In 1983, Nora Räthzel from the autonomous women's collective of the West German independent Marxist journal, Das Argument, wrote to ask me to write a 'keyword' entry for a new Marxist dictionary. An editorial group from Das Argument had undertaken an ambitious project to translate the multi-volume Dictionnaire Critique du Marxism (Labica and Benussen, 1985) into German and also to prepare a separate German supplement that brought in especially the new social movements that were not treated in the French edition.1 These movements have produced a revolution in critical social theory internationally in the last twenty years. They have also produced - and been partly produced by - revolutions in political language in the same period. As Räthzel expressed it, 'We, that is the women's editorial group, are going to suggest some keywords which are missing, and we want some others rewritten because the women do not appear where they should' (personal communication, 2 December 1983). This gentle understatement identified a major arena of feminist struggle the canonization of language, politics, and historical narratives in publishing practices, including standard reference works.
'The women do not appear where they should.' The ambiguities of the statement were potent and tempting. Here was an opportunity to participate in producing a reference text. I had up to five typed pages for my assignment: sex/gender. Foolhardy, I wrote to accept the task.
There was an immediate problem: I am anglophone, with variously workable but troubled German, French, and Spanish. This crippled language accomplishment reflects my political location in a social world distorted by US hegemonic projects and the culpable ignorance of white, especially, US citizens. English, especially American English, distinguishes between sex and gender. That distinction has cost blood in struggle in many social arenas, as the reader will see in the discussion that follows. German has a single word, Geschlecht, which is not really the same as either the English sex or gender. Further, the dictionary project, translating foreign contributors' entries into German, proposed to give each keyword in German, Chinese (both ideogram and transcription), English, French, Russian (in transcription only), and Spanish. The commingled histories of Marxism and of imperialism loomed large in that list. Each keyword would inherit those histories.
At least I knew that what was happening to sex and gender in English was not the same as what was going on with género, genre, and Geschlecht. The specific histories of women's movement in the vast global areas where these languages were part of living politics were principal reasons for the differences. The old hegemonic grammarians - including the sexologists had lost control of gender and its proliferating siblings. Europe and North America could not begin to discipline the twentieth-century fate of its imperializing languages. However, I did not have a clue what to make of my sex/gender problem in Russian or Chinese. Progressively, it became clear to me that I had rather few clues what to make of sex/gender in English, even in the United States, much less in the anglophone world. There are so many Englishes in the United States alone, and all of them suddenly seemed germane to this promised five-page text for a German Marxist dictionary that was splitting off from its French parent in order to pay attention to new social movements. My English was marked by race, generation, gender (!), region, class, education, and political history. How could that English be my matrix for sex/gender in general? Was there any such thing, even as words, much less as anything eise, as 'sex/gender in general'? Obviously not. These were not new problems for contributors to dictionaries, but I felt, well, chicken, politically chicken. But the presses roll on, and a due date was approaching. It was time to pluck out a feather and write. In the late twentieth century, after all, we are ourselves literally embodied writing technologies. That is part of the implosion of gender in sex and language, in biology and syntax, enabled by Western technoscience.
In 1985 I was moderately cheered to learn that the editorial group really wanted an entry on the sex/gender system. That helped; there was a specific textual locus for the first use of the term - Gayle Rubin's (1975) stunning essay written when she was a graduate student at the University of Michigan, 'The traffic in women: notes on the political economy of sex'. I could just trace the fate of the 'sex/gender system' in the explosion of socialist and Marxist feminist writing indebted to Rubin. That thought provided very brief consolation. First, the editors directed that each keyword had to locate itself in relation to the corpus of Marx and Engels, whether or not they used the precise words. I think Marx would have been amused at the dead hand guiding the living cursor on the video display terminal. Second, those who adopted Rubin's formulation did so out of many histories, including academic and political interests. US white socialist feminists generated the most obvious body of writing for tracing the 'sex/gender system' narrowly considered. That fact itself was a complex problem, not a solution. Much of the most provocative feminist theory in the last twenty years has insisted on the ties of sex and race in ways that problematized the birth pangs of the sex/gender system in a discourse more focused on the interweaving of gender and class.2 It has seemed very rare for feminist theory to hold race, sex/gender, and class analytically together - all the best intentions, hues of authors, and remarks in prefaces notwithstanding. In addition, there is as much reason for feminists to argue for a race/gender system as for a sex/gender system, and the two are not the same kind of analytical move. And, again, what happened to class? The evidence is building of a need for a theory of 'difference' whose geometries, paradigms, and logics break out of binaries, dialectics, and nature/culture models of any kind. Otherwise, threes will always reduce to twos, which quickly become lonely ones in the vanguard. And no one learns to count to four. These things matter politically.
Also, even though Marx and Engels - or Gayle Rubin, for that matter had not ventured into sexology, medicine, or biology for their discussions of sex/gender or the woman question, I knew I would have to do so. At the same time, it was clear that other BIG currents of modern feminist writing on sex, sexuality, and gender interlaced constantly with even the most modest interpretation of my assignment. Most of those, perhaps especially the French and British feminist psychoanalytic and literary currents, do not appear in my entry on Geschlecht. In general, the entry below focuses on writing by US feminists. That is not a trivial scandal.3
So, what follows shows the odd jumps of continual reconstructions over six years. The gaps and rough edges, as well as the generic form of an encyclopaedia entry, should all call attention to the political and conventional processes of standardization. Probably the smooth passages are the most revealing of all; they truly paper over a very contentious field. Perhaps only I needed a concrete lesson in how problematic an entry on any 'keyword' must be. But I suspect my sisters and other comrades also have at times tended simply to believe what they looked up in a reference work, instead of remembering that this form of writing is one more process for inhabiting possible worlds - tentatively, hopefully, polyvocally, and finitely. Finally, the keyword entry exceeded five typed pages, and the chicken was plucked bare. The body had become all text, and the instrument for the inscription was not a feather, but a mouse. The new genitalia of writing will supply the analyst with her metaphors, as the sex/gender system transmogrifies into other worlds of consequential, power-charged difference.
