THE U N I VE RS I TY OF MI C HI GA N COLLEGE OF LITERATURE, SCIENCE, AND THE ARTS Department of Psychology STRATEGIES OF DOMINANCE AND SOCIAL POWER Prepared by Richard Schmuck ORA Project 06340 Proceedings of a symposium sponsored by THE HARRY FRANK GUGGENHEIM FOUNDATION NEW YORK, NEW YORK Administered through OFFICE OF RESEARCH ADMINISTRATION ANN ARBOR January 1965

PREFACE This report summarizes the proceedings of a symposium on'Strategies of Dominance and Social Power." The symposium was supported by the Harry Frank Guggenheim Foundation, and reflects the interest of Mr. Guggenheim in stimulating studies of dominance as a source of motivation and as an important factor in man's relation to man. It was held at the Fair Lane Conference Center, Dearborn Campus, of The University of Michigan, on May 15-18, 1964. The ideas for the topic and the symposium originated with Mr. Guggenheim during conversations with several close associates, especially Dr. Henry Allen Moe, Mr. James Doolittle, and Mr. Charles Lindberg. Early in their deliberations, Dr. Paul Fitts, Professor of Psychology, The University of Michigan was invited to join this group, and Dr, Fitts later was asked to direct the symposium. He brought the preliminary ideas of Mro Guggenheim and his associates to The University of Michigan where a steering committee was formed made up of the following social scientists: Dr. John Atkinson, Professor of Psychology Dr. Dorwin Cartwright, Professor of Psychology Dr. Abraham Kaplan, Professor of Philosophy Dr. Rensis Likert, Professor of Psychology and Sociology Dr. Theodore Newcomb, Professor of Psychology and Sociology Dr. Anatol Rapoport, Professor of Mathematical Biology This group of scholars lead by Dr. Fitts developed several symposium aims which can be summarized as follows: (1) To stimulate interest in an important but relatively neglected topic, that of dominance; (2) to facilitate an exchange of ideas on this topic among a small group of leading scientists and scholars from various locations and disciplines; and (3) to explore freely ideas that might catalyze scientific work on this topic. Next the steering committee proposed that the symposium discussions utilize four means at least for accomplishing these aims: (1) Elaboration and specification of problems and issues in dominance and power; (2) listing of exciting ideas on the topic; ii

(5) elaboration of some research needs and strategies, and (4) some plans for future activities. After developing these symposium aims and procedures, The University of Michigan group invited several distinguished social scientists to take part. The following accepted the invitation and participated in the conference: Dr. Robert Dahl, Professor of Political Science, Yale University Dr. Karl Deutsch, Professor of Political Science, Yale University Dr. Susumu Kobe, Professor of Economics, Waseda University, Tokyo Dr. Judd Marmor (M.D.), Psychoanalyst, Beverly Hills, California Dr. Henry Allen Moe, President, New York State Historical Association Dr. David McClelland, Professor of Psychology, Harvard University Dr. Fritz Redl, Professor of Social Work, Wayne State University These scholars were joined by the steering committee and Mr. James Doolittle, Mr. Harry Frank Guggenheim, Dr. G. Edward Pendray, and Mr. Dana Draper during the symposium. The symposium had two main types of work sessions, general meetings and small group discussions. For the latter, the participants were divided into three groups composed in the following ways: Group I-Dahl, Fitts, Guggenheim, Kobe, Newcomb, and Redl; Group II-Atkinson, Deutsch, Kaplan, Likert, and Marmor; and Group III-Cartwright, McClelland, Moe, Pendray, and Rapoport. Mrs. Nancy Davis served as secretary with the assistance of Mrs. Jane Riley. I performed as symposium coordinator and recorder. Richard Schmuck Assistant Professor of Psychology The University of Michigan iii

TABLE OF CONTENTS Section Page I, DOMINANCE AND POWER: APPROACHES, FINDINGS, AND ISSUES.1 II. POWER MOTIVATION 23 III. METHODS OF STUDYING POWER MOTIVATION 32 IV. SOME PSYCHODYNAMICS OF DOMINANCE AND POWER 38 V. DOMINANCE AND POWER IN LARGE-SCALE ORGANIZATIONS 43 VI. COMMUNITY POWER 45 VII. ECONOMIC DEVELOPMENT AND POWER ORIENTATIONS 50 VIII. DOMINANCE, POWER, AND THE SOCIAL SYSTEM 52 IX. NEEDED RESEARCH ON DOMINANCE AND POWER 54 X. FUTURE ACTIVITIES, CONCLUSIONS, AND SUMMARY 60 iv

I. DOMINANCE AND POWER: APPROACHES, FINDINGS, AND ISSUES The participants started discussion in the three small groups with the task of listing some fairly well established findings on dominance and power. GROUP I Redl was the first to offer some comments in Group I, He said that although he was unable to think of any well established propositions on dominance and power, he did wish to raise some questions which are both important and unanswered by researcho The first of these had to do with the sources or mainsprings of powerful behavior in individuals. He contrasted two positionso The first was that powerful behavior emanates from an individual's motivations, frustrations, and skillso An interpretation from this point of view is that a man such as Hitler was able to dominate others because of his own great striving for power and his adroit skills in obtaining it. The second, and contrasting, point of view was that dominating behavior on the part of an individual occurs because of the commonality of frustrations in the group over which he has power. From this orientation, Hitler was seen as taking advantage of a situation, a group condition, rather than achieving domination solely through his own motivation and skill. Redl further commented that typically we are dealing with both of these at the same time. Both situational properties and personal aspects of the individual combine in the resultant domination, Furthermore, since all human behavior occurs in group settings, it is impossible to argue that individual motivations and skills alone can account for that individual's domination over others. In short, power or domination stems from the commonality of needs and unconscious wishes in the group as well as the motivations and skills of the powerful persono Redl completed his initial statements by pointing out that the term'power" has many meanings-when we use it we are often referring to a wide variety of processes-and therefore we should work toward more precise definitions during the Conference. By way of trying to list some well established propositions on dominance and power, Newcomb briefly reviewed the animal research of SchjedlerupEbbe and Murchison, These researchers discovered a "pecking order" in groups of barnyard fowls, i.e., when chickens are placed together they establish an order of dominance or a power structure. This ordering is reflected in the eating and sexual behavior of the animalso Murchison found that the important correlates of dominance among chickens were sheer size and weights Since size and weight are correlated with sex, males are bigger than females, males often tend to be more domineeringo Newcomb added that 1

perfect transitivity of power, though it would be expected if there were a complete pecking order, seldom occurred in most collectivities observed, i.e., though A could dominate B and B could dominate C, sometimes C could dominate A. Murchison argued, however, that such nontransitivity was due to chance habit-forming encounters and to some extent were matters of simple conditioning. Newcomb raised the issue of whether or not these wellestablished findings on animals apply in any way to human beings. His conclusion was that they probably do not apply in anything like so simple a fashion, but that fairly close analogies might eventually be discovered. Kobe pointed out that monkeys exhibit dominance by strutting about in front of other monkeys with their tails standing up. He added that dominance orders among monkeys, like chickens, also are very stable. Furthermore, monkeys have a system of ostracism-rejecting some of their species who have to wait until all the other monkeys finish doing something before they dare begin. Redl wondered about how the animal studies cited by Newcomb and Kobe might apply to human beings. He has been interested particularly in dominance orderings in children's groups. Furthermore, he has been fascinated by individual differences among young boys in being able to perceive accurately the dominance ordering in their gang. Redl asked that some research be done to find out why some boys (and perhaps some animals) are more accurate than others in perceiving power hierarchies. Redl contributed another area of personal interest-an area which he called the "spill-over" effect of dominance. About this, he asked, "What does it take for power to be accepted in areas other than the one in which it was originally gained?" In other words, how do we predict which people will have generalized prestige and which will have more specific prestige? From the point of view of social problems the generalized dominance is probably the more dangerous type. We tend to be fearful of a man who commands dominance over many people across various situations. Newcomb differentiated between dominance and leadership. He defined leadership as a relationship between group members which is facilitative of a group's moving toward a goal. Dominance, on the other hand, may be opposed to leadership in this sense. The dominant person is more concerned about his own power position in a group than his facilitating the group's movement toward a goal. Fitts added that from his point of view, the nature of a situation will determine, in part, the similarity between dominance and leadership. He briefly reviewed some studies on leadership in small bomber crews. Fitts pointed out, for instance, that on very dangerous missions the crews' reference was for a leader with great technical competence, regardless of his usual dominance. 2

Redl asked what people do when testing how much power they have. He pointed out that the monkey does it by holding his tail up high, and that other animals do it by viciously protecting their territories. Young boys have also been known to guard their rooms (their territories) with considerable tenacity. Two other questions raised by Redl were: (1) What are some rituals, across cultures, related to power?, and (2) What are some physical symptoms indicative of power relations? Dahl said that he has concluded from his research that there are an extraordinarily large number of conceptualizations of power in the literature. He has decided that the researcher interested in social power should not start with the concept-but instead should approach this topic with three strategies in mind: (1) First, the resea-cher should explore ways of giving operational meaning to the concept of social power. He has tried to do this using both historical and ahistorical methods. Dahl evaluated this strategy as being somewhat successful. (2) Secondly, the social scientist might focus on a small subset of types of power. Despite the frequent usage of such terms as legitimacy, authority, persuasion, coercion, and charisma, we have not gone very far in studying them. (3) Finally, after empiricalstudies and small scale analyses, rigorous attempts to provide theoretical models are needed. Today, we have too many definitions of power and not enough hypotheses and theories. Fitts asked Dahl if it will be possible to develop a theory of power which will be applicable across individual and group levels of analysis. Dahl differentiated between the short-run and the long-run views. If ten years is the short-run, then general theory will not be very useful. For the next ten years it will be better for social scientists to attack the problems surrounding dominance and power from their own special perspectiveso Over the long-run, during the next fifty years, it will be better to try to build a more comprehensive theory, building up from these circumscribed analyses. Perhaps the most important thing to be done immediately is to work on research methodology. All social scientists are agreed that they must move much farther toward the perfection of their research methods. Dahl defined power as an interpersonal relationship-a relationship between A. and B in which the actions of A. are contingent upon some actions of B. Using this kind of definition leads to some examples which might not seem appropriate within the rubric of social power. For instance, a baby has a willing mother under his control in the sense that the mother's desires to nurse and care for the child can only be satisfied by the needs 5

and satisfactions of the child. In this sense, some actions of the mother are contingent upon the actions of the baby. Newcomb agreed with this definition of power within interpersonal terms. Dahl described his theory of power as involving three major concepts: resources, motivation, and skills. He commented that his previous research has shown that the resources and skills are more often in the hands of whitecollar workers rather than those of blue collar employees. The topic of skills in gaining power stimulated an inquiry from Redl concerning the nature of charisma. Dahl said that charisma has not been studied systematically by political scientists, Redl indicated that some youngsters, just by charm alone, can get others to do something for them. Charisma is a frequent occurrence among both children and adults, but he also agreed that little is known about it. Guggenheim asked the group if all leaders or men of power have the same personality characteristics? The participants seemed to agree that leadership and power depends a great deal on the social situationo For instance, Kobe reported that leaders in Southeast Asia must come from highstatus families, have a western education, and abundant skill in meeting and entertaining westerners. Leaders in Communist countries, on the other hand, are people with little wealth and education. They indeed may start revolutions out of a motivation for power. It was added here that different types of people emerge in labor unions as the unions develop. In the early stages of growth, union leaders are agitational personality types, while as the union matures, more bureaucratic types emerge as leaders.o Fitts commented that Likert is finding a similar pattern of changing leadership and power in businesso Dahl said that perhaps the most popular theory concerning the development of power motivation comes from Lasswell and Adler. The theory states that a desire for power arises because of early deprivations of love and affection and a resulting insecurity. Redl warned against assuming a simple and sovereign theory about power motivation. Newcomb added that power behavior can begin as an instrumental activity and later become a consummatory one, a motive in its own right. Fitts felt that we might learn much about power from studying children's play and slavery. Kobe added another, pointing to the geographical or ecological determinants of power motivation, e.go, hunters are more power motivated than horticulturists. Redl asked Dahl how the political scientist would explain why some paranoids gain power and why most do not. Dahl answered by making reference to the theory of the mass society popularized by De Tocqueville and recently formalized by Kornhausero This theory states that people become more amen4t

able to being swayed by a strong leader and national symbols when they cease to participate in occupational and community associationso On the other hand, if people share norms which are anchored in the community, they probably will not be swayed easily by the paranoid leader. He gave, as an example, the way in which the Senate finally toned down Senator Joseph McCarthy because he was doing so much harm to the institution. Elite groups such as the American Senate generally deter mass movementso Fitts added that the dictator usually attempts to get rid of such elite groups if he can get control of the military. Dahl saw two developmental stages here: The legitimacy and self-confidence of the elites is undermined, and instruments of coercion are gleaned from the state. Finally, Redl felt that identification with the aggressor might be one of the mechanisms involved in dictatorships, especially those which utilize a great deal of terror. GROUP II Atkinson suggested that each participant discuss his own approach to the topic of social power. He commenced by pointing out that his approach to the topic is from the point of view of individual motivation, and that one necessary criterion of a topic is its feasibility for empirical studies. Atkinson felt that the topic of social power is quite amenable to empirical study. He went on to say that he currently is puzzled by several issues in the study of dominance and power. The first is that power is a "dirty word"1 It is analogous to sex during the Victorian age. Perhaps, as Freud did with sex, we will have to look for devious and disguised expressions of power. Because power needs and wishes are so often repressed, we will have to hunt for their indirect and unconscious expressionso The second issue which is striking is that power is a relationship between two or more individualso Therefore, we have to be concerned about both ends of the relationship; the dominator and the dominated. Finally, Atkinson said that answers to problems involved in studying power will not be forthcoming until we upgrade our methods considerably. Likert supported Atkinson's emphasis on dominance as a social relationship, adding that we must study power as it exists in organizations and large systems. Indeed, these larger systems are qualitatively different from two-person relationships. In the studies of power in industrial organizations, for instance, a man-to-man model has been most popular until recently. This model calls for organizational studies in which all procedures are understood in terms of individuas relating to inedividuals The facts are that the highest producing managers in the nation now use the face-to-face group as their building block instead of the individual. Skilled managers wield power by capitalizing on the effective powers of small-group interaction. In fact, in present day society, power increases 5

