Chapter Five
Understanding Racism

Reflections of a Comparative Historian

Historians of black-white relations in the United States often refer to "racism," but only rarely do they define the term precisely and explore its theoretical implications. Compared with the explanatory power associated with the economic and political variables operating in a specific historical situation, the cultural predisposition to stigmatize and abuse the racial "Other" is likely to be treated as a secondary phenomenon—the by-product of something else—or, alternatively, as one ingredient in an eclectic stew, the effect of which is impossible to isolate. Sometimes the term is used narrowly to refer to a set of doctrines that rose and fell in the United States between the late eighteenth century and the mid-to-late twentieth. Often, however, it is a catchall that refers to whatever was thought and done to the disadvantage of African Americans from the sixteenth century to the present. For some purposes, perhaps, nothing much is lost by inattention or lack of analytical rigor. But comparative historians need sharper tools and stronger conceptualizations; otherwise they are likely to find implicit, attitudinal racism in most times and places—a given of any situation that appears to involve "races"—or an explicit ideological racism in only a few places and for limited periods. It is high time that historians devoted the same effort to understanding "race" as a transnational social and historical phenomenon that they have sometimes applied to class, gender, and nationalism.

Postmodernists have contributed some useful new vocabulary to this effort—especially the description of race as "a social construction"

rather than an objective fact. Substantively speaking, this formulation is not so radically new as some of its proponents suggest. The concept of race as a "social fiction" rather than a biological fact has been a staple of the sociological literature on "race relations" for half a century. But the new formulation puts an emphasis on process—how and why the fiction was created—that the older language permits but does not require. Furthermore, postmodernist thought has put gender, nation, and even class in the same category of ideologically constructed fictions that race already occupied in the thought of antiracist scholars. It invites us, therefore, to probe the connections between race and other social constructions of human identity rather than making race a secondary or derived category—the mere reflection of some deeper and more fundamental reality.

Perhaps, it has been suggested, we should no longer use the term "race" at all, because to do so gives credence to the idea that there is a physical reality to which it refers. Human beings do of course differ in physical appearance, but variations in skin color have no more intrinsic or scientific significance than differences in hair color. It is the associations that people make between such visible phenomena and their entire range of interests, beliefs, and attitudes that determine whether or not "race" comes into play. My own view is that we can continue to use the term if we recognize that it refers to an ideology rather than an objective reality. The recently coined verb "to racialize" (in the noun form "racialization") is useful in referring to the process of constructing race rather than to its results.

Like some other notable ways of construing human diversity, racism has enormous historical consequences—on the same order of importance as nationalism, sexism, class consciousness, and sectarian religious zeal. Unless one is prepared to take one of these great signifiers of human diversity and make it the root of all the others—class is the most popular candidate, although gender has its advocates—the historian faces the task of showing how a number of social and cultural constructions interact rather than proving that one is more "real" or fundamental than the others.

Merely recognizing that something is socially, culturally, or ideologically constructed does not fully explain it—indeed it scarcely explains it at all. We need to know how and why it was constructed in the way that it was. Furthermore, only the most radical postmodernists would claim that something can be constructed out of nothing. What raw materials were used, where did they come from, and how well do they

serve the aims of the builders? One can use wood to make a house or a warship. The materials do not determine the function of the thing that has been constructed, but they may help to determine how well it fulfills the purposes for which it was designed. Wooden houses burn more easily than those made of brick, and wooden ships were more vulnerable to enemy cannon fire than the steel-hulled vessels that took their place in the world's navies. Science-based concepts of race may lose credibility as the result of new discoveries and shifting paradigms, but concepts of race based on the cultural differences associated with descent groups may have greater durability. The notion that race or anything else is a social and cultural construction is the beginning of an inquiry rather than the end of it.

When sociologists and historians first wrote about racism in the period between World War II and the 1960s, they generally meant an explicit ideology based on the putative scientific truth that population groups distinguishable from each other in physical appearance or ancestry were different and unequal in genetically determined mental and behavioral capabilities. As recently as 1967, for example, the British sociologist Michael Banton defined racism as "the doctrine that a man's behavior is determined by stable inherited characteristics deriving from separate racial stocks having distinctive attributes and usually considered to stand to one another in relations of superiority and inferiority." In that same year, an American sociologist, Pierre van den Berghe, described it as an ideology based on the belief that "organic, genetically transmitted differences (real or imagined), are intrinsically associated with the presence or absence of certain socially relevant abilities or characteristics."^[1]

Michael Banton, Race Relations (New York, 1967), 8; Pierre van den Berghe, Race and Racism: A Comparative Perspective (New York 1967), 11. Recognizing the narrowness and limited applicability of his definition, Banton used the term "racialism" to cover the attitudinal and institutional aspects of racial domination, and van den Berghe's perceptive treatment of the sociohistorical role of race in the United States, Brazil, Mexico, and South Africa took account of the implicit or assumed character of racism in situations where its full ideological expression was absent or muted. But neither transcended the notion that a belief in genetic or biological determinism had to be present.

Such a doctrine or ideology was used to justify or rationalize a range of policies, depending on the circumstances and aims of the racializing group. The principal possibilities were the subordination and unequal segregation of the Other, exclusion or expulsion from a community or nation, or in the most extreme case physical annihilation. Hitler's view of the Jews and what should be done with them and the southern white supremacist's conception of the African American's place in nature and society were obvious and unambiguous examples of racism in this strict sense.

