Todd Sechser and Matthew Fuhrmann’s book Nuclear Weapons and Coercive Diplomacy provides a sustained case against the use of nuclear weapons as a tool for compelling actors to do something they would not otherwise want to do. In their reviews, three eminent scholars, Kyle Beardsley, Dan Reiter and Nina Tannenwald, are united in their praise, calling the book “important,” “powerfully argued,” and “compelling.” They diverge significantly in their critiques, however.
H-Diplo | ISSF Roundtable XI-15
Matthew Fuhrmann and Todd S. Sechser. Nuclear Weapons and Coercive Diplomacy. New York: Cambridge University Press, 2017. ISBN: 9781107514515 (paperback, $29.99).
6 April 2020 | https://issforum.org/to/ir11-15
Roundtable Editor: Diane Labrosse | Commissioning Editor: Thomas Maddux | Production Editor: George Fujii
Introduction by Robert Trager, University of California, Los Angeles 2
Review by Kyle Beardsley, Duke University. 4
Review by Dan Reiter, Emory University. 7
Review by Nina Tannenwald, Brown University. 10
Response by Todd S. Sechser, University of Virginia, and Matthew Fuhrmann, Texas A&M University 12
Introduction by Robert Trager, University of California, Los Angeles
Todd Sechser and Matthew Fuhrmann’s book Nuclear Weapons and Coercive Diplomacy provides a sustained case against the use of nuclear weapons as a tool for compelling actors to do something they would not otherwise want to do. In their reviews, three eminent scholars, Kyle Beardsley, Dan Reiter and Nina Tannenwald, are united in their praise, calling the book “important,” “powerfully argued,” and “compelling.” They diverge significantly in their critiques, however.
Tannenwald begins by calling the book an “exemplary model of social science scholarship,” and notes some of its “stunning” findings, such as that nuclear-armed states are less successful at coercion than states without nuclear weapons. After reviewing some of the headline statistics from the across the work, she argues that, in spite of appearances, the authors make a constructivist case. Why are nuclear weapons ineffective for compellence? The reason is largely the norms against their use.
Tannenwald takes issue with the book with her argument that the authors should have extended their argument further. The authors suggest that “while compellence mostly fails, nuclear deterrence mostly works.” Tannenwald argues instead that the authors’ case largely applies to deterrence as well. Historical examples discussed in the book, she argues, actually show the limits of nuclear deterrence.
Reiter reflects on the conditions under which the authors’ argument implies that nuclear coercion will be effective. He makes the case that these conditions are somewhat more common than the Sechser and Fuhrmann suggest. While they argue nuclear brinksmanship rarely succeeds, they also point out that in two important cases, it did seem to influence events. Reiter also questions whether the “backlash” response to nuclear use would be seen as relatively unimportant in the highest stakes cases, and whether states would in that case also be less concerned with precedent setting. Perhaps the possibility of nuclear coercion should not be discounted in the most important moments.
Finally, Reiter notes that the book’s conclusions rest on a limited empirical record as there have been few (the exact number is 19) overt nuclear threats. When a traffic cop directs cars down a particular route it probably does not matter that she has a gun in her holster. Similarly, most of the consequences that are threatened in the international system may not invoke the ‘nuclear option’ at all.
Beardsley offers a variety of critiques of the work. One of these relates to selection mechanisms. Though he notes that the authors address how the selection of actors with different levels of resolve into the population of challenging states might affect the results, Beardsley contends that they “do not account for a potentially even more powerful selection effect—the choices over the development of each state’s nuclear arsenals.” States that are ambitious or in dangerous environments are more likely to develop nuclear weapons and these facts may bias the estimate of the effect of nuclear weapons on coercive success. Beardsley also questions whether some findings do not depend on arbitrary coding decisions, such as which state in a crisis is labeled the “challenger,” and provides a telling example in the Cuban Missile Crisis.
All the reviews recommend the work. In fact, even the most critical of them concludes that “all students of international politics should read this book.”
Todd S. Sechser is the Pamela Feinour Edmonds and Franklin S. Edmonds, Jr. Discovery Professor of Politics at the University of Virginia. His research interests include coercion in international relations, the political dynamics of nuclear weapons, and the psychological effects of political violence. He is coauthor of Nuclear Weapons and Coercive Diplomacy (Cambridge University Press, 2017), and his research has been published in International Organization, the American Journal of Political Science, International Studies Quarterly, the Journal of Conflict Resolution, and other journals.
Matthew Fuhrmann is Professor of Political Science and Presidential Impact Fellow at Texas A&M University. He was named an Andrew Carnegie Fellow in 2016 by the Carnegie Corporation of New York. He is the author of Atomic Assistance: How “Atoms for Peace” Programs Cause Nuclear Insecurity (Cornell University Press, 2012) and the coauthor of Nuclear Weapons and Coercive Diplomacy (Cambridge University Press, 2017). His work has been published in peer reviewed journals such as American Journal of Political Science, British Journal of Political Science, International Organization, International Security, International Studies Quarterly, Journal of Conflict Resolution, Journal of Peace Research, and Journal of Politics. He is a term member of the Council on Foreign Relations.
Robert F. Trager is Associate Professor of Political Science at the University of California, Los Angeles. He has also taught at Yale and Oxford Universities, held an Olin Fellowship at Harvard University, and worked in investment banking in New York. His book, Diplomacy: Communication and the Origins of International Order, was published in 2017 by Cambridge University Press. His articles have appeared in such journals as the American Political Science Review, the American Journal of Political Science, the Journal of Politics, International Organization, International Security, and Security Studies. Current research projects focus on the moral and emotional bases of foreign policy preferences, the strategic implications of artificial intelligence developments, the role of women in making democracies more pacific and the dynamics of affinity networks.
Kyle Beardsley (Ph.D., UCSD, 2006) is Associate Professor of Political Science at Duke University. He is co-director of the International Crisis Behavior data project. His research interests include the political consequences and causes of third-party involvement in peace processes, the nature of intrastate rebellion, the implications of gender power imbalances within and through post-conflict security forces, and the effects of nuclear-weapons proliferation on crisis behavior. His first book, The Mediation Dilemma, explores how third-party conflict management frequently does well in securing short-term peace but also can contribute to greater instability in the long run, especially when the third parties rely on leverage. His second book (co-authored with Sabrina Karim), Equal Opportunity Peacekeeping, examines the consequences of and potential solutions to gender power imbalances in peace operations.
