 power
early hominid evolution 37, 38 balance of 1020
pongids concept of 990--1
fossil records 49 exchange 915
Pongo gift exchange 920
evolutionary divergencies 50 inequality 1020, 1022-3
Pongo pygmaeus 20, 22 language 862, 879
Poole, FitzJohn Porter 744--5, 831-54 law 975,976
Popper, K.R. 64, 473 normative 751
popular music political economy 749
concept 694 property 1030--1
innovation 700--1 social life 737
population sources of951-5
see also demography; synthetic mode1955-7
indigenous population violence 989-94
agricultural 214-18 Prague School
biomolecular surveys 5 ethnography of speaking 866
demographic expansion 265-91
1110
INDEX
prayer primitive art

Islam 694 definition 648, 678-9
pre-Darwinian biology see biology, pre- primitive communism
Darwinian social evolution 1011-12
predation stateless societies 1043
classification 211 primitive law
predators modern comparison 963-6
animal behaviour 766 primitive societies
pregnancy social evolution 1013
food taboos 248-9 primordial language
preliteracy symbolism 389
tensions 537-9 primordialism
Premack, David and Ann James 67 ethnicity 346
animal skills transmission 333, 350- politics 706-9
63 pnvacy
cognitivism 337 minority languages 874
prescriptive marriage spatial organization 466-7, 472, 495
exchange 801-3 process orientation
preservation anthropological view 345
music 700 language 865-6
prestige social life 738
see also status procreation
division of labour 902 social life 740
food 254--5 production
language 872, 875 egalitarianism 1012
sex 822-3 kin-ordered 1043
pnce modes of 11, 197-9, 446 production
neoclassical economics 916 (continued)
Price, B. et al. 67 political economy 747
priests social relations 217
ethnicity 712 productive labour
lateral ethnie 713 definition 892
social evolution 942 profits
Prigogine, I. 520 Marxism 913
primates progress
animal behaviour 766, 768, 769 concept of 329
communication 64 sociocultural theory 143
cranial centre of gravity 47 technologicall41-2
diet 228 prohibition
ethology 440 alcoholism 413
food 228-9 kinship 783-808
gesturallanguage 63 proletariat
hominid evolution 34, 35, 38 commodities 914
kinship 773 propaganda
mimicry 368 ethnicity 711
non-human 135-6 property
ritual behaviour 617 concept of 1012
social transmission 355 kinship 797
sociali ty 7 39 music 700, 701
technology 440 private ownership 1028, 1030-1
1111
INDEX
Prosimii occupational1029
hominid evolution 35 typology 948
protein requirements Rapoport, Amos 337, 460-97
human populations 229 rate of exchange
proto-cultural phenomena labour 896
bird behaviour 166-7, 168 Rathbun, Ted 283
prototype theory rational nationalism
componential analysis 374 typology 717
psycho-archaeology rationality
technology 440 belief 563-87, 580--2
psychological space rats
definition 479 social transmission 352
psychology Ravindram, DJ. 545
belief systems 578 reading
child development 745 learning 535
culture 332-3 realism
development 744 political 987, 989
individualism 1017 reality
literacy 533 belief systems 565-6, 581
technology 440 language 861
public health specialists 227 nature of 448
puritanism social592
food behaviour 245, 256 reason
Pygmies anthropological study 332
hominid evolution 34 behaviour 563
human uniqueness debate 4-5, 26-
~zfeh (Israel) 8
burials 97 nature of 504-5
fossil records 93 reciprocal altruism
vocal tracts 126 social evolution 164
Quakers reciprocity
language 880 exchange 911-36
Quesna~F.912,914 law 966--7
Quiatt, D. 139-40 negative 922, 923-4
positive 922, 923-4
racial inequality reconstruction
comparative study 1024-5 symbolic behaviour 387-9
racial stratification record-keeping
class structure 1014 social evolution 945
Radcliffe-Brown, A.R. 861 recordings
belief systems 564 archives 700
food taboos 246 technological development 690, 692
kinship 784 Redfield, R. 965
law 963, 965, 967 redistribution
Raleigh, Waiter 1042 reciprocity 921
ranching reference
classification 212 symbolism 377, 381
Ranger, T.O. 975 reflexivity
rank temporality 504--5
gifts 918 refuse heap model
subsistence mode transition 208
1112
INDEX
regional continuity see continuity repetition

