Security Assurances and Nuclear Nonproliferation 9780804784917

Citation preview

Security Assurances and Nuclear Nonproliferation

SECURITY ASSURANCES and

NUCLEAR NONPROLIFERATION Edited by Jeffrey W. Knopf

Stanford Security Studies An Imprint of Stanford University Press Stanford, California

Stanford University Press Stanford, California © 2012 by the Board of Trustees of the Leland Stanford Junior University. All rights reserved.

No part of this book may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying and recording, or in any information storage or retrieval system without the prior written permission of Stanford University Press. Printed in the United States of America on acid-free, archival-quality paper Library of Congress Cataloging-in-Publication Data Security assurances and nuclear nonproliferation / edited by Jeffrey W. Knopf. pages cm. Includes bibliographical references and index. ISBN 978-0-8047-7827-5 (cloth: alk. paper) 1. Nuclear nonproliferation. 2. National security. 3. Security, International. 1. Knopf, Jeffrey W., editor of compilation. JZ5675.S44 2012 327.1'747-dc23

2011052155

Typeset at Stanford University Press in 10/14 Minion Special discounts for bulk quantities of Stanford Security Studies are available to corporations, professional associations, and other organizations. For details and discount information, contact the special sales department of Stanford University Press. Tel: (650) 736-1782, Fax: (650) 736-1784

CONTENTS

Preface and Acknowledgments Contributors

Vlll Xl

1 Introduction Jeffrey W. Knopf

1

PART I: THEORY AND HISTORY 2

Security Assurances: Initial Hypotheses Jeffrey W. Knopf

3 The Psychology of Assurance: An Emotional Tale Janice Gross Stein

13

39

4 The Role of Security Assurances in the Nuclear

Nonproliferation Regime John Simpson

57

PART II: CASE STUDIES 5 Libya, Nuclear Rollback, and the Role of Negative

and Positive Security Assurances Wyn Q. Bowen

89

6 Security Assurances and Iran: Assessment

and Re-conceptualization Jim Walsh

III

vi

CONTENTS

7 Maintaining Japan's Non-Nuclear Identity: The Role of u.s. Security Assurances Yuki Tatsumi

137

8 Infusing Commitment with Credibility: The Role of Security Assurances in Cementing the U.S.-ROK Alliance Scott Snyder and Joyce Lee

162

9 Assessing the Role of Security Assurances in Dealing with North Korea John S. Park

189

10 The United States and Swedish Plans to Build the

Bomb, 1945-68 Thomas Jonter 11

219

The "Model" of Ukrainian Denuclearization Sherman W. Garnett

246

PART III: CONCLUSIONS

12 Conclusions James J. Wirtz

275

Index

291

TABLES AND FIGURE

TABLES 2.1

Hypotheses on Security Assurances

6.1

Iran's Nuclear and Threat Environment, 1953-79

116

6.2

Iran's Threats, Assurances, and Nuclear Behavior, 1953-79

118

6·3

Iran's Nuclear and Threat Environment, 1979-2009

l23

6·4

Iran's Threats, Assurances, and Nuclear Behavior, 1979-2009

125

9.1 North Korea's Main Sources of Security Assurances

32

202

FIGURE 12.1 Case Outcomes by Type of Assurance

278

vii

Preface and Acknowledgments

This project emerged when two sources of inspiration came together. A course I teach provided the first spark. For several years, I have taught a graduate seminar that covers deterrence and other strategies whose aim is to influence other actors in international politics. The course includes a module on the topic of reassurance. Reassurance is a strategy in which one state tries to convince another state that it harbors no aggressive intentions toward it; the goal of reassurance is to reduce the chances of an unintended conflict between two states with an adversarial relationship. In past years, whenever it came time to update my syllabus, I could usually find relevant recent publications on all the other strategies covered by the course. In contrast, I found little new research that added to our understanding of strategies of assurance or reassurance, and I came to see my existing assigned readings as increasingly dated. In the back of my mind, I began thinking I should develop a research project to advance both theory and empirical research in this subject area. Fortunately, this vague impulse soon encountered a second key spark. Dr. Michael o. Wheeler, then director of the u.s. Defense Threat Reduction Agency's Advanced Systems and Concepts Office (DTRA-ASCO), gave a talk at the Naval Postgraduate School in which he identified security assurances as a critical topic in need of further research. Dr. Wheeler had in mind the use of security assurances as a tool for discouraging states from seeking to acquire nuclear weapons. As part of the efforts that have been made to prevent proliferation, existing nuclear-armed states have offered both "negative" assurances that they will not use their nuclear weapons to threaten non -nuclear states and "positive" commitments to assist states that are so threatened. viii

PREFACE AND ACKNOWLEDGMENTS

ix

This is a different context than that addressed by the academic literature on reassurance. Yet it struck me that the two topics had important elements in common, suggested by the fact that both employed the root term "assurance:' Both strategies share a core idea about how to influence others: both reassurance and security assurances seek to do so by alleviating some source of insecurity on the part of the recipient. By making the recipient feel less threatened, all versions of assurance aim to moderate its behavior in some way. In the realm of nuclear nonproliferation, security assurances have been intended to reduce recipients' concerns about being threatened by nuclear weapons, thereby making them feel less need to acquire a nuclear arsenal of their own. I decided that a study of the impact of security assurances could make a double contribution. First, it would help us ascertain whether assurances have been effective in slowing the spread of nuclear weapons and identify ways to make them more effective. Second, it could contribute to our understanding more generally of assurance as a type of influence strategy and bring renewed attention to the value of assurance as one option in the policy toolkit. I approached Dr. Wheeler with my idea, and with his encouragement developed it into a research proposal to DTRA-ASCO. The proposal received funding, and this book is the result. The research agenda was one that would be beyond the capacity of anyone individual, at least in any reasonable length of time, so it was clear from the outset that this project would have to be a team effort. I am very pleased and honored that the other authors included in this volume accepted my invitation to be part of the project. The project began with an authors' workshop at the Cheyenne Mountain Resort in Colorado Springs, Colorado, in August 2009. I provided an initial theory paper ahead of time to help guide the other paper writers in their research. Most important, the introductory paper included a preliminary list of hypotheses about factors that might affect whether assurances are effective, and I asked the authors of several country case studies to assess these hypotheses in their papers. At the workshop, all the paper writers received feedback from each other as well as assigned discussants. Based on feedback from the workshop plus subsequent comments from me and two external reviewers, all of the papers have gone through several rounds of revision to produce the chapters published here. For any collective endeavor such as this, the editor and authors depend upon invaluable assistance from many people and organizations. To start with, I thank DTRA-ASCO for providing the funding that made this research pos-

x PREFACE AND ACKNOWLEDGMENTS

sible. The opinions expressed in this book are, however, those of the authors and are not intended to represent the positions or policies of DTRA or the u.s. government. Many individuals also contributed to this project in various ways. First and foremost, I thank Mike Wheeler for his encouragement and support of the project and the valuable guidance he offered. I also thank former or current DTRA employees Kerry Kartchner, Jeffrey Fields, Dave Hamon, and Mike Keifer, as well as Jim Smith of the u.s. Air Force Academy, for helping facilitate the initial workshop and participating in its discussions. Several academic colleagues participated in the workshop and offered valuable inputs, and I express my gratitude to Bruno Tertrais, Etel Solingen, Michael Kraig, Jungsoo Kim, Chris Twomey, and Erik Dahl for their participation and the insights they provided. I also wish to acknowledge two anonymous external reviewers for Stanford University Press. They provided thoughtful and detailed comments on all the chapters, and this volume benefited from their very professional and constructive feedback. I also owe a special debt of gratitude to Meghan Holcomb, who handled nearly all the logistics and administrative tasks for the project. On several occasions, she helped prevent the workshop from falling through due to some snafu, and she also helped me, as P.I., to navigate the tricky waters of federal government contracting rules. In addition, I thank Brenda Lyons and Colleen Pace of Cheyenne Mountain Resort for helping make our workshop there a pleasant experience. Geoffrey Burn and Jessica Walsh of Stanford University Press have also been a joy to work with, and I deeply appreciate all their efforts to facilitate the process from initial proposal to finished book. Finally, I give a very special thanks to my partner, Christina Milburn. Any project involving so many participants plus government bureaucracy inevitably encounters speed bumps along the way. Christina's sympathy and good humor helped keep me sane whenever this project hit a rough patch. This book is dedicated to her.

Jeffrey W Knopf Santa Cruz, CA

Contributors

Wyn Bowen is Professor of Non-Proliferation and International Security at King's College London, where he is also Director of the Centre for Science and Security Studies in the Department of War Studies. His publications include Libya and Nuclear Proliferation (IISS Adelphi Paper, 2006); The Politics of Ballistic Missile Non-Proliferation (Macmillan, 2000); The Global Partnership against WMD: Success and Shortcomings of G8 Threat Reduction since 9/11 (RUSI Whitehall Paper, 2011), with Alan Heyes and Hugh Chalmers; and "Iran's Nuclear Challenge: Nine Years and Counting;' International Affairs (2011), with Jonathan Brewer. Sherman Garnett is Dean of the James Madison College of Public Affairs at Michigan State University. He served in the Office of Secretary of Defense from 1984 until 1994, working during the last few years as the Assistant Secretary for

Russia, Ukraine, and Eurasia during the crucial phase of the trilateral talks on Ukrainian denuclearization. He became a Senior Associate at the Carnegie Endowment for International Peace in 1994, publishing his Keystone in the Arch:

Ukraine and the Emerging Security Environment of Eastern and Central Europe in 1997. Thomas Jonter is Professor of International Relations in the Department of Economic History, Stockholm University. He is also chair of the ESARDA (Eu-

ropean Safeguards and Research Development Association) working group for Training and Knowledge Management. Professor Jonter has been a visiting scholar at the Stockholm International Peace Research Institute (SIPRI) and the Europe Center at the Freeman Spogli Institute for International Studies, xi

xii

CONTRIBUTORS

Stanford University. His recent publications include "The Swedish Plans to Acquire Nuclear Weapons, 1965-1968: An Analysis of the Technical Preparations;' Science and Global Security (2010); "Between Welfare and Warfare: The Rise and Fall of the 'Swedish Line' in Nuclear Engineering;' coauthored with Maja Fjaestad, in Science for Welfare and Warfare (Science History Publications, 2010); and "Nuclear Power without Nuclear Weapons;' in Terrorism and Global Insecurity: A Multidisciplinary Perspective (Linton Atlantic Books, 2009). Jeff Knopf is Professor and Chair of the Nonproliferation and Terrorism Studies Program at the Monterey Institute of International Studies. Dr. Knopf is the author of Domestic Society and International Cooperation: The Impact of Protest on u.s. Arms Control Policy (Cambridge University Press, 1998) and is a former editor of The Nonproliferation Review. His essay "The Fourth Wave in Deterrence Research" won the Bernard Brodie Prize for the best article in Contemporary Security Policy in 2010. Joyce Lee is a Research Associate in the Korean Studies Program at the Walter H. Shorenstein Asia-Pacific Research Center at Stanford University. Her research focuses on Northeast Asian security issues and U.S. foreign policy toward the region. Ms. Lee received a double Bachelors degree in Political Science and Psychology from University of Washington and a Masters degree in Public Policy from Cornell University. John S. Park is a senior adviser at the U.S. Institute of Peace and a research fellow at the Harvard Kennedy School's Belfer Center for Science and International Affairs. His publications include "North Korea, Inc.: Gaining Insights

into North Korean Regime Stability from Recent Commercial Activities" (USIP Press, April 2009 ), and "North Korea's Nuclear Policy Behavior: Deterrence and Leverage;' in The Long Shadow: Nuclear Weapons and Security in 21st Century Asia (Stanford University Press, 2008). John Simpson is Professor of International Relations at the University of

Southampton, UK, and a Visiting Professor at the Centre for Science and Security Studies at King's College London. He has also been program director of the Programme for Promoting Nuclear Nonproliferation (PPNN); a member of the UN Secretary General's Advisory Board on Disarmament Matters; and since 1999 an advisor to UK delegations to NPT meetings and conferences. Dr. Simpson's recent articles include "Is the Nuclear Non-Proliferation Treaty Fit for the Purpose?" (UNA-UK Briefing Report, 2011) and "British Nuclear

CONTRIBUTORS

xiii

Weapon Stockpiles, 1953-78: A Commentary on Technical and Political Drivers" (RUSI Journal, 2011). Scott Snyder is Senior Fellow for Korea Studies and Director of the Program

on U.S.-Korea Policy at the Council on Foreign Relations. He was formerly Director of the Center for U.S.-Korea Policy at the Asia Foundation, where he also served as Korea country representative from 2000 to 2004. Mr. Snyder has worked as an Asia specialist in the Research and Studies Program of the u.S. Institute of Peace, and served as Acting Director of the Asia Society's Contemporary Affairs Program. His books include China's Rise and the Two Koreas: Politics, Economics, Security (Lynne Rienner, 2009) and Negotiating on the Edge: North Korean Negotiating Behavior (USIP, 1999). He is also the editor of The u.S.-South Korea Alliance: Meeting New Security Challenges (Lynne Rienner, 2012) and coeditor of Paved with Good Intentions: The NGO Experience in North Korea (Praeger, 2003). Janice Gross Stein is the Belzberg Professor of Conflict Management in the De-

partment of Political Science and the Director of the Munk School for Global Affairs at the University of Toronto. She is the coauthor, with Eugene Lang, of the prize-winning The Unexpected War: Canada in Kandahar (Penguin, 2007) and coeditor, with Peter Gourevitch and David Lake, of The Credibility of Transnational NGOs (Cambridge University Press, 2012). She also edited the book Diplomacy in the Digital Age: Essays in Honour ofAmbassador Allan Gotlieb (Random House, 2011). Yuki Tatsumi is a Senior Associate with the East Asia Program at the Henry L.

Stimson Center. She is the author of Japan's National Security Policy Infrastructure: Can Tokyo Meet Washington's Expectations? (Stimson, 2008), as well as a coauthor with Andrew L. Oros of Global Security Watch: Japan (Praeger, 2010). She is also a recipient of the 2009 Yasuhiro Nakasone Incentive Award. She holds a B.A. in Liberal Arts from International Christian University in Tokyo, Japan, and an M.A. in International Economics and International Relations from the Paul H. Nitze School of Advanced International Studies (SAIS) of Johns Hopkins University. Jim Walsh is a Research Associate at the Massachusetts Institute of Technology's

Security Studies Program. He is one of a handful of Americans who have traveled to both Iran and North Korea for talks with officials about nuclear issues. Before coming to MIT, Dr. Walsh was Executive Director of the Managing the

xiv

CONTRIBUTORS

Atom project at Harvard University's John F. Kennedy School of Government and a visiting scholar at the Center for Global Security Research at Lawrence Livermore National Laboratory. He is the author of Dangerous Myths: North Korea, the United States, and the Future ofAsia (Yale University Press, 2012).

James J. Wirtz is Dean of the School of International Graduate Studies, Naval Postgraduate School, Monterey, California, and Director of the Global Center for Security Cooperation, Defense Security Cooperation Agency. He is coeditor of Complex Deterrence: Strategy in the Global Age (University of Chicago Press, 2009) and of Over the Horizon Proliferation Threats (Stanford University Press, 2012).

Security Assurances and Nuclear Nonproliferation

Introduction Jeffrey W. Knopf

In the pursuit of national security, states often rely on the threat or use of military force to defend their interests and deter challenges. Yet threat-based strategies are not always the best option. In some cases, a state may be better off seeking to give others a greater sense of security, rather than by holding their security at risk. Efforts to alleviate insecurity can be important not only in dealing with potential adversaries, but in relations with allies as well. Despite their possible value, however, security assurances-which can be defined as promises to respect or ensure the security of others-have received much less attention from policymakers and scholars than have measures for defense or deterrence. The most prominent use of security assurances in international politics has been in conjunction with efforts to prevent the spread of nuclear weapons. The 1968 Treaty on the Non-Proliferation of Nuclear Weapons, more commonly called the Non-Proliferation Treaty (NPT), created two classes of states: nuclear weapon states and non-nuclear weapon states. Only the five states that had tested nuclear weapons before the treaty was opened for signature (the United States, Russia, China, Britain, and France) could join as nuclear weapon states, while all other states were expected to join as non-nuclear weapon states. Those countries required to forswear nuclear weapons have sought to make sure that doing so would not jeopardize their security vis-a.-vis nuclear-armed states. They have requested both negative security assurances, which involve a pledge by nuclear weapon states not to use or threaten the use of nuclear weapons against non-nuclear weapon states, and positive security assurances, which involve a pledge to come to the aid of non-nuclear weapon states that are

2

JEFFREY W. KNOPF

nevertheless subject to such a threat or attack. The nuclear weapon states have offered such pledges, but not always in forms as strong as most non-nuclear weapon states would like. In addition to assurances directly associated with the nonproliferation regime, some nuclear weapon states (especially the United States) have offered positive security assurances in a bilateral format. These generally rely on extended deterrence commitments, more colloquially referred to as a nuclear umbrella. From time to time, policymakers rediscover security assurances as a potential nonproliferation tool. Commenting on Iran's progress toward a nuclear weapon capability, U.S. secretary of state Hillary Clinton suggested in July 2009 that the United States might respond by extending "a defense umbrella over the [Persian Gulf] region:' The statement was widely interpreted as an attempt to signal to Iran's Arab neighbors that they would not need to react to an Iranian bomb with nuclear weapons development efforts of their own because, if they wanted U.S. assistance, the United States would help defend them against a possible Iranian threat. l We have little knowledge, however, about whether such an assurance would be effective. The goal of this volume is to investigate the potential utility of security assurances both as a nonproliferation tool and more generally. The first three essays in the volume identify and define different concepts of assurance, mine relevant bodies of theory to develop initial hypotheses about the impact of assurances, and summarize the history of assurances in the nonproliferation regime. The next seven chapters assess the role of security assurances in a number of cases of nuclear proliferation and restraint. These empirical chapters seek to evaluate the overall impact of assurances as well as to identify conditions under which security assurances are more or less likely to be effective. Based on the research findings, the concluding chapter offers practical pointers on how to make security assurances as effective as possible in preventing nuclear proliferation. The focus on nuclear proliferation plays a dual role in this study. First, for the purpose of examining assurance strategies in general, decisions about nuclear acquisition or restraint are a useful empirical test bed. This is the only policy area in which security assurances have been widely used, creating a ready set of cases that can be selected for study. In addition, because decisions about whether or not to obtain nuclear weapons involve the highest possible national security stakes, they pose a hard test for a hypothesis that assurances can influence state behavior. If security assurances have some effectiveness in prevent-

INTRODUCTION

3

ing nuclear proliferation, this will permit confidence in an inference that they could be effective more broadly. Second, nuclear nonproliferation is a focus in this study due to its intrinsic importance. Efforts to prevent the spread of nuclear weapons have long been intended to forestall the possibility that regional rivals could end up in a nuclear war. More recently, the potential that nonstate terrorist actors or aggressive regimes could get their hands on a nuclear device has added urgency to nonproliferation efforts. While nonproliferation policy discussions often cite security assurances as a potentially relevant tool,2 there has been almost no empirical research that assesses the effectiveness of the range of available assurance strategies, including both positive and negative assurances. This volume represents the first study to use systematic empirical research to produce generalizations about the effectiveness of security assurances. ASSURANCE AS AN UNDERSTUDIED STRATEGY

Assurances are promises. They involve declarations or signals meant to conveya commitment to take or refrain from taking certain actions in the future. Specifically, assurances can be defined as attempts by one state or set of states to convince another state or set of states that the senders either will not cause or will not allow the recipients' security to be harmed. Compared with other strategies employed in international politics, assurances have not been the subject of much empirical research. Coercive strategies such as deterrence and compellence have received by far the greatest attention when it comes to assessing the ways states attempt to influence the behavior of other states. 3 In addition to the manipulation of military threats, the use of economic statecraft has been a focus of much research, including efforts to determine the effectiveness of economic sanctions and to a lesser extent the usefulness of positive incentives. 4 In recent years, the idea of "soft power" as a possible basis of influence has also elicited considerable interest. 5 In contrast, there has been little effort to develop a general theory of security assurances or to conduct systematic empirical research on the effectiveness of assurances. Most discussions of nonproliferation-related assurances focus on policy issues. Those that discuss assurances associated with the NPT typically either address the legal status of existing assurances or offer policy prescriptions for how to strengthen them. 6 Since 2008, several studies have addressed the assurance implications of extended nuclear deterrence, but again these focus mostly on contemporary policy debates rather than empirical analysis

4

JEFFREY W. KNOPF

of the impact of assurances.? Only one published study has involved empirical research on the effectiveness of the full range of nonproliferation-related assurances. In it, Bruno Tertrais surveyed a number of past cases, finding in general that positive assurances conveyed by bilateral defense treaties have a greater impact than other types of assurances. 8 The present volume seeks to go deeper by exploring the conditions under which different types of assurance are more or less likely to be effective. Some research on the causes of proliferation, which is discussed more fully in the next chapter, also analyzes security assurances. In nearly all cases, the studies consider only assurances conveyed by bilateral defense pacts and not the wider set of assurances associated with the nonproliferation regime. In addition, the findings in this literature regarding the impact of assurances have been highly contradictory, suggesting that more research is needed in any case. 9 The most relevant body of social science literature concerns the concept of reassurance. 10 Reassurance, which involves seeking to convince another state that one harbors no hostile intentions toward that state, is a form of assurance. But the range of possible assurance strategies is broader than just reassurance. In some situations, assurance is more likely to take the form of a security guarantee to an ally than a nonaggression pledge to an adversary. Because assurance can take other forms besides reassurance, work on reassurance cannot by itself provide a complete theoretical framework for examining the effectiveness of security assurances. In addition, the body of empirical research on reassurance is also limited,l1 and it has not to date focused on the possible role of reassurance in combating proliferation. In sum, as a focus for empirical research, security assurances are perhaps the least studied of all the strategies that states can utilize in efforts to influence other actors in world politics. This volume seeks to rectify that neglect. OUTLINE OF THE VOLUME

Part I of this study covers relevant theory and history. The next two chapters deal with the theory of security assurances. In Chapter 2, Knopf first reviews the different terminology used in relation to assurance in an effort to clarify definitions and show how different concepts of assurance relate to one another. This initial theory chapter then derives from relevant existing bodies of research more than a dozen hypotheses about the conditions that might affect the effectiveness of assurances. These hypotheses are investigated in the case study chapters in the volume.

INTRODUCTION

5

The next chapter is by Janice Stein, one of the leading authorities on reassurance. In it, Stein summarizes recent psychological research, especially concerning the role of emotions, to update her previous work on how psychological factors affect the use of assurance strategies. Then, in Chapter 4, John Simpson provides a history of NPT-related assurances. Simpson, who has been personally involved in NPT-related activities for many years, reviews the evolution of both positive and negative assurances in the nonproliferation regime and ongoing debates about how to strengthen such assurances. Following the reviews of theory and history, Part II of the volume contains case studies that evaluate the impact of assurances on seven specific countries. The research design for this part of the volume follows the method of "structured, focused comparison" developed by Alexander George. 12 The case-study authors were each provided in advance with a standardized list of questions about assurances and other factors to investigate in their cases. The case-study authors were also given a draft list of the hypotheses from the initial theory chapter for their consideration. The cases were selected to include both apparent nonproliferation successes and failures as well as cases where both positive and negative assurances were more relevant. There is also variation within several of the cases, with assurances being available in some time periods but not others. The emphasis was on the regions of greatest contemporary proliferation concern-the Middle East and East Asia-supplemented by some cases in which assurances have long been assumed to have played an important role. The Middle East part of the study involves chapters by Wyn Bowen on Libya and James Walsh on Iran. For Libya and Iran, negative assurances were more relevant than positive assurances, with the difference that Libya appears to be a nonproliferation success story while Iran does not. Three cases from Northeast Asia follow next: Yuki Tatsumi on Japan, Scott Snyder and Joyce Lee on South Korea, and John Park on North Korea. Japan and South Korea represent cases where positive assurances have been more relevant and apparently successful. For North Korea, like Iran, negative assurances have emerged as most relevant but to date without apparent success. The final two case studies involve analyses of Sweden by Thomas Jonter and of Ukraine by Sherman Garnett. In these cases, it has generally been assumed that assurances-primarily positive in the case of Sweden and negative in the case of Ukraine-helped to convince both countries to renounce nuclear weapons; the chapters here explore whether these assumptions are well founded.

6

JEFFREY W. KNOPF

The volume ends with a concluding chapter by James Wirtz that summarizes the case findings and the policy lessons they hold. As this project progressed, it became clear that the findings might reflect selection effects due to the cases chosen for study. First, the cases all involve countries that were at some point deemed serious proliferation risks. As a consequence, the cases nearly all involve special, country-specific assurances beyond the generalized assurances associated with the NPT. The volume may therefore underestimate the impact of NPT-related assurances because the countries willing to be satisfied by those assurances did not become part of the case sample. Second, the case selection is likely to lead to clearer conclusions concerning positive than negative assurances. There are more cases in which positive assurances were more relevant, and the negative assurance cases may overweight the rate of failure given that two of the cases (Iran and North Korea) have been among the most challenging cases for finding any effective policy tool. Future research on countries from relatively nonaligned regions, such as South America or sub-Saharan Africa, would provide a more fully representative sample from which to evaluate the assurances most closely associated with the nonproliferation regime, especially negative security assurances. Still, the cases in this volume make it possible to offer assessments about whether assurances can be effective in curbing proliferation and, if so, the conditions that affect the likelihood of success. The case study authors find that security assurances had an impact in most but not all of the cases. The success rate is greater in the cases of positive assurances, but to some degree the Libya and Ukraine cases show that negative assurances can be an important nonproliferation tool as well. The impact of assurances is typically modest rather than decisive, suggesting that they are no silver bullet. Assurances appear to be most effective when they are embedded in a larger strategy that also involves positive incentives, broader efforts at reassurance, or other measures. Domestic politics in both the recipients and senders of assurance are also important. Finally, the credibility of assurances is important and far from automatic, but the factors that produce credibility tend to be highly context dependent, although opportunities for defense cooperation or consultation often show up as an important factor. The findings in the volume lend support to the conventional wisdom that bilateral security guarantees have important advantages in dissuading proliferation. The details of the cases also reveal, however, that the influence of positive assurances rests to an important degree on the normative delegitimizing of nuclear weapons associated with the NPT. In addition, both the theory and

INTRODUCTION

7

case chapters reveal important tensions between positive and negative forms of assurance. Extended nuclear deterrence commitments associated with some forms of positive security assurance encourage nuclear weapon states to maintain robust nuclear arsenals and a readiness to use these weapons. Such a posture can make promises not to threaten the use of nuclear weapons against non-nuclear states appear insincere. Hence, the findings in this volume point to a policy dilemma: it may be difficult to gain the benefits of both positive and negative security assurances simultaneously. This makes it important to explore possible non-nuclear options for providing positive security assurances. These might include the use of conventional forces for deterrence, arms exports to bolster the recipient's defense capabilities, provision of missile defenses, or nonmilitary forms of assistance. The research findings in this volume also suggest that finding effective non-nuclear forms of positive security assurance will not be easy. CONCLUSIONS

Security assurances are an integral component of the nuclear nonproliferation regime. But a strategy of seeking to assure other states about their security could be much more widely applicable in international politics. It could help a state that is seeking to reduce the chances of conflict with a rival; to restrain its allies from provocative behavior or encourage them to stand fast; or to elicit cooperation from others in dealing with a variety of security problems, such as terrorism, piracy, or communal conflicts. When states do not believe they are risking their security, they may be more likely to adjust their policies and behavior in directions favored by an assurance provider. In contrast to the sizable literatures on other tools for influencing state behavior, such as deterrence and economic sanctions, there is little empirical research on the effectiveness of security assurances. This volume aims to advance the state of knowledge on this question. Because assurances are unlikely to be uniformly effective in all cases, this volume aims in particular to develop conditional generalizations about when assurances are most likely to have an impact. NOTES 1. Mark Landler and David E. Sanger, "Clinton Speaks of Shielding Mideast from Iran;' New York Times, July 23, 2009, AI. 2. For a brief history and analysis of different forms of security assurance, see Michael o. Wheeler, Positive and Negative Security Assurances, Project on Rethinking Arms

8

JEFFREY W. KNOPF

Control (PRAC) Paper NO.9 (College Park: University of Maryland, Center for International and Security Studies, February 1994). 3. For good reviews of the vast literature on deterrence, see Paul K. Huth, "Deterrence and International Conflict: Empirical Findings and Theoretical Debates;' Annual

Review ofPolitical Science 2 (1999); and Patrick M. Morgan, Deterrence Now (Cambridge, UK: Cambridge University Press, 2003). For research on coercive diplomacy that also summarizes earlier research, see Robert J. Art and Patrick M. Cronin, eds., The United States and Coercive Diplomacy (Washington, DC: United States Institute of Peace, 2003)· 4. For contrasting views on sanctions, compare Gary Clyde Hutbauer, Jeffrey J. Schott, Kimberly Ann Elliott, and Barbara Oegg, Economic Sanctions Reconsidered, 3rd edition (Washington, DC: Peterson Institute for International Economics, 2007); and Robert A. Pape, "Why Economic Sanctions Do Not Work;' International Security 22, no. 2 (Fall 1997). An important study of positive incentives that explicitly focuses on nonproliferation cases is Thomas Bernauer and Dieter Ruloff, eds., The Politics of Positive

Incentives in Arms Control (Columbia: University of South Carolina Press, 1999). 5. Joseph S. Nye, Jr., Soft Power: The Means to Success in World Politics (Boulder, co: Perseus Books, 2004). 6. George Bunn and Roland Timerbaev, "Security Assurances to Non-NuclearWeapon States: Possible Options for Change;' Programme for Promoting Nuclear NonProliferation (PPNN), Issue Review no. 7, September 1996; Bunn, "The Legal Status of U.S. Negative Security Assurances to Non-Nuclear-Weapon States;' Nonproliferation

Review 4, no. 3 (Spring-Summer 1997); Rebecca Johnson, "Security Assurances for Everyone: A New Approach to Deterring the Use of Nuclear Weapons;' Disarmament Diplomacy, no. 90 (Spring 2009). 7. Victor Utgoff and David Adesnik, On Strengthening and Expanding the U.S. Nuclear Umbrella to Dissuade Nuclear Proliferation, IDA Paper P-4356 (Alexandria, VA: Institute for Defense Analyses, July 2008); David S. Yost, "Assurance and U.S. Extended Deterrence in NATO;' International Affairs 85, no. 4 (July 2009); Clark A. Murdock and Jessica M. Yeats, Exploring the Nuclear Posture Implications of Extended Deterrence and Assurance (Washington, DC: Center for Strategic and International Studies [CSIS], November 2009). 8. Bruno Tertrais, "Security Assurances and the Future of Proliferation;' in Over the

Horizon Proliferation Threats, ed. James J. Wirtz and Peter R. Lavoy (Stanford: Stanford University Press, 2012). A paper from a set of conference papers available on-line also discusses the NPT and U.S. security guarantees, finding that both have some impact on limiting proliferation, but the latter has been the stronger measure. Although the paper does not provide detailed empirical evidence to support its conclusions and does not consider the possible impact of negative security assurances, it is still noteworthy for being one of the very rare studies to attempt an empirical assessment of the impact

INTRODUCTION

9

of assurances. See Sara Z. Kutchesfahani, "The Relevance of Historical Experience to Current Nuclear Proliferation Challenges;' in A Collection of Papers from the 2009 PONI Conference Series (Washington, DC: Project on Nuclear Issues, CSIS, 2010), CSIS.org/ imageslstories/poniho0728_collection_oLconference_papers_2009.pdf. 9. The studies referred to in this paragraph are discussed more fully in the next chapter. For citations, see Chapter 2, notes 17-23. 10. Richard Ned Lebow and Janice Gross Stein, "Beyond Deterrence;' Journal of So-

cial Issues 43, no. 4 (Winter 1987). 11.

This research is discussed in the next chapter. For citations, see Chapter 2, n. 12.

12. Alexander L. George and Andrew Bennett, Case Studies and Theory Development in the Social Sciences (Cambridge, MA: MIT Press, 2005).

Security Assurances: Initial Hypotheses Jeffrey W. Knopf

This chapter describes the theoretical framework that guided the research in this volume. It proceeds in two sections. It first discusses variations in the terminology related to assurance. In order to standardize terminology for the project, this section seeks to define and label different notions of assurance. It also briefly summarizes existing empirical research on the variants of assurance most directly relevant to this project. The second section of this chapter introduces preliminary hypotheses about the conditions under which security assurances are most likely to be effective. It does so by drawing on and adapting hypotheses from two bodies of literature: research on deterrence and reassurance, and research on the causes of nuclear proliferation. These hypotheses are evaluated in the case study chapters in the volume. DEFINITION OF KEY TERMS

In both academic literature and government policy documents, the terms "assurance;' "assurances;' and "reassurance" all appear-sometimes used interchangeably, sometimes with different connotations. This section seeks to standardize the definitions of these various terms. This will make it possible to develop a generic, overarching concept of assurance that can cover multiple variants, and to have more precise terms for identifying the distinct variants. This section will also show that the NPT-related distinction between positive and negative assurances can be generalized, in that all forms of assurance tend to entail either a positive commitment to come to the aid of another party or a negative commitment to refrain from taking action against it.

13

14 JEFFREY W. KNOPF

Assurance I-A Component of Deterrence: In his influential writings about coercive strategies, Thomas Schelling introduced the idea of assurance as a necessary element of deterrence or compellence. If deterrence involves a threat to impose costs on an actor if it takes an action one wishes to prevent, then logically the strategy entails a promise not to impose those costs as long as the actor refrains from taking the unwanted action.! Without this assurance, because the target must fear action will be taken against it no matter what it does, it has no incentive to comply with deterrence. Assurance in this first usage is not a separate strategy, but is instead one component of a deterrence strategy. Henceforth, it will be referred to as deterrence-related assurance. Assurance II-Alliance Commitments: A more recent usage of the term "assurance" emerged in U.S. policy in the George W. Bush administration. In its initial Quadrennial Defense Review (QDR) and National Security Strategy, the administration identified assuring allies and friends of the U.S. commitment to them as one of the goals of U.S. defense strategy.2 In the first usage of assurance, above, it is a measure directed toward potential adversaries as part of an effort to deter them. As used by the Bush administration, assurance is a strategy directed at allies, not adversaries. It can also be considered as a stand-alone strategy, rather than being only a subelement of deterrence or compellence. Unlike NPT-related security assurances, the Bush administration version of assurance is not limited to nuclear scenarios; it also implies assurance against conventional and other non-nuclear threats, as well as the potential use of non-nuclear means to provide such assurance. To distinguish it from the first type of assurance, the second form of assurance will be called alliance-related assurance. 3 This type of assurance is related to, but not identical with, extended deterrence. The latter strategy is meant to affect the calculations of adversaries, and its effectiveness is measured by whether or not it keeps them from an attack on a state's ally they might otherwise have attempted. Extended deterrence might also be intended to assure allies, but if so the relevant measure of effectiveness is whether the ally feels protected, not whether the adversary is deterred, and if the ally lacks confidence in extended deterrence it will not be assured. Hence, although there is a great deal of research on what makes extended deterrence work to deter aggression,4 this research says little about what makes allies feel assured; to date there is virtually no empirical research on what makes alliancerelated assurance effective.

SECURITY ASSURANCES: INITIAL HYPOTHESES

15

Reassurance: Reassurance is a strategy of seeking to persuade another state that one harbors no aggressive intentions toward it. It is generally put forward as a way to reduce the chances of unintended conflict. There are two main strands of literature on reassurance: one psychological, the other structural. The psychological approach emerged in the early 1960s in the work of Charles Osgood and Amitai Etzioni.5 In the mid-1980s, Richard Ned Lebow and Janice Gross Stein returned to this tradition and did extensive work to elaborate the theory behind reassurance. 6 Stein has updated her work on the role of psychological factors in reassurance for this volume. Scholars working in this tradition say states can convey reassurance by taking a series of modest unilateral cooperative initiatives or else making a larger, dramatic symbolic gesture. The persistence in or unexpected nature of cooperation is intended to overcome psychological factors that would otherwise keep the other side from altering its image of the first side as implacably hostile. The other strand of thinking about reassurance emerged from Robert Jervis's influential distinction between the deterrence and spiral models and subsequent refinements by Charles Glaser.? This line of work holds that deterrence is the appropriate policy for dealing with expansionist or opportunistic states, but is the wrong prescription and likely to set off an escalatory spiral when dealing with states motivated by insecurity. Implicit in this analysis is the idea that in situations that fit the spiral model, reassurance is the appropriate strategy. Jervis, Glaser, and others working mainly within the realist tradition link reassurance to structural conditions. This leads them to view reassurance primarily in military terms and to recommend placing an emphasis on defensive rather than offensive military capabilities (that is, defensive or nonoffensive defense).8 Lebow and Stein, in contrast, emphasize the impacts of domestic politics and especially psychology and view reassurance as a way to overcome constraints on rationality that arise from these sources. 9 For them, reassurance is not just military but includes broader political gestures and agreements. 10 Finally, taking a somewhat different approach, Andrew Kydd has sought to ground reassurance on a rational-actor foundation, using game theory to model a reassurance game. He finds signaling that is costly but not too costly to be the key, as this enables a security-seeking state to convey its lack of aggressive intent without leaving it vulnerable in case the recipient is actually motivated by greed rather than insecurityY

16

JEFFREY W. KNOPF

The effectiveness of reassurance has not been a subject of nearly as much empirical research as has deterrence, but there are some relevant studies. Unfortunately, these have reached contradictory conclusions about how well reassurance works, with even the same case sometimes being a subject of disagreementY Reassurance is clearly a form of assurance, in that it seeks to assure the other side that one is not out to harm them. But it differs from the two forms of assurance described above. Reassurance can be used as a complement or even a completely alternative strategy to deterrence, so it should be distinguished from the use of assurance as an implied element within a deterrence strategy. Reassurance is also a negative form of assurance-a promise by state A not to attack state B---so it should be distinguished from the positive form of assurance implied by a commitment to come to the defense of one's allies if they are attacked. In this sense, reassurance is narrower than the category of security assurances as a whole, meaning that research on reassurance cannot by itself supply a complete theory of assurance strategies. But in another sense reassurance is broader than NPT-related security assurances, because it applies to the conventional realm as well as the nuclear realm and even potentially to nonmilitary issues.

Security Assurances: The most common use of the term "assurances" (plural rather than singular) comes from the nuclear nonproliferation regime. Countries that give up nuclear weapons naturally want guarantees that they will not be risking their security as a result. Efforts to assure them against nuclear threats will be labeled nonproliferation-related assurances. The history of NPT-related security assurances is covered in the chapter by John Simpson, so this chapter will not go into detail on that history. Empirical research on the effectiveness on NPT-related assurances is made complicated, however, by the fact that the status of these assurances is ambiguous. Non-nuclear weapon states have requested two types of assurance. Positive security assurances involve promises to come to the aid of non-nuclear states if they are threatened or attacked by nuclear weapons. 13 Negative security assurances are promises not to use or threaten the use of nuclear weapons against non-nuclear weapon states. The nuclear weapon states would not agree to include security assurances in the text of the NPT, so existing security assurances have all arisen from commitments made outside the treaty itself. In practice, these assurances have often involved qualifications, so they have not been con-

SECURITY ASSURANCES: INITIAL HYPOTHESES

17

sidered satisfactory by all non-nuclear weapon states. As a result, there have been periodic efforts to have security assurances made universally applicable and legally binding. It is also possible to make a distinction between security assurances and security guarantees, although these terms are also sometimes used interchangeably. The term "security guarantee" is often used to connote a commitment made by a state to an ally through the mechanism of a formal defense treaty. In the nonproliferation realm, security guarantees generally imply extending a nuclear deterrent umbrella over an ally with the goal of convincing the ally that it does not need a nuclear deterrent of its own. As ordinary language would suggest, the term "guarantee" also implies a stronger commitment than an assurance. In practice, however, a security guarantee is a particular form of positive security assurance, one involving a bilateral commitment conveyed in a legally binding form, and so security guarantees are best viewed as a subset of the broader category of nonproliferation-related assurances. This chapter will use the term "NPT-related assurances" when it wishes to discuss only assurances associated with the global regime, whereas "nonproliferation-related assurances" will be used as the more inclusive term that includes bilateral security guarantees. Existing Empirical Research on Nonproliferation-Related Assurances

Much of the writing about security assurances in recent decades has been prescriptive, urging the nuclear weapon states to reaffirm and strengthen NPTrelated assurances. 14 Empirical research on the effectiveness of assurances as a nonproliferation tool is more limited. A critique of NPT-related assurances by Joseph Pilat argued that there is little evidence such assurances have had much impact, but the article was not based on any new empirical research. IS As noted in the introductory chapter, only Bruno Tertrais has attempted an overall survey of the impact of the different types of security assurances, leading him to conclude that positive assurances in the form of bilateral defense pacts have helped stem proliferation. 16 This book takes two steps to deepen the analysis contained in Tertrais's work, which is based on a brief overview of a number of cases. First, this volume examines several country cases in greater empirical detail. Second, it develops and evaluates hypotheses about factors that might be related to the effectiveness of security assurances. One other source of relevant empirical research comes from studies of the

18

JEFFREY W. KNOPF

causes of nuclear proliferation and restraint. These have sometimes touched on the role of security assurances, though none have made this a central object of the study. Within this proliferation literature, some pertinent evidence can be found in a growing body of statistical analyses. These studies, which use quantitative techniques to assess suspected causes of proliferation, all include defense pacts with nuclear weapon states as one of their variables. Unfortunately, the results are inconsistent and far from definitive. Some analyses find that a defense pact with a nuclear ally reduces the likelihood of nuclear proliferation,17 but others report the opposite correlation, implying that defense pacts with nuclear allies might actually increase the risk of proliferation. IS Further complicating matters, these findings are often not statistically significant across all the models tested, which might suggest that there is no relation between defense pacts and proliferation. The statistical studies do not attempt to measure the impact of negative security assurances, nor do they consider other possible types of positive assurance beyond a defense treaty. The fact that some tests find a modest impact for positive assurances, while others do not, might be due to variations across individual cases in the importance of such assurances. If so, the case study research in this volume might help tease out the factors that make positive assurances more or less important. Compared with the quantitative research, the case study literature on the causes of proliferation has tended to reach more skeptical conclusions about the impact of security guarantees. Much of the recent research has been concerned with challenging traditional realist-oriented security explanations for proliferation. Within a security model of proliferation, the availability of a security guarantee from a nuclear ally would be one explanation for why a state facing security threats nevertheless eschews nuclear weapons. Several studies note, however, that realist theory itself, given its emphasis on self-help, would not lead to great confidence in the willingness of states to entrust their security to an ally that might be making itself vulnerable to nuclear retaliation if it acts.19 Etel Solingen argues that the historical record also does not suggest a strong correlation. Some states with nuclear allies nevertheless pursued nuclear weapons (Britain, France, North Korea before the fall of the Soviet Union), while other states with no nuclear protectors have changed course (Argentina, Brazil, South Africa, Egypt, Libya).20 T. V. Paul, while more sympathetic to security explanations, also acknowledges that the impact of alliance ties varies across cases,21 and Jacques Hymans likewise sees no consistent correlation be-

SECURITY ASSURANCES: INITIAL HYPOTHESES

19

tween alliances and proliferation decisions. 22 Maria Rost Rublee, in contrast, is open to a limited role for nuclear allies. Although she is critical of security explanations, Rublee argues that security guarantees sometimes affect domestic debates. In countries where a growing segment of domestic opinion has embraced nonproliferation norms, the offer of a nuclear security guarantee can serve as a consolation prize to conservative and hawkish elites, leading them to abandon support for an independent nuclear program and join an emerging domestic consensus in favor of non-nuclear status. 23 None of the preceding studies dismiss security guarantees entirely. Rather, in suggesting that positive security assurances are not always decisive, these studies indicate the need for a differentiated analysis. It is possible that assurances are important to some countries in some circumstances, but less influential in other cases. If so, it will be necessary to develop contingent generalizations, and comparative analysis across cases can help identify the conditions under which security assurances are most likely to be effective. In addition, like the statistical literature, case studies of proliferation decisions have focused on positive assurances in the form of alliance commitments. Negative security assurances have largely been overlooked. The literature on proliferation will be a useful source of hypotheses about assurances, but in its current state it is far from offering an adequate understanding of the effects of assurance strategies. MOVING TOWARD GENERALIZATIONS

The preceding review of variations in terminology indicates that it is possible to identify a generic, overarching concept of assurance that contains multiple variants. In general, security assurances can be defined as any attempt by a state or group of states to convince another state or group of states that their security will not be harmed. NPT-related assurances are one form of assurance (or perhaps more accurately, one context in which assurances have been employed), but other forms and policy goals for assurances are possible. However, the NPTrelated distinction between positive and negative assurances can be generalized. Alliance-related assurance, as described in the Bush administration QDR, is a type of positive assurance. Positive assurances involve pledges to come to the assistance of the recipient if some third party threatens or attacks it. The strategy of reassurance is a type of negative assurance. Negative assurances involve pledges by the sender not to itself threaten or attack the recipient. Non-nuclear weapon states have sought both positive and negative security assurances, but, unfortunately, there can be tension between the steps required

20

JEFFREY W. KNOPF

for each. 24 Positive assurances have been strongly associated with the idea of a nuclear umbrella. The ability to extend nuclear deterrence over friends and allies implies a need to maintain an adequate-size nuclear arsenal and a posture of readiness to use those weapons. This same posture can be seen as contrary to the spirit of negative assurances. Promises not to threaten or use nuclear weapons become more convincing if nuclear weapon states signal an interest in reducing their reliance on nuclear weapons, for example by cutting their stockpiles or restricting the missions and roles assigned to those weapons. Arguments that positive security guarantees require maintaining robust nuclear capabilities could make negative assurances appear insincere. This raises an important question: can such trade-offs be avoided or at least minimized? This might be possible if positive assurances did not require extending a nuclear security guarantee. It is worth exploring whether some combination of conventional military measures, provision of missile defenses, and pledges of nonmilitary assistance would make for an effective positive assurance. In practice, the cases of positive assurances in this volume involve provision of a nuclear umbrella, so there is not much of an evidentiary base for judging whether non-nuclear alternatives could be a workable substitute. The form in which assurances are delivered might also make a difference. Existing research has mostly focused on bilateral assurances. A defense pact between a nuclear and a non-nuclear state has been taken to imply a positive security guarantee. This is not the only form in which assurances might be offered, however; beyond bilateral assurances, three other basic formats are possible. If a bilateral pledge involves a one-to-one arrangement, then other alternatives include one-to-all, all-to-one, and all-to-all arrangements (either globally or within a region). A single state might offer a generalized assurance that applies to all other states. The negative assurances provided by the nuclear weapon states have basically taken this form, with some caveats. There can also be circumstances in which all the nuclear weapon states, or some other multilateral coalition or even the international community as whole, makes a pledge directed at one individual state. The pledges made to Ukraine after the breakup of the Soviet Union are a possible example of a "many-to-one" assurance. Finally, there could be assurances that are generalized in terms of both senders and recipients. For example, a UN resolution that required all states, nuclear and non-nuclear, to offer economic, medical, and other assistance to any state that is the victim of a nuclear attack would be a globalized form of positive assurance. 25 These four possible formats for assurances will be referred to as

SECURITY ASSURANCES: INITIAL HYPOTHESES

21

bilateral, generalized individual, focused multilateral, and global (or universal) assurances. It might also be worth considering regional arrangements to be a fifth format, as some existing regional nuclear-weapon-free zones (NWFZs) have incorporated negative assurances from the nuclear weapon states. 26 This raises the obvious question of whether the format in which an assurance is provided affects its impact. For example, if North Korea can be influenced, is it more likely to respond to a focused multilateral assurance provided through the mechanism of six-party talks, or is it only going to be satisfied by a bilateral deal directly with the United States? Note that this question assumes that North Korea will require some form of country-specific assurance beyond the generalized assurances associated with the NPT. In practice, most of the cases in this volume focus on such "extra" assurances for a recipient in addition to those available through the nonproliferation regime. PRELIMINARY HYPOTHESES

A major goal of this project is to develop contingent generalizations about the circumstances under which security assurances are more or less likely to be effective. Because there has not previously been much effort to develop an empirical theory of assurances, the task of identifying generalizations will be based in large part on inductive methods. Participants in this project have examined several individual cases, with the idea that by comparing those cases some generalizations would emerge. To assist in the investigation of the cases, some preliminary hypotheses to be considered by the case study authors were made available beforehand. This section describes these initial hypotheses, which were derived from other bodies of theory relevant to thinking about security assurances. The two most relevant literatures are those on deterrence (and to a lesser extent coercive diplomacy and reassurance) and on the causes of nuclear proliferation and restraint. The hypotheses are mostly framed in terms of the effectiveness of assurances, and some clarification about how effectiveness is assessed is in order. For purposes of this study, the outcome of interest-or dependent variable-is treated as a simple dichotomy: nonproliferation success or failure, as indicated by the presence or absence of nuclear weapons or a nuclear weapon program (and recognizing the potential for uncertainty in some cases). A nonproliferation outcome can result from reversal, when a country gives up a nuclear arsenal it has acquired; abandonment, when a country shuts down a nuclear weapons development effort it has begun but not completed; or renunciation, when

22

JEFFREY W. KNOPF

a country considers but decides against initiating a nuclear weapon program. There are cases in this volume within all three categories: Ukraine is a case of reversal (with the caveat that it never had operational control of former-Soviet weapons on its soil); Libya, South Korea, and Sweden are cases of abandonment; and Japan is a case of renunciation. The two cases considered to be nonproliferation failures also vary: North Korea has tested nuclear weapons, while it remains unclear at the time this volume is going to press whether Iran has made a decision to build nuclear weapons or is still keeping its options open. In the case study chapters, the authors assess effectiveness by investigating the empirical record for evidence on whether or not the presence or absence of assurances contributed to these outcomes. Many of the hypotheses can be related to two overarching factors. Several address the credibility of assurances, focusing on factors that might make assurances more or less credible. Most of the others focus on factors associated with the recipient, ranging from its strategic interests to its internal decisionmaking processes. One rather obvious hypothesis leaps out from the literatures on both deterrence and proliferation and is presented first, then additional hypotheses are derived separately from discussions of deterrence and proliferation, respectively.

Hypothesis 1: Assurances are more likely to be effective when a target state's interest in nuclear weapons is driven to a significant degree by security concerns. This hypothesis is basically common sense. Efforts to assure a state about its security will be most relevant when a state is in fact concerned about security. This hypothesis also follows from the primary condition that affects whether reassurance or deterrence is the most appropriate strategy. Reassurance strategies, and by implication assurances in general, are most likely to be effective when dealing with an insecure state. With respect to a state with offensive motivations, assurances may be beside the point or even send a dangerous signal of weakness. The proliferation literature also suggests that this is a meaningful hypothesis. Traditionally, security concerns have been the primary explanation given for decisions to develop nuclear weapons. Increasingly, however, the literature has identified other explanations for proliferation, ranging from concerns about prestige to a variety of internal factors. If a particular proliferation decision is not motivated at least partly by security concerns, then assurances are less likely to be effective. They may not be guaranteed to fail, as other hypotheses introduced below suggest circumstances under which assurances

SECURITY ASSURANCES: INITIAL HYPOTHESES

23

might have influence even when a state's leaders are not driven by security concerns, but other things being equal they are less likely to succeed. It is also possible that the relationship might be one that would be graphed as an inverted U -shape. Assurances might be effective when applied to states with moderate levels of insecurity, but for states that perceive extreme levels of danger there may be no outside security guarantee strong enough to dissuade them from seeking their own nuclear deterrent. Hypotheses from Deterrence Theory

Factors that make deterrence work or fail have been studied extensively. Because deterrence and assurance are both influence strategies-and because some forms of positive assurance rely upon extended deterrence-some findings about deterrence might, with appropriate adaptation, also apply to assurances. There is not a unified theory of deterrence, but rather there are three different schools of thought. Rational deterrence theory CRDT), derived from neorealist theories of international relations, is the mainstream approach. Proponents of a strategic culture approach have rejected the assumption of a generic or universal rational actor, leading them to recommend a strategy that in recent U.S. policy has been labeled "tailored deterrence."27 Finally, a decisionmaking approach, favored by many of those who propose reassurance as an alternative to deterrence, stresses domestic, organizational, and psychological constraints on rationality. This chapter does not take a position on which approach is right, but rather mines all three for potentially applicable hypotheses. The primary emphasis in RDT is on credibility.28 It has long been recognized that extended deterrence poses special challenges because threats to act on behalf of another party are not as inherently credible as threats to act in self-defense. Yet making extended deterrence convincing to an adversary may not be as difficult as making the commitment credible to one's allies. During the Cold War, former British Defense Minister Denis Healey liked to comment that "it takes only five percent credibility of American retaliation to deter the Russians, but ninety-five percent credibility to reassure the Europeans:'29 The challenges may be greater still for negative assurances. Positive assurances are often extended to friends and allies, and the possibility of common threats creates an interest in upholding positive assurances. Negative assurances, in contrast, are more frequently focused on nonaligned parties or even adversaries, and the latter may be especially unlikely to believe a promise not to threaten them. Given

24

JEFFREY W. KNOPF

the obvious reasons for wondering whether assurances will be seen as credible, hypotheses 2 through 6 deal with different factors that might bolster credibility.

Hypothesis 2: Public declarations of security assurances increase their effectiveness. It is possible to communicate assurances through unpublicized diplomatic channels, but announcing assurances publicly might make them more credible. A public declaration commits a state in the eyes of other states and its own public. Failure to follow through could damage the state's reputation and its government's standing with domestic constituencies. The audience costs associated with public declarations have been found to make deterrent threats more effective,30 and the same may be true with respect to assurances. Hypothesis 3: Legally binding mechanisms will increase the effectiveness of assurances. This hypothesis is suggested by the history of policy debates about NPT-related security assurances. Non-nuclear states have often sought legally binding assurances, which suggests that they would find such assurances more credible. There are two types of legally binding commitments that could be explored. First, as suggested by existing research on proliferation, a formal defense treaty might make bilateral positive assurances more effective relative to a mere verbal declaration. Second, an international treaty or the equivalent binding the nuclear weapon states collectively to either negative or positive assurances might also be more effective than purely verbal declarations. Because no such global treaty exists, this hypothesis might be a possible explanation if negative assurances are found to be of limited effectiveness. As noted above, there are some regional nuclear-weapon-free zone treaties in which the NPT nuclear states have formally obligated themselves to negative assurances. This suggests it would be worthwhile to explore whether states in regions covered by these NWFZs have been more likely than other states to abjure nuclear weapons. Among the factors that contribute to the credibility of deterrence, resolve is often seen as the key. In RDT, estimates of resolve have been connected to three factors: intrinsic interests at stake, a state's reputation based on past behavior, and the use of commitment tactics (such as Herman Kahn's famous example of throwing your car's steering wheel out the window in a game of chicken).31 All three factors suggest possible hypotheses relevant to assurances.

SECURITY ASSURANCES: INITIAL HYPOTHESES

25

Hypothesis 4: The greater the political and economic ties between sender and recipient, the more effective security assurances will be. Statistical studies of extended deterrence have found that political and economic connections between a deterrer and its protege are a good indicator of the interests at stake for the deterring state. The greater these ties, the more likely extended deterrence is to succeed. 32 Because extended deterrence is a major element of positive assurances, the same finding is likely to apply to security assurances. Importantly, this hypothesis does not rest on the threat of nuclear retaliation per se, but rather on the fact an ally is so important to the provider of assurances that the latter is likely to come to its ally's aid, whether this be through nuclear or conventional means. Hypothesis 5: A reputation for keeping past nonproliferation or alliance commitments will increase the effectiveness of assurances. The importance of a reputation for resolve based on a state's past behavior has been an issue of considerable debate in deterrence research. 33 To the extent such a reputation matters, it suggests that a record of making good on alliance commitments will increase the credibility of positive security assurances. A different kind of reputation, for honesty in one's diplomatic dealings, has been found to be important by Anne Sartori.34 This suggests that keeping one's promises to cooperate is at least as important as following through on one's threats to punish those who challenge deterrent commitments. If this is true, then the nuclear weapon states' record of compliance with other nonproliferation obligations is likely to affect the credibility of negative security assurances. A final hypothesis regarding credibility falls in the realm of commitment tactics. In the Cold War, once the Soviet Union achieved the ability to strike the U.S. homeland, this led some to question the credibility of U.S. extended deterrence. For example, de Gaulle is widely reported to have questioned (though I cannot locate a citation for this) whether the United States would be willing to trade New York for Paris. Measures that have been taken previously to make extended deterrence commitments more credible might also be relevant to security assurances.

Hypothesis 6: Forward-deployed troops and other forms of defense cooperation will increase the effectiveness of security assurances. Forward-deployed troops were the primary commitment tactic employed by the United States during the Cold War. U.S. troops stationed in West Ger-

26

JEFFREY W. KNOPF

many and South Korea were intended to function as a trip-wire that, by requiring an invading army to engage U.S. personnel, would place great pressure on the United States to respond. If the recipient of positive security assurances is willing to accept a nuclear state's troops on its soil, such forward deployments might serve as a signal that makes such assurances more credible. After the initial draft of these hypotheses had been completed, several new studies appeared dealing with the assurance element of extended nuclear deterrence. These called attention, in addition to forward-deployed troops, to the value of consultation mechanisms and other types of defense cooperation. 35 Hence, hypothesis 6 also suggests that other forms of defense cooperation, such as joint planning, joint training, arms transfers, and opportunities for consultation, might also increase the credibility of assurances by making commitments more tangible. George Bunn has argued that, rather than the legal language of the North Atlantic Treaty, it was these tangible security relationships that convinced NATO members Germany and Italy to sign the NPT as non-nuclear states. 36 The most contentious policy question raised by this hypothesis is whether forward-deployed nuclear weapons are necessary to make positive assurances credible. Logically, nuclear weapons stationed in the United States or another nuclear weapon state should be usable for extending deterrence to allies, because a retaliatory strike could be launched from a nuclear state's soil or submarines at sea as easily as from allied territory. Perceptions are what matter here, however, and if states believe that nuclear weapons on their soil or deployed nearby are necessary, then positive assurances will not be as effective in the absence of such deployments. The preceding several hypotheses focused on the credibility of assurances; in most of the remaining hypotheses the focus shifts to factors that might affect the receptivity of the recipient. Credibility has been the primary concern in rational theory, but a second tradition of thinking about deterrence is critical of the generic rational-actor model used in RDT. It argues that countries with different histories, political systems, and strategic cultures will have different value systems, meaning that a threat that would deter the United States might not deter other countries that prioritize different values. 37 To make deterrence effective, this tradition holds, it is important to ascertain and hold at risk what the other side's leaders value most. This suggests an analogous hypothesis regarding assurances.

SECURITY ASSURANCES: INITIAL HYPOTHESES

27

Hypothesis 7: Assurances will be most effective if they are tailored to take account of unique features of the target state's culture, decision-making procedures, and leadership concerns. In short, just as deterrence can be tailored to individual cases, so too can assurances. As Robert Einhorn argues, although general theories have value, proliferation takes place "in specific-in specific countries, in specific international and domestic circumstances, and with specific persons and organizations making discrete decisions:' Hence, policies to prevent nuclear weapons development must "focus on individual countries:'38 Reflecting this perspective, two recent monographs on using extended deterrence to assure allies advocate that such "assurance must be tailored to the strategic culture, threat perceptions, values, and specific concerns of each ally:'39 If this hypothesis is correct, it implies that bilateral and focused multilateral assurances are more likely to be effective than generalized and global assurances. A decision-making approach, which rather than strategic culture emphasizes domestic, organizational, and psychological constraints on rationality, suggests two further hypotheses.

Hypothesis 8: Assurances will have to be strong enough to overcome cognitive biases in order to be effective. Embedding them in a larger strategy may be one way to do this. This is the main hypothesis suggested by decision-making research on deterrence and reassurance. Research in psychology shows that once individuals form an image of another actor, they tend to look for confirming information and discount potentially disconfirming evidence. 4o It is thus hard-but not impossible-to change an established image. To do so requires going beyond what a rational-actor analysis would imply is necessary. Repeated actions, or actions so large and surprising they cannot be ignored, may be needed to overcome cognitive biasesY One way to do this may be to supplement negative NPT assurances with a broader strategy of reassurance, or to combine positive nuclear assurances with a broader alliance-related assurance strategy. Taking additional steps that fall outside the realm of nuclear policy may be a way to challenge a target's skepticism about purely nuclear-related assurances. Hypothesis 9: Assurances can fail if domestic or alliance constraints limit the ability of a sender or senders to fully implement an assurance strategy. Most of the hypotheses presented here concern factors that might affect a

28

JEFFREY W. KNOPF

target state's response to assurances. However, the ability of the sender(s) to follow through on an intent to offer assurances can also affect the likelihood of success. Research has shown that domestic constraints can undercut the use of positive incentives, and Lebow has suggested that a leader's ability to overcome domestic constraints is likely to be important in efforts to use reassurance as wellY A recent dissertation by Jungsoo Kim has shown that objections by allies can similarly prevent states from fully implementing a reassurance strategyY This all suggests a need to pay attention to possible domestic and alliance constraints on the senders of assurances. Hypotheses from Proliferation Theory

To ascertain the effectiveness of security assurances as a nonproliferation tool, it is important to think about how assurances relate to the general factors that lead states to seek or renounce nuclear weapons. For this reason, research on the causes of proliferation and restraint represents the most directly relevant literature for identifying possible hypotheses about assurance. Security threats have long served as the standard explanation for proliferation. Observers expect states facing an existential threat, arising either from a nuclear-armed rival or from adversaries with significant conventional superiority, to be interested in nuclear weapons. This perspective is reflected in the hypotheses introduced above. To the extent proliferation is driven at least partly by security concerns, as hypothesis 1 notes, assurances become relevant. But given the reasons why states may not want to count on others for their security, additional factors such as those identified in the other hypotheses may be important in determining whether or not a given security assurance is deemed credible. If nonsecurity considerations are at work, using assurances becomes more difficult, but not necessarily impossible. A lot of recent proliferation literature has questioned the adequacy of the security explanation and called attention to other potentially relevant variables. From fairly early in the nuclear age, it was recognized that prestige and the desire for great power status could be important considerations for some states. Since then, some theorists drawing on social constructivism have argued that the norms associated with nuclear weapons can be changed. If the nonproliferation regime successfully delegitimizes nuclear weapons, then pursuing them becomes a symbol of rogue or outlaw status, not of prestige, and states concerned about their standing in the international community will reject nuclear programs. 44 Other theorists have turned to internal factors. Nuclear weapons programs

SECURITY ASSURANCES: INITIAL HYPOTHESES

29

may serve the parochial interests of some politicians, military actors, or nuclear scientists. If these actors become a sufficiently powerful domestic lobby for the bomb, developments in a state's external environment may become less relevant to its decisions about nuclear acquisition. Etel Solingen has advanced a more general coalitional theory of proliferation, based on how a state's ruling coalition relates to the international economy.45 Outward-looking, liberalizing coalitions will renounce nuclear weapons, she argues, fearing that a bomb program will lead their state to be cut off from the hoped-for benefits of participating in globalization. In contrast, inward-looking coalitions, especially if motivated by nationalism or religious fundamentalism, will be interested in nuclear weapons as a symbol of defiance and national pride. Jacques Hymans has put forward another theory that makes similar predictions, but based at the individual rather than domestic level of analysis. 46 Hymans suggests that leaders who are highly nationalistic and perceive external opposition to their state will promote bomb programs, while leaders who lack this combination of pride and fear will not be interested in nuclear weapons. Taken as a whole, the literature on proliferation suggests several additional hypotheses beyond those presented above.

Hypothesis 10: States with regional security concerns will be most interested in positive security assurances; states without such concerns will be most interested in negative assurances. T. V. Paul has sought to integrate realist and liberal theories of proliferation. On the realist side, he confirms that security is still a good explanation for many cases, but concludes that the focus has to be on the regional environment. 47 Most cases of proliferation, he believes, have derived from regional threats, not from concerns about the two Cold War superpowers. If this is true, countries in dangerous regional neighborhoods represent the cases in which positive assurances are most likely to be effective. This suggests that the role of positive assurances will merit special scrutiny in regions like Northeast Asia and the Middle East. The logical flip side of this analysis is that negative assurances will be more relevant in other cases. This includes countries that view an NPT nuclear weapon state as a direct threat. But it also includes countries that are not directly threatened. States that want to remain nonaligned and states that hope to keep their region nuclear free will be most concerned about nuclear weapon state behavior that could destabilize the situation in their region. Efforts by nu-

30

JEFFREY W. KNOPF

clear-armed countries to use nuclear threats against another state in the region or to deploy nuclear weapons elsewhere in the region might create pressures for proliferation in the region or force nonaligned states to choose a side. Such states will press for negative security assurances as a way to reduce the chances of such a scenario developing.

Hypothesis 11: The impact of assurances will depend in part on how they affect the perceived prestige and appropriateness of nuclear status. To the extent that norms matter, negative assurances are likely to be especially important. To the extent that states care about how they are perceived by others, norms associated with nuclear weapons could influence their behavior. Anything that increases the prestige or status associated with nuclear weapons, or that makes such weapons appear a normal and appropriate means of pursuing national security goals, will lead states to view proliferation as more attractive or at least acceptable. Anything that lowers the prestige, perceived utility, or legitimacy of nuclear weapons reinforces nonproliferation norms. Positive security assurances, when in the form of extended nuclear deterrence, imply that nuclear weapons have continued value. They hence tend not to bolster norms against nuclear weapons acquisition (though they might not weaken them). Negative assurances, because they imply that nuclear weapons should playa strictly limited role, are more easily compatible with efforts to delegitimize nuclear weapons. To the extent state decisions are influenced by prevailing norms, negative assurances are likely to be more effective than positive assurances in strengthening nonproliferation norms and encouraging restraint. Hypothesis 12: Assurances will be more effective if they can be utilized in a way that alters internal debates in the target in a favorable direction. At first glance, work that highlights domestic determinants of proliferation would seem to lead to pessimistic inferences about assurances. If nuclear programs are driven by internal factors, then the actions of external actors might make no difference to calculations about whether or not to continue those programs. But this depends on whether there is still internal debate about a country's nuclear program. Prior research on how international factors affect domestic politics yields certain rules of thumb for how to influence internal debates about foreign policy.48 An influence strategy should aim to strengthen or at least not undermine the position of moderates, which in this case means those who favor participation in the nonproliferation regime and oppose indigenous nuclear weapons development. At the same time, the strategy should

SECURITY ASSURANCES: INITIAL HYPOTHESES

31

aim to weaken or at least not bolster the position of hard-liners, which in this case means advocates of a nuclear weapons program. Security assurances might be crafted to do this, in three ways. First, a well-timed statement of negative assurances can be used to bolster the position of moderates. Even if the underlying reasons why a country is pursuing nuclear weapons are found in the parochial interests or belief systems of certain domestic actors, the advocates of nuclear weapons development are likely to justify the program in domestic debates by arguing that it is needed for security against external threats.49 A forceful, public commitment to negative security assurances by the state or states being portrayed as the source of a security threat can undermine the main rationale for a nuclear program being offered by its advocates, which might tilt the balance in domestic debates in favor of nonproliferation. Positive assurances offer a second possible mechanism for influencing internal debates. As noted above, Rublee contends that security guarantees can affect domestic debates when important actors in the domestic arena have already embraced nonproliferation norms. In these circumstances, positive assurances can satisfy actors who previously favored nuclear weapons development, making them willing to go along with the preference of the rest of domestic opinion to renounce the nuclear weapons option. If this is correct, it suggests that, contrary to the previous hypothesis, norms may operate in conjunction with positive rather than negative assurances in certain cases with nonproliferation outcomes. Third, a final possible use of positive assurances is indirect. If a nuclear weapon state is being depicted as the source of threat, an offer of positive security guarantees by that country is unlikely to make much difference in internal debates in the target state. If one's state is viewed as the enemy, an offer to help the other state if it is attacked by some third party is largely beside the point. But positive assurances might make a difference if they can be used in a way that undermines the arguments of hard-liners in the target state that nuclear weapons development will improve the state's security. It might be possible to do this indirectly, by offering positive security assurances to the state's regional rivals, or by recommitting to and strengthening such assurances if they already exist. If moderates can claim that their country's nuclear program has led to a deepening of ties and military cooperation among the state's potential adversaries, this might provide them with ammunition in domestic debates to argue against hard-liners' claims that nuclear weapons are good for security.

32

JEFFREY W. KNOPF

TABLE 2.1

Hypotheses on Security Assurances Hypothesis 1: Assurances are more likely to be effective when a target state's interest in nuclear weapons is driven to a significant degree by security concerns.

Hypothesis 2: Public declarations of security assurances increase their effectiveness.

Hypothesis 3: Legally binding mechanisms will increase the effectiveness of assurances.

Hypothesis 4: The greater the political and economic ties between sender and recipient, the more effective security assurances will be.

Hypothesis 5: A reputation for keeping past nonproliferation or alliance commitments will increase the effectiveness of assurances.

Hypothesis 6: Forward-deployed troops and other forms of defense cooperation will increase the effectiveness of security assurances.

Hypothesis 7: Assurances will be most effective if they are tailored to take account of unique features of the target state's culture, decision-making procedures, and leadership concerns.

Hypothesis 8: Assurances will have to be strong enough to overcome cognitive biases in order to be effective. Embedding them in a larger strategy may be one way to do this.

Hypothesis 9: Assurances can fail if domestic or alliance constraints limit the ability of a sender or senders to fully implement an assurance strategy.

Hypothesis 10: States with regional security concerns will be most interested in positive security assurances; states without such concerns will be most interested in negative assurances.

Hypothesis 11: The impact of assurances will depend in part on how they affect the perceived prestige and appropriateness of nuclear status. To the extent that norms matter, negative assurances are likely to be especially important.

Hypothesis 12: Assurances will be more effective if they can be utilized in a way that alters internal debates in the target in a favorable direction.

Hypothesis 13: Assurances will be more effective if they are packaged with positive incentives.

This will only be possible, however, if there is scope for domestic debate inside a potential proliferator along with moderates who are capable of seeing that their state's own behavior is provoking the strengthening of ties between external nuclear powers and the state's regional neighbors. Otherwise, efforts to strengthen security guarantees to a state's regional rivals might only reinforce the beliefs of the state's hard-liners that they are surrounded by enemies. Hence, it is important to investigate whether or not there is sufficient scope for internal debate in a target country before trying to influence such debate.

SECURITY ASSURANCES: INITIAL HYPOTHESES

33

Hypothesis 13: Assurances will be more effective if they are packaged with positive incentives.

This hypothesis follows from Solingen's coalitional analysis and the previous hypothesis. If there is a political competition between outward-looking and inward-looking coalitions in the target state, then the goal of an influence strategy should be to strengthen the outward-looking coalition. If Solingen is right, the central issue in the domestic debate will be different views of participation in the global economy, and not the nuclear weapons program per se. It will hence be important to validate the arguments of the outward-looking coalition that openness to the outside world will lead to economic benefits and opportunities. Offers to open markets, provide foreign investment funds, or support membership in multilateral economic institutions would all be valuable incentives to proceed down the path favored by outward-looking forces. Threats to impose economic sanctions would also motivate a liberalizing coalition to turn away from nuclear weapons development, but such threats could also strengthen the inward-looking coalition by validating its claims that the outside world is hostile. The use of positive incentives in combination with security assurances, in contrast, is more likely to create mutually reinforcing reasons for the target state to turn away from nuclear weapons development. This hypothesis has obvious similarities to hypothesis 8, above. Taken together, they suggest that an important determinant of the effectiveness of NPTrelated security assurances may be whether they are integrated appropriately into a larger strategy. The combination of NPT-related assurances, diplomatic efforts at reassurance, and economic engagement may be more potent than any one strategy on its own. The hypotheses that have been introduced in this section are summarized in Table 2.1. CONCLUSIONS

This chapter has canvassed relevant literatures to identify initial hypotheses about assurance strategies that might apply to the nonproliferation realm. The most basic observation is that assurances are most likely to be effective when a target state's potential interest in nuclear weapons is at least partly driven by security concerns. But the mere fact that a target state has security concerns does not guarantee that assurances will be effective. As with any influence strategy, an offer of assurances must be seen as credible by the recipient. This chapter has identified several factors that may affect the credibility of assurances.

