1

“Racism”: Its Core Meaning

Apart from a small number of avowed white supremacists, most Americans wish very much to avoid being called “racist.”1 Yet the moral reproach carried by the term is threatened by a current tendency to overuse it. Some feel that the word is thrown around so much that anything involving “race” that someone does not like is liable to castigation as “racist.” “Is television a racist institution?” asked an article concerning the National Association for the Advancement of Colored People’s (NAACP) criticism of prime-time network shows for having no “minority” actors in lead roles.2 A local newspaper called certain blacks “racist” for criticizing other blacks who supported a white over a black candidate for mayor.3 A white girl in Virginia said that it was “racist” for an African American teacher in her school to wear African attire.4 The Milton, Wisconsin, school board voted to retire its “Redmen” name and logo depicting a Native American wearing a headdress, because they had been criticized as “racist.”5 Merely mentioning someone’s race (or racial designation),6 using the word “Oriental” for Asians without recognizing its origins and its capacity for insult, or socializing only with members of one’s own racial group are called “racist.”

A few observers suspect that the word has lost all significant meaning. “Racism is…what black activists define it to be.…When words lose coherent meaning, they also lose the power to shame. ‘Racism,’ ‘sexism,’ and ‘homophobia’ have become such words. Labels that should horrify are simply shrugged off.”7 The Time columnist Lance Morrow sees social damage in this development: “The words ‘racism’ and ‘racist’ are a feckless indulgence, corrosive to blacks and whites alike and to relations between them.”8

“Racism” and “racist” should be reserved for certain especially serious moral failings and violations in the area of race. They should not be permitted to spread to include everything that someone might justifiably disapprove of. A major reason for what Robert Miles calls the “conceptual inflation”9 of the idea of “racism” is its having become the predominant notion used to mark morally suspect behavior, attitude, and social practice regarding race. The result is that either something is racist or it is morally in the clear.

In Boston a white police officer, as a bizarre joke and apparently with no malice intended, placed a hangman’s noose on the motorcycle of a black police officer. “Police Probe Sees No Racism,” says the headline of an article reporting the findings of an investigation into the incident.10 Perhaps the white officer was not “a racist”, nor operating from racist motives. The victim in the incident said, however, that “you cannot hang a noose like that near any black man who knows his history and say it does not have tremendous significance.”11 If our only choices are to label such an act either “racist” or “nothing to get upset about,” those who seek to call attention to any racial malfeasance will be tempted to describe it as “racist.” That overuse in turn diminishes the moral force of the word and thus contributes to a lowering of concern about both racism and other racial wrongs.

Not every instance of racial conflict, insensitivity, discomfort, miscommunication, exclusion, injustice, or ignorance should be called “racist”. Not all racial incidents are racist incidents. We need a more varied and nuanced moral vocabulary for talking about the domain of race. We need to articulate the range of values and disvalues implicated in race-based beliefs and attitudes, actions and interactions, institutions and practices. All forms of racial ills should elicit concern from responsible individuals. If someone displays racial insensitivity, but not racism, people should be able to see that for what it is. In a soccer game, a nine-year-old white boy said to one of his black teammates, “Boy, pass the ball over here,” and “was virtually accused of being a racist by the father of one of his teammates,” according to an article on the incident. (The word “virtually” itself suggests the loss of an evaluative vocabulary other than “racist” and “racism.”) The white boy was almost surely not “a racist,” and the article itself goes on to express more accurately the racial ill involved in his remark: “The word ‘boy’ is a tripwire attached to so much charged racial baggage that it is no longer safely used as a term for a prepubescent male.”12

If a policy has a racially unjust effect, or unequally affects already unequally positioned racial groups, it should be reason for concern even if there is no suspicion that it arises from racist motives or is part of an entrenched pattern strongly rooted in historical racism. For example, school lunch programs have been criticized for relying too strongly on milk, in light of African Americans’ substantial propensity toward lactose intolerance in; but no untoward motives or failures of sensitivity need have prompted this policy for it to be of moral concern. Similarly, it is troubling if prime-time TV fails adequately to reflect its viewers’ ethnoracial diversity; but the failure is not necessarily “racist.”13 Someone who exhibits a culpable ignorance about racial matters that bear on an interaction with an acquaintance or co-worker should feel a degree of shame, and resolve to correct that ignorance, without having to think she has been “racist.” We should not be faced with the choice of “racism or nothing.”

Conceptual inflation and moral overload arise from another source as well—designating as “racism” any prejudice, injustice, inferiorizing, or bigotry against human groups defined, say, by gender, disability, sexual orientation, or nationality. In The Decent Society, Avishai Margalit, an Israeli philosopher, discusses racism as the denying of dignity to any human group, and he uses as a particular test case “retarded” persons.14 This inflated use of “racism” does, certainly, pay indirect tribute to racial oppression and denial of dignity as the central form of such mistreatment in contemporary Western consciousness; and that centrality is reflected also in coinages such as “sexism,” “ableism” (discrimination against the disabled), “classism,” and “heterosexism.”15 This proliferation of other “isms” at least avoids the confusion wrought by Margalit’s conflating all of them with “racism” itself and encourages us to explore both the similarities and the differences between discrimination, exploitation, and denials of dignity based on race and those based on gender, sexual orientation, disability, national membership, and the like. But Margalit’s subsuming all these moral ills under “racism” cuts off such inquiry at the starting line, and, in so doing, contributes to a counterproductive inflation of the term “racism.”

“Racism”: A History

The term “racism” was first used by European social scientists in the 1930s to characterize and condemn the Nazi belief system, which posited the superiority of the “Aryan” race over an elaborate ranking of allegedly lesser races.16 The Nazis were not the first to make use of ideas of a hierarchy of human groups distinguished by inherent characteristics and generally distinguishable by physical appearance. As I observe in more detail in chapter 6, the idea of racial hierarchy had become firmly entrenched in American and, to a lesser extent, European thought in the nineteenth and early twentieth century. (Indeed, the Nazis were distinctly influenced by American racial thought.17)

Although many persons had opposed previous race-based ideas and the systems of social domination they rationalized, it was not until the Nazi period that the term “racism” came to express moral revulsion at such systems. The rejection of racism was fed, in addition, by developments within the sciences that, independently, had begun to throw into question the idea of a hierarchy of discrete human populations or “races”.18

Postwar revelation of the extent of Nazi atrocities helped intensify the revulsion attached to the concept of racism. At the same time, the word’s central locus of reference shifted to still-existing systems of domination, exclusion, and oppression—segregation in the United States, racialized colonialism among the European powers (in Asia, Africa, Latin America), and apartheid in South Africa.19 Retrospectively, American slavery contributed to the strongly negative moral valence of the term, especially, though not only, in the United States.

George Mosse, a historian of European racism, notes that originally the term “racism” referred not merely to an “articulation of prejudice” or a “metaphor for suppression” but to “a fully blown system of thought, an ideology like Conservatism, Liberalism, or Socialism.”20 What was rejected in the ideology or doctrine of racism was the idea that so-called racial groups, however identified, possessed inherent and inescapable characteristics and could be ranked on a scale from inferior to superior. That is, “racism” was defined as biological or quasi-biological determinism, plus inegalitarianism. What was being condemned was a system of thought; but the moral revulsion carried by the term was intimately bound up with the dehumanizing and sometimes murderous social systems the ideology was used to rationalize.

Although “racist” phenomena predated the invention of the concept itself, not every system of belief that led to the rationalization of human inequality or subordination could be called “racist.” The social system in question must have developed a conception of “race” in terms of which subordination was justified. The acceptance for centuries in Europe and North Africa of Christians enslaving Muslims, and Muslims enslaving Christians, was not premised on race and racial inferiority; and the idea that it was morally permissible to subjugate conquered people culturally distinct from the conquerors was long unrelated to any racial ideas. Slavery or subjugation, whether founded on religion or conquest, is morally repulsive, but they are not intrinsically “racist” unless enslaved groups are seen as races distinct from the race of the enslavers.

