ON PSYCHOLOGICAL OPPRESSION / SANDRA LEE BARTKY
ordinary contexts. For Hegel, Marx, and Sartre, "objectification" is an important moment in the dialectic of consciousness. My decision to treat ordinary sexual relations or even sexual desire alone as involving some objectification is based on a desire to remain within this tradition. Further, Sartre's phenomenology of sexual desire in Being and Nothingness (New York: Philosophical Library, 1966) draws heavily on a concept of objectification in an unusually compelling description of the experienced character of that state:
The caress by realizing the Other's incarnation reveals to me my own incarnation; that is, I make myself flesh in order to impel the Other to realize for-herself and for-me her own flesh, and my caresses cause my flesh to be born for me in so far as it is for the Other flesh causing her to be born as flesh. I make her enjoy my flesh through her flesh in order to compel her to feel herself flesh. And so possession truly appears as a double reciprocal incarnation, (p. 508)
What I call "objectification," Sartre here calls "incarnation," a refinement not necessary for my purposes. What he call "sadism" is incarnation without reciprocity. Most of my examples of sexual objectification would fall into the latter category. 1L Fanon, Black Skin, White Masks, p. 177. Eldridge Cleaver sounds a similar theme in Soul on Ice (New York: Dell 1968).The archetypal white man in American society, for Cleaver, is the "Omnipotent Administrator," the archetypal black man the "Super-Masculine Menial."
12. Ibid., p. 180.
13. Erik Erikson,"lnner and Outer Space: Reflections on Womanhood," Daedalus, vol. 93, 1961, pp. 582-606.
14. Miluilo Markovic, "Women's Liberation and Human Emancipation," in Women and Philosophy, pp. 165-166. In spite of this lapse and some questionable opinions conT cerning the nature of female sexuality, Markovic's paper is a most compelling defense of the claim that the emancipation of women cannot come about under capitalism.
15. Exod. 1:13-14.
16. The available clinical literature on the psychological effects of social inferiority supports this claim. Sec William H. Gricr and Price M. Cobbs, Black Rage (New York: Grosset & Dunlap, 1969); Pauline Bart, "Depression in Middle-Aged Women," in Women in Sexist Society, ed. Vivian Gornick and Barbara Moran (New York: New American Library, 1971), pp. 163-186; also Phyllis Chesler, Women and Madness (New York: Doubleday, 1972).
17. Bertell Oilman, A lienation: Marx's Conception of Man in Capitalist Society (London and New York: Cambridge University Press, 1971), p. 135.
18. Ibid. p. 143.
19. Karl Marx, The Economic and Philosophical Manuscripts of 1844, ed. Dirk J. Struik (New York: International Publishers, 1964), p.111.
20. Ibid,
21. The use of the masculine, possessive pronoun is deliberate.
FIVE FACES OF OPPRESSION
Iris Young
Someone who does not see a pane of glass does not know that he does not see it. Someone who, being placed differently, does see it does not know the other does not see it.
When our will finds expression outside ourselves in actions performed by others, we do not waste our time and our power of attention in examining whether they have consented to this. This is true for all of us. Our attention, given entirely to the success of the undertaking, is not claimed by them as long as they are docile____
Rape is a terrible caricature of love from which consent is absent. After rape, oppression is the second horror of human existence. It is a terrible caricature of obedience.
—SIMONEWEIL
I have proposed an enabling conception of justice. Justice should refer not only to distribution, but also to the institutional conditions necessary for the development and exercise of individual capacities and collective communication and cooperation. Under this conception of justice, injustice refers primarily to two forms of disabling constraints, oppression and domination. While these constraints include distributive patterns, they also involve matters which cannot easily be assimilated to the logic of distribution: decisionmaking procedures, division oflabor, and culture.
Many people in the United States would not choose the term "oppression" to name injustice in our society. For contemporary emancipatory social movements, on the other hand—socialists, radical feminists, American Indian activists, Black activists, gay and lesbian activists—oppression is a central category of political discourse. Entering the political discourse in which oppression is a central category involves adopting a general mode of analyzing and evaluating social structures and practices which is incommensurate with the language of liberal individualism that dominates political discourse in the United States.
A major political project for those of us who identify with at least one of these movements must thus be to persuade people that the discourse of oppression makes sense of much of our social experience. We are ill prepared for this task, however, because we.have no clear account of the meaning of oppression. While we find the term used often in the diverse philosophical literature spawned by radical social movements in the United States, we find little [ '     direct discussion of the meaning of the concept as used by these movements.
In this chapter I offer some explication of the concept of oppression as I understand its use by new social movements in the United States since the
■    FIVE FACES OF OPPRESSION / IRIS YOUNG
Five Faces of Oppression    ■ 39
1960s. My starting point is reflection on the conditions of the groups said by these movements to be oppressed: among others women, Blacks, Chicanos, Puerto Ricans and other Spanish-speaking Americans, American Indians, Jews, lesbians, gay men, Arabs, Asians, old people, working-class people, and the physically and mentally disabled. I aim to systematize the meaning of the concept of oppression as used by these diverse political movements, and to provide normative argument to clarify the wrongs the term names.
Obviously the above-named groups are not oppressed to the same extent or in the same ways. In the most general sense, all oppressed people suffer-some inhibition of their ability to develop and exercise their capacities and express their needs, thoughts, and feelings. In that abstract sense all oppressed people face a common condition. Beyond that, in any more specific sense, it is not possible to define a single set of criteria that describe the condition of oppression of the above groups. Consequently, attempts by theorists and activists to discover a common description or the essential causes of the oppression of all these groups have frequently led to fruitless disputes about whose oppression is more fundamental or more grave. The contexts in which members of these groups use the term oppression to describe the injustices of their situation suggest that oppression names in fact a family of concepts and conditions, which I divide into five categories: exploitation, marginalization, powerlessness, cultural imperialism, and violence.
In this chapter I explicate each of these forms of oppression. Each may entail or cause distributive injustices, but all involve issues of justice beyond distribution. In accordance with ordinary political usage, I suggest that oppression is a condition of groups. Thus before explicating the meaning of oppression, we must examine the concept of a social group.
OPPRESSION AS A STRUCTURAL CONCEPT
One reason that many people would not use the term oppression to describe injustice in our society is that they do not understand the term in the same way as do new social movements. In its traditional usage, oppression means the exercise of tyranny by a ruling group. Thus many Americans would agree with radicals in applying the term oppression to the situation of Black South Africans under apartheid. Oppression also traditionally carries a strong connotation of conquest and colonial domination. The Hebrews were oppressed in Egypt, and many uses of the term oppression in the West invoke this paradigm.
Dominant political discourse may use the term oppression to describe societies other than our own, usually Communist or purportedly Communist societies. Within this anti-Communist rhetoric both tyrannical and colonialist implications of the term appear. For the anti-Communist, Communism denotes precisely the exercise of brutal tyranny over a whole people by a few rulers, and the will to conquer the world, bringing hitherto independent peoples under that tyranny. In dominant political discourse it is not legitimate to
use the term oppression to describe our society, because oppression is the evil perpetrated by the Others.
New left social movements of the 1960s and 1970s, however, shifted the meaning of the concept of oppression. In its new usage, oppression designates the disadvantage and injustice some people suffer not because a tyrannical power coerces them, but because of the everyday practices of a well-intentioned liberal society. In this new left usage, the tyranny of a ruling group over another, as in South Africa, must certainly be called oppressive. But oppression also refers to systemic constraints on groups that are not necessarily the result of the intentions of a tyrant. Oppression in this sense is structural, rather than the result of a few people's choices or policies. Its causes are embedded in unquestioned norms, habits, and symbols, in the assumptions underlying institutional rules and the collective consequences of following those rules. It names, as Marilyn Frye puts it, "an enclosing structure of forces and barriers which tends to the immobilization and reduction of a group or category of people" (Frye, 1983, p. 11). In this extended structural sense oppression refers to the vast and deep injustices some groups suffer as a consequence of often unconscious assumptions and reactions of well-meaning people in ordinary interactions, media and cultural stereotypes, and structural features of bureaucratic hierarchies and market mechanisms—in short, the normal processes of everyday life. We cannot eliminate this structural oppression by getting rid of the rulers or making some new laws, because oppressions are systematically reproduced in major economic, political, and cultural institutions.
