A Passion for Difference

Essays in Anthropology and Gender


Polity Press

Copyright © Henrietta L. Moore 1994

The right of Henrietta L. Moore to be identified as author of this
work has been asserted in accordance with the Copyright,
Designs and Patents Act 1988.

First published in 1994 by Polity Press
in association with Blackwell Publishers Ltd.
Reprinted 1995, 1997, 2005

Polity Press
65 Bridge Street
Cambridge, CB2 1UR, UK

Polity Press
350 Main Street
Malden, MA 02148, USA

All rights reserved. Except for the quotation of short passages
for the purposes of criticism and review, no part of this publication
may be reproduced, stored in a retrieval system, or transmitted, in
any form or by any means, electronic, mechanical, photocopying,
recording or otherwise, without the prior permission of the publisher.

Except in the United States of America, this book is sold subject to
the condition that it shall not, by way of trade or otherwise, be lent,
re-sold, hired out, or otherwise circulated without the publisher’s
prior consent in any form of binding or cover other than that in
which it is published and without a similar condition including
this condition being imposed on the subsequent purchaser.

ISBN 978-0-7456-6805-5 (Multi-user ebook)

A CIP catalogue record for this book is available from the British Library.

Typeset in 10½ on 12 pt Palatino
by Best-set Typesetter Ltd., Hong Kong
Printed and bound in Great Britain by
Marston Book Services Limited, Oxford

This book is printed on acid-free paper.

For further information on Polity, visit our website:

In memory
my mother
Josephine Moore
















This book was produced during a period of leave from the London School of Economics made possible by a grant from the Leverhulme Trust. The Trust generously funded a project on The Development of Models for the Analysis of Gender in the Social Sciences’ from 1 January to 31 December 1993. I am indebted to both institutions for their help and financial support.

Chapter 1 was originally published in Feminist Review, vol. 47, 1994, as ‘Divided we stand: sex, gender and sexuality’.

Chapter 2 is a revised version of an article originally published as ‘Gendered persons: dialogues between anthropology and psychoanalysis’ in Suzette Heald and Ariane Deluz (eds), Anthropology and Psychoanalysis: An Encounter through Culture, London: Routledge, 1994.

Parts of chapter 3 originally appeared in ‘The problem of explaining violence in the social sciences’ in Peter Gow and Penelope Harvey (eds), Sex and Violence: Issues in Representation and Experience, London: Routledge, 1994.

Chapter 5 is a substantially revised version of an article originally published as ‘Gender and the modelling of the economy’ in Sutti Ortiz and Susan Lees (eds), Understanding Economic Process, Lanham: University Press of America, 1993.

Chapter 6 was originally published in Maurice Biriotti and Nicola Miller (eds), What is an Author?, Manchester: Manchester University Press, 1993.

I am very grateful to Routledge, Manchester University Press, the University Press of America and Feminist Review for permission to republish the above materials.

I have benefited enormously from the help and support of many friends and colleagues whilst writing these essays. I would particularly like to thank Michelle Stanworth, Marilyn Strathern and Megan Vaughan for their intellectual companionship, and for their criticisms which they always manage to couch in the most generous possible terms.


A Passion for Difference

Difference exerts an uncanny fascination for all of us. Contemporary social and cultural theory exhibits an obsessive concern with issues of difference, and such is the malleability of the term that almost anything can be subsumed under it.1 This passion for difference seems to be linked to its unspoken and under-theorized pair, ‘the same’ or ‘sameness’. This is not implied, of course, in the deconstructionist notion of différance, but it is implicitly there in much feminist and social science theorizing, as well as in contemporary political activism. Deciding on differences is one way of delineating identities. Difference(s) from others are frequently about forming and maintaining group boundaries. The brutal and bloody nature of this maintenance work is everywhere in evidence.

Thinking about difference entails, then, thinking about identity and/or sameness. However, these latter terms are not themselves identical. Within the academy, establishing an understanding of the relations between difference and identity has been a complex and sometimes explosive task. Feminist scholars, in particular, have been struggling with the question of how or to what degree women might be the same or similar without being identical. What is it, if anything, that we share? This book is concerned with questions of difference, sameness and sharing. It also addresses the various rhetorical forms of ‘we’ at work in feminist and anthropological writing. Who or what does ‘we’ refer to in the contemporary moment? Problems of reference here are connected to issues of belonging. Identity and difference are not so much about categorical groupings as about processes of identification and differentiation. These processes are engaged for all of us, in different ways, with the desire to belong, to be part of some community, however provisional. Belonging invokes desire, and it is in this desire that much of the passion for difference resides.

