Tailoring Deterrence by Studying Restraints at the Nuclear Brink

Matthew R. Costlow
Matthew R. Costlow is a Senior Analyst at the National Institute for Public Policy and former Special Assistant in the Office of Nuclear and Missile Defense Policy, Department of Defense.

The classic questions of the Cold War about nuclear escalation and how to potentially stop it receded for a time with the fall of the Soviet Union, but today the United States confronts these questions anew as nuclear threats have grown. U.S. officials navigating the very real dangers of an adversary’s threats of nuclear employment are unhelpfully bombarded with news reports about how this action or that policy might cause the adversary to escalate a conflict—often without voicing the other perspective: why a state may refrain from escalation. This Information Series examines this latter possibility—not because it is necessarily the most likely possibility in all cases, but because studying a state’s reasons for restraint may illuminate some factors U.S. decisionmakers and intelligence analysts can employ to better tailor deterrence threats. In short, understanding the potential reasons for restraint can help produce more effective deterrence threats to reinforce and strengthen the validity of those reasons in the adversary’s mind.

Each adversary will likely have a different set of values, goals, worldviews, risk propensities, and other unique factors relevant to deterrence, so this Information Series cannot present a universal “how to” guide for promoting adversary restraint. Its goal, instead, is to examine the political reasons why a state leadership may choose to limit its actions in two scenarios: refraining from employing nuclear weapons during a conventional conflict and trying to keep an ongoing nuclear war limited in some fashion. There are other factors pertinent to whether states act in a restrained fashion when considering nuclear employment beyond political reasons, such as operational factors (the resilience of command and control capabilities) and bureaucratic factors (whether the war plan in practice meets the intent of stated political objectives). These factors, however, are beyond the scope of this article. The assumption for the purpose of this discussion is that political leaders have the means to signal or demonstrate restraint—with the obvious caveat that in reality such an assumption may not be true; and, even if it was, the adversary still may not respond as desired.

The focus here is on potential choices made at the strategic level for restraint instead of escalation. To examine these potential choices, this Information Series is organized in six parts, with the first being an explanation about why this topic in particular is relevant today and for the foreseeable future. Even looking past today’s news headlines concerning potential Russian nuclear employment against Ukraine, U.S. officials since the Obama Administration have repeatedly stated their belief that the risk of an adversary’s nuclear employment is rising. Yet this relatively recent concern is not reflected in the current academic or strategic literature. Put simply, there is a gap between what U.S. officials are concerned about—i.e., promoting adversary restraint in a conflict—and what today’s available literature discusses.

The reasons why a state leader may choose restraint in one context may not be the same in another context, which is why it is important to examine each scenario individually. This is not to say that factors promoting restraint are mutually exclusive or bound to only apply in specific scenarios to the exclusion of others—but each scenario may dictate that some factors are uniquely relevant to a decision for restraint or escalation.

Potential Restraints in a Conflict between a Nuclear-armed State and a Non-nuclear State

The first scenario to examine is a nuclear-armed state involved in a conflict with a non-nuclear state. Even though the nuclear-armed state faces no threat of a response in kind from the non-nuclear state, there are still a number of a reasons why its leadership may refrain from employing nuclear weapons, including:

• Nuclear employment may provoke other states to enter a conflict.
  • A state may not want to risk threatening another state’s vital national interests by employing nuclear weapons. Such an act may raise the perceived stakes for other states outside the conflict and cause them to enter the conflict on the side of the non-nuclear state.
• Nuclear employment may damage relations with vital allies or partners.
  • Vital allies and partners of the nuclear-armed state may pressure it not to employ nuclear weapons as a condition for their continued good relations, and perhaps military aid. Nuclear employment against a non-nuclear state might prompt other states outside the conflict to impose severe economic sanctions, or worse, on the aggressor state’s allies and partners.
• Resorting to nuclear employment against a non-nuclear state might be seen as a sign of weakness.
  • State leaders may be reluctant to employ nuclear weapons against a non-nuclear state, especially if there is a chance it would not achieve the military or political objective, because others may perceive the act as a sign of weakness. If a state leader greatly values his reputation, or how history might perceive him, the prospect of having to resort to employing nuclear weapons may—in combination with other factors—promote restraint.
• Nuclear employment would be out of proportion to the political goals of the conflict.
  • States in the nuclear age have gone to war over claimed vital interests which inform the political objectives that leaders set for the conflict. Yet, even while leaders of nuclear-armed states have borne enormous costs in conventional war, they have not resorted to nuclear strikes because those means exceed what is needed to accomplish the political objectives at an acceptable cost. Thus, proportionality and relation to political goals appear to be particularly important factors in why a nuclear-armed state may choose to refrain from employing nuclear weapons against a non-nuclear state.
• Nuclear employment may incentivize nuclear proliferation among hostile states, thus damaging long term security.
  • If a state employed nuclear weapons against a non-nuclear state, others may reassess their deterrence requirements against the nuclear-armed state—perhaps even concluding that they need to develop their own nuclear weapons programs or, if they already have a nuclear force, expand it. While such effects may not be immediate, the long-term security implications for the nuclear-armed state may make even a potentially effective nuclear strike for short term gains appear too costly on balance.
• Escalation of a conventional conflict to nuclear employment may affect domestic support at home or abroad.
  • Dictatorships and democracies, to one extent or another, must account for their publics’ reactions to major decisions, likely including nuclear employment. For a dictatorship, like Vladimir Putin’s or Xi Jinping’s, a nuclear first use against a non-nuclear state may fracture domestic support for the conflict by raising the prospect that another nuclear-armed power might enter the conflict and impose severe costs on the aggressor’s homeland. For a democracy, political leaders must consider how their populace might react to the perceived wisdom and necessity of nuclear first use against a non-nuclear power—with the attendant consequences for re-election prospects.
• Nuclear employment in a scenario short of national survival may be perceived as immoral.
  • While unlikely to be a major factor for restraint in Russia or China, some state leaders may believe, or perceive their population believes, that nuclear employment against a non-nuclear state transgresses some moral boundary to a degree that is unacceptable under the circumstances.