Gender (English), Geschlecht (German), Genre (French), Género (Spanish)
[The root of the English, French, and Spanish words is the Latin verb, generare, to beget, and the Latin stem gener-, race or kind. An obsolete English meaning of 'to gender' is 'to copulate' (Oxford English Dictionary). The substantives 'Geschlecht', 'gender', 'genre', and 'género' refer to the notion of sort, kind, and class. In English, 'gender' has been used in this 'generic' sense continuously since at least the fourteenth century. In French, German, Spanish, and English, words for 'gender' refer to grammatical and literary categories. The modern English and German words, 'gender' and 'Geschlecht', adhere closely to concepts of sex, sexuality, sexual difference, generation, engendering, and so on, while the French and Spanish seem not to carry those meanings as readily. Words close to 'gender' are implicated in concepts of kinship, race, biological taxonomy, language, and nationality. The substantive 'Geschlecht' carries the meanings of sex, stock, race, and family, while the adjectival form 'geschlechtlich' means in English translation both sexual and generic. 'Gender' is at the heart of constructions and classifications of systems of difference. Complex differentiation and merging of terms for 'sex' and 'gender' are part of the political history of the words. Medical meanings related to 'sex' accrue to 'gender' in English progressively through the twentieth century. Medical, zoological, grammatical, and literary meanings have all been contested in modern feminisms. The shared categorical racial and sexual meanings of gender point to the interwoven modern histories of colonial, racist, and sexual oppressions in systems of bodily production and inscription and their consequent liberatory and oppositional discourses. The difficulty of accommodating racial and sexual oppressions in Marxist theories of class is paralleled in the history of the words themselves. This background is essential to understanding the resonances of the theoretical concept of the 'sex-gender system' constructed by Western anglophone feminists in the 1970s.4 In all their versions, feminist gender theories attempt to articulate the specificity of the oppressions of women in the context of cultures which make a distinction between sex and gender salient. That salience depends on a related system of meanings clustered around a family of binary pairs: nature/culture, nature/ history, natural/human, resource/product. This interdependence on a key Western political-philosophical field of binary oppositions - whether understood functionally, dialectically, structurally, or psychoanalytically - prob - lematizes claims to the universal applicability of the concepts around sex and gender; this issue is part of the current debate about the cross-cultural relevance of Euro-American versions of feminist theory (Strathern, 1988). The value of an analytical category is not necessarily annulled by critical consciousness of its historical specificity and cultural limits. But feminist concepts of gender raise sharply the problems of cultural comparison, linguistic translation, and political solidarity.]
Articulation of the problem area in the writings of Marx and Engels In a critical, political sense, the concept of gender was articulated and progressively contested and theorized in the context of the post-Second World War, feminist women's movements. The modern feminist concept for gender is not found in the writings of Marx and Engels, although their writings and other practice, and those of others in the Marxist tradition, have provided crucial tools for, as well as barriers against, the later politicization and theorization of gender. Despite important differences, all the modern feminist meanings of gender have roots in Simone de Beauvoir's claim that 'one is not born a woman' (de Beauvoir, 1949; 1952, p. 249) and in post-Second World War social conditions that have enabled constructions of women as a collective historical subject-in-process. Gender is a concept developed to contest the naturalization of sexual difference in multiple arenas of struggle. Feminist theory and practice around gender seek to explain and change historical systems of sexual difference, whereby 'men' and 'women' are socially constituted and positioned in relations of hierarchy and antagonism. Since the concept of gender is so closely related to the Western distinction between nature and society or nature and history, via the distinction between sex and gender, the relation of feminist gender theories to Marxism is tied to the fate of the concepts of nature and labour in the Marxist canon and in Western philosophy more broadly.
Traditional Marxist approaches did not lead to a political concept of gender for two major reasons: first, women, as well as 'tribal' peoples, existed unstably at the boundary of the natural and social in the seminal writings of Marx and Engels, such that their efforts to account for the subordinate position of women were undercut by the category of the natural sexual division of labour, with its ground in an unexaminable natural heterosexuality; and second, Marx and Engels theorized the economic property relation as the ground of the oppression of women in marriage, such that women's subordination could be examined in terms of the capitalist relations of class, but not in terms of a specific sexual politics between men and women. The classical location of this argument is Engels' The Origins of the Family, Private Property and the State (1884). Engels' analytic priority of the family as a mediating formation between classes and the state 'subsumed any separate consideration of the division of the sexes as an antagonistic division' (Coward, 1983, p. 160).5 Despite their insistence on the historical variability of family forms and the importance of the question of the subordination of women, Marx and Engels could not historicize sex and gender from a base of natural heterosexuality.
