U.S. DIPLOMACY

Presidential Decisionmaking Page xiii Preface

This is a study of presidential decisionmaking in foreign policy, a subject that has absorbed me since I was a graduate student in the Political Science Department at the University of Chicago in the days before Pearl Harbor. In a seminar on the psychology of politics taught by Nathan Leites I became interested in the role that a leader's personality can play in his political behavior. My research paper for the seminar was a study of Woodrow Wilson from this standpoint. Some years later, with the invaluable help of my wife, Juliette L. George, whose contribution to the enterprise went well beyond that of coauthor, the seminar paper became a biography, Woodrow Wilson and Colonel House: A Personality Study (New York: John Day Co., 1956).

By this time I had become deeply involved in foreign policy and national security research, first as a research analyst in U.S. military government in Germany and then as a member of the Social Science Department of the Rand Corporation. A continuing challenge of the many years spent at Rand was the task of bringing scholarly analysis to bear effectively upon high-level policymaking. At Rand policy research was not defined exclusively in terms of narrow applied research. Emphasis was given to the need for basic research that would improve the knowledge base and contribute to the development of theory and methodology. Before leaving Rand in 1968 to join the Political Science Department at Stanford University, I had started a project aimed at bridging the gap between theory and practice. One of its objectives was the development of policy- relevant theory in well-defined issue-areas of foreign policy. My premise was that policymakers need theory and in fact do use it in making decisions, though often they operate with primitive theories of questionable validity that are, moreover, not well articulated. Another Page xiv Preface objective was to understand better the rules of thumb and shortcuts that political leaders resort to in their efforts to make rational decisions when laboring under the well-known "cognitive limits" on rationality. These interests, which originated while I was at Rand, led eventually to a number of books and articles on related subjects: coercive diplomacy, deterrence, crisis management, the "operational codes" of political leaders, multiple advocacy, and policy- relevant forecasting.

In 1970 I attended the Conference on Coping and Adaptation held at Stanford University under the sponsorship of the Department of Psychiatry with the support of the National Institute of Mental Health. The conference stimulated me to view political decisionmaking from a different perspective, and I wrote a paper, "Adaptation to Stress in Political Decision Making: The Individual, Small Group, and Organizational Contexts," for the volume based upon the conference proceedings. This early paper, in fact, provides an important part of the framework for the present book.

In 1974 Peter Szanton, then research director of the Commission on the Organization of the Government for the Conduct of Foreign Policy, a joint presidential-congressional committee, asked me to undertake a study for the commission on a topic that it had entitled "Minimizing Irrationality" and defined as follows:

Recent work in several disciplines provides new insight into the tendencies of personal and bureaucratic factors (and in the case of crises, physiological and additional psychic factors) to distort the judgment of decision makers. Drawing on recent work in the political, behavioral and psychological sciences, this study would address two questions: (1) to what extent are current organizational, procedural and staff arrangements unnecessarily vulnerable to such pressures; (2) what alternative arrangements might either shield decision makers from such pressures or open their deliberations to others less likely to be affected by them? Answers would be sought as to arrangements both for response to crises, and for more routine decision making.

Additional definition and refinement of the study was achieved as a result of a planning conference held at Stanford University in mid-June 1974. This conference was attended by the research director of the commission, Peter Szanton, and seven consultants. They were Professor Graham Allison, Harvard University; Professor Paul Hammond, then head of the Social Science Department, the Rand Corporation; Professor Charles Hermann, associate director, the Mershon Center, Ohio State University; Professor Ole Holsti, Political Science Department, Duke University; Professor Richard T. Johnson, Graduate School of Business, Stanford University; Professor Thomas Schelling, Harvard University;

Page xv Preface and Dr. Richard Smoke, then a post-doctoral fellow with the Institute of Personality Assessment, University of California at Berkeley. Others who were invited but could not attend were Professor John Steinbruner, then at Harvard University, and Professor Robert Jervis, Political Science Department, UCLA.

The consensus that emerged from the planning conference was that the intended purposes of the study would be best addressed by focusing on possible impediments to optimal (or effective) information processing within the foreign-policymaking system in the executive branch, and that attention should be focused upon the ways in which these impediments might affect the quality of major decisions made at the highest levels of policymaking. While this definition of the study did not reject the possibility that gross irrationalities of the type that concern professional psychiatrists may occur in individuals who participate in policymaking systems, it directed attention instead to the variety of other types of impediments that can distort information processing and the evaluation of alternative options.

The study for the commission, to which a number of other scholars contributed, was entitled "Towards A More Soundly Based Foreign Policy: Making Better Use of Information." It was issued in 1975 as Appendix D in Volume 2 of the Appendixes that were published for the commission by the U.S. Government Printing Office. Having been encouraged by a number of acquaintances to prepare a publication based on this report that would be more readily available for wider use, I revised and updated it in the past year.

