Fortunes of Feminism (24 page)

5
Judith Stacey, “Sexism By a Subtler Name? Postindustrial Conditions and Postfeminist Consciousness in the Silicon Valley,”
Socialist Review
no. 96, 1987, 7–28.

6
Kath Weston,
Families We Choose: Lesbians, Gays, Kinship
, New York: Columbia University Press, 1991.

7
Nancy Fraser, “Clintonism, Welfare, and the Antisocial Wage: The Emergence of a Neoliberal Political Imaginary,”
Rethinking Marxism
6:1, Spring 1993, 9–23.

8
Some of the most sophisticated discussions are found in
Feminist Legal Theory: Readings in Law and Gender
, eds. Katharine T. Bartlett and Rosanne Kennedy, Boulder, CO: Westview Press, 1991.

9
David T. Ellwood,
Poor Support: Poverty in the American Family
, New York: Basic Books, 1988.

10
Robert Goodin,
Reasons for Welfare: The Political Theory of the Welfare State
, Princeton, NJ: Princeton University Press, 1988.

11
Not all dependencies are exploitable. In
Reasons for Welfare
, 175–6, Robert Goodin specifies the following four conditions that must be met if a dependency is to be exploitable: 1) the relationship must be asymmetrical; 2) the subordinate party must need the resource that the superordinate supplies; 3) the subordinate must depend on some particular superordinate for the supply of needed resources; 4) the superordinate must enjoy discretionary control over the resources that the subordinate needs from him/her.

12
Albert O. Hirschman,
Exit, Voice, and Loyalty: Responses to Decline in Firms, Organizations, and States
, Cambridge, MA: Harvard University Press, 1970; Susan Moller Okin,
Justice, Gender, and the Family
, New York: Basic Books, 1989; Barbara Hobson, “No Exit, No Voice: Women's Economic Dependency and the Welfare State,”
Acta Sociologica
33:3, Fall 1990, 235–50.

13
Frances Fox Piven and Richard A. Cloward,
Regulating the Poor
, New York: Random House, 1971; Gosta Esping-Andersen,
The Three Worlds of Welfare Capitalism
, Princeton, NJ: Princeton University Press, 1990.

14
Robert Goodin,
Reasons for Welfare
.

15
Edward V. Sparer, “The Right to Welfare,” in
The Rights of Americans: What They are—What They Should Be
, ed. Norman Dorsen, New York: Pantheon, 1970.

16
Ann Shola Orloff, “Gender and the Social Rights of Citizenship: The Comparative Analysis of Gender Relations and Welfare States,”
The American Sociological Review
58:3, June 1993, 303–28. The anti-exploitation objective should not be confused with current US attacks on “welfare dependency,” which are highly ideological. These attacks define “dependency” exclusively as receipt of public assistance. They ignore the ways in which such receipt can promote claimants' independence by preventing exploitable dependence on husbands and employers. For a critique of such views, see Chapter 3 of this volume, Fraser and Gordon, “A Genealogy of ‘Dependency.'”

17
Ruth Lister, “Women, Economic Dependency, and Citizenship,”
Journal of Social Policy
19:4, 1990, 445–67; Amartya Sen, “More Than 100 Million Women Are Missing,”
New York Review of Books
37:20, December 20, 1990, 61–6.

18
Lenore Weitzman,
The Divorce Revolution: The Unexpected Social Consequences for Women and Children in America
, New York: Free Press, 1985.

19
David T. Ellwood,
Poor Support
, 45.

20
Lois Bryson, “Citizenship, Caring and Commodification,” unpublished paper presented at conference on Crossing Borders: International Dialogues on Gender, Social Politics and Citizenship, Stockholm, May 27–29, 1994; Arlie Hochschild,
The Second Shift: Working Parents and the Revolution at Home
, New York: Viking Press, 1989; Juliet Schor,
The Overworked American: The Unexpected Decline of Leisure
, New York: Basic Books, 1991.

21
Ruth Lister, “Women, Economic Dependency, and Citizenship.”

22
Laura Balbo, “Crazy Quilts,” in
Women and the State
, ed. Ann Showstack Sassoon, London: Hutchinson, 1987.

23
Actually, there is a heavy ideological component in the usual view that public assistance is need-based, while social insurance is desert-based. Benefit levels in social insurance do not strictly reflect “contributions.” Moreover, all government programs are financed by “contributions” in the form of taxation. Public assistance programs are financed from general revenues, both federal and state. Welfare recipients, like others, “contribute” to these funds, for example, through payment of sales taxes. See Nancy Fraser and Linda Gordon, “Contract versus Charity: Why Is There No Social Citizenship in the United States?”
Socialist Review
22:3, July–September 1992, 45–68.

