Jump to ContentJump to Main Navigation
Women, Culture, and DevelopmentA Study of Human Capabilities$

Martha C. Nussbaum and Jonathan Glover

Print publication date: 1995

Print ISBN-13: 9780198289647

Published to Oxford Scholarship Online: November 2003

DOI: 10.1093/0198289642.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (oxford.universitypressscholarship.com). (c) Copyright Oxford University Press, 2021. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use. date: 16 June 2021

Inequalities Between the Sexes in Different Cultural Contexts

Inequalities Between the Sexes in Different Cultural Contexts

(p.274) Inequalities Between the Sexes in Different Cultural Contexts
Women, Culture, and Development

Susan Moller Okin

Oxford University Press

Abstract and Keywords

Okin argues that what is unjust in the situation of women in developing countries is not different in kind from the injustices that unequal family structures produce in developed countries. In both cases, the family, which claims to be a source of love and a school for ethical virtue, often manifests and perpetuates injustice. Contending against feminists opposed to generalizing about the context of women, Okin maintains that generalizing with proper caution and sensitivity can be a valuable and crucial part of good feminist argument, bringing to light what is salient and pervasive in unjust laws and practices.

Keywords:   family, gender roles, justice, oppression

Theories of justice, confronted with the challenges of feminism, post‐modernism, and multiculturalism, are undergoing something of a crisis. How can they be universal, principled, founded on good reasons that all can accept, and yet take account of the many differences among persons and social groups? Feminists have been in the forefront of pointing out that large numbers of persons have typically been excluded from consideration in purportedly universalist theories. And some feminists have then gone on to point out that many feminist theories, while taking account of sexist bias or omission, have neglected racist, heterosexist, class, religious, and other biases. Yet, while to some extent acknowledging this neglect, some of us discern problems with going in the direction of formulating a theory of justice entirely by listening to every individual's or group's concrete point of view and expression of its needs. Is it possible, by taking this route, to come up with any principles at all? Is it a reliable route, given the possibility of ‘false consciousness’? Doesn't stressing differences, especially cultural differences, lead to a slide towards relativism? The problem that is being grappled with is an important one. There can no longer be any doubt that many voices were not heard, while most theories of justice were being shaped. But how can all the different voices express themselves, and be heard, and still yield a coherent and workable theory of justice? This question is one to which I shall (eventually) return.

It is a daunting task for a feminist theorist accustomed to writing about justice between the sexes in the context of Western industrialized societies to venture into the subject area of women in different cultural contexts in far poorer countries. The difficulties I faced in writing this paper stemmed partly from an initially inadequate knowledge, which I have done, and am continuing to do, what I can to rectify. But they were also brought about by the current prevalence among feminists of charges of ‘essentialism’ or, as Ruth Anna Putnam names it in this volume, ‘substitutionalism’. It is ‘essentialist’, some say, to talk about women, the problems of women, and especially the problems of women ‘as such’.1 White, middle‐ and upper‐class feminists, it is alleged, have been insensitive not only to the problems (p.275) of women of other races, cultures, and religions, but even to those of women of other classes than their own. ‘Gender’, therefore, is a problematic category, unless qualified by race, class, ethnicity, religion, and so on (see Childers and hooks, 1990; Harris, 1990; Minow and Spelman, 1990; Spelman, 1988). Those who are often referred to as essentialists, and whom Ruth Anna Putman, adopting the term from Benhabib (1987) but adapting it to her own purpose, calls ‘substitutionalist’ feminists, allegedly commit the mistake of subsuming all women under the category to which they themselves belong. They ‘substitute the experience of white middle and upper class women (mainly professional women) for that of all women’ (Putnam, this volume, p. 311).

If this is so—if, when writing about women, privileged women are really talking only about themselves and their own situations, there is little chance that feminists like myself will have anything of note to say about the situations of poor women in poor countries with cultures that may differ considerably from our own. However, I shall argue that the anti‐essentialist critique is overblown, overvalued, and largely invalid; and I shall do this in part by drawing on the knowledge I have been acquiring about women whose life circumstances are in many respects different from those of the majority of western women. Thus, in this paper, I shall make a qualified defence of essentialism. Before doing so, however, I wish to acknowledge that at least some of the allegations of anti‐essentialists are valid, when applied to some feminist theories. Feminists with such pedigrees as Harriet Taylor, Charlotte Perkins Gilman, Virginia Woolf, Simone de Beauvoir and Betty Friedan all seem to have assumed, for example, in works written from the mid‐nineteenth century up to the 1960s, that the women they were liberating would have recourse to servants or (as it is so often euphemistically and impersonally put) ‘household help’. With the partial exception of Woolf, who remarks briefly on the difficult lot of maids, they did not pay attention to the ‘help’, the vast majority of whom were also, of course, women—and, in the US context, usually black women. I think, however, that such problems are far less present in the works of most recent feminists. Few contemporary feminist theorists are neglectful of differences of class and race where these are relevant to their arguments.2 (p.276) But the charges of ‘essentialism’ and ‘substitutionalism’ seem to grow ever louder.

What is largely missing from anti‐essentialist or anti‐substitutionalist writing, however, is the evidence and argument that would be required to show, first, that and how the alleged failing is present in a theory. Surely here, as in other cases, it is not unreasonable to place the burden of proof on the critic. To be convincing, she needs to show that and how the alleged ‘essentialist's’ or ‘substitutionalist's’ theory omits or burdens (which latter word I take to mean ‘rationalizes the disadvantages suffered by’) persons other than those few the substitutionalist does take account of (since they are just like him/herself). If one is going to make a case that certain groups or categories of people are excluded from a theory, one needs both to show that they are and, if possible, to suggest how their inclusion would affect the theory. We know, for example, that Rawls largely ignored issues of gender in A Theory of Justice, not only because he does not mention sex as one of the characteristics that is unknown in the original position, but also because he fails to discuss many distributional issues of enormous concern to women—including the issue of the justice or injustice of the gender‐structured family.3 One of the problems of feminist anti‐essentialism, however, is that it tends to substitute the claim ‘We're all different’ for both argument and evidence. One of the few critiques of feminist essentialism that I think is done with much success at all is put forward in a paper by Angela Harris, where she first shows how ignorance of the specifics of a culture mars a particular feminist analysis of a problem, and then demonstrates how, in (p.277) some respects, black women in the USA have had a qualitatively, rather than a simply quantitatively different experience of rape than that of white women (Harris, 1990: 594, 598–601). Even here, though, I think the anti‐essentialist critique is only partly convincing. For Harris is as much disturbed by white feminists saying that black women are ‘just like us, only moreso’ as she is by their marginalizing black women or ignoring them altogether. As I shall argue, what Harris regards as this essentialist ‘insult to black women’—that the problems of women of different races and cultures are in many respects ‘similar to those of white, western women but moreso’—is exactly the one I reach in this paper, when I apply some of our Western feminist ideas about justice to the situations of poor women in many poor countries.

In ‘Why not a Feminist Theory of Justice?’ Putnam gives as specific examples of feminist substitutionalism all the chapters of Seyla Benhabib's and Drucilla Cornell's book, Feminism as Critique, except the Introduction, and my Justice, Gender, and the Family. Before moving on to the central task of this chapter, I should like to respond, briefly, to Putnam's reading of my book. While she acknowledges that I make reference to particular problems of black women, poor women, and single mothers, she objects that ‘when Okin discusses the vulnerability of women and their children, she focuses primarily on the situation of white women, and indeed of privileged white women’ (Putnam, this volume, p. 313). But this is not so. In a chapter central to the book's argument I show, on the basis of empirical data, that contemporary American families of many different types are typically unjust, and that most women suffer vulnerability from the division of labour within marriage, from the choices they make about training and work in anticipation of gender‐structured marriage and, often, from the breakdown of marriage. Any careful reading of it confirms that the chapter—and, indeed, the book as a whole—is decidedly not focused on middle‐ or upper‐class housewives or professional women whose husbands or former husbands make a good living, as Putnam suggests.4 Moreover, when in this (p.278) most empirically based chapter of the book I do generalize—as one surely must in writing anything of this type—I usually use phrases such as ‘most women’, ‘wives tend to . . . ’, ‘in many marriages’, ‘often’, ‘typically’, or ‘except in rare cases’, when a bald (‘essentialist’) generalization would be misleading. More often than not, indeed, the exceptions are actually those more‐privileged women whose point of view Putnam claims I focus on at the expense of those of all other women.