as it is used with face-to-face groups as the smallest units. Finally, - Likert emphasized how power resides in the social system, not in individuals. Marmor felt that the discussants were using the term "power" in a much too diffuse and broad manner. He pointed out how Likert used influence, dominance, control, and power interchangeably. Marmor felt these usages were not precise enough. He said that we should distinguish between two types of motivationally different power. The first of these emanates from constructive motivation-from adaptive tendencies-from an individual with positive feelings. The second type of power stems from hostility. This latter type Marmor called "power striving." Such individuals are driven by a need which stems primarily from their feelings of being threatened. Feelings of being threatened come from perceiving oneself either as inadequate and inferior, or as weak and easily overwhelmed. Such individuals develop a striving for power in order to protect themselves. Marmor pointed out that dominance, on the other hand, comes from inner strength. Individuals are delegated dominance by others because of superior strength, intelligence, and authority. At the group level we have examples of nations which are dominant, e.g,, the United States; and those who are power driven, e.g., China, Atkinson elaborated on Marmor's distinction between power and dominance. He said that the findings of a national survey, using TAT pictures to measure power motivation, indicated that high-power motivation exists at the top and the bottom of the occupational hierarchy. Managers and the deprived people, especially Negroes, scored the highest in power motivation. Perhaps the managers fit Marmor's dominance while the deprived are more power motivated. Marmor added that we should look at differences in power motivation both within and between societies. The Zuni, for instance, in contrast to most Americans, make a point of not standing out and of being meek. Kaplan said that he was interested in three topicsO The first of these was power motivation. The second involved strategies, tactics, and procedures of getting and using power. As a philosopher, Kaplan pointed out that he was interested in the adaptation of means to ends or, in other words, the nature of rational behavior. Thus, he was interested in how men of power wield their power. Finally, Kaplan said that he was interested in the effects of using power, and the value implications. He asked "What are the grounds for our value assessments in this area? How can we improve our appraisals? What can we do to modify motivations and strategies or both, so that we get the greatest good out of power?" Likert commented that he hoped that the social scientist might be able to help the philosophers' value judgments by strategically collecting data. Deutsch pointed out that he was concerned with the application of systems theory to the study of power. His main interests in research were 6

with international systems, classes, and interest groups. At the same time, he does enjoy moving across the board from the international system to the individual. He illustrated his interest by describing an eight-by-eight matrix. The rows of this matrix were as follows: The individual personality, the small face-to-face group (up to 10), the village or tribe (100-1000), the district or small town (1000-10,000), the province or big city (100,000 to 1 million), the nation state (1 million to 100 million), the large blocs (1 billion), and the world system (3 billion). If one makes similar columns, an eight-by-eight matrix with 64 cells results with the diagonals representing internal messages. The matrix is formed when one designates that the rows represent the initiation of power and the columns represent the recipients or destinations~ Deutsch suggested four definitions or perspectives of power performance: (1) The ability to make a difference on the environment, (2) the probability of fulfilling onets own program —one's own inner intention or goals; (3) the difference between the change imposed on the environment and the change imposed on the one who is wielding the power; and (4) the probability that the power is transferable in contrast to being focused and specific. Likert said that the more overlap there is among the parts of a social system, the greater the possibility of any one part influencing another part. In such overlapping situations, communication and reciprocal influences are facilitated. Kaplan posited a curvilinear relationship between the destructive use of power and overlapping systems in contrast to Likert's suggested monotonic relationship. He argued that examples such as civil wars and divorces seem to indicate that overlapping relationships, when severed, tend toward aggressiveness and bitter conflicto Likert argued that the Civil War was bitter because the two cultures did not overlap very much. They were never really engaged in mutual and collaborative problem solving. Furthermore, it is probably true that the more any of Deutsch's eight units struggle for power unsuccessfully, the more bitter the outcome. So Likert argued that although one gets a major explosion when an overlapping relationship is destroyed, the probability is very low that it will be destroyed, Deutsch pointed out that if incompatibilities exist between two systems, the more interaction you get the more conflict you generate. On the other hand, if compatibility is neutral to positive increased contact enriches the relationship even more. The more interactions and transactions Americans had with McCarthy, the more they disliked him. By way of summary, Atkinson and Kaplan indicated that several definitions of power had been presented, especially those of Marmor and Deutsch. 7

Furthermore, the eight-by-eight matrix presented by Deutsch described the possible actors being discussed and the four aspects of the topic, motivation, strategies, effects, and limitations, gave some direction. The assumption also was made that problems at each level in the eight-by-eight matrix are at least somewhat analogous. Assertions, in other words, should hold for families, organizations, and nation states. Kaplan added that morality and law really refer to two separate parts of the problem. While both can be included under the general heading of limitations of power, morality has more to do with delimiting the content of the effects, while laws speak to limitations on strategies or means. Gandhi was unique in that he attempted to gain power by working on both the content of the effects and the nature of the strategies-both fitting within religious and legal doctrines. Kaplan directed the group to a different topic by pointing out that he was interested in understanding the mechanisms whereby an infant with a grandiose sense of power becomes an adult with feelings of powerlessness. Marmor suggested that the theory of infantile omnipotence is dubious. However, it does appear that the child has fantasies of omnipotence about his parents and that such attitudes carry over into later life. Marmor said that powerlessness is especially true of adults with regard to their feelings of control over decision-making in a nation state. Likert said that most adults do not feel powerless. He said that studies at the Survey Research Center showed that political efficacy is correlated with the extent to which a person is realistic. Deutsch added that Dahl's studies of New Haven confirm Likert's results for local problems, but not for international issues. In Dahl's survey the citizens of New Haven felt effective in regard to local political issues but very powerless concerning issues of nuclear war. Marmor felt that in the big urban metropolis most citizens do not feel that they can be influential even on the local level. Perhaps in small towns, he allowed, people may feel more powerful. Kaplan suggested that people feel powerful in relation to the amount of time they are on the sending end of an important communication. Most people, most of the time, are recipients of messages. Therefore, an individualls frequency of being a source is positively related to his feelings of efficacy. Deutsch presented an alternative hypothesis, that the speed and appropriateness of a response are important in developing feelings of efficacy in the recipients Atkinson presented a conceptual model which had proved useful in studies of achievement motivation. In such research, cognitive expectations that an act will have successful consequences are very important. The theory, applied to power behavior, would state that a tendency to respond is a function of motivation for power "multiplied"i by the cognitive 8

expectation that an act will produce a power outcome multiplied by the incentive value of procuring power in this situationo Motives here refer to a class of outcomes, while incentives are elements in the set. Growing up with a sense of weakness should lead to low expectations of being able to produce an effect. This percept of powerlessness would have an important effect on an individual's power behavior. Marmor added that this sense of weakness plus a need for security leads the child to project large amounts of power to the father, and also to God. Such perceptions are carried out at the expense of growing up and feeling powerlesso Kaplan pointed out that often men do things because there is a very low expectation of success. Moralists, for instance, such as Niebuhr, enter into events knowing that they will not succeed merely out of principle and self-integrity. In some Southeast Asian countries some people vote Communist as a protest only when they know for sure that the Communists cannot win. These are political parties of expression in contrast to the more pragmatic parties of actiono Atkinson said that issues like these are not problematic for modern motivation theory. There are many reasons for doing something; the individual has many competing tendencies. Atkinson said that the studies on achievement motivation have indicated that expectations for success are inversely related to incentive. Thus, the easier a task is, the higher the expectation of success, but the lower the incentive value. Atkinson asked whether or not this inverse relationship is true of power motivation. Kaplan answered by saying that so far as the concept of cognitive expectation goes, he believed that the powermotivated person would work hardest when the expectation of success is neither high nor lowo It is the middle range where overt behavior increases. Likert felt that it was important to consider the significance of changes in the way power is being used today. People know more about power today; they can make more choices —they have more alternatives. We should study the bases of this societal change which has seen power change from coercion to persuasion and gradually to more and more reciprocal influenceo Likert's studies indicated that the modern leader can exercise more power in an organization as he manages power in an indirect wayo The more interaction and reciprocity of influence the leader can develop, the more effective the group will beo Kaplan suggested three relationships between cognitive expectations and incentives in answer to Atkinson: (1) The relationship is negative, i.e., when expectation is low, incentive is high and vice versa. This is characteristic of achievement motivation; (2) the relationship is positive, i.e., when expectation is low, incentive is low and vice versa. This is characteristic of aggression. For instance, the bully becomes more of a bully when he has a victim to threaten; and (3) the relationship does not exist, i.e., expectation and incentive are independent. Kaplan felt that this is 9

characteristic of the power relationship. Here the man is interested in a result regardless of his expectation. Deutsch pointed out that a bully is one who comes out of an earlier stage of a low expectation of power, and now glories in contemptible victories over the weak. Marmor agreed in that a striving for power develops as a compensation for threatened inadequacies. The bully attaches high value (incentive) to a perception of the restoration of high expectations of achieving power. Success reinforces further striving for power in that it raises the expectations of more power. These early frustrations and low expectations for power, characteristic of the bully, get into Atkinson's theory as unfulfilled or frustrated tendencies. Freud wrote about this more generically as the unfulfilled wish. Marmor added that one characteristic of a person with psychopathological power striving is the insatiable quality of this motivation. The Don Juan type has to seduce many women. He has to be reassured constantly. The greedy business man has to continue to make more and more money. Behind his behavior is a sense of insecurity. The tyrant must constantly exaggerate his power. These people have an initial self-image of weakness. Marmor went on to say that most people do not have insatiable wishes for power. He feels that the Hobbesean hypothesis that all men have an inexhaustible hunger for power is invalid. This view emanates from an instinctual position which claims that all men have a drive to achieve domination over others. Hans Morganthau appears to be a modern counterpart of Hobbes. Marmor's own psychiatric orientation is adaptational; he assumes that power striving exists in some people as a means of adjusting to their environment. Kaplan suggested that we differentiate between one's perception of power and one's power behavior. Furthermore, it will be important to study the components, bases, and results of unconscious power motivation. Marmor pointed out that unconscious power motivation is sometimes seen in the sick person who tyrannizes over others, or the stutterer who uses stuttering as a means to keep others hanging on every one of his words. Deutsch supported Marmor s conjectures that the insatiable power drive is not universal. He added that throughout history, certain kinds of social arrangements have led to more or less power motivation. In situations where there was a great deal of power striving, apologists saw power as instinctual and universal. Instinctual theories are often used to perpetuate and rationalize the status quo and lead to'a circular chain of illusions. 10

GROUP III Cartwright led off the discussion by expressing his own social psychological orientation. He said that he was interested in the interrelations between the individual personalities and social arrangements-between personality and social structure. He pointed out two kinds of specific interests in this regard. The first of these dealt with the effects of social arrangements on the motivation for power, e.g., family structure and dynamics as they influence individual drives for power. The second interest was the impact that different kinds of persons have on social. power arrangements. Cartwright presented an example of the importance of social structure which has emerged from studies of dominance structures in flocks of chickens. At about eight weeks, chickens establish a stable pecking ordero This dominance hierarchy is established through small dyadic battles which soon cease. After the hierarchy has stabilized, gestures are sufficient to keep lower order chickens subordinate and higher order chickens superordinateO Cartwright went on to say that biochemistry is associated with dominance in chickens. If one injects male hormones into a chicken with very little power and puts it in contact with another chicken outside the context of the flock, the former will dominate the latter. On the other hand, when the chicken given male hormone is kept in the same flock, one notices little differential effect. One conclusion is that the social system has inertial effects, that imitation, habit, and social relations are important in the dominance hierarchies of chickens. This research, according to Cartwright, highlights the coercive effects of social systems once they are established. Rapoport supported Cartwright's observation by describing some research which he did on pecking orders in chickenso His research showed that dominance hierarchies were established by social and not biological factors. After ferreting out most of the relevant variables, Rapoport's findings emphasized the importance of habit and imitation in the development and maintenance of pecking orders in chickenso McClelland suggested a generalization about power which might apply to both animals and humans. It was that one gets power as he controls more and more resources; among chickens this may be food and sex, among humans, money and armies. Rapoport pointed out that one must differentiate between resources to be consumed and resources to be directed. Often, the most power-driven men want to direct resources, but have little taste for lavish consumption. McClelland suggested that there are probably cross-cultural differences. on consumption power and the power of direction. For instance, the Arabs appear to be interested in consuming resources. Westerners perhaps are more interested in directing resourceso 11

Cartwright offered another distinction from. the formal theories of social powero He distinguished scope of power from the domain of power. Scope of power has to do with the range of activities that one controls. Domain, on the other hand, refers to the set of people over which one has powero McClelland said that according to studies being carried out by Rudin, the peak of power imagery in children's readers occurred in this country around 1910. Recently the concern for power has been going up again, especially since the Eisenhower administrationo Rndin theorized that a cycle exists in which achievement imagery reaches a peak in a culture first, power imagery follows, and finally affiliation imagery reaches a peako After the affiliation peak Rudin is not sure what occurs, but he does think that we are moving into a power period in the UoSoA. At the same time, imagery in the Soviet Union appears to be more achievement oriented in contrast McClelland explained that imagery from childrenTs textbooks is used for such studies because it is indicative of what the significant elites are thinking. Some of these elites are doing it-achieving, and using power-while others are writing about it. Pendray pointed out that in recent years business executives have found different ways of wielding power without coercion. Persuasion through public relations and the mass media has taken over from coercion. Power, in some senses, has been made more palatable but also more insidious. Cartwright added that while power imagery has been on the rise since 1950, this is the same period during which the democratizing of business leadership has occurred. Both the basis and the means of power have been shifting in our societyo We are in the period now of the engineering of consent. Power and social engineering, however, are "dirty words" in our cultureo McClelland added that power among humans is most effective when men's mind's are changed-not when coercive external controls are used. One basic question for further research raised by Cartwright had to do with the scope of power. He asked: "How general is the scope of power? What are the processes by which power spreads and by which it transfers to new activities?" Sometimes people use symbols to increase scope over a domaino This question also raises issues about legitimate scopeo How much power, for instance, should a corporation have over the employees? The discussion changed here from defining and theorizing about power, to the conditions in a culture which lead a person to have a strong need for power. Cartwright answered by saying that the culture determines in part the types of personalities who try to gain dominance, Among inmates in a custodial prison, for instance, the prisoners who gain dorminance have been in prison longer, have more years to serve, have been charged with more violent crime, and have a higher rate of recidivism than the other prisoners. These leaders, in effect, have a comrmitment to oppose societyo 12

They most represent the values of the prison inmates and therefore achieve dominance. in therapeutically oriented prisons, by comparison, the prisoners who become dominant are the most easily treated and curable prisoners. Cartwright concluded that we get different leaders emerging according to the overall management of the prison. One generalization derived from such studies is that leaders reflect, more than the members, the basic values of the group. McClelland said that one aspect of power and dominance, which we frequently forget, is that power is a relationship. Whenever we have a leader, we also have followers. There is very little in the social scientific literature which reflects this relational view of power. Cartwright agreed, indicating that a political scientist such as Dahl is mostly concerned with the person who is leading while he, a social psychologist, is interested mostly in the people who are being led. Pendray asked about people who are interested in power and domination. What motivates them? Why is a person power-hungry? McClelland said that the psychologists know much more about achievement motivation than about power. For instance, it is known that achievement motivation in males derives from three familial conditions: (1) Very high, yet reachable, standards of achievement set for the boy. (2) Considerable parental warmth and reward for achievement. (3) Low dominance on the part of the father, which allows the boy considerable autonomy. In such families, the standards of excellence are set just a little ahead of where the child is presently performing. Cartwright suggested that perhaps one thing parents do to encourage power motivation in their child is to say things such as: "Stand up for your rights —don't let people push you around.'" McClelland felt that power differs from achievement in that the winner takes all, while achieving tends to be cumulative. The power-driven person must constantly invent new games to make sure he still has power. Once he wins, he goes on to the next one. McClelland went on to say that people with great power motivation usually do not do well in authority positions. He added that one must differentiate between power for selfish reasons and power which is derived from an institutionalized authority position. One way of studying these differences, according to Cartwright, would be to study the motivational bases and effects of various mobility patterns in a business organization. Some men are motivated for interpersonal power, others for glory, and still others for achievement. Cartwright also pointed out that high authoritarianism is not the same as power motivation. The high-authoritarian person is interested in dominating people below and being submissive to people above, Rapoport added that perhaps power motivation is similar to a highly authoritarian person who must always be on top. 13