Dan Reiter is Samuel Candler Dobbs Professor of Political Science at Emory University. He is author or editor of Crucible of Beliefs: Learning, Alliances, and World Wars (Cornell University Press, 1996), Democracies at War (with Allan Stam, Princeton, 2002), How Wars Ends (Princeton University Press, 2009), and The Sword’s Other Edge: Tradeoffs in the Pursuit of Military Effectiveness (Cambridge University Press, 2017), as well as dozens of scholarly articles. His current research projects cover foreign imposed regime change, civil-military relations, gender and conflict, alliances, military effectiveness, and other topics.
Nina Tannenwald is Director of the International Relations Program at Brown University’s Watson Institute for International and Public Affairs and a Senior Lecturer in Political Science. Her research focuses on the role of international institutions, norms and ideas in global security issues, efforts to control weapons of mass destruction, and human rights and the laws of war. Her book, The Nuclear Taboo: The United States and the Non-use of Nuclear Weapons Since 1945 was awarded the 2009 Lepgold Prize for best book in international relations. Her current research projects include the future of the nuclear normative order, the effectiveness of the laws of war, and targeted killing. In 2012-2013 she served as a Franklin Fellow in the Bureau of International Security and Nonproliferation in the U.S. State Department. She holds a master’s degree from the Columbia School of International and Public Affairs and a Ph.D. in international relations from Cornell University.
Review by Kyle Beardsley, Duke University
In Nuclear Weapons and Coercive Diplomacy, Todd Sechser and Matthew Fuhrmann ask whether states with nuclear weapons are more successful in their coercive threats—“compellent” threats which are attempts to alter another state’s behavior—than states without nuclear weapons. Through a combination of quantitative and qualitative analyses, they decisively arrive at a negative answer. “Nuclear arsenals may be good for self-defense, but they do not allow countries to dominate world politics.” (236)
As an entire package, Sechser and Fuhrmann’s book presents a compelling case against viewpoints that consider nuclear weapons to be useful for coercive statecraft. The argument and analysis will remain relevant and important as long as some states have nuclear arsenals and others strive to build their own. With that in mind, it is worth applying a critical eye to some of the elements of the project, with the hope that future scholarship will continue to improve our understanding of the advantages and limitations of nuclear arsenals as tools of statecraft. In what follows, I first provide a summary of the book’s approach, briefly comment on the merits of the project, and then raise three lines of critique.
Sechser and Fuhrmann present existing arguments from what they term the ‘Nuclear Coercionist School,’ which has concluded that nuclear weapons do indeed provide coercive advantage to the successful proliferators. They then advance a ‘Logic of Nuclear Skepticism,’ which raises three core objections to the arguments from the Coercionist School: 1) in many cases nuclear weapons are not useful or are redundant to achieving foreign policy objectives; 2) the costs to a challenger of using nuclear weapons are often too high to make a nuclear threat credible; and 3) the stakes involved in coercive threats are typically too low from the challenger’s perspective as to make the use of nuclear weapons a consideration. In addition to these objections, the Nuclear Skepticism argument also takes issue with the possibility that nuclear states are able to circumvent these credibility problems by relying on tactics of brinkmanship, which manipulate the risk of inadvertent nuclear use, as tools to compel an adversary.
The bulk of the book’s content relates to the empirical analysis, which combines a series of quantitative models that explore the potential for nuclear challengers to be more successful in their coercive threats with nineteen qualitative case studies that both illustrate a number of important cases of nuclear coercion failure and discuss how even the cases in which a nuclear state successfully coerced another state are not actually strong exemplars of nuclear coercion theory.
A Compelling Critique of Nuclear Compellence?
The preponderance of the logic and the evidence provides a convincing rebuttal of prevailing views about the potential coercive benefits of nuclear weapons. The fact that nuclear weapons—mere possession or the size of the arsenal—do not appear to be positively correlated with an ability for states to coerce other states to change their actions, and that so many cases show leaders struggle to get any coercive leverage from their nuclear arsenals, scores high in terms of persuasiveness. That being said, the following points bear on whether Sechser and Fuhrmann have settled the matter.
One line of critique takes issue with their treatment of the existing scholarship. The book claims that the Nuclear Coercionist School represents the conventional wisdom, but that betrays a certain amount of recency bias. While some recent studies, especially a prominent study by Matthew Kroenig, have advanced a coercionist argument, it is not clear that this has become conventional wisdom. In fact, in his recent book, which was published after Sechser and Fuhrmann’s, Kroenig characterizes the conventional wisdom as being rooted in the logic of mutual deterrence and thus closer to the conclusion of Sechser and Fuhrmann that nuclear weapons are rather limited in being able to do more than deter. Kroenig does not argue that his book has supplanted the earlier, foundational work by scholars such as Robert Jervis and Kenneth Waltz that established an argument of the Nuclear Revolution in which nuclear weapons are sources of restraint in international politics because of the power of mutual deterrence in the presence of second-strike capabilities. In fact, it is striking how little Sechser and Fuhrmann engage with Jervis’s arguments, even though their conclusions corroborate well, if not in full, the logic of the Nuclear Revolution.
In addition to missing an opportunity to engage with the earlier work of Jervis and Waltz, there is also not much discussion of the seminal studies of Richard Betts, Charles Glaser, and, more recently, Francis Gavin. Work by these scholars has dominated reading lists on the study of nuclear statecraft for years, and they present alternative schools of thought that do not clearly fit into a Nuclear Coercionist School or a Logic of Nuclear Skepticism. For these reasons, Nuclear Weapons and Coercive Diplomacy might not serve well as the go-to book to assign for a treatment of the evolution of thought on nuclear strategy—it would need to be paired with alternative approaches.
Another line of critique could take issue with the methods used to support the Logic of Nuclear Skepticism. Two issues in particular arise, but it is important to reiterate the earlier positive assessment that Sechser and Fuhrmann do well to pair their quantitative findings with an impressive number of case studies to bolster their claims, which do not fully hinge on the quantitative analysis. First, the book struggles with the challenge of proving a null relationship. That is, the Logic of Nuclear Skepticism posits that there should be no relationship between nuclear weapons and success in coercive diplomacy, but our tools of inference, in which we try to reject a null hypothesis in favor of an alternative hypothesis, are ill suited when the ‘null hypothesis’ is actually consistent with the expectations from the argument. Readers without social science research training might not fully grasp the fact that the statistical analyses do not provide some form of confidence with which we can conclude that nuclear weapons do not confer coercive value.