The present book differs from the original report in several respects. It includes a new introduction, "The President's Search for Quality Decisions-A Framework," in which I discuss in more detail the difficult problem of trying to understand the role of various personality variables in an executive's decisionmaking and leadership behavior. Chapter 7, "Presidential Management Styles and Models," is also entirely new. Several chapters have been considerably expanded. Chapter 5 provides a more detailed discussion of the effects of bureaucratic politics on the quality of information and advice that reach the president. The nature and role of policy-applicable theory (Chapter 13) are now discussed in detail and are illustrated with reference to crisis management, coercive diplomacy, deterrence, and d6tente.

Chapter 3, "The Importance of Beliefs and Images," has been revised to include new material on the relevance of recent developments in cognitive psychology and attribution theory for the study of political decisionmaking. Less- extensive revisions and additions have been made in other chapters.

I keenly regret that limitations of space have made it necessary to drop Page xvi Preface from the present version of the study two important specialized chapters that appeared in the original report. These are David K. Hall's chapter on "The 'Custodian-Manager' of the Policymaking Process," which is on a forthcoming Ph.D. dissertation in political science at Stanford University, and the stimulating proposal "Maintaining the Qual Decision-Making in Foreign Policy Crises," prepared by Marga G. Hermann and Charles F. Hermann.

It is perhaps useful to add a word or two about the nature of this study, which attempts an ambitious synthesis of a great deal of material that bears in one way or another on the tasks of understanding and proving the quality of presidential-level decisionmaking in the sphere of foreign policy and national security. A tremendous body of significant research and writing on decisionmaking and on the related task developing more effective policymaking procedures has accumulaed in the last several decades. This literature is voluminous and often hightly specialized. Those who have contributed to it include not only persons with practical experience in foreign policymaking, but also many specialists in the behavioral sciences--psychologists, political scientist, sociologists, economists, organizational theorists, and decision theorists. They have utilized many different kinds of data and have employed variety of research approaches in studying the behavior of individual decisionmakers, small groups engaged in problem solving or policymaking, and complex organizations.

Much of this specialized literature has been examined in preparing this study, but the use made of it has necessarily been selective. Judgmenthas been exercised regarding the relative significance of different studies for present purposes. No doubt this judgment is imperfect and will be questioned. Moreover, despite a conscientious effort to locate all sources con taining significant information, undoubtedly important materialt been missed.

The reader will want to keep in mind that while this study has benefited greatly from the contributions of academic specialists, it is not written primarily for the community of research scholars. References to relevant literature and footnotes are not lacking in this report, but they have been kept to a minimum.

It was my good fortune to have had the assistance of many others in planning and preparing the original report for the commission. In addition to the contributions to that report by David K. Hall and Margaret and Charles Hermann, Professor Robert Keohane, my colleague at Stan ford University, coauthored with me Chapter 13, "The Concept of Na tional Interest." A briefer version of what is now Chapter 14 was prepared by Dr. Richard Smoke. In addition, Joseph Atkinson, then of Page xvii Preface Yale University, and Professor Anne McMahon, then of the Sociology Department at Stanford, assisted in the preparation of Chapter 4.

Altogether, over twenty persons read one or more chapters of the original report and prepared comments. They are Joseph Atkinson, Lincoln P. Bloomfield, Davis Bobrow, Thomas V. Bonoma, Chester L. Cooper, Eric Davis, Lloyd S. Etheredge, Stephen Genco, David Hamburg, Grant Hilliker, Ole R. Holsti, Irving L. Janis, Robert Jervis, Curtis Kamman, Richard Lazarus, Joshua Lederberg, Joseph S. Nye, Jr., E. Raymond Platig, Austin Ranney, Robert L. Rothstein, John Stein- brunet, and Aaron Wildavsky.

I wish to express my appreciation to the following persons for their willingness to be interviewed when I was collecting additional data for the commission report and for sharing with me their insights into problems of policymaking or for calling relevant materials to my attention: Drs. Herbert J. Spiro, Carl Nydell, E. Raymond Platig, Pio Uliassi, and Sven Groennings-all of the State Department; Professor Arnold A. Rogow, Graduate School and University Center, City University of New York; Drs. Leonard Wainstein and John Ponturo, Institute of Defense Analysis; Drs. Andrew Marshall and Alfred Goldberg, Department of Defense; Dr. Irving M. Destler, the Brookings Institution; Dr. Bertram S. Brown, director, National Institute of Mental Health; Dr. Francis Barnes, M.D., and Dr. Bryant Wedge, M.D., both of Washington, D.C.; Dr. William Davidson, M.D., Institute for Psychiatry and Foreign Affairs, Washington, D.C.; Dr. Chester L. Cooper, Woodrow Wilson International Center, Smithsonian Institution; Professors Henry S. Rowen and Alain C. Enthoven, Graduate School of Business, Stanford University; Drs. Vincent P. Zarconne and Jared R. Tinklenberg, Psychiatry Department, Stanford University.

