Final Authorities: Nuclear Command and Control in Nuclear Competitions

Why do some nuclear powers adopt highly centralized systems for managing their nuclear forces, even when they have seemingly strong incentives to do otherwise? Why do nuclear powers differ in how they respond to vulnerability in their nuclear forces? This dissertation compares and contrasts the rules, procedures and technologies for preventing unauthorized nuclear use and ensuring responsive performance of nuclear forces in war in the United States and the Soviet Union/Russia.

Nuclear Deterrence in the Gray Zone

Great power competition is often associated with high-intensity conventional conflict in the shadow of nuclear weapons. But there are strong indications that the United States is most likely to encounter great power conflict in the form of low-intensity conflict involving attribution challenges, political warfare and a mix of military and non-military methods---a form of conflict now termed gray zone competition. What role do nuclear weapons and nuclear deterrence play in the gray zone? This project explores the relationship between nuclear weapons and gray zone competition. A particular focus is how gray zone tactics can strengthen or undermine nuclear deterrence and strategic stability. U.S. competitors could use information and influence operations to strengthen their own strategic deterrence while undermining the perceived effectiveness and credibility of U.S. nuclear and non-nuclear deterrence capabilities. This project involves data collection on use cases of gray zone conflict and analysis. It sets out to understand what the motives, tactics and implications of gray zone competition in the nuclear shadow and to develop a set of potential U.S. responses.

Under whose Umbrella? Examining Alliance Datasets and Nuclear Security Guarantees

The effectiveness of extended deterrence, especially extended nuclear deterrence, is a key variable in theories of nuclear proliferation as well as theories of conflict and escalation in the nuclear age. Yet the effects of nuclear security commitments on both proliferation and conflict are not clear, with several well-regarded large-$n$ studies yielding contradictory findings. This paper argues that some this indeterminacy may be an artifact of the way that extended nuclear deterrence is operationalized in statistical analyses. Close examination of popular data sets of defensive alliances reveal a number of troubling inconsistencies, including a significant number of agreements that do not meet the broadest definition of an international alliance. Moreover, only a small fraction of the agreements in the data are ``defensive'' in that they involve a military threat matched by an identifiable security commitment by a nuclear power. A majority of the agreements in the data instead involve security commitments to states that do not face significant insecurity. I argue that empirical analyses that do not take into account this significant variation in alliance agreements risk making theoretically and empirically dubious inferences. I replicate two recent studies and find that their findings are indeed sensitive to alternative specifications of the alliances variable.