During the past ten years or so the study of prejudice, which has figured importantly in American sociology virtually since it established itself as a separate discipline, has developed several new emphases. “Intergroup relations” are no longer studied merely as one aspect of American social life, but as a problem that exists in many countries, and indeed as a factor in international relations. In the United States, moreover, the analysis of intergroup relations is no longer restricted to racial or ethnic minorities, but now includes minorities defined by religious, cultural, occupational, or political criteria. There has been a tendency, finally, to shift our interest from a social to a psychological framework—from behavior patterns or institutions to attitudes or beliefs, from discrimination to prejudice, from a sociologist’s to a social psychologist’s view of minority-majority relations.

There is nothing sacrosanct about the traditional analysis of group prejudice, and there is no reason why another generation of social scientists should not reflect its new interests by developing a different conceptual framework; but in my opinion a good deal of the revision of the older theory has not represented an improvement. Very often basic principles underlying a study, say, of Negro-white patterns in the United States have been applied to the analysis of superficially similar relations to which they are not relevant. In this essay an attempt will be made to spell out some of these principles and thus to delimit the range of situations to which the concept of prejudice is properly applicable.

_____________

Literally, prejudice means prejudgment, a judgment before knowledge. By group prejudice, then, we should mean both (1) a judgment, an evaluation of an ethnic or other group and its individual members, and (2) an incomplete or otherwise inadequate factual base to this evaluation. How far the concept has departed from this original meaning can be best indicated by citing a few examples. Thus, four experts writing in the authoritative Handbook of Social Psychology define the word as follows: “By prejudice we mean an ethnic attitude in which the reaction tendencies are predominantly negative. In other words, for us a prejudice is simply an unfavorable ethnic attitude.”¹ According to one of the best and most widely used texts in social psychology, “A prejudice is an unfavorable attitude—a predisposition to perceive, act, think, and feel in ways that are ‘against’ rather than ‘for’ another person or group.” It is contrasted with a “predisposition toward intimacy and/or helpfulness.”² Or, according to a recent sociology text, “Prejudice refers to those negative attitudes that create a predisposition toward unfavorable responses to a racial or minority group.”³

One element common to these definitions—and more could be cited—is the substitution of “adverse judgment” for “prejudgment,” or, in other words, the removal of the concept of prejudice from a factual context. But, one could ask, might not a “negative” or unfavorable attitude be, not a prejudgment, but rather the end product of a conscientious effort to arrive at the truth? Indeed this might be so, we are told by the first of these books, but “common experience would suggest that . . . the objective support for stereotypes is at best a minor aspect of stereotypic thinking.” Such a comment, however, begs the question by identifying unfavorable attitudes as stereotypes, which by definition are not based directly or mainly on empirical data. Surely not all unfavorable attitudes concerning all ethnic groups are prejudices, in the sense of pre-factual judgments. For example, the statement that Egyptians are 77 per cent illiterate does not reflect a prejudice. On the other hand, many stereotypes (in the same sense of pre-factual judgments) are either favorable or, at worst, ambivalent. Is it insulting, per se, to call the Irish witty, or the Jews clever, or the Germans industrious?

_____________

That an attitude can be designated as prejudiced, without relating it to the relevant facts, is often merely asserted as a bald statement; when this thesis is defended, the usual argument is that since all groups are to some degree heterogeneous, a general statement about any one of them, or about any individual member, reflects a prejudice. Thus according to the liberal cliché there is only one race, the human race. Race, in any other sense than a synonym for Species, is, in Jacques Barzun’s phrase, “the modern superstition,” as though there were no empirical basis for classifying, say, Chinese, Bantus, and Englishmen into separate categories. The argument rests on the fact that racial differentiation is not sharp or permanent, that however one classifies races there will be intermediate subgroups, if not immediately then certainly after the gene pools have been re-aligned during the next historical period. On the same basis, however, it would be reasonable to deny the reality of larger biological groups, for the differentiation between species or orders or even phyla is often arbitrary in the same way and is sometimes a point of dispute among specialists in systematics. Race is a fact; the superstition, the abomination, is racism.