hypothesis temporality 521
regional geography replacement hypothesis
definition 480 art/language 99-100
regional planning modern humans 88, 90, 91-2
spatial organization 480 representation
regulative systems art 664--70
food production debate 10 collective 1017
subsistence modes 204-6 iconicity 386
rehearsals reproduction
hunting 621 collective violence 985
performance 641, 642, 643 genetic inheritance 178
Reichard, G. 864 relationships 741
Reichel-Dolmatoff, G. 375 reservations
relationships American Indians 303
joking 861 residence
labour 893 kinship 794-5
law 966--7 resonance
sex 821-5 symbolic 377
social/biological 739-54 resources
socialization 838 inequality 1010
terminologies 804--6 modern world 753
relatives normative power 990--2
classification 803-6 spatial organization 469, 472
relativism time 515
belief systems 565, 566--7, 581-2 warfare 994
cultural concept 329 restricted exchange
human uniqueness debate 29 marriage 926, 929
language 872 restricted literacy
linguistics 864 autonomous model 533
social anthropology 784 retrospective nationalism
religion ethnicity 708
art analysis 654 Reuleaux, F. 423, 439
Australian Aboriginal 697 reversibility
belief in 567-72 temporality 517, 519-20
colonial expansion I 041 revolution
division of labour 903, 904 technological concept 442
ethnicity 707 Reynolds, F. and V. 617
labour 894 rhetoric
literacy 528, 537, 543 literacy 533
Marxism 632 rhinoceros
politics 712,716--17 hunting debate 85, 86
ritual613-16 Ricardo, David 895, 912, 915
social evolution 944 Richards, A. I. 226, 227
temporality 515 Richerson, P.J. 182
remuneration Rindos, D. 207
inequality 1019 rites of passage
Renaissance music 699
art analysis 652 ritual
rent adaptation 372
theory of916 belief systems 568, 577-8
1113
INDEX
cultural significance 340--3 Sackett,J. 155,670

nature of 342-3 sacred centres
performance 613-45, 638-9, 640, ethnicity 713
641 sacrifice
religious 583 performance 632-8
training 64 1, 64 2, 64 3 purpose of 570
violence 634 Sadat, An war 1002
Rivers, W.H.R. 1022-3 Sahabi (Libya)
Riviere, Peter 792 fossil records 54
Robbins, L. 915-16 Sahara
Roberts, Simon 750-1, 753, 962-79 horticulture 102
Robertson, A.F. 908 Sahlins, Marshall
Robertson Smith, W. 569-71 inequality 1020
robotics predatory expansion 1043
technology 44 reciprocity 920, 923-4, 925, 927,
rock art 930-1
Australia 98 significance 378, 390
Rockhill, Kate 546--7 social evolution 941
role Sahul
inequality 1022 colonization 81
language 861 sea transport 94
language socialization 869 Saint Cesaire (France)
law 968 Neanderthals 85, 91, 92
role settings St Thomas Aquinas 563
definition 461-2 Salisbury, Richard 900
Romantic Movement Salzano, F. M. 49
art analysis 649 Samuelson, P.A. 916
Germany 724,725, 864 San (Bushmen)
Rosaldo, Michelle 821 behaviour debate 275--6
Rousseau,Jean-Jacques697, 1011,1019, belief systems 57 5
1054 cultural traditions 67, 98
Rowlands, M. 949, 951, 953 demography 285
Royce, A.P. 688 fertility 289
Rozin, P. 255 hominid evolution 34, 60
rubella hunting debate 86
Amazonia 308 population growth 271
Rubinstein, RobertA. 751,983-1004 ritual617
rules vocal tracts 66
kinship 783-808 sanitation
legal pluralism 978 warfare994
negotiation 1001 Santillana, G.de 390
social order 752 Sapir, E. 536, 831, 862, 864, 8
Russia Saraswathi, L.S. 545
ethnicity 716 Sartre, Jean-Paul 628, 638
modern world 723-4 satellites
trade relations 1044 spatial organization 49 3
Russian Orthodox church Sattenspiel, L. 286
literacy 541 Sauer, Car! 205
Saussure, Ferdinand de
S-structures language 862, 865
definition 378 linguistic theory 401
1114
INDEX
literacy 536 ethnocentrism 987-8

