In recent years, there has been much thought and discussion on the problem of meeting the impact of prejudice on children. The other side of the coin—the fact that children generally come across prejudice only when it has infected other children—has received less attention, but is possibly the root of the matter. Here Miriam Reimann reviews a variety of social psychological studies that have considered the question of prejudice in children—its genesis, its causes, the way it gets expressed.

_____________

A group of seven-year-olds at a progressive school were reporting what they had done over their Thanksgiving vacation. A dark-skinned Negro boy, one of the most popular children in the class, said he had gone to watch television at the home of his uncle, Jackie Robinson.

“That’s impossible! His uncle can’t be Jackie Robinson,” a white boy shouts out.

“Why is it impossible?” the teacher asks.

“Because Jackie Robinson is colored!”

This seven-year-old already knew that “colored” meant something to the world, but it meant nothing concrete to him; and his bizarre confusion is typical of children’s awareness of race. But his ability to use such general terms as “colored” will grow as he grows; in a year or so he will know that Jackie Robinson is colored, and that his schoolmate is colored, too. Whether he will also be “prejudiced” toward Jackie Robinson and his schoolmate, and just what his “prejudice” will consist of—these are more difficult questions.

In the last fifteen years, psychologists and sociologists have been exploring the origins and development of children’s attitudes to “minority” groups. In the long run, sound intercultural education programs, designed to reduce prejudice and increase the rational acceptance of differences, will have to be built on such knowledge. But the results of research are now fragmentized into a great many articles published in scientific journals, much of it unknown to the thousands of teachers and community leaders concerned with developing intercultural education. Moreover, for all the effort so far expended, our knowledge is still very incomplete, and researchers’ conclusions are, or appear to be, contradictory.

Four groups of questions are asked by those studying prejudice in children:

1. At what age do signs of prejudice appear among children? Does prejudice increase or abate with age? And what are the usual manifestations of prejudice at various age levels?
2. Is prejudice in children a social habit acquired from the culture, or is it largely a compensation for personal insecurity, springing from the needs and defects of the individual personality?
3. What social experiences most affect children’s group attitudes? The conscious lessons in tolerance of a teacher? The unconscious lessons of prejudiced parents and other adults? The unthinking remarks of other children? How does the child handle the contradictions between the viewpoints of teachers, parents, and other children?
4. What can be done about children’s prejudices? Does organized contact—in school or clubs—with members of minority groups prevent the acceptance of prejudiced stereotypes? What techniques might schools adopt for the prevention or elimination of childhood prejudices?

_____________

Some answers to the first set of questions can be found in Bruno Lasker’s Race Attitudes in Children (Holt, 1929), an impressive collection of anecdotal material and speculations about children, assembled from adults in all parts of the country. This book was not an attempt to “measure” childhood prejudice, nor even a proof of anything—except that children are prejudiced, and in many confused ways. From all parts of the country, from school teachers, parents, social workers, and group leaders, came reports and examples of prejudiced behavior.

Lasker’s informants found evidence that children were aware of other groups, and prejudiced against them, as early as in the kindergarten years. (Many educators to whom I have talked still maintain that few children under eight or ten even recognize group or color differences. Lasker’s conclusions, however, have been supported by the work of later investigators in the field, notably that of Radke, Davis, and Trager, discussed below. And what many adults take to be innocence on the part of young children may be rather a sophisticated reluctance to exhibit what is officially forbidden.) Lasker observed that what we might consider prejudice is variously signified in various age groups: “The small child is more apt to exhibit signs of fear, the child of early school age teasing and combativeness, either associated with or soon followed by a sense of ridicule—more amused than malicious—for strangers in appearance, language or manners. . . .”

Prejudiced attitudes, Lasker generalized, are not deliberately taught but are transmitted without conscious intention to the growing child by parents and other adults, and children. (Later, Eugene L. and Ruth E. Horowitz showed that the parents of prejudiced children in Tennessee specifically disclaimed teaching their children prejudice, and that older children denied their parents had had any role in forming their attitudes—“Development of Social Attitudes in Children,” Sociometry, 1938.)

