52     DIFFERENCES AND BODIES
according to one respondent: 'Sometimes I think I could throw all this work out of the window. Sometimes it gets on top of you ... I just feel 1 want to pack it all in. I want to get miles away. 1 just can't go on any more. I don't want to go home and start getting the tea, but I do.' The pressure is much the same for all. However the responses of different women ranged widely, from stoic acceptance, through fantasy escape (notably colourŕul holidays, in the imagination), to resistance by creating a shop-floor culture, to trade union militancy. (At least for a time: their union happened to be controlled by men, and when the women went on strike it let them down.)
The reproductive arena is not fixed, it can he re-shaped by social processes. Indeed it constantly is being reshaped; there is social struggle over this as well as other aspects of gender. For instance, the fertility of a woman's body means something different where contraception is effective and small families are planned, from what it means where women are designated lifelong breeders and nurturers - barefoot, pregnant and in the kitchen, as the saying goes. There is social conflict over the potential meaning of women's fertility. 'Righr-to-Ufe* militants arc not just attempting to outlaw abortion: they seek to push the whole reproductive arena into the pattern they call 'the traditional family'. It is no accident that very few Right-to-Lifc activists campaign for the one straightforward and practical solution to the problem of abortion -effective contraception.
It is possible for social practice to move gender orders in different directions, and create different relations between bodies and social structures. The liberal-fem in ist idea expressed by Maccoby and Jacklin (quoted above), that a society can choose the gender order it wants, \* sociologically naive. A society divided by conflicting interests does not 'choose' as a unit. But Maccoby and Jacklin were not mistaken in seeing a range of historical possibilities in gender relations. There are different futures towards which contemporary societies might be moved, by mobilizing social forces on a sufficiently large scale. I will return to this issue in chapter 8.
m
Gender Relations
Patterns in gender
Chapter 2 included two studies of organizations, llarrie Thome's study of American elementary schools and Dunbar Moodic's study of South African mines. Bach of these organizations had a regular set of arrangements about gender: who was recruited to do what work (most of the teachers were women, ail of the mineworkers were men); what social divisions were recognized (e.g. creating 'opposite sides' in the playground); how emotional relations were conducted (e.g. the 'mine wives*); and how these institutions were related to others (e.g. the families of the workers).
Such a pattern in gender arrangements may be called the gentler regime of an institution. Research on a very wide range of organizations has mapped their gender regimes - schools, offices, factories, armies, police forces, sporting clubs. It is clear that gender regimes arc a usual feature of organizational life.
These studies make clear that the gender regime of an institution can change - though change is often resisted. An example is the merger of two gender-segregated English secondary schools described in a very interesting ethnography by Joan Draper (1993). After the merger some boys tried to establish dominance in the new social space, some girls accepted subordination, other girls fought it. Meanwhile other boys began experimenting with gender and turned up in dyed hair, eveshadow and nail polish. The teachers found the turmoil hard to handle and some became distressed at the loss of their previously established place
54    GENDER RELATIONS
in the educational world. Over time, however, a new gender regime crystallized.
When Thome went into Oceanside Elementary School and found that most of the teachers were women, she was not exactly surprised. That is the usual arrangement in elementary schools in the United States. Similarly, Moodie was not astonished to find an all-male workforce at the Witwarersrand gold mines he investigated. That is the usual arrangement in South African mines, and in mining all over the world.
The gender regimes of these particular organizations, then, are parr of wider patterns, which also endure over time. As in chapter 1, I call these wider patterns the gender order of a society. The gender regimes of institutions usually correspond to the overall gender order, but may depart from it. This is important for change. Some institutions change quickly, others lag; or to put it another way, change often starts in one sector of society and takes time to seep through into other sectors.
When we look at a set of gender arrangements, whether the gender regime of an institution or the gender order of a whole society, we are basically looking at a set of relationships - ways that people, groups, and organizations are connected and divided. 'Gender relations' are the relationships arising in and around the reproductive arena discussed in chapter 3. Not all gender relations are direct interactions between women on one side and men on the other. The relations may be indirect - mediated, for instance, by a market, or by technologies such as TV or the Internet. Relationships may be among men, or among women, but still are gender relations - such as hierarchies of masculinity among men.
Gender relations are always being constituted in everyday life. If we don't bring it into being, gender does not exist. This point is forcibly made by eth n o method o logy, a school of sociological research concerned with what we presuppose in everyday conduct. Candace West and Don Zimmerman, in a celebrated article called 'Doing Gender' (1987), show an impressive range of ways in which everyday speech constitutes gender relations. Not only are speakers identified in terms of their gender. Relationships between them, such as dominance, deference, antagonism, solidarity, are constantly being enacted in the course of conversations which are nominally about quite different subjects.
Yet we are not free to enact gender however we like. In reality, gender practice is powerfully constrained. When I, as an Australian academic in the 2000s, relate to people in gendered ways, I am not free to use the practices of a slave-owning Athenian aristocrat of the fifth century bc. Wrong meanings would be attached to my actions, and I would doubtless find time to work out my errors in gender theory from a cell in Long Bay Gaol.
GENDER RELATIONS    55
Social theory has attempted to capture the fact of constraint and the patterns in relationships with the concept of structure. Relations among people (or among groups or institutions) would have little significance if they were randomly arranged. Patterns in these relations would matter little if they were ephemeral. It is the enduring or extensive patterns among social relations that social theory calls 'structures'.
The gender arrangements of a societ}' involve social structure in this sense. For instance, if religious, political and conversational practices all place men in authority over women, we speak of a patriarchal structure of gender relations. Or if clans of men regularly marry each others' sisters, we speak of a kinship structure of exchange.
A structure of relations does not mechanically determine how people or groups act. That was the error of deterministic marxism. But a structure of relations certainly defines possibilities and consequences. For instance, the structure of gender relations in Australian society did not fix what sexual practices Huey Brown (chapter 2) would engage in. But they gave him a definite set of possibilities. When he took up certain of them - continuing sex with men, drag, and domestic partnership - the structure of gender relations defined powerful consequences for his life, which are traced in Gary Dowsett's case study.