It is a pleasure to acknowledge my longstanding indebtedness to Dr. David A. Hamburg, M.D., formerly head of the Department of Psychiatry and Behavioral Sciences, Stanford University, and currently president of the Institute of Medicine, National Academy of Sciences, for introducing me to the study of problems of stress and coping, and for the help he has given me in conceptualizing certain aspects of the present study and in assisting me with some of the interviews.

For unfailingly cheerful and efficient secretarial and administrative help I am indebted, as so often in the past, to Mrs. Lois Renner and Mrs. Barbara Sullivan.

It should be noted, finally, that some important matters that can affect policymaking are not taken up in this report. We do not deal here with problems associated with the physical and mental health of policymakers, the effects of sleep disorders on cognitive processes, the

Page xviii Preface effects of aging on the performance of political leaders, the effects of drugs and medications taken for various purposes, the effects of "jet lag," etc. Preliminary data were gathered on some of these matters, but I became quickly evident that to provide useful and responsible analysis ofthese questions would require highly specialized competence and much more time and resources than were available for the present study. Give the importance of these matters, it is hoped that they will become the focus of a separate study at some point in the future. Alexander L. Geor

Page 1 Introduction:

The President's Search for Quality Decisions--A Framework

This book focuses on the task of improving the quality of information, analysis, and advice available to a president and his leading advisers for making important foreign-policy decisions. Before proceeding, however, it is well to recognize that the search for high-quality decisions in foreign policy, as in domestic policy, is often complicated by the need to be attentive to other considerations as well. Trade-Offs Between Quality, Acceptability, and Decisionmaking Resources

Like it or not, every president must struggle with the difficult task of attempting to harmonize his search for high- quality decisions with two other requirements for being an effective leader and decisionmaker. [Note 1] Thus, a president must be sensitive to the need to achieve sufficient consensus in support of his policies and decisions within his own administration, in Congress, and with the public. The amount of support required will vary for different presidential policies and actions as will a president's judgment of the minimal level of consensus that he needs on different occasions. [Note 2] One hopes that on many occasions there will be no serious conflict between the quality of a decision and its acceptability and hence no necessity for the president to make a difficult trade-off decision whether to sacrifice some of the quality of a preferred policy for greater acceptability or, conversely, to accept the risk of low acceptability and inadequate support in order to pursue a policy or action that he and his advisers regard as more advantageous from the standpoint of the national interest. But the need for minimal levels of support and consensus on behalf of foreign policy is an ever-present constraint on the president's ability to seek and to choose high-quality policies. [Note 3]

Similarly, a president's search for quality in his foreign-policy decisions must be sensitive to the constraints of time and to the proper use of Page 2 Introduction available policymaking resources. In many instances the search for higher-quality decision cannot or should not be prolonged insofar as the failure to make a timely decision may itself reduce the likelihood of achieving a successful outcome. Nor should the search for a higher- quality decision on one policy question be allowed to consume disproportionate share of the manpower, and the analytical and intelligence resources that must be available to attend to other urgent policy questions.

Given these important constraints on the president's ability to dedicate himself exclusively to the search for high- quality decisions, a number of dilemmas are encountered. Just as there are often trade-offs between th search for quality decisions and the need for consensus and support, so too, difficult trade-offs will be encountered between the search for greater quality and the prudent management of time and other policymaking resources. And, finally, it must be recognized tha presidential decisionmaking often encounters still another trade-off dilemma: that between a desire to achieve greater acceptability for the chosen policy and the expenditure of precious time and other policymaking resources that would be incurred in the effort to develop greater support for that policy.

We have identified three competing desiderata that impinge in varyinf degrees on presidential decisionmaking. They often create genuine dilem Figure 1. Trade-Off Dilemmas in Presidential Decisionmaking

Page 3 Introduction mas, forcing the executive to exercise difficult, often quite painful tradeoff judgments. An "effective" decisionmaker must be sensitive to all three desiderata; he is obliged to weigh their relative importance in any given situation; and he is called upon to exercise good judgment in making trade-offs among them. The resulting complexity of presidential decisionmaking is depicted in Figure 1. The Criterion of a "Quality" Decision

We have spoken thus far of the search for "high-quality" decisions without defining that desideratum. In the sphere of foreign policy this criterion is usually identified with the concept of "national interest"-i.e., a high-quality decision is one in which the president correctly weighs the national interest in a particular situation and chooses a policy or an option that is most likely to achieve national interest at acceptable cost and risk. Presumably, therefore, it is the criterion and proper understanding of national interest that should guide policymakers in their search for high-quality decisions. We will discuss later the serious limitations of the concept of national interest from this standpoint. [note 4]