McClelland thought that the power relationship has more of an exchange quality, compared to achievement. When we buy something in America we asked "How much is it?" "What is its quality?" In the power-oriented culture, the shopper and the storekeeper are interested more in establishing a reciprocal influence relationship, giving and receiving from each other. In the Arab culture, for instance, where power is more important than achievement, one engages the storekeeper in a personal conversation. In the power relationship you give something and get something. The potlatch is an example of the power game, one gives up many resources to get respect, power, and prestige. Cartwright was not sure if McClelland was really describing the power relationship. To Cartwright, the Arab examples sounded as though individuals were trying to establish affiliative relationships. According to McClelland, controlling of another's behavior does occur as you establish a relationship with the othero Reports from these first discussions were presented to all of the members of the symposium. Newcomb began by reporting that Group I tried to find some well-established findings or principles in the area of dominance and power. He pointed out that the members of Group I spoke about the findings on pecking order among fowls and the functions performed by different people in small groupso Even though these two areas were discussed in some detail, Newcomb's group decided that little really was know and well-established about dominance and social powero Group I decided that there are several issues which need more study. The first had to do with the importance of methodologyo Group I agreed that social scientists must work quite diligently during the next few years to perfect their methods in studying problems such as the motivations behind power behavior. Secondly, they decided that children's play would be a fruitful area for research on power and dominanceo Dr. Fitts pointed out that Harlow's work on chimpanzees could be used as an example of the kind of work social scientists might do with human infants. Thi-rdly, Group I talked about using historical documents for studying power-for example, documents relating to slavery. Fourthly, they decided that charisma had not been studied adequately and should be researched. Newcomb finished reporting by presenting Dahl's reply to a question raised about the definition or the locus of power. Dahl said that power always has to do with a relationship between actors; a relationship in which the behavior of one actor is contingent on the behavior of another. Subsequent to Newcomb's summary Guggenheim asked the group if social scientists know whether there are inherent qualities of dominance in human beings. Guggenheim wanted to know whether there is an instinct or an innate tendency for humans to dominate over other humans. McClelland answered that there is little research which indicates that leadership traits carry over different situations. leaders, by and large, do not have similar traitso Cartwright pointed out. however, failing to find leadership traits which go 14~

across situations does not mean they do not exist. Likert supported Cartwright's statement by saying that in his experience many people in industry seem to be able to lead across situations, indicating to him, at least, that perhaps certain personality characteristics do characterize leaders. Pendray asked the group whether there were genetically inherited characteristics for dominance. The group generally agreed that there is little information to show that such traits even exist. Guggenheim asked, "Do people always try to find another who is lower than they? Is this a universal characteristic of people?" Dahl rephrased the question by asking, "Is it true that there is a constant motivation to dominate or be dominated in human beings?" Marmor answered by saying that, in his opinion and experience, every person does not have a need to dominate. Furthermore, he added, we should differentiate between dominance and the need for power. Marmor said that dominance is an established relationship or state between two people in which one person is stronger, more intelligent, more adequately equipped, or has more authority than the other. Power, on the other hand, is similar to domination, and is measured by the number of times the person is thinking about controlling the behavior of others. According to Marmor, it is this second type-power-striving-which he'thinks the Conference should be discussing. Power-striving is rare, according'to Marmor, although history has shown that it can do a great deal of damage. Deutsch underlined Marmor's point, by emphasizing that most dominance relationships are learned through simple conditioning. The dominance hierarchy among chickens for instance, or that observed among cows, is learned by each animal being conditioned to move toward or away from every other animal, i.e., each animal learns simple approach and avoidance tendencies in response to the others, According to Deutsch, deep power-striving in human beings is much rarer than these dominance hierarchies. Cartwright said that Group III started by discussing the nature of power motivationo The first issue broughlup was whether power always involves the control of resources. Most of Group III decided that the control of resources was an important part of the power relationship. Furthermore, if this is true, the next issue for the study of power motivation is whether or not the procurement of such resources is for personal consumption or in order to control their allocation. In the case of allocation, one gains power by controlling the way other people behave or use material goods. Another issue brough up by Group III involved research on power motivation. Fitts summarized three research questions: (1) What determines the percentage of the time a person thinks about power and dominance? (2) In the study of individuals who are said to be dominant, or powermotivated, in how many situations are they dominant? (3) What are the implications for the study of power of games that are played only once, such as duels, and games which can be played many times and won and lost, such as chess? 15

Cartwright went on to say that another way of studying power motivation is to look at people who are already in positions of power in the society. One can do this by picking people in authority positions such as those in the military establishment. Here one would study what has motivated the individuals who have risen to power in obviously powerful roles. Group III discussed research on achievement motivation and pointed to the possibility of using knowledge gathered in studying the achievement motive as a guide for studying power motivationo McClelland said, however, that power motivation is quite different from achievement motivation. Power has to do with a win-lose or all-none situation and unlike achievement motivation one cannot accumulate power, Dahl, Newcomb, and Deutsch disagreed with McClelland on the latter point. They argued that power can be accumulated and that when one does lose, it does not mean that he has lost all of the power he had before losing. The fact of losing may often assist a person in winning the next time. Kaplan pointed out that it might be a norm of a small group or entire culture that once you lose, you are allowed to win the next time. Deutsch said that the loser of a political election gets at least two satisfactions; one that he becomes the standard bearer of a minority party, and still remains powerful in that regard, and another that he is still seen as a potential candidate for next time provided he does not lose too often. Dahl said that even if a politician is defeated he still possesses certain resources. Among his resources are that he is known now by more people and there are expectations that he might be elected next time. Rapoport said that at least three basic models are helpful in looking at the power relationship. In the first of these A moves B in an intended direction. In the second A is not able to move B, and in the third A moves B but in the opposite direction intended. Since these models are directional there is no reason for our viewing the power situation as an all-ornone game as McClelland sees it. Dahl said that even in any particular instance power is not necessarily all or none. For instance, in trying to influence a member of the John Birch Society to liberalism one might be successful in getting the super-patriot to move only a little in changing his mind, or a great deal. Certainly, there are gradations in the extent to which A can move B in a certain directiono Group III also discussed the possibility of content analyzing documents such as textbooks. stories, and myths in order to see how often power motivation occurs in a cultureo At the same time, such studies would allow a comparison between, or among, different types of motivation, namely, achievement, affiliation, and power motivation. One of McClelland's colleagues for instance, Dro Rudin, has traced the power motivation in the United States since 1860. He has done this by content analyzing United States textbooks; specifically he has worked with elementary school readers. Rudin has shown 16

that power imagery reached a peak in these textbooks around 1910 while it hit bottom around 1950. The thirties, of course, were also the years that the democratizing of industry was taking place in the human relations movement. According to Rudin power imagery presently is rising again in the United States. There has been a steady increase from 1930 to 1950. It would be interesting to know whether the decline of power imagery ceased after the depression and started to rise during the war yearso Rudin's studies indicate that achievement imagery was at its peak in 1890; that power imagery was at its peak in 1910, and that affiliation imagery was at its peak in 1930 in this country. Group III talked about the nature of power as a social arrangement. Such concepts as domain and scope were discussed and it was agreed that power is relational and that the relational approach has not been fully exploited. Cartwright emphasized that social scientists will have to begin to look more at power relationships as symetrical, looking simultaneously at the leader and the follower. Group III discussed the scope of power. They raised the issue as to how a social scientist might predict to a powerful person's dominance generalizing to other areas. "What are the conditions of spread of power? What are the consequences as an individual begins to increase his scope and domain of power?" Other questions were: "How is power accumulated? What are the processes by which it becomes effective?" and finally, "What is the role of ideology in power?" Ideology often serves the function of accumulating power. The ideologist talks about getting power so that he might be able to do glorious things for the population. Often, however, when power is procured the spoken ideology is forsaken. Perhaps ideology is importantly related to power because a man's power is greater when he is changing another's mind. Ideology also can reach a much larger number of people than can sheer coercion or police power, so that in the long run, the leaders of a state, an organization, or a community have greater power if an ideology has been internalized by the followers. Kaplan said that Group II had explored the definitions of power and dominance first, and later went on to explore several dimensions of the problem. Group II also, saw power as a relationship. They sharpened the concept of power by defining it as a situation in which one actor is able to produce an effect on another and fulfill his own intentions in doing so. Kaplan's group also differentiated between power performance and perceived power, indicating that this is an important distinction for psychological research. Five dimensions of social power were: discussed: Motivations, levels, strategies, limitations, and effectso 17

Two important aspects of the problems of motivation were noted. One was Marmor's differentiation between positive power, or assertiveness, and negative power, or destructiveness. Often, according to Marmor, the individual who is delegated authority, or who achieves dominance because of his powerful resources, shows constructive assertiveness. The individual who achieves power out of a sense of weakness and threat often is destructive. Such a person is compensating for weaknesses and fears which he is repressing. The second aspect had to do with Atkinson's theory of cognitive expectations and the incentive value of an outcome. According to Atkinson, the tendency to respond with power behavior is a function of a stable motivation or dominance multiplied by a cognitive expectation that one might have success in a particular situation, multiplied by the incentive value or reward value of being successful in exerting power in the situation. The last two components of this equation vary from situation to situation. The first component-called the motivation to Dominate, or M- is more stable. Kaplan pointed out three possible relationships between the cognitive expectation and the incentive value. The first he referred to as the achievement type; here cognitive expectations and incentive values are inversely related. If one expects to be successful in the achievement area, the reward value of being successful is not very high. On the other hand, if one expects to fail, then achieving has a very high reward. A second relationship between cognitive expectations and incentive value is a positive one. This is the aggressive pattern, typified by the bully wo tends to find those situations most gratifying in which he has a high expectation of success in exerting power. The third kind of relationship is one in which the two, cognitive expectations and incentive values, are independent. Kaplan felt that this pattern typified most power behavior. The person strives for power over another person regardless of how easily he sees getting it. He strives for power over the other person primarily because of the incentive value of having power over that person. General discussion commenced with Cartwright pointing out that one area of great interest for research on power is whether the feeling of personal efficacy or effectiveness is related to power motivation in any way. Do people who are low in feelings of powerfulness have high power motivation? Redl felt that the two do go together in the following fashion. He said that among teachers who have rather high power motivation there is a malady which he calls "implementational despair." As such teachers find that they are not able to influence students in the direction that they had hoped to, before starting to teach, they feel despair and feel powerless to do things. Out of this sense of powerlessness they become punitive toward the pupils, and do not do a very effective job. McClelland felt that a sense of powerlessness and high need for power, do go together. He said that the survey data which Veroff has collected at The University of Michigan show that lower-class Negroes are high in power 18

motivation. At the same time, other studies seem to indicate that lower class Negroes feel powerlessness in relation to the political structure of the communityo Atkinson pointed out that personal efficacy or feelings of power can be mapped into his equation under cognitive expectation. Cognitive expectation is simply a statement of efficacy in any particular situation. According to Atkinson, all deprived groups in the society have a strong unfulfilled tendency to dominate because of frustrations experienced earlier. According to his equation there appeared to be two ways a person can get a strong tendency to dominate. The first of these would be represented by a high cognitive expectation, the second would be represented by the unfulfilled wish as high. To accommodate this, Atkinson's equation now would read: The tendency to respond in a dominating way is a function of a stable motive for domination multiplied by a cognitive expectation that one will be successfully powerful in a given situation multiplied by the incentive value of becoming powerful, in that situation, plus the unfulfilled wishes that one has about being dominant. Current methodologies for measuring power motivation do not differentiate between the two types of high tendencies to dominate. Hence, the national survey data, which McClelland spoke about, cannot adequately distinguish between a person who has a high tendency to dominate because he has a high cognitive expectation and a person who has a tendency to dominate because he has a high unfulfilled wish. Atkinson feels that the healthy people whom Marmor had called "assertive" would be typified by having a high cognitive expectationo Whereas people who have destructive tendencies in the power area would be typified by a high unfulfilled or frustrated wish. Marmor underlined these comments of Atkinson'so He said that the healthy person who achieves dominance does it out of a sense of strength. The neurotic individual, on the other hand, needs constant reassurance, and he strives for power out of insecurity and fear. Kaplan reminded the group that high expectations for being powerful do not always lead to tendencies to dominate. He reiterated Kobe's example from the evening before, which was that some people in Southeast Asia vote Communist only when they know the Communists cannot win. They do this for protest purposes. In so doing they are an example of what Deutsch refers to as an "expression party." Deutsch refers to the Democratic and Republican parties in contrast as "action parties." Cartwright, in summing up some of Atkinson's discussion, pointed out that we would expect the highest power motivation to be found among Negroes who are optimistic about the future. Negro leaders who have rising expectations of civil liberties and rights ought to have the strongest tendencies to dominate, according to Atkinson's equation. Cartwright concluded this about optimistic Negroes because for them both cognitive expectations and unfulfilled wishes would be quite higho 19

Marmor said, in answer to an earlier question by Guggenheim, that there is no evidence that the need to dominate is a primary motive among human beings. According to Marmor, motivation to dominate is socially determined, It has its basis in either lack of gratification or insecurity, most often insecurity. There is no evidence that individuals have to dominate other individualso There does seem to be some evidence that individuals do need safety and securityo Some individuals find safety and security by striving for power constantlyO Likert said that social scientists also should be interested in the strategies of using power. Studies of managers in industry show that those who emphasize reciprocal influence through interaction in face-to-face groups are managers of more high-producing organizations than those who do not. He has found that if managers build reciprocal interaction with their employees, they have more of a total capacity to wield power. Some of what we have been talking about as power motivation, according to Dr. Likert, will have to be rewritten in terms of different strategies. The effect of motives, in short, will be different, depending on the strategies used to actualize them. Newcomb pointed out that power can often be wielded in peculiar ways. For instance, one can control another's behavior by being submissive and weak. An example of this is a person who gets sick, in order to control the behavior of another persono Moreover, one can dominate another person by loving him; or by being helpless in his presence. Deutsch, at this point, spoke about the tyranny of the weak, He said that we must look at both parts of the relational system of powero Often the dominated has considerable power over the dominator. Take, for example, a colonial country, in which the way the people in the dominated nation react to the domination of the leaders has a great deal to do with the resources that the dominators have to use in keeping the colony viable. The jail-keeper who has to watch the prisoner constantly has really lost a great deal of his power. He is expending many of the resources which he has available to him by being close to the prisoner for so many hourso Newcomb finished this part of the discussion by stating that the consequences and motivations of power behavior are often quite different, and we must differentiate between them. Next, Dahl proceeded to describe two paradigms of power relationships. The first is a model in which we start with the behavior of the controller who acts in relation to a respondento The actions of the controller change the expectations or perceptions of the respondent and the respondent behaves in a submissive fashion to the controllero In this first model we can say, with little argument, that the controller is controlling or dominating the behaviors of the respondent. In such a model the behaviors of the controller precede the actions of the respondent and the controller's behaviors are perceived by the respondento 20

