SOCIAL ACTION, ROGUE REACTION: US POST-COLD WAR NUCLEAR COUNTERPROLIFERATION STRATEGIES A DISSERTATION SUBMITTED TO THE DEPARTMENT OF POLITICAL SCIENCE AND THE COMMITTEE ON GRADUATE STUDIES OF STANFORD UNIVERSITY IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY AlexanderH.Montgomery September2005 (cid:13)c Copyright by Alexander H. Montgomery 2006 All Rights Reserved ii IcertifythatIhavereadthisdissertationandthat,inmyopin- ion,itisfullyadequateinscopeandqualityasadissertation forthedegreeofDoctorofPhilosophy. ScottD.Sagan (PrincipalAdviser) IcertifythatIhavereadthisdissertationandthat,inmyopin- ion,itisfullyadequateinscopeandqualityasadissertation forthedegreeofDoctorofPhilosophy. JohnMeyer IcertifythatIhavereadthisdissertationandthat,inmyopin- ion,itisfullyadequateinscopeandqualityasadissertation forthedegreeofDoctorofPhilosophy. LynnEden IcertifythatIhavereadthisdissertationandthat,inmyopin- ion,itisfullyadequateinscopeandqualityasadissertation forthedegreeofDoctorofPhilosophy. DavidJ.Holloway Approved for the University Committee on Graduate Stud- ies. iii iv Abstract Which US post-Cold War counterproliferation strategies towards nuclear aspirants have been most successful and why? Military capabilities, economic needs, and social desires provide both motivations for nuclear proliferation and opportunities for states to counter these motivations. I argue that strategies involving the use of military or economic tools alonearenotonlyunsuccessfulbutareactuallycounterproductive. Idemonstrateusingthe cases of North Korea and Iran that including social benefits in the form of symbolic and diplomaticgesturesiscrucialtothesuccessofthesestrategies. To generate hypotheses regarding the effects of different types of strategies on nuclear outcomes, I combine realist, liberal, and constructivist theories with spiral and deterrent models. I include the effects of three intervening mechanisms: domestic political struc- tures, positive feedback loops, and the effects of multiple-country interactions. To test these hypotheses, I break down strategies into individual actions directed at either sup- pressing the demand for or halting the supply of nuclear technologies to individual prolif- erants. I use vector autoregression to uncover action-reaction cycles and establish general interaction patterns between the United States and North Korea. I then use within-case congruence procedures to compare theoretical predictions of the effects of different types of strategies with nuclear outcomes in both North Korea and Iran. I find in both cases that feedback played an important role in amplifying US strategies and that social incentives were consistently related to success. Finally, I perform a cross-case comparison of how the structure of proliferation networks among second-tier nuclear aspirants has affected weapons development times; I find that tacit knowledge constrained the A.Q. Khan nu- clear network to a hub-and-spoke structure, minimizing its effectiveness. Together, these findings imply that to stop proliferation, social benefits should be used to target existing and potential hubs of proliferation networks; future research should find that other prolif- eration networks are similarly constrained and that social motivations play a large role in proliferationdecisions. v Acknowledgements Thisdissertationistheresultoffiveyearsoffollowinganoccasionallycircuitouspathfrom an initial inquiry into the effects of social sanctions on nuclear aspirants to the (finally) finished product. I could not have completed it without the assistance of many individuals alongtheway. Anyshortcomingsofthiswork (includingforgettingtothankthem)aremy own;anyvirtuesareinlargepartduetotheirassistance. I would first like to thank the members of my committee. Scott Sagan provided in- valuable guidance throughout the project, from compiling an initial reading list to reading the final draft on a very tight schedule; his suggestions on all parts of the dissertation have been invaluable. Lynn Eden gave copious advice on numerous occasions, spending a great deal of time on everything from brainstorming to polishing; her door was always