I wish now to put anti‐essentialist feminism to what I think is a reasonably tough test. In doing this, I take up the gauntlet that Elizabeth Spelman throws down in the course of the anti‐essentialist argument of Inessential Woman. She says, referring to the body of recent research about women that has appeared in many fields:

Rather than assuming that women must have something in common as women, these researchers should help look to see whether they do . . . Rather than first finding out what is true of some women as women and then inferring that this is true of all women . . . we have to investigate different women's lives and see what they have in common other than being female and being called ‘women.’ (Spelman, 1988: 137)

Trained as a philosopher, however, she does not seem to consider it appropriate for her to take up this challenge of actually looking at some of the new empirical evidence. Having said the above, she turns directly back to discussing Plato. Trained as a political scientist, I shall attempt to look at some comparative evidence. I'll put some Western feminist ideas about justice and equality to the test by seeing how well these theories—developed in the context of women in well‐off, industrialized countries—work when used to analyse the very different situations of some of the world's poorest women, living in poor countries.5 How well do our accounts and explanations of gender inequality stand up in the face of considerable cultural and socio‐economic differences?

1 Differences and Similarities in Gender Oppression: Poor Women in Poor Countries

Does the assumption ‘that there is a generalizable, identifiable and collectively shared experience of womanhood’ (Putnam, this volume, p. 312, (p.279) quoting Benhabib and Cornell, 1987: 13) have any validity, or is it indeed a substitutionalist myth, rightly challenged by Third World women and their spokesfeminists? Do the theories devised by First World feminists, particularly our critiques of non‐feminist theories of justice, have anything to say, in particular, to the poorest women in poor countries, or to those policy‐makers with the potential to affect their lives for better or for worse?

I shall address, in turn, four sets of issues that have been addressed both by recent feminist critics of Anglo‐American theories of justice, and by those development scholars who have in recent years concerned themselves with the neglect or distortion of the situation of women in the countries they study. First, why and how has the issue of inequality between the sexes been ignored or obscured for so long, and addressed only so recently? Secondly, why is it so important that it be addressed? Thirdly, what do we find, when we subject households or families to standards of justice: when we look at the largely hidden inequalities between the sexes? And finally, what are the policy implications of these findings?

1.1. Why Attention to Gender Is Comparatively New

In both development studies and theories of justice, the lack until quite recently of attention to gender—and in particular to systematic inequalities between the sexes—seems to be due primarily to two factors. The first is the assumption that the household (usually assumed to be male‐headed) is the appropriate unit of analysis. A clear dichotomy between the public (political and economic) and the private (domestic and personal) has been taken for granted, and only the former has been regarded as the appropriate sphere for development studies and theories of justice, respectively, to attend to. In ethical and political theories, the family is often viewed as an inappropriate context for justice, since love, altruism, or shared interests are assumed to hold sway within it. Alternatively, it is sometimes assumed or argued that the family is ‘naturally’ a realm of hierarchy and even injustice. (For a critique of such views, see Okin, 1989b, ch. 2.) Occasional theorists, like Rousseau, have made both of these claims. In economics, development and other, households have until recently simply been taken for granted as the appropriate units of analysis, on such questions as interests or the distribution of income and wealth. And sometimes here, too, the claims involve a curious mixture of hierarchy and altruism: the household ‘head’—an adult male, if one is present—is presumed to allocate resources optimally for the group as a whole (Sen, 1990b: 131, citing Becker, Samuelson, and critics). The public/private dichotomy and the assumption of the male‐headed household have many serious implications for women, as well as for children, which will be discussed below (Dasgupta, 1993; Jaquette, 1982: 283; Okin, 1989b: 10–14 and 124–33; Olsen, 1983 and 1985; Pateman, 1983; Sen, 1990b).

(p.280) The second factor contributing to the neglect of gender is the simple failure to disaggregate data or arguments by sex. In the development literature, it seems to appear simply in this form (for criticisms of this failure, see e.g. Chen, Huq, and D'Souza, 1981: 68; Jaquette, 1982: 284). In the justice literature, the neglect of women used to be obscured by the purportedly generic use of male pronouns and other referents—‘mankind’, and so on. Of late, the (rather more insidious) practice that I have called ‘false gender neutrality’ has appeared. This consists in the facially inclusive use of gender‐neutral terms (‘he or she’, ‘persons’, and so on), when the point being made is simply invalid or otherwise false if one actually applies it to women. Such superficial, terminological responses to feminist challenges not only strain credulity but sometimes result in nonsense (Okin, 1989b: esp. 10–13, 45). Despite such differences, then, the effect is the same in both literatures; women have not, until recently, been taken into account, so the inequalities between the sexes have been ignored.

The public/domestic dichotomy has serious implications for women, as I shall explain further below. It not only obscures intrahousehold inequalities. It also results in the failure to count a great deal of the work done by women as work, since all that is considered ‘work’ is what is done, for pay, in the public sphere. All of the work women do in bearing and rearing children, cleaning and maintaining households, caring for the old and the sick, and contributing in various ways to men's work, does not count as work. This is clearly one of those instances in which the situation of poor women in poor countries is not qualitatively different from that of most women in rich countries, but rather ‘similar but worse’. For even more, in some cases far more, of the work done by women (and often by children, too) in poor countries is rendered invisible, not counted, or subsumed under men's work. The work of subsistence farming, tending to animals, domestic crafts (especially if not for the market), and the often arduous and extremely time‐consuming fetching of water and fuel, are all added to the category of unrecognized work of women.6 Chen notes that women who do all these things ‘are listed [by policy‐makers] as “house‐wives”,’ even though ‘their tasks are as critical to the wellbeing of their families and to national production as are the men's.’ (Chen, 1983: 220; see also Dasgupta, 1993; Drèze and Sen, 1989: ch. 4; Jaquette, 1982; Sen, 1990b; Waring, 1989).

(p.281) 1.2. Why Does It Matter?

This may seem like a silly question. Indeed, I hope it will soon be unnecessary, but it isn't, yet. I therefore argue, at the outset of Justice, Gender, and the Family, that the omission from theories of justice of gender, and of much of women's lives, is significant for three major reasons. Each of these reasons applies at least as much to the neglect of gender in theories of development. The first is obvious: women matter, and their well‐being matters at least as much as that of men. I say ‘at least as much’ because the well‐being of children is so intrinsically bound up with that of women, as I shall explain further below. As has been documented by many feminist scholars and organizations, women in the USA and, to different degrees, women in other industrialized countries are disproportionately represented amongst the poor, the exploited and underpaid, the assaulted, and the powerless. (See, for example, Center for American Women and Politics Fact Sheet, 1989; Estrich, 1987; Okin, 1989b: esp. ch. 7; Rhode, 1989 and 1991; Walby, 1986.) As scholars of development have recently been making clear, the multiple inequalities between the sexes in a number of poor countries have not only highly detrimental, but fatal consequences for millions of women. Amartya Sen has recently argued that—because of severe sex‐based deprivations of various kinds—as many as 100 million fewer women exist in the world today, than might normally be expected on the basis of male/female mortality rates in societies less devaluing of women—not only the Western industrialized world, where women out‐number men quite significantly, but Africa, too, where the numbers are more equal. Compared with the average African sex ratio of 1.02 women to each man, Sen estimates that there are 30 million ‘missing’ women in India, and 38 million ‘missing’ in China. These statistics, Sen says, ‘form something like the tip of an iceberg much of which is hard to observe’ (Sen 1990b; see also Dasgupta, 1993; Drèze and Sen, 1989: ch. 4, Drèze and Sen, 1990: Introduction, pp. 11–14; but cf. Harriss, 1990; Wheeler and Abdullah, 1988). So here too we can reasonably say that the neglect of sex as a variable in much of the development literature until recently has obscured ‘similar, but much worse’ injustices than those that have been hidden in Anglo‐American theories of justice.