The second model is much more complex, and perhaps for this reason, more interesting. This is the model of anticipated reaction. Here, the respondent has expectations, though we do not know exactly their point of origin. These expectations lead him to behave in a certain way, To the observer this behavior looks as though it is submissive in relation to a controller. For instance, a senator who is making a decision about his vote on civil rights may consider that if he votes in a certain way he will get rewarded by the President. The President, however, has not done anything directly to the senator which would make him think this. After the senator votes as the President may have wanted him to vote we say that the President has controlled the behavior of the senator, yet this model is quite different from the first one, Here we start with expectations, move to responses, and then make the assumptionthat a controller has influenced the expectations of the respondent. Kaplan pointed out that perhaps the main difference between Dahl's paradigms is epistomological. In both models the phenomenologies of the respondents play a major role, In fact, the percepts of the respondents are viewed as determinants of their behavior. In the first model it is obvious to us that the controller has produced the percepts in the respondent. In the second one it is an open question whether or not the controller has produced the perceptions in the respondent. Certainly, in most cases there must be some reality to the perceptions of the respondent in either model. So, even though it is perhaps easier to find the reality of these perceptions in the first model because the controller has directly acted upon the respondent, this does not mean that in the second model the respondent is not reacting realistically. Perhaps, in the second model, the respondent is reacting to subtle cues or other indications emitted by the controller in past situations which do not directly bear on the issue at this moment. Dahl pointed out that perhaps in the first paradigm we can easily identify who is controlling whom, The controller is influencing the respondent. However, in the second perhaps we can see the respondent as actually controlling the controller, as in the example given, where the senator attempts to control the President. That is, by voting on the civil rights measure he influences the President to reward him. Such an example is analogous to certain jokes or humor about the rat in the laboratory who is controlling the psychologist, By pressing down a lever, the rat gets a pellet. A psychologist, therefore, has been influenced to give the rat a pellet by the actions of the rat. Such an example is probably trivial, however, and not worth pursuing. McClelland pointed out that Dahl's second model, the model of anticipated reactions, is most relevant in the study of the effects of ideologies or great men on history. Such men as Jesus or Marx have influenced the shape and form of modern history in many ways and yet have not done this directly. Their influence is ideological and therefore in the expectations of the actors. 21

Rapoport added a few remarks on the ethics involved in studying dominance and power. He pointed out that traditionally there has been a separation between "what is" and "what ought to be" in the realm of science. In the early days of the natural sciences this separation was extremely important. Many social scientists continue to make this distinction, carrying it over from the natural sciences. These scientists argue that the same distinction between description and prescription should prevail in social science. One famous sociologist, C, Wright Mills, took exception to this. Mills argued that where human phenomena are concerned, we must be quite conscious of both what is and what ought to be. "I want to echo Mills' position," Rapoport said. As social scientists and especially as social scientists studying power, we must be careful to interject values as we proceed. It is true that we might delay valuation or suspend it while we describe, but ultimately we must face the problem of finding those procedures for making man's life a better one. Values must quite consciously be dealt with as a part of the social scientific enterprise. 22

II. POWER MOTIVATION McClelland was asked to present some of his ideas on research concerning power motivation. He began by describing the methods that he has used in his research on achievement motivationo These methods include stories which subjects write in reference to Thematic Apperception Test pictures. The researcher concentrates on the ideational content presented by the respondent. Most work up to now has been done on achievement imagery. In the case of power, little has been done except for the work of Veroff at The University of Michigan. The research on achievement imagery started in 1947 with Atkinson and McClelland working together. The code for achievement imagery has been worked out very well. One assumption underlying the research is that the motives of people are reflected in their thought content. McClelland called such thoughtcontent,'"the inner concerns of people." Respondents express these inner concerns through their perceptions of ambiguous stimuli such as the pictures in the Thematic Apperception Test. McClelland wondered whether or not the work that has been done on achievement motivation could be applied to studies of power motivation. He said that much work has to be done yet on power motivation before the research reflects the sophistication of the studies on achievement. For serious work on power motivation the researcher will have to formulate coding procedures and see if these pay off by predicting power behavior. A measure is not useful unless it relates to another significant behavioral variable. Thus, McClelland argued that the main criterion that we, as scientists, should keep in mind as we work out measures of power motivation is how adequately the measure relates to other variables in which we are interestedo The problem of power for us, as American scientists, is multiplied because of the general embarrassment in the culture about the topic. Americans are relatively insensitive to power dynamicso In fact, McClelland thoughtthat perhaps we have unconscious resistances regarding power. Atkinson likened the power area, during the Twentieth Century in this country, to the topic of sex in the Victorian era. In contrast, McClelland's experiences in Tunisia have indicated that Arab cultures in general are much more power- than achievement-orientedo In our culture it is typical for people to use power in order to achieve something, or in order to compete successfully with a standardo In Arab culture it is much less likely that a person will use power to achieve. It is more likely that he will engage in power relationships simply for the gratification of engaging in them. McClelland said that achievement motivation involves doing well, accomplishing something competently, or competing with a standardo He 23

emphasized competing with a standard and striving for excellence. Power motivation, on the other hand, is characterized by attempts to influence the behavior of another person, as an end in itself. In contrast to achievement motivation, the power-oriented person wants to control others simply because he enjoys controlling the others and not for any other end. The power-oriented person's satisfactions come directly from controlling the other or from making another's behavior contingent upon his owno McClelland included, in his notion of power behavior, mutual or unilateral manipulation of another person's behavior. Marmor found some fault with this in that he thinks of power more in the sense of a dominance-submission relationship rather than the simple manipulative one described by McClelland. Marmor stressed that power is related closely to a hierarchical perception of life. This is repugnant to us, in the United States, because of the strength of democratic attitudes about interpersonal relationships. But, according to Marmor, the goal of the individual who is power-oriented is to dominate another person, to have mastery over him in some sense. McClelland said that he wanted to be a bit more generic in looking at power motivation, including even the simplest attempts of manipulation. Perhaps, in some cases, power motivation is subtly expressed. For instance, a person might get gratification out of manipulating another person, even though he does not have dominance over him completelyo McClelland talked about four different ways of expressing power behavior, These four ways enumerate methods of controlling or influencing the behaviors of another person: (1) The actor can control the respondent's behavior through reward and punishment. An example of this is, "I will give you a candy bar if you go to the store and get me a paper (2) A person can achieve dominance over another if the other identifies with him. This is the phenomenon of charismao In such situations certain magnetic influences of person A controls some of the behaviors of person B, (3) A third way of influencing the behaviors of another person is through changing that person's mind, or through restructuring that person's situation. This can be the method of ideology, persuasion, or propaganda. It is a method of indirect manipulation. An example of this is reminding a Methodist, when he begins to take a drink of alcohol, that he is a Methodist. Here one is evoking expectations of the Methodist which would reactivate his ideology. Cartwright added to this method of influence the method of manipulating the environment or so-called ecological change where resources are made unavailable. Following the example of the Methodist who drinks, Cartwright added the example of a prohibition lawo After such a law has been passed two things can occur; one is that the Methodist can be punished by the law for drinking; the second is that he might find it 24

harder to procure alcoholic beverages. It was this latter outcome which Cartwright saw as a typical effect of ecological changeo (4) Finally, McClelland mentioned "playing the norms" as a method of influence. He gave as an example, the king who orders the sun to rise. He gains influence because he has capitalized on something that he knows will occur. Thus, when A gives B a gift, assuming that there is a norm of reciprocity present, B is influenced to give A. a gift. Bargaining situations are typified by this kind of influence methodo Along these lines, McClelland mentioned that in Arab cultures there is much more giving and receiving than in the United States culture. This indicated to McClelland that giving and receiving is tied up with power behavioro Hypnosis is an example that may include playing the norms. Hypnotists often tell a person to do something he would have done anyway in order to get more control over him. Marmor felt that McClelland was talking about influence and not power with these four examples of methods of domination, Marmor said that poweris influence for the sake of making another person submissive. Deutsch indicated that we could talk about power in two ways. One is when A influences B in order to increase the dependency of B upon him. A second way is when A influences B in order to increase the independency of Bo An example of this second way might be education or therapyo McClelland felt that the dominance-submission that Marmor was talking about, or the superiorinferior relationship, is not necessarily involved in all power behavior. He said that in Arab cultures, for instance, some men seem to like to increase their scope and domain of power, and yet do not have thoughts about being in a higher status position than other peopleO McClelland added that Marmor's position seems a lot like the theory of the authoritarian personality, adding that this is another valuable position but perhaps only a specific case of a much more general phenomenon. Atkinson felt that McClelland was sometimes talking about affiliative instead of power relationships. For instance, when McClelland described the manipulative behavior of a small Arab boy, it seemed to Atkinson that the boy was trying to establish a friendship relationship with McClelland. If this were the case, then the boy could be said to be motivated for affiliation. McClelland, of course, viewed the boy as trying to gain power over him, and tried to restrict the boy's domination. According to McClelland's view, the boy was expressing power and he was resistingo Atkinson said that perhaps the boy was expressing affiliation and McClelland perceived this as power. Deutsch supported Atkinson by pointing out that McClelland had a high cost on his time; the small Arab boy had a low cost associated with his time. From the boy's point of view he was simply trying to establish a relationship with McClellando From McClellandTs point of view, a valuable resource, namely time, was being used upo Thus, the boy was taking some of 25

McClelland's power away. McClelland, therefore, saw the situation as a power relationship while the small boy saw it as a relationship of affiliation. McClelland went on to outline the differences between achievement and power motivation. About achievement motivations he said that the occupations where we find the most highly achievement-oriented people are in entrepreneurial occupations, namely promoters and salesmen. Power-oriented people, on the other hand, are in different occupations, Browning, a social scientist, showed that politicians have rather high power motivations. Also, other studies indicate that senior managers who own their own businesses are high in power motivation. Dahl pointed out that Browning also found that those politicians with moderately high power motivation were more effective in their politics than were those politicians who had very high or very low power motivations. McClelland added that ownership appears to be a very important part of power motivation and central to poweroriented behavior. Cartwright indicated that perhaps the relationship between ownership and power motivation is gradually changing in this country. He suggested that the persons with most power now are managers in bureaucracies and not people who own their own businesses. This "managerial revolution" has changed the nature of power behavior in the country. McClelland next contrasted people with more achievement and more power motivations by pointing out how each reacts to the knowledge of results. The highly achievement-oriented person wants to know how well he is doing in a very clear-cut way. He wants to have a quantitative concrete score for his performance. An achievement-oriented person likes to play games which are quantitatively scored. A power-oriented person, on the other hand, tends to be scared of quantitatively scored games. He gets his feedback when he sees another doing what he wants him to do. If another person shows that his behavior is contingent on the commands of A., then A, if he has power-motivation, feels reinforced and satisfied. Furthermore, the person who is power-motivated desires to extend the breadth and domain of his power. This, in itself, is rewarding. The "power-hungry" individual can be understood partially by the desire to extend the breadth and domain of power. If one has another under his domination and he knows that there is a very high probability that if he commands, the other will follow, then the dominator gets little gratification from dominating. On the other hand, if the dominator has less evidence that another person will do what he wants him to do, then he will get a higher reward if the other person does do what he wishes, Therefore, the power-oriented person always seeks an interpersonal situation in which there is some risk that he will not be able to influence the other person. Once he has established a perfect contingency so that his actions cause following reactions of the other, then he will look for other situations in which this perfect contingency does not exist. After he establishes another per26

feet contingency, he moves on to still another relationship. Thus the drive for power is insatiable. The power-driven individual is trying constantly to spread both his scope and domain of power. McClelland indicated that this is one reason for our having such cliches as "absolute power corrupts absolutely," or simply "power corrupts." Deutsch indicated that perhaps now McClelland was suggesting another definition of power. According to Deutsch, this definition would be that power is the reduction of uncertainty about another's behaviors. Atkinson indicated that what McClelland was saying fits nicely into his model of motivation and behavior. McClelland was talking about the incentive value of power and suggesting a relationship between incentive value and cognitive expectations. McClelland suggested that incentive value and cognitive expectations might be related inversely. When the power-oriented person finds that he can, without much effort, influence another person, the incentive value is reduced. On the other hand, when the power-oriented person finds that he will have some difficulty in influencing another, the incentive value increases. Dahl suggested that perhaps this inverse function is true for part of the power-oriented person's behavior, it may not be true in all situations as it seems to be in achievement-oriented behaviors. Fitts contrasted two types of drives. On the one hand there are the basic drives which, according to our psychological studies, appear to be satiable drives Such drives, as hunger, thirst, and warmth are satisfied or satiated by the proper environmental materials or resources. Many human drives, on the other hand-e.g., the drive for power, achievement or affiliation —are more complex and do not appear to be satiable. Perhaps some of them are, and others are not. Marmor again reminded the group that the way power is being talked about in the discussions is unclear. He felt that it is very important to differentiate between the two types of motivation underlying what seems to be power behavior. The first of these motivations emanates from strength and leads to constructive effects. The second emanates from weakness and leads to the deprecating of others and to domination in a relationship. McClelland felt that Marmor was being too restrictive in his view of power and added that Marmor's view of power motivation was appropriate mostly for the neurotically motivated person. McClelland answered by saying that he was willing to allow Marmor to fuse power motivation with aggression if this were the way he wanted to look at domination. According to McClelland it is the job of the behavioral scientist, however, to distinguish power, aggressiveness, destructiveness, and so on, and to look at each of them separately. Rapoport said that often motivations or personality have little to do with the long-term effects. He asked of what use will motivations be in 27

making evaluations in history. He gave the example of Stalin compared to Lenin, pointing out that Stalin is seen as a tyrant and a villain in Russia whereas Lenin is seen as a saint. Nevertheless, some of the things which Lenin did led to the power which Stalin was able to wield. In this sense, even though Lenin's motives may have been "pure," the effects of his behaviors were destructive. Kaplan said that perhaps motivation does enter into evaluation on a statistical basis. He said that Marmor's point is that if we can differentiate between constructive and destructive motivations for power, we might find that constructive motivations lead to higher values, and that destructive motivations lead to more negative values. In this sense, motivations will have different consequences and effects, and perhaps can be used in making evaluations. McClelland went on to compare the achievement-oriented person with the person with high need for power. He pointed out that for the achievementoriented person there is an inverse relationship between the incentive value of success and the probability of success. He hypothesized that this same inverse relationship would not hold for power motivation. The power-driven person will direct his attention to those situations where the probability of gaining power is high. Generally speaking, power-oriented persons are not risk-takers according to McClellando However, he would expect to find some power-oriented risk-takers in the population. Achievement-motivated people have confidence in themselves and prefer others to help them when they are trying to solve a problem. Achievement-oriented people will accept help in assisting them in the solution of a problem, because solving the problem is what will give them the most satisfaction. Power-oriented people on the other hand are less willing to give credit to others and often want to be the dominant person themselves. They find their gratification in controlling another's behavior, not in the final product of their actions. Cartwright pointed out that Helen Lewis did a study some years ago in which she contrasted task-orientation with self-orientation. Lewis' research indicated that the task-oriented person when helped to complete a task, had little residual tension. The self-oriented person who was helped, on the other hand, continued to feel some tension because he was not the one to complete the task. According to Cartwright, the task-oriented person would be high in achievement motivation whereas the self-oriented person would be high in power motivation. The achievement-oriented person faces situations and tries to learn from them so that he will be able to deal with similar situations in a more effective way. Power-oriented persons, on the other hand, dislike learning from experience but always want to win. Losing, for the powermotivated person, drains his reservoir of self-potency, and reminds him of the inadequacies that he is actively repressing. 28