Indeed, it appeared to Lasker that prejudice flourished despite the formal ideology of racial equality fostered in many public schools. A Midwestern high school teacher, who is quoted often in the book, claimed that “it is quite practicable to eliminate race prejudice as an active force in school matters,” but that even the complete absence of racial discrimination within the school had nothing to do with what the children did outside the school. Colored and white boys formed distinct groups on the street during the noon hours. “It is a question with me,” this teacher concluded, “whether mixed education is likely to touch the social separation of the races.” Another Midwestern teacher reported the same phenomenon in the primary grades. In their school games the children chose foreigners as partners as often as any others, but their choice of partners when they were unsupervised—while leaving the building or on the playground—was not so “democratic.” Thus, even at the ages of five to eight, these children had learned the niceties characteristic of prejudiced but well-mannered adults.

Among adolescents, Lasker found that the split between “official” and actual behavior widened. On the one hand, adolescents have a deeper understanding of history and of the ideals of the American creed. On the other, their more conscious competitiveness in scholarship and in sports, and their awareness of the sexual implications of intergroup mixing, predispose them to greater prejudice.

Lasker observed, on the basis of this and other evidence, that it is questionable whether individual contacts with members of other groups effectively offset other, prejudice-forming, influences. And if they don’t, “it is hardly worthwhile to engineer such contacts.” The public school system is commonly supposed to be America’s great democratizing influence. But clearly the unsegregated school is not the only or major influence shaping the child’s attitudes on race. If Lasker’s premise of the unconscious learning of prejudice is correct, one might have to borrow a concept from psychoanalysis and consider whether the best way to combat children’s prejudices is not by revelation of the sources of their prejudices rather than simply by ideological counter-propaganda. Or perhaps we should look toward a change in the social situation that creates these attitudes.

_____________

Dr. Lasker’s book was frankly exploratory. A later attack on this problem concentrated on the question: how does the child reconcile conflicting influences? This study was R. D. Minard’s Race Attitudes of Iowa Children (University of Iowa, 1931). Minard gave a verbal questionnaire to 1,641 Iowa children from grades seven to twelve. The questions asked were of two types: those requiring a judgment of the behavior of fictional persons, and those requiring the subject to indicate how he would act in hypothetical situations.

A typical question of the first category ran: “A young lady belonged to a sorority at a state university. There was a faint trace of Negro blood in her ancestry, although no one would ever have suspected the fact from her appearance, and the young lady was herself unaware of the fact. The truth was revealed by accident. After the discovery it was suggested that she resign from the sorority, and she did so. Question: Ought this young lady to have been asked to resign from the sorority?”

Questions of the second type inquired if the subject would just as soon have a trace of Negro blood that didn’t show, or if the subject would just as soon marry a person who did have such a trace.

The interesting fact revealed in this study is that, as the children got older, their objective judgments of situations involving members of minority groups (that is, their answers to questions of the first type) became more tolerant, but their personal responses (that is, their answers to questions of the second type) indicated greater prejudice. Apparently their personal emotional responses were somehow immune to their intellectualized attitudes.¹

_____________

We can observe how the growing child’s attitudes toward minority groups become immunized to official ideology by looking at another study, which tested the attitudes of some twelve-year-old children in Cincinnati in 1931, and retested them in succeeding years (Rose Zeligs, “Children’s Intergroup Attitudes,” Journal of Genetic Psychology, 1948). Dr. Zeligs concluded that race and nationality prejudices seem to be supported and perpetuated by patterns and stereotypes “deeply ingrained in our children’s social environment and . . . molding their attitude.” It is especially interesting to observe in the following series of responses how one girl’s prejudiced feelings were rationalized and reconciled with other conflicting feelings and opinions:

1931. The Negro isn’t of my race of people. Most of the time white people don’t associate with Negroes. They are not clean. Some of the girls are rough and not careful when they play.

1933. I don’t like the Negro race. I know the ones in our school are awful wild. They are unclean, unpleasant race to have around.

1937. In some respects I dislike the Negro intensely because of their unclean ways of living, and yet, at times I pity them because people are so prejudiced against them.