The systemic character of oppression implies that an oppressed group need not have a correlate oppressing group. While structural oppression involves relations among groups, these relations do not always fit the paradigm of conscious and intentional oppression of one group by another. Foucault (1977) suggests that to understand the meaning and operation of power in modern society we must look beyond the model of power as "sovereignty," a dyadic relation of ruler and subject, and instead analyze the exercise of power as the effect of often liberal and "humane" practices of education, bureaucratic administration, production and distribution of consumer goods, medicine, and so on. The conscious actions of many individuals daily contribute to maintaining and reproducing oppression, but those people are usually simply doing their jobs or living their lives, and do not understand themselves as agents of oppression.
I do not mean to suggest that within a system of oppression individual persons do not intentionally harm others in oppressed groups.The raped woman, the beaten Black youth, the locked-out worker, the gay man harassed on the street, are victims of intentional actions by identifiable agents. I also do not mean to deny that specific groups are beneficiaries of the oppression of other groups, and thus have an interest in their continued oppression. Indeed, for every oppressed group there is a group that is privileged in relation to that group.
The concept of oppression has been current among radicals since the 1960s partly in reaction to Marxist attempts to reduce the injustices of racism
FIVE FACES OF OPPRESSION / IRIS YOUNG
Five Faces of Oppression    ■ 41
and sexism, for example, to the effects of class domination or bourgeois ideology. Racism, sexism, ageism, homophobia, some social movements asserted, are distinct forms of oppression with their own dynamics apart from the dynamics of class, even though they may interact with class oppression. From often heated discussions among socialists, feminists, and antiracism activists in the last ten years a consensus is emerging that many different groups must be said to be oppressed in our society, and that no single form of oppression can be assigned causal or moral primacy (see Gottlieb, 1987). The same discussion has also led to the recognition that group differences cut across individual lives in a multiplicity of ways that can entail privilege and oppression for the same person in different respects. Only a plural explication of the concept of oppression can adequately capture these insights.
Accordingly, I offer below an explication of five faces of oppression as a useful set of categories and distinctions which I believe is comprehensive, in the sense that it covers all the groups said by new left social movements to be oppressed and all the ways they are oppressed. I derive the five faces of oppression from reflection on the condition of these groups. Because different factors, or combinations of factors, constitute the oppression of different groups, making their oppression irreducible, I believe it is not possible to give one essential definition of oppression. The five categories articulated in this chapter, however, are adequate to describe the oppression of any group, as well as its similarities with and differences from the oppression of other groups. But first we must ask what a group is.
THE CONCEPT OF A SOCIAL GROUP
Oppression refers to structural phenomena that immobilize or diminish a group. But what is a group? Our ordinary discourse differentiates people according to social groups such as women and men, age groups, racial and ethnic groups, religious groups, and so on. Social groups of this sort are not simply collections of people, for they are more fundamentally intertwined with the identities of the people described as belonging to them. They are a specific kind of collectivity, with specific consequences for how people understand one .another and themselves. Yet neither social theory nor philosophy has a clear and developed concept of the social group (see Turner et al., 1987).
A social group is a collective of persons differentiated from at least one other group by cultural forms, practices, or way of life. Members of a group have a specific affinity with one another because of their similar experience or way of life, which prompts them to associate with one another more than with those not identified with the group, or in a different way. Groups are an expression of social relations; a group exists only in relation to at least one other group. Group identification arises, that is, in the encounter and interaction between social collectivities that experience some differences in their
way of life and forms of association, even if they regard themselves as belonging to the same society.
As long as they associated solely among themselves, for example, an American Indian group thought of themselves only as "the people." The encounter with other American Indians created an awareness of difference; the others were named as a group, and the first group came to see themselves as a group. But social groups do not arise only from an encounter between different societies. Social processes also differentiate groups within a single society. The sexual division of labor, for example, has created social groups of women and men in all known societies. Members of each gender have a certain affinity with others in their group because of what they do or experience, and differentiate themselves from the other gender, even when members of each gender consider that they have much in common with members of the other, and consider that they belong to the same society.
Political philosophy typically has no place for a specific concept of the social group. When philosophers and political theorists discuss groups, they tend to conceive them either on the model of aggregates or on the model of associations, both of which are methodologically individualist concepts. To arrive at a specific concept of the social group it is thus useful to contrast social groups with both aggregates and associations.
An aggregate is any classification of persons according to some attribute. Persons can be aggregated according to any number of attributes—eye color, the make of car they drive, the street they live on. Some people interpret the groups that have emotional and social salience in our society as aggregates, as arbitrary classifications of persons according to such attributes as skin color, genitals, or age. George Sher, for example, treats social groups as aggregates, and uses the arbitrariness of aggregate classification as a reason not to give special attention to groups. "There are really as many groups as there are combinations of people and if we are going to ascribe claims to equal treatment to racial, sexual, and other groups with high visibility, it will be more favoritism not to ascribe similar claims to these other groups as well" (Sher, 1987, p. 256).
But "highly visible" social groups such as Blacks or women arc different from aggregates, or mere "combinations of people" (see French, 1975; Friedman and May, 1985; May, 1987, chap. 1). A social group is defined not primarily by a set of shared attributes, but by a sense of identity. What defines Black Americans as a social group is not primarily their skin color; some persons whose skin color is fairly light, for example, identify themselves as Black. Though sometimes objective attributes are a necessary condition for classifying oneself or others as belonging to a certain social group, it is identification with a certain social status, the common history that social status produces, and self-identification that define the group as a group.
Social groups are not entities that exist apart from individuals, but neither 20 are they merely arbitrary classifications of individuals according to attributes which are external to or accidental to their identities. Admitting the reality of
2 FIVE FACES OF OPPRESSION / IRIS YOUNG
Five Faces of Oppression    a 43
social groups does riot commit one to reifying collectivities, as some might argue. Group meanings partially constitute people's identities in terms of the cultural forms, social situation, and history that group members know as theirs, because these meanings have been either forced upon them or forged by them or both (cf. Fiss, 1976). Groups are real not as substances, but as forms of social relations (cf. May, 1987, pp. 22-23).
Moral theorists and political'philosophers tend to elide social groups more often with associations than with aggregates (e.g., French, 1975; May, 1987, chap. 1). By an association I mean a formally organized institution, such as a club, corporation, political party, church, college, or union. Unlike the aggregate model of groups, the association model recognizes that groups are defined by specific practices and forms of association. Nevertheless it shares a problem with the aggregate model. The aggregate model conceives the individual as prior to the collective, because it reduces the social group to a mere set of attributes attached to individuals. The association model also implicitly conceives the individual as ontologically prior to the collective, as making up, or constituting, groups.
A contract model of social relations is appropriate for conceiving associations, but not groups. Individuals constitute associations, they come together as already formed persons and set them up, establishing rules, positions and offices. The relationship of persons to associations is usually voluntary, and even when it is not, the person has nevertheless usually entered the association. The person is prior to the association also in that the person's identity and sense of self are usually regarded as prior to and relatively independent of association membership.
Groups, on the other hand, constitute individuals. A person's particular sense of history, affinity, and separateness, even the person's mode of reasoning, evaluating, and expressing feeling, are constituted partly by her or his group affinities.This does not mean that persons have no individual styles, or are unable to transcend or reject a group identity. Nor does it preclude persons from having many aspects that are independent of these group identities.