In this sense, social structure conditions practice. This does not imply that structures cause, or exist separately from, practices. The structure of gender relations has no existence outside the practices through which people and groups conduct those relations. Structures do not continue, cannot be 'enduring', unless they are reconstituted from moment to moment in social action. In this sense gender, even in its most elaborate, abstract or fantastic forms, is always an 'accomplishment', as West and Zimmerman have put it. Gender is something actually done; and done in social life, not something that exists prior to social life.
Four dimensions of gender
When the pioneering British feminist Juliet Mitchell published Woman's Estate in 1971, she argued that women's oppression involves not one, but four structures: production, reproduction, socialization and sexuality.
Why make such distinctions? Many discussions of gender do not. For instance, the feminist lawyer Catharine MacKinnon (1989), developing a theory of the state and the gender dimension of law, treats 'gender hierarchy' as a homogeneous whole. The anthropologist Gayle Rubin (1975), in a very influential model of the 'sex/gender system', treated the whole field as a single system. But when we look closely into these
56    GENDER RELATIONS
theories, it becomes clear that each prioritizes a particular kind of relationship (MacKinnon: domination; Rubin: kinship). If we were to put power relations and kinship together in a more comprehensive picture of gender, we would need at least a two-dimensional model.
There are also practical reasons for acknowledging multiple dimensions in gender relations. We often experience disparities in gender relations, as if part of our lives were working on one gender logic, and another part on a different logic. When this happens in public life, not just in personal affairs, the complexity within the gender system becomes highly visible.
For instance, the modern liberal state defines men and women as citizens, that is, as alike. But the dominant sexual code defines men and women as opposites. Meanwhile customary ideas about the division of labour in family life define women as housewives and carers of children. Accordingly women entering the public domain - trying to exercise their rights as citizens - have an uphill battle to have their authority recognized. They may try to solve this problem by becoming "honorary men1, tougher than the toughest, like Margaret Thatcher in Britain and Madeleine Albright in the United States. But most women in politics, like Hillary Clinton in the United States and Cheryl Kernot in Australia, have to struggle for credibility.
The political scientist Carole Pateman (1988) dramatized this disparity in her argument that the 'social contract* of liberal society was underpinned by a "sexual contract1, the private subordination of women to men. This gave the whole of liberal democracy the character of a 'fraternal social contract', an agreement among men. The statistics or political participation given in chapter 1 suggest this is still broadly true, around the world.
At times such disparities become so striking that they stimulate a strong cultural response. The sixteenth-century cult of 'Gloriana' is a fascinating example. Elizabeth Tudor became queen of England under rules of inheritance that preferred men but admitted women as residual heirs. She became a skilful politician, riding out rebellion and financial crisis, successfully managing deep religious tensions and the changing social forces represented in parliament - which broke out into revolution a few decades after her death. She was, in the language of the day, a strong monarch. But her authority was in flagrant contradiction with the ideas of a patriarchal society. To maintain legitimacy she and her supporters had to construct a new sexual identity (stalling endlessly on marriage negotiations, and celebrating the 'Virgin Queen*) and a mixed-gender position as leader of a new cult of nationality. In a famous speech
GENDER RELATIONS    57
she gave at the time of the invasion threat from the Spanish Armada, izabeth pur it this way:
1 know I have the body of a weak and feeble woman, bur 1 have the heart and stomach of a king, and of a king of England too, and think foul scorn that |the duke of] Parma or |thc king of] Spain, or any prince of Europe should dare to invade the borders of my realm; to which, rather than any dishonour shall grow by me, I myself will take up arms, I myself will be your general, judge, and rcwarder of every one of your ■ virtues in the field. (Ncale 1960: 302)
\n extraordinary literary cult was fostered, which by late in her reign was almost defining her as a supernatural being. This genre includes Spenser's The taerie Queene, one of the great English epic poems.
There is, then, a strong case for seeing gender relations as internally complex, as involving multiple structures. If that general case is accepted, how are we ro identify and map the structures involved?
Mitchell's original model mainly distinguished types of practice -work, child-rearing and sexuality - but also mixed these with social functions, such as 'reproduction' and 'socialization'. Aparr From some logical inconsistency, this approach has limitations. It is clear, for instance, that rather different gender relations can exist in the same kind of practice. Consider, for instance, the range of social relations involved in "sexual-it)'', as shown in Dowsett's study of Harriet Brown.
An alternative approach is to identify different social dynamics, or processes of change, and try to work back to their internal logic. This was the approach taken by classical socialism, which identified the dynamic of class struggle and worked back to a structural analysis of capitalism. It is the approach of single •structure theories of patriarchy, which starts with the political dynamic of feminism and describes the system of power and oppression that feminism confronts.
A sophisticated development of this idea was offered by Sylvia Wal by in Theorizing Patriarchy (1990J, which distinguishes six structures in contemporary patriarchy: paid employment, household production, culture, sexuality, violence, and the state. This greatly improves the kind of model seen in MacKinnon's work. Walby's model is still a model of patriarchy, that is to say, institutionalized inequality in gender relations. If we want to include in the picture of gender patterns that are not inherently unequal, we need a different formulation.
The model I suggest is a development from the one that I proposed in Gender and Power (Conncll 1987). Ir distinguishes four dimensions
58    GENDER RELATIONS
of gender, describing four main structures in the modern system of gender relations. Later in the chapter 1 will discuss how these structures change. Here I will outline them and comment on their significance.
Power relations
Power, as a dimension of gender, was central to the Women's Liberation concept of 'patriarchy', and to the social analyses that flowed from it: the idea of men as a dominant 'sex class', the analysis of rape 3S an assertion of men's power over women, and the critique of media images of women as passive, trivial and dumb.