Racist doctrines were not fully utilized to justify slavery in the Americas until the nineteenth century (though proto-racist ideas were so employed in the previous two centuries), in part because slavery was not thought to require a moral justification, and in part because the concept of “race” had not been fully developed. As we shall see also, the American “racial worldview,” to use Audrey Smedley’s expression, did not reach full development until the post-slavery period.

Many current authoritative definitions of “racism” preserve the original focus on a doctrine, ideology, theory, or cohesive set of beliefs. The Oxford English Dictionary of 1982, for example, says, “Racism is the theory that distinctive human characteristics and abilities are determined by race.” Blackwell’s 1993 Dictionary of Twentieth-Century Social Thought begins its entry on “racism” thus: “Any set of beliefs which classifies humanity into distinct collectivities, defined in terms of natural and/or cultural attributes, and ranks these attributes in a hierarchy of superiority and inferiority, can be described as ‘racist’.”21 Charles Taylor, in his influential philosophical work Sources of the Self says, “Racists have to claim that certain of the crucial moral properties of human beings are genetically determined: that some races are less intelligent, less capable of high moral consciousness, and the like.”22

But established contemporary use does not confine “racism” as a term of moral reproach to doctrines of biological superiority. In the past several decades, we have come to criticize as “racist” not only beliefs but actions, motives, attitudes, statements, symbols, images, practices, societies, and persons. Nor do we require that racist manifestations involve such doctrines. (At most they may presuppose that the targeted group had historically been the subject of such beliefs.) If it is racist to steer blacks away from buying homes in a particular neighborhood because they are seen as “undesirable,” rather than because they are seen as biologically inferior, then “racism” is not limited to a system of belief in racial inferiority.

The Native American logo mentioned at the outset of this chapter is racist because it demeans a group that has been seen in racial terms, not because anyone displaying the logo necessarily holds racist beliefs about them presently. In the summer of 1999, a young white man, Benjamin Nathaniel Smith, murdered a Korean American man and a black man and shot at several Jews and other Asian Americans, evidently out of racial hatred for these groups. These actions, and the motives that prompted them, were uniformly regarded as racist in the major media, prior to knowing whether Smith also believed in explicitly racist doctrines.23 Use of the epithets “racist” and “racism” is widely accepted regardless of knowledge about whether those so characterized believe in “races,” in the sense meant in the original definition of racism.24 For example, though the use of epithets such as “wetback,” “greaser,” and “nigger” are unproblematically referred to as “racist,” their usage does not require belief in a hierarchy of biologically based “races” (on the part of the user, hearer, or anyone else).

Even when racial bigotry or hatred do involve racist beliefs, such views of the group in question are often a product of the hatred rather than a consequence of adhering to a racist ideology. As hatred and antipathy, especially of a racial nature, are socially proscribed, the hater may be tempted to rationalize them by casting the target of her antipathy as evil, criminal, menacing, “taking over,” or the like. Here the belief is secondary to the emotion or attitude, not its basis.

The sense of wrongfulness evoked by the term “racism,” originally attached to a system of beliefs, drew its moral force from the evils of the social systems in which those beliefs were embedded. (A belief unconnected to horrible practices would be unlikely to garner such strong moral opprobrium.) Hence it was natural to extend the reach of the term to the acts, attitudes, emotions, and symbols that were integral to systems of oppression. Segregation, for example, functioned not only through the belief that blacks were racially inferior to whites, but through attitudes of contempt, scorn, prejudice, and hatred; through acts expressing such attitudes; and through particular institutions or practices (for example, blacks being referred to by their first names but being compelled to address whites of any age as “Mr.” and “Mrs.”).

The historical process by which “racism” expanded its reference from doctrines to systems, acts, and persons, and became the central term of moral condemnation, was neither simple nor inevitable. For example, in the 1930s and before, the word “racialism” was also employed to refer to the doctrine of hierarchically arrayed, biologically defined racial groups, as in Jacques Barzun’s influential 1938 book, Race, A Study in Modern Superstition.25 However, if someone tried to condemn the array of phenomena we currently call racist with the epithet “racialist,” she would be unsuccessful. It remains a term used primarily by specialists, with little popular moral cachet.26 In the postwar period, especially in the United States, attention turned to the psychological underpinnings of racist structures and behavior, and the idea of “racial prejudice” took center stage as the concept that best described those underpinnings. Gordon Allport’s magisterial The Nature of Prejudice in 1954 helped to promote this notion of prejudice as the central factor in racial ills; the idea became a prime mode by which social scientists attempted to come to grips with the moral legacy of the Holocaust and with racial segregation in the United States.27

Allport did not refer to racial prejudice as “racism,”28 perhaps in part because he was concerned with all forms of prejudice (racial, religious, ethnic, and others), and believed there to be a psychic unity among them, involving a distinction between “in-groups” and “out-groups.” Still, his work helped to prompt a focus of moral (and research) concern on various attitudes that underlay the operation of racist systems.

The Kerner Commission’s report on civil disorders in 1968, following the urban riots and uprisings in black areas in the mid-1960s, made “white racism” a central organizing idea, blaming it for the unsatisfactory conditions of life in Detroit, Los Angeles, Brooklyn, and other cities that led to the “disturbances.”29 The prestige and wide circulation of that report helped to bring the terminology of racism to a broader popular audience. The Black Power movement of the same era also tended to employ “racism” as a primary means of conceptualizing the state of black America. Black Power, an influential work of that time by Kwame Ture (then Stokely Carmichael) and Charles Hamilton, was critical of the All-portian focus on individual prejudice and bigotry, and shifted the locus of “racism” from the individual to broader social structures of inequity and to socially pervasive antiblack stereotypes and prejudices.30 They referred to this phenomenon as “institutional racism” (further discussed below), arguing that progress in reducing individual prejudice had by no means erased pervasive inequities in the life chances of blacks and whites. Both their specific analysis and the increasingly general use of “racism” further served to embed that term in the moral vocabulary that referred to objectionable phenomena of a racial nature.

As segregation and colonialism (and, more recently, apartheid in South Africa) were dismantled, it became evident that racist attitudes, actions, and statements could also occur outside of such systems. They could be directed by members of any racial group toward any other, independent of the social and economic position of either of the parties, and they seemed deserving of condemnation in their own right.

As concepts of moral disapprobation, “racism” and “racist” have definitively broadened their reach beyond doctrines of biologically based hierarchy. At the same time, current use is not sufficiently unified or stable to allow us to point to one definition as the “true meaning” of “racism.” Nevertheless, I will suggest a core meaning, rooted in the history of its use, that secures “racism” as referring to phenomena deserving of the severest moral condemnation, and that encourages us to make use of the considerable other resources our language offers us for describing and criticizing race-related ills that do not characteristically rise to the level of racism—racial insensitivity, racial conflict, racial injustice, racial ignorance, racial discomfort, and others. An agreed-upon meaning that avoids conceptual inflation and moral overload would facilitate interracial communication, and it should diminish an inhibiting fear of the dreaded charge of “racism” while also encouraging a more morally nuanced vocabulary for discussing race-related phenomena. My suggested definition of “racism” should stanch its threatened loss of moral impact, which discourages moral concern about racism and other race-related ills.

Defining “Racism”

I want to suggest that all forms of racism can be related to one of two general themes or “paradigms”: inferiorization, and antipathy.

Inferiorization is linked to historical racist doctrine and racist social systems. Slavery, segregation, imperialism, apartheid, and Nazism all treated certain groups as inferior to other groups (mostly the dominant group, although sometimes other non-dominant racial groups) by reason of their biological nature.