In terms of my own writing the question of where and to what I belong involves, as it does for others, a consideration of position and location. If I belong somewhere, then I speak and write from there, and the specifics of that location matter. But all locations are provisional, held in abeyance. One is never truly anywhere and if locations or positions are to be specified, they will always be in the plural. The crisis of location is a productive, but personally terrifying one. Both the label ‘anthropologist’ and that of ‘feminist’ remain under radical interrogation, and as a feminist anthropologist I find my relations with these terms to be strenuous, nuanced and unrelentingly complex. However, I am passionately committed, and this is what provides the energy and the propulsion for my work. The essays in this collection represent my attempts to come to grips with the questions of difference and sameness that being a feminist and an anthropologist raise.

Bodies and identities

The provisionality of positions reminds us that when we consider questions of location we are not simply trying to reinscribe an essentialism of place. Positionality is too often reduced to individual experience and/or to representation: ‘I know because I’ve been there’ and ‘I know because I am one’. These slippages are particularly troublesome when linked to grounds for authority. Anthropology and feminism share a tendency to assert that experience acts as an ontological given. This issue has been discussed in many times and places, but what worries me here is the way in which experience is sometimes reduced to its linguistic and cognitive elements, to what I know and to what I can talk about. This process of reduction encourages a view of experience which sees it as ontological, singular and fixed. Experience can and does act ontologically for all of us, but it does so through a technique of construction. This process of construction involves a recognition of the role of physical presence in establishing dialogue between individuals and groups. What is at issue is the embodied nature of identities and experience. In subsequent chapters I propose a notion of the ‘lived anatomy’ and of bodily praxis as a mode of knowledge that draws on an understanding of experience as a form of embodied intersubjectivity. The very fact of being present as an embodied subject gives a particular character to the ontology of experience which emphasizes the degree to which social interactions are embodied ones taking place in concrete space and time.

Intersubjectivity and dialogue involve situations where bodies marked through by the social, that is, by difference (race, gender, ethnicity and so on), are presented as part of identities. The uses of the body, the particular circumstances of interaction and the readings made by others are all involved in the taking up of a position or positions that form the basis for the enunciation of experience. Experience is thus intersubjective and embodied; it is not individual and fixed, but irredeemably social and processual.

The experience of being a woman or being black or being a Muslim can never be a singular one, and will always be dependent on a multiplicity of locations and positions that are constructed socially, that is, intersubjectively. The intersubjectivity of experience is not confined, of course, to physical appearances, to actual dialogue and to the concrete nature of sociological circumstance. Intersubjectivity is also about identifications and recognitions. It is about desire and the projection and introjection of images of self and others. One of the major questions here, addressed in chapter 3 through issues concerning sexuality and violence, is the problem of how we construct and acquire identities, and how well these processes are captured by current theories in the social sciences.

The individual and the social

One of the major themes running through this book concerns the relationship between anthropology and psychoanalysis. I take anthropology to task in several ways for its failure to theorize the acquisition of gender identity and the multiple nature of subjectivity. One puzzling feature of the development of feminist anthropology has been its relative neglect of the debate concerning the deconstruction of the humanist subject. Chapter 2 suggests various ways in which cross-cultural data might be used to illuminate and contribute to feminist theorizing in this area.

My discussions of gender identity and gendered subjectivity work over a series of old, but unresolved themes about the relationship of the individual to the social and vice versa. Anthropology’s emphasis on the social at the expense of the individual accounts in large part for its failure to develop a theory of the subject. However, the problem raised by cultural difference and its relation to gender difference is one about how processes of identification and recognition work. What makes the cultural discourses of gender powerful and how well do they regulate/constitute/represent people’s experience of gender in any given context?

