The Social Psychology of Groups

John W. Thibaut

• Publisher: Routledge
• Year: 2017
• ISBN:
  9781351473880
• Edition: 1
• Topic: Psychology
• Subtopic: History & Theory in Psychology
• Index: Psychology

This landmark theory of interpersonal relations and group functioning argues that the starting point for understanding social behavior is the analysis of dyadic interdependence. Such an analysis portrays the ways in which the separate and joint actions of two persons affect the quality of their lives and the survival of their relationship. The authors focus on patterns of interdependence, and on the assumption that these patterns play an important causal role in the processes, roles, and norms of relationships. This powerful theory has many applications in all the social sciences, including the study of social and moral norms; close-pair relationships; conflicts of interest and cognitive disputes; social orientations; the social evolution of economic prosperity and leadership in groups; and personal relationships.

1.

Introduction

In this book is presented a theory of interpersonal relations and group functioning. The major motivation that governed this effort was a desire to give cumulative treatment to a discussion of some persistent problems in social psychology and to answer for ourselves the question: how do the data of the field look when they are arranged by their relevance to a conceptual structure that begins with relatively simple assumptions and adds further ones only as they become necessary? If it is not entirely idiosyncratic, any success along these lines might be expected to be useful both as a guide to research and in contributing some order and simplification to an increasingly bewildering congeries of fact.

What kind of theory should one construct for this purpose? From the very first books with social psychology explicitly in the title, quite different answers to this question have been suggested. The book by the psychologist William McDougall (1908) was mainly an attempt to identify and classify the various social motives. In the same year the sociologist Edward A. Ross initiated a contrasting tradition for social psychology by defining its distinctive mission as the analysis of social interaction. In the years since 1908 many writers on social psychology have implicitly accepted the injunction of Ross to concern themselves with social interaction. To mention merely a few of these, we might point to Dashiell's (1935) comments on the social situation as a “reciprocal affair” in which each person is “stimulable and reactive,” Cottrell's (1942) reinforcement analysis of the “interact pattern,” Lewin's (1947) interpretation of dyadic interaction as a “three-stage process,” Bales's (1950) extensive discussions of interaction in small decision-making groups, and Sears's (1951) S-R references to “double contingency” in the “social episode.” It is in the tradition of social psychology represented by these works that the present book is written.

Some of the implications of Ross's injunction, to begin with social interaction, may be seen by considering the distinctive aspects of social interaction that are revealed when it is compared with the simpler situation with which individual psychology has traditionally dealt. In the typical experiment in psychology the subject is in some manner under the management of the experimenter, who controls the presentation of stimuli, the opportunities for behavior available to the subject, and, most importantly, the provision of diverse incentives and rewards for behavior. In general, by these various means, the experimenter exerts control over the behavior of the subject, and the procedures by which he does so constitute the independent variables of the experiment. The behaviors that the subject actually emits constitute, of course, the dependent variables.

In these experiments it is usually assumed that the subject has no counter-control over the experimenter. What the subject does makes no difference to the investigator (at least if he is thoroughly in his scientific role of impartial seeker of truth), and even if it does make a difference it cannot be permitted to cause him to deviate from his prearranged schedule of activities. This assumption must be valid if the two sets of variables are to have the status of independent and dependent, in which the former can be viewed as antecedent to the latter in causal efficacy.

The situation is sharply different when social interaction is considered. The simplest situation is that in which two subjects interact in response to a task set by the experimenter. The possibility is now introduced that each subject will exercise control over the other. We attempt to develop some of the substantive implications of this kind of reciprocal control in the subsequent chapters of the book. We now merely note that methodologically the complexity that is added by reciprocal control may be denoted by the loss of a clear separation between independent and dependent variables. Each subject's behavior is at the same time a response to a past behavior of the other and a stimulus to a future behavior of the other; each behavior is in part dependent variable and in part independent variable; in no clear sense is it properly either of them.

If it is true that free social interaction leads to an ambiguity about what is dependent and what is independent, and since social psychology is traditionally committed to the use of experimental methods as at least an important part of its technical repertoire, how has investigation proceeded? The answer seems to be that the problem has been largely bypassed. This will be seen more clearly in a brief description of some of the main types of social-psychological research. In all of these some degree of experimental control has been maintained.

One of the most commonly used procedures in social psychology involves the presentation to an individual subject of a social stimulus fixed by the experimenter. For example, the subject may be required to make judgments of emotions from photographs of facial expressions, to respond to the various items of an attitude questionnaire, or to indicate his most likely response to (verbal descriptions of) hypothetical situations. Similar in principle is the use of a “live” or a tape-recorded stimulus person in research on social perception and of preinstructed experimental accomplices in certain types of experiments.

