Some Social Psychological Aspects Of Conceptual Functioning

Muzafer Sherif[1]
University of Oklahoma

I am grateful to Dean Schiller Scroggs, who invited me to this conference, and to Professor Northrop and Dr. Kunz for giving me the opportunity to make some remarks on the important topic of conceptual functioning. As a social psychologist, the consideration of "the structure of concepts" is beyond my ken. My chief concern as a social psychologist is with the process of socialization and its consequences in the experience and behavior of individuals in their inter-personal and group relations. The social psychologist is concerned with concepts chiefly because concepts enter as weighty factors in the socialization process and in its consequences. From the psychological point of view, we are on firmer ground when we speak in terms of conceptual functioning, rather than in terms of the structure of concepts.

In this early stage of social psychology — for that matter of psychology in general — I believe our most fruitful efforts are in the formulating of clear, significant problems. Adequate conceptualizations and technical refinements will follow if pertinent problems are raised at the outset. During this discussion, I shall attempt to raise some problems which seem to be significant ones for a social psychological accounting of conceptual functioning.

It is necessary to add, however, that the concepts I shall be concerned with are not merely those used by an elite group of research scientists. In a psychological sense, we cannot differentiate such concepts from those used in everyday life. I shall speak of concepts as those which the individual shares with his group, small or large, and which are standardized for that group and, hence, are used as communicable symbols. In this sense, it is unthinkable to speak of concepts which are not crystallized in words or symbols.


The major approach (but not the only approach) to the study of symbolic processes by psychologists has been the studies of "concept formation." Concept formation is defined by one experimenter in the area (Smoke)[2] as "the process whereby an organism develops a symbolic response — usually, but not necessarily linguistic —which is made to the members of one class of stimulus patterns, or to an aspect of such a class, but not to other stimuli." Such reaction to members of a group or class of stimuli as equivalent or similar has been observed in simple form on the sub-human level.[3] For example, members of sub-human species have responded to triangularity per se, regardless of size or position.

With human subjects, the procedure in most concept formation studies involves repeated exposure to a series of stimuli which may be grouped according to certain physical characteristics into one class, but not into another. A nonsense syllable given by the experimenter is to be associated with the "correct" members of the class. Usually no rewards are given other than those that might accompany a good performance in the task.

There have been fruitful attempts along this traditional line of study. For example, Heidbreder[4] has demonstrated in a series of experiments that concepts of concrete objects are attained with greater ease than concepts of more abstract relationships.

The classical concept formation studies on the whole have emphasized a logical inductive process at the expense of motivational and functional variables operating in most life situations. However, a more serious inadequacy of this line of approach is the neglect of one of the most characteristic features of concepts on the human level —namely, their shared character. In human society, concepts are standardized; they become common property of the group. Concepts are accumulated in human groupings and passed to coming generations. In the terminology of Charles Morris,[5] concepts of a language are comsigns, i.e., standards common to members of a group.

Although concept formation, in the sense of a reaction to a group or class of stimuli as though they were equivalent, may be quite properly studied on the sub-human level, the formation and standardization of comsigns or language signs are peculiar to the human level.[6] The human organism is the only one in the animal kingdom capable of developing and mastering a systematic code of language signs. This capacity is basic to the qualitative variations distinguishing the behavior of man from other primates,[7] and, in fact, one may say to the appearance of human culture. Physiologically, this capacity appears to result from the increased extent and complex-

(63) -ity of the cerebral cortex in man.[8] When the factors leading to such qualitative differences between the symbolic reactions of subhuman and human organisms are ignored, we are eventually caught in a sort of reverse anthropomorphism — a "zoomorphism" in which the heterogeneous linguistic activities of man are interpreted in terms of the symbolic responses of the rat or chimpanzee in the laboratory.

Valid comparisons can perhaps be made between the generalizations of the young child beginning to talk and those of sub-human animals. The child's acquisition of his "first words" is a slow process, marked by trial and error. However, the period of laborious trial and error learning of words is comparatively brief and is followed by an almost dramatic surge in vocabulary development so characteristic that some investigators have called this period "the naming stage." In her recent survey, McCarthy[9] found the literature on vocabulary development in general agreement that "vocabulary appears to increase rather slowly at first, then, quite rapidly throughout the preschool period."

Estimates of children's vocabularies give striking evidence of this spurt in language acquisition. In M. E. Smith's study[10] of children from the ages of eight months to six years, the vocabulary increased five-fold between the eighteenth and twenty-first months. (These data may be plotted as in picture on P. 64).

In short, at some time around the second year of life, vocabulary development is sharply accelerated. This rapid increase in the acquisition of words seems to involve a generalization of greater scope than any hitherto achieved—a generalization cutting across the sensory modalities which adults can express in some such phrase as "everything has a name."