The German Ideology (Part I, Theses on Feuerbach) is the major locus for Marx and Engels' naturalization of the sexual division of labour, in their assumption of a pre-social division of labour in the sex act (heterosexual intercourse), its supposed natural corollaries in the reproductive activities of men and women in the family, and the consequent inability to place women in their relations to men unambiguously on the side of history and of the fully social. In The Economic and Philosophic Manuscripts of 1844, Marx refers to the relation of man and woman as the 'most natural relation of human being to human being' (Marx, 1964b, p. 134). This assumption persists in volume one of Capital (Marx, 1964a, p. 351). This inability fully to historicize women's labour is paradoxical in view of the purpose of The German Ideology and subsequent work to place the family centrally in history as the place where social divisions arise. The root difficulty was an inability to historicize sex itself; like nature, sex functioned analytically as a prime matter or raw material for the work of history. Relying on Marx's research on ethnographic writings (1972), Engels' Origins (1884) systematized Marx's views about the linked transitions of family, forms of property, the organization of the division of labour, and the state. Engels almost laid a basis for theorizing the specific oppressions of women in his brief assertion that a fully materialist analysis of the production and reproduction of immediate life reveals a twofold character: the production of the means of existence and 'the production of human beings themselves' (1884; 1972, p. 71). An exploration of this latter character has been the starting point for many Euro-American Marxist-feminists in their theories of the sex/gender division of labour.6
The 'woman question' was widely debated in the many European Marxist parties in the late nineteenth and early twentieth centuries. In the context of the German Social Democratic Party the other of the two most influential Marxist treatments of the position of women was written, August Bebel's Woman under Socialism (1883; orig. Women in the Past, Present and Future, 1878). Alexandra Kollontai drew on Bebel in her struggles for women's emancipation in Russia and the Soviet Union; and within German social democracy, Clara Zetkin, a leader of the International Socialist Women's Movement, developed Bebel's position in her 1889 'The Question of Women Workers and Women at the Present Time'.7
The gender identity paradigm The story of the political reformulations of gender by post-1960s Western feminists must pass through the construction of meanings and technologies of sex and gender in normalizing, liberal, interventionist-therapeutic, empiricist, and functionalist life sciences, principally in the United States, including psychology, psychoanalysis, medicine, biology, and sociology. Gender was located firmly in an individualist problematic within the broad 'incitement to discourse' (Foucault, 1976) on sexuality characteristic of bourgeois, male-dominant, and racist society. The concepts and technologies of 'gender identity' were crafted from several components: an instinctualist reading of Freud; the focus on sexual somatic-and psychopathology by the great nineteenth-century sexologists (Krafft-Ebing, Havelock Ellis) and their followers; the ongoing development of biochemical and physiological endocrinology from the 1920s; the psychobiology of sex differences growing out of comparative psychology; proliferating hypotheses of hormonal, chromosomal, and neural sexual dimorphism converging in the 1950s; and the first gender reassignment surgeries around 1960 (Linden, 1981). 'Second-wave' feminist politics around 'biological determinism' vs. 'social constructionism' and the biopolitics of sex/gender differences occur within discursive fields pre-structured by the gender identity paradigm crystallized in the 1950s and 60s. The gender identity paradigm was a functionalist and essentializing version of Simone de Beauvoir's 1940s insight that one is not born a woman. Significantly, the construction of what could count as a woman (or a man) became a problem for bourgeois functionalists and pre-feminist existentialists in the same historical post-war period in which the social foundations of women's lives in a world capitalist, male-dominant system were undergoing basic reformulations.
In 1958, the Gender Identity Research Project was established at the University of California at Los Angeles (UCLA) medical center for the study of intersexuals and transexuals. The psychoanalyst Robert Stoller's work (1968, 1976) discussed and generalized the findings of the UCLA project. Stoller (1964) introduced the term 'gender identity' to the International Psychoanalytic Congress at Stockholm in 1963. He formulated the concept of gender identity within the framework of the biology/culture distinction, such that sex was related to biology (hormones, genes, nervous system, morphology) and gender was related to culture (psychology, sociology). The product of culture's working of biology was the core, achieved, gendered person - a man or a woman. Beginning in the 1950s, the psychoendocrinologist, John Money, ultimately from the institutional base of the Johns Hopkins Medical School's Gender Identity Clinic (established 1965), with his colleague, Anke Ehrhardt, developed and popularized the interactionist version of the gender identity paradigm, in which the functionalist mix of biological and social causations made room for a myriad of 'sex/gender differences' research and therapeutic programmes, including surgery, counselling, pedagogy, social services, and so on. Money and Ehrhardt's (1972) Man and Woman, Boy and Girl became a widely used college and university textbook.
The version of the nature/culture distinction in the gender identity paradigm was part of a broad liberal reformulation of life and social sciences in the post-Second World War, Western, professional and governing élites' divestment of pre-war renditions of biological racism. These reformulations failed to interrogate the political-social history of binary categories like nature/culture, and so sex/gender, in colonialist Western discourse. This discourse structured the world as an object of knowledge in terms of the appropriation by culture of the resources of nature. Many recent oppositional, liberatory literatures have criticized this ethnocentric epistemological and linguistic dimension of the domination of those inhabiting 'natural' categories or living at the mediating boundaries of the binarisms (women, people of colour, animals, the non-human environment) (Harding, 1986, pp. 163-96; Fee, 1986). Second-wave feminists early criticized the binary logics of the nature/culture pair, including dialectical versions of the Marxist-humanist story of the domination, appropriation, or mediation of 'nature' by 'man' through 'labour'. But these efforts hesitated to extend their criticism fully to the derivative sex/gender distinction. That distinction was too useful in combating the pervasive biological determinisms constantly deployed against feminists in urgent 'sex differences' political struggles in schools, publishing houses, clinics, and so on. Fatally, in this constrained political climate, these early critiques did not focus on historicizing and culturally relativizing the 'passive' categories of sex or nature. Thus, formulations of an essential identity as a woman or a man were left analytically untouched and politically dangerous.
In the political and epistemological effort to remove women from the category of nature and to place them in culture as constructed and self-constructing social subjects in history, the concept of gender has tended to be quarantined from the infections of biological sex. Consequently, the ongoing constructions of what counts as sex or as female have been hard to theorize, except as 'bad science' where the female emerges as naturally subordinate. 'Biology' has tended to denote the body itself, rather than a social discourse open to intervention. Thus, feminists have argued against 'biological determinism' and for 'social constructionism' and in the process have been less powerful in deconstructing how bodies, including sexualized and racialized bodies, appear as objects of knowledge and sites of intervention in 'biology'. Alternatively, feminists have sometimes affirmed the categories of nature and the body as sites of resistance to the dominations of history, but the affirmations have tended to obscure the categorical and overdetermined aspect of 'nature' or the 'female body' as an oppositional ideological resource. Instead, nature has seemed simply there, a reserve to be preserved from the violations of civilization in general. Rather than marking a categorically determined pole, 'nature' or 'woman's body' too easily then means the saving core of reality distinguishable from the social impositions of patriarchy, imperialism, capitalism, racism, history, language. That repression of the construction of the category 'nature' can be and has been both used by and used against feminist efforts to theorize women's agency and status as social subjects.
Judith Butler (1989) argued that gender identity discourse is intrinsic to the fictions of heterosexual coherence, and that feminists need to learn to produce narrative legitimacy for a whole array of non-coherent genders. Gender identity discourse is also intrinsic to feminist racism, which insists on the non-reducibility and antagonistic relation of coherent women and men. The task is to 'disqualify' the analytic categories, like sex or nature, that lead to univocity. This move would expose the illusion of an interior organizing gender core and produce a field of race and gender difference open to resignification. Many feminists have resisted moves like those Butler recommends, for fear of losing a concept of agency for women as the concept of the subject withers under the attack on core identities and their constitutive fictions. Butler, however, argued that agency is an instituted practice in a field of enabling constraints. A concept of a coherent inner self, achieved (cultural) or innate (biological), is a regulatory fiction that is unnecessary - indeed, inhibitory - for feminist projects of producing and affirming complex agency and responsibility.