24
The free-rider worry is usually posed androcentrically as a worry about shirking paid employment. Little attention is paid, in contrast, to a far more widespread problem, namely, men's free-riding on women's unpaid domestic labor. A welcome exception is Peter Taylor-Gooby, “Scrounging, Moral Hazard, and Unwaged Work: Citizenship and Human Need,” unpublished typescript, 1993.

25
Employment-enabling services could be distributed according to need, desert, or citizenship, but citizenship accords best with the spirit of the model. Means-tested day care targeted for the poor cannot help but signify a failure to achieve genuine breadwinner status; and desert-based day care sets up a catch-22: one must already be employed in order to get what is needed for employment. Citizenship-based entitlement is best, then, but it must make services available to all, including to immigrants. This rules out Swedish-type arrangements, which fail to guarantee sufficient day care places and are plagued by long queues. For the Swedish problem, see Barbara Hobson, “Economic Dependency and Women's Social Citizenship: Some Thoughts on Esping-Andersen's Welfare State Regimes,” unpublished typescript, 1993.

26
That incidentally would be to break decisively with US policy, which has assumed since the New Deal that job creation is principally for men. Bill Clinton's 1992 campaign proposals for “industrial“ and “infrastructural investment“ policies were no exception in this regard. See Nancy Fraser, “Clintonism, Welfare, and the Antisocial Wage.”

27
Government could itself provide carework services in the form of public goods or it could fund marketized provision through a system of vouchers. Alternatively, employers could be mandated to provide employment-enabling services for their employees, either through vouchers or in-house arrangements. The state option means higher taxes, of course, but it may be preferable nevertheless. Mandating employer responsibility creates a disincentive to hire workers with dependents, to the likely disadvantage of women.

28
Evelyn Nakano Glenn, “From Servitude to Service Work: Historical Continuities in the Racial Division of Paid Reproductive Labor,”
Signs: Journal of Women in Culture and Society
18:1, Autumn 1992, 1–43.

29
It, too, conditions entitlement on desert and defines “contribution“ in traditional androcentric terms as employment and wage deductions.

30
Exactly what else must be provided inside the residual system will depend on the balance of entitlements outside it. If health insurance is provided universally as a citizen benefit, for example, then there need be no means-tested health system for the non-employed. If, however, mainstream health insurance is linked to employment, then a residual health care system will be necessary. The same holds for unemployment, retirement, and disability insurance. In general, the more that is provided on the basis of citizenship, instead of on the basis of desert, the less has to be provided on the basis of need. One could even say that desert-based entitlements create the necessity of need-based provision; thus, employment-linked social insurance creates the need for means-tested public assistance.

31
Peter Kilborn, “New Jobs Lack the Old Security in Time of ‘Disposable Workers,'”
New York Times
, March 15 1993, A1, A6.

32
Failing that, however, several groups are especially vulnerable to poverty in this model: those who cannot work, those who cannot get secure, permanent, full-time, good-paying jobs—disproportionately women and/or people of color; and those with heavy, hard-to-shift, unpaid carework responsibilities—disproportionately women.

33
Failing that, however, the groups mentioned in the previous note remain especially vulnerable to exploitation—by abusive men, by unfair or predatory employers, by capricious state officials.

34
Exactly how much remains depends on the government's success in eliminating discrimination and in implementing comparable worth.

35
Universal Breadwinner presumably relies on persuasion to induce men to do their fair share of unpaid work. The chances of that working would be improved if the model succeeded in promoting cultural change and in enhancing women's voice within marriage. But it is doubtful that this alone would suffice, as the Communist experience suggests.

36
Christine A. Littleton, “Reconstructing Sexual Equality,” in
Feminist Legal Theory
, eds. Katharine T. Bartlett and Rosanne Kennedy.

37
Caregiver allowances could be distributed on the basis of need, as a means-tested benefit for the poor—as they have always been in the United States. But that would contravene the spirit of Caregiver Parity. One cannot consistently claim that the caregiver life is equivalent in dignity to the breadwinner life, while supporting it only as a last-resort stop-gap against poverty. (This contradiction has always bedeviled mothers' pensions—and later Aid to Dependent Children—in the United States. Although these programs were intended by some advocates to exalt motherhood, they sent a contradictory message by virtue of being means-tested and morals-tested.) Means-tested allowances, moreover, would impede easy transitions between employment and carework. Since the aim is to make caregiving as deserving as breadwinning, caregiver allowances must be based on desert. Treated as compensation for socially necessary “service” or “work,” they alter the standard androcentric meanings of those terms.