Post-structuralist theories of the subject and of positionality are useful here because they create a space in which it is possible to talk about the different subject positions proffered by various discourses. Thus, gendered subjectivity does not have to be conceived of as a fixed and singular identity, but can be seen instead as one based on a series of subject positions, some conflicting or mutually contradictory, that are offered by different discourses. This would be all very well were it not for the problem of how to account for the fact that individuals do not always take up the subject positions offered to them. One obvious point is that the existing discourses on gender in any given context are hierarchically organized, that is, some are more powerful and have greater social sanction than others. In the United Kingdom, for example, some forms of masculinity are perhaps much easier to identify with than others because they are socially valued and accepted. Oppressed groups frequently develop their own discourses that work in contra-distinction to dominant ones, but the questions are, can people actively recognize and choose the subject positions they take up, and to what degree are they able to resist the terms of dominant discourses? Much of the debate here turns on the use of the terms ‘choice’ and ‘resistance’, and their suitability for analysing processes that are not always conscious or strategic; these issues are discussed in chapters 3, 4 and 5.

What is clear is that individuals are able to bring a considerable amount of self-reflection to bear on the practices and discourses of day-to-day living. In chapter 4 I discuss the ways in which bodily praxis can act as a form of self-reflection that does not always enter the discursive. This clearly raises issues of intentionality and agency, and a number of chapters address the question of agency and its relationship to social determinations. One problem here is how to integrate people’s self-images and self-representations with dominant cultural ideologies and/or discourses. Any approach to the analysis of agency must include a consideration of the role of fantasy and desire, both with regard to questions of compliance and resistance and in connection with the construction of a sense of self. These points are elaborated further in chapter 3.

Chapter 5 addresses the politics of identity and its relationship to notions of rights and needs. Rhetorical strategies that draw on categorical and/or stereotypical identities are put into play in circumstances where political and economic resources are at stake. I discuss the redistribution of household resources and how resource flows are determined by a field of power within which identities are constantly being reformulated in categorical terms. The power to define reality is an economic and political power. The experience for individuals of such external definitions has consequences not only for their self-images, but for the material circumstances they find themselves in. I also discuss the way in which notions of agency are predicated upon theories of rights and needs that are implicated in certain identities, and how this varies for persons of different race, gender and class.

Language and the imagination

The construction of a self in relation to other selves involves the enunciation of a series of speaking positions. The taking up of a position on an issue that directly concerns one is always difficult, but it is possible to maintain a critical reflection on one’s own experience and on the various positions/locations one chooses to adopt. The position of the anthropologist has always been ambiguous and uneasy because it has depended on a stable division between ‘us’ and ‘them’. This should not obscure the fact, however, that the anthropological ‘we’ has always been an imaginary category. In chapters 2 and 6, I discuss the fact that at the present time anthropology has at least as much trouble with the unstable nature of the category ‘us’ as it does with the category ‘them’. When cultural theorists and colonial discourse theorists discuss anthropology and its representations of the other, they frequently conflate many complex issues, not least because they appear to assume that all anthropologists are Euro-Americans. This effectively silences what many anthropologists have to say about these problems, and it erases the real difficulties in moving between the poles of a contrastive pair in order to demonstrate where the lines of difference solidify and where they break down. These points are discussed more fully in chapters 1, 2 and 7.

One issue that is raised immediately is the question of where anthropologists get their models from. I suggest in chapters 6 and 7, where I discuss writing and the anthropological imagination, that we do not spend enough time attending to the fantasies and imaginative images of the anthropologists themselves. The anthropological self, like other selves, is one made up through projection and introjection, through identification and recognition and through a desire to belong. How do we construct images of ourselves as anthropologists; and how do the resulting images mark our work and our writing?

This question finds particular force at various times in the book when I address the problem of language and, in particular, the construction of theoretical language. One problem here concerns the degree to which a certain way of conceiving of and talking about gender difference which is prevalent in the social sciences is appropriate for discussing alternative ways of modelling sex/gender difference. This issue is raised in chapters 1 and 2, where I argue that social science and psychoanalytic models are themselves based on local folk models, and the view they produce of sex/gender difference is thus a very ethnocentric one.

Difficulties in this area are compounded because anthropologists and feminists, and indeed different groups of feminists, habitually use a set of terms that they imagine have a common meaning. The terms ‘sex’, ‘gender’, ‘sexuality’, ‘gender relations’ and ‘social relations’ are used in a number of quite distinct and very different ways. The result is that, even within the feminist and anthropological communities, we spend a great deal of time talking past each other. Part of the problem can be traced to different intellectual and linguistic traditions, but the rest is probably due to the fact that our use of these terms and the metaphoric resonances they set up for us are grounded in our own bodies and our own experiences. Since we are all gendered individuals, and since we can only speak the social through our selves and through our bodies, it is clear that these terms can never refer to pure concepts.

