1. For example, power can be balanced between two states, between two alliance systems, or among many. A balance of power can be relatively cooperative if the “poles” act with conscious restraint in efforts to maintain equilibrium, or it can be highly competitive if each pole seeks to gain power at the other's expense. Alliance systems can be tightly held “empires,” or they can be looser associations of states akin to spheres of influence. Alliance leaders can try to reformulate domestic politics within their subordinate states, or they can join with other states regardless of their internal characteristics.

2. From Ruggie, I abstract three features that distinguish multilateralism from other patterns of relations between states: indivisibility, generalized organizing principles, and diffuse reciprocity. See John Gerard Ruggie, “Multilateralism: The Anatomy of an Institution,” in this issue of IO.

3. See ibid. Ruggie discusses “collective security,” which is one possible manifestation of multilateralism in security. It is not the only one, although a full collective security system would be the modal case.

4. This is consistent with the standard neorealist arguments about alliances. See Waltz, Kenneth, Theory of International Politics (Reading, Mass.: Addison-Wesley, 1979)Google Scholar; andWalt, Stephen M., the Origins of Alliances (Ithaca, N. Y.: Cornell University Press, 1987)Google Scholar.

5. By early 1948, the Soviet Union had concluded bilateral “treaties of friendship and mutual assistance” with Poland, Czechoslovakia, Romania, Hungary, and Bulgaria. The treaties bound each country to the Soviet Union and prevented them from dealing directly with one another on security issues. The Soviet army was given a free hand to modernize and reorganize the East European armed forces. See Johnson, A. Ross, “The Warsaw Pact: Soviet Military Policy in Eastern Europe,” in Terry, Sarah Meiklejohn, ed., Soviet Policy in Eastern Europe (New Haven, Conn.: Yale University Press, 1984), p. 259Google Scholar. Until the Warsaw Pact was signed in 1955, the Soviet Union did not even offer the facade of an integrated defense organization to its allies, as it had done earlier in the economic field through the Council for Mutual Economic Assistance. SeeHutchings, Robert L., Soviet-East European Relations: Consolidation and Conflict (Madison: University of Wisconsin Press, 1983)Google Scholar; and Wolfe, Thomas, Soviet Power and Europe (Baltimore, Md.: Johns Hopkins University Press, 1970)Google Scholar.

6. Indeed, the British and the French in particular might have been more content to the extent that bilateralism would have saved them from having to countenance and deal directly with a revitalized state and military force in Germany.

7. For a generalized discussion, see Lisa L. Martin, “Interests, Power, and Multilateralism,” International Organization, forthcoming.

8. The analogy is to transaction costs economic analysis. See Williamson, Oliver, ed., Industrial Organization (London: Edward Elgar, 1990)Google Scholar.

9. The United States also sacrificed legitimate claims to greater recompense for providing security to relatively exposed states, such as Turkey, than to well-protected states, such as Britain. In effect, the promise was to make an equal sacrifice for highly unequal causes.

10. I differ here in emphasis with Olson and Zeckhauser's classic model, which assumes that strategic nuclear deterrence is a public good and uses that assumption to analyze burden-sharing problems within NATO. See Olson, Mancur and Zeckhauser, Richard, “An Economic Theory of Alliances,” Review of Economics and Statistics 48 (08 1966), pp. 266–79CrossRefGoogle Scholar. My point is that the publicness of nuclear deterrence within NATO was a result of voluntary institutional choices which themselves need to be explained.

11. See Gaddis, John Lewis, Strategies of Containment (New York: Oxford University Press, 1982), p. 10Google Scholar; Blum, John Morton, ed., The Price of Vision: The Diary of Henry A. Wallace, 1942–46 (Boston: Houghton Mifflin, 1973), p. 146Google Scholar; and Schaller, Michael, The U. S. Crusade in China, 1938–1945 (New York: Columbia University Press, 1979), pp. 98–99 and 176–78Google Scholar.

12. Kimball includes both isolationism and messianic internationalism as manifestations of U. S. unilateralism, which he characterizes as “the American urge to go it alone in the event that others did not accept American demands.” See Kimball, Warren F., The Juggler: Franklin Roosevelt as Wartime Statesman (Princeton, N. J.: Princeton University Press, 1991), p. 84Google Scholar.

13. On these points, see Gaddis, Strategies of Containment, chap. 1; Dallek, Robert, Franklin Delano Roosevelt and American Foreign Policy, 1932–1945 (New York: Oxford University Press, 1979), especially pp. 390–91Google Scholar; and Larson, Deborah, Origins of Containment: A Psychological Explanation (Princeton, N. J.: Princeton University Press, 1985), chap. 2Google Scholar. An important exception was Roosevelt's summer 1943 decision not to share information about the atomic weapons program with the Soviet Union on a voluntary basis, but Roosevelt consistently resisted Churchill's efforts to expand the scope of this subpartnership to the exclusion of the Soviets. Kimball, correctly in my view, interprets Roosevelt's decision on the atomic issue as a way to hedge his bets on Stalin. See Kimball, The Juggler, chap. 5.

14. See Spykman, Nicholas John, America's Strategy in World Politics: The United States and the Balance of Power (New York: Harcourt, Brace, & World, 1942)Google Scholar; and Lippmann, Walter, U. S. Foreign Policy: Shield of the Republic (Boston: Little, Brown, 1943)Google Scholar. This argument had intellectual roots in British foreign policy thought as well; see, for example, SirMackinder, Halford, “The Geographical Pivot of History,” Geographical Journal 23 (04 1904), pp. 421–44CrossRefGoogle Scholar.

15. See Gaddis, , Strategies of Containment, pp. 14–19Google Scholar. In a major foreign policy address given on 16 June 1945, Truman pronounced that “unless there is complete understanding between [the] three great powers there will be no peace” and that the alternative, competition between two great powers, would end up “a truce-armistice, which will be just like the one we had in 1920.” Truman is quoted by Larson, in Origins of Containment, p. 141Google Scholar.

16. The original “long telegram” can be found in Foreign Relations of the United States (FRUS) 1946, vol. 6, pp. 696–709Google Scholar; Kennan later offered his own retrospective clarifications in his Memoirs, 1925–1950 (Boston: Little, Brown, 1967), pp. 292–95 and 354–67Google Scholar. Truman resisted similar conclusions for some time. Consider, for example, his reaction to Churchill's iron curtain speech, discussed by Larson, in Origins of Containment, pp. 264–65Google Scholar.

17. For example, in a speech given in Chicago on 18 November 1947, Secretary of State George Marshall argued vehemently against a bipolar spheres of influence “solution” for Europe; see Department of State Bulletin, 23 November 1947, p. 1026. For a discussion of the views of Charles Bohlen and William Bullit, two extremely hard-line critics of Soviet behavior who retained substantial influence in Washington, see Brownell, Will and Billings, Richard, So Close to Greatness: A Biography of William C. Bullit (New York: Macmillan, 1988)Google Scholar.

18. Office of Strategic Services, “American Security Interests in the European Settlement,” research and analysis report no. 2284, 29 06 1944, Modern Military Records Branch, National Archives,Washington, D.C.Google Scholar.