The second reason I have raised (in the US context) for the necessity for feminist critiques of theories of social justice is that equality of opportunity—for women and girls, but also for increasing numbers of boys—is much affected by the failure of theories of justice to address gender in‐equality. This is in part due to the greater extent of economic distress in female‐headed households. In the USA, nearly 25% of children are being raised in single female‐headed households, which include three‐fifths of all chronically poor households with children. Estimates of the proportion of households throughout the world that are headed by a single female range (p.282) from one‐fifth to one‐third, with the percentage much higher in regions with significant male out‐migration (Chen, 1983: 221; Folbre, 1992; Jaquette, 1982: 271). Many millions of children of both sexes are directly affected by the higher rate of poverty among such families.7 Theories of justice or of economic development that fail to pay attention to gender ignore this, too.

In addition, the gendered division of labour has a serious impact on the opportunities of girls and women, compared with those of boys and men, virtually regardless of economic class.8 The opportunities of females are significantly affected by the structures and practices of gendered family life, particularly by the fact that women are almost invariably primary caretakers, which has much impact on their availability for full‐time wage‐work, results in their frequently being over‐worked, and renders them less likely than men to be considered economically valuable. This factor, too, operates ‘similarly but moreso’ within poor families in many poor countries. There, too, adult women suffer—often more severely—many of the same effects of the division of labour as do women in richer countries. But, in addition, their daughters are likely to be put to work for the household at a very young age, are much less likely to be educated and to attain literacy than sons of the same households and, worst of all—less valued than their brothers—they have less chance of staying alive, because they are more deprived of food or of healthcare (Dasgupta, 1993; Drèze and Sen, 1990: ch. 4; Sen, 1990; Papanek, 1990).

Thirdly, I have argued, following in the footsteps of John Stuart Mill, that the failure to address the issue of justice within households is significant because the family is the first, and arguably the most influential, school of moral development (Mill, 1869: ch. 4; Okin, 1989b: esp. 17–23, 185–6). It is the first environment in which we experience how persons treat each other, and in which we have the potential to learn how to be just or unjust. If children see that sex difference is the occasion for obviously differential treatment, they are surely likely to be affected in their personal and moral development. They are likely to learn injustice, by absorbing the messages, if male, that they have some kind of ‘natural’ enhanced entitlement and, if female, that they are not equals, and had better get used to (p.283) being subordinated if not actually abused. As Hanna Papanek notes, in the context of poor countries: ‘Domestic groups in which age and gender difference confer power on some over others are poor environments in which to unlearn the norms of inequality,’ and ‘given the persistence of gender‐based inequalities in power, authority, and access to resources, one must conclude that socialization for gender inequality is by and large very successful’ (Papanek, 1990: 163–5, 170). The comparison of most families in rich countries with poor families in poor countries—where distinctions between the sexes often start earlier, and are much more blatant and more harmful to girls—yields, here too, the conclusion that, in the latter case, things are not so much different as ‘similar but moreso’. Many Third World families, it seems, are even worse schools of justice, because they are more extreme inculcators of the inequality of the sexes as natural and justified than their developed world equivalents. And thus there is even more need for attention to be paid to gender inequality in these contexts.

1.3. Justice in the Family

What do we find when we compare Anglo‐American feminists' findings about justice within households in their societies with recent discoveries about distributions of benefits and burdens in poor households in poor countries? Again, in many respects, the injustices of gender are quite similar.

In both situations, women's access to paid work is constrained, both by discrimination and sex segregation in the workplace and by the assumption that women are ‘naturally’ responsible for all or most of the unpaid work of the household (Bergmann, 1986; Fuchs, 1988; Gerson, 1985; Okin, 1989b: 147–56; Pateman, 1988: ch. 5; Sanday, 1974; Sen, 1990b: esp. 128–30). In both situations, women typically work longer total hours than men:

Time‐use statistics considering all work (paid and unpaid economic activity and unpaid housework) reveal that women spend more of their time working than men in all developed and developing regions except northern America and Australia, where the hours are almost equal. (United Nations Report, 1991: 81 and ch. 6 passim. See also Bergmann, 1986.)

In both situations, vastly more of women's work is not paid, and is not considered ‘productive’.9 Thus there is a wide gap between men's and (p.284) women's recorded and perceived economic participation. The perception that women's work is of less worth, largely because either unpaid or poorly paid (despite the fact that in most places they do more of it, and it is crucial to the survival of household members) contributes to women's being devalued and having less power both within the family and outside of the household (Blumstein and Schwartz, 1983; Dasgupta, 1993; Drèze and Sen, 1990: ch. 4 Okin, 1989b: ch. 7; Sanday, 1974; Sen, 1990b). This in turn compounds the likelihood that the division of labour between the sexes will continue, reinforcing women's complete or partial economic dependence upon men.

Thus women often become involved in a downward spiral of socially caused and distinctly asymmetric vulnerability (Okin, 1989b: 138; Drèze and Sen, 1989: 56–9).10 The devaluation of women's work, as well as their lesser physical strength and economic dependence upon men, in turn allows them to be subject to physical, sexual and/or psychological abuse by their husbands or other male partners (Gordon, 1988; United Nations Report, 1991: 19–20; Global Fund for Women Report, 1992). However, in many poor countries, as I have mentioned, this power differential extends beyond the abuse and overwork of women to deprivation in terms of the feeding, healthcare, and education of female children—and even to their being born or not. ‘Of 8,000 abortions in Bombay after parents learned the sex of the foetus through amniocentesis, only one would have been a boy’ (United Nations Report, 1991; see also Dasgupta, 1993; Drèze and Sen, 1989: ch. 4; Sen, 1990). In both situations, then, women's participation in work outside the household is likely to improve their status within the family. However, this is not necessarily assured. It is striking to compare Bergmann's analysis of the situation of ‘drudge wives’ in the USA, who work full‐time for pay and who also perform virtually all of the household's unpaid labour, with Sanday's finding that, in some Third World contexts, while women who do little of the work that is considered ‘productive’ have low status, many who do a great deal of it become ‘virtual slaves’ (Bergmann, 1986: ch. 11; Sanday, 1974: 201).11

Thus, though most poor women in poor countries work long hours each day, they are often economically dependent on men. This, too, is ‘similar to, but worse than’ the situation of many women in richer countries. It results from so much of their work being unpaid work, so much of their paid work being poorly paid work, and, in some cases, from men's laying claim to the wages their wives and daughters earn. Feminist critics since Ester Boserup's pioneering work in 1970 have argued that women's economic (p.285) dependency on men has been in many cases actually exacerbated by changes that development theory saw only as progressive. All too ready to assume that women were ‘housewives’ and economic dependants, mainstream theorists did not notice that central elements of ‘development’, such as technology, geographic mobility, and the conversion from subsistence to market economies were not by any means universally beneficial, but processes that frequently benefited men while cutting women out from their traditional economic and social roles and thrusting them into the modern sector ‘where they are discriminated against and exploited, often receiving cash incomes below the subsistence level . . . in turn increas[ing] female dependency’ (Jaquette, 1982; see also Boserup, 1970; Tinker, 1990: ch. 1).12 What was development for men often turned out to be regressive for women.