Human nature and politics being what they are, the abstract concept of "national interest" has many competitors when it comes to the president's own subjective conception of a high-quality decision. The policy issues with which an executive must deal often arouse in him a variety of motives and interests that are in fact extraneous to the values associated with even a very broad conception of the national interest. Foreign-policy issues and the circumstances in which they arise may activate the policymaker's personal motives and values and bring to the fore his political interests or those of his administration or political party. This is not surprising, since the way in which the president deals with a particular policy problem may have important consequences for his sense of personal well-being or his political fortunes. Thus, whether he is fully cognizant of it or not, a president may stretch the desideratum of "high quality" to include consideration of whether the decision he makes will have one or more of the following consequences:

• satisfy or frustrate his personal values; • provide an outlet for expressing deep-seated motives and impulses; • obtain approval or disapproval from persons who are significant figures in his life; • enhance or damage his self-esteem; • advance or set back his career prospects; • strengthen or weaken his political and bureaucratic resources.

Page 4 Introduction

At times, these personal and political stakes in a foreign-policy issue may lead the president in the same direction as his objective conception of where the national interest lies. But it must be recognized that they also are capable of diluting, if not distorting, his search for a high-quality decision based on the criterion of national interest. At the same time, however, various safeguards against such possibilities do exist. In the first place, the president must justify his foreign-policy decisions with reference to the criterion of the national interest in a manner that is credible to his associates, to Congress, and to public opinion. His awareness of this necessity can serve as an important brake on the tendency to allow personal needs or political interests to dilute his search for quality decisions. Then, too, much of presidential decisionmaking is institutionalized in ways that reduce or contain such tendencies. The operation of organizational, procedural, and staff arrangements in support of presidential decisionmaking serves to structure and discipline a president's choices; thereby, they reduce-though certainly they do not eliminate-the possibility that personal motives and interests will intrude into and distort his judgment. [Note 5] The Role of Personality in Decisionmaking

This study does not reject the possibility that gross irrationalities of the type that concern psychiatrists may occur in presidents and other leading participants in policymaking. [Note 6] However, this possibility is not singled out for detailed examination. Instead, the present study directs attention to other impediments that can lower the quality of information, analysis, and advice available for making important foreign-policy decisions and that can distort the president's judgment in making such choices. Before proceeding with the major focus of our inquiry, however, some additional observations may be offered regarding the role of personality factors in high-level policymaking.

Many individuals--even those who enjoy remarkably successful careers in politics, business, academia, or some other occupation--operate with complex motivational patterns that often include deficits or vulnerabilities in their self- esteem for which they attempt to obtain compensation in the pursuit of their careers and in the day-to-day performance of tasks associated with their occupations. Moreover, all individuals employ psychological devices to cope with the anxiety, fear, or guilt that they experience from time to time in certain situations.

In fact, it often is not at all difficult to find evidence that an individual's character-rooted needs--whether it be an unusual need for affection, respect, aggression, rectitude, power, security, etc.-- or his ego Page 5 Introduction defenses are being expressed in the performance of tasks associated with his position. From this it is tempting to infer that the way in which that person performed was determined in part, if not wholly, by the personal needs or motives in question. While this may be a correct explanation on occasion, it is also important to recognize that the expression of personality needs is a necessary but insufficient condition for establishing the critical causal importance of those personality factors for the behavior in question. The fact that a political leader strives for or achieves gratification of personal needs in his performance of a political role does not automatically bestow decisive causal importance to the personality variable.

Performance in executive positions such as the presidency is sensitive to a variety of constraints. A given behavior may be no more than a response to the logic of the situation--i.e., a requirement of the situation rather than a requirement of personality. Or it may be a response to the expectations of others as to how an incumbent president should behave in a particular situation or in situations of a given type. Executive behavior, in other words, is sensitive to situational and role variables as well as to character-rooted needs or to psychodynamic patterns for adapting to stressful experiences.

In addition, it must be recognized that an individual's personality system itself includes more than character-rooted needs and ego defenses (such as projection, denial, repression) that are employed to cope with anxiety, fear, and guilt. Thus, an executive's political behavior will be shaped also by a variety of cognitive beliefs (ideology, world view, beliefs about correct political strategy and tactics, etc.) that he has acquired during the course of his education, personal development, and socialization into political affairs. In other words, much of an individual's behavior as a political decisionmaker will reflect what he has learned along the way either through direct or vicarious experience and will be shaped also by the values and behavioral patterns that he has acquired by modeling himself on prestigious persons.

