An early version of this article, prepared in collaboration with Coit Blacker, was presented at the 1988 annual meeting of the American Political Science Association in Washington, D.C. I thank Jack Levy, Alexander George, Uri Bar-Joseph, Lynn Eden, George Breslauer, Michael Desch, Ted Hopf, Stephen Van Evera, Felicia Wong, Stephen Krasner, and two anonymous reviewers for helpful discussions, comments, and criticisms.

1. See, for example,Gaddis, John Lewis, “The Long Peace: Elements of Stability in the Post-War International System,” International Security 10 (Spring 1986), pp. 99–142CrossRefGoogle Scholar. For alternative views, see the recent exchange between Mueller and Jervis: Mueller, John, “The Essential Irrelevance of Nuclear Weapons: Stability in the Postwar World,”Google Scholar and Jervis, Robert, “The Political Effects of Nuclear Weapons: A Comment,” International Security 3 (Fall 1988), pp. 55–79 and 80–90Google Scholar, respectively.

2. For an excellent discussion of the wide range of postwar superpower relationships consistent with the predictions of realism, see Larson, Deborah, The Origins of Containment: A Psychological Explanation (Princeton, N.J.: Princeton University Press, 1985), pp. 19–23.Google Scholar

3. See, for example, Garthoff, Raymond L., Detente and Confrontation: American-Soviet Relations from Nixon to Reagan (Washington, D.C.: Brookings Institution, 1985)Google Scholar; Gelman, Harry, The Brezhnev Politburo and the Decline of Detente (Ithaca, N.Y.: Cornell University Press, 1984)Google Scholar; and Ulam, Adam B., Dangerous Relations: The Soviet Union in World Politics, 1970–82 (New York: Oxford University Press, 1983).Google Scholar

4. I discuss this problem at length in Explaining Cooperation in U.S.-Soviet Arms Control (Princeton, N.J.: Princeton University Press, forthcoming)Google Scholar. For a more general discussion, see Keohane, Robert O., “Theory of World Politics: Structural Realism and Beyond,” in Finifter, Ada, ed., Political Science: The State of the Discipline (Washington, D.C.: American Political Science Association, 1983), pp. 503–40.Google Scholar

5. The most widely studied example of a well-developed cooperative security arrangement between great powers in the prenuclear era is the Concert of Europe and its accompanying system of diplomacy. See, in particular, Holbraad, Carsten, The Concert of Europe: A Study in German and British International Theory, 1815–1914 (London: Longman, 1970)Google Scholar; Jervis, Robert, “From Balance to Concert: A Study of International Security Cooperation,” World Politics 38 (10 1985), pp. 58–79CrossRefGoogle Scholar; and Lauren, Paul Gordon, “Crisis Prevention in Nineteenth Century Diplomacy,” in George, Alexander, ed., Managing US-Soviet Rivalry: Problems of Crisis Prevention (Boulder, Colo.: Westview Press, 1983), pp. 31–64.Google Scholar

6. The state of mutual assured destruction does not, of course, absolutely guarantee the territorial integrity of the superpowers' homelands. States remain free to choose war, as discussed later in this article. There may yet be some incentive to cooperate to further reduce the already small possibility that nuclear war would come about through accident or inadvertence. However, the maintenance of a sufficiently large and survivable nuclear retaliatory force arguably offers a higher degree of assurance against direct attack than any preponderance of conventional defensive weapons previously available to a state.

7. For example, see Rice, Condoleeza, “SALT and the Search for a Security Regime,” in George, Alexander, Farley, Phillip, and Dallin, Alexander, eds., US-Soviet Security Cooperation (New York: Oxford University Press, 1988), pp. 293–306.Google Scholar

8. In “From Balance to Concert,” Jervis himself rejects the comparison. While he sees the Concert as exerting an autonomous impact on state expectations and behavior, he argues that the patterns of behavior in U.S.-Soviet security relations even during detente were so closely linked to immediate conceptions of self-interest that the concept of a regime is superfluous to their explanation. It is clear that the Concert included some elaborate forms of cooperation, such as joint action against revolutionary forces, which detente did not. On the other hand, I have discussed several features of U.S.-Soviet detente which not only go beyond what the Concert aspired to do in arms control and political issues but which also appear to go beyond the simple constraints of short-term self-interest.