34

JEFFREY W. KNOPF

In addition to the credibility of the provider or providers and their assurances, it may also be necessary to take account of the particular circumstances and concerns of individual recipients and tailor assurances accordingly. Other factors worth considering include internal debates in target countries as well as the influence of cognitive and emotional processes on individual decisionmaking. To deal with such factors, it may be helpful not to let NPT-related assurances stand on their own, and instead package them with other, related strategies. Although both positive and negative assurances are likely to be important, there can unfortunately be trade-offs between the two. Research on proliferation and actual U.S. policy have both tended to give the primary emphasis to positive security guarantees, but some of the hypotheses identified here suggest that in some situations negative assurances may turn out to be the more valuable nonproliferation tool. If so, it will be important to look for ways to gain the benefits of positive assurances without undermining negative assurances. With almost no empirical knowledge base to draw on, all these suggestions are necessarily speculative. The most important goal of this project, therefore, is simply to expand our empirical understanding of how security assurances have worked as a nonproliferation tool in the past. NOTES 1. Thomas C. Schelling, Arms and Influence (New Haven: Yale University Press, 1966),74· 2. U.S. Department of Defense, Quadrennial Defense Review Report, September 30, 2001; President of the United States, The National Security Strategy of the United States of America, September 2002. 3. Assurance of allies was previously labeled "reassurance" by Michael Howard ("Reassurance and Deterrence: Western Defense in the 1980s;' Foreign Affairs 61, no. 2 [Winter 1982/83]), but as described below a different meaning of the term "reassurance" has become dominant in the International Relations field; the Bush administration coinage is now more commonly employed when discussing relations with allies. 4. For more, see the sources in Chapter 1, n. 3. 5. Charles E. Osgood, An Alternative to War or Surrender (Urbana: University of Illinois Press, 1962); Amitai Etzioni, The Hard Way to Peace: A New Strategy (New York: Crowell-Collier Press, 1962). 6. Richard Ned Lebow and Janice Gross Stein, "Beyond Deterrence;' Journal of Social Issues 43, no. 4 (Winter 1987); Stein, "Deterrence and Reassurance;' in Behavior, Society and Nuclear War, vol. 2, ed. Philip Tetlock et al. (New York: Oxford University Press,

SECURITY ASSURANCES: INITIAL HYPOTHESES

35

1991); Richard Ned Lebow, "Deterrence and Reassurance: Lessons from the Cold War;'

Global Dialogue 3, no. 4 (Autumn 2001). 7. Robert Jervis, Perception and Misperception in International Politics (Princeton: Princeton University Press, 1976), ch. 3; Charles L. Glaser, Rational Theory of International Politics: The Logic of Competition and Cooperation (Princeton: Princeton University Press, 2010 ). 8. Robert Jervis, "Cooperation under the Security Dilemma;' World Politics 30, no. 2 (January 1978); Glaser, Rational Theory, esp. 64-68. 9. For an effort to integrate realist and psychological approaches to reassurance, see Paul Midford, "The Logic of Reassurance and Japan's Grand Strategy;' Security Studies 11,

no. 3 (Spring 2002). 10. Richard Ned Lebow, "Conclusions;' in Psychology and Deterrence, ed. Robert Jer-

vis, Richard Ned Lebow, and Janice Gross Stein (Baltimore, MD: Johns Hopkins University Press, 1985), 227. 11. Andrew H. Kydd, Trust and Mistrust in International Relations (Princeton: Princeton University Press, 2005). 12.

In particular, the following studies (other than Larson) have debated whether

or not Gorbachev represents an effective use of a reassurance strategy. Deborah Welch Larson, "Crisis Prevention and the Austrian State Treaty;' International Organization 41, no. 1 (Winter 1987); Richard A. Bitzinger, "Gorbachev and GRIT, 1985-1989: Did Arms Control Succeed Because of Unilateral Actions or in Spite of Them?" Contemporary Se-

curity Policy 15, no. 1 (April 1994); Alan R. Collins, "GRIT, Gorbachev and the End of the Cold War;' Review of International Studies 24, no. 2 (April 1998); Janice Gross Stein, "Image, Identity, and the Resolution of Violent Conflict;' in Turbulent Peace: The Challenges of Managing International Conflict, ed. Chester A. Crocker, Fen Osler Hampson, and Pamela Aall (Washington, DC: USIP Press, 2001); Kydd, Trust and Mistrust, ch. 8; Evan Braden Montgomery, "Breaking Out of the Security Dilemma;' International Security 31, no. 2 (Fall 2006). 13. If positive security assurances include more than just extended nuclear deter-

rence, then non-nuclear states could offer certain types of positive assurances as well. 14. See, for example, George Bunn and Jean du Preez, "More than Words: The Value

of U.S. Non-Nuclear-Use Promises;' Arms Control Today 37, no. 6 (July/August 2007). 15. Joseph F. Pilat, "Reassessing Security Assurances in a Unipolar World;' Washing-

ton Quarterly 28, no. 2 (Spring 2005). 16. Bruno Tertrais, "Security Assurances and the Future of Proliferation;' in Over the Horizon Proliferation Threats, ed. James J. Wirtz and Peter R. Lavoy (Stanford: Stanford

University Press, 2012). Tertrais presented a prepublication version of his book chapter at the initial workshop for this project in 2009. I thank him for sharing his research findings with the other participants in this project.

36

JEFFREY W. KNOPF

17. Sonali Singh and Christopher R. Way, "The Correlates of Nuclear Proliferation;'

Journal of Conflict Resolution 48, no. 6 (December 2004); Dong-Joon Jo and Erik Gartzke, "Determinants of Nuclear Weapons Proliferation;' Journal of Conflict Resolution 51, no. 1 (February 2007); Daniel Verdier, "Multilateralism, Bilateralism, and Exclusion in the

Nuclear Proliferation Regime;' International Organization 62, no. 3 (Summer 2008); Matthew Kroenig, "Importing the Bomb: Sensitive Nuclear Assistance and Nuclear Proliferation;' Journal of Conflict Resolution 53, no. 2 (April 2009); Philipp C. Bleek, "Why Do States Proliferate? Quantitative Analysis of the Exploration, Pursuit, and Acquisition of Nuclear Weapons;' in Forecasting Nuclear Proliferation in the 21st Century, vol. 1, ed. William Potter and Gaukhar Mukhatzhanova (Stanford: Stanford University Press, 2010). 18. Matthew Fuhrmann, "Spreading Temptation: Proliferation and Peaceful Nuclear

Cooperation Agreements;' International Security 34, no. 1 (Summer 2009); Harald Muller and Andreas Schmidt, "The Little Known Story of De-Proliferation: Why States Give up Nuclear Weapon Activities;' in Potter and Mukhatzhanova, Forecasting Nuclear Pro-

liferation. 19. For example, see Jacques E. C. Hymans, "Theories of Nuclear Proliferation: The

State of the Field;' Nonproliferation Review 13, no. 3 (November 2006): 456. 20. Etel Solingen, Nuclear Logics: Contrasting Paths in East Asia and the Middle East (Princeton: Princeton University Press, 2007), 12-14, 25-27, 256. 21.

T. V. Paul, Power versus Prudence: Why Nations Forgo Nuclear Weapons (Montreal

and Kingston: McGill-Queen's University Press, 2000),153-54. 22. Jacques E. C. Hymans, The Psychology of Nuclear Proliferation (Cambridge, UK:

Cambridge University Press, 2006), 42-43 n. 79. 23. Maria Rost Rublee, Nonproliferation Norms: Why States Choose Nuclear Restraint (Athens: University of Georgia Press, 2009). 24. Foran has labeled this the "security assurance paradox;' claiming that "in their strongest forms, positive and negative security assurances undermine one another:' Virginia 1. Foran, ed., Security Assurances: Implications for the NPT and Beyond (Washington, DC: Carnegie Endowment for International Peace, 1995), 4. 25. For an argument in favor of the need to move toward such universalized security

assurances, see Rebecca Johnson, "Security Assurances for Everyone: A New Approach to Deterring the Use of Nuclear Weapons;' Disarmament Diplomacy, no. 90 (Spring 2009). Harald Muller ("Between Security Council Inaction and Self-Helplessness: The Case for a Positive Security Assurances Alliance;' in Foran, Security Assurances) earlier pointed out that positive security assurances need not be limited to the nuclear weapon states but could include pledges from non-nuclear states. 26. Leonard S. Spector and Aubrie Ohlde, "Negative Security Assurances: Revisiting

the Nuclear-Weapon-Free-Zone Option;' Arms Control Today 35, no. 3 (April 2005).

SECURITY ASSURANCES: INITIAL HYPOTHESES

37

27. M. Elaine Bunn, "Can Deterrence Be Tailored?" Strategic Forum, no. 225, January 2007, available at www.ndu.edu/inss. 28. William W. Kaufmann, "The Requirements of Deterrence;' in Military Policy and National Security, ed. William W. Kaufmann (Princeton: Princeton University Press, 1956). 29. Denis Healey, The Time ofMy Life (London: Michael Joseph, 1989), 243. 30. James D. Fearon, "Domestic Audience Costs and the Escalation of International

Disputes;' American Political Science Review 88, no. 3 (September 1994). 31. Herman Kahn, On Thermonuclear War (New Brunswick, NJ: Transaction Publishers, 2007; originally published Princeton University Press, 1960), 291; Thinking about the Unthinkable (New York: Horizon Press, 1962), 45; On Escalation: Metaphors and Scenarios (New York: Praeger, 1965), 11. 32. Paul Huth and Bruce Russett, "What Makes Deterrence Work? Cases from 1900 to 1980;' World Politics 36, no. 4 (July 1984). 33. For a recent challenge to the view that a state's past behavior is important for how other states perceive the credibility of its threats, see Daryl G. Press, Calculating Credibility (Ithaca, NY: Cornell University Press, 2007). 34. Anne E. Sartori, Deterrence by Diplomacy (Princeton: Princeton University Press, 2007). 35. David S. Yost, ''Assurance and U.S. Extended Deterrence in NATO;' International Affairs 85, no. 4 (July 2009); ClarkA. Murdock and Jessica M. Yeats, Exploring the Nuclear Posture Implications of Extended Deterrence and Assurance (Washington, DC: Center for

Strategic and International Studies, November 2009); James M. Acton, "Extended Deterrence and Communicating Resolve;' Strategic Insights 8, no. 5 (December 2009). 36. George Bunn, "Security Assurances against Nuclear Attack: The Legal Framework for the NPT Extension Conference and Beyond;' in Foran, Security Assurances, 9-10. 37. See, for example, Keith B. Payne, "The Fallacies of Cold War Deterrence and a New Direction;' and Colin S. Gray, "The Reformation of Deterrence: Moving On;' both in Comparative Strategy 22, no. 5 (December 2003). 38. Robert J. Einhorn, "Identifying Nuclear Aspirants and Their Pathways to the

Bomb;' Nonproliferation Review 13, no. 3 (November 2006): 493, emphasis in original. 39. Murdock and Yeats, Exploring the Nuclear Posture Implications, 24; see also Acton, "Extended Deterrence and Communicating Resolve:' 40. Jervis, Perception and Misperception, ch. 4.

41. Larson, "Crisis Prevention and the Austrian State Treaty"; Stein, "Deterrence and Reassurance:' 42. Richard N. Haass and Meghan L. O'Sullivan, Honey and Vinegar: Incentives,

Sanctions, and Foreign Policy (Washington, DC: Brookings Institution Press, 2000),

38

JEFFREY W. KNOPF

178-81; Richard Ned Lebow, "The Deterrence Deadlock: Is There a Way Out?" in Jervis,

Lebow, and Stein, Psychology and Deterrence, 192-93. 43. Jungsoo Kim, "Reassurance Strategy: Incentives for Use and Conditions for Success;' Ph.D. diss., Naval Postgraduate School, Monterey, CA, March 2010. In particular,

Kim finds, the United States and South Korea have each sometimes prevented the other from more aggressively attempting a reassurance effort with North Korea. 44. Scott D. Sagan, "Why Do States Build Nuclear Weapons? Three Models in Search

of a Bomb;' International Security 21, no. 3 (Winter 1996/97); Rublee, Nonproliferation

Norms. 45. Solingen, Nuclear Logics. 46. Hymans, The Psychology of Nuclear Proliferation. 47. Paul, Power versus Prudence. 48. Graham T. Allison and Morton H. Halperin, "Bureaucratic Politics: A Paradigm

and Some Policy Implications;' World Politics 24, supplement (Spring 1972); Robert D. Putnam, "Diplomacy and Domestic Politics: The Logic of Two-Level Games;' Interna-

tional Organization 42, no. 3 (Summer 1988); Jack Snyder, "International Leverage on Soviet Domestic Change;' World Politics 42, no. 1 (October 1989). 49. Peter R. Lavoy, "Nuclear Myths and the Causes of Nuclear Proliferation;' Security

Studies 2, nos. 3/4 (Sept. 1993).

The Psychology of Assurance: An Emotional Tale Janice Gross Stein

Strategies of assurance in international politics are designed to deal with frightened and distrustful adversaries and allies. These strategies are also useful against aggressive or "greedy" adversaries who must be assured that if they heed deterrent warnings, they will not be attacked.! If assurance is to succeed, allies and adversaries must be reasonably confident that a status quo that makes them feel secure is likely to endure, that their security will not be jeopardized in the foreseeable future. This description makes clear the inherently psychological nature of assurance. 2 "Insecurity;' "trust;' "credibility;' "greed;' "fear;' "confidence;' and "risk" are psychological concepts that are central to the analysis of when assurance is appropriate, when it is likely to succeed, and what forms it should take. These are not objective properties of the environment but a function of how an ally or adversary feels about and understands reality. Early research looked at strategies of assurance as a way to overcome cognitive barriers to changing deeply embedded and resistant images of an adversary. That scholarship, while still relevant, has been supplemented by new research within psychology that speaks directly to the dilemmas of assurance in international politics. That new research supports earlier arguments that assurance can be a useful strategy, but shows the need to ground any theory of assurance on a much broader psychological foundation. This research also points to additional obstacles that can complicate the use of assurance. This chapter addresses only three of several strands of research in psychology that speak to dilemmas of assurance in a risky and uncertain environment. 39

40

JANICE GROSS STEIN

The first is prospect theory, which has changed our understanding of risk propensity. The second is research done in the last two decades that emphasizes the primacy of emotions as adaptive programs of action. The third speaks explicitly to the emotional component of risk, to risk as feeling. The final section of this chapter suggests how psychological concepts can help in formulating strategies of assurance. PROSPECT THEORY AND RISK

People have a preference for simplicity, are averse to ambiguity, and misunderstand probability.3 People are not intuitive probability thinkers. Their estimates depart systematically from what objective probability calculations would dictate. "Human performance suffers;' argues Tetlock, "because we are, deep down, deterministic thinkers with an aversion to probabilistic strategies that accept the inevitability of error:'4 To make matters worse, likely states of the world are difficult to estimate because we do not have repeated trials with large numbers in world politics. Foreign policy decision-makers and analysts generally do not live in a world of risk, where the likelihoods of alternative outcomes are known, but in a world of uncertainty, a world of unknown unknowns. In this world, leaders must develop their own rough estimates of risk. Prospect theory has shown that people also have risk profiles that depart from what models of rational choice would expect. We are far more averse to loss than we are gain-seeking. Foreign policy decision-makers, like people generally, are not neutral about risk. Cognitive psychology has generated robust evidence that loss is more painful than comparable gain is pleasant and that people prefer an immediate smaller gain rather than taking a chance on a larger longer-term reward. 5 People systematically overvalue losses relative to comparable gains. These propositions about risk have held up across a wide variety of cultures and situations. The impact of loss aversion on foreign policy decisions is considerable. 6 Leaders tend to be risk averse in the domain of gain and risk-acceptant in the domain of loss, such as when they face a crisis in which they are likely to lose or have lost something that matters to them. Leaders are also likely to take greater risk to protect what they already have-the "endowment effect"-than to increase their gains. They are also likely to take greater risk to reverse losses, to recapture what they once held, than they would to make new gains. And when decision-makers suffer a significant loss, they are far slower to accommodate to these losses than they are to incorporate gains. Finally, leaders reverse their

THE PSYCHOLOGY OF ASSURANCE: AN EMOTIONAL TALE

41

preferences and make different choices when problems are reframed as losses rather than gains. These general findings apply directly to foreign policy choices. The work I did earlier on Egyptian decision-making distinguished between "need" and "opportunity" to assess the likely success of assurance. This distinction hinges on the motivation of leaders, which is often opaque and difficult to assess. The distinction is more easily captured by prospect theory and its emphasis on a reference point which is often, but not always, the status quo. After the war in 1967, the status quo was not acceptable to President Sadat; his reference point was Egyptian control of the Sinai. He never "normalized" for the loss of the Sinai to Israel. Even though Israel had an obvious advantage in military capabilities, Sadat was undeterred. In a domain of loss, he designed around Israel's military strengths and launched a war in 1973. 7 Under these kinds of conditions, no strategy of assurance was likely to be effective, nor could any security regime be built, until Egypt regained the Sinai and Sadat shifted his reference point back to the status quo. Prospect theory helps in delimiting the scope conditions of strategies of assurance. Prospect theory implies much more, however, than tlIe argument that the propensity to take risk is a function of how people think about where they find themselves in relation to a reference point. It is also about feelings. People feel more pain from losses than they feel pleasure from equivalent gains. 8 It is this asymmetry in feeling that underlies loss aversion. Fear is such a powerful emotional experience in part because the pain of loss is commensurately greater than the pleasure of equivalent gain. Prospect theory suggests that one function of assurance can be to reduce tlIe chances tlIat a state frames its situation within the domain of loss. THE PRIMACY OF EMOTION

Cognitive psychologists and behavioral economists, despite their evidencebased critique of models of microeconomic rationality, generally move only one degree away from the fundamental assumption of utility-maximizing rationality. They continue to set rationality as the default and then explore the consequences of systematic "errors" and "deviations," of "constrained" or "bounded" rationality. These "deviations" only make sense against a background of a narrowly conceived concept of rationality as a calculation based on objective (or as objective as possible) estimates of probability and value. Two decades of research in neuroscience are reshaping our understanding of

42

JANICE GROSS STEIN

the relationship between emotion and cognition, and challenging this concept of rationality. Emotions are important because it is they that give value. The terms "emotion:' "affect:' "feelings:' and "moods" are not identical. I use "emotion" interchangeably with "affect" as an umbrella term to include the experience that is rooted in physiological changes in the body and the awareness of that experience. I may experience fear, for example, as a pulsating heartbeat and sweating with no conscious awareness yet that I am frightened, much less what is frightening me. A "feeling" refers to the conscious awareness that I am afraid of something specific, while "mood" generally refers to a more diffuse and unfocused experience. 9 Neuropsychology now rejects a separation between cognition and emotion as untenable. The one is embedded within the other. And by extension, rationality and emotion are interdependent, not opposite to one another. Emotion is necessary, though not sufficient, for rationality.lO Emotions create part of the yardstick for measuring the costs and benefits that are central to rational choice. In seminal research, Damasio demonstrated that patients who sustained injuries to those parts of the brain that are central to the processing of emotion were incapable of making rational decisions. ll Elliott, his patient, suffered a brain injury in that part of the brain that controls emotions, but he was then, to his doctor's astonishment, unable to distinguish among important and unimportant cues and make rational decisions. Damasio's work ignited a research program on the relationship between cognition and emotion, a program that confirms that behavior is strongly influenced by finely tuned affective systems. 12 "When these systems are damaged or perturbed by brain injury, stress, imbalance in neurotransmitters, or the 'heat of the moment: the logical-deliberative system-even if completely intact, cannot regulate behavior appropriately:'13 Rationality, in short, presupposes and indeed requires emotion. New evidence of the sizable role of automatic processes has produced a fundamental difference in the way evolutionary psychologists and economists think about preferences and behavior. As economist Colin Camerer and his colleagues put it: As economists, we are used to thinking of preferences as the starting point for human behavior and behavior as the ending point. A neuroscience perspective, in contrast, views explicit behavior as only one of many mechanisms that the brain uses to maintain homeostasis and conceives of preferences as transient state variables that ensure survival and reproduction. The traditional economic account of

THE PSYCHOLOGY OF ASSURANCE: AN EMOTIONAL TALE

43

behavior, which assumes that humans act so as to maximally satisfy their preferences, starts in the middle (or perhaps even toward the end) of the neuroscience account.!4

There is growing consensus that emotion is "first;' because it is automatic and fast, and that it plays a dominant role in shaping behavior. We know now that emotion operates in part below the threshold of conscious awareness. IS Contrary to conventional wisdom, we generally feel before we think and, what is even more surprising, often we act before we think. Not surprisingly, the conscious brain then interprets behavior that emerges from automatic, affective processes as the outcome of cognitive deliberations. 16 The human brain affectively tags virtually all objects and concepts and these emotional tags come to mind automatically when these objects and concepts are evoked. 17 People trust their immediate emotional reactions and only correct them through a comparatively laborious cognitive process after the fact. 18 In the well-known "ultimatum game;' one party proposes to a second party how to divide a fixed asset, with the proviso that both will get the proposed shares only if the second side agrees to the division. In the nuclear realm in world politics, for example, a large state might offer a smaller ally only a small part of the available assets-nuclear technology for peaceful purposes or limited access to trade in exchange for forgoing nuclear weapons. Ifleaders of the ally accept what they consider an "unfair" offer, they walk away with something, but if they reject the offer, neither side gets anything. Models of microeconomic rationality would dictate that the other party accept whatever it is offered, because "something is better than nothing" in a one-play game. But participants in multiple experiments do not play that way; they overwhelmingly reject offers that are much below their estimate of 50 percent of the total value. In a series of games where subjects agreed to functional magnetic resonance imaging (tMRIs), the insula cortex that encodes pain and odor disgust was activated when subjects received an unfair offerY That emotional response likely preceded conscious calculation and the decision to reject the offer. This strongly embedded, automatic, emotional reaction to perceived unfairness clearly has consequences for strategies toward would-be proliferators. What is framed as unfair matters. If the existing nonproliferation bargain is viewed as unfair, nuclear restraint becomes less likely, even if it would objectively be in a state's interests. Under these circumstances, one goal of assurance might be to try to establish a greater sense of fairness in the relationship between nuclear haves and have-nots by increasing the obligations of the former toward the latter.

44 JANICE GROSS STEIN

All emotions are either positive or negative and many carry action tendencies. 2o Emotions are adaptive programs of action that have evolved over time to ensure survival and reproduction. A useful way of thinking about emotion and cognition is to see affective processes as those that address the go/no-go questions, the questions that motivate approach-avoidance, while cognitive processes are those that answer true/false questions. 21 Choice, including the choice by leaders whether or not to accept assurances, clearly invokes both kinds of processes. Establishing truth claims about states of the world is usually not enough for people to make a choice. What matters to me, what I value, is an emotional as well as cognitive issue, and is important in what I decide to do, whether I go or don't. Whether or not I am treated fairly is an emotional as well as a cognitive judgment and, in this sense, emotion carries utility. The new field of"neuro-economics" is beginning to conceive of utility as something that is experienced. 22 In this sense of emotion as lived experience, it is irrational to urge people to take the emotion out of their decision-making. EMOTION AND EMOTIONS

Research has tended to examine emotion or affect as a generalized phenomenon in decision-making. Emotions, however, need to be disaggregated. The five core emotions discussed below-fear, anger, sadness, happiness, and disgust-each have a traceable and distinct impact on decision-makingY The experimental literature on the impact of discrete emotions is small but promising. A series of experiments assessed the impact of disgust and sadness, two negative emotions, on financial decisions. Disgust, a physiological experience, evokes the feeling of being too close to an indigestible object or idea and an implicit action tendency to expel current objects and avoid taking in anything new. 24 What impact, researchers wanted to know, would disgust have on buying and selling prices? Experimenters suggested that it would reduce selling prices among people who owned the object (an "expel" goal) and buying prices among participants who did not (an "avoid taking anything in" goal). The impact of disgust-in an experimental setting-was precisely as predicted. It was so strong, in fact, that it wiped out the traditional endowment effect. The dynamics of sadness are different. Sadness arises from loss and helplessness and evokes the goal of changing one's circumstances. Accordingly, sadness should reduce selling prices, but increase buying prices (as in "retail therapy"), since both buying and selling represent an opportunity for change. The experiments confirmed the hypothesized impact of sadness.

THE PSYCHOLOGY OF ASSURANCE: AN EMOTIONAL TALE

45

These experiments were performed in an unusually demanding context. They were designed to assess the "carryover effects" of specific emotions-the extent to which emotions elicited in one situation become an implicit lens for subsequent situations. 25 The strong results suggest first that different emotions have different consequences. Emotions of the same valence-both negativecan have opposing effects. Negative moods do not always suppress change, nor do they inhibit choice. Second, these results suggest that emotions can have a powerful impact on choices. Of the five core emotions, the impact of fear is the most widely studied. Fear has long been at the center of realist analyses, especially those that emphasize the security dilemma. Much of the analysis, however, takes place within a framework of microeconomic rationality and demonstrates how individually rational choices produce collectively suboptimal outcomes. Fear, in this approach, connotes only an anticipation of future risks, not the feelings aroused by those perceived dangers. 26 Neuropsychologists and behavioral economists treat fear very differently than do micro economic models. Fear is conditioned in part by our evolutionary makeup and is frequently evoked by crude or subliminal cues. Fear typically peaks just before a threat is experienced and is highly dependent on mental imagery and its vividness. Neuroscientists have also demonstrated that fear conditioning may be permanent, or at least far longer lasting than other kinds of learning. "To the extent that these differences exist between the calculus of objective risk and the determinants of fear, and to the extent that fear does play an important part in risk-related behaviors;' argue Loewenstein and his colleagues, "behavior in the face of risk is unlikely to be well-described by traditional consequentialist models:'27 When fear is strong enough, it overwhelms the usual risk-acceptance associated with loss aversion. 28 The fourth important negative emotion, anger, operates quite differently than fear. Fear arises from and evokes appraisals of uncertainty and loss of individual control, whereas anger arises from and evokes appraisals of certainty and heightened individual controp9 Anger triggered in one situation evokes more optimistic risk estimates and risk-seeking choices in unrelated situations, whereas fear does the opposite. 30 We can speculate then that angry adversaries are more likely to make risk-acceptant decisions, whereas frightened adversaries are more likely to be risk-averse and hesitant. Happiness is the one positive emotion among the core emotions. Not surprisingly, positive emotions trigger more optimistic assessments than negative

46

JANICE GROSS STEIN

emotions, whereas negative emotions trigger more pessimIstlc assessments than positive emotions, even if the source of the emotion has no relation to the target judgments. 3! The exception, as noted above, is anger, which also triggers optimistic assessments, although it does so through a different path. The impact on decisions of fear, on the one hand, and happiness and anger, on the other, is somewhat counterintuitive. Because fear is associated with higher risk and uncertainty than anger, the decisions made by frightened individuals are more risk-averse, whereas the choices made by both angry and happy individual are more risk-tolerant. 32 In an experimental setting, induced anxiety increased individuals' preference for low-risk, low-reward options. 33 Fear and anxiety tend to favor pessimistic risk assessments and cautious, risk-averse decision-making. 34 It is important, therefore, in designing strategies of assurance to understand whether an insecure ally or adversary is primarily frightened or angry, or both. Much of the policy discussion on Iran, for example, revolves implicitly around the question of whether its leadership is frightened or angry, or both simultaneously. Assurance is more likely to be effective if the other side's decision-makers are primarily frightened rather than angry. EMOTIONS AND RISK ASSESSMENT AND RISK PROPENSITY

Risk and uncertainty are the dominant characteristics of strategic decisions in global politics. Behavioral decision theory and research from neuroscience-unlike microeconomics-treat risk as feeling. The emotional reactions to a risky situation often diverge from the cognitive evaluations of risk, and when they do, the emotional reactions often exert a dominating influence on behavior.35 Research generally confirms the significant and direct impact of emotion on risk assessment and risk propensity.36 Emotion is influenced by the vividness of the possible outcomes, by the mental imagery these outcomes evoke, by the associations these outcomes evoke, and by how an outcome is described. Changes in probability within some broad midrange of values have little effect on anticipatory emotions. Feelings of fear or worry in the face of decisions under risk or uncertainty have an all-or-none characteristic; people may be sensitive to the possibility rather than the probability of negative consequences. People are likely more insensitive to probability variations for emotional and vivid than for pallid outcomes. 3? Emotions also playa critical role in forward-looking decision-making. People care about the delayed or uncertain consequences of their decisions only to the extent that thinking about these consequences evokes affect.38 In thinking

THE PSYCHOLOGY OF ASSURANCE: AN EMOTIONAL TALE

47

about risk, people anticipate the consequences of different options, but focus on what is emotionally salient to them. One possible-and compelling-explanation is that the needs of the present make themselves felt through emotion, while the needs of the future express themselves largely through thought. Most people feel the urge to satisfy their immediate needs, but they only think about satisfying their future needs. 39 In a well-known experiment, Damasio set up a game involving patients who had suffered brain injuries that impaired their capacity to process emotions and others who were not so impaired. 40 Players each had a fixed amount of money and they could pull cards from four different decks; these cards would either give or take away money. Two of the decks were low risk, low reward, while the other two were high reward but with risks that ensured even higher losses. Those whose emotional capacity was not impaired quickly began to draw cards only from the low-risk, low-reward decks, while the patients with impaired emotions drew from the high-risk, high-reward decks and consistently went bankrupt. The players with diminished emotional capacity did not take account of the future consequences of their choices. Without feelings, future consequences meant little. The risk-as-feeling argument contends that responses to risky situations result in part from direct, unmediated emotion and in part indirectly from emotion mediated through cognition. Feelings-fear, anger, sadness-respond directly to a sense of risk. People also evaluate risky alternatives through cognitive processes, based largely on the probability and desirability of associated consequences. These cognitive evaluations generate affective consequences, which also exert a reciprocal influence on cognitive evaluations. Although the two reactions are reciprocal and interdependent, they come from different sources and can, at times, diverge. When they do diverge, emotion often swamps cognition. We know now that risk preference is often determined by emotionsY EMOTION AS A PRACTICAL TOOL OF POLICY ANALYSIS

Scholars in international relations are now beginning to look carefully at the neuroscience of emotion and how it affects cooperation, war termination, the credibility of threats, and the solution of collective action problems. 42 Here I address the role of emotion in designing and developing strategies of assurance. I do not claim that emotion is the only, or even the most powerful, variable in the success of strategies of assurance. In science and in social science, we have become appropriately skeptical of any single-variable explanation of

48

JANICE GROSS STEIN

outcomes in complex environments. What is important is to demonstrate how emotion is connected to other variables that have been identified as important to the outcome of assurance strategies. Strategies of assurance designed to counter the impulse to proliferation work at multiple levels. In the first instance, they must be chosen by an individual leader or group of decision-makers who consider the impact on their own domestic constituencies, on the leaders of a would-be proliferator, and on their public. Satisfying the constraints of multiple constituencies while providing credible assurances of future security so that developing a nuclear weapons program is no longer attractive to a would-be proliferator is demanding. Knopf suggests that assurances are more likely to be effective when a state's interest in nuclear weapons is driven to some significant degree by security concerns. This hypothesis is consistent with the research on fear that predicts risk-aversion and hesitancy, creating an opening for assurance as an alternative to a risky strategy of nuclear proliferation and enhanced defense spending. If leaders are fundamentally satisfied with the status quo but worry that they may be attacked in the future, then they are more likely to be receptive to assurances that credibly promise to preserve the status quo into the future. Security-seeking states that simultaneously fear the future are ideal candidates for strategies of reassurance. A second set of hypotheses focuses on credibility and emphasizes the importance of public declarations, legally binding mechanisms, and honoring commitments. There is both an individual and a collective component to these arguments. At the individual level, we ask: why should leaders believe these assurances? What makes them credible? And how does emotion play into credibility? Drawing on the new research on emotion, Jonathan Mercer sounds an appropriately cautionary note about how one party's actions contribute to credibility in the eyes of anotherY Feelings constitute beliefs, he argues, and we do not own our credibility. Other parties do, and these beliefs are only partially affected by what we do. They are also affected by whether they support or oppose what we are doing, hate us or like us, are angry at what we have done to them in the past, and whether or not they fear us. The selection and interpretation of evidence is theirs, not ours, to make. Since they, not we, largely determine our reputation, Mercer cautions appropriately about overconfidence in the credibility of assurances or threats. Credibility, in other words, is at least in part a function of the emotions of the other. Even if the argument is somewhat overdrawn-what I do constitutes my

THE PSYCHOLOGY OF ASSURANCE: AN EMOTIONAL TALE

49

reputation, at least in part-his caution is useful, especially in the context of the overwhelming emphasis on credibility and resolve in the nuclear age, an emphasis that often distorted nuclear strategy.44 Honoring a commitment, moreover, speaks to issues of trust and fairness, issues that have deep emotional resonance. It is easy to draw the feedback loop between fulfilling a commitment and the trust that it engenders in others, which in turn creates a climate where it is easier to honor commitments. A second level embedded in these arguments is the collective. It is not only the opposing leadership that strategies of assurance seek to persuade but their domestic publics as well. Much of the literature on "costly signals" builds on the assumption that the public commitments leaders make have costly consequences at home, with their own domestic publics, as well as internationally, should they fail to honor these assurances. It is the cost to leaders with their publics, the argument goes, that makes these commitments credible. For our purposes, the prior question is how does individual emotion, an embodied experience, become collective? We need to understand how emotion moves from the individual to the collective. Epidemiological and viral models help to explain the diffusion of emotion from an individual to a larger group.4S Emotion is an individual property, but the individual is embedded in a social context, picks up cues from her environment, reacts emotionally to those cues, and diffuses emotion back to others. The spread of emotions from one individual to another is similar to other contagious processes. "Just as diseases spread through contagion, so does confidence, or lack of confidence;' argue George Akerlof and Robert Shiller.46 "Indeed confidence ... may be as contagious as any disease. Epidemics of confidence or epidemics of pessimism may arise mysteriously because there was a change in the contagion rate." Confidence is an emotional state-an indicator of optimism about and trust in the future-which can spread to others. This is one of the ways, in an uncoordinated viral process, that "collective moods;' both positive and negative, develop. An individual's mood is partly a function of the mood of others and, in this sense, it is reasonable to speak of a collective mood. This process of contagion generally works from the "bottom up" to generate shared moods and, at the extreme, herd behavior. Alternatively, elites can deliberately, through their language and use of symbols, create a collective mood that endures over time. There is little doubt that the Iranian public feels aggrieved that special demands are being made of Iran that are not being made of other nuclear powers. There

50

JANICE GROSS STEIN

is a shared sense of unfairness, injustice, and even anger that sets the context for the discussion of Iran's nuclear program. The palpable sense of unfairness is so strong that it infused the leadership struggle within Iran and constrained President Ahmadinejad's ability to accept the agreements his negotiators had crafted. In a spiraling process, Iran's behavior in turn generates increasing distrust and anger at Iran's leadership as its leaders and officials are caught repeatedly suppressing information about its programY Collective emotion can act as a significant constraint or enabler on the success of assurance. In the transition from the psychology to the sociology of emotion, emotions spread and diffuse from one individual to another and create long-lasting, enduring public moods of fear, anger, suspicion, or optimism about the future. These collective moods set the context against which strategies of assurance are heard and interpreted. In thinking about the design of assurance strategies, the dynamic from the individual to the collective and back is an obvious and necessary consideration. Knopf hypothesizes that public declarations of security assurances are likely to increase their effectiveness. This hypothesis underscores the importance of collective moods in the determination of the likelihood of successful assurance. A sentiment of public optimism and trust in particular assurances or broader confidence in the future among the public in the target state enhances the likelihood that security assurances will be effective, while a climate of anger reduces the likelihood that assurances will succeed. An important strategic target for the design of assurances should therefore be the collective mood of one's ally or adversary. Public assurances are more likely to be effective precisely because they are public, because they enter into an ongoing debate, and can help to shape the emotional climate so that it favors the moderates rather than the hard -liners who play on shared fear or anger. A third set of arguments discussed by Knopf emphasizes the other state's culture, decision-making procedures, and leadership concerns. To reframe the argument somewhat, how people are treated matters, and often process is as important as outcome. There is strong evidence from neuroscience that perceived unfairness and humiliation evoke powerful emotional responses. When this kind of treatment evokes anger, publics and leaders become less cautious and more risk-acceptant. Assurances are unlikely to provoke feelings of insult or humiliation and, consequently, are more likely to elicit a positive response. This depends, however, on the assurances-whether they are negative or positive-being perceived as fair.

THE PSYCHOLOGY OF ASSURANCE: AN EMOTIONAL TALE

51

The arguments that Etel Solingen and Jacques Hymans make are consistent with the research on emotion. 48 Inward-looking leaders, who are nationalistic and confrontational domestically and internationally, and who are likely to be angry, are not predisposed emotionally to respond to assurances. They are riskacceptant and will be among the most difficult to dissuade from proliferation. To the extent that there is competition among elites, the competition may open space for strategies of assurance with more outward-looking elites, but this can heighten threat perception by the inward-looking leadership. Iran's leadership, for example, even before the disputed 2009 election, warned against interference by outsiders to promote a "velvet" revolution. "Interference" is in the eye of the beholder. Emotion also helps us to understand how leaders think and feel about risk. Emotion matters in an uncertain world where leaders consider the risks of the options they have. Paradoxically, "greedy" leaders are less likely to indulge their appetites than some game theoretic models suggest, largely because the pleasure of gain is less than the pain of loss. "Fearful" leaders are also less likely to challenge the status quo, except when they think that they are about to lose something that they already have and that they value, or when they are seeking to recover something of value-something they care about-and have lost. The analysis of emotion helps to identify the conditions under which strategies of assurance are likely to succeed. In short, assurances are most likely to succeed when they are least needed. Finally, it is worth remembering that strategies of assurance often do not stand alone when a relationship is adversarial. Since the intentions of an adversary are almost always uncertain, leaders are simultaneously seeking to assure an adversary's leaders-should they be frightened-but also to signal resolveshould the adversary's leaders be greedy. And even when they are reasonably confident that their adversary is greedy, they seek simultaneously to signal resolve-to prevent undesired action-and assurance-that should their adversary refrain, they will be secure. Much of the debate around strategies to prevent nuclear proliferation grapples with this dilemma of simultaneous signals. As Andrew Kydd argues, leaders of status quo states (such as those trying to prevent proliferation) want to acquire a reputation for both moderation and resolve. 49 How to calibrate these signals? Game theoretic analyses suggest that leaders can mitigate the sharpness of this trade-off by developing different audience costs for failing to respond to challenges in contrast to those for engaging in unilateral aggression. 50 Analysts

52

JANICE GROSS STEIN

of the security dilemma have long argued that the optimal environment for calibrating this kind of mixed strategy is when offense and defense are distinguishable one from each other, and even better, when leaders believe that the environment favors the defenseY These kinds of environments make strategies of assurance easier to design and execute. While both logics are compelling, the design and execution of this kind of calibrated strategy is extraordinarily difficult given the power of emotions as signals. As Kaufman argues, "Information that is ... emotionally interesting or exciting will receive greater weight than its evidentiary value warrants."52 Leaders trying to signal resolve by increasing their own costs should they fail to respond will in all likelihood create a more highly charged environment among their adversary's public and leaders. That highly charged environment will decrease their adversary's receptivity to assurances of moderation. In a dynamic environment where emotions are the carriers of signals, assurances can be crowded out by demonstrations of resolve. CONCLUSION

Emotional as well as cognitive factors are critical to the identification of the appropriateness and success of positive assurances, negative assurances, and reassurance. We have paid inadequate attention to the emotional context of assurance. Three over arching arguments would help guide the analysis of assurance. First, emotion is one of the screens through which all information must pass, and it shapes the way strategies of assurance are felt as well as interpreted. Second, emotions are programs of action. They predispose people to fight or flee, to hesitate or act. Third, the way people assess options and risk is in part a function of how they feel about alternative futures. All three matter in analyzing the likely impact of assurance strategies, both among leaders and their publics. Research on emotions shows that assurance could play an important role in defusing conflict, but also that designing effective assurance strategies is challenging. Failure to consider the impact of emotion will only make the design of effective strategies even more difficult. NOTES 1.

Thomas C. Schelling, Arms and Influence (New Haven: Yale University Press,

1966),74· 2.

For convenience, I use the term "assurance" throughout this chapter to cover as-

THE PSYCHOLOGY OF ASSURANCE: AN EMOTIONAL TALE

53

surance as a component of deterrence, assurance of allies, and reassurance in the context of arms races, deterrence, and the security dilemma. See Knopf, this volume, Chapter 2, for these distinctions. 3. Philip E. Tetlock, Expert Political Judgment: How Good Is It? How Can We Know? (Princeton: Princeton University Press, 2005); Robyn M. Dawes, "Behavioral Decision Making and Judgment;' in The Handbook of Social Psychology, 4th ed., vol. 1, ed. Daniel T. Gilbert, Susan T. Fiske, and Gardner Lindzey (New York: McGraw-Hill, 1998), 497548; Paul Slovic, "Trust, Emotion, Sex, Politics, and Science: Surveying the Risk-Assessment Battlefield;' Risk Analysis 19, no. 4 (August 1999): 689-700. 4. Tetlock, Expert Political Judgment, 40. 5. Daniel Kahneman and Amos Tversky, "Prospect Theory: An Analysis of Decision under Risk;' Econometrica 47, no. 2 (March 1979): 263-91; Amos Tversky and Daniel Kahneman, ''Advances in Prospect Theory: Cumulative Representation of Uncertainty;'

Journal of Risk and Uncertainty 5, no. 4 (1992): 297-323. 6. William A. Boettcher, "The Prospects for Prospect Theory: An Empirical Evaluation of International Relations Applications of Framing and Loss Aversion;' Political

Psychology 25, no. 3 (June 2004): 331-62. 7. Janice Gross Stein, "Calculation, Miscalculation, and Conventional Deterrence I: The View from Cairo;' in Psychology and Deterrence, ed. Robert Jervis, Richard Ned Lebow, and Janice Gross Stein (Baltimore, MD: Johns Hopkins University Press, 1985), 34-59; Stein, "Deterrence and Learning in an Enduring Rivalry: Egypt and Israel, 1948-

73;' Security Studies 6, no. 1 (Autumn 1996): 104-52. 8. Jonathan Mercer, "Emotional Beliefs;' International Organization 64, no. 1 (January 2010): 1-31. 9. ''Affect'' is defined as "positively or negatively valenced subjective reactions that a person experiences at a given moment in time" (Colin Camerer, George Loewenstein, and Drazen Prelec, "Neuroeconomics: How Neuroscience Can Inform Economics;'

Journal of Economic Literature 43, no. 1 [March 2005]: 39). It is the way people represent the value of things as good or bad and, technically, is just one dimension of emotion (Rose McDermott, "The Feeling of Rationality: The Meaning of Neuroscientific Advances for Political Science;' Perspectives on Politics 2, no. 4 [December 2004]: 691-706). The two are nevertheless often used interchangeably. 10. Jonathan Mercer, "Rationality and Psychology in International Politics;' Interna-

tional Organization 59, no. 1 (Winter 2005): 77-106. 11. Antonio Damasio, Descartes' Error: Emotion, Reason, and the Human Brain (New York: Putnam, 1994). 12. Joseph LeDoux, The Emotional Brain: The Mysterious Underpinnings of Emo-

tional Life (New York: Simon and Schuster, 1996); Jaak Panksepp, Affective Neuroscience (Oxford: Oxford University Press, 1998); Edmund T. Rolls, The Brain and Emotion (New York: Oxford University Press, 1999).

54

JANICE GROSS STEIN

13. Camerer et aI., "Neuroeconomics;' 11. 14. Ibid., 27. 15. LeDoux, The Emotional Brain; Piotr Winkielman and Kent C. Berridge, "Unconscious Emotion;' Current Directions in Psychological Science 13, no. 3 (2004): 120-23. 16. Camerer et aI., "Neuroeconomics;' 26. 17. John A. Bargh, Shelly Chaiken, Paula Raymond, and Charles Hymes, "The Automatic Evaluation Effect: Unconditional Automatic Attitude Activation with a Pronunciation Task;' Journal of Experimental Social Psychology 32, no. 1 (January 1996): 104-28; Jan De Houwer, Dirk Hermans, and Paul Eelen, "Affective and Identity Priming with Episodically Associated Stimuli;' Cognition and Emotion 12, no. 2 (March 1998): 145-69. 18. Daniel T. Gilbert and Michael J. Gill, "The Momentary Realist;' Psychological

Science 11, no. 5 (September 2000): 394-98. 19. Alan G. Sanfey et aI., "The Neural Basis of Economic Decision-Making in the Ultimatum Game;' Science 300, no. 5626 (June 13, 2003): 1755-58. 20. Nico H. Frijda, "The Laws of Emotion;' American Psychologist 43, no. 5 (May 1988): 349-58; Leonard Berkowitz, "Anger;' in Handbook of Cognition and Emotion, ed. Tim Dalgleish and Mick J. Power (New York: Wiley, 1999), 411-28. 21. Robert B. Zajonc, "Feeling and Thinking: Preferences Need No Inferences;'

American Psychologist 35, no. 2 (February 1980): 151-75; Zajonc, "On the Primacy of Affect;' American Psychologist 39, no. 2 (February 1984): 117-23; Zajonc, "Emotions;' in The Handbook of Social Psychology, 4th ed., vol. 1, 591-632; Camerer et al., "Neuroeconomics;' 18. 22. Daniel Kahneman and Alan B. Krueger, "Developments in the Measurement of Subjective Well-Being;' Journal of Economic Perspectives 20, no. 1 (Winter 2006): 3-24. 23. David DeSteno, Richard E. Petty, Duane T. Wegener, and Derek D. Rucker, "Beyond Valence in the Perception of Likelihood: The Role of Emotion Specificity;' Journal

ofPersonality and Social Psychology 78, no. 3 (March 2000): 397-416; George Loewenstein and Jennifer S. Lerner, "The Role of Affect in Decision Making;' in Handbook ofAffective Sciences, ed. Richard J. Davidson, Klaus R. Scherer, and H. Hill Goldsmith (New York: Oxford University Press, 2003), 619-42. 24. Richard S. Lazarus, Emotion and Adaptation (New York: Oxford University Press, 1991). 25. Jennifer S. Lerner, Deborah A. Small, and George Loewenstein, "Heart Strings and Purse Strings: Carryover Effects of Emotions on Economic Decisions;' Psychological

Science 15, no. 5 (2004): 337-41. 26. Neta C. Crawford, "The Passion of World Politics: Propositions on Emotion and Emotional Relationships;' International Security 24, no. 4 (Spring 2000). 27. George F. Loewenstein et aI., "Risk as Feelings;' Psychological Bulletin 127, no. 2 (March 2001): 280. 28. Empirical evidence that fear can swamp loss-aversion comes from the waves of

THE PSYCHOLOGY OF ASSURANCE: AN EMOTIONAL TALE

55

panic selling that followed a steep drop in the equities markets after September 2008. After experiencing losses, investors should have been willing to take additional risks to compensate for the losses. They have done so in the past, but this time, acute fear trumped that instinct. "We are caught in a spiral in which we are so scared of losing our jobs, our savings, that fear overtakes our brains;' wrote neuroscientist Gregory Berns ("When Fear Takes Over Our Brains;' New York Times, December 7,2008, Business section, 2). When people are frightened, panic selling is not uncommon. Fear, in other words, overwhelms the usual risk-acceptance associated with loss aversion. 29. Craig A. Smith and Phoebe C. Ellsworth, "Patterns of Cognitive Appraisal and

Emotional Response Related to Taking an Exam;' Journal of Personality and Social Psy-

chology 52, no. 3 (March 1987): 475-88. 30. Jennifer S. Lerner, Roxanna M. Gonzalez, Deborah A. Small, and Baruch Fis-

chhoff, "Effects of Fear and Anger on Perceived Risks of Terrorism: A National Field Experiment;' Psychological Science 14, no. 2 (March 2003): 144-50; Camerer et aI., "Neuroeconomics:' 31. Eric J. Johnson and Amos Tversky, "Affect, Generalization, and the Perception of

Risk;' Journal of Personality and Social Psychology 45, no. 1 (1983): 20-31. 32. Larissa Z. Tiedens and Susan Linton, "Judgment under Emotional Certainty and

Uncertainty: The Effects of Specific Emotions on Information Processing;' Journal of Personality and Social Psychology 81, no. 6 (December 2001): 973-88; Jennifer S. Lerner and Dacher Keltner, "Fear, Anger, and Risk;' Journal ofPersonality and Social Psychology 81, no. 1 (2001): 146-59. 33. Rajagopal Raghunathan and Michael Tuan Pham, "All Negative Moods Are Not

Equal: Motivational Influences of Anxiety and Sadness on Decision-making;' Organiza-

tional Behavior and Human Decision Processes 79, no. 1 (July 1999): 56-77. 34. Nazanin Derakshan and Michael W. Eysenck, "Interpretive Biases for One's Own

Behavior and Physiology in High-Trait-Anxious Individuals and Repressors;' Journal of Personality and Social Psychology 73, no. 4 (October 1997): 816-25; Michael W. Vasey, Nagat EI-Hag, and Eric L. Daleiden, ''Anxiety and the Processing of Emotionally Threatening Stimuli: Distinctive Patterns of Selective Attention among High- and Low-Test Anxious Children;' Child Development 67, no. 3 (June 1996): 1173-85; Lerner and Keltner, "Fear, Anger, and Risk:' 35. Randolph M. Nesse and Richard Klaas, "Risk Perception by Patients with Anxi-

ety Disorders;' Journal of Nervous and Mental Disease 182, no. 8 (1994): 466-70. 36. Christopher K. Hsee and Elke U. Weber, ''A Fundamental Prediction Error: Self-

Other Discrepancies in Risk Preference;' Journal of Experimental Psychology: General 126, no. 1 (1997): 45-53. 37. C. Peter Bankart and Rogers Elliott, "Heart Rate and Skin Conductance in An-

ticipation of Shocks with Varying Probabilities of Occurrence;' Psychophysiology 11, no. 2 (March 1974): 160-74; George Loewenstein, Scott Rick, and Jonathan D. Cohen, "Neu-

56

JANICE GROSS STEIN

roeconomics;' Annual Review of Psychology 59 (2008): 647-72; Yuval Rottenstreich and Christopher K. Hsee, "Money, Kisses, and Electric Shocks: On the Affective Psychology of Risk;' Psychological Science 12, no. 3 (May 2001): 185-90. 38. Thomas J. Cottle and Stephen L. Klineberg, The Present of Things Future (New

York: Free Press, 1974). 39. Hersh Shefrin, Beyond Greed and Fear: Understanding Behavioral Finance and

the Psychology of Investing (New York: Oxford University Press, 2002), 141-42. 40. Antoine Bechara, Hanna Damasio, Daniel Tranel, and Antonio Damasio, "De-

ciding Advantageously before Knowing the Advantageous Strategy;' Science 28, no. 275 (1997): 1293-99· 41. Loewenstein et aI., "Neuroeconomics;' 274, 280. 42. McDermott, "The Feeling of Rationality"; Mercer, "Rationality and Psychology";

James K. Rilling et aI., ''A Neural Basis for Social Cooperation;' Neuron 35, no. 2 (July 18, 2002): 395-405; Janice Gross Stein, "Emotion and Financial Decision Making: Asset

Bubbles, Panic Selling and the 'Great Recession;" paper prepared for the American Political Science Association Annual Meeting, Toronto, September 2009, 3-6. 43. Mercer, "Emotional Beliefs:' 44. Robert Jervis, The Meaning of the Nuclear Revolution (Ithaca, NY: Cornell Uni-

versity Press, 1989). 45. Stein, "Emotion and Financial Decision Making:' 46. George A. Akerlof and Robert J. Shiller, Animal Spirits: How Human Psychology Drives the Economy, and Why It Matters for Global Capitalism (Princeton: Princeton

University Press, 2009),56. 47. Interviews of participants in a Track II process, May 2010. 48. Etel Solingen, Nuclear Logics: Contrasting Paths in East Asia and the Middle East

(Princeton: Princeton University Press, 2007); Jacques C. Hymans, The Psychology of

Nuclear Proliferation (Cambridge: Cambridge University Press, 2006). 49. Andrew Kydd, Trust and Mistrust in International Relations (Princeton: Princeton University Press, 2005). 50. Ibid., and Ken Booth and Nicholas J. Wheeler, The Security Dilemma: Fear, Co-

operation, and Trust in World Politics (New York: Palgrave Macmillan, 2008). 51. Charles L. Glaser, "The Security Dilemma Revisited;' World Politics 50, no. 1 (October 1997): 171-201, and Rational Theory of International Politics (Princeton: Princeton University Press, 2010). 52. Chaim D. Kaufmann, "Out of the Lab and into the Archives: A Method for Test-

ing Psychological Explanations of Political Decision Making;' International Studies

Quarterly 38, no. 4 (December 1994): 563.

The Role of Security Assurances in the Nuclear Nonproliferation Regime John Simpson

This chapter reviews the history of nuclear-related security assurances, focusing mainly on multilateral contexts, in order to provide background relevant to the other chapters in this volume. Requests for, and offers of, nuclear security assurances predate the emergence of the current global nonproliferation regime. Yet such assurances have never been encoded in as strong a form as many states have wanted. In recent years, the salience of security assurances appears to have declined, but the legacy of past debates still affects nonproliferation discussions. The time may therefore be ripe to think creatively about how to adapt nuclear-related security assurances to the contemporary environment. Positive commitments by nuclear weapon states (NWS) to allied or dependent non-nuclear weapon states (NNWS) potentially threatened with nuclear attack have been a feature of international relations since the late 1940S; such commitments are often called security guarantees. In parallel, decolonization led to the emergence in 1961 of a Non-Aligned Movement (NAM) seeking to enhance the collective political power of those states outside the Cold War blocs. To protect themselves from the consequences if an interbloc nuclear war occurred they advocated nuclear disarmament and the abandonment of policies of nuclear deterrence. Until these aims were achieved, the NAM sought commitments from NWS never to use nuclear weapons against NNWS (negative security assurances), and to provide assistance and support in the event of such attacks (positive security assurances). In some cases, these demands for positive assurances were also motivated by the suspected nuclear weapon programs of regional neighbors. Throughout this time period, there have been tensions between the nuclear security guarantees offered to NWS allies and the 57

58

JOHN SIMPSON

broader, legally binding security assurances sought by nonaligned states. States have sought security assurances bilaterally, regionally (especially through the mechanism of nuclear-weapon-free zones), and globally through the UN and its associated negotiating forums. In 1961 the UN General Assembly adopted an Irish resolution calling for measures to prevent the further transfer or spread of nuclear arms. Movement toward a legal nondissemination and nonproliferation instrument subsequently became a primary focus for NAM demands, and formal negotiations began in 1965 in the Eighteen Nation Disarmament Committee (ENDC). They were jointly chaired by the two bloc leaders, the United States and the USSR. To get states allied to one of the bloc leaders that had not acquired nuclear weapons to sign such a treaty, their existing positive security guarantees, in the form of extended nuclear deterrence, had to be protected. The resultant Treaty on the Non-Proliferation of Nuclear Weapons (NPT) opened for signature on July 1, 1968, and came into force on March 5, 1970. It contained no legally or politically binding nuclear security commitments or assurances, despite the demands for this from NAM states. The NPT had an initial duration of twenty-five years and provided for review conferences every five years to discuss progress in meeting treaty objectives. For the NAM states its main advantage, given that few of them could aspire to acquire nuclear weapons, was the Article VI commitment for all states to engage in "good faith" negotiations on nuclear disarmament. Only a world without nuclear weapons could remove the potential nuclear threats they faced. Other security benefits were the commitments of other NNWS not to acquire nuclear weapons and to accept International Atomic Energy Agency (IAEA) safeguards on their territory. However, the need to make at least a gesture toward NAM security concerns was recognized by the three depositary states, the United States, the United Kingdom, and the USSR. They used UN Security Council Resolution (UNSCR) 255, passed on June 19, 1968, 1 to offer NNWS a rather weak assurance of assistance if the latter were subject to nuclear attack. NAM demands for stronger security assurances subsequently migrated to four alternative forums: the NPT review process; regional nuclear-weapon-free zone arrangements; the successor forums to the ENDC (currently the Conference on Disarmament in Geneva); and the UN General Assembly (UNGA). Little movement on this issue proved possible in the NPT review process until 1995, largely due to the reluctance of NATO states to forgo nuclear use, given the perceived conventional superiority of Warsaw Pact forces.

THE ROLE OF SECURITY ASSURANCES

59

The first regional Nuclear-Weapon-Free Zone (NWFZ) Treaty, covering Latin America and the Caribbean (the Treaty of Tlatelolco), opened for signature in 1967. The treaty had a protocol providing negative security assurances to its parties which all five NWS eventually ratified. It was not until 1985 that a similar NWFZ treaty was agreed for the South Pacific, followed in 1995-96 by two more covering Southeast Asia and Africa and, in 2006, a treaty covering Central Asia. The United States, however, has not ratified the negative security assurance protocols in any of the treaties since Tlatelolco, and none of the NWS have so far ratified the protocols attached to the Southeast Asian or Central Asian zones. 2 A set of negative assurances were offered by individual NWS in the context of the UNGA Special Sessions on Disarmament in 1978 and 1982, though, with the exception of that of China, all were heavily conditioned. They were then amended and linked to a UN Security Council Resolution in 1995. This step helped secure permanent extension of the NPT, but little collective movement followed. However, both the United States and the United Kingdom strengthened and clarified their assurances unilaterally in 2010. It remains to be seen whether the latest policy developments can satisfy both allies covered by extended nuclear deterrence and nonaligned states seeking stronger assurances against nuclear threats. This chapter chronicles and analyzes the evolution of nuclear security assurances. The first section reviews the sustained demands from NAM states, especially those within the NPT, for security assurances during the last half-century. The second section contains a short analysis of the options available to enhance existing NPT-related security assurances. The third section explores the impact of recent changes in the security environment on the demands for NPT security assurances. The chapter concludes that the salience of security assurances may be decreasing, but they still remain relevant and continue to be sources of diplomatic friction. The form and content of security assurances may need to be adjusted, however, to fit current circumstances. SECURITY ASSURANCES, SECURITY GUARANTEES, AND THE NPT: A SHORT HISTORy3

Joint security arrangements can be either bilateral or multilateral in form, and have included both nonaggression agreements and positive commitments of assistance if one party is attacked. The earliest security assurances of the nuclear age involved positive commitments made via alliances. The initial nu-

60

JOHN SIMPSON

clear weapon states (the UK, U.S., and USSR) committed themselves in the early 1950S to come to the assistance of certain other states on the basis of often opaque, but apparently legally binding, defense agreements. While the commitments contained in these defense pacts were often articulated in general terms (for example, Article 5 of the North Atlantic Treaty),4 participation in them by nuclear-armed states inevitably gave them a nuclear element. Bilateral defense agreements with nuclear deterrent overtones were established, for example, between the United States and Japan,s South Korea,6 and the Philippines.? In parallel, new multilateral defense pacts involving nuclear states were created, with the United States and/or the United Kingdom as a member of the NAT,8 ANZUS,9 the RIO pact,1O CENTO,ll and SEATO,n and the Soviet Union as the leader of the WTO.13 The military and economic attractiveness of substituting nuclear for conventional weaponry led the initial nuclear states to extend their nuclear deterrent threats to cover their allies. One way of enhancing the credibility of extended deterrence was deployment of nuclear weapons on or near allied territory. This global spread of nuclear deployments reinforced concern that ownership of these weapons would inevitably spread to more states unless mechanisms to prevent this spread could be developed. Against this background, two distinct modes of thought emerged within NNWS on how best to obtain nuclear security. One was to rely on the perceived deterrent effects of the nuclear states' nuclear weapons and associated political commitments and military arrangements to prevent themselves from being the target of both conventional and nuclear aggression. The other was to regard this form of security as illusory and argue instead for the abolition of nuclear weapons. Until that occurred, the only options appeared to be to obtain legal commitments by the NWS to engage in a disarmament process, and in the interim persuade them to offer legally binding commitments to refrain from engaging in nuclear aggression, and to offer assistance to NNWS that were so threatened or attacked. NWS also began to perceive that NNWS not formally allied with them might require some form of security commitment to dissuade them from seeking a nuclear deterrent of their own. After China conducted its first nuclear test in October 1964, U.S. President Lyndon Johnson gave a speech in which he declared: "Nations that do not seek nuclear weapons can be sure that if they need our strong support against some threat of nuclear blackmail, then they will have it:'14 The Johnson administration hoped this generalized positive security

THE ROLE OF SECURITY ASSURANCES

61

assurance would help convince India not to follow China's lead. IS This vaguely worded unilateral statement did not satisfy NAM states, however, who instead pressed for such assurances to be incorporated in a formal, multilateral agreement. The negotiation from 1965 onward of what eventually became the NPT emerged from calls for a nondissemination treaty. These calls were triggered by the United States seeking to make its nuclear weapons accessible to its allies in the event of a general war in Europe. Allied concerns that the growing strategic nuclear power of the USSR would neutralize the U.S. extended nuclear deterrent capabilities had fueled moves to give the allies easier access to nuclear weapons. In the early 1960s, ideas were floated for creating a mixed-manned NATO strategic nuclear force with allied fingers on triggers via a Multilateral Force (MLF)16 or Allied Nuclear Force (ANF)Y These proposed hardware solutions were by 1965 giving way to a software fix, formalized in December 1966, in the form of a NATO Nuclear Planning Group (NPG) to provide direct national political inputs into plans for nuclear use. This substitution of a political institutional mechanism for the MLF and ANF ideas, which were now abandoned, made NPT negotiations with the USSR possible. However, U.S. bilateral and multilateral extended nuclear deterrent arrangements remained necessary if U.S. allies were to be willing to sign and ratify such a treaty. IS Above all, the perceptions of USSR conventional superiority meant that any constraints on nuclear weapon deployments would face strong NATO opposition. At the same time, many NAM states feared that they could be subject to security threats from existing NWS or new states that acquired nuclear weapons. This made the question of nuclear security assurances for states not part of an alliance an important focus of debate in the period before the signing of the NPT. The ensuing debate was complicated by the existence of six "threshold" nuclear weapon states (Argentina, Brazil, India, Israel, Pakistan, and South Africa) that appeared likely to remain outside of the treaty. This raised the issue of whether nuclear security assurances should be reserved for NPT parties alone to encourage those outside the treaty to join, or should be negotiated in the ENDC and be available to all states. Also, some in the NAM argued that, as they would voluntarily forgo their inalienable right to self-defense written into the UN Charter by signing the treaty, they wanted in return legal commitments from the NWS to provide support and assistance in the event of a threat, or actual attack, with nuclear weapons. 19

62

JOHN SIMPSON

As France chose not to join the treaty and the People's Republic of China (PRC) was then not a member of the United Nations or the ENDC, the three NWS NPT depositaries refused to commit themselves to such assurances in the text of the NPT. However, they recognized the need to provide some type of "stand alone" security commitment to treaty states, and did this through UN Security Council Resolution 255, passed on June 19, 1968, seven days after the UN General Assembly approved the NPT. 20 UNSCR 255 stated that the Security Council and "above all its nuclear weapon State permanent members, would have to act immediately in accordance with their obligations under the United Nations Charter" in response to a nuclear attack against a NNWS. It also referred to identical declarations by the three depositaries to support any NNWS party to the NPT that was the "victim of an act or an object of a threat of aggression in which nuclear weapons are used:'21 The positive security assurances conveyed by UNSCR 255 were not considered satisfactory by most NNWS because: • the Security Council could only act on the basis of transparent nuclear threats or actions; • UNSCR 255 appeared merely to reiterate existing NWS commitments under the UN Charter; • while the resolution obliged the NPT Depositary States to act if nuclear aggression occurred, rather than giving them the option of acting, it did not specify the action to be taken; • the resolution appeared to give France, the United Kingdom, the United States, and the USSR the freedom to take collective military action under this resolution without consulting either the target state or its claimed aggressor (unless it was one of them); and • France, a non-NPT signatory, could veto any Security Council resolution affecting its actions. 22 About the same time that UNSCR 255 provided the first generalized positive security assurance, the first treaty-based instrument for a negative security assurance emerged. The Latin American states had embarked on the negotiation of a nuclear-weapon-free zone in their region, and the opening of its text for signature in February 1967 preceded that of the NPT. Article 3 of Additional Protocol II of this Treaty of Tlatelolco, which the five NWS were asked to sign, stated that they were "not to use or threaten to use nuclear weapons against the contracting parties:'23 Although it was to be 1979 before all the NWS ratified

THE ROLE OF SECURITY ASSURANCES

63

this protocol, its text served as a template for incorporating similar negative security assurances in later NWFZs. Developments Following Negotiation of the NPT

It was not until the 1978 UNGA Special Session on Disarmament (UNSSOD I) that further action was taken to provide security assurances to NNWS. The NAM aim in this forum was to have the NWS commit themselves collectively to legal language similar to Protocol II of the Tlatelolco Treaty. Instead, the NWS made a series of unilateral statements that contained varying commitments:

• China stated that "at no time and in no circumstances will it be the first to use nuclear weapons";24 • France said it was "prepared to give ... [assurances of the non-use of nuclear weapons against non-nuclear States] ... , in accordance with arrangements to be negotiated, to States which constitute non-nuclear zones";25 • the Soviet Union declared that it "will never use nuclear weapons against those states which renounce the production and acquisition of such weapons and do not have them on their territories";26 • the United Kingdom stated that it "undertakes not to use nuclear weapons against ... [non-nuclear-weapon States which are parties to the Treaty on the Non- Proliferation of Nuclear Weapons and to other internationally binding commitments not to manufacture or acquire nuclear weapons] ... except in the case of an attack on the United Kingdom, its dependent territories, its armed forces or its allies by such a state in association or alliance with a nuclear-weapon state";27 and • the United States, in a similar formulation, avowed that it "will not use nuclear weapons against any non-nuclear-weapon State Party to the NPT or any comparable internationally binding commitment not to acquire nuclear explosive devices, except in the case of an attack on the United States, its territories or armed forces, or its allies, by such a state allied to a nuclearweapon state, or associated with a nuclear-weapon state in carrying out or sustaining the attack:'28 In 1982, the second session of this forum (UNSSOD II) led to China, France, and the Soviet Union making changes to their formal positions on this issue: • China stated that "at no time and under no circumstances will China be the first to use nuclear weapons, and that it undertakes unconditionally not to use or threaten to use nuclear weapons against non-nuclear countries and

64

JOHN SIMPSON

nuclear-free-zones:'29 thus extending its assurance to the latter states even after the start of nuclear hostilities by others; • France stated that "it will not use nuclear arms against a State that does not have them and that has pledged not to seek them, except if an act of aggression is carried out in association or alliance with a nuclear weapons State against France or against a State with which France has a security commitment:'30 thereby bringing its commitments closer to those made in 1978 by the United Kingdom and United States; and • the Soviet Union assumed "an obligation not to be the first to use nuclear weapons ... , [and was] prepared to conclude bilateral agreements or guarantees with States which do not possess nuclear weapons and do not have them on their territories:'3! thus joining China in giving a no-first-use commitment. NPT NAM states continued to regard these assurances as unsatisfactory in at least three respects: they were not in the form of a legally binding international treaty; they were not joint or uniform; and they applied to all states whether or not those states had joined the NPT. In particular, they expressed concern about the vague and potentially broad exception made by the three Western NWS for states acting "in association with" another NWS. As a consequence, from 1978 onward, two resolutions on security assurances started to appear regularly on the agenda of the UNGA. One, sponsored initially by the USSR and later by Bulgaria, focused on the need to conclude a legally binding convention on negative security assurances. The other, sponsored by Pakistan, stressed the need for a common formula to be included in any international instrument. The issue was raised also from 1975 onward at the quinquennial NPT Review Conferences, under the general heading of "regional issues:' Nigeria took the lead on negative security assurances, as this was seen by the African NAM membership as a method of putting pressure on other states to react to South Africa's nuclear capabilities. Egypt took the lead on positive security assurances, seeking on behalf of the Arab states to pressure the United States and its NATO allies over Israel's undeclared nuclear weaponry.32 In the 1985 NPT Review Conference Final Declaration, it was noted that negotiations were ongoing in the Conference on Disarmament (CD) on a legal instrument on negative security assurances. The same sentiments were expressed at the 1990 Review Conference, to which Egypt submitted a working paper on both negative and positive security assurances. 33 This paper asserted that nega-

THE ROLE OF SECURITY ASSURANCES

65

tive security assurances were the "most effective guarantee against the use or threat of nuclear weapons pending disarmament under effective international control." It highlighted the "existing peril" emanating from "certain [non -NPT] nuclear threshold states"-that is, Israel and South Africa, and made two specific proposals: that the Conference should call upon the Security Council to pass a resolution on the subject and agree to a legally binding instrument. It recommended that the UNSC resolution should contain: • an unconditional collective commitment by the NWS not to use or threaten to use nuclear weapons against those NPT NNWS not possessing or placing nuclear weapons on their territory; • a clear indication of the mandatory action to be adopted by the NWS and the Security Council if an NNWS NPT party was to be the object of a nuclear attack or threat of attack; • an obligation on all states to provide immediate assistance to the victim; and • an obligation on the threatening party to pay reparation or compensation to the victim. The legally binding instrument was to contain three commitments by the NPT NWS: • not to use or threaten to use nuclear weapons from their own territories against any state party that did not have nuclear weapons on its territory; • to support the imposition of sanctions against any state, party or nonparty to the NPT, which used nuclear weapons against a NNWS party to the treaty without nuclear weapons on its territories; and • to provide assistance, if requested, to any NNWS party to the treaty without nuclear weapons on its territory which had been attacked or threatened by the use of nuclear weapons. In addition, all NWS were requested to take steps to prevent any nonparty to the NPT with the capability to use nuclear weapons from attacking or threatening to attack an NPT NNWS. Although the 1990 Review Conference reached agreement to consult on this working paper prior to the 1995 NPT Review Conference, irresolvable differences between the United States and the NAM states over negotiations on a comprehensive test ban treaty (CTBT) meant that the 1990 Conference issued no Final Declaration providing authorization for this activity. Egypt at that

66

JOHN SIMPSON

point reverted to seeking progress on the issue through the CD. It submitted a working paper to that body in August 1991,34 pointing out the shortcomings of UNSCR 255 and unsuccessfully urging consultations on the subject of security assurances between all five recognized nuclear weapon states. Assurances Offered on the Path to NPT Extension

The NPT entered into force in 1970 for an initial duration of twenty-five years. The agenda for the NPT Review Conference in 1995 therefore included the treaty's future extension. This Conference took place after a period of major change in the international environment. Acute East-West nuclear confrontation had subsided with the dissolution of the Soviet Union, and the three NPT depositaries had started to reduce the numbers of their deployed nuclear warheads. China and France had both acceded to the NPT in 1992, meaning that all five NWS recognized by the treaty were now party to it. As part of an effort to ensure that former Soviet warheads would be removed from the territory of successor states to the USSR other than Russia, on December 5, 1994, three sets of security assurances were provided by the Russian Federation, the United States, and the United Kingdom to Kazakhstan, Ukraine, and Belarus. France provided a similar assurance to Ukraine. 35 Their contents paved the way for the wording of global commitments that would be offered by the NWS five months later through UNSCR 984. The 1995 NPT Review and Extension Conference, with China and France present for the first time, had the task of deciding on the future duration of the treaty. Major diplomatic efforts were made by the NWS to convince NAM members to go along with an indefinite extension. As part of this effort, the NWS sought to follow up the Egyptian proposals of 1990 and 1991 by offering enhanced security assurances to treaty parties. The route chosen was a series of individual statements at the UN on April 5-6, 1995, which were collectively acknowledged and welcomed by the passing of UNSCR 984 on Apriln. The five individual assurances were not uniform, but they did contain new and substantive commitments combining both positive and negative elements. The Chinese statement covered all NNWS,36 not just those in the NPT. It contained three commitments: no first use; no use in any circumstances against all NNWS and NWFZs; and support for action within the UNSC to provide necessary assistance to any NNWS that had come under attack with nuclear weapons, and to impose "strict and effective" sanctions upon the attacking state. It also called for the early conclusion of an international convention on

THE ROLE OF SECURITY ASSURANCES

67

no-first-use of nuclear weapons and a legal instrument providing negative security assurances. The Russian assurance was contained in a single sentence. 37 This said it would not use nuclear weapons against NNWS party to the NPT except in the case of: "an invasion or any other attack on the Russian Federation, its territory, its armed forces or other troops, its allies or a State towards which it had a security commitment, carried out or sustained by such a non-nuclear state in association or alliance with a nuclear weapon state:' This statement reversed the USSR's no-first-use pledge made in 1982; aligned Russia with the three Western NWS in excluding states "in association" with a nuclear state; and contained no reference to positive security assurances. The United States made a similar statement to the Russian one in respect of non-use against an NNWS NPT party,38 but qualified this by specifying the state had to be in compliance with the NPT to benefit from the assurance. It also indicated "its intention to provide or support immediate assistance" to any NNWS party to the NPT "that is a victim of an act of, or an object of a threat of, aggression in which nuclear weapons are used:' It listed some possible response measures and stated that in the event of such an act of aggression the Security Council should consider which means were appropriate. The United Kingdom made a statement similar to that of the United States in most respects.39 France did likewise,40 except unlike its Western counterparts, it did not spell out any measures through which the positive security assurances might be implemented. UNSCR 984 noted these national statements, invited member states to provide specific forms of assistance in the event of an act of nuclear aggression, and welcomed the actions of those states that had indicated they would provide support to, or assist, any state that was the object of nuclear aggressionY Along with other steps taken by the NWS to assuage NNWS concerns, these efforts to enhance security assurances proved adequate to secure the indefinite extension of the NPT. Desire not to weaken the nonproliferation regime led non-nuclear states to go along with making the treaty permanent despite the lack of a common, joint statement on security assurances from the NWS, the vague and modest content of the positive security assurances offered by most of them, and the failure to provide the assurances in a legally binding instrument. 42 The Conference also saw the new South African regime take over from Nigeria the role of promoting negative security assurances, while Egypt and the Arab states lost the previous guaranteed support of the African states for seeking to enhance positive security assurances at NPT meetings.