In both rich and poor countries, women who provide the sole economic support for their families often face particular hardship. Some, but not all, reasons for this are the same. Discrimination against women in access to jobs, pay, retention, and promotion are common to most countries, with obviously deleterious effects on female‐supported families. Many such women in both rich and poor countries also suffer from severe ‘time‐poverty’, since they are carrying the double burden of domestic and bread‐winning responsibilities. However, as Chen's paper in this volume shows, the situation of some poor women in poor countries is more like that of Western women in the nineteenth century than that of contemporary Western women: even though they have no other means of support, they are actually prohibited (by religious laws or oppressive cultural norms) from engaging in paid labour. For such women, it can indeed be liberating to be helped (as they have in some cases been by outsiders like Chen) to resist the sanctions invoked against them by family elders, neighbours, or powerful social leaders. Though many forms of wage‐work, especially those available to women, are hardly ‘liberating’, except in the most basic sense, women are surely distinctly less free if they are not allowed to engage in it, especially if they have no other means of support. Many employed women in Western, industrialized countries still face quite serious disapproval if they are mothers of young children or if the family's need for their wages is not perceived as significant. But at least, except in the most oppressive of families or subcultures, no one forbids them to work. By contrast, as Chen's paper and some of her other work make clear, the basic right to be allowed to make a much‐needed living for themselves and their children is still one (p.286) that many women in other cultures are denied (Chen, 1983 and this volume).

Here, then, is a real difference—an oppressive situation that few Western women still face, though the practical obstacles to their being both mothers and wage‐workers are still substantial. But to return to similarities: another that I discovered, while comparing some of our Western feminist ideas about justice with studies of the situations of poor women in poor countries, has to do with the dynamics of power within the family. I have applied to contemporary US families Albert Hirschman's theory of the effects of differential exit potentials on power within relationships (Okin, 1989b: ch. 7), and much the same point has been made by Barbara Bergmann, Victor Fuchs, and David Heer (Bergmann, 1986; Fuchs, 1988; Heer, 1963).

I have argued that while, unlike some of the social contexts within which Hirschman's theory was developed, marriage usually involves some mutual dependence and vulnerability; in crucial respects, gender‐structured marriage involves women in a downward spiral of socially caused vulnerability that is not shared by men. As I demonstrate in some detail, the typical division of labour within marriage makes wives far more likely to be exploited both within the marital relationship itself and in the workplace (Okin, 1989b: ch. 7). To a great extent, and in innumerable ways, women in our society are made vulnerable, economically and socially, by the gendered practices of marriage. They are first set up for vulnerability during their developing years by the socially reinforced expectations that they will be the primary or sole caretakers of children, and that they need to try to attract and retain the economic support of a man, to whose worklife they will be expected to give priority. They are rendered further vulnerable by the actual division of labour within almost all current marriages. They are disadvantaged at work by the fact that wage‐work is still largely structured around the assumption that ‘workers’ have wives at home. They are rendered far more vulnerable by becoming the primary caretakers of children or elderly parents, and their vulnerability peaks if their marriages dissolve and they become single parents. That this is socially and not naturally caused vulnerability—and that it is therefore a matter of injustice—is often overlooked, because of the unfounded assumption that women are inevitably the primary caretakers of the young, the sick, and the old.

Recent research on the differential socio‐economic outcomes of divorce for men and women in the USA confirms that their exit options are, in this respect, very disparate. As Lenore Weitzman concludes, from her study of divorce in California:

For most women and children, divorce brings precipitous downward mobility—both economically and socially. The reduction in income brings residential moves and inferior housing, drastically diminished or nonexistent funds for recreation and leisure, and intense pressures due to inadequate time and money. Financial hardships (p.287) in turn cause social dislocation and a loss of familiar networks for emotional support and social services, and intensify the psychological stress for women and children alike. (Weitzman, 1985)

Weitzman's striking finding, that in the first year after divorce the average ex‐wife's household income (adjusted for need) drops by 73% while the exhusband's rises by 42%, has been challenged by some critics as exaggerated. However, dubious assumptions are made by some of these critics, and other studies conducted in a number of different states confirm the fact that there is indeed a significant difference in the socio‐economic outcome of divorce for men and women (Bell, 1988; Glendon, 1987; Kay, 1987; McLindon, 1987; Okin, 1989b: esp. 160–7; Okin, 1991b; Rhode, 1989; Wishik, 1986. For critics of the ‘different outcomes’ view, see Duncan and Hoffmann, 1988). There is little doubt, then, that the typical asymmetric dependency of wives on husbands affects their potential for satisfactory exit, and thereby influences the effectiveness of their voice within marriage. It is highly likely that most wives, aware of their relatively dismal prospects after divorce, take this into consideration in deciding how firm a stand to take on, and even whether to raise, important issues that are likely to be conflictual—including the division of paid and unpaid work within the family. We simply cannot understand the distribution of power within the family without taking the differential exit factor into account, and the notion that marriage is a just relationship involving mutual vulnerability cannot survive this analysis.

Similar analyses of the dynamics of power within the family have recently been applied to the situation of women in poor countries by Dasgupta and Sen (Dasgupt, 1993; Sen, 1990b). Dasgupta uses the exit theory in explaining the ‘not‐uncommon’ desertion by men of their families during famines: ‘The man deserts [his wife] because his outside option in these circumstances emerges higher in his ranking than any feasible allocation within the household.’ (1993: 329) He regards the complex theory he employs—John Nash's game‐theoretic programme, as ‘needed if we are to make any progress in what is a profoundly complex matter, the understanding of household decisions’ (1993: 329).13 But the conclusion he reaches is very similar to mine: any factor that improves the husband's exit option or detracts from the wife's exit option thereby gives him additional voice, or bargaining power, in the relationship. Likewise, anything that improves the wife's exit option—her acquisition of human or physical capital, for example—will increase her autonomy and place her in a better bargaining position in the relationship (Dasgupt, 1993: 331–3). Thus when circumstances of severe poverty combine with a lack of paid employment opportunities for women, increasing women's dependency on men, men's power (p.288) within the family is likely to be greatly enhanced—in many cases legitimized by highly patriarchal cultural norms.

Employed in this context, the theory not only explains (much as does my employment of Hirschman's theory) the self‐reinforcing nature of women's lack of power within the family. It also points to the injustice of a situation in which the assumption that women are responsible for housework and childcare, their disadvantaged position in the paid workforce, and their physical vulnerability to male violence all contribute to giving them little bargaining room when their (or their children's) interests conflict with those of the men they live with, thereby in turn worsening their position relative to that of men.

Sen's application of the Nash model to what he terms ‘cooperative conflict’ within households in Third World societies adds another interesting dimension to the discussion (Sen, 1990b: esp. 134–40). As well as using Nash's model to analyse the effect of the respective ‘breakdown positions’ (exit potentials) on bargaining power within households, Sen stresses the importance for outcomes, and for the strength of bargaining positions, of the perceptions of the involved parties—both the perceptions of their welfare or interests and the perceptions of their respective contributions. Given a highly traditional division of labour, he argues that women's perceptions of their own welfare are likely to be far less distinct from their perceptions of the interests of their families as a whole than are men's,14 with the result that even less value will be attached to their own personal interests, within the already unequal bargaining situation. To add further to the inequality, because of the sex‐based biases about what constitutes productive work, the women's perceived contributions to the household are likely to be considerably less than their actual contributions. Thus Sen concludes that the effects of the differential exit potentials of the two sexes are likely to be further compounded, in traditional families, by these two other factors. And he notes that, much as I have argued in relation to intrahousehold power in industrialized countries, a weak breakdown position can lead to an even weaker one in the future course of a woman's own life, and has repercussions from one generation to the next. On the other hand, strengthening her breakdown position can place a woman in a position to develop it even further (Sen, 1990b: 137–8).

