Stay informed: Sign up for eNews Subscribe

Whitewashing Race The Myth of a Color-Blind Society

Read Chapter 1

Of Fish and Water

Perspectives on Racism and Privilege  

There ain't no white man in this room that will change places with me—and I'm rich. That's how good it is to be white. There's a one-legged busboy in here right now that's going: "I don't want to change. I'm gonna ride this white thing out and see where it takes me." - Chris Rock


According to a well-known philosophical maxim, the last thing a fish notices is the water. Things that are unproblematic seem natural and tend to go unnoticed. Fish take the water they swim in for granted, just as European Americans take their race as a given, as normal. White Americans may face difficulties in life—problems having to do with money, religion, or family—but race is not one of them. White Americans can be sanguine about racial matters because their race has not been (until recently) visible to the society in which they live. They cannot see how this society produces advantages for them because these benefits seem so natural that they are taken for granted, experienced as wholly legitimate.They literally do not see how race permeates America's institutions—the very rules of the game—and its distribution of opportunities and wealth.

Blacks, Latinos, and other people of color in the United States are racially visible, and everyone seems to notice their race. For them, the same culture, law, economy, institutions, and rules of the game are not so automatically comfortable and legitimate. In a white-dominated society, color brings problems. And if people of color cry foul, if they call attention to the way they are treated or to racial inequality, if they try to change the distribution of advantage, if they try to adjust the rules of the game, white Americans (whose race and racial advantage are invisible) see them as asking for special privileges. They are seen as troublemakers.

What this means is that there is no such thing as a "view from nowhere"—to use Thomas Nagel's apt phrase. People's perspectives on race reflect their experience on one side of the color line or the other. Whites routinely misperceive the reality of black lives. For example, even though blacks are about twice as likely as whites to hold low-paying jobs and are more than twice as likely to be unemployed, 50 percent of whites say the average black is about as well off as the average white person. (Blacks, on the other hand, tend to be more realistic and accurate in their perceptions of their economic status relative to whites.) If white Americans make no effort to hear the viewpoints and see the experiences of others, their awareness of their own privileged racial status will disappear. They can convince themselves that life as they experience it on their side of the color line is simply the objective truth about race. But while this allows them to take their privileged status for granted, it also distorts their understanding. This error poses serious problems for conservatives' analysis of racial inequality.

Of course, individual views within racial groups vary. Not everyone who shares the same subjective perspective will draw the same conclusions about policy. But any perspective that is unreflectively locked inside its own experience is limited, and this is particularly so when that perspective reflects the dominant culture. Failure to understand that they take whites' racial location for granted leads racial realists to ignore the ways in which race loads the dice in favor of European Americans while simultaneously restricting African Americans' access to the gaming table. White privilege, like the water that sustains fish, is invisible in their analysis.

This chapter is about perspective, and how definition—the power to name—determines perception, and ultimately, prescription. It traces the difference it makes if one group's perspective pervades almost everything, from culture to law. Apostles of the new perspective on race insist that racism is primarily a thing of the past. They come to this conclusion because they filter their evidence and their judgment through an outdated, discredited understanding of racism as intentional, obvious, and individual. These misconceptions are not unique to any particular writer or writers. Many white Americans and American institutions, including the current Supreme Court majority, hold parallel views. Because racial conservatives ignore the variability of racial reality in America, they do not recognize that racism is lodged in the structure of society, that it permeates the workings of the economic, political, educational, and legal institutions of the United States. Without that recognition, however, we will be unable to resolve the pernicious problems of race that confront us as Americans.


Conceptions of Race and Racism after the Civil Rights Revolution

In the new conventional wisdom about race, white racism is regarded as a remnant from the past because most whites no longer express bigoted attitudes or racial hatred. The Thernstroms note that despite black riots and crime in the streets in 1968, "nowhere in the voluminous polling evidence available for these years is there any sign that whites were drifting in the direction of the virulent anti-black sentiments so prevalent in the 1940s and 1950s." Indeed, the real story for most whites is that racism has almost disappeared. Marianne Means flatly asserts, "We all agree that slavery was evil. But the blood of slavery does not stain modern mainstream America." The Thernstroms concur. "White racial attitudes have truly altered," they write. "Whites with a pathological hatred of African Americans can still be found, of course. But the haters have become a tiny remnant with no influence in any important sphere of American life."

Racial realists conclude that racism has ended because of the massive change in white attitudes toward blacks over the past sixty years. For example, more than half of all whites once believed that blacks were intellectually inferior. In 1994, however, only 13 percent of whites believed that blacks had "less in-born ability to learn" than whites. Whites also used to favor school segregation by overwhelming majorities, but now 90 percent favor school integration. In the 1940s whites believed they should be favored in competition for jobs. Today, on the other hand, whites unanimously agree that "blacks and whites should have an equal chance to compete for jobs." The Thernstroms go so far as to assert that white attitudes had already changed for the better before the civil rights movement erupted in the 1960s.

To racial realists, this evidence means that the color line has been radically altered. Although many whites still accept one or more negative stereotypes about African Americans, a recent study by Paul Sniderman and Thomas Piazza asserts that only 2 percent of the population could be considered old-fashioned bigots who subscribe to a large number of racist stereotypes. Consequently, it is rare today to find cases of discrimination such as the ones involving Texaco's executives calling African Americans "black jelly beans," a member of the Dallas school board referring to African Americans as "niggers," and the "raw racism" experienced by black secret service agents in a Baltimore Denny's restaurant.

The evidence cited by racial realists indicates that they, like many whites, use a particular understanding of racism. This notion assumes that racism is motivated, crude, explicitly supremacist, and typically expressed as individual bias. Racism, in short, is a form of "prejudice." Paul Sniderman and Thomas Piazza define it as "a consistent readiness to respond negatively to a member of a group by virtue of his or her membership in the group, with the proof of prejudice being thus the repetitiveness with which a person endorses negative characterization after negative characterization."

Given this concept of racism and the use of opinion surveys to measure it, one should hardly be surprised that many people believe racism is a thing of the past. After all, virulent antiblack sentiments have diminished, formal barriers based on malicious intent have in large part been dismantled, and few Americans would accept publicly sanctioned racial barriers today. Were these its undisputed characteristics, one might be tempted to agree that racism is obsolete.

The law institutionalizes the American ideal of equality, and it provides remedies for those hurt by bias. Current law embraces the concept of racism as intentional individual prejudice, and also its corollary—that whites today are often unfairly accused. Evolving doctrine in racial discrimination cases reflects what Angela Harris has called an "essentially moralistic" view. In several reverse discrimination lawsuits, for example, the Supreme Court has explicitly worried that affirmative action plans impose unacceptable burdens on "innocent" third parties (read whites). In equal protection cases, the Court has increasingly emphasized invidious intention as a necessary element for finding actionable discrimination.

But this perspective has its critics. Twenty-five years ago, Alan Freeman documented how, after a brief period of attention to what he called a "victim perspective" in the jurisprudence of equality, the Court moved decisively to adopt a "perpetrator perspective" on issues of race. Adopting the perpetrator perspective means looking at contested race issues from the vantage point of whites. The "perpetrator perspective" in law, like the conservatives' understanding of racism, is preoccupied with white guilt or innocence. It largely ignores whether people of color have suffered injury or loss of opportunity because of their race. Other critics have raised analogous arguments, paying attention to group subordination or disadvantage. Ignoring these analyses, the courts have extended and deepened their attachment to the perpetrator perspective as the racial law of the land.