Women's Liberation recognized that patriarchal power was not just a matter of direct control of women by individual men, but was also realized impersonally through the state. A classic example, analysed in a famous article by Catharine MacKinnon (19S3), is court procedure in rape cases. Independent of any personal bias of the judge, the procedures by which rape charges arc tried effectively place the complainant rather than the defendant 'on trial'. The woman's sexual history, marital situation and motives in laying a charge are all under scrutiny.
Many attempts at legal reform have been made since, and have proved that the inbuilt biases in social assumptions and court procedure about sexual assault are very difficult to eliminate. It can still be a damaging experience for a woman to bring charges. A very public example of the difficulty occurred in Sydney in late 2000. A young woman made a complaint to police about an event during a party in the Parliament buildings, involving a sexual approach by a Member of Parliament in his office (he said rhe approach was consensual, she said it was not). The Speaker of the House (a man who belonged to the same party as the MP in question) responded by collecting derogatory evidence from an aide about the woman's behaviour on the night in question. The woman dropped the complaint, to avoid the impact of publicity on her private life. Nevertheless an official inquiry was held in a blaze of publicity, into the possibility of corrupt conduct by the Speaker, his aide and the MP. All were cleared.
Another important case of the institutionalization of power relation* is bureaucracies. Clare Burton, an Australian social scientist who also served in public life as an equal opportunity commissioner, spoke of the 'mobilization of masculine bias' in selection and promotion of staff. By this she meant the impersonal but pervasive tendency, in organizations dominated by men, to favour criteria and procedures that favour men (Burton 1987).

GENDER RELATIONS    59


Power also emerged as a major theme in Gay Liberation writing such as Dennis Altman's Homosexual: Oppression and Liberation (1972). (n this case the focus was on power applied to a specific group of men: criminalization, police harassment, economic discrimination, and violence. Gay Liberation theorists linked the oppression of gay men with the oppression of lesbians and the oppression of women generally. This argument laid the foundation for the analysis of gendered power relations among men, and the distinction of hegemonic from subordinated masculinities (Carrigan, Connell and Lee 1985) which is important in current research on men and masculinities.
Power operating through institutions, power in the form of oppression of one group by another, is an important part of the structure of gender. But there is another approach to power, popularized by the French historian and philosopher Michel Foucault (1977). Foucault was sceptical of the idea that there was a unified, central agency of power in society. Rather, he argued, power is widely dispersed, and operates intimately and diffusely. Especially if operates discursively, through the ways we talk, write and conceptualize. This diffuse but tenacious power operates close up, not at a distance. It impacts directly on people's bodies as 'discipline' as well as on their identities and sense of their place in the world.
This post-structuralist approach appealed to many feminist as well as gay theorists, who saw here a way of understanding the fine texture, as well as the strength, of gendered power. Power is present intimately. The discourse of fashion and beauty, for instance, positions women as consumers, subjects them to humiliating tests of acceptability, enforces arbitrary rules and is responsible for much unhappiness, ill health, and even some deaths (among young women whose dieting goes out of control). Yet there is no Patriarchy Central compelling women to do all this. As the 'lip gloss' in Barrie Thome's ethnography illustrates, girls and young women enter the world of fashion and beauty because they want to, because it delivers pleasures, and because the regulation and discipline are bound up with the identity they arc seeking.
Both these approaches to power contribute to our understanding of gender relations: they are not exclusive. There is both organized, institutional power and diffuse, discursive power. And both approaches raise the crucial question of resistance.
To give a full account of power relations requires an account of the way power is contested, and countervailing power is mobilized. Total domination is extremely rare; even fascist dictatorships could not accomplish that. Gendered power is no more total than other kinds.
60    GENDER RELATIONS
Oppressive laws are met by campaigns for reform - such as the most famous of all feminist campaigns, the 'suffragette' struggle for the vote. Domestic patriarchy may be weakened, or manoeuvred around, by the inhabitants of the 'red chamber' (as the classic Chinese novel put it), the women of the household.
Discursive power can also be contested or transformed. The remarkable work of the Australian educator Bronwyn Davies shows that challenges to patriarchy need not involve head-on confrontation. In Shards of Glass (1993) Davies shows how educators in the classroom can help children and youth gain control of gender discourses. Young people can learn how they are discursively positioned and regulated, and can learn to shift between, or manoeuvre among, identities.
The conditions for resistance change in history. The modern liberal state, which emerged in Europe and North America in the eighteenth and nineteenth centuries, creates possibilities for mass politics which did not exist before. Monarchical states and household patriarchies did not depend on notions of citizenship; the liberal state does. In that sense, the development of patriarchal institutions themselves created the conditions for the emergence of modern feminism.
Production relations
The 'sexual division of labour' was the first structure of gender to be recognized in social science, and remains the centre of most discussions of gender in anthropology and economics. In many societies, and in many situations, certain tasks are performed by men and others are performed by women. So, in the Aboriginal communities of the Australian central desert, hunting wallabies and kangaroos was undertaken by men, collecting root vegetables and seeds was mainly undertaken by women. In contemporary North America teaching young children is mainly done by women; in South Africa underground mining is entirely done by men.
Such divisions of labour are common throughout history and across cultures. But while gender divisions of labour are extremely common, there is not exactly the same division in different cultures or at different points of history. The same task may be 'women's work' in one context, and 'men's work' in another. Agricultural labour - digging and planting - is an important example.
A striking modern case is secretarial work. Being a clerk was originally a man's job - as described in Herman Melville's dark short story 'Bartleby the Scrivener' (1853). With the advent of the typewriter and the growing scale of office work, clerical work increasingly involved

GENDER RELATIONS    61
Somen; in fact it became archetypical 'women's work*, as Rosemary Pringle shows in Secretaries Talk (1989). But with the advent of the computer and word processing, 'the secretary' is disappearing as an occupational category. Clerical work is again, increasingly, being done by men.