Cultural differences are also often seen as non-existent; they are defined, as we have seen, as “stereotypes.” In one study, for example, Princeton undergraduates were asked to name the traits characteristic of ten different ethnic groups.⁴ The Chinese, as one instance, were designated as “superstitious, sly, conservative, tradition-loving, loyal to family ties, industrious, meditative, reserved, very religious, ignorant, deceitful, and quiet.” It is a commonplace that the Chinese family is an extraordinarily tight and resilient institution, and Sinologists would agree that the Chinese are typically—or were until very recently—“loyal to family ties.” The Chinese are also ordinarily “ignorant,” if by this we mean illiterate; but it would be difficult to determine whether they are “sly.” The stereotype thus comprises statements that are true, some that can be interpreted as true, and some that cannot be validated at all. But the authors make no attempt to differentiate in empirical terms among the twelve traits. Like other analysts of stereotypes, they implicitly maintain that a judgment of an ethnic group—any judgment—is a prejudgment.

Sometimes the irrelevance of facts to one’s judgment of a people is stated explicitly. The moral yardsticks with which we measure the relative worth of cultures, it is maintained, are themselves part of the cultures, and one may not evaluate any society except by the standards that its members use to judge themselves. Some peoples like beef, some pork, and some human flesh; it is all, as Molotov once said of fascism, a matter of taste. This version of “cultural relativism,” as it was developed by anthropologists a half century ago, was a necessary condition to understanding a primitive culture in its own terms; for if the equality of all cultures had not been postulated, a representative of Western civilization living with a tribe of two hundred naked food-gatherers would have been able to see nothing but their patent inferiority.

The classic statement of cultural relativism with respect to moral values is, perhaps, the late Ruth Benedict’s Patterns of Culture. I use it as one of the readings in my introductory sociology class, and even the dullest freshman can detect that her pretense not to judge among the three primitive peoples she describes is—a pretense. The brighter ones understand also that this was in part a comment on American culture; the contrast between the warlike Dobu or the highly competitive Kwakiutl on the one side, and the quiet, peaceable Zuni on the other, was an anthropologist’s restatement of the reactionary-progressive, aggressive-peaceloving dichotomy of the mid-1930’s. (Today, however, when a different sort of dichotomy is the fashion, the Zuni seem suspiciously “other-directed,” if not downright “conformist.”)

Two kinds of generalizations are made in the social sciences, statistical and modal. The first, statistical, is of the type, “City people have small families,” or “Jews vote Democratic”—meaning that the designated group shows proportionately more of a measurable quality than the rest of a defined population. A proscription of such generalizations on the ground that they do not apply to all individuals in the group would, if generally applied, mean the end of social science and of a good deal of physical science as well. Virtually every statement that we make about human groups, comparing the middle class with the working class or urban dwellers with rural, can be only a statement of probability. If we demanded universality of every generalization there would be none; for just as there are different kinds of Negroes so there are different kinds of city-dwellers, and just as there are persons marginal to “Negroness,” so there are suburbanites.

_____________

A modal generalization is an attempt to transcend the totality of statistical statements and describe the typical rather than the average. For example, the Chicago anthropologist Robert Redfield has distinguished two ways of life, “urban” and “folk” society; and Ruth Benedict, in The Chrysanthemum and the Sword, has characterized the Japanese as a paradoxical combination of delicate artistry and militarism. To delineate the Japanese or any other people as a “type,” however, is to attempt so tenuous an abstraction from the factual base that the result may be difficult to distinguish from a “stereotype.” Indeed, some of the studies of “national character” have been marked by gross bias (e., Brickner, Is Germany Incurable?), and others have blown up dubious Freudian hypotheses to preposterous dimensions (e., Gorer’s works on the Great Russians and the Americans).

It is not usual, however, to characterize such scholarly efforts as prejudiced. In the Handbook of Social Psychology noted above, the chapter preceding the one on prejudice reviews many of the studies on national character, and the reader passing on from one to the other may well be puzzled by the sudden change in climate. Identical generalizations—of the type, “The Irish are witty”—are discussed in the chapter on national character by referring to the object, the Irish, and in the chapter on prejudice by referring to the subject, the person making the statement. Brickner and Gorer have been sharply criticized, but the criticism has always been of their works; no one has suggested that Brickner has an “authoritarian personality” or that Gorer’s frustrations have made him aggressive.

