The Hanover Round Table and
Social Psychology of 1936
Floyd Henry Allport
IN THE days when psychologists were directing their thoughts toward the 1936 meeting of their national association, the happy suggestion was made by Dr. Steuart H. Britt that the time had come for social psychologists to hold a round table to discuss the subject matter of their science. The suggestion was well received; and at the request of Dr. Britt and the Secretary of the Association I availed myself of the privilege of serving as Chairman of such a gathering. In planning this meeting it was decided to steer a middle course between haphazard discussion from the floor and long theoretical orientations presented by a few. The plan was adopted of selecting a list of topics representing the most active research areas of social psychology, and asking specialists in these topics to present brief papers, each of which was to be followed by discussion. In the array of topics selected, as shown by the papers following, conspicuous omissions occur. Because of the necessity of conducting the entire round table within a little over two hours, and because the contributors were psychologists, it was thought best to limit ourselves to the psychological approach, omitting the sociological viewpoint and the controversy over group and individual, except where these forced themselves into the discussion. That the sociological approach inevitably entered is clearly indicated in the contributions themselves. Papers written from the standpoint of ethnology and linguistics would also have been much to the point; but these, too, had to be omitted for lack of time. The important contributions of psychopathology and psychoanalysis were ruled out for the same reason, and finally also, and with special reluctance, the applications of social psychology to current social problems. In spite of these limitations, however, the topics remaining probably represent the most direct and promising lines of attack so far as psychological investigation is concerned.
(456) The logical arrangement of the program was as follows. After an introductory survey of the topics and problems which might be considered subject matter for social psychology, attention was to be given to some of the basic human materials of inter-individual behavior, namely, to individual differences and characteristics of personality. It was planned that the next group of papers should deal with special methodological viewpoints and techniques such as observation and measurement. A final series of papers were intended to show the character of methods and results in some specialized types of situation in which the social behavior of individuals had been studied. In the assignments given the contributors, they were asked, first, to emphasize actual experimental or observational investigations, and, second, to show the bearing of the methods and findings of these researches upon the task of social psychology as understood by the contributor himself.
In order to focus the efforts of the contributors still more sharply, a tentative definition of social psychology was suggested by the Chairman in advance. Whether a participant would endorse this particular definition was made entirely optional; and, as will be seen, there were a number of criticisms of the formulation suggested. Probably this device did aid in making the discussion more pointed, and in stimulating contributors to give thought to their own definitions in their bearing upon the methodologies presented. The definition suggested had been formulated with the object of representing a fair consensus of opinion among psychologists working in the field of social psychology. "Social Psychology," according to this definition, "may be defined as a study of the behavior (or awareness) of individuals in their reactions to other individuals or in social situations, and the behaviors through which individuals stimulate one another in such situations."
So varied, interesting, and provocative were the contributions to this round-table that it was thought desirable to publish the entire series as a symposium. The Chairman and the contributors wish to thank the editors of SOCIAL FORCES for making this possible. The publication of these papers in a sociological journal is, we believe, particularly fortunate at this time. In the last few years social psychology has been vigorously developing in two somewhat diverging directions. The sociologists have studied and systematized the subject from one point of view and the psychologist from quite another; and the members of the two schools have had all too little to do with one another. Psychologists may welcome this opportunity to bring some of their contributions to the ' attention of the sociologists for whatever service they can render to the common cause.
No elaborate review is needed of the diverse and contrasting themes here presented. These able papers speak for themselves. Nevertheless there will be apparent, upon reflection, a few comparisons through which a certain unification, or at least some sense of relationships, might be achieved. The present writer hopes that no one will feel he is over-stepping his bounds as Chairman of a round-table which has now "passed into history," if he attempts, in the light of his own experience, this task of comment and interpretation.
Dr. Britt's view of the scope of social psychology is based upon a comprehensive survey and an eclectic treatment for the interests of teaching and research. Significant human problems, rather than logical definition and systematization, are the criteria upon which he would select his
(457) materials. In this respect Dr. Britt's practical approach differs from that of professor Brown, who believes that the science can now be precisely and theoretically defined, and working concepts developed for its systematic exploration. One is reminded, in Dr. Britt's proposal, of Dr. Hadley Cantril's article on "The Social Psychology of Everyday Life," and also of the association of psychologists recently formed for bringing psychology into service in the solution of modern social and economic problems. Dr. Britt's paper shows how social psychology, like other sciences, has emerged from its philosophical background and has now arrived at the stage of empirical investigation.