semiotics 368-9, 383 modern world 751
savagery sedentarization
social evolution 1011 definition 213
scalar stress sedentism
concept of 273 colonization 101-2
scale diminishing returns 280-2
classification 491 disease 284
spatial organization 485-6 fishing 205
scales of evaluation food selection 23 8
inequality 1019 herding 236
scarcity Neolithic 267-8
neoclassical economics 916 population pressure 273
wealth 915, 929 prehistoric populations 12
Schafer, R. 849 sex 827
Schechner, Richard 342-3, 613-45 subsistence systems 201-2, 207-9
schemata Seeger, Anthony 345, 346, 686--702
shame 851 Seeger, Charles 693
socialization 834-6, 846--7 segmentation
Schieffelin, B.B. 869 inequality 1020--1
Schieffelin, E.L. 697 selection
Schneider, D. 370, 372 see also natural selection
school literacy clonall85
Iran 541 prefrontal cortex 119
Schwartz, T. 836, 839 self-induced 165-8
science self-awareness
symbolic interpretation 372 human uniqueness debate 27-8
temporality 514 personhood 843
Western 565 self-rule
science fiction ethnicity 715
technology 421 selfhood
scientism identity 841-3
technology 441 selfish gene theory
Scotton, C.M. 862 phenotypes 164
scribes semantics
ethnicity 712 art 344
Scribner, S. 534 colour 390
seasonality music 345
fallowing 218 semi-fixed feature elements
food 239, 243 built environment 463, 491
nomadic pastoralism 212 semiological analysis
plant cultivation 207 art 665, 666
sedentism 273 semiology
subsistence systems 202 origins 865
Second World War 983 semiotics
secrecy anthropological role 371
myth 606--8 art analysis 660
secularism cultural behaviour 155
colonialism 1053 definition 368-9
security literacy 527
common 983-1004 material culture 335
1115
INDEX
spatial organization 337 sharecropping

style 671 division of labour 908
tropes 368 sharing
sensori-motor intelligence ethnicity 709
apprenticeship 151 experience 359-61
definition 148 foraging 228
sensory effects meals 250-3
aesthetics 344 national languages 873
sensory space subsistence systems 203
definition 479 symbolism 380
Sequoyah shopkeepers
literacy 540 market classification 933
Service, Elman 944, 945, 946, 947 sickle-cell anaemia
settings natural selection 168
definition 461-2, 463 Sidi Abderrahman (Morocco)
social complexity 490 pebble tools 82
spatial organization 492 Sierra Leone
settlements art analysis 666, 676
abstract geometrical space 478 myth 603
ekistics 480 Sigaut, Frans;ois 336, 420--52
food production debate 10 sign language
patterns 482 art analysis 664
social evolution 942 speech comparison 110-11
spatial organization 469 signs
sex AustralianAboriginal667
biology 813-16 comparative reconstruction 388
classification 816--21 definition 369
culture 813-16 hunting 620
gender 813-28 linguistic 862-3
inequality 821-5 literacy 527, 528
relationships 742-3 symbolism 380-1
subordination 821-5 Walbiri graphics 661
Western folk model816-21 Sikhs
sexual division oflabour ethnicity 713
feminist movement 822, 825 silence
sexual receptivity 27 Apache language 867
sexuality Sillitoe, P. 656
violence 633-4 silver
Shackley, M. 145 colonialism 1046--8
Shackley, S. 440 Silverstein, M. 373, 875, 878
Shakespeare, William 628-30, 632 Simmel, Georg 396, 893
shamans Simon, J. 277
improvisation 613 Simon, Paul 701
theatre 624 Simondon, G. 423
training 63 7 Simpson, G. G. 35, 37, 38, 58, 60
shame smgmg
socialization 851 non-verbal communication debate
Shanidar (Iraq) 63-4
animal domestication 235 Skelton, R.R. 70
vocal tracts 126 Skhul (Israel)
Shanks, M. 412-13
1116
INDEX
fossil records 93 interactive nature 342

vocal tracts 126 performance 626--9, 63(}-1, 632
skills social ecology
art analysis 654 definition 480
cultural significance 334--40 social evolution
division of labour 907 politics 940-57
literacy 339-40 primitive communism 1011-12
political economy 747 synthetic theory 949-57
technical438-42, 445-9 social field
training 359 definition 977
transmission of 332-3 social geography
Skinner, G.W. 935 definition 480
Skolimowski, H. 141 social learning
Skorupski, J. 579-80 skill transmission 333
skulls social life
remodelling 46 biological link 739-44
slang introduction 737-54
linguistic style 877 law 964
slavery personhood 744--7
abolition 1054--5 political economy 747-50
colonialism 307, 1046--8 social mobility
technological relationship 449-51 inequality I 030
small-group ecology social networks
definition 480 development 101-2
smallpox nomadism 474
American Indian comparison 311 spatial organization 466, 469
eradication 301 social occupancy
New World impact 299, 302 spatial organization 482
Smith, Adam 891, 892, 912, 922 social parenthood
Smith, Anthony D. 346-7,706-28 kinship 792-4
Smith, P. 272, 283 social physiology
snacks law 963
definition 250 social policy
food behaviour 252 egalitarianism 1032
Snyder, F. G. 975 social power
social analysis sources of 951
individualism 1017 social production
social anthropology technical skills 445-9
literacy 533 social psychology
relationships 739-44 complex social organization 27 3
relativism 784 identity 844-5
spatial organization 481 social relationships
studyof415 anthropological study 402-3,415-16
social bonding biological link 739-44
analogical processing 382 disease 12
social complexity hunter-gatherers 20 I
spatial organization 489-90 social storage
social control cultural transition 87-8
law 968 social structure
social drama American Indian comparison 302-3,
312-13
1117
INDEX
civilization 274 sociopolitical processes