When writing this book, I constantly came up against the limits of my language and the limits of my imagination. What these texts are about is the struggle to develop a specifically anthropological approach to feminist post-structuralist theory. They are also an attempt to provide a radical critique of anthropology from a feminist perspective. In my writing I change my position and location many times. I have tried, where possible, to reveal the lines of fracture and ambiguity in my own thinking because I have wanted to try to show how much my theorizing is marked through with the specifics of a particular feminist anthropological self.



This essay was originally presented as a paper, and since much of what it discusses turns on problems of position, location, self-representation and representativity, I have decided to leave it, as far as is possible, in its original form. Extensive use of the first-person pronoun is frowned on in the contexts in which I am used to working, but I have deliberately retained it in this text to try to convey a sense of particularity, of myself speaking in a specific context(s). The use of ‘we’ is a highly politicized act both in anthropology and in feminist contexts. Its use here is intended to convey a sense of audience, that is, of myself speaking to others. But, and much more importantly, it also operates as a mark of interrogation, a fictive unity that reveals the lines of fragmentation at the very moment when it claims affinity.1

The original impetus for this paper was a question concerning the way in which feminism had influenced or affected my own work. This perfectly reasonable request engendered in me a feeling of intense panic. My first thought was ‘Oh God, how has feminism influenced my work?’ The root of the anxiety, of course, is one about being found out, being exposed as ‘not the real thing’, ‘not a proper feminist’. The anxiety of failure and lack is not entirely confined to feminists. In fact, it is probably rather a common paranoia among academics. However, what this anxiety raises for me as a feminist is the question of positionality. Feminist politics and feminist practice have always required a clear sense of position and of the politics of location. For one thing, there has been the necessity of speaking out, declaring one’s feminist politics within the workplace or the home or the political party or wherever. In addition, the powerful, sometimes acrimonious debates within the feminist community itself have demanded that one own up as to where one locates oneself in terms of a variety of carefully drawn and demarcated internal divisions: radical feminist or socialist feminist, for example? These divisions are important because they have guided the political programmes proposed by different groups of feminists, and because they bring already politicized identities into play. They raise, therefore, what I am going to call, after Nancy Miller (1991: 20), the problem of representativity. Who and what do we represent when we speak out, and how do we negotiate the inevitable problem in the social sciences of having to speak about people whilst trying not to speak for them? The question of who speaks for whom and on what basis has given rise in feminist debate to a number of very significant divisions, one of which is the split between theory and practice. The main issue here is how to link theoretical work with political activism. Those who have not seen themselves as theorists have demanded to know what purpose theory serves for them and how readily, if at all, theory takes account of their experiences, concerns and struggles. Feminist theory has seemed to many not only arcane, but elitist, racist and/or patriarchal.

Thus, the politics of location make two things abundantly clear. First, that there is no single, homogenous body of feminist theory; and secondly, that the divisions between different groups of women, as well as between practising feminists, make it impossible to assert a commonality based on shared membership in a universal category ‘woman’. Such divisions have a particular resonance for me because I work as a social anthropologist. As it happens, I work with and across divisions of race, class, sexuality, ethnicity and religion. I question the purpose of my work, especially my theoretical writing, for the people I work with because I do not find it easy to know of what immediate use it could be to them. I frequently try to deal with this problem, at least in part, by grounding my theoretical thinking in the details of daily life and in the realities of post-colonial political economies. I do not succeed in this as often as I should like, and I tenaciously hold on to what I try to convince myself is an acceptable political position by giving as much space and time to working on issues of agricultural change, women’s labour and nutrition as I do to writing on theoretical questions. The gross imbalances of power involved in my research situation mean that at every turn the very fact of writing and talking about other people’s lives can never be clearly separated from the question of whether or not one is speaking for them. This is a perennial problem for all feminist social scientists, in spite of a commitment to feminist methodologies and participatory research. Many of my feminist colleagues are very critical of my involvement in anthropology, often projecting on to me their own anxieties about how to deal with issues of race and class, and about how to manage the increasing gap between feminist activism and the academy. I inevitably do the same to them. The most significant impact that feminism has had on my work has been to create a space in which I must continually engage with these issues of positionality and representativity. I want to take up a very small part of this theme in this essay and discuss the way in which theoretical treatments of sex, gender and sexual difference are connected to what it is that unites and what it is that divides us as women and as feminists.