68

JOHN SIMPSON

Two further developments at this meeting helped secure indefinite extension of the NPT. A resolution on the Middle East sponsored by the three depositary NWS called upon all parties to cooperate and exert "their utmost efforts" to ensure the early establishment of a "Middle East zone free of nuclear and all other weapons of mass destruction and their delivery systems:'43 The second important development was agreement on a "strengthened" review process,44 and a set of Principles and Objectives for Nuclear Nonproliferation and Disarmament. 4s The latter included a statement that "further steps should be considered to assure non-nuclear-weapon states party to the Treaty against the use or threat of use of nuclear weapons. These steps could take the form of an internationally legally binding instrument:' Important developments also took place before and after the 1995 NPT Review Conference regarding nuclear-weapon-free zones. The NWFZ Treaty of Rarotonga,46 covering the South Pacific, had been signed by the regional states in 1985. The end of apartheid and the accompanying change of regime in South Africa facilitated negotiation of an NWFZ covering the whole of Africa, and the resulting Treaty of Pelindaba47 was signed in April 1996. The Treaty of Bangkok,48 creating a NWFZ covering Southeast Asia, was signed in December 1995. Amendments to the Treaty of Tlatelolco in November 199049 enabled Argentina and Brazil, previously listed with South Africa as "threshold states:' to ratify it in 1994. These regional developments led to all the NPT NWS signing the relevant security assurance protocols to Tlatelolco, Rarotonga, and Pelindaba by 1996. The Treaty of Bangkok remained unsigned because of disagreements with the regional states over both the method of delineating the zone and whether its security assurances would still apply to states that withdrew from the treaty. Although the United States has signed the protocols to the African and South Pacific NWFZs, as of this writing the U.S. Senate has not ratified either of them. However, Secretary of State Hillary Clinton announced in a speech to the NPT Review Conference on May 3, 2010, that the Obama administration was submitting them to the Senate for ratification, and would also consult with the parties to the NWFZs in Central and Southeast Asia in an effort to allow the United States to sign their protocols as well. so Progress in negotiating NWFZs has led some observers to suggest they could be a promising mechanism for providing legally binding negative security assurances,S1 but in practice the NPT has remained the central focus for such efforts.

THE ROLE OF SECURITY ASSURANCES

69

Security Assurances, 1996-2010

The years following NPT extension proved frustrating to those NNWS hoping for rapid progress toward more uniform and binding security assurances from the NWS. The 1997 meeting of the Preparatory Committee (PrepCom) for the 2000 NPT Review Conference discussed three clusters of issues which mirrored the tasks allotted to the three Main Committees of the 1995 Conference. It was agreed that "special time" would be given at its 1998 meeting to a focused discussion of a specific topic within each cluster, one being security assurances. 52 Also, in March 1998 the CD established an Ad Hoc Committee on "effective international arrangements to assure non-nuclear-weapon States against the use, or threat of use of nuclear weapons."53 However, this committee rapidly became deadlocked in arguments over the CD's work-plan. In the late 1990S, U.S. policy began to weaken previous negative assurances. In December 1997, U.S. President Clinton issued Presidential Directive 60, which appeared to preserve the option of U.S. retaliation with nuclear weapons against an attack involving chemical or biological weapons (CBW). This was made transparent in November 1998 when U.S. Defense Secretary Cohen, speaking in a NATO context, stated that "the ambiguity involved in the issue of the use of nuclear weapons contributes to our own security, keeping any potential adversary who might use either chemical or biological [weapons] unsure of what our response would be:'54 This posture of "calculated ambiguity;' intended to stretch nuclear deterrence to cover prevention of chemical or biological attacks, appeared to apply to all states including those without nuclear weapons. A clear contradiction had emerged between the negative security assurances given in 1995 by the three NATO NWS and that organization's declaratory policy of retaining an option for first-use of nuclear weapons in CBW scenanos. The 1998 NPT PrepCom, despite the promised discussion of security assurances, ended without any substantive outcome. In 1999, South Africa circulated a working paper containing a "Draft Protocol on the Prohibition of the Use or Threat of Use of Nuclear Weapons Against Non-Nuclear-Weapon States Parties to the Treaty on the Non-Proliferation of Nuclear Weapons."55 This was skillfully crafted to contain language taken from previously agreed NWS statements, NPT documentation, and Security Council resolutions, and was a consolidation in legal form of commitments already made by NPT parties. The Draft Protocol:

70

JOHN SIMPSON

• contained both negative and positive assurances by both the NWS and NNWS parties to the NPT (reflecting the fact that even NNWS could offer positive assurances involving non-nuclear means of assistance to states threatened or attacked with nuclear weapons); • would void assurances if a state was involved in the type of armed aggression listed in the 1995 assurances of France, Russia, the United Kingdom, and the United States; • would only cover NPT NNWS in compliance with Article II of the treaty; • was to be signed without reservation or conditions; • made no mention of any form of military assistance; and • specified that all NPT parties would undertake to cooperate with the UNSC in responding to an act or threat of nuclear aggression. A second relevant development was the formation in 1998 of the New Agenda Coalition (NAC), comprising Brazil, Egypt, Ireland, New Zealand, Mexico, South Africa, and Sweden, states that bridged the traditional Eastern, Western, and NAM caucus groupings. Its focus was primarily on nuclear disarmament. Partly as a consequence, the only two topics assigned to subsidiary bodies for focused discussion at the 2000 Review Conference were disarmament and a Middle East NWFZ, but not security assurances. Security assurances still figured in the NPT review meeting in 2000, but in a way that suggested they were declining in salience. The Final Document agreed by the 2000 Review Conference prefaced its discussion of security assurances by first reaffirming that "the total elimination of nuclear weapons is the only absolute guarantee against the use or threat of use of nuclear weapons:'56 It then noted that "legally binding security assurances by the five nuclear-weapon States to the non-nuclear-weapon States parties to the ... NPT strengthen the nonproliferation regime:' The only actions requested, however, were limited to calling upon "the Preparatory Committee to make recommendations to the 2005 Review Conference on this issue;' and requesting relevant states to take steps "to bring into effect the assurances provided by nuclear-weapon-free zone treaties and their protocols:'57 The Final Document also called upon all parties to report to NPT meetings on their actions to implement a NWFZ in the Middle East, and for the UN Secretariat to collate these reports.58 The 2000 Review Conference signaled two changes in the way many NNWS were dealing with the issue of security assurances. One was that it appeared more than ever to be a secondary issue for the NAM in comparison with nuclear disarmament. The

THE ROLE OF SECURITY ASSURANCES

71

second was that the Arab states, particularly Egypt, were now giving priority to nuclear disarmament and a NWFZ in the Middle East as methods of responding to Israel's nuclear weapons, rather than seeking enhanced NPT security assurances. The issue did not go away, however, and after the 2000 Review Conference, u.s. policy generated increased consternation among those states promoting enhanced security assurances. In December 2001, in the aftermath of the 9/11 terrorist attacks, the Bush administration completed its Nuclear Posture Review (NPR). Administration briefings and leaked portions of the classified document confirmed continuing u.s. willingness to consider a nuclear response to a CBW attack. A State Department spokesperson stated: "If a weapon of mass destruction is used against the United States or its allies, we will not rule out any specific type of military response:' The NPR also hinted at a new willingness to use nuclear weapons pre-emptively. This led two observers to conclude that the "bunker-busting nuclear weapons" proposed by the previous administration would be used "to destroy biological [or chemical] weapons stored underground by an enemy. This position assumed first use of nuclear weapons in that engagement:' This position seemed to repudiate previous U.S. negative securityassurances. 59 Little discussion on security assurances occurred during the 2002 PrepCom for the 2005 NPT Review Conference, but the issue attracted more attention in the 2003 session, with South Africa's 1999 Draft Protocol being inserted into a larger working paper produced by the NAC states. 60 In the covering language, it was emphasized that negotiating such assurances under the umbrella of the NPT, rather than in the CD, would "provide a significant benefit to the Treaty parties and would be seen as an incentive to those who remain outside the NPT:' The working paper also listed the key elements that would have to be part of a legally binding instrument: • identification of the states providing the assurances and their beneficiaries; • the nature and scope of the assurances being provided; • other elements that would need to be included, such as qualifications to the assurances; • provisions on mandatory actions to be taken by the UN Security Council; and • the format in which the assurances would be provided.

72

JOHN SIMPSON

In the 2004 PrepCom session, the NAM wished to see disarmament and security assurances assigned to separate subsidiary bodies at the upcoming Review Conference. Calls for strengthened security assurances were made by China, Cuba, Indonesia, Iran, Malaysia (on behalf of the NAM), Mexico (on behalf of the NAC), and NigeriaY Some called for an unconditional legally binding document, while others stressed the more tactical need for a subsidiary body on the subject at the Review Conference. In addition, working papers were submitted by the Arab league,62 the ASEAN countries,63 China,64 and jointly by Belgium, The Netherlands, and Norway.65 All stressed the significance of security assurances to both address the concerns of NNWS and strengthen the nonproliferation regime. At the 2005 Review Conference, China, in a working paper, again sought to link legally binding negative security assurances to its no-first-use policy and re-establish an ad hoc committee on security assurances in the CD.66 Iran also took a strong stand on negative security assurances. It argued that the 1995 unilateral statements on security assurances and UNSCR 984 were "inseparable parts of the deal over the indefinite extension of the treaty"; that the effect of the u.S. 2001 Nuclear Posture Review was to place "the non-nuclear-weapon states more than ever under the real threat of use of nuclear weapons"; and that "efforts for the conclusion of a universal, unconditional and legally binding instrument on security assurances to non-nuclear-weapon states should be pursued as a matter of priority by the international community:' It also proposed creating an ad hoc committee to draft language that would enable the 20lO Review Conference to adopt a decision that "the threat or use of nuclear weapons against non-nuclear weapon states shall be prohibited:'67 Given deep divisions, especially between Iran and the United States, the 2005 Review Conference ended in failure. Although the NAM argued for separate subsidiary bodies on disarmament and security assurances, it was eventually agreed to have both subjects addressed together in the subsidiary body attached to Main Committee 1. The Conference was unable to agree on the text of a Final Document, however, and thus no recommendations for action on anything emerged from this meeting. 68 Sharp disagreements about security assurances were one of the reasons for failure of the Conference. In particular, the United States refused to engage on assurances. As a U.S. representative at the Conference stated: [Tlhe end of the Cold War has further lessened the relevance of non-use assurances from the P-5 to the security ofNPT [non-nuclear-weapon states], particularly when

THE ROLE OF SECURITY ASSURANCES

73

measured against the real nuclear threats from NPT violators and non-state actors . .. . [Llegally binding assurances sought by the majority of states have no relation to contemporary threats to the NPT. 69

Efforts to prepare for the next NPT Review did not initially appear to promise much progress. The 2007 PrepCom session was deadlocked for much of its allocated time by procedural issues raised by Iran. Both ie o and the NAC 71 put forward working papers on security assurances based on their previous positions. Italy,72 Canada,?3 China,?4 and the Republic of Korea?5 also offered their thoughts on the matter. The only action proposed was to submit to the 2010 NPT Review Conference the draft of a legal instrument, but no procedure to do this was discussed during either this or the 2008 Prep Com session. At the 2009 meeting Iran produced a working paper which again claimed that the 1995 assurances and UNSC resolution were "inseparable parts" of the 1995 extension arrangements and proposed that "the upcoming conference establish an ad hoc committee to work on a draft of a legally binding instrument on the illegality of nuclear weapons and the provision of security assurances ... and to submit the draft of the legal instrument to the next Review Conference for consideration and adoption:' It also proposed that the 2010 Review Conference should adopt a decision that "the threat or use of nuclear weapons against non-nuclear-weapon States shall be prohibited."?6 Although this Prep Com session did not agree on substantive recommendations, the final draft of its proposed "Recommendations to the Review Conference"??clearly indicated the reduced salience of security assurances. Only two sentences were devoted to this issue in one of five subparagraphs of the section on "nuclear disarmament, including specific practical measures:' These recommended two actions for the Review Conference: ''Affirm the importance of effective assurances that nuclear weapon States will not use or threaten to use nuclear weapons against non-nuclear weapon State parties. Examine ways and means to achieve additional assurances that are legally binding." By contrast, the Middle East resolution was given a separate section, as were NWFZs. Actions taken by the United States following the election of Barack Obama as president helped create greater optimism about the prospects for a successful Review Conference in 2010 and new movement on security assurances. In his speech in Prague in April 2009, Obama recommitted the United States to the goal of eventual nuclear disarmament. The Obama administration also gave fresh impetus to the issue of security assurances shortly before the 2010 Review Conference when it unveiled its April 2010 Nuclear Posture Review Report.78

74

JOHN SIMPSON

This announced that the United States was making a change to "strengthen its long-standing 'negative security assurance' by declaring that the United States will not use or threaten to use nuclear weapons against non-nuclear weapon states that are party to the NPT and in compliance with their nuclear nonproliferation agreements:' The new assurance sought to alleviate concerns about U.S. policy that arose after the 1995 NPT extension by indicating that "any state eligible for the assurance that uses chemical or biological weapons against the United States or its allies and partners would face the prospect of a conventional military response:' The NPR did not rule out, however, the possibility of "any adjustment in the assurance that may be warranted by the evolution and proliferation of the biological weapons threat:' The NPR also made explicit those states that were excluded from the assurance (NPT NNWS in noncompliance with their obligations), thereby removing the previous exception for states allied or associated with an NWS. In practice, the Obama NPR could be interpreted as offering unconditional security assurances to all states except the eight declared nuclear weapon states, Israel, and Iran. The strengthened negative security assurance in the NPR, along with other steps like the signing of the New START treaty, helped facilitate a largely successful NPT Review Conference in May 2010. The opening statement by U.S. Secretary of State Hillary Clinton, in which she announced the president would submit the protocols to the African and South Pacific NWFZ treaties to the U.S. Senate for ratification, also helped. The NPT Review Conference agreed to three Actions with respect to security assurances. Action 7 asserted that the CD in Geneva should "immediately begin" discussions on all aspects of security assurances. Action 8 committed "all nuclear-weapon states" to respect existing commitments, and those who had yet to do so to "extend security assurances to non-nuclear-weapon States parties to the Treaty:' Action 9 encouraged all "concerned States" to ratify the nuclear-weapon-free-zone treaties and their relevant protocols; to "constructively consult and cooperate" to bring into force their "relevant legally binding protocols"; and to "review any related reservations:'79 Action 7 appeared to imply a switch of the focus for future action away from an NPT-specific forum to the CD global one, while Action 8 left opaque whether future actions should be limited to the five NPT nuclear-weapon-states or these plus the three (or four if one counts North Korea) states outside the treaty. Despite agreement on these Actions, security assurances were not a major focus at the Review Conference. Instead, the main NAM state concerns in-

THE ROLE OF SECURITY ASSURANCES

75

volved what they perceived as inadequate progress on disarmament and the lack of negotiations on a Middle East zone free of weapons of mass destruction. Only intense negotiations to create an acceptable compromise on the latter issue enabled the Conference to conclude with successful adoption of a final document. 8o During the Review Conference, the United Kingdom was in the middle of a general election campaign. Toward the end of the Conference, its new foreign secretary announced the initiation of a review of the UK's declaratory policy on security assurances. When completed in October 2010, the review stated that "the UK will not use or threaten to use nuclear weapons against non-nuclear weapon states parties to the NPT. ... [T]his assurance would not apply to any state in material breach of [its NPT] nonproliferation obligations:'81 This assurance was similar in content and effect to the new U.s. one. The recent revisions to the U.s. and UK negative security assurances still do not go as far as NAM states have long requested, but, given the lower attention devoted to assurances in recent NPT Review Conferences, these efforts may be sufficient to keep the issue from becoming more fractious again for the time being. This still leaves open the question of what further steps could be taken in the years ahead. THE PRACTICAL PROBLEMS OF PROVIDING ENHANCED SECURITY ASSURANCES: AN ANALYSIS

Over the last forty years there have been multiple attempts to craft both global and regional security assurances to strengthen the attractiveness and credibility of the NPT and its linked nonproliferation regime to NNWS. The issues and possible options for enhancing them are now relatively transparent and to be found in the following areas: i. The Forms of Assurance

Four types of security assurance have been provided or proposed: • nuclear security guarantees (that is, extended nuclear deterrence) provided through multilateral or bilateral defense agreements; • regional NWFZ agreements; • unilateral negative and positive nuclear security assurances provided through declarations by NPT NWS to its NNWS states party; and • common negative and positive nuclear security assurances provided through a legally binding international convention, treaty, or agreement.

76

JOHN SIMPSON

Realistically, further development of the first option appears unlikely on a multilateral basis. It is possible that individual NWS would consider extending a nuclear umbrella to certain new states on a selective basis, but some states in the world would still be left out of such guarantees. With respect to option two, further moves to deliver negative security assurances to NNWS in a legally binding form via NWFZ agreements seem feasible, providing the u.s. Senate is prepared to ratify their protocols. Unless progress can be made on the proposed Middle East zone, however, expanding the geographic coverage of NWFZs appears unlikely for the time being. Given that some states will be left out of regional zones, many NNWS remain focused on a global approach. Their preference is for option four, the negotiation of a new international instrument. However, the Actions agreed at the 2010 Conference appear to be moving the focus for action on a global legally binding agreement from an NPT forum to the CD, whose track record of negotiating nuclear arms control agreements gives little grounds for optimism. If attention reverts to option three, the key demand by NNWS will be for the standardization and harmonization of existing NPT NWS unilateral statements. While there may be some room for moving these closer to the recent u.s. and UK reformulations, it appears unlikely that these two assurances can be merged with those of China and the Russian Federation into a single agreed joint statement, especially because the other four NWS are disinclined to embrace China's unconditional no-first-use pledge. In sum, all four types of vehicle for providing security assurances face limitations with respect to satisfying NNWS wishes. ii. The Substance and Beneficiaries of Assurances

Existing arrangements and proposals for enhancing nuclear security differ significantly over substance. These differences fall into two categories: the scope of the commitment and the beneficiaries. The first includes: • • • •

negative nuclear assurances; positive nuclear assurances; nuclear no-first-use agreements; and guarantees linked to stationing of or access to NWS nuclear weapons. The second could include:

• all current NPT (and NWFZ) parties; • states accepting integrated IAEA safeguards (that is, INFCIRC 153 plus 540),82 perhaps limited only to those declared in compliance with their obligations;

THE ROLE OF SECURITY ASSURANCES

77

• states without nuclear weapons stored or stationed on their territories; and • states without formal alliance relationships with nuclear weapon states. In the first category, following the recent revisions of declaratory policy by the United States and United Kingdom, the most promising way forward appears to be in the area of negative assurances. It has proven difficult to specify content for globalized positive assurances, and most NWS are not now willing to embrace no-first-use or to deploy nuclear weapons to the territories of allies where they are not currently stationed. In the second category, agreements of the third and fourth type seem less likely to make progress than those offering additional benefits to states of the first and second type. Limiting security assurances to NPT parties, or the subset of those with integrated IAEA safeguards, would fall in the intersection of recent trends in U.S. and UK policies and the preferences of many NPT NNWS. iii. The Four Nonparties to the NPT and Security Assurances

The current situation with security assurances has some parallels with the pre-1993 situation, when China and France were nonparties to the NPT. Today, only India, Israel, North Korea, and Pakistan are nonparties, but none are formally recognized as nuclear weapon states. This confronts NPT parties with a series of difficult, but currently hypothetical, questions in relation to security assurances. These include: • is the recognition of nuclear status implied by seeking security assurances from the NPT nonparties unacceptable to NPT parties? • would assurances offered by them through the UNGA, UNSC, or through negotiations in the CD be regarded as acceptable by NPT parties? • would it be more acceptable to obtain enhanced security assurances through a universal no-first-use agreement? and • would members of NWFZs be prepared to accept security assurances from the non-NPT states? To date, these questions have received almost no attention, as discussion of security assurances remains focused on NPT members. FUTURE SCENARIOS: THINKING OUTSIDE THE BOX

Two key questions now appear to be emerging in relation to nuclear security assurances through NPT channels and nuclear security guarantees through NWS extended deterrence commitments. One is whether the ongoing inter-

78

JOHN SIMPSON

linked debates over these issues in NPT meetings are outmoded relics from the bipolar past, and as such have little military significance today. They clearly continue to have symbolic diplomatic significance as part of the bedrock of NAM and NATO political solidarity, respectively, and moving away from existing positions may be perceived to threaten the future existence of these institutional arrangements. A second question is whether the more positive attitude toward nuclear disarmament that became evident following Obama's Prague speech in 2009 is generating a rethinking of attitudes toward these issues by both nuclear and non-nuclear NPT states. Is attention switching from nuclear security assurances to a disarmament process that could involve both NWS commitments and actions and the creation of further regional NWFZs? And is the new politics of nuclear security assurances going to be one of constructive partnership, rather than conflict and friction? Within the NGO community, some rethinking is already occurring. Rebecca Johnson of the UK-based Acronym Institute offers an interesting set of ideas on what the substance of this new politics might involve. She recognizes the inherent contradictions in the demands by NAM states for assurances, as this creates a situation in which they "still treat the NPT-recognized NWS as both primary threat and primary source of assistance:' She also highlights the need for NPT action to "make the security assurance regime apply to actors outside the NPT, without appearing to confer additional status on non-NPT nuclear weapon possessors India, Israel and Pakistan:' She rejects a no-first-use agreement as the solution, as it would "legitimize the retaliatory use of nuclear weapons when deterrence fails:' Her proposal is to follow the path created by UNSCR 1540, which sought to confront the threat of nuclear terrorism by calling on states to enact domestic legislation to criminalize those individuals and groups engaged in or assisting such activities. In place of existing policies on interstate security assurances, she would substitute a single package of integrated measures to address both nuclear nonproliferation and nuclear terrorism, and to include both NPT parties and nonparties. This would involve the international community declaring "the use of nuclear weapons is a crime against humanity;' criminalizing its use, and bringing those involved before the International Criminal Court in order to "stigmatize and, in effect outlaw the use of nuclear weapons for everyone:'83 This drive to rethink security assurances has been triggered by three sets of events. The first has been the collapse of the Eastern Bloc. With this de-

THE ROLE OF SECURITY ASSURANCES

79

velopment, the conventional threat to Europe has been largely removed, and with it the resultant military need to station nuclear weapons on the territories of NATO non-nuclear weapon state allies (although, as discussed shortly, this perspective is not shared by all NATO members).84 The second has been that nuclear security in the world outside of Europe is increasingly regionally based, giving impetus to assurances being sought at the regional rather than global level. The third driver is that only three or four states (depending on how one views North Korea)85 are nonparties to the NPT-and all have nuclear weapons. Thus the relevant question in relation to the regime and the NPT is no longer the role enhanced security assurances may play in enticing states to accede to the treaty, but the role they may play in preventing defections, as new proliferators can now only emerge from among the treaty parties. These changes resulted in two specific questions being raised in relation to the November 2010 review of NATO's Strategic Concept. One was whether current arrangements for aircraft of five NATO NNWS (Belgium, Holland, Germany, Italy, and Turkey) to deliver U.S. nuclear ordnance should be retained (and a new generation of aircraft to carry them procured). The second was whether the nuclear ordnance earmarked for use by u.S. aircraft based in Europe should continue to be stored in Europe or be repatriated back to the United States. 86 In the event, the 2010 Strategic Concept agreed in Lisbon retained the estimated 200 U.S. tactical nuclear warheads deployed in Europe, while making some modest changes in language to reduce the importance attached to such weapons. 87 However, even if NATO Europe were at some future date to move away from nuclear security guarantees, increased assertiveness by China and renewed tensions on the Korean Peninsula make it unlikely that U.S. allies in East Asia would be willing to accept a weakening of the U.S. nuclear umbrella in their region. In light of the mixture of both progress and countervailing pressures in 2009-11, two possible "business as usual" strategies built on legacy ideas appear to be ways to move policy forward. One is to develop negative nuclear security assurances and no-first-use agreements on a regional basis, with their content differing in areas such as the Middle East and Northeast Asia. A second is to push for global security assurance agreements that have reservations attached to them in connection with specific regions. It is not difficult to identify a list of possible new proposals for NPT security assurances; what is more difficult, given the historical rhetoric of NAM states and the cost-benefit calculations NWS will surely apply to new assurances, is

80

JOHN SIMPSON

to ascertain whether they are capable of implementation. Such a list might include: • more NWFZs; • a universal nuclear no-first-use agreement; • a joint declaration by the NPT NWS containing both positive and negative enhanced security assurances for NPT parties (who are in compliance with the treaty?); and • a package deal in the form of an international agreement or treaty where non-nuclear NPT parties agree to restate their pledges of non-nuclear weapon status, make nonaggression pledges, and/or accept enhanced transparency in their peaceful nuclear activities in return for strengthened negative and positive security assurances from the NPT NWS. For many states the issue of enhanced negative security assurances appears to remain on the NPT agenda largely for historic and political reasons. This may mean that declaratory policy changes offering political benefits could now be provided at minimal cost to the national security of the NWS. The 2010 U.S. enhancement of its unilateral negative security assurance and the similar step taken by the UK are prime examples of this. The U.S. NPR indicated a belief that states hostile to it and its allies will not feel free to use other weapons of mass destruction if NWS commit themselves to negative nuclear security assurances. Should this be regarded as part of an ongoing natural process of marginalization of the utility of nuclear weapons in both military and political contexts-or a sacrificing of the nuclear security future to the diplomatic present? On positive security assurances, the situation is also changing. National planning for dealing with the consequences of nuclear terrorism has made NWS better equipped to offer practical assistance in the event of nuclear use. Moreover, development of mobile conventionally armed theater missile defense systems offers an alternative hardware route to deter proliferators that might substitute for security guarantees involving extended nuclear deterrence. This could encourage collective discussions on the contemporary utility and relevance of the extended nuclear deterrence guarantees originating in the context East-West conflict. Could they be internationalized through a consultation process among the five NPT NWS, rather than offered through individual states or regional alliances? And should the future for such forces be seen in the context of a traditional world of nations-or of an evolving global network

THE ROLE OF SECURITY ASSURANCES

81

environment in which, as Johnson suggests, nuclear use is criminalized on the individual level? This chapter has chronicled the tensions between security guarantees extended to allies and the requests of nonaligned states for broader security assurances in association with the NPT and regional NWFZs. The case studies that follow focus mainly on the former. Yet the analysis in this chapter suggests that it would be a mistake to emphasize only extended nuclear deterrence to the exclusion of NPT-related security assurances. Recent efforts to enhance security assurances have been driven by long-standing pressures to improve the diplomatic atmospherics surrounding the nuclear nonproliferation regime and the NPT. The inheritance from past debates makes such policies necessary if progress is to be made in enhancing NPT safeguards and creating more effective mechanisms for dealing with noncompliance. It is vital therefore that both objectives-strengthening assurances and ensuring compliance-continue to be pursued together, rather than be seen as competitive alternatives. For the two competitive cultures operating within the NPT process (prioritizing NWS declaratory and legal commitments to disarmament versus taking practical actions to prevent new proliferation and make progress on arms reduction) must move forward in parallel if international security is to be assured, at least until thinking "outside the box" by the NGO community and others can outflank current security concepts and make security assurances and guarantees unnecessary. NOTES 1. UN Security Council Resolution 255, June 19, 1968, http://f40.iaea.org/worldatom/ Documents/Legal/unsc255.shtml. 2. Arms Control Association, Fact Sheet, "Nuclear-Weapon-Free Zones (NWFZ) at a Glance;' December 2009, http://www.armscontrol.orglfactsheets/nwfz.

3. Much of the description and analysis in this section draws on the author's personal observations while present at all NPT meetings since 1985, and on private discussions with members of relevant delegations. 4. Article 5 states: "The Parties agree that an armed attack against one or more of them in Europe or North America shall be considered an attack against them all and consequently they agree that, if such an armed attack occurs, each of them, in exercise of the right of individual or collective self-defense recognized by Article 51 of the Charter of the United Nations, will assist the Party or Parties so attacked by taking forthwith, individually and in concert with the other Parties, such action as it deems necessary, including the use of armed force, to restore and maintain the security of the North At1antic area:'

82

JOHN SIMPSON

5. U.S.-Japan Mutual Defense Treaty, March 8, 1954, and U.S.-Japan Mutual Cooperation and Security Treaty, January 19, 1960. 6. U.S.-Republic of Korea Mutual Defense Treaty, November 17, 1954. 7. U.S.-Republic of the Philippines Mutual Defense Treaty, August 30, 1951. 8. North Atlantic Treaty (NAT), April 4, 1949. NATO, the organization to implement the military aspects of this treaty, was developed during the 1950S. 9. ANZUS (Australia, New Zealand, and the United States) Treaty, April 29, 1952. 10. Inter-American Treaty of Reciprocal Assistance, September 2, 1947. 11.

Central Treaty Organization, also known as the Baghdad Pact, signed February

24,1955. Members were the UK, Iran, Iraq, Pakistan, and Turkey. 12.

South East Asian Treaty Organization, signed September 8, 1954. Members were

the UK, U.S., Australia, France, New Zealand, Pakistan, Philippines, Taiwan, and Thailand. 13. Warsaw Treaty Organization (also known as the Warsaw Pact), established May 14, 1955. Original members were the USSR, Albania, Poland, Czechoslovakia, Hungary, Bulgaria, Romania, and the German Democratic Republic. 14. Michael O. Wheeler, Positive and Negative Security Assurances, Project on Rethinking Arms Control (PRAC) Paper no. 9 (College Park: University of Maryland, Center for International and Security Studies, February 1994),6. 15. For an account of the evolution of the security assurances issue prior to the start of negotiations on the NPT, see George Bunn, "The Legal Status of U.S. Negative Security Assurances to Non-Nuclear Weapon States;' Nonproliferation Review 4, no. 3 (Spring/Summer 1997): 2-5. 16. The Multilateral Force (MLF) was an idea first proposed by the United States in 1960, and discussions on it continued within NATO through to 1965. There was considerable confusion over whether it would involve Polaris missile submarines, or the deployment of these missiles on multinationally manned surface ships. For a recent account of this debate based on UK archives, see Richard Moore, Nuclear Illusion, Nuclear

Reality: Britain, the United States and Nuclear Weapons, 1958-1964 (Basingstoke: Palgrave Macmillan, 2010). 17. The Allied Nuclear Force (ANF) was proposed by the incoming UK Labour Government in 1964 as an alternative to the MLF. 18. George Bunn, Arms Control by Committee (Stanford: Stanford University Press, 1992),59-105· 19. Emily Bailey, "The NPT and Security Guarantees," in Nuclear Non-Proliferation:

A Reference Handbook, ed. Darryl Howlett and John Simpson (Harlow, UK: Longman Current Affairs, 1992), 51-56. 20. For a succinct account of these negotiations, see Thomas Graham and Leonor Tomero, "'Obligations for Us All': NATO & Negative Security Assurances;' Disarmament

Diplomacy, no. 49, August 2000, http://www.acronym.org.uk!dd//dd49/49nato.htm.

THE ROLE OF SECURITY ASSURANCES

83

21. See note 1, above. 22. This analysis originated in a paper on "Nuclear Security Guarantees and Assurances as a Method of Reinforcing the NPT" prepared by the author for a discussion by the PPNN Core Group at their meeting in Southampton in July 1993. 23. For text, see Mountbatten Centre for International Studies (MCIS) and Center for Nonproliferation Studies (CNS), NPT Briefing Book, 2010 Edition, Section G, 1-6, accessible at www.mcis.soton.ac.uk. 24. UN General Assembly, Ad Hoc Committee of the Tenth Special Session, document A/S-1O/AC.1!I7, June 7, 1978, Annex para. 7, quoted in ibid., Section L, 1. 25. Official Records of the General Assembly, Tenth Special Session, Plenary Meetings, 27th meeting, June 30, 1978, para. 190, quoted in ibid. 26. Official Records of the General Assembly, Tenth Special Session, Plenary Meetings, 5th meeting, May 26, 1978, paras. 84 and 85, quoted in ibid., 2. 27. Official Records of the General Assembly, Tenth Special Session, Plenary Meetings, 26th meeting, June 28, 1978, para. 12, quoted in ibid. 28. A/C.1/33/7, November 17, 1978, Annex, quoted in ibid. 29. A/S-12/11, April 28, 1982, quoted in ibid., 1.

30. Official Records of the General Assembly, Twelfth Special Session, Plenary Meetings, 9th meeting, June 11, 1982, quoted in ibid. 31. Official Records of the General Assembly, Twelfth Special Session, Plenary Meetings, 12th meeting, June 12, 1982, quoted in ibid., 2. 32. Ibid., Part 1, 10. 33. NPT/CONF.IV/PC.III/19, April 27, 1990. 34. CD/SA/WP.13,August 6, 1991. 35. Programme for Promoting Nuclear Non-Proliferation (PPNN), Briefing Book,

eighth edition (MCIS, University of Southampton, 2000), Section K, 5-8, http://www. mcis.soton.ac. uk/programmes/Bb2e8fin. pdf. 36. MCIS/CNS, NPT Briefing Book, 2010 edition, Section L, 1.

37. Ibid., 2. 38. Ibid., 3. 39. Ibid., 2-3· 40. Ibid., 1-2. 41. Ibid., 3-4. 42. Bunn, "The Legal Status of U.S. Negative Security Assurances" (12), however, argues both that they are "politically binding;' and points out that the International Court of Justice in its 1996 advisory opinion on The Legality of the Threat or Use by a State of Nuclear Weapons concluded unanimously that they should be observed. 43. NPT/CONF.1995/32/RES.l, quoted in MCIS/CNS, NPT Briefing Book, 2010 edition, Section E, 2-3. 44. NPT/CONF.1995/32/DEC.1, quoted in ibid., 1.

84 JOHN SIMPSON

45. NPT/CONF.1995/32/DEC.2, quoted in ibid., 1-2. 46. MCIS/CNS, NPT Briefing Book, 2010 edition, Section G, 6-10. 47. Ibid., 10-15. 48. Ibid., 15-19. 49. Ibid., 2-5· 50. See www.state.gov/secretary/rmi2010/05h41424.htm. 51. Leonard S. Spector and Aubrie Ohlde, "Negative Security Assurances: Revisiting the Nuclear-Weapon-Free Zone Option;' Arms Control Today 35, no. 3 (April 2005). 52. John Simpson, "The Consequences of the 1997 PrepCom and Their Implications for the Review Process;' in South EastAsia: Regional Security and Nuclear Non-Proliferation (Southampton: Programme for Promoting Nuclear Non-Proliferation, 1997). 53. "Conference on Disarmament Concludes 1998 Session;' press release, DCFf350, September 10, 1998, available at http://www.fas.orginews/un/otherh998091O_dc650. html. 54. Graham and Tomero, "Obligations for Us All:' 55. NPT/CONF.2000/PC.III/9, May 11, 1999. 56. NPT/CONF.2000i28 (Part!) Article VII and the Security of non-nuclear-weapon States, Para 2. 57. Ibid. 58. Ibid., Regional Issues: The Middle East, particularly implementation of the 1995 Resolution on the Middle East, Para 7. 59. George Bunn and Jean du Preez, "More than Words: The Value of U.S. NonNuclear-Use Promises;' Arms Control Today 37, no. 6 (July/August 2007), quotes at 18. 60. NPT ICONF.2005IPC.IIIW.P.11. 61. For a listing of relevant papers, see NPT ICONF.2005.1, Final Report of the Preparatory Committee for the 2005 Review Conference. 62. NPTICONF.2005IPC.III/W.P.12. 63. Ibid., 21. 64. Ibid., 9. 65. Ibid., 25. 66. NPT/CONF.2005IW.P.7. 67· Ibid., 49. 68. Rebecca Johnson, "Politics and Protection: Why the 2005 NPT Review Conference Failed;' Disarmament Diplomacy, no. 80 (Autumn 2005). 69. Statement by Stephen G. Rademaker, quoted in Bunn and du Preez, "More than Words;' 19. 70. NPT ICONF.201O/PC.IIW.P.68,NPT ICONF.2010/PC.IIIW.P.25;NPT ICONF.20101 PC.III/W.P.2. 71. NPTICONF.201O/PC.I/W.P.11. 72. Ibid., 27.

THE ROLE OF SECURITY ASSURANCES

85

73. Ibid., 29· 74· Ibid., 43· 75. Ibid., 72. 76. NPT ICONF.201O/PC.III/W.P.2. 77. NPT ICONF.201O/PC.III/CRP 4/Rev.2. 78. Nuclear Posture Review Report, quoted in MCIS/CNS, NPT Briefing Book, 2010 edition, Section V, 32-35. 79. NPT/CONF.2010/50 (VO!'l), 22. 80. William Potter et a!., "The 2010 NPT Review Conference: Deconstructing Consensus;' CNS Special Report, June 17, 2010, http://cns.miis.edu/stories/pdfs/1006l7_npC 201O_summary.pdf. 81. "Securing Britain in an Uncertain Future: The Strategic Defence and Security Review;' HM Government Cm 7948, October 2010, Part Three: The Deterrent, 3.7, 3738. 82. IAEA Information Circular (INFCIRC) 153 describes the comprehensive safeguards agreement originally required of NPT NNWS, while INFCIRC 540 contains the Additional Protocol to strengthen safeguards developed in the early 1990S after the discovery of Iraq's clandestine nuclear weapons program. 83. Rebecca Johnson, "Security Assurances for Everyone;' Disarmament Diplomacy, no. 90 (Spring 2009). 84. Michael Chalmers, Nuclear Narratives: Reflections on Declaratory Policies, Whitehall Report 1-10 (London: RUSI, May 2010). 85. NPT states have differing views on whether the DPRK remains a party to the NPT or has withdrawn. This is because it did not take all the actions required for it to legally withdraw from the treaty. 86. Information in the public domain suggests that in practice those weapons that remain in Europe are at a low level of readiness, and in very limited numbers. Hans M. Kristensen, U.S. Nuclear Weapons in Europe: A Review of Post-Cold War Policy, Force Levels, and War Planning (Washington, DC: Natural Resources Defense Council, 2005), available at: www.nrdc.org/nuclear/euro/contents.asp. Indeed the key issue appears in practice not so much to retain them for imminent use in Europe but to sustain the freedom to return them to Europe should the need arise at some future time. 87. Arms Control Association, "Experts Call NATO Strategic Concept 'Missed Opportunity to Reduce Role of Obsolete Tactical Nukes from Europe;" November 19, 2010, http://www.armscontrol.orgipressroom/NATOMissedOp.

Libya, Nuclear Rollback, and the Role of Negative and Positive Security Assurances Wyn Q. Bowen

In 2003, the Libyan regime of Colonel Muammar Gadhafi announced that it had decided to forgo the pursuit and possession of nuclear and other unconventional weapons. The decision followed secret negotiations between Libya, the United Kingdom, and the United States. In his article on nuclear reversal, Ariel Levite notes that reversal decisions are usually influenced by multiple factors. l This was the case with Libya, leading to debate over the relative importance of various factors that contributed to its nuclear rollback, ranging from domestic economic and political pressures to the effects of external pressures and inducements, including security assurances. Given the secretive nature of nuclear decision-making, it is prudent to be cautious about the weight accorded to various factors, especially because the empirical evidence remains incomplete. Two security assurances, the first negative and the second positive, are associated with Gadhafi's decision to give up Libya's WMD programs. The first, and most important, predated the Libyan announcement of December 19, 2003. It involved what Robert Litwak describes as "an implicit assurance of regime survival"-an "assurance of non-intervention"-from Washington. 2 The second set of assurances postdated the announcement by almost two and a half years, although its origins date back to the secret negotiations that preceded Libya's announcement. These assurances took the form of a Joint Letter of Peace and Security signed by the governments of Libya and the United Kingdom in 2006. It included, among other things, a British commitment to assist Libya with its conventional defense requirements and to seek UN Security Council action if Libya were attacked or threatened with chemical or biological weapons. 89

90

WYN Q. BOWEN

This chapter considers the effect of the two sets of security assurances on Libya's rollback decision. The December 2003 announcement makes the Libya case a nonproliferation success story. Security assurances contributed to this success, although their application became possible only following significant changes in Libya's domestic political and external security context. The assurances also had to be combined with other instruments, notably economic incentives, to alter the regime's position on WMD. However, American and British actions taken against the regime in 20n-including involvement in NATO military action to protect the general population in response to the regime's violent repression of popular protest against Gadhafi's rule-highlight the fact that striking a deal in one sphere (for example, WMD) does not preclude a subsequent deterioration in relations if the regime in question later engages in other types of behavior deemed to significantly violate international norms. The chapter draws on primary source documentation and the developing literature on Libya's decision to get out of the nuclear business. 3 It is also informed by more than twenty off-the-record interviews with officials from the United States, the United Kingdom, and the International Atomic Energy Agency (IAEA) with direct knowledge of Libya's decision to renounce WMD, the implementation of the decision, and the subsequent process of re-engagement with Britain and the United States.4 Section one of this chapter describes Libya's pursuit of nuclear and other WMD, the status of the programs in 2003, and the secret negotiation process that started in 1999. The second section examines the implicit American assurance of noninterference with the regime's security as an influence on Libyan decision-making. The third section examines the largely positive assurances provided by the United Kingdom in 2006. This section covers ground hitherto untouched in the academic literature. The final section relates the Libya case to the hypotheses on security assurances formulated by Jeff Knopf.

BACKGROUND Libya signed the NPT in 1969, ratified it in 1975, and concluded a Comprehensive Safeguards Agreement with the IAEA in 1980. Despite these commitments, the Gadhafi regime pursued the clandestine acquisition of a nuclear weapons capability from the early 1970S until 2003. This included being a customer of the A. Q. Khan proliferation network during the 1990S, which enabled Libya to procure many of the building blocks of a nuclear weapon program including key elements of a uranium enrichment program. Despite importing

LIBYA, NUCLEAR ROLLBACK, AND SECURITY ASSURANCES

91

significant volumes of sensitive equipment and materials, Libya did not make significant progress in its quest to develop a nuclear weapon. This is explained in part by the country's lack of indigenous technical expertise to effectively harness the acquisitions it made. 5 Other factors limiting progress include the fact the program was not well managed, the wariness of other countries about working with Libya on nuclear issues, and the effect of the multilateral embargo on the country from the early 1990s. 6 After Libya's decision to get out of the WMD business, American and British officials took charge of inventorying, dismantling, and shipping out of Libya tlIe key elements of the nuclear weapons program, with tlIe IAEA verifying the process. On the chemical weapons front Libya's program was not as far advanced as had been suspected prior to 2003. Nevertheless, the regime possessed 24,576 metric tons of sulphur mustard, 1,400 metric tons of precursors for mustard and nerve agents, and 3,563 empty chemical bombs.? The Organization for the Prohibition of Chemical Weapons (0 PCW) took charge of verifying and supervising the destruction of the chemical program. In the biological field tlIe regime's program has been described as "little more than a plan that had made minimal progress:'B Libya also agreed to get rid of 80 SCUD-B launchers and some 240 missiles, plus five SCUD-C missiles. 9 Libya's WMD Rationale

Libya's weapons ambitions primarily reflected security and political drivers.lO As Braut-Hegghammer argues, the regime's rationale for developing nuclear weapons included an early desire-following the coup that brought Gadhafi to power in 1969-to develop a "national deterrent strategy"; a nuclear deterrent was seen to be "a technological 'fix' requiring little manpower yet enabling the regime to protect Libya's vast territory from external threats:'ll The Gadhafi regime viewed threats, however, tlIrough tlIe prism of regime security. Decision-making in Libya hence reflected the requirement of deterring and defeating tlIreats whetlIer externally or internally generated. 12 Israel's nuclear capability also appears to have influenced Libyan thinking from the early 1970S, partially in terms of security but also, and perhaps more importantly, in terms of the regime seeking to make political capital in the context of the Arab-Israeli conflictY The pursuit of nuclear weapons was evidently perceived as a means to promote a vision of Libyan leadership in tlIe Arab world. According to Braut-Hegghammer, a contributing factor appears to have been Gadhafi's "disappointment over not having been consulted by the

92

WYN Q. BOWEN

Arab war coalition in 1973:' Simultaneously, increased oil earnings accompanying the oil crisis presented the wherewithal to finance the development of Libya's nuclear sector.14 It is difficult to dispute Braut-Hegghammer's observation that, "[flrom the outset it appears that the desire to obtain nuclear weapons was primarily rooted in a political vision of how this would elevate Libya's international role and more vague assumptions about how this would help protect the Libyan regime:'ls However, Libya's deteriorating relationship with the United States during the 1980s-because of Tripoli's support for international terrorism-appeared to change the relative emphasis of the motivating factors. The American air strikes against Tripoli and Benghazi in 1986 in retaliation for terrorist acts linked to Libya were a sobering demonstration of the regime's vulnerability to external intervention, particularly because leadership targets were hit. As a result, the security rationale became more important than the political one. 16 Concerns about security, principally in the context of American hostility, continued into the 1990S. For example, in a much quoted speech in 1990, Gadhafi spoke about the 1986 attack, stating that, if Libya had then "possessed a deterrent-missiles that could reach New York-we would have hit it at the same moment!' He also argued that Libya and the wider Arab world should develop such a deterrent including nuclear weaponsY The subsequent implication of Libya in the bombing of Pan Am flight 103 over Lockerbie, Scotland, in 1988, which resulted in UN sanctions in the early 1990S on top of an existing American embargo imposed in the 1980s, increased Libyan isolation and the regime's sense of confrontation with Washington. It has been argued that the Gadhafi regime subsequently began to pursue a "two track" policy during the 1990S. On one hand this involved seeking to enhance relations with the United States and the West more generally, in order to reduce international pressure and have the sanctions removed. On the other hand it involved continuing to work on the nuclear program as a form of insurance just in case the first track failed. IS In terms of enhancing relations, the Gadhafi regime agreed in August 1998 to a proposal by the United States and the United Kingdom to put on trial two Libyans suspected of the Lockerbie bombing. The regime was rewarded with UN Security Council Resolution 1192, which provided for the suspension of multilateral sanctions when the suspects were surrendered. 19 After the suspects were handed over for trial in April 1999, UN sanctions were suspended, and in July the UK re-established diplomatic relations. In spring 1999, the United States, United Kingdom, and Libya also initiated secret negotiations to resolve

LIBYA, NUCLEAR ROLLBACK, AND SECURITY ASSURANCES

93

remaining aspects of the Lockerbie issue, such as compensation. Several rounds of talks took place over the next three years, and it was communicated to Libya on the margins that the WMD issue would need to be addressed before Washington would normalize relations. 20 Libya suggested in 1999 that it was willing to give up WMD, but for reasons explained below, unlike its 2003 announcement this offer was probably not sincere. After September 11, 2001, however, the regime began to perceive its security interests as "increasingly aligned with countries once viewed as adversaries:' with Libya and the United States in particular sharing the same sense of threat from violent Islamist groupS.21 Tripoli began to cooperate with Washington and other Western countries on antiterrorism because of shared concerns about the Al Qaeda threat. This cooperation contributed to improved relations between Libya and its long-time adversaries, the United States and Britain. In 2003, several months after Lockerbie had been resolved and UN sanctions permanently removed, secret negotiations were initiated specifically on WMD. The final outcome was Libya's rollback decision in December. What, then, was the role played by security assurances in this context? SECURITY ASSURANCE 1: IMPLICIT AND NEGATIVE

Levite observes that decisions taken by states to roll back nuclear programs involve "some diminution of the perceived utility of nuclear weapons:' He argues that the factors likely to prompt such a diminution fall into three categories: (1) changes to external security dynamics; (2) changes in the domestic context in which a regime's or country's "security and/or economic orientation" undergoes some type of alteration; and (3) "systemic or state-specific incentives" including the role of norms. 22 On this latter point he notes that, in the context of domestic change, external incentives "may tilt the balance in one direction or another:'23 Libya encompasses all three of Levite's categories. It took domestic political and economic developments, combined with external security changes, to make the context in Libya ripe for the regime to be influenced by external incentives, including an American assurance of nonintervention. The Domestic Context

Internal changes provided the essential political backdrop for the decision. During the 1990S, Libya experienced a significant economic downturn that undermined standards of living and gave impetus to the development of political opposition including growing support for militant Islamist groupings. The

94

WYN Q. BOWEN

growth in opposition, which even included an attempted military coup, posed a challenge to the regime's long-term survival. It became apparent to the regime that addressing these challenges would require addressing the underlying economic problems, and this could only be done if the UN and U.S. embargoes were ended. To achieve this, the regime moved away from its confrontational role on the international stage, including ending its involvement in terrorism and ceasing to pursue WMD. Jentleson and Whytock have highlighted some of the economic difficulties experienced by Libya. Problems included an average annual economic growth rate of less than 1 percent between 1992 and 1998, 30 percent unemployment, rampant inflation, and a significant drop in real per capita income. These problems stemmed from poor economic management, the impact of international sanctions, and reduced oil prices. The economic problems resulted in growing political instability.24 This included the appearance of a domestic Islamist opposition, which was of particular concern given the Islamist violence that neighboring regimes had been experiencing in Algeria and Egypt. These domestic challenges led to a decision by the Gadhafi regime to improve the country's international standing in order to have the UN and U.S. embargoes removed. 2s A requirement for improving the economic situation was modernizing and expanding the oil sector, and doing this depended on ending the embargoes. 26 Litwak notes that the crisis faced by Gadhafi sharpened a debate between the regime's hardliners and more pragmatic technocrats. The latter group believed the regime's confrontational foreign policy was now a major hindrance to economic development. Gadhafi himself evidently took the side of the technocrats, and starting in 1999 this led to cooperation on Lockerbie and secret negotiations with the United Kingdom and the United States. 2? Moreover, according to Braut-Hegghammer, after 9/11 Gadhafi and senior regime members decided that giving up the unconventional weapons programs so that relations could be improved with Washington, and the West more generally, was the best way to guarantee regime security.2s Importantly, during secret talks with the Clinton administration in 1999, Libyan officials communicated their willingness to "put everything on the table;' including an offer to sign the Chemical Weapons Convention and "open their [chemical] facilities to inspection:' As the chief U.S. negotiator in these talks, Martin Indyk notes, however, the chemical program was not deemed to pose an immediate problem, so the United States gave priority to "getting Libya out of terrorism and securing compensation" for families of Lockerbie victims.

LIBYA, NUCLEAR ROLLBACK, AND SECURITY ASSURANCES

95

Moreover, according to Indyk, Libya's nuclear program "barely existed" at that time. 29 Indeed, it was not until April 2000, for example, that the UK Joint Intelligence Committee began noting that "there was an evolving, and as yet incomplete, picture of the supply of uranium enrichment equipment to at least one customer in the Middle East, thought to be Libya, and evidence linking this activity to [A. Q.J Khan."30 It has since emerged that Libya took a "strategic decision" in 1995 to reinvigorate its nuclear weapon efforts, notably by working on gas centrifuge enrichment of uranium, and began receiving assistance from the network in 1997.31 This suggests that Libya's offer in 1999 to engage just on the chemical issue was a cynical effort to secure concessions from the United States while concealing an active nuclear weapons project. Although important policy changes were underway in 1999, the context was not yet right for Gadhafi to make a genuine offer to abandon Libya's nuclear weapons ambitions. External Security Dynamics

While domestic factors are pivotal to understanding Libyan rollback, alterations to external security dynamics were also important. The regime began to look very differently at the nuclear issue after the terrorist attacks of September 11, 2001. First, in terms of the regime's domestic standing, the money being spent on the nuclear program was not being put into development projects that would directly assist the Libyan people. And second, the nuclear program was seen to raise the likelihood of American intervention because of growing U.S. concerns about the potential nexus of nuclear proliferation and terrorism. 32 Ending the weapons program would reduce the chances of Libya's becoming a target of U.S. military action. Indeed, a diplomatic telegram from the U.S. embassy in Cairo on September 20, 2001, illustrated the Gadhafi regime's sense of vulnerability. It drew attention to Libya's desperation not to be seen to be in the terrorism business any longer and Gadhafi's frantic communications to other Arab leaders to get this message to the Bush administration. 33 Some Bush administration officials claimed in late 2003 and 2004 that the war in Iraq was the major influence on Gadhafi's turnaround on WMD, because of the "demonstration effect" of forceful regime change in Baghdad. However, by the time Libya publicly announced it was forgoing the pursuit and possession ofWMD, the regime had long understood it had to address this file after resolving Lockerbie if it wanted to achieve full re-engagement with the United States. This message was communicated to Tripoli toward the end of the Clinton years in 1999, and again after President Bush took office on the edges of

96

WYN Q. BOWEN

the Lockerbie negotiations in 2001. Moreover, Libya was not listed in the ''Axis of Evil" in January 2002, and the two countries began cooperating on counterterrorism issues after 9/11. While the events in Iraq may well have pushed the negotiations forward by demonstrating to Libya what could potentially happen if the regime did not continue charting a moderate course in its foreign policy, the war most likely served as an accelerant and cementer of the WMD decision rather than its primary cause. Nevertheless, given that the Libyans approached British intelligence to negotiate specifically on WMD immediately before the coalition invaded Iraq, it is prudent not to rule out the effect the war might have had in consolidating the regime's view that giving up on WMD was in its best interests. 34 Similarly, the interdiction of centrifuge equipment in October 2003 en route to Tripoli on board the BBC China can be characterized as consolidating Libya's final decision as opposed to being a principal driving factor. The interdiction made the Libyans much more forthcoming about the extent of their nuclear program during the secret negotiations, after they had held out until relatively late in the day in terms of being fully open. In short, the interdiction demonstrated that Libya's two negotiating partners had accurate and up-to-date intelligence, making it pointless to continue trying to conceal the nuclear program's true scope. 35 External Incentives

While the need to address domestic challenges, combined with external security developments, provided the necessary context for the Gadhafi regime to enter into secret negotiations with the U.S. and UK governments-first on Lockerbie and terrorism and later on WMD-external incentives were central to successfully engaging Tripoli on these issues. In this respect, a final and pivotal security-related factor involved what the United States was prepared to offer in return for Libyan movement on the WMD issue. As Litwak explains, "[T]he key to the December 2003 agreement was an implicit assurance of regime survival" that took the form of an "assurance of non-intervention" from Washington. 36 While the Gadhafi regime wanted to improve relations with the United States and the outside world, a significant amount of suspicion existed in Tripoli about American intentions. Indeed, one of the principal drivers of Libya's nuclear weapon aspirations during the 1980s and 1990S was American hostility to Tripoli as manifested in the prolonged confrontation with Washington

LIBYA, NUCLEAR ROLLBACK, AND SECURITY ASSURANCES

97

over Gadhafi's sponsorship of terrorism and, increasingly, his interest in unconventional weapons. Building trust and confidence in the United States as a negotiating partner, as well as the United Kingdom for that matter, was always going to be pivotal to reinventing the relationship. It is within this context that credible assurances over intentions became so important. Braut-Hegghammer notes that the secret negotiations initiated in 1999 to resolve the Lockerbie issue, which continued into the Bush administration, provided a vital framework for discussing other contentious issues such as WMD.37 Moreover, as Litwak argues, "Key to the Libyan breakthrough was the administration's decision, in the words of a former U.S. official, 'to take yes for an answer' -that is, to embrace Gadhafi's behavior change and to forgo the more expansive goal of regime change."38 The implicit assurance of nonintervention was a part of the changes in external security dynamics that contributed to Libya's decision. It made improved relations with the West appear a feasible way to provide for regime security. The official public statement released by Libya following the decision to disarm demonstrates that the regime understood itself to have received the requisite security assurance. According to the statement: After an agreement was reached regarding the Lockerbie case, the Libyan side ascertained that there wasn't a secret agenda on the American and British side against Libya and was assured of their good intentions towards Libya. This encouraged Libya to go forward and open the remaining files of mutual concern, beginning with weapons of mass destruction in which the American administration and the British government were continuously asking Libya to cooperate. 39

These two sentences in the official statement demonstrate that the Gadhafi regime had by December 2003 developed a significant level of trust in the intent of its negotiating partners to meet their commitments made during the secret talks, be it the removal of UN sanctions or not pursuing regime change. Indeed, the Libya case demonstrates the importance of viewing security assurances as a process or a journey, not just a destination. Gadhafi's son and then likely successor Saif aI-Islam said the Libyan leader took the WMD decision after Washington assured Gadhafi it was not planning to topple him. He said Libya was also promised economic assistance through foreign investment and access to defensive capabilities.40 Tripoli's decision was "part of a much broader engagement with the Gadhafi regime" from the late 1990S onward, during which confidence among the negotiating parties gradu-

98

WYN Q. BOWEN

ally grew as commitments made by one side were fulfilled and were then reciprocated by the otherY The Libyans were told early in the secret talks over Lockerbie that resolving this issue (putting the suspects on trial, arranging compensation for victims' families, and so forth) would end UN sanctions. They were further informed that ending u.s. sanctions and fully re-engaging with Washington required addressing WMD. Following the Lockerbie resolution in August 2002, the UN embargo was removed the following month. Importantly, lifting the UN embargo demonstrated to Tripoli that "its negotiating partners could be expected to fulfill their commitments vis-a.-vis WMD:'42 The success of the Bush administration's implicit assurance to Libya was largely determined by its track record of living up to earlier U.s. commitments. The Gadhafi regime's confidence was increased by the United Kingdom's proactive role, described more fully below, in providing assurances about u.s. and UK intentions and highlighting the positives of moving forward with disarmament. Moreover, as Jentleson and Whytock observe, the Gadhafi regime began to see itself as having common security interests with former adversaries following the terrorist attacks in 2001, due to shared concerns over the threat posed by Islamist extremists. 43 This convergence of interests led to cooperation in the field of counterterrorism, which contributed to improving relations. An examination of various factors related to the Libya case demonstrates, as Litwak has argued, that the implicit American assurance of noninterference with the security of the Gadhafi regime was probably the clinching issue in the WMD agreement of December 2003. The process of building trust and confidence during the secret negotiations from 1999 onward, first on Lockerbie and later on WMD, was pivotally important in making the American assurance credible from Libya's perspective. The reassurance against seeking regime change later proved conditional on good behavior by the Gadhafi regime in other spheres. The fact that the United States and United Kingdom supported military intervention in Libya in 2011 does not change the conclusion that the reassurance was sincere and effective in the context in which it was provided in 2003·

SECURITY ASSURANCE 2: EXPLICIT AND LARGELY POSITIVE

The Gadhafi regime was concerned from the outset about guaranteeing its security in the absence ofWMD programs. This issue featured prominently in the second set of assurances formalized with the United Kingdom some two and a half years later. In contrast to the implicit U.s. negative assurance, the

LIBYA, NUCLEAR ROLLBACK, AND SECURITY ASSURANCES

99

Joint Letter on Peace and Security involved a formal, public, and explicit set of largely positive assurances. Key provisions included the following: • The UK confirmed that "it would seek appropriate action by the Security Council in the event that it determines that the Great Socialist People's Libyan Arab Jamahiriya is subject to an act of aggression or threat of aggression in which chemical or biological weapons are used:' • The UK as a member of the Security Council agreed to "[p 1rovide support in accordance with any decision of the UNSC and the UN Charter or relevant conventions" to Libya "in the case that it is the victim of an act of aggression or threat of aggression in which chemical or biological weapons are used." • The UK pledged to "[wlork directly or through the international community to strengthen" Libya's conventional "defense capabilities so as to ensure that it is able to protect its security and the safety of its national territory from all threats:' • The two states agreed to "refrain in their international relations from the threat or use of force against the territorial integrity or political independence of either State, or in any other manner inconsistent with the purposes of the United Nations:' • The states also recalled the UK's declaration of April 6, 1995, on positive security assurances in the NPT context, which reaffirmed "the intention of the United Kingdom to seek immediate Security Council action to provide assistance, in accordance with the Charter, to any non-nuclear weapon State party to the Treaty on the Non-Proliferation of Nuclear Weapons, that is a victim of an act of aggression or an object of a threat of aggression in which nuclear weapons are used:'44 While the United Kingdom committed itself in 1968, and reaffirmed this commitment in 1995,45 to seek immediate Security Council action to assist any nonnuclear weapon state that is the subject of a nuclear weapon threat or nuclear use, the 2006 Letter expanded this specifically in the Libyan context to include chemical and biological weapon threats or use. The Need for Added Assurance

At the most basic level the Joint Letter was designed, from a British perspective, to reassure Libya about security provisions after giving up the pursuit and possession of WMD and long-range missiles. While the Letter was signed in

100

WYN Q. BOWEN

June 2006, its origins date back to the talks that resulted in Libya's rollback decision. On December 20, 2003, for example, the official Libyan statement noted, "[T]he British prime minister was continuously sending letters and personal envoys who met with the leader Muammar al-Gadafy and asked him for cooperation because this will open horizons of collaboration and enable Libya to obtain defensive weapons:'46 Moreover, Saif Al-Islam stated in December 2003 that, in addition to the American negative assurance, Britain and the United States further agreed to assist Libya economically and militarily, and even that "the West, and international society, will be responsible for the protection of Libya:'47 Following a Libyan request for military talks, officials from the UK Ministry of Defense and British Army officers visited Libya in January 2004 to discuss "conventional defense assistance, including training and joint exercises:' The British delegation sought to reassure their hosts that Libya's security would not be jeopardized by the WMD decision; rather, it would be strengthened. 48 Prime Minister Tony Blair did not himself mention defense assistance in his remarks on the Libyan breakthrough on December 19, 2003, but he said he looked forward "to developing a productive relationship" with the Gadhafi regime and Libya, adding, "Those countries who pursue such a path will find ready partners in the U.S. and the UK, as Libya will see:'49 The United Kingdom was committed from the outset, then, to nurturing positive relations with Libya. This subsequently included both informal and formal assurances about security provisions in the absence ofWMD programs. What motivated the UK government to make these commitments? Tying in Gadhafi

One factor was a desire to reduce the likelihood the Libyan leader would change his position on the WMD issue down the line. One British official noted in January 2004 that a key goal was to "lock Gaddafi in" because of his erratic and unpredictable tendencies. 50 These were on display in the year following the decision to renounce WMD. Gadhafi initially spoke about the need for other countries with WMD capabilities and development programs to abandon them and for Libya to "become an example to be followed:'51 He told the African Union summit in March 2004 that Libyan security could not be provided for by nuclear weapons, and they actually represented a danger to countries that possessed them. 52 It was not long, however, before Gadhafi became less glowing in his comments over what Libya had received from the United States by way of

LIBYA, NUCLEAR ROLLBACK, AND SECURITY ASSURANCES

101

compensation. In a Time Magazine interview in January 2005, Gadhafi stated, "Libya and the whole world expected a positive response-not just words, although they were nice words-from America and Europe:' He complained, "They promised, but we haven't seen anything yet:'53 While his public disgruntlement did not reflect the very good deal Libya actually received-the ending of sanctions and re-engagement with the United States and others-it raised the spectre, however remote, of Libya back-tracking on WMD. Gadhafi's disgruntlement also projected a negative image of what other proliferators could potentially expect to receive for giving up the pursuit and possession of such weapons. This leads to a second motivating factor for the UK's decision to negotiate the Joint Letter. The "Libya Model"

The Letter's text stated a British desire for the Libya case to be "emulated by other States:' As Kim Howells, minister of state at the Foreign and Commonwealth Office, expressed it, the Letter "will serve as an example to other states that there is a route back into the international community and the advantages of Libya's WMD decision."54 The "Libya model" had already been touted in January 2005 by Parliamentary Under-Secretary of State for Foreign and Commonwealth Affairs Bill Rammell. He told Parliament, "What has happened with Libya is a model that I hope North Korea follows, recognizing that it is possible to make progress on such issues by genuinely engaging with the international communitY:'55 From a British perspective a key element of the Libya model was fully engaging with states that voluntarily renounce WMD. Determining the scope of such engagement, however, is always going to be a subjective process. This was demonstrated in 2009 when the Scottish executive decided, despite strong U.S. objections but with Downing Street support, to release the man convicted for the Lockerbie bombing. Prime Minister Gordon Brown claimed the release of Abdelbaset al-Megrahi, declared by doctors to be terminally ill with cancer, was motivated purely on humanitarian grounds and "by the desire to bring Libya back into the international fold after the country agreed to abandon its program of weapons of mass destruction:'56 The prime minister further stated, "It is in all our interests and Britain's national interest that Libya rejoins the international community. So it was the duty of those responsible to look at all possible outcomes of the Megrahi case and their effect on our relations with Libya and on international terrorism and nuclear proliferation:'57 Although most media attention focused on the compassionate

102

WYN Q. BOWEN

argument for Megrahi's release, these statements by the prime minister show that Britain's desire to cement Libya's WMD decision and prevent any backsliding also played a role. Economic Motivations

A third motivating factor for negotiating the Joint Letter, and later releasing Megrahi, was economic. At the time the Letter was concluded in June 2006, Kim Howells stated, "We wish through this document to strengthen cooperation in several areas such as security and the fight against terrorism, and [to 1 develop economic relations."58 This statement illustrates that the British assurances were motivated, at least in part, by a desire to bolster economic relations between the two countries. The importance of economic considerations is perhaps illustrated most notably by the Megrahi affair in 2009. In early September, British Justice Secretary Jack Straw contradicted what Gordon Brown had said about the rationale for releasing the Libyan when he admitted that it was driven, in part, by a desire to secure new oil contracts. Specifically, the government had come under pressure from BP to include Megrahi in a prisoner transfer agreement out of concern that a failure to comply with Libyan demands for his release would jeopardize its interests. 59 An oil and gas exploration deal it was seeking to conclude with the regime was reportedly worth some £IS billion; it was subsequently ratified within six weeks of the UK government's change of heart on the prisoner transfer. 60 The Libya case demonstrates the potential bumps that can develop during a normalization process between states when economic incentives operate in both directions. Specifically, Tripoli used oil as a bargaining chip with London and demonstrated how economic incentives can become a double-edged sword. It is important to note that Washington was not immune to such pressure. As early as 2001 the Gadhafi regime told American firms to use their oil concessions in Libya or they risked losing them to European competitors. The State Department subsequently decided in March 2002 to allow Marathon Oil to initiate talks with the Libyans before UN and American sanctions were removed. 61 Despite Libya's economic and political isolation by the United States from the mid-1980s, the regime allowed the Oasis Group of American oil firms (Marathon Oil, Conoco-Phillips, and Amereda Hess) to keep their concessions through stand-still agreements allowing them to reinitiate operations in the future. This allowed Tripoli to retain a powerful tool in its future dealings with Washington. 62 These examples of British and American dealings with Libya

LIBYA, NUCLEAR ROLLBACK, AND SECURITY ASSURANCES

103

over oil illustrate an important lesson for other efforts to coax states away from proliferation and the sponsorship of terrorism, particularly those such as Iran that possess significant oil and gas resources. When target states control an important resource such as oil, they will retain some leverage over nuclear weapon states trying to convince them to abandon nuclear activities. The Joint Letter was not signed until June 26, 2006, two and half years after Libya's WMD decision became public and some two years after the nuclear program was dismantled. The delay is probably explained by London's desire not to put anything formally onto paper until Libya's designation as a state sponsor of terrorism had been lifted by the U.S. State Department. State announced on May 15, 2006, that Libya was being removed from its list of state sponsors of terrorism, just one and a half months before the Letter was inked. 63 Absence of a U.S. Positive Security Assurance

Although London and Washington worked closely together to bring Libya in from the cold, the Bush administration did not want to conclude a similar set of positive assurances with Tripoli. According to Arms Control Today, one administration official said on August 23, 2006, in the context of the Joint Letter, that it was "premature to consider any arrangement that would involve the United States directly:'64 This was despite talk in Washington, similar to that in London, about the "Libya model:' In March 2005, William J. Burns, then Acting Under-Secretary of State for Political Affairs and previously Assistant Secretary for Near Eastern Affairs during the secret negotiations with Libya, told Congress that Libya "has offered a powerful positive example to other countries." He added, "Libya is probably the best model for how a country can renounce the pursuit of weapons of mass destruction (WMD) and take important steps to rejoin the international community."65 Despite the fact that both capitals promoted the "Libya model;' Washington and London adopted different approaches to implementing it. At one level this reflected the fact that London had already restored full diplomatic relations with Tripoli in 1999. At another level, internal divisions within the Bush administration over whether to "reward" Libya made the process more challenging from an American perspective. Despite administration divisions it responded positively in terms of reinventing the U.S.-Libya relationship from January 2004 onward, albeit more slowly than the United Kingdom. American actions included lifting sanctions incrementally during 2004, redeploying diplomatic personnel to Libya in February 2004, and removing Libya from the state spon-

104

WYN Q. BOWEN

sors of terrorism list in June 2006. A significant brake on Washington making quicker progress was the process of negotiating a comprehensive claims settlement agreement with Libya, which was completed only in August 2008. This was followed by a visit to Libya by Secretary of State Condoleezza Rice in September 2008, and the respective deployment of ambassadors to each other's capitals in January 2009. 66 The Bush administration made no moves, however, to negotiate a set of positive security assurances with Libya before it left office. The improvement in relations was abruptly reversed in 2011 after the Gadhafi regime violently cracked down on domestic opposition, leading to U.S. and UK support for military intervention in Libya and making any question of further assurances for Gadhafi moot. IMPLICATIONS FOR THE HYPOTHESES

What does Libya tell us about the hypotheses identified in Chapter 2? It offers qualified support for hypotheses 1 and 7, stronger support for hypotheses 5, 8, and 13, and disconfirming evidence for hypotheses 2 and 12. First, Libya supports hypothesis 1, that security motivations make assurances relevant, but only in a modified form. Tripoli was not primarily concerned about assurances against nuclear threats, but was instead focused on regime survival. Hypothesis 2, concerning public declarations, is not fully supported. The Libya case demonstrates that even secret and implicit negative assurances can be a highly effective and credible incentive for a country to roll back WMD programs. A public assurance might well have made the breakthrough less likely given that the secrecy of the WMD negotiations was based in part on preventing domestic opposition to the deal from developing in the United States, Libya, and potentially elsewhere. For positive assurances, publicizing an agreement, so third countries understand the nature of the commitments, may be more important. This is because part of the rationale behind such assurances is likely to be a desire to demonstrate to others that a commitment is in place to assist the target if its security is threatened in a particular way. For example, it was in Libya's interests for its neighbors to understand Britain's commitments under the 2006 Joint Letter. The Libya case also has important lessons for hypothesis 5, about the value of a reputation for keeping one's commitments. The key to the breakthrough with Libya was the prior establishment of an environment characterized by trust where the target had developed sufficient confidence in the intent of the assuring party to live up to its commitment, in this case not to pursue regime

LIBYA, NUCLEAR ROLLBACK, AND SECURITY ASSURANCES

105

change. With Libya trust was built through an initial negotiation process on a non-WMD related issue, Lockerbie, where all sides made specific commitments and then proved the authenticity of them through subsequent actions. The u.s. reputation for keeping commitments was thus important in this case, but was based on implementing prior commitments on a different issue-namely, keeping promises to ease sanctions if Libya turned over the Lockerbie bombing suspects. However, this chapter has also illustrated that there will always be the challenge of reconciling the views of the various parties concerned over the speed at which such commitments are met, and even the specific nature of the commitments themselves. Establishing a favorable reputation may be difficult if disagreements over the specifics become aired in public, such as happened with the dispute over Megrahi's release. The Libya case also offers partial support to hypothesis 7, about tailoring, in that the assurance had to be tailored to Gadhafi's concerns about regime survival. The fact that the negative assurance also took the form of a broad reassurance against seeking regime change also supports hypothesis 8, about the value of embedding assurances in a broader strategy. With regard to hypothesis l2, about altering internal debates, an ability to influence internal debates was important but also only half the story. One enabling condition for rollback in Libya was the relatively nonpluralistic nature of its political system. While Gadhafi evidently had around him a close group of confidantes and advisers who participated in the secret negotiations over Lockerbie and WMD on his behalf, and although there was an apparent debate between hardliners and pragmatists, the Libyan leader had the ability to deliver on his decisions once taken because of his dominant influence on decision-making and implementation. Internal debate may have helped convince Gadhafi it was in his best interests to give up the WMD programs, but it was also important that once a strategic decision was taken he was in a position to deliver. The strong political control enjoyed by the Libyan leader helped ensure that he could decisively implement the decision that had been reached. Finally, the case strongly supports hypothesis 13, on the value of combining assurances with positive incentives. Libya provides a solid example of how the international community can make the pursuit of nuclear weapons prohibitively costly, in economic and political terms, through a negative incentive strategy based on targeted sanctions, political and diplomatic isolation, and export controls. When it came to the actual secret negotiations, however, this case demonstrates the utility of applying both carrots and sticks. Libya highlights

106

WYN Q. BOWEN

the importance of communicating not only the penalties for not giving up the pursuit ofWMD, but also the advantages of doing so. Given the magnitude of Libya's decision it was certainly appropriate, and in the end essential, for its negotiating partners to apply a mix of positive and negative incentives to coax the regime along the way. Regardless of the American assurance of nonintervention, it is doubtful that the regime would have negotiated the WMD deal if positive incentives had not been on the table, notably the prospect of an end to the unilateral sanctions and the chance to re-engage with Washington. CONCLUSIONS

Libya represents an important success for global nonproliferation efforts. Both negative and positive security assurances contributed to this outcome. By themselves, however, they were not adequate. Appropriate changes in Libya's domestic political and international security environments were necessary to set the stage for the use of assurances. The negative assurances were also broader than the standard NPT-related assurances, and instead took the form of reassurance by the United States that it would not seek regime change. The fact that the interaction was not exclusively bilateral, and also included the United Kingdom, was also important to facilitating the rollback of Libya's WMD programs. Finally, assurances had to be combined with other instruments, especially economic incentives, to clinch the deal. After the events of 2011, it is not clear if the Libya model still has the potential to be attractive to other "rogue state" proliferators. It will depend on whether assurances against seeking regime change are still on the table and can be made credible to dictatorial regimes worried about their survival. In the Libya case, this trust had to be built through a process of diplomatic engagement. If the Libya model is still to prove viable elsewhere, similar processes of engagement to build confidence will be required. NOTES 1. Ariel Levite, "Never Say Never Again: Nuclear Reversal Revisited;' International Security 27, no. 3 (Winter 2002/3): 68-69. 2. Robert S. Litwak, "Living with Ambiguity: Nuclear Deals with Iran and North Korea;' Survival 50, no.! (February 2008): 99. 3. The main sources on Libyan nuclear rollback to date include: Malfrid Braut-Hegghammer, "Libya's Nuclear Turnaround: Perspectives from Tripoli;' Middle East Journal 62, no.! (Winter 2008): 55-72; Wyn Q. Bowen, Libya and Nuclear Proliferation: Stepping

LIBYA, NUCLEAR ROLLBACK, AND SECURITY ASSURANCES

107

Back from the Brink, Adelphi Paper 380 (London: International Institute for Strategic Studies, May 2006); Robert Litwak, Regime Change: u.s. Strategy through the Prism of 9/11 (Washington, DC: Woodrow Wilson Centre Press, 2007); Bruce W. Jentleson and

Christopher A. Whytock, "Who 'Won' Libya? The Force-Diplomacy Debate and Its Implications for Theory and Policy;' International Security 30, no. 3 (Winter 2005/6): 47-86; George Joffe, "Libya: Who Blinked and Why;' Current History 103 (May 2004); Etel Solingen, Nuclear Logics: Contrasting Paths in East Asia and the Middle East (Princeton: Princeton University Press, 2007), 213-28. 4. These interviews were made possible by a Small Research Grant from the British Academy. The interviews were conducted on a background only, and not for attribution, basis. The chapter also benefited from feedback received at several international conferences, including the Workshop on Security Assurances, Colorado Springs, August 13-14, 2009; tile Nobel Symposium, Norway, June 25-27, 2009; and tile International Studies Association Annual Convention in Chicago, February-March 2007. 5. For a detailed analysis of the technical evolution of Libya's nuclear weapons program, see Bowen, Libya and Nuclear Proliferation. For further details on tile technical aspects of tile Libyan program, see: Director General, IAEA, Implementation of

the NPT Safeguards Agreement of the Socialist People's Libyan Arab Jamahiriya (Vienna: IAEA, February 20, 2004), http://www.iaea.org/NewsCenter/Focus/IaeaLibya/index. shtml; Director General, IAEA, Implementation of the NPT Safeguards Agreement (Vienna: IAEA, May 28, 2004), http://www.fas.org/nuke/guide/libya/iaea0504.pdf; Director General, IAEA, Implementation of the NPT Safeguards Agreement (Vienna: IAEA, August 30, 2004), http://www.iaea.org/NewsCenter/Focus/IaeaLibya/index.shtml; Director General, IAEA, Implementation of the NPT Safeguards Agreement (Vienna: IAEA, September 12, 2008), http://www.iaea.org/NewsCenter/Focus/IaeaLibya/index.shtml; Royal Malaysia Police, "Press Release by Inspector General of Police in Relation to Investigation on the Alleged Production of Components for Libya's Uranium Enrichment Program;' February 20, 2004, http://www.rmp.gov.my/rmp03/0/020040220scomi_eng.htm (accessed in 2004; link no longer active). 6. Bowen, Libya and Nuclear Proliferation. 7. Jonathan Tucker, "The Rollback of Libya's Chemical Weapons Program;' Nonpro-

liferation Review 16, no. 3 (November 2009): 373-75. 8. Ibid., 363. 9. Sharon A. Squassoni and Andrew Feickert, Disarming Libya: Weapons ofMass De-

struction, U.S. Library of Congress, Congressional Research Service, Congo Rep. RS21823, Washington, April 22, 2004, 2; "Libya Timeline: Key Events in the Lead Up to and the Aftermath of the Iraq War;' Iraq Watch, http://www.iraqwatch.orglroundtables/RT4/ Libya-Timeline.htm. 10. Bowen, Libya and Nuclear Proliferation; Braut-Hegghammer, "Libya's Nuclear Turnaround:'

108

WYN Q. BOWEN

11.