The Supreme Court's standard for white innocence is very low. Before the modern civil rights era, the Supreme Court often insisted that analysis of motive was inappropriate in constitutional adjudication. During the past several decades, however, the Court has increasingly required that plaintiffs in equal protection discrimination cases not only may, but must, probe defendants' motives. To be successful, plaintiffs must prove specific and conscious bad intentions, the equivalent of the concept of racism as personal prejudice. Under the equal protection clause of the Fourteenth Amendment, the Court holds it is not enough to show that people would reasonably know the discriminatory consequences of their actions. Nor is it enough that actors foresaw the predictable effects of their actions and still proceeded in spite of them. To gain or sustain a remedy for racial injustice, litigants must meet a very high standard: they must show specific discriminatory purpose or malice. Reva Siegel argues on the basis of credible evidence that the Court knew this was a level of responsibility plaintiffs would "rarely be able to prove."

Under congressional statutes, the role of intent is somewhat reduced. The Court has sometimes said that proof of employment discrimination may be based on a demonstration that policies have a disparate impact rather than on a showing of intent—proving, for example, that African Americans or other racial groups are more likely to be disadvantaged by an employment practice than whites. Although the courts give lip service to unintentional bias in cases involving claims of discriminatory treatment, particularly in employment, most of the governing precedents require that plaintiffs prove intentional bias. In 2001 the Supreme Court further extended that requirement. It held that under Title VI of the Civil Rights Act, which prohibits the discriminatory use of federal money, proving disparate impact would no longer be sufficient to win discrimination suits by private parties against federally funded contractors or institutions.

The Court now requires proof of invidious intention in most cases of racial discrimination. It does not apply this standard of intent in age discrimination cases where the relevant statutory language is identical to that in Title VII. In these cases, the courts have accepted a distinction between motive (a factor in causing action) and intent (a specifically proven state of mind) that is more favorable to plaintiffs alleging discrimination. In other settings that address harms caused by others, such as personal injury law, courts assess liability and compensate victims not simply for intentional harms but also for injuries caused accidentally, that is, negligently. Plaintiffs do not have to prove malice or purpose unless they seek punitive damages.

Choosing to make the specific intentions of identifiable individuals the criterion of racism is neither neutral nor appropriate. It is self-aggrandizing and misguided to judge others by their actions but ourselves only by our intentions. In Supreme Court decisions and in the minds of many whites, the relevant "ourselves" are predominantly white or, in Freeman's phrase, potential "perpetrators." Many whites want to determine whether racism exists by exploring their explicit personal intentions. If we are deciding whether to put someone in jail, then assessing his intentions may be appropriate. However, where disputes do not involve criminal charges but rather decisions about social, educational, welfare, or employment policy, questions of guilt, innocence and punishment are not the issue. No one goes to jail for discrimination. In discrimination litigation, the focus is on the legitimacy and fairness of the distribution of scarce opportunities and resources. To ameliorate injustice and achieve a more desirable state of civil affairs, it is more important to examine the problems of discrimination, injuries, and unfairness than to evaluate the culpability and motives of particular perpetrators.

The Court's narrow definition of discrimination, like the realists' equation of racism with prejudice, severely restricts what counts as bias or as evidence of bias. This definition tends to exonerate whites, blame blacks (by default), and naturalize (render unobjectionable) the broad realities of race-based subordination in the United States. This definition of racism, as we have already noted, is also empirically and conceptually flawed. It depends almost exclusively on attitudinal evidence uncovered by opinion polling. This poses two problems. First, even on its own terms, this interpretation of racism ignores significant research that shows how racist attitudes have persisted. In his recent book The Ordeal of Integration, Orlando Patterson examined a variety of evidence and concluded that "all things considered, it is reasonable to estimate that about a quarter of the Euro-American population harbors at least mildly racist feelings toward Afro-Americans and that one in five is a hard-core racist." This is not a small number. If Patterson is correct, the Thernstroms' "miscreants of the night" are hardly a fringe.

Second, by relying on survey questions written in the 1950s, this research ignores possible changes in the character of racism and is, therefore, incorrectly measuring modern expressions of it. Donald Kinder and Lynn Sanders write that "a new form of prejudice has come to prominence, one that is preoccupied with matters of moral character, informed by the virtues associated with the traditions of individualism. Today, we say, prejudice is expressed in the language of American individualism." Statements about individual failure, in other words, may be racially coded expressions of a derogatory stereotype.

There are also abundant survey data documenting the persistence of widespread racial prejudice forty years after the civil rights revolution. Many writers who use polling data to show the decline of racism cherry pick among these surveys and omit this evidence. Some of the most compelling evidence of tenacious prejudice comes from studies of residential discrimination. In 1992, the Detroit Area Survey found that 16 percent of whites said they would feel uncomfortable in a neighborhood where 8 percent of the residents were black, and nearly the same percentage said they were unwilling to move to such an area. If the black percentage rose to 20 percent, 40 percent of all whites indicated they would not move there, 30 percent said they would be uncomfortable, and 15 percent would try to leave the area. Were a neighborhood to be 53 percent black, 71 percent of whites would not wish to move there, 53 percent would try to leave, and 65 percent would be uncomfortable. A more recent study of four cities (Atlanta, Boston, Detroit, and Los Angeles) yielded similar results. Camille Zubrinsky Charles found that more than half of whites in these four cities expressed a preference for same-race neighborhoods, while blacks expressed a strong preference for integrated neighborhoods.

Contrary to the optimism of racial realists, one finds precious little evidence, even in the polling data they use, that many white Americans believe in integrated neighborhoods, especially if that means a neighborhood with more than a very few black families. Pejorative racial stereotypes are not restricted to one's choice of residence. They continue to be fundamental to (white) American culture. When the University of Chicago's National Opinion Research Center asked people to compare blacks and other ethnic groups on a number of personal traits in 1990, they discovered that 62 percent of nonblack respondents believed that blacks were lazier than other groups, 56 percent stated that they were more prone to violence, and 53 percent thought they were less intelligent. Another report suggests that white Americans are still substantially opposed to intimate contact with African Americans. In one national survey conducted in 1978, 70 percent of whites rejected interracial marriage on principle. This hardly represents the significant change in whites' attitudes trumpeted by the proponents of racial realism.

Both the meaning of survey data and the way they are used by these cheerleaders for racial progress are also problematic. Because the typical questions used to measure changes in racial attitudes essentially gauge how closely attitudes conform to the American creed enshrined in the Declaration of Independence, it is not surprising to find that most (white) Americans sound tolerant. This is because when prejudice and tolerance are evaluated by these criteria, the questions assess only whether people subscribe to American ideals. It is hardly a major discovery to find that racism has declined when individuals are asked whether they believe in equal job treatment and integrated schools. Because the ideals of equality and formal tolerance are central to American identity, most Americans know the "correct" answers to such questions. Thus, rather than representing a decline in racism, these polling data actually measure adherence to the principles of American society.

Because most surveys tap only surface commitment or verbal adherence to ideals, polling data may reveal more about the correlation between self-presentation and socioeconomic class than about the persistence of racism. When tolerance means verbalizing principles acquired through exposure to liberal middle-class institutions, lower- and working-class whites will appear to be more racist than middle-class whites. Surveys that find prejudice and intolerance declining among America's white middle class also link racist sentiments disproportionately to poor and working-class white Americans, or to the "lunatic fringe." This finding is not new. As long ago as 1966, Paul Sheatsley found that the highest scorers on his "pro-integration scale" shared three features in common: they attended college, their earnings were high, and they were professionals. But the narrow catch of this racism net reflects only its limited definition of racism. The behavior between classes may not differ much, but, unlike well-educated middle- and upper-class whites, poorly educated working-class white people are nearly precluded from this conception of "tolerance" because they have not learned the "proper" ways to present their racial views to pollsters.

