The EAI Fellows Program on Peace, Governance, and Development in East Asia, supported by the Henry Luce Foundation, and Smith College helped underwrite the research for this article. Thanks to Kathryn Weathersby at the Woodrow Wilson Center and Bernd Schaefer at the German Historical Institute in Washington, DC for making historical documentation available to me. The article has benefited from many useful comments made by students and scholars at a dozen East Asian institutions of higher learning that I visited during my EAI Fellowship tour in May 2007. I subsequently presented this work at several American universities and at the annual meeting of the American Political Science Association in 2007. Thanks especially to Stephan Haggard, Rieko Kage, Byung-Kook Kim, Dinshaw Mistry, Steve Noerper, William Potter, and two anonymous reviewers for their careful readings of earlier drafts of this article.Google Scholar

1. Herrmann, Richard K. and Choi, Jong Kun, “From Prediction to Learning: Opening Experts' Minds to Unfolding History,” International Security 31, no. 4 (Spring 2007): 132–161.CrossRefGoogle Scholar

2. Richelson, Jeffrey, Spying on the Bomb: American Nuclear Intelligence from Nazi Germany to Iran and North Korea (New York: W. W. Norton, 2006). For a systematic analysis of Richelson's evidence, see Montgomery, Alexander and Mount, Adam J., “Misunderestimation: Explaining US Failures to Predict Nuclear Weapons Programs,” paper presented at the 2006 American Political Science Association meeting. See also Froscher, Torrey C., “Anticipating Nuclear Proliferation: Insights from the Past,” Nonproliferation Review 13, no. 3 (November 2006): 467–477.Google Scholar

3. For an overview of the literature, see Scobell, Andrew, “North Korea's Nuclear Intentions,” in Lister, James M., ed., Challenges Posed by the DPRK for the Alliance and the Region (Washington, DC: Korea Economic Institute, 2005), pp. 78–95.Google Scholar

4. Cha, Victor D. and Kang, David C., Nuclear North Korea: A Debate on Engagement Strategies (New York: Columbia University Press, 2003).Google Scholar

5. This is part of a more general trend in East Asian studies. See Haggard, Stephan, “The Balance of Power, Globalization, and Democracy: International Relations Theory in Northeast Asia,” Journal of East Asian Studies 4, no. 1 (January–April 2004): 1–38. A powerful application of general social science theory to the case of North Korea is Stephan Haggard and Marcus Noland, Famine in North Korea: Markets, Aid, and Reform (New York: Columbia University Press, 2007).Google Scholar

6. Cha, and Kang, , Nuclear North Korea, p. 3.Google Scholar

7. The seminal text, recently republished with commentary from contemporary comparative foreign policy scholars, is Snyder, Richard, Bruck, H. W., and Sapin, Burton, Foreign Policy Decision-Making (Revisited) (New York: Palgrave Macmillan, 2002).Google Scholar

8. Recently, Etel Solingen has also applied a foreign policy analysis framework to the North Korean nuclear issue (Solingen, Etel, Nuclear Logics: Contrasting Paths in East Asia and the Middle East [Princeton: Princeton University Press, 2007]). Solingen's basic argument is that leaderships that are “inward-oriented” (particularly with respect to their economic development strategies) have a tendency to promote “nuclearization,” whereas leaderships that are “internationalizing” have a tendency to promote “denuclearization.” Solingen therefore naturally concludes that the DPRK, with its thoroughly inward-oriented leadership, would likely pursue nuclearization. Solingen's model is largely complementary with the model that I develop below. The major difference between them is in their definition of the dependent variable. Solingen is interested in nuclearization, which she defines as “movement toward nuclear weapons acquisition, even if it does not result in actual acquisition” (Solingen, Nuclear Logics, p. 301, note 1). By contrast, the key puzzle for this article—and indeed the key puzzle for all analysts of the DPRK nuclear issue—is whether, when, why, and how Pyongyang would make and then implement a decision to acquire the bomb. Solingen's model is silent on this puzzle.Google Scholar

9. For a brief review, see Hymans, Jacques E. C., “Theories of Nuclear Proliferation: The State of the Field,” The Nonproliferation Review 13, no. 3 (November 2006): 455–465.CrossRefGoogle Scholar

