Bridging the Gap Theory and Practice in Foreign Policy Alexander L. George Foreword by Ambassador Samuel W. Lewis UNITED STATES INSTITUTE OF PEACE PRESS Washington, D.C. GL 8029374 POLSC 10-11-93 For Emil and Irene and Lee and Jane Contents Foreword by Samuel W. Lewis ix Preface xiii Introduction xvii Part One The Gap between Knowledge and Action 1. The Two Cultures of Academia and Policymaking 3 2. The Role of Knowledge in Policymaking 19 31 Part Two The Inadequate Knowledge Base for U.S. Policy toward Iraq, 1988-91 3. Outcomes of U.S. Strategies toward Iraq 33 4. Reforming Outlaw States and Rogue Leaders 45 5. Appeasement as a Strategy for Conflict Avoidance 61 6. Why Deterrence and Reassurance Sometimes Fail 71 7. The Failure to Coerce Saddam Hussein 79 viii Contents 89 8. War Termination: Integrating Military and Political Objectives Part Three The Bridge between Knowledge and Action 105 9. Contemporary International Relations Theory 107 10. Types of Knowledge for Policymaking 115 Summary and Conclusions 135 Notes 147 Index 163 -- Foreword - Policymakers and scholars alike have long noted the existence of a significant gap between theory and practice in foreign policy. Indeed, some days the distance between the two seems more like a yawning chasm. One explanation for this gap lies in a conflict between the two different cultures of academe and government. From the standpoint of the policymaker, the scholar is “too academic,” all too often prone to abstraction and jargon. The academic can operate in a more relaxed time frame. The policymaker must nearly always act with imperfect information, before a fully satisfactory analysis is complete. He or she does not have the luxury of saying, “Other things being equal ...” Schol- ars, on the other hand, may complain that practitioners are too haphazard and ad hoc in their approaches to situations, and too ready to apply pat formulas or supposed lessons of history in uncritical ways. Practitioners place too much faith in intuitive judgment, scholars say, and may make simplistic generalizations. This gap between theory and practice in foreign policy is a subject that has long interested us at the Institute of Peace, and we have been especially fortunate to have Alexander George with us for the past two years as a distinguished fellow in the Jennings Randolph Program for International Peace. - ix x Foreword In this ground-breaking volume, George provides a pene- trating analysis of the many striking differences between the two cultures of academia and policymaking. He argues that while the gap between theoreticians and practitioners cannot be eliminated, it can be bridged. To that end, he identifies specific types of “policy-relevant knowledge” needed by the practitioner, and notes that scholars have not yet provided adequate conceptualization and general knowledge, drawn from historical experience, of many strategies and instru- ments of foreign policy. The lack of such knowledge, he dem- onstrates, was in part responsible for failures of U.S. policy towards Iraq in the period leading to the Persian Gulf War. A point that George stresses is that general knowledge about a strategy cannot substitute for, but can only aid, the judgment of the policymaker, who is often called upon to make difficult choices between competing considera- tions. Policy-relevant information can play an especially im- portant role during this phase, when the policymaker must weigh various options while at the same time taking into account other factors such as the need to muster public sup- port. But even after you build a bridge, there's no guarantee any- one will use it. We have a lot of work to do. I say “we” because it should be a shared responsibility—serious and committed scholars and those of us in relevant institutional roles need to work together to promote more meaningful interaction. We at the Institute of Peace can provide forums and catalytic sup- port. The academic community can provide more of the policy-relevant knowledge and intellectual frameworks that are needed. Practitioners can come to the table with an open mind about better utilizing all the resources available to them. In particular, we must concentrate on reaching the deci- sionmakers in ways that can get their attention. Scholars need to understand better the types of knowledge needed by poli- cymakers, look for ways to disseminate their research more effectively, and explore ways of conveying its practical impli- Foreword xi cations to senior policymakers, not just to mid-level analysts. We must together translate theory into practice on the tough new international agenda that lies before our nation. The cause of peacemaking demands nothing less. Samuel W. Lewis, President United States Institute of Peace Preface This book addresses the task of bridging the gap between theory and practice in foreign policy. This task requires me to identify the types of knowledge about international rela- tions that will be relevant and useful to those who conduct foreign policy. I have been preoccupied with this challenging task during much of my career, first during the years spent as a member of the RAND Corporation and since 1968 as a member of the Department of Political Science at Stanford University. I am grateful to the United States Institute of Peace for the award of a Distinguished Fellowship, which enabled me to pursue the project in Washington, D.C., from September 1990 through June 1992. I was delighted to find that the In- stitute shares a keen interest in developing scholarly knowl- edge for use in policymaking and works in constructive ways to encourage two-way interaction between academic scholars and policy specialists. My study has turned out to be somewhat different, and I think better, for having been pursued in the stimulating en- vironment of Washington. Closer proximity to the policy world forced me to reexamine and sharpen some of the ideas I entertained in the past. I believe I have a better understand- ing now, which I have tried to communicate in this book, xiii xiv Preface of the kinds of knowledge needed in policymaking and how such knowledge, when it is available, can contribute to policymaking More important, the preparation of the study and in par- ticular the chapters that assess the strategies the United States has employed toward Iraq since 1988 strongly confirmed' a long-standing concern that the state of existing policy- relevant knowledge is inadequate and that much additional scholarly research directed to producing such knowledge is badly needed. I was fortunate in preparing the study to have had the op- portunity to discuss U.S. policies toward Iraq with ten senior policy officials in the State Department, the Department of Defense, and the staff of the National Security Council. These individuals participated in and are knowledgeable about U.S. policy toward Iraq. They kindly read and com- mented on earlier drafts of chapters in part two of the study. I asked these policy specialists to tell me whether I had cor- rectly stated the facts and whether my analytical interpreta- tions of U.S. policy toward Iraq were reasonable. Their re- sponses to these questions were generally reassuring, and they offered additional information and useful judgments, which I have attempted to incorporate into these chapters. For understandable reasons, these officials—two no longer in the government-preferred to remain anonymous. I am indebted to Jane Holl, a specialist on war termination problems, for helpful comments on earlier drafts of chapter 8 and for allowing me to see several as yet unpublished essays on this topic. I benefited also from stimulating conversations with many foreign policy specialists who are not in the gov- ernment, though many of them previously were. Some of them kindly read and commented on earlier drafts of some of the chapters in the book. In alphabetical order they are Sanjoy Banerjee, Andrew Bennett, Robert Bowie, Dan Cald- well, Arthur Cohen, Eliot Cohen, Chester Crocker, Terry Deibel, Hugh DeSantis, Daniel Druckman, Arun Elhance, Preface XV Muhammad Faour, Juliette George, Ashraf Ghani, Richard Herrmann, Mark Hoffman, Stephen Hosmer, Fred Iklé, Martin Indyk, Bruce Jentleson, Robert Jervis, Michael Kre- pon, Steven Kull, James Laue, Joseph Lepgold, Samuel Lewis, David Little, Sean Lynn-Jones, Michael Mandelbaum, Ernest May, Alexander Moens, Joseph Montville, Patrick Morgan, Joseph Nye, Robert Pastor, Don Peretz, Alan Platt, Jerrold Post, William Quandt, Stephen Rock, Walt Rostow, Robert Rothstein, Shimon Shamir, William Simons, Richard Smoke, Louis Sohn, Janice Gross Stein, Stephen Stedman, Eric Stern, and I. William Zartman. If they read this book, I trust they will see reflected in it some of their observations and suggestions. I wish to express appreciation to Samuel Lewis, president of the Institute, and Michael Lund, director of the Jennings Randolph fellowship program, for their unflagging encour- agement of the project and insightful suggestions for im- proving the study. I also wish to thank Otto Koester and Joseph Klaits, program officers, and Barbara Cullicott, pro- gram administrator, for providing so supportive and conge- nial an environment for serious research; Dan Snodderly for his good-humored and incisive editorial suggestions; Blaine Vesely and Denise Dowdell for indispensable and efficient li- brary services; Mia Cunningham for her careful copy- editing; and Anne Cushman, Tarak Barkawi, and William Tanzola for their competent and cheerful research and secre- tarial services. Finally, as so often in the past, my wife, Juliette, provided indispensable support and understanding as well as insightful suggestions and comments. Introduction The central purpose of this study is to encourage better com- munication and closer collaboration between academic schol- ars who study foreign policy and practitioners who conduct it. Better communication requires a better understanding by both scholars and practitioners of the three types of knowl- edge needed in policymaking. In addition to reliable and timely intelligence about situational developments, policy- makers need (1) conceptualization of strategies—a conceptual framework for each of the many different strategies and in- struments available to them for attempting to influence other states. Policymakers also need (2) general, or generic, knowledge of each strategy, based on study of past experience that iden- tifies the uses and limitations of each strategy and the condi- tions on which its effective employment depends. Finally, pol- icymakers need a sophisticated, insightful understanding of each of the state-actors with whom they interact—what I shall refer to as (3) actor-specific behavioral models—in lieu of a dan- gerous tendency to assume that they can be regarded as ra- tional, unitary actors. These three types of knowledge are discussed in great de- tail in this study to make the following points: xvii xviii Introduction • Policymakers often operate with inadequate knowledge or erroneous assumptions or both regarding the strategies they seek to employ in the conduct of foreign policy. • Conceptualization and general knowledge of many foreign policy strategies and actor-specific models are inadequate in important respects and as yet poorly developed. • Much additional scholarly research in academic centers and within the government is needed to improve the knowledge base for foreign policy. Many policy specialists have a strong aversion to the idea that theory can have relevance and potential utility for pol- icymaking. In fact, as many scholars have discovered, the eyes of practitioners often glaze over at the first mention of the word “theory” in conversation. I have no desire in this study to convert policymakers into adherents of theory; in fact, I shall point out in some detail the weakness of academic theo- ries of international relations from the perspective of the types of knowledge needed in policymaking. However, it must also be said that good theories provide relevant and use- ful conceptual frameworks by means of which to understand the general requirements of a strategy and the general logic associated with its effective employment. Such theoretical- conceptual knowledge is critical for policymaking. And, as a matter of fact, all policymakers make use of some such theory and conceptual frameworks, whether consciously or not. That is, in employing a strategy, policymakers rely on as- sumptions, often tacit assumptions, about the strategy's gen- eral requirements and logic. The gap between theory and practice that forms the starting point for this study is not in policymakers' non-use of theoretical concepts, but in the fail- ure to analyze them more critically and to be more aware of the impact of theoretical-conceptual assumptions on policy- making I have discovered in my conversations with policy special- ists that if I avoid the word “theory” and speak instead of Introduction xix "generic knowledge” about foreign policy, they are much more receptive to the relevance of “generic knowledge" for policymaking. I believe this is so partly because policy special- ists recognize that generic problems exist in the conduct of foreign policy-for example, that the task of deterrence emerges repeatedly over time and with different adversaries. Therefore, policy specialists readily agree that general, or ge- neric, knowledge about the uses and limitations of a particu- lar strategy, derived from study of past experience, could be very helpful when strategy must be used in a new situation. Incidentally, whether the empirical generalizations that com- prise generic knowledge should also be regarded as a form of theory—or “laws,” as some scholars prefer to call this type of knowledge—should not get in the way of recognizing the im- portance of generic knowledge for policymaking. I have chosen to demonstrate the importance and rele- vance of all three types of knowledge in part two of this study in a very concrete manner by calling attention to the weak knowledge base that underlay five of the six strategies the United States pursued toward Iraq in 1988–91 and that con- tributed to their ineffectiveness. The reader will want to know why I speak in the title of this book of “bridging" the gap rather than "eliminating" it. The choice of words is deliberate and of considerable importance. I will argue that the gap between the three types of knowl- edge I have identified (which can be very loosely referred to as policy-relevant theory) and practice can be only bridged and not eliminated. Scholarly knowledge of this kind can have only an indirect and often limited impact on policymak- ing. Since I also claim that the contribution of these three types of knowledge is often critical for sound policy neverthe- less, I need to explain this apparent contradiction. The types of knowledge identified in this study serve as in- puts to policy analysis within the government and as aids to the judgment of policymakers. Such knowledge cannot sub- stitute for policy analysis or for the policymaker's judgment. XX Introduction Even the best conceptualization of a given foreign policy strategy and the most highly developed general knowledge of that strategy cannot substitute for competent policy analy- sis within the government, in which analysts must consider whether some version of that strategy is likely to be viable in the particular situation at hand. Similarly, such knowledge cannot substitute for the judgment policymakers must exer- cise in deciding whether to employ the strategy on a given occasion, since that judgment takes into account other rele- vant considerations not encompassed by general knowledge of the strategy. It is in this sense that scholarly knowledge has an indirect and often limited impact on policy. It is important that we understand why this is so, and this question is ad- dressed particularly in chapters 2 and 10. How, then, can policy-relevant knowledge aid policy analy- sis and the policymaker's judgment? First, it can assist in mak- ing a sound diagnosis of a problem situation; second, it can help identify an effective policy response for dealing with that problem. Thus, policy-relevant knowledge contributes to two essential functions in policymaking: the diagnostic task and the prescriptive one. I place particular emphasis on the diag- nostic contribution policy-relevant theory and knowledge are capable of making than to their ability to prescribe sound choices of policy. Correct diagnosis of a policy problem and of the context in which it occurs should precede and—as in medical practice—is usually a prerequisite for efforts to make the best choice from among treatment options. The analogy with the medical profession is an apt one, since the policy- maker, like the physician, acts as a clinician in striving to make a correct diagnosis of a problem before determining how best to prescribe for it. * It may be useful in this introduction to recall the origins, background, and framework of this study. In 1966, while still Introduction xxi a member of the RAND Corporation, I saw the need to sup- plement efforts to formulate general theories of international relations with theories that are more relevant for the conduct of foreign policy. To this end I initiated a small research pro- ject, “Theory and Practice in International Relations.” Later, upon moving to Stanford University in 1968, I elaborated the title to “Bridging the Gap between Theory and Practice in Foreign Policy,” and this topic has remained the focus of my research program and, in one way or another, of most of my studies since then. Quite early in implementing this research program I con- cluded that it would be necessary to move beyond structural realist theory, rational choice theory, and game theory ap- proaches that were (and are still) favored and being pursued by many talented scholars. These deductive approaches to theory development "black-box” both the process of policy- making and strategic interaction between states; that is, they deal with these processes by assumption. Instead, I felt it nec- essary to engage in direct but admittedly difficult empirical study of policymaking processes and strategic interaction be- tween actors. However, I do not regard deductive and empir- ical ways of approaching the task of developing international relations theory as antithetical. Rather, like many other re- searchers, I believe the development of both deductive and empirical approaches to theory development can be im- proved by trying to link them more closely. Finally, the reader should keep in mind that I use the word theory to encompass a broad range of ways of formulating knowledge that come out of the scholarly tradition. I found it useful in developing policy-relevant theory, which I regard as the type of knowledge needed for what historians used to refer to as statecraft, to distinguish between two types of theory. Substantive theory, the first type, deals with standard foreign policy undertakings and strategies such as deterrence, crisis management, coercive diplomacy, détente, war termination, mediation and dispute resolution, and Introduction xxiii broader criterion of value rationality and to normative consid- erations. Both substantive theory and process theory have important contributions to make to the improvement of foreign policy. Quite obviously, substantive knowledge of foreign affairs can have no impact on policy unless it enters into the process of policymaking. Substantive knowledge must combine with the effective structuring and management of the policymaking process in order to improve the analytic (versus the political) component of policymaking. The present study deals almost exclusively with substantive theory and the knowledge requirements of foreign policy. However, I have included in chapter 2 a realistic, even sober, view of some aspects of the policymaking process that tend to crowd out or reduce the impact of substantive knowledge. It is important for scholars who are interested in developing policy-relevant knowledge not to overintellectualize policy- making by assuming that it is or should be devoted exclu- sively to identifying and choosing high-quality policy options based on the criterion of analytic rationality. The present study addresses only some of the substantive knowledge requirements needed for the conduct of foreign policy. Not included are a host of important problems that affect the interests of individual nations and their peoples, to which policy specialists must be attentive. Among these prob- lems are proliferation of nuclear weapons and other mass de- struction capabilities, environmental and ecological prob- lems, population and demographic trends, problems of food production and distribution, water scarcities, sanitation and health problems, and emergence of nationalistic, ethnic, and religious conflicts. Well-informed, objective analyses of these problems are an essential part of the knowledge require- ments for conduct of foreign policy. Introduction XXV seek out common interests and to develop policies for pro- moting them. Statecraft includes strategies for cooperation and the building of institutions and regimes as well as for conflict avoidance, management, and resolution. The realist theory of international relations offers some im- portant guidelines for the practice of statecraft. Realist theory is not a theory of foreign policy, however (see chap- ter 9), and it provides very little of the substantive knowledge needed for the practice of statecraft. Thus, in addition to the help policymakers get from realist theory, they require so- phisticated conceptualization of the many different instru- ments of foreign policy and the various strategies that can be employed to influence other states. In addition, they need general knowledge of the uses and limitations of each policy instrument and strategy. This type of general, or generic, knowledge is most useful to policymakers when it is couched in the form of conditional generalizations; conditional general- izations identify (1) the conditions that favor successful use of each particular instrument and strategy, and (2) other condi- tions that make success very unlikely. Realist theory itself provides neither conceptualization nor generic knowledge of the many policy instruments and strat- egies. Another limitation of realist theory is that it assumes that all states can be treated as rational, unitary actors. The conduct of foreign policy, however, requires differentiated knowledge of each adversary, ally, or neutral state with which one is interacting. Therefore, the realist model must be sub- stantially supplemented if not altogether replaced by a dis- criminating actor-specific model. These three types of knowledge-conceptualization of strategies, generic knowledge, and actor-specific models, and their role in policymaking are discussed in detail in chapter 10. Earlier I indicated that the present study focuses on strat- egies and instruments of policy. However, I will not under- take the ambitious task of discussing the state of existing xxvi Introduction knowledge of all of the many policy instruments and strate- gies that states employ in interacting with each other. Instead, I will focus attention in part two on a smaller number of strat- egies and use them to illustrate the general points I wish to make about the need for policy-relevant knowledge. This study is addressed to two types of readers: those whose primary orientation derives from participation in the world of scholarship, whether primarily in academic settings or elsewhere, and those whose orientation to foreign policy has been largely influenced by having participated in the policy- making process. Accordingly, I have tried to write this book in a way that will be understood both by policy specialists and by academic scholars. In addressing complex and difficult is- sues I have tried to minimize use of abstract jargon and have made minimal use of footnotes and references. Bridging the gap between the theory and practice of for- eign policy requires in a very real sense bridging the differ- ences between the two cultures of academia and the policy- making world. It is important to recognize and be sensitive to these two cultures if we are to make further progress in de- veloping policy-relevant knowledge. This task, however diffi- cult, is by no means insurmountable. I believe that a start can be made by focusing in chapter 1 on the relationship between knowledge and action, a problem that should engage the in- terest of both scholars and policy specialists. Three questions are posed: (1) What contributions can knowledge make to policymaking? (2) What types of knowledge are most relevant for policy? (3) How can this type of knowledge be developed by scholars and research specialists and how can it be em- ployed effectively by policymakers? 1 1. The Two Cultures of Academia and Policymaking The policymaker, unlike an academic analyst, can rarely wait until all the facts are in. ... He is very often under strong pressure to do something, to take some action. ... The ca- pacity of human beings to deal with situations of vast com- plexity is very limited. The human mind needs a highly sim- plified “map” of a situation if it is going to be capable of taking any action or making a decision.