In modern Western society, gender divisions between jobs are not the whole of the gender division of labour. There is a larger division between '■work' - the realm of paid labour and production for markets - and 'home'. The whole economic sphere is culturally defined as mens world {regardless of the presence of women in it), while domestic life is defined as women's world (regardless of the presence of men in it).
The Norwegian sociologist 0ystein Holrcr (1995, 1997) argues that this division is the structural basis of the modern Western gender order. fit is what makes this system different from the gender orders of non-Western, non-capitalist societies. His point is not only that our notions of 'masculinity' and 'femininity' are closely connected with this division. Just as important, the social relations that govern work in these two spheres are different. In the economy, work is done for pay, labour is bought and sold, and the products of labour are placed on a market where profit prevails. In the home, work is done for love (or from mutual obligation), the products of labour are a gift, the logic of gift-exchange prevails. From these structural differences, Holter argues, flow characteristically different experiences for men and women - and our ideas about the different natures of men and women.
This is not exactly a distinction between 'production' and 'consumption', though that has been suggested by others as the economic core of the gender system. Domestic 'consumption' requires work, just as much as factory-based 'production'. Housewives do not spend their time lolling on couches and scoffing chocolates. Housework and childcare are hard work and the hours have remained long, despite the advent of 'labour-saving* machines like vacuum cleaners and microwave ovens. But housework and job-work are done in different social relations, as Holter correctly observes, and they consequently have very different cultural meanings.
The division of labour itself is only part of a larger process. In a modern economy the shared work of women and men is embodied in every major product, and every major service - therefore in the process of economic growth. Yet women and men are differently located in that process, and as the statistics of income in chapter 1 show, women and jmen get different benefits from it.
What can be seen here is a gendered accumulation process. Maria Mies (1986), the German socialist feminist who has formulated this issue
62    GENDER RELATIONS
most clearly, suggests that the global economy has developed through a dual process of colonization and 'housewifization*. Women in the colonized world, formerly lull participants in local non-capitalist economics, have been increasingly pressed into the 'housewife' pattern of social isolation and dependence on a male breadwinner.
Accumulation in modern economies is organized through large corporations and global markets. The gender regimes of these institutions make it possible for them to apply the products of men's and women's joint work in gendered ways. The way firms distribute corporate income - through wage structures, benefits packages, etc. - tends to favour men, especially middle-class men. The products that corporations place on the market have gender effects and gendered uses, from cosmetics to armaments.
The gendered accumulation process has many effects beyond the "economy' narrowly defined. For instance, where there is a gender division of labour in occupations - such as men being the majority in engineering and mechanical trades, women in arts-based and human service jobs - there will be a division in the education systems which prepare people tor this work. It is not surprising to find that enrolments in school courses in 'engineering studies' and "computer sciences' are overwhelmingly boys, while enrolments in 'fine arts' and 'hospitality' are mainly girls.
Emotional relations
The importance of emotional attachment in human life was made clear a hundred years ago by Sigmund Freud (1900}. Borrowing ideas from neurology but mainly learning from his own case studies, Freud showed how charges of emotion - both positive and negative - were attached, in the unconscious mind, to images of other people. His famous analysis of the "oedipus complex', the centrepiece of his theory of personalit)' development, showed how important the patterning of these attachments, or cathexes, might be. (For clear and careful definitions of these terms sec The Language of Psycho-Analyst*; Laplanche and Poutali-1973.)
In fact Freud was speaking not only about the individual mind, but also about the pattern of relationships inside an important social institution, the bourgeois family. He thus opened up for investigation the structure of emotional relations, attachments or commitments. This is an important dimension of gender, often interwoven with power and the
GENDER RELATIONS    63
division of labour (e.g. in the figures of the father and the mother), but also following its own logic.
Emotional commitments may be positive or negative, favourable or hostile towards the object. For instance, prejudice against women (misogyny), or against homosexuals (homophobia), is a definite emotional relationship. Emotional commitments may also be, as Freud emphasized, both loving and hostile at once. Ambivalence, as this state is called, is common in reality though it tends to be forgotten in gender myths and stereotypes.
A major arena of emotional attachment is sexuality. Anthropological and historical studies have made it clear that sexual relations involve culturally formed bodily relationships, not a simple biological reflex (Caplan 1987). They have a definable social structure. The main axis on which contemporary Western sexuality is organized is gender: the division between cross-gender (heterosexual) and same-gender (homosexual) relations. This distinction is so important that we treat it as defining different kinds of people ("homosexuals', 'heterosexuals'), and certain biologists go looking for a 'homosexual gene' to explain the difference. (However, no one has gone looking for the 'heterosexual gene*.)
But cross-cultural research shows that many societies do not make this distinction. They have both same-gender and cross-gender sexual encounters, but they do not arrange them the way we do, nor think they define different types of people. For instance, the 'Sambia', a community in Papua New Guinea described in a well-known ethnography by Gilbert Herdt, Guardians of the flutes (1981), treat same-gender sexualit)* as a ritual practice that all men are involved in at a particular stage of life. From a Western point of view, all Sambia men are homosexuals at one age, and all switch over to become heterosexuals at another. That is absurd, of course. From a Sambia point of view, they are simply following the normal development of maseulinity.
In contemporary Western societ)-, households are expected to be formed on the basis of romantic love, that is, a strong individual attachment between two partners. This ideal is promoted in mass media and popular fiction, and its importance is confirmed by research with groups who might be thought sceptical of it. They include the men in Gary Dowsett's study (Harriet Brown was not alone wanting to live in a loving couple); and the American college students in an ethnography by Dorothy Holland and Margaret Eisenhart, Educated in Romance (1990).