At the age of twelve, this girl gave as her first reason a strong ingroup feeling (“the Negro isn’t of my race”); as her second, social requirements (“white people don’t associate with Negroes”); as her third, a popular stereotype (“they are not clean”); and finally, almost as an afterthought, her personal experience (“some of the girls are rough . . .”). At the age of fourteen, she raised her personal experience to first place. At the age of eighteen her childhood stereotypes remained in force, but she had developed some conflict about allowing them completely to dominate her, and we hear the echo of the conflict when her answer implies that unprejudiced views are supposed to be part of the “American way.” This very awareness, however, permits her to remain prejudiced because she now feels she has taken the demands of tolerance into account. Her prejudice is legitimized by lip service to tolerance.

_____________

In recent years the personalities of prejudiced and unprejudiced persons have been studied, and significant correlations between personality and prejudice have been found. The most intensive study on this subject was done by a research group at the University of California at Berkeley, whose work on adults is reported in The Authoritarian Personality (Harper, 1950). One of the authors of this work, Else Frenkel-Brunswik, also studied a group of children aged eleven to sixteen (“A Study of Prejudice in Children,” Human Relations, 1948). And for this group, too, she found that prejudice is not an isolated sentiment but part of a complex of attitudes toward men and society; and that this general complex is in turn related to the whole emotional orientation of the individual.

Fifteen hundred children were presented with a series of about fifty slogans relating to race attitudes as well as to more general social attitudes, and on the basis of their scores were divided into an upper 25 per cent of children considered “unprejudiced” or “liberal,” and a lower 25 per cent considered “prejudiced” or, in the researcher’s term, “ethnocentric.”

The prejudiced children tended to have what might be called a conservative and self-oriented attitude toward society. For example, answering the questions, “What is wrong with America today, and how would you change America?” ethnocentric children gave such answers as:

Taxes on everything, and the cost of living.

Clean up the streets—all that garbage lying around! See that everything is in order.

The unprejudiced children, on the other hand, mentioned the “atomic bomb, the condition of the Negroes in the South, the need for world peace,” and so on.

Ethnocentric children tended also to be conservative in their attitudes toward the roles of the sexes in society. Prejudiced girls and boys both tended to agree that “girls should only learn things that are useful around the house.” An ethnocentric girl said that when girls are around boys, they should “act like a lady, not like a bunch of hoodlums. Girls should not ask boys to date. It’s not ladylike.” Ethnocentric children placed a high premium on good manners, order, social approval, cleanliness. The stress on social approval and on power converge to produce an attitude of awe toward money, which is considered as having an exaggerated power for good or evil. A typical comment of a prejudiced child was: “No dollar, no friend; have a dollar, got a friend”; and the reverse of the coin: “It [money] helps make enemies. Money is the root of all evil, they say.”

The prejudiced children’s aversion to weakness extended to those who have power over them, their parents and their teachers, from whom they demand harsh discipline and punishment for misdemeanors and failures. “For what should the hardest punishment be administered?” elicited the following replies from some of them:

Naturally for murder, the next is for not paying attention to her mother and father. She should be sent to a juvenile home for not paying attention to her parents.

Talking back, not minding; for example, if you are supposed to saw a certain amount of wood in one hour and don’t do it you should be punished for it.

The ethnocentric children looked toward their parents as a source of power rather than of love. Liberal children stressed a comradely relationship to their parents.

In short, Dr. Frenkel-Brunswik concludes that the attitudes of ethnocentric children are outgrowths of their central personality, and indeed of the type of personality depicted (in adults) in The Authoritarian Personality—though in general it is less firmly established. Their prejudice stems from their effort toward rigid conformity to what they conceive to be the values of awe-inspiring and authoritarian figures. Their rigid conformism obstructs a flexible social attitude, and facilitates the acceptance of black-or-white group stereotypes, particularly those they see dominant in society.

This analysis applies only to the extremely prejudiced children on the one hand, and those most free of prejudice on the other. We should realize there has been no such exhaustive inquiry into the personalities and attitudes of the large group that falls into neither the upper nor the lower quarters, but is in the middle ground comprising 50 per cent of the children (although a study by Milton Rokeach, “Generalized Mental Rigidity as a Factor in Ethnocentrism,” Journal of Abnormal Psychology, 1948, found a positive correlation among the entire population studied between prejudice and general mental rigidity). Conceivably, the study of milder prejudices might not reveal this psychological pattern linked with prejudice.