The achievement-oriented person has a long time-perspective. For him, the present and the future are intimately connected and planning makes up a big part of his life. The achievement-oriented person also has a long future-perspective. The power-oriented person, on the other hand, connects present and future less intimately. He is more present oriented and faces each situation as it arises. The parent who desires to control his child for later benefits is more achievement oriented, in contrast to the parent who wants to control his child so that the child will be submissive before him immediately. The latter is a power-motivated parent. Dahl felt that the successful politician in this culture is probably more achievement- than power-oriented. Because our culture suspects negative things from power orientations, we tend to screen out people with high power motivation from political posts. Therefore, in America, we should find people using political positions to achieve. McClelland contrasted the power-oriented person with the achievementoriented person in the kinds of games they like to play. The power-motivated person prefers games of strategy. The achievement-oriented person, on the other hand, prefers a game of skill. A game of strategy is typified by playing the opponent. A. game of skill is characterized by competing against a standard of excellence. Kaplan contrasted two types of chess players. One who plays the opponent —a power-oriented chess player-and the other who plays the board-the achievement-oriented player. Rapoport described games of skill as games of contest. He gave, as examples, bowling, track, football, and so on. Chess is a game of strategy. Deutsch said that the power-oriented person emphasizes opportunity and timing; whereas the person with achievement motivation looks at the long run and decides to learn gradually. Rapoport said that the power-oriented person tends to deny others opportunities to be better than he. At the same time he takes his own opportunities at the appropriate times. Cartwright indicated that perhaps the power-oriented person would try to convert games of skill into games of strategy. McClelland reported some findings from an article by Roberts and Sutton-Smith ("Child Training and Game Involvement," Ethnology, 1, No. 2, April, 1962). The article discusses cultural determinants and behavioral consequences of various games. Three types of games are discussed: games of skill, games of strategy, and games of chance. McClelland suggested that one could contrast power orientations and achievement orientations by looking at the games of strategy and skill. McClelland also contrasted some kinds of maladies which the achievementand power-oriented persons might develop. The achievement-centered person would tend to get high blood pressure and ulcers. The power-oriented person, or at least the power-oriented cultures, tend to be high in murder and suicide. Redl added that the power-oriented cultures should be high in paranoia. McClelland said that in power-oriented cultures suicide is often 29

institutionalized. Also, in great dictatorships, where power is an important issue, mechanisms, usually formally stated, were present for suicide and murder. McClelland said that wealth is a measure of success for the achievementoriented person, while wealth is a means to influence for the power-oriented person. Furthermore, the power-oriented person, in contrast to the achievement-oriented person, is more interested in maintaining a status difference between the sexes. So a power-oriented man will see men at a higher status level than women. A power-oriented woman, on the other hand, will see women as equal or superior to men. Atkinson discussed some of Veroff's findings on power motivation. He said that Veroff developed a measure of power motivation by assembling some campus leaders on the evening of an election. Each of the campus leaders was up for election that evening. He presented TAT pictures to each of these candidates under the tension of counting the ballotso The candidates wrote stories while looking at the pictures. Veroff went on to compare the stories written by these candidates for elective positions with stores written by non-leaders in relaxed conditions. He used some of the different cues he found in the stories written by the campus leaders as indications of power motivation. Veroff's studies indicate that the student who is high in power motivation is more argumentative in class and shows a higher frequency of attempts to convince others of his point of view. These data represent observations made by instructors of their students. Further, Veroff found, power-oriented persons prefer occupations in which they can be in a leadership position or in which they are looked up to by others. Further, Veroff found, the highly power-oriented people are weak in social values, e.go, they have low humanitarian concerns. McClelland said that his data indicate that great dictatorships are typified by higher power imagery and low affiliation imagery in the textbooks and storieso Veroff's work indicates that power motivation is found most frequently in the highest and lowest income groups in the population. Managers are higher than professionals in power motivation, and Negro men are higher than white men. Further, power motivation is highest among the gradeschool educated, older men, unskilled workers, poorer men, and college educated older ladies. Veroff also found that power-oriented persons tend more often to come from broken homes than the less power-motivated people. Veroff did not find a relationship between power motivation and a person's desire to lead others, or his getting a feeling of satisfaction out of leading others~ He did find, however, that power-motivated people tend to resist supervision and being led by others. This raises the question of what role the need for independence plays in power orientations, 50

Still other findings indicate that men with high power motivations have more marital problems than those with low. Women, on the other hand, with high power motivations, indicate fewer or "less serious" marital problems than women with low power motivation. Men high in power motivation complain more about spending too little time with their children than men with lower power motivation. The high power-motivated men felt they could not spend as much time as they wanted to with their children. Finally, women, high in power motivation complained more about getting angry with children than did women with lower power motivation. 51

III. METHODS OF STUDYING POWER MOTIVATION Atkinson was asked to comment on measuring power motivation. He demonstrated the TAT as a measure of power motivation~ He passed out forms to all of the participants and presented pictures which were to be stimuli for stories written by the participants. Each of the participants wrote stories about two pictures. Atkinson pointed out that stories are coded for their imagery, whether it be achievement, power, or affiliation. This is done by a step by step analysis of searching for cues for achievement, power, and affiliation. His experience has indicated that strong power stories have three or four power cues or images. Scoring for an individual is done by totaling up all of the cues found in each of the stories. Such a test presumably measures some of the thoughts on power, achievement, and affiliation of individuals. Atkinson pointed out that if more than six stories are written in succession, the incidences of power, achievement, and affiliation imagery tend to decrease. There appears to be a satiation effect working. The following instructions were given to the participants before writing the stories: "You are going to see a series of pictures and your task is to tell a story that is suggested to you by each picture. Try to imagine what is going on in each picture then tell what the situation is, what led up to the situation, what the people are thinking and feeling, and what they will do. "In other words, write as complete a story as you can. A story with plot and characters. You will have 20 seconds to look at a picture and 4 minutes to write your story about it. Write your first impressions, and work rapidly. I will keep time, and tell you when it is time to finish your story, and to get ready for the next picture. You will write on two pictures. There are no right or wrong stories, or kinds of stories, so you may feel free to write whatever story is suggested to you when you look at the picture. Spelling, punctuation, and grammar are not important. What is important is to write out as fully and as quickly as possible, the story that comes into your mind as you imagine what is going on in each picture. Notice that there is one page for writing each story. If you need more space for writing any story, use the reverse side of the paper." After the stories were written by the participants, Marmor and Dahl read some of them to the group. For the most part these stories were full of achievement imagery, with very little power imagery. McClelland, quite consciously, wrote a power story so that the participants could see what 32

one was like. He read the story to the group. "Joe hates his boss who is always telling him what to do and is currently criticizing him badly. Joe has tried out a new system of accounting which his boss doesn't like. Because of a lack of understanding on the part of the boss, he is afraid that Joe will get ahead of him and get more authority than he in the company. Joe thinks that if his boss does not shut up, leave him alone, and stop criticizing him, he will quit and join up with the Ford Motor Co. which will let him do what he wants. The boss realizes that Joe will quit and is afraid of losing business. He stops complaining and criticizing just in time. Joe wins the point he has been making. He puts in the new system of accounting, displaces the boss, and becomes Presidents" Atkinson presented Veroff's ideas on measuring power imagery. He said that in order for the over-all code of power imagery to be scored there has to be some reference to the thoughts, feelings, and actions of one of the characters in a story which indicates that the character is concerned with the control or the means of influencing a person. Power imagery can be found in the imagery about any character mentioned in the story. Evidence of concern can come from any one of three sourceso (1) There is some statement of affect surrounding the control of the means of influencing another person. A character can be feeling good about winning an argument or feeling bad because he was unable to have his way about something. Also, statements of wanting to win a point, showing dominance, gaining control, convincing someone of something or putting a point across, can be interpreted as implicit statements of effective concern about the control of the means of influence, Effective concern can also be found in statements of wanting to avoid weakness. Examples of this are being humiliated in a status position; being ashamed of incapacity to assert one's self or to become dominant; and resenting the influence of another and wanting to overcome this. Some very weak statements of concern over control of the means of influence are scored. Statements of desires to teach another person something, to inspire another person, to interest another person in something, although apparently weak in obvious power significance, should be scored. The only times when statements like those above would not be scored, would be in cases where teaching or inspiring is solicited by the person being influenced, such as statements of teachers wanting to answer questions that students have raised. Solicited advice or opinion would be scored if there is evidence for power imagery beyond mere mention of answering requests. A person trying to put across a point within a solicited advice-giving story would make the story scoreable for power imagery. 33

Statements of a person wanting another person to gain control of the means of influence, such as a person wanting someone else to win an election, would be scored as power imagery. The only other case like this which can be scored is where there is a clear identification of the person and the candidate. This kind of imagery may appear in election stories where John wants to win the election, although John is not the candidate. (2) There is a definite statement about someone doing something about controlling the means of influencing another person. Something that the character is actually doing is the only kind of imagery that can qualify as power imagery under this criterion. The character has to be disputing a position, arguing something, demanding or forcing something, trying to put a point across, giving a command, trying to convince someone of something, or punishing someone. In short, any activity in order to obtain control of the means of influencing someone is counted. Statements that are either in the passive voice or in past or future tense are scoreable, but the mere mention of dissension or a shift of opinion in a story is not sufficiently explicit for it to be scored. Someone must be explicitly dissenting or trying to influence opinion in these cases. Physical power can be used as a means of influencing but does not, by itself, imply power concern. Power imagery would not be scored for example, if it were clear that the utilization of physical power was mainly in the service of expressing hostility. Some special considerations should be noted under this second criterion. Trying to interest, teach, and inspire someone will be statements that are scoreable for power imagery under this second criterion. Trying to win an election for someone else would not be scored unless there is close identification of thle person campaigning and the person up for office. Another consideration should be recognized; sometimes it will be clear that the activity of the characters in the stories although it meets the criteria listed, is only for the purposes of arriving at the goal of some other motive. When this is so, the story should not be scored for power imagery. However, if it is not clear whether power motivation or some other motivation is the ultimate concern of the imagery, score the story for power imagery. (3) A story can be scored for power imagery if there is a statement of an interpersonal relationship which is culturally defined as one in which a superior person has control over the means of influencing another one who is subordinate. Examples of these are: boss and worker or judge and defendent. Mere mention of a superior-subordinate relationship is not enough. There has to be some mention of the activity involved in carrying out this relationship. Indeed, if the story about a boss and a worker goes on to elaborate about the affiliative bond between the two men the story should not be scored. Mere mention of the fact that a given person was influential would not be enough to allow the story to be scoredo For a story to be scored for power imagery under this criterion there has to be 54

some mention of the subordinate as well as the superior position. Either the subordinate is directly involved in the imagery or the effect of the superior on the subordinate has to be clear for the story to be scored. The parent-child relationship is not enough, in itself, to be considered a power relationship. The use of culturally defined channels of influence by the subordinate in a superior-subordinate relationship will be scored. In answer to a question by Marmor, Atkinson said that very little work has been done on healthy and unhealthy power driveso In the college populations studied, the power drives that are manifested are most often healthy. Atkinson pointed out that according to Veroff's findings, opposition to power indicates power motivation. Thus, we would expect an anarchist to be power motivated. Rapoport criticized this hypothesis, indicating that he would not think of an anarchist as a power-motivated person. Atkinson asked for some assistance in thinking about the kinds of pictures which might evoke power imagery easily. The group "brainstormed" the following ideas: An assassination scene, parent-child relationships, a teacher in a classroom setting with students looking on, a military picture in which several ranks are represented, a judge in a courtroom, a policeman walking down the street, a policeman walking in on a crap game, a hypnotist and his subject, a civil rights picture in which a Negro is being hauled out of the restaurant by the police, a picture from Mutiny on the Bounty, and a picture of a demigod such as Hitler or Mussolini. Atkinson pointed out that it has been found that a picture with too many achievement cues does not differentiate between the high achievementoriented and lower achievement-oriented people. For instance, if the picture of two track stars running in a race is presented, almost universally one finds that the subjects write achievement imagery. Similarly, in the power area, perhaps a picture of Mussolini, or some other power-oriented person, will tend to evoke power imagery from most subjects. With this in mind, we look for pictures which have a moderate amount of cues in the power, achievement, or affiliation areas. At the same time, pictures with many power cues do serve a useful function. We can use these to collect images from subjects which can later be codes used in stories written after pictures less laden with power. Redl asked, "Why limit the study of power to TAT questions, or even to projective devices? Why not extend our methodology?" He said that we should begin to use participant-observation more often in studying power situations. We could, for instance, put someone in a principalTs office and observe how disciplinary sessions are handled, or we could put someone in children's groups to observe power relationships. We could put someone in classrooms or in prisons and other such settings where power orientations are frequently expressed. Likert supported Redl s suggestions for broadening the methodologies for studying powero He pointed out that one useful 35

method in organizations has been to study perceived power and its correlates. It has been found, for instance, that the decision-making processes, productivity, and influence structures of organizations are interrelatedo Other types of projective devices were suggested in discussion. One of these was auditory materials, such as tapes or records. Here the subject would be presented with parts of conversations and then would write or tell a story about what was happening, Another idea for a projective technique is the story completion or sentence completion testo Here the subject completes a story or a sentence, both of which are initially incomplete. These techniques have been used quite effectively by psychologists in the past. Another technique discussed was to look for people who are power oriented already, Such people could be identified by psychiatric ratings, and then their motivations could be measured and analyzed0 Cartwright said that we will want to combine methods in the future using such tools as sociometric tests, interviews, observations, projective tests3 and character inventories, It would be premature to assume at this point, that the projective test is the best methodology for studying power behavior and power motivation. McClelland added that we will want to do what Murray did in the OSS studies in which he used a variety of assessment techniques to measure the criterion behaviors. Atkinson commented that the problem is not finding stimuli for projective cues, but instead is in discovering appropriate ways of coding the stories. Deutsch pointed out that the context in which the projective test is given will have an effect on the stories that are written. Atkinson agreed and said that there is an attempt in most experiments on power motivation, either to manipulate experimentally the context or to control the effects of the contexto Atkinson said that research has indicated that the ending of the story is not very important in measuring motivationo What is important about the stories are the instrumental activities discussed by the subjects. Deutsch felt it important to differentiate between two types of power orientation. He entitled these, power-oriented and the policy-oriented persons, using the language of political science. The power-oriented person wants to do something for his own self-aggrandizement; he thinks about elections only in terms of personal power and victory. The policy-oriented person, on the other hand, thinks first of the policy and secondly of himselfo In other words, if this policy-oriented person would find a way of achieving the policy without his being elected he would prefer that way just as much as he would prefer being electedo Cartwright discovered some advice to business executives in an article by Martin and Sims, "Power Tactics, Harvard Business Review, NovemberDecember, 19560 He presented these as possible items for instruments, Such instruments could be used to measure one s attitude towards powero 36

Some of these items were as follows: (1) The able executive is cautious about how he seeks and receives advice. He takes counsel only when he himself desires it, (2) In the struggles for power and influence that go on in many organizations, every executive needs a devoted following and close alliances with other executives, both on his own level and above him, if he is to protect and enhance his status and sphere of influence, (3) The wise executive maintains his flexibility and he never completely commits himself to any one position or program. (4) The executive should accept compromise as a means of settling differences with his tongue in his cheek, While appearing to alter his view, he should continue to press forward toward a clear-cut set of goals. It is frequently necessary to give ground on small matters, to delay, to move off on tangents, even to suffer reverses in order to retain power for future forward movement. 37