A related 'regulatory fiction' basic to Western concepts of gender insists that motherhood is natural and fatherhood is cultural: mothers make babies naturally, biologically. Motherhood is known on sight; fatherhood is inferred. Analysing gender concepts and practices among Melanesians, Strathern (1988, pp. 311-39) went to great pains to show both the ethnocentric quality of the self-evident Western assertion that 'women make babies' and the inferential character of all vision. She showed the productionist core of the belief that women make babies (and its pair, that man makes himself), which is intrinsic to Western formulations of sex and gender. Strathern argued that Hagen men and women do not exist in permanent states as subjects and objects within Aristotelian, Hegelian, Marxist, or Freudian frames. Hagen agency has a different dynamic and geometry. For Westerners, it is a central consequence of concepts of gender difference that a person may be turned by another person into an object and robbed of her or his status as subject. The proper state for a Western person is to have ownership of the self, to have and hold a core identity as if it were a possession. That possession may be made from various raw materials over time, that is, it may be a cultural production, or one may be born with it. Gender identity is such a possession. Not to have property in the self is not to be a subject, and so not to have agency. Agency follows different pathways for the Hagen, who as persons 'are composed of multiple gendered parts, or multiple gendered persons, who are interacting with one another as donors and recipients in maintaining the flow of elements through the body' (Douglas, 1989, p. 17). Sexist domination between persons can and does systematically occur, but it cannot be traced or addressed by the same analytical moves that would be appropriate for many Western social fields of meaning (Strathern, 1988, pp. 334-9). Butler could - cautiously - use Strathern's ethnographic arguments to illustrate one way to disperse the coherence of gender without losing the power of agency.
So, the ongoing tactical usefulness of the sex/gender distinction in life and social sciences has had dire consequences for much feminist theory, tying it to a liberal and functionalist paradigm despite repeated efforts to transcend those limits in a fully politicized and historicized concept of gender. The failure lay partly in not historicizing and relativizing sex and the historical-epistemological roots of the logic of analysis implied in the sex/gender distinction and in each member of the pair. At this level, the modern feminist limitation in theorizing and struggling for the empirical life and social sciences is similar to Marx and Engels' inability to extricate themselves from the natural sexual division of labour in heterosexuality despite their admirable project of historicizing the family.
Sex/gender differences discourse exploded in US sociological and psychological literature in the 1970s and 80s. (This is shown, for example, in the occurrence of the word gender as a keyword in the abstracts for articles indexed in Sociological Abstracts [from 0 entries between 1966 and 1970, to 724 entries between 1981 and 1985], and in Psychological Abstracts [from 50 keyword abstract entries from 1966 to 1970, to 1326 such entries from 1981 to 1985].) The explosion is part of a vigorous political and scientific contestation over the construction of sex and gender, as categories and as emergent historical realities, in which feminist writing becomes prominent about the mid-1970s, primarily in criticisms of 'biological determinism' and of sexist science and technology, especially biology and medicine. Set up within the epistemological binary framework of nature/culture and sex/ gender, many feminists (including socialist and Marxist feminists) appropriated the sex/gender distinction and the interactionist paradigm to argue for the primacy of culture-gender over biology-sex in a panoply of debates in Europe and the United States. These debates have ranged from genetic differences in mathematics ability of boys and girls, the presence and significance of sex differences in neural organization, the relevance of animal research to human behaviour, the causes of male dominance in the organization of scientific research, sexist structures and use patterns in language, sociobiology debates, struggles over the meanings of sex chromosomal abnormalities, to the similarities of racism and sexism. By the mid-1980s, a growing suspicion of the category of gender and the binarism sex/gender entered the feminist literature in these debates. That scepticism was partly an outgrowth of challenges to racism in the Euro-American women's movements, such that some of the colonial and racist roots of the framework became clearer.8
The sex-gender system Another stream of feminist sex/gender theory and politics came through appropriations of Marx and Freud read through Lacan and Levi-Strauss in an influential formulation by Gayle Rubin (1975) of the 'sex-gender system'. Her paper appeared in the first anthology of socialist/ Marxist feminist anthropology in the United States. Rubin and those indebted to her theorization adopted a version of the nature/culture distinction, but one flowing less out of US empiricist life and social science, and more from French psychoanalysis and structuralism. Rubin examined the 'domestication of women', in which human females were the raw materials for the social production of women, through the exchange systems of kinship controlled by men in the institution of human culture. She defined the sex-gender system as the system of social relations that transformed biological sexuality into products of human activity and in which the resulting historically specific sexual needs are met. She then called for a Marxian analysis of sex/gender systems as products of human activity which are changeable through political struggle. Rubin viewed the sexual division of labour and the psychological construction of desire (especially the oedipal formation) as the foundations of a system of production of human beings vesting men with rights in women which they do not have in themselves. To survive materially where men and women cannot perform the other's work and to satisfy deep structures of desire in the sex/gender system in which men exchange women, heterosexuality is obligatory. Obligatory heterosexuality is therefore central to the oppression of women.