_____________

The conclusions to be drawn from the University of California study might indicate that the ethnocentric person requires psychotherapy for his prejudices. Other studies, however, suggest that the milder prejudices of children, as seen in verbal expressions and behavior, can be modified by a less intensive attack.

Virginia M. Axline (“Play Therapy and Race Conflict in Young Children,” Journal of Abnormal and Social Psychology, 1948) has presented an extremely interesting report on a play therapy course conducted with four seven-year-old public school children, one of whom was colored. The children were what was formerly called “bad,” now “disturbed,” and were taken for a one-hour session once a week to a room for non-directed therapy under a trained therapist. The atmosphere was completely “permissive”—no behavior was prohibited except direct physical violence against persons.

The race problem came up five times, and did not arise at all during the last five meetings. Miss Axline reports the children’s statements and actions in stenographic detail, and we see how the immediate hurt reaction of a Negro girl to verbal attacks by one of the boys eventually leads him to try to make amends. Miss Axline’s conclusion was: “When one provides a situation wherein the children are given an opportunity to be themselves—and an opportunity to react in a very permissive situation, then it seems that they can more readily come to terms with their own attitudes and emotions; and in a face-to-face situation . . . they can and do assume responsibility for their attitudes. . . .”

But most children rarely if ever have the “opportunity to react” provided by the completely permissive situation of play therapy. It is a delicate task to elicit frank and uninhibited responses from children on a subject that is as embarrassing and taboo to them as sex used to be. To meet this problem, many psychologists have employed rather subtle and indirect approaches to the study of race attitudes.

Eugene Horowitz (“The Development of Attitude toward the Negro,” Archives of Psychology, 1936) showed pictures of white and colored faces to comparable groups of New York City and Southern schoolboys and asked the subjects which they preferred: “Show me which you like best; show me which you would like to live next door to you,” and so on. He found a continuous development of prejudice from year to year, but no significant differences in prejudice scores between the Northern and Southern groups, and no evident correlation of prejudice with the amount of contact by white subjects with Negro schoolmates. Among the groups tested was a class which had a very popular Negro boy in it, and several other classes in mixed public schools, but these did not diverge from the general results obtained. The only radically divergent results were found in a group of boys attending a Communist settlement house in New York, who showed no prejudice and possibly a slight pro-Negro bias.

Two years later, E. L. and R. E. Horowitz (“Development of Social Attitudes in Children,” Sociometry, 1938) conducted another “show-me” study among Tennessee boys and girls, but this time pictures of both sexes as well as both races were shown. To these children, race was a more important factor than sex in their choices; that is, a white boy, presented with a picture of a white girl and a colored boy, would be more likely to prefer the white girl, although in tests (written tests requiring the subject to state whom he would prefer to sit next to) given in Northern communities, sex had been found more important than race in comparable age groups. In other words, among Northern children one’s own sex is preferred to one’s own race; among Southern children, one’s own race is preferred to one’s own sex.

The Horowitzes suggested (“Race Attitudes,” in Characteristics of the American Negro, ed. Otto Klineberg, Harper, 1948) that while his earlier study indicates there is no apparent difference between Northern and Southern children in the prevalence of anti-Negro prejudice in the population, his later one suggests that there is a difference in the relative weight of prejudice in relation to other factors.

_____________

Summarizing the results of the studies reported to this point, it would appear that the questions asked at the beginning of the article might be answered in the following way:

1. Prejudice appears during early childhood, perhaps in the pre-school years, and increases with advancing age (Lasker, Minard).
2. Prejudice seems to be instilled by the unconscious example or teaching of the social environment formed by parents, adults, and other children (Lasker and E. L. and R. E. Horowitz, 1938).
3. Prejudice is closely tied to the basic personality of the individual (Frenkel-Brunswik).
4. Prejudice is on the whole stronger than the counter-propaganda of democratic teachers and the influence of democratic ideology, and becomes more organized and more rigid as the child grows older (Minard, Zeligs, Horowitz). Further, prejudice does not seem to be closely dependent on personal contact with members of minority groups, for whether we study children in the North or South, or children in segregated or unsegregated classes, the pattern of prejudice is more or less the same (Horowitz).