The assertion of the non-universal status of the category ‘woman’ is by now almost a commonplace. Anthropology has had a particular historical role in the development of feminist theory because of its contribution to the critical reworking of the category ‘woman’. In the 1970s feminists outside anthropology drew readily on the cross-cultural data provided by anthropological research to establish variability in gender and gender roles, and thus provide substantive content for the feminist position that gender was socially constructed and not biologically determined. However, cross-cultural variability in the social construction of gender could not and did not account for women’s universal subordination, and in order to remedy this, anthropology developed two very important comparative theories.

The first asserted that women everywhere were associated with nature, partly as a result of their reproductive functions, while men were associated with culture. It was suggested that the devaluing of nature in relation to culture accounted for the hierarchical relations between women and men (see Ortner, 1974). The second theory emphasized that women were inferior to men because they were linked to the domestic sphere, once again in consequence of their role in reproduction and child care, whilst men were associated with the public sphere of social life (see Rosaldo, 1974). These comparative theories of women’s subordination were not long-lived. The categories of nature, culture, public and private were themselves found to be historically and culturally variable, and the homologies posited between these categories and the categories of gender difference were revealed to be far from universal (see Moore, 1988: 13–30; MacCormack and Strathern, 1980; Strathern, 1984; and Rosaldo, 1980). What is important about these two comparative theories of women’s subordination is that they attempted to provide socially, as opposed to biologically, based accounts of women’s position in society and of the origins of gender difference. The preconditions for this project were, of course, that the biological and the social had already been separated from each other as explanations for the origins of gender difference. Whatever role biology was playing, it was not determining gender.

The very fact that these comparative theories were social rather than biological in their determinations opened them to critical reinterpretation by feminists of colour, feminists from the developing world and lesbian feminists. They challenged the notion of the universal category ‘woman’ and the assumption of underlying commonalities of existence for all women. Trans-cultural and trans-historical patterns of female subordination were rejected, and theoretical concepts were reformulated.2 In the social sciences, at least, this produced a crisis both about the political purpose and organization of a feminist politics which did not appear to have a coherent constituency and about the status of analytical models of gender. In general, it would probably be fair to say that many responded to the latter crisis by asserting the necessity for culturally and historically specific analyses. We could look for commonalities between well-specified situations, but we would never be able to state in advance what the consequences of the intersections of race, class and gender, for example, would be. What is interesting about this crisis is that it generated a simultaneous move towards pluralism and specificity. An enormous range of empirical outcomes and theoretical positions were produced as a result of having to reduce the scope of any model or analytical statement to a particular situation. We now recognize this development as part of a general critique of universalizing theories, metanarratives and totalizing typologies. The current debate is, of course, one about whether we locate the origins of this movement in post-structuralism and deconstructionism or in feminism.

However, as regards feminist theory in the social sciences, the shift in methods of gender analysis towards a specificity which would account for a plurality of experiences and contexts was not as radical as it seemed. One fixed position remained and that was the division between sex and gender. Gender was seen as socially constructed, but underlying that idea was a notion that although gender was not determined by biology, it was the social elaboration in specific contexts of the obvious facts of biological sex difference. It did not matter that almost everyone recognized that both biology and culture were historically and culturally variable concepts, as were the relations between them. The problem was that the elaboration of the social determinations and entailments of gender in all their specificity had effectively left the relationship between sex and gender very under-theorized.