38
In
Justice, Gender, and the Family
, Susan Okin has proposed an alternative way to fund carework. In her scheme the funds would come from what are now considered to be the earnings of the caregiver's partner. A man with a non-employed wife, for example, would receive a paycheck for one-half of “his” salary; his employer would cut a second check in the same amount payable directly to the wife. Intriguing as this idea is, one may wonder whether it is really the best way to promote a wife's independence from her husband, as it ties her income so directly to his. In addition, Okin's proposal does not provide any carework support for women without employed partners. Caregiver Parity, in contrast, provides public support for all who perform informal carework. Who, then, are its beneficiaries likely to be? With the exception of pregnancy leave, all the model's benefits are open to everyone; so men as well as women can opt for a “feminine” life. Women, however, are considerably more likely to do so. Although the model aims to make such a life costless, it includes no positive incentives for men to change. Some men, of course, may simply prefer such a life and will choose it when offered the chance; most will not, however, given current socialization and culture. We shall see, moreover, that Caregiver Parity contains some hidden disincentives to male caregiving.

39
In this respect, it resembles the Universal Breadwinner model: whatever additional essential goods are normally offered on the basis of desert must be offered here too on the basis of need.

40
Wages from full-time employment must also be sufficient to support a family with dignity.

41
Adults with neither carework nor employment records would be most vulnerable to poverty in this model; most of these would be men. Children, in contrast, would be well protected.

42
Once again, it is adults with neither carework nor employment records who are most vulnerable to exploitation in this model; and the majority of them would be men.

43
Susan Okin,
Justice, Gender, and the Family
.

44
Joan Williams, “Deconstructing Gender,” in
Feminist Legal Theory
, eds. Katharine T. Bartlett and Rosanne Kennedy.

45
Quoted in Ruth Lister, “Women, Economic Dependency, and Citizenship,” 463.

PART II

Feminism Tamed:

From Redistribution to Recognition in the Age of Identity

5

Against Symbolicism:
The Uses and Abuses of Lacanianism for Feminist Politics
^*

This chapter grew out of an experience of severe puzzlement. For several years I watched with growing incomprehension as a large and influential body of feminist scholars created an interpretation of Jacques Lacan's theory of the symbolic order, which they sought to use for feminist purposes. I myself had felt a disaffinity with Lacanian thought, as much intellectual as political. So, while many of my fellow feminists were adapting quasi-Lacanian ideas to theorize the discursive construction of subjectivity in film and literature, I was relying on alternative models to develop an account of language that could inform a feminist social theory.
¹
For a long while, I avoided any explicit, meta-theoretical discussion of these matters. I explained neither to myself nor to my colleagues why I looked to the discourse models of writers like Foucault, Bourdieu, Bakhtin, Habermas, and Gramsci instead of to those of Lacan, Kristeva, Saussure, and Derrida.
²
In this essay, I want to provide such an explanation. I will try to explain why I think feminists should eschew the versions of discourse theory that they attribute to Lacan and the related theories attributed to Julia Kristeva. I will also try to identify some places where I think we can find more satisfactory alternatives.

1. WHAT DO FEMINISTS WANT IN A DISCOURSE THEORY?

Let me begin by posing two questions: What might a theory of discourse contribute to feminism? And what, therefore, should feminists look for in a theory of discourse? I suggest that a conception of discourse can help us understand at least four things, all of which are interrelated. First, it can help us understand how people's social identities are fashioned and altered over time. Second, it can help us understand how, under conditions of inequality, social groups in the sense of collective agents are formed and unformed. Third, a conception of discourse can illuminate how the cultural hegemony of dominant groups in society is secured and contested. Fourth and finally, it can shed light on the prospects for emancipatory social change and political practice. Let me elaborate.