10. For further elaboration, see Hymans, Jacques E. C., The Psychology of Nuclear Proliferation: Identity, Emotions, and Foreign Policy (Cambridge: Cambridge University Press, 2006), especially chs. 1–2.CrossRefGoogle Scholar

11. One vote for the relevance of classical deterrence theory is cast by President Musharraf of Pakistan, who despite sitting on a sizable nuclear arsenal found US threats of bombing his country “back to the Stone Age” in the aftermath of the September 11 attacks to be highly credible. See Musharraf, Pervez, In the Line of Fire: A Memoir (New York: Free Press, 2006).Google Scholar

12. The incredible complexity of the nuclear choice, and therefore the futility of trying to squeeze it into a standard cost-benefit framework, is powerfully conveyed in Sen, Amartya, “India and the Bomb,” Journal of Peace Economics, Peace Science and Public Policy 6, no. 4 (Fall 2000): 16–34.Google Scholar

13. See Holsti, Ole, “Foreign Policy Formation Viewed Cognitively,” in Axelrod, Robert, ed., Structure of Decision (Princeton: Princeton University Press, 1976). For a clarion call for recognizing that the constraints on pure rationality are not merely informational but also cognitive in nature, see Odell, John S., “Bounded Rationality in the World Political Economy: The Nature of Decision Making,” in Andrews, David M., Henning, C. Randall, and Pauly, Louis W., eds., Governing the World's Money (Ithaca: Cornell University Press, 2002). A review of recent research on “bounded rationality,” which especially emphasizes the importance of affect and the emotions, is Daniel Kahneman, “Maps of Bounded Rationality: A Perspective on Intuitive Judgment and Choice,” Nobel Prize Lecture, December 8, 2002, available at http://nobelprize.org/nobel_prizes/economics/laureates/2002/kahnemann-lecture.pdf. Parenthetically, Kahneman's lecture shows how far his work has evolved beyond prospect theory's originally quite modest suggestions of deviations from pure rationality.Google Scholar

14. Of course, external stimuli are necessary to put the question of going nuclear on the political agenda. But given the fear emotion's tendency to magnify external threats, those stimuli need not be very large. For the full elaboration of this model, see Hymans, , The Psychology of Nuclear Proliferation, ch. 2.Google Scholar

15. Delumeau, Jean, Rassurer et protéger. Le sentiment de sécurité dans l'Occident d'autrefois (Paris: Fayard, 1989).Google Scholar

16. The specific symbolism of nuclear weapons is explored in Jervis, Robert, The Meaning of the Nuclear Revolution (Ithaca: Cornell University Press, 1989).Google Scholar

17. For more on the notion of decisions without calculations, see Rosen, Stephen P., War and Human Nature (Princeton: Princeton University Press, 2005). Also of interest on this point is Crawford, Neta C., “The Passion of World Politics: Propositions on Emotion and Emotional Relationships,” International Security 24, no. 4 (Spring 2000): 116–156.Google Scholar

18. See Hymans, , The Psychology of Nuclear Proliferation, especially pp. 195–203.Google Scholar

19. In truth, the eras of the father and the son are not so clearly distinguishable. Kim Jong Il began taking over much of the day-to-day business of the state as early as 1980; by the early-1990s nuclear crisis, it was clear that he, not his father, was calling most of the shots. Indeed, when Kim Il Sung met with Jimmy Carter during the nuclear crisis of 1994, he appeared almost as unfamiliar with his country's negotiating stance as Carter was with that of the Clinton administration. See Oberdorfer, Don, The Two Koreas: A Contemporary History (Reading, MA: Addison-Wesley, 1997), p. 328.Google Scholar

20. Note that even in regimes where the general dominance of the top leadership is not so extreme, nuclear decisionmaking tends to be highly concentrated in its hands. See Poneman, Daniel, Nuclear Power in the Developing World (Boston: Allen & Unwin, 1982), ch. 9.Google Scholar

21. Weathersby, Kathryn, “The Enigma of the North Korean Regime: Back to the Future?” in Lister, James M., ed., Challenges Posed by the DPRK for the Alliance and the Region (Washington, DC: Korea Economic Institute, 2005), p. 46. Note that Weathersby argues that Kim Il Sung's confidence began to break down in the 1980s.Google Scholar