Later in this paper, the functional significance of the child's early words will be mentioned in another context. Here it is sufficient to note that this generalization —that "things have names" — is closely related to the use and acceptance of the instrumental function of words. Speech activity becomes an effective means of satisfying the child's needs and bringing other persons into his activity for aid, comfort and play. In this period, a new and solely human striving appears, — namely, a striving to learn the names of things. It may be indicated at first by the child's own primitive sounds and gestures, such as "eh-eh-eh." Later, we see the familiar picture of the two-year-old pointing again and again to objects and repeating insistently, "What's dis?" Linguistic response acquired by this short-cut method come to supplement and, in time, to replace other forms of response. In fact, it is during this period that the symbolic activities of sub-human animals and those of human children come to a parting of the ways. At the begining of the nine-month period in which the Kellogs raised their own son and an infant chimpanzee together, the chimpanzee surpassed the boy in many respects, including response to human words. But in spite of the fact that


Figure 1.
Figure 1. Vocabulary increase from eight to forty-two months. (Adapted from M.E. Smith, (79), Table 8, p. 54). See Table I.  
Note the sharp increase in size of vocabulary beginning at about eighteen months of age.

(65) the boy was somewhat retarded linguistically in terms of established vocabulary norms, he was easily surpassing the ape around the age of two.[11]

The symbolic behavior of sub-human animals is characteristically bound to more or less discrete and immediate perceptual situations. Pre-human gestures, vocalizations, and response to human words are not removed from the motivational and emotional situations in which they occur.[12] While the symbolic behavior of the young human is similarly tied to concrete perceptual situations and routines, normal children advance to a level of conceptually symbolic functioning: It is within the bounds of known physiological facts and harmonious with behavioral observations to assume that the achievement of a conceptually symbolic level of functioning is closely related to the maturation of the human central nervous system. The validity of this relationship is indicated especially by the breakdown of conceptual functions with pathological degeneration of cortical tissue in disease (such as paresis), brain tumor, or injury.[13]

When the human child attains that generalization embodied in the notion that "things have names," which is revealed by a marked acceleration in vocabulary development around the age of two, he has taken the first major step away from the perceptually symbolic level — in which behavior is bound to immediate objects and situations —toward a conceptually symbolic level of behavior. This conceptually symbolic level of behavior is distinctly a human attainment which is basic to the understanding of the psychological development of the human individual.


Perhaps the most important and certainly one of the most distinctive effects of concepts in human behavior is their categorizing effect in relation to those objects, situations or relations to which they refer. When objects or persons, situations, etc., are subsumed under a common name, perception of and response to the various stimuli tends to be similar as determined by the meaning, i.e., by the generalization, crystallized in the name. The name functions as a category to which stimuli either belong or do not belong. Response to stimuli in the same category tends to be similar. For example, in one study[14] , young children were taught to call two very different stimulus objects by the same name. It was found that "other responses (such as reaching for the object) are more likely to generalize from one to the other than when the two objects have been given different names."

This categorizing effect of concepts has long been noted by various authors in several fields. In a more philosophical vein, we find this fact stressed by such writers as Ernst Cassirer and De Laguna.[15] Ethnologists have gathered extensive evidence substantiating this point and

( 66) some, notably Edward Sapir[16] , have emphasized it in their writings. More recently, Kluckhohn and Kelly[17] generalized as follows concerning such comparative findings: "A language is not merely an instrument of communication and for rousing the emotions. Every language is also a device for categorizng experience. The continuum of experience can be sliced very differently. We tend all too easily to assume that the distinctions which Indo-European languages (or our own particular language) force us to make are given by the world of nature. As a matter of fact, comparative linguistics shows very plainly that any speech demands unconscious conceptual selection on the part of the speaker. No human organism can respond to all the kaleidescopic stimuli which impinge upon it from the external world. What we notice, what we talk about, what we feel as important is in some part of a function of our linguistic patterns. Because these linguistic habits tend to remain as unquestioned "background phenomena," each people tends to take its fundamental categories, its unstated basic premises for granted. It is assumed that others will think the same way, for 'it's only human nature.' When others face the same body of data, but come to different conclusions, it is seldom thought that they might be proceeding from different premises. Rather it is inferred that they are 'stupid' or 'illogical' or 'obstinate'."

Research has yielded considerable evidence of the categorizing effect of concepts in various psychological functions. In perceiving and remembering, it has been found that the introduction of a concept effects subsequent response in terms of the name or symbol. For example, studies by Bartlett, Gibson, and by Carmichael, Hogan and Walter[18] have shown that if a form or object is given a familiar name, either by subject or experimenter, its reproduction is altered in the direction of the name. After surveying the relevant literature, McGranahan[19] concluded that: "The effect of language on perception appears to be to make features of the objective world that are represented by language forms stand out in greater articulation, to give greater individuality to the object or event represented, to cause similarities to be seen in things similarly represented, and in general to influence perception in the direction of the speech-forms."

It is well established that all perceptions, judgments, etc., take place within a frame of reference. The more salient factors, either internal or external, serve as anchorages or reference points in terms of which reaction is made. By their nature, language concepts provide organized and crystallized anchorages which determine response in a major way. Because of its categorizing effect, it is probably language more than any other single factor which lends to the perceptions of human adults their highly organized character.