First, consider the uses of a conception of discourse for understanding social identities. The basic idea here is that people's social identities are complexes of meanings, networks of interpretation. To have a social identity, to be a woman or a man, for example, just
is
to live and to act under a set of descriptions. These descriptions, of course, are not simply secreted by people's bodies; nor are they simply exuded by people's psyches. Rather, they are drawn from the fund of interpretive possibilities available to agents in specific societies. It follows that, in order to understand the gender dimension of social identity, it does not suffice to study biology or psychology. Instead, one must study the historically specific social practices through which cultural descriptions of gender are produced and circulated.
³

Moreover, social identities are exceedingly complex. They are knitted together from a plurality of different descriptions arising from a plurality of different signifying practices. Thus, no one is simply a woman; one is rather, for example, a white, Jewish, middle-class woman, a philosopher, a lesbian, a socialist, and a mother.
⁴
Because everyone acts in a plurality of social contexts, moreover, the different descriptions comprising any individual's social identity fade in and out of focus. Thus, one is not always a woman in the same degree; in some contexts, one's womanhood figures centrally in the set of descriptions under which one acts; in others, it is peripheral or latent.
⁵
Finally, it is not the case that people's social identities are constructed once and for all and definitively fixed. Rather, they alter over time, shifting with shifts in agents' practices and affiliations. Even the way in which one is a woman will shift—as it does, to take a dramatic example, when one becomes a feminist. In short, social identities are discursively constructed in historically specific social contexts; they are complex and plural; and they shift over time. One use of a conception of discourse for feminist theorizing, then, is in understanding social identities in their full socio-cultural complexity, thus in demystifying static, single variable, essentialist views of gender identity.

A second use of a conception of discourse for feminist theorizing is in understanding the formation of social groups. How does it happen, under conditions of domination, that people come together, arrange themselves under the banner of
collective
identities, and constitute themselves as collective social agents? How do class formation and, by analogy, gender formation occur?

Clearly, group formation involves shifts in people's social identities and therefore also in their relation to social discourse. One thing that happens here is that pre-existing strands of identities acquire a new sort of salience and centrality. These strands, previously submerged among many others, are reinscribed as the nub of new self-definitions and affiliations.
⁶
For example, in the current wave of feminist ferment, many of us who had previously been “women” in some taken-for-granted way have now become “women” in the very different sense of a discursively self-constituted political collectivity. In the process, we have remade entire regions of social discourse. We have invented new terms for describing social reality—for example, “sexism,” “sexual harassment,” “marital, date, and acquaintance rape,” “labor force sex-segregation,” “the double shift,” and “wife-battery.” We have also invented new language games such as consciousness-raising and new, institutionalized public spheres such as the Society for Women in Philosophy.
⁷
The point is that the formation of social groups proceeds by struggles over social discourse. Thus, a conception of discourse is useful here, both for understanding group formation and for coming to grips with the closely related issue of socio-cultural hegemony.

“Hegemony” is the Italian Marxist Antonio Gramsci's term for the discursive face of power. It is the power to establish the “common sense” or “doxa” of a society, the fund of self-evident descriptions of social reality that normally go without saying.
⁸
This includes the power to establish authoritative definitions of social situations and social needs, the power to define the universe of legitimate disagreement, and the power to shape the political agenda. Hegemony, then, expresses the advantaged position of dominant social groups with respect to discourse. It is a concept that allows us to recast the issues of social identity and social groups in the light of societal inequality. How do pervasive axes of dominance and subordination affect the production and circulation of social meanings? How does stratification along lines of gender, “race,” and class affect the discursive construction of social identities and the formation of social groups?

The notion of hegemony points to the intersection of power, inequality, and discourse. However, it does not entail that the ensemble of descriptions that circulate in society comprise a monolithic and seamless web, nor that dominant groups exercise an absolute, top-down control of meaning. On the contrary, “hegemony” designates a process wherein cultural authority is negotiated and contested. It presupposes that societies contain a plurality of discourses and discursive sites, a plurality of positions and perspectives from which to speak. Of course, not all of these have equal authority. Yet conflict and contestation are part of the story. Thus, one use of a conception of discourse for feminist theorizing is to shed light on the processes by which the socio-cultural hegemony of dominant groups is achieved and contested. What are the processes by which definitions and interpretations inimical to women's interests acquire cultural authority? What are the prospects for mobilizing counter-hegemonic feminist definitions and interpretations to create broad oppositional groups and alliances?

The link between these questions and emancipatory political practice is, I believe, fairly obvious. A conception of discourse that lets us examine identities, groups, and hegemony in the ways I have been describing would be of considerable use to feminist practice. It would valorize the empowering dimensions of discursive struggles without leading to “culturalist” retreats from political engagement.
⁹
In addition, the right kind of conception would counter the disabling assumption that women are just passive victims of male dominance. That assumption over-totalizes male dominance, treating men as the only social agents and rendering inconceivable our own existence as feminist theorists and activists. In contrast, the sort of conception I have been proposing would help us understand how, even under conditions of subordination, women participate in the making of culture.