The social ontology underlying many contemporary theories of justice is methodologically individualist or atomist: It presumes that the individual is ontologically prior to the social. This individualist social ontology usually goes together with a normative conception of the self as independent. The authentic self is autonomous, unified, free, and self-made, standing apart from history and affiliations, choosing its life plan entirely for itself. 25 One of the main contributions of poststructurahst philosophy has been
to expose as illusory this metaphysic of a unified self-making subjectivity, which posits the subject as an autonomous origin or an underlying substance to which attributes of gender, nationality, family role, intellectual disposition, and so on might attach. Conceiving the subject in this fashion implies conceiving consciousness as outside of and prior to language and the context of social interaction, which the subject enters. Several currents of recent philosophy challenge this deeply held Cartesian assumption. Lacaman psychoanaly-
sis, for example, and the social and philosophical theory influenced by it, conceive the self as an achievement of linguistic positioning that is always con-textualized in concrete relations with other persons, with their mixed identities (Coward and Ellis, 1977). The self is a product of social processes, not their origin. .
From a rather different perspective, Habermas indicates that theory of communicative action also must challenge the "philosophy of consciousness" which locates intentional egos as the ontological origins of social relations. A theory of communicative action conceives individual identity not as an origin but as a product of linguistic and practical interaction (Habermas, 1987, pp. 3-40). As Stephen Epstein describes it, identity is a "socialized sense of individuality, an internal organization of self-perception concerning one's relationship to social categories, that also incorporates views of the self perceived to be held by others. Identity is constituted relationally, through involvement with—and incorporation of—significant others and integration into communities" (Epstein, 1987, p. 29). Group categorization and norms are major constituents of individual identity (see Turner et al., 1987).
A person joins an association, and even if membership in it fundamentally affects one's life, one does not take that membership to define one's very identity, in the. way, for example, being Navaho might. Group affinity, on the other hand, has the character of what Martin Heidegger (1962) calls "thrown-ness": one finds oneself as a member of a group, which one experiences as always already having been. For our identities are defined in relation to how others identify us, and they do so in terms of groups which are always already associated with specific attributes, stereotypes, and norms.
From the thrownness of group affinity it does not follow that one cannot leave groups and enter new ones. Many women become lesbian after first identifying as heterosexual. Anyone who lives long enough becomes old. These cases exemplify thrownness precisely because such changes in group affinity are experienced as transformations in one's identity. Nor does it follow from the thrownness of group affinity that one cannot define the meaning of group identity for oneself; those who identify with a group can redefine the meaning and norms of group identity. The present point is only that one first finds a group identity as given, and then takes it up in a certain way. While groups may come into being, they are never founded.
Groups, I have said, exist only in relation to other groups. A group may be identified by outsiders without those so identified having any specific consciousness of themselves as a group. Sometimes a group comes to exist only because one group excludes and labels a category of persons, and those labeled come to understand themselves as group members only slowly, on the basis of their shared oppression. In Vichy France, for example, Jews who had been so assimilated that they had no specifically Jewish identity were marked as Jews by others and given a specific social status by them. These people "discovered" themselves as Jews, and then formed a group identity and affinity with one another (see Sartre, 1948). A person's group identities may be for
■    FIVE FACES OF OPPRESSION / IRIS YOUNG
the most part only a background or horizon to his or her life, becoming salient only in specific interactive contexts.
Assuming an aggregate model of groups, some people think that social groups are invidious fictions, essentializing arbitrary attributes. From this point of view problems of prejudice, stereotyping, discrimination, and exclusion exist because some people mistakenly believe that group identification makes a difference to the capacities, temperament, or virtues of group members. This individualist conception of persons and their relation to one another tends to identify oppression with group identification. Oppression, on this view, is something that happens to people when they are classified in groups. Because others identify them as a group, they are excluded and despised. Eliminating oppression thus requires ehminating groups. People should be treated as individuals, not as members of groups, and allowed to form their lives freely without stereotypes or group norms.
This book takes issue with that position. While I agree that individuals should be free to pursue life plans in their own way, it is foolish to deny the reality of groups. Despite the modern myth of a decline of parochial attachments and ascribed identities, in modern society group differentiation remains endemic. As both markets and social administration increase the web of social iriterdependency on a world scale, and as more people encounter one another as strangers in cities and states, people retain and renew ethnic, locale, age, sex, and occupational group identifications, and form new ones in the processes of encounter (cf. Ross, 1980, p. 19; Rothschild, 1981, p. 130). Even when they belong to oppressed groups, people's group identifications are often important to them, and they often feel a special affinity for others in their group. I believe that group differentiation is both an inevitable and a desirable aspect of modern social processes. Social justice, I shall argue in later chapters, requires not the melting away of differences, but institutions that promote reproduction of and respect for group differences without oppression.
Though some groups have come to be formed out of oppression, and relations of privilege and oppression structure the interactions between many groups, group differentiation is not in itself oppressive. Not all groups are oppressed. In the United States Roman Catholics are a specific social group, with distinct practices and affinities with one another, but they are no longer an oppressed group. Whether a group is oppressed depends on whether it is subject to one or more of the five conditions I shall discuss below.
The view that groups are fictions does carry an important antidetermin-ist or antiessentialist intuition. Oppression has often been perpetrated by a conceptualization of group difference in terms of unalterable essential natures that determine what group members deserve or are capable of, and that exclude groups so entirely from one another that they have no similarities or overlapping attributes. To assert that it is possible to have social group difference without oppression, it is necessary to conceptualize groups in a much more relational and fluid fashion.
Rve Faces of Oppression ■
Although social processes of affinity and differentiation produce groups, they do not give groups a substantive essence.There is no common nature that members of a group share. As aspects of a process, moreover, groups are fluid; they come into being and may fade away. Homosexual practices have existed in many societies and historical periods, for example. Gay men or lesbians have been identified as specific groups and so identified themselves, however, only in the twentieth century (see Ferguson, 1989, chap. 9; Altman, 1981).
Arising from social relations and processes, finally, group differences usually cut across one another. Especially in a large, complex, and highly differentiated society, social groups are not themselves homogeneous; but mirror in their own differentiations many of the other groups in the wider society. In American society today, for example, Blacks are not a simple, unified group with a common life. Like other racial and ethnic groups, they are differentiated by age, gender, class, sexuality, region, and nationality, any of which in a given context may become a salient group identity.
This view of group differentiation as multiple, cross-cutting, fluid, and shifting implies another critique of the model of the autonomous, unified self. In complex, highly differentiated societies like our own, all persons have multiple group identifications. The culture, perspective, and relations of privilege and oppression of these various groups, moreover, may not cohere. Thus individual persons, as constituted partly by their group affinities and relations, cannot be unified, themselves are heterogeneous and not necessarily coherent.
THE FACES OF OPPRESSION
Exploitation
The central function of Marx's theory of exploitation is to explain how class structure can exist in the absence of legally and normatively sanctioned class distinctions. In precapitalist societies domination is overt and accomplished through directly political means. In both slave society and feudal society the right to appropriate the product of the labor of others partly defines class privilege, and these societies legitimate class distinctions with ideologies of natural superiority and inferiority.
Capitalist society, on the other hand, removes traditional juridically enforced class distinctions and promotes a belief in the legal freedom of persons. Workers freely contract with employers and receive a wage; no formal mechanisms of law or custom force them to work for that employer or any employer. Thus the mystery of capitalism arises: when everyone is formally free, how can there be class domination? Why do class distinctions persist between the wealthy, who own the means of production, and the mass of people, who work for them? The theory of exploitation answers this question.
Profit, the basis of capitalist power and wealth, is a mystery if we assume that in the market goods exchange at their values. The labor theory of value
■    FIVE FACES OF OPPRESSION / IRIS YOUNG
dispels this mystery. Every commodity's value is a function of the labor time necessary for its production. Labor power is the one commodity which in the process of being consumed produces new value. Profit comes from the difference between the value of the labor performed and the value of the capacity to labor which the capitalist purchases. Profit is possible only because the owner of capital appropriates any realized surplus value.
In recent years Marxist scholars have engaged in considerable controversy about the viability of the labor theory of value this account of exploitation relies on (seeWolff, 1984, chap. 4).John Roemer (1982), for example, develops a theory of exploitation which claims to preserve the theoretical and practical purposes of Marx's theory, but without assuming a distinction between values and prices and without being restricted to a concept of abstract, homogeneous labor. My purpose here is not to engage in technical economic disputes, but to indicate the place of a concept of exploitation in a conception of oppression.