Social research of the sort outlined in the preceding paragraph need not be different, at least in degree of experimental control, from standard experimental work with individual subjects. Social interaction per se is not studied and the experimenter controls the task presented to the (individual) subjects. There does exist, though, a type of research in which social interaction clearly occurs but is not the explicit or main object of study. These are investigations in which a group or collectivity of persons replaces the individual subject and which focus on the effects of variations in aspects of the group's environment (e.g., the task or external threat) or properties of the group (e.g., leadership style or composition of membership) as they bear on the performance (or morale) of the group. These studies are basically similar to those with individual subjects described above, except that some processes exist in the group studies (social interaction) which do not exist in the individual studies. The interaction itself is not usually studied. When he is primarily interested in discovering predictable regularities between properties of the group's environment (or of the group itself) and rates of group perfomance, the investigator may choose to ignore (or to describe only casually in passing) the processes of social interaction that intervene between the independent and dependent variables. The investigator's choice may be very wise if he can safely assume that the intervening processes will contribute only a random variability that is small in relation to the potency of his independent variable and the sensitivity of his dependent one. That so many studies of this sort have been successful in discovering these over-all regularities and uniformities is surely some evidence that this assumption is frequently valid. Nevertheless, social interaction is a worthy object of study in its own right. In the chapters which follow we attempt to develop an approach to the study of interaction that permits at the same time a consideration of the more traditional research of social psychology.

The approach to be proposed takes as its independent variables the possibilities for reciprocal control possessed by the members of a collectivity. Only that control mediated by the ability to affect another person's outcomes (such as rewards, payoffs, reinforcements, and utilities) is considered, although for reasons given in Chapter 2 this limitation is not felt to be very restrictive. Thus the analysis begins with a description of the way in which two or more individuals are interdependent in achieving favorable outcomes. The dependent variables are the various aspects of a relationship that can be viewed as outgrowths of the particular pattern of interdependence present there. These include properties of the interaction process, norms, and roles.

In other words, we start with a description of the problem that confronts a number of people if a continuing relationship among them is to be viable. This problem is defined in terms of the possibilities the participants have for contributing to or detracting from each other's outcomes. Our theory, then, is in large part a set of hypotheses about the important dimensions of this problem—that is, about the significant variations in the patterning of their interdependency. The description of the relations of objective interdependency that characterize a designated set of persons dominates the discussion throughout the book and gives it whatever unity and cumulative character it may have.

Once their objective interdependency has been described, it becomes possible to determine what the members of the collectivity should do if their relationship is to be viable, stable, and optimally satisfactory. But the actual course of the interaction and its products can be congruent with the expectations the theory sets for them only if it is assumed that the actors have accurate understanding of their problems of interdependency and have the necessary insight into the required social solutions or the opportunity to develop such solutions by some learning process or transfer of training from similar situations. In dealing with this transition from the objective interdependency to interactive behavior, our interest is clearly psychological. Just as the student of perception concerns himself with discrepancies between objective phenomena and subjective representations of them, our focus is upon the perception and understanding of objective interdependency and factors affecting the veridicality of these perceptions. But let it be emphasized that these problems are meaningful only if means exist of describing the “stimulus”—the objective structure of interdependent relations. Distortions and inaccuracies of cognition are revealed by comparing actual social processes with those predicted from the objective potentialities mentioned above.

Rather simple psychological assumptions are made to account for discrepancies between the objective situation and the actual social behavior (or, for that matter, for correspondence between these two). We accept as a basic premise that most socially significant behavior will not be repeated unless it is reinforced, rewarded in some way. No stand has been taken as to the range of events that may constitute rewards for the individual. This we believe is a matter for empirical research. Some of the events reported to have reward value in studies of interpersonal relations are discussed in Chapter 3. At times we have been tempted to invoke certain special social motives (e.g., a need for power or status) to explain specific phenomena. That we resisted this temptation is not to deny the existence of such motives nor to deny the importance for social psychology of a careful analysis of them (in the manner of McClelland, et al, 1953, for example). However, since the list of such motives is still an open-ended matter and an indefinite number of plausible ones may be added, the appeal to a motive to explain any given social phenomenon seems both too “easy” and too unparsimonious. A genetic learning approach seems rather unsatisfactory at present for the same reasons. As long as the early conditions of learning are not known in complete detail, one may too easily refer to plausible learned drives or acquired reinforcements as the need arises.