Braut-Hegghammer, "Libya's Nuclear Turnaround;' 60.

12. Bowen, Libya and Nuclear Proliferation, 12. 13. Ibid., 13-14. 14. Braut-Hegghammer, "Libya's Nuclear Turnaround;' 62. 15. 16. 17. 18.

Ibid., 60. Ibid., 63-64. Bowen, Libya and Nuclear Proliferation, 21. Braut-Hegghammer, "Libya's Nuclear Turnaround;' 67.

19. 20. 21. 22.

Bowen, Libya and Nuclear Proliferation, 56. Ibid., 68-69. Jentleson and Whytock, "Who 'Won' Libya?" 78. Levite, "Never Say Never Again;' 68-69.

23· 24. 25. 26. 27. 28. 29.

Ibid., 74. Jentleson and Whytock, "Who 'Won' Libya?" 65-66. Braut-Hegghammer, "Libya's Nuclear Turnaround;' 65. Bowen, Libya and Nuclear Proliferation, 68. Litwak, "Living with Ambiguity;' 98-99. Braut-Hegghammer, "Libya's Nuclear Turnaround;' 71. Martin S. Indyk, "The Iraq War Did Not Force Gadaffi's Hand;' Financial Times,

March 9, 2004· 30. Review of Intelligence on Weapons of Mass Destruction, Report of a Committee of Privy Counsellors, HC 898 (London: Stationery Office, July 14, 2004), http://www.

archive2.official-documents.co. uk/ document/ deps/hc/hc898/898. pdf. 31. "Implementation of tile NPT Safeguards Agreement of the Socialist People's Libya Arab Jamahiriya;' Report of tile Director General to the IAEA Board of Governors, February 20,2004, GOVI2004/12, 5. 32. Braut-Hegghammer, "Libya's Nuclear Turnaround;' 71. 33. Many thanks to Isabelle Anstey for drawing this document to my attention. "Sudan and Libya Want Egyptian and Other Arab Help to Avoid Becoming a U.S. Target;' State Department Document: From American Embassy Cairo, September 20, 2001, MaRl DodD: 841934, http://www.judicialwatch.org/archive/2006/statedocqadhafi.pdf. 34. Bowen, Libya and Nuclear Proliferation, 63-64. 35. Ibid., 68-69· 36. Litwak, "Living with Ambiguity;' 99. 37. Braut-Hegghammer, "Libya's Nuclear Turnaround;' 71. 38. Litwak, Regime Change, 197. 39. "Libya Statement: 'We Ascertained There Wasn't a Secret Agenda against Us;"

The Guardian, December 20, 2003, http://www.guardian.co. uk/world/ 2003/dec/ 20/libya. politics2. 40. Litwak, Regime Change, 194-95.

LIBYA, NUCLEAR ROLLBACK, AND SECURITY ASSURANCES

109

41. Bowen, Libya and Nuclear Proliferation, 68. 42. Ibid., 65. 43. Jentleson and Whytock, "Who 'Wor! Libya?" 78. 44. Joint Letter on Peace and Security between the Great Socialist People's Libyan Arab

Jamahiriya and the United Kingdom of Great Britain and North Ireland, June 26, 2006 (signed on behalf of the UK by Minister of State at the Foreign and Commonwealth Office, Kim Howells, and for Libya by Abdulati Ibrahim Al Ibidi, Secretary for European Affairs), 1-3. 45. See "Unilateral Nuclear Assurances by Nuclear Weapon States;' in Security Assurances, Programme for Promoting Nuclear Non-Proliferation, K3, http://www.ppnn. soton.ac. uklbb2/Bb2secK. pdf. 46. The Guardian, "Libya Statement:' 47. Litwak, Regime Change, 194. 48. Philip Sherwell, "British Troops Stand By to Train Libyan Forces;' Daily Tele-

graph, January 25, 2004, http://www.telegraph.co.uklnews/worldnews/africaandindianoceanllibyal1452555/British-troops-stand-by-to-train -Libyan -forces.html. 49. "PM Welcomes Libyan WMD Announcement;' statement by PM Tony Blair, December 19, 2003, http://www.number-lQ.gov.uk. 50. Sherwell, "British Troops:' 51. He made this comment on a visit to the European Union (EU) in April 2004. Michael Thurnston, "Kadhafi Urges World to Follow His Lead, Give up Weapons of Mass Destruction;' Agence France-Presse, April 27, 2004. 52. John Bolton, "The NPT: A Crisis of Non-Compliance;' statement to the 3rd session of the Preparatory Committee for the 2005 Review Conference of the Treaty on the Non-Proliferation of Nuclear Weapons, New York City, New York, April 27, 2004, http://www.state.gov/t/u/rm/31848.htm.

53. Scott Macleod and Amany Radwan, "10 Questions for Muammar Gaddafi;' Time Magazine, January 31, 2005, http://www.time.com/time/magazine/article/o,9171,1022602,00.html. 54. Michael Nguyen, "UK Offers Libya Security Assurances;' Arms Control Today 36,

no. 7 (September 2006). 55. The Parliamentary Under-Secretary of State for Foreign and Commonwealth Affairs (Mr. Bill Rammell), House of Commons, January 19, 2005. See "Lockerbie;' Column 268WH, Hansard, Westminster Hall Debates for January 19, 2005, House of Commons Daily Debates, Parliament, http://www. parliament. the-stationeryoffice.co. uk/pal cm200405/cmhansrd/cm050119/hallindxi50119-x.htm. 56. Patrick Wintour, "Libya Oil Deals Were Factor in Megrahi Talks, Says Straw;' The

Guardian, September 5, 2009, http://www.guardian.co.uklworldI2009/sep/05/straw-admits-oil- role-megrahi -talks. 57. Patrick Wintour and Severin Carrell, "Gordon Brown Finally Admits Support

110

WYN Q. BOWEN

for Lockerbie Bomber Release;' The Guardian, September 2, 2009, http://www.guardian. co. uk/politics/ 2009/sep/ 02/ gordon -brown -megrahi -lockerbie-release. 58. "UK, Libya Agree to Work for Global Peace, Security;' Agenda Angola Press (ANGOP), June 28, 2006, http://www.portaiangop.co.ao/motixipcpt/portal!capa/ index.html. In a confirmation of progress on counterterrorism cooperation, Foreign Secretary David Miliband told the House of Commons on October 12, 2009, that "[tloday, we share information and co-operate in our efforts to disrupt and dismantle terrorist groups in Europe and north Africa, in particular al-Qaeda in the Maghreb:' See "Libya;' The Secretary of State for Foreign and Commonwealth Affairs, David Miliband, House of Commons, October 12, 2009, Daily Hansard-Debate, October 12,2009, Column 27, http://www. parliament. the-stationery-office.co. uk/pal cm200809/ cmhansrd/ cm091012/ debtext/91012-0005.htm. 59. Wintour, "Libya Oil Deals Were Factor." 60. Jason Allardyce, "Lockerbie Bomber 'Set Free for Oil;" The Sunday Times, August 30, 2009, http://www.timesonline.co.uk/tol/news/politics/article6814939.ece. 61. "Libya;' Country Analysis Briefs (Washington, DC: Energy Information Administration, U.S. Department of Energy, January 2004). 62. Milton Viorst, "The Colonel in His Labyrinth;' Foreign Affairs 78, no. 2 (March/ April 1999): 72; Greg Flakus, "U.S. Oil Companies Re-enter Libya;' Voice of America

News, January 2, 2006. 63. U.S. Department of State, "Background Note: Libya;' http://www.state.gov/r/pa/ ei/bgn/5425.htm. 64. Nguyen, "UK Offers Libya Security Assurances." 65. Statement of the Honorable William J. Burns, Acting Under-Secretary for Political Affairs, U.S. Department of State, Libya: Progress on the Path Towards Cautious Reen-

gagement, Hearing before the House Committee on International Relations, 109th Cong., 1st sess., March 16, 2005, 10, 38. 66. State Department, "Background Note: Libya:'

Security Assurances and Iran: Assessment and Re-conceptualization Jim Walsh

This chapter examines the relationship between security assurances and nuclear decision-making in Iran. Iran does not provide an ideal set of observations for testing the hypotheses on security assurances proposed in Chapter 2, but the results of the case are fairly clear. None of the hypotheses evaluated below are strongly sustained by the case history. Security threats and the available sources of assurance are not without causal or policy consequence, but the story plays differently than one might expect. The Shah, for instance, made his most earnest efforts to build a nuclear infrastructure at a time when relations with his chief protector, the United States, were arguably at their best and when the threat of a Soviet invasion had receded. The Islamic Republic, for its part, has never enjoyed an Iran-specific security assurance, and variations in its nuclear program do not seem strongly correlated with changes in its threat environment. In sum, the provision of security assurances has played only a modest role in Iranian affairs, while pride, anti-imperialism, and domestic discontent have often influenced Tehran's behavior, including its nuclear policy. The final section of this chapter uses Iran's experience as a basis for reconside ring the concept of security assurance. It explores how assurances might operate in different environments and presents an alternative way of thinking about their causal import. In particular, it suggests that in circumstances like those facing Iran, assurances might be more effective as an instrument for affecting political factors rather than for addressing security threats.

111

112

JIM WALSH

IRAN AS A TEST FOR HYPOTHESES ON SECURITY ASSURANCES

This chapter focuses on two broad questions concerning security assurances and nuclear behavior. The first is whether the provision of security assurances affects the actions of potential proliferators. This is the central causal question, and the most common hypothesis posits that security assurances reduce proliferation incentives (and associated proliferation -related behaviors) and thereby reduce the likelihood that a state will acquire nuclear weapons. A second, subsidiary but nevertheless important question concerns what, if any, attributes are associated with effective assurance. Knopf has identified a number of different characteristics that might plausibly influence the relative success of security assurances. Given the particulars of the Iranian case, six of the hypotheses identified in Chapter 2 are considered as potentially relevant: • • • • • •

Security concerns as a significant motivation (HI) Public declaration of the assurance (H2) Legal character of the assurance (H3) Political and economic ties between the parties (H4) Presence of forward-deployed troops (H6) Embedding assurance in a larger strategy (H8)

As a crude test of the hypotheses, this study reviews Iran's nuclear history and compares (a) the presence or absence of an assurance and (b) the character of any security assurance in play with (c) the government's nuclear behavior. So, for example, when assessing one of the six characteristics of a security assurance listed above, if the presence of one of these factors is associated with (1) a strong security assurance and (2) reduced interest in nuclear weapons, then the hypothesis is sustained. In addition, if the absence of the factor is associated with a weak security assurance and greater interest in nuclear weapons, this is taken as tentative support for the hypothesis. Iran's nuclear history spans two broad periods: the monarchy under the Shah (1953-79) and the Islamic Republic (1979-present). The two periods are evaluated separately. Each section starts with a review of Iran's nuclear activities during the time period, then summarizes the security assurances that were available, and then describes the threat environment in order to enable an assessment of the role of assurances.

SECURITY ASSURANCES AND IRAN

113

THE MONARCHIAL PERIOD, 1953-79 The Shah's Nuclear Program: Slow Start, Fast Finish

The Shah came to power at the dawn of Atoms for Peace. Iran signed a civil nuclear cooperation agreement with the United States in 1957 and became host to the regional Institute for Nuclear Science in 1959. That same year, the Shah established a nuclear research center at Tehran University and told the visiting President Eisenhower that he wanted a "'crash program' to obtain highly mobile forces with atomic weapons, long-range missiles, effective anti-aircraft missiles, additional air bases, and improved aircraft:'l Iran's actual progress in the nuclear field was modest, however. Unlike Israel, Egypt, and others in the region, Iran did not move aggressively to build a nuclear infrastructure. It was not until 1967 that Tehran's first reactor went critical, a modest 5MW research reactor supplied by the United States-along with 5.58kg of 93 percent enriched uranium. Soon after, Iran seemed to close the door on building a nuclear bomb. It signed the NPT in 1968 and ratified it in 1970. Nuclear technology remained a secondary priority for the king until the 1970S, when flush with oil revenues and at U.S. urging, he embarked on a plan to rapidly expand Iran's nuclear program. A key year was 1974, when the Shah created the Atomic Energy Organization of Iran (AEOI) under the direction of Dr. Akbar Etemad and bought a 10 percent stake in Eurodiff, the European enrichment consortium. By 1975, AEOI was reported to have 150 personnel "trained in physics:' In the two years from 1974 to 1976,AEOI's budget increased from roughly $39 million a year to more than $1 billion a year. Iran made ambitious plans to build more than twenty nuclear power reactors, starting with two to be constructed by West Germany at Bushehr. Iran also received help from other countries. In addition to West Germany, these included Argentina, South Africa, France, India, and the United States, though India's 1974 "peaceful nuclear explosion" complicated international nuclear commerce. 2 As an NPT signatory, Iran's program was ostensibly peaceful, but on at least two occasions in the 1970S, the Shah publicly raised the possibility that Iran would one day possess nuclear weapons-statements that were quickly retracted. Despite Iran's denials, its membership in the NPT, and its support for a regional nuclear-weapon-free zone, those around the Shah believed that his aspirations went beyond power plants and included nuclear weapons as well,3

114

JIM WALSH

By 1978, however, visions of nuclear grandeur were put aside, a victim to financial difficulties, domestic political instability, and scandal. 4 When the Shah left Iran in search of medical treatment, the dream of a nuclear Iran went with him-at least for a time. Iran's nuclear effort had been slow off the mark and ended before any of its major projects were completed. Security Assurances and the Shah

During the monarchial period, the relevant security assurances involved (1) the Baghdad Pact (later renamed CENTO, the Central Treaty Organization, after Iraq withdrew from the pact), (2) Iran's 1959 bilateral agreement with the United States, and (3) Iran's assurances to the Soviet Union in the early 1960s. None of these were explicitly nuclear-related. The Baghdad Pact was signed in 1955 and disbanded after the fall of the Shah in 1979. Organized by the United States and Britain, the pact's official members were Turkey, Iraq, Iran, Pakistan, and Britain (the United States did not formally join but offered other forms of support). The pact's members promised that they would "co-operate for their security and defence" (Article I). In addition, the parties agreed to "refrain from any interference whatsoever in each other's internal affairs ... [and] ... settle any dispute between themselves in a peaceful way" (Article II).5 The pact thus provided both positive and negative assurances to its members. The pact reflected Britain's need to reorganize its empire after World War II and America's desire to contain the Soviet Union. Whatever its aims, the Baghdad Pact/CENTO effort has been judged harshly by historians, while-more to the point-the Shah by 1965 described it as "moribund" and a "masquerade:'6 The arrangement was hampered by the vagueness of its assurances, a lack of institutionalization (most especially the absence of a shared military command), and the complexities of a multipolar region with cross-cutting rivalries. It did not help that Iraq withdrew from the pact, that member Britain joined France and Israel to attack Egypt in the Suez War in 1956, that the parties did little to help member Pakistan during its wars with India in 1965 and 1971, or that Britain decided in the late 1960s to withdraw from "east of Suez:' CENTO was long dead by the time it was formally dissolved in 1979. In contrast, Iran's bilateral relationship with the United States grew stronger over the decades. In 1959 the two countries signed a defense agreement, but Washington initially emphasized that there were limits to the American commitment, although the United States did provide Iran with political sup-

SECURITY ASSURANCES AND IRAN

115

port and armaments.? Over time the relationship deepened and reached a high point during the Nixon years. This was the era of the new Guam Doctrine. Washington wanted Iran to take a larger role in its defense, and Tehran wanted to use petrodollars to build its military and technological capacity. By 1972, Iran's military spending was four times what it spent in 1966.8 To ascertain the relevance of the foregoing assurances, this chapter next considers the threat environment that confronted the Shah. Political Context and Threat Environment

In 1941, Mohammad Reza Pahlavi ascended to the throne following the British and Soviet invasion of Iran and the removal from power of his father, Reza Shah. Though he had been king for more than a decade, the young ruler did not become Iran's dominant political actor until after the 1953 coup that toppled Prime Minister Mohammed Mossadegh. The Shah's tenure would be known for dramatic efforts to modernize Iran and for increasingly brutal, unpopular, and eccentric leadership. The Shah relied primarily on the military as his domestic source of power. Security concerns that might be relevant to an interest in nuclear weapons were plentiful. During the Shah's reign, the government faced a variety of potential threats, both external and internal. The Soviet Union, a burgeoning superpower, was situated on Iran's border. The USSR had already invaded Iran once, and the Shah told American officials that he feared that the Soviets might push again into Iran to secure access to oil and a warm-water port.9 Beginning in the early 1960s, however, an increasingly confident Shah sought to improve political and economic relations with Moscow, even while maintaining strong defense ties with Washington. Iran went so far as to offer assurances to the Soviet Union that Iran would not become a base for American or other attacks against the USSR.lO By the early 1970S, it was commonly assumed that Israel had acquired nuclear weapons (it had done so just before the 1967 War). In 1974, India conducted its first nuclear test, and most observers expected Iran's territorial neighbor, Pakistan, to ramp up its nuclear program in response. Iran and Israel had a surprisingly strong working relationship, and Iran's relations with Pakistan were not adversarial, but those inclined to worry could point to multiple nuclear programs (in some cases, arsenals) uncomfortably close to Tehran.l1 Regional actors also posed threats. The Shah was particularly suspicious of Egypt's Nasser, who came to power following the 1952 officer's revolt. Egypt's

116

JIM WALSH

revolution had dispatched King Farouk, and Nasser's brand of populist panArabism was seen as a direct challenge to the region's monarchies. The Shah feared plots hatched in Arab capitals as much as he worried about an incursion by Soviet troops.12 Iraq went from ally to potential rival following the coup that brought Qasim to power in 1958 and Iraq's withdrawal from the Baghdad Pact the following year. By the mid-1970S, Saddam Hussein had taken control and was building a substantial military capability. Iraq also initiated a nuclear program. A rising, Ba'athist military power on Iran's border had direct implications for its security. The Shah could also not discount the dangers posed by his own allies, first Britain and then the United States, as the latter took on an increasingly prominent role once the United Kingdom withdrew from "east of Suez:' These imperial (and nuclear) powers had installed the Shah because of his views on communism and oil, but it would have been reasonable for the Iranian ruler to suspect that what imperial powers give, imperial powers can take away. Finally, there was the threat that actually managed to defeat Reza Pahlavithe internal threat. The foreign-engineered coup that installed the Shah in 1953 drew the ire of former Mossadegh supporters and leftists in the Tudeh Party. On the other side of the ideological continuum, traditionalists and conservative clerics objected to the Shah's modernization initiatives, such as those aimed at improving the status of women. Large landowners were unhappy with the Shah's program of land reform. As the regime became more repressive, others joined ranks of the opposition. Even among his most loyal base of support, the military, there were those who plotted his removal. 13 Table 6.1 summarizes the threat environment during the Shah's reign. The second column lists "nuclear-relevant" threats-that is, threats for which nuclear weapons could conceivably be seen as a useful counter. This TABLE

6.1

Iran's Nuclear and Threat Environment, 1953-79

1

Period 19505 19605 19705

2

3

Nuclear-relevant threats

Regional nuclear-

4

arming states

Other threats

USSR USSR USSR

Note: Italics indicate nascent nuclear weapons program.

Egypt, Internal Israel Israel, Pakistan

Iraq,Egypt Iraq, Internal

SECURITY ASSURANCES AND IRAN

117

analysis uses a traditional specification of nuclear-relevant threat: adversaries that possess nuclear weapons, adversaries that possess a latent nuclear weapons capability, and adversaries with a two-to-one advantage in conventional capability. The Soviet Union more than qualifies as a nuclear-relevant threat, though a more nuanced reading of the Iranian-Soviet relationship in the 1960s might suggest that fears of direct Soviet attack had diminished. 14 By the 1970S, however, Iran's worries about Soviet influence again grew more intense as the USSR gained allies in the region (for example, Iraq, Egypt, Syria). The coding for column two also fails to capture the fact that while the United States and Britain were political backers, they were also potential threats-whether by coup or outright intervention. Column three inventories other countries in the region with nuclear weapons programs or nuclear arsenals. Israel and Pakistan were not Iranian adversaries at the time, but are relevant if the focus is capability rather than intentions. Column four lists non-nuclear threats, both internal and in the region. Egypt in the 1960s and Iraq in the 1970S are in italics, because both had initiated nuclear programs, though it has to be stressed that in neither case had the program gone very far. Egypt in the 1960s was actively establishing the building blocks of a nuclear program but was unsuccessful in acquiring reprocessing or enrichment technology. The program went on to suffer a deathblow following the 1967 WarY Iraq's nuclear efforts began in the mid-1970S, but it was not until the 1980s that the program progressed in any significant way. The Shah viewed both Nasser and Saddam Hussein as threats, but at this stage, they were political (Egypt) and conventional military (Iraq) threats, not nuclear threats. The Shah's internal threats were most acute early in his tenure when he had not yet consolidated power, and again later in the 1970S as popular grievances grew stronger. Assessing the Influence of Assurances

Given Iran's behavior during the monarchial period, how should one assess the performance of the central hypothesis-that security assurances reduce proliferation incentives and thereby make nuclear weapons outcomes less likely? On the one hand, it could be said that Iran received assurances and, in fact, did not become a nuclear weapons state-a simple truth that is probably too simple. The Shah had many reasons to feel threatened, but CENTO did little or nothing about most of those threats. It was especially irrelevant to

118

JIM WALSH

TABLE 6.2

Iran's Threats, Assurances, and Nuclear Behavior, 1953-79 6

2

Period

Nuclearrelevant threats

1950S

USSR

1960s

USSR

Israel

Egypt, Iraq

1970S

USSR

Israel,

Internal,

1

Regional nuclear states

Pakistan Note:

4

Security

Other threats

assurances

Egypt, Internal

CENTO, U.S.

CENTO, U.S., [USSR] CENTO, U.S., [USSR]

Iraq

Nuclear development Interest in NW but little action Little nuclear activity Robust nuclear activity

Italics indicate nascent nuclear weapons program; brackets indicate recipient of assurance.

the problems of foreign subversion and domestic opposition, which were often seen as his most pressing challenges. Still, one could rightly argue that the U.S.-Iran bilateral agreement and the assurance Iran provided the USSR allowed Iran the luxury of worrying less about Soviet military might, and the presence of other threats unaltered by additional security assurances does not diminish the importance of the assurances that were provided. The problem comes when one tries to tie this all back to Iran's nuclear behavior. Table 6.2 summarizes Iran's threats, security assurances, and nuclear activity during the Shah's reign. Columns 1-4 are the same as Table 6.1. Added are column 5 on security assurances and column 6 describing Iran's nuclear program. (In column 5 the USSR is in brackets, because Iran is the provider, not the recipient, of the assurance.) The problem for any effort to judge security assurances as important is that Iran's most active efforts in the nuclear arena occurred when the U.S.-Iranian relationship was deepening and the Soviet Union appeared to be less of a direct military threat (though more of a political threat). Moreover, Iran did very little with its nuclear program during earlier periods when threats were arguably more acute and assurance less certain. Indeed, while the Shah was said to be obsessed with military matters and thought of himself as a military expert, the primary drivers of Iranian policy were as often a desire to consolidate his domestic position and elevate his personal stature. As an American intelligence report from 1969 noted, "The Shah sees himself in the role of a latter-day Cyrus the Great who will restore to Iran at least a portion of its old glory as a power to be reckoned with in its own part of the world:'16 This observation came just as the king was about to embark on his massive nuclear program.

SECURITY ASSURANCES AND IRAN

119

So what of Stein's intriguing analysis concerning the potential role of emotions and other psychological predilections and their impact on proliferation and assurance? Stein focuses on prospect theory and the role of emotion in shaping beliefs, in particular, the emotions of fear, anger, sadness, happiness, and disgust. Iran under the Shah would appear to be a less promising case for the application of prospect theory. In the 1950S, when the external security environment was relatively more dangerous and his domestic standing more precarious, it would seem that the Shah was in a domain of losses, but his behavior during this period is more risk-averse than risk acceptant. In the mid- and late 1970S, his regional and domestic fortunes improved, but he behaved in ways that appear more risk acceptant, at least as far as the pursuit of a nuclear program was concerned. This is not to suggest that emotion played no role in the Shah's behavior. It was not fear, anger, happiness, or disgust that seemed to drive the Iranian leader as much as a desire for prestige or status. By the mid-1970S, he was secure enough and wealthy enough to finally pursue his dream of being the next Cyrus and to put behind him the embarrassing or even humiliating events that had occurred earlier in his reign. On the question of whether particular attributes enhance the effectiveness of assurances, the results are not surprising, given the lack of a robust relationship between assurances and nuclear behavior. Of the hypotheses identified above as most likely to be relevant, four appear to fare poorly during the monarchial period. CENTO included a public declaration (H2), had a legal character (H3), was characterized by political and economic ties between the parties (H4), and was part of a larger strategy (H8)-in this case, containment. Despite these attributes, CENTO was perceived as weak. Political and economic-and also military-ties also increased in the bilateral U.S.-Iran relationship in the 1970S, which according to H4 should have made assurances more credible, yet this was the period of the Shah's most aggressive efforts to advance Iran's nuclear weapons program. The role of forward-deployed troops (H6) is more complicated. Both Britain and the United States had forward-deployed troops and bases in the region, which should have given the Shah confidence that help was nearby if required, but there were no Baghdad Pact or CENTO-specific troop deployments. Moreover, the Shah did not want U.S. troops on Iranian soil. Hosting American troops would have undermined any perception of independence, upset the Soviets (violating Iran's assurance to USSR), and made it easier for a foreign power to interfere in domestic politics.

120

JIM WALSH

During the time of the Shah, security threats and the amelioration of threats by security assurances were neither trivial nor predominant. Did the security threats matter? Yes. Did the assurances help? Probably. Were threats and assurances, traditionally conceived, the central factors explaining Iranian nuclear behavior? No. Domestic politics and the Shah's personal ambitions mattered more, and the available sources of assurance were never seen as credible enough to change the Shah's mind. THE REVOLUTIONARY PERIOD, 1979-2009 The Islamic Republic's Nuclear Program: Freeze, Reconstitution, Drift, Sprint

The 1979 revolution brought a halt to the Shah's nuclear project, as both the demand for and supply of nuclear technology ended. Western countries abruptly canceled their nuclear transfers even as the Islamic Republic of Iran (IRI) moved to withdraw from Eurodiff and other joint projects. Ayatollah Khomeini's government had multiple reasons for freezing the nuclear program. To begin with, it was the Shah's program, and thus tainted by association. Add to that a scarcity of funds, the emigration of nuclear and other scientists fleeing the revolution, a distrust of foreigners (who had a visible role in the program), and the more immediate challenges of domestic political consolidation and governance, and it is little wonder the program was suspended. By the mid-1980s, however, the government's attitude changed. Despite the war with Iraq or perhaps because of it, Iran's leadership decided to reconstitute the nuclear program. In 1984, it opened a research center at Isfahan and began encouraging Iranian nuclear expatriates to return home. Even with the renewed interest, though, the program suffered a number of problems. In 1984 and again in 1985, Iraq bombed Iran's reactor site at Bushehr. Most foreign suppliers were skittish about working with Iran, especially at a time when it was at war with Iraq. Last but certainly not least, the nuclear program appeared to suffer from poor internal management. Despite help from A. Q. Khan, which began in the late 1980s, Iran's nuclear program drifted without significant accomplishment. l ? Finally in 1997, the head of the nuclear program was replaced. Gholam Reza Aghazadeh took the reins and progress followed soon after. IS In 2002, an opposition group revealed the existence of previously secret nuclear activities, including a uranium enrichment facility at Natanz. Afterward, Iran admitted to more nuclear-related activities than it had previously declared to the Inter-

SECURITY ASSURANCES AND IRAN

121

national Atomic Energy Agency (IAEA). Faced with a threat of referral to the UN Security Council (UNSC), in 2003 Tehran agreed to temporarily suspend uranium enrichment and enter negotiations with the "EU-3" (Britain, France, and Germany). In 2006 talks broke down, and the IRI resumed uranium enrichment. The IAEA referred the Iran case to the United Nations, and the Security Council imposed the first of a series of sanctions on Iran. In response, Iran continued to build centrifuges. By 2010, AEOI had constructed more than eight thousand centrifuges and had made progress with a heavy water moderated reactor at Arak. A short time later, it began to enrich uranium to a level of 19.75 percent. In September 2009, both Iran and Western governments revealed that Iran had also covertly been working to build a second uranium enrichment facility near QomY A key question is the Islamic Republic's objective: what did it hope to accomplish with its recent advances in nuclear technology? Iranian officials have consistently denied any interest in nuclear weapons, citing both religious and strategic rationales. 2o Many analysts do not believe these claims and suspect that the IRI is seeking to build nuclear weapons. They point to A. Q. Kahn's transfer of weapons-related design information and Iran's missile program as evidence of the IRI's nuclear intentions. Others suggest that Tehran may have made a "capability decision" but not a "bomb decision:'21 Security Assurances and the Islamic Republic

There is no evidence that the Islamic Republic has been provided with a security assurance from another government, nor does it appear that Tehran has sought one. Indeed for most of its thirty years, the revolutionary regime has prided itself on its independence from the very same great powers that would be in a position to offer an assurance. 22 In August 2005, the EU -3 made a proposal that contained some securityrelated provisions. These included a pledge that the UK and France would reaffirm negative security assurances announced in 1995 in connection with the NPT renewal process, as well as a restatement of support for a Middle East WMD-Free Zone. Most of the proposal, however, focused on acknowledging Iran's right to civil nuclear power and offered assistance in finding external sources of nuclear fuel. In return, Iran would have to halt its efforts to develop an indigenous fuel cycle and take steps to assure the outside world of its compliance with the NPT. Iran promptly rejected this proposal as "an insult on the Iranian nation:' It complained that the EU -3 package imposed new demands

122

JIM WALSH

on Iran without adequately guaranteeing the supply of nuclear fuel and that the security provisions involved nothing new but merely repeated "previously made general commitments."23 In 2006, China, Russia, and the United States joined the EU-3 in diplomatic efforts to address Iran's nuclear program, creating a group known as the PS+l (for the five permanent UNSC members plus Germany). Subsequent proposals from the PS+l have included promises to avoid interference in Iran's internal affairs and discussions on regional security, and have restated support for a WMD-free zone for the Middle East, but the talks themselves have centered on the nuclear issue and the political and economic benefits that an agreement might produce. In short, the United States and the Europeans have given much greater emphasis to positive incentives than to security assurances.24 Finally, the events of 2011 in Libya might have given the Supreme Leader pause about the benefits to be gained from American and British security assurances. Colonel Gadhafi struck a deal with the United States and Britain in 2003, in which Libya agreed to give up its nuclear and other illicit weapons programs in return for regaining access to the community of nations and no longer being subject to sanctions and international opprobrium. As Wyn Bowen describes in the previous chapter, an informal, behind-the-scenes assurance from the Bush administration not to seek regime change in Tripoli appears to have been an important part of the deal, but the agreement could not reasonably have been construed as license for Gadhafi to start massacring his own domestic population. After reports that Gadhafi's forces would enter the besieged Libyan town of Benghazi and murder its civilians, NATO forces intervened and eventually helped topple the Libyan leader, leading to his death at the hands of rebel forces. Although it would be difficult to argue that the 2003 nuclear deal should have protected Gadhafi under these circumstances, it would not be incredible for the government in Tehran to view events in Libya as reason to doubt the efficacy of a negative security assurance provided by the very countries it believes had previously tried to topple the Islamic Republic. In sum, the IRI has not been offered a country-specific security assurance, and it might not have found such an assurance credible had it received one. Political Context and Threat Environment

Could assurances have mattered? A quick review of the threat environment since the start of the Islamic Republic shows that security threats were abundant. The Shah's departure was followed by a fierce and bloody struggle for power. The ascent of Ayatollah Ruholla Khomenei and the clerics did not end

SECURITY ASSURANCES AND IRAN

TABLE

123

6.3

Iran's Nuclear and Threat Environment, 1979-2009 2 1

Period 1979-84 1984-97

1997-2009

Nuclear-relevant threats

Regional NW states

USSR, U.S., UK, Iraq, Israel USSR/Russia, U.S., UK, Iraq, Israel U.S., UK, Israel, Russia

Pakistan Pakistan

Pakistan

4

Other threats Internal, Saudi Arabia Saudi Arabia

Internal, Saudi Arabia

political violence; assassinations, insurgency, and executions continued. And it was not long before Iraq took advantage of the opportunity presented by Iran's unsettled condition to launch an attack on its vulnerable neighbor. Now, in addition to everything else, the inexperienced, revolutionary, and unconsolidated leadership had to fight a major war. Khomenei and the government faced hostility at home, invasion by a potent rival on Iran's border, and opposition from the United States, Britain, Russia, and virtually all the Arab world. The young regime's Islamist, revolutionary ideology and promise to "export of the revolution" only served to add to the list of its international opponents. 25 Table 6.3 summarizes the Islamic Republic's potential nuclear and non-nuclear threats. For column two, the United States, USSR, United Kingdom, and Israel can all be classified as adversaries of the IRI that also possessed nuclear weapons, and the first three also enjoyed more than a two-to-one conventional advantage. Iraq posed a latent nuclear threat until the 1991 Gulf War and the subsequent dismantling of Saddam Hussein's nuclear program. Coding the USSR/Russia as an adversary is problematic insofar as the USSR went from being an opponent immediately after the revolution to a somewhat supportive political and economic ally after the collapse of the Soviet Union. Russia has tended to act as a counterweight as the United States and Europe have adopted an increasingly oppositional stance, especially after 2006. Still, Iranians do not trust Moscow, a sentiment reinforced by Russia's invasion of Georgia in 2008. In column three, Pakistan evolves from having a weapons program to being a weapons state. Even so, Iran did not view Pakistan as a military adversary, despite Islamabad's support for Afghanistan's Taliban government. For column four, one could argue that the Tehran has faced an ongoing internal threat since its founding, though several measures would suggest that

124

JIM WALSH

domestic insecurity has been most intense in the years after the revolution, in the late 1990S, and most recently following the June 12, 2009, election. After the Iran -Iraq War and prior to June l2, most Iranians who opposed the government preferred democratic evolution to violent revolution. Since then, the picture has been unsettled and unclear. The opposition has ranged from conservative pragmatists and former pillars of the regime to reformers to those newly radicalized by events following the disputed election. Further complicating matters are the anti-Ahmadinejad Principalists, who have supported the revolutionary regime but opposed the conservative Iranian president. Column four also lists Saudi Arabia, but here again, the situation is not a simple one. Iran has more than twice the population and a substantially larger GDP than Saudi Arabia, but Riyadh spends more than twice what Tehran does on its military and enjoys a significant qualitative edge over the IRI. It is unlikely that Saudi Arabia would ever attack Iran, and almost as improbable that Iran would attack the leader of OPEC, on whom it depends economically. That still leaves room, however, for political and proxy conflicts between two countries that clearly view each other as rivals. In short, for most of the revolutionary period, the primary threats from Tehran's perspective have been domestic opponents, foreign subversion, Iraq, and more recently, the possibility of direct military attack by the United States or Israel on Iranian nuclear facilities. Of these, the Iraq threat was the most direct and severe, but fears of a Western sponsored "color revolution" have come to predominate in recent years. Assessing the Hypotheses

As with the monarchial period, it is possible to make a simple, but probably erroneous, link between assurance and nuclear behavior. One could argue that (a) the Islamic Republic has not benefited from a security assurance, (b) it is seeking nuclear weapons or at least a nuclear weapons capability, and therefore (c) the hypothesis is sustained. As with the monarchial version of this proposition, the facts in this syllogism are accurate but the causal linkages appear problematic. In general, the history of the Islamic Republic is characterized by a fairly high and constant level of threat. Yet there have been some noteworthy changes over time, including the diminution of the Iraq threat in 1991 and its eventual removal in 2003, a general improvement in Iran-Russian relations, and the decline and then return of internal threats, especially following the June 2009

SECURITY ASSURANCES AND IRAN

125

election. Variation in the u.s. threat is less straightforward. The Islamic Republic has consistently portrayed the United States as "the Great Satan:' but threat perceptions could well have spiked in the years immediately following 9/11. The United States sent troops into Afghanistan and Iraq, both of which border Iran, and in 2002 President Bush named Iran as part of the "axis of evil:' suggesting a possible U.S. objective of regime change. However, any spike in threat perceptions was probably short lived. The U.S. military became overextended in Iraq and Afghanistan, and by Bush's second term it was clear that the administration was not going to undertake any further military invasions. Against this background, the Islamic Republic's nuclear program has manifest substantial variation, most notably a very long interval of dormancy followed by rapid progress beginning in the late 1990s-a time when its chief adversary (Iraq) had been largely defanged and the United States, under President Clinton, was cautiously considering engagement. Moreover, Iran went on to produce its first enriched uranium two years after Saddam Hussein was deposed and at a time when U.S. power in the region was seen as having been diminished because of the difficulties U.S. forces were experiencing in Iraq. Table 6.4 summarizes the IRI's threat environment and nuclear activity during this time. It is not implausible that Iran's pursuit of nuclear technology is a function of its threat environment and the related fact that the international community tacitly accepted Iraq's use of chemical weapons against Iran in the 1980s.26 Indeed, the timeline suggests that the Iran-Iraq War was a catalyst for restarting the nuclear program. Still, the ongoing puzzle of Iran's nuclear behavior is not why it may have been interested in nuclear weapons, but why it waited so long to get serious about it. It took Tehran some twenty years to construct its first centrifuge, TABLE 6.4

Iran's Threats, Assurances, and Nuclear Behavior, 1979-2009 6

2

Nuclear-relevant threats

Regional NW states

Pakistan

1984-97

USSR, U.S., UK, Iraq, Israel USSR/Russia, U.S., UK, Iraq, Israel

19972009

U.S., UK, Israel, Russia

Pakistan

1

Period 1979-84

Pakistan

4

Security

Other threats

assurances

Internal, Saudi Arabia Saudi Arabia

NA

Internal, Saudi Arabia

NA

NA

Nuclear development Nuclear efforts frozen/cut Resume nuclear program, but drift, modest progress Aggressive nuclear gains

126

JIM WALSH

despite receiving help from the A. Q. Kahn network. The IRI's most dramatic progress in the nuclear field has come in the last several years. In this same interval, the probability of invasion by Iraq, Russia, or the United States was near zero and arguably at its lowest point since the 1979 revolution. (Of course, the risk of U.S. or Israeli air strikes remains, but if Iran were concerned about that, its best course to forestall such an attack would be to halt its nuclear activities.) Given that the Islamic Republic's nuclear activities are not strongly correlated with variations in the threat environment and that it has received no security assurances, none of Knopf's hypotheses can be evaluated directly in this time period. Here again, though, it is worth considering Stein's hypotheses drawn from psychology and neuroscience. As with the period of the Shah, the application of prospect theory is problematic, again because the Islamic Republic has made its most earnest efforts to pursue nuclear technology at a time when threats were relatively less robust-that is, Tehran appeared risk acceptant when in a domain of gains, and vice versa. Stein raises another variable, however, that shows promise for the Iranian case. She cites research showing that an actor will reject an otherwise rational negotiated settlement if it views its relationship with other actors as unfair. The leaders of the Islamic Republic have long expressed the feeling that Iran has been a victim of foreign interference and unfair treatment. These emotions are accompanied by the often-voiced view that Iran does not receive the respect it deserves. It would not be surprising, therefore, if the feelings of Iranian leaders influence their negotiating stances and even their nuclear behavior. In the Iranian case, these feelings of unfairness and disrespect are accompanied by a set of distinct but related emotions and beliefs that Iran not only deserves respect but, beyond that, has a special place in the international system (above other countries) and should achieve technical and political autonomy. Finally, beyond traditional explanations that focus on security threats and newer hypotheses based on psychology and neuroscience, one should consider the role of domestic and organizational politics. In the last few years, there has been a consistent effort on the part of Iranian officials to use the nuclear issue to stoke nationalism and to rally a sense of collective pride over technological accomplishment-that is, in the service of domestic politics rather than countering foreign threats. This dynamic in domestic politics may have scuttled a tentative deal in fall 2009. Iranian negotiators indicated that the country would be willing to send much of its enriched uranium abroad for processing into nuclear fuel, a step that would have slowed down Iran's ability to convert its ura-

SECURITY ASSURANCES AND IRAN

127

nium stocks into weapons-grade materials. Although Iran's hardline President Ahmadinejad was apparently initially supportive, the government repudiated the deal after it received extensive criticism at home, including from figures associated with the reform movement in the recently concluded and then still contested June election.27 OVERALL CASE FINDINGS

The case ofIran provides a limited but suggestive basis for assessing the performance of the hypotheses on security assurances. In the main, the hypotheses reviewed here are not sustained. For both the monarchial and the revolutionary phases, the regimes' most aggressive moves in the nuclear arena came at a time when threats were less pronounced than in earlier periods, and in the case of the Shah, where U.S.-Iranian relations were at an apex. The case for assurance looks somewhat stronger for the revolutionary period, where the absence of a security assurance is associated with the presence of nuclear activity, but the relationship is not as robust as it might be. One cannot help but suspect that even if Iran had faced no credible external threat, it still would have pursued a nuclear program, be it under the Shah or under the Islamic Republic. Pride, prestige, the ghost of Persian empires past, and domestic political problems might have been motivation enough. ADAPTING SECURITY ASSURANCES TO THE IRANIAN CONTEXT

Nonproliferation-related assurances, whether the generalized assurances associated with the NPT or a country-specific pledge from one or more nuclear powers, have either been unavailable to Iran or incapable of changing its nuclear behavior. Given the limits of traditional security assurances in the Iranian context, how might assurances be modified to make them more effective? Two approaches are offered. The first attempts to make assurances more relevant to the security situation Iran faces, and the second treats security assurances as a political instrument whose effect is intended to address motivations unrelated to security concerns. The Problem of Multiple External Threats

In situations where states face multiple military threats, a security assurance that addresses just one threat might not be sufficient to reduce a country's inclination to develop nuclear weapons. In theory, an assurance regarding a single

128

JIM WALSH

threat could help the beleaguered state, insofar as it might allow it to redirect energy and attention to its remaining threats. Still, it may take only one robust nuclear-relevant threat to trip an interest in nuclear weapons. In short, in the context of multiple threats, assurance regarding a single threat might be useful but not sufficient. One alternative is for a great power to provide a comprehensive or integrated security assurance, one that would apply to any potential threat posed by another state. Governments are loath to provide such broad commitments, even for allies, and so it is difficult to imagine that there would be much enthusiasm in a country such as the United States for offering such a guarantee to a perceived enemy such as Iran. Moreover, the potential great power guarantors in this situation (the United States and Russia) already have friends in the region, friends who will wonder, first, why Iran is getting what they cannot have, and, second, what such an assurance implies for the commitments they have already received. Saudi Arabia and Israel, for example, are unlikely to respond positively to an American commitment to support the defense of Iran. And they will hardly be happier if it is Russia that supplies the protection. The assurer could offer guns rather than a political commitment. This might make Iran feel more secure and have the additional effect of giving Iran's militarya stake in continued good relations with the assurer, but again many in the region will object and, once given, arms are difficult to retrieve if the recipient does not live up to its obligations. Whether the offer is arms or protection, any assurance provided may involve a security dilemma, where efforts to assure one party cause its regional rivals to feel more insecure and spur them to take action to redress the perceived imbalance. Another possibility is the establishment of a regional security architecture in which the major regional players and the great powers all participate. Implicit in any such arrangement would be an understanding that the members of the institution would refrain from armed force against other members. The goal of such an enterprise would be to increase transparency, coordination, and confidence among the regional players and to provide a forum or mechanism for managing their differences over security. Such institutions are difficult to establish and maintain and are usually less than fully satisfying even in favorable environments, and the Middle East is far from a favorable region for establishing a new security institution. That said, a regional security institution might reduce the sense of threat at the margin and may provide an instrument for preventing an escalation of tensions that could spur a renewed interest in nuclear weapons.

SECURITY ASSURANCES AND IRAN

129

None of the foregoing alternatives would appear to provide a sure-fire solution to the problem of assurance, but it may be that this particular problem, the problem of multiple external threats, is one that will have to be addressed with limited initiatives promising partial success, and that threat minimization rather than absolute assurance will have to suffice. The Problem of the Assurance Provider

With Iran, the question of who provides an assurance is also an issue. In theory, the provider should be among the strongest states in the system so that it can deliver the required protection. Belgium offering a security assurance to Iran will not count for much. The conundrum, of course, is that the countries that are the most powerful are the very ones who have, in the past, invaded or otherwise threatened Iran's governments. The paradox of "enemy as protector" is further complicated by the fact that Iran's revolutionary, Islamic government has made anti-imperialism and the threat of foreign subterfuge core elements of its ideology. Indeed, it may be reluctant to give up the anti-imperialism card a time when there is a challenge to its legitimacy at home. And as discussed above, the case of Gadhafi's ouster with the assistance of u.s. and NATO military strikes further complicates the equation. One possibility is for any arrangement to involve a third party that all the countries trust. China plays that role in Six-Party Talks. Perhaps China could playa similar role in this case despite the absence of a historical relationship with Iran, but the United States and Russia are unlikely to welcome that. It is conceivable that Turkey could also play this role; it is a militarily and geopolitically powerful state with a majority Muslim population that has good relations with both Iran and the PS+l. Of course, Turkey is no match for the United States or Russia, but it is a NATO member and neither the United States nor Russia has an interest in antagonizing Turkey. It may also be possible to design specific instrumentalities for specific problems. With regard to Iranian enrichment, for example, some analysts have proposed multinational enrichment on Iranian soil as a way past the current impasse. 28 Whatever the merits or demerits of the scheme, one of the less frequently discussed aspects of such an arrangement is the confidence it would provide to Iran that its facilities would not be subject to military attack. Neither the United States nor Israel would likely bomb a facility if it required killing scores of international personnel.

130 JIM WALSH

The Problem of Internal Threats

At first glance, it would seem that internal threats are irrelevant to and/or beyond the scope of potential security assurances provided by outside powers. The case of Iran illustrates that this is not necessarily the case. Its history and current ideological bent encourage Iran's leaders to claim that foreign plots and policies of regime change pose a threat to its continued survival. To the extent that this is a heartfelt suspicion, it poses an obstacle to any diplomatic resolution of the nuclear dispute and related controversies. It would be difficult for the IRI to believe that "the West" is actively trying to subvert the regime and simultaneously bargaining in good faith on other security issues. How might the United States and other states offer assurances that they will not interfere in Iran's domestic affairs? The United States, for example, could publicly reaffirm the 1981 Algiers Accord, which obliges the parties not to interfere in each other's domestic politics. It could demonstrably end various overt and possibly covert programs aimed at "democracy promotion;' a phrase that the Iranian government equates with regime change. It might get tough on the Mujahadeen-e-Khalq (MEK, aka People's Mujahedin of Iran, or PMOI) or other groups that Tehran views as terrorist threats, and it could set up mechanisms for information sharing and dispute management on the assumption that transparency will reduce suspicion. It is unclear whether any or all of these steps would be sufficient. They could also carry near-term and potentially long-term political costs. Promising not to seek domestic change in Iran after protesters have been hanged and young people beaten in the streets will be difficult to sell to domestic constituencies in the United States and Europe, and may be too difficult for Western political leaders to stomach themselves. It also raises the possibility that, should a postrevolutionary government come to power, it might harbor ill will toward the P5+l countries for having abandoned Iranian reformers in a cynical bid to further their own national interests. Given the potential political costs to both sides, one could imagine that any arrangement might be kept secret or at least out of the headlines. All parties would thus be free to engage in the rhetoric of their choice but with the common understanding that assurances had been given and accepted. This might be a more expedient course, but as Knopf has hypothesized, private commitments may not be viewed as sticky as public commitments, and thus the force of the assurance might be degraded.

SECURITY ASSURANCES AND IRAN

131

On the flip side, failure to make progress on the nuclear issue may open the possibility that a delegitimized regime will turn to the bomb to bolster its domestic and international standing. Prior to the 2009 domestic crisis, many analysts held the view that Iran wanted a nuclear weapons capability but had not yet made a decision to actually build a bomb. According to this view, a capability decision would allow Iran to enjoy some of the advantages of a nuclear status without paying all the costs. An ongoing crisis of legitimacy might alter that calculation, however. Security Assurances as a Political Instrument

In the preceding sections, the concept of assurance has been treated primarily as a matter of security. The aim has been to explore how traditional conceptions of assurance might be modified in the face of particular security conditions-multiple external threats, provision of assurance by formerly hostile states, and the presence of internal security threats. In this section, attention shifts from assurance as security instrument to assurance as political instrument. In this context, one can imagine circumstances in which a potential assurance recipient (1) does not face an acute, objective security threat, (2) wrongly perceives that it does not face such a threat, (3) believes it can successfully defend against such a threat, or (4) rightly perceives a threat but faces still larger challenges that take a higher priority. Even under these circumstances, a security assurance might prove to be a valuable instrument because of its political effects. At its simplest, one can conceive of a security assurance as an act that might build trust or confidence among the parties. Any such increase in the level of confidence might then have broader consequences for dispute settlement. Alternatively, a security assurance, even if not "needed;' might be considered a status gain or something that brings greater legitimacy to the recipient. In this same vein, the assurance could be treated as a deliverable, a tangible benefit that the recipient's government can point to as evidence of its negotiating success. Finally, the provision of a security assurance could have a disproportionate and positive impact on particular political actors or domestic constituencies, which in turn, could alter the distribution of votes or support for a negotiated settlement. In the case of Iran, for example, it has been rumored that Ali Larijani, the current leader of the Majlis and former nuclear negotiator, believes that Iran should receive a security assurance. 29 If this were true of Larijani or

132

JIM WALSH

another key leader, it would be a minority view among Iranian elites. Under these circumstances, the offer of a security assurance could nevertheless prove useful, if only because it might shift the position of an actor who might have a decisive influence on a negotiation outcome. It has to be added, however, that the political consequences of security assurances are not fixed. In principle, they could have either positive or negative political effects. So for example, the Iranian military or organizations within the military might see a security assurance as benefiting their interests (for example, by resulting in access to arms imports) or undermining their interests (such as by reducing the rationale for big budgets or broad jurisdiction). If a more broadly political assurance were offered and had the intended effect, this would be compatible with some of Knopf's analysis in Chapter 2. The last scenario sketched here is consistent with hypothesis 12, which holds that assurances can be effective if they alter internal debates in a favorable direction. More important, an emphasis on assurances as a political instrument supports Knopf's suggestion that NPT-related assurances might sometimes need to be embedded in a broader strategy of reassurance or alliance-related assurance in order to overcome the other side's doubts. But Knopf still tends to assume that the goal is to address underlying security concerns, while the point here is that assurances might have beneficial political effects even if the recipient's interest in nuclear activities is not primarily motivated by security. DRAWING THE RIGHT CONCLUSIONS: THE FUTURE UTILITY OF SECURITY ASSURANCES

The objective of this inquiry has been to evaluate the influence of security assurances on Iranian nuclear behavior. After assessing the historical record, it concludes that assurances did not playa major role in determining nuclear outcomes. Security assurances to the Shah, in the form of the Baghdad Pact and U.S.-Iran bilateral relationship, had little effect. Security assurances have barely been offered to the Islamic Republic, but it is also not clear how effective stronger assurances could have been, given the many concerns driving the Iranian nuclear program today. Instead, it appears that while security threats and related assurances were not without causal consequence, other factors often predominated. It would be a mistake, however, to conclude that, just because assurances did not playa significant role in the Iranian case, this means that assurances are of little value in general or could not be useful for Iran in the future. If anything,

SECURITY ASSURANCES AND IRAN

133

the example of Iran is a useful starting point for imagining ways in which security assurances can be made more relevant or more effective for a contemporary context. Iran will not be the only potential proliferator facing multiple external threats, internal threats, or the need for assurances from the very states that threatened it in the past. The Iranian record also offers the interesting case of what might be called "reverse assurance;' where the weaker party makes commitments to a stronger state in order to reduce the level of threat-for example, the Shah's promise to the USSR not to allow American bases on Iranian soil despite the obvious security commitment between the two allies. The difficulty of providing effective assurance in the circumstances analyzed here should not be underestimated. Scholars and policymakers alike need to confront these challenges directly and fashion new instrumentalities that make assurance relevant to to day's circumstances. States seeking to prevent the spread of nuclear weapons will continue to need security assurances as a policy instrument. Scholars will have to help them develop new forms and applications of assurance to accomplish that worthy aim. NOTES The author benefited greatly from the support and insights offered by a number of people, including the participants in the Security Assurances Workshop, Dan Altman, Corie Walsh, and two anonymous reviewers. 1.

For a detailed chronology of Iran's nuclear program, see Nuclear Threat Initiative

(NTI), Iran Nuclear Chronology, http://www.ntLorg/e_research/profiles/Iran/Nuclear/ chronology.html (accessed January 10, 2011). On the Shah's desire for a "crash program;' see Memorandum of Conversation, President [Eisenhower l, the Shah of Iran, Tehran, December 14, 1959, Foreign Relations of the United States, 1958-1960, vol. VII (Washington, DC: Department of State, 1993), 659 n. 2. 2. NTI, Iran Nuclear Chronology. On u.S. support for the Shah's nuclear program, see also Mustafa Kibaroglu, "Good for the Shah, Banned for the Mullahs: The West and Iran's Quest for Nuclear Power;' Middle East Journal 60, no. 2 (Spring 2006): 207-32; and Dafna Linzer, "Past Arguments Don't Square with Current Iran Policy;' Washington Post, March 27, 2005, 15. 3. NTI, Iran Nuclear Chronology; David Patrikarakos, "How Iran Went Nuclear;'

New Statesman, June 18, 2009, http://www.newstatesman.com/international-politics/2009/06/nuclear-programme-iran-world. 4. NTI, Iran Nuclear Chronology. 5. For the text of the agreement, see "Baghdad Pact; February 4, 1955;' The Avalon Project, Yale Law School, http://avalon.law.yale.edul2oth_centurylbaghdad.asp.

134 JIM WALSH

6. "Telegram from the Embassy in Iran to the Department of State, Tehran;' August 31, 1965, Foreign Relations of the United States, 1964-1968, vol. XXII, Iran (Washington, DC: Department of State), http://nasser.bibalex.orglData/USDocWebIHTMLlXXII Iran%201964-1968/www.state.gov/www/aboucstate/history/voLxxiiIj.html. For a brief account of CENTO's unhappy history, see Panagiotis Dimitrakis, "British Diplomacy and the Decline of CENTO;' Comparative Strategy 28, no. 9 (2009): 317-31. 7. "Telegram from the Department of State to the Embassy in the United Kingdom;' February 26, 1965, Foreign Relations of the United States, 1964-68, vol. XXII, Iran (Washington, DC: Department of State), http://history.state.gov/historicaldocumentsl frus1964-68v221 d70. 8. CIA, "Intelligence Memorandum: Recent Trends in Iranian Arms Procurement;' May 1972, Foreign Relations ofthe United States, 1969-76, vol. E-4, Documents on Iran and Iraq, 1969-72, Document 181 (Washington, DC: Department of State, 2006), http://history.state.gov/historicaldocuments/frus1969-76ve04/d181 (accessed October 17, 2011). 9. Memorandum of Conversation, H.LM. the Shahinshah and Amb. Henry Lodge, February 2, 1958, Foreign Relations of the United States, 1958-60, vol. VII (Washington, DC: Department of State, 1993), 535-37. 10. In 1962, the Shah and the Soviet Union exchanged notes in which Iran promised that it would not be a base for attack against the USSR. This was followed in subsequent years by economic agreements and the sale of Soviet arms to Iran. Helen Chapin Metz, ed., Iran: A Country Study (Washington, DC: GPO for the Library of Congress, 1987), http://countrystudies.us/iran/19.htm. 11. On relations between Iran and Israel, see Trita Parsi, Treacherous Alliance-The Secret Dealings of Iran, Israel and the United States (New Haven: Yale University Press,

2007). It is worth noting as well that Turkey was a member of NATO, a nuclear alliance, and had U.S. nuclear weapons stationed on its territory. 12. "Telegram from the Embassy in Iran to the Department of State;' May 29, 1958,

Foreign Relations of the United States, 1958-60, vol. VII (Washington, DC: Department of State, 1993), 550. 13. "Dispatch from the Embassy in Iran to the Department of State;' February 10, 1958,

Foreign Relations of the United States, 1958-60, vol. VII (Washington, DC: Depart-

ment of State, 1993), 537-42. 14. CIA, "Intelligence Memorandum: Recent Trends in Iranian Arms Procurement:' 15. After the 1967 War, the government froze all funding for the nuclear program. Shyam Bhatia, Nuclear Rivals in the Middle East (London: Routledge, 1988), 57-58. For a detailed discussion of Egypt's nuclear program in this period, see Jim Walsh, "Bombs Unbuilt: Power, Ideas, and Institutions in International Politics;' Ph.D. diss., MIT, May 2000, 201-9. 16. CIA, "Intelligence Report: Centers of Power in Iran;' May 1972, no. 2035/72,

SECURITY ASSURANCES AND IRAN

135

Foreign Relations of the United States, 1969-76, vol. E-4, Documents on Iran and Iraq (Washington, DC: Department of State, 1993), http://history.state.gov/historicaldocuments/frus1969-76ve04/d180. 17. NTI, Iran Nuclear Chronology; Joseph Cirincione, Jon B. Wolfsthal, and Miriam Rajikumar, Deadly Arsenals: Nuclear, Biological, and Chemical Threats, 2nd ed. (Washington, DC: Carnegie Endowment for International Peace, 2005), esp. 302-3. 18. NTI, Iran Profile: Nuclear Facilities, http://www.nti.orgle_research/profiles/Iran/ Nuclear/facilities.html. 19. Since 2003, the best sources on Iran's evolving nuclear infrastructure are the periodic reports of the IAEA ("Implementation of the NPT Safeguards Agreement in the Islamic Republic of Iran;' Report by the Director General) and the supplemental analysis offered by David Albright of the Institute for Science and International Security, which can be found at http://www.isis-online.orgl. 20. On Iran's disavowal of nuclear weapons on religious grounds, see "Iran Warns over Nuclear Impasse;' CNN, August 11, 2005, http://www.cnn.comI2005/WORLD/europe/08/1O/iran.iaea/index.html. See also Karl Vick, "In Iran, Gray Area on Nuclear Weapons: Religious View Is Not Absolute;' Washington Post, June 21, 2006, A15. During the Iran-Iraq War, Iranian religious officials were reported to have resisted the development of chemical weapons on religious grounds, despite their use by Iraq. See, for example, Javed Ali, "Chemical Weapons and the Iran-Iraq War: A Case Study in Noncompliance;, Nonproliferation Review 8, no. 1 (Spring 2001): 43-58; Joost R. Hiltermann, "Outsiders as Enablers: Consequences and Lessons from International Silence on Iraq's Use of Chemical Weapons during the Iran-Iraq War;' in Iran, Iraq, and the Legacies of War, ed. Lawrence G. Potter and Gary Sick (New York: Palgrave Macmillan, 2004), 151-66. 21. On Iran, the U.S. intelligence community has concluded that "Iran is keeping open the option to develop nuclear weapons in part by developing various nuclear capabilities that bring it closer to being able to produce such weapons, should it choose to do so. We do not know however if Iran will eventually decide to build nuclear weapons" (emphasis added). Dennis Blair, "Annual Threat Assessment of the US Intelligence Community for the Senate Select Committee on Intelligence;' February 2, 2010, 13, http://www.dni. gov/testimonies/ 20100202_testimony. pdf. 22. On decision-making in Iranian foreign and security policy, see Abbas Maleki, "Decision Making in Iran's Foreign Policy: A Heuristic Approach" (Tehran: International Institute of Caspian Studies, 2002); Jalil Roshandel, "Evolution of the DecisionMaking Process in Iranian Foreign Policy (1979-1999)" (Copenhagen: DIIS [previously COPRI], 1999); Heidar Ali Balouji, "The Process of National Security Decision Making in Iran;' in Europe and Iran: Perspectives on Nonproliferation, ed. Shannon N. Kile, SIPRI Research Report no. 21 (Stockholm: Oxford University Press, 2005), 72-96; Shahram Chubin, "Whither Iran Reform, Domestic Politics and National Security;' Adelphi Paper 342, IISS (London: Oxford University Press, 2002).

136 JIM WALSH

23. The Acronym Institute, "Disarmament Documentation: E3 Proposal to Iran, August 5, 2005;' www.acronym.org.uk!docS/0508/doc03.htm. 24. For an overview of the various proposals made to Iran, see Arms Control Association, "History of Official Proposals on the Iranian Nuclear Issue;' fact sheet, http:// www.armscontrol.orgifactsheets/Iran_Nuclear_Proposals. 25. Shireen T. Hunter, Iran and the World (Bloomington: Indian University Press, 1990); Dilip Hiro, The Iranian Labyrinth: Journeys through Theocratic Iran and Its Furies (New York: Nation Books, 2005). 26. Patrikarakos, "How Iran Went Nuclear:' 27. David E. Sanger, "Iran Said to Ignore Effort to Salvage Nuclear Deal;' New York

Times, November 9, 2009, A4. 28. See, for example, William Luers, Thomas Pickering, and James Walsh, "A Solution for the U.S.-Iran Nuclear Standoff;' New York Review of Books 55, no. 4 (March 20, 2008): 19-22. 29. It has been suggested to this author that Ali Larijani values a security assurance, though evidence to support this claim is difficult to find.