A second methods issue is that the analyses do not well account for the selection process of nuclear capabilities. Sechser and Fuhrmann do address how the selection of differently resolved actors into challenges might affect the results, but they do not account for a potentially even more powerful selection effect—the choices over the development of each state’s nuclear arsenals. States choose whether to pursue nuclear weapons with expectations over how those nuclear weapons will affect their performance in coercive diplomacy; it therefore becomes difficult to make a counterfactual claim of how states would have performed without nuclear weapons by comparing nuclear states to non-nuclear states because the nuclear states have fundamentally different prior expectations for how they are able to perform. Put more concretely, if states which have struggled, or are more ambitious, in coercive diplomacy are more likely to develop nuclear weapons, then a lack of correlation between nuclear weapons and coercive success may have nothing to do with the utility of nuclear weapons and everything to do with the underlying types of states which choose to proliferate.
A final line of critique might raise concerns about the handling of various conceptual and theoretical issues. Where, for example, is bargaining in the argument? In particular, Sechser and Fuhrmann do not well account for the problem of endogenous demands: when states do have capabilities that improve their bargaining position, they are likely to raise their demands, which should change the benchmark for what successful compellence looks like. More discussion is also needed on whether challengers and targets can really be differentiated theoretically and empirically. In the case of the Cuban Missile Crisis, Sechser and Fuhrmann treat the U.S. as the challenger and the USSR as the target state, but an alternative narrative might consider the status quo prior to the deployment of missiles to Cuba and then perceive the USSR as the challenger and as successfully coercing the US to withdraw its Jupiter missiles from Turkey and to provide assurances to the Castro regime in Cuba. Does it matter if the coercion attempt is against another nuclear power or a non-nuclear power? The major cases of failed coercion in Chapter 5 are all cases in which a nuclear state failed to coerce another nuclear state or a protégé of a nuclear state, but there is insufficient theoretical discussion about whether the most relevant domain for the Nuclear Coercionist School is in contests against nuclear or non-nuclear states. Finally, the authors do not address issues of alliances and nuclear umbrellas. Related to the previous issue, there is little discussion of how coercion might play out differently when a nuclear state challenges an ally of a nuclear state or when an ally of a nuclear state challenges another state.
In short, the book’s exposition and the analysis provide a needed corrective against notions that nuclear weapons can be useful policy instruments to improve a state’s coercive capacity. All students of international politics should read this book, as should all strategic policy planners. At the same time, the book left this reader wanting more in terms of engagement with the existing scholarship, wrestling with tough issues related to the methods of inference, and developing fully the theoretical arguments. Future scholarship might take up some of these issues and further develop our understanding.
Review by Dan Reiter, Emory University
Academics and policy-makers need to know if nuclear weapons can be useful tools of coercion. Nuclear Weapons and Coercive Diplomacy by Todd S. Sechser and Matthew Fuhrmann addresses this topic. The book builds on the extensive existing literature on nuclear coercion, providing theoretical treatment of why nuclear coercion is difficult, new primary evidence that is pertinent to some classic cases of nuclear coercion, discussion of new cases, and quantitative empirical analyses.
Their powerfully argued skepticism of the coercive power of nuclear weapons arrives at an important time. The Trump administration has expressed interest in expanding the American nuclear arsenal, and the 2018 Nuclear Posture Review speculated on uses of nuclear weapons beyond deterrence. President Donald Trump appears to believe that strong U.S. pressure, including nuclear threats, has pushed North Korea to make diplomatic concessions: “With all of the failed ‘experts’ weighing in, does anybody really believe that talks and dialogue would be going on between North and South Korea right now if I wasn’t firm, strong and willing to commit our total ‘might’ against the North.” In turn, North Korea’s past public statements suggest an interest in using nuclear threats for coercive purposes, such as ending U.S.-South Korea war games. Its state media declared in December 2017, “Ceaseless large-scale war games [by the U.S. and South Korea]… [are] creating a situation that a nuclear war may break out any moment.” Russian President Vladimir Putin’s March 2018 speech emphasized qualitative and quantitative improvements in Russia’s nuclear arsenal, developments that might make nuclear coercion seem attractive to Moscow. To all leaders considering nuclear coercion, Sechser and Fuhrmann conclusively demonstrate the limited diplomatic uses of nuclear weapons, especially beyond deterrence.
My question is: do nuclear weapons provide any coercive advantage? I agree with the book’s central argument that nuclear coercion threats face severe credibility problems, and are unlikely to be effective most of the time. However, can nuclear coercion work under any conditions? There is a difference between the 1980s view of former Secretary of Defense Robert McNamara that nuclear weapons are completely useless for anything other than deterring nuclear use, and the view that nuclear coercion is a tool in the diplomatic tool box, only useable under narrow and dire circumstances, like a fire axe under glass in an office building or rattlesnake venom in a doctor’s medical bag, but a tool of some utility nonetheless.
The first task is to discern whether the book is making the more absolutist claim that nuclear weapons can never facilitate coercion, or the more limited claim that nuclear weapons are only occasionally, even rarely, able to facilitate coercion. At some points in the book, the authors appear to make the more sweeping claim that nuclear weapons are never useful coercive tools. For example: “Thus, states may reap some benefits from their investments in atomic weapons. However, enhanced coercive leverage is not one of those perks” (15). And: “nuclear weapons do not help states throw their weight around in world politics. …Nuclear blackmail does not work” (236).
However, in other places the authors seem to allow that nuclear weapons might, at least under some circumstances, contribute to coercive success. The authors suggest that nuclear weapons might have played a role in the outcome of the Cuban Missile Crisis (203-210), though they directly argue against the claim that we might infer from the Cuban Missile Crisis that nuclear weapons are useful coercive tools (247-249). In other places, their language seems to allow for the possibility that nuclear weapons might under some conditions facilitate coercion: “These problems combine to render nuclear arsenals unhelpful in most [emphasis added] coercive contexts” (236).
Under what conditions might nuclear threats make coercive success more likely, even within the context of the authors’ theoretical argument? Though the authors shy away from directly stating when nuclear coercion might work, they do discuss factors that facilitate coercive success in general, namely the coercer’s ability to impose its will, high stakes for the coercer especially in relation to the target, and high costs of conflict for the target in relation to the coercer and in relation to the benefits at stake (30-34). Their discussions of the credibility problems of nuclear coercive threats in turn imply that nuclear coercive threats do not meet these conditions.