Though race-based antipathy is less related to the original concept of “racism,” today the term unequivocally encompasses racial bigotry, hostility, and hatred. Indeed, the racial bigot is many people’s paradigm image of “a racist.” A disturbing but illuminating example of contemporary antipathy racism occurred in Washington state in 1999. The Makah tribe of the Olympic Peninsula announced its intention to hunt for whales as a way of instilling pride and tradition in the tribe’s youth. The hunt was permitted by the government, and the tribe killed a whale in May of that year. Many non-Native American residents of the state were outraged by this act. Amidst some arguably reasonable objections were expressions of outright antipathy racism toward the Makah, and toward Native Americans more generally. One letter to the Seattle Times, for example, said, “I have a very real hatred for Native Americans now. It’s embarrassing, but I would be lying if I said it wasn’t the truth.”31

Historical systems of racism did of course inevitably involve racial antipathy as well as inferiorization. Hatred of Jews was central to Nazi philosophy; and it is impossible to understand American racism without seeing hostility to blacks and Native Americans as integral to the nexus of attitudes and emotions that shored up slavery and segregation.32

I suggest that all the various forms of racism are related to inferiorization or antipathy, and will illustrate this in the next several chapters. To simplify that discussion, I want to introduce three other general categories. Personal racism consists in racist acts, beliefs, attitudes, and behavior on the part of individual persons. Social (or sociocultural) racism comprises racist beliefs, attitudes, and stereotypes widely shared within a given population and expressed in cultural and social modes such as religion, popular entertainment, advertisements, and other media. Institutional racism refers to racial inferiorizing or antipathy perpetrated by specific social institutions such as schools, corporations, hospitals, or the criminal justice system as a totality.33

Each of the three operates in complex interaction with the others. Persistent institutional racism encourages personal belief in, or suspicion of, racial inferiority. Personal racism reflects already existing social racism, and also sustains it. Personal racism slows or blocks society’s moral impetus to change racist institutions. Some contemporary writing has attempted to reduce one of these forms to another, or to demote it in importance. For instance, in some influential writing on racism, the Ture-Hamilton notion of racism as a system of injustice and unequal advantage or power has almost entirely eclipsed the earlier Allportian focus on individual attitude and behavior.34

The view of racism as solely systemic substitutes one partial conception of racism for another. For instance, suppose a white person hates black people, but the white person is relatively isolated and powerless, does not come into contact with blacks, and generally does not even express his or her feelings to others. This hatred is unquestionably racist, yet it contributes virtually nothing to a system of unjust advantage based on race. If racism were only systemic, then such an individual would not be racist and, indeed, individuals could never be racist. It is revealing, however, that adherents of the view in question seldom follow that logic exclusively. Generally they at least implicitly acknowledge that individual actions and persons may be “racist.” They imply, however, that when an individual acts in what would ordinarily be called a racist manner, the action is to be condemned as racist only insofar as it contributes to a system of unjust racial advantage.35

This institutional conception unnecessarily narrows the reasons to condemn racism. We do better to accept the plurality of items that can be racist (beliefs, institutions, systems, attitudes, acts, and so on), without thinking that one of these need be the foundation of all the others. Personal, social, and institutional racism are each morally problematic, in at least partially distinct ways.

Inferiority and Antipathy Racism

Inferiorizing personal racism is expressed in various attitudes and behavior—disrespect, contempt, derision, derogation, demeaning. It can also, involve a developed set of beliefs about a biologically based hierarchy of races, but it need not do so. For one thing, an individual may be racist against only one racial group and have no views about others. An individual can be contemptuous toward another racial group without really believing that it and its members are inferior. A white or black person, for example, may grow up in an environment in which Mexicans and Mexican-Americans are routinely treated and regarded with contempt. On reflection she may realize that she does not really believe Mexicans to be inferior or otherwise worthy of contempt, yet her manifestations of contempt toward Mexicans will still be racist in character.36

An inferiorizing racist generally thinks that the racial other is inferior to her own group, but sometimes people believe their own group to be inferior. The film A Soldier’s Story depicts this internalized racism very powerfully.37 The film focuses on a black company at an army base during World War II, whose sergeant, Waters, hates another black soldier as representing all that is deficient in blacks. As the story unfolds, Waters reveals an extreme desire for the approval of whites. Ultimately he acknowledges a racial self-hatred that has fueled his attempts to purge his company of forms of black behavior (and types of black persons) that he regards as making “the race” look bad to whites. The film powerfully illustrates that internalized racism is the product of a society in which the group in question is devalued.38

Inferiorizing and antipathy racism are distinct. Some inferiorizing racists do not hate the target of their beliefs; they may have a paternalistic concern and feelings of kindness for persons they regard as their inferiors. This form of racism characterized some slave owners and many whites’ views of blacks during segregation. (Joel Kovel has called this form of racism “dominative.”39) The concern and kindness are misdirected, and demeaning, because the other is not seen as an equal, or even as a. full human being; it is a racist form of concern, yet is distinct from antipathy and hatred.

Conversely, not every race hater regards the target of her hatred as inferior. In the United States, antipathy toward Asians and Jews often accompanies, and is in part driven by, a kind of resentment of those seen as in some ways superior (for example, more economically successful). And some whites who hate blacks do not really regard blacks as inferior; they may fear and be hostile to them, but fear and hostility are not the same as contempt and other forms of inferiorizing (though they may accompany one another). Survey research suggests that pure inferiorizing racism toward blacks has substantially decreased since segregation, more so than antipathy racism.40 Still, the great and persistent racial inequalities in our society provide a standing encouragement to advantaged groups to see disadvantaged groups as deserving their lower status.

The two forms of personal racism are not entirely separate either. Mexican Americans are often seen as inferior and also hated as a “foreign element” allegedly usurping jobs that whites regard as their prerogative. (African Americans can also have racial hostility toward Mexican Americans for the latter reason, but are much less likely to regard them as racially inferior.) The paternalistic inferiorizing racist, such as a white segregationist, often hates those members of the racial group who do not maintain what he regards as “their place”—for example, blacks who do not engage in the deference behavior the paternalistic racist expects. Emmett Till was lynched in 1955 out of hatred directed toward a young black man who had transgressed the rules of racial deference and constraint defining him as an inferior being. Hatred characteristically surfaces toward those members of the inferiorized group who act in a manner implying they do not regard themselves as inferior to the racist.

Racial Prejudice

The idea of racial prejudice will help clarify the antipathy paradigm for personal racism. The word “prejudice” can be confusing, since its linguistic form suggests that it simply means “prejudging” something before you have a chance to really know it. Certainly prejudging can be part of racial prejudice, but as generally understood the term implies something more—a general antipathy, or animus, toward another racial group, or toward an individual because he or she is a member of that group. Gordon Allport’s classic definition of prejudice states that the antipathy is “based upon a faulty and inflexible generalization.”41 This is an important point. An individual is not prejudiced if she has good reason for antipathy toward a group—for example if she is hostile toward a family plotting to undermine her business. As Allport notes, antipathy toward Nazis is not prejudice.42

Not all antipathy is prejudice, but all antipathy toward a racial group is prejudice, as such groups are too large for anyone to have valid personal grounds for antipathy toward the entire group. Blacks or Latinos do not have grounds for antipathy toward all whites (although some people think they do). Neither blacks, Latinos, nor whites who have been personally victimized by members of the other groups have grounds for hostility toward the entire group. They are prejudiced if they have such hostility.

Prejudice need not be conscious. (The same may be said of bigotry, though stronger forms of animus may be less liable to stay hidden from consciousness.) A white individual may feel hostility toward African Americans or Asian Americans without realizing she does. Although the increasing social disapproval of racial prejudice (and other comparable group prejudices) in the past forty years or so is a positive development, it has had the effect of providing some people with an incentive to suppress acknowledgment of their prejudice.