22. Cumings, Bruce, Korea's Place in the Sun: A Modern History (New York: W. W. Norton, 1997); Hassig, Ralph and Oh, Kongdan, North Korea Through the Looking Glass (Washington, DC: Brookings Institution Press, 2000); Szalontai, Balazs, Kim Il Sung in the Khrushchev Era: Soviet-DPRK Relations and the Role of North Korean Despotism, 1953–1964 (Washington, DC: Woodrow Wilson Center Press, 2005).Google Scholar

23. Kelman, Herbert, “Social-Psychological Dimensions of International Conflict,” in Zartman, I. William and Rasmussen, J. Lewis, eds., Peacemaking in International Conflict: Methods and Techniques (Washington, DC: United States Institute of Peace Press, 1997).Google Scholar

24. Cumings, , Korea's Place in the Sun, p. 403; Scobell, , “North Korea's Strategic Intentions,” p. 14.Google Scholar

25. Cumings, , Korea's Place in the Sun, p. 403; for more elaboration on the Japanese connection, see Jager, Sheila Miyoshi, “Women, Resistance, and the Divided Nation: Women and the Romantic Rhetoric of Korean Reunification,” Journal of Asian Studies 55, no. 1 (February 1996), especially p. 45.Google Scholar

26. Vorontsov, Alexander V., “North Korea's Military-First Policy: A Curse or a Blessing,” Nautilus Institute Policy Forum Online 06-45A (June 8, 2006), available at www.nautilus.org/fora/security/0645Vorontsov.html. See also Patrick McEachern's article in this issue.Google Scholar

27. Kim, Samuel S., “Research on Korean Communism: Promise Versus Performance,” World Politics 32, no. 2 (January 1980), especially pp. 303–304.CrossRefGoogle Scholar

28. This is an inheritance from earlier Korean racial nationalist historiography, which also has influenced contemporary South Korean views. See Schmid, Andre, “Rediscovering Manchuria: Sin Ch'aeho and the Politics of Territorial History in Korea,” Journal of Asian Studies 56, no. 1 (February 1997): 26–46.Google Scholar

29. For Kim Il Sung's addresses, I relied on English translations that were produced after 1975 by the Foreign Broadcast Information Service (FBIS) Asia and Pacific Daily Report; for the joint editorials, I relied on English translations produced contemporaneously by the DPRK itself. While speeches prior to 1975 are also available in collections of Kim Il Sung's works published by the DPRK, I limit the analysis to the post-1975 period for two reasons: first, because before the mid-1970s, the notion of an indigenous DPRK nuclear bomb was quite implausible except perhaps to the truest of true believers; and, second, because for earlier years, no FBIS translation is available, and the regime is known to have a practice of rewriting old texts to bring them into conformity with contemporary ideological positions.Google Scholar

30. See Hymans, , The Psychology of Nuclear Proliferation, ch. 3.Google Scholar

31. Ibid. Google Scholar

32. George, Alexander L., Propaganda Analysis: A Study of Inferences Made from Nazi Propaganda in World War II (Evanston, IL: Row, Peterson, 1959).Google Scholar

33. Cha, and Kang, , Nuclear North Korea, p. 45. Though he is less explicit about dates than Kang, Cha also argues that the period 1989–1994 was a key turning point in causing the DPRK regime to go “double or nothing” (p. 30).Google Scholar

34. Szalontai, Balazs and Radchenko, Sergey, “North Korea's Efforts to Acquire Nuclear Technology and Nuclear Weapons: Evidence from Russian and Hungarian Archives,” CWIHP Working Paper No. 53, Cold War International History Project, August 2006.Google Scholar

35. Cumings, , Korea's Place in the Sun, p. 467. Other clear statements of DPRK nuclear weapons intent are reported in Oberdorfer, , The Two Koreas, p. 253.Google Scholar

36. Soviet Foreign Ministry memorandum, August 24, 1962, translated and reprinted in Szalontai and Radchenko, “North Korea's Efforts,” p. 33. Chol goes on to mention China's nuclear program, but clearly he is not talking only about China.Google Scholar