Some writers promoting the new orthodoxy on racial inequality also seem unaware that evidence based on broad changes in opinion is insufficient to assess a complex, multifaceted problem like the persistence of racism. The gap between what people tell survey researchers and what they actually do or believe is wide, and a very different picture emerges when one moves from political abstractions to routine behavior. Discrepancies between racial attitudes and behavior are large and pervasive. White Americans overwhelmingly endorse civil rights principles. When asked, 88 percent of whites in 1978 agreed that blacks have a right to live wherever they want to, up from 76 percent in 1970. By 1980, in fact, just 5 percent of whites were willing to tell a pollster they preferred strict segregation. Yet only 40 percent said they would vote for a law stating "a homeowner cannot refuse to sell to someone because of their race or skin color." White Americans may support the principle of fair housing, but less than half say they are willing to act on this principle. In fact, when actual patterns of racial isolation are examined, it is clear that very few whites prefer integrated to segregated neighborhoods.

American Apartheid, an award-winning study of housing segregation by Douglas Massey and Nancy Denton, reveals just how wide the gap is between attitudes and behavior. Using demographic data about where African Americans and whites actually reside, Massey and Denton demonstrated that levels of residential segregation have hardly changed since the 1960s. Applying a sophisticated index of segregation to thirty metropolitan areas with the largest black populations between 1970 and 1980, they discovered that in northern cities, "this (segregation) index averaged over 80 percent in both 1970 and 1980." The index declined a mere 4 points over the decade of the 1970s and only 2 percent during the 1980s, and most of the decline occurred in small cities with small black populations. Massey and Denton conclude that "blacks living in the heart of the ghetto are among the most isolated people on earth."

The Thernstroms challenge this conclusion, arguing that Massey and Denton exaggerate the persistence of residential segregation. But they provide no counterevidence, nor do they generate demographically grounded indices of integration. Rather, they attempt to refute Massey and Denton with an analysis that is laughable. "The strongest proof that residential segregation has been declining for a generation," they write, "comes from national surveys [that] have intermittently asked blacks and whites whether members of the other race live in the same neighborhood as they do." They find the patterns "striking" and report "fully two-thirds of all African Americans at the time (1964) said that they had white neighbors." The fact, they write, "that the figure was as high as five out of six in 1994" is evidence that residential apartheid has declined. The Thernstroms apparently imagine that people's beliefs about who lives in their neighborhood are a more accurate indication of residential segregation than measures of where and how people actually live.

Similar gaps between attitudes and behavior are found in most contexts where race is an issue. These gaps become especially obvious when the reality of one's everyday life is directly affected. Support for desegregation of schools was relatively free of cost so long as no busing was involved or one's own children attended private schools. On-the-job equality also had an all-American sound to it, especially when there were very few blacks in one's occupation. Upgrading blacks from unskilled to skilled work was a fine goal if one's own work was white-collar or professional. But as black enrollments in prestigious universities and professional schools increased, constitutional amendments eliminating affirmative action became the order of the day. When the demands of people of color hit closer to home and directly affect middle- and upper-class whites, these traditionally color-blind Americans begin to sound distinctly less tolerant and become seriously concerned with the color of people's skin.

These empirical flaws in studies purportedly demonstrating that racism has declined are compounded by fundamental conceptual problems. By now, the prejudice approach to the study of racism has been discredited and has become almost completely obsolete. The challenge to the prejudice paradigm began as early as 1958 when sociologist Herbert Blumer first argued that racism was better understood as a sense of group position than as a collection of bigoted individual attitudes. Since Blumer's groundbreaking article, a long line of sociologists, social psychologists, and legal theorists have moved beyond the outdated notion of racism employed by most advocates of color-blind ideology. Instead of locating racism in intentions, attitudes, and obviously crude supremacist expressions or in pathological individual psyches, these scholars use a more complicated conception. Their analysis assumes that racism is often unintentional, implicit, polite, and sometimes quite normal. They look for racism in behavior as well as in attitudes and find it in culturally and economically produced systems of advantage and exclusion that generate privilege for one racially defined group at the expense of another.

Using this more realistic conception of racism, it becomes apparent that those who argue racism has declined ignore critical evidence that contradicts their assumptions. Their understanding of race paints a one-sided, terribly inaccurate portrait of racism in modern America. A very different picture emerges when racism is understood as a sense of group position and as the organized accumulation of racial advantage, a system best understood by observing actual behavior.


Racial Privilege and Group Position

Because it extends far beyond individual attitudes, permeating the very structure and organization of American society, race strongly determines the ways in which Americans are treated and how they fare. White Americans, whether they know it or not, benefit as individuals and as a group from the present social pecking order. The social, political, and economic benefits of being white encourage white Americans, argues George Lipsitz, to invest in whiteness as if it were a form of venture capital and to work at increasing its value. When it comes to race, white Americans' social choices are very often molded by the relationship between whiteness and accumulated racial advantages.

The possessive investment in whiteness is like property. And as a kind of property, the value of whiteness, as Cheryl Harris points out, lies in "the unconstrained right to exclude" or to deny communities of color opportunity or the chance to accumulate assets. Exclusion, as is evident in the case of residential segregation, is a cardinal principle of white identity. To paraphrase Harris, those who possess whiteness have, until recently, been granted the legal right to exclude others from the advantages inherent in whiteness; they have accumulated wealth, power, and opportunity at the expense of the people who have been designated as not white. In this sense, the experiences of white and nonwhite Americans are intimately connected. The benefits of being white are related to the costs of being nonwhite. This is why it makes more sense to analyze racism in terms of group position rather than in terms of the bigoted attitudes of individuals.

White privilege is pervasive. Most discussions of racial inequality focus on labor markets, the criminal justice system, residential segregation, and education. But race also counts in ways that are less obvious, indeed typically invisible, to white Americans. While often unrecognized, these patterns of racial disadvantage point to the insidiously pervasive power of racism in American life. Because most Americans use such a narrow conception of racism, it is not surprising that they fail to recognize these subtle expressions of racial inequality that are woven into the fabric of society.

To see the pervasiveness of white privilege, consider first something as ordinary as consumer trade. As we noted in the introduction, blacks and other minorities are denied mortgages far more frequently than whites with comparable incomes. But even in other situations, including those where market forces would be expected to eclipse racial factors, race plays a powerful role. Researchers studying automobile dealerships in the Chicago area found, for example, that salespeople offer significantly lower sales prices to white men than to women or blacks, even when economic factors and bargaining strategies are held constant. A more recent study shows that in the 1990s blacks paid significantly more for car loans arranged through dealers than whites did, despite having comparable credit histories. Similarly, clerks in retail stores are frequently more concerned with the color of shoppers' skin than with their ability to pay. Cignal Clothing, a subsidiary of Merry-Go-Round Enterprises, for example, stamped an information form on the back of personal checks. The form included a section marked "race," and shoppers were classified "W" for white, "H" for Hispanic, and "07" for black. Sociologist Joe Feagin, drawing on thirty-seven in-depth interviews with middle-class blacks in several American cities, found widespread evidence that black shoppers were treated less respectfully than their middle-class white counterparts. "No matter how affluent and influential," he reports, "a black person cannot escape the stigma of being black even while relaxing or shopping."

Health care is another realm where significant disparities exist between blacks and whites—disparities that often mean life itself. We have already noted the wide gaps in mortality rates and access to primary care between blacks and whites. Similar disparities cut across every aspect of health and health care, and few of these differences can be fully attributed to social class or genetics. For example, the National Cancer Institute (NCI) recently reported that cancer death rates are increasing much faster for blacks than whites, sometimes by as much as twenty to one hundred times as fast. Black women are more likely than white women to die of breast cancer, even though the incidence of the disease is lower among blacks. According to the NCI report, "Black men have a cancer-death rate about 44 percent higher than that for white men." In fact, African American men between the ages of fifty and seventy are nearly three times as likely to die from prostate cancer as white men, and their prostate cancer rate is more than double that of whites.