- Robert Bowie The development of theory about international relations by academic scholars and the use of this knowledge by practi- tioners in the conduct of foreign policy have been handi- capped by the different cultures in which they have tradition- ally resided.2 Members of these two communities have been socialized in quite different professional and intellectual worlds. They generally define their interest in the subject of international relations differently. The policymaker is typi- cally preoccupied with how best to promote the national in- terest of the United States. However, many scholars believe this to be too narrow an approach to international relations and choose to work with a set of values broader than those encompassed by conventional notions of the national inter- est. An additional obstacle to bridging the two cultures is the 3 4 The Gap between Knowledge and Action reluctance of some scholars, particularly when they disagree with the government's foreign policies, to serve as “techni- cians” for the state by providing specialized knowledge that may be “misused.” Such scholars prefer the roles of critic and "unattached intellectual.”3 With the passage of time, however, the gulf between the two communities has eroded substantially. As academic schol- ars and academically trained specialists in international af- fairs have moved in and out of government service, and as former foreign policy practitioners have found places in aca- demia, a substantial intermingling of the two cultures has taken place. Many policy specialists and technical experts holding positions in the policymaking hierarchy, especially in middle levels, have had advanced training in international re- lations or in related fields. In addition to drawing on their own familiarity with scholarly knowledge, they are likely to maintain contact with academic scholars and their publica- tions. These policy specialists can thereby serve as an infor- mal bridge between the two worlds and increase the likeli- hood that relevant scholarly knowledge can be used in policymaking. Policy specialists in the government increas- ingly have tried in a variety of ways to tap and make use of the knowledge of scholars on specific policy problems. Fre- quently used for this purpose are conferences and meetings at which academic specialists meet with policy specialists to discuss a particular problem. Foundations have made impor- tant contributions in encouraging interaction between aca- demics and policy specialists; for example, the Council on Foreign Affairs has for many years conducted a fellowship program that places carefully selected young international relations scholars in the government for a year and mid- career policy specialists in academic centers of foreign policy and security research. As a result, progress is being made in understanding how scholarly research on international rela- tions can benefit from the perspective and experience of the policy world and how, in turn, policymaking can benefit from 6 The Gap between Knowledge and Action make additional progress in bridging the gap between the two cultures and their traditional perspectives. It is just this opportunity and challenge that has motivated the present study. It is well to begin by clearing some of the underbrush that still stands in the way and formulating a common ground on which productive two-way interaction can take place be- tween the theoretical approach to study of international re- lations pursued in academia and the practitioner's need for the kind of knowledge that can facilitate formulation of sounder, more effective policy. That common ground, it will be suggested here, is to be found in the relationship between knowledge and action. In the following sections I shall consider some typical res- ervations, which may or may not be wholly justified, that many (though not all) practitioners of foreign policy and many (though not all) academic international relations schol- ars have about their respective approaches to the subject.4 I hope the reader will keep in mind that this is a highly con- densed characterization and that there is considerable varia- tion in the views held by members of both groups. In part three, I offer a different set of observations regarding how the relationship between knowledge and action can be con- ceptualized so as to provide a bridge between scholars, who specialize in the production of better knowledge of interna- tional relations, and government policy specialists, whose re- sponsibility includes using general knowledge to help top policymakers make their decisions. Practitioners' Reservations about Academic Theory Academic specialists have often been struck by the fact that the eyes of policy specialists quickly glaze over at the first mention of the word “theory” or the phrase “scientific study of international relations.” This reaction is unsettling to aca- demic scholars, for they have been socialized in a world in Academia and Policymaking 7 -- which the development of theory and scientific knowledge enjoys the greatest respect and highest priority. What lies behind this glazing over of the practitioners' eyes? It is a certain image of academic efforts to theorize about international relations and policymaking. A familiar complaint is that academic scholars do not understand how policy is actually made, even when they undertake in case studies to follow the paper trail of a particular decision. Lack- ing an understanding of the process, academics tend to over- intellectualize policymaking and exaggerate the importance of analytic rationality as the criterion on the basis of which policy is, or at least ought to be, chosen (see chapter 2). Practitioners and policy specialists also question the high level of abstraction often employed in scholarly writings, and they are dubious about the relevance and utility of many of the theoretical generalizations and models put forward by academic researchers. Even when the results of scholarly research are recognized as potentially relevant, they are often under-used for another reason: not a few policy specialists exposed to the scholarly literature have concluded that most university professors seem to write largely for one another and have little inclina- tion or ability to communicate their knowledge in terms com- prehensible to policymakers. Some practitioners are acutely uncomfortable with the unfamiliar and, to them, pretentious jargon with which their activities are described, and they are perplexed by the esoteric explanations that are given for their behavior. One may recall in this connection Secretary of State Dean Acheson's tongue-in-cheek response when he be- came aware of the way his role in the U.S. decision of June 1950 to come to the assistance of South Korea had been re- ferred to in a scholarly study. Acheson took exception to being treated as a “dependent variable” in that study, observ- ing dryly that he thought it obvious that he had played the role of an “independent variable” in that crisis. - 8 The Gap between Knowledge and Action Practitioners also object to what they consider a heroic but misguided effort by academicians to put foreign policy on a scientific basis. Insisting that statecraft is an art and not a sci- ence, practitioners emphasize the role of judgment in making policy decisions. (I return to what “judgment” encompasses in chapter 2.) Some years ago George Ball, then under secre- tary of state, emphasized this point in describing the com- plexity of the problem he and his colleagues had faced dur- ing the Cuban missile crisis: “We were presented ... with an equation of compound variables and multiple unknowns. No one has yet devised a computer that will digest such raw data as was available to us and promptly print out a recommended course of action. ..."5 And other former policymakers have also emphasized that the president, unlike the academic scholar, has to make a decision in which all the variables are at play; the scholar deliberately limits the number of variables considered in order to develop a parsimonious theory. How do policy specialists respond to quantitative studies that attempt to account for war and other foreign policy out- comes by means of statistical correlations? Quantitative stud- ies may claim, for example, to “account for” (but not neces- sarily explain) 75 percent of the variance in the outcome of many instances of a particular phenomenon. Such studies may provide policymakers with a useful indication of the probability distribution, but of course this finding does not tell whether an actual case falls in that group or is among the 25 percent that are not accounted for. A more important dis- satisfaction with this type of study among policymakers stems from the fact that it typically ignores domestic political and decision-making variables. The policy relevance of such stud- ies is limited, therefore, because they do not include variables over which policymakers have some leverage that they can use to try to influence outcomes. Therefore, such studies give policymakers no help in deciding what they can do to avoid unwanted outcomes and to achieve desired ones. More generally, the gap between academic scholars and Academia and Policymaking 9 . -- "6 practitioners reflects the difference between their profes- sional missions. Academics aim at increasing general knowls edge and wisdom about international relations; practitioners are more interested in the type of knowledge that increases their ability to influence and control the course of events. In the role of policymaker, individuals often adapt by becoming action oriented: they need uncomplicated diagnoses and so- lutions in order to take action. It is not surprising that policy specialists should “want short, precise answers to problems rapidly. ... But the academic cannot give a short answer to a policymaker's question—not because the academic is long- winded or pretentious, but because [it appears to him that] that the short answer is in fact not an answer. ... Similarly, practitioners find it difficult to make much use of academic approaches such as structural realist theory and game theory, which assume that all state actors are alike and can be expected to behave in the same way in given situations, and which rest on the simple, uncomplicated assumption that states can be regarded as rational unitary actors. On the contrary, practitioners believe they need to work with actor- specific models (see chapter 10) that grasp the different inter- nal structures and behavioral patterns of each state and leader with which they must deal. Similarly, although practi- tioners can accept the academician's argument that the as- sumption of pure rationality is a powerful tool for the con- struction of theory, they are more impressed with the caveat that sophisticated academicians add, namely, that the valid- ity of such theories of course must be established by empi- rical research that examines the “fit” between the theory and reality. Practitioners have a similar reservation when encouraged to use new psychological theories, developed under carefully controlled laboratory conditions, that purport to—and in- deed do—have relevance for some aspects of foreign policy. Perhaps drawing on their memories of undergraduate psy- chology courses, policy specialists recall that experimental 10 The Gap between Knowledge and Action psychologists themselves emphasize that theories developed under laboratory conditions may or may not have external validity in real-world situations. Policy specialists also believe that there are often good rea- sons to question the results of academic studies that explain and evaluate foreign policy decisions. Policymakers suspect that scholarly studies are not always as objective as they ap- pear, and that they contain political or ideological biases. Given the proliferation of competitive “advocacy analysis” in the ever-expanding world of think tanks and academic cen- ters, policymakers can find excuses and justification for re- jecting those studies that do not fit their own policy orienta- tions and rely instead on those that do. Then, too, the data base used in scholarly studies, policy specialists argue, is often inadequate and can distort the scholar's interpretations. Not only is access to classified infor- mation for such studies limited, but available declassified in- formation does not give the whole picture and may introduce flaws in the interpretations. Similarly, data obtained from current or former policymakers are subject to a variety of limitations that scholars are often unaware of or do not ade- quately take into account. In some cases scholars rely too un- critically on information about unsuccessful policies pro- vided by lower level policy specialists who say in effect, “We had the correct analysis of the problem and the right answer, but the higher-ups didn't listen.” Finally, the practitioner is often wary of academic research theories and generalizations that purport to provide policy advice. A strong reservation of this kind was expressed by Richard Goodwin, a former policy adviser in the Kennedy and Johnson administrations, in a sharply critical review some years ago of the book Arms and Influence, written by Thomas Schelling, a leading strategic theorist.? For Goodwin, Schelling's book “raises troubling questions about the grow- ing body of social-science literature devoted to military theory.” Goodwin rejected the view that what Schelling called i Academia and Policymaking II “the diplomacy of violence” can be the subject of systematic theory, and he marshaled several familiar arguments to but- tress his conclusion. For Goodwin, history is sui generis, and to generalize from one historical crisis to another is therefore dangerous and “almost guarantees error.” In any case, the task of developing valid generalizations is, according to Goodwin, virtually insuperable. And even if some sort of generalization were somehow derived, its relevance to any immediate policy problem could not be determined, since in almost any situation policymakers must act without knowing all the facts, and the variables in the situation “are so numer- ous that they elude analysis.” These concerns led Goodwin to put forward a sober warning: “The most profound objection to this kind of strategic theory is not its limited usefulness but its danger, for it can lead us to believe we have an understand- ing of events and a control over their flow which we do not have.” I shall return to Goodwin's concern later in this chapter. Academic Scholars' Reservations about the Policy World Although specialists in the study of international relations sharply disagree among themselves over questions of method and theory, many of them hold the belief that policymakers are too aconceptual and atheoretical, even anticonceptual and antitheoretical. At the same time, academic scholars note that policymakers and their advisers often do operate with a general view of international politics influenced by their ac- quaintance with realist theory and such related concepts as national interest and balance of power. Academic specialists express concern, however, that the realist theory that shapes the mind-set of these policymakers is a warmed-over and sim- plified version of Hans Morgenthau's influential post-World War II textbook Politics Among Nations and similar efforts by others in that era to steer American foreign policy away from what they considered to be a well-meaning but naively idealist Academia and Policymaking 13 sumptions, beliefs, and maxims that are seldom raised to full consciousness and examined critically.9 Policy beliefs and maxims of this kind emerge from an individual's personal or vicarious experience of past historical events, and from les- sons passed on from earlier generations of policymakers. Such policy beliefs and historical analogies are usually rather simple, uncomplicated formulations of causal relationships. An example is the familiar proposition derived from the ex- perience of the 1930s that was and remains influential in shaping American policy, namely, “If appeasement, then World War III,” or, as it is often worded, “If we appease the opponent now we shall have to fight a much larger war against him later." To the academician, simple generalizations of this kind se- riously oversimplify the complexity of foreign policy strate- gies and outcomes. They should be replaced by conditional generalizations; thus, one needs to ask “Under what conditions is appeasement counterproductive and serves to increase the likelihood of war?” But the question should also be asked, “Under what different conditions is appeasement a useful, viable strategy for conflict resolution?” (The strategy of appease- ment is discussed in chapter 5.) Systematic study of a variety of historical cases and more thoughtful reflection, the acade- mician argues, are needed to provide discriminating condi- tional generalizations of this kind. 10 (At the same time, how- ever, it must be said, academic scholars have yet to produce a systematic, empirically grounded, and differentiated theory of appeasement.) A related criticism singles out the policymaker's habit of relying on and often misusing a single historical precedent or analogy." Policymakers and scholars who agree with Richard Goodwin maintain that history is sui generis and that each case is unique; nonetheless, in practice they often search for and use historical analogies to help them deal with their “unique” cases. Academicians see policy specialists as uninter- ested in cumulating lessons from a number of different 14 The Gap between Knowledge and Action historical cases and having no effective method for doing so. As a result, or so academicians believe, policymakers often do not take into account and digest a broad variety of relevant historical experience, and therefore what they do learn from history is of dubious validity. For these and other reasons, academic scholars believe pol- icymakers need to develop a better understanding of how their own beliefs and tacit assumptions about the interna- tional system, international politics, and other actors in the state system influence their perception of developments, their diagnoses of situations, and their judgments. Academ- ics are therefore very much in agreement with the observa- tion made many years ago by a former State Department planner, Louis Halle, that the foreign policy of a nation ad- dresses itself not to the external world per se but rather to “the image of the external world” that is in the minds of those who make foreign policy. Halle concluded his book on Amer- ican foreign policy with a sober warning: “In the degree that the image is false, actually and philosophically false, no tech- nicians, however proficient, can make the policy that is based on it sound.”12 Accordingly, academic specialists suggest, practitioners should recognize the importance of subjecting the operative beliefs and policy maxims they rely upon to scrutiny via scholarly studies of a broader range of historical experience. Finally, academic specialists believe that in choosing a pol- icy, policymakers are too often swayed by political considera- tions rather than by the results of objective analysis. Accord- ing to this view, policymakers are less inclined to choose the best policy than to choose one that commands greater politi- cal support and avoids too much controversy. Another criti- cism is that even when good policies are announced, the rea- son is largely so that the administration can say it has a policy, but low priority is assigned to it and little is done to imple- ment it. At the same time, it must be said that many academic scholars do not understand or sympathize with the fact that Academia and Policymaking 15 policymakers must deal with difficult trade-off dilemmas be- tween the quality of a policy and other desiderata. (On the latter point, see chapter 2.) I have called attention to the somewhat different conceptions of knowledge about international affairs that are valued most in academia and in the world of policymaking. Also contrib- uting to the gulf between these two cultures are the rather different socialization experiences that shape an individuals career orientation and professional style in the worlds of aca- demia and policymaking. Those who participate in policy- making learn that little can usually be accomplished by indi- viduals without a great deal of coordination, cooperation, and compromise with others. And one gets ahead in govern- ment by being able to adapt to the norms and procedures of various kinds of group work. The policymaking process blends the search for a high-quality policy decision with the need to develop a degree of consensus and support. Inter- personal and negotiating skills as well as analytic competence are at a premium. At the extreme, an individual occasionally finds it preferable not to claim credit for a good idea but in- stead to plant it inconspicuously in the process of interacting with others. On the other hand, as is well known, efforts to cooperate and coordinate often break down, and the policy process is then driven by the dynamics of bureaucratic politics. When the game of bureaucratic politics cannot be tamed by strong leadership at higher levels, information and knowledge be- come instruments of the struggle between competing policy advocates. Academia is less conducive to participation in group work. To get ahead one tries to differentiate one's ideas and schol- arly products to give them a distinctive stamp. Importance is attached to establishing an individual reputation rather than 16 The Gap between Knowledge and Action a reputation for effective participation in group work. It is not surprising that academics who enter government often find that with the individualistic professional style into which they have been socialized they encounter difficulties in the world of policymaking. However, it must also be said that other academics entering the policy world adapt all too read- ily to tempting opportunities to participate in influencing policy. They become absorbed in the internal politics of poli- cymaking and lose interest in strengthening the role of schol- arly knowledge in policymaking. In my interviews I have found that such former academics are often more useful as informants on the process and politics of policymaking than on ways of strengthening and using substantive knowledge of policy problems. A Common Ground: Knowledge and Action The most promising way to bridge the gap between the acad- emician and the policymaker, I believe, is to focus on the re- lationship between knowledge and action in the conduct of foreign policy. That relationship, of course, must be concep- tualized in a way that will speak to both academic scholars and policy specialists. What this means, more concretely, is that scholars should acquire a clear and detailed understanding of the types of knowledge policymakers need as they attempt to understand and to deal with different types of foreign pol- icy problems. Scholars can be assisted to some extent in refin- ing their understanding of the knowledge requirements of practitioners by what policy specialists themselves say they need. But since practitioners and their staff specialists may not be able to articulate knowledge requirements well enough, scholars will need to study actual policymaking closely to identify the types of theory and knowledge that are likely to be relevant. Since so many individuals who serve as policy specialists Academia and Policymaking 17 and in staff roles in the government have previously studied international relations in academic centers, they can and do serve as informal intellectual brokers between the two cul- tures. To the extent that they accept this role they are in a position to draw on and adapt the results of academic re- search for use in policymaking and, also, to familiarize aca- demic specialists with the kinds of information and knowl- edge required by policymakers. How much interest they have in doing so and how well they perform this role, however, is another question. Nonetheless, it has been useful to interview a number of present and former policy specialists and intelli- gence officers with academic backgrounds in order to draw on their experience and to obtain their suggestions for better appreciating the relationship between knowledge and action. In concluding this chapter I return to one of Richard Good- win's criticisms of Thomas Schelling's Arms and Influence. Goodwin argued that a “systematic theory” of the kind he (incorrectly) attributed to Schelling was impossible as well as dangerous. The criticism rests on a misunderstanding of sys- tematic theory as something that seeks or seems to provide policymakers with detailed, high-confidence prescriptions for action in each contingency that may arise. I will argue that such a theory does not exist and is not feasible. The choice, however, is not between detailed, high-confidence prescrip- tions for action and nothing at all. I suggest that theory and generic knowledge may most feasibly and usefully contribute to the diagnosis of the specific situations with which policy- makers must deal, rather than to prescriptions for action. My assumption is that correct diagnosis of a policy problem should precede and—as in much of medical practice—is usu- ally a prerequisite for making the best choice from among policy options. The analogy with the medical profession is an 18 The Gap between Knowledge and Action apt one, since the policymaker, too, acts as a clinician in striv- ing to make a correct diagnosis of a problem before deter- mining how best to prescribe for it. The reader will note that for Goodwin's rhetorical question whether a systematic theory is possible I have substituted three different questions: (1) What is the relationship be- tween knowledge and action? (2) What kind of knowledge is most relevant for assisting policymaking? (3) How can this type of knowledge be developed by scholars and research specialists and how can it be employed effectively by policy- makers? 2. The Role of Knowledge in Policymaking Some persons interviewed for this study expressed a sober view of the prospects for improving foreign policy by devel- oping and making available better knowledge of the various strategies and activities that enter into the conduct of foreign policy. Some who read an earlier draft of parts of this study thought I had underestimated the extent to which policy is determined by factors other than scholarly knowledge and competent policy analysis. A realistic view of how policy is ac- tually made, they cautioned, would emphasize that political considerations in the shaping of policy often constrain the impact that objective analysis can have on the decisions taken. Others emphasized that top policymakers and policy profes- sionals at lower levels are heavily influenced by their existing mind-sets in terms of how they view and deal with policy problems and in their receptivity to new ideas and analysis. In fact, scholars who have studied policymaking are well aware of these characteristics of the policy process and the constraints they impose. However, many of them reject, as I do, the implication that even an improved body of policy- relevant knowledge would make little difference in improv- ing the quality of many policy decisions and that, therefore, 19 The Role of Knowledge 21 policymakers must take into account, I have coined the term political rationality. It is admittedly difficult, then, to ensure that relevant sub- stantive knowledge will have an optimal impact on policy- making. Assuming that it does enter the policymaking pro- cess, what kind of impact can it be expected to have? The position taken in this study is that policy-relevant knowledge can be expected to make a limited and indirect, but still quite useful, contribution. General knowledge of international relations produced by scholars can be only an input to, not a substitute for, the policy analysis of a specific problem conducted within the govern- ment. Policy analysts, not academic scholars, have the difficult task of adapting the available general knowledge of a given strategy or foreign policy undertaking to the particular case at hand. General knowledge can help with making the diag- nosis of the situation that policymakers need, but the diag- nosing also requires access to specific intelligence and infor- mation about the situation. Government policy analysts and policymakers have access to more of that information than scholars outside the government have. Similarly, it should be emphasized, general knowledge of a strategy or standard foreign policy undertaking is not a substitute for, but only an aid to, the final judgment of high-level policymakers. (Incidentally, the same can be said regarding the results of intragovernmental policy analysis; it, too, is an aid to, not a substitute for, the judgment of policymakers.) This is so be- cause these policymakers take into account various factors, such as domestic political considerations, international opin- ion, and value priorities, that cannot be anticipated by schol- ars who produce generalizations about foreign policy and cannot be taken into account or adequately weighed even by policy analysts. Thus, whereas scholars and policy analysts can and should preoccupy themselves with the task of identi- fying high-quality policy options (i.e., meeting the special cri- terion of analytical rationality) high-level policymakers must exercise broader judgments that take into account a variety 22 The Gap between Knowledge and Action of additional considerations. What these additional consid- erations are will become clearer as we turn now to a discus- sion of what is meant by the “judgment” of policymakers to which policy-relevant knowledge attempts to contribute. Judgment in Policymaking The importance of judgment is often emphasized in discus- sions of policymaking. Individuals who have had experience in policymaking regard judgment as a critical and indeed necessary ingredient of policymaking, and for this reason, they maintain, policymaking must be regarded as an art and not a science. Yet the concept of judgment is seldom expli- cated or studied and, therefore, despite its centrality in pol- icymaking it remains ambiguous. Since judgment plays so im- portant a role in bringing knowledge to bear on action, it is necessary to try to identify more clearly what judgment con- sists of. One way to proceed with this task is to replace the global notion of judgment with an identification of those aspects of decisional problems about which policymakers exercise judg- ment. At least seven different types of judgment need to be distinguished from one another. A policymaker may feel ob- ligated to make more than one type of judgment in arriving at a particular decision, but it is unlikely that all or most of the following types of judgment would enter into a particular decision. 1. Trade-off judgments. One of the most important decisions high-level policymakers are often obliged to make concerns the trade-off between seeking to maximize the analytical quality of the policy to be chosen and the need to obtain suf- ficient support for the policy option that is finally chosen. An- other familiar, often difficult trade-off problem arises from having to decide how much time and policymaking resources to allocate to the effort to identify the best possible policy op- The Role of Knowledge 23 Search for high-quality decisions Trade-off Trade-off Need for acceptability, consensus, support Trade-off Prudent management of time and other policymaking resources Figure 2.1. Trade-off dilemmas in policymaking. (Adapted from A. L. George, Presidential Decisionmaking in Foreign Policy: The Effective Use of In- formation and Advice [Boulder, CO: Westview Press, 1980), p. 2.) tion. A third trade-off problem arises from having to decide how much political capital, influence resources, and time to expend in an effort to increase the level of support for the option to be chosen. The three trade-off problems are depicted in figure 2.1. Academic specialists can easily fall into the error of think- ing about high-quality policy decisions in too narrow a frame- work. Policymakers have to deal with the tension that often exists between policy quality and the need to choose a policy that commands enough support. Very often a measure of quality has to be sacrificed in favor of a decision that will get the kind of political support within and outside the adminis- tration that is necessary if the policy is to have a chance of being sustained. Another trade-off in political policymaking is the one be- tween the quality of the decision and the policymaker's sen- sible use of time and of analytical and political resources. A policymaker who spends a tremendous amount of time The Role of Knowledge 25 jeopardize the success of the weaker variant of the strategy chosen in order to gain additional support. Of course, the need for support may sometimes push the policymaker in the other direction—toward adopting a stronger variant of a strategy when a milder one would be more appropriate. There is another way theory and generic knowledge can contribute to a better understanding of the costs and risks of trade-offs between quality and consensus. As is well known, bargaining often takes place among advocates of different policy options, and at times the dynamics of bargaining weaken the role of objective analysis in the search for an op- tion that different members of the policymaking group can agree on. Good analysis of a policy problem can equip pol- icymakers to anticipate what kinds and degrees of effective- ness a high-quality option is likely to lose if trade-offs are made during the bargaining in order to gain broader sup- port. In this view, analysis is not a substitute for bargaining but serves to inform and discipline the bargaining process in a way that helps prevent ending up with a compromised policy that is likely to prove ineffectual. 