37. The report pointedly remarks that the North Korean translator chose not to render the latter sentence into German. Botschaft der DDR in der KVDR, “Aktenvermerk über ein Abendessen in der Residenz, das vom Gen. Botschafter Henke und seiner Gattin am 6. 12. 1969 in der Zeit von 19.30 Uhr bis 22.30 Uhr gegeben wurde,” Pyongyang, December 13, 1969, document viewed at the German Historical Institute in Washington, DC. My translation.Google Scholar

38. Hungarian Foreign Ministry memorandum, February 16, 1976, translated and reprinted in Szalontai, and Radchenko, , “North Korea's Efforts,” p. 55.Google Scholar

39. Szalontai, and Radchenko, , “North Korea's Efforts,” p. 10.Google Scholar

40. Ibid., pp. 2, 25.Google Scholar

41. The DPRK's report to the IAEA had not completely denied that these efforts had taken place, but it had been untruthful about their extent.Google Scholar

42. See Wit, Joel S., Poneman, Daniel B., and Gallucci, Robert L., Going Critical: The First North Korean Nuclear Crisis (Washington, DC: Brookings Institution Press, 2004).Google Scholar

43. Richelson, , Spying on the Bomb, pp. 522–524. The State Department was somewhat more circumspect about this judgment.Google Scholar

44. Cha, and Kang, , Nuclear North Korea, p. 145.Google Scholar

45. For estimates of the DPRK plutonium stockpile, see the reports by David Albright and the Institute for Science and International Security, available at www.isis-online.org/publications/dprk/index.html.Google Scholar

46. The inadequate attention paid to difficult measurement tasks in the nuclear issue area is explored in Eden, Lynn, Whole World on Fire: Organizations, Knowledge, and Nuclear Weapons Devastation (Ithaca: Cornell University Press, 2004).Google Scholar

47. MacKenzie, Donald and Spinardi, Graham, “Tacit Knowledge, Weapons Design, and the Uninvention of Nuclear Weapons,” American Journal of Sociology 101, no. 1 (July 1995): 44–99.Google Scholar

48. See Commission on the Intelligence Communities of the United States Regarding Weapons of Mass Destruction, Report to the President, March 31, 2005, available at www.wmd.gov/report, ch. 2.Google Scholar

49. For Sagan's point of view, see Sagan, Scott D. and Waltz, Kenneth N., The Spread of Nuclear Weapons: A Debate Renewed (New York: W. W. Norton, 2002).Google Scholar

50. See Clapham, Christopher S., Third World Politics: An Introduction (Madison: University of Wisconsin Press, 1985). In this issue, Walter C. Clemens Jr. offers a broadly similar Weberian take on the DPRK case.Google Scholar

51. Chehabi, Houchang E. and Linz, Juan J., eds., Sultanistic Regimes (Baltimore: Johns Hopkins University Press, 1998), p. 9.Google Scholar

52. Stephan Haggard and Marcus Noland second the notion of the regime's extreme personalism in terms of economic policy. Haggard, and Noland, , Famine in North Korea, p. 227. For a great deal of instructive detail, see Martin, Bradley K., Under the Loving Care of the Fatherly Leader: North Korea and the Kim Dynasty (New York: Thomas Dunne Books [St. Martin's Griffin], 2006).Google Scholar

53. Noland, Marcus, “Transition from the Bottom-Up: Institutional Change in North Korea,” Working Paper, Institute for International Economics, March 20, 2006. Noland's identification of the DPRK as “patrimonial”—albeit “with a more efficient state apparatus than, for example, Iraq under Saddam Hussein”—is on p. 12.Google Scholar

54. The “Confucian” bureaucracy of imperial China, with its rigorous, objective, knowledge-based bureaucratic system of selection and advancement, clearly bears very little resemblance to the thoroughly politicized structures built by the Kims.Google Scholar