Higher death rates for blacks diagnosed with cancer are a recent development. In the 1930s, blacks were only half as likely as whites to die of lung cancer. Since 1950, however, the rate of lung cancer deaths among black men has increased at three times the rate for white men, and age-adjusted figures reveal that the rate was actually 40 percent higher among black men by the 1970s. An increase in smoking rates is not the likely culprit behind the change. Exposure to environmental toxins and carcinogens, which are disproportionately located in poor and minority communities, is one important reason for the racial disparities in cancer mortality rates. Differential access to screening, prevention, and treatment is another reason for the disparities. One of the chief reasons black women are more likely to die of breast cancer is that they are not diagnosed until the disease has reached an advanced and more lethal stage. A study of operable non-small cell lung cancer found that the rate of surgery for black patients was 12.7 percent lower than that for whites with the same diagnosis. The authors of this study concluded that "the lower survival rate among black patients . . . is largely explained by the lower rate of surgical treatment among blacks." Racial differences in mortality rates for cervical cancer remain significant even after adjusting for age and poverty, and are likely attributable to disparities in screening and diagnosis.

Racial disparities in mortality rates for stroke and coronary heart disease are also significant. The black mortality rate for strokes is 80 percent higher than the white rate and the black mortality rate for coronary heart disease is 40 percent higher. Racial differences in hypertension are well documented and are particularly pronounced among low-income African Americans. One study rejected the common assumption that hypertension among blacks is genetic, concluding that socioenvironmental factors like the stresses of low job status and income are responsible for the different rates of hypertension.

Access to sophisticated diagnostic and treatment procedures for coronary heart disease and related ailments also accounts for significant health differences between blacks and whites. Once differences in age, sex, health care payer, income, and diagnoses for all admissions for circulatory disease or chest pains to Massachusetts hospitals had been accounted for, a 1985 study found that whites underwent significantly more angiography and coronary artery bypass grafting than blacks. More recent studies confirm the results. One study, for example, found that after controlling for differences in age, gender, disease severity, comorbidity, geography, and availability of cardiac facilities, blacks were 60 percent less likely to have had coronary angioplasty or coronary bypass surgery and 50 percent less likely to have had thrombolytic therapy. Similarly, researchers who investigated stroke treatments found that "white patients were approximately 50 percent more likely to receive imaging than were black patients"; they also found that of patients deemed appropriate for carotid endarterectomy, two-thirds of white patients but only half of blacks underwent the surgery.

Ironically, amputation of a lower limb is the one advanced procedure that blacks receive far more often than whites. African Americans are more likely to have such last-resort procedures because of inadequate treatment of hypertension and diabetes—illnesses that reflect inadequate care and treatment. This is a perfect illustration of how disaccumulation works: small deficits in health care add up over time, leading to the disaccumulation of health and a perverse outcome.

Neither income nor social class adequately explains these differences in mortality rates and treatment. Rather, the burden of evidence contained in these studies indicates that race is a crucial variable. A recent National Bureau of Economic Research study, for example, found that income inequality between racial groups—not income inequality within racial groups—explains the differences in mortality rates.

Race has a powerful and widespread impact on health treatment and thus health outcomes. Blacks and Latinos are less likely than whites to have access to basic health insurance. Another serious obstacle to quality care for black and Latino patients is that minority doctors, who typically treat disproportionate numbers of minority patients, have greater difficulty than white physicians securing authorization for care. Nationwide, about one-third of black and Latino doctors report difficulty obtaining necessary hospital admissions, compared to one-quarter of white physicians. Racial differences in infant mortality and prenatal care are also linked to a perverse version of racial profiling. Hospitals and clinics with high proportions of minority patients often conduct more systematic and intrusive screening for drug abuse and sexually transmitted disease than do those that treat white women, even though that pattern is not justified by prevailing rates of substance abuse. This in turn discourages many black women from seeking needed prenatal care. Another study found that low-income African American mothers in Chicago who reported being the victims of racial discrimination were twice as likely to give birth to very low-weight babies compared to mothers reporting no discrimination.

Racial bias is another important source of the differences in the ways life-threatening diseases are treated. Recent evidence suggests that racial stereotyping, and even discrimination, influence doctors' treatment recommendations for patients. K. Schulman and his colleagues asked doctors to respond to videotaped interviews with "patients" who were actually actors with identical medical histories and symptoms. Only the race and gender of the actors were different. Doctors turned out to be significantly less likely to refer black women for aggressive treatment of cardiac symptoms than other categories of patients with the same symptoms. Doctors were also asked about their perceptions of patients' personal characteristics. Black male actor-patients, whose symptoms and comments were identical to white male actor-patients, were perceived to be less intelligent, less likely to participate in treatment decisions, and more likely to miss appointments. Doctors in the study thought that both black men and women would be less likely to benefit from invasive procedures than their white counterparts, less likely to comply with doctors' instructions, and more likely to come from low socioeconomic backgrounds. In other words, where actor-patients were identical except for race, black patients were usually seen as low-income members of an inferior group.

Although few doctors may be intentionally racist, not very many are immune to America's racial history and the resulting cognitive bias. In his pathbreaking article on unconscious racism, Charles Lawrence III has observed that "[racism] is part of our common historical experience and . . . culture. It arises from the assumptions we have learned to make about the world, ourselves, and others as well as from the patterns of our fundamental social activities." Because doctors, health insurance officials who authorize treatment procedures, and grievance hearing officers exercise considerable discretion, there is ample room for cognitive bias and stereotypes to influence their decisions. Discretion arises because only a small proportion of medical treatments are scientifically validated, because experts have differences of opinion about appropriate treatment, and because approaches must be individualized for the specific characteristics of each patient. Discretion is inescapable in medicine. But combined with other sources of racial bias, it accentuates differences in treatment and health care. This pattern of racially biased discretion is similar to patterns in the criminal justice system, another institution whose practitioners wield wide powers of discretion.

Sports, a third arena in which race matters, is perceived by many as one of the most meritocratic, color-blind institutions in American life. If there is any realm in which the color line should have disappeared by now, it is professional sports, where measures of achievement are supposedly obvious, numerical, and uncontested. Yet even though 79 percent of National Basketball Association (NBA) players in the 1996-97 season were black, 76 percent of the head coaches were white. By 2001, the proportion of white coaches had dropped to 66 percent, as ten NBA head coaches were black. And although 66 percent of the National Football League (NFL) players in the 1996-97 seasonwere black, 90 percent of the head coaches were white. By the 2000-2001 season, the numbers had not changed; there were still only three African American head coaches, accounting for 10 percent of NFL coaches.

The situation is not much different in college sports. Sixty-one percent of Division I-A male basketball players were black in the 1996-97 season, but 81.5 percent of the head coaches were white. The numbers barely changed at the end of the 2001 season, as the proportion of white head basketball coaches decreased to 78 percent. And although 52 percent of the Division I-A football players were black during the 1999-2000 season, 92.8 percent of the coaches were white. By 2001, nearly 97 percent of the head coaching positions had gone to whites.

These discrepancies are unlikely to even out anytime soon. After the 1996-97 college football season, there were twenty-five openings for head coach of Division I-A teams. Only one of those schools—New Mexico State University—even interviewed a black candidate. During the 1997 and 1998 seasons, thirteen head coaches were named in the NFL, a turnover of almost 50 percent in the thirty-team league. Not one of the replacements was black. The situation did not change much in the next three years. Although the NFL turnover rate was 75 percent between 1998 and 2001, only one African American was hired as a head coach.

Can these racial discrepancies be explained by the concept of merit? Some may think these head coaches got their jobs because they had the best records. The evidence, however, does not support this explanation. There have been only four black head coaches in the history of the NFL. Each of them has either played for or coached on a Super Bowl championship team, or was a college conference coach of the year. By contrast, as of 2001 only thirteen of the twenty-seven white NFL head coaches held this distinction. An analysis of the turnover among NFL coaches at the end of the 1997-98 season makes it obvious that merit is not the sole criterion for being a head coach. The potential pool of blacks has included (to name just a very visible few) Johnny Roland, all-American running back and Pro-Bowler who has been an NFL assistant coach for twenty-two years; Art Shell, former NFL Pro-Bowler with a 56-41 record as head coach of the Raiders and currently an NFL assistant coach; and Sherman Lewis, ten-year offensive coordinator (next in line to head coach) for the Green Bay Packers and an NFL assistant coach for twenty-nine years.