Maintaining Japan's Non-Nuclear Identity: The Role of U.S. Security Assurances Yuki Tatsumi

Japan has a unique profile among the non-nuclear weapon states in the world: it voluntarily renounced the possession of nuclear weapons despite having one of the world's most advanced civil nuclear energy programs and the technological capacity to become a nuclear-armed state. Moreover, Japan's renunciation of nuclear weapons is also a part of its post-World War II national identity, an identity that still shapes Japan's self-image today. Since the 1990S, however, rising tensions on the Korean Peninsula over North Korea's nuclear program have prompted debate within Japan regarding its decades-old non-nuclear policy. While there is little sign that Japan will abandon this policy in the near future, frequently expressed concerns about Japan's joining the nuclear club make it important to consider what contributed to Japan's decision to forswear nuclear weapons and the role security assurances have played in bringing about and maintaining this decision. For Japan, the most important assurance has been a U.S. security guarantee in the form of a bilateral alliance and extended nuclear deterrence. Given the considerable impact that the Fukushima Dai-ichi nuclear power plant accident has had on the Japanese public's attitude toward nuclear power, antinuclear sentiment seems to have regained strength in Japan since the March 2011 earthquake and tsunami triggered widespread damage and radiation releases at the four power stations in Fukushima Dai-ichi. Still, given that Japan continues to face nuclear weapon threats in its neighborhood, the question of whether Japan might go nuclear remains important. This chapter proceeds in four sections. First, it traces the evolution of Japan's non-nuclear policy since 1945. Second, it describes the security assurances 137

138 YUKI TATSUMI

that have been offered to Japan and the role those assurances have played in Japan's decision to renounce nuclear weapons and maintain this decision. In this context, it reviews past debates within Japan on a nuclear option. Third, it discusses how different security assurance hypotheses may be applicable to explain Japan's behavior. The chapter ends by exploring the key factors likely to shape future debates in Japan on the possession of nuclear weapons, including how perceptions of the reliability of u.s. security assurances may affect those debates. The chapter concludes that Japan's non-nuclear weapon status has multiple causes. In addition to the positive security assurance from the United States, Tokyo's recognition of the negative regional security implications of going nuclear and domestic antinuclear sentiment have also been important. This suggests that a future weakening of the U.S. security guarantee might not automatically lead Japan to reconsider its decision. Yet attention to the credibility of the security assurance still remains a prudent nonproliferation investment. JAPAN'S NUCLEAR POLICY: A SNAPSHOT

Japan's nuclear policy has been grounded in the "Three Non-Nuclear Principles (Hikaku San-gensoku):, These three principles, introduced by Prime Minister Eisaku Sato in 1967, declare that Japan will not "produce, possess, or allow the introduction of nuclear weapons onto Japanese soil."! Sato subsequently placed these principles in the larger context of "Four Pillars of Nuclear Policy (Kaku Yon Seisaku)": (1) promotion of the peaceful use of nuclear energy; (2) efforts in global nuclear disarmament; (3) reliance on U.S. extended deterrence; and (4) steadfast maintenance of the Three Non -Nuclear Principles. 2 Based on these fundamental principles, Japan's postwar nuclear policy has had three key dimensions: renunciation of nuclear weapons, reliance on U.S. extended nuclear deterrence, and commitment to develop a civil nuclear program. These three dimensions have been closely interconnected through both Japan's commitment to the nuclear nonproliferation regime and its establishment of a domestic legal framework. Renunciation of nuclear weapons has been the most pronounced dimension of Japan's nuclear policy. In this context, support for nuclear disarmament has been a prominent feature of Japan's postwar diplomacy. Tokyo considers the nuclear Non-Proliferation Treaty (NPT)-and International Atomic Energy Agency (IAEA) safeguards and the Comprehensive Test-Ban Treaty (CTBT) that support the NPT-as the most important foundation for international

MAINTAINING JAPAN'S NON-NUCLEAR IDENTITY

139

nuclear disarmament efforts. Japan has also historically subjected itself to what is probably the largest share of IAEA inspections, and in December 1999 it became the first state with significant nuclear activities to implement the Additional Protocol (adopted by the IAEA in May 1997), permitting expanded inspection activities. It has also annually submitted a draft resolution on nuclear disarmament to the UN General Assembly since 1994. Finally, in 2008, Japan cosponsored with the government of Australia the International Commission on Nuclear Non-Proliferation and Disarmament (ICCND), producing a report that sought to identify an achievable path to the eventual elimination of nuclear arms.3 Japan's decision to abjure nuclear weapons came after some hesitation, however. While the NPT opened for signature in 1968, Japan did not sign it until 1970, and it took Tokyo six years to ratify the treaty. The most important factor behind this delay was Japan's concern that NPT safeguards would hinder its nuclear power industry, especially given that a Japan poor in natural energy resources had decided to rely on nuclear power as a major energy source. Japan acceded to the NPT only after the IAEA produced a safeguards agreement addressing Japan's concerns. 4 Since then, promoting nuclear disarmament has been one of the key elements in Japanese foreign policy. 5 While less dramatic than its decisions regarding nuclear weapons, Japan's commitment to develop its civil nuclear program has the longest history. In fact, even prior to adhering to the non-nuclear policy principles, Japan had committed itself to the peaceful use of nuclear energy by signing on to a series of international treaties and agreements in order to develop an indigenous civilian nuclear power industry. In November 1955, Japan signed a bilateral agreement with the United States that allowed the latter to loan highly enriched uranium to Japan for nuclear reactor research. The agreement imposed the conditions that (1) the spent fuel would be returned to the United States, and (2) Japan would provide an after-use report on the loaned uranium. 6 Japan and the United States took this initial arrangement a step further when the two countries signed the Agreement for Cooperation between Japan and the United States Concerning the Peaceful Uses of Nuclear Energy in February 1968. 7 Japan also became one of the first nations to join the IAEA when it was established in 1957, thereby reaffirming its commitment to the peaceful use of nuclear power. Japan has institutionalized its commitment to the peaceful use of nuclear energy through a series of legislative acts since the 1950S. The 1955 Atomic Energy Basic Law (Genshi-ryoku Kihon-ho) obligates the Japanese government to limit

140

YUKI TATSUMI

all research, development, and use of nuclear energy only to "peaceful purposes:'8 The Basic Law provides the principles on which additional laws have been enacted to regulate Japan's civil nuclear power activities. Japan's reliance on u.s. extended nuclear deterrence has been another important dimension of its nuclear policy. However, concerns about North Korea's nuclear program and China's augmentation of its nuclear forces, combined with a recent u.s. policy of decreasing its dependence on nuclear weapons, have prompted questions regarding the credibility of u.s. extended nuclear deterrence among Japanese political leaders, government officials, and other intellectuals in recent years. Subsequent sections of this chapter analyze the u.s. security assurance and current perceptions of it in Japan. The coexistence of these three dimensions makes Japan's non-nuclear policy unique. On the one hand, as the remainder of this chapter will show, Japan's renunciation of nuclear weapons remains strong. On the other hand, Japan's reliance on u.s. extended nuclear deterrence for its security from nuclear threats exposes it to a dilemma of how a country that is ultimately protected by the nuclear weapons of its ally can push for international nuclear disarmament. Japan's civil nuclear energy program further complicates its non-nuclear policy. Japan has been among the world's largest users of nuclear energy, ranking third behind only the United States and France. 9 It has a robust civil nuclear energy program with fifty-five light water power reactors, operated by ten electric power companies, and at least prior to the 2011 nuclear accident had plans to build more plants in the future. lo In the "Action Plan for a Low Carbon-Emission Society" (Tei- Tanso Shakai zukuri Koudou Keikaku), which was adopted by the Aso Cabinet in July 2008, Japan placed nuclear energy as "the core of the efforts to achieve a low carbon emission society:'ll Japan's pursuit of a closed nuclear fuel cycle in its civil nuclear energy program has been the feature that most prompts concern in nonproliferation circles. A closed, complete fuel cycle can create a ready supply of materials that can be used to construct a nuclear weapon, putting Japan's nonproliferation commitments into questionY Finally, Japan's choice to become a non-nuclear weapon state has deep psychological underpinnings in Japan's postwar identity as a heiwa kokka ("peace state"),13 as well as "the only country that has been a victim of an atomic bomb" (yui-itsu no hibaku-koku).14 As then Foreign Minister Hirofumi Nakasone succinctly put it: ''As Japan suffered nuclear catastrophes in Hiroshima and Nagasaki, the country knows the horror of nuclear devastation from its own experience .... I believe it is Japan's mission to convey to all people around the world

MAINTAINING JAPAN'S NON-NUCLEAR IDENTITY

141

the facts of the calamity of nuclear bombings that happened in August 1945 in Hiroshima and Nagasaki, across the boundaries of various political viewpoints and ideologies:'15 The resilience of this postwar identity in Japan-and the so-called nuclear allergy that accompanies it-goes a long way to supplement security assurances provided by the United States in encouraging Japan to maintain its non-nuclear policy. THE SECURITY ASSURANCE PROVIDED TO JAPAN AND ITS IMPACT

The primary source of security assurance for Japan is the U.S.-Japan alliance. Postwar Japan pursued a national strategy anchored by the "Yoshida Doctrine;' in which military power plays very little role. Formulated by the influential postwar Prime Minister Shigeru Yoshida, this strategy gave overriding priority to economic growth. Under this doctrine, Japan's national security policy rested on three core principles: the minimal role of its military power, reliance on the bilateral alliance with the United States for its security, and efforts to achieve security through other means of national power. 16 As a result, Japanese security became closely linked to its security relationship with the United States. The 1951 Security Treaty (replaced in 1960 by the Treaty of Mutual Cooperation and Security) permitted the United States to have military bases on Japanese territory and committed it to defend Japan in the event of an attack. In particular, Article 5 of the Mutual Security Treaty obligates the United States to come to Japan's defense when it faces external aggression. This U.S. defense commitment of Japan has always been understood to include nuclear extended deterrence (more commonly referred to as a "nuclear umbrella"), creating the foundation of the U.S. positive security assurance to Japan. Indeed, as noted earlier, "the Four Pillars of Non-Nuclear Policy" set forth by Prime Minister Eisaku Sato explicitly include "reliance on U.S. extended deterrence" as a part ofJapan's non-nuclear policy. 17 Japan has not relied solely on the U.S. security guarantee, but rather views it and the international nuclear nonproliferation regime as both being important. To Japan, the NPT and IAEA inspections augment the bilateral security assurance provided by the United States by helping reduce the number of countries that might otherwise go nuclear. IS The security assurance provided by the United States, particularly its "nuclear umbrella;' has played an important role in Japan maintaining its policy of renouncing nuclear weapons. Even though Japan's constitution, largely drafted

142

YUKI TATSUMI

by u.s. occupying forces after World War II, denied Japan the right of belligerency and prohibited Japan from building its defense capability beyond what is minimally necessary for self-defense, the Japanese government has never interpreted its constitution as ruling out the option of an independent nuclearweapons program. In 1957, Prime Minister Nobusuke Kishi announced that, while Japan would not seek nuclear weapons, a government review had concluded that the possession of nuclear arms was not necessarily unconstitutionaU 9 Kishi's successor, Hayato Ikeda, preferring Japan to pursue economic growth over a military buildup, explored the option of acquiring nuclear weapons as a means to reduce Japan's defense budget. 2o Even after Japan ratified the NPT in 1976, the Japanese government argued in 1978 that its constitution did not prohibit the possession of nuclear weapons so long as those weapons did not exceed what was "minimally necessary" to defend Japan-a claim that Tokyo reiterated in 1982 and has upheld since thenY However, whenever Japan has studied a possible nuclear weapons option in the past, potential damage to its alliance with the United States has been a key factor that led Japan to conclude that it would not be in Japan's interest to seek an indigenous nuclear weapon program. There have been two known occasions when the Japanese government has quietly but seriously explored the option of acquiring an independent nuclear capability. Its first deliberation took place under the watch of Prime Minister Sato from the late 1960s through the mid-1970S. Sato was driven primarily by his concern about China's first nuclear test in 1964, which he raised with President Lyndon Johnson during their meeting in 1965. 22 In response, the United States reiterated its security assurance in a Joint Communique that "reaffirmed the United States' ... commitment ... to defend Japan:'23 Sato also realized, however, that NPT negotiations then underway would close the door on a nuclear option for Japan. He therefore decided that Japan should seriously explore whether to pursue a nuclear option while it still could. Sato assigned the Study Group on Democracy-a nonprofit organization closely connected with the Cabinet Intelligence and Research Office (CIRO)to conduct research entitled "Basic Study on Japan's Nuclear Policy" (Nihon no Kaku Seisaku ni kansuru Kiso-teki Kenkyu). The Study Group produced two reports: one on Japan's technological and financial capacity to produce nuclear weapons and the other on the policy implications should the Japanese government decide to acquire a nuclear capability.24 In its first report in September 1968, the Study Group concluded that while

MAINTAINING JAPAN'S NON-NUCLEAR IDENTITY

143

it would be feasible for Japan to acquire the capacity to produce a very limited number of nuclear weapons, it would be extremely difficult for Japan to actually develop its own nuclear force. The difficulties the report listed included anticipated strong opposition from the public, restrictions on reprocessing plutonium to which Japan was subject as a result of IAEA regulations, the financial and other resources that would be required, and the challenges for Japan to develop other technologies required to build nuclear forces (for example, precision guidance systems).25 The Study Group's second report, in January 1970, examined the expected challenges should Japan choose to acquire an independent nuclear capability. The report, identifying Japan's vulnerability to retaliatory attacks and potential international isolation as problems with going nuclear, concluded that becoming a nuclear weapon state would not improve Japan's national security. The report argued that while acquiring nuclear forces could heighten the national sense of pride in the short term, it would harm Japan's place in the international community in the end. 26 Under Sato's watch, the Japanese government conducted two more studies on the nuclear option. One was conducted by the Ministry of Foreign Affairs (MOFA). Their deliberations produced an internal document, Waga Kuni no Gaiko Seisaku Taiko ("Guidelines of Japan's Foreign Policy"), in 1969, which recommended that Japan should uphold its non-nuclear policy while maintaining the economic and technological capability to produce nuclear weaponsY The other study was conducted by the Japan Defense Agency (JDA) in 1970. This internal study was initiated by JDA Director-General Yasuhiro Nakasone (who later became prime minister in 1982). The JDA internal study concluded that: a. It would take about five years at maximum to develop a nuclear weapon program. It would also require an investment of 200 billion yen (approximately 40 percent of the FY 1970 defense budget); and b. A nuclear umbrella provided by the United States remained the most realistic and effective means to protect Japan. 28 Nakasone's initiation of this study had resulted from the Nixon administration's announcement of the Guam Doctrine in 1969, which sought to shift the burden of conventional defense to America's allies in Asia. In the ensuing years, America's withdrawal from Vietnam cast new doubt on the reliability of the U.S. security guarantee. The Japanese Diet also entered a heated debate about

144

YUKI TATSUMI

whether to ratify the NPT. In these circumstances, Japan sought a renewed expression of the u.s. commitment. Foreign Minister (and later prime minister) Kiichi Miyazawa traveled to Washington in mid-April 1975 to meet Secretary of State Henry Kissinger. Miyazawa obtained a public statement from the u.S. government that described U.S. nuclear weapons as a deterrent to threats against Japan and pledged U.S. action to defend Japan in the event of either conventional or nuclear attack. The meeting also led to the creation of a new joint U.S.-Japan Defense Cooperation Committee. 29 These actions reassured Japan, which reiterated its non-nuclear policy and its reliance on U.S. nuclear deterrence in its 1976 National Defense Program Outline (NDPO).30 The second known study of Japan's nuclear option was initiated by the JDA in 1994-95. The study was conducted at the time when the North Korean nuclear crisis in 1993-94 and China's nuclear test in 1995 reminded Japanese officials of the threat of nuclear weapons that Japan continued to face despite the end of Cold War. In addition, the NPT was set to expire in 1995. Japan initially expressed reluctance to support indefinite extension of the treaty, putting Tokyo at odds with other advanced industrial nations. Moreover, an intensifying friction with the United States over trade issues generated concern that such tensions with Washington could spill over into security relations and weaken U.S. defense commitments to Japan. The JDA internal study produced ''A Report Concerning the Problems of the Proliferation of Weapons of Mass Destruction:' The study concluded that Japan would not benefit by seeking to become a nuclear weapon state for the following reasons: a. A nuclear Japan is likely to cause devastating damage to the U.S.-Japan alliance, because such a decision would indicate Tokyo's distrust of the U.S. security guarantee; b. A nuclear Japan will accelerate discrediting (and even disintegration) of the global nonproliferation regime centered on the NPT; c. Japan, if it chooses to become a nuclear weapon state, must withdraw from the NPT, which is likely to cause a vastly negative impact on Japan's international reputation; d. Strong public resistance within Japan is anticipated; and e. The financial cost of building and sustaining the infrastructure for an independent nuclear capability is far greater than what the Japanese government can afford. 3!

MAINTAINING JAPAN'S NON-NUCLEAR IDENTITY

145

The report also considered a worst-case scenario in which both the u.s. security guarantee and the NPT regime collapsed. The report concluded that, even in these circumstances, going nuclear would not be in Japan's interests. Similar to the 1970 report, the study argued that such a step could trigger regional arms races and, given its population density, still leave Japan vulnerable in the event of a nuclear weapons exchange. 32 Also similar to what occurred in the 1960s and 1970S, internal debates over Japan's nuclear option in the 1990S led Japanese officials to take steps to reaffirm and strengthen the alliance with the United States. During President William Clinton's visit to Japan in April 1996, Clinton and Japanese Prime Minister Ryutaro Hashimoto issued a U.S.-Japan Joint Declaration on Security, which reiterated the U.S. commitment to defend Japan and proclaimed the continued value of the Mutual Security Treaty and U.S. military presence. Based on an agreement reached at the summit, the two countries negotiated a revision of the Guidelines for U.S.-Japan Defense Cooperation, completed in September 1997. These steps satisfied the Japanese officials that for the time being that U.S. security assurances remained solid. 33 In Chapter 2 of this volume, Jeff Knopf suggests that security assurances are more likely to be effective when a country's interest in nuclear weapons is driven largely by its security concerns. Japan's postwar policy of maintaining non-nuclear weapon status seems to be an ideal case to validate Knopf's argument. Japan's case suggests that this security argument may have limits, however. The study conducted by the JDA in 1994-95 is particularly salient to illustrate this anomaly. As discussed above, the report gave consideration to a worst-case scenario in which the U.S. security guarantee and the NPT regime-two key factors around which Japanese government has built its nonnuclear policy-both collapsed. It is significant that the JDA report concluded that Japan would be better advised not to seek to acquire nuclear weapons even under such circumstances. This suggests that the U.S. security assurance may be only one of the drivers behind Japan's steadfast maintenance of its non-nuclear weapon state status. Regional security considerations and Japan's "nuclear allergy" have independent effects, which are buttressed by the U.S. security assurance. In Chapter 3 of this volume, Janice Stein discusses the importance of giving consideration to the roles that emotions play in a country's decisions. In this context, she cautions against overconfidence in the credibility of assurances or

146

YUKI TATSUMI

threats. She also discusses how the "collective emotion" of a nation can act as either a constraint or enabler on the success of assurance. Her arguments in these regards are relevant in examining Japan's decision-making. As discussed above, the 1994-95 JDA study primarily referred to the credibility of u.S. extended nuclear deterrence as the basis for its argument against Japan pursuing an independent nuclear capability. It further pointed to the risk of a regional arms race and its implications given Japan's geostrategic vulnerabilities as the reason for advising against Japan's going nuclear, even if the u.S. security assurance and the NPT regime both collapsed. Using the terminology introduced by Stein, one can argue that it is Japan's "collective emotions" demonstrated through its confidence in U.S. extended nuclear deterrence that enable successful assurance vis-a.-vis Japan. Stein further suggests that fear leads to risk-averse behavior. Following this premise, one could argue that Japan's "fear" of potential risk-risk of an arms race in the case of the JDA report-encourages Tokyo to avoid risk by maintaining its status quo as a non-nuclear weapon state. More important, one can also argue that the "nuclear allergy" among the Japanese public is an important collective emotion that deeply affects Japan's postwar non-nuclear policy. In fact, some argue that Japan's decision to remain as a non-nuclear weapon state is a result of a complicated political process that emerged from the intimate interaction between two forces within Japan-rational policy calculation at the leadership level and emotional attachment to the memory of Hiroshima and Nagasaki among the Japanese public. 34 In other words, one can argue that Japan's current non-nuclear policy is supported by collective emotions at two levels (confidence in U.S. extended nuclear deterrence among the policy elites and antinuclear sentiment among the public), which function as a strong enabler for successful maintenance of the U.S. security assurance toward Japan. EVALUATING THE HYPOTHESES

So far, this chapter has reviewed the evolution of Japan's non-nuclear policy and the known past efforts by the Japanese government to re-examine this long-standing policy, all of which have led to the conclusion that it is not in Japan's interest to reverse its non-nuclear policy. The chapter has also discussed how the U.S. positive security assurance toward Japan, demonstrated through Washington maintaining a bilateral alliance with Tokyo and extending a nuclear deterrent umbrella, does not alone account for Japan's steadfast com-

MAINTAINING JAPAN'S NON-NUCLEAR IDENTITY

147

mitment to its non-nuclear weapon state status. However, the examination of past Japanese government efforts to revisit the decision not to pursue an independent nuclear capability demonstrates that much of Japan's behavior can be explained by some of the security assurance hypotheses introduced in Chapter 2 of this volume. Among the thirteen hypotheses presented, the following hypotheses seem particularly useful in explaining Japanese behavior:

Hypothesis 2: Public declarations of security assurances increase their effectiveness. Developments in Japan in both the Cold War and post-Cold War years support this hypothesis. As shown above, at various points in the 1960s, 1970S, and 1990S, Japanese officials sought to affirm the u.s. commitment by eliciting public statements that the u.s. government still attached great value to maintaining the U.S.-Japan alliance. More recently, when Japan felt a renewed sense of threat after Pyongyang's nuclear test in 2006, U.S. Secretary of State Condoleezza Rice made a swift public statement on Washington's commitment to provide extended nuclear deterrence to Japan, which quelled Japan's anxiety at least temporarily.35 Rice's statement was particularly important because, as described below, it came at a time when high-profile politicians in Japan began explicitly considering Japan's pursuing an independent nuclear capability. At all these points, public declarations of the continued U.S. commitment to maintain its "nuclear umbrella" over Japan allowed Japanese leaders to reconfirm Japan's non-nuclear policy vis-a.-vis the Japanese public.

Hypothesis 3: Legally binding mechanisms will increase the effectiveness of assurances. In the case of Japan, the prevailing view has been that Japan receives security assurances primarily on a bilateral basis. The U.S.-Japan Mutual Security Treaty serves as the binding mechanism suggested by this hypothesis. In particular, Article Five of the U.S.-Japan Security Treaty obligates the United States to defend Japan if there is an act of aggression against Japan. 36 In the minds of Japanese defense officials, U.S. extended deterrence is an important part of U.S. capability to defend Japan. U.S.-Japan bilateral agreements on civil nuclear cooperation have provided an additional legally binding framework to discourage Japan from pursing indigenous nuclear weapon capability. One could argue that the NPT also serves as a multilateral legal mechanism to provide security assurances. Although Japan tends to see the NPT regime more as a framework through which it promotes the normative cause of nuclear

148

YUKI TATSUMI

disarmament than as a source of security assurances, the NPT regime does provide assurance to Japan that there is a multilateral legally binding mechanism that attempts to prevent the proliferation of nuclear weapons.

Hypothesis 4: The greater the political and economic ties between sender and recipient, the more effective security assurances will be. Japan and the United States fit this proposition extremely well. The two countries have established a six-decade-long relationship that closely intertwines not only their similar security concerns in East Asia (most notably North Korea) but their global economic interests as well. Under such circumstances, it is extremely hard for Japan to obtain an independent nuclear capability without considering the impact of its decision on its relationship with Washington. This suggests that assurances are not only a way to ease a recipient's concerns about security, but can also become a source of leverage for the provider. An implicit threat to withdraw a security guarantee can become a form of coercive pressure. This leverage may be a two-way street, however. As the case history shows, when Japan detected the possibility of a weakening U.S. presence in East Asia, Japan often has successfully drawn out public statements to reaffirm the U.S. defense commitment by hinting that Tokyo may choose to go nuclear in the absence of a credible U.S. security assurance. Hypothesis 5: A reputation for keeping past nonproliferation or alliance commitments will increase the effectiveness of assurances. In case of Japan, hypothesis 5 is closely related to hypotheses 2 and 4. When Japan has worried about U.S. alliance commitments, such as during the Vietnam period, the credibility of the U.S. security guarantee has suffered, and Japan has taken steps to elicit public reaffirmations by the U.S. government. But hypothesis 5 has also worked in a reverse way. The idea that a reputation for keeping past nonproliferation or alliance commitments is important describes well Japan's own concern for its image as a close and committed U.S. ally and a strong supporter of nuclear nonproliferation. Hypothesis 6: Forward-deployed troops and other forms of defense cooperation will increase the effectiveness of security assurances. Evaluating this hypothesis is not straightforward, as U.S. military bases in Japan, and especially on Okinawa, have become a source of great controversy and domestic opposition in Japan. Yet Japan has also sought to strengthen its defense relations with the United States whenever its concern about nuclear

MAINTAINING JAPAN'S NON-NUCLEAR IDENTITY

149

threats has risen. In particular, Japanese officials have valued opportunities to participate in and shape the terms of defense cooperation with the United States. Hypothesis 6 hence appears relevant in explaining Japan's behavior.

Hypothesis 11: The impact of assurances will depend in part on how they affect the perceived prestige and appropriateness of nuclear status. To the extent that norms matter, negative assurances are likely to be more important overall than positive assurance. This hypothesis states that anything that boosts the status or prestige associated with nuclear weapons will make proliferation attractive. Japan's case shows that the hypothesis also works the other way around: in the case of Japan, its positive perception of its status as a non-nuclear state makes acquisition of a nuclear weapon capability an unattractive option. The hypothesis is not perfectly supported, however. While norms have had an important impact, they have not led Japan to give priority to negative security assurances; instead, positive assurances have been the greater concern. This hypothesis also explains the persisting perception gap between Japan and the international community over Japan's intentions. For instance, Tokyo's civilian nuclear power program, particularly its plutonium reprocessing efforts, often raises questions about Japan's commitment to its non-nuclear policy. In the minds of many Japanese, however, its history is enough evidence to prove Japan's strong commitment to renouncing nuclear weapons. For the majority of Japanese, foreign questions about the intent of Japan's civil program are so far from reality that they are regarded as a symbol of the outside world's lack of understanding of Japan.37 POST-9/11 NUCLEAR DEBATES IN JAPAN

As examined in the previous section, various security assurance hypotheses help explain Japan's decision to maintain its non-nuclear weapon state status. There is a limitation to the explanatory power of these hypotheses, however. The re-emergence of the nuclear debate in Japan since the 2000S, which is examined in this section, revolves around the credibility of the extended deterrence offered by the United States. This may suggest limits to hypotheses 2 and 3, in that public and legally binding assurances do not seem to carry as much weight as they once did. The consideration of "collective emotions" discussed by Stein in Chapter 3 is relevant in this regard, as it focuses on how the changes in "collective emotions" among the Japanese public may change the context in

150 YUKI TATSUMI

which the statements by u.s. and Japanese policy elites on u.s. security guarantees are heard and interpreted. Despite the official Japanese government position of upholding its nonnuclear policy, the debate over whether Japan should seek an independent nuclear capability has been lingering in public discourse since the 1990S. In 1999, for instance, former parliamentary Vice Defense Minister Shingo Nishimura stated during an interview that Japan, in light of the developments in East Asia, might have to study the option of possessing nuclear weapons. 38 While then Prime Minister Keizo Obuchi promptly dismissed Nishimura to demonstrate the Japanese government did not want to appear to be wavering on its nonnuclear commitments, the incident showed that some political leaders continue to argue for Japan to reconsider whether to seek an independent nuclear capability. Since the 2000S, lack of progress in the attempt to convince North Korea to give up its nuclear program and an increasing concern over China's military capability have been driving the debate over Japan's nuclear option. Pyongyang's missile test in July 2006 and nuclear test in October 2006 only encouraged this trend, leading the debate on Japan's nuclear option in political, journalistic, and academic circles to be played out publicly in an unprecedented manner. North Korea's second nuclear test in May 2009 and multiple provocations in 2010 fueled further concerns in Japan. There are three major differences in the post-9/11 nuclear debate in Japan in comparison to the Cold War and immediate post-Cold War periods. First and foremost, there is a significantly higher degree of political tolerance for those who entertain the idea of Japan's going nuclear; previously, a politician's reference to Japan's potential nuclear capability or questioning of Japan's nonnuclear policy meant an immediate dismissal of his position either in the government or in the political party. In post-9/11 nuclear debates, that began to change. For instance, in April 2002, the then leader of Japan's main opposition party, Ichiro Ozawa, stated in a speech, "With the plutonium reserve from nuclear power plants, Japan has enough to produce thousands of warheads:'39 Unlike past cases in which politicians made similar statements, Ozawa did not suffer any consequence afterward. A former chief cabinet secretary, Yasuo Fukuda, also stayed in his position when he mentioned the possibility of Japan's pursuing a nuclear option in case the international security environment deteriorates. 4o Furthermore, on October 15, 2006, within a week after North Korea's initial nuclear test, Shoichi Nakagawa, while serving as chairman of the Policy

MAINTAINING JAPAN'S NON-NUCLEAR IDENTITY

151

Affairs Research Council of the then ruling Liberal Democratic Party (LDP), suggested that Japan should seriously discuss whether the nuclear option would be in Japan's interest. 41 Despite a great deal of media criticism, he was not forced to resign his position in the LDP.42 Nakagawa repeated the suggestion in a speech in spring 2009 after a North Korean missile launch. 43 Following Nakagawa's 2006 statement, then Foreign Minister Taro Aso argued that it would be important for Japan to discuss all options, including the possession of nuclear weapons. He further rejected the notion that a nuclear debate is somehow taboo, arguing that "Japan is a free country that does not control people's opinions:' Again, despite severe criticism in the mass media suggesting these comments were "inappropriate:' Aso was neither fired nor forced to resign. 44 The second difference is the comparatively more reasonable tone of debates. Although the most recent debate over Japan's nuclear option is still spearheaded primarily by those who can be categorized as "right-wing:'45 their arguments have grown more pragmatic. For instance, many of them, unlike during the Cold War, do not support Japan's abandoning its alliance with the United States and remilitarizing to develop a full range of offensive capabilities including nuclear weapons. Rather, they propose a range of options for Japan that is far from that extreme. For instance, Masato Ushio, a conservative journalist who specializes in military affairs, suggests that Japan, while maintaining a strong alliance with the United States, gradually abandon the Three Non-Nuclear Principles. He proposes that Japan revise the principle of not allowing the introduction of nuclear weapons on Japanese land, thereby permitting the United States to deploy its nuclear weapons in Japan. 46 Kan Ito, another conservative political analyst, advocates Japan's acquiring a nuclear deterrence capability that is minimally required for Japan to defend itself, but he rejects the idea of Japan's becoming a major military power. 47 Even Shoichi Nakagawa, who has often been criticized for his extremely conservative views on security matters, says that the premise for Japan to have such a debate must be that Japan never engages in a war of aggression. 48 Third, the moderates have entered the debate. For instance, Toshiyuki Shikata, a respected retired Japan Ground Self-Defense Force officer who is known for his strong support for the U.S.-Japan alliance, advocates a robust discussion in Japan over nuclear weapons. While opposing the idea of Japan's pursuing an independent nuclear capability, Shikata believes that such a debate will allow Japan to actively engage with Washington over the alliance relationship and thereby ensure the credibility of the U.S. nuclear umbrella. 49 Satoshi Morimoto,

152

YUKI TATSUMI

a security policy expert in Japan who also strongly supports the U.S.-Japan alliance, while arguing the acquisition of nuclear capability would not serve Japan's interest, maintains that Japan needs to have a more thorough debate over its nuclear option, both inside and outside the government. 50 Another factor encouraging these nuclear debates in Japan, particularly since Pyongyang's nuclear test in 2006, is the seemingly softened u.S. attitude toward Japan's internal debates on a nuclear option. American observers used to describe the prospect of Japan's going nuclear as something that would be "most worrisome" for u.S. interests in East Asia. 51 However, as the nuclear threat from North Korea is becoming more tangible, there is a growing acknowledgment among U.S. policymakers of Japan's sense of insecurity. Some American observers have now begun to call for a U.S.-Japan dialogue on nuclear issues as a means for the United States to reassure Japan of the efficacy of Washington's nuclear umbrella while also helping the Japanese carry out a rational assessment of whether acquiring nuclear weapons will be in their interestY Furthermore, post-9/n poll results suggest a slight shift in public sentiment on the issue in Japan. To be sure, antinuclear sentiment remains widespread among the Japanese public. However, the polls taken following North Korea's nuclear tests suggest that the Japanese public (1) regards the discussion on nuclear weapons as less of a taboo, and (2) has lower confidence in the international nuclear nonproliferation regime. For instance, an opinion poll taken by the Yomiuri Shimbun in November 2006 shows that 45 percent of respondents agreed that Japan should have a more open discussion on nuclear weapons. 53 Similarly, in an opinion poll conducted by the Mainichi Shimbun in November 2006, while only 14 percent advocated Japan's acquiring nuclear capability, 61 percent considered it acceptable to discuss nuclear options. 54 Furthermore, in a January 2008 survey conducted by the MOFA on Japan's disarmament and nonproliferation policy, less than half (47.5 percent) of the respondents thought that the NPT was helpful in international peace and stability. Among those who did not think the NPT was helpful, almost 73 percent identified the lingering nuclear weapon problems with North Korea and Iran as the reason for their answer. 55 Still, Japan's non-nuclear policy continues to enjoy broad support from the general public, despite the intensifying nuclear crisis in North Korea. For instance, in November 2006, a public opinion poll conducted by the Yomiuri Shimbun indicated that approximately 80 percent of respondents thought that Japan should uphold the Three Non-Nuclear Principles. 56 A survey by the Asahi

MAINTAINING JAPAN'S NON-NUCLEAR IDENTITY

153

Shimbun in October 2006, immediately after North Korea's first nuclear test, also showed that approximately 80 percent of the respondents supported continuing to uphold the Three Non-Nuclear Principles. 57 Furthermore, the public opinion polls conducted by the MOFA in January 2008 found that 73.5 percent of respondents believed that Japan should take a leadership role in education and enlightenment in the area of nuclear disarmament and nonproliferation. Approximately 89 percent of those who thought Japan should lead such an effort identified Japan's responsibility "as the only victim of an atomic bomb" as the reason for their response. 58

In fact, following the accident at the Fukushima Dai-ichi nuclear power plant, the Japanese public's aversion to even civilian nuclear power seems to be getting stronger. In an opinion poll conducted by Asahi Shimbun in June 2011-a mere three months after the tragic Great Eastern Japan Earthquake and the resulting problems at the Fukushima Dai-ichi nuclear power plant-74 percent of the respondents supported an eventual Datsu Gen-Patsu ("Graduation from Nuclear Power Plants").59 Even in the opinion poll that Mainichi Shimbun conducted more than two months later, in late August 2011, 74 percent of the respondents still answered that Japan should try to reduce its dependence on nuclear power plants in the long run.60 It remains to be seen whether the resurfacing of "anti-nuclear power plant" sentiment among the Japanese public will have a long-term effect on its collective emotions regarding Japan's nuclear options, but at the time of this writing it seems to have renewed popular sentiment against nuclear weapons. CONCLUSION

This chapter has examined Japan's policy of remaining a non-nuclear weapon state and the role that security assurance has played in Japan's decision to maintain this policy. It has also reviewed the post-9/11 re-emergence of the debate over Japan's nuclear option. The findings have implications for the likelihood of Japan's continuing to uphold its non-nuclear policy. Most important, it should be noted that the security assurance-specifically the extended deterrence offered through the U.S.-Japan alliance-has proven effective in helping convince Japan to maintain its non-nuclear policy. As the internal government deliberations on seeking an indigenous nuclear capability in the past clearly show, the existence of u.s. extended deterrence has provided a strategic rationale for Japan to forgo the nuclear option. Yet this security guarantee has not been the only important factor. The "nuclear allergy;' the

154

YUKI TATSUMI

perception held among the vast majority of Japanese that a non-nuclear policy positively affects the country's image abroad, and government concerns that a nuclear program would provoke other regional states, also buttress Tokyo's rationale for not seeking an independent nuclear capability. Because of these other factors, there would be considerable inertia behind remaining non-nuclear regardless of what happens with the u.s. security guarantee. At the same time, however, it remains an open question what Japan might do in the event that Japan's confidence in u.s. extended deterrence were completely lost, especially if other factors also moved in a negative direction, such as Japan's concluding that acquisition of nuclear weapons would lead to a greater international reputation. It is noteworthy, therefore, that confidence in U.S. extended deterrence has become an issue for Japanese policymakers in the context of developments in two policy areas: North Korea and the Obama administration's strong support for nuclear disarmament, including nuclear reductions called for in the New START treaty with Russia. In regard to the developments on North Korea's nuclear program, there has been a growing fear within Japan about whether Washington takes Japan's security concerns seriously. This sentiment intensified when the United States, in the last years of the Bush administration, decided to de-list North Korea as a state sponsor of terrorism despite Japan's strong opposition. When Secretary of Defense Robert Gates commented in late March 2009 that the United States did not intend to shoot down a North Korean missile unless it was heading toward the U.S. homeland, it aggravated Tokyo's anxiety on U.S. defense commitments to Japan.6J In the area of nuclear disarmament, President Obama's announcement of his commitment to move toward a nuclear-free world presents a serious dilemma for Japan. On the one hand, given Japan's position as a long-time advocate for nuclear disarmament, Obama's commitment to nuclear disarmament is welcome news. On the other hand, Tokyo's reliance on Washington's "nuclear umbrella" makes Japan concerned that the United States may move toward that goal too quickly. In particular, faced with an imminent nuclear threat presented by North Korea and a long-term concern about China's nuclear weapon capability, some policymakers express concerns that if nuclear reduction proceeds too quickly, the credibility of U.S. extended deterrence may be called into question. 62 A change in Japan's political landscape may also help trigger change in Tokyo's past behavior in an attempt to reconcile its dilemma. Japan experienced

MAINTAINING JAPAN'S NON-NUCLEAR IDENTITY

155

a real regime change for the first time in fifty-five years when the Democratic Party of Japan (DPJ) won an overwhelming majority in the House of Representatives election in August 2009, and DPJ leader Yukio Hatoyama became prime minister in September 2009. Although he was forced to resign less than a year after taking office, the DPJ retained power and subsequently elected Naoto Kan and then, when Kan had to resign over his handling of the nuclear disaster, Yoshihiko Noda as the DPJ party leader and prime minister. For however long it remains in power, it is highly unlikely that the DPJ government will change Tokyo's position on its nuclear policy. If anything, the DPJ seems more committed to pursue nuclear disarmament and tightly hold on to the Three Non-Nuclear Principles. When u.S. President Barak Obama met then Prime Minister Yukio Hatoyama in Tokyo in November 2009, one of the issue-specific joint statements they released was on nuclear disarmament. 63 At the same time, however, the DPJ government differentiates itself from its predecessors (which were formed by the Liberal Democratic Party and its coalition partners) by questioning some of the most basic elements in the U.S.Japan alliance. For instance, the Hatoyama government questioned the rationale for keeping a considerable U.S. military presence in Okinawa, and for this reason it initially postponed the decision on whether to uphold the existing bilateral agreement on the relocation of Marine Corps Air Station Futenma on Okinawa. While Hatoyama ultimately decided to proceed with the original relocation, this broke his campaign promise and elicited considerable criticism that led to his resignation. In addition, the MOFA, under the direction of then Foreign Minister Katsuya Okada, also conducted an internal investigation of the so-called secret agreement between Japan and the United States on the introduction of U.S. nuclear weapons to Okinawa during the Cold War, questioning why such an important matter of Japanese foreign policy was never discussed publicly.64 While it is highly unlikely that any DPJ government will decide that Japan should pursue an independent capability, the Hatoyama administration established a precedent in questioning the rationale for maintaining the status quo, which may continue and possibly lead, for instance, to another assessment of Japan's nuclear option in the future. As the only country to have had the atomic bomb used against it, "to go, or not to go nuclear" is a serious question for Japan-one that can fundamentally impact its postwar national identity. However, recent developments in the security environment have prompted Japan to discuss the nuclear question in

156 YUKI TATSUMI

an unprecedented manner. As long as one major factor-Japan's confidence in the credibility of u.s. extended deterrence-remains undecided, observers of Japan's nuclear debate are advised to pay even closer attention to the debate within Japan. For the United States, this also means that the security assurance has become an issue that requires greater attention in its alliance management vis-it-vis Japan. The challenge for Washington lies in ensuring that U.S. extended deterrence remains credible in the eyes of the Japanese. Japan needs to be convinced that the United States has both the will and capability to defend Japan if military attacks are attempted against Japan. In addition, Japan needs to be confident that sufficient consultation will take place between the U.S. and Japanese governments on the nuclear issue. 65 In this context, the 2009 decision between the two governments to hold a regular bilateral consultation on nuclear issues is an encouraging development. 66 As of this writing, extended deterrence has always been a part of the discussion whenever officials from both sides have met for consultation since then. 67 When foreign and defense ministers of the two countries met for the Security Consultative Committee on June 21, 2011, a wide range of nuclear policy-related issues-nuclear deterrence, the safety of the civil nuclear program, North Korea's and Iran's nuclear programs, and efforts in nuclear disarmament-were identified as the key issues for further alliance discussion. 68 Such a consultation could contribute a great deal to assuring the Japanese government of the credibility of U.S. extended deterrence, the core of the U.S. security assurance to Japan. Alongside the impact of antinuclear sentiment among the Japanese public, the U.S. security assurance has served as an effective means to maintain Japan's non-nuclear policy. In the post-9i11 world in which the international nonproliferation regime faces many challenges, how to maintain a robust security assurance will be one of the major alliance management issues for policymakers in both Tokyo and Washington. Trying to do this without sacrificing other policy agendas currently shared by the two capitals, such as those around nuclear disarmament, will require especially careful policy coordination. NOTES The author would like to thank Jenny Shin, who, as a summer intern, provided valuable assistance in finalizing this chapter. 1.

Prime Minister Eisaku Sato's statement in response to a question by Representative

MAINTAINING JAPAN'S NON-NUCLEAR IDENTITY

157

Tomomi Narita, Budget Committee, House of Representatives, 57th Ordinary Session, December 11, 1967, http://kokkai.ndl.go.jp/SENTAKU/syugiin/057/0514/05712110514002a. html (accessed May 1,2009). In practice, Japan accepted what might be seen as an exception to the third principle. It permitted u.S. Navy ships to make port calls without asking the United States to state whether those ships carried nuclear weapons. 2. Prime Minister Eisaku Sato, in response to a question by Saburo Eda, 58th Ordinary Session, January 30, 1968, http://kokkaLndl.go.jp/SENTAKU/syugiin/058/000l/ main.html (accessed August 19, 2011). 3. Ministry of Foreign Affairs of Japan (MOFA), Heisei 2o-nen Nihon no Gun-shuku Fu-kakusan Gaiko ("Japan's Disarmament and Nonproliferation Diplomacy 2008"), 3033, http://www.mofa.go.jp/mofaj/gaiko/ gun_hakusho/ 2008/pdfs/holll.3. pdf (accessed June 10, 2009); Shinichi Ogawa and Michael Schiffer, "Japan's Plutonium Reprocessing Dilemma;' Arms Control Today 35, no. 8 (October 2005), 21; International Commission on Nuclear Non-Proliferation and Disarmament, http://www.icnnd.org (accessed May 5,2010). 4. T. V. Paul, Power versus Prudence: Why Nations Forgo Nuclear Weapons (Montreal: McGill-Queen's University Press, 2000), 49-50; Maria Rost Rublee, Nonproliferation Norms: Why States Choose Nuclear Restraint (Athens: University of Georgia Press, 2009), 66-68. 5. MOFA, Heisei 2o-nen Nihon no Gunshuku Fukakusan Gaiko (Tokyo: Ministry of Foreign Affairs of Japan, 2008),30-89. 6. Since then, the agreement was revised and replaced by subsequent U.S.-Japan Agreements for Cooperation Concerning Peaceful Uses of Nuclear Energy. See Heisei 2-nen ban Genshiryoku Hakusho (''Atomic Energy White Paper 1990"), http://www.aec. go.jp/jicst/NC/about/hakusho/WP1990/sb2090102.htm (accessed June 2, 2009). 7. The agreement was replaced by a new Agreement for Cooperation between Japan and the United States Concerning the Peaceful Uses of Nuclear Energy in 1988. See http://www.anzenkakuho.mext.go.jp/lawlist/pdf/s630702_05.pdf (accessed August 20, 2008). 8. Genshi-ryoku Kihon-ho (''Atomic Energy Basic Law"), Article Two, http://law.egov.go.jp/htmldata/S30/S30H0186.html (accessed February 3,2008). 9. U.S. Energy Information Administration, "World Net Nuclear Electric Power Generation, 1980-2007;' http://www.eia.doe.gov/fuelnuclear.html (accessed May 5, 2010). 10. Emma Chanlett-Avery and Mary Beth Nikitin, Japan's Nuclear Future: Policy Debate, Prospects, and U.S. Interests, CRS Report for Congress, RL 34487 (February 19, 2009),3. 11. Atomic Energy Commission. Heisei 2o-nen Genshiryoku Hakusho ("White Paper on Atomic Energy 2008"), 10, http://www.aec.go.jp/jicst/NC/about/hakusho/ hakush0200811.pdf (accessed May 20,2009).

158 YUKI TATSUMI

12. Ogawa and Schiffer, "Japan's Plutonium Reprocessing Dilemma." 13. Mike M. Mochizuki, "Japan Tests the Nuclear Taboo;' Nonproliferation Review 14, no. 2 (July 2007): 306. 14. See, for instance, MOFA, Heisei 20 nen-do Nihon no Gun-shuku Fu-kakusan Gaiko ("Japan's Disarmament and Nonproliferation Diplomacy 2008"), 30, http://www. mofa.go.jp/mofaj/gaiko/gun_hakusho/2008/pdfs/ho1l13.pdf (accessed June 10, 2009). 15. MOFA, "Statement by Mr. Hirofumi Nakasone: 'Conditions toward Zero-11 Benchmarks for Global Nuclear Disarmament;" April 27, 2009, http://www.mofa.go.jp/ policy/un/ disarmament/ arms/ state0904.html (accessed August 1, 2011). 16. Christopher W. Hughes, Japan's Security Agenda: Military, Economic & Environ-

mental Dimensions (Boulder, CO: Lynne Rienner Publishers, 2004), 122-23. 17. Policy address by Prime Minister Eisaku Sato, 58th Ordinary Session, January 27, 1968, http://kokkaLndl.go.jp/SENTAKU/syugiin/058/0001/main.html (accessed May 1,2009)· 18. Anthony DiFilippo, Japan's Nuclear Disarmament Policy and the U.S. Security

Umbrella (New York: Palgrave Macmillan, 2006), 35. 19. Kurt M. Campbell and Tsuyoshi Sunohara, "Japan: Thinking the Unthinkable;' in The Nuclear Tipping Point: Why States Reconsider Their Nuclear Choices, ed. Kurt M. Campbell, Robert J. Einhorn, and Mitchell B. Reiss (Washington, DC: Brookings Institution Press, 2004), 221. 20. Michael J. Green and Katsuhisa Furukawa, "Japan: New Nuclear Realism;' in The Long Shadow: Nuclear Weapons and Security in 21st Century Asia, ed. Muthiah Alagappa (Stanford: Stanford University Press, 2008), 349. 21. Statement by Hideo Sanada, Director of Cabinet Legislative Bureau, at the Budget Committee of the House of Councillors, March 11, 1978; Statement by Reijiro Tsunoda, Director of Cabinet Legislative Bureau, at the Budget Committee of the House of Councillors, April 5, 1982. Extracted from Heisei 19-nendo Ban Bouei Handbook ("Handbook for Defense 200l") (Tokyo: Asagumo Shimbun-sha, 2007), 603-5. 22. Kenneth B. Pyle, Japan Rising: The Resurgence of Japanese Power and Purpose (New York: Public Affairs, 2007), 251. 23. Quoted in Rublee, Nonproliferation Norms, 61. 24. The two Study Group reports were never released in their entirety to the public. Summaries of both, however, were reported by Asahi Shimbun on November 13, 1994. For quotations from the summary of the two Study Group Reports reported in Asahi

Shimbun, see http://homepage.mac.com/ehara~en/jealous~ay/nuclear_armament. html (accessed December 8, 2008). For a later analysis that contains additional information, see Yuri Kase, "The Costs and Benefits of Japan's Nuclearization: An Insight into the 1968/70 Internal Report;' Nonproliferation Review 8, no. 2 (Summer 2001). 25. Asahi Shimbun, November 13, 1994, http://homepage.mac.com/ehara~en/jeal­ ous_gay/nucleacarmament.html (accessed December 8, 2008).

MAINTAINING JAPAN'S NON-NUCLEAR IDENTITY

159

26. Ibid. 27. Green and Furukawa, "Japan: New Nuclear Realism;' 351. 28. Ibid. 29. John E. Endicott, "The 1975-76 Debate over Ratification of the NPT in Japan;'

Asian Survey 17, no. 3 (March 1977): 281-82; Rublee, Nonproliferation Norms, 67. 30. Green and Furukawa, "Japan: New Nuclear Realism;' 352. 31. Llewelyn Hughes, "Why Japan Will Not Go Nuclear (Yet);' International Security 31, no. 4 (Spring 2007): 76-78. 32. Ibid., 79; Campbell and Sunohara, "Japan: Thinking the Unthinkable;' 227-28. 33. Green and Furukawa, "Japan: New Nuclear Realism;' 353; MOFA, Joint Statement: U.S.-Japan Security Consultative Committee, http://www.mofa.go.jplregion/namerica/us/ security/defense.html (accessed May 5, 2010 ). 34. See, for example, Nobumasa Akiyama. "The Socio-Political Roots of Japan's Non-Nuclear Posture" in Japan's Nuclear Option: Security, Politics and Policy in the 21st Century, ed. Benjamin L. Self and Jeffrey W. Thompson (Washington, DC: Henry L Stimson Center, 2003), 64-91. 35. Yomiuri Shim bun, October 18, 2006, http://www.yomiuri.co.jp/feature/fe7000/ news/20061018i201.htm (accessed December 21, 2009). 36. The Treaty of Mutual Cooperation and Security between Japan and the

United States of America, January 19, 1960, Article Five, http://www.ioc.u-tokyo.ac.jp/ -worldjpn/documentsltexts/docs!19600119.T1E.html (accessed December 20, 2009). 37. Katsuhisa Furukawa, "Nuclear Option, Arms Control and Extended Deterrence: In Search of a New Framework for Japan's Nuclear Policy;' in Self and Thompson, Ja-

pan's Nuclear Option, 95. 38. Andrew L. Oros, "Godzilla's Return: The New Nuclear Politics in Insecure Japan;' in Self and Thompson, Japan's Nuclear Option, 50-51. 39. Kyodo News, April 6, 2002. 40. Yomiuri Shimbun, June 1 and June 3, 2002. 41. Shoichi Nakagawa, Jiritsu shita Kokka no Kaku-busou Rongi ("Nuclear Debate of an Independent Nation"), Voice (December 2006): 46-51. 42. For instance, Nihon Keizai Shimbun (October 16, 2006) criticized his statement as "unwise:'

43. Kyodo News, "Nakagawa Floats Sobering Option: Going Nuclear;' April 20, 2009, http://search.japantimes.co.jp/cgi-bin/nll20090420a3.html (accessed September 28, 2011). 44. Asahi Shimbun, October 20, 2006. 45. Sometimes referred to as "neo-autonomists" or "nationalists", they argue that Japan needs a more independent security policy, implicitly suggesting that Japan keep an open mind about acquiring nuclear weapons. See, for example, Rust M. Deming, "Japan's Constitution and Defense Policy: Entering a New Era?" Strategic Forum, no. 213

160

YUKI TATSUMI

(Institute for National Strategic Studies, National Defense University, November 2004), http://www.ndu.edu/inss/strforum/SF213/SF213_Final.pdf (accessed December 8, 2008); and Richard J. Samuels, "Securing Japan: The Current Discourse;' Journal of Japanese

Studies 33, no. 1 (Winter 2007). 46. Masato Ushio, San Gensoku wa Dankai-teki Haiki-wo ("Three Principles to Be Relinquished Gradually"), Will (January 2007): 77-79. 47. Kan Ito, Jikoku no Bouei ni Sekinin wo Moteru Atarimae no Kuni-ni ("Toward a Normal Country That Can Be Responsible for Its Own Security"), in Nihon Kakubusou no Ron-ten ("Points of Debate in Japan's Nuclear Option") (Tokyo: PHP Kenkyu-sho, 2006), 189-251. 48. Nakagawa, "Nuclear Debate of an Independent Nation;' 48. 49. See, for instance, Toshiyuki Shikata, Gunjiteki Tsu-ru kara Seijiteki Tsu-ru he ("From a Military to a Political Tool"), Will (January 2007): 38-45. 50. Satoshi Morimoto, Nihon Bouei Saiko-ron ("Japan's Defense Revisited") (Tokyo: Kairyu-sha, 2008), 266-78. 51. See, for example, Ted Carpenter, "It's Time to Revise U.S.-Japan Security Relationship;' USA Today Magazine 125, no. 2620 (January 1997). 52. See, for instance, Brad Glosserman, "Nuclear Basic for the Alliance;' PacNet

Newsletter, no. 21 (April 7, 2007). 53. Yomiuri Shimbun, November Opinion Polls, conducted November 11-12, 2006, http://www.mansfieldfdn.org/pollsI2006/poll-06-18.htm (accessed July 2, 2009). 54. Kaku Hoyuu 'No,' Giron wa Younin (" 'No' to the Possession of Nuclear Weapons, but OK to Discuss"), Mainichi Shimbun, November 27, 2006. 55. MOFA, Nihon no Gunshuku Fukakusan Gaikou ni kansuru Ishiki Chosa ("Opinion Polls on Japan's Disarmament and Nonproliferation Diplomacy"), conducted January 11-27, 2008, http://www.mofa.go.jp/mofaj/gaiko/chosa/i_chosa.html (accessed June 15,2009)·

56. Yomiuri Shimbun, November Opinion Polls, conducted November 11-12, 2006, http://www.mansfieldfdn.org/pollsI2006/poll-06-18.htm (accessed July 2, 2009).

57. Kita no Kaku-jikken de Towareru Senshu Bouei: Tekichi Kougeki, Kaku Busou-ron Fujou ("Exclusively Defense-Oriented Posture Questioned in the Wake of North's Nuclear Test: Enemy Strike Capability and Nuclear Option Resurfaces"), Bloomberg News, May 29, 2009, http://www.bloomberg.com/apps/news?pid=71000003&sid=aolBZVXCa qcw (accessed July 6, 2009). 58. MOFA, Nihon no Gunshuku Fukakusan Gaikou ni kansuru Ishiki Chosa ("Opinion Polls on Japan's Disarmament and Nonproliferation Diplomacy"), conducted January 11-27, 2008, http://www.mofa.go.jp/mofaj/gaiko/chosa/i_chosa.html (accessed June 15,2009)·

59. Shorai-teki ni 'Datsu Gen-Patsu' Sansei 74%: Asahi Shimbun Yoron Chosa ("74% Supportive of Eventual 'Graduation from Nuclear Power Plants': Asahi Shimbun Pub-

MAINTAINING JAPAN'S NON-NUCLEAR IDENTITY

161

lic Opinion Poll"), Asahi Shimbun, June 13, 2011,http://www.asahLcom/national/update/ 0613/TKY201106130401.html. 60. Mainichi Yoron Chosa: Gen-Patsu 'Jikan kakete sakugen subekida' 74% ("Mainichi

Public Opinion Poll: 74% Think Nuclear Power Plants 'Need to Be Reduced over a Long Run"'), Mainichi Shimbun, August 21, 2011, http://mainichLjp/select/seiji/news/ 20110822koooomo10088000c.html. 61. See, for example, Yoshihisa Komori, Kita Misairu Hassha Nichi-bei Doumei no Aratana 'Shinjitsu' ("North's Missile Launch: A New 'Truth' about the U.S.-Japan Al-

liance"), Sankei Shimbun, April 6, 2009, http://www.iza.ne.jp/news/newsarticle/world/ americal239703/ (accessed June 1 2009). 62. Keiko Iizuka, Kitachosen no Bousou: Nichi-bei Doumei Kyoka no tame Imakoso Kaku no Kasa no Giron wo ("Lashing Out by North Korea: Discussion of the Nuclear

Umbrella Needed Now More than Ever for the Sake of Strengthening the U.S.-Japan Alliance"), WEDGE (July 2009): 10-13; Yukiko Satoh, "Reinforcing American Extended Deterrence for Japan: An Essential Step for Nuclear Disarmament;' AJISS-Commentary, no. 57, February 3,2009. 63. The United States-Japan Joint Statement toward a World without Nuclear Weapons, November 13, 2009, http://www.whitehouse.gov/the-press-office/united-states-

japan-joint-statement-toward-a-world-without-nuclear-weapons (accessed January 10, 2010). 64. Japan Times Online, "Secret Pacts Existed; Denials 'Dishonest;" March 10, 2010,

http://search.japantimes.co.jp/cgi-bin/nll2010031Oa1.html (accessed May 5,2010). 65. Iizuka, Kitachosen no Bousou, 13. 66. Nichi-bei 'Kaku no Kasa' Koushiki Kyogi de Icchi ("U.S. and Japan Agreed on an

Official Nuclear Consultation"), Sankei Shimbun, July 18, 2009, http://sankeLjp.msn. com/politics/poliCY/090718/plc0907181745003-m.htm (accessed July 19, 2009). 67. "U.S.-Japan Begin Discussions on Security Issues Including Nuclear Umbrella;' Japan Today, March 18, 2010, http://www.japantoday.com/category/politics/view/us-

japan-begin -debate-on -security-issues-including-nuclear-umbrella (accessed May 26, 2010); interview with a Ministry of Defense official, Tokyo, March 19, 2010. 68. Joint Statement of the Security Consultative Committee: Toward a Deeper and Broader U.S.-Japan Alliance: Building on 50 Years of Partnership, June 21, 2011, http://

www.mod.go.jp/j/approach/anpo/201106_2pluS2/js1_e.html.

Infusing Commitment with Credibility: The Role of Security Assurances in Cementing the U.S.-ROK Alliance Scott Snyder and Joyce Lee

The Republic of Korea (ROK) explored a nuclear weapons option in the 1970S, and since the 2000S there have been new public calls in South Korea for it to develop nuclear arms in response to the advance of North Korea's nuclear weapons program. A positive security assurance provided through its alliance with the United States has so far helped keep South Korea on the path of nuclear restraint, but the effectiveness of American assurances has been far from automatic. That the U.S.-ROK alliance has endured since the signing of the U.S.-ROK Mutual Defense Treaty in October 1953 and its entry into force in November 1954 is an indication of its extraordinary success. But under the surface, the relationship has always contained elements of tension and distrust. These frictions have required persistent efforts by both parties, but especially the United States, to ensure that each side would continue to feel it could rely on the alliance partnership. Despite other changes over time, the threat from the Democratic People's Republic of Korea (DPRK) has persisted as South Korea's primary security concern. For the United States, the Mutual Defense Treaty was occasioned by mixed motives: a need to provide reassurance to South Korea regarding the ongoing U.S. commitment following the Korean War, as well as to provide a mechanism for both deterrence against North Korea and restraint against actions by South Korea that might lead to renewed conflict.! The treaty has succeeded in deterring North Korean aggression and ensuring peace and security in the Korean Peninsula; however, the treaty by itself has periodically been insufficient to assure South Korea of Washington's commitment to its defense. Despite American assurances, South Korea has persistently harbored doubts that U.S. political 162

INFUSING COMMITMENT WITH CREDIBILITY

163

commitments would prove to be sufficient and has worried that America's legal commitments under the treaty might be subject to loopholes in its application. 2 This chapter will explore the role of security assurances in four phases of alliance interaction: (1) the rising distrust felt by the Park Chung-hee administration regarding u.s. alliance commitments in the context of the Nixon Doctrine and the u.s. opening to China in the early 1970S; (2) the restoration of confidence in the alliance during the administrations of Chun Doo-hwan and Ronald Reagan in the early 1980s; (3) the challenge to the alliance posed by the apparent divergence in U.S.-ROK security priorities in the 1990S and early 2000S, first in the context of the U.S.-DPRK Agreed Framework and subsequently in the context of the Sunshine Policy of reconciliation with North Korea under the Kim Dae-jung and Roh Moo-hyun administrations; and (4) the "restoration" of alliance ties under the Lee Myung-bak administration. The chapter will assess the conditions under which assurances have been effective, and will consider the implications of these findings for the hypotheses presented in Chapter 2. This chapter will conclude by considering how changes in the alliance might shape the nature and impact of assurances going forward. CREDIBILITY OF U.S. SECURITY ASSURANCES UNDER SIEGE: THE PARK CHUNG-HEE PERIOD

One of the most dramatic periods of tension in the alliance was the nearcollapse during the 1970S of the credibility of u.s. security commitments in the eyes of Park Chung-hee, South Korea's third president, who took power through a military coup in the early 1960s. The deployment of u.S. tactical weapons starting in 1958 had served as a symbol of U.S. commitment to South Korea's defense. 3 The U.S. policy of neither confirming nor denying nuclear weapons deployments, however, created ambiguities that may have left doubts in South Korea regarding American intentions even despite these deployments. 4 In the face of what was still in the 1970S a militarily more powerful North Korea, changes in u.S. foreign policy starting in the late 1960s fueled Park's anxieties regarding South Korea's vulnerability. In response, Park pursued two tracks. On the one hand, although South Korea had signed the NPT as a nonnuclear state in 1968, Park initiated a covert nuclear weapons program. On the other hand, Park doggedly demanded continued assurances of u.S. defense commitments. Throughout the 1970S, South Korea found the U.S. policy of extended deter-

164

SNYDER AND LEE

rence to be inconsistent and ambiguous. 5 Three major developments in u.s. policy in Asia especially caught Koreans by surprise. First, the Nixon Doctrine and other policy initiatives begun by President Richard Nixon contributed to growing doubts among government officials about the credibility of the u.s. security commitment. Announced on Guam in 1969, the Nixon Doctrine called on u.s. allies in Asia to take on a greater share of the burden for their own defense. President Nixon's subsequent decisions to pursue Sino-American rapprochement and u.S.-Soviet detente further shocked South Korean leaders. They worried that the United States might abandon South Korea to an aggressively hostile North Korea in order to expedite Sino-American normalization and the u.S.-Soviet detente process. 6 Both the Nixon and Carter administrations appeared to pursue a policy of disengagement from South Korea as a matter of expediency-in order to remove any possibility of a clash with either the Soviet Union or the People's Republic of China (PRC) in Asia. South Koreans noted with concern that U.S. direct military assistance to South Korea declined sharply after 1969 and ceased completely in 1978. 7 U.S. withdrawal from South Vietnam and the ignominious end to the Vietnam War added to South Korean security concerns. As a demonstration of the limits of U.S. power, America's failure in the Vietnam War increased South Korean fears that a U.S. security guarantee would not be realized and that it would need to prepare for abandonment by the United States. 8 Second, U.S. efforts to reduce the number of troops in South Korea also strained the alliance. The first step came with President Nixon's withdrawal of the U.S. 7th Army Division of roughly 20,000 troops from Korea and relocation of the 2nd Division away from the demilitarized zone (DMZ) in 1971, moves that caused great concern to Park Chung-hee. 9 This adjustment meant that, for the first time since the end of the Korean War, South Korean forces were the primary line of defense for South Korea against the North. Park believed that the South Korean Army was simply incapable of defending the country by itself,1O and Nixon's move exacerbated his fears that the ROK Army might have to fight alone against a North Korean attack. In response to the withdrawal plan, Park asserted that the presence of U.S. troops was "absolutely necessary" until South Korea developed its own capabilities to defend itself from North Korean aggression. l1 A more extensive U.S. force withdrawal plan was proposed by President Jimmy Carter in 1977 along with plans to reduce and eventually remove U.S. nuclear weapons in South Korea. Carter linked the withdrawal plan in part to stepped-up criticism of South Korea's human rights record. The anticipated

INFUSING COMMITMENT WITH CREDIBILITY

165

removal of u.s. tactical nuclear weapons from South Korea was interpreted as a clear erosion of u.s. promises regarding the defense of South Korea. Even greater than the actual impact on military capabilities was the psychological and political impact, given South Korean views of North Korea's superior military strength. In a 1977 press briefing, the press secretary of the Park government said, "Since the U.S. already had expressed an official intention to withdraw the U.S. forces from South Korea, we will try to cooperate with the u.S. government. But, it should be noted that South Korea is in strong need of tactical nuclear weapons for achieving deterrence by retaliation:'12 This statement illustrated the direct connection between the erosion of the credibility of U.S. commitments to defend South Korea against North Korean aggression and South Korean pursuit of nuclear weapons. Third, as a result of a shift in U.S. nuclear strategy in the 1960s from "massive retaliation" to "flexible response;' the strategic gap in U.S.-ROK perceptions widened further. The ROK viewed the deployment of U.S. tactical nuclear weapons in the South, including atomic landmines along the DMZ, as a tripwire to get the United States to launch an all-out attack against North Korea as soon as any invasion by the North began. In contrast, the United States had adopted the strategy of flexible response in part to delay the need for immediate nuclear retaliation in the case of an attack by an American adversary. To the Park Chung-hee government, the military strategy of immediate nuclear retaliation was a vital factor for its national security, especially given South Korea's continuing sense of vulnerability to the North Korean military. Despite assurances by U.S. government officials that immediate nuclear retaliation would be part of the U.S. response, South Korean authorities were doubtful that the United States would keep such a strategy when it did not match its military doctrine in Europe. In the minds of South Korean policymakers, it was necessary to develop an alternative means to defend the country against a war with North Korea in case of U.S. abandonment.13 The fear of abandonment was the main factor behind South Korea's alarmed reaction to U.S. force withdrawal plans during the Park period. 14 This fear cata1yzed Park Chung-hee's pursuit of an independent nuclear deterrent. Park even said in an interview in June 1975 that "if the United States withdraws the nuclear umbrella, South Korea will develop indigenous nuclear weapons for the purpose of national security:'15 Park could make such a claim because he had established a secret government committee to explore the nuclear option in 1970. A production plan focused on plutonium was completed in late 1973, and

166

SNYDER AND LEE

soon thereafter South Korea entered into negotiations to purchase a spent-fuel reprocessing facility from France and a heavy water reactor from Canada. 16 The quest for security was the primary motivation for South Korea's choice to pursue development of nuclear weapons. Until the 1980s, North Korea's military capability outweighed that of South Korea both qualitatively and quantitatively. The military balance favored South Korea only if one included U.S. troopsY The Park regime perceived a deteriorating external security environment that was made more dangerous by the perceived inconsistency of U.S. commitments to South Korea. Park Chung-hee's nuclear decision was directly linked to his concerns about the regional security environment and his doubts about America's commitment to defend South Korea. 18 When the United States determined that South Korea had initiated a covert nuclear weapons program, top officials in the Gerald Ford administration, including Secretary of State Henry Kissinger, threatened to cut off U.S.-ROK security ties and halt approximately $500 million in loans and loan guarantees for South Korea's ambitious civil nuclear energy program. 19 To South Korea, the presence of U.S. forces in Korea was the most compelling feature of U.S. security assurances. In response to the Park government's pledge to terminate the nuclear weapon effort, the United States promised to maintain deployments of U.S. Forces Korea (USFK) for as long as the security environment requiredunder the condition that South Korea continue to forgo nuclear weapons. One tangible result of U.S. pressures in this period is that South Korea ratified the NPT in 1975, seven years after signing it, although it actually continued work on the nuclear weapons program. A similar cycle followed President Carter's announced plan to withdraw U.S. troops, except this time South Korea used the threat of nuclearization as a stick to secure U.S. assurances, openly hinting that it might develop nuclear weapons in a bid to get Carter to reverse the withdrawal decision. 20 The United States again applied economic pressure, threatening to cancel loans from the ExportImport BankY Under pressure from South Korea as well as the U.S. military and Congress, and faced with new intelligence that upgraded the assessment of the North Korean military threat, Carter set aside his plans for further troop reductions. The United States also agreed in 1978 to establish a Combined Forces Command (CFC) that would integrate control of the two allies' military forces in South Korea. Finally, Carter offered to restore U.S. economic and military assistance, including increased sales of U.S. jet fighters, as long as South Korea terminated the nuclear weapons program. 22

INFUSING COMMITMENT WITH CREDIBILITY

167

The main factor that prevented South Korea from going nuclear in the Park years was a combination of u.S. coercion and reassurance, including the threat to remove U.S. forces completely from South Korea and the promise that U.S. forces would remain in Korea if the nuclear decision was reversed. These carrots and sticks had the effect of slowing rather than ending the South Korean effort. Indeed, the biggest incentive for Park's decision to keep working on a nuclear option was his concern regarding U.S. force withdrawals. Likewise, Park Chung-hee's threats to pursue nuclear weapons constituted a form of coercion through which he was able to extract continuing American commitments to South Korea's defense, including pledges to use nuclear weapons in the event that South Korea might be threatened militarily. Although Park announced in 1977 that South Korea would not develop nuclear weapons, he continued a clandestine program until his assassination in October 1979 by the head of Korea's intelligence agency.23 According to former government officials under President Park, South Korea's first atomic bomb was to be completed by early 1981. 24 As described in the next section, Chun Doo-hwan finally dismantled the nuclear weapon program when he assumed power following a military coup in the aftermath of the Park assassination. 25 CREDIBILITY OF U.S. ASSURANCES RESTORED: REVITALIZED ALLIANCE COOPERATION UNDER CHUN AND ROH

The Park Chung-hee government's attempt to develop nuclear weapons shook the alliance to its core, but in the end resulted in enhanced security assurances from the United States. The security environment surrounding South Korea improved noticeably during the Chun Doo-hwan and Roh Tae-woo periods as South Korea made great strides relative to North Korea economically and militarily while the North's economy sputtered. But most important, credible steps to renew U.S. commitments restored South Korean confidence in the U.S. security guarantee. As President Ronald Reagan took office in 1981, he downplayed the public emphasis on human rights concerns of his predecessor,26 and undertook efforts to make the U.S. security guarantee for South Korea more credible. Reagan promised no further withdrawals of American forces in Korea, nullified Carter's decision on tactical nuclear weapons, which would have withdrawn remaining U.S. nuclear weapons by 1982, and provided South Korea with a large number of advanced weapon systemsY Two summit meetings between presidents Chun and Reagan in 1981 and 1983

168

SNYDER AND LEE

marked a significant turning point for South Korea's thinking about its nuclear weapons pursuits. The joint communique issued at the end of the Reagan visit in 1983 described, for the first time, the security of South Korea as "pivotal" to the stability of Northeast Asia and as "vital" to the security of the United States.28 The Reagan-Chun communique signed during the 1981 summit also stressed that South Korea would have to be a "full participant" in any U.S. discussions with North Korea for reunification of the country, addressing South Korean fear of being left out of its own unification equation. 29 Joint security cooperation increased, and the United States agreed to give the South more advanced military technology.30 A 1982 joint communique from a meeting of defense ministers also explicitly "confirmed ... the United States' nuclear umbrella."31 At the same time, the Reagan administration made it clear that the improved security relationship was conditional on South Korea's fully ending its pursuit of nuclear weapons. With the credibility of U.S. alliance assurances now restored and not wanting to risk the relationship with the United States, President Chun is believed to have canceled the nuclear program altogether. 32 Another major reassurance mechanism established during this period was the Team Spirit Exercise, initiated in 1976. Through these exercises, the South Korean military was integrated into U.S. nuclear strategy. By 1983, Team Spirit became the largest U.S. field training exercise with any ally.33 The 1982 exercise included an overflight of a U.S. B-52, which the Washington Post described as a "symbol of the U.S. nuclear punch that could top off any conventional defense of the South:'34 Apart from the strengthened U.S. assurances, the international political environment also favored the security of South Korea in the late 1980s as China and the Soviet Union abandoned their hostility toward South Korea. 35 South Korea gained confidence both in the U.S. security commitments and in its own military capability. By the late 1980s, South Korean military capabilities surpassed those of North Korea. South Korea also went through an important domestic transition to democracy in this period. In December 1987, Roh Tae-woo, a close associate of Chun Doo-hwan, won the country's first democratic election, keeping power in the hands of the ruling conservative party. In July 1988, Roh announced his desire to improve relations with North Korea, leading to the start of talks between the two Koreas the following year. South Korea appeared satisfied with U.S. security commitments and the credibility of the U.S. nuclear umbrella in the post-Park Chung-hee period, even as the Chun and Roh governments seemed to put lower priority on U.S.

