MAD AND THE NUCLEAR DETERRENCE PROBLEM
THE CENTRAL military reality confronting American practitioners of nuclear deterrence--the reality that underlies debates over coupling, counterforce, and strategic defenses--is the mutual societal vulnerability which has resulted from superpower development of invulnerable capabilities for inflicting mutual assured destruction. Excellent accounts of the development of Soviet and American nuclear arms programs and of the evolution of nuclear thinking are available elsewhere; 1 the aim of the present discussion is simply to examine the logical deterrence conundrum posed by MAD.
Beginning in the mid- 1950s, it became clear to American planners that the Soviet Union would soon possess the ability to respond to U.S. use of nuclear weapons by destroying American cities and that, even if the United States struck first and attempted to blunt Soviet retribution, the United States would be unable to prevent substantial damage to its society in an all-out war between the superpowers. By the early 1960s, the risk of millions of American casualties in a nuclear exchange was large enough to constrain U.S. options for using nuclear forces during times of heightened tensions. 2 By the time the Soviets achieved rough nuclear parity with the United States in the late 1960s, both sides had arsenals more than large and secure enough to guarantee each an invulnerable ability to devastate the other's society.
While MAD thus meant that the United States had the ability to devastate the Soviet Union no matter what steps the Soviets might take, MAD also raised the possibility that the United States might be deterred from executing its deterrent threats: so long as U.S. cities are