Still another important component of the total personality of a political decisionmaker is the set of skills that he has acquired, which, he has come to believe, provide him with relevant tools and resources for effectively meeting the role demands of a political leader and an executive. These skills, and the successful use of them on previous occasions, provide the individual with a sense of personal efficacy for addressing at least some of the role tasks associated with executive leadership. These examples indicate that the task of assessing the impact of an executive's personality on his performance is complex and thereby caution against the common tendency to explain too much of an individual's per- Page 6 Introduction formance in terms of personality flaws. Personality itself, as indicated, is not a simple concept, but rather an intricate system comprised of the individual's character-rooted needs and motives, his sense of self (or "identity") and the various self-esteem vulnerabilities and needs that accompany it, the skills and adapative patterns that the individual has acquired for relating to other individuals and for coping with or mastering a variety of tasks and situations, and the cognitive beliefs and knowledge in terms of which he organizes his experiences, plans his activities, and diagnoses and responds to new situations. Awareness of the complexity of personality--whether its complexity is expressed in these terms or by using other concepts--should serve to discourage the tendency to explain behavior by going for "the jugular of the unconscious" and other forms of psychological reductionism that bypass attention to the situational, institutional, and role contexts in which the executive functions.

Having said this, however, it is still the case that much of value can be learned by studying presidents from the standpoint of the fit or lack of fit between various components of personality and the different role and situational requirements of the enormously complex job of being president. [Note 7] Fascinating evidence indicates that a number of individuals who served as president were aware to some extent, in some cases even to a considerable extent, of a lack of fit between some of their personality characteristics and some of the demands of the presidential role. Just as brilliant performance of some presidential tasks can be traced to a "fit" between personality components and role requirements, so, too, can inadequate performance of some presidential tasks be traced to "malfits" between personality and role. The example of William Howard Taft, a poignant one from a human standpoint, will suffice to illustrate the point.[Note 8]

The fact that Taft's personal qualities did not fit well the requirements of the presidential role was evident to him as well as to others even before he became president. He wrote to his wife in 1906, while he was a member of Theodore Roosevelt's cabinet, saying that "politics, when I am in it, makes me sick" (p. 76). Nonetheless, Taft's ambitious wife helped Roosevelt drag Taft into running as the Republican presidential candidate in 1908. Taft's eighty-year-old mother, on the other hand, knew her son well enough to write to him in 1907, trying to discourage him from agreeing to become Roosevelt's hand-appointed successor: she pointedly observed that "Roosevelt is a good fighter and enjoys it, but the malice of politics would make you miserable" (p. 76). This proved to be an accurate prediction!

Taft also had a temperamental antipathy to self-dramatization, and Page 7 Introduction this caused him simply to ignore reporters during his first weeks in the White House. Taft's abhorrence of drama or emotion discouraged him from trying to learn the techniques of public relations. He had no ability and little interest in shaping the news as an instrument of power and leadership. One biographer notes that at times Taft was reluctant to read the newspapers for fear that he would have to comment!

Taft also ruled out many other standard techniques of influence. He was loathe to use patronage or to appeal to the public over the heads of Congress. "His creed was--don't play politics, fight for what is right, and seek to be a source of unity rather than of discord" (p. 89).

For a variety of reasons Taft did not think that he could be a particularly effective president. He lacked a sense of efficacy--i.e., the skills and resources--relevant to many of the role requirements of the presidency. Not surprisingly, he succumbed to the temptation to cope with this lack of fit between personality and role by frequently lapsing into passivity and complacency. His friend Archie Butt wrote that Taft "believes that many things left to themselves will bring about the same result as if he took a hand in their settlement" (p. 92).

Psychologists note that individuals caught in an unpleasant situation from which they cannot remove themselves physically often resort to removing themselves psychologically. There are many evidences of Taft's effort to "escape the field" in order to minimize the unpleasant aspects of the job of being president. As one biographer puts it: "He ate

too much, wanted to sleep too long, preferred playing bridge to work, and was always reluctant to leave the golf course . . . . He loved to travel because it got him out of Washington and away from problems but complained that he was expected to make speeches on his tours" (p. 95).

If there was a lack of fit between certain components of Taft's personality and the requirements of the presidency, interestingly enough the opposite was the case when he became chief justice, a position for which he was much better suited temperamentally and for which there was a much better fit between personality and role requirements. Thus, as chief justice, Taft did not experience the self-doubts that had plagued him in the presidency. As chief justice, far from attempting to "escape the field," he worked harder and longer and was much happier than he had been as president. Taft was a much stronger and more effective chief justice because his motives, incentives, and skills were more fitted for a judicial rather than a political position. The interaction between personality, situation, and role can take many forms, and it is by no means the case that the impact of personality on performance of the executive role is always deleterious. In fact, the op- Page 8 Introduction posite may be the case. Thus, personal motives and needs aroused in a particular situation may serve to alert the individual to perceive more acutely the political opportunities (or threats) presented by that situation. The possibility that such a situation affords for gratifying a personalaity need may also help to energize the individual to meet difficult situational tasks and role requirements. As a result, personality factors may facilitate and improve the individual's performance of role re quirements.