However, it is possible that certain narrow conditions could arise in which nuclear coercive threats could increase the likelihood of coercive success. To make this point, let’s work through the heart of the authors’ nuclear skepticism theory (45-56) to argue that conditions might exist under which nuclear weapons might facilitate coercive success. The authors first argue that nuclear threats are redundant, in the sense that especially for great powers there is little that nuclear weapons can do to a target that conventional forces could not accomplish anyway. They assert, however, that nuclear weapons might still be useful for a great power to destroy underground, hardened targets, and that nuclear weapons would be useful for non-great powers with smaller conventional militaries. They also allow for the possibility that nuclear threats offer the unique opportunity for brinkmanship (that is, conventional arms cannot nearly as easily facilitate brinkmanship), creating the risk of escalation to catastrophe, which in theory could provide coercive pressure. The authors are dismissive of the utility of nuclear brinkmanship but not completely so, arguing that “Nuclear brinkmanship rarely [emphasis added] succeeds,” consistent with their observation that nuclear brinkmanship probably played at least some role in the successful coercion of the Soviet Union during the Cuban Missile Crisis (203). They also note that Israeli nuclear brinkmanship during the 1973 Yom Kippur War might have contributed to the American decision to resupply Israel (220-221). This again opens the door for the view that nuclear brinkmanship might provide additional coercive pressure under at least some circumstances.
The second plank of their nuclear skepticism theory is that nuclear coercive threats are unlikely to succeed because of the high costs of nuclear coercion attempts, specifically a direct political backlash to nuclear coercive threats which encourages other states to balance against the coercer or acquire their own nuclear arsenals, thus setting precedent. However, conditions could arise in which these factors might not undermine the credibility of a coercive threat. Global public backlash might not be an important factor if the stakes were considered high enough. Certainly, the George W. Bush administration paid little attention to the millions of Europeans who marched against the impending invasion of Iraq in late 2002 and early 2003. Regarding the reactions of other states, these factors also might take a back seat to more immediate security concerns, and backlash might also be muted if coercion is undertaken to advance global stability interests and restore the status quo (such as turning back territorial aggression). Further, successful nuclear coercion might actually reverse proliferation incentives if it reassures the nerves of skittish allies. Certainly, a leading motive of the Kennedy administration during the Cuban Missile Crisis, the leading example of successful nuclear coercion, was that failure to turn back the missile threat would undermine U.S. credibility with its allies. Currently, an American failure to turn back North Korean aggression could stimulate nuclear proliferation incentives in South Korea.
Regarding precedent setting, the authors argue that an attempt at coercion would set a precedent, legitimating future nuclear blackmail. However, this is an argument as to why states might not attempt nuclear coercion, not why such coercion might fail. They also argue that because the use of nuclear weapons might legitimate others’ use, this would pressure coercers not to actually follow through on a nuclear threat, thereby undermining the legitimacy of the threat. Though this point has merit, concerns about legitimating future use might be secondary of the stakes of the current crisis are high enough.
The third plank of their theory is low stakes. The authors make the valid point that coercion attempts frequently concern matters of low to moderate stakes, “rarely threaten[ing] the coercer’s vital interests” (50). This is true, but it allows for the possibility that if the stakes were high enough, nuclear coercion threats might be sufficiently credible to work. The authors also observe that credibility for coercers is intrinsically lower, because actors fear losses more than they value gains. The implication of this point is that coercive threats are most likely to be credible (and effective) if they demand the de facto restoration of the status quo, such as demanding that an aggressor hand back a recently seized piece of territory. Coercion in the Cuban Missile Crisis worked in part because the U.S. demand was for a de facto restoration of the status quo, a nuclear-free Cuba.
More concretely, can we imagine plausible circumstances in the near future in which nuclear weapons coercion might work, allowing for the authors’ criteria for successful coercion, and the critiques of their own nuclear skepticism theory? The U.S. may face such circumstances, specifically a non-great power engaged in territorial aggression that gravely threatened American national security interests, including a North Korean attack on South Korean territory, or Iranian aggression on Persian Gulf nations. In these cases, overwhelming American nuclear superiority, coupled with some regional ballistic missile defense, would boost the credibility of the threat. The stakes would be high, maintaining the territorial integrity of a critical American ally or maintaining global access to oil reserves. Public backlash would likely be muted, as the U.S. would be reacting to a dangerous rogue state that was violating international norms of state sovereignty. The U.S. would also be attempting to restore the status quo, pre-aggression borders, further boosting the credibility of the threat.
That said, nuclear threats would not necessarily help against the range of plausible territorial aggressive acts. They likely would not work to turn back the Chinese capture of reefs in the South China Sea because of the size and sophistication of China’s nuclear arsenal, the non-vital stakes of the conflict, and the fact that the legal disputes over the South China Sea mean there is no clear status quo. Similarly, nuclear threats might not work against Russian aggression against a non-NATO member like Ukraine, given low U.S. interests and Russian nuclear power.
All that said, let me offer a few caveats. First, even under the most optimal conditions, nuclear threats might make coercion more likely to succeed, and could not guarantee success. Second, these and any other policy recommendations are based on a very limited empirical record. There have been few cases of nuclear coercion in which overt nuclear threats were made, and even in those cases it is difficult to conclude the degree to which nuclear threats affected the target’s behavior. Third, even if one believed that nuclear threats make coercion more likely to succeed, it is a dangerous game to play, as there is the possibility that nuclear threats might raises the risks of preemptive or accidental nuclear war.
Sechser and Fuhrmann offer a convincing argument that nuclear weapons are not broadly useful tools of coercion, though it is unclear whether they believe that nuclear weapons never aid coercion, or aid coercion only rarely. Even within the confines of their theoretical argument, under the right set of narrow conditions nuclear weapons could help coercion succeed. Foreign policy-makers still might make a place for nuclear coercion in their diplomatic and military toolbox.
Review by Nina Tannenwald, Brown University
This important book analyzes whether nuclear coercive threats in international politics are effective; that is, whether a nuclear-armed state is able to effectively coerce other states into doing what it wants. The book shows, through extensive statistical and historical analyses, that nuclear coercive threats are rarely successful and that nuclear coercion has little credibility in international politics. While nuclear deterrence is credible, nuclear compellence or coercion – a threat to get a state to do something it is currently not doing—is not.
This is a forceful argument on a big and important topic with direct contemporary policy relevance. The supporting evidence is compelling. The book is an exemplary model of social science scholarship in its sharp theorizing, clear writing, and impressive use of mixed methods, drawing on both statistical work and nineteen historical case studies. The especially clear discussion of coercion theory at the beginning of the book is useful to assign to undergraduates.