Where this pattern holds, sexual attachment is now the main basis of household formation. The cultural dominance of the West has meant a
64    GENDER RELATIONS
shift, in many post-colonial situations, from the choice oř a marriage partner by one's parents to the choice oř a partner by personal attraction - romantic love. The resulting tensions arc explored in the recent comedy East is East, a film about an Anglo-Pakistani family struggling about arranged marriages, Muslim tradition, and British working-class realities. Curmudgeonly conservatives warned that the shift from marriages arranged by wise parents to marriages contracted by impetuous youth risked the collapse of a household when the sexual interest died. The historically startling level of divorce in the United States - where according to very recent sample survey data, 43 per cent or first marriages end in separation or divorce within fifteen years - shows thev were right.
Emotional relations are also found in the workplace (and not just in the form of office sex). Rosemary Pringle's study, already mentioned, shows how emotional relations with bosses help to construct the very job of 'secretary'. Arlie Hochschild's classic The Managed Heart (1983) analyses emotional labour in the modern economy. There are many jobs where producing a particular emotional relationship with a customer is central to the work being done. These are, typically, gender-typed jobs. Hochschild's main examples are airline hostesses, a job where workers are trained to produce sympathy and induce relaxation; and telephone debt collectors, a job where workers must display aggression and induce fear. Hochschild argues that this kind of labour is becoming more common with the expansion of service industries. If so, alienated relations based on commercialized feelings and gender stereotypes may be increasingly important in modern life.
Hostile emotional relationships are not only symbolic, like the ones enacted by Hochschild's debt collectors. They may involve all too real practices of oppression. Stephen Tomsen's (1998) study or homophobic killings, for instance, shows two major patterns of conduct. One is gang attacks in public places by young men who go looking for gender deviants to punish, a process that depends on mutual encouragement in the group. The other is killings by individuals in private. Some or these involve a violent response to a sexual approach (and perhaps to the killers' own desires) which they think calls their masculinity into question. Both patterns may result in killings of extreme brutality.
Emotional relations go beyond the ŕace-ro-ŕace. Nationalism, as Joane Nagel (1998) points out, constantly uses gender imagery in constructing national solidarities. We are all familiar with the "family of the nation, the "father or his country', the heroic soldier dying to protect his womenfolk, 'Mother Russia', the nation as goddess. It is no accident that, as
GENDER RELATIONS    65
Irina Novikova (chapter 2 above) and Svetlana Slapšak (2000) show, new nationalisms in the former communist countries of eastern Europe are reasserting highly traditional gender images.
Symbolic relations
All social practice involves interpreting the world. As post-structuralists observe, nothing human is 'outside' discourse. Society is unavoidably a world of meanings. At the same time, meanings bear the traces of the social processes by which they were made. This is the fundamental point made by the sociology of knowledge. Cultural systems bear particular social interests, and grow out of historically specific ways of lire.
This point applies to gender meanings. Whenever we speak of 'a woman' or 'a man', we call into play a tremendous system of understandings, implications, overtones and allusions that have accumulated through our cultural history. The 'meanings'' of these words are enormously greater than the biological categories of male and female. When the Papua New Guinea highland community studied by Marilyn Strath-ern (1978) say 'our clan is a clan of men', they do not mean that the clan entirely consists of males. When an American football coach yells at his losing team that they are 'a bunch of women', he does not mean they can now get pregnant. But both are saying something meaningful, and in their contexts, important.
The best-known model of the structure of symbolism in gender derives from the French psychoanalyst Jacques Lacan. I.acan's analysis of the phallus as master-symbol gave rise to an interpretation of language as 'phallocentric\ a system in which the place of authority, the privileged subjectivity, is always that of the masculine. The potentially infinite play of meaning in language is fixed by the phallic point or reference; culture itself embodies the 'law of the father'. If that is so, the only way to contest patriarchal meanings is to escape known forms of language. Hence feminist thinkers in the 1970s, such as Kaviere Gauthier, developed an interest in women's w-riting as an oppositional practice that had to subvert the laws of culture. (For translations of Gauthier, and other French feminists on this question, see Marks and de Courtivron 1981.)
Chris Weedon (1987) wonders how feminist theory could have adopted so deterministic a psychology, which gives no room for opposition, only for escape. There are certainly other schools oř psychoanalysis which offer more open-ended accounts of gender and suggest more possibilities for action. Nevertheless the dichotomous gender structuring of culture is important, and the Lacanian approach gives us some inkling
66    GENDER RELATIONS
or why patriarchal gender arrangements are so difficult to abolish. To do so involves uprooting, not just a few intolerant attitudes, but a whole system of communication and meaning. Queen Elizabeth, addressing her men at Tilbury, acknowledged 'the body of a weak and feeble woman', but claimed 'the heart and stomach of a king*. She could not have reversed her symbolism, and claimed 'the heart and stomach of a woman', if she were to motivate her troops to fight.
Though language - speech and writing - is the most analysed site of symbolic gender relations, it is not the only one. Gender symbolism also operates in dress, makeup, gesture, in photography and film, and in more impersonal forms of culture such as the built environment.
Elizabeth Wilson's (1987) elegant study of fashion. Adorned in Dreams, shows that women's and men's styles of dress not only symbolize gender difference, but are also a site of struggle over what women and men are allowed to do. The famous 'bloomers' of nineteenth-cen tury dress reform were connected with the struggle to expand the rights of women. For a short while bloomers were adopted by suffrage activists. They were jeered at by conservatives because they symbolized emancipated women (not that they changed women's activities in practice). Similarly in the 1960s the new fashion styles were connected with young women's demand for sexual freedom, and were duly denounced as licentious. Jean Shrimpton, a visiting British fashion model, created a media scandal in Australia by going to the Melbourne Cup races in a mini-dress and - an unforgivable offence - without gloves!
Rosa Linda Fregoso's The Bronze Screen (1993) illustrates the play of gender relations in film - in this case, films produced by Chicana/ Chicano film-makers, about the community of Mexican affiliation in the south-western USA. Chicano (men) film-makers, Fregoso observes, have nor demeaned their women characters, but they have not given them an active role in discourse. Only with the advent of woman film-makers did films start to explore generational difference, language, religion and relationships from women's standpoints, and show some of the tensions and ambiguities in women's position and responses. Architectural design also reflects assumptions about gender dichotomy and gendered spaces, and grows out of the designers' gendered experience. Annmarie Adams and Peta Tancred in Designing Women (2000), a study of gender and architecture in Canada, found that the imagery in professional journals persistently associated women with interiors, especially domestic interiors, but presented architects as 'powerful, virile, and masculine'. However, this pattern of marginaiization changed as women arrived in the profession, and became influential in establishing the modernist style.