IV. SOME PSYCHODYNAMICS OF DOMINANCE AND POWER Marmor presented some ideas on the psychodynamics of dominance and power. He said that there were at least three kinds of things which he wanted to note concerning power motivation. The first of these was that we are in need of genetic studies. We need more studies on the roots of the need for power and on the personality development during the early years of people who are driven for power. Secondly, Marmor thinks that we should keep in mind the difference between the task-oriented person and the individual who is person oriented. The first of these McClelland has called achievement motivated, the second, power motivated. This is the same distinction Marmor wishes to maintain. It is important to point out, when thinking about power behavior, that the person with power motivation wants to dominate other people for the sake of dominating them. The power-motivated person does not care much about getting the task completed. He gains his gratifications from dominating another person. Finally, Marmor said, that the compulsive drive for power emanates primarily from authoritarian home settings. In such families power, rather than love, is the main source of security for youngsters. Marmor went on to contrast two types of families in which the poweroriented child develops. The first of these families is the so-called authoritarian family in which the repression of hostility is usual. A child in such families is taught not to express aggressionand hostility, especially in the presence of powerful adults. The child also is rewarded for finding scapegoats on which he can vent his suppressed hostilities. A second type of home mentioned by Marmor, is the broken family. In such families there is a lack of any definite focus of security. The best way for a child to get along is by learning to manipulate others. For such children people become objects to manipulate and they are dehumanized, Such people generally feel themselves to be deprived and unfairly treated by life. Finally, an extreme example of these two types of development, is the psychopathic personality type. Such a person has a poor conscience or superego, and power becomes an overriding theme in his life. Morality and ethics are used purely for the sake of manipulating other people. Marmor felt that it was very important from the motivational standpoint that we not confuse Jesus and Hitler. Motivationally speaking, as well as evaluationally, Jesus and Hitler were quite different people. Marmor said that the main part of the personality which is distinctively different between the compulsive and healthy, power-driven people, is the nature of the superego. A compulsively driven power-oriented person has weaker controls and much more destructive content in his fragmented superego. A healthier person has more constructive content in his superego and a much 38

different set of values. Marmor contrasted two kinds of radicals which he had had in therapy. One was motivated mainly by hatred, the second was motivated mainly from love. The first of these, the man motivated by hatred, grew up in a broken home with a very authoritarian fathero He was in constant conflict with his father and always opposed to authority figures in general. Moreover, he was against the status quo, whether it be proximal to him in his immediate neighborhood, or more distal, such as the nation state. Furthermore, he was untrustworthy both in his marital relations and his relations with his childreno He thoughthe loved humanity, but according to Marmor he hated people. At the same time he voiced beliefs which were related to apparently good causes and a desire to help the poor people in the countryo The second radical was motivated mainly by love. He grew up in a very large family with a devoted mother. For most of his early life his mother emphasized the importance of an equal distribution of love and resources, She encouraged him to spread the love and resources equally among all of the siblings. He became an idealistic gentleman with good human relationships, warm friendships, and a loving disposition. It is true that sometimes he said such things as "down with the tools of Wall street'" or "down with the capitalists," but, according to Marmor, his heart bled for underprivileged people. He was sincerely dedicated to helping people and to the elimination of poverty and prejudiceo He would tend to be cruel only if he were driven to it, according to Marmor. Guggenheim asked Marmor if these men had been different during childhood. Marmor said that he had not done therapy on them as children, but he would certainly predict that they were quite different children. Marmor emphasized that life circumstances especially early life circumstances, are the origins of power motivations and other relevant social motivations. Kaplan reminded Marmor that both the nature of one's motivations and the content of one's superego are important psychological determinants of power outcomes. One may have to look at the nature of the superego before making a successful prediction about the constructive or destructive nature of the power behavior. Nevertheless, we can predict that the most destructive types of power behaviors will be manifested by individuals for whom power is an end rather than a means. Pendray asked Marmor for the difference between dominance and power. Marmor answered that dominance is a state of a relationship between two people. It is a relationship which is not necessarily sought, but which exists. Marmor pointed out that he uses the biological definition of dominance, Here, we have a relationship in which one of the parties is dominant because of some superior quality~ Among animals this superior quality might be strength or speed. Among human beings we speak of status, 39

role, authority, high intelligence, and accomplishments. A person who is dominant is not necessarily power motivated. In fact, most of the time dominant people have no real need to be power driven. For instance, the president of a university, the secure parent, or a teacher, may all be dominant but may not be power motivated. The power-driven person, on the other hand, gets pleasure out of dominating others. He gets a great deal of pleasure out of pulling rank, or throwing his weight around. He has an active need to control and subordinate others. Power has many meanings, but to Marmor the need for power means the need to dominate, the need to subordinate other people. Marmor went on to say that exercising power, or the strategies of power is a separate issue. Dahl and Atkinson reminded the group of the relational character of power and dominance. Dominance, according to Dahl, is a habitual relationship between two actors in which one is assumed to be superordinate and the other subordinate. According to Newcomb, such a relationship is a behavioral one. Atkinson said that when studying power, it is important to take into consideration both the motives of the dominator and the motives of the dominated person. Furthermore, we must look at the multiplicity of motives involved in power behavior. Atkinson said that we might be observing the presence of a very strong motive such as Marmor has described as power striving, or we might be observing the absence of something, such as the superego, when observing power behavior. Atkinson felt it overly simple to use only the power motive when trying to understand the determinants of power behavior. Marmor said that he liked the idea of looking at motivations as they multiply to determine individual behavior. He said that, for instance, he might have been talking about high need for power and low need for affiliation when he was giving a description of the person with the destructive power behavior. Indeed, we might say that the unhealthy and neurotically power-driven person has high need for power and low need for affiliation. Marmor said that his theory of human behavior can be termed an adaptive theory; every human being attempts to adapt to his setting by achieving a sense of safety and a sense of pleasure. The way individuals do this is different, and determined somewhat by acculturation. In a democratic society it is assumed that people should be able to achieve safety and pleasure from others rather easily. Some individuals have difficulty because of their familial upbringing, in finding safety and pleasure. They tend to adapt to their environments by becoming power oriented; by looking for situations of safety and pleasure through the manipulation of other people. Some of these people can be said to be under acculturated, i.e., they have underdeveloped consciences. They do not play the societal rules according to how they have been set up and shared by most of the individuals in the society. Such power-oriented behavior grows out of a lack 40

of personal social controls. Marmor agreed that we can look both at the pushes to power-oriented behavior and the pulls, the motivations and the superego effects. Deutsch felt it useful to try to combine several motivational systems into one. He first presented categories from Lasswell and Kaplan, Power and Society. Deutsch said that it would be useful to consider each of these categories as being relevant to power behavior: Power, deference, enlighten ment, wealth, skill, affection, righteousness, spontaneity, well-being, and security. Deutsch added spontaneity to the Lasswell-Kaplan list. It is similar, according to Deutsch, to the need for independence. Deutsch said that Atkinson and McClelland's motivational categories can be matched up with this Lasswell-Kaplan system in the following way. The need for power is similar to the Lasswell-Kaplan categories of power and deference, achievement is covered by enlightenment, wealth, and skill; and finally, affiliation is covered by affection, righteousness, and spontaneity. Deutsch also mentioned the taxonomic system developed by Parsons. According to Deutsch the way one can match up the Parson's system with that of Lasswell-Kaplan and that of Atkinson-McClelland is the following: Parson's category entitled "goal attainment" is similar to "need for power," both of which are related to "power and deference." Parson's "adaptation" is similar to the "need for achievement" which is covered in the LasswellKaplan system by "enlightenment, wealth, and skill." Parson's category of "integration" is similar to the McClelland-Atkinson category of "need for affiliation" which overlaps with "affection, righteousness, and spontaneity" from the Lasswell-Kaplan system. Finally, Parson's category of "pattern maintenance" is similar to Lasswell-Kaplan categories of "well-being and sewcurity." Newcomb raised a question about the usefulness of such typologies. He said that psychology is full of dead taxonomies. Rapoport, agreeing with Newcomb, said that this was a very nice catalog, but where does it get the researcher? Deutsch answered that these categories have already proven to be very useful in political science and sociology, and presumably the McClelland-Atkinson category has been useful in psychology. He further said that if we keep on doing research in different directions, without trying to combine it into a single system, we will not get very far. Deutsch said that these categories can be usefully used in machine scoring where one sets up a code and uses an IBM machine to code stories. Many predictions would be made from these categories. According to Deutsch they are not atheoretical; each of them has its own theoretical system built up behind it, and each is based on a very complicated conceptual system. McClelland said that in 1953, Parsons, Shils, and others, had attempted to do the saime thing that Deutsch is attempting. The product from their meetings was Toward a General Theory of Action and is now quite popular. 41

McClelland said that he has somewhat of a tense feeling about such activities in that they do not seem to get very far. He said that he thinks h ug;rees with Stouffer who said we must have more modest empirical approaches to the problems of social science, and wait for some years before we get into general theories. Doolittle spoke very briefly about leadership in the military. He felt that there are five very important personal attributes for the military leader to have. These were: (1) courage, (2) intelligence, (3) integrity, (4) humanity, and (5) spirituality. These attributes should help the military man to acquire from his men (1) respect, (2) admiration, and (3) affection. Doolittle felt that respect was the most important of these three with admiration as next most important. Doolittle went on to describe McArthur as a Dominant man (with a capital D), Marshall as a dominant man (with a small d), and Eisenhower as an influential man. 42

V. DOMINANCE AND POWER IN LARGE-SCALE ORGANIZATIONS Likert elaborated on some of his own research on organizations. He said that the manner in which power is used and leadership is performed affects organizational productivityo His findings suggest that as a country industrializes, different ways in which power is exercised occur. For productivity to be high in an industrialized society two-way influence must occur more than one-way influence. Direct uses of power leads to low control and influence in an organization and low performance especially in highly industrialized countries. Likert pointed out that in organizations that use overlapping group structures where men at all levels in the organization interact with the level above and the level below in making decisions, higher productivity exists. An organization with overlapping group problem-solving across its hierarchical levels has high productivity. Both the use of group problemsolving and listening to the problems of men at all levels of the organization lead to higher productivity. Furthermore, the amount of influence, overall, in an organization increases as overlapping problem-solving is used. When the men at lower levels in the organization find it easy to take grievances upward or consider their peers as being helpful, productivity seems to be high. Such decision-making with men at the higher level in an organization tends to lead to high morale, cohesiveness, and productivity. Likert said that recent research indicates that as changes in the character of organizational decision-making create overlapping communication structures and more reciprocal influence, high worker involvement and higher productivity occur. Furthermore, happiness with the organization along with high performance goals lead to high productivity. In organizations where reciprocal influences occur often across levels, happiness and performance goals tend to be high. In such organizations we also find much less power behavior among the leaders. Likert went on to say that many companies are not convinced of this because of the way in which they judge success. Generally speaking, companies look at success from the point of view of their accounting reports which are short-ranged. Managers of companies find that by putting the screws on workers for a small duration of time they are able to raise production quickly. Likert pointed out that these managers seldom look at the long-range effectso He said that certain hostilities and negative attitudes build up within the workers and show up perhaps one, two, or even three years later in terms of labor disputes and unrest in the company. Marmor added that it sounds like Likert was saying that good leadership leads rather than pushes. 43

Kobe said that in some countries such reciprocal influence in organizations works well. However, in countries which do not have much industrialization, it would be dangerous to put the reciprocal influence idea into practice. Adverse effects, unrest, and low productivity might result. Marmor pointed out that Likert is talking in organizational language about something which we have observed in terms of family life. He said that children tend to be most obedient when they identify with an a.ttra.ctive adult. Further, children tend to follow adults more under conditions of reward in contrast to punishment. Fear of punishment is not a good method, generally speaking, for getting a child to identify with the adult. At the same time the adult cannot be totally permissive. He must engage the child in various forms of problem solving so that the child will learn limits. Marmor added that this seems to be what Likert is talking about in organizations. Limits are set through mutual problem solving while effort is made on the part of top management to make production goals attractive and happiness an important part of the workers' lives. Marmor added that Likert's study appears to be an application of the old adage that "one can catch more flies with honey than with vinegar." Cartwright pointed out that according to Likert's findings, organizations with high reciprocal influence are ones in which the workers have more influence and in which the managers have more influence, so that influence rises regardless of the level in the organization. In such organizations, workers and managers feel involved, and as though they are carrying out important jobs for production. It goes without saying, according to Cartwright, that communication seems to be the important link here. Communication is almost always smooth in productive organizations. 44

VI. COMMUNITY POWER Dahl began by presenting some material from his book on the New Haven community entitled, Who Governs? Unlike the earlier sessions, he pointed out that his work is not motivationally oriented but focuses on community power and has been carried out from the perspective of a political scientist. Furthermore, Dahl preferred the term influence, though he felt it legitimate to use power and influence interchangeably. He set for his job a description and explanation of the distribution of influence, the patterns of influence, some characteristics of the influentials, and changes in any of these. His emphasis was on influence in regard to the actions of local government in New Haven. As Cartwright pointed out, Dahl mostly is concerned with the scope of power. Dahl's research was concerned with overt decisions in the government which affected many people. The important variable, then, was final outcome on issues which were visible. Dahl focused on three issues in his research. The first of these had to do with decisions concerning the public schools~ This issue was chosen because of its widespread interest in New Haven and because school expenses make up the largest part of the city budget. The second issue studied was decisions on nominations for local offices. Here Dahl studied the recruitment process. Finally, Dahl studied decisions on urban redevelopment. This was a major activity involving both the business and governmental segments of the community. In fact, the city charter was being considered for change because of the urban renewal issue, while Dahl was carrying out his research. Dahl pointed out that these issues satisfied only part of his academic interests. He also studied recruitment patterns into political jobs; the political structure, rules of the political process, and finally, the local ideology or belief system about government. Dahl said that these issues are increasingly more difficult to operationalize compared with the issues involving the schools, nominations, and urban renewal. The approach of many political analysts is to look at a community such as New Haven from the point of view of a traditional theory. One such theoretical orientation has to do with the nature of democratic institutions. Here, the role of elections and the impact of the electorate are emphasized. The problem for such theorists generally, is how one can measure the influence of the electorate on the policy makers. The second perspective, quite different from the first, emphasizes eliteso Such writers as Pareto, Mosca, Co W. Mills, Marx, and the muckrakers saw power held by a very few people with common interests and goals who are highly coordinated and inter45

connected. Dahl did not begin his research with either of these orientations. Instead he sought some more generic concepts in order to conceptualize power in the community. Dahl assumes that influence is a function of resources. Resources are things which are used to produce responses from others. They are such things as money, status (one's own or his manipulation of others), and the control over information, either through expert knowledge in the bureaucracies or through the newspapers and radio stations. Another resource is legality or legitimacy. The individual with the legitimacy is able to influence another because the other willingly accepts and expects to be influenced. The dominator is thought to be correct morally in using his influence. Dahl assumes that the more of these resources one has, the more political power or influence he can wield. However, one must add the concept of utilization here. It is not enough to have access to the resources. For instance, some wealthy people collect paintings while others collect politicians. This notion of utilization, who uses his recources to get influence, overlaps with the motivational categories discussed by McClelland and Atkinson. Finally, Dahl assumes that efficiency in using the resources is also needed to predict to influence behavior. Efficiency refers to the capacity for using resources well. Thus, according to Dahl: Resources + utilization + efficiency - Power (influence). Dahl went on to discuss the methods he used in his research. He said that political scientists are years behind the psychologists in using sophisticated methodology. Partly because of this, Dahl's research team resorted to a variety of methods. These methods included the following: (1) Historical records were used. Starting in 1796 the names of incumbents are available along with their educational, social, and occupational characteristics in New Haven. Few studies have been done on the political history of New Haven even though many good records are available. (2) Participant observation was a second method used by Dahl's research team. Here, Raymond Wolfinger, a graduate assistant at that time, observed in both the Mayor's office and the office of the Director of Urban Redevelopment. Both men were willing to have a social scientist in their office so long as: he would exclude himself from sensitive high-level meetings, Dahl would never publish anything that either of these public administrators had not seen, and furthermore would never publish anything that could be traced to them. Wolfinger felt that regardless of these restrictions, he was able to uncover a great deal of what went on at important meetings. 46

