Social Responsibility and the Group

Muzafer Sherif

IN MODERN DIFFERENTIATED SOCIETIES with complex division of labor, problems of social responsibility assert themselves as mutual interdependence of men and groups becomes increasingly compelling. They are enormous problems, encompassing on the one hand the area of social control and, on the other, the development of "social conscience." The many facets of social responsibility include psychiatric, psychological, sociological, cultural, political, economic, and industrial issues. Collaboration and coordination of various disciplines are required to meet its challenge.

Therefore, it would be pretentious for a representative of any one of these disciplines to discuss the entire topic. I am a social-

( 220) psychologist interested in research on human relations within and between groups. My attempts are directed toward integrating findings and conclusions on these topics. But surely it would be impertinent for me, from the angle of social psychology alone, to write prescriptions or offer a panacea for the social responsibilities of those who actually plan and work with groups.

First I shall raise the problem of social responsibility and delineate that facet which lies within my area of research as a social psychologist. Then I shall depart from the realm of the abstract to analyze this facet of the problem in terms of experiments on relations within and between groups which were undertaken specifically to study responsibility, solidarity, and morale as social-psychological problems.

The term "responsibility" is used in several senses. Therefore, we have to specify the exact sense in which we shall use it in this presentation. In one sense, responsibility implies being accountable to others under the coercion of external pressures, sanctions, and correctives. In this sense, the problem of responsibility in actual social life leads immediately to analysis in administrative, legal, and power terms-terms which are beyond my ken as a social psychologist.

In another sense, social responsibility implies inner promptings and inner urgency. In this sense, the individual feels accountable to uphold certain values because he feels certain things should be done, certain values should be upheld, certain aims must be achieved, and certain desirable standards in human relations must be observed. This is the kind of social responsibility with which this paper is concerned.

Social responsibility in this sense is not an abstract quality or state of mind that descends on individuals every once in a while. Social responsibility, or lack of it, is an aspect of human relations in all its forms, whether within groups or between groups. Therefore, realistic analysis of responsibility traces this aspect of human relations as it develops in reciprocal interactions of man with man in their respective groups and between their respective groups.

One familiar approach to the analysis of inner responsibility or

( 221) "conscience" finds its origins in the restraints and frustrations of the individual's instinctive impulses early in life by the coercive authority of adults. This approach is characteristic especially of orthodox psychoanalysis. At another extreme, there are sociologists who consider the development of responsibility almost as a process of pouring social standards or values into a passive individual recipient.

Those of you in social group work are in a fortunate position to see the rich actualities of life which the research literature merely abstracts. From either vantage point, it is apparent that neither a strictly individualistic account or a strictly sociocultural account can do justice to the actualities of inner responsibility, social conscience, feelings of solidarity and morale, which are complex psychological formations unique to man. This fact is recognized by leaders in group work, like Eduard Lindeman, Wilbur I. Newstetter, and Grace Coyle, as well as by recent trends in social psychology.

The modern trend in social-psychological theory recognizes that the "personal" and the "social" are everywhere intertwined in human relations and in personal development. Psychological formations like inner responsibility, conscience, and feelings of solidarity are joint products of influences stemming from within the individual and stemming from his sociocultural setting. Influences from the individual himself and influences external to him are always functionally related.

Therefore, the worn-out conceptions which place the individual and the group at opposite poles, considering the group as necessarily the enemy of individual development, can only impede clarification of social responsibility and related phenomena, such as group identification, group leadership, and solidarity.

Social responsibility, in the sense of inner accountability to one's fellow men in personal and group relations, emerges when relevant values become part and parcel of what he considers as his "self." Feelings of responsibility from within are constituents of the individual's self-image. The formation of a self-image by the individual, with highly cherished personal values toward his

( 222) neighbors and toward groups he deals with, is a product of his free participation in group activities with other human beings toward goals which he shares with them.