cosmology 583 literacy 542-54
evolution 941, 950 solidarity
gifts 918 division of labour 903-4
group systems 794--7 ethnicity 709
kinship 783-808 language 879
labour 893 law 965
law 963 mechanical967, 1016
literacy 339 music 697
order 752 Solo River Oava)
sedentism 272-3 H. erectus 82
social transmission song
cultural knowledge 839 see also music
information 351-4 totemism 599-600
social-symbolic manipulation sorcery
cultural role 372-3 belief systems 568
socialism Palaeolithic 614, 615
gift exchange 920 Sousa Santos, B.de 976
sociality South Africa
questions of737-9 inequality 1024--5
socialization South America
cultural identity 399 colonialism 1046--8
identity 831-54 South American Indians see American
kinship 784 Indians
language 868-9 space
literacy 550 concept of 478-9
personhood 745-6 organization see spatial organization
society Spain
concepts of 737-8 colonialism 1046--8
definition 274 language 873-4
inequality 1010--35 New World impact 303
warfare 992-4 spatial organization
socio-cultural tradition built environment 460-97
tool-use 140-5 conceptual framework 337
sociobiology material expression 493-4, 495, 496
aggression 985 prehistoric art 384
animality /humanity comparison 22 scales 490-1,492
definition 480 study of 479-83
ethnicity 707 specialization
evolutionary theory 176 division of labour 899-900, 901,
genetic evolution 164-5, 191 902-5
natural selection 741 free-market economies 906
optimal foraging theory 277-83 geography 445
reproductive fitness 290 law 964
sociolinguistics species
language 862 concept of 4, 23-4, 27-8
literacy 533, 536--7 gene pools 25
sociology variation 17-18
division of labour 1016 specificity
meal-sharing behaviour 253 artefacts 408-9
occupational status 1029-30 speech
1118
INDEX
see also language status

anatomical components 111 see also prestige
childhood 849 classification 1029
evolution 127-9 inequality 1010
H. habilis 63 labour 894--5
hominid evolution 46 language 861, 869
humanity 8, llO-ll lateral ethnie 714
physiology lll-16 music 699
production ll6--19 spatial organization 469
statistical analysis 688 stateless societies 1043
transmission rate 110 status symbols
writing comparison 536-7 environment 337-8
Spencer, Herbert 142,944,946 food 248, 254--5
Sperber, D. 370--2, 373 Stegmiiller, W. 521
Speth, J. 280 Steklis, H.D. 64
spice trade Stengers, I. 520
colonialism 1045-6 Sterkfontein (South Africa)
Spiro, M.E. 789, 835 fossil records 37, 56, 57
sports Steward,Julian 944,946
efficacy 448-9 stimulants
skills analysis 440--1 music 696
Sraffa, P. 914 Stone Age
Stalin, J. 1029 big-game hunting 279
Standard English Stone, R. 687
identity 875 stone technology
standard language colonization 94-5
identity 875-6 encephalization 81
staple finance Late Stone Age, Africa 96
social power 953, 954 stone tool-makers
staple foods hominid evolution 63, 69-72
nutrition 236-7, 239-42 stone tools
starvation Acheulean 81
warfare 993 cultural transition 91
state H. erectus 82
see also nation states Stonehenge (England) 4ll, 514
demotic ethnie 720 story-telling
formation 965 hunting 621
inequality 1020 Strathern, Marilyn 400, 820, 823
institutions 7 50 stratification
non-European polities 1041-2 civilization 274 stratification
social life 738 (continued)
society 945-7 division of labour 903
theory of origins 27 4 inequality 1010
typology 948-9 law 973
warfare711-12 race 1025
state level organization social evolution 946
health and nutrition 283 social order 753
sedentism 273-4 stratified society
stateless societies typology 948
inequality 1020 Strauss, Anselm 99 5-6
world order 1042-4 Strauss, C. 83 5
1119
INDEX
Street, Brian V. 339-40, 527-54 filter function 114, 115