If the sexual property system were reorganized in such a way that men did not have overriding rights in women (if there was no exchange of women) and if there were no gender, the entire Oedipal drama would be a relic. In short, feminism must call for a revolution in kinship. (Rubin, 1975, p. 199)
Adrienne Rich (1980) also theorized compulsory heterosexuality to be at the root of the oppression of women. Rich figured 'the lesbian continuum' as a potent metaphor for grounding a new sisterhood. For Rich, marriage resistance in a cross-historical sweep was a defining practice constituting the lesbian continuum. Monique Wittig (1981) developed an independent argument that also foregrounded the centrality of obligatory heterosexuality in the oppression of women. In a formulation which its authors saw as the explanation for the decisive break with traditional Marxism of the Movement pour la Libération des Femmes (MLF) in France, the group associated with Wittig argued that all women belong to a class constituted by the hierarchical social relation of sexual difference that gives men ideological, political and economic power over women (Editors of Questions feministes, 1980).9 What makes a woman is a specific relation of appropriation by a man. Like race, sex is an 'imaginary' formation of the kind that produces reality, including bodies then perceived as prior to all construction. 'Woman' only exists as this kind of imaginary being, while women are the product of a social relation of appropriation, naturalized as sex. A feminist is one who fights for women as a class and for the disappearance of that class. The key struggle is for the destruction of the social system of heterosexuality, because 'sex' is the naturalized political category that founds society as heterosexual. All the social sciences based on the category of 'sex' (most of them) must be overthrown. In this view, lesbians are not 'women' because they are outside the political economy of heterosexuality. Lesbian society destroys women as a natural group (Wittig, 1981).
Thus, theorized in three different frames, withdrawal from marriage was central to Rubin's, Rich's, and Wittig's political visions in the 1970s and early 80s. Marriage encapsulated and reproduced the antagonistic relation of the two coherent social groups, men and women. In all three formulations both the binary of nature/culture and the dynamic of productionism enabled the further analysis. Withdrawal of women from the marriage economy was a potent figure and politics for withdrawal from men, and therefore for the self-constitution of women as personal and historical subjects outside the institution of culture by men in the exchange and appropriation of the products (including babies) of women. To be a subject in the Western sense meant reconstituting women outside the relations of objectification (as gift, commodity, object of desire) and appropriation (of babies, sex, services). The category-defining relation of men and women in objectification, exchange, and appropriation, which was the theoretical key to the category 'gender' in major bodies of feminist theory by white women in this period, was one of the moves that made an understanding of the race/gender or race/sex system and the barriers to cross-racial 'sisterhood' hard for white feminists analytically to grasp.
However, these formulations had the powerful virtue of foregrounding and legitimating lesbianism at the heart of feminism. The figure of the lesbian has been repeatedly at the contentious, generative centre of feminist debate (King, 1986). Audre Lorde put the black lesbian at the heart of her understanding of the 'house of difference':
Being women together was not enough. We were different. Being gay-girls together was not enough. We were different. Being Black together was not enough. We were different. Being Black women together was not enough. We were different. Being Black dykes together was not enough. We were different . . . It was a while before we came to realize that our place was the very house of difference rather than the security of any one particular difference. (Lorde, 1982, p. 226)
This concept of difference grounded much US multi-cultural feminist theorizing on gender in the late 1980s.
There have been many uses and criticisms of Rubin's sex-gender system. In an article at the centre of much Euro-American Marxist and socialist-feminist debate, Hartmann (1981) insisted that patriarchy was not simply an ideology, as Juliet Mitchell seemed to argue in her seminal 'Women: the Longest Revolution' (1966) and its expansion in Women's Estate (1971), but a material system that could be defined 'as a set of social relations between men, which have a material base, and which, though hierarchical, establish or create interdependence and solidarity among men that enable them to dominate women' (Hartmann, 1981, p. 14). Within this frame, Hartmann attempted to explain the partnership of patriarchy and capital and the failure of male-dominated socialist labour movements to prioritize sexism. Hartmann used Rubin's concept of the sex-gender system to call for an understanding of the mode of production of human beings in patriarchal social relations through male control of women's labour power.
In the debate stimulated by Hartmann's thesis, Iris Young (1981) criticized the 'dual systems' approach to capital and patriarchy, which were then allied in the oppressions of class and gender. Note how race, including an interrogation of white racial positioning, remained an unexplored system in these formulations. Young argued that 'patriarchal relations are internally related to production relations as a whole' (1981, p. 49), such that a focus on the gender division of labour could reveal the dynamics of a single system of oppression. In addition to waged labour, the gender division of labour also included the excluded and unhistoricized labour categories in Marx and Engels, that is, bearing and rearing children, caring for the sick, cooking, housework, and sex-work like prostitution, in order to bring gender and women's specific situation to the centre of historical materialist analysis. In this theory, since the gender division of labour was also the first division of labour, one must give an account of the emergence of class society out of changes in the gender division of labour. Such an analysis does not posit that all women have a common, unified situation; but it makes the historically differentiated positions of women central. If capitalism and patriarchy are a single system, called capitalist patriarchy, then the struggle against class and gender oppressions must be unified. The struggle is the obligation of men and women, although autonomous women's organization would remain a practical necessity. This theory is a good example of strongly rationalist, modernist approaches, for which the 'postmodern' moves of the disaggregation of metaphors of single systems in favour of complex open fields of criss-crossing plays of domination, privilege, and difference appeared very threatening. Young's 1981 work was also a good example of the power of modernist approaches in specific circumstances to provide political direction.
In exploring the epistemological consequences of a feminist historical materialism, Nancy Hartsock (19833,13) also concentrated on the categories that Marxism had been unable to historicize: (1) women's sensuous labour in the making of human beings through child-bearing and raising; and (2) women's nurturing and subsistence labour of all kinds. But Hartsock rejected the terminology of the gender division of labour in favour of the sexual division of labour, in order to emphasize the bodily dimensions of women's activity. Hartsock was also critical of Rubin's formulation of the sex-gender system because it emphasized the exchange system of kinship at the expense of a materialist analysis of the labour process that grounded women's potential construction of a revolutionary standpoint. Hartsock relied on versions of Marxist humanism embedded in the story of human self-formation in the sensuous mediations of nature and humanity through labour. In showing how women's lives differed systematically from men's, she aimed to establish the ground for a feminist materialist standpoint, which would be an engaged position and vision, from which the real relations of domination could be unmasked and a liberatory reality struggled for. She called for exploration of the relations between the exchange abstraction and abstract masculinity in the hostile systems of power characterizing phallocratic worlds. Several other Marxist feminists have contributed to intertwined and independent versions of feminist standpoint theory, where the debate on the sex/gender division of labour is a central issue. Fundamental to the debate is a progressive problematization of the category labour, or its extensions in Marxist-feminist meanings of reproduction, for efforts to theorize women's active agency and status as subjects in history.10 Collins (1989a) adapted standpoint theory to characterize the foundations of black feminist thought in the self-defined perspective of black women on their own oppression.