Since the late 1960s, however, scholars and activists in the United States have tended to apply the term to attitudes and practices viewed as objectively harmful to the interests and aspirations of people previously designated as racially inferior, even though an explicit doctrine of

innate racial differences is no longer invoked as a rationale. The discovery of "institutional"—as opposed to "attitudinal"—racism has broadened the concept to include the discrimination that persists because institutions operate on the basis of seemingly color-blind rules and procedures that in fact deny equal opportunity to members of minority groups. But to maintain that such discrimination is truly unintentional and does not involve individual attitudes would undermine the conception of racism as an ideological construction and make it synonymous with the statistical inequality and apparent social inefficiency of any group with a sense of racial or ethnic identity, whatever the actual causes of its situation might be.

The concept of racism remains relevant in such cases only if it can be established that members of one identifiable group falsely assume that the members of another such group perform less well than themselves because of their inherent or deep-seated inadequacies rather than their disadvantages in an unjust social system. This standard would still cover most of what has been described as institutional or structural racism. Whites who oppose programs that seek to secure equal opportunities for blacks may deny with varying degrees of sincerity that they consider blacks to be genetically inferior to themselves, but can anyone really doubt that if the inhabitants of our inner cities were white rather than black or Latino, more would be done to alleviate the conditions that breed gang warfare, crime and a variety of other social problems? It also seems likely that support for the death penalty would not be as strong as it is in the United States today if a disproportionate number of those on death rows did not happen to be black.

Evidence in fact abounds that actual prejudice continues to operate on all levels of American society. The persistence of white supremacist attitudes is manifested in a toleration of police brutality against minorities, black-white disparities in judicial punishments for the same crimes, housing and mortgage discrimination, and glass ceilings in corporate personnel policies. It is therefore premature and misleading to maintain that attitudes and ideologies are no longer the problem. If American institutions operated in the way they are supposed to operate, we would still have class and status inequality, but racism would no longer be the appropriate diagnosis for such conditions.

Racism is not the only form of injustice and inequality in the world, and if we wish to think about it in useful and productive ways, we have to be able to distinguish it from other varieties of human nastiness. Rather than enlarge the concept of racism to include structural

inequalities that are independent of racialization, it would be more fruitful to expand the concept to include modes of thought and behavior that we do not usually think of as racist because they do not involve the classic racist doctrine of biological logical inferiority signified by color.

Two recent books illustrate the difference between the old and new versions of racist ideology. The Bell Curve by Charles Murray and Richard Herrnstein is an effort to revive old fashioned biological racism. In their view blacks are simply less well-endowed intellectually than whites. Dinesh D'Souza's The End of Racism eschews the genetic determinism of abilities in favor of a cultural determinism that does much the same work. D'Souza attributes African American "failure" to a "dysfunctional" group culture and then uses this judgment to support a contention that racism is not the source of black underachievement. Stereotyping and stigmatizing a racialized group on the basis of cultural rather than biological inferiority provides a new rationale for discrimination rather than a basis for combating it. This "new racism" is not really unprecedented. Cultural rather than biological determinism was the official justification for apartheid in South Africa. Furthermore, D'Souza's line of argument recalls the rationale for black enslavement and subordination that preceded the growth of scientific racist doctrines in the late eighteenth and nineteenth centuries.^[2]

Richard J. Herrnstein and Charles Murray, The Bell Curve: Intelligence and Class Structure in American Life (New York, 1994); Dinesh D'Souza, The End of Racism: Principles for a Multiracial Society (New York, 1995). See also my review of D'Souza, New York Review of Books, 19 October 1995, 10-15.

According to sociologists and historians who draw a sharp distinction between racism and ethnocentrism, it is anachronistic to describe as racist the prejudice and discrimination against the Other that emerged before Western ethnological thinkers articulated naturalistic conceptions of human diversity and inequality. In an early effort to grapple with the relationship between the ideologies rationalizing slavery in the early colonial period and those employed in the nineteenth century, I made a distinction between ideological and "societal" racism. In the latter category, I included practices that treated a subaltern group as if it were inherently inferior to the socially dominant group even though an explicit doctrine of innate racial differences had not yet been promulgated and widely accepted. But there was of course an ideological basis for such discrimination. An alternative to identifying racism by its effects rather than by the consciousness that produced it is to seek the common element in the differing ideological formulations that have sustained white or European domination over people of color. If the term racism is to apply, I now believe, its association with the specific form of biological determinism that justified slavery and segregation in

the nineteenth and twentieth centuries must be regarded as fortuitous rather than essential.^[3]

George M. Fredrickson, The Arrogance of Race: Historical Perspectives on Slavery, Racism, and Social Inequality (Middletown, Conn., 1988), 189-205. This essay on "The Social Origins of American Racism" was originally published in 1971.

What we need for comparative historical analysis is a theoretical understanding of racism that is broad enough to take account of contemporary ideologies that stigmatize the Other without appealing to racist science as well as attitudes that preceded the elaboration of classic racist doctrine. But in seeking such breadth we must be careful to avoid giving credence to the view that racism is an essential or primordial human response to diversity, something that inevitably takes place when groups that we would define as racially different come into contact. It must be remembered that we are doing the defining and that, as the historical record shows, the categories we use did not always exist and were in fact constructed or invented by our ancestors. There is a strong, and perhaps unavoidable, tendency in any society to disparage those defined as Other or alien. Such negative stereotyping may be an inescapable component of identity and boundary maintenance. But who the outsiders are and how much or in what ways they are despised and mistreated are the products of history, not of basic human instincts.