Prejudice may be nonconscious in another sense as well. An individual may be hostile toward a certain group but not realize the racial basis for her hostility. For example, she may dislike a certain style of personal interaction no matter who exhibits it, without recognizing that her reason for doing so is that she generally associates that style with black people, and dislikes black people.43

Moreover, as Allport suggests, prejudice generally involves not only dislike of a group but a faulty view of it as well. Someone who is prejudiced against Mexicans, or Mexican Americans, generally both dislikes members of that group and regards them as having negative characteristics that justify her dislike—being lazy, dirty, dangerous, or unfairly taking “her people’s” jobs.44 Classic understandings of prejudice assumed that these faulty conceptions always stem from ignorance of the group, arising from lack of contact with it. Allport himself did not make this error; he realized that contact with another group does not always lead to understanding. Research on school desegregation suggests that some whites become more prejudiced when they are put in schools with blacks than they were before they knew any blacks.45 In any case, what it means to “know” another group well enough to avoid faulty generalizations about it is a very complex matter. At the least, it involves getting to know some members of the group both as individuals and as members of the group. It always means having a lived understanding of the great diversity within any racial group. For true “knowing,” other conditions must no doubt be met as well.46

Yet Allport is not correct to build into the definition of prejudice “based upon a faulty and inflexible generalization.” For one thing, generalizations about the group could be after-the-fact rationalizations of prejudice, rather than the actual basis of the prejudice. In addition, someone could be hostile toward a group, but when asked why would say “I don’t know; I just don’t like them.” Such a person may be suppressing some conception of the group that he is reluctant to state. Or he may just have an irrational antipathy. We would still call that person “prejudiced”. He may have some rudimentary negative images of the group that accompany his antipathy, but these do not rise to the cognitive status of Allport’s “generalization.”47

Thus prejudice is a kind of antipathy, toward a race-defined group, and would by my definition appear to count as a form of “racism”. Indeed racial prejudice is often called “racism.”48 But, “prejudice” often implies dislike or antipathy, but not necessarily hatred or strong antipathy. Intensely negative affects are beyond prejudice—for example the hatred shown by some Serbs in driving Kosovar Albanians from their homes and territory in the spring of 1999. It seems absurdly understated to say that the Serbs “were prejudiced against the Albanians.”

Because prejudice is, in general, a less malevolent attitude than hatred and intense hostility, it is less morally evil. It is not clear whether we should call weaker forms of racial prejudice “racism.” These forms should, in any case, evoke opprobrium in their own right.

Categorial Drift

We have looked at two ways that “racism” and “racist” have been conceptually inflated and morally overloaded, diminishing their usefulness and force as concepts expressing moral reproach: the tendency to apply them to every malfeasance in the racial area, and to use them as general concepts for all forms of group discrimination, oppression, or denial of dignity. We are now in a position to discuss a third such devaluation of these words, which results from their undifferentiated use in regard to very different entities: beliefs, acts, attitudes, statements, symbols, feelings, motives, and persons. I call this confusion “categorial drift.” For instance, a person commits one racist act and is called “a racist,” or makes a racist statement and is assumed to be doing so from a racist motive. Here we have drift from one category of racism to a second, generally more objectionable one.

A discussion of categorial drift will enable us to clarify how personal racism in the different categories (acts, statements, persons, and so on) relates to the antipathy and inferiorization themes. Let us begin with an example.

Ms. Verano is a white fourth grade teacher. She feels comfortable with all the children in her very racially-mixed class. She holds all students to equally high standards of performance. But, though she has never admitted this to herself, she is not really comfortable with most of the black parents. She does not dislike blacks, nor does she think they are inferior. She is not particularly familiar with African American culture, however, knows very few blacks other than her students, and is not confident about her ability to communicate with blacks. As a result Ms. Verano is somewhat defensive when speaking with black parents in conferences, and is not able to listen to their concerns and viewpoints about their children as well as she does with parents in other racial groups. Because she does not glean as much information from the black parents about their children as she does from the other parents, she is not able to do as much for these children as for the others in her class. Ms. Verano does not recognize any of this, however.

Is Ms. Verano a racist? Is she prompted by racist motives?

Motives, Acts, and People

Racist motives are those based in antipathy toward or inferiorizing of a racial group, the latter category including contempt, derision, disrespect, and the like. Racist acts are diverse, and one type is an act prompted by a racist motive. Acts that make use of racist statements, jokes, symbols, or images, even if the person performing the act is not motivated by antipathy or an inferiorizing attitude, may be called “racist”. For example, an individual may tell a racist joke in order to go along with a group, in order to feel accepted, or merely to get into the spirit of an occasion, without holding the inferiorizing or hostile attitude expressed in the joke. Yet acts whose motives are racist seem to me more definitively racist than those whose motives are not, even if the latter in some other way do involve racism.

A racist person is not merely someone who commits one racist act or acts on a racist motive on a small number of occasions. Motives and attitudes such as bigotry, antipathy, and contempt must be embedded in the person’s psychological makeup as traits of character. In this sense, being racist is like being hateful, dishonest, or cruel, in implying an ingrained pattern of thought and feeling, as well as action.

Just as someone can act cruelly or dishonestly on one or several occasions without being a cruel or dishonest person, so someone can act in a racist manner on some occasions without being “a racist.” It is a much stronger condemnation to say that a person is “a racist” than to say that some of her actions are racist. Unfortunately, current public discourse frequently conflates these two things, as if everyone who engaged in some racist behavior were a full-fledged racist. People confusedly and inadequately defend themselves against the charge that they have committed a racist act by saying “I’m not a racist.”

The New York Times ran a long article on Al Jolson, the entertainer of the 1920s and 1930s who was best known for performing in blackface. “Was Jolson a racist?” the Times asked, and answered by saying, “Although he was guilty of many faults, Jolson showed no sign of ethnic hatred.”49 But the racial issue in Jolson’s performances has little or nothing to do with whether Jolson was a racial bigot, or even with whether he intended to demean blacks by utilizing the demeaning black stereotypes that were the staple of blackface minstrelsy. The article goes on to pinpoint the real issue, which is independent of Jolson’s own character or motives: “Black-face evokes memories of…an age in which white entertainers used the makeup to ridicule black Americans while brazenly borrowing from the rich black musical traditions that were rarely allowed direct expression in mainstream [that is, white] society.”50

It is true, and fortunate, that some people who are racist in their character can learn not to be, if they disavow their racism and go through a process of personal struggle. An inspiring example of such a transformation is C. P. Ellis, a former Klansman who became antiracist, largely through being forced to work on a school board with a black woman from his town.51 Of course, a person who merely commits a racist act is a much better candidate for avoiding racism in the future than is an ingrained racist. The former may just be thoughtless or ignorant about the nature or effects of her racist acts, or her antipathy may not run deep nor be securely lodged in her psychological makeup. Without being overly sanguine about the prospects of reducing attitudinal and motivational racial prejudice, I would suggest that racial animus rooted in someone’s character is more difficult to erase than that which is not.52

In her uncomfortable reactions to black parents, Ms. Verano, the white fourth grade teacher, is neither a racist nor does she act from racist motives. She holds no antipathy toward blacks. She is not an inferiorizing racist. Though uncomfortable with black adults, she is not racially prejudiced against them.

Symbols, Jokes, and Remarks

Symbols, jokes, images, epithets, and remarks can be racist in their own right, apart from people’s motives in using them. What makes them racist is that they either directly or indirectly portray a racial group in an inferiorized manner (for example, as stupid or otherwise worthy of contempt, disrespect, or derision), or in a manner rendering them appropriate targets of hatred, antipathy, or dislike. An example of direct symbolic racism is the logo of the Cleveland Indians baseball team—a grinning, moronic-looking Native American. As nonrepresentational symbols often involve no explicit portrayal of a group, a symbol can be indirectly racist—the swastika, for example—when its origin and cultural meaning are associated with racist portrayals, hostility, or inferiorizing. A racist epithet may involve no distinct portrayal of a racial group, and attribute no specific characteristics to the group; epithets such as “spic,” “kike,” “nigger” are racist, however, because they are generally understood to be derogatory and insulting.

Many symbols are racist whether those who display them are aware of this or not. Like the swastika, the Confederate battle flag is such a racist symbol.53 The flag symbolizes the Confederacy as a defender of the “Southern way of life” that involved slavery as an essential component. The contemporary history of the flag’s presence in official venues and incorporation within some official southern state flags is a further manifestation of its racist meaning. In 1956, for example, Georgia incorporated the “stars and bars” into its state flag as a sign of defiance of the Brown v. Board of Education school integration ruling two years previous. Mississippi has done so as well. In 1961 and 1962, in the midst of controversy over segregation, Alabama and South Carolina, respectively, raised the Confederate flag over their state capitols (beneath the American flag) for a similar reason. In the summer of 2000, partly in response to a national boycott called by the NAACP, the South Carolina legislature voted to remove the flag from the dome atop the statehouse, and Georgia has demoted the Confederate emblem to a much smaller place within its state flag.