Who was chosen? One thirty-four-year-old with eleven years of coaching experience, two of which were as offensive coordinator, and a forty-two-year-old with four years of experience as an NFL assistant coach and one year as a college head coach. Each of these men had been an assistant coach under Sherman Lewis, who was passed over. Also chosen as head coaches were a former head coach whose previous four years produced records of 8-8, 7-9, 7-9, and 2-6, and ten men over the age of fifty-five with an average record of 6-10. Only one member of this latter "old boys' club" had made the playoffs the season before. All were white. It appears that race matters more than merit in hiring NFL head coaches.

According to a report released in October 2002, African Americans in the NFL are the last hired as head coaches and the first fired. Few of them, the report found, were involved in the interview process. Since 1920, the league has hired more than four hundred head coaches and, as of the end of the 2002 season, eight of them (2 percent) have been African American. "When you see a Denny Green fired after the record he has built and then not get a new job," said attorney Cyrus Mehri, "or Marvin Lewis coach the best defense ever, win a Super Bowl and two years later not have a head job, you know that something is wrong."

A similar pattern is apparent in baseball careers. A study of lifetime pitching and batting averages by sports sociologists at Northeastern University shows that black players have to out-hit and out-pitch their white counterparts by substantial margins to win and keep their jobs. Mere journeymen can have long and profitable careers as long as they are white, but among African Americans, only stellar and above-average players will succeed. Perhaps this explains why there are so few black managers in major league baseball. Baseball typically hires managers, coaches, and front office personnel from the echelon of "good but not great" players. Because most of these players just happen to be white, black ballplayers have difficulty becoming coaches.

Professional sports are not atypical in this regard. In a national project examining the hiring practices of large law firms, Harvard University legal scholar David Wilkins observed that, as in baseball, black applicants with average grades are less likely to be hired than whites with the same records. Black partners are much more likely than whites to be Harvard or Yale graduates. The "black superstar" requirement is most obvious at the most prestigious firms. As one partner at an elite Chicago firm told the researchers, his firm sets "higher standards for minority hires than for whites. If you are not from Harvard, Yale, or the University of Chicago, youare not adequate. You're not taken seriously."

As these examples indicate, race counts very heavily in the ways Americans are treated. Being white, as Chris Rock's fictional one-legged busboy recognized, has its advantages, and being nonwhite has its disadvantages. The problem of race in America is not that people come in different colors; the problem is that people are treated differently according to their color. The most important feature of being white, then, is not pigment, melanin, or skin color. It is, rather, the very close connection between being white and having improved economic opportunities and life chances.


From White Advantage to Racial Subordination: The Reciprocal Nature of Racism

The experiences of white and nonwhite Americans are intimately connected. The benefits of being white are related to the costs of being nonwhite. White Americans are privileged because they benefit from the present social order. As individuals and as a group, they derive advantages from the ways in which race limits the lives of people of color, whether they know it or not.

Because critics of color-conscious policies measure the decline of racism by the absence of crude personal prejudice, they do not recognize or take account of these potent realities. White coaches benefit from the higher standard to which black coaches are held. White Americans' chances of receiving loans are significantly enhanced when competition from people of color is reduced. When white men can buy new cars at markups one-third to half those offered to black men and women, their advantage (estimated at a collective $150 million annually by Yale professor Ian Ayres) is underwritten by race. In an era of cost pressure and scarcity in health care, the white advantage could be said to extend to the gift of life itself.

When economic and political resources are scarce, as most are, the relationship between whites and nonwhites may be zero-sum. Many white Americans are sure their children will lose when people of color demand their fair share of admission to elite universities or professional schools. For them, simply having to compete without the hidden benefits of being white is a significant hardship. Jennifer Hochschild articulates this concern elegantly: "As the number of contestants for a fixed number of prizes increases," she writes, "the chances of winning decrease. The arithmetic is simple: As blacks gain chances, whites lose certainty."

Wins and losses look quite different from opposite sides of the racial divide. They also look different depending on time frame and basis of judgment. Sometimes whites fear that an outcome is zero-sum even if it may not truly be. Access to education looks like a zero-sum game, at least in the short run, as prestigious universities allocate limited places. But in the long run, failure to include people of color will harm everyone by limiting economic growth as well as by intensifying racial strife. Wins and losses can be calculated for a large group to which one belongs (like a race), for one's subsegment of the American population (such as an occupation), or for an individual. These different ways of judging who wins and loses, along with fear and mutual suspicion, make it difficult to assess outcomes consistently. Although they may not recognize it, whites and blacks sometimes find themselves in a lose-lose relationship. No one benefits, for example, when black youths go to jail because of a failure to invest in community social support systems. And if race was recognized and its consequences assessed instead of being ignored, perhaps policies with win-win results could be forged more often.

Whatever might be possible in a better future, today's race hierarchy is a powerful force. Thus whites, aware or not, misguided or not, typically resist change because their privileged status comes with (unearned) advantages. White Americans who believe they will lose if blacks gain are prone to oppose policies designed to reduce racial inequalities. Donald Kinder and Lynn Sanders point out that "insofar as interests figure prominently in white opinion on race, it is through the threats blacks appear to pose to whites' collective well-being. " Perhaps this explains why so many white American men think only of their short-term group interests and therefore oppose affirmative action policies. Because affirmative action eliminates the special advantages they have enjoyed historically, many white men believe they have something to lose when these policies are adopted. They believe this even though there is little evidence that white men lose jobs due to affirmative action.

Racism is related not only to actual privilege. It also entails a commitment to maintain relative group status. What matters is the magnitude or degree of difference that white Americans have learned to expect and maintain in relationship to people of color. A telltale illustration of this occurred when federal officials were trying to desegregate southern hospitals in the 1960s. A southern senator convinced officials in the Office of Equal Health Opportunity to create an exception to the desegregation policy. The exemption he created allowed doctors to place white patients in segregated rooms if physicians were willing to certify that integration would be detrimental to the patient's medical condition. Although very few doctors took advantage of this opportunity, the policy was tantamount to creating a new disease that afflicted whites: racism.

Housing segregation is another, more pervasive, instance of whites establishing status differentials based on race. One expert reported "Whites prefer and are willing to pay more for segregation than blacks are willing [or able] to pay for integration." White people's apprehensions about living in racially mixed neighborhoods underscore this investment in relative group position. A large number of white Americans believe that property values decline as blacks move into a neighborhood. According to a Newsday poll, 58 percent of Long Island's whites felt this way, and another survey found that 40 percent of Detroit's white population also subscribed to this notion. Because a home is viewed not only as a major investment but also as a symbol of one's worth, Massey and Denton contend "these views imply that whites perceive blacks to be a direct threat to their social status." Stanley Greenberg's study of working-class white voters in Michigan confirms this interpretation. "Blacks constitute the explanation for their vulnerability," he writes, "and for almost everything that has gone wrong in their lives: not being black is what constitutes being middle class; not living with blacks is what makes a neighborhood a decent place to live." Bobo and Zubrinsky provide a dramatic example of this expression of racism. Using data from a large multiethnic survey in Los Angeles, they found that, "as the affective difference that whites prefer to maintain between themselves and members of minority groups rises, so does the level of opposition to racial residential integration."