19. Of course, Kennan did not believe that the United States was as vulnerable to this as was the Soviet Union, but he did not view the United States as immune. See, for example, “George Kennan to Cecil B. Lyon, 13 October 1947,” Department of State Policy Planning Staff Records, Chronological File, Box 33, Diplomatic Branch, National Archives, in which Kennan argued that the “first and primary element of ‘containment'… [is the] encouragement and development of other forces resistant to communism” and that “it should be a cardinal point of our policy to see to it that other elements of independent power are developed on the Eurasian land mass as rapidly as possible.” John D. Hickerson, director of the State Department Office of European Affairs would later write that the notion was to foster a European “third force,” which he characterized as “a real European organization strong enough to say no both to the Soviet Union and to the United States.” See “Hickerson Memorandum, 21 January 1948,” FRUS 1948, vol. 3, p. 11Google Scholar.

20. Kennan's memos show that although he had few illusions about the resistance which his ideas concerning the restoration of German power would encounter, he viewed the alternative of a bipolar world as certainly more dangerous. In that vein, Kennan would later write somewhat sardonically that “the only thing wrong with Hitler's new order was that it was Hitler's.” See “George Kennan to Dean Acheson, 18 October 1949,” Policy Planning Staff Records, Box 32, Diplomatic Branch, National Archives. Kennan thought that the U.S. government could successfully combine reassurance, persuasion, and gentle coercion to convince others to accept German power in the context of a united Europe.

21. The toleration of diversity was a key theme in Kennan's fundamental optimism about the United States and the strategy of containment. As Kennan stated in his October 1948 speech to the Naval War College, the Soviet government's inability to tolerate diversity would turn out to be the “weakest and most vulnerable point in the Kremlin armor”; quoted by Gaddis in Strategies of Containment, pp. 43–44.

22. Gaddis, , Strategies of Containment, p. 42Google Scholar.

23. See Gaddis, John Lewis, The Long Peace: Inquiries into the History of the Cold War (New York: Oxford University Press, 1987), pp. 43, 57–58, and 64Google Scholar. I discuss the State Department's evolving position on military guarantees in the next section of my article.

24. Kennan fully expected that an “independent” Europe would find its interests coincident with those of the United States on most of the important issues. The vision was of a U.S. West European partnership of sorts, one that would be robust enough to withstand the Soviet threat but still sufficiently contingent to restrain the United States. For a different view that places stronger emphasis on the importance of lingering American isolationism, see Howard, Michael, “Introduction,” in Riste, Olav, ed., Western Security: The Formative Years (New York: Columbia University Press, 1985), pp. 11–22Google Scholar.

25. Both of these assumptions underlay Kennan's argument in the “long telegram.” They were also consistent with the U.S. military's assessments of Soviet army capabilities following the war. See Poole, Walter S., “From Conciliation to Containment: The JCS and the Coming of the Cold War, 1945–6,” Military Affairs 42 (02 1978), pp. 12–15CrossRefGoogle Scholar.

26. In late February 1947, Acheson told his journalist friend Louis Fischer that he was convinced that “the thing is not so urgent in Turkey but in Greece it is a matter of days”; quoted by Larson, in Origins of Containment, p. 303Google Scholar. On 26 February 1947, Marshall gave Truman a memo reporting a similar consensus among the Departments of State, War, and Navy and indicating that “we should take immediate steps to extend all possible aid to Greece”; see FRUS 1947, vol. 5, pp. 58–59Google Scholar. It was also agreed that the fall of Greece to the communists would probably not be an isolated event. Marshall, for example, warned congressional leaders in late February that “we are faced with the first crisis of a series which might extend Soviet domination to Europe, the Middle East, and Asia”; quoted by Larson in Origins of Containment, p. 306.

27. See, in particular, Acheson's, executive session testimony in U.S. Congress, Senate Foreign Relations Committee, Legislative Origins of the Truman Doctrine, 80th Congress, 1st sess., 1973, pp. 17 and 21–22Google Scholar.

28. See “Informational Objectives and Main Themes,” FRUS 1947, vol. 5, pp. 77–78Google Scholar.

29. “Clayton Memorandum, ‘The European Crisis,’ 27 May 1947,” FRUS 1947, vol. 3, p. 231Google Scholar. See also the testimony of General Lincoln, George A., member of the War Department general staff, in U.S. Congress, Senate Foreign Relations Committee, Legislative Origins of the Truman Doctrine, p. 160Google Scholar. For the military assessment of U.S. power and Soviet power at this time, see “JCS 1769/1, 29 April 1947,” FRUS 1947, vol. 1, pp. 734–41Google Scholar.

30. Clayton's first memo, dated 5 March 1947, is quoted and discussed by Garwood, Ellen C. in Will Clayton: A Short Biography (Austin: University of Texas Press, 1958), pp. 115–18Google Scholar. Clayton's second memo, which was prepared in late May 1947 as input for Marshall in drafting his 5 June 1947 speech, is abstracted in “George Kennan to Dean Acheson,” FRUS 1947, vol. 3, pp. 231–32Google Scholar.

31. It is notable that Marshall bypassed Clayton, who as Undersecretary of State for Economic Affairs had at least equal claim to the assignment, and instead called on Kennan and the policy planning staff to head the central study on the Program for European Recovery. The policy planning staff argued that the European countries should themselves author the plan. See Gimbel, John, The Origins of the Marshall Plan (Stanford, Calif.: Stanford University Press, 1976), pp. 199–203Google Scholar.

32. “Report of the Policy Planning Staff,” FRUS 1947, vol. 1, p. 772Google Scholar.

33. “George Kennan to Cecil B. Lyon, 13 October 1947,” Department of State Policy Planning Staff Records, Chronological File, Box 33, Diplomatic Branch, National Archives; emphasis added.

34. See U.S. Congress, Senate Committee on Foreign Relations, European Recovery Program, 80th Congress, 2d sess., January 1948. There were few objections to this scenario during the hearings. For an isolated example, see the comments of Senator Bourke Hickenlooper (p. 490), who doubted that the Europeans could sustain the requisite military capabilities under any circumstances.

35. Many of the factors favoring bilateralism that I discussed in the previous section would have applied to Marshall Plan aid as well. Apart from the gain to the United States of being able to differentiate among aid recipients, the bilateral alternative would have significantly reduced transaction costs, since the multilateral solution required getting sixteen states to agree on a comprehensive plan. As it turned out, an agreement to establish the OEEC (the forerunner to the Organization for Economic Cooperation and Development, or OECD) was not completed until April 1948. For a different view, see DePorte, A. W., Europe Between the Superpowers (New Haven, Conn.: Yale University Press, 1979), chap. 7Google Scholar.

36. In The Long Peace, p. 69, Gaddis quotes Bevin telling his cabinet that “the closest association with the United States is essential” for defense, despite the fact “that such a policy might well require the subordination of British and European interests to those of the United States.”

37. See “Inverchapel to Marshall, 13 January 1948,” FRUS 1948, vol. 3, pp. 4–5Google Scholar. See also “Memorandum of Conversation by Lovett,” p. 13; “Inverchapel to Lovett,” pp. 14–15; and “Lovett to Inverchapel,” p. 17.

38. For a description of the French concern expressed at the July 1947 meeting on Marshall Plan aid, see Ireland, Timothy, Creating the Entangling Alliance: The Origins of the North Atlantic Treaty Organization (Westport, Conn.: Greenwood Press, 1981), pp. 48–58Google Scholar.

39. The United States made only minor concessions in the interest of bolstering a fragile government in Paris. See ibid., pp. 50 and 60–63.