Sandra Harding (1983) took account of the feminist theoretical flowering as a reflection of a heightening of lived contradictions in the sex-gender system, such that fundamental change can now be struggled for. In extending her approach to the sex-gender system to The Science Question in Feminism (1986), Harding stressed three variously interrelated elements of gender: (1) a fundamental category through which meaning is ascribed to everything, (2) a way of organizing social relations, and (3) a structure of personal identity. Disaggregating these three elements has been part of coming to understand the complexity and problematic value of politics based on gender identities. Using the sex-gender system to explore post-Second World War politics of sexual identity in gay movements, Jeffrey Escoffier (1985) argued for a need to theorize the emergence and limitations of new forms of political subjectivity, in order to develop a committed, positioned politics without metaphysical identity closures. Haraway's (1985) 'Manifesto for Cyborgs' (see this volume, pp. 149-81) developed similar arguments in order to explore Marxist-feminist politics addressed to women's positionings in multi-national science- and technology-mediated social, cultural, and technical systems.
In another theoretical development indebted to Marxism, while critical of both it and of the language of gender, Catherine MacKinnon (1982, p. 515) argued that
Sexuality is to feminism what work is to marxism: that which is most one's own, yet most taken away . . Sexuality is that social process which creates, organizes, expresses, and directs desire, creating the social beings we know as women and men, as their relations create society . . . As the organized expropriation of the work of some for the benefit of others defines a class - workers - the organized expropriation of the sexuality of some for the use of other defines the sex, woman.
MacKinnon's position has been central to controversial approaches to political action in much of the US movement against pornography, defined as violence against women and/or as a violation of women's civil rights; that is, a refusal to women, via their construction as woman, of the status of citizen. MacKinnon saw the construction of woman as the material and ideological construction of the object of another's desire. Thus women are not simply alienated from the product of their labour; in so far as they exist as 'woman', that is to say, sex objects, they are not even potentially historical subjects. 'For women, there is no distinction between objectification and alienation because women have not authored objectifications, we have been them' (1982, pp. 253-4). The epistemological and political consequences of this position are far reaching and have been extremely controversial. For MacKinnon, the production of women is the production of a very material illusion, 'woman'. Unpacking this material illusion, which is women's lived reality, requires a politics of consciousness-raising, the specific form of feminist politics in MacKinnon's frame. 'Sexuality determines gender', and 'women's sexuality is its use, just as our femaleness is its alterity' (p. 243). Like independent formulations in Lacanian feminisms, MacKinnon's position has been fruitful in theorizing processes of representation, in which 'power to create the world from one's point of view is power in its male form' (p. 249).
In an analysis of the gendering of violence sympathetic to MacKinnon's, but drawing on different theoretical and political resources, Teresa de Lauretis's (1984, 1985) approaches to representation led her to view gender as the unexamined tragic flaw of modern and postmodern theories of culture, whose faultline is the heterosexual contract. De Lauretis defined gender as the social construction of 'woman' and 'man' and the semiotic production of subjectivity; gender has to do with 'the history, practices, and imbrication of meaning and experience'; that is, with the 'mutually constitutive effects in semiosis of the outer world of social reality with the inner world of subjectivity' (1984, pp. 158-86). De Lauretis drew on Charles Peirce's theories of semiosis to develop an approach to 'experience', one of the most problematic notions in modern feminism, that takes account both of experience's intimate embodiment and its mediation through signifying practices. Experience is never im-mediately accessible. Her efforts have been particularly helpful in understanding and contesting inscriptions of gender in cinema and other areas where the idea that gender is an embodied semiotic difference is crucial and empowering. Differentiating technologies of gender from Foucault's formulation of technologies of sex, de Lauretis identified a specific feminist gendered subject position within sex/gender systems. Her formulation echoed Lorde's understanding of the inhabitant of the house of difference: 'The female subject of feminism is one constructed across a multiplicity of discourses, positions, and meanings, which are often in conflict with one another and inherendy (historically) contradictory' (de Lauretis, 1987, pp. ix-x).
Offering a very different theory of consciousness and the production of meanings from MacKinnon or de Lauretis, Hartsock's (1983 a) exploration of the sexual division of labour drew on anglophone versions of psychoanalysis that were particularly important in US feminist theory, that is, object relations theory as developed especially by Nancy Chodorow (1978). Without adopting Rubin's Lacanian theories of always fragmentary sexed subjectivity, Chodorow adopted the concept of the sex-gender system in her study of the social organization of parenting, which produced women more capable of non-hostile relationality than men, but which also perpetuated the subordinate position of women through their production as people who are structured for mothering in patriarchy. Preferring an object relations psychoanalysis over a Lacanian version is related to neighbouring concepts like 'gender identity', with its empirical social science web of meanings, over 'acquisition of positions of sexed subjectivity', with this concept's immersion in Continental cultural/textual theory. Although criticized as an essentializing of woman-as-relational, Chodorow's feminist object relations theory has been immensely influential, having been adapted to explore a wide range of social phenomena. Drawing on and criticizing Lawrence Kohlberg's Kantian theories, Gilligan (1982) also argued for women's greater contextual consciousness and resistance to universalizing abstractions, for example in moral reasoning.
Evelyn Keller developed a version of object relations theory to theorize systematic epistemological, psychic, and organizational masculine dominance of natural science (Keller, 1985). Keller foregrounded the logical mistake of equating women with gender.11 Gender is a system of social, symbolic, and psychic relations, in which men and women are differentially positioned. Looking at the expression of gender as a cognitive experience, in which masculine psychic individuation produces an investment in impersonality, objectification, and domination, Keller described her project as an effort to understand the 'science-gender system' (p. 8). Emphasizing social construction and concentrating on psychodynamic aspects of that construction, Keller took as her subject 'not women per se, or even women and science: it is the making of men, women, and science, or, more precisely, how the making of men and women has affected the making of science' (p. 4). Her goal was to work for science as a human project, not a masculine one. She phrased her question as, 'Is sex to gender as nature is to science?' (Keller, 1987).