The statistical studies analyze how successful nuclear-armed states are at coercion compared with non-nuclear states. The findings contradict many of the assumptions in the literature about the coercive value of nuclear weapons. Of the 210 compellent (coercive) threats in the statistical data set, nuclear-armed states succeeded in their coercive efforts just ten times, or in just under 5 percent of the cases. Overall, in a stunning finding, nuclear-armed states were less successful at coercion than states without nuclear weapons (20 percent of the time versus 32 percent). Further, having a nuclear monopoly over the adversary provided no advantage. States possessing a nuclear monopoly prevailed in their coercive threats just 16 percent of the time, while states without nuclear weapons prevailed 33 percent of the time.
The authors also engage in a detailed qualitative analysis of all nineteen cases of explicit attempts at nuclear coercion—nine cases of failure and ten cases of success (the leading case of success is the 1962 Cuban missile crisis)—to trace out how the coercive effort played out. These historical case studies, drawing on secondary and primary source material, form a substantial part of the book and are well done. The authors make a serious effort to deal with alternative explanations, are transparent about the sometimes mixed nature of the evidence and about their efforts to weigh it, and arrive at appropriately nuanced conclusions. The case studies are a pleasure to read.
Why are nuclear coercive threats mostly ineffective, even against states that could not retaliate with nuclear weapons? The answer is that the costs of a nuclear first strike are extremely high, including not only military costs but also political, economic, and normative costs. As the authors write, there would be “direct backlash” (48) against the violation of “an important international opprobrium” against any first use of nuclear weapons (93). Indeed, although constructivism does not make an explicit appearance in this book, and the authors are not self-defined constructivists, this is actually a book of closet constructivism. Issues of norms, legitimacy, and precedent play a significant role in the explanation.
In order to see this, it is useful to look more closely at the explanation. Sechser and Fuhrmann have a three-part answer as to why nuclear coercive threats are incredible. First, in many cases, the coercer can accomplish its goals by relying on conventional military power (and nuclear weapons are not useful for dealing with human rights violations or terrorists). Second is the very high costs of a nuclear attack. Finally, in many cases, the stakes are low and do not justify the costs of a nuclear first strike.
Of these three elements, the heavy lifting in the explanation is really done by the second variable, the high costs of a nuclear first strike. As they write, “The costliness of nuclear threats is an important reason nuclear weapons are poor tools of credibility” (257). In contrast to the “nuclear coercionist” school that they are criticizing, Sechser and Fuhrmann define “costs” broadly. They argue that first use of nuclear weapons would likely have not only military costs but “direct backlash” in the form of outraged public opinion, domestic political costs for leaders, economic costs and sanctions, and diplomatic costs including possibly disrupted alliance relations. “The international community may shun nuclear users,” they note (48). In explaining why the nuclear-armed challengers in forty-one cases in the quantitative data set failed to compel their adversaries and why nuclear weapons played little role, Sechser and Fuhrmann argue that the stakes “were not worth the price of violating an important international opprobrium and risking the wrath of friends and allies” (93). Further, the first use of nuclear weapons “might set a dangerous precedent – namely that it is acceptable to use the bomb to settle international disputes” (50). In the long run it could encourage nuclear proliferation.
Together, these factors help to explain why there are relatively few cases of explicit nuclear coercion. Out of a total 210 cases of coercion, only 19 of these (9%) actually involved some kind of explicit nuclear coercive threat. According to the authors, “Like targets, challengers recognized that coercive nuclear threats are usually incredible” (78), because of the high costs, including “the international uproar” if a state used nuclear weapons first.
In short, the key explanation for why nuclear threats are incredible is that use of nuclear weapons is not perceived as legitimate. Use would provoke a tremendous political backlash and there would be significant international opprobrium against any state that launched a nuclear first strike.
Two additional points are worth mentioning. One has to do with the issue of signaling. The authors highlight signaling problems extensively in the case studies. One of the book’s arguments for why nuclear coercion is difficult is that brinkmanship signals are often not received or are misperceived. Yet, the main argument of the book suggests that effective signaling is mostly beside the point. In the end, if the nuclear threats are incredible anyway, even a signal that is effectively conveyed will likely be unsuccessful.
A second issue is whether the argument has implications for deterrence, not simply compellence. Sechser and Fuhrmann argue that it does not, because they seek to make a clear distinction between deterrence and compellence. In their view, while compellence mostly fails, nuclear deterrence mostly works. Yet here, I think, the authors miss an opportunity to cite additional evidence from the deterrence side of the nuclear threat picture that strengthens their larger point about the circumscribed value of nuclear weapons and threats.
At the end of the book, for example, they push back (correctly, in my view) against arguments of deterrence skeptics that nuclear weapons are not useful even for deterrence (255). Skeptics often give as evidence for the failure of nuclear deterrence the fact that non-nuclear states have attacked nuclear powers on several occasions (for example, Syria and Egypt attacked Israel in 1973; Iraq attacked Israel and U.S. troops with Scud missiles in the 1991 Gulf war). Although Sechser and Fuhrmann reject the skeptics’ general critique of deterrence, they avoid explaining directly these obvious cases of deterrence failure. Yet these kinds of cases do suggest the limits of nuclear deterrence in ways that are entirely consistent with the book’s central argument about the incredibility of nuclear coercive threats, and thus provide additional support for their argument. That is, the high costs and illegitimacy of a nuclear first strike undermine not only attempts at nuclear compellence but even deterrence against non-nuclear states.
One of the important contributions of this significant piece of social science research is that it has a lot to say to the nuclear policy community. A key lesson is that the conditions for nuclear coercive success are extremely narrow: the challenger must have a weak conventional military; it must be impervious to costs of all kinds, including military, political, economic, diplomatic, normative, and legitimacy; and the stakes have to be very high. The analysis also makes clear that nuclear superiority does not get you much. As the authors write, “there is no relationship between nuclear superiority and coercive victories” (254). Finally, the analysis reinforces the importance of maintaining and strengthening norms against first use of nuclear weapons. Indeed, the book ends on a strong constructivist note. As the authors summarize in the conclusion, “Coercive nuclear threats have lacked credibility for the first seven decades of the nuclear age in part because of the backlash that would follow any use of nuclear weapons for coercive purposes. Yet if leaders someday decide that this backlash is no longer likely, then the barriers to using nuclear weapons for coercion would begin to weaken. If norms against using nuclear weapons dissolve, or if the use of nuclear weapons becomes broadly acceptable, then nuclear weapons could become more useful political instruments” (257).
Given the alarming revival of nuclear brinkmanship over the last couple of years, and the downward pressure on norms of nuclear restraint, this book should be urgent reading in key foreign and defense ministries around the world, and especially in Washington, D.C.