9. It is true that U.S.-Soviet relations underwent a considerable “thaw” in the wake of this crisis and that the signing of the Limited Test Ban Treaty in 1963 was probably a closely related concomitant of that process. Nevertheless, this does not explain the move to a much more wide-ranging and extensive detente process in the early 1970s, which in fact followed a period of relative calm and stability in U.S.-Soviet relations.

10. Waltz, Kenneth, Theory of International Politics (Reading, Mass.: Addison-Wesley, 1979), p. 79.Google Scholar

11. Ibid., pp. 180–81.

12. On this point, see ibid., pp. 176–83; and Gilpin, Robert, War and Change in World Politics (New York: Cambridge University Press, 1981), pp. 213–19.CrossRefGoogle Scholar

13. The cities and the manufacturing, transportation, energy, and communication facilities of the two states are equally vulnerable. Moreover, there is virtually no technological prospect that either state will soon develop, let alone successfully deploy, a strategic defense system capable of providing a meaningful degree of protection against nuclear attack. For a comprehensive review, see Office of Technology Assessment, U.S. Congress, Ballistic Missile Defense Technologies (Washington, D.C.: Government Printing Office, 1985).Google Scholar

14. In Theory of International Politics, p. 93Google Scholar, Waltz notes that unit-level actors in anarchic systems cannot engage in specialization: “The states that are the units of international political systems are not formally differentiated by the functions they perform.”

15. I return to this issue when I discuss some of the “positive” management tasks that were performed under detente and additional tasks that may be part of a new detente in the 1990s.

16. See Waltz, , Theory of International Politics, chap. 9.Google Scholar

17. Ibid., p. 209.

18. See Kissinger, Henry, White House Years (Boston: Little, Brown, 1979), especially chap. 5.Google Scholar

19. The usual caveats apply to this treatment of “American decision makers.” Obviously, the decision-making elite is not a rational unitary actor, nor are perceptions of interest unanimously shared. My discussion focuses on mainstream beliefs and on the points of consensus that were shared among the majority of top decision makers.

20. See the text of Brezhnev's speech in honor of the fiftieth anniversary of the Revolution: “Fifty Years of Great Victories of Socialism: Report by Comrade L. I. Brezhnev, General Secretary of the CPSU Central Committee,” Pravda, 4 November 1967.

21. The caveats discussed in footnote 19 also apply here.

22. See Breslauer, George, “Why Detente Failed,” in George, Managing US–Soviet Rivalry, pp. 319–40Google Scholar. Others have offered interpretations of Soviet intentions as more aggressive, claiming that detente was seen in the Kremlin from the start as a fundamentally offensive policy that would “lull” the West into a false sense of security and permit Moscow to take advantage of the unilateral restraint that would result. For an argument attributing “offensive detente” to domestic and party politics in the Kremlin, see Gelman, , The Brezhnev Politburo and the Decline of Detente.Google Scholar

23. The Soviet leadership drew direct links between the development of Soviet “defensive might” and the emergence of greater “realism” in U.S. foreign policy in scores of major speeches between 1967 and the mid-1970s. For representative statements, see reports of Brezhnev's speeches entitled “Fifty Years of Great Victories of Socialism” and “For Strengthening the Solidarity of Communists: For a New Upswing in the Anti-Imperialist Struggle,” Pravda, 4 November 1967 and 8 June 1969. This argument was also expressed in analytic terms by Arbatov and others; see Arbatov, Georgi A., “An Event of World Significance,” USA, 08 1972, pp. 3–12.Google Scholar

24. By the late 1960s, American strategic theorists were largely agreed that ABM systems posed the primary threat to achieving a robust and stable nuclear balance in which neither side would be compelled to escalate the race in offensive systems or to use nuclear weapons first in a crisis. For a concise discussion of the logic of these arguments and their historical evolution, see Haley, P. Edward, Keithly, David M., and Merritt, Jack, eds., Nuclear Strategy, Arms Control, and the Future (Boulder, Colo.: Westview Press, 1985).Google Scholar

25. See the discussion by Smith, Gerard in Doubletalk: The Story of SALT I (New York: Doubleday, 1980), pp. 93–94 and 107–8.Google Scholar

26. See Holloway, David, The Soviet Union and the Arms Race (New Haven, Conn.: Yale University Press, 1983)Google Scholar. See also Garthoff, Raymond, “The Soviet Military and SALT,” in Valenta, Jiri and Potter, William, eds., Soviet Decisionmaking for National Security (London: Allen & Unwin, 1984), p. 152.Google Scholar