GENDER RELATIONS    67
Symbolic relations in gender include the rules for 'gender attribution' studied by ethnomethodologists. Here we move below the level at which gender categories ordinarily appear, to consider how a person (or action) gets assigned to a gender category. These rules are normally taken for granted in everyday life. But they are painstakingly studied by cross-dressers and transsexuals hoping to 'pass', which requires one to produce an effect of naturalness by deliberate action. Accordingly, transsexuals have appeared to psychiatrists and ethnomethodologists as a kind of natural experiment exposing the cultural underpinnings of the gender system (Kessler and McKenna 1978).
But things get complicated when the transsexuals read the psychiatrists1 and ethnomethodologists' books - as some now do. As a warning against over-simplified views of gender, transsexualism itself has now become a gender category, and to a certain extent a sexual subculture. You can buy the international Tranny Guide (Vicky Lee 1999) to learn how to do it (with serious advice on body care, how to present at the job, etc.) and how to get in touch with the cross-dressing scene around the world. You can even check this scene out on the Internet (try www.wayout-publishing.com). In a recent book Viviane Namaste (2000) urges attention to the real-Hfe situations and experiences of transsexual and transgendered people - which tend to be 'erased' by queer theory, social science and medicine alike.
The tranny scene is determinedly upbeat, but there is a dark side to violating the cultural categories. Transgender people often face ostracism, loss of jobs, and family hostility, as well as major difficulties in sexual relations. Some have to support themselves by sex work such as stripping and prostitution. As Harriet found, there is a certain clientele of 'straight' men who are excited by transsexuals. But this does not mean they respect them. Roberta Perkins's pioneering book presenting the voices of Transsexuals in Sydney includes Naomi, a stripper who remarked:
I think men have a definite dislike for women in general, that's why women are raped and bashed, and strippers are up there to provide an outlet for this dislike by the yelling of profanities at them. Transsexuals are lower down than women according to men, and look how many men sexually abuse transsexuals. (1983: 73)
Naomi's point about abusive men relates not only to the cultural relations of gender but also to power relations, in the form of sexual violence. She also implies something about production relations -
68    GENDER RELATIONS
straight men have the economic resources to be the clients of these services. And of course her remarks relate to emotional relations, in terms of sexual desire and hatred. So all four structures of gender are present in this one situation.
This is usual. In distinguishing four structures of gender relations, I do not mean to suggest they operate in separate compartments or life. They are constantly intermingled and interacting in practice. I distinguish structures analytically because tracing out their logic helps in understanding an extremely complex reality. This does not imply that reality itself comes in boxes. Naomi, for one, knows that.
Gender as history
Ideologies of 'natural difference' have drawn much of their force from the traditional belief that gender never changes. Adam delved and Eve span. Men must work and Women must weep, Boys will be Boys. Serious analysis begins with the recognition that exactly the opposite is true: everything about gender is historical.
What does 'historical' mean? In the whole story of life on earth, human history represents a new process of change. Some time in the last half-million years, social dynamics replaced organic evolution as the central mechanism of change in our biosphere. Sociobiologists and evolutionary psychologists are nor absurd in asking how human society is related to the evolution of the natural world. The same question was the centre of nineteenth-century sociology, when books with titles like Social Evolution {Kidd 1898) were best-sellers. But these authors, over-anxious to prove the continuity of evolution, miss the deep change in the process of change. A radically new dynamic was introduced when the collective capacities of humans could he mobilized by social relations. This is why human society, and not organic evolution, can produce cloth, pottery, ziggurats, irrigated rice-fields, rock music and gravity-wave detectors-
Some biological features of human ancestors were certainly preconditions of this change. The open architecture (to borrow a computer term) of the human hand, brain, and speech apparatus makes an immense range of applications possible. The human body, equipped with arm and hand, cannot scratch as sharply as a cat, dig as well as a wombat, swim as fast as a seal, manipulate as delicately as a monkey, or crush as powerfully as a bear. But it can do all those things moderately well; and it can make tools to do them all very well. This multiplies the capacities of any one person. Yet rhe greatest human invention
GENDER RELATIONS    69
of all is other human beings. We not only create social relations, we teach new generations to operate in, and build on, the social relations already existing. With cumulating effects over time, social relations multiply the capacities of any individual body on the astonishing scale we see all around us. So great a multiplication, ironically, that it now threatens human life by nuclear war or environmental disaster.
The horizon in time where history appears is also the horizon of gender. In the broadest perspective, gender represents the transformation of the system of sexual reproduction by social action. Human collective capacities, organized through social relations, lead to entirely new possibilities. Some are for creativity and pleasure. For instance, sexuality is constructed in culture, and the world of love and eroticism becomes possible. Some are for subjection and exploitation. Patriarchy becomes possible, along with family property, bride-price, convents and prostitution.
Above this horizon is the history of gender: the course or events that has produced the actual gender orders we live in. The history of gender includes the history of practices, and transformations of the body in practice. It includes the production and transformation of the categories of gender. We know these are not fixed; new categories ('the homosexual', 'the housewife') appear and others decline. The history of gender includes the gender regimes of institutions and the gender orders of societies.
This is, in principle, a world history. That idea was first formulated in the nineteenth century, in debates about 'origins' which invented the idea of a primitive matriarchy. The search for origins was resumed in the 1970s in the debate about patriarchy unleashed by Women's Liberation. The search is futile. As the French feminist Christine Delphy (1984) showed in a brilliant critique, origins stories are nor history but are a form of myth-making. They create myths in which later social arrangements are explained (and often justified) by a mechanism 'discovered' at the point of origin.