In learning situations of human adults and children who have gained some instrumental mastery of language, symbolic factors usually enter at the outset because of their involvement in the perceptual process. In a good many experiments on human learning, linguistic factors are, therefore, uncontrolled. The investigator often presumes that only those factors which he introduces into the situation are operating. When experimentally introduced into the situation, linguistic factors alter the usual course of the learning process observed on the sub-human level. Trial and error behavior is considerably reduced or even eliminated. For example, when human subjects are instructed of the sequence of stimuli in a conditioning experiment (e.g. a green light followed by shock), the number of trials exerts no consistent effect on the strength of the conditioned response. Learning may occur in one trial. Similarly, the learned response may be extinguished in one trial by new verbal cues.[20]

Resistance to conditioning may increase with the entrance of symbolic factors. Genetically, the speed of conditioning has been found to decrease around the ages of three to five years when the child achieves instrumental mastery of his verbal processes.[21]

Other experiments have demonstrated that conceptual factors substantially increase the speed and efficiency of discrimination[22] , maze learning[23] , learning of motor skills [24] , and problem solving.[25]

In an ingenious experiment Pyles[26] found that the learning of three-dimensional objects by children ages two to seven years was directly influenced by whether or not the objects were named. The greatest difficulty was encountered in learning unnamed objects. Next in order of difficulty came objects assigned nonsense names by the experimenter. The greatest ease in learning was found for objects named as familiar animals.

In summary, after a child has achieved the generalization that "everything has a name," the learning process can be short-circuited. The trial-and-error encounters with stimuli characteristics of learning by sub-human animals and human infants give way to a predominantly deductive process in which responses to whole groups of objects and persons are learned by the acquisition of a name or concept. It is this telescoping of the learning process which makes possible the inculcation of so many social concepts in the form of short-cut dictums. The child's concepts of objects, persons, and relationships in the social world are frequently attained more by the dictums of adults than by contact with the actual stimulus situ-

(68) -ations. For example, young children, even though they have had no contact with Hindus or Turks, acquire concepts of Hindus and Turks with appropriate accompanying attitudes. The mere fact that a particular person or object is placed in one category rather than another has unmistakable psychological consequences. But in the case of most social stimuli, the consequences are more far-reaching, because their conceptual categorization establishes the social world as the child comes to see it.

It is this acquisition of concepts which makes possible the formation of attitudes toward the many objects, persons, and situations in the child's world. Through the formation of such attitudes, the child is eventually enabled to relate himself psychologically to his environment. In early infancy the child does not clearly distinguish between his own body and its desires and the external world. This distinction cannot be made accurately until crystallized in the concepts of "me", "mine", and "I". A host of distinctions must be made —for example, the ability to locate events accurately in the past, present, or future — an achievement which occurs very gradually. Further conceptualization of persons and objects with the accompanying formation of appropriate attitudes leads to the formation of a personal identity having definite psychological relationships with other persons and groups. These attitudes are, then, ego attitudes, largely formed in relation to standardized concepts or norms.[27] They determine the personal identity of the individual as he experiences it and constitute his ego or self. Until such conceptual distinctions are made, the child cannot begin to set up future standards or goals for his own behavior. According to Gesell, the setting of goals for the future does not begin until around the age of seventy-two months[28] , that is, after a considerable degree of development of conceptual functioning.

This ego development — the acquisition of concepts pertaining to objects, persons and relationships in the social world and the formation of attitudes psychologically relating the individual to these stimuli — constitutes the main core of human socialization. The work of ethnologists convincingly shows that the individual becomes altruistic, individualistic, competitive, favorably inclined toward this group of people, antagonistic toward that group, as determined by the norms or concepts existing in the society of which he is a member.

This idea is best expressed in the imposing work of Gardner Murphy on Personality, which is the most comprehensive synthesis of what little we know on personality to date.[29] The self is not an innate entity which is present at birth. Its development is a naturalistic process as the child comes into contact with his environment. The main lines of the development of self are presented by Gardner Murphy in three stages. At its earliest stage, the development of self may be characterized as primarily perceptual — this is when one's own body is clearly delineated from surrounding objects; when attitudes and identifications develop in relation only to persons who

(69) are immediately present within his perceptual range, and who are instrumental in satisfying his needs. Then, as his ability to comprehend qualities develops, the self begins to acquire qualities or traits. Later, as his conceptual functioning develops, his self becomes more and more a classificatory system. This classificatory system implies drawing family lines, lines of school membership, church membership, ethnic group membership, etc., with all the appropriate attitudes that go with each one of them.

From the consideration of the vastly significant effects of concepts in human psychological development, it does not follow, as some might conclude, that every reaction, social response, or psychological process is determined solely or even chiefly by verbal factors. Every reaction is jointly shaped by internal or internalized factors and by external stimulating conditions operating at the time in their functional relationship. As we shall see in more detail in a moment, it is obvious that the conceptual classifications in a society and the generalizations made by children are not based primarily on logical processes or objective common properties of stimuli. In the great proportion of instances, motivational and affective factors enter to determine the stimuli subsumed under a concept and, hence, to determine its meaning. On the other hand, it is factually erroneous to consider the process one of mere subjective categorization. This is the frequent implication of uncritical attempts to demonstrate the relativity of concepts in different societies. Objective stimuli almost always enter into the drawing of conceptual distinctions. Further, when the discrepancy between behavior and objective reality becomes too great, there are forces from society (in the case of the child) or within society tending to bring the two into closer correspondence.

