The Problem of Inconsistency in Intergroup Relations

Muzafer Sherif

THE PROBLEM OF CONSISTENCY and inconsistency in intergroup relations is not a separate issue. Emphasis of this point is necessary if we are to achieve an adequate handling of the problem in research and in practice. As in many other areas, research efforts in this area are doomed to failure so long as the problem is taken as an isolated topic of investigation. Nor will such an approach be of substantial help to men seriously concerned with doing something about this vital matter in human relationships. In fact, practitioners and men of good will in education are rightly becoming suspicious of the value of the prescriptions "handed down" to them by the research people. Too often, such prescriptions fall pitifully short of the scope and diversity of the concrete experiences faced in the field of realities —realities which are not infrequently grim.

In this paper, I shall not try my hand at writing a rounded social psychological account of consistency-inconsistency in intergroup relations. An adequate social psychology of this serious issue can come only after we learn more about the underlying phenomena which have been more familiar topics of study for the social psychologist. Certainly the problems of attitude, attitude change, group membership, and ego functioning are among them. The aim of this paper is to relate briefly the topic at hand to these more familiar basic problems and to call attention to a few implications derived on this basis.

The issue of consistency-inconsistency in intergroup relations is another current issue which can be handled only as part and parcel of persistent major problems to which it is organically related.[1] To start with the obvious fact, such inconsistencies in intergroup relations cannot be considered apart from the reality of the institution of social distance (prejudice)-the standardized scale of social distances at which one group is placed in relation to other groups. Many-by no means all-campers in an interracial camp, caught temporarily in the atmosphere of the camp situation, may reduce the social distances at which the groups of fellow campers are placed, thus revealing reduction in prejudice. But when these same campers return to their own customary group setting, they may and do (as evidence indicates) return to their roles as good members of their own group. The out-groups which were treated perhaps with fairness or even cordiality in the camp situation are put back on the social distance scale in the positions standardized in the in-group.

( 33)

It is evident, then, that in order to have any understanding of the inconsistency of behavior in such situations, we must first have a clear picture of the institution of social distance. The necessity of keeping in mind the prevailing social distance scale becomes evident when one examines the extensive, but all too frequently one-sided studies and practical efforts aimed at reducing or eliminating prejudices. On the whole, such methods as contact (even contact in group situations), information, instruction, resort to mass media of communication and other propaganda devices have been disappointingly unrewarding. In some cases, such well-meant efforts have produced bitter fruit-different kinds of "boomerang" effects. In our opinion, these disappointing results are due to the one-sidedness of the approaches used. The factors which investigators overlooked proved to be more potent than those with which they happened to be preoccupied.

In the case of consistency-inconsistency in intergroup relations as well, it may be rewarding to postpone for a time concentration on the topic itself and to look first at the implications of the social psychological phenomena that underlie prejudice.

An undue emphasis on individual childhood frustrations and other frustrations, on an authoritarian atmosphere of the family and of other agencies of child training, on aggressive tendencies thus engendered in the development of individuals leads to some variety of "displacement" theory of prejudice. All these are genuine problems to be taken up in later stages of the analysis. However, exclusive preoccupation with such part-problems is doing gross injustice to the broad scope of the topic. At this date, it should have been accepted as a well established fact that prejudice is a standardized social institution for the country like any other social institution. Its existence is quite independent of the particular life histories of the individuals in terms of their degree of contact with groups in question, information about them, etc. This fact has come out time and again in studies such as those of Bogardus, F. H. Allport, Meltzer, Zeligs and Hendrickson, Hartley, Murphy and Likert, and others. As good members of their group, the relatively better adjusted majority of the in-group shares the institution of social distance. Response to the individuals belonging to various out-groups is regulated by the position of those groups on the social distance scale. The concentration on individual frustrations, aggressions and other individual factors can shed light on a limited area--the area of the degree of intensity with which the particular individual reveals his prejudice.

An adequate account of prejudice as a social institution goes well beyond the scope of competence of the psychologist. With all the social, economic, political, religious factors coming into the picture, this task lies largely in the domain of the social scientist. Therefore, the psychologist has to learn from the social scientists in this area before he can do justice to the topic.

