As much as any single figure could, Stoller put the distinction between sex and gender on the map for writers and researchers in the humanities and the social sciences. But if his basic ideas quickly became commonplace, they were also soon being used in ways that he could not have anticipated. With the tremendous revival of feminist politics in North America and western Europe in the late 1960s came renewed attempts to understand and contest the social disadvantages experienced by women and Stoller’s separation of sex from gender was pressed into service as the cutting edge of a critique of male domination. So, when Kate Millett began to outline her theory of patriarchy in Sexual Politics (1977 [1970]), which was one of the founding texts of second-wave feminism, she drew upon Stoller’s work to under­score her argument that ‘male and female are really two cultures’ since his evidence seemed to cast doubt upon ‘the validity and permanence of psycho-sexual identity’ as a fact of life. Yet, in staking this claim, Millett was actually moving in quite the opposite direction to Stoller’s own highly individualistic psychoanalytic theorizing; for, when she rephrased his distinction to read ‘sex is biological, gender psychological, and therefore cultural’ (my emphasis), she was but one step away from mapping the opposition between sex and gender on to that between nature and culture (Millett 1977: 29—31).

As we shall see in a moment, there is good reason to question the apparent obviousness of this equation, but it still tends to govern the meanings ascribed to the sex/gender distinction even today. Locating gender within the many-sided realm of culture became the primary means of challenging the supposed inevitabil­ity of women’s subordination, part of what the historian Joan Scott, looking back over more than a decade of feminist research, has called ‘a genuine historicization and deconstruction’ of masculinity and femininity that sought to minimize or reduce human biology’s capacity to underpin the spuriously ‘fixed and permanent quality’ of these terms (Scott 1988). ‘Gender’, accord­ing to Scott’s pithy definition, is simply ‘a social category imposed on a sexed body’. Perhaps the most influential attempt to define the relationship between sex and gender through the contrast between nature and culture occurred within feminist anthro­pology, notably in Gayle Rubin’s 1975 essay ‘The Traffic in Women: Notes on the “Political Economy” of Sex’. In a wide – ranging theoretical and cross-cultural analysis, Rubin argued that every known society has what she dubs ‘a sex/gender system’, that is:

a set of arrangements by which the biological raw material of human sex and procreation is shaped by human, social intervention and satisfied in a conventional manner, no matter how bizarre some of the conventions may be.

(Rubin 1975: 165)

Just as hunger may be satisfied by any number of different kinds of food, each of them ‘culturally defined and obtained’, so, in any given society, sex too is filtered through the culturally dominant codes that regulate the behaviour acceptable in men and women. But these codes police not only ‘the social relations of sexuality’; they also determine the social division between the sexes, the basis upon which men and women are placed into ‘mutually exclusive categories’. Pointing to what she saw as the arbitrariness inherent in such classificatory logics, Rubin insisted that:

Men and women are, of course, different. But they are not as different as day and night, earth and sky, yin and yang, life and death. In fact, from the standpoint of nature, men and women are closer to each other than either is to anything else – for instance, mountains, kangaroos, or coconut palms. The idea that men and women are more different from one another than either is from anything else must come from somewhere other than nature. . . Far from being an expression of natural differences, exclusive gender identity is the suppression of natural similarities. It requires repression: in men, of whatever is the local version of ‘feminine’ traits; in women, of the local definition of ‘masculine’ traits. The division of the sexes has the effect of repressing some of the personality characteristics of virtually everyone, men and women.

(Rubin 1975: 179-80)

On this view the goal of cultural critique is not only the unmasking of a restrictive and fundamentally flawed conception of nature, but also the liberation of a true, because more genuinely natural, human diversity from the chains of social convention. By clearing away these arbitrary and artificial cultural obstacles, it might even be possible to imagine ‘the overthrow of gender itself’ — or at least, this prospect seems to be implicit in Rubin’s argument (Butler 1990: 75).

Rubin’s essay remains one of the most remarkable attempts to think through the causes of gender inequalities, constructing a systematic theoretical framework that links work, kinship and politics. Drawing upon insights from Marxist economics, psycho­analytic accounts of identity and anthropological studies of marriage and the family, Rubin shows how men typically ‘have certain rights in their female kin’, whereas ‘women do not have the same rights either to themselves or to their male kin’ and may be used as bridewealth, trophies, gifts and even ‘traded, bought, and sold’. Yet the notion that such oppressive kinship systems represent ‘an imposition of social ends upon a part of the natural world’, including the translation of sex into gender, assumes that sex and nature are somehow unproblematically given and exist outside the particular stock of cultural knowledge which makes one society so different from another (Rubin 1975: 175—6).