40. In less than a year, Soviet-backed communist parties had gained nearly exclusive control over governments in Budapest, Sofia, Bucharest, Warsaw, and Prague. While substantial influence over foreign policy might have been consistent with the American view of “loose spheres,” the infiltration and exclusive control of domestic politics by forceful subversion, secret police activity, and the reorganization of the military under the Soviet army was not.

41. In an emergency telegram sent right after the coup, Marshall said that he now envisioned a long-term and direct U.S. involvement in Europe, including “protracted security guarantees.” See “Marshall to Douglas, 28 February 1948,” FRUS 1948, vol. 2, p. 101Google Scholar. Marshall was supported by Undersecretary of State Robert Lovett and by Hickerson, both of whom now agreed that the time had come for the United States to take on formal security treaty obligations in Europe. See Ireland, Creating the Entangling Alliance, p. 80.

42. One report recommended that the United States make even its informal support contingent on the willingness of the Brussels Pact members to expand their organization to include, among other West European nations, Germany. See “Butler Memorandum, 19 March 1948,” FRUS 1948, vol. 3, pp. 58–59Google Scholar; and “Report of the Policy Planning Staff, 23 March 1948,” FRUS 1948, vol. 3, pp. 62–63Google Scholar.

43. And, most important, only a formal U.S. commitment would calm the French about eventual German involvement. See Ireland, Creating the Entangling Alliance, pp. 83–85.

44. This language was a compromise outcome of discussions between Marshall, Lovett, John Foster Dulles, and Senators Arthur H. Vandenberg and Tom Connally. Vandenberg worried that even this much would encourage European dependence on the United States, but he nevertheless came close to accepting the administration's preferences, as put forward in an NSC report of 13 April 1948. See “Souers to the NSC,” FRUS 1948, vol. 3, pp. 86–87Google Scholar. The Vandenberg resolution, which passed the Senate on 11 June 1948, incorporates similar language.

45. See Gaddis, , Strategies of Containment, p. 72Google Scholar. The French government took a similar position on 20 May 1948, after the Senate Foreign Relations Committee informed the French of the precise language specifying the nature of the U.S. commitment that it was putting together for the Vandenberg resolution. See “Douglas to Marshall,” FRUS 1948, vol. 2, pp. 266–68Google Scholar.

46. Even this was qualified: Lovett, following policy recommendations set out in an NSC report of 28 June 1948, told the British that the United States would not offer even an informal commitment until the Western Union (the defense organ of the Brussels Pact) moved to organize itself more effectively. See Ireland, Creating the Entangling Alliance, pp. 100–103; and Kennan, Memoirs, 1925–1950, p. 407.

47. “Memorandum of Conversation by Bohlen,” FRUS 1948, vol. 3, p. 206Google Scholar.

48. “Memorandum by the Participants in the Washington Security Talks, July 6 to September 9, Submitted to Their Respective Governments for Study and Comment,” FRUS 1948, vol. 3, p. 239Google Scholar.

49. The compromise draft reflected the American position almost entirely on these important points. For a comparison of the Europeans’ preferred wording, the Americans’ preferred wording, and the eventual compromise, see Ireland, Creating the Entangling Alliance, p. 106.

50. See “Considerations Affecting the Conclusion of a North Atlantic Treaty,” FRUS 1948, vol. 3, pp. 284–88Google Scholar. Kennan continued to argue this line until he left the policy planning staff at the end of 1949. As late as April 1949, he initiated a new study aimed at determining “whether the emergence of a unified Western Europe postulates the formation of a third world power of approximately equal strength to the United States and the Soviet Union.” Kennan was quoted by Gaddis, in The Long Peace, p. 67Google Scholar. See also “Bruce to Acheson, 22 October 1949,” FRUS 1949, vol. 4, p. 343Google Scholar.

51. See, in particular, Articles 3 and 5 of the North Atlantic Treaty, reproduced in The North Atlantic Treaty Organization: Facts and Figures (Brussels: NATO Information Service, 1989), pp. 376–78Google Scholar.

52. The United States even rejected pleas from the French government to participate in defense planning groups for the western Europe region, northern Europe region, and southern Europewestern Mediterranean region, limiting its activities to the groups that would draw up defense plans for the North Atlantic Ocean region and the Canada-U.S. region. See “Report of the Working Group on Organization to the North Atlantic Council,” FRUS 1949, vol. 4, pp. 322–36Google Scholar.

53. The invasion lent some credibility to the argument that the West might anticipate a surprise attack in Europe once Soviet army capabilities had been adequately expanded. On this point, see “NSC-68, United States Objectives and Programs for National Security, 14 April 1950,” FRUS 1950, vol. 1, pp. 251 and 267Google Scholar. For at least a short time in June and early July 1950, the notion that Korea might be a “feint” devised to lure American forces and tie them down in the periphery while the Soviets prepared for an attack on Europe was taken seriously in Washington and elsewhere. See Gaddis, , Strategies of Containment, p. 110Google Scholar.

54. The Council Deputies, made up of deputies to the foreign ministers of each of the NATO countries, was established in May 1950 to meet in continuous session and act essentially as a secretariat for the North Atlantic Council.

55. The New York Times, 10 September 1950, p. 1.

56. The North Atlantic Council Resolution of 26 September 1950 is discussed in “Acheson to Webb,” FRUS 1950, vol. 3, p. 350Google Scholar.

57. See “Report by the North Atlantic Military Committee to the North Atlantic Defense Committee,” FRUS 1950, vol. 3, p. 557Google Scholar; and“U.S. Delegation Minutes of Second Meeting of Sixth Session of North Atlantic Treaty Council with Defense Ministers,” FRUS 1950, vol. 3, pp. 595–96Google Scholar. Acheson informed the Congress in February 1951 that the United States would send four additional divisions to add to the two already on occupation duty in Germany.

58. The reorganizations took place in December 1950, April 1951, and June 1951, respectively.

59. NATO forces grew from about fifteen divisions and fewer than 1,000 aircraft in early 1951 to about thirty-five divisions and about 3,000 aircraft by December 1951. See Ismay, Lord, NATO: The First Five Years, 1949–1954 (Paris: NATO, 1954), pp. 40 and 101–2Google Scholar.

60. The French preferred to use the European Coal and Steel Community, rather than NATO, as a way of incorporating West Germany into Europe. Another alternative at this juncture would have been for the United States to cut bilateral deals with Germany and with France on separate terms, as I suggested earlier. See McGeehan, Robert, The German Re-armament Question (Urbana: University of Illinois Press, 1971), p. 161Google Scholar.

61. It is well known that Eisenhower feared that inappropriately large defense expenditures threatened to cripple the U.S. economy. At the same time, the new president had a strong desire to regain a level of positive initiative in American foreign policy, which he felt had deteriorated under Truman. See, for example, Huntington, Samuel, The Common Defense (New York: Columbia University Press, 1961), pp. 64–88Google Scholar; and Gaddis, , Strategies of Containment, pp. 146–47Google Scholar.

62. The North Atlantic Council later set a goal of thirty nuclear-armed, active-duty divisions to defend the European central region. See Gaddis, Strategies of Containment, pp. 148–50; and Huntington, The Common Defense, pp. 80–81. See also Enthoven, Alain C. and Smith, K. Wayne, How Much Is Enough ? Shaping the Defense Program, 1961–1969 (New York: Harper & Row, 1971), pp. 120–21Google Scholar.