structural anthropology length 116
definition 374 reconstruction 121, 122, 123-7
structural linguistics speech components 111, 113
symbolism 369 speech production 128
structuralism surplus-value theory
cooking 243 Marxism 920
food 226, 246 Suya
inequality 1019 music 345, 693-701
language 865 swallowing
myth 592 human speech 114
ordering process 401 Swanscombe (England)
paradigms 373 cultural traditions 84
relationship terminology 784 Swartkrans (South Africa)
symbolic role 372 fossil records 56, 57
technology 440 stone tools debate 71
style swidden agriculture
art analysis 670-2 Amazonia 308
sublimation domestication 208
ritual violence 634 fallowing 218
subordination home-bases 205
sex and gender 821-5 spatial organization 469
subsistence West Africa 167
see also diet; syllabic writing
food; literacy 529
nutrition symbolic manipulation
demography 10-13 concordant structure 379
food production debate 11 symbolic mechanism
human species 4 conceptual system 372
local220-l definition 371
modes 197-221 symbolic resonance
ownership 953 meaning 377
social evolution 950 symbolism
Sudan anthropological goals 370-3
warfare 993 art analysis 659
sugar art/language 99-100
colonialism 1046--8 artefacts 409
suicide behaviour 101, 381-92
American Indian comparison 312-13 belief systems 567-8
Sumatra 20 categorization 367-70
Sumerians collective violence 989
ethnicity 709 comparative reconstruction 387-9
literacy 528 cooking 245
Sunghir (Russia) cultural concept 329, 333-4
Upper Palaeolithic 96 cultural relationship 366--92
superstructures differentiation 598
Marxist theory 141 division of labour 903
supralaryngeal vocal tracts ethnicity 707, 710, 711
animal comparison 119, 120, 121 ethnology 375
Australopithecine 126 food selection 237-8, 243-5
Broca's aphasia 116-17 formal methodology 373-5
1120
INDEX
human uniqueness debate 27 Tambiah, S. 246, 577-8

hunting societies 98 Tanaka, J. 67
labour 894 Tannen, D. 877
language 101, 862, 864-5 Taoism
latent functions 470 belief systems 584
linguistic theory 373 cooking 245
meaning 376-7 Taung (South Africa)
nationalism 725 fossil records 37, 56, 57
negotiations 1001-3, 1003 Tawney, R.H. 1011, 1032
ontogenesis 386-7 taxation
origins 380-1 alcohol413
prehistoric 383-6 colonialism 1052
space 478 world order 1040
spatial organization 337, 469 taxonomy
structural concordance 378-9 biologicall8, 24
temporality 519 N eo lithic 1017
theatre 614 relationship terminologies 804
tool-use 156 Taylor, F.W. 441
symbols Taylor, R. 314
definition 366, 369 technical intelligence
literacy 528 knowledge 438-42
synchrony technical lineages
modern/traditional debate 504, 522 concept of 435
syntagma workings 434--8
definition 373 technical paths
syntax operational sequences 427-9
animal comparison 110 technical skills
brain mechanisms 116-19 social production 445-9
Broca's area 8 technics
evolution 127-9 science of 420--4
hierarchy 148 slavery 449-51
neural networks 108 technique
tool-use 146-7 concept of 436
synthetic theory technography
evolution 162-5 concept of 336
systematics definition 423
classification 38-9, 58 technology 420-52
Hominoidea 35-6 agriculture 270
systems ecology big-game hunting 279-80
environment 198 concept of technography 330
definition 134, 422-3
taboos equipment debate 197
food 246--50 future of 451-2
Tabouret-Keller, A. 872 human context 140--1
Tabun cave (Israel) human evolution 8 technology
stone technology 94 (continued)
vocal tracts 126 individualism 1017
tails military 990
genetic modification 18 music 690-3
human uniqueness debate 27 population growth 268-9
human-animal comparison 15-19 social evolution 941, 941-3, 942-3
1121
INDEX
subsistence systems 200 theories of mind