2. Judgments of political side-effects and opportunity costs. Still another type of trade-off dilemma must be recognized. The criterion of analytical rationality is applied most comfortably by the policymaker when the problem in question is “bounded”—that is, when it is insulated from other policy issues that are already on the agenda or may soon emerge. But many policy issues are imbedded in broader political and policy contexts. When this is so, the policymaker finds it nec- essary to consider what effect his choice on a particular policy issue will have on his overall political standing and on other parts of his overall policy program. In choosing what to do in situations of this kind the policymaker is guided not exclu- sively or even primarily by the dictates of analytical rationality but may be heavily influenced by his judgment of the political side-effects and opportunity costs of his choices. Of course, The Role of Knowledge 29 - --- “weak” theories and general knowledge of foreign policy problems are relevant nonetheless, and that their absence or the failure to make use of them in policymaking can mean the success or failure of policy.? Part two highlights this point by showing that lack of adequate conceptualization and gen- eral knowledge of several of the strategies employed by the United States toward Iraq in 1988–91 had something to do with the failure or mixed results of these policies. - Part Two The Inadequate knowledge Base for U.S. Policy toward Iraq, 1988-91 3. Outcomes of U.S. Strategies toward Iraq During the Iran-Iraq War the United States strongly tilted toward Iraq, furnishing it with substantial economic assist- ance and large amounts of indirect military aid to prevent Iran from achieving a hegemonic position in the Gulf and spreading its radical Islamic fundamentalism. This policy was based on familiar balance-of-power considerations, and it was successful. But at different stages in its relations with Iraq after the Iran-Iraq War, the United States employed a num- ber of strategies that either failed to accomplish their pur- pose or produced mixed results. This chapter contains an overview of the development of U.S. policy toward Iraq be- tween 1988 and 1991 and brief discussions of those strategies and their outcomes. With the end of the Iran Iraq War in August 1988, the question arose whether a change in U.S. policy was desirable or required. A few mid-level policy specialists in the Bush ad- ministration argued that balance-of-power considerations now dictated ending economic and military assistance to Iraq, since it had emerged from its successful war to constitute a potential new threat to regional stability. These estimates were based not merely on balance-of-power logic but also on 33 Outcomes of U.S. Strategies 35 doubt wished to continue its policy of wooing Iraq away from the Soviet Union. Although Saddam Hussein was generally regarded as a “bad actor,” some encouragement was derived from the fact that he had moderated some aspects of his policies and be- havior during the war with Iran. And it was hoped that Sad- dam might be disposed, if treated in a conciliatory and friendly way, to alter whatever hostility he may still have har- bored to the conventional norms and practices of the inter- national system and to evolve into a responsible, constructive actor in the Middle East. Some policy officials, but not all, thought Saddam could be “reformed” insofar as his external behavior might become more moderate and more acceptable, leaving aside the much less likely possibility of a moderation of his internally repressive regime. Accordingly, to the extent that top officials in the adminis- tration gave serious thought to the matter, they did not judge it either necessary or advisable to curtail aid to Iraq and shift to a policy of containment and deterrence. Indeed, relations with Iraq were not a priority consideration for the president and his closest advisers during this period. Their attention was focused on the profound changes in the Soviet Union and Eastern Europe. As Don Oberdorfer notes, “Without attention at these high levels, the chances for a policy shift... are slim.”2 In reflecting today on how they perceived the situation at that time, administration officials may well believe that to have shifted to a policy of containment and confrontation of Iraq was not a viable option; they may well believe that they really had no choice at the time but to continue the policy of constructive engagement and friendship toward Iraq. Even though policy officials may not have explicitly stated that the United States should help Saddam because he was redeem- able, Washington's actions toward Iraq included elements not only of an effort to resocialize and reform a rogue leader but also elements of appeasement. (I use the term appeasement 36 U.S. Policy toward Iraq here in its original, nonpejorative sense. Conciliation and some connotations of constructive engagement are roughly equivalent in meaning to appeasement.) To the extent that policymakers even thought of undertaking a more confrontational policy toward Iraq, such a policy was regarded as not only unneces- sary but also lacking domestic and international support, par- ticularly among friendly Arab states in the region. Be that as it may, the continuation of the policy of friend- ship appears to have been buttressed by two closely related tacit assumptions: first, that Iraq was still needed to balance the threat that Iran, though badly damaged by the war, would continue to pose to its neighbors and to U.S. interests in the region; and second, that by continuing to emphasize its desire for friendly relations with Saddam the U.S. could re- form him. 3 After the Gulf crisis erupted, administration leaders ac- knowledged that their policy had been to try to resocialize Saddam Hussein and that it had failed. President Bush was quoted as saying there had been “some reason to believe that perhaps improved relations with the West would modify his behavior.” On another occasion after the bombing of Iraq started, the president stated to a reporter, “We tried the peaceful route; we tried working with him and changing [him] through contact. The lesson is clear in this case that it didn't work.” More recently, in a press conference held on May 27, 1992, President Bush defended his pre-Gulf War policy toward Iraq, saying that his administration tried to work with Saddam "on grain credits and things of this nature to avoid aggressive action. And it failed. That approach, holding out a hand, trying to get him to renounce terrorism and join the family of nations, didn't work.” And the presi- dent's special assistant for national security affairs, Brent Scowcroft, also acknowledged that the object of the earlier approach to Saddam was “to make this guy a reasonably re- sponsible member of the international community." Members of the administration began to have some doubts 94 Outcomes of U.S. Strategies 37 about the policy of friendship with Iraq in early 1990 when, in his speech of February 24, Saddam Hussein attributed hos- tile intentions to the United States and objected to its military presence in the Gulf and its role in the Middle East. A review of policy toward Iraq was undertaken in the spring by the Interagency Deputies Committee, headed by deputy national security adviser Robert Gates. That the policy was not work- ing as it should and that it required some adjustments was recognized. A few inducements to Iraq for closer ties with the United States were removed, but it was agreed that the option of friendship with Iraq should be kept open, and most forms of assistance were continued.” Participants in the policy review found it difficult to under- stand why Saddam had resorted to hostile rhetoric and accu- sations against the United States. At the same time, however, they could not see any reason for not continuing to hope that Saddam would moderate his behavior. It was decided to try to reassure the Iraqi leader that the United States wished to continue the policy of friendship, that it did not harbor hostile intentions toward Iraq, and that its military presence in the Gulf was not aimed at Iraq. At the same time, it was thought advisable to convey to Saddam diplomatically that the quality of the relationship between the two countries in the future would depend on Iraq's behavior. As assistant secretary of state John Kelly put it in his congressional testi- mony in late April, Iraq's difficult behavior was recognized and posed a challenge, but it remained important to provide Iraq with an opportunity “to reverse this deterioration of re- lations.”This message was conveyed in several ways—includ- ing a visit to Baghdad in May by Richard Haass, the National Security Council (NSC) specialist on Near East and South Asian affairs—but always in the context of Washington's de- sire to continue to seek friendship with Iraq. When Saddam began openly menacing Kuwait in mid-July 1990 and then mobilized large forces on the border, the ad- ministration finally took a more sober view of his intentions. 38 U.S. Policy toward Iraq A belated ad hoc effort was mounted to deter him in case he was planning to attack Kuwait. But the administration was able to put together only a weak deterrence strategy and, moreover, coupled its bland deterrence effort with official re- assurances to Saddam that it continued to desire friendly re- lations—a dual strategy of deterrence and reassurance that failed to achieve its objective. After Saddam's forces quickly overran Kuwait, the admin- istration responded by mounting a strong deterrent effort to dissuade a further move against Saudi Arabia. Whether Sad- dam actually entertained such a plan and needed to be de- terred remains uncertain, but he did not move against Saudi Arabia. From the administration's standpoint, the threat to Saudi Arabia lay not exclusively in the possibility of invasion. By taking over Kuwait and keeping sizeable Iraqi forces there, Saddam was in a position to intimidate Saudi leaders and influence them to go along with his wishes. As for the objective of liberating Kuwait, the administra- tion took the lead in obtaining UN Security Council approval for employing coercive diplomacy to persuade Saddam to re- move his forces from Kuwait. Initially backed by a compre- hensive set of economic sanctions, this variant of coercive di- plomacy was later judged to be inadequate for achieving a timely and reliable success, and the idea of a “preventive war” to drastically weaken Iraq's military capabilities gained sup- port. Without committing itself to this course of action, the administration decided to augment U.S. forces in the Gulf to create an “offensive capability,” and it succeeded in persuad- ing the Security Council in late November to adopt a new res- olution embracing the strongest variant of coercive diplo- macy, one that gave Saddam an ultimatum backed by the threat of overwhelming force. To the surprise of many, though not all, officials, the ulti- matum failed to induce Saddam to pull out of Kuwait. The war that followed was remarkably successful from the mili- tary standpoint, and policymakers were soon obliged to con- sider how the political objectives of the war could be fur- Outcomes of U.S. Strategies 39 thered by military strategy and by the terms for ending the war. It was decided for various reasons to forgo a demand for Saddam's ouster (a move within the United Nations to declare him a war criminal subject to an international trial had been set aside earlier) and also to forgo military occupation of the country. The way Washington dealt with the termination of the war led quickly to disappointment and questioning of its strategy. (Note that U.S. News and World Report titled its book on the Gulf War Triumph without Victory.) It would be quite incorrect to go so far as to say that the U.S.-led coalition won the war but lost the peace. But important questions re- main as to the nature of the peace, the fate of Saddam, and Iraq's future. Since the end of the war the United States and its coalition partners have returned to the strategy of coercive diplomacy based on a continuation of economic sanctions to ensure Iraq's compliance with UN terms and, at least in the case of the United States, to bring about Saddam Hussein's downfall. Many months after the end of the war, the administration undertook a belated effort to organize a broad-based opposi- tion to Saddam Hussein composed largely of Iraqi dissidents living outside the country. And in August 1992 the adminis- tration imposed a “no-fly zone” over southern Iraq, ostensi- bly to shield dissident Shi’ites from Iraqi air attacks but also to increase pressure for Saddam's ouster. If he survives nonetheless, the remaining option will be containment and deterrence of a much weakened Iraq. More remote but not inconceivable is the possibility that eventually Saddam may be seen as a useful actor in the regional balance of power in case Iran emerges as a new threat. This brief review of U.S. policies toward Iraq in the past dec- ade has highlighted the fact that several of the strategies em- ployed to deal with that country and its leader at different junctures failed altogether. The list of strategic failures Outcomes of U.S. Strategies 41 7 failure of policy rested exclusively on whether the U.S. image of Saddam Hussein was correct or whether the quality of the conceptualization and general knowledge of the strategies that were employed was defective. Available information from open sources and interviews with several policy special- ists also suggests a fundamental weakness, if indeed not a fail- ure, of intelligence on critical matters throughout the period examined. In October of 1989, a year after the end of the Iran-Iraq War, the U.S. intelligence community assured the administration in a formal national intelligence estimate (NIE) that for the next two to three years Saddam Hussein would not pursue aggressive policies toward any of his neigh- bors but would focus on rebuilding his country and its econ- omy." Washington was well aware that Saddam Hussein's gov- ernment was laboring under severe economic pressures as a result of its prolonged, costly war with Iraq. Intelligence was very slow, however, in asking why, then, Saddam was continu- ing to spend so much money on his ambitious military pro- grams. It would have been possible to estimate that he was spending about a third of Iraq's gross national product on these military programs rather than giving greater priority, as the NIE assumed, to rebuilding his country and its internal economy, and to ask what this spending implied about his ambitions and plans. Coupled with this intelligence gap was a misplaced sense of assurance that we knew what we needed to know with regard to the approximate size and progress of Iraq's strategic weapons program and that we could reliably assess its current military capabilities. On the other hand, the five months that intervened between the onset of the crisis in early August 1990 and the onset of the air war in mid- January 1991 were effectively used by intelligence services to develop better information on targets in Iraq. The weakness of the intelligence effort was perhaps mag- nified by the calculated, determined policy of friendship the administration was bent on pursuing toward Iraq. We know from studies of past cases that a firmly established policy pro- vides a framework and atmosphere that can subtly discour- 42 U.S. Policy toward Iraq age intelligence from contradicting existing policy and re- duce policy specialists' receptivity to information that calls ex- isting policy into question. The policy framework of pursuing friendship with Saddam may well have reduced incentives to ask more probing questions about Saddam Hussein's longer range intentions. The commitment at the highest levels of the administration to pursue friendship with Saddam Hus- sein may have discouraged intelligence and policy specialists from a more vigorous challenge to the policy of continuing to supply considerable economic and indirect military aid to Iraq. Later, after Iraqi forces occupied Kuwait and the United Nations imposed economic sanctions, the intelligence com- munity appears to have lacked the kind of knowledge of Iraq's economy necessary for developing confident, reliable assessments of the impact economic sanctions were likely to have. The administration's lack of confidence in its ability to predict the effect of sanctions and to assess the speed with which Iraq's military would be able to produce weapons of mass destruction played a role in its decision to shift to an ultimatum backed by the threat of war. In other respects, too, inadequate intelligence made policy and its implementation more difficult. Limited knowledge of Muslim culture, psychology, and ways of thought and expres- sion added to the difficulty of inferring Saddam Hussein's real beliefs, calculations, and intentions from his unusual rhetorical style and added uncertainty to efforts to influence him. Dealing with leaders from a culture so different from our own added to the familiar problem of ensuring that what one says is being understood and is taken seriously. **** It is not my purpose in the chapters that follow to provide a detailed historical account of U.S. policy toward Iraq during this period. Nor is it my purpose to contribute to political crit- icism of the administration's policy of friendship toward Iraq 4. Reforming Outlaw States and Rogue Leaders The objective of resocializing and reforming a “rogue” leader, as the experience with Saddam Hussein indicates, is not easily achieved. The administration's effort to do so prob- ably failed for a number of reasons. It appears to have been based on a defective conceptualization of the strategy, an in- correct image of Saddam Hussein, and a questionable and inconsistent application of the strategy of rewards and pun- ishments. In addition, for reasons already noted in the dis- cussion of the weakness of U.S. intelligence in chapter 3, Washington responded sluggishly to indications that the strategy was not working and that it had to be either re- vamped or replaced by containment and deterrence. In part, too, though this was not the fundamental problem, Washington's sluggish response owed something to the fact that top-level policymakers were intensely preoccupied with developments in the Soviet Union and Europe and, as they admitted later, did not give enough attention to growing in- dications of Saddam's intentions. The predominant mind-set of U.S. officials was that Iraq would continue to be preoccupied for some time with the onerous task of recovering from its destructive war with Iran. 45 46 U.S. Policy toward Iraq Accordingly, Iraq was regarded as relatively unimportant in terms of policy priorities. I have been told that in much of 1990, before mid-July, the deterioration of relations with Iraq was not even the top priority of Near East policy specialists. Some were more concerned with the growing tension be- tween India and Pakistan, which was complicated by the dan- ger of additional nuclear proliferation. Other Middle East policy specialists were preoccupied with the Arab-Israeli con- flict and were focusing on how to get a peace process started. There is ample reason to believe that the administration's policy toward Saddam Hussein prior to the Gulf crisis was based squarely on the assumption that he could be encour- aged to play a constructive, responsible role. This assumption was supported by evidence that Saddam had displayed re- strained behavior during the Iran-Iraq War. This restraint seemed to indicate that he was moving toward greater mod- eration in attitudes and behavior. This image of Saddam evidently played a fundamental role in the adoption of NSC Directive 26, October 1989, which set the basis for U.S. policy toward Iraq. As characterized by Don Oberdorfer, NSC Directive 26 had as its underlying premise that Baghdad had emerged from the war with Iran “pre- pared to play a more constructive international role.”! Ex- cerpts from the directive, declassified and made available to Congress in May 1992, stated that “normal relations between the U.S. and Iraq would serve our longer-term interests and promote stability in both the Gulf and the Middle East.” The directive recognized important differences with Iraq over chemical and biological weapons, its nuclear aspirations, the state of human rights in Iraq, and its meddling in Lebanon and other Middle East states, but nonetheless held that “the U.S. Government should propose economic and political in- centives for Iraq to moderate its behavior and to increase our influence with Iraq."And, specifically, U.S. companies were to be encouraged to participate in the postwar reconstruction of Iraq as long as they did not abet nuclear proliferation. Reforming Outlaw States 47 Various indications in Saddam Hussein's behavior during the Iran-Iraq War encouraged U.S. policy to move in this direction: -- - • Iraq began to distance itself from the Soviet Union's policy in various Third World countries. • Iraq became less hostile to the Arab-Israeli peace process. • Saddam expelled some (though not all) leading terrorists. • Iraq apologized and eventually paid $27 million in a com- pensation package for families of the thirty-seven victims of the Iraqi missile attack on the USS Stark in the Persian Gulf. • Iraq agreed to restoring diplomatic relations with the United States and treated American officials in Iraq in a strikingly friendly way. • Iraq eased restrictions on foreign travel. • Access was given to Iraqi Kurdistan.3 Iraq offered a substantial market for sale of U.S. agricul- tural products and investment opportunities. Washington's perception that Iraq was moving in a moder- ate direction was reinforced by leaders of friendly Arab states who attached importance to the change in Saddam's behav- ior. The danger of mistaking a leader's tactically motivated good behavior as a sign of a more fundamental change is a familiar one in international relations. Later, Saddam's speech of February 24, 1990, in which he attributed hostile intentions to the United States and objected to its naval presence in the Gulf was regarded by many mem- bers of the Bush administration as an emotional overreaction to a Voice of America (VOA) broadcast in Arabic that criti- cized secret police abuses in Iraq and other countries and seemed to call for overthrow of Saddam's regime. This inter- pretation was held even by some area experts, whereas a few policy specialists took a more sober view of Saddam's speech, regarding it as a well-considered statement of his aspirations and his estimate of the strategic situation in the region. Lack Reforming Outlaw States 49 Outlaw states and their rogue leaders refuse to accept and abide by some of the most important norms and practices of the international system. Such states may seek to dominate and reshape the system to their own liking, and they may aim at global or regional hegemony. Some of them resort to prac- tices such as state terrorism and taking as hostages citizens or official representatives of other states. The Great Powers traditionally have accepted some re- sponsibility for maintaining an orderly international system. Their incentive to find ways of coping with the threat to or- der by revolutionary powers, outlaw states, and rogue leaders is understandably accentuated when their own important na- tional interests are threatened by the aims and behavior of such actors. It should be noted, however, that there exists no clear and commonly accepted definition of an outlaw or rogue state. These concepts have no standing in international law, and the United Nations works imperfectly to single out such offenders and deal with them. In fact, members of the international community may disagree among themselves whether the behavior of a certain state justifies its being re- garded as an outlaw and treated as a pariah. Even behavior that violates a particular norm may be condoned by some as an understandable way of pursuing legitimate grievances or ambitions. Much of the task of recognizing and coping with outlaws, then, is undertaken by individual states, usually one or more of the Great Powers, which have a stake in preserving the system that they have helped to create and that they sub- scribe to, as well as in protecting interests threatened or dam- aged by an outlaw. At the same time, it should be recognized that efforts by one or more states to cope with outlaws do not always win agreement and support from other states; reso- cialization of the rogue leader then becomes all the more dif- ficult. What strategies, then, are available for dealing with revo- lutionary and outlaw states and their rogue leaders? Which strategies have been tried in the past and with what results? I Reforming Outlaw States 51 demanding meaningful changes in policy and behavior in re- turn for each concession or benefit—is safer and likely to be more effective than pure appeasement in achieving resociali- zation in the long run. In scholarly writings conditional reci- procity is usually treated as a tactic to be employed in negoti- ating a particular issue or in encouraging changes in one or more of an adversary's policies. Here, however, we point to its strategic use as part of a long-range effort for bringing about fundamental change in the nature of the outlaw state and its leadership, that is, the gradual replacement of its antipathy to the norms and practices of the international system with attitudes and behavior more supportive of that system. In other words, conditional reciprocity may be used as a lever for implementing the strategy of behavior modification, to which we will turn shortly, and the objective of resocializing the outlaw state and reforming its rogue leadership. At the same time, one should keep in mind that conditional reci- procity can also be used for the lesser objective of inducing a change in the policies of another actor, whether that actor is an outlaw state or a responsible adherent of the norms and practices of the international system. In any case, the strategy of resocialization and the levers it employs must be conceptualized in a sophisticated way and carefully implemented. This is easier said than done, in part because we have as yet virtually no systematic analyses of past efforts of this kind. Nonetheless, it is possible to differentiate the use of re- wards and punishments in a strategy of resocialization from the use of rewards and punishments in two other strategies, GRIT (graduated reciprocation in tension reduction) and “tit for tat,” which have different and more limited aims than the resocialization strategy. GRIT is not a strategy for resocializ- ing and reforming outlaw states. Rather, it has the much more limited aim of removing distrust between states and thereby paving the way for a relaxation of tensions. GRIT attempts to do so by taking a series of meaningful conciliatory Reforming Outlaw States 53 adversary does not reciprocate, it is highly questionable whether additional positive moves would be consistent with the strategy of conditional reciprocity. This somewhat abstract conceptual discussion of alterna- tive strategies is useful only up to a point in policymaking. There are uncertainties in gauging whether the adversary is likely to be more receptive to one approach than to another or, indeed, to any of them. Policymakers may have to operate without reliable knowledge of the opponent's receptivity and likely response. And it may be difficult to correctly interpret the adversary's response. Intelligence sources and diplomatic communication may be helpful in reducing these uncertain- ties but are not likely always to eliminate them. Thus, willingness to experiment and rely on trial and error may be necessary. However, the differences among the strat- egies should not be ignored or blurred in practice. For ex- ample, it is possible that at various times the Bush administra- tion's policy of friendship toward Saddam Hussein blurred the important differences among GRIT, bribes, conditional reciprocity, and behavior modification. To the extent that blurring occurred, it would have further complicated the al- ready difficult task of evaluating the efficacy of the policy of friendship and taking appropriate corrective measures. As for the time-honored, if not always effective, practice of tit for tat, it received fresh attention during the Cold War as a possible strategy for eliciting cooperative behavior between actors whose mutual interests call for cooperating to avoid the worst possible outcome for both but who cannot easily do so because they are caught in a “prisoners' dilemma” (PD) situation. The relationship between a Great Power and the outlaw state it is attempting to reform, however, is not at all similar to the relationship between actors caught in a PD. The PD game is built on the premise that in a given situation the two sides have a latent interest in cooperating to avoid the worst possible outcome of their interaction; the challenge of the game for them is to act toward each other in ways that 54 U.S. Policy toward Iraq secure the better outcome that both prefer. The results of a computer simulation devised by Robert Axelrod indicated that in repeated plays of the PD game the tit-for-tat strategy performed best in achieving cooperation. This strategy bears a resemblance to some forms of conditional reciprocity in calling for each side to reward a conciliatory move by the other with a conciliatory move of its own and for responding to a hostile move with a negative one of its own until the two sides eventually converge in trading only positive moves; hence “cooperation” is established. 