To this point I have discussed the production of sexual character as if each milieu were independent of all others. It is time to bring into the analysis the structures that interrelate milieux (chapter 6) and their historical composition into a gender order for the society as a whole (chapter 7).

To start with the structure of power, workplace studies like those discussed in chapter 5 show that face-to-face relations are strongly conditioned by the general power situation between employers and employees and its materialization in particular labour processes. A notable case is the job of personal ‘secretary’ in business. An apparently very individualized relationship of mutual dependence and trust between the executive (generally a man) and the personal secretary (almost always a woman) in fact rests on sharp differences of income, the industrial vulnerability of the employee, and the overall social power and authority of men. A specific version of femininity is called for, in which technical competence and the social presentation of attractiveness, social skill and interpersonal compliance are fused. This kind of femininity has to be produced, and is by the informal training documented in Chris Griffin’s study of British girls moving from school into office work.

The power hierarchy among men in the industrial enterprise is clear enough, from managers and professionals at the top to unqualified manual workers at the bottom. In sharp contrast to the situation of personal secretaries, the men in manual industrial work are often in situations that allow a countervailing solidarity (one of the bases of unionism) and with it a rejection of the masculinity of the dominant group. John Lippert provides a striking description of the aggressive, sometimes violent, heterosexual masculinity produced among motor manufacturing workers in Detroit. The description can be matched in other countries: Meredith Burgmann’s account of ‘machismo’ among radical buil­ders’ labourers in Sydney and Paul Willis’s account of masculine ‘shop-floor culture’ among metal workers in Birmingham. The common elements are a cult of masculinity centring on physical prowess, and sexual contempt directed at managers, and men in office work generally, as being effete.

These examples also point to the gender structuring of pro­duction. Elements of sexual character are embedded in the distinctive sets of practices sometimes called ‘occupational cultures’. Professionalism is a case in point. The combination of theoretical knowledge with technical expertise is central to a profession’s claim to competence and to a monopoly of practice. This has been constructed historically as a form of masculinity: emotionally flat, centred on a specialized skill, insistent on professional esteem and technically based dominance over other workers, and requiring for its highest (specialist) development the complete freedom from childcare and domestic work provided by having wives and maids to do it. The masculine character of professionalism has been supported by the simplest possible mechanism, the exclusion of women. Women have had a long struggle even to get the basic training, and are still effectively excluded from professions like accountancy and engineering.

In manual trades, and manual work more broadly, the claim to competence is rather different. Here the most competent are not the most specialized but the most versatile — those with a range of skills, able to tackle any job that offers. This too is often constructed as a form of masculinity dependent on a domestic division of labour. Tradesmen have often been prepared to move around from place to place, even from country to country, to increase their range of experience, the wife’s willingness to stay or go being assumed. Fathers have taken care to provide their sons with a range of skills as insurance against economic fluctuations. To quote another British working-class autobiography, from a miner’s son called Fred Broughton who grew up in the years before World War I, ‘Father used to say, “I shall not leave you much money, but I will teach you every job, then you can always get work”. He showed us every job in the garden and on the farm, including how to get stone in the quarry and trim it and build stone walls.’

The construction of nursing as an element of the sexual division of labour, an occupation blending a particular version of femininity with the technical requirements of the job, has already been discussed.

Finally the structure of cathexis is involved. This is the most obvious of structural determinations of sexual character because of the prominence of heterosexual couple relationships in everyday life. It is folklore that ‘opposites attract’. One of the most familiar features of sexual display is behaviour and clothing that emphasizes stereotyped sex differences. Studs display their biceps and pectorals, suave charmers grow their pencil moustaches; ‘girls’ emphasize their vulnerability in tight skirts and high-heeled shoes, sheer stockings and make-up that is constantly in need of repair. So much emotion is adrift around these marks of difference that they can get cathected in their own right, as argued in chapter 5. These stereotypes are so familiar that it is necessary to stress that they are not the whole story. Alongside the Errol Flynns and John Waynes are figures like Cary Grant, whose appeal is specially as a model of sympathetic (though not effeminate) masculinity. In a study of images of masculinity in Australian television, Glen Lewis has pointed to the prominence of‘soft’ men as presenters, especially in daytime programs directed at women.

Desire may be organized around identification and similarity rather than around difference. Homosexual love is the obvious case. The attempt to reduce this to attraction-of-opposites by assuming it is based on a butch/femme pattern is now generally discredited. Gay liberation theory lays emphasis instead on the solidarity created by love between women or between men. The point is that there are many more possibilities than the standard dichotomy or complete structurelessness. Works like Pat Califia’s Sapphistry explore a variety of erotic constructions of femininity (homosexuality still presupposes gender division) based on identi­fication and shared experience; the same can be done for masculinity.

There is a related possibility among heterosexual people, for powerful desire can exist between those whose character structure is similar. An interplay between identification and reciprocity, and a literal playing with similarity and difference, becomes possible as a basis of eroticism. On such a basis heterosexual masculinity and femininity might be recomposed as various kinds of psychological hermaphroditism, a possibility I will return to in chapter 13.

To sum up: it is possible to see how each of the major structures impinges on the way femininity and masculinity are formed in particular milieux. Conversely, these structures must be seen as the vehicles for the constitution of femininity and masculinity as collective patterns on a scale far beyond that of an individual setting. In the terms proposed in Part II, we have moved from particular gender regimes to the society-wide gender order. The question now to be faced is how, at the level of a whole society, the elements are composed, interrelated and ordered.