Response by Todd S. Sechser, University of Virginia, and Matthew Fuhrmann, Texas A&M University
We thank Kyle Beardsley, Dan Reiter, Nina Tannenwald, and Robert Trager for their thoughtful and constructive comments on our book Nuclear Weapons and Coercive Diplomacy. We are grateful to have such a distinguished group of scholars engage with our work. Their comments boil down to four central points, which we address in this response.
Can nuclear coercion ever work?
The research we present in our book suggests that nuclear weapons historically have been poor tools of compellence. But Reiter raises a key question: are there conditions under which nuclear coercion might be more successful? Indeed, we address this question in the final chapter (256-258). Our theory identifies three conditions under which nuclear blackmail might succeed in the future.
First, conventionally weak states may be better positioned to use nuclear threats for coercive objectives. As we argue in the book, nuclear weapons are often redundant to a coercer’s conventional capabilities, making it less plausible that the coercer would be willing to pay the significant strategic costs of using nuclear weapons (47). Why use nuclear weapons, after all, when conventional forces can get the job done? But if a coercer’s conventional capabilities are limited, then nuclear weapons might be more critical to achieving its objectives. A coercive nuclear threat from Pakistan, for example, might be more credible than one from the United States simply because Pakistan has fewer conventional alternatives for achieving its military aims.
Second, for a coercive nuclear threat to succeed, the stakes of a crisis must be of the highest national importance to the coercer. In most cases of coercion, this condition is not met. Coercion typically is not about defending against threats to national survival; it is about making changes to a status quo that the coercer has sometimes endured for many years. In other words, coercers are usually quite capable of living without whatever territorial or policy changes they are demanding, even if they would prefer not to. This cuts against the credibility of threats to use nuclear weapons in support of those demands.
However, under narrow conditions, national survival might be at stake for a coercer. If a state’s territorial integrity or regime survival were in jeopardy, a nuclear threat would become much more credible. For example, in a scenario where thousands of Indian troops had overrun Pakistani defenses and had begun marching towards Islamabad, a Pakistani threat to strike New Delhi with nuclear weapons would have to be taken seriously. Indian leaders surely understand this, which is why India is unlikely to launch a full-scale invasion of Pakistan.
Third, nuclear blackmail could become more effective in the future if leaders come to believe that the use of nuclear weapons would not prompt a significant international backlash. Since the early years of the nuclear age, leaders have believed that using nuclear weapons for aggression could encourage nuclear proliferation, trigger counterbalancing coalitions, invite economic sanctions, or bring about other adverse outcomes. This is why nuclear coercion is costly for coercers, even against nonnuclear countries that cannot retaliate in kind. But if the use of nuclear weapons becomes normalized and the world becomes more tolerant of nuclear blackmail, coercive nuclear threats could in turn become more credible.
While we agree with Reiter that it is possible to envision conditions under which nuclear coercion would be effective, we disagree with his assertion that our conclusions are based on “a very limited empirical record.” Quite the contrary: our conclusions are based on analyses of virtually the entire nuclear age, including hundreds of compellent threats and territorial disputes in which nuclear powers were involved. Moreover, in our book we conduct in-depth assessments of 19 episodes in which leaders employed nuclear brinkmanship tactics. While Reiter is correct that overt nuclear threats are rare, we are unaware of any theory suggesting that nuclear coercion can succeed only if the threat to use nuclear weapons is made explicitly. Rather, theories of nuclear coercion emphasize how brinkmanship, nuclear superiority, or even the mere possession of nuclear weapons bolster the effectiveness of coercive threats. As one recent study put it, “nuclear weapons loom in the background of any contest with a nuclear state.” The existing empirical record offers more than adequate information to test these theories.
What is the conventional wisdom about nuclear coercion?
Beardsley wonders whether the nuclear coercionist school, which holds that nuclear weapons are effective tools of compellence, is really the conventional wisdom today. Indeed, he is correct to note that other scholars have questioned the coercive utility of nuclear weapons as well. Yet it is important to note that many scholars and policymakers have argued the opposite. Matthew Kroenig’s recent book, The Logic of American Nuclear Strategy, which Beardsley references, is only the most recent example. From the early days of the nuclear age, experts and government officials have espoused the coercive benefits of nuclear weapons.
Consider a few examples. Bernard Baruch, an adviser to President Truman on nuclear issues, believed that the U.S. nuclear monopoly gave Washington coercive leverage over the Soviet Union: “America can get what it wants if she insists on it. After all, we’ve got [the bomb] and they haven’t and won’t for a long time to come.” Robert Pape argued in his 1996 book Bombing to Win that “even if the coercer’s nuclear resources are limited, the prospect of damage far worse than the most intense conventional assault will likely coerce all but the most resolute defenders.” Michael Horowitz and Dan Reiter wrote in a 2001 study that “coercion is more likely to work…if the coercer enjoys a unilateral nuclear advantage.” And Beardsley himself made a similar claim in an important 2009 paper with Victor Asal: “Nuclear weapons provide more than prestige, they provide leverage. They are useful in coercive diplomacy, and this must be central to any explanation of why states acquire them.”
Which view is the ‘true’ conventional wisdom is, however, somewhat beside the point. What is clear is that there is a widespread belief, spanning across decades and continuing to the present day, that nuclear weapons are useful for compellence. This belief persists in academia as well as in policy circles, even if it is not universally held. Evaluating its claims is the central purpose of our book.
Beardsley suggests that we could have engaged more with earlier scholarship on nuclear statecraft. As he notes, several scholars have previously expressed skepticism about the coercive utility of nuclear weapons: Robert Jervis, for example, doubts that nuclear weapons can be used as “a lever to change the status quo.” Yet our work differs from this earlier scholarship in several ways.
First, our theory offers a different theoretical rationale for why nuclear coercion rarely works. While Jervis and others reach this same conclusion, one of their key arguments is simply that compellence is difficult. But knowing that coercion is hard does not necessarily undermine the position of the nuclear coercionist school, since nuclear powers might nevertheless succeed more frequently than non-nuclear states. By contrast, we explain why nuclear coercion in particular is difficult, and why coercive threats from nuclear states are not more likely to succeed.
Second, the earlier studies mentioned by Beardsley consider nuclear coercion largely in the narrow context of the superpower rivalry during the Cold War. Indeed, the risk of nuclear retaliation was one of the key reasons these scholars were pessimistic about the utility of nuclear coercion: Jervis, for example, argued that the risk of “all-out war” between the United States and Soviet Union prevented them from using nuclear threats to coerce each other. But nuclear threats could be used in a much broader context, either against smaller nuclear states or against nonnuclear adversaries. Jervis’ theory of the nuclear revolution, Charles Glaser’s analysis of strategic nuclear policy, and other theoretical perspectives were not developed to address these circumstances. Our theory, however, explains why nuclear coercion is unlikely to work even when the target cannot retaliate in kind.