I therefore agree with historian Barbara Fields that "race" and all the ideas and attitudes associated with it are shaped and sustained by social contexts that change over time and are not therefore the reflection of some "transhistorical" impulse that is rooted in objective human differences. But Fields and other historians who argue that "class" is real and "race" is not are captives of a theory of social relationships that privileges one form of social inequality over others in a manner that does violence to the actual history of human inequality. If class is defined very loosely as "the inequality of human beings from the standpoint of social power," there is no disagreement, although it is hard to make analytical use of such an all-encompassing passing concept. Problems emerge when Fields goes on to invoke "the more rigorous Marxian definition involving social relations of production." Is it really true, as she claims, that class in this sense "can assert itself independently of people's consciousness" while race cannot? People certainly do differ in economic power and position, but such differences have literally no meaning until they enter people's consciousness and are interpreted in some fashion. The specifically Marxist conception of two essential and perpetually antagonistic classes is not a necessary deduction from the existence of economic inequality in a capitalist society. There could be more than two essential classes, and classes do not have to be viewed as inevitably at war. Class in my view is as much a

historical and social construct as race, which also builds on differences that actually exist but are not meaningful until constructed into an ideology of differences that serves the purposes of a social group.^[4]

Barbara J. Fields, "Ideology and Race in American History," in Region, Race, and Reconstruction: Essays in Honor of C. Vann Woodward, ed. J. Morgan Kousser and James M. McPherson (New York, 1982), 150-151.

Besides differing in their relation to the means of production or to the market, people really do differ in physical characteristics that are subject to classification, immediate or remote ancestry, and cultural traits associated with belonging to a historically defined community. It is my contention that race and racism derive from the act of interpreting or construing such noneconomic differences to create a sense of group solidarity or peoplehood that becomes the basis for assertions of dominance or privileged status over those considered outside of the group. (If the process stopped at group identity and solidarity, we would have ethnocentrism but not racism.) This way of grounding group power and inequality is not simply a variation on the larger theme of class domination. It has independent roots and consequences. Class refers to the fact that every developed society has an economy and that some people have more access to its fruits than others. Race arises from the equally pervasive fact that all human beings have some sense of family or kinship, a way of differentiating those with whom one has real or fictive "blood ties" from those who are unrelated. Ethnicity can be viewed as extended kinship, and race can be seen as an inflation and elaboration of the notion that my "family" not only is better than yours but also has special rights and privileges as a consequence.

Although race and class are both historical inventions—creative interpretations or ideological constructions based on two ways of conceiving human diversity—it would be a mistake to infer that, once invented, they do not become durable and enormously influential ways of perceiving the world. The construction of class may lead to class conflict, revolution, and socialist societies. The construction of race may lead to secession in defense of racial slavery, the creation of social orders based on racial caste, or to gas chambers for stigmatized peoples.

A useful way of comprehending the relationship of race and class is Max Weber's writings on social hierarchy as a general phenomenon and stratification based on ethnicity or race as a special case. For Weber, "status," or the unequal assignment of honor and prestige to individuals or groups, may vary independently from "class," which he defines as the economic advantage or disadvantage that comes from objective relationships to a capitalistic market. Status may be based on aristocratic descent or the ability to maintain a prestigious life-style, but

in multiethnic societies it can also be derived simply from membership in an ethnic or racial group that has a history of dominance over other groups and seeks to preserve that relationship.^[5]

Max Weber, Economy and Society: An Outline of Interpretive Sociology, ed. Guenther Roth and Claus Wittich (Berkeley, 1978), 385-398, 926-939.

Using Weber's concept of "ethnic status" to get at the nature of racism allows us to sidestep an unprofitable debate on the difference between race and ethnicity. As Donald L. Horowitz has argued persuasively in a broad-ranging comparative study of "ethnic conflict," the designation of people by skin color and the mistreatment of them on that basis has no special features that would distinguish it in any definitive theoretical way from group domination based on religion, culture, or the simple belief that some people have defective ancestry. It is only because modern Western liberalism often assumes that it is relatively easy for people to change their religion or culture and be assimilated into a group other than the one in which they were born that the distinction has become important.^[6]

Donald L. Horowitz, Ethnic Groups in Conflict (Berkeley, 1985), 41-54.

Even in the West the de facto ascription of ethnic status has often been derived from something other than skin color or reputed nonwhite ancestry. Northern Ireland, for example, has the essential characteristics of what sociologist John Rex calls a "race relations situation." A Catholic could certainly convert to Protestantism; however, it is not only extremely unlikely that one would do so but also doubtful whether one could thereby win full acceptance into the Protestant community. Racial consciousness is also what what makes the Palestinian-Israeli conflict so difficult to resolve, despite the lack of clear physical differences between the antagonistic ethnic communities. A large proportion of the population of Israel are Jews who found themselves increasingly oppressed in various Arab societies, and the Arab minority in Israel, despite having the right to vote, is not eligible for full citizenship in a Jewish ethnic state.^[7]

See John Rex, "The Concept of Race in Sociological Theory," in Race and Racialism, ed. Sam Zubaida (London, 1970), 35-55.