temporality 515 shared experience 359-61
Western conceptions 136 socialization 850
technonature Thessalia (Greece)
definition 423 animal domestication 235
teeth third parties
hominization 44--5 law 971
impaction 121 Third World
Tehuacan (Mexico) collective violence 988
agriculture 236 modern nations 722-3
Telefolmin post-colonialism 1062-3
art analysis 664 warfare994
templates Thomas, Robert J. 907
global analysis 390 Thomsen, C.J. 442
myth 594 Thoreau, H.D. 441
structuralism 378-9 thought
temporality 505-8 modes of 572-5
tool-making 386 threat
tempo negotiation technique 998
Western time 511 thumbs
temporality hand comparison 145
anthropological theory 330 Tibetans
artefacts 409-15 ethnicity 715
clock time 512-16 Tikopia
dualism 516--22 art analysis 653
everyday life 508-12 Tilley, C. 412-13
identity 411-13 time
organization 465-6 see also temporality
study of 338-9 art analysis 664
templates 505-8 Hopi 864
terminologies Nuer 863
relationship 804-6 organization 465
territoriality perceptions of 503-23
definition 484-6 spatial organization 469
gift exchange 923 time-space
spatial organization 472 definition 617
territorialization timing process
politics 711 Western world 509-10, 511
territory to as
rootedness 715 art analysis 66 7, 668
test-tube babies Tobias, Philip V.
motherhood 791-2 hominid evolution 6, 33-72
texts Linnaeus interpretation 33-5
ritual613 speech 8
Thailand Tocqueville,Alexisde 1013,1014,1016
language 881-2 Tokelau Islands
theatre blood pressure study 316-17
dramatic role 342-3 tongue
functions 613 human 113, 122, 123-4, 125
hunter-gatherers 616--17,618,619-20 tool-making
shamanic techniques 624 Acheulean 383
1122
INDEX
human evolution 7-8 social power 955

iconicity 386 spices 1045-6
prefrontal cortex 119 sugar I 046--8
stone 69-72 tradition
tool-using distinction 134 art analysis 657-8
tool-use authority 973
agriculture 277 colonialism 1051
big-game hunting 279-80 division of labour 904
foraging 228 environment 337
seed-processing 266 law 963, 964--5, 975-6
slavery 450--1 music 699-700
symbolic capacity 382 power 991
tool-making distinction 134 temporality 514
tools tool-making 154
definition 133-4 transmission of 351
human anatomy 145-6 traditional societies
human behaviour 133-5 anthropological study 504, 517
human cognition 146--50 belief systems 564
human culture 151-7 traditionalism
iconic signs !56 ethnicity 708
indexical signs !55 non-infectious disease 314
natural history tradition 137-40 spatial organization 490
non-human behaviour 133-7, 150, traffic accidents
!56 man-made disease 12
socio-cultural tradition 140-5 Traill, A. 66
Torralba (Spain) training
hunting 85 division of labour 899
Torres Straits expedition 657 modern humans 359-60
totemism ritual 641, 642, 643
food taboos 246-8 transcience
language of 341 artefacts 413-15
naming systems 595-8 translation
Toynbee, A. 442 belief 575-8
traders transmission
market classification 933-4 cultural knowledge 839
trading transportation
see also barter; social evolution 945
exchange treaties
barter 929 ethnicity 713
complex social organization 27 3 trial-and error
demographic expansion 267-8 tool-use !50
disease relationship 300 tribes
early hominids 220 inequality 1020
European expansion 1044--5 social evolution 1013
fishing 234--5 society 944--5
gold 1046--8 tribute
market classification 932-3 social power 951
market places 931-6 Trinil Oava)
opium 1050 Pithecanthropus 82
si! ver 1046--8 Trinkaus, E. 92
slavery 1046-8 Trivers, R. 164,191
1123
INDEX
Trobriand Islanders military involvement 988

art analysis 664 music 698
belief systems 571-2, 576, 577 occupations 1029
exchange 928 power structure 1028
fatherhood 789-90, 793, 805 Universal Declaration of Human Rights
kula 916 23
law 966--7 unproductive labour
social power 951 definition 892
tropes unskilled labour
definition 368 capitalism 905-6
tropical disease see disease gift exchange 930
tropical rain forests Upper Palaeolithic
colonization policy 307-9 abstraction 391
tuberculosis art 383-5
Amazonia 308 cultural traditions 96-9
American Indian comparison 311 cultural transition 93
New World impact 302 Europe 92, 94
Tuc d'Audoubert (France) hunting 86--8, 204, 209
Palaeolithic 614 population growth rates 266
Tukano symbolic organization 367
symbolism 375 uprightness
Turkana (Kenya) see Lake Turkana ... hominid evolution 40, 41-2,43,61
Turnbull, C. 253 Uraha (Malawi)
Turner, Victor 567,604, 871,969,972 fossil site 57
ritual626-9, 630,631-2,639 urban design
Tylor, Edward 142 spatial organization 480
Tyson, Edward 34 urban ecology
definition 480
Ubediya Oordan) urban geography
stone tools 81-2 definition 480
Uehara, S. 138 Urban Revolution
U exkiill, J. von 468 concept of 442
Union of Soviet Socialist Republics demographic expansion 267
collapse of 988 social evolution 942
negotiations 1000--1 urbanization
occupations 1029 Third World 723
power structure 1028 Ury, William 998, 999
umqueness use-value
human25-30 definition 915
United Kingdom utility
colonialism 1047-8 political economy 748
labour 894--5 technological progress 142
language 876 wealth 915, 929
United States of America uxorilocal residence
ethnicity 723 rules 795, 796, 797
foreign policy 990, 1001, 1059
indigenous peoples comparison 309- Vai
13 literacy 534--5
inequality 1014, 1024--5 value
language 875, 876, 877 egalitarian 1031-2
life expectancy rates 304--6 inequality 1019-20
1124
INDEX
music 698-9 wage-labour