Marx's theory of exploitation lacks an explicitly normative meaning, even though the judgment that workers are exploited clearly has normative as well as descriptive power in that theory (Buchanan, 1982, chap. 3). C. B. Macpher-son (1973, chap. 3) reconstructs this theory of exploitation in a more explicitly normative form. The injustice of capitalist society consists in the fact that some people exercise their capacities under the control, according to the purposes, and for the benefit of other people. Through private ownership of the means of production, and through markets that allocate labor and the ability to buy goods, capitalism systematically transfers the powers of some persons to others, thereby augmenting the power of the latter. In this process of the transfer of powers, according to Macpherson, the capitalist class acquires and maintains an ability to extract benefits from workers. Not only are powers transferred from workers to capitalists, but also the powers of workers diminish by more than the amount of transfer, because workers suffer material deprivation and a loss of control, and hence are deprived of important elements of self-respect. Justice, then, requires eliminating the institutional forms that enable and enforce this process of transference and replacing them with institutional forms that enable all to develop and use their capacities in a way that does not inhibit, but rather can enhance, similar development and use in others.
The central insight expressed in the concept of exploitation, then, is that this oppression occurs through a steady process of the transfer of the results of this labor of one social group to benefit another. The injustice of class division does not consist only in the distributive fact that some people have great wealth while most people have little (cf. Buchanan, 1982, pp. 44-49; Holm-strom 1977). Exploitation enacts a structural relation between social groups. Social rules about what work is, who does what for whom, how work is compensated, and the social process by which the results of work are appropriated operate to enact relations of power and inequality. These relations are produced and reproduced through a systematic process in which the energies of the have-nots are continuously expended to maintain and augment the power, status, and wealth of the haves.
Five Faces of Oppression ■
Many writers have cogendy argued that the Marxist concept of exploitation is too narrow to encompass all forms of domination and oppression (Gid-dens, 1981, p. 242; Brittan and Maynard, 1984, p. 93; Murphy, 1985; Bowles and Gintis, 1986, pp. 20-24). In particular, the Marxist concept of class leaves important phenomena of sexual and racial oppression unexplained. Does this mean that sexual and racial oppression are nonexploitative, and that we should reserve wholly distinct categories for these oppressions? Or can the concept of exploitation be broadened to include other ways in which the labor and energy expenditure of one group benefits another, and reproduces a relation of domination between them?
Feminists have had little difficulty showing that women's oppression consists partly in a systematic and unreciprocated transfer of powers from women to men. Women's oppression consists not merely in an inequality of status, power, and wealth resulting from men's excluding them from privileged activities. The freedom, power, status, and self-realization of men is possible precisely because women work for them. Gender exploitation has two aspects, transfer of the fruits of material labor to men and transfer of nurturing and sexual energies to men.
Christine Delphy (1984), for example, describes marriage as a class relation in which women's labor benefits men without comparable remuneration. She makes it clear that the exploitation consists not in the sort of work that women do in the home, for this might include various kinds of tasks, but in the fact that they perform tasks for someone on whom they are dependent. Thus, for example, in most systems of agricultural production in the world, men take to market the goods women have produced, and more often than not men receive the status and often the entire income from this labor.
With the concept of sex-affective production, Ann Ferguson (1979; 1984; 1989, chap. 4) identifies another form of the transference of women's energies to men. Women provide men and children with emotional care and provide men with sexual satisfaction, and as a group receive relatively little of either from men (cf. Brittan and Maynard, pp. 142-48).The gender socialization of women makes us tend to be more attentive to interactive dynamics than men, and makes women good at providing empathy and support for people's feelings and at smoothing over interactive tensions. Both men and women look to women as nurturers of their personal lives, and women frequently complain that when they look to men for emotional support they do not receive it (Easton, 1978).The norms of heterosexuality, moreover, are oriented around male pleasure, and consequently many women receive little satisfaction from their sexual interaction with men (Gottlieb, 1984).
Most feminist theories of gender exploitation have concentrated on the institutional structure of the patriarchal family. Recently, however, feminists have begun to explore relations of gender exploitation enacted in the contemporary workplace and through the state. Carol Brown argues that as men have removed themselves from responsibility for children, many women have become dependent on the state for subsistence as they continue to bear nearly total responsibility for childrearing (Brown, 1981; cf. Boris and Bardaglio,
■    FIVE FACES OF OPPRESSION / IRIS YOUNG
1983; A. Ferguson, 1984).This creates a new system of the exploitation of women's domestic labor mediated by state institutions, which she calls public patriarchy.
In twentieth-century capitalist economies the workplaces that women have been entering in increasing numbers serve as another important site of gender exploitation. David Alexander (1987) argues that typically feminine jobs involve gender-based tasks requiring sexual labor, nurturing, caring for others' bodies, or smoothing over workplace tensions. In these ways women's energies are expended in jobs that enhance the status of, please, or comfort others, usually men; and these gender-based labors of waitresses, clerical workers, nurses, and other caretakers often go unnoticed and undercompensated.
To summarize, women are exploited in the Marxist sense to the degree that they are wage workers. Some have argued that women's domestic labor also represents a form of capitalist class exploitation insofar as it is labor covered by the wages a family receives. As a group, however, women undergo specific forms of gender exploitation in which their energies and power are expended, often unnoticed and unacknowledged, usually to benefit men by releasing them for more important and creative work, enhancing tiieir status or the environment around them, or providing them with sexual or emotional service.
o Race is a structure of oppression at least as basic as class or gender. Are there, then, racially specific forms of exploitation? There is no doubt that racialized groups in the United States, especially Blacks and Latinos, are oppressed through capitalist superexploitation resulting from a segmented labor market that tends to reserve skilled, high-paying, unionized jobs for whites. There is wide disagreement about whether such superexploitation benefits whites as a group or only benefits the capitalist class (see Reich, 1981), and I do not intend to enter into that dispute here.
However one answers the question about capitalist superexploitation of racialized groups, is it possible to conceptualize a form of exploitation that is racially specific on analogy with the gender-specific forms just discussed? I suggest that the category of menial labor might supply a means for such conceptualization. In its derivation "menial" designates the labor of servants. Wherever there is racism, there is the assumption, more or less enforced, that members of the oppressed racial groups are or ought to be servants of those, or some of those, in the privileged group. In most white racist societies this means that many white people have dark- or yellow-skinned domestic servants, and in the United States today there remains significant racial structuring of private household service. But in the United States today much service labor has gone public: anyone who goes to a good hotel or a good restaurant can have servants. Servants often attend the daily—and nightly—activities of business executives, government officials, and other high-status professionals. In our society there remains strong cultural pressure to fill servant jobs—bellhop, porter, chambermaid, busboy, and so on—with Black and Latino workers. These jobs entail a transfer of energies whereby the servers enhance the status of the served.
Five Faces of Oppression b
Menial labor usually refers not only to service, however, but also to any servile, unskilled, low-paying work lacking in autonomy, in which a person is subject to taking orders from many people. Menial work tends to be auxiliary-work, instrumental to the work of others, where those others receive primary recognition for doing the job. Laborers on a construction site for example are at the beck arid call of welders, electricians, carpenters, and other skilled workers, who receive recognition for the job done. In the United States explicit racial discriminatiorl once reserved menial work for Blacks, Chicanos, American Indians, and Chinese, and menial work still tends to be linked to Black and Latino workers (Symanski, 1985). I offer this category of menial labor as a form of racially specific exploitation, as a provisional category in need of exploration.
The injustice of exploitation is most frequently understood on a distributive model. For example, though he does not offer an explicit definition of the concept, by "exploitation" Bruce Ackerman seems to mean a seriously unequal distribution of wealth, income, and other resources that is group based and structurally persistent (Ackerman, 1980, chap. 8). John Roemer's definition is narrower and more rigorous: "An agent is exploited when the amount of labor embodied in any bundle of goods he could receive, in a feasible distribution of society's net product, is less than the labor he expended" (Roemer, 1982, p. 122). This definition too turns the conceptual focus from institutional relations and processes to distributive outcomes.