Even when the empirical context of prejudice is acknowledged as pertinent, it is not certain that the relevant facts are indeed being examined to judge whether a person’s attitudes are prejudgments. One article tells us, for instance, that group prejudice is “a type of stereotype which does not correspond with the facts”—for example, “dislike or disgust for the ‘dirty Japs,’ the ‘Red menace,’ the not-to-be-trusted Catholics or capitalists or labor unions or foreigners.”⁵ The very list of examples indicates how difficult it may be in the social sciences to come by “the” facts. Although not arranged in an order to suggest this, the list constitutes a continuum from broad, heterogeneous aggregates (“foreigners”) to clearly defined political organizations with rather definite purposes (trade unions, world Communism). The difference between the two is fundamental. Antipathy to “foreigners,” since it cannot be based on knowledge of all of them as a group, must be prejudice; and it makes good sense to look for the cause of such prejudice in the psyche of the subject. Hatred of trade unions or of world Communism may also be and often is a neurotic symptom; but in this case the object is clearly enough defined to make possible an unambiguous attitude among the well informed, and given certain values this could be sharply negative without being in the slightest a prejudgment.

If we define group prejudice as a judgment about a group before the facts concerning it are known, it is thus possible to demonstrate the existence of prejudice with reasonable standards of reliability. The “facts” may be demonstrably false, as with the legend about the Protocols of the Elders of Zion. Or the “group” may be so vague as to lack any definable characteristics, as with “foreigners.” Or the sentiment may be so strongly expressed (“All Germans are pigs”) as to suggest no relation to empirical evidence. In many specific cases, however, it is extremely difficult to distinguish prejudice from a tentative judgment based on the inadequate facts that happen to be available. In order to make this distinction in general terms, it is necessary to state a bit more definitely what we mean by a group.

_____________

What is a minority group? A statistical category constituting less than half—persons whose last name begins with “A,” for example—does not comprise a minority in a sociological sense. There must also be some feeling of group coherence and a specific pattern of interaction with the rest of the population. But a minority is still a group relatively small in numbers. It confuses the issue to equate this originally statistical term with the political concept of subordination or the sociological one of discrimination. Nevertheless, such an identification is frequently made, either by implication or explicitly. As Louis Wirth has put it, “The people whom we regard as a minority may actually, from a numerical standpoint, be a majority.”⁶

Such an extension of the meaning of the word is not only semantically inelegant but also conceptually impermissible, if only because it glosses over the important distinction between democratic and non-democratic societies. The very idea of a minority suffering from discrimination implies a democratic moral judgment. In most of history, as well as in most of the world today, the overwhelming social division has been between a small ruling elite and a vast ruled mass, with the latter not significantly differentiated by the possession of civil rights or their lack. All of the 97 per cent of the Soviet population not members of the Communist party, for instance, suffer from what in democratic terms would be considered discrimination, but this is no reason for designating this 97 per cent a “minority.” The “American dilemma,” as Gunnar Myrdal termed it, reflects a conflict between the democratic ideal of equal opportunity for all and the actual inequality. In this sense, there is no Nazi and no Soviet dilemma. It is only in a democratic state that what Tocqueville called “the tyranny of the majority” can exist, and “minority problems” in the traditional sense are thus restricted to democratic societies.

The most significant minority in American society, both numerically and in other respects, is the Negro tenth of the population, and the analysis of Negro-white relations often has been used as a model for analyzing other intergroup relations. Negroes differ from most other minorities, however, in three fundamental respects:

1. It is meaningful to set the Negro tenth off against the non-Negro nine-tenths—or, in Wirth’s terms, the “minority” against the “dominant.” Lutherans, for example, also comprise a small proportion of the American population, but it would not ordinarily be appropriate to designate the rest simply as “non-Lutherans.” In this case, the division obviously should not be into two sectors but into the many religious denominations in the United States. While there are also intermediate groups between Negroes and non-Negroes, both other minorities and whites committed to the fight against discrimination, the dichotomy in this case reflects the most significant feature of Negro-white relations. A minority, that is to say, is a relatively small group that is opposed by a relatively unified majority.

This distinction has been ignored in many recent works, for example one on religious life and “discrimination” in the Netherlands—W. Goddijm’s Catholic Minority and Protestant Dominant. This study illustrates very well the dangers involved in transporting sociological concepts from one culture to another. Dutch Catholics, who constitute about 38 per cent of the population, are the largest coherent group in the political and social life of the country. The remaining 62 per cent, too heterogeneous to be considered a unit, are not “dominant”; and the Dutch Catholics are not a “minority” but a plurality.