INFUSING COMMITMENT WITH CREDIBILITY

169

forces as a means of national security as South Korea's own military power grew stronger. 36 A retired Korean general who served as a staff member for the CFC recalled, ''As long as the Reagan administration continued to reassure the u.S. security guarantee in various ways, there was no reason for the Chun and Roh governments to seek indigenous nuclear weapons at the expense of the firm U.S.-ROK security alliance."37 Withdrawal of U.S. Nuclear Weapons and the Denuclearization Agreement

In October 1991, the United States suddenly announced its decision to remove all its nuclear weapons from South Korea, including land- and sea-based tactical weapons and air- or bomber-delivered weapons. The move resulted from a decision by President George H. W. Bush, announced in late September, to initiate a worldwide drawdown of U.S. tactical nuclear weapons. 38 Seoul's first reaction was to reject the idea, calling it "unrealistic;' but President Roh Tae-woo eventually accepted the withdrawal, hoping that this decision would pave the way for North Korea to accept international inspections on its nuclear facilities. In a declaration issued on November 8, Roh stated that South Korea would not build, possess, store, deploy, or use nuclear weapons or seek enrichment or reprocessing capability, and he called on North Korea to match his declaration. These moves had the desired effect, and, on November 26, Pyongyang announced that it would allow inspection of its declared nuclear installations if Washington allowed inspections of its bases to guarantee that all U.S. nuclear weapons were withdrawn from South Korea. 39 As of mid-December 1991, the last U.S. nuclear weapons had been removed from South Korea, and Roh declared publicly that no U.S. nuclear weapons were deployed or stored in South Korea, calling on North Korea to immediately abandon its nuclear development program. 40 On the last day of the month, South and North Korea agreed to a Joint Declaration on Denuclearization of the Korean Peninsula, which committed the two sides to the terms of Roh's unilateral declaration in November (as well as not to test nuclear weapons). Following the withdrawal of U.S. nuclear forces from South Korea, North Korea signed a long-delayed nuclear safeguards agreement on January 30,1992, with the International Atomic Energy Agency (IAEA)Y When the United States decided to remove its tactical nuclear weapons, it took pains to reassure South Korea that the nuclear umbrella would remain intact. The absence of nuclear weapons on the ground in Korea would be made

170

SNYDER AND LEE

up for by long-range nuclear weapons deployed in the United States and on board ballistic missile submarines deployed in the Pacific OceanY Given these and prior assurances, the nuclear weapons withdrawal was not interpreted as an erosion of the U.S. commitment to Korea,43 and did not seem to weaken the U.S.-ROK alliance. THE IMPACT OF DIVERGING U.S.-ROK PRIORITIES AND SOUTH KOREAN DOMESTIC CHANGE

New engagement by both the United States and South Korea with the North, greater South Korean self-confidence flowing from its economic development, and the effects of South Korea's democratic consolidation together resulted in greater volatility in U.S.-ROK alliance relations in the 1990S and early 2000S. The external context for the alliance was transformed by new post-Cold War opportunities for engagement with North Korea. These introduced new potential for tensions between Washington and Seoul and presented greater U.S.ROK coordination responsibilities, challenges that required the two allies to engage in mutual reassurance. The first post-Cold War test of the alliance came as a result of differences between Washington and Seoul over how to deal with North Korea. These differences emerged partly as a result of the Clinton administration's decision to negotiate directly with the DPRK over the nuclear issue, a step that deviated from a long-standing U.S. policy of taking a back seat to Seoul on all interactions with North Korea. IAEA inspections starting in 1992 had raised questions about the accuracy of North Korea's declared activities, leading North Korea to announce in March 1993 its intention to withdraw from the NPT. Although the DPRK did not act to make good on its withdrawal announcement at the end of a three-month wait required by the treaty, tensions rose again in May 1994 when North Korea removed fuel rods from its five megawatt reactor in the absence ofIAEA supervision. The Clinton administration began planning possible military options, putting North Korea and the United States on a course for conflict. This was forestalled when former President Jimmy Carter created space for renewed U.S.-DPRK negotiations during a meeting with Kim Il-sung in Pyongyang in June OfI994. Subsequent U.S.-DPRK talks resulted in negotiation of the Agreed Framework in Geneva in October 1994. In the course of the negotiations, North Korea suspended its withdrawal from the NPT and agreed to freeze activities at its nuclear reactor and plutonium reprocessing site. Despite this apparent success, the bilateral framework for negotiations was

INFUSING COMMITMENT WITH CREDIBILITY

171

frustrating to South Korean President Kim Young-sam (elected in 1992) because South Korea was excluded from U.S.-DPRK talks on issues that directly affected South Korean security. Such a structure induced distrust between the allies, despite high levels of bureaucratic coordination between Washington and Seoul. In the course of the Agreed Framework negotiations, moreover, the United States provided negative security assurances to North Korea. 44 The U.S.ROK alliance was inevitably strained by the apparent contradiction between a new U.S. pledge not to use or threaten to use nuclear weapons against the DPRK and the goal of the alliance to deter North Korea. The U.S.-ROK alliance faced additional challenges under presidents Kim Dae-jung and Roh Moo-hyun. Kim Dae-jung's election in December 1997 represented South Korea's first peaceful transition of power to the opposition progressive party, and the new government pursued diplomatic engagement with the North through what it called "the Sunshine Policy:' The perceived progress in inter-Korean relations that followed raised questions in some quarters about the continued viability and/or need for the alliance with the United States. These doubts were accompanied by a desire on the part of South Korea to take a more independent role in diplomacy. Some progressives began to view the alliance as more of a constraint and a burden than a benefit, a perception that was reflected in conflicting views between the two countries over how to deal with North Korea and how to share the cost of keeping the U.S. troops stationed in Korea. 45 The first ever inter-Korean summit in June of 2000 had a clear impact on South Korean public opinion toward the North as well as toward the future of the alliance. According to a survey conducted by Donga Ilbo immediately following the summit, less than 10 percent of respondents viewed North Korean ruler Kim Jong-il as a dictator, compared with more than 34 percent prior to the summit. 46 Expectations for Korean reunification also rose among Korean students following the summitY To the extent that North Korea no longer seemed to be a threat, the continued viability of the U.S.-ROK alliance was called into question. Another survey result from September 2000 suggested that 64 percent of respondents favored either an immediate withdrawal or a partial reduction ofUSFK. 48 During this period of rising Korean public hostility toward USFK as a symbol of the alliance, it became necessary for Kim Dae-jung to provide assurances to the United States regarding the value attached to the alliance by South Korea. Kim publicly emphasized that improvements in inter-Korean relations were

172

SNYDER AND LEE

only possible with the U.S.-ROK military alliance and that the continued presence of USFK would be vital not only for security but also for the diplomatic and economic strength of South Korea. 49 This is a rare case in the history of the alliance in which assurances flowed from the object of protection to the partner providing security guarantees. Kim Dae-jung's efforts to engage North Korea were received poorly at the beginning of the George W. Bush administration. While Kim's policy was based on the possibility of promoting trust-building reconciliation with North Korea and changing North Korea's core behaviors through economic integration with the outside world, the Bush administration saw the Sunshine Policy as a oneway giving of concessions to an untrustworthy dictatorship. 50 The Bush administration's extreme skepticism toward both North Korea and the Sunshine Policy led to serious levels of mistrust between the two administrations. U.S.-ROK mistrust gradually came to be manifest in South Korean public concerns that the United States might attack North Korea militarily, thereby entrapping the South into an unwanted and unnecessary conflict. These South Korean views were especially strong following President Bush's naming of North Korea as part of the "axis of evil" in the run-up to the U.S. invasion of Iraq.51 An incident in 2002 in which a U.S. armored vehicle accidentally struck and killed two South Korean schoolgirls added to public mistrust of the ROK's American ally. President Bush's reassurance to South Korea that the United States had no plans to attack North Korea helped to ease concerns in Seoul, but the levels of mistrust between Washington and Seoul had corrosive effects on the public mood in which the institutions of alliance cooperation continued to function. Nor did South Korean concerns about Bush administration intentions entirely abate. Kim Dae-jung's successor Roh Moo-hyun, another South Korean progressive, suggested that he was more worried about whether the United States would strike North Korea than whether North Korea would start a war with South Korea. 52 The South Korean public was also widely reported to perceive the United States as a greater source of threat to peace and stability than North Korea in 2004, despite earlier revelations of North Korea's covert nuclear weapons pursuits. 53 These revelations, announced publicly after a meeting in October 2002 at which the United States claimed North Korea had admitted seeking to construct a secret uranium enrichment plant, led to the unraveling of the U.S.DPRK Agreed Framework. 54 The DPRK responded by kicking out IAEA inspectors, resuming operation of its plutonium production facilities, and an-

INFUSING COMMITMENT WITH CREDIBILITY

173

nouncing-and this time making good on-its intention to withdraw from the NPT. North Korea's renewed production of fissile materials in early 2003 challenged an underlying premise of Kim Dae-jung's Sunshine Policy that the North Korean nuclear issue was under control, putting U.S. and South Korean priorities directly at odds with each other. While the main objective of u.S. policy toward North Korea was a resolution to the North Korean nuclear crisis, the priority goal of South Korea's DPRK policy was preservation of peace and stability on the peninsula. In addition to diverging views of the threat from North Korea, another challenge to cooperation between the two allies emerged as South Korea sought greater equality with and independence from the United States both militarily and politically. These factors led to changes in the alliance after Roh Moo-hyun became president in early 2003. As a result of South Korea's rising confidence in its military capability and economic stability, the Roh Moo-hyun administration sought to readjust the U.S.-ROK alliance and give South Korea greater control over its own security forces. Negotiations on revised operational control arrangements produced an agreement to dissolve the CFC by April of 2012 and replace it with separate command structures under which the ROK would take a leading role and U.S. forces would playa supporting role. 55 Due to concerns arising from subsequent North Korean behavior, however, in June 2010 the date for dissolving the CFC was pushed back to December 2015. Some of these concerns arose from North Korea's progress in its nuclear weapons program. The North Koreans in October 2006 conducted their first test of a nuclear device, raising the need for additional forms of assurance from the U.S. government. Although the Roh administration played down the South Korean reaction to the test and even refused to join the U.S.-led Proliferation Security Initiative (PSI) in the wake of the test, Secretary of State Condoleezza Rice traveled to Seoul and Tokyo within two weeks of the test and offered strong statements of assurance regarding the U.S. commitment to extended deterrence. 56 Despite the Roh administration's low-key approach to North Korea's test, former South Korean defense ministers reportedly asked for the return of U.S. nuclear weapons removed from South Korea in 1991, and some conservatives called for South Korea to develop its own independent nuclear capability.57 At the annual Security Consultative Meeting of defense officials, which took place not long after the test, South Koreans asked for the term "extended deterrence" to be included in the joint communique. The United States agreed, with some reluctance, and the final text "offered assurances of firm U.S. com-

174 SNYDER AND LEE

mitment and immediate support to the ROK, including continuation of the extended deterrence offered by the u.s. nuclear umbrella:'58 Even during a period of diverging priorities between the United States and South Korea, renewed U.S. assurances proved to be relevant. LEE MYUNG-BAK'S EFFORTS TO "RESTORE"THE ALLIANCE

After conservatives won the December 2007 election, the new president, Lee Myung-bak, came into office determined to "restore" the U.S.-ROK alliance following years of perceived tensions with the United States under Kim Dae-jung and Roh Moo-hyun. After repeatedly emphasizing that "restoring the U.S.-ROK alliance based on the established friendship" is a top priority of his administration,59 Lee traveled to the United States as his first foreign destination, where he was hosted by President Bush at Camp David. The Joint Statement of the summit expressed the two sides' commitment to the U.S.-ROK alliance as a "strategic alliance that seeks to enlarge common interests on the basis of universal values and strong truSt:'60 This declaration might be viewed as an exercise by both sides in mutual reassurance, designed to reaffirm the alliance as the foundation for close partnership on both North Korea and broader regional and security issues. Inter-Korean relations, in contrast, deteriorated early in the Lee Myung-bak administration in part due to the North's negative reaction to Lee's efforts to introduce greater reciprocity into relations with North Korea. Reports of Kim Jong-il's ill health and the initial stages of a leadership succession process also appeared to be factors behind a series of North Korean provocative actions in early 2009, including the launch of a multistage rocket and a second nuclear test on May 25,2009. 61 North Korea's 2009 missile and nuclear tests drew strong condemnation from the UN Security Council, but also stimulated a need for further assurance measures in the U.S.-ROK alliance. Notably, the Lee government, believing that the South needed greater protection from North Korean threats of nuclear blackmail, requested a written pledge of the U.S. commitment to the defense of South Korea. This was included in a joint vision statement produced by a summit meeting between Lee and recently elected U.S. President Barack Obama in June 2009. Meeting less than a month after North Korea's second nuclear test, the two presidents reaffirmed the U.S.-ROK security alliance and discussed ways to cooperate on denuclearization of the Korean Peninsula. The joint vision statement reaffirmed the provision of U.S. positive security assurances to South Korea, maintaining

INFUSING COMMITMENT WITH CREDIBILITY

175

that "the continuing commitment of extended deterrence, including the u.s. nuclear umbrella, reinforces this assurance:'62 Keith Payne states, "There could hardly be a stronger allied statement of the perceived value of u.s. nuclear weapons for the continued assurance of allies or a more explicit rejection of u.s. ambiguity in its extended deterrence commitments:'63 North Korea's nuclear tests catalyzed a debate within South Korea over whether it could count on the u.s. nuclear umbrella for its security. Less than a week after the first North Korean nuclear test in October 2006, South Korean public opinion showed a dramatic shift, with nearly two-thirds of South Koreans favoring nuclear acquisition. According to a ]oongAng Ilbo poll,64 65 percent of respondents said that South Korea should develop nuclear weapons to protect itself. This view was also articulated by Kim Dae-joong, a leading columnist for Chosun Ilbo. He argued that it would be difficult for South Korea, in the face of a nuclear-armed North Korea, not to reconsider its position on nuclear weapons, in the process questioning the u.S. sense of responsibility for the future security of South Korea. 65 In a less radical position, in the months following the second nuclear test, some academics and retired military officers called for redeployment of U.S. tactical nuclear weapons to South Korea. 66 Further North Korean provocations in 2010 resulted in the possibility of redeployment being raised even by high-ranking government officials. In March, South Korea blamed the North for sinking an ROK naval vessel, the Cheonan. In November, Pyongyang revealed a new uranium enrichment facility it had constructed to a delegation of American experts. Shortly thereafter, North Korea shelled a South Korean island, killing four. After the revelation of the uranium enrichment facility, South Korean Defense Minister Kim Tae-young told parliament that the government would consider asking the United States to redeploy tactical nuclear weapons. 67 The Lee government has so far rejected such a step, however. In April 2010, not long after the Cheonan incident, Foreign Minister Yu Myung-hwan ruled out asking for redeployment because it would be contrary to u.S. global nuclear policy objectives. 68 The government also quickly disavowed Defense Minister Kim's suggestion that redeployment was under consideration. A government spokesperson explained that redeployment would violate earlier agreements to denuclearize the Korean Peninsula. 69 North Korea's provocations did raise the importance, however, of the United States providing political assurances to the South Korean public. In this context, there is no question about the political need to include U.S. extended de-

176

SNYDER AND LEE

terrence pledges in the 2009 joint vision statement, which clarifies in writing the u.s. commitment at the highest levels to extended deterrence and provision of a nuclear umbrella for South Korea. This kind of assurance is designed to impress upon the South Korean public that the costs of going nuclear loom larger to South Koreans than the actual risks of refraining from it. But, in practice, the statement did not provide a further commitment to extended deterrence and u.S. defense of South Korea beyond that already reflected in the Mutual Defense Treaty or restated in prior U.S.-ROK Military Committee Meeting (MCM) and Security Consultative Meeting (SCM) statements following North Korea's first nuclear test in 2006. So it is fair to ask whether the need for reassurance regarding the u.S. commitment to South Korea's defense following North Korea's second nuclear test sheds greater light on continuing South Korean doubts about American commitments or on the readiness of the United States to provide assurances to South Korea at an early stage as a means by which to shore up the credibility of the U.S. security commitment to South Korea. Following the North's shelling of Yeonpyeong Island, the two sides agreed in December 2010 to establish a joint Extended Deterrence Policy Committee.7° This committee provides a mechanism for Seoul and Washington to discuss the functioning of extended deterrence against the North Korean nuclear threat. U.S. assurances have not been purely nuclear, however, but also include a conventional dimension. The two sides held new joint military exercises, for example, following the shelling incident. An important issue yet to be addressed is whether South Korea and the United States would have the same thresholds in terms of expectations regarding the circumstances under which the U.S. defense commitment-including aspects of extended deterrence-would be exercised in practice. In the course of preparing the 2010 Nuclear Posture Review, the Obama administration carefully considered the concerns of the governments of South Korea and Japan in an effort to ensure the credibility of extended deterrence commitments to allies while at the same time minimizing the role of nuclear weapons as a component of U.S. national defense strategy.71 It also remains unclear how the various parties assess the significance and impact of the U.S. extended deterrence commitment to South Korea. Is it regarded primarily as an assurance measure targeted at South Korea or as a deterrence measure targeted at the North? The ostensible reason for South Korea to request such an assurance would be to counter any possible move by the North to extort South Korea by utilizing the threat of a nuclear strike, but from a U.S.

INFUSING COMMITMENT WITH CREDIBILITY

177

perspective, the main target of security assurances was South Korea itself; any impact on North Korea beyond bolstering the u.S. commitment to deterrence is most likely ancillary. A further complication regarding u.S. provision of positive assurances to South Korea through extended deterrence is the impact of those assurances to the South on North Korea's perception of the negative security assurances that the United States has periodically offered in a variety of forms, both in the context of the 1994 Agreed Framework and a later Six-Party Joint Statement in September 2005. What can be said is that u.S. security assurances to the ROK have been and remain important, but have also required periodic efforts to infuse them with credibility. What does this history tell us about the hypotheses introduced by Jeff Knopf? TESTING HYPOTHESES

Examination of the four phases of the U.S.-ROK alliance covered in this case study yields several insights regarding the following hypotheses from Chapter 2:

Hypothesis 1: Assurances are more likely to be effective when a target state's interest in nuclear weapons is driven to a significant degree by security concerns. This hypothesis is supported, but with qualifications. Security can be a motivation for exploring nuclear weapons without assurances necessarily being effective or even the only option to address security concerns. The review of the Park Chung-hee period suggests that South Korea's confidence in its own security was set primarily by the level of threat it perceived, and only secondarily by whether sufficient assurances were provided to meet that threat, not vice versa. The Park government's security concerns were focused on the superior North Korean military power, but its interest in nuclear weapons was also affected by its questions about whether or not U.S. commitments to the defense of South Korea were credible. As suggested by Janice Stein in Chapter 3, South Korean security fears may have complicated the American task of providing assurances by driving up the threshold for what would be necessary to successfully assure South Korea beyond "rational" calculations of what should have been sufficient. The United States used both South Korea's fear of abandonment and assurances of U.S. continued military presence to reverse South Korea's drive for an indigenous nuclear program. A nuclear guarantee was not the only consideration in the South's sense of security. When the South Korean military buildup eventually became superior to that of North Korea and North Korean conventional power was no

178

SNYDER AND LEE

longer considered a serious threat to South Korea in the post-Park periods, u.S. deployment of nuclear weapons in South Korea became less important to South Korea's defense strategy. The complete removal of these weapons, in 1991, seemed to have no clear impact on the effectiveness of U.S. assurance (in contrast to some NATO arguments in 2011 against u.S. removal of tactical nuclear weapons). The United States helped South Korea maintain a sense of security by increasing its sales of advanced conventional weapons in the 1990S and 2000S. An improved inter-Korean relationship in the years following the removal of u.S. nuclear weapons also added to the South Korean government's perceived level of security. As was evident during periods of reconciliation, South Korea's relations with the North and its perception of the North Korean threat also affected its alliance relationship with the United States. Based on the premise that South Korea will not ponder going nuclear, its relationship with the North and how it perceives the North Korean threat will have a greater impact on its sense of security than the strength of the security assurances provided by the United States. The critical factor is the recipient's perception: whether it is deterrence to an enemy or assurance to a friend, what makes security assurances effective is not in the hands of the provider but in the eyes of the recipient. South Korea's calculations regarding the development of nuclear weapons will be based on its own assessments of the degree of North Korean threat and the U.S. commitment.

Hypothesis 2: Public declarations of security assurances increase their effectiveness. Hypothesis 3: Legally binding mechanisms will increase the effectiveness of assurances. The perceived credibility of assurances is critical, but legally binding mechanisms such as the U.S.-ROK Mutual Defense Treaty are at most a necessary condition and not alone sufficient to assure credibility. The United States, as a provider of assurances, had to continuously ensure the credibility of its commitment. On several occasions, this required making public declarations that the U.S. commitment to South Korea remained strong. As a way to underscore U.S. credibility, for example, the inclusion of the extended deterrence commitment in the 2009 U.S.-ROK joint vision statement appears to have been effective at the government level, but not fully with the South Korean public. One challenge going forward is how to institute follow-up measures that

INFUSING COMMITMENT WITH CREDIBILITY

179

are effective in convincing the South Korean public that U.S. extended deterrence commitments remain credible. Dissatisfaction still pervades South Korea among those who feel that even the joint vision statement is not enough.72 In the South Korean case, public declarations have served as a way to buttress legally binding mechanisms, but there are inherent limits to what even the combination can accomplish. As Shim Sangsun argues, "The security of a small power cannot be assured by only a security guarantee, such as a nuclear umbrella. Under the anarchy of international politics, small powers are always aware that they may be sacrificed in case of conflicts between and among great powers."73

Hypothesis 4: The greater the political and economic ties between sender and recipient, the more effective security assurances will be. This hypothesis is supported, but not only because extensive ties make extended deterrence commitments more credible. These same ties can also be a source of coercive leverage for the security provider. South Korea's economic dependence upon the United States and its allies was a critical factor in South Korea's decision to abandon its nuclear project. When the United States discovered the existence of a secret nuclear development project under the Park regime, the Ford administration warned that continuing such program would result in a loss of American Export-Import Bank loans for South Korea's civilian nuclear program and "jeopardize" the U.S.-ROK security relationship.74 The United States was South Korea's largest trading partner and if it persuaded Japan, the second largest partner, to join in the sanctions, the economic damage to South Korea would have been devastating.7 5 A major consideration for South Korea therefore was not to risk creating economic hardship that would be detrimental to its political stability and economic growth. In these conditions, the economic and political pressures from the United States complemented the impact of U.S. assurances in Korea.

Hypothesis 6: Forward-deployed troops and other forms of defense cooperation will increase the effectiveness of security assurances. Assurances backed by concrete actions, such as forward-deployed troops, are a critical tool by which to underscore the credibility of alliance commitments. The United States used forward-deployed troops as an assurance that the alliance cooperation would be sufficient to meet South Korea's security needs both with Park Chung-hee and with Chun Doo-hwan. Although one cannot deny that U.S. nuclear weapons, rather than the U.S. troop presence, were viewed by the Seoul government as the ultimate guarantor of South Korea's security dur-

180

SNYDER AND LEE

ing the Park Chung-hee period,76 the withdrawal of the u.s. nuclear weapons in the early 1990S did not weaken the credibility of assurances due to an increased effort by the u.s. government to provide reassurances that were meaningful to South Korea. The Carter administration's willingness to create new forms of defense cooperation such as the CFC and the Reagan administration's commitment to maintain forward-deployed troops increased the effectiveness of security assurances during these periods. It is clear that the Chun and Roh administrations' embrace of nuclear reversal was largely attributable to Washington's continued reaffirmation of a security guarantee that was demonstrated not only by words but also by strengthened military cooperation between the two countries. Following North Korea's nuclear tests in 2006 and 2009, and again after its various provocations in 2010, the United States has provided additional assurances and has promoted the strengthening of alliance cooperation against North Korea. Although these assurances were provided against the backdrop of a reconfiguration of U.S. forces on the Korean Peninsula, the United States has taken actions in support of its commitments that have thus far had the political effect of providing assurance to its South Korean counterparts. The establishment of a bilateral Extended Deterrence Policy Committee to discuss implementation of the joint vision statement is one welcome example, although as discussed above these measures have not fully alleviated the concerns of the South Korean public.

Hypothesis 7: Assurances will be most effective if they are tailored to take account of unique features of the target state's culture, decision-making procedures, and leadership concerns. Assurance measures in the four phases ofU.S.-ROK alliance were tailored to take account of unique features of South Korea's leadership concerns and decision-making procedures. The Park government's growth-oriented, export-led model convinced policymakers that the benefits from continued political and economic ties with the United States outweighed South Korea's security interests in developing an independent nuclear arsenal. The U.S. use of a combination of assurances and economic threats made the Park regime recognize that continuing the existing nuclear program, such as efforts to acquire plutonium reprocessing capability, would seriously undermine South Korea's economic strategy of export-led industrialization. In dealing with the authoritarian Chun administration, President Reagan

INFUSING COMMITMENT WITH CREDIBILITY

181

made gains by de-emphasizing the human rights issue, at least publicly, although Chun's dictatorship concerned many in the u.s. government. 77 Chun's and Roh's efforts to gain diplomatic recognition from the United States and the outside world by securing the 1988 summer Olympics for Seoul and achieving UN membership in 1991 likely also contributed to South Korea's willingness to forgo a nuclear weapons option. Economic modernization was also driving changes in South Korea's capacity to contribute to its own self-defense and its level of confidence vis-a-vis a flagging North Korea. Thus, both the Chun and Roh governments viewed denuclearization and continued reliance on U.S. security assurances as an optimal choice for the security of South Korea. During the Sunshine Policy period, mutual assurances between the United States and South Korea were necessary as a foundation for perpetuating trust between alliance partners in the context of South Korea's steps toward reconciliation with North Korea and its own democratization.

Hypothesis 10: States with regional security concerns will be most interested in positive security assurances. The bottom-line requirement in the South Korean debate over its national security was its desire for a positive security assurance by a stronger ally, namely the United States, as regional security, primarily the North Korean threat, remained its core concern. U.S. extended deterrence has been an enduring feature of its security commitment to South Korea that has assisted South Korea in deciding to forswear its own nuclear development. U.S. commitments to extended deterrence have taken on greater significance following North Korea's nuclear tests. The written pledge in the 2009 joint vision statement thus reaffirmed the positive security assurances given to South Korea. With these security guarantees, South Korea may even be better off without its own nuclear weapons because it is economically costly and politically risky to develop nuclear weapons programs when a country can enjoy protection by a stronger nuclear weapon state. A special challenge in recent years for U.S. policy has been how to provide the right mix of positive security assurances to South Korea to assure the credibility of the alliance and uphold deterrence while also providing negative security assurances to North Korea. Since the 1990S, one delicate issue is the question of how U.S. provision of negative security assurances to North Korea might influence U.S. positive security commitments to South Korea. This apparent contradiction has become more than hypothetical since North Korea's first nuclear

182

SNYDER AND LEE

test in 2006. On the one hand, the United States reiterated its commitment to extended deterrence on behalf of South Korea, but on the other hand, the September 2005 Six-Party Joint Statement contained negative security assurances that the United States had no intent to attack North Korea. The issue is less whether or not the operational aspects of these two types of assurances are contradictory since one might argue that the U.S. negative security assurance is not operative if North Korea is not meeting its obligations under the Six-Party talks. But the respective commitments underscore the necessity of a parallel and holistic approach between the United States and South Korea to possible reconciliation with North Korea.

Hypothesis 13: Assurances will be more effective if they are packaged with positive incentives. Seoul's decision to abandon its nuclear project and its continued reluctance to reconsider a nuclear option in the face of a rising North Korean nuclear threat reveal how positive-and also negative-incentives packaged with assurances can persuade a country, already on the road to nuclear proliferation, to reverse its course. Park Chung-hee and his successors faced various dilemmas in deciding whether to continue or forgo nuclear weapons development. Among many considerations, promises by presidents Ford, Carter, and Reagan to suspend U.S. troop withdrawals were a critical factor in persuading Seoul to back off from nuclear acquisition. South Korea gave up its nuclear program in return for a continued presence of U.S. troops on the peninsula that would act as a tripwire to deter North Korean aggression, upgraded military assistance that included sales of advanced conventional weapons, and closer economic and diplomatic ties with the United States and with the rest of the world. To the South Korean calculation, the benefits coming from these positive incentives far outweighed the benefits of being an isolated nuclear power. CONCLUSION

This study of the role of security assurances during various phases in the U.S.ROK alliance has illustrated a variety of different types, purposes, and effects of assurances as a means by which to underscore the credibility of U.S. defense commitments under the U.S.-ROK Mutual Defense Treaty. Over the history of the U.S.-ROK alliance, South Korean relations with the United States changed from a wholly dependent relationship to one in which South Korea has a distinct capacity, role, and set of interests. During the period when South Korea

INFUSING COMMITMENT WITH CREDIBILITY

183

exclusively depended on its alliance relations with the United States, the South Korean government constantly sought U.S. assurances of its alliance commitments. As South Korea became more confident in its own capabilities as a result of its economic development and democratization, new "internal" challenges to the credibility of alliance commitments emerged on both sides, requiring assurance that the United States would not entrap South Korea into a potential conflict, either on or off the peninsula. In addition, traditional positive assurances in the form of U.S. extended deterrence commitments have played an important role in discouraging South Korean efforts to pursue nuclear weapons capabilities. An interesting feature of this case is that, while South Korea has clearly valued the U.S. nuclear umbrella, the relevant positive assurances have not been exclusively nuclear in nature. Forward-deployed troops, joint military consultations and exercises, and the offer of advanced conventional capabilities have also been important sources of assurance, especially since the United States withdrew its tactical nuclear weapons in 1991. The credibility and intent behind U.S. commitments have often been subject to doubt. Because U.S. extended deterrence commitments under the alliance remain untested in practice, those commitments may still be subject to question among South Koreans. Nonetheless, there appears to be little likelihood that South Korea would pursue a nuclear weapons option at the current stage. U.S. efforts to align its assurances to both South and North Korea are potentially more problematic. U.S. signals meant to reinforce positive security assurances to South Korea in response to North Korea's nuclear tests, given while the United States is also providing negative security assurances to the North that the United States has no "hostile intent" to attack the North, may be subject to misinterpretation and confusion within the respective targets in each case. This is particularly the case because regardless of the intent behind the assurances offered, the credibility of assurances is in the eye of the beholder. The interaction between U.S. positive security assurances to South Korea and negative security assurances to North Korea, including the intent of the messages and their respective interpretations by the two Koreas, is a subject that will require more in -depth study in the future. A related question is the extent to which U.S. positive security assurances to South Korea have resulted in improved deterrence of the North. As both South Korea and the United States have simultaneously engaged in dialogue and deterrence with North Korea, it is important to understand in greater detail the nature and effects of assurances in order to avoid any possible misunderstandings or miscalculations among the concerned parties.

184

SNYDER AND LEE

NOTES 1. William J. Taylor, Jr., "Challenges to ROK-U.S. Security Relations;' in The Future of South Korean- U.S. Security Relations, ed. William J. Taylor ( Boulder, CO: Westview Press, 1989), 114. See also Gilbert Rozman and Shin-hwa Lee, eds., South Korean Strategic Thought toward Asia (New York: Palgrave Macmillan, 2008). 2. Mitchell Reiss, Without the Bomb: The Politics of Nuclear Nonproliferation (New

York: Columbia University Press, 1988), 85. 3. Peter Hayes, Pacific Powderkeg: American Nuclear Dilemmas in Korea (Lexington, MA: Lexington Books, 1991), 35. 4. Franklin B. Weinstein, "The Nuclear Dimension;' in The Security of Korea: U.S.

and Japanese Perspectives on the 1980s, ed. Franklin Weinstein and Fuji Kamiya (Boulder, CO: Westview Press, 1980), 108. 5. Suh Sung, Unbroken Spirits: Nineteen Years in South Korea's Gulag (Lanham: Rowman and Littlefield, 2001), 23. 6. Chae-Jin Lee, A Troubled Peace: U.S. Policy and Two Koreas (Baltimore, MD: Johns Hopkins University Press, 2006), 73. 7. Etel Solingen, Nuclear Logics: Contrasting Paths in East Asia and the Middle East (Princeton: Princeton University Press, 2007), 84. 8. Sangsun Shim, "The Causes of South Korea's Nuclear Choices: A Case Study in Nonproliferation;' Ph.D. dissertation, University of Maryland, 2003, ProQuest Dissertations and Theses, UMI #3112610, 48. 9. Hyung-A Kim, Korea's Development under Park Chung-hee: Rapid Industrialization 1961-79 (New York: Routledge, 2004), 109. 10. Don Oberdorfer, The Two Koreas: A Contemporary History (Boston: AddisonWesley, 1997), 68. 11. "Seoul Chief Terms US Troops Vital;' New York Times, June 24, 1970, 1. 12.

Seoul Shinmun, May 18, 1977, as quoted in Shim, "The Causes of South Korea's

Nuclear Choices;' 48. 13. Shim, "The Causes of South Korea's Nuclear Choices;' 44-48. For more details

on U.S. weapons deployed in South Korea, see Hans M. Kristensen, ''A History of U.S. Nuclear Weapons in South Korea;' September 28, 2005, http://www.nukestrat.com/ korea/koreahistory.htm. 14. Alon Levkowitz, "The Seventh Withdrawal: Has the US Forces' Journey back Home from Korea Begun?" International Relations of the Asia-Pacific 8, no. 2 (May 2008): 142 • 15. Rowland Evans and Robert Novak, "Korea: Park's Inflexibility;' Washington Post,

June 12, 1975. 16. Jonathan D. Pollack and Mitchell B. Reiss, "South Korea: The Tyranny of Geography and the Vexations of History;' in The Nuclear Tipping Point: Why States Reconsider

INFUSING COMMITMENT WITH CREDIBILITY

185

Their Nuclear Choices, ed. Kurt M. Campbell, Robert J. Einhorn, and Mitchell B. Reiss (Washington, DC: Brookings Institution Press, 2004), 262. 17. Solingen, Nuclear Logics, 84. 18. Scott Snyder, "South Korean Nuclear Decision Making;' in Forecasting Nuclear

Proliferation in the 21st Century: Volume 2, A Comparative Perspective, ed. William C. Potter (Stanford: Stanford University Press, 2010),158-81. 19. T. V. Paul, Power versus Prudence: Why Nations Forgo Nuclear Weapons (Montreal and Kingston: McGill-Queen's University Press, 2000),121. 20. Michael J. Engelhardt, "Rewarding Nonproliferation: The South and North Korean Cases;' Nonproliferation Review 3, no. 3 (Spring-Summer 1996): 32. 21. Pollack and Reiss, "South Korea;' 263. 22. Kang Choi and Joon-Sung Park, "South Korea: Fears of Abandonment and Entrapment;' in The Long Shadow: Nuclear Weapons and Security in 21st Century Asia, ed. Muthiah Alagappa (Stanford: Stanford University Press, 2008),377-78; Michael J. Siler, "U.S. Nuclear Nonproliferation Policy in the Northeast Asian Region during the Cold War: The South Korean Case;' East Asian Studies 16, nos. 3-4 (AutumnIWinter 1998): 71-74. 23. Oberdorfer, The Two Koreas, 73. 24. Sang Hun Choe, "South Korea Was Close to A-Bomb Development in Early 1980s;' Associated Press, October 5, 1995. 25. Pollack and Reiss, "South Korea;' 263. 26. Human rights criticism of South Korea had risen sharply in 1980 as a result of the Kwangju massacre in May, in which troops killed perhaps 1,000 to 2,000 prodemocracy demonstrators, and the arrest of Kim Dae-jung, then the leading dissident politician. For Reagan administration reasoning for working with Chun, see "Seeing Human Rights in the 'Proper Manner': The Reagan-Chun Summit of February 1981;' National Security Archive Electronic Briefing Book no. 306, ed. Robert Wampler, posted February 2, 2010, http://www.gwu.edu/-nsarchiv/NSAEBB/NSAEBB306/index.htm. 27. Shim, "The Causes of South Korea's Nuclear Choices;' 111-16. 28. Brian Bridges, Korea and the West (New York: Routledge and Kegan Paul, 1986), 60. 29. Boca Raton News, "Reagan Says Troops to Stay in South Korea;' February 3,1981,

http://news.google.com/newspapers?nid=1291&dat=19810203&id=X8kPAAAAIBAJ&sji d=W40DAAAAIBAJ &pg=6601,740816. 30. Bridges, Korea and the West, 61. 31. Michael Getler, "Viewing Maneuvers from the 'Royal Box'; President of South Korea, Weinberger Observe Exercises under Very Tight Security;' Washington Post, April 1,1982, A20. 32. Siler, "U.S. Nuclear Nonproliferation Policy;' 75-78.

186

SNYDER AND LEE

33. UN Command in South Korea, "History of Team Spirit Exercises;' Seoul, March 1986,3, as quoted in Oberdorfer, The Two Koreas, 93. 34. Getler, "Viewing Maneuvers from the 'Royal Box: " 35. Shim, "The Causes of South Korea's Nuclear Choices;' 14. 36. Ibid., 81. 37. Interview by Shim Sangsun, quoted in Shim, "The Causes of Soutil Korea's Nuclear Choices;' 119. 38. "South Korea-A-Arms Pullout Plan Reported;' New York Times, October 19,1991, 4·

39. Nuclear Information Project, "The Witildrawal of Nuclear Weapons from Soutil Korea;' Nuclear Brief, September 28, 2005, http://www.nukestrat.com/korea/withdrawal. htm (accessed September 1, 2009); Paul, Power versus Prudence, 123-24. 40. Robin Bulman, "No A-Arms in S. Korea, Roh Says;' Washington Post, December 19,1991. 41. Federation of American Scientists, "Nuclear Weapons Program-Nortl1 Korea;' http://www.fas.org/nuke/guide/dprklnuke/index.html (accessed May 25, 2010). 42. Nuclear Information Project, "The US Nuclear Umbrella over South Korea;' http://www.nukestrat.com/korea/umbrella.htm (accessed September 1, 2009). 43. J. Sterngold, "Seoul Says It Now Has No Nuclear Arms;' New York Times, December 19, 1991. 44. Appendix to the "Agreed Framework between the United States of America and the Democratic People's Republic of Korea;' October 21, 1994, http://www.atomicarchive.com/Docs/Deterrence/AgreedFramework.shtml. 45. Kongdan Oh, "U.S. -RO K: The Forgotten Alliance;' Brookings Northeast Asia Commentary, no. 22, http://WWW3.brookings.edu/opinions/2008/1O_south_korea_oh.aspx. 46. Donga Ilbo, May 31 and June 15, as cited in Scott Snyder, "The End of History, the Rise of Ideology, and the Pursuit of Inter-Korean Reconciliation;' in Korea Briefing 2000-2001: First Steps toward Reconciliation and Reunification, ed. Kongdan Oh and Ralph Hassig (Armonk, NY: Asia Society, 2002), 34-35. 47. Mann-Gil Han, "Role of Education in National Unification;' Korea Focus 9, no. 2 (2001): 134. 48. Lee Nae-young and Jeong Han-wool, "Anti-Americanism and tile ROK-U.S. Alliance;' East Asian Review 15, no. 4 (Winter 2003): 31, fig. 4. 49. Sung-Joo Han, "The Inter-Korean Relationship and Regional Security;' in Ko-

rean Security Dynamics in Transition, ed. Kyung-Ae Park and Dalchoong Kim (New York: Palgrave, 2001), 196. 50. White House Office of the Press Secretary, "Remarks by President Bush and President Kim Dae-jung of Soutil Korea;' March 7, 2001, http://usinfo.org/wf-archive/2001/010307/epi}02.htm.

INFUSING COMMITMENT WITH CREDIBILITY

187

51. Scott Snyder, "Pursuing a Comprehensive Vision for the U.S.-South Korea Alliance;' Center for Strategic and International Studies, April 2009. 52. David R. Sands, "S. Korea Opposes Attack on North;' Washington Times, February 14, 2003, AI. 53. "Number One Threat to South Korea's Security Is the United States;' Chosun Ilbo, January 12, 2004, http://english.chosun.com/w21data/html/news!200401!200401120029. html. 54. Paul Kerr, "North Korea Admits Secret Nuclear Weapons Program;' Arms Control Today 32, no. 9 (November 2002). 55. Yonhap News, "South Korea to Reclaim Wartime OPCON in April 2012;' February 23, 2007.

56. The Guardian, "Rice Vows to Back Allies over North Korea Crisis;' October 18, 2006, http://www.guardian.co. uk/world/ 2006/ octh8/northkorea. usa. 57. The Korea Times, "US Nuclear Umbrella: Double-Edged Sword for South Korea;'

June 24, 2009, http://www.koreatimes.co.kr/www/news/nation!2009/06/120_47427.htmi. 58. Choi and Park, "South Korea;' 381; quotation from 401 n. 42. 59. The Korea Times, "President Elect Vows Creative Diplomacy;' December 19,

2007· 60. "Statement of ROK -US Summit;' White House Office of the Press Secretary, August 6, 2008. 61. Scott Snyder, "Changes in Seoul's North Korean Policy and Implications for Pyongyang's Inter-Korean Diplomacy;' paper presented to the Conference on "Emerging Issues of North Korean Foreign Policy;' Centre for Korean Research, University of British Columbia, Vancouver, BC, June 25-26, 2009, 7. 62. White House Office of the Press Secretary, "Joint Vision for the Alliance of the United States of American and the Republic of Korea;' June 16, 2009. 63. Keith B. Payne, "On Nuclear Deterrence and Assurance;' Strategic Studies Quar-

terly 3, no. 1 (2009): 55. Dr. Payne served as a deputy assistant secretary of defense in the Bush administration from 2002 to 2003. 64. Chang-un Shin, "In Poll, 78% Say Engagement Policy Should Change;' JoongAng

Daily, October 11, 2006. 65. Kim Dae-joong, ''Abandoned at a Nuclear Crossroads;' Chosun Ilbo, editorial, March 12, 2007. 66. Lee Jong-Heon, "Calls for Nuclear Weapons in South Korea;' UPI, October 21, 2009· 67. Christian Oliver and Daniel Denby, "Seoul Raises Spectre of Return of US Nuclear Arms;' Financial Times, November 22, 2010, www.ft.com. 68. Jung Sung-ki, "Seoul Won't Ask U.S. to Redeploy Nukes;' Korea Times, April 21, 2010, http://www.koreatimes.co.kr/www/news/nation/201O/05/205_64622.html.

188

SNYDER AND LEE

69. "S. Korea Denies Seeking Redeployment of U.S. Tactical Nuclear Weapons;' Yon-

hap News, November 23, 2010. 70. "S. Korea, U.S. Launch Joint Committee to Deter N. Korea's Nuclear Threats;'

Yonhap News, December 13, 2010. 71. Scott Snyder, "Finding a Balance between Assurances and Abolition: South Ko-

rean Views of the Nuclear Posture Review;' Nonproliferation Review 18, no. 1 (March 2011): 147-63. 72. Tae-woo Kim, "ROK-US Defense Cooperation against the North Korean Nu-

clear Threat: Strengthening Extended Deterrence;' in The US-ROK Alliance Relations

for the 21st Century, ed. Jung-Ho Bae and Abraham Denmark (Seoul: Korea Institute for National Unification, 2009), 181-220. 73. Shim, "The Causes of South Korea's Nuclear Choices;' 255. 74. Paul, Power versus Prudence, 121.

75. Engelhardt, "Rewarding Nonproliferation;' 32; Reiss, Without the Bomb, 96. 76. Reiss, Without the Bomb, 106.

77. Bridges, Korea and the West, 61.

Assessing the Role of Security Assurances in Dealing with North Korea John S. Park

After conducting nuclear weapon tests in 2006 and 2009, North Korea has increasingly insisted that it has become the world's ninth nuclear weapon state. Many interpretations of North Korean nuclear decision-making leading to this situation are inadequate in explaining Pyongyang's nuclear behavior. While hostile u.s. policy is frequently cited by North Korea as the reason it needs a nuclear deterrent, a careful examination of nuclear behavior by the Democratic People's Republic of Korea (DPRK) reveals that even when the United States has offered explicit security assurances, North Korea has continued with its nuclear weapons development. This behavior raises important questions about our understanding of security assurances and North Korea's stated policy goals. This chapter examines the relationship between security assurances and North Korean nuclear decision -making by focusing on four key areas. l The first is an examination of key geopolitical shocks that had a major impact on the North Korean regime. Recent monographs portray Pyongyang as being highly effective at applying a consistent, proactive, and perfidious strategy of using its nuclear program to coerce other countries to accede to its demands. 2 North Korean responses to geopolitical events related to security assurances reveal a significantly different picture than the one depicted in this literature. The second area covered by this chapter is an analysis of the main sources of security assurances for North Korea over its history. Two treaties signed in 1961, with dIe Soviet Union and People's Republic of China (PRC), provided positive security assurances. In the early 1990S with the end of the Cold War and dIe collapse of the Soviet Union, it became clear that Pyongyang could no longer rely on Moscow's security assurance. During this period, North Korea began seek189

190 JOHN S. PARK

ing a new source of assurances, focusing its efforts on the United States. North Korea received U.S. negative security assurances in the 1994 Agreed Framework and a 2005 joint statement emanating from the Six-Party Talks, as well as a more general reassurance in a 2000 joint communique. Third, this chapter assesses this volume's hypotheses on security assurances based on how North Korea reacted to geopolitical shocks. Although assurances have clearly not succeeded in persuading Pyongyang to abandon its nuclear weapons program, the case suggests that they have the potential to moderate North Korea's behavior under the right conditions. Of Jeffrey Knopf's initial hypotheses regarding security assurances, the analysis of North Korean behavior at least partially sustains hypotheses 4,7,8,9,10,12, and 13, though in some cases only with modifications. Security assurances appear most likely to be effective when they are tailored to North Korea's particular concerns, including with threats outside its region; embedded in a larger strategy that incorporates positive incentives; capable of influencing internal debates; and not hamstrung by the provider's domestic or alliance constraints. The fourth and final part of the analysis explores the conditions under which security assurances may be most effective in dealing with North Korea in the future. In its negotiations with the United States during the second Bush administration, North Korea sought a package deal through which Pyongyang would receive tangible assurances for regime survival. However, China may be supplanting the United States as the most likely provider of assurances. Since the mid-2000s, concerted Chinese interactions with North Korea through the Communist Party of China (CPC)-Workers' Party of Korea (WPK) channel have enabled the North Koreans to attain significant portions of their desired package deal from Beijing. The pace of deal-making in this CPC-WPK channel accelerated as North Korea's relations stalled with both the Obama administration in Washington and the Lee Myung-bak administration in Seoul. It is unclear how developments in late 2010, including North Korea's revelation of a new uranium enrichment plant and shelling of a South Korean island, will affect PRC-DPRK relations. The picture has been further unsettled by a leadership succession following the death of Kim Jong-il in late 2011. However, China's strategic interest in offering a package deal including security assurances that could restrain North Korea's behavior has continued undiminished. This chapter concludes that the Chinese approach appears to be the most viable among the feasible options for using assurances, but even with this approach the prospects of getting North Korea to give up nuclear weapons are not good.

ASSESSING THE ROLE OF SECURITY ASSURANCES

191

NORTH KOREA'S REACTIONS TO GEOPOLITICAL SHOCKS: FOUR KEY WAVES

North Korea has not been able to develop and implement a coherent, longterm strategy with regard to its nuclear policy. The source of this inability has been a series of sudden, traumatic geopolitical events that have forced North Korea to scramble to find a way to maintain regime survival. In doing so, North Korea has become adept at formulating and pursuing short-term tactics to adapt to and survive each new geopolitical shock. In many theoretical analyses of North Korean nuclear policy goals and behavior, there is an implicit assumption that the North Korean leadership is skilled at formulating strategy and pursuing its long-term policy goals by using the threat of the country's nuclear program. An examination of North Korean nuclear behavior during four key geopolitical periods affecting the regime will reveal a state that has been more effective from a tactical standpoint than a strategic one in achieving the immediate goal of regime survival. This North Korean behavior has implications for the applicability of security assurances. Since the establishment of the North Korean state in 1948, it has primarily existed as an aid-recipient country. In this respect, pursuingjuche,3 the concept of self-sufficiency, served as a practical method to conceal the externally dependent nature of the North Korean regime. At the core of North Korea's ability to survive has been the ability to react rapidly following a geopolitical shock to the regime. These shocks can be likened to a tidal wave that leaves North Korea scrambling to secure a replacement aid patron following the loss of the previous one. The tactical manipulation of the opportunities that emerge with each wave, however small, has enabled the regime to ensure its survival from wave to wave. This behavior occurs from a position of weakness, rather than strength. This tactical response behavior will be explored more closely in the next section. In this section, the four key waves that have hit North Korea are identified and the manner in which North Korea used its nuclear policy to react to each wave to ensure short-term survival is described. Wave 1: The Sino-Soviet Split and Rivalry

In the immediate aftermath of the Korean War, the severely weakened North Korean state faced a formidable U.S. military deterrent and a growing South Korean defensive capability. Seeking to maintain a strategic buffer state, the Soviet Union increased its military aid to the remnants of the North

192 JOHN S. PARK

Korean army in the mid-1950s. In the late 1950S, North Korea also received growing military aid from China. The Sino-Soviet split during the following decade became the first geopolitical shock to affect North Korea's approach to nuclear matters. As the Sino-Soviet rivalry deepened in the 1960s, the conventional view is that North Korea benefited immensely by playing one patron off of the other. North Korea, however, was not as masterful as is commonly believed. As a RAND study in October 1984 points out: Although the DPRK's geostrategic importance affords it a certain amount of influence, North Korea has not been able to translate this importance into anything more than very cautious and conditional support from its two principal patrons. This is particularly true concerning matters touching directly upon great power interests. Far from a North Korean tail wagging the Russian or Chinese dog, therefore, North Korea has constantly had to scramble to adjust to policies of the USSR or the PRC which were often adopted for reasons having nothing to do with Pyongyang but which had an important effect upon it. The actual leverage of the DPRK has been extremely limited. 4

For North Korea, the Sino-Soviet rivalry caused frequent disruptions in aid from its patrons. The absence of consistent aid flows made domestic production more attractive to Pyongyang. This is about the time the DPRK initiated its nuclear program, with important technical assistance from the Soviet Union. Not knowing how much it could count on such assistance continuing, North Korea increasingly focused on plutonium-based nuclear technology during this period. North Korea reaped a strong benefit from pursuing the plutonium path-the opportunity to develop local capabilities. In 1962, North Korea began construction of an IRT-2000 research reactor it procured from the Soviet Union. In the early 1970S, North Korea used "indigenous technology to expand the capacity of the IRT-2000 research reactor:'5 Although North Korea periodically received nuclear components, equipment, and training from the Soviet Union, it also indigenized its nuclear program as much as it could. Its graphite-moderated nuclear reactor at Yongbyon, which the DPRK began building in 1979-80, was "built to run on locally mined and milled uranium:'6 North Korea also built a facility at Yongbyon for reprocessing plutonium from the reactor's spent fuel. These efforts to maximize indigenous capacity helped the DPRK weather shifts and disruptions in aid relationships during the period of Sino-Soviet rivalry.

ASSESSING THE ROLE OF SECURITY ASSURANCES

193

Wave 2: South Korea's Nordpolitik

In the late 1980s and early 1990S, the Republic of Korea (ROK) embarked on a highly effective diplomatic campaign in which the South established diplomatic relations with the Soviet Union in 1990 and with China in 1992. Offering lucrative trade and investment incentives to Moscow and Beijing, Seoul drastically altered the traditional patron-client relationships that Pyongyang had with the Soviets and the Chinese. These South Korean diplomatic achievements constituted a major shock to the North Korean regime because they threatened its vital, though limited, connections to the international system. North Korea's sense of betrayal by Moscow and Beijing was compounded by further steep reductions in trade, loans, and aid, especially from the Soviet Union as chronic economic inefficiencies and a debilitating arms race took their toll on the Soviets. Starting in the late 1980s, North Korea's economy experienced dramatic contractions due to shrinking imports and major reductions in exports. Following the collapse of the Soviet Union in 1991 and the end of the Cold War, North Korea was no longer able to enjoy any of the last vestiges of "friendship prices" from or preferential arrangements with its Soviet and Chinese allies. During this period, North Korea increased its nuclear activity at the Yongbyon Nuclear Complex. In 1989, North Korea shut down and refueled its 5 MW(e) reactor. North Korea then reprocessed the spent fuel to weapons-grade plutonium in its Radiochemical Laboratory. At the same time, the end of the Cold War created a possible opening to alter relations with the United States. In late 1991, the United States withdrew its tactical nuclear weapons deployed in South Korea as part of a global policy of reducing its foreign deployments of such weapons. At a time of increased isolation resulting from Seoul's successful Nordpolitik and the collapse of its Soviet patron, North Korea now did something on the diplomatic front it had not done in more than four decades. In January 1992, Kim Yong-sun, international director of the WPK, met with Arnold Kanter, U.S. under secretary of state for political affairs, in New York in the highest level meeting between the United States and North Korea since the Korean War armistice talks.? With U.S. suspicion growing that North Korea was using its nuclear program to develop a nuclear arsenal, Washington focused on getting the North Koreans to sign a safeguards agreement with the International Atomic Energy Agency (IAEA). Although North Korea had acceded to the NPT in 1985, it had not concluded an IAEA safeguards agreement as required by the treaty. Such an agreement was needed to facilitate IAEA inspections of the suspicious ac-

194 JOHN S. PARK

tivities taking place at the Yongbyon facilities. Prior to the New York meeting, President George H. W. Bush's administration was deeply divided about even convening a meeting with a country that the United States had long held to be untrustworthy. 8 On the eve of this meeting, Nordpolitik's shadow loomed large. Seoul's innovative policy had induced a drastic reduction in North Korea's remaining commercial ties with its communist allies. The disintegration of the Soviet Union accelerated and compounded this trend. The loss of key patrons dealt a heavy blow to the viability of the aid-dependent North Korean economy. Opportunities to engage the United States offered a counterbalance to this dire situation. The Kanter-Kim meeting initiated an on-again, off-again interaction between Pyongyang and Washington that has persisted. From the U.S. point of view, the Kanter-Kim meeting achieved its immediate objective, as it persuaded North Korea to accept IAEA safeguards. The resulting inspections, however, set the stage for a new nuclear crisis. Wave 3: Collapse of the Agreed Framework

After IAEA inspectors questioned North Korea's initial declaration of its past nuclear activities, a serious nuclear crisis unfolded in 1993-94. Efforts to resolve the crisis led to Pyongyang's first significant engagement with the United States, culminating in the Agreed Framework. In this accord, signed in October 1994, the North Korean leadership agreed to a gradual, step-by-step approach intended to ultimately lead to a nuclear weapons-free Korean Peninsula. This would begin with an immediate freeze on work at the Yongbyon Nuclear Complex. In return, North Korea would obtain deliveries of fuel oil, the construction of two light water reactors (LWRs), and at the end of the process normalized ties with Washington. When the Agreed Framework unraveled instead, this became a third geopolitical shock to the North Korean regime. From Pyongyang's perspective, the gradual format of the Agreed Framework was ideal for enabling the regime to test U.S. intentions. Rather than immediate nuclear dismantlement, a series of steps had to be taken for that end goal to be achieved. Should the relevant end goals for North Korea not be delivered-that is, the completion of the LWR project and normalized relations-then the nuclear freeze implemented at Yongbyon could be undone. In the end, that is what took place. Even while the Clinton administration remained in office, implementation of the Agreed Framework proved slow and incomplete. This resulted initially

ASSESSING THE ROLE OF SECURITY ASSURANCES

195

from congressional skepticism and reluctance to fund the deal, later reinforced by increasing administration suspicions of North Korea's intentions. The remaining vestiges of the Agreed Framework unraveled during the first term of the George W. Bush administration. Having publicly stated during the presidential campaign that he "despised and loathed" North Korean leader Kim Jong-il, one of Bush's first actions as president was ordering a thorough review of America's North Korea policy.9 As a result, while the heavy fuel oil shipments initially continued, the Bush administration effectively halted the reactor project, which was at the heart of the Agreed Framework. In a meeting in fall 2002, described more fully below, the administration also challenged North Korea with evidence of a secret uranium enrichment project, which it deemed a violation of the Agreed Framework by the DPRK. With these developments, the Agreed Framework collapsed. North Korea responded by reprocessing the 8,017 fuel rods that had been sitting in storage ponds and restarting its 5 MW( e) reactor. Wave 4: Advent of the Bush Doctrine

Overlapping with the collapse of the Agreed Framework, a fourth shock for North Korea emerged from the aftermath of the 9/11 attacks when the Bush administration began to speak openly about taking more aggressive action to eliminate possible future threats from so-called rogue states. In January 2002, President Bush named North Korea as part of an "axis of evil:' The following year, the United States invaded Iraq, another member of the axis of evil, with the avowed goal of regime change. Washington sought to use the new preemption-centered Bush Doctrine to increase the coercive pressure on the Kim Jongil regime. At the same time, the first-term Bush administration set stricter conditions for its participation in future negotiations with Pyongyang. The administration refused to enter into substantive negotiations with the Kim Jong-il regime until complete, verifiable, irreversible disarmament (CVID) had been confirmed in North Korea. Pyongyang insisted on the precise opposite-that is, negotiations and concessions first, CVID later. Although the North Koreans and Americans held talks in a trilateral venue hosted by the Chinese in late April 2003 to end the nuclear deadlock, the official North Korean and U.S. statements indicated that the above policy stances largely remained unchanged. lO North Korea's resumption of work in spring 2003 at its plutonium production facilities can be seen as much as a response to these broader shifts in U.S.

196 JOHN S. PARK

doctrine and negotiating terms as to the collapse of the Agreed Framework. North Korea's nuclear activities were not simply a national defense measure, however, but also a part of the North's bargaining tactics as it dealt with the four geopolitical shocks described above. The next section analyzes the North's bargaining in each of the four waves. FOUR NORTH KOREAN RESPONSES

In each of these waves, North Korea used its nuclear program for one or both of two purposes: to achieve deterrence objectives and/or to bargain for a particular short-term outcome in order to ensure regime survival in the immediate aftermath of the geopolitical shock. In pursuing these goals, North Korea has benefited from divergence in the responses of affected countries in the wake of each wave. It has been able to calibrate and tailor its tactical techniques to each situation and country. Adapting to the Sino-Soviet Split and Rivalry

During the period of the Sino-Soviet split and the ensuing rivalry in the 1960s, North Korea bargained with the Soviets to secure the components, equipment, and training required for developing a nascent nuclear program. At the same time, the Soviets increased their acquisition of North Korean deposits of uranium and monazite that contributed to the rapidly growing Soviet nuclear industryY These negotiations transpired against the background of North Korea's first ever positive security assurance-embodied in the newly agreed Soviet-DPRK Treaty of Friendship, Cooperation and Mutual Assistance in 1961. As the Sino-Soviet rivalry grew, North Korea could no longer rely on stable sources of aid and assistance. An example of this recurring predicament was Soviet Premier Nikita Khrushchev's decision to decrease Soviet military and economic aid to North Korea in the early 1960s. Soviet concerns that Kim Il-sung was becoming "a pro-Chinese Maoist" contributed to doubts about providing nuclear technology to a suspect North Korean ally. As a result, Moscow sought to reduce nascent Soviet-North Korean nuclear cooperationY As the schism between Beijing and Moscow deepened, North Korea began a concerted effort to explore ways to make its nuclear program more indigenous. While the split created some limited opportunities for Pyongyang, it also revealed North Korea's vulnerability to patrons whose focus and aid shifted depending on larger geopolitical movements. An indigenous, juche-oriented nuclear program and infrastructure hence became a priority for the North Ko-

ASSESSING THE ROLE OF SECURITY ASSURANCES

197

rean leadership.13 Given North Korea's initial geostrategic value to the Soviets as the Cold War raged in the early 1960s and its abundant deposits of uranium ore and monazite, Pyongyang was eventually able to bargain with Moscow for the nuclear assistance required to establish and develop its nuclear program. Surviving South Korea's Nordpolitik

When South Korea effectively established diplomatic ties with Communist bloc countries under Nordpolitik, North Korea's growing diplomatic isolation created panic in Pyongyang. The North Korean leadership became so alarmed because of the significant policy concessions South Korea elicited from the North's chief allies. In return for a U.S.$3 billion trade credit and direct aid package to Moscow, Seoul gained an enormous diplomatic and security payoff. The Soviets pledged to support South Korea's admission to the United Nations, agreed to cease supplying North Korea with offensive weapon systems, and most significantly, guaranteed that it would halt all assistance to North Korea's nuclear program. 14 The landmark visit by Soviet leader Mikhail Gorbachev to Beijing in May 1989 compounded the impact of Nordpolitik on North Korea. The formal restoration of party-to-party relations between the two countries brought an end to the Sino-Soviet split. As a direct result, the "value of North Korea as a strategic asset to both China and the Soviet Union" dropped significantly. IS Pyongyang's critical ability to manipulate the rivalry between its titan patrons had evaporated. In this radically transformed international landscape, Pyongyang began displaying strong interest in what it could receive from Washington in return for bargaining away its nascent nuclear weapons program. Having increased its nuclear activity and production of weapons-grade plutonium in the late 1980s and early 1990S, North Korea sought to use its enlarged nuclear program as a bargaining chip sufficient to garner massive concessions-particularly normalized ties with Washington-that would facilitate peaceful coexistence. 16 As U.S. demands grew for greater North Korean disclosure in the wake of the IAEA discovery of discrepancies in Pyongyang's initial declaration to the Agency, tensions rose significantly on the Korean Peninsula. The DPRK's attempt to bargain with Washington during this tense period was initially ineffective because the Clinton administration remained adamant about Pyongyang's adherence to its international nonproliferation regime obligations. By mid-1994, Pyongyang began implementing a blackmail tactic whereby

198 JOHN S. PARK

it began to unload the spent fuel rods from its 5 MW(e) reactor and extract nuclear material for reprocessing. In May 1994, the IAEA confirmed that Pyongyang's removal of spent fuel took place in the absence of international monitors. In defiance, North Korea announced on June 13, 1994, that it was withdrawing from membership in the IAEA. In the aftermath of these actions IAEA inspectors left North KoreaY By increasing the stakes, Pyongyang tried to force Washington into nuclear negotiations. Rather than bring Washington to the negotiating table, however, Pyongyang's nuclear blackmail backfired and led to the referral of the nuclear inspection issue to the UN Security Council. North Korea's nuclear brinkmanship and the U.S. demands for full compliance with IAEA safeguards eventually led to a standoff; Washington came within hours of authorizing U.S. military mobilization for the Korean theater of operations. An eleventh-hour meeting in Pyongyang between former President Jimmy Carter and North Korean leader Kim II-sung narrowly averted open hostilities between the United States and North Korea. Carter's intervention represented a lucky break for North Korea, not a success for its bargaining tactics. In the ensuing months of intense nuclear talks in Geneva and New York, the North Koreans sought a grand bargain with the Americans. 18 The outcome of these arduous talks was the 1994 Agreed Framework, described above. For Pyongyang, the promise of eventual normalized relations with the United States was an especially prized part of the deal. Adapting to a Post-Agreed Framework Order

With the sweeping electoral victory of the Republican Party in November 1994, powerful opposition to the Agreed Framework solidified in the U.S. Con-

gress and implementation of the nuclear agreement began to encounter major delays. A key component of the deal was the annual delivery to North Korea of 500,000 tons of "heavy oil for heating and electricity generation:'19 U.S. delays in supplying the heavy fuel oil began to arouse considerable alarm in Pyongyang, as did the lack of progress in upgrading bilateral ties. North Korea's concern about the accord grew as more delays emerged regarding the LWR project turnkey components and the related construction timetable. As setbacks in the Agreed Framework's implementation increased, Pyongyang embarked on a clandestine uranium enrichment program. 20 U.S. intelligence pointing to the existence of a highly enriched uranium (HEU) program reignited nuclear tensions with Washington. Almost eight years after the sign-

ASSESSING THE ROLE OF SECURITY ASSURANCES

199

ing of the Agreed Framework, James Kelly, assistant secretary of state for East Asian and Pacific affairs, arrived in Pyongyang in October 2002 for the first U.S.-North Korean meeting since President George W. Bush's electoral victory in 2000.21 In his meeting with then First Vice Foreign Minister Kang Sok-ju, Kelly confronted his host with U.S. intelligence data reportedly showing that the North Koreans had been cheating on the previous nuclear accord. Caught by surprise by Kelly's allegations that North Korea had violated the Agreed Framework by pursuing a uranium-enrichment program, Kang is reported to have asserted North Korea's right under the NPT to conduct such activity. In addition to asserting this right, Kang warned that North Korea had "more powerful things as well:'22 In a repetition of previous crises, North Korea had again been caught cheating, just as it had in the early 1990S when it submitted an initial inventory declaration to the IAEA that understated the amount of plutonium North Korea had reprocessed. The Kelly-Kang meeting initiated a series of events that ended with the demise of the Agreed FrameworkY In the immediate aftermath of the collapse of the Agreed Framework, Pyongyang began denying the existence of an HEU program and called for the resumption of the 1994 nuclear accord's implementation. Concluding that Pyongyang had lied and cheated on its previous agreement, the Bush administration demanded CVID prior to any negotiations about concessions to North Korea. Pyongyang responded by expelling IAEA inspectors in December 2002, then fully reprocessing spent fuel rods at its Yongbyon storage site and restarting its 5 MW(e) reactor. In January 2003, the DPRK announced its withdrawal from the NPT. North Korea had returned to nuclear bargaining with the United States. Using the Bush Doctrine's Rigidity to Gain Seoul's and Beijing's Assistance

In the post-9/11 era, North Korea's nuclear bargaining tactic produced a significantly different response from the United States than it had in the early 1990S. This time, the North's nuclear activities did not create a sense of urgency to broker a deal with Pyongyang and preserve the nonproliferation regime. Instead, the United States had developed a new approach to dealing with proliferation threats, a posture that became known as the Bush Doctrine. Unveiled by President Bush in June 2002, the Bush Doctrine emphasized preemptive action to disarm "dictators with weapons of mass destruction [who 1can deliver those weapons on missiles or secretly provide them to terrorist allies:'24

200

JOHN S. PARK

Rather than back down, Pyongyang chose to raise the nuclear stakes. At a U.S.-DPRK-PRC trilateral meeting in Beijing in April 2003, North Korean negotiator Li Gun privately informed Assistant Secretary of State Kelly that North Korea possessed nuclear weapons. This was likely intended in part to serve a deterrent purpose against any contemplated U.S. preemptive attack. In addition, however, Li reportedly threatened that North Korea may "choose to export or 'demonstrate' them and would base its decision on whether Washington meets a slew of demands, including lots more aid:'25 During this period of escalating nuclear tensions, North Korea's nuclear bargaining tactic had a profound impact on its neighbors to the south and the north. From Seoul's and Beijing's viewpoint, the increasing possibility that North Korea's escalation of tensions with Washington would trigger unilateral U.S. action made the North's nuclear threats a source of grave concern. Each had a major stake in preserving stability in North Korea-for Seoul, doing so is a prerequisite to implementing its plan for gradual economic integration and eventual reunification, whereas for Beijing, preserving stability prevents what would be a serious disruption to its internal economic development efforts. For both South Korea and China, the situation in North Korea had become, in many ways, an extension of domestic policy.26 As a result, both Seoul and Beijing intensified their diplomatic efforts to stabilize the nuclear crisis. For Seoul, its priority was rapidly assembling a large aid package to entice Pyongyang away from its nuclear confrontation with Washington. The Bush Doctrine alarmed Seoul as much as it did Pyongyang. Soon after his February 2003 inauguration, President Roh Moo-hyun had initiated a more proactive ROK role in inter-Korean relations to directly counter Washington's hard-line stance. In sharp contrast to Washington's refusal to negotiate with Pyongyang, Roh sent his senior envoys to meet with the North Korean leadership. Roh's envoys reportedly conveyed detailed plans for unprecedented inter-Korean economic relations with "aid on a massive scale" if North Korea were to work toward a resolution of the nuclear crisisY After the April bombshell, Seoul increased these efforts to entice Pyongyang away from provoking a national security-obsessed United States. In Beijing's eyes, the danger posed by North Korea's nuclear weapons program was more symbolic than actual. Chinese officials argued that even a North Korean nuclear test would not equate to weaponization because Pyongyang had not achieved the ability to mate a miniaturized warhead to a proven delivery system. 28 Based on the developmental experiences of the nuclear weapon states,

ASSESSING THE ROLE OF SECURITY ASSURANCES

201

North Korea would require multiple nuclear tests to further refine the potency of its program. Chinese officials were skeptical about Washington's assertion that North Korea had a near-deployable nuclear arsenaP9 Concerned that Washington had overstated its threat assessment of North Korea's nuclear capability, Beijing responded by increasing its multilateral diplomatic efforts to deal with the nuclear crisis. These efforts culminated in the first round of Six-Party Talks in Beijing in August 2003 with representatives from the United States, China, North Korea, South Korea, Japan, and Russia. The Bush administration agreed to participate, despite its deep skepticism about North Korea, because it preferred multilateral talks to the direct bilateral talks sought by Pyongyang. It also thought that North Korea's regional neighbors would have to be involved if the effort to apply pressure on the DPRK regime were to be successful. During this period, North Korea's short-term needs for regime survivalfood aid, energy supplies, and economic assistance-were disproportionately met by China, with additional support from South Korea. While directed at Washington, North Korea's nuclear bargaining tactic yielded enormous aid and de facto political protection from Beijing and Seoul-for the latter, to the detriment of its alliance with the United States. Taking full advantage of the major policy differences among China, South Korea, and the United States, North Korea's nuclear bargaining tactic enabled the Kim Jong-il regime to once again turn an extremely weak hand into a highly effective formula for short-term survival. Cognizant that Washington's rigid CVID policy stance would prevent resumption of the LWR project or heavy fuel oil shipments, Pyongyang focused its attention elsewhere. This review of four geopolitical shocks and North Korea's responses demonstrates how Pyongyang reacted with tactical behavior aimed at ensuring regime survival in the short run. North Korea has used its nuclear weapon program as part of its bargaining tactics in this context. In addition, the DPRK has seen nuclear weapons as a deterrent to possible threats from the United States. Given these motivations for the nuclear program, to what extent have security assurances played a role in moderating Pyongyang's pursuit of nuclear weapons? NORTH KOREA'S MAIN SOURCES OF SECURITY ASSURANCES

North Korea's nuclear decision-making and reactions to key geopolitical events have taken place against a backdrop of changing security assurances. Table 9.1 summarizes the main sources of North Korea's security assurances.

202

JOHN S. PARK

TABLE 9.1

North Korea's Main Sources of Security Assurances Year

Source

Type

1961

1994

Soviet-DPRK Treaty of Friendship, Cooperation & Mutual Assistance Sino-D PRK Treaty of Friendship, Cooperation & Mutual Assistance U.S.-DPRK Agreed Framework

2000

U.S.-DPRK Joint Communique

2005

Six-Party Talks Joint Statement of Principles

Positive Security Assurance Positive Security Assurance Negative Security Assurance Reassurance (no "hostile intent") Negative Security Assurance

1961

They include two treaties, an agreement, a joint communique, and a joint statement that involve the Soviet Union, China, and the United States at different points in time, and cover both positive and negative forms of assurance. Pyongyang's Treaties with Moscow and Beijing: Key Sources of Positive Security Assurances

For much of North Korea's history, the Soviet-DPRK Treaty of Friendship, Cooperation and Mutual Assistance (1961) and the Sino-DPRK Treaty of Friendship, Cooperation and Mutual Assistance (1961) constituted the two primary sources of security assurances for North Korea. Along with other provisions, both treaties included mutual defense clauses committing one party to aid the other if it were attacked. While both provided a positive security assurance to Pyongyang, the former represented the larger pillar in buttressing North Korea's economy and security. In the late 1980s, Seoul embarked on its highly effective Nordpolitik campaign that culminated in the establishment of diplomatic relations with the Soviet Union in 1990 and China in 1992. By offering massive trade and investment incentives to Moscow and Beijing, Seoul altered traditional patron-client ties on which Pyongyang depended. In September 1990, Soviet Foreign Minister Eduard Shevardnadze flew to Pyongyang to inform the North Koreans of imminent Soviet-ROK diplomatic normalization. Although the 1961 treaty still remained in effect, this effectively marked the end of a comprehensive economic, political, and security relationship that Pyongyang had had with Moscow for decades. After the breakup of the Soviet Union, the Russian Federation proved unwilling to maintain all of the commitments of its Soviet predecessor to the

ASSESSING THE ROLE OF SECURITY ASSURANCES

203

DPRK. In September 1995, the Russian Foreign Ministry informed North Korea that Moscow would not renew the 1961 Soviet-North Korean treaty beyond its September 1996 termination date, and instead proposed a new, downgraded treaty. Moscow considered Article I of the 1961 treaty, which had obliged the Soviet Union to provide military assistance in defense of North Korea, to be "non -operative:'30 Although not as significant, the Sino-DPRK treaty provided another source of a positive security assurance to Pyongyang. Like the Soviet-DPRK treaty, the Sino-DPRK one provided a critical-though not consistent-lifeline for Pyongyang after it came into effect in the early 1960s. After Beijing's establishment of bilateral relations with Seoul in 1992 and the end of Moscow's pledge to come to the defense of North Korea in 1996, Pyongyang's last remaining positive security assurance largely existed only on paper. Growing economic interdependence between China and South Korea, as well as between China and the United States, greatly reduced the likelihood that China would come to the defense of its paper ally. China prioritized preventing instability on the Korean Peninsula that could jeopardize Chinese internal economic development goals. After a series of North Korean acts of brinkmanship in the mid-2000s, Beijing warned that should North Korea provoke a retaliatory attack by South Korea or the United States, China would not come to the assistance ofPyongyang. 31 Beijing also began a concerted effort to negotiate a revision of the mutual defense clause of their 1961 treaty-a change that Pyongyang has repeatedly avoided. Although Beijing initiated a high-level effort in October 2005 to stabilize North Korea by offering assistance with and incentives for undertaking economic reform, Pyongyang did not fully engage. Pyongyang appeared instead to focus its attention on acquiring a nuclear deterrent capability. In October 2006, it conducted its first nuclear test. In a period in which it lacked a functioning positive security assurance, North Korea declared itself a nuclear power and began negotiating in a more strident manner in both Six-Party Talks and bilateral contexts. A second nuclear test followed in May 2009. That these moves to achieve nuclear weapon status occurred at a time when North Korea lacked a comprehensive security, economic, and diplomatic relationship with any external power is quite suggestive. This focus on comprehensive relationships underscores that the 1990S were North Korea's most challenging decade. In 1996, North Korea lost its main source (the former Soviet Union) of a positive security assurance and experienced a major devaluation of the remaining source (China). Compound-

204 JOHN S. PARK

ing this situation was the disappearance of the last vestiges of North Korea's preferential trading arrangements with its communist patrons. While the 1961 Soviet-DPRK defense pact operated, a nuclear weapon state (Soviet Union) offered a credible positive security assurance to a non-nuclear weapon state (North Korea). This positive assurance was not fully effective, as the DPRK made concerted efforts to develop indigenous nuclear capabilities in the 1980s, well before the Soviet collapse. Yet North Korea did not cross the threshold to testing a nuclear device while the assurance remained in place. The decline and termination of this external security guarantee prompted increased interest in North Korea in developing its own nuclear deterrent. North Korea accelerated its nuclear weapons program during a period in which North Korea sought to fill a positive security assurance void that essentially had begun in the early 1990S. The U.S.-DPRK Agreed Framework: Pyongyang's First Negative Security Assurance

The United States is the only nuclear weapon state regarded as a threat by North Korea. The decline and loss of positive security assurances from its allies made a negative assurance from its main potential adversary a possible alternative. The United States has since offered several assurances, starting with the negotiations leading to the Agreed Framework. In talks in June and July 1993, the United States agreed to a set of principles for a deal that included a negative security assurance. This was then formally incorporated into the Agreed Framework. With respect to the security assurance, the agreement stipulated that "the U.S. will provide formal assurances to the DPRK, against the threat or use of nuclear weapons by the U.S:'32 This negative security assurance served as an important nascent confidence-building measure for two countries that were accustomed to a highly adversarial relationship. During this period, North Korea froze its nuclear activities at its Yongbyon reactor and reprocessing facility. It also suspended construction of larger reactors that would have been able to produce more plutonium. The Agreed Framework supplemented the negative security assurance with positive incentives, including energy assistance. As part of this, the United States led an international consortium tasked with building two LWRs in North Korea. The Korean Peninsula Energy Development Organization (KEDO) fostered the development of core working partnerships among North Koreans, South Koreans, and Americans. Despite chronic delays in implementation, the

ASSESSING THE ROLE OF SECURITY ASSURANCES

205

KEDO project facilitated many formative experiences of cooperation among traditional adversaries, until November 2002, when the Bush administration convinced KEDO members to halt further work on the LWR project. The U.S.-DPRK Joint Communique: No "Hostile Intent"

U.S.-DPRK tensions rose again in the late 1990S mainly as a result of North Korea's missile development activities, specifically the 1998 launch of a longrange missile on a flight trajectory over Japan. After the missile flight-test (described as a satellite launch by North Korea), President Clinton appointed former Defense Secretary William Perry to lead a review of u.S. policy toward North Korea. Upon completion of the review in fall 1999, the Clinton administration re-engaged North Korea based on the recommendations of the Perry Report.33 Bilateral diplomatic initiatives led to an unprecedented meeting on October 10, 2000, in the White House between President Clinton and Kim Jong-il's special envoy, Vice Marshall Jo Myong-rok. It resulted in a joint communique that included a key statement: ''As a crucial first step, the two sides stated that neither government would have hostile intent toward the other and confirmed the commitment of both governments to make every effort in the future to build a new relationship free from past enmity:'34 The joint communique effectively restarted U.S.-DPRK negotiations. The Clinton-Jo meeting laid the foundation for Secretary of State Madeleine Albright to hold meetings with Kim Jong-il during October 23-24, 2000, in Pyongyang. This period of bilateral negotiations, however, did not last long. The electoral change from the Clinton administration to the George W. Bush administration in January 2001 highlighted to the North Koreans the lack of continuity in U.S. interactions with Pyongyang. The joint communique was an example of reassurance, which Knopf describes as "a type of negative assurance:'35 It was not explicitly nuclear-related, but instead sought to convey a more general absence of "hostile intent:' Interestingly, the reassurance was bidirectional in this case, indicating that both sides perceived a potential threat from the other. The Six-Party Talks Joint Statement of Principles: A Reaffirmed Negative Security Assurance

Following the collapse of the Agreed Framework and hardening of U.S.North Korea relations in President Bush's first term, his second term witnessed a new attempt at engagement using the Six-Party Talks. The fourth round of talks produced an agreement on principles intended to make possible North

206

JOHN S. PARK

Korea's denuclearization. Concluded on September 19, 2005, the Joint Statement of Principles also contained a negative security assurance: "The United States affirmed that it has no nuclear weapons on the Korean Peninsula and has no intention to attack or invade the DPRK with nuclear or conventional weapons:' The Joint Statement also contained a more general statement of reassurance as well: "The DPRK and the United States undertook to respect each other's sovereignty, exist peacefully together, and take steps to normalize their relations subject to their respective bilateral policies:'36 The negative security assurance in the Joint Statement of Principles is useful in tracking North Korea's shifting policy statements and preferences. Despite seeking this type of security assurance from the United States since the early days of the first George W. Bush administration, North Korea showed a lack of interest almost as soon as it received it. In subsequent U.S.-DPRK talks, it became clear to the United States that North Korea wanted a "regime security assurance" from Washington. 37 The specific details and implications of this North Korean demand will be assessed below. The Obama Administration's Nuclear Posture Review

The Six-Party Talks produced two agreements in 2007 on steps to implement the joint principles agreed to in September 2005. A final deal to complete denuclearization of the Korean Peninsula could not be reached, however, before the Bush administration left office. North Korea then presented the new Obama administration with a series of new challenges, including a long-range missile test in April 2009 and its second nuclear test in May 2009. The Obama administration has not offered a new security assurance specifically focused on North Korea, thus leaving the 2005 Joint Statement as the last security assurance concluded in talks with the North Korean government. The Obama administration did, however, attempt to gain traction from a negative security assurance in a different context. In April 2010, the administration released its Nuclear Posture Review (NPR). The NPR revised the existing generalized negative security assurance that had been U.S. policy since 1978 to remove an exception for states allied with a nuclear weapon state. The NPR stated that the United States would not threaten or use nuclear weapons against any non-nuclear weapon state that is both party to the NPT and in compliance with its NPT obligations. 38 Administration officials, including Defense Secretary Robert Gates, explained publicly that one goal of the revised negative security assurance was

ASSESSING THE ROLE OF SECURITY ASSURANCES

207

to provide an incentive for North Korea, and also Iran, to give up their nuclear weapons programs and come into full compliance with the NPT. North Korea quickly dismissed the new NPR. A DPRK foreign ministry spokesperson claimed that the new U.S. policy meant North Korea was still subjected to nuclear threats and was therefore no different from the Bush administration's preemption doctrine. Pyongyang also complained that the NPR represented a backtracking on the negative security assurance of the 2005 Joint Statement. 39 Although it will be difficult to make any security assurance effective, North Korea's response to the 2010 NPR suggests that it is likely to prefer a bilateral assurance focused explicitly on its concerns over a more generalized NPT-related assurance. TESTING HYPOTHESES ON SECURITY ASSURANCES

The preceding analysis of North Korean behavior at least partially sustains Knopf's initial hypotheses 4,7, 8, 9, 10, l2, and 13 regarding security assurances, though it also suggests possible modifications or caveats to some of these.