The key element in ethnicity is descent, and ethnic status emerges when a group of people with a real or fictive common ancestry assert their dominance over those who are believed to be of a different and inferior ancestry. The Burakhumin of Japan are descended from a caste that once engaged in occupations considered unclean or impure. Although they differ scarcely at all in phenotype or general culture from other Japanese, the discrimination against them, on grounds of descent alone, closely resembles the color discrimination of Western societies. One might conclude, therefore, that racism, or something virtually indistinguishable from it, has no essential relation to skin color or other obvious physical characteristics and need not even be based on significant

cultural differences. The essential element is the belief, however justified or rationalized, in the critical importance of differing lines of descent and the use of that belief to establish or validate social inequality.^[8]

On the Barakhumin as a minority caste analogous to African Americans in their social position and the problems they face, see John Ogbu, Minority Education and Caste (New York, 1978).

Ethnic status—the sense of being top dog because of one's ancestry—may come from the conquest or earlier enslavement of other ethnic groups or simply from being the original inhabitants of an immigrant-receiving society. To a degree that Weber did not anticipate, such a sense of social superiority could also develop in societies that considered themselves ethnically homogeneous, placed great value on this lack of diversity, and were therefore unwilling to receive ethnic strangers into their national community—one thinks of the history of Australia and Japan, for example. Lest we fall into essentialism, however, we have to bear in mind that the operative group definitions and boundaries are not fixed but are in fact constructed or reconstructed in response to changing historical circumstances. At the same time, we must avoid going to the opposite extreme and overestimating how easily they change or how directly responsive they are to short-term historical developments. Constructed racial categorizations may endure for very long periods, as the career of the patently illogical "one-drop rule" for defining African American ethnicity clearly exemplifies.^[9]

See F. James Davis, Who Is Black? One Nation's Definition (University Park, Pa., 1991).

Racism, then, can be defined as an ethnic group's assertion or maintenance of a privileged and protected status vis à vis members of another group or groups who are thought, because of defective ancestry, to possess a set of socially relevant characteristics that disqualify them from full membership in a community or citizenship in a nation-state. A racist society functions like a private club in which the membership conceives of itself in a certain way and excludes those who do not fit in. (This analogy is especially apt, because under the "black ball" system not all members have to be strongly prejudiced against an applicant for membership; they merely have to defer to the prejudices of others.) Such a sense of ascribed identity and entitlement naturally inclines its beneficiaries to defend their group position if they believe it to be threatened. Many years ago the Weberian sociologist Herbert Blumer caught the essence of racism when he described race prejudice as an anxious sense of "group position."^[10]

Fredrickson, The Arrogance of Race, 210-211; Herbert Blumer, "Race Prejudice as a Sense of Group Position," Pacific Sociological Review 1 (1958), 3-7.

In contrast to traditional ways of defining racism, this definition puts less emphasis on how the alleged deficiencies of the Other are described and explained and more on how a group defines itself and its prerogatives. The essence of racism is caught by such old American expressions

as "give him a white man's chance" or "she's white, free, and twenty one." Historian David Roediger has recently suggested that the key to racism in the United States is an understanding of the meaning of "whiteness" rather than blackness. If we examine the concept of whiteness, we find that it has no specific cultural content, but exists solely as an indicator of higher status than those designated as black or colored.^[11]

See David R. Roediger, The Wages of Whiteness: Race in the Making of the American Working Class (London, 1991); and Toward the Abolition of Whiteness: Essays on Race, Politics, and Working Class History (London, 1994).

I would simply add that an ascribed status of whiteness signifies that the bearer comes from preferred stock, whereas one of blackness or brownness implies a debased and unworthy ancestry. Whites may find it natural that blacks are underrepresented in positions of prominence because such roles have always been occupied by people like themselves and are therefore legitimately inherited from their ancestors—even if they have no firm basis for believing that blacks would not perform as well as their own "kin." This sense of inherited entitlement may be the hard core of racism. When combined, as it usually is, with motives of a more mundane and material kind, it becomes a powerful source of social and cultural identity.

Racism as a general phenomenon is not therefore defined by any specific set of beliefs about what makes a given minority undeserving of equal treatment. We know from the history of anti-Semitism and anti-Japanese discrimination in the United States that racism of a virulent sort can be directed at groups believed to be superior, at least in their competitive efficiency, to an in-group seeking to protect its position. Using this definition, we would have no problem in considering the South African regime of the late 1980s to be racist even though it was edging toward a willingness to "share" power with Africans and refrained from invoking innate racial superiority to justify its presumption that whites must retain de facto social and economic dominance in a reformed, "multiracial" South Africa. Similarly, those opponents of antidiscrimination or affirmative action programs in the United States who, implicitly or explicitly, base their resistance on fears of losing something to which they feel entitled by ethnicity or ancestry can be described as racists despite the fact that they talk about acquired culture or competence rather than genetics. Even Brazil, that allegedly most nonracist of color-differentiated societies, displays an element of status consciousness based on ancestry, as is clearly evident when a person of dark complexion finds that he or she must have more money or education than a white to attain a comparable social position.