This conception of individual development and group functioning as intimately related processes lies at the basis of the experiments I shall report to you in very brief form. Such an integrative approach, which conceives of individual and group as functionally interdependent, has received repeated confirmation in sociological studies of small groups and in various psychological studies. Here, I can mention just one of the psychological studies, namely, the impressive series carried out by the Swiss psychologist jean Piaget, on the development of moral judgment and responsibility.[1]

Piaget studied the developmental transition in the moral judgment of the individual. In one study, one group of children six to eight years of age and one group nine to twelve years of age were asked what kinds of behavior they considered unfair or wrong. The younger children considered acts unfair or wrong which were forbidden by their parents and other grownups. At the ages of six to eight years, they seldom noted acts of inequality as unfair. Nor were acts of social injustice considered unfair by these children. In order to have a notion of inequality of treatment among individuals, the child must recognize and participate in reciprocal relations with others, especially age-mates; he has to be aware of mutual responsibilities and feel obligations prompted from within.

On the other hand, the older children of nine to twelve very seldom reported forbidden acts as unfair, but considered 73 percent of the cases of inequality as unfair. A few cases of social injustice arising from lack of equity were also called unfair by these older children. Piaget concluded that through participation in interaction with others on a reciprocal basis, the child's moral values develop from those imposed by others, particularly grownups, to a growing realization of mutual obligations and inner responsibility to others.

In short, such abstract conceptions as social responsibility, solidarity, and morale arise from the reciprocal relations of man to man and group to group. Apart from such functional relations

( 223) in small and large groups, social responsibility, solidarity, and morale have little real meaning. Therefore, as one concrete illustration I turn now to our research on actual reciprocal relations in which social responsibility and solidarity arise and operate.

Since 1948 a program of research has been under way under my direction to test some hypotheses derived from sociological, psychological, and everyday-life accounts of group life. This research program included three large-scale experiments carried out in 1949, 1953, and 1954.[2] The conclusions in this paper are based on the 1949 and 1954 experiments.

An over-all aim of these experiments was to create conditions in which groups would form naturally and to observe how an individual became an active and responsible part of the group. We were interested in testing hypotheses concerning essential conditions for the development of inner responsibility to group values on the part of individual members. This aim required that we start with individuals who did not know each other. As the first stage in the experiments, groups were formed and their interaction observed over a period of time.

Our interest in group values was not confined to desired means and ends within the confines of one group. We were equally concerned with values and attitudes pertaining to behavior toward other groups. The next problem, therefore, concerned the limiting conditions for the formation of friendly or hostile evaluations of other groups and friendly or hostile treatment of their members.

In the next stage of the experiments, accordingly, the groups which were formed at the initial step were brought into contact. Since friction between groups is the most challenging problem in intergroup relations today, the first contact between groups took place in competitive and mutually frustrating conditions.

( 224)

Finally, in the 1954 experiment, an attempt was made to reduce the friction between groups engendered by conditions of rivalry and mutual frustration. In other words, the aim was to establish conditions for interaction between groups conducive to willing and active cooperation with another group, which had formerly been seen as an "enemy."

From my own limited contact with the literature and practice of social group work, I believe you may find that procedures used in these experiments have a good deal in common with methods advanced by some of the leaders in the group work field. The chief difference is that in the experiments I shall briefly describe, methods and techniques were employed in situations planned to test certain hypotheses in a systematic fashion.

In the first place, every effort was made that the situation appear natural and lifelike to the individuals participating in the experiment. The settings of these experiments were summer camps. The subjects were eleven-and-twelve-year-old pre-adolescent boys, who find camping activities fascinating.

The primary technique for securing data was that of the participant observer. The subjects were not aware that their behavior was constantly being observed. The experimental staff appeared to them to be regular personnel of the camp. No staff member appeared as a psychologist or investigator so far as the subjects were concerned. There is both empirical and experimental evidence that individuals are mindful of being observed, especially by persons with authority and especially if their actions relate to personal concerns and personal relations with others.[3]

There is always the possibility, of course, that observers may be highly selective in reporting in spite of all the precautions strictly observed. Therefore, observations were cross-checked with several other techniques. Candid recordings and candid pictures were made whenever possible without cluttering the natural flow of interaction. Sociometric choices were obtained. At crucial points we took advantage of developments in the laboratory which per-

(225) -mit one to obtain evidence of a person's attitudes indirectly, without his awareness of the purpose of the task .[4] Conclusions which I shall draw here are based on results cross-checked by two or more of these techniques.

