Ethnic and Racial Studies Vol. 21, no. 4 (1998)


Howard Winant


In the complex cross-currents of the post-civil rights era, what is racism? Is it the same old thing or has it changed in response to the changing dynamics of race itself in the post-civil rights era? To answer such questions, to understand the meaning of racism today, to take an informed and politically effective stand in such complex crosscurrents, is no easy matter.

Before we even tackle the matter of racism, we must first develop a working understanding of what we mean by race. This is not so easy either. Today we recognize that the concept of race is problematic, that the meaning of race is socially constructed and politically contested. This is a hard-won recognition, one which has obtained fairly generally only since WWII (Omi and Winant 1994).

But obviously problematizing race is not enough. We must steer between the Scylla of thinking that race is a mere illusion, mere ideology (in the sense of false consciousness) on the one hand, and the Charybdis of thinking that race is something objective and fixed. Both of these positions have their temptations, and by no means only for those who would deny the significance of race. The former position ("race as illusion") is upheld today not only by neoconservatives but also by radical theorists of race such as Anthony Appiah and Barbara Fields. In the work of these scholars, whatever its other merits, there is little recognition of the autonomy and depth of racialization in the U.S. The latter view ("race as "objective") is accepted not only by biological determinists and scientific racists, but also by many social scientists (some of them quite progressive), for example William Julius Wilson, Milton Gordon, or Michael Banton. In the work of these analysts, whatever its other merits, there is little recognition of the socially constructed, politically contested meaning of racial categories, of racial identity, of racialized experience.

In contrast to these approaches, Michael Omi and I have proposed the theory of racial formation, which looks at race not only as the subject of struggle and contest at the level of social structure, but also as a contested theme at the level of social signification, of the production of meanings. By the former we mean such issues as the racial dimensions of social stratification and distribution, of institutional arrangements, political systems, laws, etc. By the latter we mean the ways in which race is culturally figured and represented, the manner in which race comes to be meaningful as a descriptor of group or individual identity, social issues, and experience.

We have sought to theorize racial formation as a permanent process in which historically situated projects interact: in the clash and conflict, as well as the accommodation and overlap of these projects, human bodies and consciousness, as well as social institutions and structures, are represented and organized. We argue that in any given historical context, racial signification and racial structuration are ineluctably linked. To represent, interpret, or signify upon race, then, to assign meaning to it, is at least implicitly and often explicitly to locate it in social structural terms.

The linkage between culture and structure, which is at the core of the racial formation process, gives racial projects their coherence and unity. Thus, once it is argued that the U.S. is inherently a "white man's country" (as in certain far right racial discourses), or that race is a spurious anachronism beneath the notice of the state (as in neoconservative positions), or that racial difference is a matter of "self determination" (as in certain radical racial discourses), the appropriate political orientation, economic and social programs, etc., follow rather quickly.

 The reverse is also true: when organizations, institutions, or state agencies advocate or resist a certain racial policy or practice, when they mobilize politically along racial lines, they necessarily engage in racial signification, at least implicitly and usually explicitly. Thus when the Supreme Court rules that individualism and meritocracy are the only legitimate criteria for employment decisions or university admissions today, it inevitably and simultaneously represents race as illusory and spurious. Let me give some other examples. Consider the implications when spokespersons for the Aryan Nations or the Church of the Creator (two U.S. fascist groups) propose setting aside areas of the country for "whites only": in this structural initiative they simultaneously represent race as a natural, invariant, biological difference. On the polar opposite end of the political spectrum, consider what happens when radical democratic organizations, such as the Highlander Center or UNITE -- the newly merged needle trades union -- engage in community or labor organizing that seeks both to build multiracial organizations and to recognize the relevance of distinct racialized experiences among their constituents. Here too, in this effort to mobilize politically, to change the social structure, they necessarily represent race in terms of decenteredness, flexibility, and the relative permanence of difference, embracing the DuBoisian synthesis of full democracy and racial "conservation" (Du Bois 1995 [1897]).

Keep this idea of racial formation via racial projects on hold while the discussion is refocused on racism; I shall return to the dynamics of racial formation presently. In what follows I shall first discuss the transformations that have affected the concept of racism since the ambiguous triumph of the civil rights "revolution" in the mid-1960s. Next, I shall offer an account of contemporary racism which I believe more adequately addresses present conditions. Finally, in a brief conclusion, I offer some thoughts about the changing link between racial formation and racism as the millenium approaches.