Chodorow's early work was developed in the context of a related series of sociological and anthropological papers theorizing a key role for the public/private division in the subordination of women (Rosaldo and Lamphere, 1974). In that collection, Rosaldo argued the universal salience of the limitation of women to the domestic realm, while power was vested in the space men inhabit, called public. Sherry Ortner connected that approach to her structuralist analysis of the proposition that women are to nature as men are to culture. Many Euro-American feminist efforts to articulate the social positioning of women that followed Woman, Culture, and Society and Toward an Anthropology of Women (Reiter, 1975), both strategically published in the mid-1970s, were deeply influenced by the universalizing and powerful theories of sex and gender of those early collections. In anthropology as a discipline, criticisms and other outgrowths of the early formulations were rich, leading both to extensive cross-cultural study of gender symbolisms and to fundamental rejection of the universal applicability of the nature/ culture pair. Within the disciplines, there was growing criticism of universalizing explanations as an instance of mistaking the analytical tool for the reality (MacCormack and Strathern, 1980; Rosaldo, 1980; Ortner and Whitehead 1981; Rubin, 1984). As feminist anthropology moved away from its early formulations, they none the less persisted in much feminist discourse outside anthropological disciplinary circles, as if the mid-1970s positions were permanently authoritative feminist anthropological theory, rather than a discursive node in a specific political-historical-disciplinary moment.
The universalizing power of the sex-gender system and the analytical split between public and private were also sharply criticized politically, especially by women of colour, as part of the ethnocentric and imperializing tendencies of European and Euro-American feminisms. The category of gender obscured or subordinated all the other 'others'. Efforts to use Western or 'white' concepts of gender to characterize a 'Third World Woman' often resulted in reproducing orientalist, racist, and colonialist discourse (Mohanty, 1984; Amos et al., 1984). Furthermore, US 'women of colour', itself a complex and contested political construction of sexed identities, produced critical theory about the production of systems of hierarchical differences, in which race, nationality, sex, and class were intertwined, both in the nineteenth and early twentieth centuries and from the earliest days of the women's movements that emerged from the 1960s civil rights and anti-war movements.12 These theories of the social positioning of women ground and organize 'generic' feminist theory, in which concepts like 'the house of difference' (Lorde), 'oppositional consciousness' (Sandoval), 'womanism' (Walker), 'shuttle from center to margin' (Spivak), 'Third World feminism' (Moraga and Smith), 'el mundo zurdo' (Anzaldúa and Moraga), 'la mestiza' (Anzaldúa), 'racially-structured patriarchal capitalism' (Bhavnani and Coulson, 1986), and 'inappropriate/d other' (Trinh, 1986-7, 1989) structure the field of feminist discourse, as it decodes what counts as a 'woman' within as well as outside 'feminism'. Complexly related figures have emerged also in feminist writing by 'white' women: 'sex-political classes' (Sofoulis, 1987); 'cyborg' (Haraway, 1985 and this vol. pp. 149-81); the female subject of feminism (de Lauretis, 1987).
In the early 1980s, Kitchen Table: Women of Color Press was established in New York and began to publish the critical theoretical and other writings of radical women of colour. This development must be seen in the context of international publishing in many genres by women writing into consciousness the stories of their constructions, and thereby destabilizing the canons of Western feminism, as well as those of many other discourses. As the heterogeneous and critical subject positions of 'women of colour' were progressively elaborated in diverse publishing practices, the status of 'white' or 'Western' also was more readily seen as a contestable location and not as a given ethnicity, race, or inescapable destiny. Thus, 'white' women could be called to account for their active positioning.
Rubin's 1975 theory of the sex/gender system explained the complementarity of the sexes (obligatory heterosexuality) and the oppression of women by men through the central premise of the exchange of women in the founding of culture through kinship. But what happens to this approach when women are not positioned in similar ways in the institution of kinship? In particular, what happens to the idea of gender if whole groups of women and men are positioned outside the institution of kinship altogether, but in relation to the kinship systems of another, dominant group? Carby (1987), Spillers (1987), and Hurtado (1989) interrogated the concept of gender through an exploration of the history and consequences of these matters.
Carby clarified how in the New World, and specifically in the United States, black women were not constituted as 'woman', as white women were. Instead, black women were constituted simultaneously racially and sexually as marked female (animal, sexualized and without rights), but not as woman (human, potential wife, conduit for the name of the father) - in a specific institution, slavery, that excluded them from 'culture' defined as the circulation of signs through the system of marriage. If kinship vested men with rights in women that they did not have in themselves, slavery abolished kinship for one group in a legal discourse that produced whole groups of people as alienable property (Spillers, 1987). MacKinnon (1982, 1987) defined woman as an imaginary figure, the object of another's desire, made real. The 'imaginary' figures made real in slave discourse were objects in another sense that made them different from either the Marxist figure of the alienated labourer or the 'unmodified' feminist figure of the object of desire. Free women in US white patriarchy were exchanged in a system that oppressed them, but white women inherited black women and men. As Hurtado (1989, p. 841) noted, in the nineteenth century prominent white feminists were married to white men, while black feminists were owned by white men. In a racist patriarchy, white men's 'need' for racially pure offspring positioned free and unfree women in incompatible, asymmetrical symbolic and social spaces.
The female slave was marked with these differences in a most literal fashion - the flesh was turned inside out, 'add[ing] a lexical dimension to the narratives of woman in culture and society' (Spillers, 1987, pp. 67-8). These differences did not end with formal emancipation; they have had definitive consequences into the late twentieth century and will continue to do so until racism as a founding institution of the New World is ended. Spillers called these founding relations of captivity and literal mutilation 'an American grammar' (p. 68). Under conditions of the New World conquest, of slavery, and of their consequences up to the present, 'the lexis of reproduction, desire, naming, mothering, fathering, etc. [are] all thrown into extreme crisis' {p. 76). 'Gendering, in its coeval reference to African-American women, insinuates an implicit and unresolved puzzle both within currrent feminist discourse and within those discursive communities that investipte the problematics of culture' (p. 78).
