No one knows precisely when and where gender was initially used to refer to the social and cultural aspects of sexual difference, but it is clear that the term was already current in sexology by the early 1960s. For example, gender does not figure in Alex Comfort’s post-war overview of Sexual Behaviour in Society (1950) until the book was revised for publication some thirteen years later (under the new title Sex in Society), when the author added a brief discussion of ‘gender roles’. Significantly, this was placed in a chapter on ‘The Biological Background’ to human sexuality, in which Comfort stressed the difficulty in knowing the extent to which our sexual behaviour was instinctive, given ‘the far greater importance of higher mental functions in man’ than in the other animal species (Comfort 1963: 34). This explains why:

The ‘gender role’ which an individual adopts – ‘manly’ or ‘womanly’ – according to the standards of his culture, is oddly enough almost wholly learned, and little if at all built in; in fact, the gender role learned by the age of two years is for most individuals almost irreversible, even if it runs counter to the physical sex of the subject.

(Comfort 1963: 42)

‘Gender’ is used here to index the wide variation in styles of behaviour between societies, but it also suggests that within them the degree of choice is fairly limited. By making his readers aware of these cultural differences, Comfort hoped to demystify human sexuality and so help to release them from what he believed were unnecessary and irrational sexual taboos. At the same time, however, his more guarded emphasis upon the irreversibility of gender roles seems to tell against the promise of any easily accessible path to sexual liberation. For liberal humanists like Comfort it was often hard to understand why sexual enlighten­ment should appear to lag so far behind the other modes of social and technological progress.

Probably the most thorough attempt to theorize the distinction between sex and gender in this period is to be found in the writings of the psychoanalyst and anthropologist Robert J. Stoller, whose book Sex and Gender: On the Development of Masculinity and Femininity appeared in 1968. Stoller located the starting – point for his work in Freud’s paper on ‘The Psychogenesis of a Case of Homosexuality in a Woman’ (1920) which argued that a person’s physical sexual attributes, mental attitudes and objects of desire could ‘vary independently of one another’; so that ‘a man with predominantly male characteristics and also masculine in his erotic life may still be inverted in respect to his object, loving only men instead of women’ (Freud 1979: 9.398—9). In a similar vein, Stoller used the term ‘gender’ to signal the complexities of those ‘tremendous areas of behaviour, feelings, thoughts, and fantasies that are related to the sexes and yet do not have primarily biological connotations’ (Stoller 1968: ix). Not only do we tend to confuse sex and gender, however. We also assume too readily that the various components of gender are mutually reinforcing, whereas in fact they may well pull in different directions.

In addition to separating sex from gender analytically, therefore, Stoller distinguished between ‘gender role’ and ‘gender identity’ in order to indicate that one’s inner and outer life may be deeply conflicted or fail to coincide. The gender role that one plays out before others may offer little clue as to who one feels oneself to be, and consequently in Stoller’s theory the very definition of gender identity is founded upon the possibility of an inner discord, a kind of non-identity with one’s sexual being:

Gender identity starts with the knowledge and awareness, whether conscious or unconscious, that one belongs to one sex and not the other, though as one develops, gender identity becomes much more complicated, so that, for example, one may sense himself as not only a male but a masculine man or an effeminate man or even as a man who fantasies being a woman.

(Stoller 1968: 10)

By the end of this sentence gender identity has begun to sound like a form of alienation and, perhaps not surprisingly, much of Stoller’s clinical work was devoted to examining cases in which people felt uncertain as to whether they were ‘really’ masculine or feminine. According to Stoller the first few years of childhood usually gave rise to a stubborn core of gender identity and this could sometimes clash with subsequent experiences, values or wishes. But such conflicts were rarely simple. In a discussion of those patients whose genital abnormalities meant that they were brought up ‘in an atmosphere of parental doubt’, Stoller suggested that it was possible for them to develop ‘a hermaphroditic gender consciousness’, a unique core identity that recognized the division of the world into two sexes, while feeling that he or she was a social and a sexual misfit who ‘belongs to neither’ (Stoller 1968: 33-4). On the other hand, not all psychic conflicts were necessarily destructive. Commenting on the case of a patient who was a transvestite, Stoller observed that part of the man’s pleasure in heightening his sense of femininity through cross-dressing came from a simultaneous awareness of also being male. Here ‘the two aspects of gender identity’ — ‘the later one, I am feminine, and the earlier core identity, I am (nonetheless) a male — were equally ‘essential to his perversion’ (Stoller 1968: 40). To manipulate the limits of gender, to play with and upon a sense of the contra­dictions of identity, was, for this subject, to secure the conditions of the most intense sexual enjoyment.

The continuing legacy of Stoller’s work (and that of the Gender Identity Research Clinic which he set up in Los Angeles in the 1960s) can still be seen in the American Psychiatric Association’s official diagnostic manual. There one can find entries on ‘gender dysphoria’ or ‘gender identity disorder’, defined as a ‘persistent discomfort about one’s assigned sex or a sense of inappropriateness in the gender role of that sex’, a syndrome that often temporarily affects children — though no clue is offered as to why this should be so (American Psychiatric Association, DSM-IV 1994: 533).

While inevitably lacking the psychological subtleties of Stoller’s detailed case histories, these brief, abstracted symptomatologies do point up the weaknesses of his own narrowly clinical approach. Stoller had little time for broader social analyses of gender, dismissing Foucault’s ‘cultural history’ of sexuality as nothing more than ‘a story’, based on ‘shards’ and ‘fragments’, that ‘leaves out the information that the intimate study of individuals can add’ (Stoller 1996: 16—18).