10 However, unlike tit for tat, which is a symmetrical game, resocialization is an asym- metrical game; one actor attempts to bring about fundamen- tal changes in the attitudes as well as the behavior of the other. The use of rewards and punishments in resocialization strategy has to be much more refined and more finely cali- brated. Efforts to use conditional reciprocity on behalf of the re- socialization objective are more likely to make headway when leaders of the outlaw state have begun to question the results of their antipathy to certain norms and practices of the inter- national system and, having become somewhat disenchanted with their earlier policies, are now willing to question the as- sumptions on which those policies were based. Consideration should be given to building into the practice of conditional reciprocity “tests” designed to find out whether the leaders of the outlaw state are genuinely moving toward abandoning earlier hostile attitudes and are ready to accept the norms and constraints of the international system. If they are not, the conclusion may be justified that the op- posing leaders cannot be resocialized and that the only alter- native is containment or efforts to bring about their replace- ment by more tractable leaders. In employing conditional reciprocity as a lever, what one "gives” the outlaw state and what one demands in return re- quire strategic planning. A series of incremental steps must be planned or improvised, yet the strategy must be imple- Reforming Outlaw States 55 - -- mented flexibly on the basis of monitoring and feedback. There must be awareness of the risks of the strategy and ways of minimizing and controlling those risks, and sensitivity to indications that the strategy is not working and needs prompt reassessment. 11 In all these respects it appears that the administration's ef- forts vis-à-vis Saddam fell short of what was needed and con- tributed to failure. What, then, are some of the risks of the strategy of condi- tional reciprocity and ways of minimizing them? It is not yet possible to derive firm answers to this question from studies of historical cases in which something like the strategy of con- ditional reciprocity was employed. In the meanwhile, by drawing on general principles of behavior modification and learning theory, some hypotheses can be formulated as to the risks of the strategy and possible ways of minimizing or avoid- ing them. 12 In listing these hypotheses here, I also raise the question whether these risks were adequately understood in the administration's policy toward Saddam Hussein. 1. Concessions and benefits bestowed should not be linked merely with general injunctions to improve behavior, and they should not be provided simply on the basis of the out- law's vague assurances of better behavior. Rather, the offer of benefits should be coupled with a demand (however diplo- matically conveyed) for quite specific changes in behavior that the outlaw state understands and agrees to. This ap- proach is consistent with a cardinal principle of the psycho- logical technique of behavior modification, which emphasizes that the therapist must identify for the subject the specific behavior that is to be extinguished and the more appropriate, acceptable specific behavior that should replace it. (Of course, it is possible that the outlaw state will refuse to accept the linkage of benefits to be received with some or all of the behavior changes demanded.) 2. Benefits generally should not be bestowed on an outlaw state in advance for reciprocity at some later date. Doing so 56 U.S. Policy toward Iraq violates another basic principle of behavior modification, which emphasizes positive reinforcement by means of a re- ward after the subject has performed required behavior and rejects the alternative practice of offering a bribe in advance to elicit the required behavior.13 3. The concessions and benefits bestowed on an outlaw state should be capable of being withdrawn or at least termi- nated if its leaders renege on their part of the reciprocal ar- rangement. If the concessions are not reversible, they should be in the nature of acceptable losses and the outlaw state should be punished in some other way for its delinquency. 4. Insofar as possible, concessions and benefits should give leaders of the outlaw state and its people a stake in continuing the process of conditional reciprocity and an awareness of the advantages of accepting and participating in the interna- tional system. (This is probably what Henry Kissinger had in mind when, during the détente of the early 1970s, he spoke of weaving a "web of incentives” to encourage Soviet leaders to enter into playing a more “constructive” role in interna- tional affairs). I have provided a provisional sketch of conditional reci- procity, its general requirements, and some of its risks. It should be obvious that this strategy is not assured of success and that its chances of succeeding may depend on a slow, in- cremental, patient application of conditional reciprocity. In addition we must recognize the following three complicating factors that may jeopardize efforts to pursue this strategy, all of which probably played a role in U.S. policy toward Iraq during and after the Iran-Iraq War. 1. The Great Power antagonist may need the outlaw state's support to orchestrate an effective balance of power against an aggressive third party. 2. The Great Power may mistake tactically motivated good behavior by the outlaw state as evidence of a strategic change for the better in that state's orientation to the norms of the international system. Reforming Outlaw States 57 3. Even a coherent, well-conceptualized long-range policy for attempting to resocialize the outlaw state may not be im- plemented consistently for various reasons. For example, the Great Power may be distracted by other foreign policy prob- lems; obtaining and maintaining domestic and international understanding and support for the long-range resocializa- tion policy may be difficult; bureaucratic officials may fail to implement policy fully or to correctly understand the policy laid down by top policymakers; and intra-administration dis- agreements on specific policies toward the outlaw state may undermine a more purposeful and consistent use of rewards and punishments. (These difficulties of implementation, of course, are not unique to the task of carrying out a policy of resocialization; they are also encountered in the conduct of foreign policy more generally.) I noted earlier the absence of any systematic scholarly study of past efforts to reform outlaw states and to draw their leaders into acceptance of the norms and practices of the in- ternational system. The several hypotheses provided in this chapter about the requirements and modalities of resociali- zation need to be assessed through comparative studies of past efforts of this kind, some successful and others not. The absorption of Kemal Ataturk’s Turkey into the international system is an example of successful integration of what was regarded initially, particularly by the British, as a possible outlaw state or, at least, as one situated outside the interna- tional community. In the contemporary era, efforts to deal with Iran, Vietnam, and Cambodia will be worth studying from this standpoint. The Nixon-Kissinger détente policy probably constitutes an example of a flawed version of the strategy of resocializa- tion insofar as its objectives included the long-range one of encouraging the Soviets to mend their ways and enter into a new “constructive relationship” with the United States. The development of a more constructive relationship between the two superpowers was to serve as the foundation for a new 58 U.S. Policy toward Iraq international system—what Nixon vaguely referred to as “a stable structure of peace.” However, as many commentators noted, Nixon and Kissinger do not appear to have clearly conceptualized or elaborated what they had in mind in this respect. To be sure, the grand strategy for achieving this long-range objective combined rewarding the Soviets for good behavior with punishing them for unacceptable behav- ior. In other words, it was a carrot-and-stick strategy that at- tempted to employ, although imperfectly, behavior modifica- tion and conditional reciprocity. The conciliatory component of the strategy offered the Soviet Union a number of benefits it prized: the possibility of greater trade and more access to western credits, grain, and technology; the possibility of en- hanced international status and recognition as a superpower equal to the United States; and the possibility of agree- ing to the Soviets' long-standing desire for formal recognition of the territorial changes in Eastern Europe and acceptance of the Soviet Union's dominant position in that area. In return, Nixon and Kissinger hoped that once the Soviet Union acquired a strong stake in the détente process it would act with restraint in the Third World lest it jeopardize bene- fits it was receiving from the evolving relationship. In the meantime, when the Soviets misbehaved in the Third World, Nixon attempted to react sharply. In this context U.S. leaders urged on the Soviets in general terms the necessity to adhere to a new set of norms and rules of conduct for restraining competition and avoiding conflict throughout the world. The underlying premise was that if these efforts were effective, not only would such norms and rules evolve over time but they would eventually be internalized by Soviet leaders and shape their behavior thereafter. The strategy of resocialization in this case was flawed both conceptually and in implementation. Aside from attempting to weave a web of incentives to induce restraint in Soviet for- eign policy—or, as one commentator put it, to create a new type of Soviet self-containment—it was not clear what re- 60 U.S. Policy toward Iraq for its own interests but also with their challenge to the norms and practices of the international system. More systematic knowledge regarding the uses, limitations, and risks of the strategy of attempting to reform an outlaw state is not merely of historical or theoretical interest. Rather, it has considerable relevance for contemporary U.S. foreign policy. For example, in early 1992 the administration for- mally reviewed U.S. policy toward Iran in order to consider adopting a strategy of constructive engagement that would entail lifting some economic sanctions. According to the New York Times, the policy review, completed in April, concluded that any gesture that “might be politically meaningful in Teh- ran-lifting the ban on oil sales to America, for example, would have been politically impossible at home. On the other hand, a reward small enough to be painless in American po- litical terms, such as lifting the ban on exports of carpets and pistachios, would have seemed too petty to Tehran.” The pol- icy review's conclusion that the time was not propitious for adopting a new policy is said to have been influenced by the earlier failure of constructive engagement toward Iraq. Ac- cording to the New York Times, “even those analysts who de- fend the use of incentives to moderate behavior are be- wildered about how to treat Iran,” recognizing that the Iranian government's moves to curb radical elements and to expand ties with the West may be only a tactical maneuver that could be reversed when Iran succeeded in reconstruc- ting its economy. 15 5. Appeasement as a Strategy for Conflict Avoidance To show stubborn unyieldness to an opponent who possesses a real sense of grievance over specific issues may be as dan- gerous as to make concessions to an opponent whose ambi- tions are endless.-Evan Luard! Elements of appeasement can be seen in the policy of friend- ship the United States pursued toward Iraq until the Persian Gulf crisis erupted in late July 1990. Whether one wishes to call it appeasement or prefers the term conciliation or construc- tive engagement, a review of that policy warrants the observa- tion that it was neither well conceptualized nor well imple- mented. In any case, it will be useful to discuss here the nature of appeasement and its uses and limitations. Many students of international relations have noted that the failure of the western powers to thwart the early aggres- sive moves of Japan and Hitler's Germany in the 1930s had a profound impact on subsequent American foreign policy at- titudes. The lesson of Chamberlain's abortive effort to ap- pease Hitler at Munich in 1938 has been deeply etched in the consciousness of successive generations of policymakers and foreign policy specialists. There is ample evidence that on nu- merous occasions since the end of World War II American 61 62 U.S. Policy toward Iraq policymakers have been influenced by the simple proposition “If appeasement, then World War III.” “Appeasement” became such an invidious phenomenon that little effort was made by policymakers and scholars alike to differentiate between those special conditions under which appeasement was indeed a misguided and dangerous policy and other conditions when it could eliminate the sources of conflict and the possibility of war with another state. Seven- teen years after Fred Iklé deplored the undiscriminating condemnation of appeasement, the distinguished Australian international relations scholar J. L. Richardson could still observe that the contribution of contemporary theories of in- ternational relations to the study of appeasement “has been disappointingly limited.”? At the same time, however, Rich- ardson's own contribution focused almost exclusively on Chamberlain's abortive effort to appease Hitler. A serious scholarly effort is only now under way to make a systematic comparative analysis of a variety of historical cases. The purpose of that analysis is to identify the conditions under which appeasement is a viable conflict avoidance strat- egy and other conditions in which appeasement is misguided and may well contribute to the eventual onset of war.3 The classic definition of appeasement is a simple one. In the language of diplomacy employed in the European balance-of-power system, appeasement referred to a policy of attempting to reduce tension between two states by the me- thodical removal of the principal causes of conflict between them. In this sense appeasement was regarded as a strategy for eliminating the potential for war in a conflict-ridden re- lationship between two states. Whereas the classic definition refers to the removal of all the principal causes of conflict in the relationship, partial appeasement is also possible, leaving some sources of conflict untouched. The definition of ap- peasement leaves unanswered the important question of how to go about removing causes of conflict in a relationship with 66 U.S. Policy toward Iraq dealing with an outlaw state whose leaders essentially reject the norms and practices of the international system and are disposed to behave in ways that will undermine the order and stability of the system. Appeasement of such actors is neither desirable nor feasible, given their destructive orientation to the existing international system (see chapter 4). On the other hand, when the adversary is not an outlaw but advances either revisionist or expansionist claims, the basic policy choices are conciliation, deterrence, or some combina- tion of the two. Appeasement need not and often should not attempt to satisfy all of the revisionist or expansionist aims of the other party in a single grand settlement. There may be advantage in conducting appeasement in a careful, incre- mental fashion. Appeasement can thereby be incorporated into a strategy of conditional reciprocity by means of which one secures compensating concessions or assurances of one kind or another from the adversary. Appeasement can aim at either a limited reduction or a comprehensive elimination of the conflict potential in the re- lationship with the adversary. Partial appeasement—that is, appeasing some of the adversary's demands—may be re- garded as an acceptable short-term strategy for defusing a war-threatening crisis. But even if partial appeasement is suc- cessful in this respect, other sources of conflict in the relation- ship remain that may provoke new crises and military conflict in the future. Or appeasement can have the more ambitious objective of thoroughly removing, either in one move or in- crementally, all major sources of conflict in the relationship. Another useful distinction is between passive appeasement and active appeasement. Passive appeasement is the practice of not explicitly agreeing to, but not opposing, the adversary's alteration of a status quo situation-for example, France's ac- ceptance of Hitler's occupation of the Rhineland. Active ap- peasement, on the other hand, is a diplomatic agreement that acquiesces to a forthcoming alteration of the situation for Appeasement 67 example, the Munich agreement with Hitler, which accepted the forthcoming incorporation of the Sudetenland portion of Czechoslovakia. In the absence of systematic research on past efforts to em- ploy appeasement, there is no basis for articulating an oper- ational theory or generic knowledge of this strategy that might help policymakers decide whether appeasement and conciliation would be a viable strategy. But practitioners should consider a number of questions when deciding be- tween conciliation and deterrence in the face of demands for a change in the status quo. 1. Are the adversary's objectives revisionist or expansion- ist? If expansionist, are they of a limited, acceptable char- acter? 2. Will appeasement of the adversary whet his appetite and encourage him to pursue new expansionist aims? Will the adversary view concessions as evidence of goodwill, friendship, and recognition of the legitimacy of his revisionist claims, or as evidence of irresolution and weakness and there- fore a temptation to seek greater gains? 3. Can the adversary be appeased in such a way as to avoid giving the impression at home and abroad that one has yielded to blackmail? Will appeasement of the adversary re- sult in serious damage to one's reputation in the eyes of other states and encourage them to advance revisionist or expan- sionist demands of their own? 4. How can one limit or control the various risks of ap- peasing another state? By drawing a line as to the extent of concessions that will be made? By appeasing individual claims incrementally? By obtaining credible formal assurances from the adversary that his demands for changes in the status quo are limited? Can tests be devised to assess the scope of the adversary's intentions? 5. Is the expected benefit of appeasing the adversary lim- ited to the short-term objective of avoiding a crisis or war? Or 68 U.S. Policy toward Iraq can short-term appeasement on a specific issue be built into a longer range strategy of turning the entire conflictful rela- tionship into a cooperative one? 6. Is reliance on deterrence instead of appeasement a bet- ter strategy for coping with the adversary's hopes and de- mands for a change in the status quo? Will successful deter- rence induce the adversary to give up hopes and efforts for changing the status quo in the future? Or will it only strengthen his motivation and lead him to prepare for chal- lenging deterrence more effectively in the future? Beyond its psychological impact, will the change in the status quo in the adversary's favor that is being considered significantly alter the relative power balance? The last of these questions highlights the need for studying a variety of historical cases to better understand the conditions under which appeasement or deterrence is likely to be a more effective conflict avoidance strategy. In an adaptation of one of Stephen Rock’s suggestions, four possible situations and scenarios can be identified for analytical purposes, though policymakers can expect difficulty judging, especially at first, which of these four possibilities correctly identifies the case at hand:6 1. Either appeasement or deterrence can succeed, at least in the short run. A possible example is the Falkland Islands crisis, in which the British might have avoided the invasion of Argentina and the ensuing war through either a more robust deterrence effort or timely appeasement. 2. Neither deterrence nor appeasement is likely to suc- ceed, because the adversary is bent on employing military force. A possible example is Hitler's determination to go to war against Poland in the autumn of 1939. 3. Only deterrence can succeed, because the adversary would respond to appeasement by generating new demands. Appeasement 69 A possible example is Chamberlain's appeasement of Hitler on the Sudetenland question, which did not prevent him from occupying the rest of Czechoslovakia later. 4. Only appeasement can succeed, either because the de- fender lacks the capability or will or both to mount a robust deterrence effort or, if war breaks out, to pursue it effectively. A possible example is what Barbara Tuchman regards as En- gland's “missed opportunity” to appease and thereby retain its American colonies.? Comparative studies of successful and unsuccessful ap- peasement can help identify the conditions under which ap- peasement may be a viable strategy, the risks of the strategy, and ways of coping with the risks. Such insights can be gained, for example, by comparing British appeasement pol- icy toward Hitler with the gradual improvement in Anglo- French relations after the Fashoda crisis of 1898 and with Chancellor Willy Brandt's successful Ostpolitik policy toward the Soviet Union and Eastern Europe. Chamberlain's effort to employ appeasement as a tool for the larger objective of creating the conditions for a new structuring of peace and stability in Europe was, according to one interpretation, vi- tiated by his “impatience and contempt for the cautionary ex- pedients of traditional diplomacy": While between 1933 and 1937, the British Foreign Office had sought, however imperfectly, to establish the bona fides of the dictators and to determine whether there were common in- terests and values that would serve as a basis for cooperation before undertaking a systematic removal of specific causes of conflict, Chamberlain saw no point in these preliminaries and soon got rid of those people, like Eden and Vansittart, who insisted on them. [He) heedlessly plunged ahead to what should have been the final stage of a laborious process, on the mistaken assumption that Hitler and Mussolini shared his ob- jectives and that a demonstration of his readiness to satisfy their demands would bring the peace that he desired. 8 In contrast, Willy Brandt initiated his Ostpolitik policy with a clear notion not only of the changes he wished to achieve 70 U.S. Policy toward Iraq but also of the difficulties and risks that lay in his path. His policy embraced the possibility of making major concessions to the Soviet Union and the East German regime and, hence, certainly contained elements of appeasement. But Brandt skillfully employed the strategy of conditional reciprocity to orchestrate a complex process to ensure that he would gain the important benefits he sought in return. “Unlike Cham- berlain in his dealings with Hitler, Brandt did not rush the process in order to move as quickly as possible into ap- peasement.” Rather, Brandt minimized and controlled the risks “by moving one step at a time and then waiting until the situation ripened before moving ahead again.”9 A survey of the history of international relations should easily identify many other instances in which appeasement and conciliation either worked or did not, thereby enabling scholars to produce the conditional generalizations that will refine our understanding of when this strategy is and is not likely to be a viable one and how best to implement it. 10 Deterrence and Reassurance 73 not allow American troops on their soil Saddam may have believed that, therefore, there could not be a strong U.S. mil- itary response to his move against Kuwait. As in earlier historical cases-for example, the North Ko- rean attack on South Korea in June 1950—when the aggres- sor judges a commitment to be lacking, the most rational strategy available to him for challenging the status quo is the fait accompli—the quick, decisive use of ample force to achieve the objective before the potential defender of the state subjected to attack can reconsider its policy and decide to intervene. And this is, indeed, the strategy of aggression Saddam chose to follow.2 Even at the time, it was clear that Washington faced diffi- cult obstacles in trying to mount a stronger deterrence effort. As a senior administration official later explained to report- ers, “We were reluctant to draw a line in the sand.” Continu- ing, he added that one could not envisage at the time that the American public and Congress would support the deploy- ment of substantial American military force for deterrence purposes over a dispute involving twenty miles of desert ter- ritory. Besides, it was clear that the local Arab states, in partic- ular Saudi Arabia, would not have allowed the stationing of U.S. troops on their territory to deter an attack on Kuwait- all the more so, because Saudi leaders did not believe it would occur. And, as the senior official went on to point out, the administration did not wish to risk asserting a deterrent threat it was unprepared or unable to carry out if its bluff were called: “The basic principle is not to make threats you can't deliver on. That was one reason why there was a certain degree of hedging on what was said.”: This case, then, illus- trates some fundamental difficulties that can be encountered in efforts to employ deterrence strategy. The lesson is a sobering one and points to the limitations of deterrence strategy. What the United States was willing and able to do after the aggression against Kuwait, it was not able to threaten to do on behalf of_deterrence. before_the 74 U.S. Policy toward Iraq invasion occurred. It would be unjustified to read history backward, as critics are wont to do, and argue that the admin- istration should have been able and willing to threaten on be- half of deterrence what it was able and willing to do to oppose the aggression after it occurred. The present case illustrates once again a lesson that emerged from earlier studies calling attention to the limita- tions of deterrence strategy as an instrument of foreign pol- icy. Deterrence cannot be counted on to compensate for a flawed foreign policy based on erroneous assumptions, such as the administration's prewar policy toward Saddam. A ma- jor power such as the United States, with its complex bu- reaucracy and decision-making process, cannot quickly or easily reverse direction and quickly improvise a strategy of deterrence when the adversary suddenly and unexpectedly poses the threat of aggression. Once a policy is established toward another state it acquires a momentum of its own that cannot be easily reversed. As a number of observers have noted, the Bush administration's response to the July crisis in the Gulf was influenced by its long-standing policy of devel- oping better relations with Hussein's Iraq. Prior to the July crisis the administration was operating on the premise that the best way to handle Saddam and moderate his behavior was through offering conditional incentives and an improve- ment in relations in return for his cooperation and good be- havior. (Whether this policy was clearly and consistently con- veyed is another matter.) In other words, the essence of the policy was an attempt to resocialize the outlaw Iraqi state and reform its leader. That policy was based on what turned out to be dubious assumptions about Saddam's aspirations, inten- tions, and mind-set-assumptions that were not subjected to critical examination until it was too late (see chapter 4). Washington's sudden concern over Saddam's intentions in late July and its move toward deterrence did not succeed al- together in overcoming the momentum of its earlier policy. Even while signaling its weak deterrence effort, the adminis- Deterrence and Reassurance 75 1 tration added reassurances that it wanted to remain on friendly terms with Iraq. Such reassurances could only fur- ther dilute the already weak deterrence effort. One is reminded of the equally misguided reliance on re- assurances of friendship that President Harry Truman and Secretary of State Dean Acheson conveyed to China in the autumn of 1950 when the Chinese leaders threatened to in- tervene in the Korean War if U.S. forces pursued the de- feated North Korean forces beyond the thirty-eighth parallel into North Korea. In employing deterrence, a correct image of the adversary's behavioral patterns is often critically im- portant. The adversary's mind-set and calculations are not easily or reliably accessible. We can illustrate this point by briefly considering two scenarios (both drawn from historical cases), elements of which may apply also to Saddam's calcula- tions leading to the attack on Kuwait. In one scenario the adversary whom one is trying to deter may feel strong domestic or international pressure to act to change the status quo now. Such was President Nasser's situ- ation in the crisis with Israel in 1967 when, according to one interpretation, he was pressed to move from one provocative action to another against Israeli security interests until, con- trary to his wishes and expectations at the beginning of the crisis, the situation escalated to war. In this scenario, the ad- versary is led to take greater and greater risks, either know- ingly or by unrealistically minimizing the risks. Deterring an adversary under these circumstances becomes appreciably more difficult, all the more so if the deterrer is not aware of the domestic and international pressures driving the adver- sary; the deterrer may assume incorrectly that the adversary is “rational” and that strong, credible deterrence is enough to induce prudence and caution. In a second scenario the adversary may feel obliged to act against the status quo because he is operating with an exag- gerated, even quite unrealistic, perception of the threat. The perception may be that the other state is considering or will 7. The Failure to Coerce Saddam Hussein Some high-level U.S. policymakers and other foreign policy specialists were genuinely surprised that coercive diplomacy failed to persuade Saddam to withdraw his forces from Ku- wait. Coercive diplomacy failed even though U.S. policymak- ers had a good conceptual understanding of the strategy and its requirements and acted accordingly. The explanation for the failure lies elsewhere, but the exact explanation cannot be easily determined. The answer lies buried in the values, be- liefs, perceptions, and calculations that influenced Saddam Hussein's judgment and led to his decision to fight rather than accept the ultimatum of January 15, 1991. Before proceeding I will discuss the general concept of coercive diplomacy and indicate the limited though useful contribution it makes in helping policymakers devise a spe- cific strategy that fits the particular situation at hand well enough to promise success. The general concept or model of coercive diplomacy identifies the major ingredients of the strategy and the logic that must be created in the mind of the adversary if it is to succeed. The general concept of coercive diplomacy becomes a strategy only when the policymaker gives specific content to the four components of coercive 79 Failure to Coerce Saddam 81 Security Council in early October carefully considered the utility and risks of subjecting Saddam to an ultimatum. (Evi- dently, therefore, the possibility of replacing sanctions with an ultimatum was a serious option at an early stage in the development of the crisis.) Drawing on historical experience (i.e., what I refer to elsewhere in this study as generic knowl- edge), the group appears to have identified four possible risks of an ultimatum: 1. The opponent may think you are bluffing. 