(3) The third method for data collection was identifying the important decisions of high-level people through interviewing and city records. Attempts were made to reconstruct complete records concerning what people were involved in important decision making, what these people took into consideration, and how they proceeded. Later, data. were tabulated on who initiated the policies which eventually were carried through. From such data a. measure of actualized influence was acquired. (4) The fourth method was a survey of the general population carried out in order to accumulate some basic data. on political participation in New Haven. Little, if anything, new was uncovered here that had not been found in other surveys. (5) An analysis of voting by wards was done as a fifth method. Emphasis here was on historical changes. Quite a few were found over the last sixty-year period. (6) Finally, allthe names listed on school boards, and urban development boards were put into a list. The characteristics of the people in this list were analyzed and considered to be the characteristics of the political leadership in New Haven. Dahl discussed his findings by first presenting some results from the historical analysis. From 1790 to 1830 the official leaders of New Haven were a homogeneous ruling elite. They came from established Yankee families, went to college at Yale, and were usually either lawyers or Congregational ministers. There were also a very few Episcopalians in the group. The members of this elite were the people of highest education, highest wealth and status, and of highest control over the religious and educational institutions, During this time in New Haven, more than at any other, one could speak of a "power elite." Dahl called the years 1830 to 1899 the "Horatio Alger period." This was a period of dominance for the entrepreneurial class. These men came from Yankee stock; however, they also came from poor farm familieso A typical example was a boy who was orphaned on a farm at age 12; he became an apprentice to a carpenter, became a skilled carpenter, next became a carriage maker, sold carriages, and finally became mayor. From 1899 to 1930, ethnic groups formed the dominant part of New Haven culture. This was the period of the new man. During this time the first non-Yankee became mayor. He was an Irish Democrat born in County Cork. During this period the Republicans also became an ethnic party. While the Democrats had many Irish Catholics, the Republicans attracted many working class Jews and Italians. 47

During the most recent years in New Haven, power is in the hands of between five and forty people. Only about eight people can be said to have very direct influence. According to the data, however, different people have influence on the different issues of urban renewal, the schools and political nominations. These influentials are typified by different socioeconomic backgrounds. Leaders involved in urban redevelopment come from the highest social classes. Those involved in public school work come from the second highest classes. Private schools drain off participation of the highest social classes. Finally, political nominations are carried out by the lowest socioeconomic group. However, in no case, is a blue-collar worker involved in any of these influential groups. From 1945 to 1951, these three groups worked quite autonomously. There was no strong coordination among them. The mayor at that time, an Italian Republican, was a very cautious man. He was more interested in his own status, than he was in policy. His information was generally fragmentary and he did not use Yale as a source of knowledge. In 1951, a new mayor was put in office, an Irish Democrato He was a highly energetic and ambitious man with a raging set of ulcers. He has always used Yale as a reference group. He brought into office some dynamic, expert administrators, and became deeply involved in all three of the issues Dahl studied. Coordination was much greater across the three decision-making groups during this period. Dahl raised the question here of how to account for these historical changes. Several things are relevant. First, there is a change in the access of different groups in the population to resources; and secondly, some dissatisfaction with the kinds of people who are in office periodically occurs. For instance, in 18.18 there was a good deal of dissatisfaction with the gentlemen's oligarchy. The dissatisfaction took the form of a religious controversy. In 1818 a coalition formed among the upper class, the Methodists, and the artisans in an attempt to oust the Congregationalists. In.1899 there was a revolt of the lower ethnic groups against the power figures. At that time, the Republicans pulled Jews and Italians into the party to form a minority coalition. Why do some people use their resources for power, while others do not? Most citizens do not use the resources they have to influence policy. Only a small proportion of people seriously try to exert influence, Marmor commented that this is probably because many people are unwilling to assume responsibility. This reluctance to assume responsibility is quite common, according to Marmor. This may be one of the important distinctions between the leaders and the leado Leaders, in contrast to followers, have a willingness, and perhaps a need, to assume responsibility. Dahl pointed out that there are major differences between the way the citizens of New Haven answer questions about democracy at the abstract and 48

concrete levelso The people generally agree with beliefs in democracy, and minority rights when these are stated at the abstract and general level. However, they often disagree with statements posed at the concrete level; e.g,, should an atheist be allowed to run for public office? Dahl explained that our rights persist because of the way liberal attitudes about democracy are distributed in the population. The willingness to see rights realized goes up with education, income, and occupation. At the apex of the community power structure there is the highest degree of consensus on maintaining the belief system about democracyo The more powerless people, objectively speaking, are least likely to support the rights of others. Dahl said that a.. leader's degree of freedom as leader, depends on how much his goals run counter to central goals of people in the population. New Haven has passed through periods in which resources were distributed unequally to unified elites, and unequally to elites which were not unified. Dahl felt that the current trend is toward the second of theseo He felt this to be true even of nondemocratic societies. McClelland asked about what kinds of people are powerful for each issue in New Haven. Dahl said that there are about twenty influentials in the schools who are middle-class people. Usually there is a trade union leader, a business man, a leader of the Negro professional class, a chemist, and a lawyer. These people almost always have children in school. They typically have very high educational aspirations for their children and themselves. Sometimes, one will gain access to the school board by first being the PTA president. The school board is now divided according to ethnic and religious backgrounds. The board has a majority of middle-class Catholics (who are not against public education), and usually at least one Negro, Jew, and Protestant. McClelland pointed out that politics is really like an occupation. The introduction to politics is done through gradual socialization. Dahl said that for most people the rewards in politics are really very small. He added that Americans appear to be more interested in politics that people from other countries. There is more professionalism in American politics than we think. All sorts of motivations are tapped in politics making any analysis limited that simply looks at power motivation as the key to political involvement. 49

VII. ECONOMIC DEVELOPMENT AND POWER ORIENTATIONS Kobe spoke next from the standpoint of his involvement in studies of economic development. He drew the following diagram on the blackboard (using some of the ideas of Rostow at MIT): Income per-capita /Prolonged and Take off / sustained growth Traditional Pre-conditions period society History Kobe said that the modern, industrialized nations appear to have gone through a history much like that pictured in the above diagram. First, there is a traditional period, a feudal economy usually, in which things are very stable. Then follows a period called "pre-conditions" during which the society is readied for a change; next the "take-off" period, and finally mass consumption and high productivity. According to Kobe, one cannot predict when the take-off period will occur. This is the case because during the first two periods, traditional and pre-conditions, one finds waves in the graph. Such waves make it difficult to know if a future period will be a take-off period or merely a temporary upswing. Along with these stages of economic growth, cultural and social psychological phenomena are occurring at the same time. During the traditional period, for instance, resignation and contentedness are very high in the population. The people are quite satisfied, take things as they are, and generally feel that you cannot do anything about changing the present state of affairs. During this traditional period, Kobe thinks that power motivation is very high. Furthermore he argued that achievement motivation becomes more prominent during the take-off period. He said that it is his job to try to assist underdeveloped countries to "take-off," In Viet Nam, for instance, the leaders are power oriented. American aid, in contrast, is postulated on achievement motivation; Americans give advice and technical know-how. Such information does not fit the most prominent motivational orientation of these people. It appeared to Kobe that problems like this might be solved if we understood how to change power motivations to achievement orientation. 50

In Japan before and during World War II, power motivation was very widespread. Feudalism was very strong and this kind of a system supported power orientationso After 1860 symbols of dominance and submission were widespread in business, culture, and the language of Japan. In the language, for example, there were fifteen ways of saying "you", me?, and "I". These different ways were hierarchical; one spoke to a superior differently from a peer who was spoken to differently from an inferior. Since World War II many changes have occurred in Japan making it more achievement than power oriented. One such change has been in the language used in radio and television broadcasts. The language of radio and television is a mixture of the old language, mixed in such a way that all hierarchical varieties of pronouns and verbs are used by the speaker, Kobe went on to point out that Japan did have high rates of suicide. He reminded the group that McClelland had argued that suicide and homicide were very high in power-oriented societies. In Japan the suicide rate goes up with age; however, there is a peak for women between 18 and 22. This appears to be a maladjustment period for the women. The old society is still very strong while the new achievement-oriented society is beginning to exert its pressuresO Suicide occured often in Japan because pacts or contracts were made on a man-to-man or person-to-person basiso If one is unable to carry out a contract, merely getting permission from the other person may be enough to break ito However, if one causes another person great personal damage by not carrying out a contract, the most extreme thing one can do to make amends is to kill himself. Another reason for suicide in Japan is that it is very difficult to get away from others physically once one is ostracized. In Thailand, on the other hand, it is possible to survive years alone in the jungleo But all of this is changing in Japan. The post-war industrialists are very different types. They are achievement oriented, like the American businessmen. The flower arrangements, the Kabuki, the puppet shows, and the tea ceremonies are all changing rapidly. Power orientations are decreasing and achievement orientations are increasing. It appears that the economy, the psychological orientations, the culture, and the language are all intermingled and connected with these power and achievement motives. The question still remains for students of economic change; where does one begin and how does one proceed to bring about economic development in these more power and tradition oriented cultures? 51

VIII. DOMINANCE, POWER, AND THE SOCIAL SYSTEM Deutsch spoke about conceptualizing power in relation to the social system. He presented a matrix in which the abscissa represented goal attainment. He defined goalattainment as achieving some"external end" in the objective world. He said that a person or social system never seeks only one end. Therefore, in the model he was describing, a variety of goals are present. The dimension of goal attainment or objective power was also called "effectiveness" by Deutsch. He talked about this dimension as the probability of attaining a concrete end in the external world. A second dimension, the ordinate of the matrix, was referred to as the probability of converting another actor's behavior into an extension of one's own self-system. This, Deutsch called "subjective power" or the integration function. He then characterized forms of power behavior by using the matrix. After describing the matrix Deutsch added the following ideas on power and dominance: (1) Power is a function of the communication channels in a social system. It is also a function of the education of the members in the social system. Power is dependent on one's ability to form and maintain a goal fulfilling program-a program is maintained through education, and this, in turn, is achieved through communication. Social systems endure in the same form over many years only when the powerful people in them have educated the people to accept their dominance as legitimate. (2) Power is dependent upon memory. A system, whether it be an individual or a social system, with poor memory cannot retain power. (3) There is a contradiction between maintaining power and taking in new information. Deutsch said that an inverse relationship exists between the power and learning capacity. Learning, he said, is equal to a change in the inner resources of a system and occurs as a response to alteration or messages from the outside world. Adlai Stevenson, for instance, lacked power because he constantly needed more information. A military action leader often fails, not because he lacks power, but because he lacks information. The very power that he employs keeps him from a great deal of information. Thus, power in systems or in individuals often breaks down because of lack of information. (4) There is a conflict between power and rationality. Often one in power cannot make completely rational decisions. The effect on rationality of various power situations should be studied. 52

(5) Power implies greater specialization and greater specialization leads to less adaptability. A heavy committment to control over resources reduces the adaptability of the power structure to new resources. (6) Power is a symbol for a relationship. It is semiquantitative and it shows a similarity to other currencies. At this point Deutsch talked for some length on the analogy of power to money. Money and power, he said, are important for getting scarce resources. They can be seen as being parallel in many instances. Money and power both gain control over skills and resources, and conversely, when one has control over skill and resources he is wealthy, according to Deutsch. Wealth can be used both in terms of power or money. Money procures things not usually available or not easily available. If a goal is scarce, money becomes valuable. In both of these instances we can substitute the term "power" according to Deutsch. Newcomb pointed out that the analogy between power and money is limited, however. Deutsch agreed and said that we must first find the similarity between two concepts before we go into the differences. If we start with the differences we probably will not perceive all of the minute differences. Deutsch completed remarks by expressing the need for empirical measures of the concepts discussed during the symposium. For instance, we need to find ways of measuring "goal attainment." How can we quantify compliant behavior? How can we measure the probability of obedience in a social system or an individual? How effective is terror as a method of power manipulation? Do we know how much support a dictator needs from the population before he will be able to run the country? What happens when terror is used? Deutsch felt that terror has severe limitations as a strategy of power. Cartwright said that he also felt that the means by which power is used are very important for research. In response to a question about the effect of terror, Deutsch answered that terror probably causes three responses in a population: (1) intimidation, (2) increased recruitment of opposition to the power elite, and (3) some identification with the aggressor. Deutsch felt that intimidation would occur in the highest percentage of cases, that recruitment of the opposition would occur next most often and that identification with the aggressor would occur least often. Deutsch completed his remarks by saying that the government in a social system exists in order to form, institutionalize, and teach habits which will make the system more stable. A social system will not be stable or a government will not remain in power unless people internalize rules. A system will not persist unless, through the communication media, the people are educated and learn to follow a set of norms, rules, and laws. 55

IX. NEEDED RESEARCH ON DOMINANCE AND POWER The participants divided into three groups again in order to discuss the sorts of research or further studies which would contribute significantly to dominance and social. power,. Each group came up with a list of topics. GROUP I (1) More studies of children in all aspects of dominance and power should be done. Group I called for studies of play, on the implications of animal research for the behavior of children, and on delinquency. (2) Further research is needed on measurement techniques. Such research would involve a look at the kinds of data which various techniques yield. Attempts should be made to perfect the existing techniques, experiment with new ones, and continuously try to find better ways of collecting data~ One technique which needs considerably more work, would be the participant-observer~ Furthermore, we should find techniques for studying a variety of groups and organizational settings where power is manifested, eogo, business, and the military~ Finally, research techniques for studying power cross-culturally need more development, (3) The relationships between conflict and power should be studied. These relationships are important because generally power is one way that conflicts in social organizations are resolved. Studies should be carried out on conflict solutions that are related to different types of dominance and powero (4) Changes in the types of people, who aspire to or who get into power positions, should be studied. There probably are differences in the kinds of personality types that have arisen to power at different points in history according to the socio-cultural characteristics that existed at those times. (5) The situational and ecological determinants of power should be studied. Most of the time in the Conference had been spent in looking at the motivational determinants of power striving. There are situational and organizational conditions which tend to influence people to seek power. (6) The importance of the frame of reference for the self-image of people in different power situations should be studied. For instance, politicians often need to ignore what some people will think of them in

order to gain power. An increase in the dependency one has on a certain frame of reference should decrease the amount of power he can achieve. A study of the changing reference groups of a power seeker is needed. (7) Language as a source of detecting forms of dominance and power should be studied. Kobe's comments about the Japanese language especially in the feudal days, was an important contribution here. Kobe thought it would be more important to study the spoken tongue than the written language. (8) Identify and study the malignant power seeker, looking deeply into his conscious and unconscious motivations and his aocio-psychological backgroundO (9) A study of economic development and its relation to power motivation should be done. Which is causal, the motivational condition or the economic situation of the nation? Another related question is how do entrepreneurial types develop? Why do Japanese, Chinese, and Germans have different orientations toward achievement and power? What are some of the socio-cultural determinants of these different orientations? (10) The differences in motivation for power, before and after the sources and skills of power are available, should be studied. (11) An examination should be made of those people who are easily dominated; how they develop, what their personality dynamics are, and how such dynamics might be changed. GROUP II (1) Studies concerning the evaluation of power by persons or groups who exercise power or those who do not exercise it, should be carried out. These would include national surveys of attitudes about power. Furthermore, power as a concept should be studied from the point of view of different groups in the society. (2) The studies of personality determinants of dominance and power are needed and very important. Such studies would focus on the developmental conditions or the socio-psychological factors which lead to the occurrence of power motivation in individuals. Furthermore, the combination of characteristics in the leader and the led which increase the probability of a dominance relationship, should be studied. This would involve looking at dominance as a relationship and studying two sets of motives. The first would concern power striving, the second would concern submissiveness needs or the desire to be led. 55