Hypothesis 4: The greater the political and economic ties between sender and recipient, the more effective security assurances will be. For much of its history, North Korea relied upon positive security assurances from the Soviet Union and, to a lesser extent, China. Pyongyang seemed to regard those assurances as most credible in the periods when it had close political ties and extensive economic relations with its communist allies. When those allies reduced their economic assistance or adjusted their diplomacy to more favorable relations with South Korea, North Korea began to invest more seriously in developing its own nuclear capabilities. When it came to judging the credibility of assurances, North Korea took its cues less from public declarations, legally binding commitments, and various forms of defense cooperation than it did from the overall quality of its political and economic relationships with its main patrons. Hypothesis 7: Assurances will be most effective if they are tailored to take account of unique features of the target state's culture, decision-making procedures, and leadership concerns. The 1961 Soviet-DPRK treaty provided a potent security assurance to North Korea during most of the period in which it operated. Given that the North Korean state system was essentially modeled on the Soviet one, Soviet officials could tailor their assurances to fit the circumstances of North Korea as the re-

208

JOHN S. PARK

cipient. As long as the Soviet Union existed, this security assurance was at least moderately effective. This situation changed dramatically as the Soviet Union began to buckle in the late 1980s. With the dissolution of the Soviet Union in 1991, North Korea's sole functioning security assurance became meaningless. As a result, North Korea sought a new security assurance as part of a package deal tailored to the Kim II-sung and then Kim Jong-il regime. This period saw the advent of the 1994 Agreed Framework and the repeated phases of stop-and-go implementation. Given that the Clinton administration had been on the brink of military action before the Carter trip in June 1994, a negative security assurance from the United States was certainly relevant to Pyongyang's concerns. Following the emergence of the Bush Doctrine, the DPRK again sought a negative security assurance from the United States. This time, however, the security assurances contained in the September 2005 agreement were not effective-and neither was the general NPT-related negative assurance contained in the Obama NPR. These nuclear-focused assurances did not address Pyongyang's larger concerns about regime survival and hence were not appropriately tailored from Pyongyang's perspective.

Hypothesis 8: Assurances will have to be strong enough to overcome cognitive biases in order to be effective. Embedding them in a larger strategy may be one way to do this. Recent developments in China's approach suggest there may be value in embedding traditional security assurances in a broader strategy, although the reasons for this are not related to the possibility that cognitive factors will lead a recipient to discount the sincerity of assurances from an adversary. Although the 1961 Sino-DPRK treaty endures, Beijing now keeps the security assurance clause ambiguous. In research interviews in March 2010, prior to North Korea's sinking of the South Korean naval vessel Cheonan and its shelling of Yeonpyeong Island, Chinese government policy analysts had explained that Beijing hopes to derive two benefits from this ambiguity.40 The first is that it may deter a major North Korean provocation against the United States or South Korea. If North Korea is uncertain about Chinese security assistance in the event its provocation triggers a military confrontation, then Pyongyang may refrain from committing the provocation. The second benefit is that Beijing is able to downplay any vestiges of the early Cold War period when relations between China and North Korea were described as being as close as lips to teeth. It is

ASSESSING THE ROLE OF SECURITY ASSURANCES

209

not known as of this writing whether China is reassessing these expectations in light of North Korean provocations during 2010. China's overall strategy, however, has incorporated a positive component that is separate from military security concerns. Since the mid-2000s, the CPC has sought to deepen its working relationship with the WPK. Embedding a security assurance in Beijing's overall approach, which involves seeking a comprehensive relationship with Pyongyang through the CPC-WPK channel, constitutes what senior CPC leaders perceive as a winning strategy. In this case, the security assurance is a narrowly focused positive assurance-that is, the CPC will work to ensure the survival of the WPK. The broader comprehensive strategy is designed to further China's primary goal of sustainable economic development by effectively addressing a chronic instability factor in Northeast Asia-North Korea. By building up the WPK's capabilities, the CPC seeks to create a viable organization in North Korea that can competently carry out economic reform and opening. If adding political and economic assistance to traditional security assurances proves effective, the reasons would be different from those hypothesized by Knopf, which focused on overcoming psychological biases. But the larger point of hypothesis 8-that assurances may be made more effective by embedding them in a larger strategy-would be supported.

Hypothesis 9: Assurances can fail if domestic or alliance constraints limit the ability of a sender or senders to fully implement an assurance strategy. In the case of the Soviet-DPRK mutual defense treaty, North Korea's major security assurance failed because of the ultimate domestic constraint-the sender peacefully collapsed and was no longer able to implement an assurance strategy. While the official end of this assurance strategy is marked as 1996 when Russia allowed the treaty to expire, it ended, in effect, with the disintegration of the Soviet Union in 1991. Negative security assurances offered by the United States have also faced constraints. In the 1990S, Congress proved reluctant to fully fund U.S. commitments under the Agreed Framework. After the conservative candidate Lee Myung-bak became president of South Korea in February 2008, his proposed CVID-based Grand Bargain with North Korea limited, in practice, how far the Bush and Obama administrations could press ahead diplomatically.41 These domestic and alliance constraints did not prevent the United States from offering negative security assurances, but they did hamper the potential U.S. ability to conclude and implement larger package deals in which assurances were embedded.

210

JOHN S. PARK

Hypothesis 10: States with regional security concerns will be most interested in positive security assurances; states without such concerns will be most interested in negative assurances. For North Korea, the greatest security concern is the United States, not South Korea or other regional neighbors. This has led to North Korean demands for negative assurances, as the hypothesis suggests, yet this hypothesis needs qualification. While the common perception in policy circles in Washington is that North Korea seeks a negative security assurance from the United States-that is, a nonaggression agreement from the United States that it will not attack North Korea-a close reading of the demands and offers North Korea made during U.S.-DPRK talks in the second Bush administration paints a different picture. Then Assistant Secretary of State for East Asian and Pacific Affairs Christopher Hill noted that in his negotiations with the North Koreans in 2006-7, his counterparts, citing the "U.S. hostile policy:' pointed out that should Washington commit to the survival of the North Korean regime there would be no need for a North Korean nuclear arsenal. Thus, although North Korea is interested in a negative assurance, what it wants goes beyond an assurance against the use of nuclear weapons or even against conventional military action. It wants a guarantee the United States will not do anything that might undermine the regime in Pyongyang. The fundamental problem with this demand is that there is no possibility that any Democratic or Republican administration would be willing or able to guarantee the survival of the North Korean regime. Even if North Korea were to peacefully denuclearize, as long as it maintains its system of governance, the host of human rights issues it poses would prevent any form of guaranteeing the survival of the regime, let alone official recognition of the North Korean government.

Hypothesis 12: Assurances will be more effective if they can be utilized in a way that alters internal debates in the target in a favorable direction. This hypothesis cannot be confirmed based on what has happened to date, but it is part of the thinking guiding Chinese strategy. The CPC's main objective is to persuade North Korea to adopt the path of economic reform and opening. Assurances from Beijing through the CPC-WPK channel are intended to facilitate capacity-building in the ranks of the WPK. By doing so, Beijing seeks to alter internal debates in Pyongyang. The CPC's hope is that the North Korean leadership will gradually move away from nuclear weapons development as it achieves sustainable progress in its economic reform efforts. Overall, Beijing has been working to create a stable security environment on the Korean Peninsula conducive to major economic reform and development activities.

ASSESSING THE ROLE OF SECURITY ASSURANCES

211

Hypothesis 13: Assurances will be more effective if they are packaged with positive incentives.

This would appear to be the lesson of the Agreed Framework. Although there is much debate about whether the 1994 deal should be considered a partial success or purely a failure, it is certainly the case that North Korea restrained its nuclear behavior more while the Agreed Framework remained in effect than it has since. For eight years, North Korea froze its plutonium activities and maintained its memberships in the NPT and IAEA. And the Agreed Framework was above all a package deal, combining its negative security assurance with economic and diplomatic incentives. China's recent strategy reflects a similar calculation. In addition to the CPC's specific use of assurances to alter internal debates in North Korea to prioritize economic reform, it also seeks to combine its assurances with positive incentives. These incentives come in the form of economic development projects, predominantly in mineral resource-rich North Hamgyong province bordering China's Jilin province. 42 With respect to Beijing's economic engagement with North Korea, Chinese government policy analysts point out that this Chinese policy is largely modeled on the sunshine policy pursued by the progressive South Korean governments of presidents Kim Dae-jung and Roh Moo-hyun during most of the 2000S. Beijing deems sustained economic engagement as the most effective way to facilitate substantive economic transformation inside North Korea. Over time, this Chinese policy is intended to foster a strategic North Korean reorientation away from nuclear weapons and toward economic reform and opening in order to bolster the long-term viability of the North Korean regime. Cognizant of how long a time commitment this endeavor will require and the likelihood of multiple challenges along the way, Beijing is willing to apply its own version of strategic patience. 43 Chinese analysts note that Beijing feels confident that it possesses the one factor crucial for success of the sunshine policy that the progressive South Korean governments ran out of-time. 44 CONDITIONS UNDER WHICH SECURITY ASSURANCES MAY BE EFFECTIVE IN DEALING WITH NORTH KOREA The Chinese Package Deal-A Winning Combination?

Based on what North Korea demanded in negotiations with the United States during the second Bush administration, a picture of a different potential package deal emerges with Pyongyang's insistence on tangible assurances for

212

JOHN S. PARK

DPRK regime survival at the center, but with China rather than the United States as the package provider. Since the mid-2000s, concerted Chinese interactions with North Korea through the CPC-WPK channel show how the Chinese have been seeking to incentivize North Korea to accept their version of a package deal that encompasses DPRK regime survival and economic development assistance from China first and North Korea's full denuclearization later. Although the Chinese have expressed a willingness to deepen the SinoDPRK relationship, Pyongyang has stated its preference for Washington as the provider of a unique regime security assurance. However, after North Korea's relations with both the Obama and Lee Myung-bak administrations stalled, the pace of deal making in the CPC-WPK relationship accelerated. Chinese support for North Korea's regime security appears to have deepened with Beijing's increased assistance to the Kim Jong-il regime via the party channel during an accelerated DPRK leadership succession process. 45 In examining China's progress in dealing with North Korea, it is clear how Beijing's tailored security assurance-specifically, a unique regime security assurance-had a greater impact as an integral part of a comprehensive package deal rather than on a standalone basis. North Korea tends to view security assurances on a stand-alone basis as just a "piece of paper:'46 What is unique about China's regime security assurance to North Korea is that at its core the CPC is pledging to ensure the survival of the WPK. This highly nuanced pledge represents an attempt to effect a major transformation in the way Pyongyang frames future Sino-DPRK relations and how the two parties cooperate within this reformulated dynamic. Senior CPC officials sought to convince the Kim Jong-illeadership group that Beijing would not recognize a third generation of the Kim family. To get around this impasse, the Kim Jong-il regime appears to have framed the succession process not as a function of power transferring from the father to the third son, Kim Jong-eun, but rather from one generation of the WPK to the next. Doing so allowed the CPC to pledge to support this rising WPK generation to bolster prospects for North Korean economic reform and opening-the key tasks that Beijing views as critical to setting North Korea on a more stable foundationY The other elements of the Chinese package deal reinforce the viability and credibility of the regime security assurance. These elements comprise economic development projects in North Korea where the Chinese are supplying the capital and equipment. This type of development activity is largely focused where most of North Korea's mineral resources are located. For China, the

ASSESSING THE ROLE OF SECURITY ASSURANCES

213

commercial benefits derived from this development work are significant. Given that a major priority for the central government in Beijing is supplying China's northeastern provinces with energy resources to fuel regional economic development, North Korean coal plays a strategic role in bolstering China's energy security. The combination offered in the Chinese package deal-a North Korean regime security assurance and progress in economic development-has the potential to be a win-win for both countries. Future success, however, will depend on whether North Korean provocations exhaust Chinese patience before the strategy has a chance to play out, and whether the United States and South Korea can be persuaded to accept the Chinese approach and refrain from escalating tensions with Pyongyang. CONCLUSION

This chapter finds that security assurances have had some ability to moderate North Korea's pursuit of a nuclear weapons capability, although they have not succeeded in producing a lasting nonproliferation result. Though it is difficult to meet the demanding conditions for success in this case, assurances are most likely to have an impact on North Korea when they are embedded in a larger package, tailored to address the DPRK's concerns, and capable of influencing internal political developments. While North Korea's scrambling to ensure regime survival following geopolitical shocks has been constant, the sources of an effective security assurance have fluctuated. The positive security assurance that Pyongyang received from Moscow in accordance with their 1961 defense treaty ended, in practice, after the demise of the provider in 1991. North Korea accelerated its nuclear weapons development program in the early 1990s-precisely at the time when its security assurance effectively disappeared. Is this correlation or coincidence? A close analysis of North Korean nuclear policy behavior during this and subsequent periods indicates the former. Creativity in tailoring a security assurance to North Korea is a major determinant of the reclusive country's receptivity and reciprocity. In this endeavor, the CPC seems to have designed a security assurance that closely matches North Korea's measurements-that is, a regime security assurance. In many respects, this is what North Korea enjoyed with the Soviet Union. The Soviet positive security assurance was combined with a preferential trading system that enabled the Kim Il-sung regime to enjoy its golden era in the 197os-the peak of Soviet power and largesse.

214

JOHN S. PARK

The key question is whether the United States can structure a robust security assurance that can be a part of a package deal to effect peaceful nuclear rollback in North Korea. The 2005 Joint Statement of Principles of the Six-Party Talks represents the closest that the United States has come to realizing a comprehensive settlement with North Korea. With senior U.S. officials skeptical that North Korea will denuclearize in the short term, it is difficult to envision a situation-even if the Six-Party Talks resume-in which the United States would go beyond the security assurance embedded in the Joint Statement of Principles to incentivize North Korea to dismantle its nuclear arsenal. Even if there were signs of North Korean intentions to initiate early stage denuclearization activities in return for a regime security assurance from the United States, there would be no political will or ability in Washington to support a U.S. guarantee of the survival of the North Korean regime. A negative security assurance is the maximum that the United States could offer-and already has offered-to the North Koreans. The Chinese innovation on the regime security assurance model is reflected in the CPC's bailout of the WPK in October 2009 during Premier Wen Jiabao's visit to Pyongyang. There to celebrate the sixtieth anniversary of the founding of Sino-DPRK diplomatic relations, the premier brought an entourage including the commerce minister and the chief of the National Development and Reform Commission-the chief architects and implementers of Chinese internal economic development-in addition to senior military officers and CPC International Liaison Department officials. During the visit, Premier Wen signed a series of innocuous-sounding bilateral "economic development, tourism, and education" deals. In practice, these deals collectively represent the conduit through which the CPC will bolster the stability and viability of the WPK and reinforce the potency of a possible Chinese package deal. 48 The implications of China's use of this unique form of security assurance as a policy instrument are profound. China has a virtual monopoly on the application of this policy instrument because no other actor can provide, let alone maintain, a regime security assurance in the same manner. Significantly, the CPC is offering a regime security assurance via the WPK with no direct linkage to immediate progress on North Korean denuclearization. No U.S. administration could even suggest such an arrangement. The Chinese bailout of North Korea casts a growing shadow over efforts to realize a nuclear-weapons-free Korean Peninsula. Cocooned by the Chinesebacked Six-Party Talks process and the deepening Two-Party approach with Beijing, North Korea could undergo a metamorphosis and emerge as a nuclear

ASSESSING THE ROLE OF SECURITY ASSURANCES

215

state with a fragmented but functional economy. In this respect, China's regime security assurance will have been the key catalyst for North Korea's attainment of its definition of being a "strong and prosperous nation:' It is difficult to imagine anything that the United States can plausibly offer to counter this unfolding reality. Only North Korea itself might derail this process. If Pyongyang continues its provocative actions, China may no longer be willing to take the risks associated with ensuring its ally's security. If so, interactions between North and South Korea could shift to a more coercive track, possibly even open conflict. Security assurances to North Korea would then be relegated to the sidelines. NOTES

The views expressed in this paper are those of the author and do not reflect those of the u.s. Institute of Peace, which does not advocate specific policies. 1. In analyzing North Korea's nuclear behavior in relation to security assurances, the author has drawn insights from U.S.-North Korea Track II meetings in which he participated and interviews he conducted with government policy advisors working on their respective countries' approaches to resolving the DPRK nuclear issue. A series of Track II meetings in which the author participated was initiated by North Korea policy analysts at the Kennedy School of Government's Belfer Center for Science and International Affairs in late 2002; it convened senior North Korean diplomats and u.S. congressional officials to foster policy discussions on how to negotiate North Korean denuclearization. The author has augmented insights derived from the Track II meetings with his ongoing interviews and meetings, since 2004, with North Korea-focused government policy advisers in Beijing; Washington, DC; Seoul; and Tokyo. 2. Victor Cha, ''A Nuclear Fission: The North Korea Debate in Washington;' Harvard International Review 25, no. 4 (Winter 2004); Yong Chool Ha and Chaesung Chun, "North Korea's Brinkmanship and the Task to Solve the 'Nuclear Dilemma;" Asian Per-

spective 34, no. 1 (2010). 3. "While the term juche is commonly translated from the Korean language as 'selfreliance; it does not lend itself to any single, precise definition. Depending on the context in which it is used it can mean national identity, self-reliance, national pride or national assertiveness. The four guiding principles of juche are autonomy or identity in ideology, independence in politics, self-sufficiency in economy, and reliance on Korea's own forces in national defence:' Soyoung Kwon, "Survival of the North Korean Regime and Changing Legitimation Modes;' Mario Einaudi Center for International Studies, Cornell University, Working Paper no. 07-05, October 2005, 1. Though there was a period of solid industrial output in the second half of the 1960s,juche's achievements were extolled in later years to an extent that bordered on myth.

216

JOHN S. PARK

4. Harry Gelman and Norman Levin, The Future of Soviet-North Korean Relations

(Santa Monica, CA: RAND Corporation, October 1984), 2. 5. Korea Atomic Energy Research Institute, "Pukhanui Haekkwallyon Yon'guhwaltong;' cited in Nuclear Threat Initiative, NT! Country Overviews, North Korea Nuclear Chronology, www.nti.orglprofiles/NK/Nulcear/46_89.html. 6. Susan Rosegrant and Michael Watkins, "Carrots, Sticks, and Question Marks: Negotiating the North Korean Nuclear Crisis (A)" (Cambridge, MA: John F. Kennedy School of Government Case Program, 1995), 5. 7. Mitchell Reiss, Bridled Ambition: Why Countries Constrain Their Nuclear Capabilities (Washington, DC: Woodrow Wilson Center, 1995), 239. 8. Don Oberdorfer, The Two Koreas: A Contemporary History (New York: Basic Books, 1997), 265-66. 9. Bob Woodward, Bush at War (New York: Simon and Schuster, 2002), 340. 10. David E. Sanger, "Bush Takes No-Budge Stand in Talks with North Korea;' New

York Times, April 17, 2003; Hyun-jin Seo, "US: North Korea Must Abandon Nukes First;' Korea Herald, April 23, 2003. 11. Alexandre Mansourov, "North Korea's Road to the Atomic Bomb;' International Journal of Korean Unification Studies 13, no. 1 (2004): 27. 12. Ibid., 36. 13. Balazs Szalontai and Sergey Radchenko, "North Korea's Efforts to Acquire Nuclear Technology and Nuclear Weapons: Evidence from Russian and Hungarian Archives;' Working Paper no. 53, Cold War International History Project (Washington, DC: Woodrow Wilson International Center for Scholars, August 2006), 28. 14. Miles Kahler, "Strategic Uses of Economic Interdependence;' in A Changing Korea in Regional and Global Contexts, ed. Lee-Jay Cho, Chung-Si Ahn, and Choong

Nam Kim (Seoul: East-West Center and Seoul National University Press, 2004), 175. 15. Ibid., 176. 16. Interviews conducted by the author with PRC government policy advisers, Beijing, August 2-6, 2004. 17. "Chronology of U.S.-North Korean Nuclear and Missile Diplomacy;' Arms Control Association (Washington, DC, October 2010), http://www.armscontrol.org/factsheets/dprkchron.asp. 18. Nicholas Eberstadt, an economist at the American Enterprise Institute, argues that for North Korea it is imperative to continually upgrade the threat posed by its weapons of mass destruction and its nuclear program in order to extort desired concessions from the international community. Nicholas Eberstadt, "Hastening Korean Reunification;' Foreign Affairs 76, no. 2 (March/April 1997): 86-87. 19. Joel S. Wit, Daniel B. Poneman, and Robert L. Gallucci, Going Critical: The First North Korean Nuclear Crisis (Washington, DC: Brookings Institution Press, 2004), 356. 20. Gary Samore, "The Korean Nuclear Issue;' Survival 45, no. 1 (Spring 2003): 10.

ASSESSING THE ROLE OF SECURITY ASSURANCES

217

21. Wit et aI., Going Critical, 371. 22. Jonathan D. Pollack, "The United States, North Korea, and the End of the Agreed Framework;' Naval War College Review 56, no. 3 (Summer 2003): 8. 23. Wit et al., Going Critical, 371. 24. President George W. Bush, "Graduation Speech at West Point;' Office of the White House Press Secretary, June 1, 2002. 25. Nicholas Eberstadt, "Reckless Driving: Kim Jong Ii's Erratic Moves Strengthen the Hawks' Case for Regime Change;' TIME Asia 161, no. 18, May 12, 2003. 26. Interviews conducted by the author with ROK foreign ministry policy advisers in Seoul, and PRC Central Party School and Ministry of State Security policy experts in Beijing, June-August 2004. 27. Soo-Jeong Lee, "South Korea Seeks Easing of Nuclear Crisis at Economic Talks with North Korea;' Associated Press, February 12, 2003. 28. Other security analysts in Northeast Asia shared such an assessment. "We cannot rule out the possibility that North Korea has produced a couple of primitive nuclear explosive devices. However, to assemble a nuclear warhead that can be mounted on a missile, nuclear testing will be required to make a smaller and lighter warhead:' Shinichi Ogawa, "TMD and Northeast Asian Security;' Nautilus Institute Working Paper (rev. ed.), September 28, 2000. 29. Hu Xuan, "Diplomacy Needed to Ease Tension;' China Daily, January 3,2003. 30. Shannon Kile, ''Appendix 2: Chronology of Principal Defence and Security-related Agreements and Initiatives Involving the Russian Federation and Asian Countries, 1992-99;' in Russia and Asia: The Emerging Security Agenda, ed. Gennady Chufrin (Oxford: Oxford University Press, 1999). 31. Interviews conducted by the author with PRC government think tank analysts, Beijing, March 8-9, 2009. 32. Agreed Framework between the United States ofAmerica and the Democratic People's Republic of Korea (Geneva: October 21, 1994). 33. Review of United States Policy toward North Korea: Findings and Recommendations, Unclassified Report by Dr. William J. Perry, U.S. North Korea Policy Coordinator

and Special Advisor to the President and the Secretary of State. Washington, DC, October 12, 1999, http://www.state.gov/www/regions/eap/991012_northkorea_rpt.htmi. 34. U.S.-DPRK Joint Communique, Washington, DC, October 12, 2000, http://www. fas.orginews/dprk/2000/dprk-oolO12a.htm. 35. Knopf, "Security Assurances: Initial Hypotheses;' Chapter 2, this volume. 36. Joint Statement of the Fourth Round of the Six-Party Talks, Beijing, September 19, 2005, http://www.fmprc.gov.cn/engltopics/dslbj/t212707.htm. 37. Victor Cha, "What Do They Really Want?: Obama's North Korea Conundrum;'

Washington Quarterly 32, no. 4 (October 2009): 126. 38. U.S. Department of Defense, Nuclear Posture Review Report, April 2010, 15.

218

JOHN S. PARK

39. Cheon Seongwhun, "Changing Dynamics of U.S. Extended Nuclear Deterrence on the Korean Peninsula;' Nautilus Institute, Special Report, November 10, 2010, http:// www.nautilus.org/publications/essays/napsnet/reports/changing-dynamics-of-u.s.-extended-nuclear-deterrence-on -the-korean -peninsula. 40. Interviews conducted by the author with PRC government think tank analysts, Beijing, March 8-10, 2010. 41. On the role of alliance constraints in limiting U.S. and ROK reassurance efforts

directed at North Korea, see Jungsoo Kim, "Reassurance Strategy: Incentives for Use and Conditions for Success;' Ph.D. diss., Naval Postgraduate School, Monterey, CA, March 2010.

42. John S. Park, "North Korea's Leadership Succession: The China Factor;' USIP

Briefing, September 2010. 43. Senior Obama administration officials call their policy toward North Korea "strategic patience:' After Pyongyang responded to the administration's initial overtures

for a continuation of denuclearization negotiations by launching a long-range missile in April 2009 and conducting its second nuclear test in May 2009, the administration remains resolved that Pyongyang has "to make the first move to reengage and that it won't be granted any concessions." Glenn Kessler, ''Analysis: North Korea Tests U.S. Policy of 'Strategic Patience;" Washington Post, May 27, 2010. 44. Interview conducted by the author with a senior research fellow from the PRC Ministry of Foreign Affairs think tank, Washington, DC, August 15, 2010. 45. Park, "North Korea's Leadership Succession:' 46. Interviews conducted by the author with PRC government think tank analysts, Beijing, March 8-10, 2010. 47. Ibid.

48. The ardent Chinese belief is that after experiencing the initial benefits of economic reform, the DPRK regime will eventually realize that giving up its nuclear arsenal and joining the international community will enhance its long-term viability. Interview conducted by the author with a senior research fellow from the PRC Ministry of Foreign Affairs think tank, Washington, DC, August 15, 2010.

The United States and Swedish Plans to Build the Bomb, 1945-68 Thomas Jonter

Why did Sweden choose, in the late 1960s, to abandon its long-standing nuclear weapons plans? In a previous paper, I have argued that the key factor leading to Sweden's abstaining from acquiring nuclear weapons was its prior decision to make the country's nuclear weapons project a part of the civilian nuclear energy program.! Seeking to produce weapons-grade plutonium in connection with a national goal of self-sufficiency in nuclear power meant trying to create a wholly domestic production cycle, a technically complicated and time-consuming process. This drawn-out process influenced the nuclear weapons plans negatively in three ways. First, it made possible the mobilization of political opposition against the nuclear weapons plans, with public opinion and parliamentary discussions moving gradually in the direction of saying no to Swedish nuclear weapons. Second, arms control talks between the United States and the Soviet Union and international initiatives to prevent the spread of nuclear weapons from the mid-1950S onward also affected the Swedish public debate and strengthened the arguments against Swedish nuclear weapons acquisition. Third, as a consequence of its decision to integrate the production of nuclear weapons within the civilian nuclear energy program, Sweden, despite contrary intentions, grew dependent on U.S. technology. This technological dependence provided the United States with leverage to dissuade Sweden from using its civilian program for producing weapons-grade plutonium. This chapter analyzes the third aspect of this explanation-that is, Sweden's technological dependence on the United States. It argues that Sweden became more and more dependent on U.S. assistance both in nuclear energy and in military hardware, and this dependence was used by the U.S. government to 219

220

THOMAS JONTER

steer Swedish decision-makers away from using the country's civilian program for the production of weapons-grade plutonium. The cooperation in the military field between the two states has been interpreted by many observers, including two Swedish government commissions, as amounting to the extension, on Washington's part, of informal security guarantees to Sweden.2 The so-called Commission on Neutrality, an official Swedish investigation of U.S.Swedish military cooperation in the years 1945-69, concluded that, during the Cold War, the United States was prepared to defend Sweden in a war against the Soviet Union. 3 This conclusion raises a further question that is especially pertinent to the study of which this chapter is a part: did this informal security guarantee also include an assurance that Sweden would be protected by the U.S. nuclear umbrella, and if so, was that a reason for Sweden to give up its own nuclear weapons ambitions? This chapter identifies indirect evidence that suggests a gentlemen's agreement between U.S. and Swedish military officials, to the effect that Sweden was placed under the U.S. nuclear umbrella. It concludes that this tacit agreement made it easier for Sweden to turn away from nuclear weapons, but was not a decisive factor in why Sweden gave up its nuclear weapons plans. In the penultimate section of this chapter, the findings of this case study are compared against the hypotheses presented in Chapter 2. The Swedish case fits hypotheses on security, defense cooperation, and tailoring. It also provides some support for, while suggesting modifications to, hypotheses concerning norms, the importance of influencing internal debates, and the value of combining assurances with positive incentives. Finally, Sweden is a disconfirming case for hypotheses concerning public and legally binding commitments. THE BEGINNINGS OF SWEDEN'S BOMB PROGRAM

By the end of World War II, the postwar Swedish foreign policy position was already well established: non participation in military alliances in peacetime aiming at neutrality in the event of war. A Swedish nuclear weapons program initially appeared consistent with this foreign policy orientation. In 1945, shortly after the United States dropped two nuclear weapons over Japan, the newly formed Swedish National Defense Research Establishment (FOA) was requested by the supreme commander of the armed forces to gather information about this new weapon of mass destruction. As part of the assignment, FOA was also charged with investigating the possibilities for Sweden of producing its own nuclear weapons. 4 Those among the Swedish political elite who ad-

THE UNITED STATES AND SWEDISH PLANS TO BUILD THE BOMB

221

vocated nuclear weapons production maintained that Swedish nuclear weapons would be necessary for deterring the Soviet Union from attacking Sweden and for guaranteeing a strong defense that would help uphold Sweden's policy of nonalignment. The first comprehensive FOA study of the possible production of nuclear weapons was completed in 1948. The study concluded that plutonium would be preferable to uranium-235 (U-235) as a nuclear material for explosive devices. The study further concluded that it would take about eight years, and probably longer, to produce a nuclear weapon. 5 A cooperation agreement between FOA and AB Atomenergi (AE), a state-controlled company that was responsible for civilian nuclear energy development, was signed in 1949. FOA was put in charge of the construction of the nuclear device and studies of its effects. AE would be responsible for efforts to produce weapons-grade plutonium, and as part of this effort would investigate the possibilities of production or procurement of heavy water that would not be subject to inspections by the supplying country. AE was also assigned the building of reactors and a reprocessing plant, and the manufacture of fuel elements to be used in the reactors for production of weapons-grade plutonium. For this reason, Sweden chose a heavy water technology that would allow reactors to be loaded with natural uranium that did not require prior enrichment. Within the framework of this plan for self-sufficiency in nuclear energy, it was primarily the extensive though low-quality kolm-type shales in central Sweden that were of interest for the domestic production of uranium. 6 In the early 1950S, Sweden also signed contracts with Norway giving Sweden access to heavy water in exchange for uranium'?

u.s. NUCLEAR POLICY TOWARD SWEDEN, 1945-59: INCREASING HELP, BUT WITH STRINGS ATTACHED

Despite the goal of self-sufficiency, Sweden could not complete a nuclear program relying solely on its own resources and know-how, and so began looking for outside assistance, including from the United States. Between 1945 and 1959, U.S. policy evolved from denying assistance to providing extensive help to Sweden's atomic energy program. To gain this help, however, Sweden had to agree to constraints that limited its freedom of maneuver in the nuclear field. From 1945 until the summer of 1948, the U.S. government followed a strict policy toward Sweden of prohibiting the export of most nuclear materials, equipment, and technical data. In addition, Washington sought to persuade the Swedes not to launch domestic uranium production, particularly not of weap-

222

THOMAS JONTER

ons-grade quality. Beginning in the summer of 1948, however, the u.s. Atomic Energy Commission (AEC) eased the restrictive policy somewhat to allow Swedish researchers access to some formerly classified material and permit the sale of certain equipment. One reason for this policy shift was that the United States wanted to uphold a certain level of flexibility in its dealings with friendly countries until the contours of postwar Western political cooperation became clearer. 8 Despite its official neutrality, later government investigations revealed that Sweden in fact pursued far-reaching cooperation with the Western European powers and the United States throughout the postwar era, beginning in the late 1940S. For example, Swedish airfields were adjusted to be able to accommodate U.S. bombers, and Sweden and the United States also shared military intelligence. 9 In early 1952, the United States established a basic policy toward Sweden as part of NSC-121. This policy accepted that Sweden would not join NATO, but it anticipated continued defense and intelligence cooperation and saw advantages in helping Sweden bolster its conventional defense capabilities. This document remained the basic U.S. policy guidance for the rest of the decade. lO The limited assistance offered by the United States in the late 1940S and early 1950S did not create any significant obstacles to Sweden's plans for an independent nuclear program. This would change by the mid-1950s, when U.S. policy shifted in the direction of supporting peaceful nuclear technology development. This enabled Sweden to receive greater U.S. assistance, which ironically had the effect of making it harder for Sweden to develop an independent nuclear capability. The new U.S. policy grew out of the "Atoms for Peace" program announced by President Eisenhower in December 1953. The basic idea was that the nuclear powers would cooperate to assist other states in the development of civilian nuclear energy in return for promises that the assistance would not be exploited to develop nuclear weapons. From that point on, transfer of nuclear materials to other countries was allowed-even in the form of highly enriched uranium and plutonium-provided that the receiving country committed itself to not using the acquired nuclear material for nuclear weapons production. 11 For Sweden's part, the Atoms for Peace program meant that the country was now allowed to buy materials and equipment from the United States that had formerly been subject to export prohibition. Apart from allowing friendly countries to buy these wares at favorable prices, the program also gave Sweden and other cooperative nations access to financial assistance for nuclear energy research. In return, the countries receiving assistance and signing cooperation agreements committed themselves to fulfilling certain security requirements.

THE UNITED STATES AND SWEDISH PLANS TO BUILD THE BOMB

223

First of all, exported equipment and materials could not be used for the production of nuclear weapons. Nor could the products received be resold or lent to a third country for purposes of nuclear weapons production. The United States was entitled, according to the cooperation agreements, to conduct inspections aimed at ascertaining that its partners were living up to the agreedupon requirements. Negotiations between Sweden and the United States focused on Swedish access to previously classified technical information and the possible procurement of uranium enriched up to 20 percent U -235 (a threshold set because below this level of enrichment uranium is generally considered not to be weapons usable). Swedish officials sought U.S.-origin enriched uranium as a relatively low-cost fuel for planned commercial nuclear energy reactorsY The pursuit of an inexpensive fuel for commercial nuclear power plants had a negative impact on the possibility of producing nuclear weapons, since the nuclear weapons plans required that Swedish reactors be loaded with uranium produced in Sweden and therefore not subject to inspections or legal restrictions against use in a nuclear bomb program. The eagerness of Swedish researchers and technicians to develop Swedish nuclear energy using enriched uranium imported from the United States put Washington in the position of being able to exploit Sweden's materials-related dependency. The responsible U.S. officials were able, step by step and as part of the evolving cooperation, to lay down requirements and raise demands in accordance with U.S. global nuclear energy policy that gradually lessened Sweden's room for maneuver. On January 18,1956, the United States and Sweden signed an agreement on civilian nuclear energy cooperation. The two parties agreed to exchange information regarding the construction, operation, and development of research reactors. The United States pledged to deliver up to six kilograms ofU-235 enriched to a maximum degree of 20 percent, with possible further quantities to be delivered if the AEC deemed them necessary for the continued effective operation of the reactors. The Swedish government committed itself to providing the AEC with continuous information regarding nuclear energy developments in Sweden. The agreement forbade Sweden from using any equipment and materials involved in the agreement for the development of nuclear weaponsY Through this and other agreements, the United States was able to gain oversight and control over important aspects of Sweden's nuclear energy development, and, most important, to prevent exported materials and equipment from being used for military purposes.

224

THOMAS JONTER

In addition to enriched uranium, heavy water also became the subject of negotiations and procurement agreements between the United States and Sweden. During the summer of 1955, the Swedish diplomat Carl Douglas raised the matter when he inquired about the possibilities of receiving financial assistance from the United States for the building of reactors. 14 In late November 1955, the AE research director, Sigvard Eklund, and CEO, Harry Brynielson, stated their wish to buy heavy water from the United States in order to be able to suspend their agreement with Norway, or at least parts of it, because the price of U.S. heavy water was much lower.ls The United States proved amenable, and, on May 16, 1957, an agreement was signed regulating the purchase of twenty-six tons of heavy water from the United States. 16 Already by June, the Swedish foreign ministry had asked the AEC if this quantity might be raised to fifty tons, within the framework of the same contractual conditions. The AEC replied that this could not be done on the basis of a mere amendment to the existing agreement due to inspection and security requirements. Instead, the AEC asked the Swedes to bring up the matter in a new round of negotiations aimed at reaching a new agreement that would contain rules and requirements regarding safeguards that had been worked out by the AEC. 17 This was important because the initial twenty-six tons of heavy water that Sweden procured were not subject to U.S. inspection control. 18 However, if AE were to receive the additional twenty-four tons, this would mean that the whole fifty- ton volume of procured heavy water would be subject to future inspection control. This condition created some irritation and hesitation on the Swedish side. 19 For the Swedes, the goal was to reduce their dependence on the United States while also minimizing the U.S. rights of inspection. Swedish officials ultimately chose not to return to the negotiating table to discuss another purchase of heavy water from the United States. If Sweden were to allow all the heavy water purchased from the United States to include the right of inspection, it would be impossible to integrate any form of production of weapons-grade plutonium within the civilian program. During the fall of 1956, Sweden also made a request to acquire an additional eight kilograms of U -235 (but this time enriched to 90 percent, which is weapons-grade) within the framework of the January 1956 agreement. The request showed how conscious Swedish officials were of the rules of the game and of what was required if Sweden were to be able to acquire additional quantities of enriched uranium. This is evident in a memorandum produced by the Delegatio-

THE UNITED STATES AND SWEDISH PLANS TO BUILD THE BOMB

225

nen for atomenergifragor (Delegation for Atomic Energy Issues, DFA), an atomic energy advisory committee composed mainly of industry representatives. 2o In proposing guidelines that Sweden should follow in the upcoming negotiations with the AEC, the DFA stated that Swedish negotiators should highlight the fact that eight kilograms of U -235 is not sufficient to produce a nuclear bomb. The idea was to clearly communicate that Sweden had no intention of using the acquired nuclear materials for the production of nuclear weapons. During the negotiations with the United States, which were conducted by the foreign ministry, the Swedes also submitted an application for financial assistance to the tune of $350,000 for the construction of Sweden's second reactor, the R2, in Studsvik. The R1, a small, heavy water research reactor located in Stockholm, had gone critical in 1954. R2 would be a light water research reactor used for test purposes, but Sweden also hoped to learn things from operating R2 that would be helpful in a bomb-development program. The application for U.S. financial assistance contained a "full set of data concerning R2;' as requested by the AECY In these negotiations, the United States demanded that a new and comprehensive bilateral cooperation agreement be signed. The AEC wanted the agreement to cover a period of ten years, and for Sweden to commit itself to fulfilling certain security requirements (in the form of U.S. inspections carried out on Swedish soil). In return, Sweden would receive U-235 (at 90 percent enrichment) on advantageous terms. The Swedish government maintained that an amendment to the already existing agreement should suffice, a position the United States did not accept. In the end, Sweden agreed to the U.S. demands, and the new agreement was signed on April 25, 1958.22 According to the terms of the agreement, Sweden would be eligible to buy or lease 200 kilograms ofU-235 (20 percent enriched). The AEC would also allow for eight of these 200 kilograms to be enriched to 90 percent for use in a test reactor. The agreement also required "that when any source or special nuclear material received from the United States requires reprocessing, such reprocessing shall be performed either in [Atomic Energy] Commission facilities or in facilities acceptable to the Commission:'23 This was meant to preclude Sweden from using the reactor fuel as a basis for obtaining nuclear materials for use in a bomb program. On April 28, 1958, the AEC declared itself willing to contribute $350,000 to the construction of R2 in Studsvik. The contribution would be disbursed upon completion of the facility.24 In 1960, the R2 reactor went critical. By the end of

226

THOMAS JONTER

the 1950S, as a result of the developments described here, Sweden had gained access to considerable materials and assistance from the United States. To do so, however, Sweden had agreed to conditions that increasingly constrained its pursuit of nuclear weapons. DEVELOPMENTS IN SWEDEN'S NUCLEAR WEAPONS RESEARCH AND DOMESTIC POLITICS

Even as the collaboration with the United States became more extensive, the nuclear weapons research at FaA continued, moving into a more intensive phase. In 1953, the second FaA study saw the light of day, and in 1955 the third study of nuclear weapons production was completed. By 1955, nuclear weapons weighing only about 100 kilograms (kg) were now being discussed, devices far lighter than those FaA had previously envisioned. The new devices, which came to be known as tactical nuclear weapons, were regarded as easily transportable and easy to use in both missiles and torpedoes. Based on the changing technology, the 1955 study concluded that it was technically feasible from then on for Sweden to produce a nuclear weapon, given access to plutonium. 25 Public debate in Sweden also took off starting in 1954. That year, the supreme commander released a report (OB-54) that recommended Sweden obtain nuclear weapons. 26 The supreme commander specifically called for adding tactical nuclear weapons to Sweden's defense capabilities, but not strategic systems. Although the opposition Conservative Party responded favorably, others did not. In the following years, the opposition toward the Swedish nuclear weapons plans grew outside and inside the Parliament, and even within the ruling Social Democratic Party. In 1956, the National Federation of Social Democratic Women adopted a resolution against acquiring nuclear arms. Facing potential electoral vulnerabilities, Prime Minister Tage Erlander felt that he had to strike a balance between opponents and proponents of nuclear weapons in his own party in order to keep the ruling coalition together. He argued at meetings with his party members that a decision on nuclear weapons should be postponed until 1958, since the government did not yet have sufficient knowledge regarding the technical prerequisites for nuclear weapons production. He also said that Sweden should not make international disarmament talks more complicated or undermine its own prodisarmament diplomatic stance by deciding to produce nuclear weapons. He also expressed concern that a Swedish bomb would likely lead to further global proliferationY With this strategy, Erlander made an explicit connection between promot-

THE UNITED STATES AND SWEDISH PLANS TO BUILD THE BOMB

227

ing technical progress intended to make domestic production of nuclear weapons feasible and international developments in the nuclear disarmament area. In other words, if the international negotiations proved unsuccessful, the policy of postponing the decision to go nuclear would no longer be followed. If the technical prerequisites were then at hand, a decision to go forward with weapons production would likely be made. By presenting the issue in this fashion, and satisfying both the yes and the no factions at least temporarily, Erlander gained some freedom of action. Opponents of Swedish nuclear weapons could continue to mobilize opposition, while working toward and hoping for breakthroughs in the international disarmament arena. Proponents were given the opportunity to continue technical preparations. Due to developments in Sweden's relations with the United States, however, in practical terms the flexibility needed to pursue a nuclear option would continue to diminish. DEVELOPMENTS IN COMMERCIAL AND MILITARY COOPERATION WITH THE UNITED STATES

Both commercial and military cooperation with the United States contributed to restricting Sweden's options. On the commercial side, the more Sweden's nuclear R&D became dependent on U.S. assistance, the more the United States could use its leverage to steer Sweden away from its nuclear weapons plans. One decisive measure taken by the United States was lowering the price of enriched uranium at the end of the 1950S, which reduced the fuel costs for running light water facilities. 28 This measure enabled private companies in states that had nuclear reactor plans to start investing in light water technology, since they would not need to spend fortunes on developing methods for enriching or processing uranium themselves. The U.S. action benefited U.S. economic interests, by helping American companies put their light water technology on the international market as a more economically favorable and reliable reactor system than the heavy water system. In addition, though, this U.S. action also had two potentially constraining effects on proliferation. Light water reactors are less useful for bomb programs than heavy water reactors because it is harder to obtain weapons-grade materials from their spent fuel. In addition, any purchases of enriched uranium from the United States would impose new safeguards on the recipient. Sweden was one of the countries enticed to consider switching to greater use of light water reactors. When the United States drastically lowered the prices of uranium, an investigation was made in 1959 by the atomic advisory commit-

228

THOMAS JONTER

tee, the DFA. The investigation estimated that Swedish production of uranium would cost 70 percent more than uranium imported from the United States. The DFA experts, who by and large represented industry interests, were in favor of importing uranium, even though it would imply restrictions in the form of U.S. inspections. 29 As a consequence of the U.S. price change, private industry now saw real opportunities for light water technology in Sweden. The Swedish military technological dependence on the United States could also be used by the Americans as a means to steer the Swedes away from their nuclear weapons plans. When Swedish military officials and diplomats made requests to purchase advanced radar systems and receive technical assistance to develop missiles in the mid-1950s, the U.S. administration was able to use this to advance U.S. nonproliferation goals. For example, in talks between U.S. and Swedish diplomats and military officials concerning cooperation in the missile field, the United States convinced the Swedish representatives not to develop a domestic manufacturing system because of high production costs. As a result of these negotiations, the Swedes-instead of manufacturing their own missiles-were allowed to purchase and even produce on license U.S. missile systems such as the Falcon, Hawk, and Sidewinder. In the eyes of the Swedish military establishment, the goal was to develop a strong defense capability on a rather small budget, and when the Americans opened the door to the possible export of advanced weapons systems it was considered as a great advantage. On the other hand, these compromises, made by a previously more or less independent Swedish defense force, carried their price. One price that Sweden had to pay was an agreement that the imported U.S. missiles could not be equipped with nuclear devices. 3D All in all, this military cooperation with the United States created a measure of Swedish dependence that in turn curtailed Sweden's room to maneuver. The increasingly intimate cooperation between the two states in the area of military technology led to the use of formal and, even more so, informal channels to communicate what was permissible for Sweden if it wished to see this cooperation continue. 3l SWEDISH INQUIRIES REGARDING THE PURCHASE OF U.S. NUCLEAR WEAPONS

In the second half of the 1950S, Swedish politicians and defense experts realized that national production of nuclear weapons would cost much more than had been estimated four to five years earlier. Consequently, Swedish officials began exploring the possibility of acquiring nuclear weapons from the United

THE UNITED STATES AND SWEDISH PLANS TO BUILD THE BOMB

229

States. The Swedish defense establishment assumed that even though Sweden was not a member of NATO, it would be in the U.S. national interest to have a Swedish military that was sufficiently strong to deter a Soviet attackY These talks should be understood in the context of Washington's intention, at this time, to equip its NATO allies with tactical nuclear warheads. For example, when the Swedish ambassador to the United States, Erik Boheman, discussed these matters with the State Department official C. Burke Elbrick in June 1957, he referred to "dual-purpose weapons such as the 'Nikes' and 'Honest Johns' recently offered to Norway and Denmark trusting, as Norway and Denmark must, that atomic warheads would be forthcoming in case of need:'33 U.S. officials recognized that most of the missile systems being requested by Sweden (the main exception being the Sidewinder) were dual-capable and hence could be armed with an atomic warhead. They clearly perceived that these Swedish requests involved an assumption that U.S. warheads might be provided in the event of a future emergency, such as a Soviet invasion. In the 1957-59 period, certain officials in the Defense and State departments suggested that the United States should be favorably disposed to seeing if some arrangement could be worked out to give Sweden future access to U.S. warheads. 34 In the end, the U.S. administration reacted negatively to these Swedish inquiries. In internal deliberations, U.S. officials recognized that there were both advantages and disadvantages to providing Sweden with nuclear weapons capability. If the United States were to maintain an unbending restrictive policy toward Sweden, there was an increased risk that the Swedes would initiate their own production of nuclear weapons and the United States would lose its existing leverage over Swedish nuclear research. In May 1959, this matter was dealt with at a high political level in the State Department. After lengthy discussion, the assessment from the State Department was that this was not a time to enter into negotiations with Sweden. The issue discussed by State Department officials was whether it would be at all possible to provide Sweden with nuclear weapons considering the restrictions of the U.S. Atomic Energy Act, which permitted the U.S. government to contribute to other nations' nuclear weapons capability only if the country in question had a mutual defense agreement with United States. Although the Atomic Energy Act sanctioned the transfer of nuclear weapons only to NATO member countries, it contained another provision, Section 144(b), allowing for non -NATO member countries to receive nuclear weapons-related training and technical data. However, this required that there be an existing nuclear mate-

230

THOMAS JONTER

rials supply agreement in place and "that other nation be participating with the United States pursuant to an international arrangement by substantial and material contributions to the mutual defense and security."35 The responsible officials in the State Department now asked themselves whether Section 144 (b) could be applied to Sweden without there being "significant change in Swedish foreign policy." Purely theoretically, they argued, it might be possible to have this law apply even without the United States signing a defense agreement with Sweden. If the United States were to pursue a policy of providing Sweden with nuclear weapons capability while preventing the country from initiating its own production of nuclear weapons, however, certain restrictive conditions would need to obtain-namely, "bilateral agreement providing either (a) for joint control over use of nuclear-capable weapons purchased by Sweden, or (b) Swedish undertaking to use such nuclear-capable weapons only with nuclear warheads provided by U.S:' The State Department asked the U.S. embassy to analyze whether such a major departure from Swedish neutrality was foreseeable. 36 Although this analysis suggested that the United States remained open to the idea of helping Sweden acquire nuclear weapons, it also shows that the U.S. government did not foresee a way to do so that was both consistent with U.S. law and likely to be acceptable to Stockholm. In Sweden, meanwhile, work continued on independently developing a nuclear capability, but the domestic political context began shifting against the program. PROTECTION RESEARCH AND THE DECLINE OF MILITARY ENTHUSIASM FOR A SWEDISH BOMB

By 1958, Swedish nuclear weapons research had reached a point where a political decision on the issue could be made. Design work was well under way on two planned heavy water reactors that were of special interest for the plans to produce weapons-grade plutonium: the Agesta (R3) Nuclear Power Station south of Stockholm and Marviken (R4) close to the city of Norrkoping.37 The issue had turned into a rather complicated business for Erlander and his government. Growing criticism within the Social Democratic Party and in the media, along with a grassroots campaign against the nuclear weapons plans carried out by a newly founded organization called AMSA (Action Group against Swedish Atomic Weapons), forced Erlander to work out a compromise that could be accepted by proponents and opponents alike. Two options were presented to Parliament for consideration. One, known as the "device

THE UNITED STATES AND SWEDISH PLANS TO BUILD THE BOMB

231

program:' was to be followed if Sweden chose to acquire nuclear weapons; the other, the "protection program:' was to be pursued if Parliament said no to nuclear weapons development. This latter program would focus research on defensive preparations against possible nuclear attack rather than design of a Swedish nuclear device. The government adopted the recommendation of the defense bill drafting committee and maintained that Sweden for the time being still did not have to make a decision on nuclear weapons. A compromise bill, approved by Parliament in July 1958, gave FaA more funds to conduct defense-related research but postponed any decision on nuclear weapons development. In practice, this meant that the protection program was approved and the device program rejected. The primary argument for postponing a definitive decision was that Sweden could continue to monitor international developments concerning nuclear weapons and conduct nuclear-related research at the same time. On this view, Sweden would have time to switch to the device track if international arms control talks failed or the international security situation developed in a more threatening direction. 38 The evidence indicates that Prime Minister Erlander began to have doubts about equipping the Swedish military with nuclear weapons as early as 1957. 39 But Erlander prioritized the achievement of broad political consensus on the nuclear weapons issue. In late 1958, Erlander appointed members of both factions within his Social Democratic Party to an Atomic Weapons Committee, chaired by himself, to seek a consensus position. Its report, a year later, concluded that the current circumstances predominantly favored not producing nuclear weapons. But the committee did not call for a change in policy. A final decision would still be postponed while protection-related research continued. Bomb opponents were pleased that official government policy still did not support building nuclear weapons, while proponents realized that the concept of protection research served, for practical purposes, as a cover for the continuation of technical preparations that would enable Sweden to move quickly to nuclear weapons design should policy change. Parliament continued to support this "freedom of action" policy in subsequent years. 40 Although research continued, the military now started to question its previous support for the program. During 1961, the Swedish military command worked to formulate a new defense plan. The formerly strong consensus, within the military, in favor of equipping the Swedish defense forces with nuclear weapons was now beginning to disintegrate. There were several reasons for

232

THOMAS JONTER

this new disunity. One reason was the competition between different branches of the armed services. The army and the navy feared that they would lose out in the struggle for budget appropriations; the air force was expected to be awarded the bulk of additional nuclear-related budgetary resources since the nuclear weapons were primarily to be mounted on aircraft. But even within the air force there were growing doubts about the advantages of nuclear weapons possession. The air force had other costly projects to defend, among them the development of a new fighter aircraft. If the nuclear weapons plans were realized, one consequence might be the shelving of these other important future projects. Because of this intramilitary disunity, the nuclear weapons issue was not dealt with specifically in the defense review OB-62. Instead, the matter was delegated to a special, secret review board, the so-called Nuclear Device Group, a move that enabled the military command to maintain a united frontY In February 1962, the Nuclear Device Group presented its findings. In the published report, the group observed that the previous u.s. nuclear weapons doctrine of massive retaliation was no longer operative. The Kennedy administration had put forward new ideas about tlIe importance of conventional weaponry, as a way to avoid immediate escalation to nuclear use in a possible future war. This observation hinted that perhaps Sweden would do better to emphasize investments in conventional forces. For practical purposes, the report's findings implied a retreat from tlIe military's previous position that Sweden's defense forces must be equipped with nuclear weapons in order to achieve an adequate deterrent capacity. Despite this apparent retreat, however, the report still underscored the importance of maintaining freedom of action, and attention was still focused on tlIe possible production of tactical nuclear weaponry. The report concluded it would be politically impractical to obtain nuclear weapons from the United States, meaning that any nuclear weapons would have to be produced domestically and effectively closing the book on earlier Swedish efforts to acquire access to U.S. warheads. In its fallback proposal, the group envisioned producing 100 tactical devices, the first of which could be ready by 1972, unless steps were taken to accelerate the process. 42 Like tlIe Nuclear Device Group, the OB-62 report emphasized the enhancement of Sweden's conventional military capabilities, since an attack against Sweden would, at the time, in all likelihood be carried out witlI conventional weaponry. The report maintained that Swedish nuclear weapons remained a

THE UNITED STATES AND SWEDISH PLANS TO BUILD THE BOMB

233

viable option, but for practical purposes even the military was now discounting this possibilityY By the time of the supreme commander's next report, in 1965, the military retreated even further from its earlier advocacy of building nuclear weapons. The 1965 report stated that an acquisition decision would be a political matter, and merely requested funds to continue nuclear-related research. 44 Why did the Swedish military stop pushing for nuclear weapons? It was partly because of an increasing lack of support for such a program in Swedish domestic politics (described more fully below), but it likely also reflected awareness of important changes in U.S. policy in the early 1960s, including a decision to assist Sweden in the event of a Soviet attack.

u.s. NUCLEAR POLICY TOWARD SWEDEN, 1960-62: INFORMAL COMMUNICATION OF A SECURITY ASSURANCE

During 1960, the U.S. government adopted a firm policy of opposition to Swedish acquisition of nuclear weapons. Yet it also recognized a potential need to assist Sweden in the event of Soviet aggression, a decision that opened the door to a possible security guarantee. The key meeting took place on April 1, 1960, when President Eisenhower presided over a National Security Council (NSC) discussion of policy toward Scandinavia. A draft of new policy guidance for the region contained several items relevant to U.S. policy on Sweden's nuclear weapons plans. The resulting policy guidance, NSC 6006/1 issued on April 6, replaced the earlier NSC-121. The new guidance made it clear that the Swedish neutral position was accepted by the United States as something that would not preclude making Swedish defense efforts complementary to those of NATO: Under the present circumstances, Sweden's membership in NATO is not necessary to Western defense. It would contribute to the over-all defensive strength of the Western powers for Sweden to modernize its defense posture and to establish in Sweden an early warning air control and advanced weapons systems (without nuclear warheads) which are compatible with and complementary to those planned for installation in the territory of neighboring U.S. allies. 45

In other words, the United States wished to assist Sweden in building up a stronger conventional defense capability in a way that would de facto integrate Sweden militarily into the NATO framework without altering its political nonalignment. Against this backdrop, it would be unwise for Sweden to exhaust its limited financial and technical resources in order to acquire nuclear weapons, since such weapons already existed within NATO. Therefore, the NSC now took a firm position on the Swedish inquiries about the purchase of U.S. nuclear

234

THOMAS JONTER

warheads as well as its independent bomb program: "[Do 1not provide nuclear warheads; and discourage Sweden from producing its own nuclear weapons:'46 Although the Eisenhower administration made a firm decision to oppose all possible routes to Sweden acquiring the bomb, it also concluded that if Sweden were to be attacked by the Soviet Union, it would be in the national interest of the United States to assist Sweden. The draft NSC policy document discussed at the April 1 meeting contained a paragraph suggesting that the United States undertake preparations to offer such assistance, but this provision was not immediately adopted because of opposition from some of the president's advisors. For example, Secretary of State Herter was critical and "wondered how we could cooperate in the defense of Sweden without informing Sweden and the NATO countries in advance:' He also claimed that such a policy might lead to a situation, if the information became known, in which other neutral countries like Austria would request the same type of protectionY Further analysis of the consequences of such a security position was undertaken by the State Department and the Joint Chiefs of Staff jointly, leading to agreement on modified language. On November 10, 1960, the president approved the modified paragraph with the proviso that the planning of U.S. assistance to Sweden should be carried out on a unilateral basis, and not within the NATO policy framework. The official policy stated: In the event of general war with Soviet Bloc (a) seek to prevent Sweden, as long as it remains neutral, from giving any assistance to the Soviet Bloc, and (b) encourage and assist Sweden, without prejudice to U. S. commitments to NATO to resist Soviet Bloc attack against Sweden. In the event of Soviet Bloc aggression against Sweden alone, be prepared to come to the assistance of Sweden as a part of NATO or UN response to the aggression.'8

In 1962, a new version of "Guidelines for Policy and Operations-Sweden" replaced the NSC 6006/1 document from 1960.49 The most noteworthy change was a new formulation of what the United States intended to do if the Soviet Union attacked Sweden: "In the event of Soviet Bloc aggression against Sweden alone, we should undertake to come to the assistance of Sweden as a part of NATO or UN response to the aggression:' The formulation "be prepared" had been replaced by "should undertake;' which implied a stronger commitment. These statements have been interpreted by a number of researchers as constituting a U.S. security guarantee extended to Sweden. If Sweden were to abandon its nuclear weapons plans, the United States would in turn promise to place Sweden under the U.S. nuclear umbrella. 50 A Swedish security policy

THE UNITED STATES AND SWEDISH PLANS TO BUILD THE BOMB

235

commission report, published in 2002, refers to a 1962 document which said, in the commission's view, that Sweden would enjoy u.s. protection in the event of war and that Sweden was, for all practical purposes, under the umbrella of u.s. nuclear forces. 51 There is no documentation to prove that there was a formal security guarantee given by the u.s. government to the Swedish government. In fact, such a formal security assurance between the two states was more or less impossible to sign for both parties. First, an explicit security guarantee of that kind would have been risky for the United States. Why promise the government of nonaligned Sweden something that the United States might not be able to provide in a crisis situation? Such a guarantee would also create problems for the United States in relations with other NATO countries. Why should a neutral country be given the same security assurances as NATO members? It was therefore decided that the U.S. military assistance should be provided on a unilateral basis and not planned for within the NATO framework. Second, it is unlikely that the Swedish government, on its part, would have entered into a formal agreement with the United States given the official Swedish policy of nonalignment. If it was not a formal agreement, what was it? Indirect evidence suggests that it was an informal security assurance that was not communicated through intergovernmental channels. Rather, it appears there was a gentlemen's agreement between the United States and Sweden, which was communicated from leading U.S. military personnel to the Swedish military command. The common defense planning and expanded military technological cooperation between the United States and Sweden during the 1960s also led to the setting up of secret, direct lines of communication between the military command structures of both countries. In fact, much of Swedish military planning from 1960 onward was based on the notion that the United States would come to the rescue if Sweden were attacked by the Soviet Union. For example, in a strategic study carried out by the Swedish defense command in 1960, it was stated that Sweden would be able to count on nuclear weapons assistance from the United States within two days of a Soviet attack against Sweden with nuclear weapons. The basis for this planning was the idea that Sweden had a place under the U.S. nuclear umbrella. 52 Most intriguingly, the diary of Supreme Commander Nils Swedlund contains a vague mention in May 1960 that there seems to have been some statement from U.S. sources that the United States would assist Sweden with nuclear weapons in a war situation. In the diary, Swedlund asks himself how this statement should be interpreted. 53

236

THOMAS JONTER

The common defense planning and expanded military technological cooperation between the United States and Sweden during the 1960s might also explain why the Swedish military changed positions so swiftly on the nuclear weapons issue, and opted to concentrate on the development of Sweden's conventional defense capability. Strategic reasoning in botlI tlIe OB-62 and Nuclear Device Group reports was similar to that put forward in the 1960 and 1962 NSC documents, suggesting that the Swedish military was aware of and adjusting itself to the new, firmer position taken by tlIe United States. It is also likely that the Swedish government would have been aware of any gentlemen's agreement between the U.S. and Swedish military apparatuses. Both sides had much to gain from this arrangement. The United States did not need to issue a formal security guarantee that it might not be in a position to uphold in all circumstances, and Sweden did not have to make commitments that might jeopardize its policy of nonalignment. SWEDEN ABANDONS THE NUCLEAR WEAPONS OPTION

During the 1960s Sweden became increasingly involved in international efforts to halt the spread of nuclear weapons and advance the cause of nuclear disarmament. In 1962, Sweden became a member of the newly founded Eighteen Nation Disarmament Committee (ENDC). Sweden was also one of the first states to sign the Partial Test Ban Treaty (PTBT) in 1963, and when the ENDC became involved in negotiations for a treaty banning the spread of nuclear weapons, Swedish diplomats contributed strongly to the process that led to the final draft of the NPT.54 The international commitment to nuclear disarmament served to strengthen the arguments against Swedish nuclear weapons and enabled government opponents like Foreign Minister Osten Unden and Swedish antinuclear activists such as AMSA and the National Federation of Social Democratic Women to take more assertive positions. This international commitment to nuclear disarmament affected the domestic Swedish political debate to the extent that even the Conservative Party stopped openly promoting a decision to produce nuclear weapons, now merely calling for continued research. Swedish public opinion also shifted between 1957 and 1961 from majority support for building the bomb to majority opposition to acquiring nuclear weapons. 55 As a result of the international efforts to prevent nuclear proliferation and promote nuclear disarmament, an international norm against nuclear weapons started to emerge, according to Maria Rost Rublee. 56 Rublee's argument is sup-

THE UNITED STATES AND SWEDISH PLANS TO BUILD THE BOMB

237

ported by this study, although with the evidence presented here it cannot be determined whether Swedish politicians and the military were more influenced by the emerging nonproliferation norm per se or by other arguments raised in connection with the international talks on nuclear weapons. Whatever we choose to call the explanatory factor, the impact on Swedish decision-makers was the same: the international commitment to nuclear disarmament reinforced the domestic pressures against developing nuclear arms, which is also the core of Rublee's argument. The conflict between Swedish civilian and military nuclear objectives gradually became more pronounced as a function of economic realities. After 1965, when Sweden ordered its first light water reactor, the civilian nuclear energy program proceeded with little regard for military requirements. In 1966, an agreement was signed between the United States and Sweden in which the United States guaranteed the delivery of enriched uranium to Sweden until 1996. The intention was to load the planned Marviken reactor (which was later canceled) with enriched uranium. This would have resulted in all Swedish reactors being subjected to safeguards,S? and meant that a strictly military reactor program would have to be set up if Sweden were to be able to produce nuclear weapons in future. For this reason, the military leadership felt that it had to pursue a policy of phased procurement, if the policy of freedom of action was to be sustained. The government, however, in its 1966 planning refused to continue this policy, and thus Sweden's nuclear weapons planning was, for practical purposes, discontinued. 58 In 1968, the Swedish Parliament voted no to Swedish nuclear weapons, and the government signed the NPT in August of that year. Sweden's nuclear weapons aspirations were now definitely dead and buried. Finally, in 1970, Sweden ratified the NPT. As part of the rationale for forswearing nuclear weapons, government officials cited their belief that Sweden would be covered by a nuclear umbrella. When a governmental commission on defense appointed in 1965 issued its report in early 1968, it argued that acquiring nuclear weapons was not in Sweden's interests and that postponing a final decision no longer made sense. The parliamentary bill that renounced nuclear weapons largely adopted this commission report. Among other reasons for rejecting nuclear arms, the report stated that if either superpower attacked Sweden, it was likely to occur in the context of a broader East-West conflict and would therefore risk nuclear retaliation by the other side. Sweden was therefore in practice covered by the superpowers' nuclear umbrellas. 59

238

THOMAS JONTER

IMPLICATIONS FOR THE HYPOTHESES ON ASSURANCES

This case has mixed implications for the hypotheses presented by Knopf in Chapter 2. It implicitly supports hypothesis 1, as Sweden's interest in the bomb was largely the result of security concerns. The Swedish case is dis confirming evidence, however, for hypotheses 2 and 3, which suggest that publicly stated and legally binding assurances are more effective. In the case of nonaligned Sweden, the United States could not offer a public or legally binding commitment. The fact that U.S. assurances were conveyed privately and informally, or perhaps left entirely tacit, was essential. This same feature of the case provides some support for hypotheses 6 and 7, which concern defense cooperation and tailoring of assurances. Increased defense cooperation between the United States and Sweden is what made a security commitment possible, and only by accepting Sweden's policy of neutrality could the United States find a way to convey positive assurances. The case also provides partial support, while suggesting possible amendments, to tlIree other hypotlIeses. HypotlIesis 11 states: "The impact of assurances will depend in part on how tlIey affect tlIe perceived prestige and appropriateness of nuclear status. To the extent that norms matter, negative assurances are likely to be especially importane' It became more and more important for Sweden, from the mid-1950s on, to identify itself as a country that was engaged in tlIe struggle to abolish or reduce tlIe number of nuclear weapons. The more important nonproliferation norms became in tlIe international debate, tlIe more strongly Sweden promoted international cooperation to prevent the spread of nuclear weapons. For this reason, U.S. arguments that a Swedish bomb would encourage otlIer states to follow suit and complicate negotiations aimed at limiting the nuclear arms race found a receptive audience inside Sweden. Hypothesis 11 is not fully supported, however. AltlIough it might be argued tlIat Sweden was influenced by emerging nonproliferation norms, this did not make negative assurances important. Rather, Sweden still sought positive assurances. Hypothesis 12 also has some bearing on the Swedish case: "Assurances will be more effective if they can be utilized in a way that alters internal debates in the target in a favorable direction:' It seems that positive assurances informally communicated from U.S. to Swedish military officials affected the trajectory of internal debate. They did so more by reducing military interest in nuclear weapons, however, tlIan by bolstering tlIe position of antinuclear voices among Swedish decision-makers.

THE UNITED STATES AND SWEDISH PLANS TO BUILD THE BOMB

239

Hypothesis 13, ''Assurances will be more effective if they are packaged with positive incentives:' also receives some support from the Swedish case, but not exactly in the way envisioned by Knopf. In this case, economic incentives functioned in a more coercive manner-tlIat is, the threat that the United States might withdraw assistance was more important than the promise of new assistance. On the Swedish side, the perception was that the price Sweden must pay in exchange for receiving much -needed military technology and nuclear energy assistance was the shelving of its nuclear weapons plans. CONCLUSIONS

The choice to integrate potential Swedish production of nuclear weapons within the civilian nuclear energy program meant tlIat Sweden, in spite of its intentions, became dependent on U.S. technology. This technological dependence increased over time, and afforded the United States the opportunity to steer Sweden away from using its civilian program for the production of weapons-grade plutonium. The most important step taken by the United States was the lowering of the price of enriched uranium in the late 1950S. As a result of this measure, light water reactor technology became an economically attractive alternative to the heavy water program for producing nuclear power in Sweden, thus opening the door for private business to build reactors. In tlIe end, even the reactor that was meant to operate with heavy water and produce plutonium for weapons was instead loaded with enriched uranium from the United States as a result of economic considerations. From then on, the only way to acquire nuclear weapons would have been to start a separate military program for tlIe production of weapons-grade plutonium witlI no ties to assistance from the United States. In tlIe early-to-mid 1960s, such a policy had no prospects of being adopted, for a number of reasons. First, leading politicians, especially in the Social Dem0cratic government, had by tlIen changed their minds about acquiring nuclear weapons, as had most of tlIe Swedish public. Second, Swedish dependence on U.S. technology would in practical terms make it impossible to develop a wholly separate military nuclear weapons program unless Sweden were willing to break away from that dependence. Third, even leading Swedish military officials had begun questioning tlIe idea of equipping the Swedish defense with nuclear weapons. This changed view among the Swedish military needs to be understood against the backdrop of the increasing connection of Swedish defense planning to U.S. and NATO strategy. In this new security policy context,

240

THOMAS JONTER

it seemed meaningless to drain Sweden's limited financial resources in order to develop nuclear weapons, when Sweden was for practical purposes integrated into the U.S. and NATO security system. The Swedish military was convinced that Sweden was covered by the u.S. nuclear umbrella, and they were therefore willing to give up their nuclear weapons plans. From the u.S. perspective, a security assurance communicated to the Swedish military could be considered as a final piece in the U.S. policy of steering Sweden away from its nuclear weapons plans. Even if previous actions to create and leverage a dependence on U.S. technology-especially through lowering of the price of enriched uranium-had been rather successful, there was a risk, even if small, that the Swedes would launch their own military nuclear weapons program. By communicating an informal assurance to include Sweden under the U.S. nuclear umbrella, the risk of Sweden's going nuclear was reduced. In the eyes of the United States, it was not yet obvious that the Swedish political elite was moving away from a policy of acquiring nuclear weapons, and even if it had been it still made sense to help Sweden avoid being a possible object of Soviet coercion or aggression. The Swedish government probably had information to suggest that there was a gentlemen's agreement between the u.S. and Swedish militaryestablishments. However, it was in both the u.S. and the Swedish governments' interests to play the game through military channels. The United States was not forced to provide Sweden with a formal security guarantee that it might not have been in a position to uphold in a crisis situation and that would complicate relations with other European states. Sweden, for its part, was not required to make commitments that might force it to give up its policy of neutrality. Official government statements never mentioned a U.S. policy commitment to Sweden, but the fact they expressed such confidence that Sweden was covered by a nuclear umbrella may be in part the result of learning through informal channels that the United States had adopted a policy to assist Sweden in the event of war. Positive security assurances were not the most important factor in the Swedish decision to abandon work on nuclear weapons. International efforts to abolish and prevent the spread of nuclear weapons, domestic opposition, and technological dependence on the United States-arising in part from the original decision to tie the military program to civilian nuclear energy developmentall mattered more. But assurances did have a modest impact through reinforcing these other factors. While dependence on the United States increased the costs of an independent nuclear weapons program, a perceived security guar-

THE UNITED STATES AND SWEDISH PLANS TO BUILD THE BOMB

241

antee reduced the necessity of pursuing that program. The belief in a nuclear umbrella also made it easier for the main proponents of a Swedish bomb in the Conservative Party and the military to accept the changed attitudes of the other political parties and the public in favor of renouncing nuclear weapons. NOTES 1. "Why Sweden Did Not Build the Bomb, 1945-1968;' paper presented at the Nobel Symposium on "Peace, Stability, and Nuclear Order;' Oscarsborg, Dr0bak, Norway, June 25-27, 2009. For prior analyses of the Swedish technical preparations to acquire nuclear weapons, see also other publications by the same author: Thomas Jonter, "The Swedish Plans to Acquire Nuclear Weapons, 1945-1968: An Analysis of the Technical Preparations;' Science and Global Security 18, no. 2 (2010); Sverige, USA och kiimenergin.

Framviixten av en svensk kiimiimneskontro1l1945-1995 ("Sweden, the United States and Nuclear Energy: The Emergence of Swedish Nuclear Materials Control 1945-1995"), Swedish Nuclear Power Inspectorate (SKI), SKI Report 99:21, May 1999; Sweden and the Bomb: The Swedish Plans to Acquire Nuclear Weapons, 1945-1972, SKI Report 01:33, September 2001; Nuclear Weapons Research in Sweden: Co-operation between Civilian and Military Research, 1947-1972, SKI Report 02:18, May 2002. Unless otherwise indicated, the information and analysis in the rest of this chapter draws from the papers and reports listed here. 2. Two government commissions have investigated Swedish-U.S. military cooperation during the Cold War. The first is titled Om kriget kommit: farberedleser for mottagande av militiirt bistand 1949-1969 (Statens offentliga utredningar ["Official Government Reports"; hereinafter, SOU], 1994). There is an English summary of the report,

Had There Been a War . .. Preparations for the Reception ofMilitary Assistance 1949-1969. Report of the Commission on Neutrality (Stockholm: Fritzes, 1994). The second government study is titled Fred och siikerhetr. Svensk siikerhetspolitik 1969-1989: Slutbetiinkamde. Dell ("Peace and Security: Swedish Security Policy 1969-1989:' Final report. Part 1) (SOU, 2002: 108). See also Jonter, Det amerikanska sparet. En undersakning om IB:s bildande och eventuella kopplingar till USA ("The American Trace: An Investigation on the Establishment ofIB and Its Eventual Connection to the United States") (SOU, 2002: 95); Mikael Nilsson, Tools ofHegemony: Military Technology and Swedish-American Rela-

tions 1945-1962 (Stockholm: Santerus Academic Press, 2007). 3. Om kriget kommit, 11. 4. Jonter, Sweden and the Bomb, 21. 5. Swedish National Defense Research Institute (FOA) archive, Outgoing documents 1948 B IV, vol. 4, H 35:2

6. Gunnar Skogmar, De nya malmfiilten. Det svenska uranet och inledningen till efterkrigstidens neutralitetspolitik ("The New Ore Fields: Swedish Uranium and the Intro-

242

THOMAS JONTER

duction of the Post -war Neutrality Policy") (Goteborg: Statsvetenskapliga institutionen, 1997)· 7. Jonter, Sverige, USA och karnenergin. 8. "From Gullion to Cumming;' RSASSEAM, box 82, U.S. National Archives (NA), Washington, DC.