Some white Southerners, perhaps entirely sincerely, purport to regard the Confederate flag as a merely regional symbol, or a symbol of the heritage of that region. In a widely publicized incident in the early 1990s, a white Harvard University student hung a Confederate flag from her dormitory window, saying that it symbolized warmth and community associated with her southern home and heritage.54 Unquestionably the Confederate flag is a regional symbol, and, moreover, it may well not convey a racism-infused message to all who display or value it. But a symbol can be both regional and racist. One who understands the history of the Confederacy and the use of its flag, as well as the flag’s recent historical use as a sign of defiance against federally-enforced desegregation, can not pretend the flag does not symbolize slavery, segregation, or black subordination more generally.55 Southerners who wish to distance themselves from nostalgia for a way of life built on slavery or segregation must find a different regional symbol, as, indeed, some southerners are attempting to do.56

The public meanings of flags, monuments, and other symbols can be multiple or contested. Symbols can mean different things in different contexts and to different groups, and the meanings can change over time.57 The shamrock is a traditional Irish and Irish American ethnic symbol. But in Boston, some African Americans see its current display in traditionally Irish areas such as South Boston, which have been compelled, against the wishes of many of its white residents, to undergo residential integration in the past two decades, as a sign of exclusion.58 Although these blacks might attach a meaning to a particular display of the shamrock, they do not thereby deny that it is an Irish American ethnic symbol. In the Confederate flag dispute, by contrast, some whites wish to deny the racist meanings of the flag.59

The making of a racist joke or the displaying of a racist symbol does not make someone “a racist.” It does not even necessarily signify a racist attitude or belief. One may use a racist symbol, make a racist remark, or tell a racist joke without realizing it, though such ignorance is more generally found in children than adults. For example, a child may wear his brother’s hat with a Confederate flag symbol, without knowing what that symbol means, because he likes the way it looks or admires his brother. As mentioned earlier, someone may retell a racially offensive joke she has heard just to get a laugh, without recognizing that the joke is racist.

Lack of such knowledge generally has no bearing on whether the symbol or joke is itself racist. However, it does bear on the user’s moral responsibility for using or telling it. Again, there is likely to be an asymmetry between adults and children in this regard. Adults generally know that a joke or symbol is racist. An adult who uses what he knows is charged racial language but then acts surprised that someone in the targeted group is offended often appears disingenuous.60 We continually hear of radio and television announcers, and politicians too, excusing their use of offensive and racist language, symbols, jokes, and remarks on the grounds that they did not intend for them to be racist, or to offend. The California state assemblyman William Knight distributed to his fellow Republican legislators a poem, written in a style that demeaned native Spanish-speakers’ way of speaking English and implied that Mexican immigrants are mostly illegal, have too many children, come to the United States to exploit a generous welfare system, are contemptuous of white people, and intend to take over the United States. When the Latino legislative caucus complained that the poem was “racist,” Knight declined to apologize but said he thought the poem clever and funny, and that it was not intended to offend anyone.61 It is impossible to believe that Mr. Knight failed to recognize that the poem was offensive, although he may well have believed that none of the persons whom he intended to see it would be offended. Be that as it may, since the poem is so patently racist and offensive, Assemblyman Knight is morally at fault for distributing it, even if he did not realize that it was. In general, people beyond a certain age should recognize what is racist; their moral responsibility for perpetrating racism when they do not recognize it is analogous to the fault of citizens who cause injury through negligence rather than direct intent. They could be reasonably expected to recognize or anticipate the moral damage. Except for people with extraordinarily sheltered lives and upbringings, ignorance of racism does not absolve one from responsibility, although, everything else being equal, intentionally demeaning a racial group is morally more blameworthy than unintentionally doing so.

So not intending to be racist does not absolve a speaker from having made a racist remark, nor, generally, of moral responsibility for doing so. If a racist remark is made, it is entirely appropriate and understandable for a member of the targeted group to be offended, whether the user meant it as racist or not.

Symbols and Offense

Often racist symbols, remarks, and jokes are discussed as if they were objectionable only because they cause offense to some group. This puts the cart before the horse. Their inherent racism is generally why members of the group targeted by the racism are offended.62

Moreover, those in the targeted group are not the only persons who have reason to object to racist expressions. Every person has reason to object to racism, just as we all have reason to object to cruelty, dishonesty, and the violation of human rights, no matter who suffers from them. Students often think it is “weird” or in some way pretentious or inauthentic—merely trying to be “politically correct”—if someone objects to a racist remark that demeans a group other than his own. They expect someone from the targeted group to object, but no one else. Moral outrage and offense should not be racially balkanized in this way.

Racism is not the only valid grounds for offense, either. Mike Royko, the tough-talking columnist for the Chicago Tribune, displays an instructive array of confusions on this point in a column, “Time to Be Color Blind to All Words of Hatred.” Royko defends a cartoon intended to make the valid point that the use of words (not only actions) can be racist. Louis Farrakhan’s attacks on Jews were the intended example. The cartoon showed a Ku Klux Klan member holding a paper with Farrakhan’s remark, “You can’t be racist by talking—only by acting.” The Klan member says, “That nigger makes a lot of sense.”63

Many readers took offense at the cartoon—at Farrakhan being labeled a “nigger”—and Royko appears to have regarded these readers as morons who can not understand that the context and intent of a word’s use affects its moral standing. Obviously there is an important moral difference between mentioning the “n”-word in the course of criticizing it, and actually using the word as a racial insult. But Royko should understand that any use of this word by a white person in application to a black person, even if satiric and ironic, is offensive to most blacks and is generally inappropriate. As used by whites, “nigger” is a unique word in its congealing the set of attitudes embedded in white supremacy that expressed contempt, hatred, derision, and exclusion toward blacks.64 The legacy of white supremacy, has not been sufficiently transcended for whites simply to appropriate the term to make valid points through satire. The word is still hurtful and offensive, even if Royko is certainly correct that it has different significances in different contexts. (The use of the word by young blacks as a term of affection and bonding is a context-bound use that has confused many whites, and is objected to by many, especially older, blacks.65) Its use in the cartoon is not itself racist; but it is objectionable.

Certainly not every time someone feels offended by a racial term is the offense warranted. Meanings and intentions can be misinterpreted, and people can be oversensitive. Nor is it possible to draw a sharp line between warranted and unwarranted offense; sometimes potential offense should be avoided even when unwarranted. Perhaps a white Washington, D.C., official should not have used the word “niggardly” in the presence of blacks, even though that word has no historical or linguistic relation to “nigger,” apart from sounding like it. Still, it is not the causing of offense only that can be the source of moral objection, but the racially offensive remark itself.

Finally, people frequently fail to realize that it is wrong to tell a racist joke even though no member of the target group is present, because such jokes cause harm by contributing to hurtful, false, and degrading views of particular groups.

At the same time, not all remarks, jokes, and symbols that have a racial significance are necessarily racist. Jokes that depend on a stereotype of Italians as loving pizza or whites as a bit uptight may offend; but they are not racist. They do not portray their targets in a degraded or seriously deficient light. The characterization of something as racist must be done with care, partly to avoid emotionally and morally overloading a situation that does not warrant it, partly to assure that other ills and missteps in the racial arena garner their appropriate claim on our consciences, and partly to protect the severe opprobrium that currently attaches to the epithet “racist.”