63. Eisenhower used phrases such as “general mobilization,” “regimentation,” and “garrison state” to capture his concerns about the impact that a prolonged bipolar cold war might have on U.S. democratic institutions. He worried that “all that we are striving to defend would be weakened and… could disappear” under the unmitigated pressure of this kind of confrontation. Gaddis, quotes Eisenhower, in a summary discussion in Strategies of Containment, pp. 133–35Google Scholar.

64. “Memorandum of Conference Between the President and Andrew Goodpaster, 2 October 1956,” White House Office (WHO), Office of the Staff Secretary (OSS), International Trips and Meetings (ITM), Box 3, Eisenhower Presidential Library (EPL), Abilene, Kansas.

65. As Steinbruner put it, Eisenhower firmly believed that a situation in which “nations which had long dominated world politics had been eased into subordinate relationships and no longer controlled the forces upon which the defense of their people and sovereign territory rested… did not square with the tradition of Europe and its historical sense of identity.” See Steinbruner, John D., The Cybernetic Theory of Decision (Princeton, N.J.: Princeton University Press), p. 171Google Scholar.

66. Press conference of 3 February 1960, reported in The New York Times, 4 February 1960, p. 1.

67. In the President's view, the Europeans’ ability to free ride on the American security commitment had in fact led to some of the consequences that Kennan had predicted and feared. Eisenhower saw the French rejection of the European Defense Community, which he had “swore, prayed, almost wept for,” as a critical turning point in this story. See “Memorandum of Conversation Between the President and Paul-Henri Spaak, November 1959,” WHO, OSS, ITM, Box 5, EPL.

68. “Transcript of Press Conference, 3 February 1960,” Ann Whitman File, Press Conference Series, Box 3, EPL.

69. “Memorandum of Conference with the President, 13 September 1960,” WHO, OSS, Subject Series, Alphabetical Subseries, Box 4, EPL. Goodpaster added that Eisenhower stood firm for “the collective handling of nuclear weapons” within the alliance because he “thought it clear that we must carry out cooperation of political significance with the others if we wish the alliance to be healthy.”

70. See the President's communication of 11 June 1957, in U.S. Congress, House Committee on Foreign Affairs, Report on Foreign Policy and Mutual Security, 85th Congress, 1st sess., 1957; and “Committee Report,” WHO, Office of the Special Assistant for National Security Affairs (OSANSA), Special Assistant Series, Subject Subseries, Box 7, EPL.

71. “Memorandum for Goodpaster of Meeting with Congressman Carl Durham, Chairman of the JCAE, 7 December 1957,” WHO, OSS, Subject Series, Alphabetical Subseries, Box 4, EPL.

72. Macmillan“s statement, made during a television interview in February 1958, is quoted by Freedman, Lawrence in The Evolution of Nuclear Strategy (New York: St. Martin's Press, 1981), p. 311Google Scholar. For Clement Atlee, joint possession of the bomb also symbolized the Anglo-American special relationship. SeeSchwartz, David N., NATO's Nuclear Dilemmas (Washington, D.C.: Brookings Institution, 1983), pp. 59–61Google Scholar.

73. Mendes-France is quoted by Newhouse, John in War and Peace in the Nuclear Age (New York: Knopf, 1989), p. 131Google Scholar.

74. See Gaulle, Charles de, Memoirs of Hope, trans. Kilmartin, Terence (London: Weidenfeld & Nicolson, 1971), pp. 208–9Google Scholar. See also Osgood, Robert E., NATO: The Entangling Alliance (Chicago: University of Chicago Press, 1962), pp. 276–96Google Scholar.

75. “Memorandum of Conference with the President, 8 August 1960,” WHO, OSS, ITM, Box 5, EPL. Eisenhower went on to remind NATO SACEUR Lauris Norstad that the nuclear weapons issue was not in his mind different from any other in alliance relations and that “we should be as generous with our allies in this matter as we think they should be in other questions regarding the alliance.”

76. “Memorandum of Conference with the President, 12 December 1958,” WHO, OSS, Subject Series, State Department Subseries, Box 2, EPL.

77. In January 1953-seven months before the Soviets tested an H-bomb-the NSC issued report no. 141, in which it informed the President that the Soviet Union would have from three hundred to six hundred bombs by 1955 and “that the net capability of the Soviet Union to injure the United States must already be measured in terms of many millions of casualties.” See Betts, Richard, Nuclear Blackmail and Nuclear Balance (Washington, D.C.: Brookings Institution, 1987), p. 150Google Scholar. See also pp. 164–69, in which Betts notes that Eisenhower never showed much confidence in the ability of a preemptive strike to reduce American casualties. In any event, Eisenhower severely doubted that he or any other American leader would ever be willing to launch a first strike: “This would not only be against our traditions but it would appear to be impossible that any such thing would occur.” See the private diary entry on 23 January 1956, p. 2, Ann Whitman File, Dwight David Eisenhower (DDE) Diaries, Box 12, EPL.

78. In Nuclear Blackmail and Nuclear Balance, Betts gives a balanced assessment of the doctrine of massive retaliation without cartooning it. In public, Eisenhower sometimes claimed that limited nuclear war was in fact possible. But in his private diaries and in the notes of his meetings, he consistently argued that any use of nuclear weapons would escalate to general war. He was particularly adamant on this point when it came to Europe. See Kistiakowsky, George B., A Scientist at the White House: The Private Diary of President Eisenhower's Special Assistant for Science and Technology (Cambridge, Mass.: Harvard University Press, 1976), p. 400CrossRefGoogle Scholar; Gaddis, , Strategies of Containment, p. 175Google Scholar; and Duffield, John, “The Evolution of NATO's Conventional Force Posture,” Ph.D. diss., Princeton University, Princeton, N.J., 1989, chap. 5Google Scholar. In a discussion with NATO Secretary-General Paul-Henri Spaak in October 1960, the President waived off questions about nonnuclear or limited conflicts in Europe as “academic.” See “Memorandum of Conference with the President, 3 October 1960,” WHO, OSS, ITM, Box 5, NATO 1959–60, EPL.

79. Consider, for example, Dulles's argument that while European security and safety were ensured by American nuclear weapons, “psychological and morale factors” demanded something more. The political value of “countries which themselves are subject to… intermediate missiles having their own intermediate missiles with which to hit back is an element that has to be weighed in the scales.” See “Dulles Background Briefing for American News Correspondents, 15 December 1957,” WHO, OSS, ITM, Box 4, EPL.

80. Eisenhower saw “Atoms for Peace” as an important part of this blueprint, but the European Atomic Energy Agency, founded in February 1957, was only a first step: “If the six countries set up an integrated institution possessing effective control… in the field of peaceful uses of atomic energy, control over military uses of atomic energy by these six countries would be simplified.” See “Memo of Lewis L. Strauss (Chairman of Atomic Energy Commission) to the President, 25 January 1956,” WHO, OSS, Subject Series, Alphabetical Subseries, Box 5, EPL; and “Dulles to Eisenhower, ‘European Integration and Atomic Energy,’ January 1956,” WHO, OSS, Alphabetical Series, Atomic Energy Commission, Box 3, EPL.