2. Negroes are set off from the white majority by skin color and hair texture; and such racial characteristics, according to the unanimous judgment of present-day biologists, are all superficial, irrelevant to the roles that Negroes could play in American society. Races do not differ in intelligence, diligence, “criminality,” or other psychic qualities of comparable significance. The racist doctrine that Negroes are inferior in such respects, however, has led to cultural patterns by which they are made inferior. If, for example, Negroes are defined as innately unintelligent, then they need have only second-rate schools—and the inadequate training they get there results, indeed, in an over-all lower level of intelligence. That is to say, the segregation of Negroes from white society depends on only two factors: the superstition of racism, and the social policies that follow from this superstition.

3. The same point can be made positively rather than negatively: the American Negro is wholly assimilationist. His many demands can be summed up in a single one—that he become invisible in white society. He wants his ghettos to merge into the rest of the cities they now disfigure; he wants the differentials by occupation, by mortality, by illegitimacy, by any other social statistic, to become a thing of the past; he even wants (if this were possible) his color to disappear. No American of European stock, not even a descendant of the first colonists, is so thoroughly embedded in American culture as the typical Negro, so lacking in vestigial emotional ties to the “home country.” (To be sure, some Negro intellectuals now have a sympathy with such Western-educated Africans as Nkrumah, Premier of Ghana, but this is not a bond with another country.)

_____________

But simply to disappear is not the usual aspiration of a minority. In Europe, on the contrary, the typical relation has been the minority’s attempt to retain intact its language or other cultural attributes, in spite of the dominant group’s efforts to absorb it completely. An American who studies the reports of the League of Nations or the United Nations commissions on minority questions is likely to be somewhat startled by them; for in the judgment of these international bodies, “the” minority problem is typified by the European pattern rather than by Negro-white relations in the United States. This estimate would seem to be justified, moreover, for even the other minorities in American society tend to one degree or another to approach the European type.

Most minorities in the United States, thus, identify with the general American culture in terms of “universals,” but want to maintain their own “alternative” values.⁷ For example, members of the various churches are at one with the rest of the population in upholding the democratic credo, but differ with respect to their several dogmas. In the United States such a differentiation is regarded as not merely permissible but good; but even a society that considers cultural variation to be healthy cannot permit it over an infinite range. Thus, the Mormon practice of polygamy was declared to be illegitimate, even for this very small sect. Whether in the particular case this was a correct decision is not the question; the point is rather that some general values are felt to be so absolute that no alternative to them can be tolerated, and if the line is not drawn at polygamy then it would have to be at cannibalism or ritual murder or what have you.

Sometimes minority groups, on the other hand, consider their different values of such fundamental importance that, by one means or another, they try to impose such values on the rest of the population, to transform them from alternatives to universals. Thus, there has been an attempt on the part of Fundamentalist Protestants to make Prohibition the law of the land; on the part of Catholics to regulate all marriages in terms of their dogma on divorce and birth control; on the part of Zionists to influence American foreign policy in the Middle East. Such activities are completely legitimate; they are the stuff of which democracy is made.

What is not legitimate, however, is to behave like a pressure group and at the same time try to retain the protection morally due a minority. This kind of opportunism has worked well for ethnic and religious minorities, and now occupational groups have also begun to define themselves as minorities. We are told by an officer of the National Grange that farmers, who now comprise only 13 per cent of the American population, are therefore a “minority group,” and that higher education “must shoulder increasing responsibility for helping minority groups acquaint the vast majority of the population, especially the thought leaders, with the problems and characteristics of their particular segment of the business and professional world.”⁸ That is, American colleges ought to set themselves up as propaganda agents for the farmers; but if for farmers, then why not also for plumbers, and physicians, and service-station attendants—and professors of sociology?

That one’s attitude toward various groups should differ according to what these groups stand for would not seem to be a very remarkable statement. In an area less dominated by sentimentality and confusion than the analysis of ethnic relations, it would be no contribution to point out that to fight discrimination against Negroes, to disagree with Catholics about the moral legitimacy of birth control, and to abhor and oppose Communism, are not inconsistent attitudes simply because all three are “minority” groups. In a very large proportion of recent writing on this subject, however, such a patent distinction has been passed over, either explicitly or by drawing our attention elsewhere.