If Weber's concept of ethnic status helps us to understand racism in a general and theoretical way, it is not sufficient in itself to make sense

of the history of racism in different societies. The contribution of ethnic consciousness to the grounding of social hierarchies is a variable and not a constant. Weber acknowledged that in modern capitalist societies class can overwhelm or subsume status, and some of his followers took this to mean that "modernization" meant the decline of "ascription" based on race or ethnicity in favor of a social hierarchy determined purely by "achievement." But it is obvious to anyone reading the newspapers or watching television that domineering and even genocidal ethnicities persist and flourish in the modern world. The analysis of particular cases often reveals that the material and political interests of an elite are responsible for arousing the latent ethnic consciousness of the masses. Hence it is not a matter of differentiating sharply between "class" and "race" situations. All that Weber's theory really requires is that one make an analytical distinction and not reduce one tendency to the other. Demagogues may encourage racism for their own ulterior ends, but they could hardly succeed in doing so unless the attitudes of the people they sought to influence provided fertile ground for sowing the seeds of ethnic hatred.

The histories of the United States, Brazil, and South Africa suggest that the color-coded variety of racism varies significantly in intensity and in function within specific social structures, economies, and cultures. What accounts for differences in the nature, strength, and consequences of racism in these societies? How do we explain the growth or decline of racist attitudes and policies within a nation's history? Anyone who has lived in the United States for the past four or five decades should realize that racism—or status consciousness based on race—changes over time in its strength and capacity to shape a social order. Blacks are far from equal in American society, but their status has improved in significant ways, and white racism, while still very much alive, has declined in power and intensity. Recent signs of retrogression have led some frustrated proponents of full racial equality to charge that "nothing has changed," but few African Americans who grew up in the South before the 1960s would find it of little significance that they are no longer disfranchised, segregated in public places, and exposed to lynch law. How do we account for such changes?

Variability and change in the salience of ethnic status and consciousness depend to a considerable extent, it would appear, on variations or changes in the power relationship among ethnic groups. To the degree that an oppressed and stigmatized group can somehow gain in physical resources, political power, and cultural recognition or prestige,

it can induce or force a dominant group to share some of its rights and privileges. This in turn can gradually erode the material and even the psychological foundations for the sense of a Herrenvolk, or dominant race, that it has clearly defined borders and a collective sense of entitlement. Unfortunately, the process is reversible; loss of power, for whatever reason, normally entails a loss of status or prestige. Emancipation from slavery or other forms of directly coerced labor does not by itself empower a group to challenge its subordinate status and the stigma that continues to be associated with servile ancestry, but it does unsettle the power equation by opening new possibilities for action to challenge the racial order as well as new dangers of marginalization, expulsion, or even extermination.

Many historical examples can be offered from the history of the United States, Brazil, and South Africa to show how political or economic power affects ethnic status. One of the most important reasons why free people of color in Brazil had greater opportunities for upward mobility than their American counterparts in the era of slavery, and were thus in a position to win a greater degree of social acceptance, was the vital role that they played in the plantation economy as growers of foodstuffs, herders of livestock, and catchers of escaped slaves. The "free Negroes" of the Old South could not play such a role because there was a large population of nonslaveholding whites to service the plantation economy. But the acquisition of political power by southern African Americans during the Reconstruction era gave them, during the relatively brief period when they could exercise their right to suffrage, an influence over public policy greater than that enjoyed by freedpeople in Brazil after their emancipation, which was completed in 1889. (Since the right to vote in the Brazilian republic depended on literacy until quite recently, most ex-slaves and their descendants have been prevented from exercising it.) In the Reconstruction period in the South, laborers' lien laws, giving the worker priority over the merchant in the division of the planter's crop after the harvest, were a tangible result of this temporary gain in political leverage.^[12]

See Carl Degler, Neither Black nor White: Slavery and Race Relations in Brazil and the United States (New York, 1971); Marvin Harris, Patterns of Race in the Americas (New York, 1964); and Eric Foner, Reconstruction: America's Unfinished Revolution (New York, 1988).

But nothing is more dangerous for a racialized caste fighting for equality than a partial and precarious accession of power. Caste consciousness dies hard, as the example of India's tortuous efforts to elevate the status of Untouchables demonstrates. Consequently, the new order must be firmly and consistently enforced over a long period if a successful backlash is to be prevented. Such persistence might require a national government acting resolutely in the name of an egalitarian ideology

that has the capacity to override the racist attitudes of the majority. Otherwise, popular resistance to minority rights can readily erase most of the gains made under an earlier dispensation of national power. The decline and fall of Radical Reconstruction in the southern states is the prime example of such a retreat in the face of resurgent racism. The assertion of federal power on behalf of equal rights was inadequate and short lived, partly because the ideology available to sanction the effort was not rooted firmly in the kind of radical democratic values needed to sustain it.

The subsequent disfranchisement of southern blacks after white supremacists regained control was congruent with efforts to place them at the mercy of employers or landlords and restrict their opportunities to acquire wealth and property or to follow occupations other than sharecropper, laborer, or servant. The partial success of this effort made it possible for whites to stereotype turn-of-the century blacks as radically and irremediably inferior. Only when blacks migrated in great numbers to the relatively freer atmosphere of the North after 1914 did they begin to acquire the resources and political influence to challenge the Jim Crow system and begin to elevate their ethnic status in ways that eventually impelled whites to abandon their claims to a racial hierarchy sustained by law.