There is, of course, a danger that the arousal of powerful personality needs may distort the individual's reality testing and his judgment of whether and how the situation threatens important policy objectives or offers opportunities for furthering those goals. Highly relevant in this regard is the strength and operation of the individual's ego controls-- that is, his ability to regulate the expression of personal needs, anxieties, and defenses in order to prevent them from distorting his effort to appraise situations realistically and deal with them appropriately. It is important to recognize that the effectiveness of ego controls is not to be gauged by their success in denying any expression at all to persoal needs in one's political functioning. This would be impossible in any case. Rather, what is more likely is that in the process of maturation most successful adult personalities have developed ego controls that are generally strong enough to curb the expression of personal needs unless the behavior that would gratify those needs is also consistent with the requirements of the situation and the demands of the role task. In other words, an individual's personality needs may be subjected to effectiive reality-testing and disciplined by that person's awareness of role and situational requirements before they are allowed expression in behavior. Personality needs can be said to have a dysfunctional, adverse effect on the performance of official duties only when there is reason believe that ego controls have failed and that the needs and motives have led him to a distorted or inadequate perception of role and situational requirements, or to a choice of inferior ways of meeting those quirements. [Note9]

In addition to this preventive control function, the so-called "ego component of personality has a positive function as well. During the course of an individual's maturation and development, he typically formulates a variety of constructive strategies for harnessing in more positive directions the expression of deep-seated personal needs and motives. [Note 10]

Some years ago, when political scientists were exposed to early developments in ego psychology, they were struck by the fact that Page 9 Introduction arousal of a leader's anxieties and ego defenses could severely impair his ability to deal rationally with a situation. As a result, political scientists--and some psychiatrists and psychologists as well -- tended to regard any display of ego defenses (such as denial, projection, or rationalization) by a political leader in a stressful situation as a telltale sign that his ability to cope rationally and effectively with that situation had been impaired. Decisions made under these circumstances were regarded with suspicion, and any inadequacy perceived in the substance of the decision was explained as being the unfortunate by-product of the leader's resort to ego defenses to cope with his anxieties. Explanatory hypotheses of this kind often oversimplified and distorted the role of unconscious emotional factors in decisionmaking. {Note 11]

Recent developments in ego psychology and in studies of the nature of coping processes offer political scientists a much more refined and discriminating understanding of these matters. Many of the classical ego-defense mechanisms

can be used constructively by an individual in the total process of coping. Defensive operations such as withdrawal, denial, or projection do not necessarily preclude intelligent and reasonable adaptation to a difficult situation. Rather, these defensive maneuvers may give the individual time to regroup ego resources and provide him with the short-run, tactical ego support that sustains him momentarily until he can return to employing more constructive ego capacities such as information seeking, role rehearsal, or planning in order to deal with the problem. [Note 12]

Psychological aspects of policymaking are important, but the present study will attempt to accommodate them within a social-psychological framework rather than by means of a narrow psychiatric approach. It must be immediately noted, however, that there does not exist a single social-psychological approach that deals comprehensively and in an integrated manner with all of the possible impediments to information processing in a complex policymaking system. Accordingly, by necessity the present study has had to formulate an eclectic framework of its own within which to discuss and evaluate the many different kinds of findings and theories that are relevant to one or another aspect of the overall problem addressed. [Note 13]

The first part of our framework identifies the five procedural tasks that must be well performed within a policymaking system if the top executive is to receive information, analysis, and advice of reasonably good quality. The second component of the framework provides a refined and differentiated model of the policymaking system that will enable the reader to see more clearly the sources of impediments to the performance of these five critical tasks. Page 10 Introduction Five Critical Procedural Tasks in Effective Decisionmaking

Certain procedures generally are considered to be desirable in decisionmaking. While adherence to these procedures does not guarantee high-quality decisions in any given instance, nor does the failure to do so necessarily result in a poor decision,, most specialists in decisionmaking believe that adherence to these procedures is likely to increase the probability of obtaining decisions of higher quality.

While specialists employ somewhat different terminologies in describing the desirable characteristics of decisionmaking procedure, there is substantial agreement that a policymaking process should accomplish the following tasks: 1. Ensure that sufficient information about the situation at hand is obtained and that it is analyzed adequately so that

it provides policymakers with an incisive and valid diagnosis of the problem. 2. Facilitate consideration of all the major values and interests affected by the policy issue at hand. Thus, the initial

objectives established to guide development and appraisal of optiion should be examined to determine whether they express adequately the values and interests imbedded in the problem and, if necessary, objectives and goals should be reformulated.

3. Assure a search for a relatively wide range of options and a reasonably thorough evaluation of the expected consequence of each. The possible costs and risks of an option as well as its expected or hoped for benefits should be carefully assessed; uncettainties affecting these calculations should be identified, analyzed, and taken into account before determining the prefer course of action.

4. Provide for careful consideration of the problems that may arise in implementing the options under consideration; such evaluations should be taken into account in weighing the attractiveness of the options.

5. Maintain receptivity to indications that current policies are not working out well, and cultivate an ability to learn from experience.