One way of avoiding this rather obvious differentiation has been, as we have noted, to concentrate on the allegedly prejudiced person rather than on the object of his alleged prejudice. Hatred of Negroes is typically irrational, and anti-Catholicism or anti-Communism is often irrational. If we conceive of prejudice as the expression of irrationality of the subject and devote our main effort to understanding the mental process that results in such an attitude, it is a matter of relative indifference what the object of the prejudice may be. This emphasis, of course, has been in part a reflection of the professional interests of psychologists, but it is remarkable, too, how much it has overlapped with a tolerance of Soviet totalitarianism. From the point of view of Communists and their sympathizers, it has been very convenient to analyze, say, anti-Semitism and anti-Communism—though not usually anti-Catholicism—as analogous expressions of a sick mind.

_____________

Perhaps the most important of the early expressions of the emphasis on the subjective and psychological was John Dollard’s Frustration and Aggression, issued by the Institute of Human Relations at Yale University in 1939. The thesis of this work, that the frustration of a desire typically leads to aggressive attitudes and behavior toward those regarded as actual or potential frustrators, was developed so broadly that the authors very nearly succeeded in bringing all of human history within its confines. The theme is introduced with a solo piccolo: little James wants an ice-cream cone, is denied it, has a temper tantrum. This reaction might have been avoided by putting “the aggressive response into a response-sequence that leads to non-rewards”—or, translating, by punishing James. As he grows up, however, he will find that life is full of frustrations and that in many cases the aggressive behavior allegedly resulting from them is not non-rewarded. Thus, with the theme fully orchestrated: “In Marxian doctrine, the theories of the class-struggle and of the nature of the state depend to some extent . . . on the frustration-aggression principle. . . . In his The State and Revolution, [Lenin] states plainly: ‘The State is the product and the manifestation of the irreconcilability of class antagonisms. The State arises when, where, and to the extent that the class antagonisms cannot be objectively reconciled.’ ” The Leninist theory of the state as the “special repressive force” is thus subsumed under Freud’s theory that all culture frustrates the primal libidinous desires of man.

A theory that encompasses everything from ice-cream cones to class struggle would not leave out group prejudice: “The existence of a social prejudice against a group of people is evidence, first, that those who have the prejudice have been frustrated and, secondly, that they are expressing their aggression or part of it in fairly uniform fashion.” However, this frustration-aggression theory, even if we accept it as valid, tells us nothing about prejudice. Frustration is a universal component of human existence and the interesting question is not whether within each psyche this frustration develops into aggression, but why some of the aggression should be expressed in some “fairly uniform fashion.” The theory that a minority group is a “scapegoat” implies that the scapegoat might just as well have been anyone else, not only that the victim is innocent of the charges made but that the charge has no relevance to the social context.

Thus, if the anti-Soviet feeling in the United States is due to psychically generated aggression, then the problem is to find a substitute scapegoat. “If one can find a villain who is opposing cooperative behavior and get the group to attack that villain, then cooperative behavior might emerge” (Professor Erich Lindemann, Harvard University). “Couldn’t the scapegoat be the Congressmen who oppose the Marshall Plan? . . . Here is the possibility for an alternative scapegoat” (Professor Bernard T. Feld, Massachusetts Institute of Technology). “Isn’t it conceivable that our objective should be to recognize the scapegoating mechanism and the projectivity mechanism? . . . For instance, you expose some of the Red-baiters, and show just what’s biting them. Or you analyze a good case of race prejudice” (Professor Gordon W. All-port, Harvard University). This whole discussion—revolving around the question, “Is our attitude toward Russia based on deliberately falsified information, or on misinterpretation?” (Dr. Ralph W. Burhoe, executive officer of the American Academy of Arts and Sciences)—is a fascinating document.⁹

The most influential application of psychological concepts to the analysis of prejudice is The Authoritarian Personality, by T. W. Adorno and his associates. The thesis of this work, in brief, is that different patterns of child-rearing in the United States have led to two contrasting types of personality, designated as “authoritarian” and “democratic,” and that prejudice or the lack of it is essentially a reflection of these two psyches. “Historical factors or economic forces operating in our society to promote or to diminish ethnic prejudices are clearly beyond the scope of our investigation.” In the eight years since it was published, The Authoritarian Personality has been subjected to a barrage of criticism and paid the homage of many imitative studies; a bibliography of works discussing it—which is not even a complete one—runs to 230 items.¹⁰ It is impossible to do justice to so extensive a body of work here; I will only repeat the one point I am seeking to make—that the concentration by these psychologists on the allegedly prejudiced person has also facilitated a confusion between prejudice and an unfavorable judgment objectively grounded.