In this connection, we must note that concepts themselves are at the outset by-products of interaction of individuals in human groups, in their efforts to satisfy their needs and to deal with their environment. It seems that the vocabulary of a group or society, hence its classifications of things, reflects the practical activities of the group in dealing with life processes.[30] It can be expected that in relatively isolated small human societies, the range of vocabulary will be proportional to the range of objects, persons, situations, etc., which have functional significance in their on-going vital activities. Of course, in relation to complex societies, this point would have to be altered substantially, especially because of the acquisition of words from other societies, and other complicating factors in the complex culture.

It has frequently been observed that the concepts in primitive societies differ widely from those made in the languages of more developed societies. As Hocart [31] rightly pointed out in 1912, such differences are not indicative of the inferior generalizing power of the primitive mind or some other peculiarity of primitive mentality. Rather such differential classifications stem from differences in the

(70) vital activities carried on by the groups in specific environments; they are standardized to deal with such differences. It is not possible in this paper to do more than suggest examples of such differential classifications. (I am grateful especially to Dr. Clelland Ford of Yale University for suggesting much of the body of literature we have studied in this connection.)

As an example, the Masai of East Africa are a nomadic group whose chief occupation is cattle raising. The Masai have as many as seventeen terms for cattle — separate words for a cow with one calf or two calves, a sterile cow, etc.[32] The chief crop of the Ifugao of the Phillipines is rice, and it is called by twenty different names during various stages of planting, growth, harvesting, etc.[33] In the Solomon Islands, one of the staples is nuts. Two nuts so alike in appearance as to seem identical except in size, are given different names. These names indicate important distinctions, for the nuts —so vital in the economy — have different seasons, are gathered differently, are cracked and preserved differently. In fact, in terms of the activities related to them, the two nuts are similar only after they are roasted or made into pudding. And then, they are called by the same name.[34]

While gardening and animal raising groups may have many names for those plants or animals vital in their economic life, they may have only very vague concepts of objects not important to them. Malinowski [35] noted that a plant or bird with no value to the group for the larder or clothing was dismissed with some phrase, such as "Oh, that is just 'bush"', or "merely a flying animal." On the other hand, in those groups in which hunting or gathering are the chief means of livelihood, especially against great odds as among the Arunta of Australia or the Siriono of Bolivia, names have been standardized for almost all the neighboring flora and fauna.[36]

In our study of five Turkish villages with differential degrees of isolation from modern industrial life, we found a general relationship between the mode of life and the concepts relating to such fundamental functions as timing and estimating distances. Concepts of time and distance in the most isolated villages were vague and inaccurate. The greater the functional contact with modern industrial life, the more precise and accurate the concepts of time and distance became.[37]

Of course, it is not necessary to go outside of one's own society to find examples of concepts arising in the course of group activities. Concepts of obvious functional value to the users form in gangs of youngsters and adolescents in large cities, as reported by

(71) Thrasher.[38] In prisons, at least a portion of the argo originates in the prison situation and reveals concepts related to the more prominent aspects of the depriving prison life.[39] In the British Armed Services, 240 neologisms—entirely new words—were reported to be standardized during the early years of the recent war.[40] The sources, forms, and phonemic properties of such new words are, of course, complicated problems which lie beyond the ken of the psychologist. They are rather subjects for the students of language.

Now I shall take a brief glance at the functional significance of children's vocabulary development. In the activities of the human individual, as well as in group interaction, symbolic responses are by-products, first. As Lewis has clearly pointed out,[41] language activity is at first merely an accompaniment to other forms of response in the child's efforts to satisfy his needs, secure aid, comfort, etc. Only gradually does speech replace other responses. The child's earliest sounds are primarily an expression of his needs and affective state and are directed only secondarily toward objects or persons.[42] As the child discovers that his sounds have consequences on the course of his activities and begins to use them instrumentally, objective reference becomes more prominent. Still, the words of young children may be used in such an extensive fashion that they are practically unintelligible to an adult. Children's words take on approximate dictionary meanings very gradually, and only then with considerable correction and pressure from adults and with the child's emerging desire to conform to adult usages. This fact is well illustrated by the definitions acceptable at various age levels on the Stanford-Binet test of intelligence. At the lowest age levels, acceptable definitions are simply in terms of the object's use. Thus, an orange may be defined as "we eat it for breakfast." The series of studies conducted by Piaget[43] on children's language development revealed and traced these developments in fascinating detail. This intimate link between motivation and concepts does not suddenly break with the attainment of conventional usages. Psychologically, concepts are all connotative in some degree. The denotative and connotative character of concepts cannot be separated from a psychological point of view. This is entirely harmonious with what is known of other psychological functions. For example, perceptions are not merely cognitive affairs. They are always jointly determined by internal affective and motivational factors and by external factors. Of course, the relative contribution of affective and cognitive factors varies in different instances.

In some cases, children standardize names or labels peculiarly their own for objects which have functional value in their activities. For example, Watt[44] reports that a sixteen-month-old child standardized the label "yo-yo" for objects which he could carry by handles.

(72) Another group of portable objects—those without handles—was called "go-gos."