Jeffrey Reiman argues that such a distributive understanding of exploitation reduces the injustice of class processes to a function of the inequality of the productive assets classes own. This misses, according to Reiman, the relationship offeree between capitalists and workers, the fact that the unequal exchange in question occurs within coercive structures that give workers few options (Reiman, 1987; cf. Buchanan, 1982, pp. 44-49; Holmstrom, 1977). The injustice of exploitation consists in social processes that bring about a transfer of energies from one group to another to produce unequal distributions, and in the way in which social institutions enable a few to accumulate while they constrain many more. The injustices of exploitation cannot be eliminated by redistribution of goods, for as long as institutionalized practices and structural relations remain unaltered, the process of transfer will re-create an unequal'distribution of benefits. Bringing about justice where there is exploitation requires reorganization of institutions and practices of decisionmaking, alteration of the division of labor, and similar measure of institutional, structural, and cultural change.
Marginalization
increasingly in the United States racial oppression occurs in the form of marginalization rather than exploitation. Marginals are people the system of labor cannot or will not use. Not only in Third World capitalist countries, but also in most Western capitalist societies, there is a growing underclass of people permanently confined to lives of social marginahty, most of whom are racially
FIVE FACES OF OPPRESSION / IRIS YOUNG
Five Faces of Oppression
marked—Blacks or Indians in Latin America, and Blacks, East Indians, Eastern Europeans, or North Africans in Europe.
Marginalization is by no means the fate only of racially marked groups, however. In the United States a shamefully large proportion of the population is marginal: old people, and increasingly people who are not very old but get laid off from their jobs and cannot find new work: young people, especially Black oř Latino, who cannot find first or second jobs; many single mothers and their children; other people involuntarily unemployed; many mentally and physically disabled people; American Indians, especially those on reservations.
Marginalization is perhaps the most dangerous form of oppression. A whole category of people is expelled from useful participation in social life and thus potentially subjected to severe material deprivation and even extermination. The material deprivation marginalization often causes is certainly unjust, especially in a society where others háve plenty. Contemporary advanced capitalist societies have in principle acknowledged the injustice of material deprivation caused by marginalization, and have taken some steps to address it by providing welfare payments and services. The continuance of this welfare state is by no means assured, and in most welfare/State societies, especially the United States, welfare redistributions do not eliminate large-scale suffering and deprivation.
Material deprivation, which can be addressed by redistributive social policies, is not, however, the extent of the harm caused by marginalization. Two categories of injustice beyond distribution are associated with marginality in advanced capitalist societies. First, the provision of welfare itself produces new injustice by depriving those dependent on it of rights and freedoms that others have. Second, even when material deprivation is somewhat mitigated by the welfare state, marginalization is unjust because it blocks the opportunity to exercise capacities in socially defined and recognized ways. I shall explicate each of these in turn.
Liberalism has traditionally asserted the right of all rational autonomous agents to equal citizenship. Early bourgeois liberalism explicitly excluded from citizenship all those whose reason was questionable or not fully developed, and all those not independent (Pateman, 1988, chap. 3; cf. Bowles and Gintis, 1986, chap. 2). Thus poor people, women, the mad and the feebleminded, and children were explicitly excluded from citizenship, and many of these were housed in institutions modeled on the modern prison: poorhouses, insane asylums, schools.
Today the exclusion of dependent persons from equal citizenship rights is only barely hidden beneath the surface. Because they depend on bureaucratic institutions for support or services, the old, the poor, and the mentally or physically disabled are subject to patronizing, punitive, demeaning, and arbitrary treatment by the policies and people associated with welfare bureaucracies. Being a dependent in our society implies being legitimately subject to the often arbitrary and invasive authority of social service providers and other public and private administrators, who enforce rules with which the marginal must comply, and otherwise exercise power over the conditions of their lives.
In meeting needs of the marginalized, often with aid of social scientific disciplines, welfare agencies also construct the needs themselves. Medical and social service professionals know what is good for those they serve, and the marginals and dependents themselves do not have the right to claim to know what is good for them (Fraser, 1987a; K. Ferguson, 1984, chap. 4). Dependency in our society thus implies, as it has in all liberal societies, a sufficient Warrant to suspend basic rights to privacy, respect, and individual choice.
Although dependency produces conditions of injustice in our society, dependency in itself need not be oppressive. One cannot imagine a society in which some people would not need to be dependent on others at least some of the time:chtidren, sick people, women recovering from childbirth, old people who have become frail, depressed or otherwise emotionally needy persons, have the moral right to depend on others for subsistence and support.
An important contribution of feminist moral theory has been to question the deeply held assumption that moral agency and full citizenship require that a person be autonomous and independent. Feminists have exposed this assumption as inappropriately individualistic and derived from a specifically male experience of social relations, which values competition and solitary achievement (see Gilligan, 1982; Friedman, 1985). Female experience of social relations, arising both from women's typical domestic care responsibilities and from the kinds of paid work that many women do, tends to recognize dependence as a basic human condition (cf. Hartsock, 1983, chap. 10). Whereas on the autonomy model a just society would as much as possible give people the opportunity to be independent, the feminist model envisions justice as according respect and participation in decisionmaking to those who are dependent as well as to those who are independent (Held, 1987). Dependency should not be a reason to be deprived of choice and respect, and much of the oppression many marginals experience would be lessened if a less individualistic model of rights prevailed.
Marginalization does not cease to be oppressive when one has shelter and food. Many old people, for example, have sufficient means to live comfortably but remain oppressed in their marginal status. Even if marginals were provided a comfortable material life within institutions that respected their freedom and dignity, injustices of-marginality would remain in the form of uselessness, boredom, and lack of self-respect. Most of our society's productive and recognized activities take place in contexts of organized social cooperation, and social structures and processes that close persons out of participation in such social cooperation are unjust. Thus while marginalization definitely entails serious issues of distributive justice, it also involves the deprivation of cultural, practical, and institutionalized conditions for exercising capacities in a context of recognition and interaction.
The fact of marginalization raises basic structural issues of justice, in particular concerning the appropriateness of a connection between participation in productive activities of social cooperation, on the one hand, and access to the means of consumption, on the other. As marginalization is increasing, with no sign of abatement, some social policy analysts have introduced the
12    ■    FIVE FACES OF OPPRESSION / IRIS YOUNG
idea of a "social wage" as a guaranteed socially provided income not tied to the wage system. Restructuring of productive activity to address a right of participation, however, implies organizing some socially productive activity outside of the wage system (see Offe, 1985, pp. 95-100), through public works or self-employed collectives.
Powerlessness
65 As I have indicated, the Marxist idea of class is important because it helps reveal the structure of exploitation: that some people have their power and wealth because they profit from the labor of others. For this reason I reject the claim some make that a traditional class exploitation model fails to capture the structure of contemporary society. It remains the case that the labor of most people in the society augments the power of relatively few. Despite their differences from nonprofessional workers, most professional workers are still not members of the capitalist class. Professional labor either involves exploitative transfers to capitalists or supplies important conditions for such transfers. Professional workers are in an ambiguous class position, it is true, because they also benefit from the exploitation of nonprofessional workers.
While it is false to claim that a division between capitalist and working classes no longer describes our society, it is also false to say that class relations have remained unaltered since the nineteenth century. An adequate conception of oppression cannot ignore the experience of social division reflected in the colloquial distinction between the ''middle class" and the "working class," a division structured by the social division of labor between professionals and nonprofessionals. Professionals are privileged in relation to nonprofessionals, by virtue of their position in the division of labor and the status it carries. Nonprofessionals suffer a form of oppression in addition to exploitation, which I call powerlessness.