The theory connecting exit potential with power within the household, then, whether in its more or its less mathematical form, seems to be at least as applicable to the situations of very poor women in poor countries as it is to relatively well‐off women in rich countries.15 Indeed, one must surely say, (p.289) in this case, too, ‘similar, but much worse’. The distribution of power within the family is an even more crucial issue in the former situation than the latter. For the stakes are undeniably higher—no less than life or death for more than 100 million women, as has recently been shown (Drèze and Sen, 1990: ch. 4; Sen, 1990a).

1.4. Policy Implications

Some of the solutions to all these problems that have been suggested recently by scholars addressing the situation of poor women in poor countries, quite closely resemble solutions that have also been proposed by Western feminists primarily concentrating on their own societies. (By ‘solutions to problems’ I mean to refer to both what theorists and social scientists need to do, to rectify their analyses, and what policy‐makers need to do, to try to solve the social problems themselves.) First, the dichotomization of public and domestic spheres must be strongly challenged, both in theory and in practice. In the context of the industrialized world, feminist theorists have argued that the myriad interconnections between life within the household and life outside of it must be recognized, and that theorists need to develop and advocate, and policy‐makers to implement, programmes that will reduce the cycle of vulnerability that presently pursues women from home to work and back home again (Okin, 1989b: 124–33 and ch. 8; Olsen, 1983 and 1985; Pateman, 1983 and 1988). Women must be taken seriously as wage‐workers, and the workplace, schools, and other relevant institutions restructured so as to recognize the fact that most workers are, at various times in their lives, also parents, and carers for the elderly and the sick. Childcare and other domestic and nurturing work must be regarded as real work, not as the private, ‘natural’, and politically irrelevant responsibilities of women. Social policies reflecting this need to be mandated. Some examples are childbirth and parental leaves, flexible working hours, subsidized high‐quality daycare, properly enforced and adequate child‐support payments for single parents.

The public/private dichotomy needs to be challenged just as much in less developed countries, though some of the policies that are feasible in richer countries are at present economically impracticable in poorer ones. But they are resisted by those in power in both contexts, due to the persistence of beliefs that women's domestic roles and responsibilities are ‘natural’ and virtually unquestionable. Mary Anderson and Martha Chen, after long experience working to improve the lives of poor women in poor countries, (p.290) report that they have ‘learned one clear lesson from the dilemmas posed by . . . basic sexism, inherent value conflicts, and cultural blind spots’:

That is, when we emphasize women's equality with men, and their ‘rights’ to an equal share of the benefits of development, we meet continual resistance both in our own development assistance agencies and amongst the powers that be in the recipient countries . . . On the other hand, when we avoid value discussions and emphasize that women are economic producers in their roles both inside and outside their households, much of the resistance has faded. When we can demonstrate that development projects which take the gender factor into account are more apt to succeed in meeting their goals than are projects which ignore it, people who are committed to development are less frequently defensive. (Anderson and Chen, 1988)

As Chen writes, in the context of poor rural regions: ‘so long as policy‐makers make the artificial distinction between the farm and the household, between paid work and unpaid work, between productive and domestic work, women will continue to be overlooked.’ (Chen, 1983: 220) Challenging the dichotomy will also point attention to the inequities that occur within households—abuse, food, healthcare. As Papanek argues, ‘Given a focus on socialization for inequality, power relations within the household—as a central theme in examining the dynamics of households—deserve special attention.’ (Papanek, 1990: 170)

Secondly, it follows from the above that the unit of analysis both for studies and for much policy‐making must be the individual, not the (male‐headed) household.16 As feminists in industrialized countries have argued, the simultaneous assumption but neglect of all that goes on within the household obscures a great deal of inequality between the sexes, which traverses the supposed separation between public and private life. Noting that, given the greater political voice of men, public decisions affecting the poor in poor countries are often ‘guided by male preferences, not [frequently conflicting] female needs’, Dasgupta concludes that:

the maximization of well‐being as a model for explaining household behaviour must be rejected . . . . Even though it is often difficult to design and effect it, the target of public policy should be persons, not households . . . . Governments need to be conscious of the household as a resource allocation mechanism. (Dasgupt, 1993: 336)

Given the effect of paid work on women's position and bargaining power within the family—including its greater influence in Third World countries on matters of life or death—it is clear that policy‐makers should improve women's employment or self‐employment opportunities. In all countries, (p.291) the assumption that ‘workers’ are men (with wives at home) has obscured the need for social structures that make being a parent consistent with being a worker. And, especially since women are even more likely in poor countries than richer ones to be providing the sole or principal support for their households, as Chen points out, they require as much access as men to credit, skills training, labour markets, technologies—and, I would add, equal pay for their work (Chen, 1983: 221). Policies facilitating women's full economic participation and productivity are needed increasingly for the survival of their households, for women's overall socio‐economic status, and to enhance their bargaining position within their families. As Drèze and Sen say, ‘important policy implications’ follow from the ‘considerable evidence that greater involvement with outside work and paid employment does tend to go with less anti‐female bias in intra‐family distribution.’ (Drèze and Sen, 1989: 58) Because of the quite pervasive unequal treatment of female children, and its tragic consequences, the need for equal treatment of women by policy‐makers is in some countries far more urgent than the need of most women in richer countries—but again, the issue is not so much different as ‘similar but moreso’.

Finally, I shall speculate about two different ways of thinking about justice between the sexes in cultures very different from ours. I have tried to show that, for feminists thinking about justice, Rawls's theory, if revised so as to include women, the family, and issues of gender justice, has a great deal to be said for it, and that the veil of ignorance is particularly important (Okin, 1989a, 1989b). If everyone were to speak only from his or her own point of view, it is unclear that we could come up with any principles of justice at all. But the very presence of the veil, which hides from those in the original position any particular knowledge of the personal characteristics or social position they will have in the society for which they are designing principles of justice, forces them to take into account as many voices as possible, and especially to be concerned with those of the least well‐off. It enables us to reconcile the requirement that a theory of justice be universalizable with the seemingly conflicting requirement that it take account of the multiple differences among human beings.

In place of what she regards as ‘substitutionalist’ feminism, Putnam proposes an ‘interactive’ (some might call it ‘dialogic’) feminism: ‘that we listen to the voices of women of colour and women of a different class, and that we appropriate what we hear’ (Putnam, this volume, p. 315).17 Listening and discussing have much to recommend them; they are fundamental (p.292) to democracy in the best sense of the word. And sometimes, when especially oppressed women are heard, their cry for justice is clear—as in the case of the women Chen both studied and helped, who were quite clear that being allowed to leave the domestic sphere in order to earn wages would improve their situations considerably. Other studies of specific groups of Third World women confirm that many who are wage labourers are well aware of the enhanced status they have thereby gained. But some whose caste status or religious rules prohibit them from working outside of their homes are divided within their own minds between accepting the prevailing devalued status of the work they do at home (even if for the market), and experiencing discontent on account of the contradiction they face between female respectability as defined in their culture and their own potential to be more highly valued and less exploited and dependent.18

Moreover, in some cases, women actively participate in, and appear to endorse unambivalently, practices that seem to outside critics to be highly oppressive. Must we conclude that what is not perceived as oppressive is not oppressive? Or can we reasonably conclude that we do not always find out what is just by asking persons who seem to be suffering injustices what they want? Oppressed people have often internalized their oppression so well that they have no sense of what they are justly entitled to as human beings. This is certainly often the case with gender inequalities, in many different cultural contexts. As Papanek writes: ‘The clear perception of disadvantages . . . requires conscious rejection of the social norms and cultural ideal that perpetuate inequalities and the use of different criteria—perhaps from another actual or idealized society—in order to assess inequality as a prelude for action.’ (Papanek, 1990: 164–5) And as Sen states: ‘There is much evidence in history that acute inequalities often survive precisely by making allies out of the deprived. The underdog comes to accept the legitimacy of the unequal order and becomes an implicit accomplice . . . It can be a serious error to take the absence of protests and questioning of inequality as evidence of the absence of that inequality (or of the nonviability of that question).’ (Sen 1990b: 126) People may be seriously deprived yet relatively cheerful—in the ‘small mercies’ situation, where their lack of discontent is based on the ‘unquestioning acceptance of their culture's priorities and evaluations of persons' worth’ (Sen, 1990b: 127–8). Thus deprivations sometimes become gagged and muffled. But it would be ethically deeply mistaken to attach a correspondingly small value to the loss of people's well‐being because of their using such survival strategies.