open. John Meyer brought a very different perspective to the table; his ability to offer concise and pertinent advice within a day of every draft is unparalleled. I would like to thank Steve Krasner for comments on early drafts, and David Holloway for readily stepping in as a committee member late in the process. I would also like to thank Woody Powell for chairingmydefense. I received invaluable help from many seminar participants. Ted Hopf and Steve Biddle offeredusefultipsonanearlyversionofChapters2-4givenasapaperatthe2004meeting oftheAmericanPoliticalScienceAssociation. ChaimBraun,ChristopherChyba,andDean Wilkening offered helpful feedback on an early draft of Chapter 6 given as a paper at the Stanford Center for International Cooperation and Security (CISAC). Gili Drori provided extensive comments on a version of my prospectus presented at CISAC. I would like to thankthemaswellastherestoftheparticipantsoftheResearchSeminaratCISAC. MostofmydissertationwritingoccurredattheKennedySchoolatHarvardUniversity, where I was a pre-doctoral fellow for two years. I would particularly like to thank Jim Walsh, Matt Bunn, John Park, Anthony Wier, and Chen Zak at Managing the Atom and SteveMillerandSeanLynn-JonesattheInternationalSecurityProgramfordiscussionson vi NorthKoreaandIranduringmytimeatBelfer. IwouldalsoliketothankDianeMcCreefor extensiveeditingonChapter6,whichisforthcomingasanarticleinInternationalSecurity. IalsoreceivedmanycommentsfromtherestoftheparticipantsoftheManagingtheAtom Project and the International Security Program seminars at the Belfer Center for Science andInternationalAffairs. Financial assistance was provided by a Graduate Research Fellowship from the Na- tional Science Foundation, the Political Science department at Stanford University, the StanfordCenterforInternationalSecurityandCooperationattheFreemanSpogliInstitute for International Studies, and the Belfer Center for Science and International Affairs, John F.KennedySchoolofGovernment,HarvardUniversity. IwouldliketothankJeanetteLee- Oderman, graduate student administrator in Political Science at Stanford, for sorting out innumerablebureaucraticcomplicationsaswellasgeneralsupport. I am fortunate enough to have many colleagues who are also good friends; they pro- vided support during the long process of researching and writing. My friends Paul Mac- DonaldandStacieGoddardgavemeearlycommentsonimportantpartsofthedissertation and helped to keep me sane while I was in Cambridge. Ron Hassner, Taylor Fravel, and Todd Sechser, my immediate predecessors in security studies at Stanford and fellow fel- lows in Cambridge, were equally good with their advice and company. Karthika Sasiku- mar, fellow student of nonproliferation, gave me suggestions on multiple subjects. Emilie Hafner-Burton,DanNexon,andJaniceBiallyMatternprovidedencouragementduringthe researching and writing process. Adrienne LeBas was a wonderful commiserator during thefinalfewmonthsofwriting. Finally,Iwouldliketothankmyfriendsandrelativesfortheirsupportandpatienceover the years, both preceding and during the overlong dissertation process. My parents, Judy andDoug,havebeenverysupportiveduringtheentireprocess,despiteanydoubtstheymay have had about how long it was going to take to finish the final product. My grandparents Bill, Jeanne, Edison, and Ruth have been very encouraging of my work whether it was readable or not. Without their support as well as the support of my entire family and friends(especiallyBrianandOrion),thisdissertationwouldnothavebeenpossible. vii Contents Abstract v Acknowledgements vi ListofTables xi ListofFigures xii 1 Introduction 1 1.1 DefiningtheQuestion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 1.2 CaseSelection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 1.3 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 2 TheoryandMethod 17 2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 2.2 Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 2.3 Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 2.4 Evidence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 2.5 TheTechnologyofProliferation . . . . . . . . . . . . . . . . . . . . . . . 