 The understanding we have of racism, an understanding which was forged in the 1960s, is now severely deficient. A quarter century of sociopolitical struggle has rendered it inadequate to the demands of the present. At the same time, I would hardly wish to argue (in the manner of neoconservatives) that racism itself has been largely eliminated in the post- civil rights era. But although we are pretty sure that racism continues to exist, indeed flourish, we are less than certain about what it means today.

 In fact, since the ambiguous triumph of the civil rights movement in the mid-1960s, clarity about what racism means has been slipping away. The concept entered the lexicon of "common sense" only in the 1960s. Before that, although the term had surfaced occasionally, the problem of racial injustice and inequality was generally understood in a more limited fashion, as a matter of prejudiced attitudes or bigotry on the one hand, and of discriminatory actions on the other.

Solutions, it was believed, would therefore involve two elements: first, the overcoming of prejudiced attitudes through the achievement of tolerance, the acceptance of "brotherhood" etc.; and second, the passage of laws which prohibited discrimination with respect to access to public accommodations, jobs, education, etc. Social scientific work tended to focus on the origins of prejudiced attitudes (Adorno et al 1950, Allport 1954), on the interests served by discrimination (Rose 1948, Becker 1957, Thurow 1969), and on the ways in which prejudice and discrimination combined or conflicted with each other (Merton 1949).

 The early civil rights movement explicitly reflected such views. In its espousal of integration and its quest for a "beloved community" it sought to overcome racial prejudice. In its litigation activities and agitation for civil rights legislation it sought to challenge discriminatory practices.

 The later 1960s, however, signalled a sharp break with this vision. The emergence of the slogan "black power" (and soon after, of "brown power," "red power," and "yellow power"), the wave of riots that swept the urban ghettos from 1964 to 1968, and the founding of radical movement organizations of nationalist and marxist orientation, all coincided with the recognition that racial inequality and injustice had much deeper roots. They were not simply the product of prejudice, nor was discrimination only a matter of intentionally informed action. Rather, prejudice was an almost unavoidable outcome of patterns of socialization which were "bred in the bone," affecting not only whites but even minorities themselves. Discrimination, far from manifesting itself only (or even principally) through individual actions or conscious policies, was a structural feature of U.S. society, a product of centuries of systematic exclusion, exploitation, and also cultural assaults of various types upon racially defined minorities.

 It was this combination of relationships, prejudice, discrimination, and structural inequality (AKA "institutional racism"), which defined the concept of racism at the end of the 1960s. Without a doubt, such a synthesis was an advance over previous conceptions. Its very comprehensiveness was better suited to the rising tide of movement activity and critique of white supremacy. Notably, its emphasis on the structural dimensions of racism allowed it to address the intransigence which racial injustice and inequality continued to exhibit, even after discrimination had supposedly been outlawed and bigoted expression stigmatized.

But such an approach also had clear limitations. As Robert Miles has argued (1989), it tended to "inflate" the concept of racism to a point at which it lost precision. If the "institutional" component of racism were so pervasive and deeply rooted, it became difficult to recognize what accomplishments the movement had achieved or what progress civil rights reforms represented. How, under these conditions, could one validate the premises of political action aimed at racial justice and greater substantive social equality? If institutional racism were indeed so ubiquitous, it became difficult to affirm the existence of any democracy at all where race was concerned. The result was a levelling critique which denied any distinction between the Jim Crow era (or even the whole longue duree of racism beginning with European conquest and leading through racial slavery and Jim Crow), and the present. Similarly, if the prejudice component of racism were so deeply inbred, it became difficult to account for the apparent racial hybridity and cultural interpenetration that characterizes civil society in the U.S., as evidenced not only by the shaping of popular mores, values, language, and style, for example, but also by the millions of people, white and black (and neither white nor black) who occupy interstitial and ambiguous racial positions. The result of the "inflation" of the concept of racism was thus a deep pessimism about any efforts to overcome racial barriers: in the workplace, the community, or any other sphere of lived experience. An overly comprehensive view of racism, then, potentially served as a selffulfilling prophecy.

Yet the alternative view, which surfaced with a vengeance in the 1970s and urged a return to the conception of racism held before the movement's "radical turn," was equally inadequate. This was the neoconservative project, which deliberately restricted its attention to injury done to the individual as opposed to the group, and to advocacy of a "color-blind" racial policy. Such an approach reduced race to ethnicity, and almost entirely neglected the continuing organization of social inequality and oppression along racial lines. Worse yet, it tended to rationalize racial injustice as a supposedly natural outcome of group attributes in competition (Sowell 1983).