27. For examples, see Sidelnikov, Colonel I., “Peaceful Coexistence and the Security of the Peoples,” Krasnaya Zvezda, 14 08 1973Google Scholar. See also Trofimenko, Henry, Changing Attitudes Toward Deterrence, ACIS Working Paper no. 25 (Los Angeles: Center for International and Strategic Studies, 1980)Google Scholar. Trofimenko's paper is notable for what it suggests about Soviet views of the problem rather than for what it has to say about U.S. attitudes (its ostensible focus). For a detailed analysis, see Azrael, Jeremy R., The Soviet Civilian Leadership and the Military High Command, 1976–1986 (Santa Monica, Calif.: Rand Corporation, 1987), p. 13.Google Scholar

28. In 1970, the Galosh deployment was halted at sixty-four launchers arrayed in four complexes. See Prados, John, The Soviet Estimate: U.S. Intelligence and Soviet Strategic Forces (Princeton, N.J.: Princeton University Press, 1986), pp. 155–56 and 169Google Scholar. This was a significant reduction from what U.S. intelligence had estimated as the originally planned deployment. See Freedman, Lawrence, U.S. Intelligence and the Soviet Strategic Threat (Princeton, N.J.: Princeton University Press, 1986), pp. 87–90Google Scholar. For a discussion of the Soviet leadership's interest in a complete ban on ABM deployments, see Smith, , Doubletalk, pp. 116–25.Google Scholar

29. Soviet concerns on this score have been and continue to be a standard feature of military writers' commentaries. During the early 1980s, Chief of the General Staff Marshal Nikolai Ogarkov wrote that the strategic military balance during the early 1960s was an unacceptable condition with extremely disadvantageous consequences for Soviet interests. Ogarkov believed that as late as 1963, the United States was capable of launching a disarming first strike against the Soviet Union. He argued that escaping from military inferiority had been a critical achievement and should remain an unassailable priority of the Kremlin's policy to ensure that the condition of essential equality was maintained. See “Victory and the Present Day,” Izvestia, 9 May 1983Google Scholar; and “Provide a Reliable Defense for Peace,” Pravda, 23 September 1983.Google Scholar

30. For evidence of the flurry of debate, see “Soviet Hints Shift on a Missile Pact,” The New York Times, 17 February 1967Google Scholar; “Soviet ABM Shift Denied,” Washington Post, 18 February 1967Google Scholar; and “Comments by Marshal R. Malinovsky,” Pravda, 23 February 1967Google Scholar. For a general discussion, see Garthoff, Raymond L., “Mutual Deterrence and Strategic Arms Limitation in Soviet Policy,” International Security 3 (Summer 1978), pp. 112–45.CrossRefGoogle Scholar

31. For the text of the BPA, see Blacker, Coit and Duffy, Gloria, eds., International Arms Control: Issues and Agreements (Stanford, Calif.: Stanford University Press, 1985)Google Scholar. Soviet leaders argued explicitly that the BPA was the most important achievement of the detente era; they seldom referred to the ABM treaty in such terms. See, for example, “On the Fiftieth Anniversary of the Union of Soviet Socialist Republics: Report by Comrade L. I. Brezhnev, General Secretary of the CPSU Central Committee,” Pravda, 22 December 1972Google Scholar; and “On the Path of October Toward New Victories for the Cause of Communism and Peace: Report by Comrade A. A. Gromyko at the Ceremonial Session Devoted to the Fifty-Seventh Anniversary of the Great October Socialist Revolution,” Pravda, 7 November 1974Google Scholar. Academician Arbatov made essentially the same argument in a number of pieces; see, for example, “On Soviet-American Relations,” Kommunist, no. 3, February 1973, pp. 1–4Google Scholar. In contrast, American leaders placed primary emphasis on the SALT process and on the concrete achievements of the ABM treaty as the cornerstone of the detente relationship. The BPA, at least when it was signed, was not thought of as a particularly significant document. See Kissinger's discussion of the BPA in White House Years, pp. 1131–32, 1150–51, 1250, and 1253Google Scholar. See also Garthoff's discussion in Detente and Confrontation, pp. 290–98.Google Scholar

32. Quoted in Legvold, Robert, “The Concept of Power and Security in Soviet History,” Prospects of Soviet Power in the 1980s, part 1, Adelphi Paper no. 151 (London: International Institute of Strategic Studies, 1979), p. 6.Google Scholar

33. For the text of the agreement, see U.S. Arms Control and Disarmament Agency, Arms Control and Disarmament Agreements: Texts and Histories of Negotiations (Washington, D.C.: Government Printing Office, 1980), pp. 159–60.Google Scholar

34. This points to two interesting questions for further research: Under what circumstances would an agreement based on complementary interests actually contribute to the deterioration of the particular conditions that made it possible in the first place? Is it possible that an agreement based on complementary interests can transform either the environment or the parties' conceptions of their interests so that their interests become shared?