The Politics of Racists and Nonracists

Because white privilege is invisible, it is common to describe "racists" and "nonracists" as very different kinds of people. Racists are characterized by the Thernstroms and other racial realists as deeply prejudiced individuals who express "raw racism," "people who can and will do horrendous things." Nonracists, on the other hand, are said to accept the principles of the civil rights movement and display few, if any, traces of prejudice. In this view, racists today are the exception and nonracists the rule. White Americans may disagree with blacks about appropriate civil rights policies—46 percent of whites, for example, think government should "ensure fair treatment of blacks," compared to 90 percent of blacks—but supposedly their opposition has nothing to do with racism. Instead, as Paul Sniderman and Thomas Piazza insist, these differences are understood as a matter of principle. "The politics of race," they write, "now has a moral bite to it that it previously lacked; for it is no longer simply a matter of rejecting prejudice in favor of the [American] creed but of rejecting key elements of the creed itself."

Conservatives like the Thernstroms make nonracism the norm and racism the exception. But drawing any sharp line between racists and nonracists is a slippery business. No doubt some racists are a disturbed bunch of people whose crude talk about people of color (as well as about women, Jews, and homosexuals) is repulsively frightening. What is striking, though, is the similarity between the behavior of those who voice blatantly racist sentiments and the so-called nonracist discourse and politics of self-styled conservatives and centrists. Putative nonracists often act like racists. Until recently, for example, former Senate majority leader Trent Lott (R-Miss.) and Congressman Bob Barr (R-Ga.) were closely associated with the Council of Conservative Citizens, a right-wing, prowhite political group. Before the Washington Post exposed this group's racist views, Lott told its members, "The people in this room stand for the right principles and the right philosophy." This was not the first nor the last time Lott expressed sentiments that blurred the distinction between conservatism and not-so-subtle racist appeals. But a later statement cost him his position as Senate majority leader. "I want to say this about my state," Lott said, at a celebration of Senator Strom Thurmond's one hundredth birthday in December 2002. "[When Thurmond] ran for president we voted for him. We're proud of it. And if the rest of the country had followed our lead, we wouldn't have had all these problems over all these years, either." What was Lott so proud of and to which problems was he referring? Senator Thurmond left little to the imagination in his 1948 campaign against Harry Truman. "On the question of social intermingling of the races," Thurmond declared, "we draw the line. And all the laws of Washington and all the bayonets of the Army cannot force the Negro into our homes, into our schools, our churches and our places of recreation and amusement." Lott is hardly the only respectable Senate conservative who smudges the line between racists and nonracists. Asked in 1994 by one of his Montana constituents, "How can you live back there [in Washington, D.C.] with all those niggers?" Senator Conrad Burns recalls he told the rancher it was "a hell of a challenge." Three years earlier the senator invited a group of lobbyists to join him at an auction. Asked what was being auctioned, he answered, "Slaves." Nor does one need to be white to conflate the meaning of racist and nonracist. "Supporting segregation need not be racist," black conservative Ward Connerly is quoted as saying. "One can believe in segregation and believe in equality of the races."

Because politicians use coded language, the assumed differences between bigots and nonbigots are sometimes difficult to locate. It was not Klansmen who put an anti-immigrant initiative on the ballot in California. It was so-called moderate Republican men. And Republican politicians have repeatedly succumbed to the temptation to run race-baiting campaigns. It was not George Wallace who poisoned the 1988 presidential campaign with the notorious Willie Horton ads but an establishment Republican. And it was not a member of the KKK who defended the Confederacy to the Southern Partisan, a neo-Confederate magazine. It was John Ashcroft, the current United States Attorney General. People who do not show up as bigots in attitude surveys sometimes behave like bigots.

When a theory assumes bigots and nonbigots are quite different but does not distinguish between them very well, how should one differentiate between "racists" and "nonracists"? Does one focus on the differences between racists and nonracists, or on their similarities? Does one define racism as virulent antiblack sentiments and a pathological hatred of African Americans or, to use Melanie Kaye-Kantrowitz's words, as "a system that normalizes, honors, and rewards whiteness?" Does one treat racists as exceptional or normal? Does one treat "racist" accounts of white supremacy as lunacy, or merely as expressions of American self-portraiture from another era? One approach finds racism in only a tiny remnant of the white population who explicitly endorse prejudiced beliefs; the other casts a wider net, finding expressions of racism among corporate executives, national politicians, and university regents.

Arguments that demonize racism and treat it as the exception lose sight of the complicated and subtle workings of being white in America. A focus on obvious bigotry, crude verbal performance, and political practices may make American "nonracists" feel better about themselves. But it also produces a false sense of security. Because it ignores culturally acceptable sophisticated forms of racism, this perspective is unable to detect the "nonracist" ways that being white works to the advantage of European Americans. Opponents of policies that undermine white people's privileges do not use Klan ideology to justify their opposition. Instead, they invoke the principles of American political beliefs. Not everyone who opposes color-conscious policies does so with the intention of defending white privilege. But one cannot assume, as all too many critics of color-conscious policies do, that opposition to affirmative action is based entirely on the principles of fair play and individual merit. Much of the opposition is based on resentment toward blacks, and this resentment is driven by a fear (conscious or not) that the interests of whites as a group are jeopardized by color-conscious policies. Because color-blind policies are cast as a defense of individualism, the group interests at stake are concealed. But this move poses a more insidious problem than the raw racism of bigots. People voicing virulent antiblack sentiments are an easy target, but restricting racism to them leaves the institutionalized benefits of being white invisible and untouched.


Racism and Law: The Maintenance of White Privilege

The law and legal institutions normalize white advantage by articulating and enforcing cultural norms, which help to maintain the racial hierarchy in the United States. At first, this seems odd. After all, in the 1950s and 1960s, federal courts helped dismantle state-sanctioned racism. The courts, however, have been ineffective in addressing contemporary racial inequality because equal protection doctrine treats individual bigotry as the core of racism. The law's insistence that intention is the sine qua non of race discrimination matches the opinion of many Americans. But this search for individual blame is psychologically naïve, and it obscures the complex sources and relationships that produce racial inequality. As Angela Harris explains,

Translated into constitutional law, this model . . . works to identify intentional wrongdoers . . . but leaves untouched unconscious racism, everyday cognitive bias and institutional structures that faithfully perpetuate patterns of racial subordination. As the legal structures that continue to disadvantage people of color become increasingly "race-neutral'" in a constitutional sense, the moral model of discrimination facilitates both the denunciation of bigotry and the maintenance of existing distributions of wealth and power. (Footnotes omitted; emphasis added.)

The face of racial subordination today is residential segregation, unequal loan policies, differential police stops, divergent medical care and schooling, variation in criminal sentencing, and disparate administration of the death penalty. Absent a smoking gun of intentionality, constitutional challenges to these forms of racial inequality are impossible.

In addition to the intention requirement, the Supreme Court's response to proposed remedies for racism poses another formidable obstacle to meaningful change. When private or public organizations set out to correct historic racial disparities, they typically institute some race-conscious remedial plan. But because such plans classify people based on race, the courts routinely strike them down. Even though these race-conscious plans aim to help subordinated groups, the courts believe they constitute reverse discrimination. Under the resulting color-blind norm, lawyers rarely succeed in justifying affirmative action plans that seek to remedy actual racial disparities and societal discrimination. As Reva Siegel points out, the result is that "doctrines of heightened [judicial] scrutiny function primarily to constrain legislatures from adopting policies designed to reduce race and gender stratification, while doctrines of discriminatory purpose offer only weak constraints on the forms of facially neutral state action that continue to perpetuate the racial and gender stratification of American society."

The irony is palpable: how did the Court arrive at a position where the antiracism doctrines of fifty years ago are now the barriers that protect racial inequality? Angela Harris explains it as fear and unwillingness to "contemplate large-scale projects of political, economic and cultural redistribution and the dramatic transformation of social institutions and practices that would result from a complete renunciation of American white supremacy." Reva Siegel argues that the Court got into this trap because, like the proverbial generals, society always directs moral outrage at the previous forms of subordination. Tracking her thesis through the entire history of American race law, Siegel suggests that this "past-wars" approach encourages moral smugness about earlier eras while ignoring problems in the present.

