The ebbing of the radical tide in the 1920s drained practical urgency from these debates about gender. In the next generation the main developments were academic. The ‘woman question’ in

politics had already produced a response from the new sciences of psychology and sociology. One line of research asked what were the psychological differences between women and men and how they were caused. Starting in the United States about the turn of the century, ‘sex difference’ research gradually accumulated in huge quantity, though highly variable quality. In the 1930s this tradition intersected with the new technology of standardized attitude and personality tests, in attempts to measure ‘masculinity/ femininity’ directly as a psychological trait. Paper-and-pencil scales of masculinity and femininity (M/F) were devised and soon put to work diagnosing gender deviance.

Gender-scaling was on the face of it neutral as to the sources of the characteristics being ‘scaled’. Academic social science was addressing that question on other terms. Jessie Taft developed the idea of women’s cultural marginalization, an approach notable because it placed power and exclusion at the centre of a social analysis of gender. The main line of academic thought, however, followed another path with the propagation in the 1930s of the concept of‘social role’. The notion of a socially provided script for individual behaviour, first learned and then enacted, was easily applied to gender.

By the 1940s the terms ‘sex role’, ‘male role’ and ‘female role’ were in use. By the end of that decade American sociologists such as Mirra Komarovsky and Talcott Parsons had spelt out a functional theory of sex roles and the cultural contradictions surrounding them. These ideas converged with the growing industries of counselling, marriage guidance, psychotherapy and welfare case work. The concept of a normative ‘sex role’ and various patterns of ‘deviance’ from it became enormously influen­tial, providing a practical warrant for intervention to straighten out deviants, and theoretical justification for the ‘helping professions’ as a whole. ‘Sex role’ has remained the central category of academic thought about gender ever since, with the sex-difference literature gradually slipping under the ‘role’ rubric as well.

In the meantime psychoanalysis had stirred anthropology in new directions. Freud and his follower Geza Roheim argued that the Oedipus complex was universal, some form of it appearing in every culture. In the 1920s and 1930s writers like Bronislaw Malinowski in Sex and Repression in Savage Society and Margaret Mead in Sex and Temperament in Three Primitive Societies drew on their own fieldwork to argue a general case for a connection

between social structure and the emotional dynamic of sexuality. Malinowski, truer to Freud, centred his argument on the functional necessity of repression and the elaboration of kinship customs as a means to that end. Mead was more interested in the emotional colouring of a whole culture, and her approach was formative in the ‘culture-and-personality’ school of American anthropology. Perhaps the most important effect of their work was simply to document the range of ways in which different cultures dealt with sex and gender. The exotica of life in the Trobriand Islands, Samoa and New Guinea dramatized for Westerners the idea of social scripting; it was difficult in the face of this to take any aspect of gender relations for granted.

By the mid-century a number of intellectual currents were converging and the stage was set for reflective synthesis. Three theorists published major works, covering remarkably similar issues, within five years of each other; one starting from field anthropology, one from theoretical sociology and one from existent­ial phenomenology. With them the social analysis of gender took its contemporary form.

Mead’s Male and Female, Talcott Parsons’s essays in Family Socialization and Interaction Process and Simone de Beauvoir’s The Second Sex had very different intellectual programs and politics. That perhaps makes what they had in common more notable. All took a psychoanalytic view of the making of personality. All tried to integrate this (though on different terms) with analysis of the division of labour conceived mainly in terms of sex roles or gender roles. Parsons was most systematic about this, as ‘role’ had become a fundamental term in his general sociology. In all three authors the sense of the social contingency of sexual character and gender relations had gone very deep. Mead’s deliberate cross-cultural contrasts dramatized the point most. But it was also assumed by Parsons, who in earlier essays had laid emphasis on the modernization of the female role in American society, and by de Beauvoir, who attempted a phenomenology of different kinds of femininity. Yet all three theorists tried to limit the free play that a complete sociologizing of gender would imply. Parsons did this by appeal to functional imperatives of society, Mead (the most conservative on this issue) by appeal to some rather ill-defined biological regularities in human development and de Beauvoir by the self/other structure of relations between men and women. All three defined gender patterns mainly in terms of relationships within the nuclear family, which all took to be, effectively, universal.

What de Beauvoir saw, and the others did not, was the dimension of power running through these relationships. Mead and Parsons, to put it in a phrase, synthesized the field of gender around the idea of custom and social stability. De Beauvoir synthesized it around the subordinationjaf^^era^n.

Tn thtTshort term the former was me-more influential approach. Parsons’s analysis of the family, particularly his distinction between ‘expressive’ and ‘instrumental’ roles, laid the basis for a conservative sociology of gender which took its place in the great expansion of American social science in the 1950s and 1960s. Its themes were the necessity of the nuclear family, the problems of personal adjustment to sex roles and techniques of intervention to keep the family in good repair. With ‘family’ and ‘sex role’ conflated, the actual focus of most of the resulting research was women as wives – and-mothers (the ‘female role’). Sex-difference and gender-scaling studies continued and were generally taken to confirm the role paradigm. Despite Parsons’s prestige, the area remained rather an academic backwater during these decades. It produced some notable pieces of field research, such as Komarovsky’s Blue-Collar Marriage and Young and Willmott’s Family and Kinship in East London. While these studies had considerable influence in social work and sometimes on social policy, they had little impact on social theory or the intellectual world of social science in general. It was not until the arrival of the new feminism at the end of the 1960s that a wider interest in gender was kindled, and then it was Simone de Beauvoir’s perspective that became central.