_____________

Only a naive person would be unable to distinguish in principle between minorities like the Negroes and those typified by Communists, but unfortunately many of those who write on the subject of ethnic relations are naive about Soviet society. Among American social scientists (apart from a handful of experts), the ignorance that still remains about the Soviet Union is great, and—in spite of the tremendous bibliography on the subject—nowhere greater than in the matter of the Soviet minorities. One need have read very little on ethnic relations in the United States, particularly in books published before about 1950, to become surfeited with statements—even in works of well-known and generally responsible scholars—that the Soviet government follows a policy toward its minorities “which accords with the best scientific knowledge and the most enlightened moral principles.”¹¹ The same sentiment, that “the Russians have welcomed cultural differences and they have refused to treat them as inferiorities,” was disseminated to a mass audience of more than a million readers in the form of a Public Affairs pamphlet.¹² For persons interested in combating ethnic prejudice, the clear implication of the alleged lack of it in the Soviet Union was that our tolerance must be extended to include both that enlightened regime and its American representatives.

_____________

Let me sum up. With respect to ethnic minorities, it is generally true that the problems surrounding group relations derive less from actual differences than from the way these differences are interpreted. But there is a real and significant difference between, for example, Communism and democracy, and to understand why a man is anti-Communist it is necessary first to look not into his mind but at Soviet society. The tolerance that in a democratic society is morally due ethnic minorities cannot legitimately be extended to totalitarian groups.

The difference is a fundamental one, and the way to maintain it is to analyze alleged prejudices in a factual context. That is, I am pleading for the abandonment of the definition of prejudice that has become usual in recent years: simply a hostile attitude; and a return to an understanding of it as an attitude based on incomplete knowledge.

_____________

¹ John Harding, Bernard Kutner, Harold Proshansky, and Isidor Chein, “Prejudice and Ethnic Relations,” in Handbook of Social Psychology, ed. Gardner Lindzey (1954), II, 1021-1061.

² Theodore M. Newcomb, Social Psychology (1950), pp. 574-575.

³ T. Lynn Smith, Social Problems (1955), pp. 423-424.

⁴ Daniel Katz and Kenneth W. Braly, “Verbal Stereotypes and Racial Prejudice,” in Readings in Social Psychology, ed. Guy E. Swanson, Theodore M. Newcomb, and Eugene L. Hartley (1952).

⁵ Ronald Lippitt and Marian Radke, “New Trends in the investigation of Prejudice,” Annals of the American Academy of Political and Social Science, 244 (March 1946), 167-176.

⁶ Louis Wirth, “The Problem of Minority Groups,” in The Science of Man in the World Crisis, ed. Ralph Linton (1945), pp. 347-372.

⁷ The terms are from Ralph Linton's Study of Man (1936). I have discussed the differentiation at greater length in an earlier article in COMMENTARY, “Immigration and Acculturation” (November 1956).

⁸ Roy E. Battles, “Higher Education and Agriculture,” in Higher Education and the Society It Serves, ed. Raymond F. Howes (1956), pp. 44-55.

⁹ See Journal of Social Issues, 4:1 (Winter 1948), pp. 21-41.

¹⁰ Richard Christie and Peggy Cook, “A Guide to Published Literature Relating to the Authoritarian Personality through 1956,” Journal of Psychology, 45 (1958), pp. 171-191.

¹¹ Wirth, op. cit . Worth's authorities for this remarkable dictum were two—Bernhard J. Stern, who till his death was at least a close sympathizer with the American Communist party, and the chief Soviet expert on the subject, the author of Marxism and the National and Colonial Question, Joseph Stalin.

¹²Ruth Benedict and Gene Weltfish, The Races of Mankind (1943), p. 26. The sentence has been deleted from the 16th edition (1956), but this continues to discuss ethnic prejudice as though it were limited to Nazi Germany and the United States.