Starting at the foundations of the subject, Dr. Miles' paper reminds us that we are dealing with human material, and that we must therefore take into account a basic biometric and psychometric law, the law, namely, of individual differences and their distribution according to the normal probability distribution. No account of societal phenomena, of groups, customs, traditions, institutions, or culture patterns, will be psychologically accurate unless a place is given to the natural individual variations of age, sex, class, race, temperament, and personality.
With Dr. Murphy's suggestive paper, we come to a new theme in the study of personality and social relationships. From one point of view our very perception of the personality of another individual varies with the social situation in which he is active at the moment. By reference to testing results, Dr. Murphy shows how our program of measuring intelligence and personality may mislead us unless we take this fact into account. A real service is here rendered by revealing a critique which must be applied by those who assume that we must first know something about personalities and then see what happens to them during their social adjustments. This social-situation view of personality is linked to Gestalt psychology and the "field-theory" of Dr. Brown, on the one hand, and to the teachings of the sociologists upon the other. It might be pointed out, however, that there is still place for a careful distinction between our perception of the personality of another and that personality itself as representing, at least hypothetically, some fundamental or underlying consistencies possessed by the individual himself, and shown through varying social situations if we but put ourselves in a position to see them. Much evidence, of course, is being accumulated by the "characteristic factor" exponents of personality. Methodologically it is possible either to look at the problem as Dr. Murphy has done, or to attack it from the opposite direction. Neither methodology invalidates the other. And in suggesting a new and interesting approach to individuals in social interaction, Dr. Murphy has made a substantial contribution.
Turning from the individual human materials to methods of study, we find that Dr. Goodenough's paper takes us at once to the heart of this problem. When we think of the vast quantity of literature written from the behavioristic viewpoint, we are struck by the way in which the authors of this literature have overlooked a fundamental problem of scientific method. Behavior is not a simple, tangible phenomena, readily measurable, like the specific gravity of a liquid or the secretion of a gland. Its observation and record may involve many subjective elements and other errors of interpretation. Until, therefore, some criterion can be established as to the reliability of behavior observations and methods for insuring accuracy, a true science of human behavior cannot
(458) be developed. And this criticism applies as forcibly to behavior in the social field as to any other form. The service rendered by Dr. Goodenough and her associates in the field of child behavior is, therefore, of great significance. Her conclusions, though somewhat negative and a bit discouraging, serve only to heighten our awareness of the necessity of making the development of a sound methodology and technique of observation our first major problem. The paper by F. H. Allport offers some further suggestions, from a slightly different approach, using not frequency categories in units of time or the provision-results of behavior, but telic or teleonomic units. Another possible aid in the problem of securing reliable observations of behavior may come to light through a re-definition of the elementary social conditions or situations in which the behavior is observed to occur.
Attitude studies have developed at a rapid pace in recent years. They afford us interesting content as well as new techniques in the multi-individual field. Dr. Katz's paper summarizes for us some of the most important of these possibilities. In many societal situations, attitudes, that is, "readinesses to respond " must be conceded to be of as much significance as the overt responses themselves. These "readinesses" are significant phenomena of political science, class allegiance, group conflict, and social organization in general. A closer coöperation is needed between psychologists, who are developing ingenious techniques of attitude measurement, and sociologists, who are competent to point out the areas for their significant use. Dr. Katz's paper indicates a relationship between the "patterning" of attitudes and the aspect of personality stressed by Dr. Murphy. It also contains the important implication that societal situations, which have been given collective or superorganic, cultural names by sociologists, such as classes, derivative groups, and customs, can be more precisely and quantitatively described as statistical aggregates of individuals whose mode of distribution falls at a certain point upon the attitude continuum. And in this implication a whole new field of research is opened up, a field in which sociologists and psychologists can join hands and work effectively together. This is the view which has been developed by F. H. Allport in his paper upon "The Observation of Societal Behaviors" and elsewhere.
Another significant rapprochement between psychologists and sociologists, employing a different set of concepts derived in part from Gestalt psychology and the work of Kurt Lewin, is indicated in the paper of Dr. J. F. Brown. Though acknowledging that human beings are the ultimate units from which we must derive our indices and measurements, Dr. Brown has conceived individuals as acting within, or as parts of, a dynamic field in which they occupy a kind of psychological space. "Social psychological events," he says, "occur in social fields where the activity at any locus within the field is determined by the total social field structure." Changes in the field dynamics, tensions, barriers, and other concepts appropriate in fieldtheory are used to make intelligible the topological positions and activities of individuals. This theory seems to offer another way of conceiving those phenomena which sociologists have usually treated in
(459) terms of groups or collectivities. There need be here no rivalry of concepts or methods, no jealous guarding of professional labels. Any method which promises illumination should be freely experimented with, without asking to what discipline it belongs. We should remember, of course, that the field-theory treatment will be oriented from the standpoint of a system rather than that of individuals in their own full range of interests, purposes, and traits of personality. This fact, however, does not in the least detract from the claim of Dr. Brown to have his position recognized as the basis for a true science. It only shows from how many different angles, and with what differences of interest and value, the problems of social psychology can be approached. As Dr. Brown concedes, the serviceability of his field-theory rests largely upon the devising of some new kind of units and methods of measurement appropriate to his topological concepts. He is probably justified in doubting the applicability to his problem of any psychological attitude measurements thus far developed. In some ways the dynamic of the field situation resembles the "continuum of effect" discussed by F. H. Allport. It is possible that further development of scales based on teleonomic units in terms of the whole grouping or system may give us something more relevant to the problem than the familiar attitude-scaling devices. If this should be the case, the field-theory and the project of observing the societal behaviors of individuals might be effectively combined for purposes of research.