Recent work in anthropology has returned to this question of the relationship between sex and gender. Sylvia Yanagisako and Jane Collier (1987) have suggested that the radical separation of sex and gender characteristic of feminist anthropology is a specific and rather pervasive ethnocentrism. They argue that it is part of a western folk model which dominates anthropological theorizing and, like so many of the other binary categorizations in anthropology – nature/culture, public/private – it does not stand up to cross-cultural examination. In many ways this simply marks the impact of neo-Foucauldian thinking in anthropology. It is worth recalling here Foucault’s argument in The History of Sexuality (vol. i) 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:

the notion of ‘sex’ made it possible to group together, in an artificial unity, anatomical elements, biological functions, conducts, sensations, and pleasures, and it enabled one to make use of this fictitious unity as a causal principle, an omnipresent meaning; sex was thus able to function as a unique signifier and as a universal signified. (1978: 154)

Foucault’s basic argument is that the notion of ‘sex’ does not exist prior to its determination within a discourse in which its constellations of meanings are specified, and that therefore bodies have no ‘sex’ outside discourses in which they are designated as sexed. Consequently, the construction of fixed binary sexes, with fixed categorical differences, is the effect of a specific discourse. What is more, if binary sex is an effect of discourse, then it cannot be considered as a unitary essentialism and, more importantly, it cannot be recognized as invariant or natural. This is, in essence, the argument Thomas Laqueur makes so elegantly in his recent book (1990) and two quite radical positions follow from this point.

First, in terms of anthropological discourse the distinction between sex and gender on which feminist anthropology has rested its case falls away. As Judith Butler (1990) points out in her reading of the above passage from Foucault, perhaps there is no distinction to be made between sex and gender after all. The second point, which follows from the first, is that, as Yanagisako and Collier (1987) assert, we cannot necessarily assume that binary biological sex everywhere provides the universal basis for the cultural categories ‘male’ and ‘female’. If gender constructs are culturally variable, then so are the categories of sexual difference. This is not the first time in anthropology or anywhere else that the fixed binary categories of sex have been interrogated; one only needs to point to the research that exists on ‘the third sex’, hermaphrodism and androgeny.3 But recent work in anthropology has a rather different purpose.

We know that the recognition of anatomical differences between women and men does not necessarily produce a discrete, fixed, binary categorization of sex in the manner of western discourse. Ethnographic material suggests that the differences between women and men which people in other cultures naturalize and locate in the human body and in features of the physical and cosmological environment are not necessarily those which correspond to the constellation of features on which western discourse bases its categorizations. For example, the social differences between women and men may be located in the body as natural differences, as in situations described by anthropologists working in Nepal, where the differences between the female and the male are conceived of as the difference between flesh and bone.4 However, these differences of gender are said to be located in all bodies, thus collapsing the distinction between sexed bodies and socially constructed genders usually maintained in anthropological discourse. The female and the male, as flesh and bone, are necessary features of bodily identity. This produces a discursive space where theories of social (gender) difference are grounded in the physiology of the body, and thus function as part of the biological facts of sex difference.

This is, of course, very close to Foucault’s own project, which is concerned with how it is that sexual differences and the category of sex are constructed within discourse as necessary features of bodily identity. In western discourse, it appears, it is not just that we need to have a body in order to have a sex, but that we need a sex so as to have a body. This rather strange way of thinking, of modelling the relationship between bodies and the categories of sexual difference, is precisely that which is most readily undermined by ethnographic material. Many of the differences which concern people around the world are internal to bodies, that is, within them rather than between them. The question is, are we to speak of these differences as differences of sex or of gender? This point is difficult to grasp for many of us because we have the gravest difficulty in understanding categories of sex and notions of sexual difference which do not correspond neatly to discrete physical bodies already designated as sexually differentiated. Sex, then, as far as we understand it within the terms of western discourse, is something which differentiates between bodies, while gender is the set of variable social constructions placed upon those differentiated bodies. It is precisely this formula which obscures rather than illuminates when it comes to the cross-cultural analysis of sex, sexual difference and gender. In many instances, as I have already suggested, gender differences are internal to all bodies and are part of the process through which bodies are sexed. In such situations it is far from apparent how we should distinguish sex from gender, and, even more problematic, it is unclear exactly what gender as a concept or a category refers to. This argument is quite different from those which have been made about the ‘third sex’, hermaphrodism and androgeny.

The instability – potential instability – of the category ‘gender’ in cross-cultural analysis is an alarming prospect. When we talk in general terms about discourses on gender and on the relationship between sex and gender, even if by this we only really mean to say different ideas about sex and gender, we still have to ask ourselves, whose discourses are we referring to? At one time anthropology subscribed to the view that each culture had its own model of gender, its own definitions of the categories female and male. This view, which was much reinforced by a predominantly Durkheimian view of culture and by the kind of liberal cultural relativism still prevalent in the discipline, has changed in recent years as anthropologists have moved towards working with models of culture which stress conflict and indeterminacy, and as they concentrate more on the differences within cultures as opposed to simply between them.5 However, it does not solve the problem of how to link what we might call dominant cultural models of gender to the specific experiences and situations of particular groups or individuals within that social context. This is not, of course, a problem which is confined to anthropology, but it raises once again the problems of positionality and representativity.