Such unique standardizations by young children have also been observed in the case of children in groups. Jespersen [45] reports that twin boys five and a half years old who were frequently left to shift for themselves developed a whole set of words. When left alone, they "conversed pretty freely and in completely unintelligible gibberish." As a matter of fact, unique standardizations of words by twins is common enough to be noted in standard texts on "speech defects."[46] I am grateful to Professor John W. Keys of the Department of Speech at the University of Oklahoma for calling this fact to my attention and also for reporting his own observations of the peculiar words of twins. In one family, the twins developed several special words. One word "tedaden" was used in the children's communications to mean "climb upon." Another word, "ding-a-ding" was used to mean "trade" or "you give me what you have and I'll give you what I have".

Of course, such childhood standardizations ordinarily pass into oblivion, partly through adult pressure and in part because the motivations and events so crystallized by the children are more or less specific to situations in which they have participated and are not shared to any degree by others.

With such facts as these in mind, we could expect that names or labels are likely to be produced when an individual or a group of individuals attempts to deal with unnamed objects, situations, or relationships having motivational value to them. Such a tendency was found in a study we made of young children. Specifically, pre-school children were placed individually and in small groups in a situation in which the only means of securing a much-desired toy was by a verbal response. Since the children were all from the lower economic levels (most of them came from a school in a poorer section of New Haven), it was possible to present toys which were highly prized and which were also unnamed by the children. Adult pressure favoring one kind of verbal response in preference to another was at a minimum. Yet, after the children faced this situation individually several times, all but two children out of twenty-two standardized a label for at least one of the toys. In the group situations, labels were standardized by all of the groups and were shared by every individual member. No doubt the standardization of names in the group situations was achieved so readily because they were advantageous in coordinating the play activities of the group.

So far, I have touched briefly upon several topics related to the social psychology of conceptual functioning — namely, the greatly accelerated development of vocabulary around the age of two in the human infant, the structuring and categorizing effects of concepts or language symbols on perceptions, learning, motives, ego development, etc. Also, I pointed briefly to the functional significance

(73) of concepts or symbols in relation to the vital motives of the individual and the group. These topics raise fundamental problems in any serious psychological theorizing particularly in the field of social psychology. We could, therefore, continue this paper from now on along one of several equally significant alternatives.

In view of the particular interest of this conference in integrated education, I propose to devote the rest of this paper to an alternative which has serious practical, and hence educational, consequences in human relations as well.

This educational issue which is rather widely discussed by ethnologists, sociologists, psychologists, as well as by semanticists in recent years, is related to the well-known fact of the structuring and categorizing effect of concepts or symbols. It is mainly through the vehicle of language (concepts or symbols) of his group that the individual acquires the culture and techniques of his particular civilization, whatever this particular civilization may be; that his perceptions, learning, and attitudes are structured and categorized, subject, of course, to variations due to individual differences. Alongside of these achievements made possible by language, we must also note that it is through the vehicle of language, through concepts and symbols, that he acquires the established traditional prejudices, stereotypes, superstitions, erroneous generalizations concerning nature around him and people within and without the range of his perception. Through the existing concepts of the group, which he acquires through the years of childhood and adolescence in the process of socialization, his mind is already made up for him about things, about the nature of human relations, about goodness or badness of people, about the superiority or inferiority of this or that group. Through the established concepts in daily usage in his group, his attitudes about things and people are already made for him, in many cases, even before his actual contacts with the referents of these concepts. Even after he has a chance to come into contact with the situations and people in question, his perceptions, his evaluations in the face of actual contacts are colored by his already existing attitudes mainly derived through the concepts of his group. For these vital concepts of everyday life concerning relations between man and woman, between professions, between groups and classes of people, are not mere cognitive affairs, scientifically attained on solidly established facts and their functional relations. These concepts embody demarcation lines and heavily-loaded positive and negative attributes or characteristics. It is for this reason that the attempts to bring about positive changes in attitudes by bringing together individuals belonging to antagonistic groups have been, on the whole, singularly unrewarding. In view of the indications of the accumulating evidence in the psychology of perception and attitudes today, it would be surprising if the attempts to bring about significant shifts in attitudes on the basis of actual contacts alone were effective. For we do not see with our eyes alone, and hear with our ears alone, but we see and hear, at the same time, with our attitudes, of which the major ones are verbally acquired. And, I suspect, the scientists, even research scientists, do not constitute an exception to this rule, especially when they are outside of their own lines of specialization.

I have time to illustrate the idea presented in the preceeding par-

(74) -agraph only with one example. It deals with the concepts related to the psychological nature of the two sexes, male and female, prevalent in the Middletown, intensively studied by Professor Robert Lynd of Columbia University. In Lynd's words, "(But) this culture says not only that men and women do different things; they are different kinds of people. Men are stronger, bolder, less pure, less refined, more logical, more reasonable, more given to seeing things in the large, but at home needing coddling and reassurance, `like little boys.' Women are more delicate, stronger in sympathy, understanding and insight, less mechanically adept, more immersed in petty detail and in personalities, and given to `getting emotional over things."' (P. 176f.)[47] Naturally, these conceptions of the psychological nature of man and woman come in as important factors in determining the reciprocal attitudes and roles of the growing boy and girl.

Some of the most serious instances of the stereotyping effects of the standardized concepts are in the field of inter-group relations. Consider such concepts as Gentile, Jew, American, Turk, White Negro, Christian, Mohammedan, and for that matter, labor, capital, strike, production, etc. Any individual classified under any of these terms, or even the very mention of these words, immediately arouses positive or negative qualities in the eyes of the person facing him, as determined by the particular characterization of these terms in his respective group.

