xi, 578 p. A print copy of this thesis is available through the UO Libraries. Search the library catalog for the location and call number. / Rational deterrence theory posits that deterrence is more likely to be successful when a state credibly communicates to its adversary that it has both the capability and intent to retaliate against threats. Yet, second-generation nuclear states, which often exist in severe security environments, have largely adopted postures of nuclear ambiguity where they do not acknowledge their nuclear weapons capabilities or the circumstances under which they would use them. To date, research has been insufficiently comparative. While some existing research offers explanations for the ambiguous nuclear postures of individual countries, it does not permit us to draw inferences across cases and assess relative explanatory power. Through comparison, both within and across cases, this project develops a more general explanation of why nuclear states choose ambiguity over a visible deterrence posture. To this end, this project analyzes the nuclear postures of three countries: Israel, India, and Pakistan. Using process tracing and the congruence procedure methodology, I assess the relative validity of existing explanations for each case and then compare these findings across the three cases. This research suggests that regional security environments, characterized by disparities in power, create strong incentives for states to acquire nuclear weapons capabilities for deterrence as well as to retain an ambiguous posture. In particular, an ambiguous posture enables regional states to avoid the costs and dangers of competitive nuclear development vis-à-vis their adversaries. The three cases also suggest that patron state pressures for non-proliferation, which combine threats and incentives, are another important constraint on the nuclear posture of second-generation nuclear states. Other variables--such as the international non-proliferation regime, domestic political interests, and the personal moral reservations of some state leaders--play some role to varying degrees in individual cases. However, these effects are limited both within the broad history of individual cases as well as in cross-case comparison of the three states. Understanding these constraints is helpful for evaluating the efficacy of policy tools designed to prevent the spread of nuclear weapons as well as how to manage crises and conflicts between regional nuclear-armed states. / Committee in charge: Lars Skalnes, Chairperson, Political Science; Jane Kellett Cramer, Member, Political Science; Ronald Mitchell, Member, Political Science; Shaul Cohen, Outside Member, Geography