2. The opponent may take the ultimatum seriously but de- cide to initiate war himself rather than accept the demand made on him (as the Japanese did in deciding to initiate war against the United States in December 1941). 3. The opponent may reject the ultimatum because he re- gards acceptance of it as humiliating, incompatible with honor, too damaging politically, and the like. 4. The opponent may neither accept nor reject the ulti- matum outright but try to defuse its coercive impact by ac- cepting the demands partially or with qualifications. This working group undertook the difficult task of estimat- ing whether any of these risks was likely to materialize if an ultimatum were employed against Saddam and how these risks might be minimized, if not avoided altogether. To this end, area experts were consulted. The full results of the group's analysis have not been made available for this study.2 It appears that the major concern of policymakers at first was to make the threat of punishment so credible that Saddam would be extremely unlikely to regard it as a bluff. At the same time it appears that the working group was sensitive to the possibility that too explicit and harsh an ultimatum might harden his resistance; accordingly, an effort was made to give coercive diplomacy the character of a game of “chicken” that would last for some time. Later, however, as the January 15 deadline for compliance approached, the fourth risk of using ultimatums became a significant concern for the administra- tion. The possibility that Saddam might announce partial or 82 U.S. Policy toward Iraq conditional withdrawal from Kuwait was referred to by well- informed journalists as the administration's “nightmare sce- nario.” Very detailed contingency plans were made to deal with such a development and were discussed with coalition partners. The prior question, of course, was why the administration considered it necessary to move from economic sanctions to a tough ultimatum backed by the threat of war. This decision was made not only because of the difficulty of making reliable intelligence estimates of the effect of sanctions on Iraq's econ- omy (see chapter 3). In addition, there was concern that sanc- tions would take too long and might not be effective, and that it might be difficult to maintain the international coalition over a long period of time. Pressure to bring the crisis to a head quickly was encouraged by knowledge that weather conditions during the forthcoming summer would make combat difficult. Also, there was considerable pressure from Saudi Arabia to move to a threat of military action. Finally, the rationale for shifting to the ultimatum was strongly sup- ported by personality assessments of Saddam Hussein that encouraged the belief that only an ultimatum backed by the threat of war had a chance of impressing him with the need to back down. However, not all U.S. policymakers were sanguine that coercive diplomacy would be successful.” Besides, some mem- bers of the administration believed that success in coercing Saddam Hussein to withdraw from Kuwait would be an un- satisfactory outcome to the crisis: it would leave him in power, his military forces intact, and Iraq free to pursue its military programs for development of weapons of mass destruction. From this standpoint, failure of coercive diplomacy could be countenanced in that it would provide an opportunity to use military force to remove Saddam from power, destroy his military forces, and end Iraq's weapons development pro- grams. Let us consider more closely the image of Saddam held by Failure to Coerce Saddam 85 6 threat of war, Post added that cultural factors had probably contributed to such miscalculations. “In the Arab world,” Post observed, “having the courage to fight a superior foe can bring political victory, even through a military defeat. . . Post also suggests that the dynamics of the crisis may have led to an inflation of Saddam's heroic self-image: “Intoxi- cated by the elixir of power and the acclaim of the Palestin- ians and the radical Arab masses, Saddam may well have been on a euphoric high and optimistically overestimated his chances for success.” Moreover, by publicly committing him- self during the crisis to the Palestinian cause, thereby gaining widespread approval in the Arab world and enhancing his self-image, Saddam may have painted himself into a psycho- logical corner. It became difficult thereafter to retreat with- out dishonor. According to this interpretation, Saddam thereby lost the capacity he had demonstrated on many pre- vious occasions to retreat when necessary. A number of conclusions and lessons can be drawn from the failure of coercive diplomacy on this occasion. The case illustrates the importance of having an accurate profile of the adversary toward whom coercive pressure is directed. Ironi- cally, however, it also illustrates at the same time the difficulty of fine-tuning the strategy so as to activate precisely that facet of the adversary's makeup that would ensure the desired re- sponse. Even when an insightful, sophisticated psychological model of the adversary is available, is supplemented by a knowledge of the adversary's political culture and political system, and is used appropriately in policymaking, there can be no guarantee of success. The adversary's reluctance to un- dertake a humiliating, costly retreat (one of the risks of an ultimatum referred to earlier) can activate psychological ten- dencies in him to look for indications that the threat is not credible or that the coercing state lacks the political will to engage in a costly war. Similarly, if the adversary falls prone to the familiar tendency toward wishful thinking his estimate of the relative military strength of the two sides will be 86 U.S. Policy toward Iraq distorted. As cognitive psychology has repeatedly and per- suasively emphasized, a person tends to give greater, often uncritical weight to new information that supports an exist- ing policy or preference and tends to discount evidence that challenges his existing mind-set. As a result, the adversary may make critical miscalculations that feed on his distaste for drawing the conclusion that it is really in his best interest to meet the demands made on him. Such miscalculations can be abetted if the adversary's mind-set includes, as Saddam's evi- dently did, an image of the coercing state as lacking the polit- ical will to engage in a tough battle in which it would have to take heavy casualties. The failure of coercive diplomacy in the Gulf crisis also calls attention to the impact the adversary's self-image can have on his calculations and judgment when he is forced to decide whether to meet the demands made on him. That Saddam entertained an inflated image of himself as a heroic leader destined to transform the Arab world was known to those who have studied him. What was not anticipated, Post suggests, is that Saddam's self-image may have been so mag- nified by developments during the crisis that retreating be- came more difficult for him. In sum, the lesson is that the outcome of coercive diplo- macy may depend on psychological, cultural, and political variables operating on the adversary, which may be difficult to foresee and difficult to deal with to ensure the success of the strategy The Gulf crisis was a tough case for coercive diplomacy for a number of other reasons as well. These reasons emerge sharply in a comparison of the Gulf crisis, in which the strat- egy failed, with the Cuban missile crisis, in which it succeeded (see table 7.1). In the missile crisis Kennedy and Khrushchev could cooperate to avoid war because neither leader believed that their disagreement approximated a zero-sum contest, and because their image of war, should it occur, was that of a nuclear catastrophe. In contrast, President Bush and Sad- Failure to Coerce Saddam 87 Table 7.1. Six Variables That Help Explain Success or Failure of Coercive Diplomacy. Variables That Favor Success Cuban Missile Gulf of Coercive Diplomacy Crisis Crisis + + + + Non-zero-sum view of the conflict Overwhelmingly negative image of war Carrot as well as stick Asymmetry of motivation favoring state employing coercive diplomacy Opponent's fear of unacceptable “punishment” for noncompliance No significant misperceptions or miscalculations + + “+” = variable present; variable absent. dam Hussein tended to see their conflict in zero-sum terms, and this tendency was reinforced by the highly invidious im- age each had of the other. Moreover, unlike Kennedy and Khrushchev, who were horrified by the possibility that, if mis- managed, the crisis could lead to thermonuclear war, Bush and Saddam held an image of the outcome, costs, and conse- quences of war that was not distasteful enough to motivate them to seek a compromise settlement. Incentives for coop- erating to avoid war were lacking on both sides. Indeed, it is not far-fetched to characterize the Bush administration's pol- icy as coercive diplomacy without fear of the consequences of failure. In the Cuban crisis, moreover, Kennedy coupled his ulti- matum with a substantial carrot: an agreement not to invade Cuba and a secret agreement to remove the U.S. Jupiter mis- siles from Turkey. In the Gulf crisis Bush relied solely on the “stick” and offered no carrot for a compromise settlement, insisting that there be no reward for aggression, although the ingredients for face saving were available to Saddam. In the 88 U.S. Policy toward Iraq Cuban crisis, in contrast to the Gulf crisis, neither side had any significant misperceptions or miscalculations during the crisis that might have led to war. In the Cuban crisis, again unlike the Gulf case, Kennedy operated with an image of Khrushchev as a leader who was capable of retreating, which he was able to capitalize on in orchestrating an effective carrot-and-stick variant of the ultimatum. Two other psychological variables that appear to have been important in past cases of coercive diplomacy were not pres- ent in the Gulf crisis. First, Kennedy succeeded in impressing Khrushchev that getting the missiles out of Cuba was more important to the United States than keeping the missiles there was to the Soviet Union. Evidently Bush did not suc- ceed, despite considerable efforts, to convince Saddam that getting Iraqi forces out of Kuwait was more important to the United States than refusing to remove them under threat of war was to Iraq. Second, Kennedy created fear of unaccept- able escalation of the crisis in Khrushchev's mind, but Bush did not succeed in creating fear of unacceptable punishment in Saddam's mind for noncompliance with the ultimatum.? A final comment on the experience with coercive diplo- macy in the Gulf crisis seems appropriate. The administra- tion's willingness to contemplate use of force was a factor from an early stage in the crisis, when President Bush stated that Iraqi occupation of Kuwait was unacceptable. There- after, American policy was driven as much by the objective of creating and maintaining an international coalition under the aegis of the UN Security Council as it was by the desire to persuade Saddam Hussein to withdraw from Kuwait, if not more. Although the strategy of coercive diplomacy had little chance of success, the attempt to employ it in the hope of avoiding war was necessary for building and maintaining in- ternational and domestic support for the objective of liberat- ing Kuwait. Ironically, the failure of coercive diplomacy was necessary to gain support for war when war became the last resort. 8. War Termination: Integrating Military and Political Objectives The ending of the Gulf War constitutes an unusual, in some ways a unique, case of war termination. In some respects it resembles the ending of the war with Nazi Germany and Ja- pan: the enemy was decisively defeated, and terms for a cease-fire were not negotiated with, but imposed on, the ad- versary. But the terms imposed on Iraq, unlike those for Ger- many and Japan, were not those of unconditional surrender, military occupation of the entire country, and replacement of the existing regime with a military government. Rather, for various reasons the United States was determined to avoid too much involvement in postwar Iraq. Though the UN coa- lition's military and political war aims in the Gulf War were extensive, they were limited in comparison with the terms im- posed on Germany and Japan. General theory and knowledge about war termination have been slow to emerge from scholarly studies. This deficiency in scholarship is largely due to the fact that there are many different types of war, different types of cease-fires, and dif- ferent types of war termination. One might add that in wars that end in military success for one side, the type of postwar relationship the victorious power wishes to develop with the 89 War Termination 91 tives constrain the possibility of more effective use of force to achieve political objectives? To be sure, the allied coalition in the Gulf War was re- soundingly successful in achieving all of its military objec- tives. Kuwait was liberated, the Iraqi army was routed, Iraq's military capability and its infrastructure were greatly weak- ened. Achievement of these military objectives also accom- plished the most important of the allied coalition's political objectives. Thus, the regional threat posed by Iraq's powerful military forces was sharply reduced. And the general objec- tive of establishing peace and security in the Middle East, called for by Security Council Resolution 678 of November 29, 1990, which had authorized use of force to liberate Ku- wait—was furthered by stiff cease-fire terms that imposed on Iraq the obligation to destroy all nuclear, chemical, and bio- logical weapons and production facilities as well as missiles capable of reaching the territory of other countries in the re- gion. Iraq was obliged to accept and cooperate with UN ef- forts to identify all such capabilities and to monitor and carry out their destruction. (Because it is beyond the scope of this study I will not discuss the difficulties that had to be overcome in implementing these terms.) However, an important political objective—the removal of Saddam Hussein from power—was not accomplished. To be sure, this objective was not explicitly authorized by the Secu- rity Council. Nor was it an official military objective of the Bush administration. That is, it was not among the military objectives formally assigned to General Schwarzkopf's forces; accordingly, military strategy did not include pursuit, cap- ture, or elimination of the Iraqi leader. Nonetheless, there can be no question that Saddam's ouster was keenly desired by the Bush administration. During the war, Washington urged the people of Iraq to force the dicta- tor out, a theme purveyed also in psychological warfare. Al- though Saddam's arrest or removal was not made a condition 92 U.S. Policy toward Iraq for a cease-fire, the cease-fire provisions were designed not only to achieve the other political-military objectives already mentioned but also to maintain pressure for Saddam's over- throw. Washington frequently hinted that Iraq's punishment would be gentler and its recovery quicker if Saddam were re- placed.” And the administration made last-minute efforts be- fore the cease-fire of February 28 to use two specially pre- pared deep-penetration bombs to “get” Saddam in one of his hardened bunkers. 4 As time passed after the end of the war, the administra- tion's failure to eliminate Saddam became a major source of political criticism at home. In part, such criticism was strengthened by expectations that had been created by Presi- dent Bush's earlier efforts to galvanize domestic support for the war by depicting Saddam as another Hitler. Not only was Washington criticized for having stopped the war before se- curing Saddam's removal, it was also criticized for standing by and allowing Saddam's forces to put down the Kurdish and Shi'ite rebellions that quickly erupted after the cease-fire. Difficulties in integrating military objectives and political objectives arise particularly in limited wars—such as Korea, Vietnam, and the Gulf-rather than in total wars such as those waged against Nazi Germany and Japan. By definition, limited wars are fought for limited objectives, which in turn are pursued by limited military means. In some limited wars, political leaders impose constraints that prevent military leaders from using optimal military strategy and tactics, and the result may be a gap between mil- itary strategy and political objectives. The Gulf War did not exhibit this particular difficulty, since the much-criticized de- cision to declare a cease-fire before U.S. troops completed the envelopment of Basra was not really consequential for achievement of war aims. As a result, it is true, some Iraqi forces escaped and could participate later in quelling the Shi'ite rebellion, but that outcome was not inconsistent with the UN coalition's limited political war aims. Besides, some War Termination 93 -- five hundred thousand additional Iraqi troops remained rel- atively undamaged in northern Iraq.5 In undertaking to end the Gulf War, U.S. policymakers had to be sensitive to its idiosyncratic features, but they were not able to anticipate and master all of them. Indeed, policy plan- ners were aware that many difficulties could arise in terminat- ing the war against Iraq and began serious contingency plan- ning for war termination as soon as the war began. They were aware, for example, of the classic difficulties and risks asso- ciated with cease-fires. But they did not anticipate the rebel- lions of the Kurds and Shi’ites almost immediately after the temporary cease-fire was declared on February 28 and the difficult problems and dilemmas these developments would create. Two questions therefore arise. Why did not the victo- rious allied armies pursue their advance into Iraq all the way to Baghdad to ensure the removal of Saddam from power? And why did not allied forces intervene to prevent the crush- ing of the Kurdish and Shi’ite rebellions? The brief answer to these questions is that to do so would have required Washington and its allies to escalate the mili- tary and political objectives they had set for themselves. There were, in fact, important limitations to the political ob- jectives. It is true that the UN coalition did intend to appre- ciably weaken Iraq militarily so that it would not constitute a threat to other states in the region. However, at the same time, it did not wish to weaken Iraq to the point of being un- able to resist pressures from Iran and Syria. Rather, although a much-weakened Iraq would emerge from the war, it should be capable of contributing to maintaining a regional balance of power. Accordingly, the UN coalition did not seek-in fact, it wished to avoid—a breakup of the existing Iraqi state into several small entities or a Lebanonization of the internal situ- ation within the country. For that reason, Washington's atti- tude was ambivalent toward the Shi'ite and Kurdish rebel- lions. This development, unexpected in war termination --- - 1 94 U.S. Policy toward Iraq planning, created severe policy dilemmas for the United States and the UN coalition. The rebellions were welcomed insofar as they might contribute to undermining Saddam's ability to stay in power, but if they were too successful they might set into motion the disintegration of Iraq and trigger possible moves into parts of Iraqi territory by Iranian, Syrian, and Turkish forces. The United States was not willing to do much to influence the outcome of the rebellions. The policy dilemma created by the rebellions for Washing- ton was compounded by their effect on the role the UN coa- lition hoped the Iraqi military would play in unseating Sad- dam. Now, quite unexpectedly, the Iraqi military had the difficult task of repressing large-scale rebellions in both the north and the south. Washington and its allies believed they had no choice but to accept the Iraqi military's fierce efforts, under Saddam's command, to suppress the rebellions. The alternative, as they saw it, was to accept the possibility that the rebellions would lead to a disintegration of the country that would be highly inconsistent with the allied coalition's politi- cal war aims. In other words, an unexpected and politically embarrassing convergence of interest in containing the rebel- lions emerged between the United States and Saddam. The rebellions and the need to suppress them gave Saddam a new lease on life. The rebellions also complicated and undermined for the time being any possibility that elements of the Iraqi military might move, as had been hoped, against Saddam. The un- stable internal situation in Iraq also discouraged the United States from offering additional incentives to the Iraqi military and the Ba'thist party to move against Saddam. Although the Iraqi military had to be denounced for its activity against the rebels and punished for violating the cease-fire prohibition on use of fixed-wing aircraft, Washington was basically un- willing to stand in the way of the efforts the Iraqi armed forces were making to contain the rebellions at least well enough to prevent disintegration of the state. As for the Iraqi 96 U.S. Policy toward Iraq of moving into North Korea to liquidate the Communist gov- ernment in Pyongyang. Flushed with its success in defeating the North Korean army, the Truman administration suc- cumbed to the temptation and pressure to shift from a lim- ited to a total war against Communist North Korea, where- upon the Chinese intervened and the United States found itself plunged into a prolonged and costly war. In the Gulf crisis the situation differed, but the weight of the histori- cal analogy was felt and reinforced the other major political- diplomatic constraints that argued for hewing to the limited political and military objectives the UN Security Council had authorized instead of seeking support for broadening them. As for the idea of sending U.S. forces all the way to Bagh- dad in the hope that doing so might lead to Saddam's ouster, this military option had been considered in the administra- tion's war planning and had been firmly rejected. This judg- ment, too, was strongly influenced by the U.S. post-Vietnam military doctrine that warned against military intervention on behalf of ambitious, amorphous political objectives. Accord- ing to the doctrine, force should be used only for achieving strictly military objectives and it should be employed in ac- cord with professional military judgment. The United States should avoid becoming involved militarily in situations in which it could not be successful in a relatively short time with minimal casualties. To send ground forces to Baghdad, it was believed, would entail substantial political and military risks of that kind. In late February, after coalition forces had routed the Iraqi army in the Kuwait theater and in southern Iraq, going on to Baghdad to get rid of Saddam and perhaps overthrow the regime would have been possible from a military standpoint but might well have resulted in substantial U.S. casualties and a prolonged involvement of American forces in a highly un- stable, volatile internal situation. (Relevant is the fact that approximately one-half million Iraqi troops, including a War Termination 97 number of Republican Guard divisions based in northern Iraq, had thus far suffered little damage.) This is not to imply that top civilian leaders in the Bush administration wanted deeper involvement and had to be dis- suaded by the military chiefs. Most civilian leaders evidently shared their inhibitions and, also for broader political and diplomatic reasons, were strongly determined to minimize U.S. involvement in postwar Iraq. Domestic political support for the war was geared to limited military objectives and might erode rapidly if Washington embarked on more ambi- tious goals. Nor could UN and coalition support for a more far-reaching military campaign be counted on. Besides, when the cease-fire was declared on February 28 further military action to bring about Saddam's downfall seemed unnecessary. It was widely believed within the admin- istration and elsewhere that Saddam could not survive in power after suffering so savage and humiliating a defeat; the devastating war would surely weaken Saddam's control over his military and his other instruments of repression so that the encouragement given by the UN coalition to get rid of him would bring about the desired result within a reasonably short time. If Saddam survived for the time being, pressure to remove him would be enhanced by postwar continuation of the stiff economic sanctions against Iraq. Could the administration have acted more effectively to en- sure Saddam's downfall and to cope with the policy dilemmas created by the rebellions? Critics of the administration's deci- sion to halt the war “prematurely” benefit, of course, from hindsight, and it can be said that they either do not under- stand or do not give enough weight to the considerations that led Washington to limit its political and military objectives. For some critics it would have been preferable to discourage or (if necessary) prevent the Iraqi army from repressing the rebellions and to accept the possibility that Iraq might break up into several entities. Other critics argue that the coalition War Termination 99 option of letting allied forces remain longer in southern Iraq to create additional pressure and leverage for Saddam's re- moval. Second, did the conservatism of the post-Vietnam U.S. mil- itary doctrine, as manifested in its insistence that force be em- ployed only to achieve strictly military objectives, unduly limit the possibility of more effective use of force and threats of force to achieve broader objectives in the Gulf War? Third, why were U.S. and allied intelligence services un- able to foresee the likelihood that after the defeat inflicted on Iraqi forces the Kurds and Shi'ites would rise in rebellion? Should Washington have established closer working relations with these groups during the early stages of the Gulf crisis? Since the administration had called on the people of Iraq to rise up against Saddam, did it not have a responsibility to pro- tect the Kurdish and Shi'ite rebels? 