Thus have we arrived at today's dilemmas. In the post-civil rights era U.S. society has undergone a substantial modification of the previously far more rigid lines of exclusion and segregation, permitting real mobility for more favored sectors (that is, certain class-based segments) of racially-defined minority groups. This period has also witnessed the substantial diversification of the North American population, in the aftermath of the 1965 reform of immigration laws. Panethnic phenomena have increased among Asians, Latinos, and Native Americans, reconstituting the US racial panorama in a multipolar (as opposed to the old bipolar) direction. Racial identity has been problematized (at least somewhat) for whites -- a fact which has its dangers but also reflects progress -- and the movements to which the black struggle gave initial impetus, notably feminism and gay liberation in their many forms, have developed to the point where a whole range of cross-cutting subjectivities and tensions (as well as new alliances) have been framed.

 But the post-civil rights era has also witnessed a significant racial reaction. The racial reaction has rearticulated the demands for equality and justice made by the black movement and its allies in a conservative discourse of individualism, competition, and laissez-faire. We must recognize that it is this "new right" discourse which is hegemonic today, and that in these terms racism is rendered invisible and marginalized. It is treated as largely an artifact of the past.


 Today, then, the absence of a clear "common sense" understanding of what racism means has become a significant obstacle to efforts aimed at challenging it. As usual there are different interpretations -- different racial projects -- in conflict with one another over the very meaning and structure of racism. It is common to find the view, especially among whites (but also among nonwhites), that we must somehow get "beyond" race in order to overcome racism. For example, I often hear in my classes comments such as "I don't care if someone is black, white, green or purple; a person's just a person to me...," etc. This implies that racism is equivalent to color- consciousness and consequently nonracism must be a lack of color- consciousness. We should recognize that this type of idea, however naive, is a true product of the civil rights era, notably the movement's early, "liberal" years.

On the other hand, I hear from other students (from my black and brown students particularly, but by no means only from them), that racism is a "system of power." This idea implies that only whites have power, and thus only they can be racists. We should also recognize the origins of this idea, which exhibits a different but no less dangerous naivete -- for it is highly problematic to assert that racially-defined minorities are powerless in the contemporary U.S. -- in the radicalized later years of the civil rights era.

Given this crisis of meaning, and in the absence of any "common sense" understanding, does the concept of racism retain any validity? If so, what view of racism should we adopt? Is a more coherent theoretical approach possible? I believe it is.

 Recall my discussion of racial formation theory at this point. Let us recognize that, like race, racism has changed over time. It is obvious that the attitudes, practices, and institutions of the epochs of slavery, say, or of Jim Crow, no longer exist today. Employing a similar logic, it is reasonable to question whether concepts of racism which were developed in the early days of the post-civil rights era, when the limitations of both moderate reform and militant racial radicalism of various types had not yet been encountered, could possibly remain adequate to explain circumstances and conflicts a quarter-century later.

Racial formation theory also allows us to differentiate between race and racism. The two concepts should not be used interchangeably. I have argued that race has no fixed meaning, but is constructed and transformed sociohistorically through competing political projects, through the necessary and ineluctable link between the structural and cultural dimensions of race in the U.S. This emphasis on projects allows us to refocus our understanding of racism as well, for racism can now be seen as characterizing some, but not all, racial projects.

Today, a racial project can be defined as racist if it creates or reproduces hierarchical social structures based on essentialized racial categories. This approach recognizes the importance of locating racism within a fluid and contested history of racially based social structures and discourses. It allows us to recognize that there can be no timeless and absolute standard for what constitutes racism, because social structures undergo reform (and reaction) and discourses are always subject to rearticulation. This definition therefore does not invest the concept of racism with any permanent content, but instead sees racism as a property of certain political projects that link the representation and organization of race -- that engage in the "work" of racial formation. Such an approach focuses on the "work" essentialism does for domination, and the "need" domination displays to essentialize the subordinated.

It is also important to distinguish racial awareness from racial essentialism. Attribution of merits or faults, allocation of values or resources, and/or representations of individuals or groups on the basis of racial categories should not be considered racist in and of themselves. Such projects may in fact be quite benign. Of course, any of these projects may be considered racist, but only if they meet the criteria I have just outlined: in other words, essentialization and subordination (which are always linked) must be present.

 Consider the following examples. First, a discursive one: the statement "Today, many Asian Americans are highly entrepreneurial." Second, a structural one: the organization of an association of, say, black accountants. The first racial project, in my view, signifies or represents a racial category ("Asian Americans"), and locates that representation within the social structure of the contemporary U.S. (in regard to business, class issues, socialization, etc.). It does not, however, essentialize; it is qualified in time ("today") and in respect to overgeneralization ("many"). The second racial project is organizational or social structural, and therefore must engage in racial signification. Black accountants, the organizers might maintain, have certain common experiences and characteristics, can offer each other certain support, etc. The effort to organize such a group is not in and of itself antagonistic to other groups; it does not aim at others' subordination, but only at members' well-being and uplift.