55. In this issue, Patrick McEachern makes a spirited case for interpreting the DPRK as a system in transition from “totalitarianism” to “institutional pluralism.” In particular, he claims that policymaking is increasingly influenced by state “technocrats,” notably in the foreign ministry and economic affairs. But to date there is little solid evidence of technocratic administration in North Korea, even if more state officials have university degrees than used to be the case. The International Country Risk Guide gives the DPRK a Bureaucratic Quality score of 0, the lowest possible score—and one shared by only nine other states in the entire world. And even McEachern admits that Kim Jong Il, who still plays the “central role,” has been steadily “personalizing his bureaucracy,” including the military. This suggests a trend away from technocratic administration.Google Scholar

56. Note that the DPRK's sultanism and its oppositional nationalism, though empirically intertwined, are analytically two separate aspects of its makeup; there is no logical requirement that sultanism and oppositional nationalism (and thus nuclear weapons ambitions) must go together.Google Scholar

57. For more on the basic tasks of management, see Milgrom, Paul and Roberts, John, Economics, Organization and Management (Englewood Cliffs, NJ: Prentice Hall, 1992).Google Scholar

58. The identification of a regime as sultanistic means that standard political science “bureaucratic politics” models will be useless for analyzing its decisionmaking. In sultanistic regimes, due to the penetration of state institutions by the top leader, there is no such thing as bureaucratic self-interest; bureaucrats pursue only personal self-interest.Google Scholar

59. The WMD Commission Report refers to Libya as an “inept bungler, the court jester among the band of nations seeking biological or nuclear capabilities” (ch. 2).Google Scholar

60. Woods, Kevin, Lacey, James, and Murray, Williamson, “Saddam's Delusions: The View from the Inside,” Foreign Affairs 85, no. 3 (May–June 2006), pp. 2–26.Google Scholar

61. For the Romania analogy, see Chehabi, and Linz, , Sultanistic Regimes, pp. 9, 35; and Haggard, and Noland, , Famine in North Korea, p. 211.Google Scholar

62. France-Presse, Agence, “Romania Produced Plutonium Under Ceausescu: IAEA Sources,” June 17, 1992 (accessed on Lexis-Nexis).Google Scholar

63. Koenig, Robert, “Science Emerges from the ‘Dark Age’ of the Ceausescus,” Science 280, no. 5371 (1998): 1829.Google Scholar

64. “A CANDU Fiasco in Romania,” Canadian Broadcasting Corporation television report, available at archives.cbc.ca/IDC-1-75-104-907/science_technology/candu/clip9. Note that due to a tiff over money and the Ceausescu regime's penchant for secrecy, the Canadians were largely sidelined from the project not long after it began.Google Scholar

65. Anderson, Donald, e-mail communication with the author, May 8, 2006.Google Scholar

66. For more on the Chinese success story—and on its numerous brushes with failure—see Feigenbaum, Evan, China's Techno-Warriors: National Security and Strategic Competition from the Nuclear to the Information Age (Stanford: Stanford University Press, 2003); Lewis, John Wilson and Litai, Xue, China Builds the Bomb (Stanford: Stanford University Press, 1988), and Ostrov, Benjamin C., Conquering Resources: The Growth and Decline of the PLA's Science and Technology Commission for National Defense (Armonk, NY: M. E. Sharpe, 1991).Google Scholar

67. See Garwin, Richard L. and Von Hippel, Frank N., “A Technical Analysis of North Korea's October 9 Nuclear Test,” Arms Control Today, available at www.armscontrol.org/act/2006_11/NKTestAnalysis.asp; and Kang, Jungmin and Hayes, Peter, “Technical Analysis of the DPRK Nuclear Test,” Policy Forum Online 06-89A (October 20, 2006), available at www.nautilus.org/fora/security/0689HayesKang.html.Google Scholar

68. Kang, and Hayes, , “Technical Analysis of the DPRK Nuclear Test.” Google Scholar

69. Reuters, , “CIA Says North Korea Nuclear Test Failed,” March 28, 2007.Google Scholar

70. On the regime's basic economic woes, see Haggard, and Noland, , Famine in North Korea, especially ch. 8. Haggard and Noland rightly note that the reasons for these woes may be political and institutional more than intellectual, but whatever their source they seem incredibly durable.Google Scholar