Racist Beliefs and Racist Believings

As we have seen, the original definition of “racism” referred to a system of belief, regarding the biological inferiority of certain races to others, but current understandings of racism are not confined to and do not depend on such belief. What does it mean, then, for a belief to be racist? There are two distinct questions here: (1) when is it racist for an individual to believe something? (call this “racist believing,”); (2) what makes a proposition—the content of a belief—racist? (call this “a racist proposition [or belief]”).66

Racist believing is a state of belief to which one is led because of a racist attitude or sentiment. For example, some white people who espouse strong adherence to grades and test scores as the sole legitimate basis for selecting students for admission to college do so because they are prejudiced against blacks or Latinos, rather than because they have arrived at their belief through a dispassionate assessment of its pros and cons. In such cases it would be accurate to say that the individuals are racist in holding the belief. One test of the psychic basis of an individual’s espoused belief in test scores is whether she accepts the large number of Asian students who will be and have been admitted to top colleges when test scores are the major basis for admissions. If she shows hesitation on this matter—saying for example that the Asian students are “grinds” and are not sufficiently “well-rounded”—her inconsistency is grounds for suspicion of a racist motivation underlying her resistance to race-sensitive admission plans. (Even here, however, plain self-interest rather than personal racism may be operating in the resistance to admitting blacks and Latinos with lower test scores and to admitting Asian Americans with high scores.)

This is a case of a “racist believing” in a nonracist proposition. A racist proposition is one whose content is of a racial group characterized as deserving of racial antipathy or inferiorization: Examples are “Jews are evil and scheming,” “Blacks are intellectually inferior,” “Mexicans are lazy.”

An individual may hold a racist belief or proposition without recognizing this. Because beliefs are not necessarily consciously endorsed thoughts, a person may unthinkingly adopt a racist view of a particular group from her family, or from the society around her, without ever consciously endorsing it, being committed to it, or even contemplating it. Suppose, for example, that a friend of such an individual points out that she seems to be making the assumption that blacks are prone to violence and Mexicans lazy, and challenges her on her basis for holding these beliefs. The friend draws her attention to many nonviolent blacks and hard-working Mexicans. She realizes that she has held these racist beliefs without any basis, and she gives them up. In such a case, I think we want to say that the individual is not a racist and was not a racist even when she held the beliefs. The beliefs themselves were genuinely racist, but belief in them had not been deeply enough rooted in her psychological makeup to make her “a racist.”67 She must be genuinely committed to and invested in racist beliefs in order to be a racist.68 I do not mean to imply, however, that it is typical of someone holding a racist belief to be so minimally invested in it that she abandons it immediately in the face of valid counterargument; indeed it is not typical.

If one can hold racist beliefs without recognizing them and without being psychically invested in them, it is also possible to be unconsciously invested in such beliefs. An individual may believe blacks to be, as a group, intellectually deficient, and he may manifest this belief in his low expectations of blacks, in surprise that certain blacks achieve at a high level, and the like—all the while being entirely unaware that he holds this belief. Unfortunately this is true of many teachers (not only white teachers) in regard to black and Latino students. Psychic investment in a destructively racist belief is reason to call an individual “a racist”; but his unawareness of this investment seems to me to be a reason not to do so. However we resolve this matter, the commitment to a racist belief, and the discriminatory acts that flow therefrom, are cause for moral concern.

Some accounts of racist beliefs or propositions define the racism of a belief not by the content of the belief but by its effects. David Wellman, for example, defines racist beliefs as “culturally sanctioned beliefs which, regardless of the intentions involved, defend the advantages whites have because of the subordinated position of racial minorities.”69 Wellman may be conflating the racism of believing these propositions with the racism of the propositions themselves, but his view does have the virtue of highlighting the fact that beliefs without explicit racial content may nevertheless, when acted on, affect the relative economic or social position of racial groups. For example, the belief that college admission should be based solely or predominantly on grades and test scores has generally had the effect, when translated into policy, of decreasing Latino and African-American admission to elite schools. Thus the belief as put into practice contributes to racial inequality and racial injustice.70 This important fact about the belief and its corresponding policy may provide a basis for objecting to the policy.

But this does not make the belief itself a racist belief. Certainly the proposition that admissions decisions should be based on grades and test scores is not racist in itself. Nor is it necessarily racist to believe it. Perhaps admission policies based solely on test scores and grades do intensify educational and social inequality.71 This does not clinch the argument as to whether they are advisable or not, all things considered. One has to weigh the values and disvalues on both sides. Some argue that the gain from the use of “color blind” admissions criteria outweighs further harm to already disadvantaged populations.72 I disagree with this argument. However, calling the belief in question, or the believing of it, “racist” emotionally overloads a discussion of the validity or worthiness of the belief and leads us to think that because a view may have undesirable racial implications, it should be dismissed prior to discussion.

The charge of racism can be made only after determining that the motives of those espousing the belief are racist (not always an easy matter). Wellman’s definition could be taken to imply that everyone who holds such nonracist (but inequality-intensifying) belief does so for racist reasons.

“Institutional Racism”

“Institutional racism” is often understood to refer to a practice that is itself free of racial bias but in its implementation has a disproportionately negative effect on subordinate racial groups.73 The seniority system in the work place is often cited as an example: since previous employment discrimination has allowed whites to accumulate much more seniority than blacks and Latinos, a system of greater job protection and rewards for seniority has disproportionately negative impact on these racial groups. With less seniority, black and Latino workers are those most vulnerable to layoffs due to recession or downsizing. Yet the principle of seniority itself is entirely race-neutral; it simply favors workers with the greatest longevity in the company. (Nor was seniority originally adopted as a covert way to exclude or oppress blacks or Latinos.)

The seniority system does indeed hinder the progress of disadvantaged minorities. It perpetuates an existing racial injustice (unequal employment opportunity) caused by past practices of racial discrimination (excluding blacks and Latinos from workplaces and unions, providing them with inferior education, and so on). Seniority also places a limit on the effectiveness of affirmative action hiring programs. Even if a special attempt is made to hire blacks and Latinos, seniority renders the newer hirees most vulnerable to layoffs.

All these are valid and useful ways to express what is morally troubling about the practice of seniority. But seniority is a valuable institution in many ways, providing job security and a bulwark against management arbitrariness, discouraging destructive competitiveness among workers, and the like.74 (And to the extent that black and Latino workers attain seniority, they share in its benefits as well.) In an overall assessment of seniority, its benefits must be weighed against its race-related costs. Some modification of seniority to protect racial minority jobs would seem a reasonable compromise, honoring the moral pull on both sides.75 Calling seniority “institutional racism,” however, both implies that the practice was, or is, driven by racist motives (despite that implication’s being disclaimed in the definition), and tars it with an opprobrium that implies that it could never be morally acceptable.

Another practice that is often called “institutional racism” is employers’ recruiting by word of mouth among current workers in the company rather than advertising in job listings. There is a nonracial business rationale for doing so: it saves publicity costs, and it garners job applicants for whom a reliable worker has vouched, thus saving on the costs of assessing job suitability in a larger group of unknown applicants. Nevertheless, such recruiting has a disparate racial impact similar to that of seniority; in many occupations blacks and Latinos constitute a smaller proportion of the workforce than their percentage in the population, and workers’ networks are generally race-specific. Thus word-of-mouth recruiting perpetuates racial injustice and sustains the legacy of racial discrimination. Unlike seniority, however, it has less going for it ethically; it merely reduces business costs. Hence, in the service of racial justice, it should generally be abandoned; employers should seek qualified black and Latino applicants, or at least not employ practices that discourage them. But, again, “institutional racism” does not seem a helpful description, either in implying that racist motivations infect the working of processes that are in fact free of racial bias or in generating a judgment of overall moral opprobrium prior to examining the ethical pros and cons.76

It might be argued that word-of-mouth recruiting is sometimes employed with the intent to screen out potential black employees (even if that intent is not acknowledged).77 In this case such recruiting would be a racist practice; it would amount to racial discrimination (see chapter 4).78 The effects may be the same as when the practice is not intentionally discriminatory, but I suggest that this commonality is best expressed by saying that both practices equally sustain racial injustice, or have a deleterious impact on black job seekers, not by saying they are both examples of “institutional racism.” The second practice is (arguably) racist; the first is not.

The concept of institutional racism was invented as a way of recognizing that inequalities can be sustained, or intensified, independent of racist attitudes and motives, and even in accordance with “normal operating procedures” in the society. This is indeed an important racial and ethical insight. But it is better expressed, I suggest, by recognizing that some otherwise perhaps ethically sound practices can have the effect of sustaining racial injustice, rather than collapsing this distinction by implying that the processes are themselves necessarily morally illegitimate due to their being infected with racist intentions.