In producing groups, in producing conflict between them, and in reducing that conflict, no special lecture methods or discussion methods or leadership training were introduced. On the contrary, the adult staff was instructed not to intervene in interaction or in decisions made by the boys until they wanted help in carrying out their decisions. Thus, the methods in the experiment resembled closely the enabling method in social group work described by Wilson and Ryland [5] and by Friedlander.[6] I take it that the main objective of the enabling method is to assist group members as they take responsibility in planning and making decisions, and not to take over these leadership and initiative functions from the group.

The planning of the experiment in successive stages was possible by controlling and altering conditions external to the groups. This procedure is not, I believe, foreign to the theory of group work. In 1935 Newstetter wrote of "guidance and stimulation primarily through influence on the social and physical setting of the group rather than through the direct personal influence or authority of the worker." [7] In each stage of the experiment, problem situations were introduced appropriate to the chief characteristics of the experimental conditions of that stage. These problem situations were introduced quite naturally. Before and during the experiment, we studied the interests and concerns of the subjects. The problem situations had high appeal value, and they were real to these boys. Therefore, they could not be ignored easily by any individual. For example, the problem of getting a

( 226) meal on their own initiative was introduced when participating individuals were hungry. Or, a boat was placed near their cabin, which was some distance from the water. Moving the boat they all wanted to take to the water required coordination of everyone's activities and efforts.

When problem situations are immediate, compelling, and embody a highly desired goal, group members do initiate discussion, do plan and carry through their plans. In this process, discussion has a place, planning has a place, action has a place, and even lecture or information has a place. The sequence of these related activities need not be the same in all cases.

The experiments proper started with the selection of subjects by rigorous criteria. These criteria were necessary to rule out the possibility that results could be explained in terms of background differences among the boys, social maladjustment, undue childhood frustrations, or previous interpersonal relations with one another. The subjects were not acquainted prior to the experiments. They were healthy, normal boys around the ages of eleven and twelve, socially well adjusted in school and neighborhood, and academically successful. They came from settled, well-adjusted families of middle- or lower-middle-class and Protestant affiliations. No subject came from a broken home. The mean I.Q. was above average.

The subjects were divided between two separate cabins at some distance from one another, in such a way that the boys in each cabin were similar to those in the other cabin in size, athletic skills, and the like. During the first stage of the experiment, these two groups were kept separate.

Our hypothesis was that when these individuals interacted with one another repeatedly in problem situations which embodied highly appealing goals and which required interdependent activities, they would organize themselves, on an informal basis, as a group without exhortations from outside to pull together. A definite group structure or organization would take shape, which would be revealed through observers' ratings of status and through sociometric choices. Certain ways of doing things, certain preferences for activities, certain desirable ways of behaving and de-

 (227) -sirable ends to pursue, would become standardized in each group. In other words, social norms or values would be shared and standardized in each group. Finally, as a group organization took shape and values were standardized, individuals in each group would function in a responsible way as group members with particular roles and would hold themselves accountable to maintain and advance the values of the group.

Formal programming during this period was kept to a minimum. As Newstetter [8] and Coyle [9] have noted, a formal program handed down is not the most effective means for development of group initiative or morale. As much as possible, the boys were allowed to engage in activities they preferred at the times they chose and as often as they wished (limited, of course, by health considerations). During these activities, compelling problem situations arose which required coordinated efforts. (I have mentioned the problem of transporting a boat to the water.) The boys were fascinated with camping out. Tents had to be assembled and pitched. There were meals to be prepared in the woods, from ingredients furnished in the bulk, such as hamburger, chunks of meat, soft drink powders, and uncut watermelon. Division of labor was required. Evening campfires presented the problem of entertainment, which the boys supplied each other.

As a result of repeated interaction in such situations, groups did begin to form. The boys pooled their efforts, divided up tasks, became acquainted with each other's particular personal characteristics. After about a week, the independent ratings of observers on two successive days of the status or social positions in each group were highly correlated (a rank order of .92 for one group and .98 for the other). Leaders emerged in each group. Some boys sifted to the bottom. Others became lieutenants or attained other positions in the group through proficiency in some activity.