political economy 748 Marxism 913
theories of916-17, 929 wages
variation capitalism 905
individuals 1016 colonialism 1050
inequality 1023-4 division of labour 899, 907-8
natural selection 7 41 free-market economies 906-7
varna productive labour 892, 895
caste system 1026--7 Wagner, D. 547
vegeculture Wagner, Roy
seed-culture comparison 215-16 culture 368, 370, 376--7
verbal art myth 594,609-10
ethnopoetics 870 Wahgi
vernacular literacy dancers 675, 676-7
education 539 Wahrhaftig, A. 991
vertical commodities Waiapi
markets 932 music 696
vertical ethnie Walbiri
politics 713-17 art analysis 660, 665-6
videotapes graphic signs 661
technological development 690, 692, Walens, S. 602
700 Wales, H.G.Q 882
Vietnam War 983 Wallace, A.R. 101
violence war
American Indian comparison 312 definition 986
anthropology of 984--7 war games
collective 983-1004 collective violence 989
ethnocentrism 987-8 warfare
hormones 813 collective violence 985-6
hunting 621 colonialism 1043-4
performance 632-8 colonization 7 53
power 989-94 descent groups 797
social order 751 health impact 301
virgin soil populations inter-state 711-12
Amazonia 308 lateral ethnie 714
colonization 300 law 972
virilocal residence modern nations 722
rules 795, 797 politics 711-12
vocal tracts social order 751
see also supralaryngeal vocal tracts social power 951-2
hominid evolution 46, 48 Warner, Lloyd 1025
morphology 65, 66 warnors
speech debate 63 lateral ethnie 713
vocalization water
definitions 69 5 warfare 994
Voltaire 1054 water clocks
voluntaristic nationalism Egypt514
typology 717 Watergate incident 627
Vygotsky, L. 833, 868 Watts, E.S. 67
wealth
Waddington, C.H. 165-6, 168, 190 belief systems 573 wealth (continued)
1125
INDEX
civilization 274 Widdowson, E.M. 227

commodities 912-15 Wilder, Harrison Hawthorne 36-7
comparative views 911-12 Williams, F.E. 606
gifts 915-28 Wilson, E.O. 351
goods 915-28 Wilson, M. 377-8
inequality 1010 Winch, P. 581
political economy 748 Winnefeld, J.A. 989
social power 953 witchcraft
weamng belief systems 572-4, 581-2
food selection 23 8 Witherspoon, G. 664
mortality rates 285 Wittgenstein, L.J.J. 787
pre-Neolithic diet 289 Wola
weapons art analysis 656
art analysis 663 Wolf, Eric 1043
collective violence 984 Wolpoff, M.H. 92
colonialism 1050 woman marnage
weavmg alliances 799
division of labour 906 women
markets 902 commodities 925
Weber,Max division of labour 899-900, 901,
booty capitalism 1046 902-5
law 964--5, 976 food behaviour 247,252-3
politics 711 inequality 1022-3
religion 570 language 876-7
social evolution 942 relationships 742-3
states 1041 sex 814-15, 821, 824
Weiner, A. 411 warfare 993, 994
Weiner,James 334,341-2,591-610 Western image 14,26
Weismann barrier Woodburn,J. 1012, 1032
ecological inheritance 189-90 words
multiple-level evolution 184 material culture 335
Weiss, K. 286 ontogenesis 387
Wernicke's area symbolic class 370
hominid evolution 48 written 337
modern humans 62, 63 work
speech production 117, 118 co-operation 891-908
West lndies measurement systems 897-9
language 868 political economy 747
Westcott, R.W. 64 workings
Westermarck, E. 800 technology 436
Western nationalism workshops
definition 717 performance 641, 642, 643
Western time World Bank
comparison 505-7, 508-12 Brazilian development policy 309
White, Leslie 141,245,942 world order
White, LynnJr. 423, 424, 434 colonialism 1040--63
White, M. 846 world view
Whitehead, H. 822-3 language 861, 863-5
Whiting, J.W.M. and B.B. 833 Worsley, Peter 753, 988, 1040--63
Whittaker, E. 841 Wrangham, R. 355
Whor~B.L. 505-7,864,868 Wright, H. 945
1126
INDEX
writing deer dancers 614,616