10 Finally, a few general remarks that will perhaps provide a framework for considering the difficult question of integrat- ing military and political objectives in limited conflicts of this kind. It is misleading and conceptually incorrect to pose the problem, as is often done, as a problem of matching military objectives and political objectives. Such a formulation as- sumes that—and begs the question whether—acceptable mil- itary means can always be devised that, if successful, will achieve the political objectives of a limited war. In fact, often there are political aims that cannot be realized solely via vic- tory on the battlefield. Not only are there inescapable limits on the utility of force as an instrument of policy, but unfore- seen consequences of military victory and unexpected devel- opments thereafter—as in the Gulf War—can handicap the ability of even sophisticated statesmen to convert military vic- tory into a full-blown political success. The achievement of 100 U.S. Policy toward Iraq political war aims in some conflicts, if indeed they are achiev- able to begin with, often requires diplomatic and other skills, but sometimes even these will not suffice. In sum, military success may be a necessary condition for achieving ambitious political war aims, but it is not a sufficient condition for doing so. Other variables and unpredictable de- velopments often come into play in the complicated “end game” that determines the political outcome of a limited war such as that in the Persian Gulf. In this war, as in some others, the end of the fighting begins a new phase in which the pur- suit of some of the political objectives of the war continues. In planning for such a war the critical question is how, and to what extent, will a particular kind of military strategy em- power the victor to achieve all of his postwar political objec- tives. My purpose has been to identify and analyze problems of war termination, difficulties that are not easily anticipated or avoided. Analysis focused on the difficulty of converting battlefield success into full achievement of political war aims, a problem that frequently arises in war termination and that needs more systematic comparative study than it has thus far received. Concluding Note to Part Two Part two has focused on critical developments in U.S policy toward Iraq from the end of its war with Iran in August 1988 through the termination of the Gulf War in late February 1991. At different points in this period the United States em- ployed six different strategies toward Iraq. Five of the strate- gies proved to be unsuccessful and the other resulted in a mixed outcome. My purpose has not been to write definitive, detailed his- tories of these efforts to influence Iraq or to contribute to political criticism of the administration for those of its strate- gies that were unsuccessful. Rather, my objective has been to provide analytical evaluation of the relationship of knowl- edge to action in the conduct of foreign policy. I have tried to implement this general objective by focusing on U.S. policy toward Iraq to show more concretely the relevance of con- ceptual and generic knowledge of strategies for the task of the policymaker. Accordingly, in each of the preceding chapters I asked whether adequate theoretical conceptualization and generic knowledge in fact existed for each of these strategies on which policymakers might draw. My finding is that the state of systematic knowledge was altogether lacking or deficient for four of the strategies—resocialization of outlaw states and 101 Concluding Note to Part Two 103 together by the United States and its coalition partners. It failed for reasons that have largely to do with the way in which Saddam Hussein perceived the crisis and its stakes and calculated and miscalculated the outcome. The failure of coercive diplomacy in this case adds to and reinforces lessons derived from previous efforts to employ this strategy, some of which have succeeded and others of which have failed. The three types of policy-relevant knowledge that were outlined in the Introduction have been alluded to in the fore- going analysis of the administration's strategies toward Iraq. These are (1) a general, abstract conceptualization of the strategy, (2) generalizations (or generic knowledge) about the conditions that favor the success of the strategy and the ob- stacles that it may encounter, and (3) actor-specific knowl- edge of the behavioral patterns of the adversary. These types of knowledge are discussed in much greater detail in chapter 10, along with the contributions they are ca- pable of making to policymaking. But first I believe it will be useful to discuss in chapter 9 the state of contemporary inter- national relations theory in order to call attention to the lim- ited contribution it makes to foreign policymaking. The larger purpose of parts two and three is to emphasize that much work remains to be done to develop policy- relevant theory and knowledge about the many different strategies and instruments of diplomacy employed in the conduct of foreign policy. Adequate scholarly knowledge does not exist for many of these instruments of policy. And even when it does exist for particular strategies, more system- atic understanding is needed of the problems of implement- ing them in different kinds of situations. 9. Contemporary International Relations Theory --- In this chapter, I will discuss the state of international rela- tions theory and comment on its uses and limitations from the standpoint of the knowledge requirements for conduct of foreign policy. Chapter 10 deals with other approaches that offer greater policy relevance, and the Summary and Conclusions contains general implications for scholarly re- search that aims at producing policy-relevant theories. What, then, is the state of contemporary international re- lations theory and what does it contribute to the knowledge base for conducting foreign policy? Expectations in this re- spect should not be excessive, since, as is well known, theory necessarily abstracts and simplifies reality to some extent. Moreover, as we have been reminded by the difficulties schol- ars have encountered in trying to explain the failure to pre- dict the end of the Cold War, theory must often struggle to catch up with changing realities; in the field of international relations, theory generally does better explaining what has happened than predicting what will happen.' For various reasons, therefore, we should not expect too much from gen- eral theories of international relations, for they do not at- tempt, and by no means claim, to provide all the knowledge needed for the conduct of foreign policy. 107 108 The Bridge between Knowledge and Action The study of international relations lacks a single compre- hensive, coherent theory. Rather, theorizing is fragmented, much of it is not very well developed, and (no surprise) the field continues to be marked by often intense controversy. Much of the “Third Debate” currently being waged by inter- national relations theorists is focused on epistemological and normative issues. Rarely do participants in this debate ask what difference the positions they advocate makes for the de- velopment of policy-relevant knowledge. Indeed, many scholars in the field show little interest in bridging the gap between theory and practice. At the same time, however, other research on international relations theory is evolving in important directions as scholars pay increasing attention to developing knowledge that has policy relevance, broaden the agenda of problems to which theory should be applicable, and undertake more sophisticated and systematic study of the importance of cognitive, organizational, and political var- iables. 3 The dominant academic theory of international politics, variously referred to as structural realism or sometimes neo- realism-deals only with basic structural features of the in- ternational system: the “anarchical” nature of the system, the relative distribution of power, and the importance of the bal- ance of power. This theory draws on the tradition of political realism, which has ancient roots in the writing of Thucydides, and attempts to transform classical realism into a scientific- deductive theory that focuses on the structure of the in- ternational system. The most authoritative and influential statement of neorealist theory is Kenneth Waltz's Theory of In- ternational Politics. It should be noted that Waltz's theory is a leaner version of the classical realist theory that was formu- lated many years ago by Hans Morgenthau. It avoids ques- tionable assumptions, ambiguities, and contradictions that Waltz and other scholars discerned in Morgenthau's writings with respect to the central concepts of power, national inter- est, and balance of power.4 At the same time, Waltz's book has Contemporary International Relations Theory 109 stimulated a considerable debate regarding the uses and lim- itations of structural realist theory and numerous suggestions for broadening the scope of international relations theory.5 Classical realist theory was more comprehensive in scope and more ambitious in its aims than Waltz's leaner neorealist theory. Morgenthau, for example, claimed not only that his version of political realism could explain and predict the play of international politics but also that it could and ought to guide statesmen engaged in the conduct of foreign policy. In other words, Morgenthau's theory was also a theory of state- craft. In contrast, narrowing the scope of Morgenthau's theory, Waltz attempted to convert it into a scientifically re- spectable deductive theory, one that relies on structural attri- butes of the international system to generate predictions and provide a basis for explaining outcomes of international pol- itics. We need to consider how well it achieves these goals and how much it contributes to the knowledge base that is neces- sary for conducting and understanding foreign policy. Although cast in the form of a deductive theory, structural realism is not a full-fledged deductive theory, because its key variables and hypotheses have not been "operationalized” so that outcomes can be predicted in specific cases. In conse- quence, structural realist theory can make only general, probabilistic predictions. But since such predictions lack grounding in systematic empirical analysis of the observed relationship between relative capabilities and outcomes in a large and presumably representative sample of interactions between states, the theory cannot express probability in statis- tical terms and is little more than a statement of likelihood. Nor does structural realism do much by way of identifying the conditions under which it expects its predictions to mate- rialize. (See the discussion in chapter 10 about probabilistic predictions versus conditional predictions.) As the theory's foremost proponents acknowledge, the few structural variables encompassed by the theory operate not as determinants of statesmen's choices of policy, but merely as 110 The Bridge between Knowledge and Action constraints, though certainly important constraints, on those choices. In effect the theory warns that the relative power of states in the international system should be properly at- tended to by statesmen in conducting foreign policy, because their failure to respect the power advantage of an adversary will likely lead to their state being punished—either by being eliminated as an actor in the system or by incurring costly setbacks in interactions with other states. This general warn- ing is indeed highly relevant and useful, but policymakers are still left to decide as best they can how, and by how much, their calculation of utility and risk should be influenced by power inferiority when they consider whether and how to pursue their interests in specific situations. And, as will be noted later, leaders of weak states—not deterred by the gen- eral warning conveyed by structural realism—often find strategies for advancing their interests in interactions with stronger states with some success. Particularly relevant to this discussion of the relationship of knowledge to action, therefore, is the fact that structural realist (neorealist) theory is not a theory of foreign policy. This point was explicitly acknowledged by Waltz, who warned against expecting his theory to “explain the particular poli- cies of states” and regarded it as an error “to mistake a theory of international politics for a theory of foreign policy.” Waltz also acknowledged that structural realist theory “makes as- sumptions about the interests and motives of states, rather than explaining them.” That he regards structural realism as a theory of constraints on foreign policy rather than a theory of foreign policy was made clear in the additional observation that “what it [structural realist theory) does explain are the constraints that confine all states."7 Although structural realist theory is not a theory of foreign policy, it is nonetheless useful for present purposes to iden- tify its limitations from the standpoint of the knowledge base that is needed for the conduct of foreign policy. “Power” re- mains an elusive concept in the theory. The emphasis on the Contemporary International Relations Theory III importance of the differences in power among states fails to take into account that not all the capabilities a state possesses come into play and influence the outcome of its interactions with other states. The theory fails to distinguish—and cannot distinguish between what might be called the totality of gross capabilities a state possesses and the often much more limited usable options that its leaders can employ or wish to employ in particular situations. And yet the distinction be- tween gross capabilities and usable options is often critical for understanding why powerful states do less well in military conflicts and trading disputes with weaker states than with stronger states. Much of the power resources a strong state possesses may simply not be relevant or usable in some dis- putes with weaker states (or, indeed, with other powerful states). Weaker states often do surprisingly well in disputes with stronger states insofar as they enjoy asymmetry of moti- vation, à variable that can compensate for inferiority in over- all power capabilities. What is at stake in a particular dispute may be more important to the weaker state than to its stronger rival; that is, the “balance of interests” may work to the advantage of the weaker state and enable it to obtain a better outcome than could be predicted by the relative distri- bution of raw power capabilities. Thus, if it is true (as structural realist theory warns) that weaker states can be punished if they ignore power inferior- ity with strong states, it is also true (but structural realist theory fails to add) that strong states can be punished if they ignore the asymmetry of motivation that weaker states some- times enjoy in disputes with stronger states. Nor should one overlook that strong states often have multiple, global foreign policy interests that compete with and place in perspective the particular interests involved in disputes with weaker states. The more powerful state may not attach enough im- portance to these particular interests to warrant a heavy ex- penditure of resources to achieve a maximum payoff in the dispute with a weaker but highly motivated adversary. 114 The Bridge between Knowledge and Action limitations of theories of international relations, because this one is currently predominant in academic circles. How, then, can progress be made in developing the addi- tional knowledge needed for the study and conduct of for- eign policy? One approach that scholars have taken is to de- velop empirically derived generalizations about the ways in which each of these additional variables—domestic politics, bureaucratic politics, ideology and so on-influences the process of policymaking, the content of decisions, the imple- mentation of policies, and their outcomes. Knowledge of this kind can indeed sensitize policy specialists, if they have not already been alerted through experience, to the ways in which foreign policy can be influenced for better or worse by each of these variables. However, theories and generaliza- tions that focus narrowly on but one of these individual causal variables are of limited utility, among other reasons be- cause they are not linked with variation in situational contexts and with strategies and instruments of policy that practition- ers employ in attempting to influence outcomes in their inter- actions with other states. The discussion now turns to theo- ries and knowledge about strategies, but first it should be acknowledged, as was noted in the Introduction, that policy- makers need many other types of knowledge as well, for ex- ample, scholarly studies of global trends and problems. 116 The Bridge between Knowledge and Action One is that the policymaking process is often driven by in- ternal and international political forces that squeeze out or limit the impact of relevant substantive knowledge. Indeed, for this reason some of those interviewed for this study expressed a sober view of the prospects for improving pol- icymaking by developing and making available better knowl- edge of various types of foreign policy strategies and actions. However, one cannot demand assurance that relevant schol- arly knowledge will be used in policymaking as a precondi- tion for attempting to improve the knowledge base. A second reason is that general knowledge of international re- lations produced by scholars can only be an input to, not a substitute for, policy analysis of a specific problem conducted within the govern- ment. Policy analysts, not academic scholars, have the difficult task of adapting the available general knowledge about a given strategy or a foreign policy undertaking to the particu- lar case at hand. To do so, policy analysts must draw on avail- able information about the particular case; such information comes from various intelligence and journalistic sources, not from bodies of theory or general knowledge of foreign af- fairs. Third, generic knowledge of a strategy or of a standard foreign policy undertaking is not a substitute for, but only an aid to, the final judgment of top policymakers. The same generally can be said regarding the results of policy analysis; it, too, is an aid, not a substitute for the judgment of the top policymakers.' This is so because top policymakers typically must take into account factors such as domestic political considerations, interna- tional opinion, value priorities, and short-term versus long- term payoffs, which either were not considered by policy analysts or were not given the same weight that top policy- makers assign to them. It will be useful at this point to recall some of the observa- tions made in chapter 2 about the policymaker's exercise of judgment. Whereas scholars and policy analysts can and should concern themselves with identifying a high-quality policy option, top policymakers have to deal with the difficult Types of Knowledge 117 trade-off between doing what they can to enhance the quality of a policy and the need for obtaining sufficient consensus and support for the policy option they eventually choose. Also, top policymakers have to decide how much time and how much of the limited pool of resources to allocate to each of these efforts. In other words, a distinction can be made between effective and rational decision making. Decision making is effective when the policymaker deals reasonably well with trade-offs between quality, support, and time and other resources. Ra- tional decision making, on the other hand, reflects the schol- ar's and the policy analyst's effort to come up with a high- quality policy decision without reference to these trade-offs and the various political considerations with which the pol- icymaker must deal. Although scholars have provided a num- ber of models of rational decision making, there exists no theory of effective decision making that continues to rely upon ad hoc judgments by top policymakers. Three types of knowledge can help policymakers decide whether and how to employ a particular strategy. Scholarly studies by academics, research and intelligence specialists within the government, and other analysts are a major way of accumulating these types of knowledge: (1) An abstract con- ceptual model (or a quasi-deductive theory) of each strategy. (2) General (or “generic”) knowledge of the conditions that favor the success of a strategy and, conversely, the conditions that make its success unlikely. (3) An actor-specific behavioral model of the adversary. Abstract Conceptual Models of Strategies An abstract conceptual model of a strategy-such as deter- rence, coercive diplomacy, crisis management, war termina- tion, détente, appeasement, dispute resolution, or coopera- 118 The Bridge between Knowledge and Action tion-identifies the critical variables of that strategy and the general logic that is associated with successful use of that in- strument of policy. Let us take the relatively simple concept of deterrence to illustrate these points. The threat to respond to actions that an adversary may be thinking of taking against a state's interests is a critical variable of deterrence theory. A deterrence policy may specify the threatened actions or refer to them in a deliberately ambiguous manner. Similarly, the threatened response to such actions, should they occur, may be specified or left open. However, the logic of deterrence requires that the threat to respond to an action against a state's interests should be sufficiently credible and potent to persuade the adversary that the costs and risks of the action he is or may be contemplating outweigh the expected gains. The logic of this abstract deterrence model, therefore, rests on the general assumption that the opponent is rational and able to calculate his benefits, costs, and risks correctly. Abstract conceptual models exist or can be formulated for other strategies as well. Two limitations of the usefulness of such models for policymaking should be noted. First, the ab- stract model is not itself a strategy but merely the starting point for constructing a strategy. The usefulness of an ab- stract model for policymaking is limited to providing the ba- sic framework for understanding the general requirements for designing and implementing a strategy. The abstract model identifies only the general logic—that is, the desired impact on the adversary's calculations and behavior—that is needed for the strategy to be successful. But the abstract model does not indicate precisely what the policymaker has to do to induce that logic into the adversary's calculations. To achieve that result the policymaker has to tailor the abstract model into a specific strategy for the particular situation at hand and for the behavioral characteristics of that particular adversary. One example should indicate what needs to be done to move from an abstract conceptual model to a specific strat- Types of Knowledge 119 egy. This example concerns coercive diplomacy, a strategy that relies on threats to induce an adversary to stop or undo a hostile action he is engaged in. To convert the abstract model of coercive diplomacy into a specific strategy the pol- icymaker has to specify each of the four variable-components of the general model: 1. What demand to make on the opponent. 2. Whether and how to create a sense of urgency for com- pliance with the demand. 3. How to create and convey a threat of punishment for noncompliance that will be sufficiently credible and potent to persuade the adversary that compliance is in his best interest. 4. Whether to couple the threat with a positive induce- ment (i.e., a “carrot”) to make compliance easier for the ad- versary and, if so, what kind and how much of an inducement to convey These variable-components of the abstract model of coer- cive diplomacy may be likened to blank lines that the policy- maker must fill in when designing a specific strategy of coer- cive diplomacy. (For further discussion of coercive diplomacy see chapter 7.) The second limiting characteristic of abstract conceptual models for policymaking is that they are typically not full- fledged deductive theories. A fully developed deductive theory of deterrence or coercive diplomacy, if properly con- structed, could be used to predict whether the strategy would succeed or fail in a particular situation. That contribution would be invaluable to policymakers, since it would remove uncertainty and guesswork. However, to have this capability the abstract model (which is, in effect, an incomplete or quasi- deductive theory) would have to be “operationalized”—that is, all its variable-components as well as the interaction among them would have to be capable of being specified and mea- sured. Fully developed deductive theories of foreign policy strategies do not exist and will be difficult to develop. And it Types of Knowledge 121 other forms that knowledge of the efficacy of a strategy can take is needed at this point. Ideally, one would like to discover universal, deterministic generalizations (or laws) that would tell us that whenever a particular condition or set of condi- tions is present the strategy will succeed. Such universal gen- eralizations, taking the form “If A, then B,” posit that the presence of condition A is sufficient for the occurrence of B. However, given the complexity of international relations phe- nomena, it is most unlikely that valid universal generaliza- tions will be discovered either through empirical research or by developing full-fledged deductive theories. We are somewhat more likely to discover a different type of relationship, one in which A is a necessary, but not a suffi- cient, condition for the occurrence of B. But attempted gen- eralizations of this kind are fragile, since logically the discov- ery of only one case in which B occurred in the absence of A falsifies the proposition that A is a necessary condition for B's occurrence. Moreover, it is quite possible that a necessary condition actually contributes little to the causal explanation of the outcome. That is, the missing additional conditions may be much more important than the necessary one for the occurrence of B. In any case, available research indicates that the outcomes of strategies are sensitive to many conditions and that no single causal pattern explains all successes and no single causal pattern explains all failures. Therefore, we need to go beyond the search for a single conditional generalization that can distinguish all successes from all failures—a quest that is not likely to be achieved—to developing conditional general- izations and causal patterns for different types of success and for different types of failures of a given strategy. Given the complexity surrounding the phenomena we are interested in, it is sensible to set aside or at least play down the search for necessary or sufficient conditions. It is prefer- able instead to regard those conditions that do seem to have causal importance in explaining some successes of a strategy 128 The Bridge between Knowledge and Action ent recommendations as to how the United States should re- spond. The essence of the hard-line position was that the So- viets were engaged in an offensive move that posed serious dangers to the entire western position in Europe. For the western powers to agree to negotiate or to present an image of flexibility in the face of Khrushchev's deadline would be to encourage the Soviets to the Soviets to press for the realization of their far- reaching objectives. The hard-line group urged a firm re- sponse coupling refusal of negotiations with deployment of additional forces to Europe and a call-up of reserves. In contrast, and consistent with its more moderate general image of the Soviet opponent, the other group of advisers offered a quite different assessment of the threat and a differ- ent set of recommendations. They believed that the Soviets were engaged in an essentially defensive operation in Berlin, aimed at consolidating their control over Eastern Europe. Accordingly, these soft-line advisers urged the president to start negotiations immediately in order to avoid a dangerous confrontation and also to convey to Moscow a willingness to reduce the irritants that West Berlin was imposing on Soviet interests in Eastern Europe. Fearing that the intransigent po- sition advocated by the hard-line school could lead the Sovi- ets to make desperate moves, advocates of the softer line fa- vored immediate negotiations to assure the Soviets that an effort would be made to meet their legitimate security in- terests. In the Cuban missile crisis the image of the opponent played an important role both in the onset of the crisis and in its resolution. Khrushchev's defective image of Kennedy- whom he viewed as a young, inexperienced, weak leader who could be pushed around and, at the same time, a rational per- son who would not risk war to get the missiles out—is widely believed, even by some analysts in the Soviet Union, to have played a role in his underestimation of the risks of the missile gambit. On the other hand, Kennedy's correct image of the Soviet leader—as one who was capable of correcting his mis- Types of Knowledge 129 take and withdrawing the missiles if made aware of the risks of refusing to do so—played a role in his choice of the strat- egy of coercive persuasion rather than military force to re- move the missiles. After discovering the missiles in Cuba, U.S. policymakers had to choose between (a) using military action to remove the missiles, and (b) applying coercive diploma- tic pressure to persuade Khrushchev to remove them peace- fully. Critical to the policy choice was the question whether Khrushchev and his advisers were capable of a diplomatic re- treat. Or had the deployment of missiles into Cuba pro- ceeded so far, and were Soviet leaders so committed to the daring venture, that for them there could be no turning back? An answer to this question was suggested by knowledge of an important aspect of the Soviet political style derived from studies of past Soviet behavior and Soviet political doc- trine. 