Siegel exposes serious inconsistencies in the Court's reasoning about race. When it strikes down race-conscious remedial plans, the Court employs what she calls a "thin" conception of race (race-as-morphological-accident, race-as-analogous-to-blood-type). Using this thin understanding of race, the Court rejects the arguments advanced by advocates of diversity and affirmative action who employ racial classification as a proxy for differences in history, culture, and experience. It sees those arguments as impermissibly stereotyping racial groups. But when minority plaintiffs challenge state policies that create or support racially disparate outcomes in housing, employment, criminal justice, and schools, this same Court uses a "thick" conception of race to justify leaving those outcomes undisturbed. For example, in the Croson case the Court characterized the small number of minority contractors as the "natural" result of different occupational preferences among racial groups. This thick view of race allowed the Court to conclude that the differences in racial proportions were unobjectionable. The Court's inconsistent use of these thick and thin conceptions of race, Siegel argues, creates a lose-lose world for advocates of racial equality.


Color-Blind or Color-Coded Law?

In rejecting race-conscious classifications or remedies, the Court adheres to a jurisprudence of color-blindness that made sense in the 1950s and 1960s when segregation was legal and was based on a rigid system of racial classification. Color-blindness undermined and transformed that system. But fifty years later when state-sanctioned racial segregation is illegal and people of color have still to achieve truly equal opportunity with white Americans, the color-blind ideal actually impedes efforts necessary to eliminate racial inequality. Formal color-blindness fails to recognize or address the deeply rooted institutional practices and long-term disaccumulation that sustains racial inequality. Color-blind ideology is no longer a weapon that challenges racial inequality. Instead, it has become a powerful sword and a near-impenetrable shield, almost a civic religion, that actually promotes the unequal racial status quo.

The law and legal culture remain critical tools for dismantling racial inequality. But the law today does not speak from a genuinely color-blind vantage point. Despite having completed the vital task of eliminating Jim Crow racial classifications, legal institutions still operate with a perspective that remains perceptually, analytically, and functionally color-coded. The color is white.

Some examples can illustrate how the justice system remains color coded. Taken-for-granted white privilege explains how one unusually public-spirited citizen could refuse to vote for someone she saw as an extraordinarily qualified young black attorney who was running for judge in a community whose population is more than half minority but whose sitting judges and magistrates were white. What was the citizen's reason? She feared the candidate would be "biased toward the community." The fact that all the sitting judges were white was "normal" and therefore invisible to this white voter. The candidate's black skin and the majority-black community, on the other hand, were palpable.

Selection of grand jurors is another example. Law professor Ian Haney-Lopez found that even though Mexican Americans numbered one of every seven persons in Los Angeles County during the 1960s, they amounted to only one of every fifty-eight Los Angeles County grand jurors.Using judges' sworn testimony about their practices for nominating grand jurors, Haney-Lopez found that "nine out of ten nominees came from within the judges' own social circles—83 percent of nominees were friends, neighbors, family members, spouses of acquaintances, or comembers of clubs, organizations, or churches, and [and a few were recommended] by someone within those same circles or a fellow judge."The judges emphatically denied that discriminatory intent had anything to do with their choices, and this is most likely true. Nevertheless, regardless of their intentions, the judges' unselfconscious bias produced a degree of racial apartheid in the grand juries. Superior court judges in Los Angeles County nominated 1,690 grand jurors between 1959 and 1969; only 47 of the nominees were Mexican Americans. And of the 233 nominees who were actually seated, only 4 were Mexican Americans.The number of Mexican American grand jurors trebled to more than 6 percent of the total by the 1990s. By then, however, Latinos made up almost 41 percent of Los Angeles's population.

Invisible white advantage also explains how a white "gum-chewing, tennis shoe wearing" clerk in an exclusive Manhattan shop could feel it was appropriate to refuse to "buzz in" an elegant African American law professor doing her Christmas shopping. The editors of the journal that published the law professor's shopping story insisted on omitting all references to personal traits like skin color. Their grounds? They believed that not mentioning race (being color-blind) was necessary to being objective. The irony, of course, is that the story made no sense unless the parties' races were identified.

Other examples show that the experience, perspective, and privilege of white Americans permeate substantive law and policy. Lawyers, particularly influential lawyers, are overwhelmingly white. The law these (mostly white) lawyers have created has important strengths, but it also reflects their (mostly white) perspective on the world. From criminal to constitutional law, from federalism to family law, from immigration to original intent doctrine, the law reflects and endorses the views, needs, and advantages of the "normal" white perspective.

White perspective is not the product of skin color but of culture and experience. We speak of the white perspective because it is the perspective most often held by whites and the institutions they construct and dominate. It is the perspective of the namers, the controllers, the holders of "natural" privilege and invisible power, those who can take for granted the advantages of the status quo. Through experience and disciplined reflection, some whites expand, if not escape, the perspective of whiteness. For reasons of identification or advantage, some nonwhites may embrace it. Both, however, are exceptions to the typical taken-for-granted, normal and unreflexive white perspective.

Lawyers articulate and apply concepts like reasonableness, harm, culpability, desert, and merit. While their perspectives on these important ideas are shaped in part by their experience, that experience is filtered through the lens of their white perspective. This standpoint shapes their view of what voting arrangements are fair. It shapes the analysis and criteria of relevance for just administration of the death penalty. It shapes the priority accorded to hate speech as compared to "fire-crying" or national security under free speech law. It shapes whether accented speech undermines job qualifications. These modern examples are as much a result of an unarticulated white perspective as was the historical conclusion that when a white person was mistaken for being black, a serious compensable injury had occurred, but when the opposite happened, compensation was not legally appropriate.

White perspective sets the standards for probable cause or reasonable suspicion. It assesses institutional arrangements and personal behavior, deciding when confessions or consents to search are voluntary. It decides whether reasonable people feel free to refuse police requests and "go about their business." White perspective weighs the appropriate responses of reasonable persons, and the permissible latitude of reasonable force. It assesses the severity of crack cocaine offenses (which mostly involve poor blacks) as compared to crimes involving powder cocaine (which mostly involve middle-class whites). And it sentences offenders using crack to more time in prison than powder cocaine users, even when they possess the same amount of cocaine.

Some white lawyers, judges, and professors even erase race from the writing of the Constitution and the formation of the nation. Some urge courts to measure constitutional rights by the "original intent" of the framers without acknowledging the founders' racism. Many of the founders, Rogers Smith has shown, understood themselves to be the "bearers of a superior culture or racial heritage [that] . . . had obvious value in preserving the supremacy of the white, propertied, European-descended but largely native born male gentry who were the chief architects of the new governments." Despite this history, commentators analyzing the constitutional framework of American federalism act as if these attitudes were unimportant when the nation's so-called neutral framework of rights and power was created. They neglect the powerful shaping force of slavery and race in the very structure of our government.

A final example comes from the heart of constitutional law. Constitutional lawyers and scholars attribute the origins of the foundational principle of judicial review to the Supreme Court's decision in Marbury v. Madison. But as federal appellate judge John Noonan observes, "[Marbury] was an empty declaration. The power asserted was not used. The power asserted was not used throughout Marshall's lifetime. For the next two generations the power asserted turned out to be mere huff and puff. . . . The first fruit of the great declaration was Dred Scott." (Emphasis added.) The Court's decision in the Dred Scott case returned an escaped slave to his former owner. By upholding slavery, the Court asserted its authority to strike down federal laws, helping to precipitate the Civil War. Nevertheless, constitutional analysts downplay the role of slavery in the evolution of the principle of judicial review. Instead, they cite Marbury's reputable and lofty rhetoric rather than the slavery-tainted Dred Scott decision—even when that means ignoring the case that first gave the doctrine some real bite.