In view of the stereotyping and even blinding effects of such concepts prevalent in everyday life situations for millions of people, the educational or practical problem is self-evident. This huge educational problem cannot be even adequately approached until we attain scientifically valid concepts and generalizations in major areas of human relations and unify them into an integrated account of human relations. After achieving this integrated account of human relations in terms of universally valid scientific concepts, and not in terms of the special conditions of one nation alone, the first educational task, it seems to me, is to make them prevail, not only in the citadels of advanced learning, in colleges and universities, but through all stages of education, in the nursery, at home, in play, in the club and professional organizations, etc.

In the light of even meager evidence concerning the effects of attitudes and ego-involvements, one can say with some assurance, that the introduction of courses in scientific method or courses in the analysis of concepts in colleges will not safeguard against the stereotyping effects of the concepts coming down from the past.

In this connection it is relevant to question what per cent of young people do change their concepts acquired in their home and communities on the basis of courses in biology, psychology, and sociology taken in their school years.

It is relevant to question how many students change their distance or prejudice scale acquired in the community with which they identify themselves, after learning in their anthropology courses the loose popular usage of the concept of "race". How many of them modify, on the basis of comparative data of an ethnology course,

(75) the conception of "human nature" they picked up in the image of their particular culture? The weight of evidence as presented by the studies of F. H. Allport, Eugene Hartley, and others, indicates that there is little difference between college and non-college populations in these matters.

By the time students come to high school, not to mention college and university, their ideas concerning human relations, both on the inter-personal and inter-group level, are already formed as prescribed by the existing concepts of their group. Their conceptions of their roles as man and woman, their conceptions of what are proper and improper personal relations, how distant or close given groups should be held in relation to themselves, are already formed as prescribed by the definitions of their group in these matters. We may question again the effectiveness of the scientific concepts and generalizations concerning these very matters taught in psychology or sociology courses. I suspect with good reason that such scientific concepts and generalizations superimposed in the formal school courses (which are taken in many cases simply to add so many points toward graduation) have only limited effects. Whereas, the attitudes and identifications taken in more naturally, almost like the act of breathing, from the widely prevalent concepts of their group concerning these very matters are the ones, on the whole, which have greater psychological weight and, hence, are likely the ones to be carried into real life situations in school and thereafter. For the attitudes derived from the widely prevalent concepts of our group are not viewed as so many bits of information; they do become a part of ourselves; they constitute the more or less stable anchorages of our personal identity. It is not surprising, therefore, that when confronted with concepts widely at variance with the established attitudes concerning human relations, the students feel uneasy, even insecure, about things. Most of them carry over the prevalent attitudes acquired from their communities when they settle down as citizens in their respective communities. I must add, however, that there are a few who try to readjust their outlooks in terms of the scientific concepts taught in their courses and acquired in the intellectual atmosphere of the school. But these few are the ones who become personally involved in the subject matter of the new concepts.

As an antidote to the stereotyping effects of concepts, it is suggested that the abstract nature of general concepts be stressed, and reactions be based on actually observable events and cases. It is stressed by semanticists especially that there is no such an entity as the generalized cow or tree. What actually exists are Cow 1, Cow 2, etc. Likewise, there is not such an entity as the Negro, which is a concept that comes rather high on the abstraction ladder; but actually there are Negro 1, Negro 2, etc., each representing individual differences — intelligent ones, dumb ones, good ones and bad ones, as is the case in any other group. This process analysis is carried further and further. For example, to stress the fact that even a single individual is not a fixed entity throughout the course of his life history, it is suggested that we think and react to Mr. X not as Mr. X in a cut-and-dried way, but instead as Mr. X 1949 and Mr. X 1960, etc.

The efforts of semanticists in this direction are certainly laud-

(76) -able. But it does not really give us a realistic antidote against the stereotyping effect of concepts which lead us to various erroneous and harmful types of behavior.

The semantic antidote exemplified above overlooks a basic psychological tendency of human beings. This basic psychological tendency of the individual is to think and react in relation to objects, situations, and people in terms of generalized standards of concepts—in short, in terms of their established frames of reference. This is true both in relation to situations which are designated as merely cognitive as well as situations loaded with value properties. This basic psychological tendency of reacting in terms of already existing generalized standards or concepts is responsible for the relative consistency of the individual personality under varying conditions from day to day. On the other hand, it is responsible also for the stereotyping, narrowing-down effects when concepts not founded on fact come into play. We would be in a continuous state of suspensea state which is painful—if we were to postpone judgment, evaluation, etc., until all the relevant data were made available from day to day in so many varied situations. The exceptional research worker, who at times has to stay in suspense until the necessary data are in before he can formulate a statement, knows too well, I think, with what great efforts this suspense is achieved and at what price in nervous strain. It is too much to expect this achievement from millions of citizens who are rightly less troubled with suspending judgment until all the relevant facts are available. Life would be miserable, indeed, if people were to spend their time and energy on Cow 1, Cow 2, etc. Even highly trained research scientists usually do not practice this when they are not in the field of their special training. They usually react as other citizens in their community. It would be very instructive from this point of view to collect pronouncements of physical scientists of various nations during the last twenty years, for example, concerning the nature and goals of human relations, and to find what per cent of these pronouncements reflect scientifically valid concepts and what per cent are concepts reflecting their particular identifications with particular groups.