Spillers foregrounded the point that free men and women inherited their name from the father, who in turn had rights in his minor children and wife that they did not have in themselves, but he did not own them in the full sense of alienable property. Unfree men and women inherited their condition from their mother, who in turn specifically did not control their children. They had no name in the sense theorized by Lévi-Strauss or Lacan. Slave mothers could not transmit a name; they could not be wives; they were outside the system of marriage exchange. Slaves were unpositioned, unfixed, in a system of names; they were, specifically, unlocated and so disposable. In these discursive frames, white women were not legally or symbolically fully human; slaves were not legally or symbolically human at all. 'In this absence from a subject position, the captured sexualities provide a physical and biological expression of "otherness'" (Spillers, 1987, p. 67). To give birth (unfreely) to the heirs of property is not the same thing as to give birth (unfreely) to property (Carby, 1987, p. 53).
This little difference is part of the reason that 'reproductive rights' for women of colour in the US prominentiy hinge on comprehensive control of children - for example, their freedom from destruction through lynching, imprisonment, infant mortality, forced pregnancy, coercive sterilization, inadequate housing, racist education, or drug addiction (Hurtado, 1989, p. 853). For white women the concept of property in the self, the ownership of one's own body, in relation to reproductive freedom has more readily focused on the field of events around conception, pregnancy, abortion, and birth, because the system of white patriarchy turned on the control of legitimate children and the consequent constitution of white females as woman, To have or not have children then becomes literally a subject-defining choice for women. Black women specifically - and the women subjected to the conquest of the New World in general - faced a broader social field of reproductive unfreedom, in which their children did not inherit the status of human in the founding hegemonic discourses of US society. The problem of the black mother in this context is not simply her own status as subject, but also the status of her children and her sexual partners, male and female. Small wonder that the image of uplifting the race and the refusal of the categorical separation of men and women - without flinching from an analysis of coloured and white sexist oppression - have been prominent in New World black feminist discourse (Carby, 1987, pp. 6-7; hooks, 1981, 1984).
The positionings of African-American women are not the same as those of other women of colour; each condition of oppression requires specific analysis that refuses the separations but insists on the non-identities of race, sex, and class. These matters make starkly clear why an adequate feminist theory of gender must simultaneously be a theory of racial difference in specific historical conditions of production and reproduction. They also make clear why a theory and practice of sisterhood cannot be grounded in shared positionings in a system of sexual difference and the cross-cultural structural antagonism between coherent categories called women and men. Finally, they make clear why feminist theory produced by women of colour has constructed alternative discourses of womanhood that disrupt the humanisms of many Western discursive traditions.
[I]t is our task to make a place for this different social subject. In so doing we are less interested in joining the ranks of gendered femaleness than gaining the insurgent ground as female social subject. Actually claiming the monstrosity of a female with the potential to 'name' . . . 'Sapphire' might rewrite after all a radically different text of female empowerment. (Spillers, 1987, p. 80)
While contributing fundamentally to the breakup of any master subject location, the politics of 'difference' emerging from this and other complex reconstructings of concepts of social subjectivity and their associated writing practices is deeply opposed to levelling relativisms. Non-feminist theory in the human sciences has tended to identify the breakup of 'coherent' or masterful subjectivity as the 'death of the subject'. Like others in newly unstably subjugated positions, many feminists resist this formulation of the project and question its emergence at just the moment when raced/sexed/ colonized speakers begin 'for the first time', that is, they claim an originary authority to represent themselves in institutionalized publishing practices and other kinds of self-constituting practice. Feminist deconstructions of the 'subject' have been fundamental, and they are not nostalgic for masterful coherence. Instead, necessarily political accounts of constructed embodiments, like feminist theories of gendered racial subjectivities, have to take affirmative and critical account of emergent, differentiating, self-representing, contradictory social subjectivities, with their claims on action, knowledge, and belief. The point involves the commitment to transformative social change, the moment of hope embedded in feminist theories of gender and other emergent discourses about the breakup of masterful subjectivity and the emergence of inappropriate/d others (Trinh, 1986-7, 1989).
The multiple academic and other institutional roots of the literal (written) category 'gender', feminist and otherwise, sketched in this entry have been part of the race-hierarchical system of relations that obscures the publications by women of colour because of their origin, language, genre - in short, 'marginality', 'alterity', and 'difference' as seen from the 'unmarked' positions of hegemonic and imperializing ('white') theory. But 'alterity' and 'difference' are precisely what 'gender' is 'grammatically' about, a fact that constitutes feminism as a politics defined by its fields of contestation and repeated refusals of master theories. 'Gender' was developed as a category to explore what counts as a 'woman', to problematize the previously taken-for-granted. If feminist theories of gender followed from Simone de Beauvoir's thesis that one is not born a woman, with all the consequences of that insight, in the light of Marxism and psychoanalysis, for understanding that any finally coherent subject is a fantasy, and that personal and collective identity is precariously and constantly socially reconstituted (Coward, 1983, p. 265), then the title of bell hooks's provocative book, echoing the great nineteenth-century black feminist and abolitionist, Sojourner Truth, Ain't I a Woman (1981), bristles with irony, as the identity of 'woman' is both claimed and deconstructed simultaneously. Struggle over the agents, memories, and terms of these reconstitutions is at the heart of feminist sex/gender politics.
The refusal to become or to remain a 'gendered' man or a woman, then, is an eminently political insistence on emerging from the nightmare of the all-too-real, imaginary narrative of sex and race. Finally and ironically, the political and explanatory power of the 'social' category of gender depends upon historicizing the categories of sex, flesh, body, biology, race, and nature in such a way that the binary, universalizing opposition that spawned the concept of the sex/gender system at a particular time and place in feminist theory implodes into articulated, differentiated, accountable, located, and consequential theories of embodiment, where nature is no longer imagined and enacted as resource to culture or sex to gender. Here is my location for a utopian intersection of heterogeneous, multi-cultural, 'Western' (coloured, white, European, American, Asian, African, Pacific) feminist theories of gender hatched in odd siblingship with contradictory, hostile, fruitful, inherited binary dualisms. Phallogocentrism was the egg ovulated by the master subject, the brooding hen to the permanent chickens of history. But into the nest with that literal-minded egg has been placed the germ of a phoenix that will speak in all the tongues of a world turned upside down.