2. LACANIANISM AND THE LIMITS OF STRUCTURALISM

In light of the foregoing, what sort of conception of discourse will be useful for feminist theorizing? What sort of conception best illuminates social identities, group formation, hegemony, and emancipatory practice?

In the postwar period, two approaches to theorizing language became influential among political theorists. The first is the
structuralist model
, which studies language as a symbolic system or code. Derived from Saussure, this model is presupposed in the version of Lacanian theory I shall be concerned with here; in addition, it is abstractly negated but not entirely superseded in deconstruction and in related forms of French “women's writing.” The second influential approach to theorizing language may be called the
pragmatics model
, which studies language at the level of discourses, as historically specific social practices of communication. Espoused by such thinkers as Mikhail Bakhtin, Michel Foucault, and Pierre Bourdieu, this model is operative in some but not all dimensions of the work of Julia Kristeva and Luce Irigaray. In the present section of this chapter, I shall argue that the first, structuralist model is of only limited usefulness for feminist theorizing.

Let me begin by noting that there are good
prima facie
reasons for feminists to be suspicious of the structuralist model. This model constructs its object of study by abstracting from exactly what we need to focus on, namely, the social practice and social context of communication. Indeed, the abstraction from practice and context are among the founding gestures of Saussurean linguistics. Saussure began by splitting signification into
langue
, the symbolic system or code, and
parole
, speakers' uses of language in communicative practice or speech. He then made the first of these,
langue
, the proper object of the new science of linguistics, and relegated the second,
parole
, to the status of a devalued remainder.
¹⁰
At the same time, Saussure insisted that the study of
langue
be synchronic rather than diachronic; he thereby posited his object of study as static and atemporal, abstracting it from historical change. Finally, the founder of structuralist linguistics posited that
langue
was indeed a single system; he made its unity and systematicity consist in the putative fact that every signifier, every material, signifying element of the code, derives its meaning positionally through its difference from all of the others.

Together, these founding operations render the structuralist approach of limited utility for feminist purposes.
¹¹
Because it abstracts from
parole
, the structuralist model brackets questions of practice, agency, and the speaking subject. Thus, it cannot shed light on the discursive practices through which social identities and social groups are formed. Because this approach brackets the diachronic, moreover, it will not tell us anything about shifts in identities and affiliations over time. Similarly, because it abstracts from the social context of communication, the model brackets issues of power and inequality. Thus, it cannot illuminate the processes by which cultural hegemony is secured and contested. Finally, because the model theorizes the fund of available linguistic meanings as a single symbolic system, it lends itself to a monolithic view of signification that denies tensions and contradictions among social meanings. In short, by reducing discourse to a “symbolic system,” the structuralist model evacuates social agency, social conflict, and social practice.
¹²

Let me now try to illustrate these problems by means of a brief discussion of Lacanianism. By “Lacanianism,” I do not mean the actual thought of Jacques Lacan, which is far too complex to tackle here. I mean, rather, an ideal-typical neo-structuralist reading of Lacan that is widely credited among English-speaking feminists.
¹³
In discussing “Lacanianism,” I shall bracket the question of the fidelity of this reading, which could be faulted for overemphasizing the influence of Saussure at the expense of other, countervailing influences, such as Hegel.
¹⁴
For my purposes, however, this ideal-typical, Saussurean reading of Lacan is useful precisely because it evinces with unusual clarity the difficulties that beset many conceptions of discourse that are widely considered “poststructuralist” but that remain wedded in important respects to structuralism. Because their attempts to break free of structuralism remain abstract, such conceptions tend finally to recycle it. Lacanianism, as discussed here, is a paradigm case of “neo-structuralism.”
¹⁵

At first sight, neo-structuralist Lacanianism seems to promise some advantages for feminist theorizing. By conjoining the Freudian problematic of the construction of gendered subjectivity to the Saussurean model of structural linguistics, it seems to provide each with its needed corrective. The introduction of the Freudian problematic promises to supply the speaking subject that is missing in Saussure and thereby to reopen the excluded questions about identity, speech, and social practice. Conversely, the use of the Saussurean model promises to remedy some of Freud's deficiencies. By insisting that gender identity is
discursively
constructed, Lacanianism appears to eliminate lingering vestiges of biologism in Freud, to treat gender as socio-cultural all the way down, and to render it in principle more open to change.