81. Eisenhower was supported in this vision most deeply by members of the State Department, although sometimes for different reasons and with different emphases. Dulles generally shared the President's view that a European nuclear force was the best alternative to either independent national arsenals or “neutralism” on the part of the allies. See, for example, Dulles's, statement of 17 April 1958 to the Joint Committee on Atomic Energy, reported in Department of State Bulletin, 5 05 1958, pp. 741–42Google Scholar. Europeanists in the State Department were particularly worried about Germany, and they took seriously Adenauer's protestations that his country was being singled out as less than equal. The possibility that a frustrated Germany might seek an independent nuclear force and then repudiate the West in favor of a separate peace with the Soviet Union was viewedseriously. European integration was the most promising way to solve the German problem, and the sharing of nuclear weapons in a multilateral force was one way to bring that about. See the “Bowie Report” of August 1960, officially entitled “The North Atlantic Nations: Tasks for the 1960s,” WHO, Office of the Special Assistant for Disarmament, Box 9, EPL; NSC doc. no. 5727, 1959, “Draft Statement of U.S. Policy on Germany,” WHO, NSC Series, Box 23, EPL; NSC doc. no. 6017, 8 November 1960, “NATO in the 1960s,” WHO, NSC Series, Box 11, EPL; and Kelleher, Catherine McArdle, Germany and the Politics of Nuclear Weapons (New York: Columbia University Press, 1975), pp. 21–29 and 90–94Google Scholar.

82. Testimony of Lauris Norstad, in U.S. Congress, Senate Committee on Foreign Relations, Mutual Security Act of 1958, 82d Congress, 2d sess., 1958, p. 187.

83. See Dulles's, speech entitled “The Evolution of Foreign Policy,” reported in Department of State Bulletin, 25 01 1954, pp. 107–10Google Scholar. Academics such as William Kaufmann, Bernard Brodie, and Henry Kissinger responded with critiques of massive retaliation that were later reprinted in Kaufmann, William W., ed., Military Policy and National Security (Princeton, N.J.: Princeton University Press, 1956)Google Scholar. Generals Maxwell Taylor and Curtis LeMay were key players in the military debate. See, for example, Taylor, , The Uncertain Trumpet (New York: Harper & Row, 1979), pp. 178–80. For a summary of the arguments, see Osgood, NATO: The Entangling Alliance, pp. 145–46Google Scholar.

84. Particularly troubling was the U.S. government's lukewarm response to Khrushchev's rather overt nuclear threats against the allies. On this point, see for example, Spier, Hans, “Soviet Atomic Blackmail and the North Atlantic Alliance,” World Politics 9 (04 1957), pp. 307–28CrossRefGoogle Scholar.

85. Steinbruner, , The Cybernetic Theory of Decision, p. 174Google Scholar.

86. Dulles is quoted by Ambrose, Stephen E. in Eisenhower (New York: Simon & Schuster, 1983), p. 405Google Scholar. As reported in The New York Times, 17 July 1957, p. 6, Dulles later justified the stockpile plan with the argument that the allies should not be left “in a position of suppliants… for the use of atomic weapons.… We do not ourselves want to be in a position where our allies are wholly dependent upon us. We don't think this is a healthy relationship.” Eisenhower quickly defended Dulles on this point, saying that the Europeans “ought to have the right, the opportunity, and the capability of responding in kind” to a nuclear strike. See Public Papers of Dwight D. Eisenhower, 1957 (Washington, D.C.: Government Printing Office, 1958), p. 550Google Scholar.

87. Steinbruner notes that “military planners had seriously considered directly providing the allies with both missiles and warheads as a logical extension of liberalization of weapons control in the United States.” See The Cybernetic Theory of Decision, p. 175; emphasis added.

88. For a summary discussion, see Trachtenburg, Marc, History and Strategy (Princeton, N.J.: Princeton University Press, 1991), pp. 184–86Google Scholar.

89. See Herkin, Gregg, The Winning Weapon: The Atomic Bomb in the Cold War (New York: Vintage Press, 1982), pp. 114–36Google Scholar. Originally drafted to secure civilian control over atomic technology that would promote peaceful applications and international cooperation, the McMahon Act was later amended as a result of the spy scandal and increasing concern about Soviet-American relations in the summer of 1946. Herkin notes, indicative of the change in mood surrounding the bill, that a section earlier entitled “Dissemination of Information” was changed to “Control of Information” and that the military was left with greater responsibility for the stockpile of fissionable materials than had been originally planned.

90. See The New York Times, 16 February 1957, p. 1; and Boutwell, Jeffrey D., Doty, Paul, and Treverton, Gregory F., eds., The Nuclear Confrontation in Europe (London: Croon Helm, 1985), pp. 12–13Google Scholar.

91. See Armacost, Michael H., The Politics of Weapons Innovation: The Thor-Jupiter Controversy (New York: Columbia University Press, 1969)Google Scholar; and The New York Times, 13 April 1957, pp. 1 and 6.

92. Dulles seems to have believed that Adenauer's feelings were on the whole representative of the mood in other European capitals. See “Memorandum of Conversation with Chancellor Adenauer, Paris, 14 December 1957,” Dulles Papers, General Correspondence and Memos, Box 1, EPL. Trachtenburg discusses German nuclear aspirations and American responses in History and Strategy, pp. 180–85.

93. See “Memorandum of Conversation with Chancellor Adenauer, Paris, 14 December 1957,” Dulles Papers, General Correspondence and Memos, Box 1, EPL; and “Memorandum of Conference with Secretary of Defense, Service Secretaries, Joint Chiefs, and Special Assistant for National Security Affairs, 17 June 1958,” WHO, OSANSA, Special Assistant Series, Subject Subseries, Box 7, EPL.

94. The resolution was vague on operational details, leaving it to NATO military authorities to “prepare a general plan for the posture of these weapons to be decided upon by the council later.” See “Heads of Government Meeting in Paris, 18 December 1957,” WHO, OSS, ITM, Box 5, EPL. After a heated domestic debate, the Federal Republic agreed in the summer of 1958 to take a large number of tactical nuclear systems under the dual-key arrangement. See Cioc, Mark, Atomica, Pax: The Nuclear Defense Debate in West Germany During the Adenauer Era (New York: Columbia University Press, 1988), chap. 4Google Scholar.

95. Dulles, is quoted by Armacost, in The Politics of Weapons Innovation, p. 188Google Scholar. With limited range (about 1,500 miles), liquid fuel, and above-ground basing, the Thor had all the disadvantages of being a vulnerable and tempting target for preemption by Soviet missiles.

96. The set requirement was for between three hundred and seven hundred of such missiles. See Steinbruner, The Cybernetic Theory of Decision, pp. 176–77.

97. Ibid., p. 184.

98. Dulles also argued that if the United States did not act soon, additional European states would almost certainly develop national nuclear forces on their own, and this would signal the end of both NATO and the emerging plan for European integration. Dulles's arguments are reported in Department of State Bulletin, 5 May 1958, p. 741. See also the testimony of Robert Murphy, in U.S. Congress, Joint Committee on Atomic Energy, Hearings Amending the Atomic Energy Act of 1954, 85th Congress, 2d sess., 1958.

99. The JCAE authorized the administration to transfer only enough information that allied troops would be ready to fit warheads to missiles and fire them under dual-key control in the event of war. It approved minimal substantive sharing of weapons design information, and then only with Britain. Of course, the cat was already out of the bag, since Britain had just tested its own H-bomb in 1957. See Steinbruner, The Cybernetic Theory of Decision, p. 181; and the comments by Senator John Pastore, chairman of the JCAE, in Congressional Record, 85th Congress, 2d sess., 1958, p. 11927.