Third, while other scholars have argued that nuclear coercion does not work, most have not attempted to test their arguments empirically. Moreover, the few studies that have done so have significant limitations. Our book aims to provide a comprehensive assessment of the historical record, drawing from data about more than 200 compellent threats, more than 300 military confrontations over disputed territory, and roughly 1,500 rounds of negotiations in territorial disputes to evaluate the efficacy of nuclear coercion. While our book is hardly the first to suggest that nuclear weapons are not useful for compellence, it provides the most comprehensive evidence to date in support of this view.
What does the theory imply for nuclear deterrence?
Tannenwald suggests that our book carries additional implications for deterrence. She observes that states have occasionally attacked nuclear powers: in 1973, for example, Egypt and Syria attacked Israel, launching the Yom Kippur War. India and Pakistan likewise fought the Kargil War in 1999 despite both states’ possession of nuclear arsenals. Tannenwald argues that our theory can explain these events, suggesting that nuclear threats lacked credibility in these cases partly because of the high costs of using nuclear weapons. While our book focuses on the credibility of nuclear weapons strictly in the context of compellence, we agree that our theory may not necessarily be limited to compellence alone. Future research should investigate how expectations of international backlash have shaped the credibility of nuclear deterrent threats in these and other cases.
Beardsley highlights two methodological issues with which any study of nuclear coercion must grapple. First, a country’s military capabilities may influence the demands it makes in coercive diplomacy. It could be the case, then, that nuclear powers fail at coercion because they make more ambitious demands than non-nuclear states.
We deployed a number of strategies to address this issue in the book (88-92). First, we examined whether nuclear powers tend to make larger demands than their non-nuclear counterparts in coercive diplomacy. We found that, in general, they do not. Second, we examined the success rates of nuclear and nonnuclear coercers in separate analyses of low- and high-stakes crises. This makes for a clearer apples-to-apples comparison since the stakes are similar for both types of coercers. We found that nuclear states lacked an advantage in both low- and high-stakes crises. This increases our confidence that nuclear states do not fail more often in coercive diplomacy because they systematically demand more. Third, we modeled coercive diplomacy as a two-stage process, accounting for the factors that cause a state to make compellent threats in the first place. Finally, we leveraged cases in which countries obtained nuclear weapons after a dispute had already begun but before it was resolved – what we call nuclear “shocks” (98). These cases address the objection that states capitulate quietly to nuclear coercers before the dispute can be observed. In cases of mid-dispute nuclear shocks, the dispute has already begun, so any concessions made by opponents of nuclear states would be observed rather than hidden, increasing our confidence that we are not systematically missing cases of successful nuclear coercion. Individually, none of these strategies can fully eliminate all barriers to causal inference. Collectively, however, they weaken alternative interpretations of our findings.
A second objection that Beardsley raises is that states choose to acquire nuclear weapons; they do not become nuclear by random chance. The factors that determine which states acquire nuclear weapons therefore could account for patterns in how they use them. Beardsley wonders, in particular, whether the states that seek nuclear weapons might be especially bad at coercion. If states pursue nuclear weapons because they have struggled in coercive diplomacy in the past, our results could simply reflect differences between countries that proliferate and those that do not.
This is a novel suggestion, and one that is well worth exploring. On its face, it seems plausible that some nuclear proliferators have been motivated by coercive objectives: China’s failure to retake Taiwan in the mid-1950s, for example, might have played a role in its decision to acquire nuclear weapons. Likewise, North Korea’s nuclear program may have been motivated by its failure to reunite the Korean peninsula. Yet the evidence suggests instead that it was the fear of nuclear blackmail, rather than the intent to engage in it, that motivated China, North Korea, and other countries to acquire nuclear weapons. For example, Chairman Mao Zedong’s decision to initiate China’s nuclear program was driven mainly by American threats to use nuclear weapons on several occasions in the early 1950s. Bolstering this view is the fact that China did not, as was feared, launch a campaign of attempted coercion upon acquiring nuclear weapons in 1964. And while U.S. officials have assessed that “‘coercive’ is perhaps the best way to think about how [North Korean supreme leader] Kim Jong-un is prepared to potentially use [nuclear] weapons,” Kim’s declaration in 2016 that North Korea “will not use nuclear weapons first unless aggressive hostile forces use nuclear weapons to invade on our sovereignty” is not consistent with this interpretation. And the cases of Britain, France, Israel, South Africa, and India do not seem to square with the view that states proliferate because they fail at coercion. Nevertheless, further research is needed to more rigorously assess this connection.
We agree with our critics on an important point: understanding the coercive effects of nuclear weapons is essential. Today the United States, Russia, and China are investing billions of dollars to modernize, refurbish, and expand their nuclear arsenals. North Korea continues to increase its nuclear capabilities, and states such as Iran, Saudi Arabia, or Japan may one day reevaluate their nuclear options. As the world enters a period of renewed great power competition, it is important for scholars and practitioners to understand how nuclear weapons shape the dynamics of coercive diplomacy. Yet our book certainly should not be the last word on this subject; we hope that our research helps catalyze future research on the political effects of nuclear weapons.
 Matthew Kroenig, “Nuclear Superiority and the Balance of Resolve: Explaining Nuclear Crisis Outcomes,” International Organization 67:1 (2013): 141-171, DOI: https://doi.org/10.1017/S0020818312000367.
 Kroenig, The Logic of American Nuclear Strategy: Why Strategic Superiority Matters (New York: Oxford University Press, 2018): 2.
 Robert Jervis. The Meaning of the Nuclear Revolution: Statecraft and the Prospect of Armageddon (Ithaca: Cornell University Press, 1989); Kenneth N. Waltz, “Nuclear Myths and Political Realities,” American Political Science Review 84:3 (1990): 730-745, DOI: https://doi.org/10.2307/1962764.
 Richard K. Betts, Nuclear Blackmail and Nuclear Balance (Washington, D.C.: Brookings Institution Press, 1987); Charles L. Glaser, Analyzing Strategic Nuclear Policy (Princeton: Princeton University Press, 1990); Francis J. Gavin, Nuclear Statecraft: History and Strategy in America’s Atomic Age (Ithaca: Cornell University Press, 2012).