In the United States, as in other advanced capitalist countries, most workplaces are not organized democratically, direct participation in public policy decisions is rare, and policy implementation is for the most part hierarchical, imposing rules on bureaucrats and citizens. Thus most people in these societies do not regularly participate in making decisions that affect the conditions of their lives and actions, and in this sense most people lack significant power. At the same time, domination in modern society is enacted through the widely dispersed powers of many agents mediating the decisions of others. To that extent many people have some power in relation to others, even though they lack the power to decide policies or results. The powerless are those who lack authority or power even in this mediated sense, those over whom power is exercised without their exercising it; the powerless are situated so that they must take orders and rarely have the right to give them. Powerlessness also designates a position in the division of labor and the concomitant social position that allow persons little opportunity to develop and exercise skills.The powerless have little or no work autonomy, exercise little creativity or judgment in
Five Faces of Oppression
53
their work, have no technical expertise or authority, express themselves awkwardly, especially in public or bureaucratic settings, and do not command respect. Powerlessness names the oppressive situations Sennett and Cobb (1972) describe in their famous study of working-class men.
This powerless status is perhaps best described negatively: the powerless lack the authority, status, and sense of self that professionals tend to have. The status privilege of professionals has three aspects, the lack of which produces oppression for nonprofessionals.
First, acquiring and practicing a profession has an expansive, progressive character. Being professional usually requires a college education and the acquisition of a specialized knowledge that entails working with symbols and concepts. Professionals experience progress first in acquiring the expertise, and then in the course of professional advancement and the rise in status. The life of the nonprofessional by comparison is powerless in the sense that it lacks this orientation toward the progressive development of capacities and avenues for recognition.
Second, while many professionals have supervisors and cannot directly io influence many decisions or the actions of many people, most nevertheless have considerable day-to-day work autonomy. Professionals usually have some authority over others, moreover—either over workers they supervise, or over auxiliaries, or over clients. Nonprofessionals, on the other hand, lack autonomy, and in both their working and their consumer—client lives often stand under the authority of professionals.
Though based on a division of labor between "mental" and "manual" work, the distinction between "middle class" and "working class" designates a division not ony in working life, but also in nearly all aspects of social life. Professionals and nonprofessionals belong to different cultures in the United States. The two groups tend to live in segregated neighborhoods or even different towns, a process itself mediated by planners, zoning officials, and real estate people. The groups tend to have different tastes in food, decor, clothes, music, and vacations, and often different health and educational needs. Members of each group socialize for the most part with others in the same status group. While there is some intergroup mobility between generations, for the most part the children of professionals become professionals and the children of nonprofessionals do not.
Thus, third, the privileges of the professional extend beyond the workplace to a whole way of life. I call this way of life "respectability." To treat people with respect is to be prepared to listen to what they have to say or to do what they request because they have some authority, expertise, or influence. The norms of respectability in our society are associated specifically with professional culture. Professional dress, speech, tastes, demeanor, all connote respectability. Generally professionals expect and receive respect from others. In restaurants, banks, hotels, real estate offices, and many other such public places, as well as in the media, professionals typically receive more respectful treatment than nonprofessionals. For this reason nonprofessionals
■    FIVE FACES OF OPPRESSION / IRIS YOUNG
seeking a loan or a job, or to buy a house or a car, will often try to look "professional" and "respectable" in those settings.
The privilege of this professional respectability appears starkly in the dynamics of racism and sexism. In daily interchange women and men of color must prove their respectability. At first they are often not treated by strangers with respectful distance or deference. Once people discover that this woman or that Puerto Rican man is a college teacher or a business executive, however, they often behave more respectfully toward her or him. Working-class white men, on the other hand, are often treated with respect until their working-class status is revealed.
I have discussed several injustices associated with powerlessness: inhibition in the development of one's capacities, lack of decisionmaking power in one's working life, and exposure to disrespectful treatment because of the status one occupies. These injustices have distributional consequences, but are more fundamentally matters of the division of labor. The oppression of powerlessness brings into question the division of labor basic to all industrial societies: the social division between those who plan and those who execute.
Cultural Imperialism
Exploitation, marginalization, and powerlessness all refer to relations of power and oppression that occur by virtue of the social division of labor—who works for whom, who does not work, and how the content of work defines one institutional position relative to others. These three categories refer to structural and institutional relations that delimit people's material lives, including but not restricted to the resources they have access to and the concrete opportunities they have or do not have to develop and exercise their capacities. These kinds of oppression are a matter of concrete power in relation to others—of who benefits from whom, and who is dispensable.
Recent theorists of movements of group liberation, notably feminist and Black liberation theorists, have also given prominence to a rather different form of oppression, which following Lugones and Spelman (1983) I shall call cultural imperialism. To experience cultural imperialism means to experience how the dominant meanings of a society render the particular perspective of one's own group invisible at the same time as they stereotype one's group and mark it out as the Other.
Cultural imperialism involves the universalization of a dominant group's experience and culture, and its establishment as the norm. Some groups have exclusive or primary access to what Nancy Fraser (1987b) calls the means of interpretation and communication in a society. As a consequence, the dominant cultural products of the society, that is, those most widely disseminated, express the experience, values, goals, and achievements of these groups. Often without noticing they do so, the dominant groups project their own experience as a representative of humanity as such. Cultural products also express the dominant group's perspective on and interpretation of events and ele-
Five Faces of Oppression ■
ments in the society, including other groups in the society, insofar as they attain cultural status at all.
An encounter with other groups, however, can challenge the dominant group's claim to universality. The dominant group reinforces its position by bringing the other groups under the measure of its dominant norms. Consequently, the difference of women from men, American Indians or Africans from Europeans,Jews from Christians, homosexuals from heterosexuals, workers from professionals, becomes reconstructed largely as deviance and inferiority. Since only the dominant group's cultural expressions receive wide dissemination, their cultural expressions become the normal, or the universal, and thereby the unremarkable. Given the normality of its own cultural expressions and identity, the dominant group constructs the differences which some groups exhibit as lack and negation. These groups become marked as Other.
The culturally dominated undergo a paradoxical oppression, in that they are both marked by stereotypes and at the same time rendered invisible. As remarkable, deviant beings, the culturally imperialized are stamped with an essence. The stereotypes confine them to a nature which is often attached in some way to their bodies, and which thus cannot easily be denied. These stereotypes so permeate the society that they are not noticed as contestable. Just as everyone knows that the earth goes around the sun, so everyone knows that gay people are promiscuous, that Indians are alcoholics, and that women are good with children. White males, on the other hand, insofar as they escape group marking, can be individuals.
Those living under cultural imperialism find themselves defined from the outside, positioned, placed, by a network of dominant meanings they experience as arising from elsewhere, from those with whom they do not identify and who do not identify with them. Consequently, the dominant culture's stereotyped and inferiorized images of the group must be internalized by group members at least to the extent that they are forced to react to behavior of others influenced by those images. This creates for the culturally oppressed the experience that W. E. B. Du Bois called "double consciousness"—"this sense of always looking at one's self through the eyes of others, of measuring one's soul by the tape of a world that looks on in amused contempt and pity" (Du Bois, 1969 [1903],p. 45). Double consciousness arises when the oppressed subject refuses to coincide with these devalued, objectified, stereotyped visions of herself or himself. While the subject desires recognition as human, capable of activity, full of hope and possibility, she receives from the dominant culture only the judgment that she is different, marked, or inferior.
The group defined by the dominant culture as deviant, as a stereotyped Other, is culturally different from the dominant group, because the status of Otherness creates specific experiences not shared by the dominant group, and because culturally oppressed groups also are often socially segregated and occupy specific positions in the social division of labor. Members of such groups express their specific group experiences and interpretations of the world to
■    FIVE FACES OF OPPRESSION / IRIS YOUNG
one another, developing and perpetuating their own culture. Double consciousness, then, occurs because one finds one's being defined by two cultures: a dominant and a subordinate culture. Because they can affirm and recognize one another as sharing similar experiences and perspectives on social life, people in culturally imperialized groups can often maintain a sense of positive subjectivity.