9 Readiness to retreat in order to extricate oneself from a situation of overwhelming danger was one of the cardinal maxims of the old Bolshevik doctrine. A good Bolshevik had to know when the time had come for retreat and not let false considerations of pride or emotion get in the way. In numerous other ways during the Cuban crisis, knowl- edge of the opponent provided by the State Department spe- cialist on the Soviet Union, Llewellyn Thompson, was of con- siderable help to Kennedy's advisory group in interpreting Soviet actions and in gauging how Khrushchev and his advis- ers were likely to react to possible U.S. moves. In addition to a correct image of the opponent, policymak- ers need to understand the adversary's image of them. This var- iable is often alluded to in emphasizing the importance of trust and of overcoming invidious, distorted stereotypes that disputing parties have of each other. One's self-image, which exercises a subtle influence in shaping one's foreign policy aspirations, goals, and behavior, is seldom the same image that is perceived by the adversary and that influences his per- ceptions and behavior. Today there is no disagreement among practitioners and Types of Knowledge 131 of the adversary that include information on how decisions are made, who the policy influentials are, and what psycho- logical, cultural, and political variables (which may be quite idiosyncratic) shape and influence that adversary's goals, per- ceptions, calculations, and behavior. Only area specialists can be expected to provide such specialized models. I have discussed the types of knowledge that can be devel- oped about each of the many undertakings and strategies employed from time to time to further foreign policy objec- tives in relations with other states. I have also indicated the contribution to policy analysis and policymaking these types of knowledge are capable of making. Both the development of this knowledge base and its use in policymaking remain quite uneven. These inadequacies were evident in the discussion in part two of strategies the United States employed, mostly without success, in different phases of its relations with Saddam Hussein's Iraq. In employing sev- eral of these strategies (resocialization of an outlaw state, ap- peasement, reassurance, and war termination) policymakers had to operate without a well-developed conceptualization of the task and without generic knowledge of the conditions under which it might succeed. The discussion in part two also called attention to the limitations of the actor-specific model of Saddam Hussein with which U.S. policymakers operated. The observation that the knowledge base for the strategies we have examined suffers from various inadequacies applies also to many other instruments of statecraft and other for- eign policy undertakings. It is true, for example, of knowl- edge about conflict and dispute resolution techniques. There are exceptions, of course. As already noted, progress has been made in conceptualizing an understanding of the con- ditions that facilitate getting parties to a dispute to the table. And the concept of “ripeness” of conflicts for mediation has Types of Knowledge 133 Finally, the results of policy analysis go to top policymakers who exercise a broader judgment that takes into account the various trade-offs noted earlier as well as other variables and considerations that were either absent or weighted differently in the policy analysis. The job of policy analysts is to provide an analytic judgment as to what is likely to be the best policy option. The policymaker, however, has to exercise a broader political judgment as to what option is most appropriate or most acceptable in the circumstances. We should also recog- nize that policymakers need not and often do not wait pas- sively for policy analysts to complete their work and commu- nicate their results. In fact, top policymakers may take the lead in expressing provisional policy preferences and setting boundaries and objectives for policy analysis. Also, there can be considerable discussion back and forth between top pol- icymakers and policy analysts. Top policymakers themselves sometimes take on major tasks of policy analysis. When they do so without access to or Abstract conceptual model of a strategy Specific intelligence and information about the problem situation Generic knowledge of the strategy POLICY ANALYSIS Policy- makers' judgment Actor-specific behavioral model Figure 10.1. Ways in which the three types of knowledge contribute to policy analysis. 134 The Bridge between Knowledge and Action much use of the types of knowledge we have discussed, we may say that top policymakers substitute their own under- standing regarding the strategy in question for the abstract conceptual model of the strategy available to policy special- ists, their own assumptions of what conditions favor its suc- cess for the available generic knowledge, and their own image of the opponent for the actor-specific behavioral model pol- icy analysts have at their disposal. It is also possible, of course, that top policymakers are already conversant with these types of knowledge or are made conversant with them during the policymaking process. The contribution of knowledge to policy analysis and to the judgment of top policymakers is depicted in figure 10.1. 136 Summary and Conclusions each other and focus instead on their shared interest in better understanding the relationship between knowledge and ac- tion. They will find embedded in this relationship three more specific questions that need to be addressed: (1) What kinds of contribution can scholarly knowledge make to policymak- ing? (2) What types of knowledge are most relevant for pol- icy? (3) How can these types of knowledge be developed by scholars and research specialists, and how can such knowl- edge be employed in policymaking? Provisional answers to these questions have been provided in this study, but before summarizing them I will restate the major conclusion of chapter 9. Broadening the Scope of International Relations Theory The dominant and most prestigious scholarly theory of inter- national relations is structural realist theory, sometimes called neorealist theory. In chapter 9, I called attention to the lim- ited scope of this theory from the perspective of the knowl- edge base required for the conduct of foreign policy. Indeed, proponents of structural realism explicitly acknowledge that it is not a theory of foreign policy—although a theory of for- eign policy is precisely what policymakers need. Clearly, scholars who wish to contribute to the develop- ment of theory and knowledge that is relevant for the con- duct of foreign policy will have to go beyond realist theory. In the present study I have advocated that systematic research be directed toward developing better conceptual and empiri- cal knowledge of each of the many strategies and different types of undertakings that form so much of the substance of foreign policy. Also, I have urged development of more so- phisticated actor-specific behavioral models of actors in inter- national politics to replace the simple assumption made by realist theory and other academic approaches that states are rational, unitary actors. By urging a revival of research on foreign policy strategies 1 38 Summary and Conclusions Generic knowledge • Generic knowledge (i.e., empirical laws and causal pat- terns) is derived from systematic comparison of explana- tions of variance in the outcomes of past efforts to make use of a particular strategy. • Generic knowledge is more useful when it takes the form of conditional generalizations rather than probabilistic generalizations. Actor-specific behavioral models • Models of this kind are needed to replace the generalized rationality often attributed by policymakers to other actors whose attitudes and behavior they are trying to influence. • The knowledge and judgment of area specialists is partic- ularly relevant for developing actor-specific models. Adequate scholarly knowledge of the conceptual and ge- neric types does not yet exist for many of the standard strat- egies and instruments of policy. And, although there are ex- ceptions, the quality of knowledge of the motivations and behavioral styles of other actors in international politics is un- even and problematic at best. Central Themes of the Study Further progress in bridging the gap between theory and practice requires that scholars take a realistic view of the lim- ited, indirect, and yet important impact that scholarly knowl- edge about foreign policy can have on policymaking. In ad- dressing this question three central themes were advanced. First, although the three types of policy-relevant knowl- edge identified can indeed help bridge the gap, they cannot eliminate it. Rather, scholarly knowledge is best conceptual- ized as an input to policy analysis of specific issues within the government and as an aid, not a substitute for, the judgments 142 Summary and Conclusions generalizations is a relatively simple research task. For ex- ample, despite the considerable research effort of many scholars over the years on the question “Do arms races lead to war?” a recent review of this literature tells us that “there is still no well-developed theory that describes the circum- stances under which arms races will or will not lead to war. Nor is there a theory that provides a reliable guide for poli- cymakers."3 However, increasingly sophisticated research on this question does show that arms races are neither a neces- sary condition for the occurrence of war (since wars do some- times occur in the absence of a prior arms race) nor a suffi- cient condition (since an arms race is not always followed by war). Additional study should result at least in identifying a number of conditions that can be said to favor the likelihood of war and perhaps in identifying ways policymakers can re- duce or control the likelihood of an arms race resulting in war. Sixth, but by no means least in importance, in attempting to develop conditional generalizations scholars should con- sider whether the phenomenon in question is characterized by equifinality, a term from general systems theory that means similar outcomes in different cases of a phenomenon can have different causal explanations. An example of equifinal- ity was the discovery that deterrence can fail in several differ- ent ways, leading to the identification of three different causal patterns leading to deterrence failure. 4 Another ex- ample of equifinality emerged in the identification of several different paths to “inadvertent war” (a war that occurs even though neither side wanted or expected it at the beginning of the crisis). The phenomenon of equifinality in fact pervades much of international relations and indeed many other areas of life, as John Stuart Mill recognized so many years ago in his System of Logic. Mill warned that the methods of agreement and dif- ference he outlined were not applicable to many social phe- nomena, because their occurrence was subject to “plurality of causes.” Equifinality has important implications not only for Summary and Conclusions 143 the form that causal knowledge of foreign policy outcomes often must take but also for research strategy aimed at devel- oping causal theory. Scholars should not assume, as they often do, that the task of developing theory and causal knowledge consists in finding a single causal generalization for all instances of an undertaking that have resulted in a sim- ilar outcome. Rather, the research task will be better pursued and be more fruitful if the investigator is alert to discovering different causal patterns that lead to a similar outcome. In summing up, we may recall from chapter 1 Richard Good- win's criticism of Schelling's Arms and Influence. Goodwin ar- gued that a “systematic theory” of the kind he inferred (in- correctly) that Schelling was proposing is impossible as well as dangerous. The criticism rests on a misunderstanding of systematic theory as something that seeks to provide, or seems to provide, the policymaker with detailed, high- confidence prescriptions for action in each contingency that may arise. Indeed, such a theory does not exist and is not feasible. But the choice is not between (a) detailed, precise, high-confidence prescriptions for action and (b) nothing. In- stead of focusing primarily on the prescriptive function of theory I have emphasized as more feasible and useful the contribution that the three types of policy-relevant knowl- edge identified in this study can make to the diagnosis of spe- cific situations that have to be made before policymakers de- cide what to do. I also noted the limited but still useful contribution these types of knowledge can make to policy analysis and to decisions on whether to undertake a particu- lar policy and how to design and implement it. -- Implications for the Policymaking Community This study lends strong support to the observation that top policymakers often operate with inadequate conceptual and 144 Summary and Conclusions generic knowledge of strategies they employ in conducting foreign policy and that such policies are often based on an inaccurate image of the adversary. Not only do these gaps and inadequacies in the knowledge base for foreign policy need to be recognized, but remedial steps need to be taken. This study also questions the view that policy failures are invariably to be explained by inadequate or faulty intelli- gence provided to top policymakers and that the remedy lies in improving the quality of intelligence. Such an explanation oversimplifies both the problem and the solution to it. Ad- mittedly, the weak intelligence available on Iraq during the period 1988–91 contributed to the failure of the strategies employed (see chapter 3). But the more fundamental defi- ciency was the poor conceptualization of several of the strat- egies and inadequate knowledge of the requirements for making effective use of them. Only a few brief observations and suggestions will be of- fered here regarding remedial steps for improving the knowledge base; to go further would require extending the focus of the present study considerably. First, the intelligence community cannot be expected to develop the conceptual and generic knowledge of strategies that policymakers need. The research needed for this purpose requires a particular kind of analytical perspective on foreign affairs and research methods that is more typically possessed by academic scholars than by intelligence specialists. A partial exception is with re- spect to what I have referred to as actor-specific behavioral models. Important contributions to developing such models can be made by intelligence specialists with area expertise and other qualifications, and by area specialists elsewhere in the government. How well they have done so in the past requires a broader study than the present one. The challenge here is to find ways of making better use of the resources and spe- cialists within the government and drawing more effectively on specialists outside the government. This objective is surely worthy of serious attention within the government. Notes Introduction 1. For a more detailed statement describing my research pro- gram see A. L. George, “Bridging the Gap between Theory and Practice,” in James Rosenau (ed.), In Search of Global Patterns (New York: Free Press, 1976), pp. 114-119. Discussions of policy- relevant theory appear in A. L. George, D. K. Hall, and W. E. Si- mons, Limits of Coercive Diplomacy (Boston: Little, Brown, 1971), pp. ix-xviii; A. L. George and R. Smoke, Deterrence in American Foreign Policy; Theory and Practice (New York: Columbia University Press, 1974), pp. 616-642; R. Smoke and A. L. George, “Theory for Pol- icy in International Relations,” Policy Sciences (December 1974). My work on “process theory” is summarized in A. L. George, Presiden- tial Decisionmaking in Foreign Policy: The Effective Use of Information and Advice (Boulder, CO: Westview Press, 1980). Chapter 6 of that book identifies and illustrates various “malfunctions” of the policy- making process. The problem of evaluating policy decisions is dis- cussed in “Criteria for Evaluation of Decisionmaking,” Global Per- spectives, Vol. 2, No. 1 (Spring 1984): 58–69. 1. Academia and Policymaking 1. Robert Bowie, as quoted in Ernest R. May (ed.), Knowing One's Enemies: Intelligence Assessment Before the Two World Wars (Princeton: Princeton University Press, 1984), pp. 3-4. 2. I use the term practitioners broadly to include specialists who make different kinds of contributions to the policymaking process. 147 148 Notes to Pages 4-10 In addition to top policymakers, they include advisers, policy ana- lysts, intelligence specialists, and functional and area experts. There are important differences in perspective toward scholarly research and its relevance to policymaking between political-level officials (elected or appointed) and career professionals who serve at lower levels in the government. The former are more action- oriented, the latter more likely to give attention to the analysis of policy issues and to be more receptive to the work and views of academic scholars. 3. Carol H. Weiss with Michael J. Bucuvalas, Social Science Re- search and Decision-making (New York: Columbia University Press, 1980), p. 2. 4. For an analysis along similar lines see Robert L. Rothstein, Planning, Prediction, and Policymaking in Foreign Affairs: Theory and Practice (Boston: Little, Brown, 1972). Observations similar to some of those in this chapter regarding the “two cultures” are also pre- sented in Daniel Druckman and P. Terrence Hopmann, “Behav- ioral Aspects of Negotiations on Mutual Security,” in P. E. Tetlock et al. (eds.), Behavior, Society, and Nuclear War, Vol. 1 (New York: Ox- ford University Press, 1989), pp. 85–173; Nathan Caplan, Andrea Morrison, and Russell J. Stambaugh, The Use of Social Science Knowl- edge in Policy Decisions at the National Level (Ann Arbor: University of Michigan, 1975); and Weiss with Bucuvalas, Social Science Research and Decision-making (see n. 3). See also Harold Guetzkow's report on twenty interviews he conducted with policy specialists in various parts of the government during the summer of 1980 to identify their perceived need for knowledge about alliance behavior. (“Sur- vey of Policy Community” in Michael Don Ward, Research Gaps in Alliance Dynamics, Denver: Graduate School of International Stud- ies, University of Denver Monograph Series in World Affairs, 1982, Vol. 19, Book 1, pp. 71–83.) 5. George Ball, “Lawyers and Diplomats,” address before the New York Lawyers' Association, New York City, December 13, 1962; Department of State Bulletin, December 31, 1962, pp. 987– 991. 6. Jerrold M. Post and Raphael Ezekial, “Worlds in Collision: The Uneasy Relationship between the Counter-terroristic Policy Community and the Academic Community, Terrorism, Vol. 11. 7. Schelling's book was published by Yale University Press in 1966. Goodwin's review appeared in the New Yorker, February 17, 1968. Notes to Pages 36-47 151 Sciolino, The Outlaw State (New York: John Wiley, 1991); Paul A. Gigot, “A Great American Screw-Up; The U.S. and Iraq, 1980– 1990,” The National Interest (Winter 1990–91): 3-10; U.S. News and World Report, Triumph without Victory: The Unreported History of the Persian Gulf War (New York: Times Books, Random House, 1992); and the three-part series of articles by Douglass Frantz and Murray Waas in the Los Angeles Times, beginning February 23, 1992. See also Richard N. Haass, “One Year after the Gulf War: Prospects for Peace,” U.S. Department of State, Dispatch, Vol. 3, No. 15. 4. Time Magazine, October 1, 1990; Oberdorfer, “Missed Sig- nals”; Washington Post, July 16, 1991, and May 29, 1992. 5. Interview with policy official (April 9, 1992), who stated that, in retrospect, implementation of specific licensing of exports to Iraq had clearly been too lax; oversight and coherence and control over individual export decisions were lacking. On the Deputies Committee policy review see also U.S. News and World Report, Triumph without Victory, pp. 18-19. 6. As quoted by Oberdorfer, “Missed Signals,” p. 36; see also U.S. News and World Report, Triumph without Victory, p. 19. 7. Testimony by Central Intelligence Agency (CIA) director Robert M. Gates before the House Banking Committee on May 8, 1992, as reported in the Washington Post, May 9, 1992. Gates added that by late spring 1990, the CIA had begun to provide informa- tion of Iraq's “huge military buildup," but he did not say whether the intelligence estimate of Iraq's intentions was changed before Iraq's invasion of Kuwait in early August. 4. Reforming Outlaw States 1. Oberdorfer, “Missed Signals”; see also the publications by Paul Gigot, Elaine Sciolino, U.S. News and World Report, and Frantz and Waas (see chap. 3, n. 3). This account of U.S. policy toward Iraq is supported also by interviews of my own with several policy specialists. 2. Elaine Sciolino, New York Times, May 29, 1992; see also Jeffrey Smith, Washington Post, May 29, 1992. 3. The foregoing points are from journalistic accounts and anal- yses that included interviews with policymakers and from a few in- terviews of my own with policy specialists in the State Department and the staff of the NSC. 4. To regard Saddam's response to the VOA broadcast as an 154 Notes to Pages 62-71 versity Press, 1971), p. 110; J. L. Richardson, “New Perspectives on Appeasement: Some Implications for International Relations," World Politics, Vol. 40, No. 3 (April 1988): 312. 3. Professor Stephen R. Rock, Vassar College, is doing this study. Drawing on research in progress, Rock presented a paper, “When Appeasement Worked: British Conciliation of the United States, 1895–1905," at the 1988 meeting of the Northeastern Political Sci- ence Association. This paper contains a number of general hypoth- eses, which Rock will assess in studying other cases of successful and abortive appeasement. 4. The great European powers made effective use of appease- ment in many instances. See, for example, Paul Kennedy, “The Tradition of Appeasement in British Foreign Policy,” British Journal of International Studies, Vol. 2 (October 1976). 5. For a more detailed discussion see G. A. Craig and A. L. George, Force and Statecraft, 2d ed. (New York: Oxford University Press, 1990), pp. 134-144, 247–251, 255-257. 6. Rock, “When Appeasement Worked,” pp. 19–20 (see n. 3). 7. Barbara Tuchman, The March of Folly (New York: Knopf, 1984), chap. 4; cited in Rock, “When Appeasement Worked,” p. 17. A related, fifth scenario is suggested by Stephen Stedman (in a per- sonal communication): a situation in which appeasement might have succeeded at an early stage in the dispute but became more difficult because reliance on deterrence or force hardened the ad- versary's position. 8. Craig and George, Force and Statecraft, p. 254; see also pp. 93-99. 9. Craig and George, Force and Statecraft, pp. 257–260. 10. See, for example, Rock, “When Appeasement Worked”; Kennedy, “The Tradition of Appeasement in British Foreign Pol- icy”; and Charalambos Papasatiriou, Grand Strategy of the Byzantine Empire, Ph.D. dissertation, Department of Political Science, Stan- ford University, 1991. 6. Deterrence and Reassurance 1. Useful accounts and interpretations appear in a number of published sources, including various contemporary journalistic re- ports not cited here; see, for example, Oberdorfer, “Missed Sig- nals,” and Sciolino, The Outlaw State (see chap. 3, n. 3). A particu- larly well-documented and incisive analysis is provided by Janice Notes to Pages 73–77 155 Gross Stein, “Deterrence and Compellance in the Gulf, 1990– 1991: A Failed or Impossible Task?” International Security, Vol. 17, No. 2 (Fall 1992): 147–179. 2. Another strategy for testing and challenging deterrence is available when the challenger (a) believes that the defender's deter- rence commitment is soft and may be eroded under carefully con- trolled pressure, or (b) believes he has a way of bypassing the de- fender's commitment by trying to change the situation in ways the defender cannot respond to effectively. This is the “controlled pres- sure” strategy. On the other hand, when the challenger is not sure whether a deterrence commitment exists he can “test” to find out by a “limited probe” strategy that can be terminated if the defender responds by making it clear he does have a commitment and will fight. For detailed discussion and historical examples of these two strategies see George and Smoke, Deterrence in American Foreign Pol- icy, chap. 18 (see Introduction, n. 1). 3. Quoted in Elaine Sciolino with Michael Gordon, Washington Post, September 23, 1990. 4. For an account of the Truman administration's erroneous as- sessment of Chinese intentions and its gross mishandling of the Chinese threat to intervene in the Korean War see, for example, George and Smoke, Deterrence in American Foreign Policy, pp. 184- 234 (see Introduction, n. 1); see also Allen Whiting, “The U.S.- China War in Korea,” in A. L. George (ed.), Avoiding War: Problems of Crisis Management (Boulder, CO: Westview Press, 1991), pp. 103– 125. Whiting's chapter updates with new material on U.S. and Chi- nese decision making his earlier classic study, China Crosses the Yalu (Stanford: Stanford University Press, 1968). 5. These interpretations of Mao's thinking, advanced many years ago by scholars (see Whiting, China Crosses the Yalu), are sup- ported by the recent publication of a secret cable from Mao to Sta- lin on October 2, 1950 (New York Times, February 26, 1992). 6. The best discussion of various reassurance strategies that may be adopted instead of or in conjunction with deterrence is Janice Gross Stein's “Deterrence and Reassurance,” in Philip E. Tetlock et al., Behavior, Society, and Nuclear War (New York: Oxford University Press, 1991), Vol. 2, pp. 8–72. See also the earlier article by Janice Gross Stein, “Reassurance in International Conflict Management,” Politcial Science Quarterly, Vol. 106, No. 3 (Fall 1991): 431–451, and the one by Richard Ned Lebow and Janice Gross Stein, “Beyond Deterrence,” Journal of Social Issues, Vol. 43, No. 4 (1987): 5–71. 