Recent developments in South Africa also reflect changes in the racial power equation that have culminated in the presidency of Nelson Mandela and the achievement of one-person one-vote, a process that began with the government's decision in 1989–1990 to negotiate with the African National Congress. These advances are due in large part to the leverage that black protesters gained in the 1980s over the South African economy—internationally through the antiapartheid movement's promotion of sanctions and disinvestment, and domestically through their growing influence as organized workers or consumers and their resistance to white rule, which made the country virtually ungovernable and thus undermined the security and prosperity of the white minority.

Such examples show that racism, however tenacious, is not a constant and unalterable influence on public action in ethnically divided societies. For Weber, status was only one of three analytically distinguishable but overlapping and interacting sources of social power and inequality. Others were "class" as determined by objective relationships to the market and "party," meaning ability to influence public decisions through political organization and access to suffrage and officeholding.

As Weberian social scientists have demonstrated, inequalities of class, status, and party do not exactly coincide. A main historical dynamic is the interaction among these forms of social power—that is, the manner in which social stratification of one kind conflicts with the others or reinforces them, as the case may be.^[13]

For an example of the effective use of the Weberian triad to analyze race politics, see Ira Katznelson, Black Men, White Cities: Race, Politics, and Migration in the United States, 1900-1930, and Britain, 1948-1968 (London, 1973).

Under a system of racial slavery, there is of course little or no contradiction; the three Weberian categories coincide almost perfectly. Blacks in the Old South had almost no access to social prestige, government, or the marketplace. After emancipation, however, the three types of inequality could vary independently. At the turn of the century, the Jim Crow era reached its height—almost, but not quite, approximating the early pattern of total subordination. Because there was now an emerging black middle class, albeit restricted to a segregated economy, there was no longer such a close fit between class and racial caste as during the slave era. Furthermore, as we have seen, the exclusion of southern blacks from American politics was mitigated to some extent by accelerating migration to the North where the right to vote and hold office persisted.

But the ethnic status of blacks in the nation as a whole may have been at a low point in the period between 1910 and the Great Depression, as reflected in the generally unfavorable or derogatory stereotypes projected by the dominant culture and in the pervasiveness of social segregation and discrimination. In 1913, the federal civil service was segregated. Also in that period the blatantly racist film Birth of a Nation achieved great popularity; the United States officially admonished French authorities during World War I to discourage fraternization between French civilians—especially women—and black soldiers out of deference to the belief of most white Americans in black inferiority and social unacceptability; and bloody race riots and a resurgent Ku Klux Klan characterized the immediate postwar years. But this was also the period when the NAACP won its first court victories and came close to getting antilynching legislation through Congress. In addition to limited political leverage, black migration to the North brought greater access to industrial jobs and better educational facilities.

The persistence of a sharply defined ethnic status hierarchy in the United States between World War I and the 1940s—a time when blacks were making some economic and political advances—does not prove that ethnic status is unaffected by changes in political and economic empowerment. In the long term, as developments in the 1950s and

1960s demonstrated, substantial and durable gains in one respect can be translated into gains in the others. Racism gains much of its strength and legitimacy from ingrained cultural attitudes, which of course change more slowly than the social and economic structures with which they were once directly and transparently associated. Furthermore, the extent to which racism is rooted in associational preferences derived from an expanded notion of family and kinship probably means that racist thought and behavior is to some degree inevitable in any ethnically diverse society. Racism can, however, be made generally disreputable, thus eliminating most of its socioeconomic effects and transforming its traditional targets into quasi-kin—or even actual kin to the extent that intermarriage becomes acceptable.

The salience and social power of racism clearly vary over time and may decrease significantly in response to changing historical circumstances. It is difficult to maintain at the core of one's identity a culturally sanctioned sense of status in the face of substantial changes in the class, prestige, and power position of a subordinate group. Racism does have a life of its own, but it cannot persist without changing its character and gradually losing some of its force in the face of dramatic and durable improvements in the economic and political strength of a disadvantaged ethnic group. Some might argue that this is getting the cart before the horse, that the empowerment of an oppressed minority requires a radical change in the status-affirming attitudes of a dominant group. But what I am in fact advancing is a kind of interactionist or feedback model of change. Increases in power affect attitudes, and changing attitudes open access to power.

How, one might ask, does such a process get started? Studies of the history of race relations in several societies suggest that something extraneous to the racial order may occur, some larger economic, political, or ideological development that calls for adjustment by the society as a whole in ways that have accidental or unintended advantages for subordinated status groups. The independent effect of nationalism, especially when loyalty to a nation in peril can be construed as requiring an inclusive rather than an exclusive sense of the national community, may be of prime importance.^[14]

On the complicated and sometimes antagonistic relationship between nationalism and racism, see Benedict Anderson, Imagined Communities: Reflections on the Origin and Spread of Nationalism (London, 1983).

Major wars or intense international competition among nations can have such a catalytic effect on race relations. The Paraguayan war of 1865–1870 speeded Brazil on the path to slave emancipation because it became necessary to use thousands of slaves as soldiers and to reward them for their participation by freeing them. Black Americans were of

course freed from bondage as the result of a Civil War that was fought primarily for the preservation of the federal union and not for their liberation. As a result of the necessities and opportunities of war, emancipation and the use of black troops became a means to the end of national integrity. The victories of the civil rights movement a century later were aided, perhaps decisively, by the belief of influential and powerful whites that Jim Crow was a serious liability in America's competition with the Soviet Union for the "hearts and minds" of Africa and Asia. White South Africa became serious about dismantling apartheid and negotiating with African nationalists because black resistance and the threat of international sanctions raised fears that, unless something was done to accommodate blacks within a capitalistic framework, a future South Africa would have no place at all for an affluent and acquisitive white minority.