Each group appropriated certain places and things as their property. One group named their swimming place and campfire, and put up signs to identify them. The other discovered an

( 228) athletic field and worked hard to improve it. They called it "ours." Each group adopted a name and stenciled it on their shirts. In the 1954 experiment, one group called themselves the "Rattlers"; the other, the "Eagles." Each group had their own favorite songs, favorite games, and favored ways of carrying out their activities. One group spontaneously organized a system of swimmer escorts and a protective circle in the water to encourage all members to learn to swim and dive well. In spite of their similarity and similar location, these desired modes of behavior differed in the two groups. In one, toughness was highly prized. In the other, swear words were frowned upon. These qualities became personal characteristics of each member which he exhibited whenever occasion arose.

In short, group norms or values were shared and standardized in each group. Because these values emerged in the course of activities which the individuals themselves initiated, planned, and carried out, the group values became their values. The group was seen as their group. Undesirable modes of behavior in terms of the values or norms of their group were recognized with disapproval, frosty silence, or scorn. Thus all came to know what were the desired modes of behavior and the desirable ends to pursue. They came to consider these as their own personal desires and own personal preferences. Thus they upheld them with a sense of inner responsibility.

These findings concerning inner responsibility to the values of a group, I believe, are not surprising to you who work with groups. They verify repeated findings in sociological literature. They support the conclusion reached by Piaget in his extensive studies of the development of moral values. As you will recall, Piaget concluded that mere compliance with rules and values imposed by outside authority gave way to free and autonomous pursuit of values only when a child participated as an equal with other children on the basis of genuine reciprocity.

Once groups were formed in our experiments, we brought them into contact with each other under conditions which were conducive to friction between them. The essential characteristic of experimental conditions in this stage was that goals could be at

( 229) -tained by one group only at the expense of the other. This condition was arranged quite naturally, since each group was eager to engage in competitive sports. A tournament of games was arranged, as though we were acceding to their repeated requests. Handsome prizes were to be given to the winning group. The Rattlers, who had appropriated and prepared the athletic field, made it available for the contests.

Members of both groups pitched in with all their energies and determination to win. One group held a prayer for victory before each contest. They devised special strategies to outwit and outmaneuver the other group.

Our hypothesis was that in conditions which were competitive and mutually frustrating, members of each group would develop hostile attitudes toward, and unfavorable stereotypes of, the other group and its members. In other words, responsibility to their own group would, under these conditions, imply hostility to the other group.

At first, good sportmanship was evident on both sides. But during the course of victories, defeats, and renewed efforts to win, intergroup friction developed. A series of mutually frustrating events arose during the tournament. Following a defeat, the Eagles stayed behind on the athletic field and burned the banner which the Rattlers had left there. When the Rattlers discovered their burned flag the next morning, they were outraged and determined to punish the guilty parties. It is important, I think, to note that their reaction was not simply one of blind rage. They planned a strategy, on their own initiative, for determining whether the Eagles were guilty and for starting a fight if they were. This strategy was carried out as planned, with both groups scuffling over each other's flags.

From this time on, name-calling and hostile acts marked the contact between the two groups. The Rattlers raided the Eagles' cabin, taking the Eagle leader's blue jeans and painting on them the legend "The Last of the Eagles." These were carried like a banner.

Each group appeared to the other as an enemy. Members of the out-group were labeled "sneaks," "cheaters," or called by even

( 230) more undesirable names. Attitudes of social distance between the groups became so marked that neither wanted anything further to do with the other. Of course, different individuals expressed their hostility toward the out-group in different ways. But in one way or another, every individual manifested the antagonism shared by his group. Furthermore, this antagonism, the stereotyped notions of individuals in the other group, and the great social distance between them persisted after the contests were over and the experimental stage was concluded.

If there are any doubts of the zeal displayed by these boys in supporting their group in its positive aims toward victory and its efforts to defeat and humiliate the other group, I believe the pictures taken on the spot will dispel them. In the 1949 study, photographs of each group were taken immediately after the tournaments ended. No one can mistake the joyful faces of the victors or the downcast expressions of the losers. Victory of one's group aroused personal joy. Defeat of one's group meant personal frustration.