Color-Blind or Color-Coded Merit

If racial perspective affects the law, then the process for choosing who will be lawyers is significant. Is the process that selects candidates for professional legal training color-blind? Admission to law schools claims to be based on merit. Merit, however, is not a freestanding or self-defining concept. Merit must be merit in reference to something, for some purpose, based on some set of judgments and justifications. Traditionally, law schools have used Law School Admission Test (LSAT) scores and undergraduate grade point averages (GPAs) as proxies for merit. Schools choose these indicators because they correlate with law students' grades in the first year of law school. Law schools use other kinds of information, but in mostly unstructured and discretionary ways. The academic indicators are by far the most decisive factors, with the LSAT playing a crucial role. When merit is defined as excellence in test-taking, however, the selection process is not as color-blind as it claims to be.

Given their role as professional schools, it seems odd that law schools rely almost exclusively on academic measures of merit to choose students. Law schools train and credential lawyers. The mission of law schools is much more focused than that of colleges and universities. Law schools primarily prepare students for professional work. Only 2 to 3 percent of graduates from elite schools enter academic careers, but, ironically, law schools place more weight on academic indicators in admissions than academic departments do.

Law school graduates hold jobs that require intellectual and analytic skills; they use and apply knowledge. Academic skills are important in professional performance, but they are not the only indicators of professional achievement. If legal rather than academic jobs are the aim of most law graduates, then some of the criteria of merit should measure the capacity for outstanding performance in legal work. But law schools do not even attempt to assess these capacities. There is reason to believe that this choice about how to define merit disproportionately excludes students of color.

Decades ago, in Griggs v. Duke Power, the Supreme Court held that devices used to screen potential employees must be job related. The Court recognized that where access to education was unequal, the workforce would be unnecessarily distorted by race if employers required applicants to hold academic credentials that had little or no demonstrated relevance to successful job performance. The situation of law school admissions is more subtle and complex, but it is closely related. Unlike the academic credentials required in Griggs, academic intellectual skills are related both to law school and to lawyering. But extending the Griggs reasoning, one might still ask whether academic credentials are the only ones related to being a good lawyer. Effective lawyers must also have abilities such as problem solving skills, people skills, persuasiveness, the capacity to inspire trust, communication skills, tenacity, and goal orientation. Using the approach taken by the Court in Griggs, then, one might object to law schools' heavy reliance on one relevant factor (academic potential) to the exclusion of others that are equally job related in determining which applicants merit admission.

Seats in law schools are not jobs, but the links to jobs and Griggs are closer than they might first appear. Law is a state-licensed professional monopoly. The state delegates responsibility to the organized bar for certifying professional competence (through the bar exam and requirements for continuing education) and for maintaining professional discipline. These activities of the bar are important, but attaining a law school education is the pivotal step in becoming a lawyer. Thus, law schools act as the primary screening device for the job of lawyer. If one applied the reasoning used in Griggs, it would be unjustified to focus almost exclusively on academic as opposed to job-related criteria in selecting students for this professional education.

This argument becomes compelling when the racial consequences of conventional admissions criteria are examined. Social science research shows that job success is correlated with a variety of factors. Even for jobs with significant intellectual content such as the law, "paper and pencil tests" of aptitude or achievement are not highly correlated with on-the-job success. Performance on standardized academic tests does, however, correlate with race. Whites generally do better on paper and pencil tests and similar academic indicators than do blacks or Latinos. Successful performance on the job, however, is much more similar among racial groups. Therefore, reliance on paper and pencil tests will predictably create greater racial disparities in admissions than would a system that also adds in other types of predictors of successful job performance. Even though paper and pencil tests and conventional academic indicators deserve weight in measuring merit, overuse of those criteria and underuse of other important criteria produces racial disparities in selection that are disproportionate and unjustified. Christopher Jencks calls this type of racial unfairness "selection system bias." Selection system bias pervades law school admissions practices. The result is that whites are advantaged at the expense of persons of color. Put another way, the processes of exclusion and inclusion used by law schools are not simply color-blind systems that measure "objective merit." Rather, law schools make choices about whom to admit on the basis of debatable criteria that are arguably color-coded. By using such limited criteria, law schools will fill their classes with white students and make it much more likely that the legal profession and the law will continue to reflect a white perspective.


Beyond Color-Blindness

In recent years, some whites have begun to recognize that they, too, have a race, that being white may not equal colorlessness, normality, or neutrality. Once their race becomes marked, whites will have the opportunity to observe what they could not see before: race and the pervasive patterns of stratification with which it correlates. The Thernstroms spend hundreds of pages asserting that racism is (nearly) dead, and that if only guilty whites and unreasonably angry blacks would stop ranting about race, color-blindness would be within our grasp. Yet toward the end of their book, even they admit that whites almost always notice blackness: "Whites are able to shed their racial identity. . . . They had all the power. . . . Part of the package of privileges that came with being white was the liberty to think in individual terms. Blacks . . . were always black." The Thernstroms note that whites have been racially invisible because they have had "all the power," but they do not recommend giving blacks power as a way to equalize racial invisibility. Instead, like other racial realists, they want formal, lip-service color-blindness without any shift in power. Would racelessness have the same meanings for blacks that do not have power as it does for whites that do? Not likely. It is power that confirms and normalizes the particular perspective of white Americans. It is dominance that allows racial invisibility. The ultimate benefit of racial power is the right to make one's advantages seem simply the natural order of things.

That unacknowledged perspective of white America radiates throughout contemporary color-blind racial discourse. The racial realist seeks to transcend racial conflict by banishing blackness and the consciousness of racial inequality that accompanies it. Racial realists could transcend racial conflict by naming whiteness and the privilege that accompanies it. But this possibility remains unexamined. Acknowledging and banishing white advantage is never considered. Nor do they propose that race be made less visible by redistributing white power, by diversifying white dominance of political, social, intellectual, academic, and economic institutions. Instead, racial realists urge color-blindness, which, in effect, "whitewashes" the racial status quo.

At the center of the debate over race in America is the question of what perspective we will use to define racism and the social policies necessary to end it. From what vantage point will problems be named and solutions found? Defining racism is not a semantic or theoretical issue. Narrowing the concept to purposeful individual bigotry is highly advantageous for whites. It locates racism in America's past. It labels black anger and white guilt as equally inappropriate. It renders most whites innocent. It blocks most governmental efforts to reduce racial subordination and isolation. And, most important, it protects and naturalizes the racial status quo. Advocates of color-blind policies do not address these issues. Nor do they admit that their conclusions mainly express the white perspective that comes naturally to them and to many other Americans. They ignore the possibility that different racial perspectives could exist. Yet only by acknowledging these profound differences in perspective can one begin to address the durable racial inequality of American society. To assume that a color-blind perspective is the remedy is to be blind to color. It is to lose sight of the reality that in contemporary America, color has consequences for a person's status and well-being.

The idea that racism is simply a collection of intentionally bigoted individual attitudes is fundamentally flawed, both theoretically and empirically. It uses assumptions that are not supported by empirical evidence, it ignores the collective dimensions of racism, and its conclusions are dictated by its vantage point. We have introduced an alternative concept of racism that rests on very different assumptions and looks to different sorts of empirical evidence to assess the persistence of racism in America. With this conception in place, a very different picture emerges of the state of racism in America.

In subsequent chapters we critically analyze the increasingly popular view that racism is obsolete and that the persistence of durable racial inequality is attributable to individual failure on the part of blacks, Latinos, and other people of color. We examine the unstated "domain assumptions" that guide the questions raised by this understanding, the data used to answer them, and the claims that follow. We show that optimistic reports of racial progress are overstated and hollow. Using our alternative understanding of racism to systematically investigate the persistence of inequality in labor markets, education, the criminal justice system, and politics, we arrive at very different conclusions. While less optimistic, our analysis is more realistic, accurate and, we think, more useful for constructing policies that reduce racial inequalities and find common ground to bridge the racial worlds that still divide America.