156 Notes to Pages 78-82 7. In this connection Stein argues in a recent paper (“Deterrence and Compellance"; see n. 1) that Saddam Hussein was convinced that the United States was masterminding a plot against his regime and, therefore, that it would have been very difficult, if not impos- sible, for the United States to devise a strategy of reassurance that could have altered his decision to attack Kuwait. Stein also calls at- tention to the weakness of the reassurances that Washington con- veyed, which did not address the economic difficulties and needs, driven in part by his expensive military programs, that Saddam be- gan voicing with increasing urgency in the spring of 1990. Nor does it appear that Washington urged the Kuwaiti government to be more forthcoming in dealing with Saddam's demands. Accord- ing to another interpretation of Saddam Hussein's decision to in- vade Kuwait, he did so largely as a result of acute economic diffi- culties and the prospect that without a massive increment of new money he would be unable to afford the level of spending needed to maintain the momentum of his massive weapons program. A similar interpretation of Saddam's image of a fundamentally hos- tile United States is given in Paul K. Davis and John Arquilla, Deter- ring or Coercing Opponents in Crisis: Lessons from the War with Saddam Hussein (Santa Monica, CA: The RAND Corporation, 1991). 7. Failure to Coerce Saddam 1. The most satisfactory analytical study of why the administra- tion moved from sanctions to the threat of war and why the ulti- matum failed is Richard K. Herrmann, “Coercive Diplomacy and the Crisis over Kuwait: 1990–1992,” in A. L. George (ed.), The Lim- its of Coercive Diplomacy, 2d ed. (forthcoming). Important insights and interpretations are to be found in Shibley Telhami, “Between Theory and Fact: Explaining American Behavior in the Gulf War," unpublished manuscript, 1992. 2. The information presented here is provisional and somewhat speculative, yet informed by an interview with a senior policy spe- cialist who participated in the working group's activities. 3. According to Bob Woodward, the CIA issued a special pre- Christmas national intelligence estimate that concluded that Sad- dam would pull out once he realized the size of the force arrayed against him. On the other hand, a Defense Intelligence Agency specialist, Pat Lang, suspected the opposite. In mid-December Gen. Colin Powell thought that Saddam would withdraw from Kuwait at the last minute, but Secretary (of Defense) Dick Cheney disagreed. 158 Notes to Pages 92-108 4. U.S. News and World Report, Triumph without Victory, pp. 3-6 (see chap. 3, n. 3). 5. A detailed account of the decision to declare a cease-fire be- fore the envelopment of the Iraqi troops in the Basra area was com- pleted is given in U.S. News and World Report, Triumph without Vic- tory, pp. 308–398 and 399–403. Important details are provided in General H. Norman Schwarzkopf's It Doesn't Take a Hero (New York: Linda Grey/Bantam Books, 1992), pp. 468–472. 6. According to Bob Woodward, The Commanders (see chap. 7, n. 3), General Powell conveyed to General Schwarzkopf the need for restraint in inflicting damage on Iraq: “Powell was convinced that it would not be in the U.S. interest to have a totally defeated Iraq with no capability to defend itself” (p. 374). See also Newsweek, April 22, 1991: “Indeed, the United States intentionally refrained from destroying the entire Iraq Army ... so that enough soldiers would be left to topple the dictator and hold the country together. Under new and stable leadership, the theory went, Iraq would not disintegrate into another Lebanon.” 7. Newsweek, April 22, 1991. 8. General Schwarzkopf also minimizes the importance of the Iraqi helicopter gunships: “The tanks and artillery of the twenty- four Iraqi divisions that never entered the Kuwaiti war zone were having a far more devastating effect on the insurgents [than the Iraqi helicopters) (It Doesn't Take a Hero, p. 489). 9. General Schwarzkopf, on the other hand, states in his book that “the option of going all the way to Baghdad was never consid- ered” (It Doesn't Take a Hero, p. 497). What he apparently means is that choosing this option was not seriously considered, and he sum- marizes the various reasons for not attempting to seize Baghdad. 10. In this connection, it has been suggested that the administra- tion's reluctance to support the Shi'ites in southern Iraq stemmed in part from a misplaced concern that they were like the Shi’ites in Iran. 9. Contemporary International Relations Theory 1. For a good discussion of why prediction is so difficult in inter- national relations, see Robert Jervis, “The Future of World Politics: Will It Resemble the Past?” International Security, Vol. 16, No. 3 (Winter 1991–92): 39–73. 2. For a commentary on the Third Debate now in progress see, Notes to Pages 108-111 159 for example, Yale H. Ferguson and Richard W. Mansback, “Be- tween Celebration and Despair: Constructive Suggestions for Fu- ture International Theory,” International Studies Quarterly, Vol. 35, No. 4 (December 1991): 363–386. 3. For useful analytical commentaries on the state of research on international relations see Ole R. Holsti, “Models of International Relations and Foreign Policy,” Diplomatic History, Vol. 13, No. 1 (Winter 1989): 15–43; and Stephen M. Walt, “The Renaissance of Security Studies, International Studies Quarterly, Vol. 35 (1991): 211–239. 4. Kenneth Waltz, Theory of International Politics (New York: McGraw-Hill, 1979). Waltz provides an interesting account of his reservations about classical realist theory in his “Realist Thought and Neorealist Theory,” in Robert L. Rothstein (ed.), The Evolution of Theory in International Relations (University of South Carolina Press, 1991), pp. 21–39. 5. For an incisive discussion see Robert O. Keohane, “Realism, Neorealism, and the Study of World Politics,” in Keohane (ed.), Neorealism and Its Critics (New York: Columbia University Press, 1986). There are a number of other critical evaluations of the real- ist paradigm. See, for example, John A. Vasquez, The Power of Power Politics: A Critique (New Brunswick, NJ: Rutgers University Press, 1983). 6. For an example of a full-fledged deductive theory, one that operationalizes key variables, see Bruce Bueno de Mesquita, The War Trap (New Haven: Yale University Press, 1981), which does generate preductions for each of a large number of specific histor- ical cases. This study is unusual and, whatever reservations one may have about the implementation and results, is particularly val- uable for its explicit specification of what is required to operation- alize a deductive theory. 7. Waltz, Theory of International Politics, pp. 121–122. 8. See, for example, Vinod K. Aggarwal, Liberal Protectionism: The International Politics of Organized Textile Trade (Berkeley: Univer- sity of California Press, 1985); and David B. Yoffie, Power and Protec- tionism: Strategies of the Newly Industrialized Countries (New York: Co- lumbia University Press, 1983). See also Robert Keohane, “The Big Influence of Small Allies,” Foreign Affairs, No. 2 (1971): 161–182. My initial formulation of the distinction between gross capabilities and usable options appeared in George et al., Limits of Coercive Di- plomacy, p. 8 (see Introduction, n. 1). 160 Notes to Pages 116-123 10. Types of Knowledge 1. Charles Hitch, one of the founders of modern systems analy- sis, emphasized many years ago that the results of even the best systems analysis study should be regarded as an aid to the prepa- ration of policy decisions and not a substitute for the judgment of the policymaker. 2. Quite unusual in the scholarly literature on international con- flict is the operationalized deductive theory of war initiation deci- sions presented in Bruce Bueno de Mesquita, The War Trap (New Haven: Yale University Press, 1981). The author substitutes for the general assumption of rational decision making a more refined concept of “positive utility,” which he operationalizes. However, the resulting deductive theory does not predict war initiation but, rather, merely that positive utility will be present if war is initiated. Accordingly, although advancing the general claim that he has pro- duced a theory of the causes of war, the author qualifies it substan- tially by conceding that positive utility is a necessary but not a suffi- cient condition for war initiation. Since there are numerous cases in which positive utility was present but the actor did not initiate war, positive utility cannot be regarded as a sufficient condition for war initiation. There are also numerous cases in the study in which war was initiated in the absence of positive utility, which calls into question whether it can be regarded as a necessary condition for war initiation. The question has been raised whether the occur- rence of war, versus the initiation of war by one side, can be ade- quately studied by focusing exclusively on one actor's decision mak- ing. A theory of war seems to require the development of two-sided strategic interaction models. Even such models would have to con- tend with the familiar observation that the outcomes of strategic interactions are indeterminate. 3. For a description of the historical explanation method see A. L. George, “Case Studies and Theory Development: The Method of Structured, Focused Comparison,” in Paul Gordon Lau- ren (ed.), Diplomacy: New Approaches in History, Theory, and Policy (New York: Free Press, 1979), pp. 43–68. 4. George Downs, “Arms Races and War,” in Philip E. Tetlock et al. (eds.), Behavior, Society, and Nuclear War (New York: Oxford Uni- versity Press, 1991), Vol. 2. 5. I. William Zartman, Ripe for Resolution: Conflict and Intervention in Africa (New York: Oxford University Press, 1985). Several years later, applying Zartman's theory of ripeness to a study of several efforts to mediate the civil war in Rhodesia, Stephen Stedman of- Notes to Pages 123-132 161 fered several refinements and amendments to the theory in Peace- making in Civil War: International Mediation in Zimbabwe, 1974–1980 (Boulder, CO: Lynne Rienner, 1991). More recently, Zartman ad- dressed the question whether the concept of “ripeness," as he ini- tially formulated it, is too culturally determined and needs to be reformulated to encompass variations. See I. William Zartman, “Beyond the Hurting Stalemate” (Paper presented at the Thirty- third Annual Convention, International Studies Association, At- lanta, March 31 to April 4, 1992). On “ripeness” see also Richard N. Haass, Conflicts Unending: the United States and Regional Disputes (New Haven: Yale University Press, 1990). 6. Janice Gross Stein (ed.), Getting to the Table: The Process of Inter- national Prenegotiation (Baltimore: Johns Hopkins Press, 1989); and, particularly, an unpublished draft paper by Stein, “The Ne- gotiating Table: Foreground and Background" (prepared for the United States Institute of Peace), which the author has kindly al- lowed me to see. 7. For a detailed discussion see George and Smoke, Deterrence in American Foreign Policy (see Introduction, n. 1); and George, Forceful Persuasion (see chap. 7, n. 7). 8. For example, Cathleen S. Fisher, “The Preconditions of Con- fidence-Building: Lessons from the European Experience” (un- published manuscript, 1992); and Alan Platt (ed.), Arms Control and Confidence Building in the Middle East (Washington, DC: United States Institute of Peace Press, 1992). 9. See, for example, the writings of George Kennan and Nathan C. Leites, A Study of Bolshevism (Glencoe, Ill.: Free Press, 1953). 10. This point is emphasized and some of the components of an actor-specific behavioral model are identified in Richard Herr- mann's current writings. See, for example, “Policy-Relevant Theory and the Challenge of Diagnosis: the End of the Cold War as a Case Study” (unpublished manuscript, 1992). See also James N. Rosenau's call for the development of “single-country” theories in “Toward Single-Country Theories of Foreign Policy: The Case of the USSR,” in C. F. Hermann, C. W. Kegley, Jr., and J. R. Rosenau (eds.), New Directions in the Study of Foreign Policy (Boston: Allen & Unwin, 1987), pp. 53–74. 11. Richard H. Solomon, “Political Culture and Diplomacy in the Twenty-first Century,” in Richard J. Samuels and Myron Weiner (eds.), The Political Culture of Foreign Area and International Studies (Washington, DC: Brassey's [US], 1992), p. 147. 12. Lynn Wagner has produced an extremely useful inventory --- 162 Notes to Pages 140-142 and discussion of the defining characteristics of each of these strat- egies in “Processes for Impasse Resolution,” working paper 91-43 (Laxenburg, Austria: International Institute for Applied Systems Analysis, December 1991). Summary and Conclusions 1. A conditional generalization is “plausible” if it is not contra- dicted by available evidence and if at least some evidence supports it. For a detailed discussion of the limits of scientific verification in social science and the need to settle for degrees of “acceptable” ver- ification, see chapter 10 in Charles E. Lindblom, Inquiry and Change: The Troubled Attempt to Understand and Shape Society (New Haven, CT: Yale University Press, 1990). 2. For a fuller statement and illustration of this point see George et al., Limits of Coercive Diplomacy, p. xvi (see Introduction, n. 1); and George, “Case Studies and Theory Development,” pp. 59–60 (see chap. 10, n. 3). 3. Downs, “Arms Races and War,” p. 75 (see chap. 10, n. 4). 4. George and Smoke, Deterrence in American Foreign Policy (see Introduction, n. 1). 5. George (ed.), Avoiding War (see chap. 6, n. 4). 164 Index Culture academic, 16 interpretation affected by, 42 policymaking, 15 and reactions to ultimata, 85 Classified information, 10 Coercive diplomacy abstract conceptual model of, 119 components of, 79–80, 119 conceptualizations of, 102 conditional generalizations of, 124 Cuban missile crisis vs. Gulf War, 86-88 failure in Gulf War, 79, 102-103 outcome variables, 86, 87, 124 partial compliance with, 81–82 role of, 79-80 threat assessment, 84 ultimatum in, 80–84 Cold War strategies, 53 Conditional generalizations, 13, 43, 120-125 equifinality, 142–143 Conditional reciprocity, 50–53 assessing success of, 54-55 complicating factors, 56-57 risks of, 55-56 structure of, 55-56 in U.S.-Soviet détente Conflict management. See also Appeasement assessing adversary, 65-66, 67–68 research-practice integration in study of, 5 Consensus and policy quality, 23-25 in professional culture of aca- demia, 16 in professional culture of policy- making, 15 Containment of outlaw state, 50 of Soviet Union, 50, 59 Council on Foreign Affairs, 4 Crisis management, substantive theory in, xxi-xxii Cuban missile crisis, 8, 63, 128– 129 coercive diplomacy in, 86–88 Decision-making. See also Individ- ual judgment effective vs. rational, 117 quality of, compromised by re- source availability, 23–24, 117 timing of, 28 Democracy, effecting policy changes in, 34 Détente in progression of diplomacy, 63-65 U.S.-Soviet, 56, 57–59, 63-65 Deterrence abstract conceptual model of, 118 appeasement vs., 65, 67–68 applied against Iraq, 71–77 comparative analysis, 73, 75–76 conceptualizations of, 102 conditional generalizations of, 123-124 limitations of, 73–74 reassurance and, 77–78 role of theory in, xxi-xxii threat assessment, 75-77 Economics in foreign policy research, 5 sanctions against Iraq, 38, 42, 80-82 Egypt, in U.S.-Iraq diplomacy, 72 Entente, 63 Equifinality, 142–143 Falkland Islands War, 68 Fitzwater, Marlin, 95 Foundations, philanthropic, 4 Game theory, xxi, 130 Gates, Robert, 37 Index 165 successes of allied coalition, 91 termination, features of, 89, 90-91 U.S. policy toward Iraq before, 33-37, 45-46 Haass, Richard, 37 Halle, Louis, 14 History, scholar's vs. policymaker's analysis of, 13-14 Hitler, Adolf, 59, 61, 66, 67, 68– 69, 126-127 General systems theory, 142 Generic knowledge. See also Theory of appeasement strategies, 67 of conditions favoring strategy, 120-125 in Gulf War policy, 101–103, 131 role in analyzing strategy, 43 role in policymaking, xvii, xviii- xix, 125, 131-134, 138 Goals Gulf War, 38–39, 91-92, 97–99 political vs. military, 92–93, 99– 100 settling for less, 27 short-term vs. long-term bene- fits, 26-27 Goodwin, Richard, 10-11, 17-18, 143 Gorbachev, Mikhail, 59, 130 Graduated reciprocation in tension reduction (GRIT), 51-53 GRIT. See Graduated reciprocation in tension reduction Gulf War allied coalition goals, 93, 95 cease-fire, 92, 93 coercive diplomacy in, 80–88, 102-103 diplomatic efforts to prevent, 38-39 failure to remove Saddam Hus- sein, 91–92, 97-99 generic knowledge base, 101– 103, 131 goals of, 38–39, 91–92, 97–99 intelligence failures, 40–42, 45 invasion of Kuwait, 37–38, 40, 41, 71, 72–73, 91 Korean War comparison, 95–96 military, vs. political, goals, 92-93 policy options after liberation of Kuwait, 95–99 political objectives, 90–93 rebellion in Iraq after, 90, 92, 93–95, 97 Ideology analysis tainted by, 10 role in foreign policy, 12 Individual judgment defining success, 27 political side-effects, 25–26 risk-benefit analysis, 26 role in policymaking, 8, 21–22, 28-29, 116-117 short-term vs. long-term bene- fits, 26-27 timing, 28 trade-offs, 22-25 types of, 22–28 value complexity in, 27–28 Information management/analysis assessing adversary, 65-66, 67-68 classified information, 10 in coercive diplomacy, 85 contribution to resocialization strategies, 53 cultural issues in, 42 flaws in, affecting analysis, 10, 144 intelligence failures in Gulf War, 40-42, 45 interpretation of Saddam Hus- sein's 1990 speech, 47–48 process theory and, xxii role in developing actor-specific models, 126, 144 Intelligence. See Information man- agement/analysis 166 Index International law, 5 outlaw states and, 49 Iran, 59, 60 Iran-Iraq War, 33–34, 36 Iraq. See also Gulf War; Saddam Hussein Ba'thist party, 94 coercive diplomacy against, 80-88 concessions to, for leaving Ku- wait, 83-84 after Gulf War, 39 Gulf War termination, 89 intelligence appraisals of, 41-42, 45 invasion of Kuwait, 38-39 Iran-Iraq War, 33–34, 36 military spending by, 41 perception of, in U.S., 34–35, 40-42, 45-46 rebellion in, after Gulf War, 90, 92, 93-95, 97 troops on Kuwait border, 37– 38 U.S. aid to, 34-35 U.S. policy after Iran-Iraq War, 33–37, 45-46 U.S. resocialization strategies, 45-48 Israel, 65, 75, 123 abstract conceptual models, 117– 120, 137 conditional generalizations, 120-125 types of, xvii, 114, 117, 131–134, 137-138 Korean War, xxii, 7, 73, 75, 76, 92, 95–96, 127 Kurdistan/Kurdish people, 47, 90, 92, 93-94 Kuwait, invasion of Iraqi preparations, 37–38, 41 United Nations response, 91 U.S. efforts to deter, 72–73 U.S. response to, 38, 40, 71 Learning theory, 55 Libya, 59 Mediation, indications for success, 122–123, 131-132 Mill, John Stuart, 142 Morgenthau, Hans, 11, 108, 109 Nasser, Gamal Abdel, 75 National Security Council assessment of ultimatum in Iraq conflict, 80-81 Directive 26 (October 1989), 46 Neorealist theory, 108-114 Nixon, Richard, détente policy, 57–59, 63-65 North Korea, xxii, 7, 59, 73, 75, 76, 92, 95–96, 127 Judgment. See Individual judg- ment Kahn, Herman, 5 Kaufmann, William, 5 Kelly, John, 37 Kennan, George, 59 Kennedy, John, 86–88, 127-129 Khrushchev, Nikita, 86-88, 127– 129 Kissinger, Henry, 24, 48 détente policy, 56, 57–59, 63-65 Knowledge. See also Generic knowledge Oberdorfer, Don, 35, 46 Outlaw states appeasement strategy applied to, 50 conditional reciprocity with, 50– 51, 54-57 current examples, 59-60 defining, 48–49 prisoners' dilemma, 53–54 strategies for dealing with, 49-50 Index 167 as science, 8, 22 short-term vs. long-term bene- fits, 26-27 statecraft and, xxiv-xxv structural realism in, 109–110, 112, 113 substantive theory of, xxi-xxii, xxiii, 20, 115-117 technical, vs. value, rationality in, xxii-xxiii theory in, lack of, 11-15 theory types, xxi-xxiii, 107–108, 114 timing, 28 traditional views, 6-11 value complexity in, 27–28 Political rationality, 21 Politics constraining military action, 92-93 effects of appeasement strate- gies, 67 Perception. See also Actor-specific models in coercive diplomacy, 80, 87–88 role in foreign policy, 14 Saddam Hussein's susceptibility to ultimata, 82-83 in strategies of deterrence, 75– 77 Pieczenik, Steve, 130 Policymaking abstract conceptual models for analyzing, 117–120, 137 analytical rationality in, 20–21 conditional generalizations in, 13, 43, 120-125 in democratic governments, 34 and generalizing, 10-11, 17–18, 143 integration with academia, 3-6, 16-17, 135-136 knowledge types, xvii, xix-xx, 9, 114, 117, 125, 131-134, 137-138 overintellectualization of, 20 policy analysis and, 133–134 political influence in, 14–15, 20- 21, 23-25 process theory in, xxii-xxiii, 20, 115 professional culture of, 15 realist theory of, 12 resource analysis, 22–24 risk-benefit analysis, 26 role of actor-specific models, 125-126, 138 role of conditional generaliza- tions, 125 role of individual judgment, 20- 21, 116-117 role of scholarship, xix-xx, xxiii, xxiv, 19-20, 21, 116, 130– 131, 132–134, 136-137, 138-145 role of theory, xvii, xviii-xix, xxv, 9, 17–18, 138, 140- 141 in goals of coercive diplomacy, 88 in policy analysis, 23-25 in policymaking, 14–15, 20–21 war and, 90, 96–97, 99-100 Politics Among Nations, by Hans Morgenthau, 11 Post, Jerrold, 84–85 Prisoners' dilemma, 53–54 Probability, prediction via, 109 Psychology, 9-10, 77. See also Actor-specific models adversary's, in coercive diplo- macy, 85-86 cultural, 42 of deterrence-reassurance, 77-78 of Saddam Hussein, 77–78, 83, 84-86 RAND Corporation, xix, 5 Rapprochement, 63 Rational choice theory, 130 Realist theory, 11-12, 108-114 168 Index Realist theory (cont.) neorealist theory vs. classical, 108-109 Reassurance strategies, 43, 77–78, 102 Research. See also Scholars/scholar- ship developing conditional general- izations, 120, 122 empirical vs. deductive, xxi flawed data in, 10 ideological bias in, 10 integration with practice, 3-6, 16-17, 135-136 practitioner views of, 7 quantitative, in foreign policy, 8 Resocialization, strategies for. See also Conditional reciprocity comparative analysis, 57 complications in, 56-57 conceptualizations of, 5, 43, 101-102 Iraq, basis for applying, 45–48 Nazi Germany, 59 Nixon-Kissinger détente strat- egy, 56, 57-59 outlaw states, 49-50 prisoners' dilemma, 53-54 targeted at Saddam Hussein, 34–35, 36, 37, 40 Resources allocating, in policymaking pro- cess, 22-24 relative strengths of conflicting states, 109-112 Revolutionary states, 48 Rewards. See also Conditional reci- procity behavior modification via, 55 bribes vs., 52 timing of, 55-56 Rock, Stephen, 68 Ross, Dennis, 84 anti-U.S. rhetoric, 37, 47–48 failure of ultimatum against, 84 ouster of, efforts at, 39, 91–92, 94-95, 97-99 perception of U.S. policy, 72–73, 77-78 psychological study of, 83, 84– 86, 126 seen as reformable, 34–35, 36, 37, 40, 45, 74 seen as susceptible to threat, 82-83 threats against Kuwait, 71 Saudi Arabia, 38, 72, 73, 82 Schelling, Thomas, 10–11, 17, 143 Scholars/scholarship analytic rationality, 7 integration with policymaking, 3-6, 16-17, 135-136 international relations theory, 107-108 language of, 7, 140 practitioners' traditional views of, 6-11 professional culture of, 15-16 role of, xix-xx, xxiii, xxiv, 9, 19-20, 21, 28–29, 116, 130-131, 132-134, 136- 137, 138-145 and trade-off dilemmas, 24-25 traditional approach to foreign policy, 3-4 traditional views of foreign pol- icy practice, 11-15 understanding practitioner needs, 16 Science foreign policy as, 8, 22 psychological, 9-10 Scowcroft, Brent, 36 Security issues, substantive theory in research on, xxii Shi'ite rebellion, 90, 92, 93–94, 97–98, 99 Sihanouk, Norodom, 130 Solomon, Richard, 130 South Korea, 73 Sadat, Anwar, 30 Saddam Hussein. See also Gulf War; Iraq Index 169 with outlaw states, 49-50 reassurance, 77–78 in statecraft, xxv theory and, xviii-xxii, 20 tit-for-tat, 53–54 trade-offs in decision-making, 24-25 U.S. efforts to reform Iraq, 33– 37, 39-40 U.S. policy toward Iraq, 101- 103 U.S. response to Iraqi threats against Kuwait, 38, 40, 71 Structural realism, 108-114 as deductive theory, 109 limitations of, 112 predictive capability, 112-113 relative power of states, 109– 111 relevance to foreign policy, 110, 112, 113 Syria, 59 Soviet Union, 56 containment policy toward, 59 influencing U.S.-Iraq relations, 35 in Ostpolitik policy, 69–70 perception of, in Berlin crisis, 127-128 perception of, in Cuban missile crisis, 128-129 relations with Iraq, 47 U.S. détente policy, 56, 57–59, 63-65 Stalin, Joseph, 126-127 Stark, USS, 47 Statecraft goals of policy in, xxiv—xxv realist theory and, xxv, 109 as science, 8 strategies in, XXV theory in, xxi Stein, Janice Gross, 123 Stimson Center, 124 Strategies. See also Appeasement; Coercive diplomacy; Resociali- zation, strategies for abstract conceptual models, 117– 120, 137 analytical tools, 117 behavior modification, 50, 52, 55, 58 Cold War, 53 conceptual framework for, xvii conditional generalizations of, 120 conditional reciprocity, 50–58 confidence-building, 124 deterrence, 72-77 development of, in policymak- ing, 131-132 failures in United States' Iraq policy, 39–42, 45, 53, 56-57 flexibility in, 54-55 generic knowledge of, 43, 120– 125, 138 graduated reciprocation in ten- sion reduction, 51–53 Theory. See also Generic knowl- edge abstract conceptual models, 117– 120, 137 deductive vs. empirical, xxi, 109 general systems, 142 international relations, 107–108, 136-137 lack of, in policymaking, 11-14 and policy analysis, 10–11, 17– 18, 143 and practice, xix-xx, xxiii practitioners' views of, 6-11, 140-141 process, xxii-xxiii, 20, 115 realist, 11-12, 108-114 role of, xvii, xviii-xx, 17–18, 116, 140-141 and strategy, 43 structural realism, 108-114 substantive, xxi-xxii, xxiii, 20, 115-117 and trade-off dilemmas, 24-25 ! 170 Index Theory of International Politics, by Kenneth Waltz, 108-109 Thompson, Llewellyn, 129 Thucydides, 108 Tit-for-tat strategy, 53–54 Trade-off judgments, 22–25 Truman, Harry, 75 Tuchman, Barbara, 69 reassurance of Iraq, 77–78 rebellion within Iraq, 93–94, 97–98, 99 Saddam Hussein, efforts to re- move, 91-92, 94-95, 97-99 Saudi Arabia, 38 sudden change in, 34, 74 Values, in policy analysis, 27–28 Vietnam, 57 Vietnam War, xxii, 89, 92, 96 Ultimatum as coercive diplomacy, 80 Cuban missile crisis vs. Gulf War, 87-88 efficacy against Iraq, 80–84 partial compliance with, 81–82 possible responses to, 81 United Arab Emirates, 72 United Nations Gulf War goals, 89, 90, 93 response to Kuwait invasion, 38, 91 in U.S. Gulf War diplomacy, 88 United States policy Berlin crisis, 127–128 coercive diplomacy, 80–88, 86-88 détente with Soviet Union, 56, 57–59, 63-65 deterrence of Iraq, 71-77 failures in, toward Iraq, 39–42, 56-57, 102–103 generic knowledge in Gulf War decisions, 101-103 before Gulf War, towards Iraq, 33–38, 45-48 Gulf War goals, 38–39, 88 Gulf War termination, 93 Iran, 60 after Iran Iraq War, 33–37 Iraq, 77–78, 89, 93–95, 101–103 and Kuwait invasion, 38, 40 after Kuwait liberation, 95-99 Waltz, Kenneth, 108–109, 110 War. See also War termination appeasement and, 13, 61-62 conditional generalizations re- garding, 122 as continuation of politics, 90 Gulf, 38–39 Korean, 73, 75, 76, 92, 95-96, 127 limited vs. total, 92 military, vs. political, objectives, 92–93, 99-100 Six-Day, 75 Vietnam, xxii, 89, 92, 96 War termination comparative analysis, 89–90 conceptualization of strategies, 43 goal setting, 100 Gulf War, 89–99 World Restored, A, by Henry Kissin- ger, 48 World War II, 92 actor-specific models in, 126- 127 structural realist theory applied to, 113 termination of, 89 Zartman, I. William, 122–123 Zero-sum equation, 86–87 United States Institute of Peace The United States Institute of Peace is an independent, nonpartisan federal institu- tion created and funded by Congress to strengthen the nation's capacity to pro- mote the peaceful resolution of international conflict. Established in 1984, the In- stitute meets its congressional mandate through an array of programs, including grants, fellowships, conferences and study groups, library services, publications, and other educational activities. The Institute's Board of Directors is appointed by the President of the United States and confirmed by the Senate. Board of Directors Chester A. Crocker (Chairman), Distinguished Research Professor of Diplo- macy, School of Foreign Service, Georgetown University Max M. Kampelman, Esq. (Vice Chairman), Fried, Frank, Harris, Shriver and Jacobson, Washington Dennis L. Bark, Senior Fellow, Hoover Institution on War, Revolution, and Peace, Stanford University Thomas E. Harvey, Senior Vice President and General Counsel, Corporation for Public Broadcasting Theodore M. Hesburgh, President Emeritus, University of Notre Dame William R. Kintner, Professor Emeritus of Political Science, University of Pennsylvania Christopher H. Phillips, former U.S. Ambassador to Brunei Elspeth Davies Rostow, Stiles Professor of American Studies Emerita, Lyndon B. Johnson School of Public Affairs, University of Texas Mary Louise Smith, civic leader; former member, Board of Directors, lowa Peace Institute W. Scott Thompson, Professor of International Politics, Fletcher School of Law and Diplomacy, Tufts University Allen Weinstein, President, Center for Democracy, Washington Members ex officio Paul G. Cerjan, Lieutenant General, U.S. Army, and President, National Defense University Stephen J. Hadley, Assistant Secretary of Defense for International Security Policy Ronald F. Lehman II, Director, U.S. Arms Control and Disarmament Agency Melvyn Levitsky, Assistant Secretary of State for International Narcotics Matters Samuel W. Lewis, President, United States Institute of Peace (nonvoting)