This admirably succinct monograph makes a useful contribution to our understanding of certain policy tools the United States and the Soviet Union adopted to try to curb the spread of nuclear weapons. Nonproliferation efforts have been based in part on international institutions such as the Non-Proliferation Treaty (NPT). Analysts of a realist bent, however, have long been skeptical that such endeavors can effectively prevent proliferation. Realists view security concerns as the driving force behind nuclear weapons programs, and they argue that the only way to stop proliferation is to provide states with security guarantees that obviate their need to acquire nuclear arms. The United States, with its network of defense treaties, has been the primary source of such security guarantees, particularly through efforts to extend nuclear deterrence—or a “nuclear umbrella”—over its allies. If the assurances provided by an alliance with the United States do not do the job, then Washington can also apply coercive pressure on its allies by threatening to withdraw its protection. In recent years, Alexander Lanoszka points out, numerous studies have identified this combination of assurance and coercion managed by the United States as the primary source of nonproliferation success.

Lanoszka seeks to challenge this emerging conventional wisdom, and he does so persuasively. He examines the relationship between alliances and nuclear proliferation through a set of historical case studies. He examines in depth three key cases: West Germany, Japan, and South Korea. Lanoszka conducted archival research on each of these cases, and the primary sources he cites add nuance to existing understandings of all three. He supplements the three core cases with brief discussions of five additional cases: Great Britain, France, Norway, Australia, and Taiwan.

Lanoszka begins with the observation that alliances can be effective under some conditions but not others and advances two propositions about the circumstances under which alliances are likely to inhibit states from exploring a nuclear option. He identifies one key condition with respect to assurance and another with respect to coercion. Regarding assurances, Lanoszka argues that alliances are not automatically effective; the simple fact of signing a defense treaty with the United States does not by itself prevent proliferation. Rather, security assurances are most effective when the United States places forward-deployed conventional forces in a country or its region. This provides a valuable clue about the timing of when some states began seeking a nuclear option. Lanoszka finds that an interest in nuclear weapons was most likely to be sparked when countries came to doubt the U.S. conventional commitment. This happened in several cases after the United States announced a decision to reduce troops based in a state or its region, a step that led an ally to fear abandonment by its superpower protector.

Once a U.S. ally begins seeking a nuclear capability, Washington may have to consider applying coercive pressure to get its ally to commit to nonproliferation. Lanoszka argues, however, that the United States typically finds it difficult to generate sufficient leverage to get its allies to halt nuclear activities. When the United States is able to coerce its allies successfully, this usually does not result from threats to withdraw military support but instead arises because the ally depends on the United States economically or technologically. An ally's vulnerability to U.S. threats to impose economic sanctions or cut off the supply of nuclear technology is the key condition for when coercion can be effective.

Combining the arguments about assurance and coercion leads to what Lanoszka describes as his “core message” (pp. 9, 153) or main “takeaway”: “alliances are better for deterring states from engaging in nuclear proliferation-related behaviors than for compelling them to give up their nuclear weapon programs” (p. 151). That is, if the United States sets up an alliance backed by credible conventional force commitments, the security assurance this conveys can prevent a state from starting a nuclear weapons program. But once a state starts working to develop a nuclear option, it becomes much harder to pressure that state to stop.

One nice feature of the book is that Lanoszka does not feel a need to claim that alliances are always the most important explanatory factor. In many of the cases, developments in domestic politics, such as the election of a new leader, evolving norms concerning the prestige attached to nuclear weapons, or the worldviews and ambitions of individual leaders, have a greater impact on state decisions about nonproliferation. Lanoszka depends on these alternative explanations for several of the cases in which a state eventually ratified the NPT and gave up its nuclear weapons–related activities. These alternatives enable him to conclude that, even in some cases that had nonproliferation outcomes, U.S. coercive pressures were still not highly effective. Key countries, he suggests, gave up the bomb despite U.S. pressure, not because of it.

This book is rich with important policy implications. With respect to North Korea, for example, many U.S. analysts advocate seeking to get China to apply pressure on its neighbor. But, as Lanoszka points out, if it is hard for a nuclear weapons state to coerce an ally to give up a nuclear program, then “experts might be overestimating China's ability to restrain its ally” (p. 156).

Although the analysis in Atomic Assurance is largely persuasive, it is at times too narrow and under-theorized. By focusing solely on alliances, the book has little to say about when other policy tools might be more appropriate for stemming proliferation. For instance, if Lanoszka is right about the impact of changes in conventional force postures, recent signals by the Trump administration that it wants to draw down U.S. forces in Germany could be seen as increasing the risk that Germany will pursue a nuclear option. This is unlikely. Germany has long since internalized its status as a non-nuclear weapon state. U.S. alliances and the assurances they provide remain important to nonproliferation, but they are not the only factor that matters. The future of the NPT is also critical for efforts to stop the spread of nuclear weapons, and policymakers should not allow a scholarly focus on alliances to mislead them into neglecting other nonproliferation policy tools.

The focus on alliances also creates some tunnel vision on the theory side. Lanoszka focuses on a vehicle for providing assurances—a defense pact—rather than on the strategy of assurance itself. This makes it harder to theorize about the qualities that make an assurance credible. I previously directed a project to examine the conditions under which assurances can forestall proliferation, and the findings were compiled in a volume I edited, Security Assurances and Nuclear Nonproliferation (Stanford, CA: Stanford University Press, 2012). One conclusion of that project is that assurances can be effective even when the recipient does not have a formal defense treaty with the United States. Some countries that have refrained from developing nuclear weapons, such as Sweden and Saudi Arabia, have not sought an alliance with the United States, yet both these countries could have reasonable expectations that the United States would come to their aid should they be attacked. Adding case studies of states that are not members of any alliance could have deepened the analysis of conditions that make assurances effective.

A similar observation applies to Lanoszka's argument about the importance of in-theater conventional forces. They seem most important as a symbol of the U.S. commitment, but commitment can be conveyed in other ways, too. The study of security assurances that I directed found that states care about being consulted. Mechanisms that formalize consultation procedures thus increase the confidence that states have in their ally's commitment beyond what is provided by forward deployments of troops.

Even though Lanoszka's recognition of the roles of domestic politics and individual leaders is welcome, he treats these as separate factors that provide potential alternative explanations to the effects of alliances. The book might have gained additional insights if Lanoszka had considered possible interactions across these levels of analysis. Maria Rost Rublee, for instance, has shown in Nonproliferation Norms: Why States Choose Nuclear Restraint (Athens, GA: University of Georgia Press, 2009) that security assurances can interact with domestic political debates. When internal opinion is moving toward nuclear renunciation, the offer of assurances by a nuclear-armed state can provide an acceptable fallback position for more hawkish figures in the debate, enabling them to accept the emerging domestic consensus against pursuing nuclear weapons. Greater attentiveness to interactions across external, domestic, and individual-level factors might have yielded other insights of this type.

No book has all the answers, but Lanoszka's conclusion that “the record of military alliances in curbing nuclear proliferation is … mixed and ambiguous” (p. 147) is an important finding about the limits of a major element of the nonproliferation policy toolkit. The case studies in Atomic Alliance also make their own contribution. They show just how slow and convoluted the process that leads to a nonproliferation outcome can be. Theoretical models and statistical analyses can leave the impression that there is an automatic correlation between the existence of a defense pact and a state renouncing nuclear arms. By tracing the inner workings of policymaking in a set of important cases, Lanoszka's book provides a valuable reminder of just how messy and uncertain nuclear decision-making is in real life.