It is the psychologist's job to study how the individual acquires the

(34) attitude of prejudice in the process of his becoming or being a member—and a good member—of his family, school, church, club, union, business organization, and social class. This is a specific case of the acquisition and functioning of an individual's attitudes in relation to any social institution or norm. As was convincingly established by Horowitz over a decade ago, prejudice is derived from membership in a community and not primarily by contact with the individuals of the group against whom prejudice is directed.[2]

Since the attitude of prejudice is an attitude which in our eyes defines our standing in relation to the members of other groups, since it is a factor in appraising the worth of other people in relation to us, it is an ego-attitude. This fact is borne out by the findings of the Hartleys and the Clarks that the ego must reach some certain degree of development before a consistent pattern of prejudice is established in the individual. In other words, the development of prejudice and ego development go hand in hand. The individual's attitudes of prejudice are consistent parts of his ego constellation. The theoretical elaboration of the above statements and factual evidence on which they are based is presented elsewhere.[3]

Reactions of the individual in intergroup relations are ego-involved reactions. The behavior of the individual in relation to the members of any out-group is not determined in a major way by the specific properties of the stimuli at hand (including the persons in question), but is altered, distorted, accentuated, minimized, etc., largely as determined by the interiorized values on the social distance scale. The perceptual selectivity and distortions so frequently observed in intergroup relations are the outcome of ego-involvements.

But the ego-attitudes concerning various groups are not the only constituent of the ego. Other ego-attitudes function as part of the ego at a given time-ego—attitudes related to membership in the family, in church, school, club, clique, gang, sex, business, union, etc. There are also more specific ego-attitudes defining the individual's specific role or relative position in such groups. These various ego-attitudes, which may be activated singly or in some combination by the properties of the specific situation at hand, may or may not be compatible. If they are incompatible, the individual will be reacting one way in one situation and the opposite way (contradictory way, if you will) in another situation, as determined by the requirements of his different roles in the two different situations. For example, take the painful dilemma in which the modern Professional woman is frequently caught. As a woman (in the feminine role), she is supposed to and she wants to react in a certain way. As a lady, she wants to observe certain niceties. As an engineer, doctor, or research worker (in the professional role), she has to and wants to live up to the expectations of her professional

(35) colleagues. In the privacy of her home with her husband or child, she may behave in a consistent way in her feminine role without contradicting herself. In sorority reunions or at purely social meetings, she may consistently behave like a lady. In her office or in a professional meeting, she may be her professional self more or less consistently. But it is a common occurrence in highly complicated modern societies to have this person caught in a situation which puts contradictory demands on her as a woman, as a lady, and as a professional person. Caught in such a situation, she will probably contradict either her womanly self, or ladylike self, or professional self, depending on the relative strengths of her various selves and the significance of the situation to her. This is especially true if she is a person of more than ordinary intensity.

Let us apply this kind of analysis to the contradictory behavior of the Detroit union members cited by Dr. Jahoda. These union members, who were taught and practiced non-segregation in their union activities, "actively participated in the race riots in 1943." If these union members had been nothing but the good and staunch union members that they were, they would not have participated in the race riot or in any act of discrimination. But they were also members of a neighborhood group, a church, ethnic group, and--as they are reminded in so many ways, directly or indirectlyof-- a "race" which they have learned stands at definite distances in relation to other "races" and groups. When this self (as member of an ethnic or racial group) was aroused, they acted this time as staunch members of their "racial" group. They followed the dictates of being "regular guys" in this situation no matter how contradictory such dictates were to other roles. In fact, they probably acted with a feeling of righteous indignation against those whom they considered to be the offenders.

Another illustration of such contradictory ego-attitudes is the case of young factory girls who, after "discussing the goals of life came to a unanimous decision that ‘being happy' was the most desirable goal." During this discussion, the girls were shown two pictures-one of a smiling working class girl, the other a serious looking wealthy girl. From the smiling appearance, the girls characterized the working class girl as happy, while the serious look indicated to them that the wealthy girl was not happy. Yet, their aspired identification was with the leisure class girl. The contradictory reaction of these girls was due to the fact that they were, of course unwittingly, using different and contradictory premises. When discussing the goals of life in a general way, they naturally chose happiness as the goal of life. Almost anyone, living under any social system, would choose happiness as the goal of life. But when faced with the concrete alternative of identifying themselves either with the serious looking wealthy girl or the smiling working girl, their persistent and intense yearnings were activated. In the culture in which they live, through movies, schools, novels, radio, society columns in the papers, face-to-face and indirect media of influence, their aspirations in the matter of identification are well ingrained. Thus

( 36) their aspired self has strongly become the wealthy girl. She is the embodiment of the finishing school girl, the debutante, the movie star, so omnipresent in every conceivable means of mass communication as the significant, successful, desirable person.