 U.S. Department of Defense, Office of the Secretary of Defense, Nuclear Postutre Review, February 2018, n.p., 2018, https://media.defense.gov/2018/Feb/02/2001872886/-1/-1/1/2018-NUCLEAR-POSTURE-REVIEW-FINAL-REPORT.PDF, vii.
 Donald J. Trump, Twitter post, 4 January 2018, 3:32 a.m., https://twitter.com/realdonaldtrump/status/948879774277128197?lang=en. Kaitlyn Schallhorn, “Trump on North Korea, from ‘Rocket Man’ to ‘fire and fury,’” Fox News, 1 June 2018, http://www.foxnews.com/politics/2018/04/18/trump-on-north-korea-from-rocket-man-to-fire-and-fury.html.
 Benjamin Hall, “North Korea says US-South Korea war games intensifies nuclear war threat,” Fox News, 4 December 2017, http://www.foxnews.com/world/2017/12/04/north-korea-says-us-south-korea-war-games-intensifies-nuclear-war-threat.html.
 I am focusing on overt nuclear threats such as verbal statements and nuclear alerts, leaving aside the possibility that the mere existence of a coercer’s nuclear arsenal boosts its chances of success, even if no overt nuclear threats or mobilizations are made.
 Though, the authors in their conclusion argue against inferring that the Cuban Missile Crisis experience demonstrates the utility of nuclear coercion.
 See also Todd S. Sechser and Matthew Fuhrmann, “Crisis Bargaining and Nuclear Blackmail,” International Organization 67:1 (2013): 173-195, DOI: https://doi.org/10.1017/S0020818312000392; Todd S. Sechser and Matthew Fuhrmann, “Nuclear Weapons Are (Still) Poor Instruments of Blackmail: A Reply to Francis J. Gavin’s Critique,” H-Diplo/ISSF Forum 2 (2015): 55-62, https://issforum.org/forums/2-what-we-talk-about-when-we-talk-about-nuclear-weapons; Sechser and Fuhrmann, “The Madman Myth: Trump and the Bomb,” H-Diplo/ISSF Policy Series (2017), https://issforum.org/roundtables/policy/1-5W-madman; and Fuhrmann and Sechser, “Nuclear Blackmail: The Threat from North Korea and Iran,” in Caroline Dorminey and Eric Gomez, eds., America’s Nuclear Crossroads: A Forward-Looking Anthology (Washington, D.C.: Cato Institute, 2019), 69-80.
 Leaders might expect these outcomes because the use of nuclear weapons “is not perceived as legitimate,” as Tannenwald suggests, but legitimacy need not be the only reason: backlash could also be a strategic response by states hoping to deter further nuclear blackmail attempts.
 For example, see Nina Tannenwald, “How Strong is the Nuclear Taboo Today?” Washington Quarterly 41:3 (2018): 89-109, DOI: https://doi.org/10.1080/0163660X.2018.1520553.
 For example, Thomas C. Schelling, The Strategy of Conflict (Cambridge: Harvard University Press, 1960), and Robert Powell, Nuclear Deterrence Theory: The Search for Credibility (New York: Cambridge University Press, 1990).
 For example, Richard K. Betts, Nuclear Blackmail and Nuclear Balance (Washington, D.C.: Brookings Institution, 1987); Matthew Kroenig, The Logic of American Nuclear Strategy (New York: Oxford University Press, 2018).
 Erik Gartzke, Jeffrey Kaplow, and Rupal Mehta, “The Determinants of Nuclear Force Structure,” Journal of Conflict Resolution 58:3 (2014): 481-508, DOI: https://doi.org/10.1177/0022002713509054. Similarly, Beardsley and Asal argue that “the possession of nuclear weapons helps states to succeed in their confrontations with other states even when they do not ‘use’ these weapons.” See Kyle Beardsley and Victor Asal, “Winning with the Bomb,” Journal of Conflict Resolution 53:2 (2009): 278-301, DOI: https://doi.org/10.1177/0022002708330386.
 Kroenig, The Logic of American Nuclear Strategy: Why Strategic Superiority Matters (New York: Oxford University Press, 2018).
 Quoted in David E. Lilienthal, The Journals of David E. Lilienthal, Vol. 2: The Atomic Energy Years, 1945-1950 (New York: Harper and Row, 1964).
 Beardsley and Asal, “Winning with the Bomb,” 297.
 Robert Jervis, The Meaning of the Nuclear Revolution: Statecraft and the Prospect of Armageddon (Ithaca: Cornell University Press, 1989), 29.
 Jervis, The Meaning of the Nuclear Revolution, 31.
 Jervis, The Meaning of the Nuclear Revolution and Charles L. Glaser. Analyzing Strategic Nuclear Policy (Princeton: Princeton University Press, 1990).
 See the discussion in Sechser and Fuhrmann, “Crisis Bargaining and Nuclear Blackmail,” 179-181.
 T.V. Paul, Asymmetric Conflicts: War Initiation by Weaker Powers (New York: Cambridge University Press, 1994); Paul C. Avey, Tempting Fate: Why Nonnuclear States Confront Nuclear Opponents (Ithaca: Cornell University Press, 2019).
 In some of our other work, we have addressed the deterrent effects of nuclear weapons. See Fuhrmann and Sechser, “Signaling Alliance Commitments: Hand-Tying and Sunk Costs in Extended Nuclear Deterrence,” American Journal of Political Science 58:4 (2014): 919-935, DOI: https://doi.org/10.1111/ajps.12082; Sechser, “Sharing the Bomb: How Foreign Nuclear Deployments Shape Nonproliferation and Deterrence,” Nonproliferation Review 23:3-4 (2016): 443-458, DOI: https://doi.org/10.1080/10736700.2016.1259062; Fuhrmann, “On Extended Nuclear Deterrence,” Diplomacy and Statecraft 29:1 (2018): 51-73, DOI: https://doi.org/10.1080/09592296.2017.1420526.
 For example, see John Wilson and Xue Litai, China Builds the Bomb (Palo Alto: Stanford University Press, 1988), especially chapter 3.
 C-SPAN, “CIA Director Mike Pompeo on Intelligence and National Security” (23 January 2018), https://www.c-span.org/video/?440140-1/cia-director-mike-pompeo-assessment-russia-changed.
 Quoted in Eric Talmadge, “North Korea Will Not Use Its Nuclear Weapons First, Kim Jong-un Tells Congress,” Independent, 8 May2016, https://www.independent.co.uk/news/world/asia/north-korea-will-not-use-its-nuclear-weapons-first-kim-jong-un-tells-congress-a7018906.html.