¹ Rational decision theory is fundamental to modern economic analysis and to the techniques of systems analysis which have been developed over the past 30 years. The influential work by Hitch, Charles and McKean, Roland, The Economics of Defense in the Nuclear Age (New York: Atheneum 1965Google Scholar), provides a textbook on the logic involved, as do Herman Kahn's On Thermonuclear War (Princeton: Princeton University Press 1961Google Scholar) and Thomas Schilling's The Strategy of Conflict (Cambridge: Harvard University Press 1960Google Scholar). Allison, Graham, in Essence of Decision (Boston: Little, Brown & Company 1972Google Scholar), has documented the pervasive influence of rational assumptions while attacking their most simplistic embodiment in highly aggregated models of government behavior. This has stimulated a discussion regarding the degree to which political analysis should disaggregate governments—whether, for example, the Soviet Union is to be interpreted as a single, coherent entity or as a disaggregated collection of organizational units, different pressure groups, and differing individuals. It is now widely conceded that the degree of disaggregation does powerfully affect the character of political analysis, and that reasonable disaggregation is required to achieve adequate understanding of modern governments. Disaggregation does not change the fundamental character of rational logic, however. Even highly sophisticated models of bureaucratic and political behavior are generally built upon the assumption that constituent actors are driven by rational calculations of their institutional and personal interests.

The conceptions of the decision process under discussion in this paper have to do with basic theories that can be worked out at any level of aggregation. Very simple examples of rational analysis are used for illustrative purposes, but the same argument applies to the more sophisticated, more disaggregated forms.

² Fred Iklé, in ”Can Deterrence Last Out the Century?” Foreign Affairs, Vol. 51 (January 1973), 267CrossRefGoogle Scholar–85, discusses the conservative assumptions used in damage calculations.

³ The aggregate measure in terms of equivalent megatonnage depends upon how the bomber force is counted. A frequently quoted public estimate can be found in the Brookings budget review: Blechman, Barry M., Gramlich, Edward M., and Hartman, Robert W., Setting National Priorities: The 7975 Budget (Washington, D.C., Brookings Institution 1974Google Scholar), III.

⁴ U.S. Congress, House, Committee on Armed Services, Hearings on Military Posture, 90th Cong., 2nd sess., 1968, pp. 8507 ft.

⁵ 90% is obviously the high end of the range of possible estimates. It includes additional forces procured to insure that the residual force surviving a massive first strike —the most catastrophic failure of deterrence—would be of sufficient size to impose unacceptable damage.

⁶ $20 billion is again the high end of the range of possible estimates. It is derived from calculations done by the Brookings Institution, which include a full accounting of direct costs and attribute a portion of administrative defense costs to the strategic forces. See Fried, Edward R., Rivlin, Alice M., Schultze, Charles C., and Teeter, Nancy H., Setting National Priorities: The 1974 Budget (Washington, D.C., Brookings Institution 1973), 296Google Scholar. Official DoD budget figures, which do not include indirect costs and even exclude some categories of direct cost, have averaged under $10 billion per year for the strategic forces.

⁷ Two recent articles by Albert Wohlstetter attack some of the more simplistic models of an arms race, and present data to indicate that the U.S.-U.S.S.R. arms competition, whatever its actual character, has not produced monotonic increases in some standard force measurements—e.g., budget levels, gross megatonnage, and equivalent megaton-nage. See Wohlstetter, Albert, ”Is There a Strategic Arms Race?” Foreign Policy Nos. 15 and 16 (Summer and Fall 1974Google Scholar). When destructive capability is measured more directly, however, with qualitative improvements in accuracy and delivery-vehicle flexibility taken into account, it is clear that the capacity of both sides has increased monotonically; few would deny that there is some form of interaction between the opposing strategic forces of the United States and the Soviet Union.

⁸ The message is reasonably clearly conveyed in the 1974 Posture Statement, U.S Department of Defense, Report of the Secretary of Defense James R. Schlesinger to the Congress on the FY 1975 Defense Budget and the FY 1975–79 Defense Program [here after referred to as Posture Statement], March 4, 1974, p. 40 f.

⁹ This argument is developed in more detail in Steinbruner, John and Carter, Barry, ”The Organizational Dimensions of the Strategic Posture: The Case for Reform,” Daedalus (Summer 1975), 131Google Scholar–54; issued as Vol. 104, No. 3 of the Proceedings of the American Academy of Arts and Sciences.

¹⁰ Thomas C. Schelling (fn. i).

¹¹ The term ”limited counterforce” is applied to the currently articulated defense policy by Greenwood, Ted and Nacht, Michael, ”The New Nuclear Debate: Sense or Nonsense?” Foreign Affairs, Vol. 52 (July 1974), 761CrossRefGoogle Scholar–80. The doctrine is officially articulated in the Posture Statement (fn. 8).

¹² ”Low” here is a relative term whose reference is the annihilating damage that modern nuclear arsenals could produce if fully utilized. It remains true that even a very limited nuclear attack would produce levels of damage properly judged enormous if compared to the standards of previous warfare. Nonetheless, the problem of deterrence persists as long as the strategic forces have not yet been used at full strength.

¹³ he United States program and the relative force balances are discussed and estimated by Secretary Schlesinger in the 1974 Posture Statement (fn. 8). Increases in accuracy and in the explosive power which can be achieved by a weapon of given weight and size serve to increase the capability of conducting counterforce attacks against hardened land-based missile sites. Since both sides significantly rely on hardened land-based missiles for their retaliatory threat, such developments seem to jeopardize their primary deterrent posture—particularly under the principle of conservative planning.

¹⁴ The logic of cybernetic decision theory and the distinction between that and the rational theory of decision is discussed in Steinbruner, The Cybernetic Theory of Decision (Princeton: Princeton University Press 1974Google Scholar).

¹⁵ Ibid., chaps. 3 and 4. ”Schelling (fn. 1).

¹⁷ See Wohlstetter, Albert, ”The Delicate Balance of Terror,” Foreign Affairs, xxxvn (January 1959), 211CrossRefGoogle Scholar–35, for an influential articulation of this argument.

¹⁸ The word ”escalation,” which is often used to label this phenomenon, belongs to the rational tradition and in popular connotation does not convey the dynamic involved. Escalation frequently suggests a more or less consciously intended sequence of increasingly violent attacks meant to put pressure on the opponent for bargaining purposes. ”Evolution” or even some appropriately grotesque construct such as ”cybela-tion” would better suggest the unwitting process involved.

¹⁹ The Guns of August (New York: Macmillan 1962Google Scholar).

²⁰ See, for example, Leitenberg, Milton, ”The Race to Oblivion,” Bulletin of the Atomic Scientists, xxx (September 1974), 12Google Scholarft.

²¹ Steinbruner (fn. 14), chap. 4.