35. These expectations were made explicit in statements by Gerard Smith, Henry Kissinger, and others during the SALT I negotiations, immediately following the May 1972 summit, and for some time thereafter. For example, in comments at a Moscow press conference immediately preceding the signing of SALT I, Smith claimed that the negotiating record of the treaty implied “a commitment [on the part of the Soviet Union] not to build any more of these ICBMs that have concerned us [and] a recognition that the deterrent forces of both sides are not going to be challenged.” See White House Press Release, 26 May 1972, in U.S. Arms Control and Disarmament Agency, Documents on Disarmament, 1972 (Washington, D.C.: Government Printing Office, 05 1974), p. 212Google Scholar. Kissinger, also alluding to the issue of counterforce weapons, wrote later that SALT I and its aftermath “gave us the opportunity to determine whether detente was a tactic or a new turn in Soviet policy.” See Kissinger, , White House Years, pp. 1244–45.Google Scholar

36. The first and most poignant demonstration of the fact that U.S. decision makers had not acceded to the idea of full political equality for the Soviet Union came in the diplomacy surrounding the Yom Kippur War of 1973, not long after the SALT I treaty came into force. Kissinger's policy of exclusionary diplomacy was a great success in further reducing Soviet influence in the Middle East. Moscow complained bitterly but to no avail that American behavior in this case was inconsistent with the understandings of detente.

37. See U.S. Arms Control and Disarmament Agency, World Military Expenditures and Arms Transfers (Washington, D.C.: Government Printing Office, 1985), p. 85.Google Scholar

38. The countervailing strategy is described in Report of Secretary of Defense Harold Brown to the Congress on the FY 1981 Budget, FY 1982 Authorization Request and FY 1981–85 Defense Programs (Washington, D.C.: Government Printing Office, 1980), pp. 65–70Google Scholar. It is true that targeting doctrine has not varied over the years to the extent that declaratory policy has. Nuclear targeting plans have always included counterforce options. This does not, however, vitiate the importance of a radical change in declaratory policy. If nothing else, the shift in U.S. strategic thought had an impact on research and development priorities as well as procurement and basing decisions for nuclear forces. See Sagan, Scott, Moving Targets (Princeton, N.J.: Princeton University Press, 1989), pp. 10–58.Google Scholar

39. This theme pervaded Soviet rhetoric during the period from 1976 to 1978. For examples, see “Comrade L. I. Brezhnev's Speech at the Plenary Session of the CPSU Central Committee on 25 October 1976,” Pravda, 26 October 1976Google Scholar; “In a Friendly Atmosphere,” Pravda, 1 December 1976Google Scholar; and Arbatov, G. A., “Big Lie of Detente's Opponents,” Pravda, 5 February 1977.Google Scholar

40. For the text of the Tula speech, see “Outstanding Exploit of the Defenders of Tula: Ceremonial Meeting Dedicated to the Presentation of the Gold Star Medal to the City,” Pravda, 19 January 1977Google Scholar. Lawrence Caldwell has pointed out to me that Sovietologists remain divided over whether the Tula speech and the subsequent “line” represented disinformation on Brezhnev's part, a disingenuous effort to “cool” the Americans down through calming rhetoric, or a serious reappraisal of the relationship between military competitiveness and political detente. Much evidence supports the latter position. For additional statements echoing the Tula line, see “The Report of the CPSU Central Committee to the Twenty-Sixth Congress of the Communist Party of the Soviet Union and the Party's Immediate Tasks in the Fields of Domestic and Foreign Policy: Delivered by Comrade L. I. Brezhnev, General Secretary of the CPSU Central Committee, on 23 February 1981,” Pravda, 24 February 1981Google Scholar; and “A Mighty Factor in the Peace and Security of Peoples: Marshal D. F. Ustinov, Member of the Politburo of the CPSU Central Committee and USSR Minister of Defense,” Pravda, 23 February 1983Google Scholar. Events since that time and particularly since the appointment of Mikhail Gorbachev as General Secretary provide additional evidence of Soviet learning on this score.