The theory that we propose seems on balance to be primarily a functionalistic one. The central concern is with the solutions that must be found to problems created by interdependency. These solutions are implicitly evaluated against standards set by the conditions necessary for group viability. Thus the focus is first of all upon what is functional from the point of view of the group or social relationship. However, an equally strong interest in the adaptations of the individual follows immediately. Because the existence of the group is based solely upon the participation and satisfaction of the individuals comprising it, the group functionalism becomes an individual functionalism. The ultimate analysis then is in terms of the vicissitudes of individuals as they try out various adaptations to the problems confronting them. The adaptations of interest to social psychology are necessarily carried on jointly because the common problems of interdependency impinge upon all of the members, though in somewhat different ways. The adjustment one individual makes affects the adjustments the others must make, which in turn require readjustment of the first, and so on. Nevertheless, whatever the social processes by which individual adaptations are coordinated and reconciled, the implications of the adaptations for the collectivity are constantly in the forefront of the theoretical purview.

In the analysis that follows we begin with the two-person relationship, the dyad. We so begin in order first to attempt to understand the simplest of social phenomena by endeavoring to be as clear and as explicit as we can about the conditions necessary for the formation of a dyadic relationship and about the interpersonal relations manifested there. Our bias on this point is apparent: we assume that if we can achieve a clear understanding of the dyad we can subsequently extend our understanding to encompass the problems of larger and more complex social relationships. Hence we devote what may appear to some readers as a disproportionate amount of space to the problem of the functioning of the dyad: most of the nine chapters that make up Part I of the book. The basic concepts that we use throughout the book are introduced, defined, and illustrated in our analysis of the dyad. It is our conviction that these concepts have general applicability beyond the dyad. We attempt later on to document this conviction in our discussion of complex relationships in Part II.

PART I

Dyadic Relationships

2.

Analysis and Concepts

In this chapter are presented the basic elements of a theory of small groups. Since these elements are used in analyzing and developing the topics covered in all later chapters, an understanding of the present chapter is a prerequisite for reading the later ones.

The building of any theory of social behavior requires making a great many simplifying assumptions. The phenomena considered are entirely too complex to be dealt with in their raw form. Many of the nuances and variations in social behavior must be disregarded if the human observer with all of his limitations is to detect any recurrent patterns. We have tried to be as explicit as possible in introducing our assumptions. However, in this very explicitness it may strike the reader that we have overly simplified complex and important matters. He is likely to find that we have ignored or diminished the importance of events or principles that he has always regarded as the essence of social interaction. To this it can only be said that our simplifications have led to a formulation that we have found to be useful in understanding and ordering the subject matter of group psychology. To evaluate the validity of this testimony, the reader must be willing to endure the hardships of learning the concepts presented in this chapter and to assess for himself whether any clarification is thereby brought to the subsequent topics. We would be the last to deny that other sets of assumptions and concepts are valuable in group psychology, nor would we deny that other and more ingenious approaches could be imagined. To put it simply, we could not find the time to consider what the list of sufficiently plausible theoretical systems might include nor even to evaluate thoroughly the alternative systems that have presented themselves to us at the various points in this undertaking.

Our conceptualization of the dyad, a two-person relationship, begins with an analysis of interaction and of its consequences for the two individuals concerned. The major analytic technique used throughout the book is a matrix formed by taking account of all the behaviors the two individuals might enact together. Each cell in this matrix represents one of the possible parts of the interaction between the two and summarizes the consequences for each person of that possible event. Although consequences can be analyzed and measured in many ways, we have found it desirable to distinguish positive components (rewards) from negative components (costs). The many factors affecting the rewards and costs associated with each portion of the matrix are described and note is taken of certain sequential effects that are not handled systematically in the present scheme.

The actual course of interaction between two individuals is viewed as only partially predictable from the matrix. Initial interactions in a forming relationship are viewed as explorations which sample only a few of the many possibilities. Interaction is continued only if the experienced consequences are found to meet the standards of acceptability that both individuals develop by virtue of their experience with other relationships. Several such standards that an individual may apply are identified, and these are related to such phenomena as attraction, dependence, and status. These concepts then set the stage for the more intensive analysis carried forward in subsequent chapters.

To clarify the ways in which the matrix of outcomes may be used in the prediction of behavior, we conclude the chapter by discussing in some detail the conditions under which it may be so used and anticipating briefly some of the further applications of the matrix to proble...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright Page
• Introduction to the Transaction Edition
• Preface
• Acknowledgments
• Table of Contents
• Chapter 1. Introduction
• Part 1. Dyadic Relationships
• Part 2. Complex Relationships
• Bibliography
• Name Index
• Subject Index