What is implied for the educator by these theories of the source of childhood prejudices? The implications of Frenkel-Brunswik’s personality-and-prejudice correlation might be that the educator is powerless to alter the consequences of unsound parent-child relationships. Another study, which adheres more closely to the social-influences line of Lasker, Horowitz, and Zeligs, however, has a more hopeful view, and offers positive suggestions for possible educational techniques to deal with children’s prejudices. This is the study of Marian Radke, Helen Trager, and Hadassah Davis, conducted under the auspices of the Bureau for Intercultural Education (“Social Perceptions and Attitudes of Children,” Genetic Psychology Monograph. Series, 1949). The conclusions in regard to personality and prejudice made by these authors are directly opposed to those of Frenkel-Brunswik: “ . . . conformity to environmental standards and expectations rather than individual securities or insecurities would appear to be the root of the child’s earliest content and valences for social groups. . . .”

Radke, Trager, and Davis presented two hundred and fifty Philadelphia school children, aged five to eight, with a series of eight pictures about which they were asked to make up stories and express attitudes. In one picture, for example, a colored child is seen standing aside from a group of white children playing together. In another, two boys are seen emerging from a synagogue, while four boys stand down the street watching them. The premise of the test is that the subjects project their own feelings and attitudes in discussing the pictures.

Many subjects at the beginning of the interview were uneasy and seemed to wish to avoid the questions of race and religion, but became actively interested when confidence in the interviewer was established. Reserve and conflict were particularly apparent among the Negro children tested; at one moment, they rejected the white group and at the next rejected their own. Jewish children valued their own group much higher than did Negro children, and expressed their valuations more emphatically than did Christian whites; they also referred to their group more frequently than did children of other groups, and almost half of the Jewish children projected “Jewish” into pictures where there was no such identification by the tester.

The study found an increase, with age, in the percentage of children expressing prejudice and showing an awareness of group tensions. The presence or absence of minority groups in the subjects’ neighborhood or school had little effect on their attitudes. The authors concluded that it was not contact with minorities, but contact with prevailing social attitudes toward minorities, that was the foundation for prejudice.

Radke, Davis, and Trager found in the children’s early inhibitions and later eagerness in regard to race discussions a clue to the kind of pedagogy required. Their fears, fantasies, curiosities, and misconceptions must not be suppressed but openly discussed. “A rule of silence about differences,” they wrote, “not only fails the child in not helping him to achieve a better understanding . . . but the silence may also be perceived by the child as tacit agreement with societal prejudices.” The authors also argued that the specific problems of prejudice which the child meets in his school, his neighborhood, and his home must be discussed frankly. The successes of the play therapy course by Axline reported above would seem to support such a program.

The author’s own inquiries into New York City educators’ attitudes toward intercultural education indicate that these suggestions are not widely followed. Whether because of fear of antagonizing some members of the adult community or because of the theory that such discussions only aggravate children’s prejudices, many educators carefully refrain from mentioning to the children that there are significant differences among various groups, that there is some degree of prejudice in almost every community and institution in the nation, and that the children themselves are frequently the objects or the subjects (or both) of prejudice. In many schools there seems to prevail a kind of liberal over-optimism that may be as ineffectual in intercultural education as a reactionary fatalism. The ignoring of differences is expected to lead to the early disappearance of all the problems they create. In effect, the children are told there is no real problem, at least not where they are concerned—whereas, as a matter of fact, there is.

_____________

The Philadelphia experimenters concluded that prejudice is largely a matter of social imposition upon the individual; the California group, as we saw, held that it is a response of the individual personality to inadequate social and emotional satisfactions. Both studies were carefully conducted and in neither does it appear that there were unwarranted generalizations of the empirical conclusions.

However, there is one way of reconciling the apparent contradictions of the California and Philadelphia studies while accepting their results. There is within our society a very wide range offered to the individual between tolerance and prejudice. He may choose any one of a number of positions, or several positions simultaneously or successively, without departing from what is regarded as the cultural norm. The “normally” prejudiced person accepts and perpetuates on the one hand a formal ideology of tolerance (“live and let live, equal opportunities for all, the American creed”) and, on the other, a fluctuating set of invidious stereotypes. Which social attitudes he will accept as his own depends on personal as well as social factors. And within the “normal” range of prejudice and tolerance, it is probably not easy to distinguish the personality elements from the complicated interplay of social influences.