Neither of these racial projects, then, can fairly be labelled racist. Of course, racial representations may be biased or misinterpret their subjects, just as racially-based organizational efforts may be unfair or unjustifiably exclusive. If such were the case, if for instance in our first example the statement in question read "Asian Americans are naturally entrepreneurial," this would by my criterion be racist. Similarly, if the effort to organize black accountants had as its rationale the raiding of clients from nonblack accountants, it would by my criterion be racist as well.

 Proceeding with this standard, to allocate values or resources -- let us say, academic scholarships to racially-defined minority students -- is not racist, since no essentializing/subordinating standard is at work here. Scholarships are awarded to Rotarians, children of insurance company employees, and residents of the Pittsburgh metropolitan area. Why then should they not also be offered, in particular cases, to blacks or Chicanos or Native Americans? The latter categories are no more suspect than the former ones.

 What if scholarships were offered only to whites? Such action would be suspect, not on the grounds of essentialism, but on those of domination and subordination, since the logic of such a racial project would be to reproduce an existing racial hierarchy.

Let us take an example that is much on our minds today: the effort to invalidate affirmative action programs on the grounds that these constitute "reverse discrimina tion." This project would, I think, be vulnerable to criticism under the criterion of racism I am developing here. This is because, as we have learned in the post-civil rights era, it is possible to reproduce racial categories even while ostensibly repudiating them. The preservation of racial hierarchy may operate through an essentializing logic that dissembles or operates subtextually:

The scenario...reads as follows: when science apologizes and says there is no such thing, all talk of "race" must cease. Hence "race," as a recently emergent, unifying, and forceful sign of difference in the service of the "Other," is held up to scientific ridicule as, ironically, "unscientific." A proudly emergent sense of ethnic diversity in the service of the new world arrangements is disparaged by white male science as the most foolish sort of anachronism (Baker 1985, 385; emphasis original).
The familiar "code word" phenomenon, that is, the subtextual signification of race, has much the same effect. Thus the claim, first made in 1896 and recently elevated to nearly hegemonic jurisprudential doctrine, that "our Constitution is color-blind," can in fact be understood in two ways. It can mean, as Justice Harlan evidently intended in his ringing dissent in the Plessy case, and as the early civil rights movement clearly understood it as well, that the power of the state should not be used to enforce invidious racial distinctions. But it can also mean that the power of the state should not be used to uproot those distinctions either. Based on the criteria I have advanced here, I suggest that despite its anti-essentialist appearance, the "color-blind" denial of the significance of race is in fact an essentializing representation of race, an "erasure" of race, so to speak, which in the present-day U.S. is generally linked to the perpetuation of racial hierarchy. It is, then, a form of racism, a type of racist project (Gotanda, 1995).

In order to identify a social project as racist using the criterion I have proposed here, one must demonstrate a link between essentializing representations of race and hierarchical social structures. Such a link might be revealed in efforts to protect dominant interests, framed in racial terms, from democratizing racial initiatives. For example: changing to at-large voting systems when minority voters threaten to achieve significant representation. But such a link might also consist of efforts simply to reverse the roles of racially dominant and racially subordinate. In melanin theories of racial superiority (Welsing 1991), for example, or in the racial ontology of the Nation of Islam with its mad scientist Dr. Yacub, we see racist projects which have a black provenance. Racism is not necessarily white, though in the nature of things, it is more often so. It inheres in those political projects that link racial essentialism and racial hierarchy, wherever and however that link is forged.


Although we can conclude that racism is not invariably white, we must also recognize that today, as in the past, there is a hegemonic racial project -- that of the "new right" -- which in general defends white racial privilege. It employs a particular interpretive schema, a particular logic of racial representation, to justify a hierarchical racial order in which, albeit more imperfectly than in the past, dark skin still correlates with subordination, and subordinate status often, though not always, is still represented in racial terms.

 Furthermore, a key problem of racism, today as in the past, is its denial, or flattening, of difference within the categories it represents in essentialist fashion. Members of racially- defined subordinate groups have for a long time faced practices of exclusion, discrimination, and even of outright extermination. Such groups are thus forced to band together in order to defend their interests (if not, in some instances, their very lives). Following this argument, such "strategic essentialism" cannot be equated with the essentialism practiced in service of hierarchical social structures. Nor would it prevent the interrogation of internal group differences, though these are sometimes overridden by the imperative for group "conservation," to use Du Bois's term.