A real history of gender begins with the recognition that the later course of events is not contained in any founding moment.
Rather, an open-ended social process is involved, which must be studied in all its complexity by patient examination of the historical records: the archaeological deposits, the written sources, the oral traditions. Local history of this kind has flourished for several decades, being one of the main branches of Women's Studies. It has produced superb work, such as Family Fortunes by Leonore Davidoff and Catherine Hall (1987), a social history of gender in the English middle class of the industrial revolution. A world history of gender has taken longer to
70    GENDER RELATIONS
develop, but now seems to be emerging from two -taiTing points. Qni is the archaeological reconstruction of gender relations in prehistory and ancient urban cultures (Gero and Conkey 1991). The other is the study of gender relations in modern imperialism, the global process which has at last reversed the proliferation of cultures and begun to create a single world society.
Recognizing the deeply historical character of gender has an important intellectual and political consequence. If a structure can come into existence, it can also go out of existence. The history of gender may have an end.
There are several ways in which gender relations might cease to be important conditions of social life. They might be weakened by an internal uncoupling, so that gender patterns in one domain of practice cease to reinforce those in another. Alternatively, gender relations might be overwhelmed by some other historical dynamic. This was expected by marxists like Alexandra Kollontai, who thought that proletarianization and socialist revolution would end the oppression of women. In our day, the total triumph of the market might do the job.
Finally, gender relations might be extinguished by a deliberate de-gendering, in which the reach of gender structure is reduced to zero. A de-gendering logic is found in some current feminist strategies, such as equal opportunity and anti-discrimination policies. Not all feminists agree with the dc-gendering approach, and not all theorists assume a complete de-gendering of society is possible. Even if it is impractical, however, a gender-free society remains an important conceptual benchmark for thinking about change.
The process of change
Most discussions of why gender arrangements change have focused on external pressures on the gender order: changing technology, urban life, the demands of capitalism, mass communications, secularism, modernization or Westernization.
It is true that these social forces can produce change in gender relations. But gender relations also have internal tendencies towards change. Further, some of the 'external1 forces are gendered from the start (for instance, the capitalist economic system}. In this discussion I will focus on the dynamics of change that arise within gender relations.
Post-structuralist theory has recognized internal tendencies towards change by arguing that gender categories are inherently unstable. For
GENOER RELATIONS    71
instance, the uncertain and contested character of the category 'women' is a theme of Judith Butler's well-known book Gender Trouble {1990). Gender identities are produced discursively. But meanings in discourse are not fixed. Indeed they are inherently unstable, incapable of being fixed.
Further, there is no fixed connection between discursive identifies and the bodies to which those identities refer. The signifier is able to float free, in a play of meanings and pleasures. That is sometimes thought to be a general feature of 'postmodern' life, and it certainly suggests that gender identities can be played with, taken up and abandoned, unpacked and recombined. This has been a theme in the 'queer theory' of the 1990s and in other applications of post-structuralist and postmodern ideas.
There are several difficulties with a concept of generalized instability. It can be made true by definition, but in that case is not interesting. If it is open to empirical checking, then it is difficult to avoid the fact that in some historical situations gender identities and relations change slowly, in other situations they change explosively. A good example is Irina Novikova's account of the Soviet and post-Soviet gender orders (chapter 2 above). Nor does a concept of generalized instability give any grip on why some people would want to change gender arrangements, while others would resist changes. This is a question of central importance for the politics of gender. It raises the issue of the differing material interests that different groups have in an unequal society - a question hard to formulate in a purely discursive theory.
The post-structuralist approach is helpful in emphasizing that identities are always historically constructed and in principle open to change; but we need a more specific theory to understand how change occurs. The key is to recognize that structures develop crisis tendencies, that is, internal contradictions or tendencies that undermine current patterns and force change in the structure itself.
This approach to change - which draws from German critical theory, especially Jürgen Habermas (1976) - allows us to distinguish periods when pressures for change are well controlled, or are gradually building, from periods when crisis tendencies erupt into actual crisis and force rapid change. It also allows us to identify interests that can be mobilized for and against change, by examining where different groups are located in the structure under pressure, and how they have come into being within that structure. Crisis tendencies can be identified in each of the four structures of gender relations identified earlier in this chapter.
Power relations show the most spectacular recent change. A global movement for the emancipation of women has appeared, challenging
72    GENDER RELATIONS
men's control of institutions as well as men's power in the intimate spheres of sexuality and the family.
The main crisis tendency here has often heen noticed. There is an underlying contradiction between the subordination of women to men in patriarchal homes and workplaces, and the abstract equality between women and men which is presupposed by citizenship and markets. Over the last two centuries this contradiction has sharpened, as the liberal state has developed, and market relations have come to dominate the economy.
Women are the main group subordinated in patriarchal power structures and so have a structural interest in change. Feminist movements, mobilizing women, have been energized by this contradiction and have used it to break down inequality. They have persistently claimed "rights' in the public sphere and used those rights to challenge oppression in private spheres. The campaign against domestic violence is a notable example (see Rebecca Dobash and Russell Dobash 1992). This campaign, claiming human rights to safety and freedom from fear, has used the patriarchal state itself to enforce these rights when violated by husbands and de factos.
Production relations have also been the site of massive change. Through the second half of the twentieth century there was a worldwide incorporation of women's labour into the marker economy. In the industrialized countries this took the form of a huge growth in married women's 'workforce participation' rates, especially in the service sector. In the developing world it took the form of an even more massive move into cities and into market-based agriculture. By the end of the century adult women's workforce participation was over 90 per cent of the men's rate in Cambodia, Ghana, Tanzania, Vietnam, Malawi, Rwanda, Mozambique, Burundi, Guinea, Benin and Sweden - and not far behind that in other parts of Scandinavia, eastern Europe, the former Soviet Union, China, central and west Africa.