‘Sex is sex,’ writes Rubin, ‘but what counts as sex is… culturally determined and obtained’ (Rubin 1975: 165). One of the primary lessons of Foucault’s History of Sexuality, however, has been that there is no simple sense in which ‘sex is sex’, and that our ideas and beliefs about sexuality have been revolutionized over the last hundred years — indeed, they are still changing. This disquieting observation need not commit us to a naive relativism, a conviction that there can be no such thing as objective knowledge. But it does entail recognizing that what has counted as truth or error has varied enormously over time and that the history of this distinction will always also be a history of those cultural practices such as science, medicine and law within which evidence and proof have been deployed and contested.

Sex and gender are therefore intimately related, but not because one is ‘natural’ while the other represents its transformation into ‘culture’. Rather, both are inescapably cultural categories that refer to ways of describing and understanding human bodies and human relationships, our relationship to our selves and to others. Sex and gender necessarily overlap, sometimes confusingly so. What once was baldly called a ‘sex change operation’ is now, not entirely euphemistically, known as ‘gender reassignment’, a term that reflects the growing instability of the body’s contours in many contemporary societies, its increasing malleability or openness to reinvention, whether through drugs, dress, discipline or surgery. Of course, there are limits to who or what we might become, though these are not always the limits we might expect. In English law, for example, regardless of how much one’s body may have changed its shape or form since birth, it was not possible to alter one’s legal status from male to female, or vice versa until 2003. In this respect, legality — and not, as Freud once wrote, ‘anatomy’ — is destiny.

As a rough approximation we might say that ‘sex’ is the name we give to the language through which we speak and come to know our desires, while ‘gender’ denotes the cultural practices or cultural media that enable these desires to be played out. In her important book Gender Trouble, Judith Butler has argued that gender is a symbolic form of ‘public action’ whose recurrence allows for our recognition as desiring and desirable subjects. For Butler:

gender is an identity tenuously constituted in time, instituted in an

exterior space through a stylized repetition of acts. The effect of gender

is produced through the stylization of the body and, hence, must be understood as the mundane way in which bodily gestures, movements, and styles of various kinds constitute the illusion of an abiding gendered self.

(Butler 1990: 140)

According to Butler’s theatrical metaphor, gender is fragile, provisional, unstable, the sum total of its appearances rather than the expression of a unifying core. Masculinity or femininity come in many transient guises, all of them in some measure unfinished or incomplete. And this is as true historically, when one considers the range of competing definitions of what it has meant to be a man or a woman, as it is true individually, when one remembers the difficulties in growing into and sustaining an identity. ‘Thus,’ as Freud noted, ‘we speak of a person, whether male or female, as behaving in a masculine way in one connection and in a feminine way in another’ (Freud 1973: 2.147). Yet he failed to add that we also disagree among ourselves as to what is appropriately masculine behaviour in one case or acceptably feminine in another.

Butler’s claim that gender is primarily an act of signification or representation can sound as if gender is a matter of choice, of picking up and discarding identities at will. Butler has herself cautioned against this popular, but deeply misguided reading of Gender Trouble:

The bad reading goes something like this: I can get up in the morning, look in my closet, and decide which gender I want to be today. I can take out a piece of clothing and change my gender, stylize it, and then that evening I can change it again and be something radically other, so that what you get is something like the commodification of gender, and the understanding of taking on a gender as a kind of consumerism.

(Kotz 1992: 83)

The flaw in this picture, says Butler, lies in its failure to take into account the contradictory mode in which we inhabit our sense of gender, not as an identity that we freely embrace, but one that we also struggle against, that sustains us at the same time as it constrains us. Like the everyday use of language from which it partly derives, gender underpins our capacity to make decisions and act upon them, while constantly slipping out of our control and ensnaring us in complex webs of meaning that no single individual can ever hope to master.

But the false image of the subject who selects her gender for herself is at least correct in suggesting that there are many genders. So various are the different conceptions of masculinity and femininity that emerge from the miscellany of sites and settings in modern societies, that we can justifiably refer to them in the plural as masculinities and femininities (see Connell 1987; 1995). How wide is the range of variation within and between these genders? The answer to this question will largely depend upon contingencies of time and place, but nevertheless critics have continued to disagree about how the problem should be theorized. Thus Teresa de Lauretis has claimed that today’s representations of gender are produced by a number of distinct ‘technologies of gender’ such as cinema or advertising and that we, as gendered subjects, can be seen to be ‘constructed across a multiplicity of discourses, positions, and meanings, which are often in conflict with one another’ (de Lauretis 1987: x). For de Lauretis these discursive contradictions may actually provide a breathing space, a moment in which new gender identities might begin to be fashioned. By contrast, though writing in roughly the same period and starting from similar analytic assumptions, Chantal Mouffe has argued that ‘despite their heterogeneity, discourses and practices do not take place in isolation’ but interact with one another to create ‘a common effect’. As a result ‘the feminine’ is invariably set up ‘as a subordinated pole to the masculine’, a process in which ‘the symbolism linked in a given society to the feminine condition plays a fundamental role’ (Mouffe 1983: 141). There can be no alleviation of gendered inequalities unless this symbolism is successfully confronted.