One set of difficulties here is about how the experiences of race, sexuality and class, as well as other forms of salient difference, transform the experience of gender. But there are additional problems about how we are to conceptualize and analyse the overdetermined relationships between dominant and sub-dominant discourses on gender, the body, sexuality and sexual difference. These questions become particularly acute when we acknowledge that they are crucial not only in and for our work, but in and for our lives. What relationship do feminist understandings of gender have to dominant gender models and ideologies; can the former ever be entirely free of the latter; is this what we are striving for? This is a matter of subjectivity and self-identity, as well as a matter of politics. When we are busy discussing other people’s discourses on gender, their views about the body, their gender identities and subjectivities, how easy do we find it to produce the kind of analysis which we would like to see applied to ourselves?

As Adrienne Rich remarked:

Perhaps we need a moratorium on saying ‘the body’. For it’s also possible to abstract ‘the body’. When I write ‘the body’, I see nothing in particular. To write ‘my body’ plunges me into lived experience, particularity … To say ‘the body’ lifts me away from what has given me primary perspective. To say ‘my body’ reduces the temptation to grandiose assertions. (1986: 215)

By ‘grandiose assertions’ Rich means presumably universalizing, comparative theories. As a lesbian feminist, Rich is only too well aware that the dominant discourses on gender, the body and sexuality prevalent in her own cultural setting do not fit her personal understanding of these categories and/or processes very closely. Lesbians, like many other groups, have evolved their own discourses, what some have termed sub-dominant or alternative discourses, on these issues. It is on this basis that writers talk of different kinds of experience – ‘the lesbian experience’ or ‘the black experience’, for example – and seek in terms of feminist theory to establish the grounds for theoretical approaches based on positionality and representativity. However, the problem is not just how to recognize the existence of specific groups who may have alternative perspectives and may not subscribe to dominant discourses within any particular setting. The more pressing problem with regard to gender, the body and sexual difference is to work out what bearing social and cultural discourses have on individual experience.

This is, of course, simply a modern version of an old problem in sociology and anthropology about the relationship between the individual and society. In anthropology this problem has often been run in terms of the relationship between dominant cultural symbols and the individual’s understanding and interpretation of them. This is a key issue in feminist theory, where feminist standpoint theory invites us to take women’s experiences as a starting point for analysis (see, for example, Harding, 1987, and chapter 4 of this volume). Standpoint theory assumes that women have a different perspective from men, and that different groups of women will also differ in their standpoints. In this sense it privileges groups over individuals, but a more radical reading of its premises would suggest that we all of us have different experiences and understandings of cultural discourses, symbols and institutions. The question is how much any of us share with each other.

The specific and the universal, the particular and the comparative – how are these two polarities to be brought into conjunction with each other? I have always been a supporter of the specific and the particular over the universal and the comparative, and I have always assumed that this is the result of my experience of research in Africa. However, I was listening to Catherine MacKinnon lecturing recently on women and human rights.6 MacKinnon holds to a radical feminist version of standpoint theory; in her work she constantly emphasizes what it is that women, in the global sense, share, and her work has been extensively criticized on precisely this point. She was talking about the mass rape and enforced impregnation of women in Croatia and Bosnia-Herzegovina. She argued simply that these crimes have been and continue to be practised on women in many different times and places, and without erasing or ignoring the specifics of what is going on in the former Yugoslavia, it is important to recognize that women suffer these crimes at the hands of men and they do so because they are women. Women are in fact universal in their particularity. It was very hard at that moment to deny the force of her argument, or even to think of any compelling reason why I should ever have disagreed with it. Women do fear sexual violence. If we want some empirical justification for such a universalizing assertion, it is only a matter of looking at the various women’s grassroots organizations around the world and at what they are campaigning against.

Rosi Braidotti, starting from very different assumptions, makes an argument which has strong parallels with MacKinnon’s. She speaks of a vision of women as a collective singularity,