43 2.5.1 ThePlutoniumRoute . . . . . . . . . . . . . . . . . . . . . . . . . 46 2.5.2 TheUraniumRoute . . . . . . . . . . . . . . . . . . . . . . . . . . 50 2.6 Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 3 SocialAction,WhatReaction? 57 3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 3.2 QuantitativeMethodology . . . . . . . . . . . . . . . . . . . . . . . . . . 60 3.2.1 DataCoding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 3.2.2 EventScaling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 viii 3.2.3 VectorAutoregression(VAR) . . . . . . . . . . . . . . . . . . . . 64 3.2.4 CrisisPeriodsandTime . . . . . . . . . . . . . . . . . . . . . . . 68 3.3 US–DPRKInteractions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 3.3.1 TheFirstNorthKoreanNuclearCrisis . . . . . . . . . . . . . . . . 74 3.3.2 TheSecondNorthKoreanNuclearCrisis . . . . . . . . . . . . . . 79 3.4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 3.A LagLengthsandCoefficients . . . . . . . . . . . . . . . . . . . . . . . . . 83 3.B EventCodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86 4 NorthKorea 91 4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 4.2 TheNorthKoreanNuclearProgram,1950–2003 . . . . . . . . . . . . . . . 93 4.2.1 NorthKoreanMotivations . . . . . . . . . . . . . . . . . . . . . . 93 4.2.2 TheOriginsoftheProgram . . . . . . . . . . . . . . . . . . . . . 94 4.2.3 ThePost-ColdWarProgram . . . . . . . . . . . . . . . . . . . . . 97 4.3 NorthKoreanActions,USReactions,1990–2003 . . . . . . . . . . . . . . 105 4.3.1 TheFirstNorthKoreanNuclearCrisis . . . . . . . . . . . . . . . . 105 4.3.2 TheSecondNorthKoreanNuclearCrisis . . . . . . . . . . . . . . 126 4.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148 5 Iran 153 5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153 5.2 TheRootsoftheIranianNuclearProgram,1957–2003 . . . . . . . . . . . 155 5.2.1 IranianMotivations . . . . . . . . . . . . . . . . . . . . . . . . . . 155 5.2.2 TheOriginsoftheProgram . . . . . . . . . . . . . . . . . . . . . 162 5.2.3 ThePost-RevolutionProgram . . . . . . . . . . . . . . . . . . . . 164 5.3 IranianActions,USReactions,1979–2002 . . . . . . . . . . . . . . . . . . 169 5.3.1 RussiaandBushehr . . . . . . . . . . . . . . . . . . . . . . . . . . 171 5.3.2 China . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178 5.3.3 OtherStates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181 5.3.4 BilateralActions . . . . . . . . . . . . . . . . . . . . . . . . . . . 183 ix 5.4 IranianClandestineAction,1979-2002 . . . . . . . . . . . . . . . . . . . . 185 5.4.1 IntelligenceEstimates . . . . . . . . . . . . . . . . . . . . . . . . 185 5.4.2 DomesticandClandestineUraniumExperiments . . . . . . . . . . 187 5.4.3 IAEAinspections . . . . . . . . . . . . . . . . . . . . . . . . . . . 190 5.5 TheIranianNuclearCrisis,2002–2003 . . . . . . . . . . . . . . . . . . . . 191 5.6 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 198 6 RinginginProliferation 201 6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201 6.2 ProliferationDeterminism . . . . . . . . . . . . . . . . . . . . . . . . . . 202 6.3 NewProliferators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205 6.3.1 NuclearNetworks: LeapfroggingorFallingDown? . . . . . . . . . 206 6.3.2 TheIrrelevanceofRegimeType . . . . . . . . . . . . . . . . . . . 211 6.4 ProliferationNetworks . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216 6.4.1 TheStructureofProliferationNetworks . . . . . . . . . . . . . . . 217 6.4.2 TacitKnowledgeandtheSpreadofNuclearWeapons . . . . . . . . 222 6.5 PastandFutureCounterproliferationEfforts . . . . . . . . . . . . . . . . . 226 6.6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229 7 Conclusion 231 7.1 PolicyImplications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231 7.1.1 NorthKorea . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232 7.1.2 Pakistan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 234 7.1.3 Iran . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235 7.2 ResearchImplications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237 Bibliography 241 Periodicals 269 x