Introduction

Jane Vaynman

The risk of nuclear use has increased, not only because of growing tensions with North Korea and Russia, but more broadly due to shifts in several underlying drivers of nuclear conflict—or, conversely, nuclear restraint. The authors of the two essays in this occasional paper, the second set prepared for the American Academy of Arts and Sciences project on “Meeting the Challenges of the New Nuclear Age,” identify two such drivers: deteriorating norms and increasingly complex security relationships. Nina Tannenwald’s paper, “The Great Unraveling: The Future of the Nuclear Normative Order,” argues that norms of nuclear non-use have deteriorated, both with respect to a global nuclear order and within relationships among nuclear states. James Acton’s paper, “Technology, Doctrine, and the Risk of Nuclear War,” focuses on the evolution of several strategic relationships, including the rise of multipolarity, crisis escalation, and the blurring of lines between nuclear and nonnuclear capabilities.

The authors come to similar conclusions for different reasons, and taken together their views suggest a rather bleak picture, with limited opportunities for reducing risks. They observe some of the same phenomena in the contemporary security environment and agree about the implications. Changes in nuclear doctrines, for example, increase the possibility of nuclear use as nuclear weapons become more “usable,” both normatively and practically. Indeed, the authors would likely agree with one another’s views, having chosen merely to focus on different dimensions of a multifaceted problem. However, there are also notable contrasts in these two approaches to understanding nuclear risks. While Tannenwald considers positions taken by nonnuclear states, Acton’s sense of the “nuclear order” focuses on relationships between the nuclear powers. Perceptions and understandings play a central role in both analyses, but for Tannenwald expectations may be informed by normative beliefs and ethical considerations, while Acton focuses on states’ inabilities to observe fully an adversary’s intentions and capabilities. These views are not contradictory and taken together reveal a broader picture for considering nuclear use dynamics.

Tannenwald discusses a series of related nuclear norms: deterrence, non-use, nonproliferation, non-explosive-testing, and disarmament. These norms are contested and face challenges from domestic politics, evolving nuclear doctrines, and international confrontations. Of these, the deterrence and disarmament discussions raise the most interesting questions. The deterrence regime argument holds that nuclear weapons are only for the purposes of deterrence, not for use in conflict. Of course, deterrence only works with a credible threat of use, but here the threat is made under the condition of extreme existential risk, with use only in retaliation to a strike against the state. Under this kind of deterrence-only norm, the United States and the Soviet Union took steps to lower incentives for first strike and diminish the chances of use due to misperception. In some ways, the deterrence norm is a mild version of a norm of non-use. Tannenwald identifies a decline in traditional understandings of deterrence, which implies a higher likelihood of use.

The ethical opposition to nuclear weapons, articulated today by proponents of disarmament and the participants of the humanitarian consequences movement, holds that nuclear use is illegitimate on moral grounds, which leads to a fundamental objection to any role for nuclear weapons, including deterrence. Though the paper also separates non-use (or a nuclear taboo norm) from the “disarmament norm,” they are deeply related, as Tannenwald notes, “the spread, strengthening, and internalization of the taboo have long been seen as a step on the route to disarmament.” Tannenwald observes a curious duality in the current status of these interlaced norms. In many respects, there is a decline in commitments to non-use evident in increased references to nuclear use by political leaders and the lowering of nuclear use thresholds in military doctrines. But the Treaty on the Prohibition of Nuclear Weapons signed in July 2017 also suggests that some nonnuclear states support a normative push for disarmament. An implication of Tannenwald’s assessment is that in the near future we might see an even sharper global contestation between ideas of nuclear weapons as effective and even prestigious military tools and nuclear weapons as morally illegitimate.

Tannenwald’s discussion of deterrence norms alongside moral opposition to nuclear weapons and disarmament suggests a tension between various options for next steps. On one hand, Tannenwald argues in favor of a global no-first-use commitment and a recommitment by nuclear states to deterrence. On the other hand, Tannenwald also takes seriously the calls for disarmament and seeks “frank conversations about the morality of deterrence.” Can states commit to a deterrence norm and also question the ethics of that norm? Does a commitment to no-first-use contradict efforts to delegitimize nuclear weapons, or is it a step toward that ultimate direction?

Tannenwald’s paper suggests, though does not explicitly argue, that a commitment to disarmament is in some ways in opposition to a milder non-use norm consistent with deterrence. There may be an important trade-off between these approaches: eventual disarmament could delegitimize the risk-reducing elements of nuclear deterrence today, but a renewed commitment to deterrence could make arguments for eventual disarmament increasingly difficult to make.

James Acton attributes the increased possibility of nuclear use to material rather than normative drivers. Acton identifies multipolarity and crisis escalation as the two key situations in which changes in technology and doctrine are today creating increased risk. His paper does not draw an explicit link between them, raising some questions as to whether global conditions have effects on crisis stability. The specific characteristics of contemporary military technology, which blur the line between nuclear and conventional capabilities, and the way these technologies have been incorporated into military doctrines come into play specifically when considering the potential for crises to escalate to nuclear use.

At the end of his essay, Acton notes, “the drivers behind the growing likelihood of nuclear use are a mix of the old and the new.” The statement is a good reflection of many of the dynamics he identifies in his paper. First, the triangular nature of today’s multipolar relationships increases the dangers of rivalries that are far from new. India and Pakistan continue to respond to one another’s buildup in nuclear arsenals and missiles, but now China has become increasingly attentive to India’s efforts. U.S. nuclear strategy remains bilaterally oriented toward Russia on one hand and China on the other, but international and domestic pressures for further rounds of nuclear reductions increasingly seek to include both Russia and China.