I would suggest confining the concept of “institutional racism” to what might more accurately be called “racist institutions”—specific schools or workplaces, systems, such as health care, criminal justice, and education, and the practices and processes therein which perpetrate racial inferiorization or antipathy.79

Specific institutions can be racist in two importantly distinct ways. First, their policies or practices can be intentionally racist. A clear case of such institutional racism was the subject of a 1969 Supreme Court case, Gaston County v. United States.80 Prior to 1965, Gaston County, South Carolina (like most southern voting districts) employed literacy tests as a condition for entitlement to vote. The tests were employed in an explicitly racially discriminatory manner, with blacks given lower scores than whites regardless of their actual performance on the tests. The 1965 Voting Rights Act struck down the use of literacy tests in most contexts, but a clause in the Act allowed a county to reinstate the tests if they administered them in a fair and unbiased manner. Gaston County attempted to conform to this requirement by restructuring its tests to be rid of racial bias in both content and administration. But because blacks had been systematically subjected to an inferior education under the segregated school systems, which still existed, they failed the literacy tests at a higher rate than whites. The Supreme Court ruled that this situation constituted depriving blacks of the equal right to vote that the Voting Rights Act was meant to ensure.

It is plausible to suppose that Gaston County sought that very result—that many blacks would be deprived of a vote (although the Court’s ruling did not depend on this implication). If so, Gaston County’s voting system was a racist institution both before and after the adjustment in the literacy tests, and intentional injustice was committed against blacks in both contexts and by both procedures. The means by which this injustice was perpetrated differed in the two contexts, but in the latter instance the county officials intended a racially discriminatory result, while using a race-neutral and facially unbiased means to achieve it. (One can imagine an unusually clueless but nonracist individual taking a job related to administering the tests without realizing this intent. Such an individual would then participate in institutional racism without herself being racist in her motives, attitudes, or beliefs.)

Intentionally racist institutions are of course not only a thing of the past. In June 2000, American General Life and Accident Insurance Co., one of the nation’s largest life insurance companies, agreed to pay $206 million to settle allegations that it had overcharged millions of mostly poor, black customers for burial insurance because of their race.81 In November 2000, Coca-Cola agreed to pay more than $156 million to current and former employees of color alleging racial discrimination and, in an unusual concession in employment discrimination cases, agreed to grant broad discrimination-monitoring powers to a panel of outsiders.82

A second way in which a specific institution or institutional practice can be racist is quite different, although the difference is one of degree. Here the institution has no official or intentional policy of racism or racial discrimination, yet the actual functioning of the institution involves racism or racial discrimination. A high school whose staff manifest contempt toward their black or Latino students would be a racist institution, as would a school whose staff are not actually contemptuous but nevertheless have low expectations of the students, essentially regarding them as racially inferior, and, as a result, providing them with inadequate education. Both such schools may well intend, as a matter of official policy, to do well by, and be nonracist in their treatment of, their students. Indeed, the principal and various staff members may even genuinely believe that they are doing as well as can be expected. These well-intentioned people may be sincere, and sincerely self-deceived. Racist institutions of this sort do not merely reflect the previously existing racist attitudes of their staff. Institutions possess an internal dynamic—an “institutional culture”—that can push not yet determinate attitudes in a more or less racist direction. In this sense the racism of institutions is not merely derivative from the preexisting racism of its individual members.

Institutions or practices that commit unwitting injustices against a racial group because of its race are institutionally racist in the unintentional sense. The direct committing of an injustice (even if unwitting) is morally different from employing a racially unbiased policy that has the effect of perpetuating a prior injustice. Not every employee of racist institutions (intentional or unintentional) must manifest racism, in order for an institution to be racist; some teachers in such schools, for instance, may treat students well and have appropriate expectations of them as learners. No precise line can be drawn as to the extent or pervasiveness of racist attitudes and behaviors within an institution in order for it to be appropriately called a “racist institution.”

A final caveat about the central insight underlying the origination of the concept of “institutional racism”—that processes not driven by racial considerations can have racially deleterious effects. In many cases, such processes are economic in character. For example, the disappearance of fairly well-paying and secure blue-collar jobs for minimally-educated workers has had a deleterious impact on the job prospects of poorly-schooled black workers living in urban centers.83 But if we view this development only in relation to its racial impact, we overlook its effect on all low-skilled poorly educated workers of any race. The ever-widening earnings gap between high-skilled and low-skilled workers of all races through the 1990s and beyond is a deeply troubling and pervasive form of class-based injustice plaguing the American polity. Its racial injustice (that is, its disproportionate impact on blacks and, to a lesser extent, Latinos) is only one part of the problem. It is true that the term “racial injustice” has largely disappeared from public discourse in the United States; but adequate language for class-based injustice has attenuated even more. Social-welfare policies serving all poor and working-class people have weakened since the 1980s,84 and general, race-independent inequalities in virtually every domain of life—access to health care, income, wealth, education, housing—reached record levels in the 1990s. As William Julius Wilson said in 1996, “The emphasis is on personal responsibility, not inequities in the larger society.”85 The moral importance of these injustices against people of all races should not get lost in a focus on their disproportionate impact on blacks and Latinos, a development encouraged by the more familiar use of “institutional racism.”

Why Is Racism a Moral Evil?

Different vices have different moral valences. It is worse to be cruel than to be inconsiderate, to be dishonest than to be thoughtless. Why is this? It may have something to do with the social or interpersonal damage done. Dishonesty is more corrosive to both interpersonal relations and social trust than is inconsiderateness. Why does “racism” connote the strong moral opprobrium I am concerned to preserve, against conceptual inflation, in its meaning and use? In its inferiorizing forms, it violates fundamental moral norms of respect, equality, and recognition of the dignity of other persons. This violation is exemplified both in attitudes that regard others as inferior beings, as well as in social systems that deny dignity. In its antipathy forms, it exemplifies unworthy and destructive sentiments and attitudes—bigotry, antipathy, hatred, malevolence.

But the moral wrong of race-based violations can not lie solely in their violating general moral norms, that is, violations that carry the same force for any victim. Otherwise, showing contempt for someone based on her race would have the same moral status as doing so because she has bad taste. What is it about racially-based violations of these human norms that intensifies the moral wrong involved.86

The additional opprobrium is racism’s integral tie to the social and systematic horrors of slavery, apartheid, Nazism, colonialism, segregation, imperialism, and the shameful treatment of Native Americans in the United States—all race-based systems of oppression.87 U.S. law recognizes that racially based wrongs are more serious than other similar wrongs by calling race a particularly “invidious” distinction. Because racial distinctions have been the source of the most heinous forms of systemic mistreatment, American law requires any policy that makes racial distinctions to pass the most stringent level of scrutiny as to whether its likely benefits outweigh its presumed wrongs. (The legal status of race as a basis of discrimination will be further discussed in chapter 4.)

“Racism” draws its moral valence from this historical context in two ways. First, the mere fact that these historical systems were based on race provides some of that opprobrium, even if current instances of racism no longer take place in the direct context of, for example, segregation, apartheid, or slavery. Because no historical systems have degraded whole groups of people because they were thought to have bad taste, scorning someone on these grounds, while certainly wrong, does not carry that historically-weighted moral opprobrium.

The second connection between the strong negative valence of “racism” and historical systems of subjugation is that we continue to live with the legacy of those systems. Dismantling the legal and other structural underpinnings of apartheid, slavery, and segregation did not automatically lead to justice for the subordinated groups. Racial injustice remains characteristic of Western society as a whole, and of some nations more profoundly than others. Were this injustice to be overcome, the epithet “racism” would lose some of its force (though not all, since racial prejudice and antipathy survives even in contexts of economic and civic equality). But current relations between racial groups continue to reflect past history.

Consider anti-Irish prejudice. At one point in American history, especially in the early years of substantial Irish immigration (1820–60), the Irish were seen as a degraded type of human being, almost as low on the scale of social esteem as blacks.88 Nor have prejudice and demeaning images of Irish people fully disappeared; the stereotype of Irish people as drunks still lives on in mainstream culture.