Obviously, any abstract concept of racism is severely put to the test by the untidy world of reality. Yet I believe that it is imperative to meet that test at the level of theory, and indeed at the level of practice that ought to flow from theory, just as we must meet the test in our everyday lives.

 Today we live in a situation in which "the old is dying and the new cannot be born," in which formerly unquestioned white supremacy is now questioned. It is a situation in which an anti- racist counter-tradition in politics and culture has made significant gains. But despite all the changes wrought by this anti-racist project, this radical democratic initiative which derives from the postwar black movement, it has not been possible to overthrow the deeply rooted belief that the US is still, as the phrase goes, a "white man's country." It has not been possible fully to transform the social, political, economic, and cultural institutions that afford systematic privileges to whites. It has not been possible to alter the displacement of the burdens and problems of the society (such as unemployment, undereducation, poverty, and disease) onto the shoulders of nonwhites.

 Thus the racial dualism that Du Bois identified nearly 100 years ago continues to operate. Recall that he characterized the black experience as as a conflict between "...two souls, two thoughts, two unreconciled strivings" (Du Bois 1995 [1903]). This now applies, albeit in very different ways, to everybody. The full exposition of this point is beyond the scope of this essay, but I have discussed it more fully elsewhere (Winant 1997, Winant 1994). Suffice it to say here that, as a society and as individuals, we both uphold and resist white supremacy. We experience both our particular privilege or subalternality, and, to the extent we can, we resist it.

 Confronting racism in such a situation is difficult. It's a moving target, a contested terrain. Inevitably, as a society, as political movements, and as individuals, we have to make lots of mistakes; we have to see our action and our thought, our praxis, in pragmatic terms. Because racism changes and develops, because it is simultaneously a vast phenomenon framed by epochal historical developments, and a moment-to moment experiential reality, we can never expect fully to capture it theoretically. Nor can we expect that it will ever be fully overcome. That doesn't mean, however, that we are free to desist from trying.


 Adorno, T.W. et al, TheAuthoritarian Personality (New York: Harper and Bros., 1950).

 Allport, Gordon W., The Nature of Prejudice (Cambridge, MA: Addison Wesley, 1954).

 Baker, Houston A., Jr., "Caliban's Triple Play," in Henry Louis Gates, Jr., ed., "Race," Writing, and Difference (Chicago: University of Chicago Press, 1986).

 Becker, Gary, The Economics of Discrimination (Chicago: University of Chicago Press, 1957).

 Blauner, Bob, "Talking Past One Another: Black and White Languages of Race," The American Prospect 10 (1992).

 Du Bois, W.E.B., "Of Our Spiritual Strivings" (1903), in David Levering Lewis, ed., W.E.B. Du Bois: A Reader (New York: Henry Holt, 1995).

 Du Bois, W.E.B., "The Conservation of Races" (1897), in David Levering Lewis, ed., W.E.B. Du Bois: A Reader (New York: Henry Holt, 1995).

 Fuss, Diana, Essentially Speaking: Feminism, Nature, and Difference (New York: Routledge, 1989).

 Gotanda, Neil, "A Critique of 'Our Constitution is Color-Blind,'" in Kimberle Crenshaw et al, eds., Critical Race Theory: The Key Writings That Formed the Movement (New York: The New Press, 1995).

 Merton, Robert K., "Discrimination and the American Creed," in Robert W. MacIver, ed., Discrimination and the National Welfare (New York: Harper and Brothers, 1949).

 Miles, Robert, Racism (New York: Routledge, 1989).

 Omi, Michael, and Howard Winant, Racial Formation in the United States: From the 1960s to the 1990s, 2nd. ed (New York: Routledge, 1994).

 Rose, Arnold M., The Negro in America (New York: Harper and Brothers, 1948).

 Sowell, Thomas, The Economics and Politics of Race: An International Perspective (New York: Quill, 1983).

 Thurow, Lester, Poverty and Discrimination (Washington DC: The Brookings Institution, 1969).

 Welsing, Frances Cress, The Isis Papers: The Keys to the Colors (Chicago: Third World Press, 1991).

Winant, Howard, Racial Conditions: Politics, Theory, Comparisons (Minneapolis: University of Minnesota Press, 1994).

Winant, Howard, "Racial Dualism at Century's End," in Wahneema Lubiano, ed., The House That Race Built (New York: Random House, 1997).