9. Om kriget kommit. Paul M. Cole, Sweden without the Bomb: The Conduct of a Nuclear-Capable Nation without Nuclear Weapons (Santa Monica, CA: RAND, 1994), 233ff. 11. Gunnar Skogmar, Sambandet mellan militart och civilt utnyttjande av atomenergin i amerikansk utrikespolitik 1945-1973 ("The Interdependence between the Military 10.

and Civil Use of Atomic Energy in American Foreign Policy 1945-1973"), Ph.D. diss., Lund University, 1979, 71 and passim. 12. "From Dulles to Amembassy;' May 19, 1955, 611.589715-1955; "Memorandum of Conversation;' May 24,1955,611.5897-2455, NA. 13. "Sveriges overenskommelser med frammande makter" (Swedish Treaty Series),

sO 1956:67-68, Stockholm 1956. 14. "Memorandum of Conversation;' November 16, 1955, RG 59, SASES, box 436, NA. 15. "From Robinson to Smith;' November 30, 1955, Rg 59, SASES, box 436, NA. 16. "From Boheman to Unden;' May 20, 1957, Eteh 1957-1960 01:3, Studwiks arkiv (Archive of Studsvik, hereinafter SA), Studsvik AB, Nykoping. 17. "From Odhner to UD;' May 20, 1957, Eth 1957-1960 01:3, SA. 18. "Mojligheterna att hiHia R3/Adam inspektionsfri" ("Possibilities to Keep R31 Adam Inspection-free"), February 5,1959; ''AE Utredningar om Tungt vatten 1957-1967, 1970-1974" (''AE Investigations on Heavy Water 1957-1967, 1970-1974"); "Uran 1956-1962, Allmant 1957-1959 Prognoser 1960" ("Uranium 1956-1962, Miscellaneous 1957-1959 Prognosis"), Direktionsarkivet, StudsvikAB, Nykoping. 19. For example, the publicly owned corporation Vattenfall, which would be responsible for the future operation of the projected heavy water reactors and facilities, complained about these restrictions in a letter to the Swedish Authority for Control of Nuclear Energy (DFA, "Vattenfall till DFA;' October 8, 1957, H6156, Hp 120, Utrikesdepartementets arkiv [Archive of the Ministry of Foreign Affairs; hereinafter, UA]). 20. The DFA grew out of an earlier advisory body, the Atornkommitten (AK, Atomic Committee). In the mid-1950S, theAK's responsibility was split into two functions: DFA was assigned responsibility for regulating safety and issuing licenses in the nuclear energy field, while "Statens nld for atomforskning" was responsible for advising on basic research. The DFA was transformed into the Swedish Nuclear Power Inspectorate (SKI) in 1974. 21. "P.M. angaende eventuell sarskild forhandling med United States Atomic Energy Commission (AEC) om 90%-igt uran 235 till reaktorn R2" ("Memorandum Concerning Eventual Special Negotiation with the United States Atomic Energy Commission [AEC] on 90% Uranium to the Reactor R2"), October 7, 1957, 22/1/93, Hp 120, UA.

THE UNITED STATES AND SWEDISH PLANS TO BUILD THE BOMB

243

22. "From the Swedish Embassy in Washington to Minister of Foreign Affairs Osten

Unden;' April 25, 1958, Eteh 1957-1960 01:3, SA. 23. "From AEC to the President of the United States;' March 27, 1958, SASES, box 43 6,NA. 24. "From Boheman to Unden;' April 28, 1958, Eteh 1957-1960 01:3, SA. 25. "Utredning av betingelserna for framstallning av atomvapen i Sverige" ("Inquiry

into the Conditions for the Production of Atomic Weapons in Sweden"), by Torsten Magnusson, November 25, 1955, 87- H 163:1-2IA, the FOA archive. 26. Stefan Lindstrom, Hela nationens tacksamhet: svensk forskningspolitik pa atomenergiomradet 1945-1956 ("The Entire Nation's Gratitude: Swedish Research Policy in the Atomic Energy Field 1945-1956);' Ph.D. diss., Stockholm University, 1991, 161. 27. Anna-Greta Hoadley Nilsson, Atomvapnet som partiproblem ("The Atomic Weapon as a Party Problem") (Stockholm: Almqvist and Wiksell International, 1989), 26-27. 28. This step was the most important U.S. action to hinder nuclear weapons pro-

duction in Sweden, according to Bo Aier, former CEO for AB Atomenergi. Interview with Bo Aler, January 18, 2002. 29. Jonter, Sverige, USA och kiimenergin, 23. 30. Nilsson, Tools of Hegemony, 343-54, 401-15; see also U.S. State Department, "Guidelines for Policy and Operations-Sweden" from 1962, where it says: "21. Help

Sweden maintain a strong defense force by enabling her to purchase modern weapons system not dependent upon nuclear warheads from U.S. production sources;' June 1962, RG 59, Records of the Policy Planning Staff 1962, Lot File 69 D 121, Box 218, NA. 31. Jonter, Det amerikanska spa ret; Nilsson, Tools of Hegemony. 32. Jonter, Sverige, USA och kiimenergin, 30-38. 33. "Conversation with the Swedish Ambassador;' June 26, 1957, Special Assistant to

the Secretary for Energy and Outer Space. Records Relating to Atomic Energy Matters, 1944-63, Box 437, NA. 34. Cole, Sweden without the Bomb, 36-41, 44-53. 35. May 15, 1959, SASES, box 437, NA. 36. Ibid. 37. The Agesta reactor came on line in 1964 and operated until 1974. Marviken never became operational and work on it ceased in 1970. 38. Jonter, Sweden and the Bomb, 42-46. 39. Nilsson, Atomvapnet som partiproblem, 43. Some see signs of hesitancy even earlier, in comments in Parliament in 1955 when Erlander said he would not like to see Sweden become the fourth nuclear weapon state, but others believe that Erlander was

initially pronuclear and only changed his mind later. Compare Jerome Garris, "Sweden's Debate on the Proliferation of Nuclear Weapons;' Cooperation and Conflict 8 (1973): 192; and Eric Arnett, "Norms and Nuclear Proliferation: Sweden's Lessons for Assessing Iran;' Nonproliferation Review 5, no. 2 (Winter 1998): 35-36.

244

THOMAS JONTER

40. Garris, "Sweden's Debate;' 195-98; Cole, Sweden without the Bomb, 96-100. 41. Wilhelm Agrell, Svenska JOrintelsevapen. Utvecklingen av kemiska och nukleara

stridsmedel 1928-70 ("Swedish Weapons of Mass Destruction: The Development of Chemical and Nuclear Weapons") (Lund: Historiska media, 2002), 281-87. 42. "Karnladdningsgruppens betankande" ("Report of the Nuclear Device Group"), HHo06. 43. Wilhelm Agrell, Alliansfrihet och atombomber: kontinuitet och forandring i den svenska JOrsvarsdoktrinen 1945-1982 ("Nonalignment and Atomic Bombs: Continuity and Change in the Swedish Defense Doctrine 1945-1982") (Stockholm: Liber f6rlag, 1985),237. 44. Garris, "Sweden's Debate;' 199. 45. "U.S. Policy toward Scandinavia (Denmark, Norway and Sweden);' April 6, 1960, RG 273, NSC 6006/1, box 51, NA. 46. Ibid. 47. Discussion at the 439th Meeting of the National Security Council, Friday, April 1,1960. The document was quoted from the appendix to the Commission of Neutrality,

Om kriget kommit, 11. 48. "Statement of U.S. Policy toward Scandinavia (Denmark, Norway and Sweden;' NSC 6006/1, April 6, 1960, in U.S. Department of State, Foreign Relations of the United

States, 1958-1960: Western Europe, vol. 8, part 2, 679. 49. Notice to holders ofNSC 6006/1 by Bromley Smith, "Rescission ofNSC 6006/1, U.S. Policy toward Scandinavia (Denmark, Norway and Sweden), dated April 6,1960;' May 2, 1962, RG 59, Records Relating to Department of State participation in the Operations Coordinating Board and the NSC, 1947-1963, Lot File 63 D 351, Box 99, NA. 50. Agrell, Svenska forintelsevapen, 297-306; Nilsson, Tools of Hegemony; Simon Moores, "'Neutral on Our Side': U.S.-Swedish Military and Security Relations during the Eisenhower Administration;' Ph.D. diss., London School of Economics, September 2004,220-21. 51. Fred och sakerhet. Svensk sakerhetspolitik 1969-89 (SOU, 2002: 108), part 1, 221-23. For a lengthy discussion on how the 1962 policy guidelines should be interpreted in

Sweden, see Jerker Widen, Vaktare, ombud, kritiker. Sverige i amerikanskt sakerhetstank-

ande 1961-1968 ("Guardian, Ombudsman, Critic: Sweden in American Security Thinking 1961-1968") (Stockholm: Santerus Academic Press, 2009). 52. Agrell, Svenska forintelsevapen, 303; Widen, Vaktare, ombud, kritiker, 63-93. 53. "PM 17/5 60" (Pro memoria, May 17, 1960), "Nils Swedlunds arkiv;' vol. 2 C, Krigsarkivet (Military Archives), Stockholm. 54. Mitchell Reiss, Without the Bomb: The Politics of Nuclear Nonproliferation (New York: Columbia University, 1988), 67. 55. Garris, "Sweden's Debate;' 198. For more on Swedish disarmament policy in the beginning of the 1960s, see Stellan Andersson, Den JOrsta grinden: svensk nedrustning-

THE UNITED STATES AND SWEDISH PLANS TO BUILD THE BOMB

245

spolitik 1961-1963 ("The First Gate: Swedish Disarmament Policy 1961-1963") (Stockholm: Santerus, 2004), 86. 56. Maria Rost Rublee, Nonproliferation Norms: Why States Choose Nuclear Restraint (Athens: University of Georgia Press, 2009),34-52. 57. R2 and R3 were already under safeguards as a result of the 1958 agreement to purchase U.S. enriched uranium. 58. Jonter, Sweden and the Bomb, 62-63. 59. Garris, "Sweden's Debate;' 200; Martin Frehm, "Sweden;' in Non-proliferation:

The Why and the Wherefore, ed. Jozef Goldblat (London: Taylor and Francis, 1985), 21718.

The "Model" of Ukrainian Denuclearization Sherman W. Garnett

When voters in Ukraine approved an independence referendum on December 1, 1991, helping precipitate the final breakup of the Soviet Union, the newly established state was, as many have put it, "born nuclear:' This new state inherited what would have been the third largest nuclear arsenal on the planet, but could or would these weapons become Ukrainian? Proponents on both sides of this question could be found both inside and outside Ukraine. Yet within a little more than two years, Ukraine reached a Trilateral Agreement with Russia and the United States under which it agreed to give up nuclear weapons. Within four and a half years, Ukraine announced that the last nuclear weapons on its soil had been removed to Russia. Security assurances were crucial in this process, but they were only effective because they were defined broadly and used in combination with a package of measures designed to bolster Ukrainian independence, address the new state's security anxieties, and support tlIe new state and the denuclearization process, in particular economically. This chapter describes the "model" for denuclearization established by the Ukraine case and its implications for this volume's hypotheses on assurances and for future nonproliferation efforts. The word "model" is in quotation marks because, although widely and correctly seen as a nonproliferation success, Ukraine's decision to abandon its nuclear arsenal is a highly unusual case. Five factors complicated for all sides the task of understanding the problem and negotiating a solution, and they may affect Ukraine's relevance to future proliferation cases. These factors, to be discussed in turn, were the sequencing in which proliferation preceded statehood, the question of operational control, the relation between nuclear issues 246

THE "MODEL" OF UKRAINIAN DENUCLEARIZATION

247

and state-building, the legacy of Cold War nuclear arms control treaties, and the wide-ranging nature of the u.s. policy response. First, Ukraine was a major nuclear power before it was a full-fledged state, reversing the normal nonproliferation dynamics. When Ukraine announced its independence, 130 SS-19 and 46 SS-24 missiles from the Soviet armed forces were deployed within its borders. Air bases within Ukraine hosted an additional 44 heavy bombers and their air-launched cruise missiles. By the counting rules of the July 1991 Strategic Arms Reduction Treaty (START), these systems had associated with them more than 1,900 strategic nuclear warheads. In addition, Ukraine also had an estimated 2,275 to 4,200 tactical warheads on its territory.l After the Soviet Union collapsed, these systems and their basing sites and command and control facilities remained under the control of the Moscow-based Strategic Rocket Forces. However, they could not be isolated from Ukrainian political and military influence, regulation, or ultimately even direct control forever. Ukraine thus had a full-fledged nuclear capacity before it had a set of basic laws or even a fully staffed foreign ministry. Second, the ambiguous status of these nuclear weapons and untested assumptions about the post-Soviet space made some observers doubt whether Ukrainian nuclear weapons were an issue, even as others, such as John Mearsheimer, argued that Ukraine should and would become a major nuclear power.2 While experts may question whether Ukraine could have ever seized operational control of this arsenal, there is little doubt that the Ukrainian leadership could have acted in a way that would have kept the issue of nuclear weapons there alive to the present day. All that was needed for a stalemate or worse was a more recalcitrant Ukraine, a less accommodating Russia, and a less far-sighted effort by the United States. Third, Ukraine's potential nuclear status was inextricably linked in tlIe minds of the Ukrainian leadership to tlIe task of building a new and stable state. For Ukraine, nuclear weapons were less important than the overarching goal of achieving sovereignty and sustaining it over time in the uncertain environment created by the fall of the USSR. However, for Russia, tlIe United States, and most of the rest of the world, avoiding the emergence of new nuclear powers in Ukraine, Belarus, and Kazakhstan was tlIe more urgent matter. The fact that Ukraine was preoccupied by the challenges of state-building, however, required those aiming for Ukrainian nuclear disarmament to take up a range of issues well beyond the arena of Cold War arms control and nonproliferation efforts. Ukraine provided a dress rehearsal for a messier world, in which familiar non-

248

SHERMAN W. GARNETT

proliferation problems are often entwined in issues of state-building, defense assistance, economic aid, and regional stability. Fourth, Ukraine's inherited nuclear arsenal was for Russia and the United States both a traditional proliferation challenge and a complication for nuclear arms control agreements such as the START treaties. START I had been signed by the Soviet Union, raising the question of how the Soviet successor states that hosted nuclear weapons on their territory, including Belarus and Kazakhstan along with Russia and Ukraine, would be incorporated into the arms control arrangement. The overwhelming preference of both Moscow and Washington was for Russia to be the sole inheritor of the Soviet Union's nuclear status, while the other three states would move beyond START to quickly join the NPT as non-nuclear parties. Just as Russian and U.S. policymakers had to come up to speed on domestic and state-building issues in Ukraine, so Ukrainian policymakers had to become expert on strategic arms control. Fifth, this intermingling of nuclear and other political, economic, and security issues created a final agreement that included many items not traditionally addressed in nonproliferation cases. It compelled all sides to adopt a broader understanding of the issues and an expansive policy response. Many policymakers in the United States and elsewhere would have been happier to observe from afar the processes of state-building on the post-Soviet space. The circumstances of Ukraine's potential arsenal forced Washington to improvise and expand its Cold War nuclear and nonproliferation horizons. It created a policy process that inspired relevant officials to move beyond the confines of anyone policy agenda, even as that process remained clear about denuclearization as the overriding aim. The result was a better agreement than could have been obtained if Russia and the United States had focused solely on the tried-andtrue issues of arms control and nonproliferation. Links between nuclear nonproliferation and regional politics, economics, and security were not new, nor were they unique to the Ukrainian case. The United States, however, had never before gone so far in linking its nuclear policy to a broad-based engagement in the politics, economics, and security of another state that is not an ally. In creating a process and ultimately an agreement, Washington adapted and expanded tools from different policy arenas. Security assurances turned out to be an especially flexible tool, precisely because they had both a concrete meaning relevant to the core issue of denuclearization but also could become an umbrella for a wider U.S. security engagement in Ukraine. Security assurances enabled the United States to enter in the middle of an evolving security rela-

THE "MODEL" OF UKRAINIAN DENUCLEARIZATION

249

tionship between Kyiv and Moscow, where paramount security and nonproliferation interests were clearly at stake, without either requiring the United States to have to declare Ukrainian independence a lost cause or forcing Washington to surrender its nuclear goals or hopes for a better relationship with Russia. With these circumstances in mind, this chapter explores the challenges, negotiations, and policy adjustments that led to Ukrainian denuclearization. In keeping with the focus of this volume, it examines the role of the security assurances offered to Ukraine. Discussion starts with the domestic and foreign policy context shaping Ukraine's view of the nuclear issue; then moves to the internal Ukrainian nuclear debate and the trilateral negotiation that gave rise to the security assurances and related agreements; and then summarizes the security assurances themselves, including the way they are shaping security and politics in the region. Though treating the period from pre-independence to the implementation of the denuclearization agreement and beyond, this chapter focuses on the year and a half of intense negotiations that led to the 1994 Trilateral Agreement. 3 The assurances in this agreement were primarily of a negative character but went beyond nuclear-related assurances to incorporate pledges against conventional military and economic threats. In addition, by incorporating the assurances in a multilateral context, the denuclearization agreement with Ukraine implied possible positive assurances that outside powers would assist Ukraine in managing its bilateral relations with Russia. This chapter concludes by evaluating implications of the case for the hypotheses derived in Chapter 2 and assessing whether or not Ukrainian denuclearization can serve as a model for addressing other nonproliferation challenges. THE DOMESTIC AND FOREIGN POLICY CONTEXT

In the case of Ukraine, an existing nuclear arsenal acquired a state. From the very first, the nuclear question thus intruded on Ukrainian efforts to build a state and establish stable foreign relations, particularly with Russia. Domestic and foreign policy matters in turn influenced the way Ukrainian leaders saw the nuclear question. Ukrainian domestic forces shaped the nuclear debate and the negotiation of security assurances in three primary ways: by encouraging moderation, by bringing economic issues to the fore, and by making state-building a priority. First, Ukraine's ethnic, political, and economic diversity created incentives for moderation in Ukrainian security policy in general, and nuclear policy in particular. On the nuclear issue and on many other security matters, this di-

250

SHERMAN W. GARNETT

versity forced the marginalization of extremist views. In the case of Ukrainian denuclearization, both nuclear hawks and those arguing for reintegration with Russia were politically irrelevant. Far from being a fifth column, the large ethnic Russian minority in Ukraine was fully integrated into the political and economic life of the country.4 Polls from May and August 1993 showed that, while a majority of Ukrainian citizens considered Russian policy toward Ukraine as coercive, only a third of the population supported the retention of nuclear weapons as a deterrent. 5 Because it would involve siding with one extreme in domestic opinion, a nuclear option would have been destabilizing internally. This fact was understood from the very beginning by the Ukrainian executive branch, though a wide range of commentators sought to portray Ukraine's ethnic and geographical divides as a fatal weakness. 6 Second, the deepening economic crisis gradually became the dominant political issue in Ukraine. Massive strikes led by coal miners in 1993 made this abundantly clear. The steady decline of the Ukrainian economy following independence undermined those who argued that military security, not the economy, should be at the center of the political debate. Ukraine needed outside political and economic assistance to address its internal needs and to establish stability in Ukrainian-Russian relations. Ordinary Ukrainians and leaders alike understood that this assistance would not be forthcoming from the United States if the country attempted to retain nuclear weapons. Third, throughout this period, the task of state-building dominated Ukrainian politics. Though some politicians tried to cast nuclear weapons as symbols and guarantors of Ukrainian sovereignty, and nearly all were content to use them as bargaining chips, the nuclear issue never became a true faultline in Ukrainian politics. The economy, the domestic struggle for power, and institutional stability were in the end the true security issues, for which nuclear weapons had little relevance, then or now. Turning to the foreign policy context, Russia was Ukraine's central partner, interlocutor, and problem. Russia was the key external factor at the heart of the Ukrainian nuclear debate. However, beneath the surface tensions, the relationship was often more cordial and pragmatic than the rhetoric has suggested and many in the West have feared. For example, in the early 1990S, the Russian government kept Crimean separatists at arm's length at a time when its own nationalist organizations and local Crimean political movements loudly proclaimed the need for a referendum on independence. Over the past twenty years, Russia has repeatedly used its dominant position on existing debts or

THE "MODEL" OF UKRAINIAN DENUCLEARIZATION

251

energy, but it has just as often accepted compromise solutions that prevented these issues from reaching a serious crisis. Most important, Moscow did not or could not wield its oil and gas as an instrument of coercion on nuclear matters.? At bottom, Russia regards Ukraine as a state that ought to be its partner and which lies within its zone of primary interests. Throughout much of the period of the nuclear negotiations, Russia's own problems and diminished capacity prevented it from succeeding in various attempts to integrate Ukraine more fully into the Commonwealth of Independent States (CIS) or to make Ukraine part of its special zone of interests. In the 1990S, Russia was simply not in a position to effect this kind of change of Ukrainian policy, though Russia's re-emergence under Putin and Medvedev suggests a new round of the game is underway. For Ukraine, the central problem was its relative weakness visa-vis Russia. Bilateral negotiations over key questions inevitably reflected this distribution of power: Ukraine did all it could to resist Russian-imposed solutions but could not obtain the permanent settlement it desired. This dynamic opened the door for the United States to play the role of a third-party balancer and mediator. However, such a role for Washington was unthinkable if Ukraine sought to remain a nuclear power. The basic tensions in the relationship and the difficulties the two sides faced in achieving a bilateral solution to the nuclear issue were well illustrated at the September 1993 Russian-Ukrainian summit meeting at Massandra, in the Crimea. Russia and Ukraine had been attempting since the collapse of the Soviet Union to negotiate a wide range of issues bilaterally. Massandra was hence overloaded with unresolved issues, including the nuclear question, the division and stationing of the Soviet Black Sea Fleet, mounting Ukrainian debt to Russia, and the recognition of Ukraine's borders. At this summit, Russian President Boris Yeltsin and his advisors exerted considerable pressure on Ukrainian President Leonid Kravchuk to agree to a comprehensive deal on the Black Sea Fleet, nuclear disarmament, and debt relief. There is evidence that senior Ukrainian officials agreed to such a package in advance. The centerpiece of this Russian package was a swap of at least partial debt forgiveness for Ukraine's share of dIe Black Sea Fleet, leaving the entire fleet to Russia while continuing to base it in what is now the Ukrainian port of SevastopoLB Once this package became public, it split the Ukrainian government so severely that any hope for the agreement quickly unraveled. Then Prime Minister Leonid Kuchma, one of those who favored the swap of the Ukrainian share of

252

SHERMAN W. GARNETT

the Black Sea Fleet for debt relief, resigned in early September. An open split also developed between Kravchuk and his defense minister, Konstantin Morozov, who resigned in October 1993. Morozov urged the immediate withdrawal of the Russian portion of the fleet from Crimea. In the aftermath of Massandra, the Russian side saw that Ukrainian weakness was not simply something to be exploited but rather a potential source of unwanted instability within Ukraine, with obvious and negative consequences for the Russian-Ukrainian relationship. Both Russia and Ukraine came away from the Massandra Summit understanding that for the Ukrainian side to move forward, it needed financial and other benefits that could only be provided in a multilateral framework. Kravchuk's experience at Massandra likely played a key role in his decision to accelerate negotiations with the United States and seek a trilateral framework for nuclear and security issues. THE NUCLEAR ISSUE: FROM INDEPENDENCE TO THE TRILATERAL AGREEMENT

The nuclear question in Ukraine unfolded in three distinct phases: an opening "anti-nuclear" phase influenced by memories of the Chernobyl nuclear power plant accident; a second period of tough Ukrainian bargaining, including claims and threats that raised questions in Washington about Kyiv's commitment to denuclearization; and, finally, the period of implementing the obligations of the Trilateral Agreement. The first phase lasted from preindependence until approximately mid-1992. Even before Ukraine achieved independence, the Ukrainian parliament, the Verkhovna Rada, issued a declaration of state sovereignty in July 1990 in which it proclaimed Ukraine's intention to become a non-nuclear state, a pledge it reaffirmed in October 1991. Washington linked its diplomatic recognition of Ukraine, which it granted at the end of 1991, to fulfillment of this pledge. Agreements reached in late December at two summit meetings of the recently formed CIS appeared to set Ukraine on a non-nuclear course. In declarations from meetings in Alma Ata and Minsk, Ukraine, Kazakhstan, and Belarus pledged to remove all tactical and strategic nuclear weapons from their territories and to join the NPT as non-nuclear weapon states. Some members of the Rada began to question the wisdom of these agreements, which the Rada never ratified. One complaint was that Ukraine had given away the tactical warheads within its territory while getting nothing in return. Kyiv now began requesting compensation for the highly enriched ura-

THE "MODEL" OF UKRAINIAN DENUCLEARIZATION

253

nium (HEU) contained in the warheads, and in March 1992 called for a suspension in the process of removing tactical weapons. Russia basically ignored this call, completing the withdrawal of all tactical weapons in early May, a move that caused Ukraine to dig in its heels that much harder on the issue of strategic warheads. The George H. W. Bush administration had little patience for this foot-dragging, and it pressured President Kravchuk to agree to a legally binding commitment that would ensure a non-nuclear Ukraine. An agreement to clarify the status of Soviet successor states in the START treaty provided the opportunity to extract such a commitment. In the Lisbon Protocol to START I, signed on May 23, 1992, the United States agreed that Russia, Ukraine, Kazakhstan, and Belarus were successors to the Soviet Union, and they all agreed to be bound by START. The protocol also required Ukraine, Kazakhstan, and Belarus to join the NPT as non-nuclear states "in the shortest possible time:' In separate side letters, these three states also promised to return all nuclear weapons on their territories to Russia. 9 In Washington, the agreement on the Lisbon Protocol appeared to end the issue, but just as the principle was resolved, the Ukrainian leadership came face to face with the realities of statehood. Internally, economic hardship and regional tensions loomed large. The relationship with Russia hit its first real hurdles, with tensions surfacing in spring 1992 over the Black Sea Fleet, a Russian-oriented political movement in Crimea, and other issues. Further, the leadership was slowly coming to understand the massive costs associated with fulfilling its nuclear commitments, such as those involved in dismantling strategic systems and transporting the warheads to Russia. The first phase of the nuclear question in Ukraine thus ended at about the time many in the West assumed the nuclear question was resolved. The second phase of questioning and negotiation soon followed, in which the key players in the Ukrainian leadership looked in earnest at the country's economic and security needs and came to value nuclear weapons as a potential bargaining chip. This phase lasted from mid-1992 until the conclusion of the Trilateral Agreement in January 1994. It featured the emergence of strong pronuclear sentiment among some in the nationalist parties. The Ukrainian leadership at times made use of this sentiment in its policy pronouncements, to the consternation of Washington. During this time, the Ukrainian government also made public its requirements for financial assistance and security guarantees. The third phase involved parliamentary concurrence in the Trilateral Agree-

254

SHERMAN W. GARNETT

ment and subsequent obligations. It began in January 1994 and lasted as a nuclear policy issue until the removal of the last warhead from Ukraine in 1996 or perhaps until Ukraine became fully compliant with START II in 2001. However, as a question of regional security and internal economic and political transformation, this phase continues to the present day. Highlights of this period include the Rada's November 16, 1994, vote to ratify accession to the NPT as a non-nuclear weapon state, the physical dismantling and transfer of the nuclear stockpile, and a burgeoning u.S.-Ukrainian security relationship. Our immediate concern is with the second phase, from mid-1992 to the conclusion of the Trilateral Agreement in 1994, along with follow-up steps to secure the Rada's acceptance. This period witnessed numerous bumps in the road: disruptions in the negotiation and implementation of Russian-Ukrainian agreements on nuclear disarmament, Ukrainian steps to assert administrative control over missile bases in Ukraine, and other statements and actions that increased anxiety in Moscow and Washington. These actions often had a source other than Ukrainian nuclear ambitions. I have argued elsewhere that the Kravchuk government-as opposed to various parties and individuals-never seriously contemplated permanent retention of nuclear weapons, particularly after the Russian withdrawal of all tactical nuclear weapons stationed in Ukraine in the spring of 1992.10 Each of these steps away from what in Washington would be defined as a smooth negotiation may be directly traced to the internal and external pressures on Ukraine described above. They grew out of Ukraine's twin preoccupations of building a state and asserting its full sovereignty and military independence from Russia. In late 1992, the Ukrainian government made plain that, while it sought to rid itself of nuclear weapons and accede to the NPT as a non-nuclear state, nuclear disarmament in Ukraine would depend on economic assistance and what it called security "guarantees:' by which it seemed to mean something that would be specifically focused on Ukraine and stronger than the NPT-related security assurances. Ukraine linked what Washington most wanted in the nuclear sphere to what it most wanted in its quest to survive and stabilize as an independent country. These two issues-Ukrainian insistence on security "guarantees" (and U.S. work on a package of security assurances that would be acceptable to Kyiv in lieu of formal guarantees), and financial compensation and assistance-determined the diplomatic rhythm of 1993 in two ways: first, the Ukrainians themselves had to define what they wanted from Russia and the United States; and, second, Ukraine, Russia, and the United States had to fash-

THE "MODEL" OF UKRAINIAN DENUCLEARIZATION

255

ion a framework for Ukrainian nuclear disarmament that incorporated within reason Ukraine's conditions. Throughout much of 1993, both processes went on simultaneously. The Internal Process of Defining Security Assurances

Ukraine first began seeking security assurances in spring 1992. A parliamentary resolution passed in April suggested that Ukraine should seek security "guarantees" as a condition of giving up nuclear weapons, and the government subsequently conveyed such a request to Washington. The U.S. response referred to the positive security assurances offered in 1968 to all non-nuclear states through UN Security Council Resolution 255, suggesting that Washington had no desire to go beyond the NPT framework to give additional country-specific assurances to Ukraine. The issue also came up in Ukraine's discussions with Russia. Moscow offered potential assurances in January and June 1993. These were made dependent on certain conditions however, including Ukraine's continued membership in the CIS. As Ukraine had begun to distance itself from the CIS out of fear of Russian domination of the organization, the conditions Russia attached to its offers of assurances rendered them unacceptable to Kyiv. During the rest of the year, Ukraine wrestled with how to define a format for assurances that would be acceptable to it and also capable of gaining U.S. and Russian assent. l l Executive-legislative relations in Ukraine were central to this process. Throughout 1993, even while it was negotiating with Russia and the United States on a basic framework for disarmament, the Ukrainian executive branch was engaged in a half-struggle, half-conspiracy with the senior leadership of the Rada to define Ukraine's bottom line on security. After the early period of antinuclear declarations, the Rada began raising doubts about the pace and even the wisdom of disarmament. The Rada essentially served in 1993 as the "bad cop;' increasing the executive branch's leverage in negotiations with Russia and the United States and helping to define the terms under which Ukraine's security concerns could be met. The Rada was neither a continuation of its rubber stamp communist predecessor nor a full-blown representative institution. The Rada reflected real party, regional, and electoral interests, but filtered through still corrupt and oligarchic politics impenetrable to outsiders. At the time of the nuclear negotiations, the Rada often appeared both an unruly opponent of the executive branch's attempts to conclude a nuclear deal and its co-conspirator. Throughout late 1992 and 1993, the Rada repeatedly delayed action on the

256

SHERMAN W. GARNETT

START I Treaty and NPT accession, formally postponing consideration of the START I Treaty in February 1993. Members of the Rada argued that the two treaties should be treated separately, despite the earlier agreement to treat the two together in the Lisbon Protocols. By mid-1993, Washington had resigned itself to the prospect that the treaties would be treated separately and hoped to use pressure to ensure that action on NPT followed START ratification as soon as possible. The Rada did not return to the START treaty until November. As delays became more frequent and nationalists within the Rada grew more vocal, Western observers often overlooked the way in which the Rada, at least at senior levels, maintained old communist traditions of control that left most of the crucial decisions to be decided by a small leadership group with close ties to the executive branch. At no time were the strongest foes of disarmament in control of the Rada. Supporters of delay or alteration of Ukraine's position on nuclear disarmament represented different parties and regions and hence saw the nuclear issue in different ways. Such a diverse group could not be united on a platform of permanent possession of nuclear weapons, but could be united on the desire to use the weapons as bargaining leverage. Throughout 1993, various members and coalitions of the Rada spoke out on the question of nuclear disarmament. For example, in April 1993, 162 deputies issued an open letter "on Ukraine's nuclear status:' The letter underscored Ukraine's status as a successor to the USSR, including "as a nuclear power:' It confirmed Ukraine's "right of ownership of the nuclear weapons on its territory:' an issue of supreme importance, not as a security issue but to strengthen Ukraine's position that it was owed compensation for the HEU in the tactical and strategic warheads stationed in Ukraine. The letter offered no specifics on how the Ukrainians defined security guarantees but stressed the importance of "state independence, national security and territorial integrity:'J2 This letter, noted more in the West for its strident assertion of Ukrainian ownership of nuclear weapons, nevertheless served to make plain to Washington that a more comprehensive framework than the Lisbon Protocol would have to be found. In August 1993, a group of parliamentarians, led by Deputy Speaker Vasily Durdinets and including key members of the Rada's foreign and defense committees' visited Washington. This group reflected the divisions within Ukraine, leaving U.S. interlocutors with conflicting observations: first, that Ukraine would not push ahead with full nuclear disarmament without a comprehensive framework that addressed its security concerns and requirements for as-

THE "MODEL" OF UKRAINIAN DENUCLEARIZATION

257

sistance; second, that there were nevertheless differences of opinion within the Rada on what would constitute acceptable security assurances, with only the nationalist-oriented members setting conditions that could not be met; and, finally, that the senior leadership of the Rada wanted u.s. involvement in a negotiated package that reflected Ukraine's conditions. In November 1993, the Rada ratified START I with a series of reservations that in effect provided the specifics of these conditions. The Rada's resolution declared that Ukraine was not bound by Article Five of the Lisbon Protocol, which had committed Ukraine to become a party to the NPT as a non-nuclear weapon state "in the shortest possible time:' It reasserted Ukraine's claim to ownership of the weapons and set forth a series of conditions that would have to be met before it would agree to exchange instruments of ratification. These conditions included an interpretation of START I that did not obligate Ukraine to dismantle all nuclear systems on its territory, but instead specified that only 42 percent of the strategic warheads had to be eliminated under the treaty. Yet the document also defined the basic requirements for technical assistance and financial compensation. It insisted that Ukraine needed "reliable security guarantees" but did not define them, leaving considerable maneuvering room to the executive branch on this issue. It also requested the president of Ukraine to negotiate with other parties on these and other issues and to confirm and control the schedule for elimination of the nuclear systems in Ukraine.13 Kravchuk and other senior officials quickly reassured the United States, Russia, and the world that the Rada's action was not the final word on the matter. Indeed, Kravchuk explained the vote as a reflection of Ukraine's need for stronger security assurances: "[We 1 fear the territorial claims that are being expressed in certain nationalist circles in Russia. [The Radal vote reflects the opinion-which is widely held here-that nuclear weapons are a means of protection against threats to the integrity of Ukrainian territory:'14 For the West, the resolution presented problems that initially obscured its potential value as a roadmap. The resolution could not be taken by the United States or Russia as a legally acceptable ratification of START 1. The dilemma in Washington was clear. Should this text be read as shutting the door or opening it? A case could be made that the Rada had finally gone too far, overturning the Lisbon Protocol and seeking the right to retain a significant portion of the nuclear warheads on its territory. However, the text also contained specific conditions that could be read as a formal negotiating proposal on security assurances, compensation, and other

258

SHERMAN W. GARNETT

key issues. In several places, it made plain that these conditions were not final or gave the president of Ukraine latitude to negotiate further. Even in point six of the resolution, which claimed Ukraine was bound by START to eliminate only a portion of the weapons on its territory, there was language stating that this formulation "did not preclude the possibility of the elimination of additional delivery vehicles and warheads according to procedures that may be determined by Ukraine:' Also, points five and eleven explicitly asked the president of Ukraine to negotiate with other parties. These points were important in Washington's assessment of the resolution. Washington could never accept the Rada's rejection of Article Five of the Lisbon Protocol as a permanent statement of Ukraine's obligations, but it could explore the basis of an interim deal that would lead to a reconsideration of this statement. In essence, Washington acknowledged that START and NPT had been temporarily delinked and that an interim agreement might be required for that linkage to be restored. It is clear from subsequent negotiations and the conclusion of the Trilateral Agreement that Washington, in the end, chose to interpret the Rada's action as something that could be overcome. The Diplomatic Negotiations 15

Throughout 1993 and until former Prime Minister Leonid Kuchma's victory in the presidential elections of July 1994, decisions within the Ukrainian leadership regarding negotiations on nuclear disarmament with Russia and the United States, as well as other matters of nuclear policy, rested with Kravchuk himself and a small group of senior leaders and advisors. In the critical stages of the negotiation of the Trilateral Framework, this group included Deputy Prime Minister Valeriy Shmarov, Deputy Foreign Minister Boris Tarasyuk, and the president's foreign policy advisor, Anton Buteyko. Though the Kravchuk government deserves much criticism for its lack of persistence on economic policy, it showed considerable energy and skill in pursuing a trilateral settlement that achieved its basic goals. To reach agreement in January 1994, it was necessary to construct a diplomatic framework for negotiations. However, to construct this framework, relations between Washington and Kyiv, which had steadily grown more tense on account of the nuclear issue, had to be improved. Improved relations were facilitated by a shift in U.S. policy in spring 1993. When the new Clinton administration took office in January, it initially continued the previous Bush administration's emphasis on applying pressure tactics to get Ukraine to denuclearize.

THE "MODEL" OF UKRAINIAN DENUCLEARIZATION

259

By April, the administration decided that this approach was not working, and it launched an interagency review of Ukraine policy. The review concluded that the United States should engage more actively with Ukraine and broaden the agenda beyond the nuclear issue, including offering stronger relationships with both the United States and NATO.16 In May 1993, U.S. Ambassador-at-Large Strobe Talbott visited Kyiv to discuss a "turning of the page" in u.S.-Ukrainian relations. U.S.-Ukrainian discussions focused not simply on outstanding nuclear matters but also on economic assistance, expanded military and defense ties, and a renewed political relationship between the United States and Ukraine. In essence, the United States sought to sketch the kind of relationship that could be established between Washington and Kyiv once the nuclear problems were resolved. This initial visit could not possibly reverse months of mutual suspicion, but it did begin a process of negotiation that brought senior levels of both governments together. Secretary of Defense Les Aspin traveled to Kyiv the following month. Secretary Aspin came directly from a meeting with Russian Minister of Defense Grachev, in which the two sides discussed U.S. proposals for early dismantling of nuclear systems in Ukraine. These proposals included arrangements for international monitoring designed to meet Kyiv's concerns about ensuring that transfer to Russia would be followed by actual dismantling of the warheads. Satisfied by these developments, Kravchuk took unilateral steps to begin deactivation of SS-19 missiles in the summer of 1993. Ukrainian Minister of Defense Morozov visited Washington in July, where the United States offered financial assistance to help cover SS-19 dismantlement costs. Regular diplomatic consultations and correspondence on economic, political, and defense matters continued. By the time of Secretary of State Warren Christopher's visit to Kyiv at the end of October 1993, the u.S.-Ukrainian dialogue had been restored and a genuine trilateral negotiating process including Russia, Ukraine, and the United States was in overdrive. By the time the Rada ratified START, with the various attached conditions discussed above, in November 1993, the basic diplomatic structure for negotiation was in place and many of the proposals that eventually found their way into the provisions of the Trilateral Agreement had been tested. In Ukraine, the growing sense of crisis, brought on by economic and political demands of striking miners in June and the Massandra Summit in September, made the need for a trilateral deal more pressing. When Washington reacted angrily to the reservations attached to the Rada's vote on START, President Kravchuk

260 SHERMAN W. GARNETT

took steps to demonstrate his goodwill. Most important, the government announced on December 20, 1993, that Ukraine would deactivate twenty SS-24s. From my interviews with participants in the mid-1990S, I believe that Kravchuk would never have risked taking the SS-24 step without some prior understanding with at least the senior leadership of the Rada. This Ukrainian executive action helped solidify the view in Washington that negotiations with the executive branch were worth continuing, and that Kravchuk and his government would find a way to deal with the Rada. Moscow was also interested in a deal that would preserve the basic framework of Massandra and begin the process of nuclear disarmament in Ukraine. As a result, both capitals committed senior officials to a concerted diplomatic effort, resulting in a breakthrough agreement. The Trilateral Agreement

The Trilateral Agreement was signed on January 14, 1994, in Moscow by presidents Clinton, Kravchuk, and Yeltsin. This agreement explicitly linked Ukraine's nuclear disarmament to its broader economic requirements and security concerns. The nuclear portions of the agreement committed Ukraine to the "elimination of all nuclear weapons, including strategic offensive arms, located on its territory ... during the seven year period as provided by the START I Treaty." Ukraine agreed in particular that "all nuclear warheads will be transferred ... to Russia" and that "all SS-24S on the territory of Ukraine will be deactivated within 10 months by having their warheads removed:' Within the same period, "at least 200 nuclear warheads from RS-18 (SS-19) and RS-22 (SS-24) missiles will be transferred from Ukraine to Russia for dismantling:'!7 In return, Ukraine was guaranteed compensation for the highly enriched uranium, beginning with 100 tons of low enriched uranium from Russia for use in nuclear power plants, underwritten by a U.S. advance payment to Russia of $60 million. In a side letter, Russia also promised to cancel some of Ukraine's debts as compensation for the HEU in the previously withdrawn tactical warheads. The United States also promised further economic and technical aid to Ukraine. The agreement reaffirmed Ukraine's commitment to complete nuclear disarmament, with Ukraine promising in a side letter to transfer all remaining nuclear warheads to Russia within three years. Kravchuk also agreed to keep the pressure on the Rada to accede to the NPT as a non-nuclear state. Of greater immediate significance, however, was that the sides agreed on concrete interim

THE "MODEL" OF UKRAINIAN DENUCLEARIZATION

261

steps including the early deactivation of all SS-24s, the beginning of the transfer of warheads to Russia, and Russian compensation for Ukraine. Unlike previous agreements, this meant that the Trilateral Agreement provided performance standards against which Ukrainian and Russian behavior could be judged. This agreement also contained security assurances that, while not breaking new ground in terms of their specific language, provided the linkage Kyiv sought between the nuclear issue and the wider range of security and state-building issues it saw as crucial. THE SECURITY ASSURANCES

In addition to Russia and the United States, Britain said that it would also be a party to the security assurances offered to Ukraine. The three states promised to seek UN Security Council assistance in the event Ukraine was a victim of nuclear threats or attack. They also pledged not to use nuclear weapons against Ukraine unless it was involved in an attack on one of them in association with a nuclear-armed state. The security assurances would only come into effect once Ukraine joined the NPT as a non-nuclear state. In negotiating the Trilateral Agreement, however, these assurances so familiar to the nonproliferation community, the so-called negative and positive security assurances, were never very controversial and were, in fact, of little interest to Ukrainian policymakers. These policymakers rightly believed that Ukrainian security would be tested well before the threat or actual use of nuclear weapons. In effect, Ukraine sought assurances that it could not obtain bilaterally that would help it establish and reinforce basic rules of the game with Russia. In this way, Kyiv hoped, for example, to prevent challenges to Ukraine's territorial integrity or Moscow's coercive use of Ukraine's energy dependence. Further, it wanted an international mechanism, preferably a binding international treaty, that would ensure that as Russian power grew, Ukraine would not be left to enforce these new rules with Russia on its own. Ukraine also sought to broaden its political and economic relationship with the West. This broadening did come about in the aftermath of the January 1994 Trilateral Agreement, with Western aid one of the key pillars supporting Kuchma's first attempt, following his election that summer, at economic reform. Ukraine understood it could not obtain everything, but seen in the light of Ukraine's immediate problems and needs, receiving such assurances represented a real success for KyiV. 18

262

SHERMAN W. GARNETT

The Texts of the Key Assurances

For the Ukrainian leadership the overriding concern was to regulate present and future relations with Russia. Therefore, beyond the agreement's reiteration of existing NPT-related positive and negative nuclear assurances, three additional assurances were critical: territorial integrity, non-use of force, and freedom from economic coercion. All three were drawn from existing commitments, such as the UN Charter and the Helsinki Final Act of the Conference on Security and Cooperation in Europe (CSCE). Territorial Integrity: The agreement reaffirmed the commitment of the three NPT depositary states "to Ukraine, in accordance with the principles of the CSCE Final Act, to respect the independence and sovereignty and the existing borders of Ukraine:' The new formula broke the link in Russia's previously offered assurances between recognition of Ukrainian territorial integrity and its participation in the Commonwealth of Independent States.i9 This first assurance was in many ways the most important for Ukraine, precisely because it won Russian acceptance, and the support of outside powers (via U.S. and UK inclusion in the assurance), for the strongest language to date on the inviolability of Ukraine's borders. Non- Use ofForce: The Tripartite Agreement also reaffirmed the obligation of Russia, the United States, and Great Britain "to refrain from the threat or use of force against the territorial integrity or political independence of Ukraine:' The likelihood of a Russian-Ukrainian war has been frequently exaggerated. Russian military weakness, overstretch, and internal problems made such a move unlikely in the 1990S and still a daunting military operation today. Ukraine's real concern lay elsewhere. With the Tripartite Agreement's assurances, Ukraine sought to make plain that Russia must relate to Ukraine as Russia does to all sovereign states. Economic Coercion: The agreement committed the three states to "refrain from economic coercion:' As Ukraine fell into hyperinflation and had to recognize the reality of its economic weakness, energy dependence on Russia, and mounting debt owed mainly to Russia, it feared that its weakness created levers for Russian coercion. In Russian-Ukrainian negotiations over the debt issue throughout the period, Russia repeatedly (and legitimately) sought some form of repayment. Moscow regularly proposed debt for asset swaps, including the notion of a controlling interest in the oil pipeline and various refineries on Ukrainian territory. The assurance in the agreement provided some interna-

THE "MODEL" OF UKRAINIAN DENUCLEARIZATION

263

tional support for limiting the exploitation of a bad economic situation. Moreover, the nuclear breakthrough brought the United States into the UkrainianRussian settlement of the debt issue as an interested observer. The United States has in fact been an active participant directly and through its allies at every important turn in the Ukrainian economy ever since. Multilateral Support for the Assurances: The final assurance in the memorandum committed all parties-Russia, the United States, Great Britain, and Ukraine-"to consult in the event a situation arises which raises a question concerning these commitments." This formal commitment to a multilateral consultation helped to shape Russian perceptions of its options in matters it once considered largely bilateral. One can see Russian reassertion under Putin and Medvedev of the earlier view of the special role for Russia in the former Soviet space-now called a "region of privileged interest"-as exactly the kind of claim that these security assurances and the subsequent bilateral and multilateral efforts by the United States, NATO, and other actors were designed to constrain. Though the assurance no longer stands alone and perhaps does not look as important as it did in 1994, it is likely that without it there would be far more uncertainty over the future of the post-Soviet space than there is at present. The Trilateral Process

The Trilateral Agreement was created by a trilateral negotiating process that remained in place to oversee the implementation of the agreement. In hindsight it is clear that such a multilateral forum would not be to Moscow's liking. On issues other than nuclear disarmament, Russia vigorously pursued oilier options that did not include the United States: the CIS; various subgroups of the former Soviet states, such as the Commonwealth Security Treaty Organization; and even groups that included China, such as the Shanghai Cooperation Organization. However, even as the trilateral negotiations have receded as an active process, it should be recalled that the forum was as important as the text of the assurances themselves. It was itself a kind of additional assurance iliat, while falling short of an absolute guarantee from the United States, Ukraine had elicited Washington's engagement (and that of the West in general) in a broad range of questions formerly seen by Moscow as bilateral. This U.S.-UkrainianRussian framework proved its worth by making possible a broad agreement that likely would have not come into being at all or been delayed for years ifleft to bilateral channels. U.S. involvement, though not radically changing the basic

264

SHERMAN W. GARNETT

Russian-Ukrainian nuclear provisions that emerged from the Massandra Summit, brought stability to both the conduct and outcome of the negotiations. In addition, the momentum of the trilateral process unlocked the door to greater bilateral and multilateral cooperation between the United States, NATO, and Ukraine. It inspired a renewed dialogue between Kyiv and Washington on political relations, security cooperation, and economic reform and assistance. In March 1994 Kravchuk visited Washington, where the Clinton administration promised to increase aid to, and seek an overall improvement of relations with, Ukraine. 20 MAKING DENUCLEARIZATION A REALITY

Following the signing of the Trilateral Agreement, Kravchuk and his government lobbied hard to obtain Rada approval for the commitments they had made. On February 3, 1994, the Rada passed a resolution that removed the reservations it had placed on its ratification of START I the previous November. Kravchuk could not obtain ratification of the NPT before he was defeated in the July election, but his successor, Leonid Kuchma, put great energy into efforts to gain the Rada's approval. Further U.S. and European promises of aid helped. So too did a commitment to put the security assurances of the Trilateral Agreement into a formal memorandum, whose text was provided to the Rada before its vote. On November 16, the Rada voted by an overwhelming margin to accede to the NPT as a non-nuclear weapon state. Ukraine deposited its instrument of ratification on December 5, 1994, at a CSCE summit meeting in Budapest. The United States, Britain, and Russia presented their signed memorandum on security assurances at the same time, to which Ukraine added its signature. France also provided security assurances in a separate document, and China issued a statement recognizing Ukraine's sovereignty and territorial integrity. At the same summit meeting, the four parties to START I also exchanged their instruments of ratification for that treaty, bringing its entry into force. All nuclear warheads were removed from Ukraine by June 1996, even earlier than the deadline it had agreed to. By the end of 2002, all the ICBMs and strategic bombers Ukraine had inherited had also been dismantled, transferred to Russia, or otherwise rendered nonusable for military purposesY President Kravchuk began the trilateral negotiations seeking "security guarantees;' not assurances. In particular, the Ukrainians urged parties to follow the model of the 1955 Austrian State Treaty and to create a legally binding, in-

THE "MODEL" OF UKRAINIAN DENUCLEARIZATION

265

ternational treaty, in which Ukraine would be a party and Ukrainian neutrality would be internationally guaranteed. This approach was consistently rebuffed, not only by Moscow but by Washington as well. The United States was unwilling to undertake new treaty obligations, and the Helsinki Final Act and subsequent CSCE documents on territorial integrity and the like, which formed the basis for the assurances in the Trilateral Agreement, were all of a "politically binding" nature. In this area, the Ukrainians clearly fell short of what they wanted. However, subsequent Ukrainian and U.S. policies have sketched out a potential place for Ukraine in NATO, though not always with support from all the NATO European members. If this ever comes to fruition and Ukraine obtains the alliance assurances it wanted, the Trilateral Agreement and Ukraine's acceptance of denuclearization will have been a critical turning point in making NATO membership a possibility. All of this makes it hard to determine whether the assurances provided to Ukraine should be considered primarily negative or positive. The texts of the key assurances were mostly phrased in negative terms-pledging signatories against the use of force or economic coercion. Yet Ukraine's leaders seem to have seen the multilateral process and commitments to future consultations, along with the offer of a range of cooperative defense and denuclearization programs, as hinting at a more positive type of assurance. Both the negative assurances in the texts of the key agreements and the positive commitments implied in the subtext were important. IMPLICATIONS FOR THEORY AND POLICYMAKERS

The Ukraine case is consistent with several hypotheses outlined by Knopf in Chapter 2. It provides qualified support for the first three hypotheses. As suggested by hypothesis 1, Ukraine had security concerns that made it interested in assurances, but these concerns might not have led to an independent program to develop nuclear weapons had Ukraine not inherited part of the Soviet arsenal. Ukraine also wanted security guarantees that were public and legally binding, as proposed in hypotheses 2 and 3, but in the end assurances that were politically rather than legally binding proved sufficient. These political assurances, however, incorporated a much larger package of political, economic, and security support and created policy momentum within the U.S. government for additional support. The case more strongly supports hypotheses 7, 8, and 13. Hypothesis 7 states that assurances may need to be tailored to the particular circumstances of the

266

SHERMAN W. GARNETT

target state. This was clearly true for Ukraine. Initial reaffirmations of general NPT-related assurances by the United States and Russia did not satisfy Ukraine. Only an agreement containing assurances tailored to address Ukraine's broader security concerns proved adequate. Moreover, these assurances led to diplomatic engagement from the United States and its allies that helped Ukraine manage its relations with Russia. This fact also provides some confirmation for hypothesis 8, about the value of embedding NPT assurances in a larger strategy of reassurance. Because Ukraine was less concerned about nuclear threats than conventional and economic threats, assurances had to be broadened from the NPT context. Finally, hypothesis 13, about the value of combining assurances with positive incentives, receives strong support in this case. Both economic aid and security assurances were likely necessary to bring about Ukraine's agreement to denuclearize, as it is hard to imagine Ukraine would have agreed if it had been offered only one but not the other.22 The findings of the case with regard to the hypotheses presented in Knopf's theory chapter do not, however, fully capture the important policy lessons of this case. Looking back, all sides including Ukraine benefited from denuclearization. The regional and global benefits of ensuring a single legitimate nuclear successor state from the breakup of the Soviet Union are obvious, but denuclearization also made Ukraine's subsequent path to a stable and sustained state much easier. Denuclearization extended Ukraine's "breathing space" and allowed it to secure its independence and make its own choices about its political, economic, and security future. The agreement on nuclear weapons gave rise in the years that followed to Russian-Ukrainian agreements on territorial recognition and the Black Sea Fleet, as well as new opportunities for Ukraine for discussions with, and potential membership in, Europe's core institutions. While relations with a much stronger and more determined Russia remain a problem, a Ukraine that was a nuclear pariah could never have attracted the outside support that it had at the time and continues to receive. The Trilateral Agreement created-during a time of uncertainty and potential chaos-a multilateral process that engaged the United States and its partners on issues of political and economic reforms, regional tensions, and bilateral disputes. It sent a strong signal of the centrality of the Ukrainian denuclearization effort and remained an effective mechanism as the United States went on to support various attempts at Ukrainian economic reform and ambitious military-to-military cooperation. It also provided a wide range of technical assistance, going beyond that embedded in the denuclearization process itself,

THE "MODEL" OF UKRAINIAN DENUCLEARIZATION

267

such as defining a Ukrainian role in the multilateral space station. The United States became a strong advocate for Ukraine within NATO, culminating in the efforts of the last Bush administration to open the door to Ukraine for eventual membership. This creation of a broad-based mechanism for dealing with the problem of proliferation in the broadest political, economic, and security context is perhaps the most important legacy of the Ukrainian case, especially for U.S. policymakers and nonproliferation advocacy groups. Even if the particular circumstances of the Ukrainian case do not appear to match those of future proliferation cases, the model of trilateral diplomacy suggests itself for both the internal organization of the U.S. response and, with modifications tailored to the individual players of a future scenario, for the international negotiations themselves. Ukraine presented a case that was unusual in its particulars, but far from unusual in terms of the uncertainties of post-Cold War diplomacy. The trilateral process required that the proliferation and arms control communities in Washington talk with one another and also with regional, economic, and military experts and policymakers. The breakup of the former USSR also presented new national formations and regional complexities that called for new perspectives and approaches. Initially, there were few seasoned Ukrainian hands around (given Ukraine's recent appearance as a state) and powerful forces urging the United States to see the issue narrowly as a proliferation case. That the U.S. policy process encouraged the emergence of country and regional expertise and included such perspectives in the mix is an important lesson. There were indeed moments when that expertise offered another perspective, one that argued for continued engagement and flexibility when other perspectives suggested a harder line. Regional expertise is not infallible and could have been wrong, but those making the hard political choices needed to hear an expanded range of voices and benefited from the input of those able to read the various sides and see potential opportunities for continued engagement. The expanded policy process and negotiating framework also fostered innovation and creativity. The United States created a comprehensive package of measures to facilitate the dismantlement of weapons, compensate Kyiv for its ownership claims to HEU, provide new sources of economic aid, and engage in a comprehensive political and security dialogue. It is easy to see the things outside the main proliferation focus as distractions or even dodges, but these "external" issues turned out to be crucial for the Ukrainian leadership and might well be crucial for a future potential proliferator. One should also not forget

268

SHERMAN W. GARNETT

that this expanded package also included an account of the consequences of moving away from denuclearization. It never wavered in seeking a denuclearized Ukraine. It contained both carrots and sticks. The traditional instruments of the nonproliferation regime are powerful tools, but nonproliferation efforts can benefit when the standard NPT toolkit is augmented by the full range of diplomatic tools brought to bear in the Ukrainian case. Finally, the unusual case may now be the norm, making the Ukraine "model" relevant to other proliferation challenges despite the case's idiosyncratic features. The fall of the Soviet Union, the powerful trends we label "globalization:' the emergence of new powers and new threats, all help ensure that future proliferation problems are likely to share the uncertainty and connections to broader social, political, and economic issues of the Ukraine case. Security assurances were vital to bringing about a nonproliferation success in this case, but only because they were construed broadly and incorporated into a broader approach. In tackling future proliferation challenges, we should take inspiration from the example of creativity and flexibility in a policy process that, while never wavering in its desired outcome, explored new perspectives and new approaches to the all too familiar and vexing problem of the spread of nuclear weapons. NOTES

The author would like to thank Paige Steen for last-minute editing and research help. 1. Joseph Cirincione, Jon B. Wolfsthal, and Mirian Rajijumar, Deadly Arsenals, 2nd ed. (Washington, DC: Carnegie Endowment for International Peace, 2005), 373; Nuclear Threat Initiative (NTI), "Ukraine Profile: Nuclear Overview;' hUp:llwww.ntLorg/e_research/profiles/Ukraine/Nuclear/index.htrnl (accessed April 25, 2010). 2. John Mearsheimer, "The Case for a Ukrainian Nuclear Deterrent;' Foreign Affairs 72, no. 3 (Summer 1993): 50-66. 3. The author was both a participant in and then an outside analyst of Ukrainian denuclearization, serving in the Office of Russia, Ukraine, and Eurasia in the Office of the Secretary of Defense until March 1994 and writing a book on Ukraine afterward (Keystone in the Arch: Ukraine and the Emerging Security Environment of Eastern and Central Europe [Washington, DC: Carnegie Endowment, 1997]). I draw heavily from the research for that book, as well as from the following articles of mine from the mid1990S for the historical sketch of the internal debate and negotiations between the key states provided in this chapter: "The Sources and Conduct of Ukrainian Nuclear Policy: November 1992 to January 1994.," in Nuclear Challenges for Russia and the New States of

THE "MODEL" OF UKRAINIAN DENUCLEARIZATION

269

Eurasia, ed. George Quester (New York: M. E. Sharpe, 1995), 125-51; and "Ukraine and the NPT;' Arms Control Today 25, no. 1 (January-February 1995): 7-12. My work at that time brought me into contact with the key officials in all three countries who worked on Ukrainian denuclearization, officials I later interviewed for my book. The insights on the attitudes of the Ukrainian political leadership in this chapter derive from interviews with senior officials in the Foreign and Defense ministries, National Security Council, presidential staff, and the Rada. This period also was rich in contemporary reporting, especially that contained in Radio Free Europe/Radio Liberty (RFE/RL) Research Reports before their demise. Key articles from this source that inform the chapter's analysis include John W. R. Lepingwell, "Ukraine, Russia and the Control of Nuclear Weapons;' vol. 2, no. 8, February 19, 1993, 4-20; "Beyond START: Ukrainian-Russian Negotiations;' vol. 2, no. 8, February 19, 1993, 46-58; "Negotiations over Nuclear Weapons: The Past as Prologue?" vol. 3, no. 4, January 28, 1994, 1-12; "The Trilateral Agreement on Nuclear Weapons;' vol. 3, no. 4, January 28, 1994, 12-20; and Bohdan Nahaylo, "The Shaping of Ukrainian Attitudes toward Nuclear Arms;' vol. 2, no. 8, February 13, 1993, 21-45. 4. According to the 1989 Soviet census, more than 11 million ethnic Russians lived in Ukraine. In the 2001 Ukrainian census, more than 8 million identified themselves as Russian. The Russian population is concentrated in the Eastern regions (Donetsk, Luhansk, Kharkov, Zaporozh'ye), Crimea, and Odessa. Only in Crimea do the Russians make up a majority, with the Russian population in the other regions of traditional Russian settlement in the east and south ranging from around 15 percent to nearly 40 percent in Luhansk. For an extended analysis of the way the ethnic and geographic diversity of Ukraine was a strength, not a weakness, see my Keystone in the Arch, 11-39. 5. See the results of two surveys reported in Ukrayina Moloda, May 21, 1993, and

Holos Ukrayiny, August 6, 1993. 6. Interviews with senior Foreign Ministry, Defense, and National Security officials, 1994-96. Similar talks with members of the Rada yielded a greater diversity of opinion, depending on the perceived threat from Russia. However, officials in the executive branch working on denuclearization and other key bilateral issues with Moscow understood that a final rejection of denuclearization in favor of a nuclear Ukraine would not only create problems with Moscow but could also have an adverse impact on the Crimea and other regions with large Russian populations. Then Defense Minster Morozov, whom I interviewed several times and met during the Trilateral process, spoke candidly of wrestling with the option of keeping nuclear weapons and wondering whether, after the return of these weapons, the diplomatic structure being built with Russia and the United States, including the security assurances, would be enough. For contemporary examples of predictions of Ukraine's imminent collapse, often seeing Ukraine's ethnic and geographical divide as the country's Achilles' heel, see Daniel Williams and R. Jeffrey Smith, "U.S. Intelligence Sees Economic Plight Leading to Breakup of Ukraine;' Wash-

ington Post, January 25, 1994; Rowland Evans and Robert Novak, "West Is Still Unsure

270

SHERMAN W. GARNETT

How to Aid Ukraine;' Chicago Sun- Times, June 9, 1994; Eugene B. Rumer, "Will Ukraine Return to Russia;' Foreign Policy 96 (Fall 1994): 129-44; F. Stephen Larabee, "Ukraine: Europe's Next Crisis?" Arms Control Today 24, no. 6 (July/August 1994): 14-19. For a response rooted in a detailed look at Ukraine's history and demography, see Roman Szporluk, Russia, Ukraine, and the Break-Up of the Soviet Union (Stanford: Hoover Institution Press, 2000). 7. During personal interviews in Kyiv and Washington in the mid-1990S, I learned that senior Ukrainian interlocutors during the period of the Trilateral talks often asserted to their U.S. counterparts that existing webs of mutual benefit among energy elites in both Russia and Ukraine-elites often responsible for diverting profits into private hands-would prevent a shut-off of gas to Ukraine. Whether such a mechanism in fact prevented Moscow's use of the energy weapon or whether Ukraine was lucky remains a question for future research. 8. Former Ukrainian Defense Minister Morozov's memoirs accuse Deputy Prime Minister Shmarov of secretly previewing the idea in Moscow and complains of Prime Minister Kuchma's support for the idea at Massandra. Morozov's memoirs were circulated as a campaign pamphlet in Kyiv in early 1994 and published in Ukrainska hazeta (1-4) in 1994. 9. For greater detail on the first phase in Ukraine's nuclear policy evolution, see Mitchell Reiss, Bridled Ambition: Why Countries Constrain Their Nuclear Capabilities (Washington, DC: Woodrow Wilson Center Press, 1995), 93-99; Lesya Gak, "Denuclearization and Ukraine: Lessons for the Future;' Nonproliferation Review 11, no. 1 (Spring 1994): 107, 116-20; and William J. Long and Suzette R. Grillot, "Ideas, Beliefs, and Nuclear Policies: The Cases of South Africa and Ukraine;' Nonproliferation Review 7, no. 1 (Spring 2000): 24-40. 10. See Garnett, Keystone in the Arch, 113-24, and "The Sources and Conduct of Ukrainian Nuclear Policy;' 125-51. 11. Gak, "Denuclearization and Ukraine;' 112-14; Reiss, Bridled Ambition, 100-102. 12. The letter reads in part: "[Elven prior to the ratification of START I, a whole complex of problems needs to be resolved. This applies in particular to the question of compensation for the nuclear materials that were taken out of the warheads of the tactical nuclear weapons that had been transferred from Ukraine to Russia is the spring of 1992, to the guarantees of destroying these weapons by Russia, and to the enormous financial expenditure on the reduction of the nuclear potential. ... At the same time it would be a mistake to agree to promises of insignificant monetary compensation in exchange for Ukraine's immediate nuclear disarmament. The question of nuclear disarmament, state independence, national security, and territorial integrity cannot become an object for bargaining or 'monetary compensations:" The full text may be found in FBIS Daily Report: Central Eurasia, April 30, 1993, 51. 13. The text of the resolution may be found in Holos Ukrayiny, November 20, 1993.

THE "MODEL" OF UKRAINIAN DENUCLEARIZATION

271

14. Quoted in Reiss, Bridled Ambition, 113. 15. A key u.s. actor on Strobe Talbott's staff and later ambassador to Ukraine, Steve Pifer, has recently provided his own detailed and insightful overview of the Trilateral negotiations in The Trilateral Process: The United States, Ukraine, Russia and Nuclear Weapons, Arms Control Series Paper 6 (Washington, DC: Brookings Institution, May 2011). 16. Reiss, Bridled Ambition, 106. 17. The text of the Trilateral Agreement may be found in RFEIRL Research Report, vol. 3, no. 4, January 28,1994,14-15. 18. For an eloquently reasoned statement of the opposite point of view that emerged in the contemporary analysis of tiIe Trilateral Agreement, see Steve E. Miller, "Ukraine's Flawed Nuclear Diplomacy;' Nonproliferation Review 1, no. 3 (Spring-Summer 1994): 47-53. Miller argues that tiIe texts of the assurances are really just reaffirmations of undertakings Ukraine already enjoyed as a CSCE member state. 19. Before the Trilateral Agreement, tiIe key documents with regard to Russian recognition of the territorial integrity of Ukraine made recognition conditional. For example, tiIe December 1991 agreement establishing tiIe Commonwealth of Independent States (CIS) linked territorial recognition to states remaining within the CIS (TASS, December 9, 1991). During the period of Ukraine's initial questioning of its earlier statements on disarmament, tiIe Russian parliament voted twice (January and May 1992) to reconsider the constitutionality of tiIe transfer of Crimea to Ukraine in 1954 and in November 1992. On the details of Russian-Ukrainian borders and the complicating factor of Romania, see Keystone in the Arch, 57-61, 91-94. 20. William J. Clinton, the President's News Conference witiI President Leonid Kravchuk of Ukraine, March 4, 1994, http://www.presidency.ucsb.edu/ws/index.php?p id=49755#axzZ1YB7POAkz. 21. Reiss, Bridled Ambition, 118-21 and 173 n. 121; Cirincione et aI., Deadly Arsenals, 377-79; Nuclear Threat Initiative, "Ukraine Profile: Nuclear Overview:' 22. For an assessment of tiIe role of positive incentives in the Ukraine case, see Thomas Bernauer, Stefan Brem, and Roy Suter, "The Denuclearization of Ukraine;' in The Politics of Positive Incentives in Arms Control, ed. Thomas Bernauer and Dieter Ruloff (Columbia: University of South Carolina Press, 1999).

Conclusions James J. Wirtz

Strategies of assurance seek to influence another state by addressing potential sources of insecurity that might be motivating its behavior. As Jeff Knopf observes in Chapter 2, "assurance" and related terms have been applied to several strategic contexts in international relations. The term can refer to a subcomponent of deterrence, to an effort to demonstrate one's reliability to allies, or to a broad effort to reassure a potential adversary that one harbors no hostile intentions toward it. In practice, the most deliberate use of assurances has been in relation to nuclear nonproliferation. By examining the use of assurances in combating proliferation, the chapters in this volume have sought to advance our understanding of the effectiveness of security assurances both in general and specifically as a nonproliferation tool. Understanding the relationship between security assurances and efforts to stop the proliferation of nuclear weapons is important to scholars and policymakers alike. For scholars, security assurances highlight the way domestic and international incentives and disincentives interact to cause governments to embrace or discard nuclear weapons. An understanding of security assurances would thus shed light on the nature of global nuclear politics. For policymakers, understanding how and when security assurances are likely to slow a nuclear weapons development program or stop a decision to acquire a nuclear arsenal would create opportunities to halt proliferation. Without insight into the way several target audiences might perceive security assurances, policymakers run the risk of applying the "right" policy in the "wrong" situation, thereby exacerbating a latent arms race. Security assurances are commonplace in in275

276

JAMES J. WIRTZ

ternational affairs; yet we lack a systematic appreciation of the way they affect incentives to obtain a nuclear arsenal. Scholars still debate if security assurances have a practical impact on national decisions to acquire nuclear weapons. Nonproliferation efforts involve both positive and negative security assurances. Positive assurances constitute a promise to take action on the part of another actor in the event of some mutually recognized untoward action. Although there are modest positive assurances associated with the global nonproliferation regime, most attention has been devoted to positive assurances in the form of security guarantees offered through an alliance relationship. These generally are associated with extended nuclear deterrence and imply a promise to retaliate with nuclear weapons if the protected party suffers a nuclear attack. Positive assurances are said to bolster nuclear nonproliferation because they provide an incentive to a government to forgo developing a nuclear capability. According to this logic, under the protection of another state's "nuclear umbrella:' a state could benefit from a nuclear deterrent without actually possessing nuclear weapons. The international community also benefits because a specific state's security objectives are achieved without fundamentally undermining nonproliferation objectives. Although they are sometimes contingent, negative assurances reflect a promise on the part of a nuclear-armed state not to use its nuclear arsenal against nations that do not possess nuclear weapons. Negative assurances actually constitute the real "grand bargain" embodied in the Non-Proliferation Treaty (NPT): by forgoing nuclear weapons, states can "opt out" of the nuclear arms race. Non-nuclear weapon states that abide by nonproliferation norms are no longer faced with the direct threat of nuclear war. From a theoretical, normative, and practical perspective, however, the two types of security assurances are not entirely fungible. Advocates of nuclear disarmament, for example, highlight the importance of negative assurances, because these do not require nuclear threats to be successful. The nonproliferation community generally looks askance at positive security guarantees because they seem to undermine nonproliferation norms even though their primary objective is to halt the spread of nuclear weapons. As a result, negative security assurances are generally associated with the nonproliferation regime, especially multilateral agreements such as the Nuclear Non-Proliferation Treaty and regional nuclear-weapon-free zones. By contrast, positive assurances are generally related to promises of extended nuclear deterrence made by national governments to other states or alliances. Positive assurances ultimately rest on the threat of using force, if not nuclear weapons. This bifurcation can take on an ideological tinge.

CONCLUSIONS

277

Those who see value in nuclear deterrence tend to see value in positive assurances, while those who champion disarmament tend to see value in negative assurances. As a result, questions remain unanswered about how to integrate both positive and negative assurances in the quest to devise a theoretically sophisticated and effective international effort to stop or reverse nuclear proliferation. This is not a hypothetical issue. As Scott Snyder and Joyce Lee noted in their discussion of U.S. positive security assurances to South Korea, negative security assurances conveyed to North Korea create the need for a "parallel and holistic approach" when it comes to fostering nonproliferation on the Korean Peninsula. 1 In other words, policy has unfolded without adequate knowledge about how or even if positive and negative security assurances can form an effective way of rolling back the threat of nuclear war in Northeast Asia. The contributors to Security Assurances and Nuclear Nonproliferation have taken an important step in advancing our understanding of assurance strategies by presenting insights into how positive and negative security assurances have affected the calculus of different governments when it comes to decisions about pursing a nuclear program. Collectively, their chapters create a basis for that "parallel and holistic approach" that policymakers will need to maximize the impact of positive and negative security assurances-policies that are theoretically distinct but nevertheless unfold simultaneously in international relations. Their efforts make it possible to offer some preliminary judgments about the effectiveness of security assurances, the usefulness of the hypotheses on assurances offered at the outset of the volume, and the relevance of positive and negative security assurances to address today's proliferation challenges. The clearest theme that emerges across the chapters is the contextual nature of assurances. Assurance strategies have been most effective when they are part of a larger package involving positive incentives or broader efforts at reassurance, or when they supplement other developments such as good political relations among allies or a country's embrace of norms against nuclear weapons. ARE SECURITY ASSURANCES EFFECTIVE?

The contributors to the volume judged that security assurances were effective in contributing to nonproliferation outcomes in the cases of Libya, Japan, South Korea, Sweden, and Ukraine, and to some extent in North Korea during the Cold War. Assurances were not successful in curtailing North Korea's nuclear program in recent decades or Iran's nuclear ambitions under the Shah. Assurances have not been given a strong test in the case of Iran since the 1979

278

JAMES J. WIRTZ

NONPROLIFERATION OUTCOME

Success

'z"

u

Positive

Japan South Korea

Negative

Libya

Failure

-