a double-edged sword: nuclear deterrence and nuclear caution

Most empirical studies of the effects of nuclear weapons onlyconsider whether a state at a given time has nuclear arms ornot, but there are strong reasons to think that the effects of astate’s nuclear arsenal are conditioned by a range of othervariables including how nuclear weapons are managed. iargue that the effect of nuclear weapons in non-existentialdisputes is determined by what i call nuclear risk: the overalllikelihood that these weapons may be used withoutauthorization by political leaders. using a formal model, ihypothesize that higher nuclear risk leads to greater deterrentpower, but it also makes the leaders of the nuclear state morecautious if their deterrence fails. i test these two hypothesesusing the correlates of war data. i first measure nuclear riskusing a simple index, and then propose a novel bayesiantechnique for imputing risk. both hypotheses are borne out byempirical results.