No general symposium on social psychology, whether psychological or sociological in aspect, can come to a close without some treatment of the problem of culture. As the social scientists have shown, culture may be considered from a superorganic or super-individual point of view, and indices may be devised to show its trends, changes, lags, and cycles. Sociologists themselves would, however, be willing to admit that culture also has its significant psychological aspect. Even material culture is based in large part upon the habits of individuals who make or use the cultural equipment. Those who have been interested only in culture per se have been inclined to regard it as something which overlies and modifies human behavior in such a way that original and native human characteristics can scarcely be seen. From this view it is but a step to the acceptance of the complete culture determinism of the behavior of individuals. When carried to this point the culture theory becomes unserviceable, first, because of the metaphysical assumptions involved, and, second, because it permits no way of explaining or predicting exceptional behaviors, qualitative differences in cultural development and change, and the emergence of novel forms of culture-content. The theory that culture displaces and supersedes biological and psychological laws of human activity seems absurd on its face. In order to discover what part human beings really do play in forming and using their culture pattern, we must discover what the behavior of human beings might be if divested of all culture. We cannot learn this directly because there are no human tribes without a culture of some sort. Recourse must therefore be had to a comparative study using as control groups the lower animals, particularly those most closely related to men.
Such is the theme which is ably presented by Dr. Maslow. Through reference to his own experiments he has indicated how the behavior involved in the biological relationships of the lower animals may be seen in its native form, unmodified by the overlay of human cultural
(460) habits and institutions. Comparisons dealing with similar biological relationships of human beings can thus be traced. This problem is the same as that which has been raised, in a different form, by F. H. Allport's observations of societal behaviors. Once we have discovered the pattern of the range and frequencies of noncultural human activities as they might be performed in a society practically without transmitted culture, we can then apply this list as a criterion to the actual behaviors which we see individuals performing in the institutional and organized life of modern society. In this way we may be able not only to describe culture itself in biological and psychological terms, but to get at the very bottom of the sociologists' problem of the maladjustment between human nature and culture.
In two other papers of the symposium, those of Dr. Cantril and Dr. Robinson, the psychological implications of culture are developed further. Dr. Cantril presents the relation of modern Technology and social organization to the social behavior of individuals, while Dr. Robinson discusses the societal standards of behavior in various vocations and professions. The idea of social norms is here brought to the front. Dr. Cantril's paper implies a degree . of cultural determinism, in that he stresses social and cultural situations in their influence upon the individual. Modern techniques of printing, for example, carry with them a direct prestige suggestion and impression of universality. Modern devices of communication, such as the cinema and the radio, effect behavior immediately in the direction of norms promulgated through these media. If we could translate these norms and the people promulgating and behaving according to them into terms of a field of tensions, the problem might be merged with that of Dr. Brown's field-theory. Or again, we may remember that all these cultural standards are developed by and "impinge upon" human organisms who have fundamentally those types of biological, non-cultural behavior which Dr. Maslow is trying to discover, and who differ from one another as individuals in the manner indicated by Dr. Miles, these differences playing a part, as suggested by F. H. Allport, in the slope of the J-curve of conforming behavior.
We have here an interesting contrast in orientation. Dr. Cantril seems to view the problem from an introspective aspect in which social organization and cultural norms seem to become a reality immediately accepted by the individual and operative in modifying his behavior. Viewed from this standpoint, societal norms indeed seem to be determining agents. In order to establish them as realities, however, we must operate in a purely mental, or implicit, plane, and must accept the limitations of the introspective, or mentalistic school. To agree that individuals accept their culture as a super-organic force or agency is by no means to prove that such a super-organic agency really exists. The question of existence, however, is a metaphysical one and need not concern us here. The point at issue really is a methodological one. If we are not satisfied with this subjectivistic reality of culture and wish to define cultural phenomena in an objective, explicit fashion, we can do so by substituting for conceptual social norms the measurements of actually observed behaviors. When we do this, the norm fades from our view as an entity, and in its place we have only a statistical distribution of behaviors, with a steep mode and abruptly descending slopes. This is the position which F. H. Allport has taken in his "observation of societal behaviors." These considerations, however, by no
(461) means discredit Dr. Cantril's view. The approach which he suggests may be as practicable as any other. Our task here is not to pass judgment upon the value of any particular formulation, but to suggest the range and variety of methods which are open for us to follow.