100. See Steinbruner, , The Cybernetic Theory of Decision, p. 182Google Scholar. In a briefing immediately following the December 1957 NATO meeting, Dulles had foreshadowed the administration's intentions with regard to the stockpile when he commented that the United States would retain control over the warheads for legal purposes and would retain the right to withhold consent but added that in “certain unspecified contingencies consent is automatically given.” See “Dulles Background Briefing, 20 December 1957,” WHO, OSS, ITM, Box 5, EPL.

101. The negotiations were stuck over two issues. The first was whether the consortium would have a mandate to produce additional missiles for national forces once the SACEUR's requirements had been met, and the second was whether the United States would be committed to supplying warheads for any such missiles. See Schwartz, NATO's Nuclear Dilemmas, pp. 76–77; and Steinbruner, The Cybernetic Theory of Decision, pp. 184–85.

102. See “Memorandum of Conference with the President, 9 June 1959,” WHO, OSS, ITM, Box 5, EPL. The NSC endorsed the plan in August 1959, despite some expressed concerns about proliferation. The argument was that the allies were quite likely to attain nuclear weapons in any event and that the best outcome from the U.S. perspective was for this to happen in the context of “NATO arrangements… for holding custody of and controlling the use of nuclear weapons.” See NSC doc. no. 5819, “Status of Mutual Security Programs as of 30 June 1958,” WHO, OSS, NSC Series, Status of Projects Subseries, Box 8, EPL; and NSC doc. no. 5906/1, 5 August 1959, “Basic National Security Policy,” WHO, OSANSA, NSC Series, Policy Papers Subseries, Box 27, EPL.

103. See “Pasadena Speech, 11 December 1959,” Papers of General Lauris Norstad, Policy Papers, Box 1, EPL. In the summer of 1959, the President's Committee to Study the U.S. Military Assistance Program, a committee under the chairmanship of U.S. Ambassador to NATO William Draper, argued that the SACEUR's requirement for a NATO IRBM would be best met by Polaris. This long-range, solid-fueled, submarine-launched ballistic missile had been developed by the United States and could readily be adapted to land-mobile launching platforms. SHAPE agreed with the committee's argument, and by the end of 1959 the plan was for the European consortium to produce Polaris missiles from American blueprints and deploy the missiles on trucks and railroad cars. Polaris could easily strike targets deep within the Soviet Union from deployment sites in the Federal Republic.

104. See The New York Times, 3 February 1960, p. 1.

105. “Transcript of Press Conference of 3 February 1960,” Ann Whitman File, Press Conference Series, Box 9, EPL.

106. In a letter to Khrushchev on 12 March 1960, Eisenhower wrote that “states with major industrial capabilities cannot be expected to be satisfied indefinitely with a situation in which nuclear weapons were uncontrolled and they themselves do not have such weapons for their defense.” For a summary of Eisenhower's letter, see “Telegram from Herter to U.S. Mission to NATO, 16 March 1960,” Norstad Papers, Atomic-Nuclear Policy, Box 85, EPL. The President also told Goodpaster that “he had simply said in the press conference what he believed and what he said before.” See “Memorandum of Conference with the President, 10 February 1960,” WHO, OSS, Presidential Subseries, Box 4, EPL. When Herter reminded Eisenhower of the need for “new legislation to authorize transfer of atomic weapons to our allies,” the President responded with the “fixed idea that we should treat our allies properly” and that the JCAE's opposition was a source of “great trouble” in this. See “Memorandum of Conference with the President, 5 February 1960,” WHO, OSS, Subject Series, State Department Subseries, Box 4, EPL.

107. See The New York Times, 5 February 1960, pp. 1 and 12; and The New York Times, 10 February 1960, p. 4. Representative Chet Hollifield, the senior Democrat, warned the President that he was facing “the greatest debate of our generation.”

108. See “Memorandum of Conference with the President, 12 September 1960,” Ann Whitman File, DDE Diaries, Box 53, EPL.

109. Herter said that the White House had not finalized “concrete intentions” for sharing weapons with the allies; and Norstad, while admitting that the MRBM plan was designed to transform NATO into “a multilateral fourth nuclear power,” portrayed it as an “idea for discussion rather than a firm proposal.” See The New York Times, 11 March 1960, p. 1.

110. See “Talking Paper for Secretary Gates: MRBMs for NATO, 1 April 1960,” WHO, OSS, Subject Series, Defense Department Subseries, Box 2, EPL. According to Herter, the United States here “made a specific offer in fulfillment of the 1957 offer to aid in a second-generation IRBM program for NATO.” See “Telegram from Department of State to Embassies in London, Paris, Bonn, 4 June 1960,” WHO, OSS, ITM, Box 5, EPL. Eisenhower intended for launch authority to rest with the SACEUR but in “discussing the possibility that Congress might seek to stipulate that the post of Supreme Allied Commander be reserved to an American as a condition for providing nuclear weapons, the President said such a condition could not be justified and should not be contemplated.” See “Memorandum of Conference with the President: NATO Atomic Force, 4 October 1960,” WHO, OSS, Subject Series, State Department Subseries, Box 4, EPL.

111. In the Defense Department, there was early concern over the details of military arrangements for the MRBM force, such as details concerning its possible vulnerability to preemptive strike or to seizure. Later, the Defense Department would press for greater precision on the questions of control, ownership, and authorization to fire the missiles.

112. See Steinbruner, , The Cybernetic Theory of Decision, p. 188Google Scholar. Others in the State Department also worried that this might be just enough to push Germany toward a national nuclear effort, imitating the French. On Smith's logic, a Germany that was “forced” to acquire its own nuclear weapons in opposition to U.S. and NATO policy might go further than de Gaulle and actually turn away from the West to make a separate deal with the Soviet Union.

113. See Bowie's report of August 1960, entitled “The North Atlantic Nations: Tasks for the 1960s,” WHO, Office of the Special Assistant for Disarmament, Box 9, EPL. A more fully declassified version of this report is at the National Archives, although much of the detail on nuclear strategy and nuclear sharing has been removed.

114. There were several ambiguous conditions attached for moving on to the second stage, including the notion that the participating states would have to agree on a precise formula for authorizing a launch in a manner that would make the weapons invulnerable to being commandeered by any one country. In a later discussion, Bowie added the phrase “should the Europeans desire it” as an important condition for moving to the second stage. See “Memorandum of Conference with the President, 3 October 1960,” WHO, OSS, ITM, Box 5, EPL. The report itself suggested that fulfillment of new goals on conventional forces should also be a prerequisite.

115. Where Eisenhower was consistently specific was in his rejection of any bilateral alternatives for nuclear sharing. He told Norstad in no uncertain terms that “Germany, France and Britain would all want such weapons. They should be handled as NATO weapons.” See “Memorandum of Conference with the President, 3 August 1960,” WHO, OSS, Subject Series, State Department Subseries, Box 4, EPL. See also “Memorandum of Conference with the President, 18 August 1960,” WHO, OSS, Subject Series, State Department Subseries, Box 4, EPL.