built environment 473-4 hunting 621
environmental impact 337 ritual643
learning 535 Yayo (Chad)
letters 539 fossil site 57
linguistics 408 Yin-Yee Ko 546
origins 528-9 Yolngu
social evolution 944 art analysis 665-6, 674, 675-7
speech comparison 536-7
symbolic categorization 367 Zapotec
systems 529-32 child mortality rates 304
Wynn, Thomas 8, 133-58, 383 labour 896, 898
Wynne-Edwards, V. C. Zawi Chemi (Iraq)
cultural models 269-70 animal domestication 235
Zhoukoudian (China)
Yanagisako, S. 816-18, 820, 822 H. erectus 82
Yanomami ornament 98
disease 308 Zinjanthropus
Yaqui classification 59
1127

Ingold Ed Encyclopedia of Anthropology Part 3

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Ingold Ed Encyclopedia of Anthropology Part 3

Uploaded by

Copyright:

Available Formats

COMPANION ENCYCLOPEDIA

London and New York

A catalog record for this book is available on request.

PART 11: CULTURE 327

13. Why animals have neither culture nor history

PART Ill: SOCIAL LIFE 735

33. Exchange and reciprocity

phrase which nicely combines the notion of encompassing a whole field of

THE SCOPE OF ANTHROPOLOGY

known as 'ethnology') was linked to emergent nationalist movements of the late

combination, for example within University Departments, is an anachronism

data of observation in every branch of anthropology have one thing in common:

and wisdom that these people impart to the fieldworker, sharpened as it is by

HUMANITY, CULTURE AND SOCIAL LIFE

particular arguments, we should never forget that the pursuit of

volume-including the definition of issues to be covered by individual articles,

sometimes provocatively towards the future. Many contributors, moreover, deal

individual topics is, I believe, unparalleled in contemporary anthropological

Manchester (1965-6), Commonwealth Visiting Professor at the University of

Civilization ( 1989), and senior editor of Paleopathology at the Origins of

ROBIN DUNBAR was educated at Oxford University and at the University of

The Evolution ofHuman Society (1987).

Europe and in Australia. He is the author of The Palaeolithic Settlement of

further fieldwork among northern Finnish farmers, he has written extensively

Ethnology at the University of Oxford. He has carried out anthropological

In 1993 he was appointed to a five-year Research Fellowship at the Institute of

Department of Anthropology at the University of Rochester. His theoretical

music of Brazilian Indians as well as on issues of land and cultural equity. He is

connection with development programmes. His latest research involves the

INTRODUCTION TO SOCIAL LIFE

of its opposition, alternately, to such notions as individual, community and

to be social life at all? Recent discussions have thrown up three kinds of

SOCIAL AND BIOLOGICAL RELATIONSHIPS

in a certain disparity between their respective views of what a relationship 15,

particular patterns and modes of relationship may have evolved through a

anthropologists, that cultural designations of kinship are arbitrarily

women-was said to belong to a normative order, superimposed upon a given

people dwell, as a preliminary to their reinscription on the level of its cultural

To characterize it as 'non-Western' is perhaps misleading, since several

rectptents of social knowledge which 'descends' upon them from an

from a structuralist perspective, would be dismissed as merely idiosyncratic or

first, they are evaluated in terms of a qualitative rather than a quantitative

evolution of complexity in human society, from the small, intimate bands of

DISPUTE, NEGOTIATION AND SOCIAL ORDER

advisers, as an instrument of domination by indirect rule. In the fifth and final

standards can be maintained without a system of imposed regulation: that is,

between, leading in aggregate to the observed pattern of social stratification. In

disastrous environmental consequences of its own expansion. Notwithstanding

Gregory, C.A. (1982) Gifts and Commodities, London: Academic Press.

SOCIALITY AMONG HUMANS AND

The tendency to impute human motivations and forms of social organization to

THE PROBLEM OF SOCIALITY

This tendency to view sociality in terms of the disposition of animals in

mutually antagonistic relationships and do defend their territories against

SOME BACKGROUND THEORY

In a conventional biological context, the material that is passed on from one

as a whole is intrinsically dynamic and can be expected to change continuously

WHY LIVE IN GROUPS?

Table 1 Main benefits gained from group-living, with some

1 Parental care Jackals, foxes, most birds

THE STRUCTURE OF SOCIAL GROUPS

mutual interest in a common resource (Harcourt 1989). In other words,

SOME SPECIAL ORGANIZING PRINCIPLES

from a common ancestor), any copy of a shared gene that is passed on by a