41. There were, of course, individuals within and outside the administration who continued to believe that SDI could and should focus on developing systems that were capable of complete area defense and would fulfill President Reagan's original vision of rendering nuclear weapons “impotent and obsolete.” By 1986, however, they were a distinct minority. Funding priorities within the research program had also been progressively reoriented away from “visionary” high-technology concepts and toward more mundane technologies that could be deployed in the near term for hard site defense or limited protection of cities against accidental attack. By the end of the decade, few Americans or Soviets believed that SDI would in the foreseeable future pose a serious challenge to the condition of mutual vulnerability and deterrence.

42. Public pressure in the United States and in Europe—symbolized most poignantly by the growth of the “nuclear freeze” movement in the United States—obviously played a considerable role in bringing an otherwise reluctant American administration back to the bargaining table with the Soviets. But this does not weaken the power of a structural argument, since structural constraints can affect state behavior by a variety of means. Terminology aside, we should expect structural variables to express themselves in some fashion through domestic politics and decision makers' individual perceptions of interest. For an excellent theoretical discussion of this issue, see Gourevitch, Peter, “The Second Image Reversed,” International Organization 32 (Autumn 1978), pp. 881–912.CrossRefGoogle Scholar

43. The problem of relative gains vis-à-vis allies is reduced to the extent that power is highly concentrated in the poles and the risks of exit from one sphere to the other are small.

44. See May, Michael M., Bing, George F., and Steinbrunner, John D., Strategic Arms Reductions (Washington, D.C.: Brookings Institution, 1988), especially pp. 18–25.Google Scholar

45. For a report of Secretary of State James Baker's proposals, see “US May Tell Soviets: Let's Share Secrets,” The New York Times, 21 April 1989, p. 8.Google Scholar

46. The 1987 INF agreement was notable for its intrusive verification procedures that go far beyond anything envisioned by the original SALT model. Still, the superpowers have not yet concluded a substantial agreement on strategic forces or arrived at a mutually acceptable formula for thinking about modern strategic defensive systems. Yet mutual deterrence is as robust at present as it has ever been, and the probability of nuclear war remains infinitesimally low. If a “grand compromise” on strategic offensive and defensive systems were negotiated in the near future, this would constitute further evidence of cooperation beyond the bounds of shared aversion.

47. Discussing the American position on negotiations in regional conflicts, Undersecretary of State for Political Affairs Michael H. Armacost said that “our approach has been to work on all the issues across the board, make progress wherever we can, and conclude agreements when in their own terms, they meet US goals and interests.” This comes close to an explicit repudiation of linkage. See his address before the General Federation of Women's Clubs, 22 June 1988, reported in Bureau of Public Affairs, U.S. Department of State, Current Policy 1089: Regional Issues and US-Soviet Relations (Washington, D.C.: Government Printing Office, 07 1988).Google Scholar

48. Ibid. The text of the pact as signed by South Africa, Cuba, and Angola is reprinted in The New York Times, 14 December 1988, p. A14Google Scholar. While the United States acted as primary broker in the Angola case, it has been widely reported that Moscow placed considerable pressure on its reluctant Cuban allies to agree to moderate their demands, and this led to withdrawal of foreign forces. See the report of Elaine Sciolino's interview with Secretary of State Shultz, George, The New York Times, 18 December 1988, section 1, p. 22.Google Scholar

49. This is obviously not true of all cases, but the fact that it is true of some is further evidence in favor of my alternative concept of structure. The Soviets, who must be particularly reluctant to invest resources in Third World areas of marginal significance, have nonetheless taken steps toward solving problems rather than just pulling out in areas where there has seemed to be some possibility for success. For an example of how this applies in the case of Ethiopia, see “The Road Downhill from Makale,” The Economist, 18 March 1989, p. 35Google Scholar. For a similar example in the case of South Africa, see “Soviets, in Shift, Press for Accord in South Africa,” The New York Times, 16 March 1989, p. A1.Google Scholar

50. In Theory of International Politics, pp. 209–10Google Scholar, Waltz refers to the “inability of the Soviet Union, whatever its inclinations, to contribute much to the management of the nonmilitary affairs of the world.” We could add several other global issues of pressing importance that might be the subject of more extensive superpower-led cooperation in the near future, although they do not start with the letter “p.”