Thus for any random population—and such a random population was studied in the Philadelphia research—it is not easy to find a correlation between personality factors and prejudice. Most people will indeed pick it up “out of the air,” from the “social atmosphere.” And for these people, their prejudices will be virulent or mild, as the situation demands. But those persons who have really developed the kind of personality that requires prejudice to balance it—that is, the authoritarian personality—will develop more consistent and perhaps more violent attitudes, and will show up in studies as extremely prejudiced. Consequently, when we approach the limits of the scale of prejudice, we will find personality and prejudice linked together. From this point of view, the “liberal” child and the “ethnocentric” child, as they appear in Dr. Frenkel-Brunswik’s study, are deviants from a norm of mild prejudice, since the completely secure and loving family held responsible for the former, the completely unprejudiced child, is probably as rare in our society as the completely hostile and rejecting family that is believed to produce the latter.

_____________

Whether “normally prejudiced” children can be taught finally to resolve the conflicts between a formal creed of tolerance and stereotyped prejudicial attitudes in favor of the former, is a question that educators like to answer in the affirmative, but that psychologists have not answered at all—unless we take their more general comments on attitudes.

Thus, Gordon Allport writes (“Attitudes,” in A Handbook of Social Psychology, ed. Carl Murchison, Clark University Press, 1935): “An attitude seldom contains all of the experience which is relevant to it, and seldom changes as rapidly as a faithful following of experience would require. . . . Because they save both time and effort, stereotyped attitudes offer great resistance to change. They resist the inroads of new contradictory experience and are retained as long as they satisfy and protect the individual.”

In such a situation, what is the value of intercultural education, and what type of formal education is most valuable? To my knowledge, no follow-up tests have been made on children subjected to the various programs of intercultural education. The employment of such projective techniques as those used by Radke, Davis, and Trager, on a before-and-after basis, would seem to be an excellent means of testing the effectiveness of different pedagogical procedures—including the one suggested by these authors.

The subject of children’s prejudices is also deficient in another type of study—one that would consider it in a non-laboratory atmosphere, in the ordinary relations of social life, in terms of social behavior (physical and verbal) rather than verbalization, i. e., the individual’s report on his behavior. The laboratory emphasis may be responsible for the fact that more general social and economic causes of prejudices have only rarely been considered.

While a few studies have failed to discover any simple correlation between economic status and prejudice, no studies—that is, in the sphere of children’s prejudices—have been made on the relation of change in economic status, and particularly loss of economic security, to prejudice. Indeed, the whole larger social background of prejudice—in the school, on the street, at home—is often filled in rather shallowly, if at all, in these studies. Yet many students believe that it is only in a time of economic crisis and general insecurity that prejudice becomes dangerous. A long-term study of the relation between economic factors and prejudice might give us some insight into the dynamics of the frightening shifts we have witnessed in our time from a relatively tolerant to a violently discriminatory social atmosphere. The fragmentary research we have summarized in this article will go on, we may hope, to fill in this social and economic setting.

_____________

¹ This study—and indeed, almost all those we will report upon—may raise in the reader's mind the serious question as to how one can distinguish discriminating behavior (rationally motivated) in reference to different ethnic groups from prejudice (motivated by emotional bias). For example, if a young person is realistically aware that friendships with Negroes will lead to “trouble” with his family or hamper his success, and therefore does not become friendly with Negroes—is this prejudice? All of these studies take it for granted that any expressions indicating that one would treat Negroes differently, or that one accepts a common view (“stereotype”) of Negroes, are expressions of prejudice. It is possible these may simply be means of accommodation to an environment which makes truly non-discriminatory behavior towards Negroes almost impossible. The writer believes that the measures of “prejudice” usually resorted to in psychological studies really do measure prejudice and not this realistic type of “discriminating” behavior; yet it is perhaps worthwhile to raise this question in the reader's mind.