Throughout all this interesting array of methodologies, there remains one simple, clear-cut problem-situation, which may well be regarded as the central task of social psychology. From a historical standpoint, it is the one methodology which has offered from the very beginning a clear possibility of an experimentally controlled approach. In this respect it offers perhaps our best hope of bringing social psychology within the domain of the more exact sciences. This problem raises the direct question-in what way is the behavior or achievement of an individual working or acting in the presence of other individuals different from his behavior or achievement in comparable activities performed alone? Some of the numerous possible ramifications of this problem-situation are pointed out by Dr. Dashiell; and his paper is a forceful challenge to us to return to some of these fundamental but neglected researches.
Coming back to the definition suggested as our starting point, it will be recalled that social psychology may be defined as the science which deals with the behavior of individuals stimulating or acting in response to other individuals, or in social situations. The methods of observing such behavior have been treated in the papers by Dr. Goodenough and F. H. Allport. Consideration has been given to the kinds or the content of the behaviors we are studying by Britt, Katz, F. H. Allport, and Maslow. The fact that individuals differ in their social behavior, and that they may be regarded as personalities acting in social situations, has been discussed by Miles and Murphy. The importance of the presence and behavior of others, that is, of the environing social situation in which the individual acts, has been well stressed in the contributions of Murphy, Brown, Cantril, and Robinson. And finally the logic of the whole process, that is, the setting up of experiments actually to determine the differential of the individual's behavior as between solitary and social situations, has been set forth by Dashiell.
If one were to ask what basic trends, if any, appear from a survey of these varying social psychologies of 1936, attention might be directed to two dominant and contrasting points of view. There is, first, the particularistic approach, in which specific factors are studied and an attempt made to correlate them and to establish generalizations whereby we can predict one of them in terms of the other. For example, we may refer to the connection of certain attitudes in individuals with other attitudes, as mentioned in Dr. Katz's paper, or to the behavior of individuals working at a mental task together as compared with their behavior alone (Dashiell). Or again we might note the incidence of certain types of behavior as performed by certain individuals or in certain social situations (Goodenough). This is perhaps the' general view of science in which the search is for a law, and a law is regarded as a statement of some predictable relationship between two events or things. The other viewpoint is one which does not try to relate two particulars or establish laws and predictions for certain types of happenings, but attempts, instead, to achieve the conceptualization of a system which will most economically relate all the elements of a cultural or societal field. Brown, Cantril, Robinson, and Murphy are exponents of this school. Coherence within the system, and deduction
(462) and verification of the particular elements, rather than lawful prediction of particulars, are the goals of such a science.
This dichotomy of viewpoint is an old one which may be found to run through various other sciences. It is, for example, the controversy between the particularism of structural and behavior psychologies on the one hand and the configurationism of Gestalt psychology upon the other. In physiology and physics there is also the possibility of treating phenomena either as a closed, conceptual system or through the experimental isolation and relating of particular variables. Perhaps for an explanation of these two contrasting predilections in science, we must go back to the ultimate purpose or values of the investigator, or to some little understood differences of specific temperament.
One is also tempted to regard this dichotomy, as it newly emerges in social psychology, as a modern restatement of the old controversy between the individual and the group. If this is true, the sociologists' position in the future may be identified with the systematic or conceptualized approach. Instead of talking in terms of group or institution, he may speak of a social field. This reduction of groups to a system of inter-relations of measurable units may turn out to be a methodological improvement over the non-measurable collective entities previously employed. Whether this new approach, however, would be adequate for all the phenomena in which sociologists are interested is a question which we cannot yet decide. If this contrast between science as the relationship of particulars and science as conceptualization should replace the old controversy of individual and group, it would be a happy outcome; for then various persistent and insoluble metaphysical problems, such as the comparative reality of group and individual, could be laid permanently aside. The problem would become one of methodology instead of metaphysics, and our conclusions would be based not merely upon philosophy, but upon the results of empirical investigation.
As to the relative merits of these two approaches, the particularistic and the conceptually systematic, time alone can tell. Our ultimate criterion must be the success with which they enable us to understand, to predict, and to adapt to human purposes, the behaviors of individuals in an environment of their fellow men.