These examples suggest that in times of national peril or catastrophe, inclusive forms of nationalism, sometimes encouraged by a belief that survival on any terms available requires a redefinition of citizenship, may prove stronger than ethnic status consciousness and open the way to lowering or even eliminating barriers to the participation and empowerment of oppressed racial groups. Such a stretching of the boundaries of national solidarity is not an unambiguous victory over prejudice against the Other, however. It can be a concomitant of intensified international conflict, in which case the burden of Otherness is shifted to some extent from the shoulders of the usual domestic scapegoats to those of foreign nationals.

A more fundamental and less contingent force that undermines traditional racial hierarchies and the status claims they engender are long-term trends in the structure and value systems of modern societies away from "ascription" and toward "achievement" as a basis for status and power. We need not adopt the naive view that these trends are irresistible or that industrial capitalism is a direct and automatic solvent of ethnic stratification; the long career of apartheid in South Africa and of group conflict in Northern Ireland show that this is not the case. Yet we can, nonetheless, recognize that ethnic hierarchies become more problematic and vulnerable when they are the only form of ascribed status that persists in an open and publicly sanctioned way. A key function of explicit racist ideology, in its heyday between the mid—nineteenth century and the Second World War, was to rationalize the conspicuous exceptions to the prevailing Western norm of a liberal, open-class society

that could be found in places like South Africa, the American South, and various European colonies.

But the period since World War II has seen an international revulsion against racism, inspired in large part by the increasing role in international affairs and organizations played by the emerging nations of Asia and Africa. Another important impetus for changing attitudes is the internationalization of capitalist enterprise, a development that has made overt racial prejudices a liability for those who would seek to compete with the Japanese and other Asians for world markets. Vast international inequalities that correlate roughly with color persist, but the current tendency to talk of the resulting conflict as pitting rich nations or parts of the world against poor ones may reflect the nature of this struggle better than the language of race that would have been more appropriate in the age of conquest and colonization.

The trend toward a worldwide struggle based on "class" in the Weberian sense, or between those who have a favored access to markets and scarce resources and those who do not, is to some extent paralleled within industrialized nations with strong traditions of racial or ethnic inequality. It is clear to many observers of contemporary Brazil that the central issue is the vast differential between a rich minority and an impoverished majority. The fact that people of darker skin are disproportionately represented among the poor is evidence of a long history of slavery and racial prejudice but may not be the most productive way to think about the current situation. Emancipating the poor as such would seem to be the main challenge, despite the exposure of subtle but persistent forms of discrimination that has recently compelled Brazilians to recognize that they do not in fact have a "racial democracy."^[15]

See I. K. Sundiatta, "Late Twentieth Century Patterns of Race Relations in Brazil and the United States," Phylon 48 (1987), 62-76.

Even in South Africa an assessment of the prospects for class conflict or accommodation may give a better sense of the forces currently at work than an exclusively racial or ethnic view of the struggle. Racism gave this version of capitalist industrialization its peculiarly segmented quality. But two theoretically color-blind ideologies, free market capitalism and Marxian socialism, came to dominate the two sides in the conflict by the 1980s, gaining in strength at the expense of a statist and corporatist doctrine of white supremacy on the one hand and a racially defined black nationalism on the other. Workers and employers alike seemed increasingly ready to view the struggle in class terms. Contrary to what one might expect from Marxist-Leninist theory, such a redefinition

or reconstruction of the situation increased the prospects for a peaceful transition of power; for history shows that class adjustments and compromises are easier to bring off than the reconciliation of groups that view their differences as primarily ethnic or racial. But South Africa has a long way to go before the dream of "nonracial democracy" is fully achieved. Whites continue to dominate economic life, and many undoubtedly believe that their European ancestry still brings certain entitlements. They may cling to the hope that they can keep the substance of white power and privilege by giving up the trappings and allowing a middle-class black minority to share their advantages.^[16]

On the ideological shifts and their significance, see Stanley Greenberg, Legitimating the Illegitimate: State, Markets, and Resistance in South Africa (Berkeley, 1987), and Julie Frederikse, The Unbreakable Thread: Non-racialism in South Africa (Bloomington, Ind., 1990).

What of current black-white relations in the United States? Historically speaking, racism—in Weberian terms, Euro-American status consciousness—has tended to predominate over any consciousness of class that transcends racial categories. The inability of the southern Populist movement of the late nineteenth century to build an interracial political coalition in the face of its opponents' appeals to racial solidarity is a well-known example of this tendency. Another is the notorious difficulty of uniting white and black workers in a collective struggle for class interests, as reflected in the long history of antiblack discrimination by organized labor and the failure of socialist movements to attract substantial black support. During the height of the great depression, when it appeared to many that there were unprecedented opportunities for class action across racial lines, W.E.B. Du Bois was driven, despite his sympathy for Marxism, to espouse the economic self-segregation of blacks, because he despaired of the capacity of white workers to overcome their cultural racism. Three decades later, after the civil rights movement had freed southern blacks from de jure segregation and de facto disfranchisement, advocates of black power and black nationalism came to a similarly pessimistic assessment of American society's ability to overcome racial segmentation. The Kerner Report's 1968 description of the nation as "moving toward two societies, one black and one white—separate and unequal" reflected a general sense in the late 1960s that the United States was still a society stratified by race or ethnic status and not merely by economic or class differences.^[17]

W.E.B. Du Bois, Dusk of Dawn: An Essay toward the Autobiography of a Race Concept (New York, 1940), chap. 7; Report of the National Advisory Commission on Civil Disorders (New York, 1968), 1.