Perhaps the most significant conclusion to be drawn from this stage of the experiment is that inner responsibility and accountability to one's own group do not guarantee responsible treatment of members of other groups, if the values of one's own group require hostility and social distance to the out-groups. Under conditions such as these, being a responsible group member means active and willing participation in hostile and aggressive actions toward the out-group.

As the final stage in the 1954 experiment, conditions were introduced which were designed to reduce friction between the groups and to replace the negative evaluation of the out-group with attitudes of friendship and cooperation. As a first step, the groups were brought into contact in activities which were satisfying for members of each group, but did not require interdependent activities involving both groups. This phase was introduced to see if removing the conditions which had produced hostility between the groups would reduce the hostility. The groups ate together in the same dining hall; they attended a movie in the same building; they carried out various entertainments side

( 231) by side. However, their antagonism, the name-calling, and derogation of the out-group continued. They used these contact situations as opportunities to intensify intergroup hostility. For example, at meals they shouted unfavorable names at each other, even throwing food and other objects.

Following this series of intergroup contacts, the main effort to reduce intergroup antagonism was made. Our hypothesis was that if the conflicting groups were brought into contact in conditions embodying goals which were compelling and desirable to both groups, but which could not be achieved by the efforts and resources of one group separately, they would tend to cooperate toward the common goal. Such goals may be termed "superordinate goals," in the sense that they are desired by members of both groups but require that both groups pull together to attain them.

It was too much to expect that the friction, stereotypes, and social distance between groups would disappear in one episode. Therefore, the next hypothesis predicted that a series of interaction situations embodying superordinate goals would have a cumulative effect in reducing conflict.

A series of problem situations embodying superordinate goals was introduced. There was no preaching, no lecturing, no adult-led discussion. The problem situations were varied, but all had an essential feature in common. All of them involved goals that became focal for both groups. The goals were urgent to these individuals; they could not be ignored. Yet their attainment clearly depended on communication, planning, and joint action by members of both groups.

The problem situations embodying superordinate goals included joint efforts to cope with a water shortage that affected all individuals and could not help being compelling. They included securing a much-desired movie, which could not be obtained by either group alone. They included an overnight camp-out where several situations arose, as planned, that required their joint efforts.

On the camp-out, for example, the truck, which was the only available transportation by which they could go for their food, stalled (as planned) just when the boys were getting hungry. It

(232) was a large truck, too large to be pushed even by both groups. The two groups had to work out a plan together. The plan adopted employed the big tug-of-war rope which had been used but a short time before in the bitter rivalry during the tournament. This time, however, both groups pulled together on two lines of rope tied to the truck, and succeeded in getting it started.

At first, cooperation between the groups was confined to the problem situations embodying superordinate goals. Following attainment of the goal, the old bickering and name-calling tended to reappear. Gradually, however, intergroup cooperation was extended by the individuals themselves to widening areas of activities. The first was agreement to take turns going first into the mess hall. Following the successful attempt to start the truck, the two groups prepared a meal together. At length, spontaneous cooperation and friendship between the groups extended to joint entertainments around the campfire, which the boys organized themselves. Even the order of performance was the result of their own planning.

When the experiment was over, boys of both groups came to the staff requesting that they all go home in the same bus, instead of in the separate buses in which they had arrived. This request was granted. The bus made a rest stop on the way home. One group had five dollars which they had won in one of the contests. Their plans for spending it at camp had not worked out. At the refreshment stand, their leader suggested that the five dollars be spent on malted milks for all the boys in both groups. All of his group agreed. This meant that malted milks for everyone would be paid for by one group, but that everyone would have to buy his own sandwiches and other treats. The group which contributed the money was fully aware that the five dollars would have paid for everything that they themselves would have ordered. Nevertheless, they preferred to share it.

The conclusion was clear that cooperation between groups prompted by a series of superordinate goals had the cumulative effect of reducing friction between groups and engendering friendship and favorable evaluations of the other group. This conclusion was supported both by observation and by other results

( 233) which were treated statistically and found significant. Friendship choices were extended for the first time to members of the outgroup. Ratings of the personal characteristics or stereotypes of the other group shifted from largely unfavorable ratings to largely favorable ratings. The categorical verdicts that "all of the other group are stinkers, cheats, sneaky" and the like fell to almost nothing.