Coming to terms with very little is no recipe for social justice. It is, (p.293) I believe, quite justifiable for those not thoroughly imbued with the inegalitarian norms of a culture to come forth as constructive critics of these norms. But critics originating from outside of a culture need not be distant or detached. They are much more likely to come up with helpful and relevant criticism if they find out as much as they can about the culture and the meanings of its practices and differential allocations of resources from its members themselves. Understanding these people's own perceptions of their situations is extremely important. (For a good example of this, see Boddy, 1982.) But the aim of this endeavour, in cases where serious inequalities exist, should not be simply to understand, but rather to do so with a view to politicizing the deprived so that they can begin to ask new questions about their cultural norms, with a view to improving their situation. Given this proviso, then, committed outsiders may often be better analysts and critics of social injustice than those who live within the relevant culture. And what they might well do, as part of their constructive criticism, is to try to encourage those within the culture to think about some of its oppressive, or at least questionable, practices from various points of view, including that of the least advantaged. In essence, then, after engaging in dialogue in order to understand the practices as well as possible, they would encourage those within the culture to engage in Rawlsian theorizing—to try to imagine themselves in the original position, not knowing who in the social order they were to be once the veil of ignorance were lifted.

Let us think for a moment about what light this might shed on some of the most cruel or most oppressive institutions and practices that historically or currently have been used to ‘brand’ women—foot‐binding, clitoridectomy and purdah. As Papanek shows, ‘well socialized’ women in cultures with such practices internalize them as necessary to successful female development. Even though, in the case of the former two practices, these women may retain vivid memories of their own intense pain, they perpetuate the cruelties, inflicting them or at least allowing them to be inflicted on their own daughters (Papanek, 1990; see also Boddy, 1982). Now clearly a theory of human flourishing, such as Nussbaum and Sen have been developing, would have no trouble delegitimizing such practices. But given the choice of a Rawlsian outlook or an ‘interactive feminist’ one, as defined by Putnam, I'd choose the former any day. For in the latter, speaking from their own standpoint alone, well‐socialized members of the oppressed group are all too likely to rationalize the cruelties. The men, who perceive themselves as benefiting from them, speaking from their standpoint alone, are even less likely to object. But behind the veil of ignorance, is it not much more likely that both the oppressors and the oppressed would have second thoughts? What Muslim man is likely to take the chance of spending his life in seclusion and dependency, sweltering in head‐to‐toe solid black clothing, or being forbidden to earn a living by the rules of purdah? What pre‐revolutionary Chinese man would have cast his vote for (p.294) the breaking of toes and hobbling through life, if he well might be the one with the toes and the crippled life? What man would endorse gross genital mutilation, not knowing whose genitals? And the women in these cultures, required to think of such practices from a male as well as a female perspective, might thereby, with a little distance, gain more notion of just how, rather than perfecting femininity, they perpetuate the subordination of women to men.

Martha Nussbaum writes of what happens when outsiders, instead of trying to maintain some critical distance, turn to what amounts to the worship of difference. Citing examples of sophisticated Western scholars who, in their reverence for the integrity of cultures, defend such practices as the isolation of menstruating women, and criticize Western ‘intrusions’ into other cultures such as the provision of typhoid vaccine, she finds this strange and disturbing phenomenon:

Highly intelligent people, people deeply committed to the good of women and men in developing countries, people who think of themselves as progressive and feminist and anti‐racist . . . taking up positions that converge with the positions of reaction, oppression, and sexism. Under the banner of their radically and politically correct ‘anti‐essentialism’ march ancient religious taboos, the luxury of the pampered husband, ill health, ignorance, and death. (Nussbaum, this volume, p. 66).

As Nussbaum later concludes, ‘identification need not ignore concrete local differences: in fact, at its best, it demands a searching analysis of differences, in order that the general good be appropriately realized in the concrete case. But the learning about and from the other is motivated . . . by the conviction that the other is one of us.’ As the work of some feminist scholars of development shows, using the concept of gender and refusing to let differences gag us or fragment our analyses does not mean that we should overgeneralize, or try to apply ‘standardized’ solutions to the problems of women in different circumstances. Chen argues for the value of a situation‐by‐situation analysis of women's roles and constraints, before plans can be made and programmes designed. And Papanek, too, shows how helping to educate women to awareness of their oppression requires quite deep and specific knowledge of the relevant culture (Chen, 1983 and this volume; Papanek, 1990).

Thus I conclude that gender itself is an extremely important category of analysis, and that we ought not to allow feminist thinking about injustice to be paralysed by differences among women. So long as we are careful, and develop our judgements in the light of empirical evidence, it is possible to generalize about many aspects of inequality between the sexes. From place to place, from class to class, from race to race, and from culture to culture, we find similarities in the specifics of these inequalities, in their causes and effects, although often not in their extent or their severity.

(p.295) Bibliography

Bibliography references:

Anderson, Mary, and Chen, Martha (1988). ‘Integrating WID or Restructuring Development?’, Paper prepared for WID Colloquium on Gender and Development Cooperation.

Bell, Rosalyn B. (1988). ‘Alimony and the Financially Dependent Spouse in Montgomery County, Maryland’, Family Law Quarterly, 22, 3.

Benhabib, Seyla (1987). ‘The Generalized and the Concrete Other’, in Benhabib and Cornell (see below).

—— and Cornell, Drucilla (1987). ‘Introduction: Beyond the Politics of Gender’, in Feminism as Critique. Minneapolis, Minn.: University of Minnesota Press.

Bergmann, Barbara R. (1986). The Economic Emergence of Women. New York: Basic Books.

Blumstein, Philip, and Schwartz, Pepper (1983). American Couples. New York: Morrow.

Boddy, Janice (1982). ‘Womb as oasis: the symbolic context of Pharaonic circumcision in rural Northern Sudan’, American Ethnologist, 9, 682–98.

Boserup, Ester (1970/1986). Women's Role in Economic Development. London: Allen and Unwin.

Center for American Women and Politics Fact Sheet (1989). Women in Elective Office.

Chen, Lincoln C., Huq, Emdadul, and D'Souza, Stan (1981). ‘Sex Bias in the Family Allocation of Food and Health Care in Rural Bangladesh’, Population and Development Review, 7, 55–70.

Chen, Martha Alter (1983). A Quiet Revolution: Women in Transition in Rural Bangladesh. Cambridge, Mass.: Schenkman Publishing Co.

—— (1993). ‘A Matter of Survival: Women's Right to Employment in India and Bangladesh’, this volume.

Childers, Mary, and hooks, bell (1990). ‘A Conversation about Race and Class’, in Marianne Hirsch and Evelyn Fox Keller (eds.), Conflicts in Feminism. New York: Routledge, Chapman and Hall.

Dasgupta, Partha (1993). An Inquiry into Well‐Being and Destitution. Oxford: Clarendon Press.

Drèze, Jean, and Sen, Amartya (1989). Hunger and Public Action. Oxford: Clarendon Press.

——, eds. (1990). The Political Economy of Hunger, Vol. I: Entitlement and Well‐Being. Oxford: Clarendon Press.

Duncan Saul, and Hoffman, Greg (1988). ‘What Are the Economic Consequences of Divorce?’, Demography, 25, 4.