116. The President told Bowie that he could “easily convert Adenauer” and that he could sell the plan to the British “in terms that they would be going back to the balance of power, contributing their wisdom, experience, and sturdiness to European affairs.” He noted that de Gaulle would be more difficult, but with an “intensive effort” he might also be brought on board. See “Memorandum of Conference with the President, 3 August 1960,” WHO, OSS, Subject Series, State Department Subseries, Box 4, EPL; and “Dillon to American Embassy in Paris, 19 August 1960,” WHO, OSS, ITM, Box 5, EPL.

117. See “NATO in the 1960s: U.S. Policy Considerations-9 September 1960 Draft, European Region,” WHO, OSS, ITM, Box 5, EPL. This paper favored the MRBM plan as a political gesture and a “symbol of NATO unity” to “assure that nations gain the self-respect and stamina to withstand Soviet bloc threats.”

118. See “Memorandum of Conference with the President, 12 September 1960,” Ann Whitman File, DDE Diaries, Staff Notes, Box 53, EPL. Chairman McCone agreed with Eisenhower “that we would not be able to have vitality on the part of our European partners as long as we refuse to give them weapons.” He also told the President that a carefully designed multilateral scheme might be acceptable to the JCAE. Eisenhower complained in response about the JCAE's power to intervene, saying “if the President had in the field of nuclear affairs the same authority the Commander in Chief has in other security affairs, the problem could be readily resolved.”

119. See “Telegram 1024 from Paris, 10 September 1960,” WHO, OSS, ITM, Box 5, EPL. Adenauer told Norstad directly that “Europe must have something in the atomic field.” Norstad responded that the “U.S. people are not going to turn over atomic weapons to any country for independent use” but offered the MRBM consortium plan as the obvious alternative. Adenauer “demonstrated great enthusiasm for the scheme.” When Spaak questioned Norstad about operational details of control over the MRBMs, Norstad reminded him that this was unimportant, since the plan was primarily a political initiative rather than a military one. He “pointed out that NATO had made much progress without answering unanswerable questions of exactly how the alliance goes to war and thought still further progress could be achieved without doing so.”

120. See “Memorandum of Conference with the President: NATO MRBM Force, 3 October 1960,” WHO, OSS, Subject Series, State Department Subseries, Box 4, EPL; and “Memorandum of Conference with the President, 3 October 1960,” WHO, OSS, ITM, Box 5, EPL. On Eisenhower's continuing frustration with the JCAE, see “Memorandum of Conference with the President, 8 November 1960,” WHO, OSS, Subject Series, Alphabetical Subseries, Box 4, EPL.

121. See “Memorandum of Conference with the President: NATO Atomic Force, 4 October 1960,” WHO, OSS, Subject Series, State Department Subseries, Box 4, EPL. Eisenhower brushed off Spaak's queries about operational issues for the force. These were unimportant, since “the obligation for the United States to act under NATO provisions specifying that an attack on one was an attack on all was clear and there was no doubt it would be observed by the United States.” His focus remained squarely on the “psychological benefit to more specifically reassuring arrangements” and on how this would “raise the morale of the NATO members.” See also “Memorandum of Conference with the President: NATO MRBM Force, 3 October 1960,” WHO, OSS, Subject Series, State Department Subseries, Box 4, EPL.

122. As it was, Eisenhower told Spaak that he would work to get congressional approval for the plan, although he again remarked in frustration that this was only necessary because “Congress had reserved for itself as far back as 1947 certain prerogatives which should belong in the executive branch.” See “Memorandum of Conference with the President: NATO Atomic Force, 4 October 1960,” WHO, OSS, Subject Series, State Department Subseries, Box 4, EPL.

123. While he professed reluctance to paint his successor into a corner on the issue, Eisenhower did make a concerted effort to set the terms of the debate for Kennedy in a way that would force the new president to confront the question of nuclear sharing early in his term.

124. See “Record of Action, NSC Meeting 467, Action no. 2334, 17 November 1960,” WHO, NSC Files, Meetings with the President, Box 5, EPL. The President also sent a personal note to Adenauer on the subject. See “President Eisenhower to Chancellor Adenauer: Cable to American Embassy, Bonn, no. 961, 26 November 1960,” WHO, OSS, ITM, Box 6, EPL.

125. See “Memo by Herter for Upcoming Ministerial Meeting,” WHO, OSS, ITM, Box 5, EPL.

126. The only caveat was that “a suitable formula to govern decision on use would have to be developed to maximize the effectiveness of this force as a deterrent and to establish its multilateral character.” See “Secretary's Statement at NATO Ministerial Meeting Under Long-Range Planning, 7 December 1960,” WHO, OSS, ITM, Box 5, EPL.

127. See ibid. See also The New York Times, 17 and 19 December 1960.

128. See Boutwell, , Doty, , and Treverton, , The Nuclear Confrontation in Europe, p. 34Google Scholar.

129. See, for example, Kennedy's speech reported in The New York Times, 9 February 1961, pp. 1 and 5. Sorensen labeled Kennedy's attitude as “pragmatic,” which meant that the President “did not look upon either the alliance or Atlantic harmony as an end in itself.… [Europe was] a necessary but not always welcome partner whose cooperation he could not always obtain, whose opinions he could not always accept, and with whom an uneasy relationship seemed inevitable.” See Sorensen, Theodore C., Kennedy (New York: Harper & Row, 1965), pp. 562–63Google Scholar.

130. By 1960, the broad consensus held by members of the national security community working inside and outside the U.S. government was that something had to be done to revamp massive retaliation and add “flexibility of response” to U.S. strategy. Eisenhower, through his last days in office, was a holdout. See “Memorandum of Conference with the President, 29 November 1960,” WHO, OSANSA, Special Assistant Series, Presidential Subseries, Box 5, EPL. Kennedy had long been a critic of massive retaliation. As a presidential candidate in 1960, he labeled it a dangerous and incredible doctrine that put the United States “into a corner where the only choice is between all or nothing at all, world devastation or submission.” See Kennedy, John F., The Strategy of Peace, ed. Nevins, Allan (New York: Harper & Row, 1960), p. 84Google Scholar.

131. See “Foreign Policy Considerations Bearing on U.S. Defense Posture, 4 February 1961,” National Security Files, Departments and Agencies, Box 273, John F. Kennedy Presidential Library (JFKL), Boston, Mass.

132. See Wohlstetter, Albert, The Delicate Balance of Terror,” Foreign Affairs 37 (01 1959), pp. 211–35CrossRefGoogle Scholar. Wohlstetter roundly criticized Eisenhower's position and argued that the United States had to think seriously about and act on the requirements for fighting a nuclear war in order to maintain deterrence.

133. See Weber, Steve, “Interactive Learning in U.S.-Soviet Arms Control,” in Breslauer, George W. and Tetlock, Philip E., eds., Learning in US and Soviet Foreign Policy (Boulder, Colo.: Westview Press, 1991), pp. 784–824Google Scholar.

134. See “A Review of North Atlantic Problems for the Future, 1 March 1961,” National Security Files, Regional Security, NATO, General, Box 221, JFKL; and “NATO and the Atlantic Nations, NSAM-40, 21 April 1961,” National Security Files, National Security Action Memoranda, Box 329, JFKL.

135. NSAM-40 labeled it “most important” to the United States “that use of nuclear weapons by the forces of other powers in Europe should be subject to U.S. veto and control.” It also indicated the following: “The United States should urge the UK to commit its strategic forces to NATO… and over the long run, it would be desirable if the British decided to phase out of the nuclear deterrent business.… The United States should not assist the French to attain a nuclear weapons capability, but should seek to respond to the French interest in matters nuclear in the other ways indicated above [ i. e., the ways indicated for the British].”