The basic psychological tendency to perceive, to evaluate and react always in terms of certain generalized standards, anchorages, or symbols can be illustrated by various experiments in the fields of judgment, perception, attitudes, etc. Especially in cases of stimulus situations which are vague or unstructured and lack sufficient objective cues, the subjects do not announce that they do not have the necessary objective data to arrive at judgments, to structure perceptions, to formulate conclusions; on the contrary, they do form judgments, they do structure perceptions and draw conclusions, usually on the basis of anchorages or standards they themselves provide internally. For example, experiments utilizing the autokinetic situation have given us significant leads along these lines. Here, I can take time to cite only one concrete example, deliberately from the field of human relations. Dr. Eugene Hartley, in one of his studies of prejudice,[48] inserted names on non-existent groups to the list of actual groups to be rated by his subjects along

(77) a social distance scale. The subjects were students of some Eastern universities, one of them being one of the oldest and best known institutions in America. The subjects did not stop in rating when they came to the names of non-existent groups. They did not say, "we never heard of these groups and therefore we cannot rate them." They went ahead and rated them, and their ratings were, on the whole, on the general level of their prejudice toward other groups. In other words, they carried over their generalized level of prejudice to non-existent groups. Thus these university students, including those from the famous university who are among the better trained youngsters in scientific methods, took definite stands concerning even mere names to which they were exposed for the first time. It seems to me that the results of this study, which serves as a paradigm of several studies, stands in bold relief as a warning against the unreality of such attempt exemplified in the Cow 1, Cow 2, instead of a generalized Cow approach.

It seems to me that the conclusion to be drawn from all these considerations is evident. Concepts enter into psychological functioning of the individual as categorizing and structuring factors. Because of their weight as categorizing and structuring factors, concepts which are not scientifically validated have a stereotyping and blinding effect on the experience and behavior of individuals. At least a great majority of the concepts in everyday use by millions of people today are concepts which are not scientifically valid. The effective antidote is not to superimpose on them in the course of education the meager facts in the sciences of human relations as they exist today. In view of the general categorizing tendency of concepts at the human level of psychological functioning, the approach that suggests as an antidote taking every single case on its own merits or demerits is unrealistic. If concepts are in use as standardized and shared property of the group, they cannot help being effective. Therefore, the only effective way of getting rid of concepts which lead us into blind alleys is to replace them with scientifically valid ones, all the way through. The field of human relations is still in a primitive stage. Therefore, it is in this huge vital field that the attainment of scientifically valid concepts is most urgent. It may be that, if an integrated scheme of valid concepts is attained in the field of human relations, the existing dichotomy between science and ethics, between theory and practice, between heart and reason, will evaporate into thin air, and future generations will not be bothered with major problems of concepts vs. values in their educational policies.