With these procedural desiderata of a good policymaking system in mind, we turn now to differentiating and refining the internal com[pmemts of a policymaking system so as to identify better the sources of impediments to the performance of these procedural tasks. Page 11 Introduction The Three Interrelated Subsystems of the Policymaking System: Individual, Small Group, and Organization

Efforts to find ways of improving the performance of a policymaking system will be facilitated by employing a more complex and differentiated model of that system that takes into account individual, group, and organizational behavior

within the executive branch. Efforts at rational calculation of policy take place in three interrelated contexts or subsystems within the policymaking system: the individual context (e.g., the chief executive, secretary of state); the small group context of the face-to-face relationships into which the executive enters with a relatively small number of advisers; and the organizational context of hierarchically organized and coordinated processes involving the various departments and agencies concerned with foreign-policy matters in the executive branch. [Note 14]

The investigator must take into account the fact that these three subsystems are interrelated. Thus, the individual- executive, the small decisionmaking group, and the organization units often interact with each other in producing foreign-policy outputs. Each of these three subsystems is capable of generating a special set of adaptive and maladaptive ways of coping with the cognitive and procedural tasks of decisionmaking. Part of the research task is to identify distinctive dynamic processes of this kind associated with each of these subsystems. It also will be necessary to consider how the distinctive dynamic processes associated with each subsystem can adversely or favorably affect information processing within the other two subsystems.

The task outlined here is a formidable one. Despite advances in relevant portions of theories of individual psychology, small group dynamics, and organizational behavior, the linkage and synthesis of these three theories is still primitive. A much more satisfactory integration of relevant elements of individual psychology and of group dynamics than is presently available will be needed eventually for identifying the variety of interaction patterns between the individual-executive and small group that can affect efforts to cope with the different kinds of stress that impinge on foreign-policy behavior. Much the same has to be said regarding the need for a better integration of theories of small group dynamics and organizational behavior and theories of executive leadership and organizational behavior. Within the broad conceptual framework that has been outlined, the study proceeds in two parts. In Part 1 the sources of impediments to information processing are identified and illustrated in (1) the dynamics of individual decisionmaking behavior, focusing in particular on the top ex- Page 12 Introduction ecutive; (2) the way in which the top executive structures the small policymaking or advisory group with which he interacts, manages its internal processes, and regulates its communications with outsiders; and (3) the patterns of behavior that emerge in the rest of the executive branch and how these patterns affect the flow of information and advice upwards to the top executive and his small advisory group and the implementation of top-level policies and decisions.

From these sources spring impediments to performance within the policymaking system of the five critically important procedural tasks. But knowledge of the sources of impediments is not sufficient by itself for practical purposes; what is needed, in addition, is a list of the symptoms of inadequate information processing to which these impediments give rise.

Since the sources of impediments cannot be eliminated easily, the practical task is to institute preventive measures to reduce or minimize the impact that they are allowed to have upon policymaking. For this purpose, knowledge of the symptoms of inadequate information processing and appraisal identified in Part One can be helpful. It can sensitize those who participate in policymaking and those who are responsible for maintaining effective procedures for policymaking to quickly spot indications that information processing has become defective and to take corrective action. Part Two of the study takes up various ways in which the policymaking system can be structured and managed in order to avoid impediments to information processing or at least to minimize their harmful effects on the workings of the policymaking process. Notes

1. This section draws upon two books by Norman R. F. Maier, Problem- Solving Discussions and Conferences (New York: McGraw-Hill, 1963) and Problem Solving and Creativity in IndivMuals and Groups (Belmont, Calif.: Brooks/Cole, 1970); Victor H. Vroom and Philip W. Yetton, Leadership and Decision Making (Pittsburgh, Pa.: University of Pittsburgh Press, 1973); Ole R. Holsti and Alexander L. George, "The Effects of Stress on the Performance of Foreign Policy-makers," in Cornelius P. Cotter, ed., Political Science Annual, vol. 6 (Indianapolis: Bobbs-Merrill, 1975), pp. 273-74. It should be kept in mind that a president's search for quality decisions in foreign policy proceeds in the context of an extremely complex set of duties and responsibilities. An incisive discussion of the president's overall "job description" is provided by Thomas E. Cronin, The State of the Presidency (Boston: Little, Brown & Co., 1975), Chapter 9.

