NUCLEAR WEAPONS AND CONFLICT LIMITATION
IN THE Introduction, in reviewing recent work on deterrence theory, we observed that the direct application of insights from empirical study of conventional deterrence might provide misleading answers about behavior in situations of nuclear deterrence. It is necessary and appropriate, therefore, carefully to consider the novel characteristics of nuclear weapons and their implication for the general logic developed in the preceding chapters. Such a consideration in turn will force us to examine the concept of conflict limitation.
What changes in the nature and practice of coercion have occurred as a result of the development of nuclear weapons and of invulnerable nuclear arsenals? What is it about nuclear war that makes it different from conventional war--and what does this imply about threats of using nuclear weapons?
It should be apparent that these questions cannot be answered on the basis of empirical examination. The necessary evidence does not exist. Nonetheless, one can reason deductively that nuclear weapons logically do make a difference--that nuclear weapons have greatly decreased the difficulty of imposing negative sanctions. It is not that the nature of pain has qualitatively changed or that a physical possi-