The problem of crisis escalation is likewise a mix of past and present, newly exacerbated by dual conventional and nuclear technologies. Acton’s assessment of the doctrinal and technological shifts suggests that states’ efforts effectively to threaten and coerce with conventional means—perhaps even decreasing the reliance on nuclear weapons—actually increase the potential for escalation to nuclear use. The military doctrines of several nuclear states have intentionally focused on creating escalation through conventional weapons. However, conventional escalation could be misperceived as entering the nuclear domain. Even more importantly, some of these conventional strategies intentionally threaten assets that are also relevant for nuclear capabilities. Acton cites the examples of command-and-control centers, early-warning satellites, and nuclear forces colocated with conventional ones. The increasing development of dual-use delivery systems also exacerbates the escalation problem, as states could mistake conventional capabilities for nuclear ones and interpret military moves by an adversary as far more aggressive nuclear signals.

Despite the pessimism of both Tannenwald’s and Acton’s essays, they suggest several areas that warrant further thinking, including arms control focused on restraining escalatory behaviors, the intersection of nonnuclear technology and norms, and domestic politics as a limiting factor for debate on nuclear weapon policy.

First, the authors are highly skeptical about prospects for future arms control, but other elements of their essays suggest we should perhaps revisit arms control ideas even at a time when they are not politically popular. Tannenwald’s proposal for a no-first-use regime, especially one adopted through international agreement, is a form of arms control that focuses on establishing rules about allowable behaviors. Compliance with a no-first-use commitment might involve openness about certain types of deployments, or demonstrations of doctrinal integration of the no-first-use concept in military planning. Tannenwald’s analysis of the decline of arms control actually suggests that perhaps efforts should focus on designing arms control that promotes common understandings and expectations, while avoiding the kinds of restraints on capabilities or on freedom to develop technologies that often raise domestic political opposition. While efforts to restrain behaviors will likewise face an uphill battle, and the United States has indeed rejected commitments to impose international laws on domestic policies, the potentially different constellations of domestic support and opposition could create opportunities for such agreements that are at the very least worth exploring.

In addition to measures Acton proposes on internal organizational reform of governments and militaries, his paper inspires ideas for cooperative efforts that likewise focus on specifying behaviors rather than limiting capabilities. Acton’s analysis implies the challenge is not always one of organizational failure to recognize a perceptions problem, but rather one of states having strategic interests that create opportunities for misperception (and consequently nuclear escalation) as they address other legitimate security concerns. Acton notes specific cases of possible misidentification of nuclear versus conventional capabilities, and in doing so he also points out specific opportunities to mitigate misperception. If, for example, there is a potential inadvertent escalation problem at the deployment stage of Chinese DF-26 missiles, then this is an area in which the United States and China could pursue confidence-building measures and information exchange. While even such limited arms control may founder for a host of reasons (including the technical and political difficulties Acton notes), the identification of how a mutually disadvantaged misperception could occur should motivate at least an attempt to design cooperative responses.

Second, it is clear scholars and policy-makers will have to contend with the challenge of how norms develop, or perhaps fail to develop, around emerging nonnuclear military technologies that increase the risk of escalation into nuclear conflict. Two factors may undermine the development of restraint norms in this space. As Acton discusses, states are developing nonnuclear technologies in response to perceived security threats and with use in mind, and the technology is integrated into doctrine and deployment. Increasing the risk of escalation may be an unintentional—or even intentional—externality of these capabilities, but neither negates its purpose within military strategy, making “non-use” norms unlikely to take hold. Further, there are many diverse ideas on how escalation could occur under new scenarios that involve long-range, high-precision conventional missiles, autonomous weapons, or cyber capabilities. The lack of consensus is unsurprising, but also inhibits both norm development and norm promotion by actors who might seek to mitigate escalation risks. Efforts at developing common understanding about even a small subset of nonnuclear technology could aid in the growth of norms around high-risk behaviors or communication during crises involving these capabilities.

Finally, thinking about mitigating risks of nuclear use need not be mired in the present. The current domestic political opposition to arms control, or as Tannenwald puts it, a “disdain,” is not necessarily the sign of a long-term trend. Throughout the Cold War and even before, some arms control agreements took years to negotiate, and long periods elapsed without agreements at all, only to have opportunities emerge later. Even without current negotiations, the New Strategic Arms Reduction Treaty (New START) and the Iran nuclear deal are relatively recent in the time frame of arms control developments, and both have substantial domestic support even in the face of political opposition. Looking ahead, President Donald Trump would not be the first world leader to oppose international cooperation efforts, who is replaced later by someone who favors them. Though it would be an unlikely bet, Trump would also not be the first leader to start with an anti-agreements policy but eventually come to see cooperation as beneficial. At the same time, the papers in this collection suggest warnings that domestic support for both nuclear restraint and international cooperation could decline even further, perhaps if nuclear weapons become entrenched in nationalist movements. The challenges to both normative and material nuclear orders identified in these essays are unlikely to decline in the short term. Thinking about ways to mitigate the risks of nuclear use should likewise seek to move beyond today’s political environment.