Page 13 Introduction

2. Different types of decisions would appear to call for different ranking of these criteria. For example, a nuclear test ban treaty requires broad support in Congress and the Pentagon to be effective, its consequences unfold rather slowly, and it is easily reversible should other adherents be found in violation of its provisions; it therefore appears to place a high premium on acceptance, and less on quality, time, or expenditure of resources. In contrast, decisions such as those made during the Cuban missile crisis seem to require primary emphasis on quality and time, whereas concern for acceptance and expenditure of resources are probably less critical considerations, at least in the short run. The important point is that only in the most trivial decisions can policymakers maximize quality and acceptance, while minimizing the expenditure of time and resources. 3. For a fuller discussion of the need for a widely shared consensus on the fundamentals of foreign policy if the

president is to be able to conduct a coherent, consistent policy over a period of time, see Alexander L. George, "Domestic Constraints on Regime Change in U.S. Foreign Policy: The Need for Policy Legitimacy" (in press). 4. See Chapter 13. 5. This thesis is ably argued by Sidney Verba, "Assumptions of Rationality and Non-Rationality in Models of the

International System," Worm Politics 14 (1961):93-117. 6. This section draws from a more detailed discussion in Alexander L. George, "Assessing Presidential

Character," World Politics 26 (1974):234-82. See also A. L. George, "Power as a Compensatory Value for Political Leaders," Journal of Social Issues 24 (July 1968); and Lloyd S. Etheredge, A World of Men: The Private Sources of American Foreign Policy (Cambridge, Mass.: MIT Press, 1978).

7. A useful framework for conceptualizing the relationship between personality and role in terms of "fit" or "malfit" is presented in Edwin J. Thomas, "Role Theory, Personality, and the Individual," in E. R. Borgata and W. Lambert, eds., Handbook of Personality Theory and Research (Chicago: Rand McNally, 1968). See also Michael Maccoby, The New Corporate Leaders." The Gamesman (New York: Simon and Schuster, 1976).

8. The brief sketch of William Howard Taft presented here is drawn from the longer profile of him in Erwin C. Hargrove, Presidential Leadership: Personality and Political Style (New York: Macmillan, 1966), pp. 77-96. Personality variables are treated sensitively and, with the recent experience of presidential abuse of power in the Vietnam war and the Watergate affair in mind, also with reference to the problem of securing greater accountability in presidential performance in Hargrove's more recent book, The Power of the Modern Presidency (New York: Knopf, 1974). In this book, one of the most important revisionist reassessments of the strong, "heroic" model of the presidency, Hargrove reexamines and modifies his earlier assumption that effective performance in the presidency requires an individual with a "power-oriented" personality. On this point see also Richard Neustadt's post-Watergate reflections in his Presidential Power: The Politics of Leadership with Reflections on Johnson andNixon (New York: Wiley, 1976). 1 have discussed some of these issues in an Page 14 Introduction unpublished paper, "The 'Power Problem' of the Presidency" (February 1976).

9. Complicating factors also should be noted: (1) the role and situational requirements that impinge on a political actor may contain conflicting or ambiguous elements that make it more difficult for him to exercise effective control and constructive regulation of his personality needs; (2) the politician's role may itself include aberrant requirements that activate personality motives and needs that ordinarily are kept under control. In other words, as Willard Gaylin, M.D., has emphasized, the nature of politics and what it takes to be successful in politics-as in business-may attract sociopathic and paranoid personality types: "The capacity to be ruthless, driving and immoral, if also combined with intelligence and imagination, can be a winning combination in politics as well as commerce. . Sociopathic and paranoid personality traits that are most dangerous in people of power are precisely those characteristics most suitable for the attainment of power in a competitive culture such as ours" ("What's Normal?," New York Times Magazine, 1 April 1973).

10. As Brewster Smith, Jerome Bruner, and Robert White noted many years ago, the early psychoanalytic account of ego defense mechanisms "failed to mention the tremendous importance of constructive strategies [employed by 'normal' individuals] as a means of avoiding the vicissitudes that make crippling defenses necessary .... [They] often prevent things from occurring that might disrupt them or, more positively, they.., plan events in such a way [so as to] operate effectively " (Opinions and Personality [New York: Wiley, 1956], p. 283; see also p. 22).

11. For a systematic discussion of the circumstances under which ego- defensive needs are likely to manifest themselves in an actor's political behavior, see Fred 1. Greenstein, Personality and Politics: Problems of Evidence, Inference, and Conceptualization (Chicago: Markham, 1969), pp. 57-61.

12. For a fuller discussion, see George, "Adaptation to Stress in Political Decision-making: The Individual, Small- Group, and Organizational Contexts," in G. V. Coelho, D. A. Hamburg, and J. Adams, eds., Coping and Adaptation (New York: Basic Books, 1974). See also Holsti and George, "The Effects of Stress."

13. The framework employed in the present study draws upon that developed in the author's earlier publications: see George, "Adaptation to Stress." For a recent review of social psychological approaches to decisionmaking, see Samuel A. Kirkpatrick, "Psychological Views of Decision Making," in Cornelius P. Cotter, ed., Political Science Annual, vol. 6 (Indianapolis: Bobbs-Merrill, 1975).

14. It should be noted that we will not attempt to deal with still other subsystems that interact with the three noted here-i.e., Congress, public opinion, and allied and neutral governments. Similarly, while much of the analysis that follows presumably relates to decisionmaking in other national contexts as well, parts of it may have more limited applicability for other nations (e.g., in small countries that have small or virtually nonexistent foreign-policy bureaucracies).