Ten years later, however, a leading black sociologist, William Julius Wilson, argued that race was declining in significance and that the situation of blacks in American society could now best be approached in terms of class. Wilson, who refined and elaborated his argument in a

later work on the black "underclass," based his case primarily on the growth of a substantial black middle class that he believed was being successfully integrated into the larger American middle class. This was in sharp contrast to the earlier situation of black elites who had been condemned by racism to seek higher status exclusively within the segregated African American community. But the price of this desegregation of elites was that blacks who could not qualify for middle-class opportunities because they lacked skills, education, and employment possibilities were stranded in the ghettos without middle-class leadership or behavioral examples. Consequently, their condition had worsened, and the nation faced the major social problem summed up in the phrase "black underclass." Underclass disabilities, Wilson concluded, were primarily a matter of class rather than race and needed to be addressed as such, mainly through social democratic or New Deal-type policies.^[18]

William Julius Wilson, The Declining Significance of Race: Blacks and Changing American Institutions (Chicago, 1978); and The Truly Disadvantaged: The Inner City, the Underclass, and Public Policy Chicago, 1987).

Wilson's theories are controversial and have been sharply criticized by sociologists and historians who believe that racism is not only alive and well but perhaps even stronger than ever. My own opinion is that the class-based, social democratic policies Wilson recommends would do much to alleviate inner-city poverty and demoralization. Also his distinction between race and class as variable determinants of black inequality is a valuable clarification of the theoretical issues involved in the study of American inequality. But Wilson overstated his case for the decline of racism, at least in his original formulations, and has not always made it sufficiently clear that he has identified an uneven and reversible trend rather than an accomplished reality.

It is certainly true that the black middle class suffers less than in the past from specifically racial discrimination, but the affirmative action policies that made such advancement possible are now in grave danger. Black achievers and aspirants for middle-class status on our campuses sometimes face harassment from white students who resent what they view as special privileges or unfair advantages for African Americans, and subtle but effective forms of discrimination persist in access to skilled jobs and white-collar employment, the competition for status and influence within corporate bureaucracies, housing opportunities, and the treatment of black customers in stores, restaurants, and other places of public access. A majority of white Americans still regard interracial marriage as undesirable and a substantial minority would like to outlaw it. Middle-class assimilation, in other words, is not as complete or as certain as Wilson sometimes implies.

Furthermore, it would be hard to deny that the black underclass is feared and despised by many whites not merely for its poverty and statistical propensity to commit crimes or use drugs but also for reasons of race. The Willie Horton stereotype, as employed by the Bush campaign in 1988, was racially charged and not merely the product of class anxieties. What has changed is the ability of education and wealth to compensate to some extent for the stigma of African appearance and ancestry. (Although it might be hard to convince a middle-class African American trying to hail a taxicab in a big city of this fact.) But to be both poor and black is to be doubly disadvantaged. Until this changes, it cannot truly be said that race is no longer significant and that the United States can confront its inequalities exclusively in terms of class.

Wilson's problematic interpretation of current race relations may, however, have some value as prophecy. The deepening economic deprivation and insecurity that relatively large numbers of blacks share with somewhat smaller proportions of other racial and ethnic groups make it conceivable that a sense of class division could eventually eclipse race consciousness as the main source of public conflict in American society. Status based on race and the politics of status protection stubbornly persist and may even increase in hard times—as recent events suggest—but they lack ideological legitimacy and no longer, as in the past, clearly sustain a functional segmentation of labor based on race. Opportunities for the construction of class and the deconstruction of race may now exist to an unprecedented degree, because blacks are no longer consistently and categorically relegated to lower-caste occupations, a development that may change basic social alignments and make class-based responses to the growth of social and economic inequality more likely.

But a sense of racial status or caste is very difficult to eradicate, and the opposite extreme of a homegrown fascism based on a heightened Euro-American status consciousness could also result from the struggle over a shrinking economic pie. Working- and lower-middle-class whites who are struggling to make ends meet or to find and keep a decent job have a tendency, especially when nudged in this direction by right-wing politicians, to blame their hardships on blacks and other racialized groups—which are allegedly benefiting unfairly from affirmative action or being supported in idleness and immorality by the welfare system—rather than on the behavior of large corporations and the operations of the market economy. But it may not be too much to hope that such appeals to ethnic status consciousness will eventually wear thin in the face

of growing evidence that blacks have nothing to do with declining wages and emigrating jobs.

A comparative historical perspective permits the hope, if not the confident expectation, that a plausible combination of circumstances and initiatives could lead to the end of racism as a significant determinant of inequality in the United States. Antiracists should be prepared to capitalize on the opportunities for class-based progressive action that will undoubtedly present themselves. To paraphrase Martin Luther King, Jr., Will it matter so much what kind of ship our ancestors came over in, when we realize that we are all in the same boat now?