Estrich, Susan (1987). Real Rape. Cambridge, Mass.: Harvard University Press.

Folbre, Nancy (1992). ‘Mothers on Their Own: Policy Issues for Developing Countries’, Paper prepared for The Population Council: International Center for Research on Women.

Fuchs, Victor (1988). Women's Quest for Economic Equality. Cambridge, Mass.: Harvard University Press.

Gerson, Kathleen (1985). Hard Choices: How Women Decide About Work, Career, and Motherhood. Berkeley, Calif.: University of California Press. (p.296)

Glendon, Mary Ann (1987). Abortion and Divorce in Western Law. Cambridge, Mass.: Harvard University Press.

Global Fund for Women Report (1992). Ending Violence Against Women: A Resource Guide. Menlo Park, Calif.

Gordon, Linda (1988). Heroes of Their Own Lives. New York: Viking.

Harris, Angela P. (1990). ‘Race and Essentialism in Feminist Legal Theory’, Stanford Law Review, 42, 581–616.

Harriss, Barbara (1990). ‘The Intrafamilial Distribution of Hunger in South Asia’, ch. 10 of Drèze and Sen (see above).

Heer, David (1963). ‘The Measurement and Bases of Family Power: An Overview’, Marriage and Family Living, 25, 2.

Hochschild, Arlie (1989). The Second Shift: Working Parents and the Revolution at Home. New York: Viking.

hooks, bell. (1984). Feminist Theory: From Margin to Center. Boston: South End Press.

Jaquette, Jane S. (1982). ‘Women and Modernization Theory: A Decade of Feminist Criticism’, World Politics, 34, 267–84.

Kay, Herma Hill (1987). ‘Equality and Difference: A Perspective on No‐fault Divorce and its Aftermath’, University of Cincinnati Law Review, 56, 1.

McLindon, James B. (1987). ‘Separate but Unequal: The Economic Disaster of Divorce for Women and Children’, Family Law Quarterly, 21, 3.

Mansbridge, Jane (1993). ‘Feminism and Democratic Community’, in Democratic Community, ed. John Chapman and Ian Shapiro. New York: New York University Press.

Mill, John Stuart (1869), reprinted 1988 (ed. S. Okin). The Subjection of Women. Indianapolis, Ind.: Hackett.

Minow, Martha, and Spelman, Elizabeth V. (1990). ‘In Context’, Southern California Law Review, 63, 6, 1597–1652.

Okin, Susan Moller (1989a). ‘Reason and Feeling in Thinking about Justice’, Ethics, 99, 2, 229–49.

—— (1989b). Justice, Gender and the Family. New York: Basic Books.

—— (1991), ‘Economic Equality After Divorce’, Dissent, Summer, 383–7.

Olsen, Frances (1983). ‘The Family and the Market: A Study of Ideology and Legal Reform’, Harvard Law Review, 96, 7.

—— (1985). ‘The Myth of State Intervention in the Family’, University of Michigan Journal of Law Reform, 18, 4.

Papanek, Hanna (1990). ‘To Each Less Than She Needs, From Each More Than She Can Do: Allocations, Entitlements, and Value’, in Tinker (see below).

Pateman, Carole (1983). ‘Feminist Critiques of the Public/Private Dichotomy’, in Public and Private in Social Life, ed. Stanley Benn and Gerald Gaus (London: Croom Helm);

also in Pateman, The Disorder of Women (Stanford, Calif.: Stanford University Press, 1989).

—— (1988). The Sexual Contract. Stanford, Calif.: Stanford University Press.

Putnam, Ruth Anna (1993). ‘Why not a Feminist Theory of Justice?’, this volume. RAWLS, JOHN (1971). A Theory of Justice. Cambridge, Mass.: Harvard University Press.

Rhode, Deborah (1989). Justice and Gender. Cambridge, Mass.: Harvard University Press. (p.297)

—— (1991). ‘The “No‐Problem” Problem: Feminist Challenges and Cultural Change’, Yale Law Journal, 100, 6, 1731–93.

Rich, Adrienne (1980). ‘Compulsory Heterosexuality and Lesbian Existence’, Signs, 5, 4.

Sanday, Peggy R. (1974). ‘Female Status in the Public Domain’, in Michelle Zimbalist Rosaldo and Louise Lamphere (eds.), Women, Culture, and Society. Stanford, Calif.: Stanford University Press.

Sen, Amartya (1990a). ‘More than 100 Million Women Are Missing’, New York Review of Books, 20 Dec.

—— (1990b). ‘Gender and Co‐operative Conflicts’, in Tinker (see below).

Spelman, Elizabeth V. (1988). Inessential Woman: Problems of Exclusion in Feminist Thought. Boston: Beacon Press.

Tinker, Irene, ed. (1990). Persistent Inequalities: Women and World Development. New York and Oxford: Oxford University Press.

United Nations Report (1991). The World's Women: Trends and Statistics, 1970–1990. New York: United Nations Publication.

Walby, Sylvia (1986). Patriarchy at Work: Patriarchal and Capitalist Relations in Employment. Minneapolis, Minn.: University of Minnesota Press.

Waring, Marilyn (1989). If Women Counted: A New Feminist Economics. San Francisco, Calif.: Harper and Row.

Weitzman, Lenore J. (1985). The Divorce Revolution: The Unexpected Social and Economic Consequences for Women and Children in America. New York: The Free Press.

Wheeler, E. F., and Abdullah, M. (1988). ‘Food Allocation within the Family: Response to Fluctuating Food Supply and Food Needs’, in I. de Garine and G. A. Harrison, Coping with Uncertainty in Food Supply. Oxford: Clarendon Press.

Whitehead, Ann (1990). ‘Rural Women and Food Production in Sub‐Saharan Africa’, ch. 11 of Drèze and Sen (see above).

Wishik, Heather Ruth (1986). ‘Economics of Divorce: An Exploratory Study’, Family Law Quarterly, 20, 1.


A shorter version of this paper has been published in Political Theory 22, February 1994. I am grateful to Elisabeth Hansot and Martha Nussbaum for comments on an earlier draft.

(1) ‘Essentialism’, employed in the context of feminist theory, seems to have two principal meanings. The other usage refers to the tendency to regard certain characteristics or capacities as ‘essentially’ female, in the sense that they are unalterable. Used in this second way, essentialism is very close to, if not always identical with, biological determinism.

(2) Sexual orientation is another matter; it is far more difficult to refute claims that lesbian women are neglected in much feminist theory. See Rich, 1980 for one of the most trenchant critiques of earlier feminist theories from this perspective. In the context of this volume, however, the issue of sexual orientation seems less able to be grasped than other issues of difference. As the evidence and argument that follow will indicate, most women in poor countries would seem to have little or no opportunity to live as lesbians. It is therefore impossible to gauge how many might wish to, or how they would wish to do so, if they could. Traditional, including religious, taboos, added to compulsory or virtually compulsory marriage (often at a very young age) and dependency on men, seems likely to make lesbian existence far more impossible, even unthinkable, for many Third World women than it is for Western women. This is, undoubtedly, an oppressive situation, but one that I cannot discuss further here.