71. And the regime had also stopped inviting Soviet scientific visitors already by the early 1970s. Of course, since the end of the Cold War the DPRK may have recruited former Soviet scientists to help its program along, though its legendary xenophobia leads one to suspect that it would not put itself in a position of dependence on them. See Zhebin, Alexander, “A Political History of Soviet–North Korean Nuclear Cooperation,” in Moltz, James Clay and Mansourov, Alexandre Y., eds., The North Korean Nuclear Program: Security, Strategy, and New Perspectives from Russia (New York: Routledge, 2000), pp. 29–30.Google Scholar

72. Mansourov, Alexandre, “The Natural Disasters of the Mid-1990s and Their Impact on the Implementation of the Agreed Framework,” in Moltz, and Mansourov, , The North Korean Nuclear Program, p. 78.Google Scholar

73. “North Korea Restarts Nuclear Reactor,” Global Security Newswire, June 4, 2007, available at www.nti.org. See also “North Korea Reactor in Poor Shape,” Global Security Newswire, March 30, 2007.Google Scholar

74. It is often claimed that the DPRK's missile program proves its capacity to succeed at high-tech R&D efforts. But the missile program is no world-beater. In the decade and a half since the DPRK's first, marginally successful, intermediate-range Nodong missile test of 1993, Iran and Pakistan as well as the DPRK have tested variants of the missile. About half of these tests have been modest successes and the other half disasters. As for the DPRK's longer-range missile development efforts—theoretically of much more concern to the United States—the booster rockets in the DPRK's “satellite launcher” (Taepodong-1) test of 1998 and in its long-range, multistage Paektusan missile (Taepodong-2) test of 2006 “failed dismally,” according to noted defense analyst Anthony Cordesman. Cordesman continues with a stern lesson for the scaremongers: “In the real world, it is only possible to talk about missile performance once a system is actually deployed and tested, its warhead is known, and enough firings have taken place to confirm actual operational capability. Computer models can help, but they have proved to be wrong again, and again, and again. Speculating about guidance platforms, warhead type of weight, the size of the booster, and other technical factors is guesswork—not fact” (Anthony Cordesman, “North Korea's Missile Tests: Saber Rattling or Rocket's Red Glare?” CSIS publication, July 5, 2006, available at www.csis.org/media/csis/pubs/060705_cordesman_korea.pdf). For more on the DPRK's 1993 and 1998 tests, see Bermudez, Joseph S., “A History of Ballistic Missile Development in the DPRK,” Center for Nonproliferation Studies Occasional Paper No. 2, 1999; for more on Iran's program and the DPRK connection, see Mistry, Dinshaw, “Assessing Iran's Missile Capabilities,” Arms Control Today, October 2007.Google Scholar

75. See Rosen, , War and Human Nature, especially p. 156.Google Scholar

76. N.B.: As I write these words, it is mid-January 2008.Google Scholar

77. Hamper, , but not prevent. “Disablement” is a neologism that has been invented in the context of the Six-Party Talks to indicate a step beyond a mere “freeze,” but not all the way to “dismantlement.” Theoretically, it would take about a year of effort to bring “disabled” nuclear facilities back into operation.Google Scholar

78. And as noted previously, McEachern in this issue argues that it may not be the case.Google Scholar

79. Commission on the Intelligence Communities of the United States Regarding Weapons of Mass Destruction, Report to the President, March 31, 2005, ch. 2.Google Scholar

80. Note, however, that qualitative interpretation must also be taken into account in making this judgment. Indeed, it is inescapable that qualitative choices will drive the quantitative results. These issues are explored in Hymans, The Psychology of Nuclear Proliferation, ch. 3; a complete explanation of coding procedures is also given in ch. 3 and in the book's appendix.Google Scholar

81. Quantitatively, this relationship is expressed as follows: (# of references to the key comparison other)/(# of references to key comparison other + # of references to wider communities in which we and they play a part). For heuristic purposes, one can think of a score greater than 0.5 as reflecting an “oppositional” identity.Google Scholar

82. Quantitatively, this relationship is expressed as follows: (# of “naked” references to key comparison other)/(# of “naked” references to key comparison other + # of “screened” references to key comparison other). For heuristic purposes, one can think of a score greater than 0.5 as reflecting a “nationalist” national identity.Google Scholar