  1. Carolyn W. Sherif has collaborated with me both in the selection of material and in the writing of this paper.
    The work on this paper, including the collection of material and the preliminary study of naming mentioned in brief, was done in 1947-48 while the author was a Rockefeller Foundation Research Fellow at Yale University. I am grateful to Professor Carl I. Hovland, Chairman of the Department of Psychology at Yale, and to the Rockefeller Foundation for making this study possible.
  2. K. L. Smoke, "Concept Formation." Encyclopedia of Psychology. P. L. Harriman, edit. New York Philosophical Library, 1946, p. 97.
  3. C. T. Morgan, Physiological Psychology. New York: McGraw-Hill, 1943, pp. 109-111.
  4. E. Heidbreder, "Toward a Dynamic Psychology of Cognition." Psychol. Rev., 1946, LII, pp. 1-22.
  5. Charles Morris, Signs, Language and Behavior. New York: Prentice-Hall, 1946.
  6. R. M. Yerkes, Chimpanzees, A Laboratory Colony. New Haven: Yale University Press, 1943. T. C. Schneirla, "Psychology, Comparative," Encyclopedia Britannica, 1948.
  7. T. C. Schneirla, "Levels in the Psychological Capacities of Animals," Encyclopedia Britannica, 1948.
  8. T. C. Schneirla, "Psychology, Comparative," op. cit.
  9. D. McCarthy, "Language Development in Children." Chapter X, pp. 476-581 in L. Carmichael edit., Manual of Child Psychology. New York: Wiley, 1946.
  10. M. E. Smith, "An Investigation of the Development of the Sentence and the Extent of Vocabulary in Young Children," Univ. of Iowa Studies in Child Welfare, III, No. 5, 1926.
  11. W. N. Kellog and L. A. Kellog, The Ape and the Child. New York: McGraw-Hill, 1933.
  12. T. C. Schneirla, "Psychology, Comparative," op. cit.
  13. T. C. Schneirla, "Psychology, Comparative," op. cit.
  14. N. E. Miller and J. Dollard, Social Learning and Imitation. New Haven: Yale University Press, 1941.
  15. Ernst Cassirer, Language and Myth. New York, 1946. G. A. De Laguna, Speech, Its Function and Development. New Haven: Yale University Press, 1927.
  16. Edward Sapir, "The Status of Linguistics a® a Science," in Selected Writings of Edward Sapir in Language, Culture, and Personality, edited by David G. Mandelbaum. Berkeley: University of California Press, 1949.
  17. Clyde Kluckhohn and W. H. Kelly, "The Concepts of Culture," in R. Linton (edit.) The Science of Man in the World Crisis. New York: Columbia University Press, 1946. Pp. 100-101.
  18. F. C. Bartlett, Remembering. Cambridge England: Cambridge University Press, 1932. J. J. Gibson, "The Reproduction of Visually Perceived Forma," J. Exper. Psychol., 1939, 12, pp. 1-39. L. L. Carmichael, H. Hogan, and A. Walter, "An Experimental Study of the Effect of Language on the Reproduction of Visually Perceived Form," J. Exper. Psychol., 1932, Pp. 73-86.
  19. D. V. McGranahan, "The Psychology of Language," Psychol. Bull., 1936, 33. Pp. 202.
  20. S. W. Cook and R. E. Harris, "The Verbal Conditioning of the Galvanic Skin Reflex," J. Exper. Psychol., 1937, 21. Pp. 202-210.
  21. G. H. S. Razran, "Conditioned Responses: An Experimental Study and A Theoretical Analysis," Arch. Psychol., 1936. No. 191.
  22. E. R. Hilgard, A. A. Campbell, and W. N. Sears, “Development of Discrimination With or Without Verbal Report." Amer. J. Psychol., 1997, XLIX. Pp. 564-580.
  23. C. J. Warden, "The Relative Economy of Various Modes of Attack in the Mastery of the Stylus Maze." J. Exper. Psychol., 1924, VII. Pp. 243-276.
  24. F. L. Goodenough and C. R. Brian, "Certain Factors Underlying the Acquisition of Motor Skill by Pre-School Children." J. Exper. Psychol., 1929, XII. Pp. 127-166.
  25. L. Welch and L. Long, "Comparison of the Reasoning Ability of Two Age Groups," J. Genet. Psychol., 1943, LXII. Pp. 63-76. L. Long and L. Welch, "Reasoning Ability of Young Children," J. Psychol., 1941, II. Pp. 21-44 ; and "Influence of Level of Abstraction on Reasoning Ability," J. Psychol., 1942, XIII. Pp. 41-69.
  26. M. K. Pyles, "Verbalization as a Factor in Learning." Child Development, 1932, III. Pp. 108-113.
  27. Muzafer Sherif and H. Cantril, The Psychology of Ego-Involvements, New York: Wiley, 1947.
  28. A. Gesell and F. L. Ilg. Infant and Child in the Culture of Today, New York Harper, 1948. Pp. 340.
  29. Gardner Murphy, Personality, New York: Harper, 1947.
  30. B. Malinowski, "The Problem of Meaning in Primitive Languages," Supplement I in C. K. Ogden and I. A. Richards, The Meaning of Meaning, New York: Harcourt, Brace, 1930. A. M. Hocart, "The 'Psychological Interpretation of Language"'. Brit. J. Psychol., 1912, V. Pp. 267-279.
  31. Hocart, Ibid.
  32. F. Merker, Die Masai, Berlin, 1904.
  33. R. F. Barton, The Half Way Sun. New York: Brewer and Warren, 1930.
  34. Hocart, op. cit., p. 272.
  35. Malinowski, op. cit.
  36. B. Spencer and F. J. Gillen, The Arunta: A Study of a Stone Age People. London 1927. A. M. Holmberg, "The Siriono: A Study of the Effect of Hunger Frustration on the Culture of a Semi-Nomadic Bolivian Indian Society." Doctorate dissertation on file in library, Yale University, July, 1946. P. 121.
  37. Muzafer Sherif, An Outline of Social Psychology. New York: Harper, 1948. Pp. 874-386.
  38. F. M. Thrasher. The Gang. Chicago: Univ. of Chicago Press, 1927.
  39. D. Clemmer, The Prison Community. Boston: Christopher, 1940.
  40. J. L. Hunt and A. G. Pringle, Service Slang. London, 1945.
  41. M. M. Lewis, Infant Speech. New York: Harcourt. Brace, 1936.
  42. M. M. Lewis, Ibid. Pp. 143-144.   McCarthy, op. cit. P. 604.
  43. J. Piaget, Language and Thought of the Child. London: Kegan Paul, French, Trubner, 1928.
  44. A. F. Watt, The Language and Mental Development of Children. London: Harrop, 1944.P. 87.
  45. O. Jespersen, Language. Its Nature, Development and Origin. New York: Holt. 1923. Pp. 186-186
  46. M. F. Berry and J. Eisenson, The Defective in Speech. New York: Crofts, 1945. Pp. 277-278. C. Van Riper, Speech Correction Principles and Methods. New York: Prentice-Hall, 1947. P. 105.
  47. Robert S. and H. M. Lynd, Middleton in Transition. New York: Harcourt, Brace, 1987.
  48. Eugene L. Hartley, Studies in Prejudice. New York: Kings Crown Press. 1946.

Valid HTML 4.01 Strict Valid CSS2