There is an underlying contradiction between the equal contribution to social labour by women and men (bearing in mind unpaid as well as paid work} and the gendered appropriation of the products of social labour. The gendered appropriation is seen in the unequal incomes of women and men as groups, the better conditions and career prospects men generally have, and the patriarchal inheritance of wealth. {It is still the general rule in big business that sons may take over the company but daughters hardly ever do.)
Women have a general interest in changing this. One consequence is that women workers make up a growing proportion of union members,
GENDER RELATIONS    73
and union militants. The story of women's struggle to establish their presence in the union movement is told in Suzanne Franzway's Sexual Politics and Greedy Institutions (forthcoming). It is significant that the latest two presidents of the Australian Council of Trade Unions (the unions' peak organization in that country) have been women. But the turbulence of the gendered accumulation process, and its interplay with class and colonial relations, create complex economic situations. An important consequence is that some women - and often the most influential - have an interest in resisting economic reform, because this would disturb the corporate system from which they benefit.
Emotional relations have also seen important recent changes in the industrialized countries. Though lesbians and gay men are still subject to homophobic abuse and violence, homosexual sexuality has to a certain extent achieved legitimacy as an alternative within the heterosexual order. Visible gay and lesbian communities now exist in many cities, antidiscrimination and anti-defamation laws exist in a number of countries, and there is a limited representation of gay and lesbian communities in some political systems and in some areas of policymaking (e.g. in relation to the HIV/AIDS epidemic). As Dennis Altman (1982) pointed out, gay and lesbian communities have achieved a position in some ways resembling that of ethnic minorities.
This is a partial resolution of a long-standing contradiction. The patriarchal gender order prohibits some forms of emotional attachment and pleasure which its own gender arrangements (e.g. homo-social institutions, the oedipal family) produce. A related logic operates within heterosexual relations. The constantly growing incitement to sexual activity (what conservatives call 'permissiveness') contradicts the continuing definition of women as sexually passive, as the objects of men's desire and seduction. The result has been an uneven pattern of change, seen in surveys of sexual behaviour, where women's sexual repertoire ha* been growing but the 'double standard' for women and men remains.
Symbolic relations are the home ground of generalized-instability arguments, which centre on the discursive construction of identifies. If might therefore seem difficult to define crisis tendencies here. But what has made it possible to recognize unstable identities is a tendency towards crisis in the legitimation of patriarchy.
Patriarchy has long been legitimated by belief systems which picture gender as a timeless, unchanging division - whether laid down bv God or fixed by the genes - which makes 'woman's place' the right place for ever and ever. Over the last century and a half, social and intellectual movements have chipped away at these assumptions: from the woman
74    GENDER RELATIONS
suffrage movement and psychoanalysis ro Gay Liberation and post-structuralism. Natural-difference ideas remain very influential in popular culture. But over time their capacity to form the unquestioned common sense of society has been undermined. In an era when 'sex changes' arc reported in the media, governments have Equal Opportunity targets, and global conferences on gender reform occur, it is difficult to rake for granted a timeless male/female opposition.
A vast change in presuppositions has thus occurred in the cultural life of the industrial (and many indusrrializing) countries. A hundred years ago those who claimed equality for women, or rights for homosexuals, had to justify the claim against presuppositions to the contrary. Now those who deny equality or rights have to justify their denial against a presumption for equality and a presumption that change can occur. The boot is on the other foot.
This discussion has focused on crisis tendencies on the large scale. It is also possible for crisis tendencies to emerge on the small scale - in personal life and in intimate relationships. Crisis tendencies arise when personal practice is structured around commitments which are both urgent and contradictory. The classic case is the incompatible desires and fears of the young child in the 'oedipaľ crisis, which Freud thought the basis of all later neuroses. We do not have to accept Freud's theory to agree that contradictions often arise in personal life, and drive change in a person's trajectory through life.
These changes may be individual and produce nothing but eccentricity. But they may also move in parallel with other lives, and this can result in sustainable change. The Women's Liberation movement of the 1960s and 1970s was not just a public event; it was fuelled by contradictions in the personal lives of women, especially in their relations with men. Narratives by women from this movement (e.g. those collected by Barbara Laslett and Barrie Thorne, 1997) show how the similarity of these experiences was recognized and became a basis of solidarity. Their actions, in turn, stimulated changes in the trajectories or certain men. One consequence was the 'fair families' of the 1980s and 1990s, whose Story has recently been explored by Barbara Risman (1998).
Since the involvement of the body in gender relations is a social process, crisis tendencies may also arise at the level of the body. Freud s classic analysis of 'hysteria' recognized precisely that: a bodily effect (e.g. a cough, or a paralysed arm) whose cause was a psychological conflict. The bodily effects may be much rougher than Dr Freud's genteel patients were used to. Asserting masculinity, in a poor neighbourhood or a
GENDER RELATIONS    75
factory or on the road, may result in violence, industrial injury, or road death. I noted in chapter 3 how factory work consumes the workers' bodies, and how exemplary masculinity in professional sport produces over-use, injury, and long-term bodily damage.
Crisis tendencies may even affect bodily sensations. As Lynne Segal observes in Straight Sex (1994), there have been many difficulties in heterosexual relations connected with the new feminism. They are not necessarily produced by feminism - arguably, by the same crisis tendencies that gave rise to feminism. Similarly, the violations of gender boundaries in transsexuality do not just occur in people's heads. They often involve bodily sensations such as hallucinations of a body of the other sex, or a sensation of being trapped within the wrong body - sec, for instance, Katherine Cummings' account of her transsexual experience in Katherine's Diary (1992).
Thus crisis tendencies in gender emerge on the large scale and on the small. All four structures contain crisis tendencies; but they are not the same tendencies, and they do not necessarily develop at the same pace or mature at the same time. There is, thus, complexity and unevenness in the process of historical change. It is not surprising that gender orders are far from homogeneous, and that gender politics are complicated and turbulent.

-