Cultural imperialism involves the paradox of experiencing oneself as invisible at the same time that one is marked out as different. The invisibility comes about when dominant groups fail to recognize the perspective embodied in their cultural expressions as a perspective. These dominant cultural expressions often simply have little place for the experience of other groups, at most only mentioning or referring to them in stereotyped or marginalized ways. This, then is the injustice of cultural imperialism: that the oppressed group's own experience and interpretation of social life finds little expression that touches the dominant culture, while that same culture imposes on the oppressed group its experience and interpretation of social life.
/
Violence /
Finally, many groups suffer the oppression of systematic violence. Members of some groups live with the knowledge that they must fear random, unprovoked attacks on their persons or property, which have no motive but to damage, humiliate, or destroy the person. In American society women, Blacks, Asians, Arabs, gay men, and lesbians live under such threats of violence, and in at least some regions Jews, Puerto Paeans, Chicanos, and other Spanish-speaking Americans must fear such violence as well. Physical violence against these groups is shockingly frequent. Rape Crisis Center networks estimate that more than one-third of all American women experience an attempted or successful sexual assault in their lifetimes. Manning Marable (1984, pp. 238—41) catalogues a large number of incidents of racist violence and terror against Blacks in the United States between 1980 and 1982. He cites dozens of incidents of the severe beating, killing, or rape of Blacks by police officers on duty, in which the police involved were acquitted of any wrongdoing. In 1981, moreover, there were at least five hundred documented cases of random white teenage violence against Blacks.Violence against gay men and lesbians is not only common, but has been increasing in the last five years. While the frequency of physical attack on members of these and other racially or sexually marked groups is very disturbing, I also include in this category less severe incidents of harassment, intimidation, or ridicule simply for the purpose of degrading, humiliating, or stigmatizing group members.
Given the frequency of such violence in our society, why are theories of justice usually silent about it? I think the reason is that theorists do not typically take such incidents of violence and harassment as matters of social injustice. No moral theorist would deny that such acts are very wrong. But unless all immoralities are injustices, they might wonder, why should such acts be
Five Faces of Oppression B
interpreted as symptoms of social, injustice? Acts of violence or petty harassment are committed by particular individuals, often extremists, deviants, or the mentally unsound. How then can they be said to involve the sorts of institutional issues I have said are properly the subject of justice?
What makes violence a face of oppression is less the particular acts themselves, though these are often utterly horrible, than the social context surrounding them, which makes them possible and even acceptable. What makes violence a phenomenon of social injustice, and not merely an individual moral wrong, is its systemic character, its existence as a social practice.
Violence is systemic because it is directed at members of a group simply because they are members of that group. Any woman, for example, has a reason to fear rape. Regardless of what a Black man has done to escape the oppressions of marginality of powerlessness, he hves knowing he is subject to attack or harassment. The oppression of violence consists not only in direct victimization, but in the daily knowledge shared by all members of oppressed groups that they are liable to violation, solely on account of their group identity. Just living under such a threat of attack on oneself or family or friends deprives the oppressed of freedom and dignity, and needlessly expends their energy.
Violence is a social practice. It is a social given that everyone knows happens and will happen again. It is always at the horizon of social imagination, even for those who do not perpetrate it. According to the prevailing social logic, some circumstances make such violence more "called for" than others. The idea of rape will occur to many men who pick up a hitch-hiking woman; the idea of hounding or teasing a gay man on their dorm floor will occur to many straight male college students. Often several persons inflict the violence together, especially in all-male groupings. Sometimes violators set out looking for people to beat up, rape, or taunt. This rule-bound, social, and often premeditated character makes violence against groups a social practice.
Group violence approaches legitimacy, moreover, in the sense that it is tolerated. Often third parties find it unsurprising because it happens fre-quendy and lies as a constant possibility at the horizon of the social imagination. Even when they are caught, those who perpetrate acts of group-directed violence or harassment often receive light or no punishment. To that extent society renders their acts acceptable.
An important aspect of random, systematic violence is its irrationality. Xenophobic violence differs from the violence of states or ruling-class repression. Repressive violence has a rational, albeit evil, motive: rulers use it as a coercive tool to maintain their power. Many accounts of racist, sexist, or homophobic violence attempt to explain its motivation as a desire to maintain group privilege or domination. I do not doubt that fear of violence often functions to keep oppressed groups subordinate, but I do not think xenophobic violence is rationally motivated in the way that, for example, violence against strikers is.
■    FIVE FACES OF OPPRESSION / IRIS YOUNG
Five Faces of Oppression
On the contrary, the violation of rape, beating, killing, and harassment of women, people of color, gays, and other marked groups is motivated by fear or hatred of those groups. Sometimes the motive may be a simple will to power, to victimize those marked as vulnerable by the very social fact that they are subject to violence. If so,-this motive is secondary in the sense that it depends on a social practice of group violence. Violence-causing fear or hatred of the other at least partly involves insecurities on the part of the violators: its irrationality suggests that unconscious processes are at work.
Cultural imperialism, moreover, itself intersects with violence. The culturally imperialized may reject the dominant meanings and attempt to assert their own subjectivity, or the fact of their cultural difference may put the lie to the dominant cultures implicit claim to universality.The dissonance generated by such a challenge to.the hegemonic cultural meanings can also be a source of irrational violence.
Violence is a form of injustice that a distributive understanding of justice seems ill equipped to capture. This may be why contemporary discussions of justice rarely mention it. I have argued that group-directed violence is institutionalized and systemic. To the degree that institutions and social practices encourage, tolerate, or enable the perpetration of violence against members of specific groups, those institutions and practices are unjust and should be reformed. Such reform may require the redistribution of resources or positions, but in large part can come only through a change in cultural images, stereotypes, and the mundane reproduction of relations of dominance and aversion in the gestures of everyday life.
APPLYING THE CRITERIA
Social theories that construct oppression as a unified phenomenon usually either leave out groups that even the theorists think are oppressed, or leave out important ways in which groups are oppressed. Black liberation theorists and feminist theorists have argued persuasively, for example, that Marxism's reduction of all oppressions to class oppression leaves out much about the specific oppression of Blacks and women. By pluralizing the category of oppression in the way explained in this chapter, social theory can avoid the exclusive and oversimplifiying effects of such reductionism.
I have avoided pluralizing the category in the way some others have done, by constructing an account of separate systems of oppression for each oppressed group: racism, sexism, classism, heterosexism, ageism, and so on. There is a double problem with considering each group s oppression a unified and distinct structure or system. On the one hand, this way of conceiving oppression fails to accommodate the similarities and overlaps in the oppressions of different groups. On the other hand, it falsely represents the situation of all group members as the same.
I have arrived at the five faces of oppression—exploitation, marginalization, powerlessness, cultural imperialism, and violence—as the best way to avoid such exclusions and reductions. They function as criteria for determining whether individuals and groups are oppressed, rather than as a full theory of oppression. I believe that these criteria are objective. They provide a means of refuting some people's belief that their group is oppressed when it is not, as well as a means of persuading others that a group is oppressed when they doubt it. Each criterion can be operationalized; each can be applied through the assessment of observable behavior, status relationships, distributions, texts and other cultural artifacts. I have no illusions that such assessments can be value-neutral. But these criteria can nevertheless serve as means of evaluating claims that a group is oppressed, or adjudicating disputes about whether or how a group is oppressed.
The presence of any of these five conditions is sufficient for calling a group oppressed. But different group oppressions exhibit different combinations of these forms, as do different individuals in the groups. Nearly all, if not all, groups said by contemporary social movements to be oppressed suffer cultural imperialism. The other oppressions they experience vary. Working-class people are exploited and powerless, for example, but if employed and white do not experience marginalization and violence. Gay men, on the other hand, are not qua gay exploited or powerless, but they experience severe cultural imperialism and violence. Similarly, Jews and Arabs as groups are victims of cultural imperialism and violence, though many members of these groups also suffer exploitation or powerlessness. Old people are oppressed by marginalization and cultural imperialism, and this is also true of physically and mentally disabled people. As a group women are subject to gender-based exploitation, powerlessness, cultural imperialism, and violence. Racism in the United States condemns many Blacks and Latinos to marginalization, and puts many more at risk, even though many members of these groups escape that condition; members of these groups often suffer all five forms of oppression.