On the basis of these experiments, certain conclusions and implications for understanding social responsibility are evident. Responsibility is necessarily in relation to other people with whom one has some relationship. Behavior is evaluated as responsible or as not responsible in terms of values or norms which are recognized by at least two people. Typically, responsibility in social life is evaluated in terms of the values of the groups to which a person belongs. The values upheld by the "responsible" person are those of his reference groups, such as his family, school., play group, gang, club, work group, professional group, community, and nation.

The experiments summarized support the proposition that when individuals interact in activities embodying common goals with high appeal value, a group organization and group values are produced. As a participant in the planning, discussions, decisions, and actions in which the group and its values formed, the individual member takes the group decisions and values as his own personal decisions and values. He behaves in responsible fashion in terms of the status arrangements and values of the group and holds himself personally accountable to do so.

But when his group is in conflict with another group, the individual takes responsibility in furthering intergroup friction and in derogation of the antagonist. In the experiments, healthy, normal individuals zealously pursued a course which was notably responsible in terms of their own group values and aims, but flagrantly irresponsible in terms of the other group.

When the two groups faced superordinate goals which both desired, but neither could attain separately, then individuals in both groups began to experience feelings of friendship and inner responsibility in cooperating with the other group. These inner

( 234) promptings were not the outcome of inherent or basic personality traits. These personal promptings and urges grew because the values of one's own group had changed to friendliness and helpfulness toward the other group. Manifestations of personal characteristics along these lines became the concern of each individual with a sense of inner urgency.

This is a demonstration in miniature of the fact that groups can develop responsibility to one another when aims common to both are immediate rather than abstract, and require the efforts and resources of the groups in question. In terms of the individual, such conditions widen the horizons of personal responsibility. But they do not lessen the inner prompting to be accountable. For the values of friendship and cooperative's with other groups are values of his own group, just as much as norms of social distance and prejudice. Just as he does other values of his group, he considers them his values, his personal preferences and desires.

Surely our investigation of social responsibility should not stop within the confines of one group, one neighborhood, or one community. Modern differentiated societies require that we turn increasingly in research and practice to the problems of social responsibility to other groups in ever widening circles.


  1. Jean Piaget, The Moral Judgment of the Child (London: Kegan, Paul, 1932).
  2. The first experiment carried out in 1949 is reported in M. Sherif and C. W. Sherif, Groups in Harmony and Tension (New York: Harper & Brothers, 1953). It was sponsored by the Yale Attitude Change Program and the American Jewish Committee. The 1953 and 19J4 experiments were carried out with a grant from the Rockefeller Foundation. Results of these experiments are reported in M. Sherif, B. Jack White, and O. J. Harvey, "Status in Experimentally Produced Groups," American Journal of Sociology, LX (1955) 370-49; and in M. Sherif et al., Experimental Study of Positive and Negative Intergroup Attitudes between Experimentally Produced Groups (Noran, Okla.: University of Oklahoma, 1954; multilithed).
  3. For example, see F. B. Miller, "'Resistentialism' in Applied Social Research," Human Organization, XII (1954), 5-8; S. Wapner and T. G. Alper, "The Effect of an Audience on Behavior in a Choice Situation," Journal of Abnormal and Social Psychology, XLVII (1952), 222-29.
  4. For a convenient summary of the experiments and these techniques, see Chapters 6 and 9 in BI. Sherif and C. W. Sherif, An Outline of Social Psychology (rev. ed.; New York: Harper & Brothers, 1956). Other research using indirect or disguised techniques for assessing attitudes is reported in Chapters 13, 15, and 17 of that book.
  5. Gertrude Wilson and Gladys Ryland, Social Group Work Practice (Boston: Houghton Mifflin, (1949), pp. 60-61.
  6. W. A. Friedlander, Introduction to Social Welfare (New York: Prentice-Hall, 1955), p. 419.
  7. Wilber I. Newstetter, "What Is Social Group Work?" in Proceedings of the National Conference of Social Work, 1935 (Chicago: University of Chicago Press, 1935), p. 294.
  8. Ibid.
  9. Grace L. Coyle, Group Work with American Youth (New York: Harper & Brothers, 1948).

Valid HTML 4.01 Strict Valid CSS2