These illustrations are sufficient to clarify the main point. What is designated as the ego (self) of the individual really consists of a host of attitudes learned in relation to persons, things, institutions, groups, norms with which he has been in contact. Especially in highly complicated modern societies of the West today, it is not a unitary ego which is involved in various situations. In one situation, one ego-attitude is involved; in a different situation, another ego-attitude or set of attitudes is activated as determined largely by the demands of the particular situation. As we have seen, various ego-attitudes of the same individual need not be compatible. When ego-attitudes are incompatible, the resulting reactions are usually contradictory to each other. (Therefore, the indiscriminate use of such terms as "ego-structure," as though the ego of the average modern man were a unitary, integrated whole, simply perpetuates a conventional psychological fallacy.) Herein lies, in our opinion, the psychological basis of most inconsistencies of individuals' reactions, especially in intergroup relations. As a good member of a union which has taken a definite anti-discrimination policy, the individual does not dare to and will not indulge in acts of discrimination, much less race rioting. But, like many of his fellow union members, he is other things at the same time. He is also a member of an ethnic group. In fact, he felt himself a member of an ethnic group (with all its auxiliary memberships) before he came to know himself as a union member. It is no wonder that in a conflicting situation, he reacts in a way contradictory to his union loyalties.

Anyone who keeps a clear picture of ego development in mind will find some well meaning attempts to better group relations rather naive. A boy goes to an interracial camp. It is unrealistic to assume that the camp situation is more than a small transitory segment of his life. He is primarily a member of a family, an ethnic group, a play group-these are the groups which contribute heavily in the composition of his ego. A boy who behaved nicely in such a camp situation can hardly be expected to behave in the same way in his school, street or family situation when he knows that he will be laughed at, punished or ostracized for such behavior.

The modern version of attitude change studies by "restructuring perception" is, of course, hued on one important conclusion-that group situations can be created in which learned ways of perceiving and categorizing can be altered. But it is scarcely realistic to assume that perceptions "restructured" in a new group situation or attitudes altered in a liberal situation will become the lasting perceptual reactions or attitudes of the individual. The individual does not stay in such especially created situations for long. The objective situation in which he is customarily situated is the angle from which he sees and reacts to the world. To claim greater

(37) validity for situations in which perceptions are restructured than for the more or less permanent objective situations is a dangerous subjective tendency in psychology. Rather the results of such studies provide clear proof that perceptual lines or categories drawn in group situations are not immutable and can be changed. This is one thing. The problem of the durability of such changes and of making them last is the great problem challenging us all.

The inconsistent behavior in intergroup relations is not due only to external "cross pressures." However, such "cross pressures" certainly do play a part in some cases. For example, a white boy may want to play with Negro boys because he heard or read something that this was the right thing to do. In some cases, he will not even try to put the idea into practice out of fear of correctives from his playmates, family, and other grownups.

But in many cases, the external "cross pressures" need not be the cause of inconsistent behavior in intergroup relations. Our idea of what we are, what other groups are, what is desirable for a person to be, what is a desirable position to occupy, who are desirable persons to associate with, who are the persons who should be put at a distance, are derived in their major outlines from our reference groups. However, in time they seem to be our very own. The individual is simply not aware of their derivation. As Hartley indicated, white children, especially older ones, are often unable to say how they acquired their prejudices. The impact of external pressures is not even needed to regulate behavior under these circumstances.

In summary, the issue of consistency and inconsistency in intergroup relations is not a separate issue. Any adequate approach to the problem, both in research and in practice, requires that it be taken as part and parcel of persistent major problems to which it is related. We have to keep in mind the picture of the institution of social distance and the implications of the psychology of ego development and functioning. An adequate account cannot be achieved by psychologists alone; it requires the close collaboration of social scientists in various areas. We have to make revisions. on our hasty generalizations concerning attitude change derived mainly from studies done in artificial, transitory situations. With these considerations in mind at every step of the way, research on the issue of consistency and inconsistency in intergroup relations can be conducted in proper perspective.


  1. Sherif, M., "The necessity of considering current issues as part and parcel of persistent major problems," Int. J. Opin. & Attit. Res., 1948, 2, 63-68.
  2. Horowitz, E. L., "The development of attitudes toward the Negro," Archives of Psychol., 1936, No. 194.
  3. Sherif, M. and H. Cantril, The Psychology of Ego-involvements, N.Y.: Wiley, 1947; and M. Sherif, An Outline of Social Psychology, N.Y.: Harper, 1948.

Valid HTML 4.01 Strict Valid CSS2