(3) Putnam charges John Rawls, in A Theory of Justice, with an even more extreme substitutionalism that results from concern only with persons like himself. I, among others, have criticized Rawls for virtually leaving both women and the family out of A Theory of Justice, and for assuming a gendered division of labour. However, I think Putnam's statement that ‘the single voice which is heard in the original position is . . . the voice of a white upper or middle class adult male who is heterosexual . . . from a Christian background . . . in perfect health and not old’ is quite unsubstantiated. Rawls explicitly says that the veil of ignorance is to hide, among other things, one's ‘place in society, his class position or social status . . . his fortune in the distribution of natural assets and abilities, his intelligence and strength and the like . . . his conception of the good . . . ’ How, then, can the voice issuing forth from this position be so singular a voice as Putnam claims it is? Given that the parties in the original position are also ‘rational’ (in the economic sense) and ‘mutually disinterested’, we must ask why, if they were upper‐ or middle‐class heterosexual, healthy Christians, would they be concerned with the well‐being of anyone unlike themselves? Why are considerable portions of Rawls's theory focused on the situation of ‘the least advantaged’? How do we explain his concern with such issues as the distributions of primary goods between classes and between the more and the less talented? How, especially, do we explain the derivation of the difference principle? And how do we explain Rawls's concern with the issue of toleration—especially religious toleration and toleration of those whom the majority find hard to tolerate? Surely the preoccupation of his theory with these issues reflects the very purpose for which he invented the veil of ignorance in the first place—so that many different points of view would be considered. As Rawls explains it, the veil exists so as to ‘nullify the effects of specific contingencies which put men at odds and tempt them to exploit social and natural circumstances to their own advantage’ (Rawls, 1971: 136). This interpretation of Rawls's theory as designed to take differences between persons into account, rather than to ignore them, is developed in Okin, 1989a.

(4) The chapter, ‘Vulnerability by Marriage’, is thirty‐five pages long. On all but six pages (two of which are taken up with summarizing a couple of non‐feminist theories), I refer to differences amongst US women themselves and/or their socio‐economic situations. In the end notes, many more such distinctions are drawn. The differences in personal characteristics of women attended to in the text include race, age, and sexual orientation. In addition to six explicit references to racial differences in the text of the chapter, much that is said about poverty and single female parenting is obviously more likely to apply to women of colour than to white women. The situational differences, also quite frequently referred to, include whether women themselves and/or their husbands or male partners are earning much or little; whether women are single, separated, divorced or married; whether or not they are mothers; whether or not they are single parents; whether they do full‐time or part‐time wage‐work or do not work outside of the home; whether their work is professional work or not; whether they do it voluntarily or out of economic necessity; and so on. The policy recommendations made in the final chapter also confirm that the book is not a work of substitutionalist feminism. So does much of the rest of the book, especially ch. 3, where I criticize Alasdair MacIntyre's and Michael Walzer's communitarian theories on the grounds that the ‘traditions’ and ‘shared meanings’ they defend frequently fail to represent the needs and/or interests of many oppressed groups.

(5) I focus, though by no means exclusively, on my own recent book, for several reasons: because in Putnam's paper, she finds it such a good example of substitutionalist theories that are not widely applicable and cannot cope with differences; because it's one of the furthest developed attempts at a feminist theory of justice so far (though Iris Young's Justice and the Politics of Difference (Princeton, NJ: Princeton University Press, 1990) develops feminist ideas about justice at least as far, although along the lines of stressing difference); and because I was curious to see whether my analysis (both of theory and society)—developed in the context of the USA—was applicable to the very different situations of poor women in poor countries.

(6) This is so in spite of the fact that the detailed division of labour between the sexes varies considerably from culture to culture. As Jane Mansbridge has recently written, in a discussion of gratuitous gendering: ‘Among the Aleut of North America, for example, only women are allowed to butcher animals. But among the Ingalik of North America, only men are allowed to butcher animals. Among the Suku of Africa, only the women can plant crops and only the men can make baskets. But among the Kaffa of the Circum‐Mediterranean, only the men can plant crops and only the women can make baskets. . . . ’ (Mansbridge, forthcoming) However, the work done by women is less likely to be ‘outside’ work, or to be paid or valued. Her analysis is derived from data in George P. Murdoch and Caterina Provost, ‘Factors in the Division of Labor by Sex: A Cross‐Cultural Analysis’, Ethnology, 12 (1973), 203–25.

(7) Poverty is both a relative and an absolute term. The poorest households in poor countries are absolutely as well as relatively poor, and can be easily pushed below subsistence by any number of natural, social or personal catastrophes. Poverty in rich countries is more often relative poverty (though there is serious malnutrition currently in the USA, for example, and drug abuse, with all its related ills, including AIDS, is highly correlated with poverty). Relative poverty, though less directly life‐threatening, can however be very painful, especially for children living in societies that are not only highly consumer‐oriented, but in which many opportunities—for good healthcare, decent education, the development of talents, pursuit of interests, and so on—are seriously limited for those from poor families.

(8) By saying ‘virtually regardless of economic class’, I am not implying that their opportunities are not also differentially constrained depending on their position in the class structure. What I mean is that numerous constraints of being female in a sexist society are likely to be experienced by women of all classes.

(9) See Dasgupta, 1993 on the effect of members' perceived ‘usefulness’ on the allocation of goods within poor families. Western as well as non‐Western studies show us that women's work is already likely to be regarded as less useful—even when it is just as necessary to family well‐being. So when women are really made less useful (by convention or lack of employment opportunities), this problem is compounded. Dasgupta questions simple measures of usefulness, such as paid employment, in the case of girls (1993: 309). Where young poor women aren't entitled to parental assets, and their outside employment opportunities are severely restricted, the only significant ‘employment’ for them is as child‐bearers and housekeepers—so marriage becomes especially valued even though its conditions may be highly oppressive.

(10) In Justice, Gender, and the Family, I referred to this as a cycle, rather than a downward spiral. However, Louise Tilly has recently pointed out to me that the latter term is more accurate, since a cycle ends up where it started, and what I am describing is a deteriorating condition.

(11) There is conflicting evidence on this matter. See, for example, Papanek, 1990: 166–8.

(12) This seems similar to changes in the work and socio‐economic status of women in Western Europe in the 16th to 18th centuries. The separation of much of production from the domestic sphere constructed the idea of men as breadwinners (who were therefore thought to need to be paid a wage on which a family could subsist) and the idea that women and children were dependent upon men (and could therefore be paid below‐subsistence wages). In fact, many working‐class women, then as now, either had no man to depend on, or one who earned insufficient to support a family.

(13) Sen (1990b) also uses the Nash model, though he does not preclude other ways of analysing such bargaining situations.

(14) The evidence he cites for this point is from a study of Indian women: V. Das and R. Nicholas (1981), ‘ “Welfare” and “Well‐Being” in South Asian Societies’, ACLS‐SSRC Joint Committee on South Asia (New York: Social Science Research Council).

(15) As Sen says: ‘Some disadvantages of women would apply in both types of situations. For example, frequent pregnancy and persistent child rearing . . . must make the outcome of cooperative conflicts less favorable to women through worse breakdown position and a lower ability to make a perceived contribution to the economic fortunes of the family. Other disadvantages are much more specific to the nature of the community, for example, greater illiteracy and less higher education of women in most developing—and some developed—countries today, and these too would tend to make the breakdown positions worse for women.’ He adds: ‘The “perceived interest response” and the “perceived contribution response”, can be tremendously more regressive for women in some societies.’ (1990b: 137)

(16) This point seems to have been first explicitly made in the context of policy by George Bernard Shaw, who argued in 1928 that the state should require all adults to work and should allocate an equal portion of income to each person—man, women, and child (The Intelligent Woman's Guide to Socialism and Capitalism (1928; reprinted New Brunswick, NJ, 1984)).

(17) As Joan Tronto has pointed out to me, Putnam's use of the word ‘appropriate’ is odd in this context, given that she is arguing against a kind of intellectual imperialism on the part of privileged Western feminists, and for the necessity of listening to the voices of women of colour and of different classes. What is the benefit for these women of their speaking if what they say is ‘appropriated’ by feminist theorists?

(18) Chen in this volume; Zarina Bhatty, ‘Economic Role and Status of Women: A Case Study of Women in the Beedi Industry in Allahabad’, ILO Working Paper, Geneva (quoted in Sen, 1990b); Maria Mies, Lacemakers in Narsapur: Indian Housewives Produce for the World Market (London: Zed Press, 1982), cf. 173–4 with 157.