The Irish, however, are now entirely integrated into the fabric of American life as full equals; no general exclusion or stigma attached to being Irish exists in the general culture. Blacks’ and Native Americans’ situation is quite different. They are both still stigmatized and inferiorized groups. (With exceptions in certain parts of the nation, Native Americans are more marginalized but less stigmatized than blacks). Thus the accusation of being “anti-Irish” rightly carries much less moral force than being “antiblack” or “anti-Native American.” The stereotype of Irish as drunkards carries much less social power than does that of blacks and Native Americans as lazy, because the latter is a cultural representation still very much interwoven with the current unequal socioeconomic status of blacks and Native Americans as groups.89

Still, stereotyping of Irish people is damaging to individuals and the Irish as a group, leading to demeaning portrayals and even to corrosive internalizing of the stereotypes. That it is not morally equivalent to racism does not mean that it is not a serious moral ill. To think otherwise is to be blinded by the “racism or nothing” outlook.

Degrees of Racism

My account of racism, rooting it in its history, is meant to help us distinguish genuine racism from lesser racial ills and infractions. The critique of “categorial drift” turns on acknowledging moral degree also. It is a more serious moral reproach to call a person a racist than to call one of his acts racist. It is generally less morally serious to use a racist symbol innocently than to do so with explicit intent to demean, harm, or insult. And so on. And some acts, persons, and participants are more racist than others. People sometimes treat “racist,” as a concept, like “pregnant”—either you are or you aren’t. But there can be degrees of racism, just as there can be degrees of dishonesty and cruelty. A film can be mildly racist, or viciously racist—for instance, in its portrayal of Arabs, or blacks, as well as in some overall sense. An individual can be somewhat hostile or intensely hostile toward Mexicans. Some disputes about racism founder on a felt need to make racism an “all or nothing” matter. But, like most vices, racism comes in degrees, and it is worse to be more rather than less racist, even if it is bad to be racist in whatever degree.

The all-or-nothing approach prevents us from seeing the complexities in an individual’s overall view of a racial other. Some whites may harbor distinctly racist prejudices and feelings toward some blacks, or blacks in general, yet also genuinely admire, appreciate, and be attracted to other blacks, or to characteristics they perceive in blacks. Though racism sometimes does function in an all-or-nothing way, in which the racial other is seen wholly and unambivalently through the negative lens of racism,90 often it does not. In the interest of accuracy and of facilitating communication about these vexing matters, we would do well to recognize such complexity. As with all moralism, simplistic judgments in the racial domain lead to polarized positions, rendering fruitful communication difficult.

Once one moves away from the general idea of “racism” as a kind of large undifferentiated thing, an “impersonal force,” as Bob Blauner once referred to it,91 it is obvious that not all forms and instances of racism are equally heinous. An act of racist violence is worse than telling a racist joke. Believing in the human inferiority of a racial group is not as morally evil as acting on that belief in order to deprive the group of its rights. Harboring racist feelings that one never expresses is not as morally bad as expressing them whenever one has the chance.

But if the term “racism” carries such different moral valences in its many manifestations, what remains of the idea that it is always a term of strong moral opprobrium, for the reasons mentioned earlier? The answer, I think, is that the opprobrium operates within categories of racist manifestations. Thus racist beliefs are particularly vile types of beliefs: racist symbols, particularly vile symbols, and so forth. But the comparative vileness does not operate across categories. Racist belief is not necessarily more objectionable morally than harmful nonracist behavior.92

Selective Racism

Our dominant concept of personal racism and prejudice is the targeting of an entire racial group, where animus or disrespect directed toward one member is grounded in animus or disrespect toward the group itself.93 Much of what we call “racism” or “racial prejudice,” however, does not actually work this way. In contemporary America, for example, nonblacks direct much stronger animus toward black men than toward black women, toward poor blacks than toward well-to-do blacks.94 Asian women tend to be the object of more demeaning stereotypes than Asian men—as passive, compliant sex objects, for example. Ms. Verano, the white fourth-grade teacher, exhibits an analogous attitude of selectiveness in her discomfort with black adults but not black children. So let us call personal racism that is differentially directed toward distinct subgroups within a given racial group “selective racism.”

An individual can be a target of demeaning or hostility by reason of membership in more than one group. That Asian women are subject to greater demeaning than Asian men may reflect both racism and sexism; the “compliant sex object” stereotype is clearly a combination of the two. Black women, too, suffer from sexism as well as racism. But this dual prejudice is not selective racism, for it involves a uniform negative view of a racially defined group, combined with a gender prejudice. Selective racism is involved when one subgroup is stigmatized more than another—black men more than black women, for example.

How can there be such a thing as selective racism? If we have as our model of racism that the entire group must be targeted, then what I am calling “selective racism” might seem not to be racism. Suppose, for example, that a white person feels positively toward older black persons of either sex and negative only toward young black men. She might say, “I’m not prejudiced against blacks themselves, so I’m not racist. I just don’t like young black men. This might be some kind of prejudice, but it is not racial.”95

But some forms of, or explanations for, this individual’s type of animus does involve animus toward the entire group. She may harbor mild prejudice toward all blacks, yet feel positively toward the elderly; so, on balance, she feels positively toward elderly blacks. Yet, in order for this form of differential racial antipathy to be present, the subject must feel less positively toward elderly blacks than toward elderly nonblacks, even if her feelings toward elderly blacks are, taken as a whole, positive. Otherwise how could she have a negative affect toward blacks in general?

This explanation can not, however, apply to all forms of selective antipathy to young black males, for it is not plausible to postulate a characteristic that every black who is not a young black male possesses, toward which the subject in question feels favorably. Rather, what seems much more common is that the category “young black males” evokes a distinctively negative prejudice. Two explanations suggest themselves. First, the subject may think of this group as quintessentially black, the archetype of blackness. That she regards other black subgroups as not fully black shows that her prejudice is directed against blackness.96 The subject only accepts blacks whom she does not take to be fully black.97 A second explanation is that the prejudice is triggered only by the confluence of characteristics, not by any of them in isolation. The racial dimension “kicks in” only when youth and maleness are also present. The prejudice is not merely a sum of the prejudices against those characteristics singly.98

The first explanation of selectivity more clearly involves a form of racism or racial prejudice, as blackness per se is the target of hostility. The second is racism in the sense that the target’s race is a necessary feature of the prejudice; but it differs from familiar understandings of racism or racial prejudice because the entire group (or blackness in general) is not a target of prejudice. Hence “selective racism,” or “selective prejudice,” may express this form.

To name an act or a person “racism” or “racist” is particularly severe condemnation. But the terms are in danger of losing their moral force, for they have been subject to conceptual inflation (overexpansive usage) and moral overload (covering morally too diverse phenomena), thus inhibiting honest interracial exchange.

The conceptual inflation arises from three sources. First, “racism” and “racist” have come to be applied to virtually anything that goes wrong in the domain of race, leaving us with an apparent choice between calling something “racist” or seeing it as of no moral concern at all. This development is both a cause and a product of an attenuation of other ways to describe problematic racial phenomena (racial discomfort, racial ignorance, racial insensitivity, racial injustice). I suggest that we breathe new life into those linguistic resources. Second, “racism” has begun to be used for all forms of group discrimination, denials of dignity, and stigmatization (on the basis of sex, disability, sexual orientation, and religion). Third, people have become sloppy about the category of item they name as “racist.” This “categorial drift” tends to up the moral ante.

Despite their current ubiquity, the terms “racism” and “racist” are of relatively recent vintage. Their history (since their coinage in the 1930s) suggests that two distinct themes should be considered the core of racism: antipathy toward, and inferiorization of, a racial group. The moral force of the terms lies in both the violation of fundamental moral norms and in their relation to the evils of historical systems of oppression within which racist phenomena have been embedded.

The all-or-nothing way of thinking about racial malfeasance distorts our understanding of racism itself. Racism admits of degrees. Persons, motives, portrayals, and statements can be more and less racist (than others). And racist attitudes and prejudices can be targeted to some subpopulations of a racial group more than others.99