(3) Studies of individuals in natural social groups should be carried out in order to understand the development of dominance and power relationships in such groups. Such organizations as military groups, college dormitory groups, and fraternity groups could be studied. If such studies were carried out, some leverage could be gotten in a fast way on the way social structure and power merge in a social organization. (4) A great deal of descriptive data on power structures of groups and organizations of different sizes are needed. Most of the studies on power structures have been done on communities. Studies of power in organizations and bureaucracies should be carried out also. (5) Changes in the functions of power, both past and anticipated, as a consequence of industrial and social changes, should be studied. Such questions can be studied as, "How will power be manifested in the family during industrial changes? How will power be manifested in small clubs; how will it be manifested in social organizations; and how will it be manifested in nation states as industrialization occurs?" (6) Studies concerning the concentrations of power and how people get power in organizations should be carried out. Such studies would focus on the ways in which people get power over the needed resources of others. (7) Studies on power styles or strategies should be done. These could be carried out in experimental games such as those used by Rapoport. In the experimental games, power styles would be studied as a function of personality variables. GROUP III (1) Research should be done on developing measures of power motivation and there should be further exploration of the individual and social correlates of power motivation. (2) Studies should be done on the determinants of the style of power which an individual useso Such studies would be concerned with the determinants of strategies of power. (3) Biophysical and biochemical studies should be done on the biological correlates of dominance. Studies on both physique and psychosomatic illnesses would be appropriate. Furthermore, studies on the genetic factors leading to various forms of dominance in animals would be useful. (4) Dominance in various occupational fields could be studied. Here one would look at the forms that power takes in various occupations.

(5) Personality studies of people in various social positions of dominance and power should be carried out. Instead of studying the developmental characteristics of power motivation here one would study those people who are already in positions of authority, dominance, and power. (6) Personality studies should be carried out on submissive people or people who are easily dominated, (7) Studies of cultures which contrast in their emphasis on dominance and power should be done. Such studies would investigate the socio-cultural characteristics of different nations as determinants of different styles of power and different evaluations of power. (8) Cross-cultural comparisons should be made on the way in which dominance is carried out. Three additional lists of recommendations for further research were generated by Atkinson, Kaplan, and McClellando Atkinson said that it is impossible to talk about an individualts dominance or power relationships outside of the social situation. Therefore, even though his recommendations pertained to studies on personality dynamics, dominance and power must be seen as socio-psychological in nature. Dominance should be conceived as a relationship between individuals in a particular social situation. The basic question which Atkinson saw for research was, "What combination of personality characteristics determine the nature of a dominantsubmissive relationship between two people?" He enumerated the following kinds of research objectives: (1) Develop measures of power motivation and related dispositions in leaders and the led. (2) Make comparisons of the personality characteristics of people in dominant and subordinate positions. (3) Study the parameters of the arena in which individuals express dominance and power. (4) Study the means by which power is expressedo For instance, one might use experimental games to study the way in which different individuals plan their strategy and employ power. (5) Study the biological and biochemical correlates and genetic factors of dominance and power. (6) Study the changes in types of people who aspire to, and attain, dominance at various times in history. 57

(7) Study the malignant power seekers, the malignant submitter, and the pathological withdrawer, (8) Study the patterning of people in dominant and submissive relationships. Atkinsonts recommendations characterized the area of personality or individual dynamics in relation to dominance and power. Deutsch commented that one could take all of Atkinson's recommendations and add parallel ones from the point of view of social systems or interpersonal relationshipso For instance, where Atkinson recommended the development of measures of power motivation and related dispositions in the leader or the led, the social system theorist would recommend the development of measures of power behavior among power elites and the masses. Kaplan attempted to summarize those research problems which cut across individuals and systems, He was looking for miscellaneous items or catchall categories. The ones that he came up with included: (1) Research on the measures of power is needed. This includes both conceptual and definitional problems. For such research, an interdisciplinary team would be valuable. (2) Kaplan also mentioned the importance of studies on the strategies of power. Here he had in mind those formal patterns or informal styles which individuals use as they attempt to manipulate others. Furthermore, he felt that some of the mathematical models, such as those employed in graph theory, could be used for an understanding of the formal properties of strategies, (3) Kaplan pointed out that research tenchiques need much more development. Techniques, such as participant observation, interviewing, or various other forms of measurement, need to be worked on considerably more especially in studies of power and dominance. (4) Forms of verbal and nonverbal communication as they are significant in power relationships should be studied. This would involve the social scientist with oral idioms and in looking at the models of verbal expression in power relationships. The material offered by Kobe on the Japanese language served as an example. McClelland tried to gather together some of the research ideas which had to do with a variety of points of view. All of these, however, included the notion of the determinants and consequences of power. (1) Studies should be done on the various ways in which power is defined and used in different social settings and cultures. 58

(2) Studies should be done on evaluations of power relationships by the social scientist as well as people in general. (5) Studies should be done on the development of power motivation and power behavior in individuals. These should be carried out especially on children. Furthermore, studies should be done on natural groups such as fraternities and military groups observing the dominance and power structures as they emerge. Finally, experimental groups should be manipulated in such a fashion so as to understand the development of dominance and power structures in groups. (4) Studies should be done on the effects of certain ecological and social settings on power motivation in individuals. (5) Historical studies should be done in which changes of the sociocultural and economic forces in a society, and how they relate to changes in achievement, power, and affiliation motivations, are analyzed. Along these lines, changes in power orientation would be especially interesting in relation to changes in the socio-political structure of a nation. (6) The coding of the means or strategies of influence should be done more carefully in the future. Two phenomena would be observed: the formal properties of strategy, and the determinants of choosing one strategy over another. (7) Studies of power should be done relative to the performance characteristics of power. For instance, what are the skill requirements to gain power in various situations? In such studies one would observe the overt manifestations of power behavior. (8) Studies should be done on societal constraints on power. Systems of law, morality, religion, and custom set the limits on the way power can be used in a social system. The way in which these structures set the limits within a system should be studied more carefully. Legal restraints, for instance, are common in most modern societies, how these restraints curb motivations for power is a basic question for research. 59

X. FUTURE ACTIVITIES, CONCLUSIONS, AND SUMMARY The participants looked to the future by proposing some possible ways of pursuing studies of dominance and power. They came up with several suggestions. (1) Financial resources should be made available to a variety of researchers from different disciplines instead of one man or one single team. (2) Some sort of a task force is needed for systematizing and coordinating the materials on dominance and power. One idea was a specialized information center, about which Cartwright spoke. The specialized information center would coordinate and disseminate knowledge on dominance and power. Social scientists would send their published data and materials to such a center. The center would characterize, coordinate, and finally disseminate these materials to other interested scientists. (3) Deutsch said that perhaps three different kinds of research groups could be set up. The first of these would work on motivational problems or the problems of personality and power. The second group would do studies on small groups and organizations as related to power; the third such group would work on political systemso He suggested that the three groups could meet separately over time with some liason and coordination. For instance, the three groups perhaps could get together once a year to pool the information which they were accumulating during that year. (4) It was suggested that experts in the fields of anthropology, psychology, sociology, philosophy, and political science be found who are interested in the areas of dominance and power and that financial support be offered to these men for pursuing their research interests in more detail. (5) A model, like that adopted by the Russell Sage Foundation, was discussed. Such a model might be followed by a group of researchers with the aid of a financial pool, doing some research themselves, and at the same time seeking other independent researchers throughout the country and offering funds to them for relevant research. Perhaps the specialized information center could also do some work on looking at the proposals of people throughout the country for their own independent research on power and dominance Likert pointed out that this Symposium could be viewed as a broad planning session, and that other kinds of conferences should follow it. Deutsch felt that at least two kinds of conferences would be appropriate. 60

The first would focus on measures of power motivation and power behavior. Such a conference would emphasize some of the empirical methods which might be used to measure the concepts of dominance and power. The second conference would take place some two or three years after the present Symposium. The topic for such a conference would be "Sharing the Pool of Data on Power" which have accumulated up to that point. McClelland indicated some skepticism about conferences without some more planning and funds for research. Cartwright indicated that some sort of an integration function should be performed by a few people-something like what had been called up to this point a specialized information-exchange center. Such a group of people would work on the research literature on power and dominance in an attempt to bring some order to it and to indicate its strong and weak points. Cartwright said that such an exchange center would not be just a. repository. Operations would be done on new data, or analyses would be made of data which had already been analyzed and published. McClelland felt that at least three tasks might be valuable to carry out. The first of these would be building a bibliography, which would be the sort of thing that Cartwright and Schmuck did for the Conference. Here, a person or a group would scan the relevant social science literature, seeking out studies on dominance and power, putting them together in some systematic way, and finally criticizing them. A second task that might be done would be to commission monographs on various phases or problems on dominance and power. This particular function, McClelland felt, was not as important as building a bibliography. The third notion would be to commission research on power and dominance. McClelland felt that the latter would be the most important thing that the group could do. Deutsch felt that at least three sorts of tasks might be done with proper funding. He agreed that the first task that should be done is a. review of the research that has been carried out up to this point. Here, one might commission people in the biological sciences, anthropology, political science, psychology, sociology, and so on, to do reviews which are relevant to their particular areas of interest. At the same time, researchers from different nations or different parts of the world could be commissioned to do reports on research being carried out in their part of the world. The second sort of task that might be done would be to finance research projects. Such finances would be granted to those people who propose projects which appear to a qualified committee to be quite relevant to the problems of dominance and power. Finally, the money could be used to support the preparation of monographs which report research that has been done on dominance and powero Cartwright underlined the importance of the integrative function. He said that this has often been the point at which support from foundations has fallen down. He referred to a project by Raven, supported by the Office 61

of Nav'al Research. Raven did an annotated bibliography on research on small groups and after the report was finished no funds were available for its dissemination. As a consequence, Raven's bibliography and cards which he had punched in the preparation of that bibliography are not being disseminated to c-her social scientists. Cartwright emphasized the importance of having fuknus available for the dissemination-integration of information on dominance and power. Cartwright went on to point out two basic ways to stimulate interest in the field of dominance and power, The first of these would be to have people who are interested in various fields, such as philosophy, psychology, anthropology, and sociology, put together materials in their fields on power and dominance, to criticize those materials, and to try to find implications for future work, Secondly, he felt that a committee should make money available for research. He added that funds should be made available for many different types of researches at different levels, and different career stages-pre-doctora.l or post-doctoral, and older, more experienced researchers. He felt that perhaps it will be best to start out with the older, more experienced researchers, and then to make more funds available in later years to younger candidates. The participants decided that they needed a. summary statement for the Conference which could be circulated to newspapers and other news media.. Cartwright said that it should be pointed out that better scientific knowledge will contribute to the solution of many of man's problems in relation to man. The illumination of this area of dominance and power should be especially worthwhile in solving some of the dilemmas man has faced through history. Further, he said that social science has now reached a level of sophistication where research on these problems can be useful. Kaplan said that he was surprised to learn during the course of the Conference, the little that is known in this area of dominance and power. Dahl supported Kaplan's conclusion by saying that only small gains have been made in recent years and there is much farther to go in understanding the dynamics of dominance and power. Likert pointed out that for the small amount of time that has been spent on the problems of dominance and power, social scientists have done well, but there is much more to be done. Redl emphasized that this is an area. in need of cross-disciplinary work. Anthropologists, sociologists, political scientists, psychologists, philosophers, and so onY should be encouraged to work on the problems. Deutsch mentioned three points that could be included in a. summary statement. First, that power and dominance are interrelated and interdependent processes quite susceptible to clarification by research~ Second, that the problems around power and dominance should be studied both at the individual and social-system levels of analysis. At this point social scientists are well equipped to start such a study. Third, that operational 62

definitions for power, dominance, and related processes, are now available, and scientists from various fields can bring their special research knowledge to bear on these problems. Fitts added a few other points. First, he emphasized the reciprocal relationships involved in the power relationship. He said that social scientists must study both the motivations of theleader and the motivations and cognitive processes of the followers. Second, at the system level, social scientists should look at organizations from the point of view of how they create positions in which men gain power or are trained to be power seekerso Finally, Fitts added that psychologists need to study power in relation to other motives such as achievement and affiliation. The complexity of motives, especially in leaders should be researched, since power motivation may not be the simple or sovereign motive involved, After considerable discussion, a summary statement was completedo This statement stood as a general summary of the conference and read as follows: "The Symposium participants agree that the study of dominance and power in society is of major significance in the further understanding of man's relation to man. Man often strives to dominate or control the behavior of others who in turn accept or resist such regulation. These power relations have important consequences in family and community life, in business and politics, and in national and international affairso "The study of dominance and power offers a means of comprehending better the origins of individual motivation and the nature of human developmento It reveals ways in which socialized man places controls over irresponsible uses of power by others, and accepts legitimate authority. In the long run, results of research in this field should indicate how this knowledge can be applied for useful ends, including the development of better leaders and more effective institutions on the one hand, and the reduction of crime, delinquency, and malignant forms of expression of power on the other. "To date, though much valuable research has been done on this topic in many fields of science, the degree of understanding so far achieved is relatively small in relation to the amount of knowledge needed for real comprehension and application, and in proportion to the potentialities of available methods of scientific studyo "The Symposium, therefore, recommends more intensive support of research by anthropologists, political scientists, psychologists, sociologists, and others on the subject of dominance and power as a. broad field calling for the approaches and techniques of many social 63

and behavioral scientists. It is proposed that this research be undertaken with such objectives as the following: "(i) Analyzing the conceptions of power as used by historians, scholars, theorists, and empirical scientists. "(2) Improving measures of individual motivations regarding the exercise or acceptance of dominance or social power. "(3) Developing useful measures of the functions and limits of power, and the means of exercising dominance in social and political systems, and at different stages in history. "(4) Analyzing the origins and development of power motivation in individuals and of power needs in social systemso "(5) Studying various manifestations of social power throughout the individual's lifespan, and in various cultures and social insitutions, and the results achieved from the use of different forms of power. "(6) Determining the consequences of power in various interpersonal, social and political contexts, and 1(7) Suggesting ways in which the motivation for dominance may be chanelled and redirected for socially desirable purposes." 64