Applying these five criteria to the situation ofgroups makes it possible to compare oppressions without reducing them to a common essence or claiming that one is more fundamental than another. One can compare the ways in which a particular form of oppression appears in different groups. For exam^ pie, while the operations of cultural imperialism are often experienced in similar fashion by different groups, there are also important differences. Oné can compare the combinations of oppressions groups experience, or the intensity of those oppressions. Thus w;th these criteria one can plausibly claim that one group is more oppressed than another without reducing all oppres-. sions to a single scale.
Why are particular groups oppressed in the way they are? Are there any causal connections among the five forms of oppression? Causal or explanatory questions such as these are beyond the scope of this discussion. While I think general social theory has a place, causal explanation must always be
■    FIVE FACES OF OPPRESSION / IRIS YOUNG
particular and historical. Thus an explanatory account of why a particular group is oppressed in the ways that it is must trace the history and current structure of particular social relations. Such concrete historical and structural explanations will often show causal connections among the different forms of oppression experienced by a group. The cultural imperialism in which white men make stereotypical assumptions about and refuse to recognize the values of Blacks or women, for example, contributes to the marginalization and powerlesshess many Blacks and women suffer. But cultural imperialism does not always have these effects.
I
Five Faces of Oppression
REFERENCES
Ackerman, Bruce. 1980. Social Justice and the Liberal State, New Haven: Yale University Press.
Alexander, David. 1987. "Gendered Job Traits and Women's Occupations."
Ph. D. dissertation, Economics, University of Massachusetts. Boris, Ellen and Peter Bardaglio. 1983. "The Transformation of Patriarchy:
The Historic Role of the State." In Irene Diamond, ed., Families, Politics
and Public Policy. New York: Longman. Bowles, Samuel and Herbert Gintis. 1986. Democracy and Capitalism. New
York: Basic. "
Brittan, Arthur and Mary Maynard. 1984. Sexism, Racism and Oppression. Oxford: Blackwell.
Brown, Carol. 1981."Mothers, Fathers and Children: From Private to Public Patriarchy." In Lydia Sargent, ed., Women and Revolution. Boston: South End.
Buchanan, Allen. 1982. Marx and Justice. Totowa, N.J.: Roman and Allanheld.
Coward, Rosalind and John Ellis. 1977. Language and Materialism. London: Routledeg and Kegan Paul.
Delphy, Christine. 1984. Close to Home: A Materialist Analysis of Women's Oppression. Amherst: University of Massachusetts Press.
DuBois.W.E.B. 1969 [1903). Souls of Black Folk. New York: New American Library.
Easton, Barbara. 1978. "Feminism and the Contemporary Family." Socialist Review 39 (May/June): 11-36.
Epstein, Steven. 1987. "Gay Politics, Ethnic Identity: The Limits of Social Constructionism." Socialist Review 17 (May—August) 9—54.
Ferguson, Ann. 1984. "On Conceiving Motherhood and Sexuality: A Feminist Materialist Approach." In Joyce Trebilcot, ed., Mothering: Essays in Feminist Tlieory. Totowa, N.J.: Rowman and Allanheld.
-. 1989. Blood at the Root. London: Pandora.
Ferguson, Kathy. 1984. The Feminist Case against Bureauracy. Philadelphia: Temple University Press. Fiss, Owen. 1976. "Groups and the Equal Protection Clause." Philosophy and
Public Affairs 5 (Winter): 107-76. Foucault, Michel. 1977. Discipline and Punish. New York: Pantheon. Fraser, Nancy. 1987a "Women, Welfare, and the Politics of Need Interpretation." Hypatia: A Journal of Feminist Philosophy 2 (Winter): 103—22. -. 1987b. "Social Movements vs. Disciplinary Bureaucracies: The Discourse of Social Needs. "CHS Occasional Paper No. 8. Center for Humanistic Studies, University of Minnesota. French, Peter., 1975. "Types of Collectivities and Blame." The Personalist 56 (Spring): 1960-69.
Friedman, Marilyn. 1985. "Care and Context in Moral Reasoning." In Carol Harding, ed., Moral Dilemmas: Philosophical and Psychological Issues in the Development of Moral Reasoning. Chicago: Precedent.
■    FIVE FACES OF OPPRESSION / IRIS YOUNG
Friedman, Marilyn and Larry May. 1985. "Harming Women as a Group." Social Theory and Practice 11 (Summer): 297—34.
Frye, Marilyn. 1983. "Oppression." In The Politics of Reality. Trumansburg, N.Y.: Crossing Press.
Giddens, Anthony. 1981. A Contemporary Critique of Historical Materialism. Berkeley and Los Angeles: University of California Press.
Gilhgan, Carol. 1982. In a Different Voice. Cambridge: Harvard University Press.
Gotdieb, Rhonda. 1984. "The Political Economy of Sexuality." Review of Radical Political Economy 16 (Spring): 143—65.
Gottlieb, Roger. 1987. History and Subjectivity. Philadelphia: Temple University Press.
Habermas, Jiirgen. 1987. The Theory of Communicative Competence. Vol. 2: Life-world and System. Boston: Beacon.
Hartsock, Nancy. 1983. Money, Sex and Power. New York: Longman.
Heidegger, Martin. 1962. Being and Time. New York: Harper and Row.
Held, Virginia. 1987. "A Non-Contractual Society." In Marsha Hanen and Kai Nielsen, eds., Science, Morality and Feminist Theory. Calgary: University of Calgary Press.
Holmstrom, Nancy. 1977. "Exploitation." Canadian Journal of Philosophy 7 (June): 353-69. j
Lugones, Maria C. and Elizabeth V. Spelman. 1983. "Have We Got a Theory for You! Feminist Theory, Cultural Imperialism and the Demand for 'the Woman's Voice.'" Women's Studies International Forum 6:573—81.
Macpherson, C. B. 1973. Democratic Theory: Essays in Retrieval. Oxford: Oxford University Press.
Marable, Manning. 1984. Race, Reform and Rebellion: The Second Reconstruction in Black America. 1945—82. Jackson: University Press of Mississippi.
May, Larry. 1987. The Morality of Groups: Collective Responsibility, Group-Based Harm, and Corporate Rights. Notre Dame: Notre Dame University Press.
Murphy, Raymond. 1985. "Exploitation or Exclusion?" Sociology 19 (May): 225-43.
Offe, Claus. 1985. Disorganized Capitalism. Cambridge: MIT Press. Pateman, Carole. 1988. The Sexual Contract. Stanford: Stanford University Press.
Reich, Michael. 1981. Racial Inequality. Princeton: Princeton University Press.
Reiman, Jeffrey. 1987. "Exploitation, Force, and the Moral Assessment of Capitalism: Thoughts on Roemer and Cohen." Philosophy and Public Affairs 16 (Winter):-3-41.
Roemer, John. 1982. A General Theory of Exploitation and Class. Cambridge:
Harvard University Press. Ross, Jeffrey. 1980. Introduction to Jeffrey Ross and Ann Baker Cottrell, eds.,
The Mobilization of Collective Identity. Lanham, Md.: University Press of
America.
Rothschildjoseph. 1981. Ethnopolitics. New York: Columbia University Press. Sartre, Jean-Paul. 19'48. Anti-Semite and Jew. New York: Schocken.
(
Five Faces of Oppression H
Sennett, Richard, and Jonathan Cobb. 1972. The Hidden Injuries of Class. New York: Vintage.
Sher, George. 1987."Groups and the Constitution." In Gertrude Ezorsky, ed., Moral Rights in the Workplace. Albany: State University of New York Press.
Symanski, Al. 1985. "The Structure of Race." Review of Radical Political Economy 17 (Winter): 106-20.
Turner, John O, Michael A. Hogg, Penelope V. Oakes, Stephen D. Rucher, and Margaret S. Wethrell. 1987. Rediscovering the Social Group:A Self-Categorization Theory. Oxford: Blackwell.
Wolff, Robert Paul. 1,984. Understanding Marx. Princeton: Princeton University Press.