136. See “NATO and the Atlantic Nations, NSAM-40, 21 April 1961.”

137. Accordingly, Kennedy began quickly to back off from Eisenhower's MLF proposal. Speaking at the Canadian Parliament in May, the President offered to commit five Polaris submarines to NATO at once but only to “look to the possibility of eventually establishing a NATO seaborne force which would be truly multilateral in ownership and control, if this should be desired and found feasible by the allies, once NATO's non-nuclear goals have been achieved.” See The New York Times, 18 May 1961, p. 12.

138. See Betts, Nuclear Blackmail and Nuclear Balance.

139. See Schlesinger, Arthur M., A Thousand Days: John F. Kennedy in the White House (Boston: Houghton Mifflin, 1965), pp. 370–88Google Scholar; and Sorensen, Kennedy, pp. 584–86. Existing contingency plans for defending Berlin were built on the assumption that nuclear weapons would come into play early. McNamara had thought about this problem before the crisis started and had warned the President in May that a defense of Berlin meant almost immediate resort to nuclear war.

140. Schelling, Thomas, “Nuclear Strategy in the Berlin Crisis,” 5 07 1961Google Scholar memo provided to me by the author.

141. Ibid.; emphasis in original text. Schelling argued that any use of nuclear weapons outside the “master plan” would constitute “noise that might drown the message” and thus could not be tolerated.

142. See Stromseth, Jane E., Origins of Flexible Response: NATO's Debate over Strategy in the 1960's (New York: St. Martin's Press, 1988), p. 38CrossRefGoogle Scholar.

143. See “Remarks by Secretary McNamara, NATO Ministerial Meeting, 5 May 1962, Restricted Session,” pp. 18–19, a document released on 17 August 1979 under the Freedom of Information Act. In June 1962, McNamara presented a version of this argument in a speech at the University of Michigan, Ann Arbor.

144. These excerpts from McNamara's speech at the University of Michigan were quoted in The New York Times, 7 June 1962.

145. See Kelleher, , Germany and the Politics of Nuclear Weapons, p. 161Google Scholar. PALs are basically “electronic combination locks” that inhibit the arming of warheads without proper authorization.

146. Finletter tried to reassure the allies that U.S. strategic forces would meet the requirements for nuclear deterrence. To deploy MRBMs in Europe would be expensive and at best redundant; the force was said to be unnecessary if it did not devolve control and positively dangerous if it did. See Steinbruner, The Cybernetic Theory of Decision, p. 207.

147. Henry Owen, the deputy director of the policy planning staff, became a central voice in this debate when he prepared an April 1962 paper arguing that Kennedy“s speech before the Canadian Parliament the previous year could provide a basis for a new MLF proposal.

148. The administration continued to offer the vague promise that it would relinquish American control over the force at some point in the future should the Europeans achieve an unspecified degree of unity. See, for example, the statements made by Rusk and McNamara before the December 1962 NATO meeting, as quoted in The New York Times, 15 December 1962, p. 1.

149. See, for example, Schlesinger, , A Thousand Days, p. 855Google Scholar. For an alternative view, see the following articles by Buchan, Alastair: “The Reform of NATO,” Foreign Affairs 40 (01 1962), pp. 165–82CrossRefGoogle Scholar; and “The MLF: A Study in Alliance Politics,” International Affairs 40 (10 1964), pp. 619–37Google Scholar.

150. See Steinbruner, , The Cybernetic Theory of Decision, especially p. 256Google Scholar.

151. This reflects, I think, a standing tradition (frequently associated with John Stuart Mill among Americans) in liberal thought about the inherent virtues of diversity, more than a direct transplant from what we now call economic liberalism.

152. If we look only at the case of the late 1940s, this conclusion might seem tempting. The interpretation would go something like this: Once U.S. policymakers recognized that the Soviets were creating an inflexible alliance structure and that the threat to the West might be immediate and military, even the U.S. government, with its vast power resources, had little freedom to pursue multilateral ideals in its relations with the West European allies. Multilateralism was an easy victim to the imperative of providing the most security in the most efficient way possible.

153. For an interesting argument specifying conditions under which a hegemon is likely to make more limited investments of power resources in its allies, see Gowa, Joanne, “Rational Hegemons, Excludable Goods, and Small Groups,” World Politics 61 (04 1989), pp. 307–24CrossRefGoogle Scholar.

154. Structural theories do not accumulate confirmation simply by correlation with historical outcomes. To claim that an historical event is explained by a theory or that a case supports a theory's claims, the process or causal chain that takes independent variables to dependent variables in the theory should be demonstrably reproduced in the empirical case. Otherwise, the most one can possibly say is that the evidence does not disconfirm the theory, and this is an extremely weak claim when a theory permits a wide range of outcomes. I discuss this in more detail in my Cooperation and Discord in US-Soviet Arms Control (Princeton, N.J.: Princeton University Press, 1991), chaps. 1, 3, and 7Google Scholar. For a different position, see Achen, Christopher and Snidal, Duncan, “Rational Deterrence Theory and Comparative Case Studies,” World Politics 61 (01 1989), pp. 143–69CrossRefGoogle Scholar.

155. Indeed, Nixon and Kissinger later circumvented restrictions with regard to the sharing of nuclear information between the United States and France. On this point, see Ullman, Richard H., “The Covert French Connection,” Foreign Policy 75 (Summer 1989), pp. 3–33CrossRefGoogle Scholar.

156. By “positive cooperation,” I mean cooperation that is more than simply avoidance of shared aversion. See my “Realism, Detente, and Nuclear Weapons,” International Organization 44 (Winter 1990), pp. 55–82CrossRefGoogle Scholar. There were always some compromises, most obviously in the case of Turkey.

157. That is, the United States is powerful and ideologically compatible, yet relatively far away and lacking in imperial pretensions. It has nuclear weapons. Finally, it has over the course of the cold war established a firm reputation for a willingness to make sacrifices for allies when necessary. I discuss further the desirability of having the United States as a bilateral ally in “The US, the Soviet Union, and Regional Conflicts After the Cold War,” in Breslauer, George W., Kriesler, Harry, and Ward, Benjamin, eds., Beyond the Cold War: Conflict and Cooperation in the Third World (Berkeley, Calif.: Institute of International Studies, 1991), pp. 382–408Google Scholar.

158. See the London Declaration on a Transformed North Atlantic Alliance, which was issued by the heads of state and government who participated in the meeting of the North Atlantic Council in London on 5 and 6 July 1990. I discuss NATO's reformulated strategy and evolving institutional structures in “Does NATO Have a Future?” in Crawford, Beverly, ed., The Future of European Security (Berkeley, Calif.: Institute of International Studies, 1992)Google Scholar.

159. Here, I combine Keohane's and Grieco's arguments about institutions. See Keohane, Robert, “International Institutions: Two Approaches,” International Studies Quarterly 32 (12 1988), pp. 379–96CrossRefGoogle Scholar; and Grieco, Joseph M., “Anarchy and the Limits of Cooperation: A Realist Critique of the Newest Liberal Institutionalism,” International Organization 42 (Summer 1988), pp. 485–507CrossRefGoogle Scholar.