Potential Restraints in a Conventional Conflict between Two Nuclear-armed States

Another important scenario to examine is when two nuclear-armed states engage in a conventional conflict and consider employing nuclear weapons. In this case, as in the previous scenario, a number of factors may promote restraint from nuclear employment, without excluding those already mentioned. These include:

• Nuclear employment could lead to uncontrolled escalation.
  • The uncertain course of war in general, and nuclear war in particular, may provide state leaders with a persuasive reason for restraint. Whatever gains a state may have made during the conflict may be put at greater risk with nuclear employment; or, whatever losses it may have suffered may only be compounded with nuclear employment. If political leaders have any ambitions beyond the conflict they are engaging in, they must first survive that conflict—a potential reason for refraining from nuclear employment and its unknown consequences.
• Nuclear employment could reduce the chance for a favorable political settlement.
  • A state leader may refrain from employing nuclear weapons against a nuclear-armed opponent because nuclear use might cause the opponent to expand its war aims and efforts, ultimately reducing the chance for a favorable political settlement. Nuclear first use may be one of the key catalysts for subverting the prospects for lasting peace since it could help change the political calculus of other states that then seek, as the Allies did in World War II, a policy of “unconditional surrender.”
• A state’s leadership may perceive weaknesses in its military capabilities that were unknown when the conventional conflict began—sowing doubt about a nuclear attack’s efficacy and likelihood of success.
  • An ongoing conflict may reveal serious vulnerabilities or weaknesses in a state’s military such that those deficiencies begin to cause a state’s leaders to doubt whether a nuclear attack will be effective. A poor performance in a conventional conflict might indicate that the military may be ill-equipped to carry out nuclear strikes at an acceptable cost.
• A nuclear attack might fail because of the opponent’s active defenses.
  • Given the proliferation and improvement of air and missile defenses, and the “no fail” nature of a nuclear strike, state leaders may refrain from nuclear employment if military leaders express significant doubt about the probability of a strike’s success. An unsuccessful nuclear strike would provide no benefits and might still provoke an unacceptable response, the worst of both worlds.
• If the opponent responds with a comparable strike, the relative damage will be disadvantageous.
  • If a state leader employs nuclear weapons against the opponent, and the opponent responds in kind, although the states may suffer similar levels of damage the kinds of targets that are destroyed may be more important to one state’s theory of victory and chances for success. That is, a tit-for-tat exchange may be too costly to initiate, even if the exchange stayed limited.
• The opponent’s nuclear signaling, short of employment, may demonstrate more resolve than expected.
  • Misperceptions can play a major role in decision making before and during a conflict, making signals of resolve between two nuclear-armed powers particularly important. If a state leader believes the opponent is unwilling to employ nuclear weapons for some reason, then an opponent’s demonstration of resolve via nuclear means short of employment (raising alert levels, dispersing launch platforms, etc.) may—in combination with other factors—cause the state leader to reassess the opponent’s resolve, thus promoting restraint.

Can Nuclear War be Limited?

Before examining the most relevant factors that could potentially promote restraint during a limited nuclear war (i.e., below the level of unrestrained general nuclear war), it is important to address the long-debated theoretical question of whether nuclear war can indeed stay limited. On this point, classic scholars of nuclear deterrence such as Thomas Schelling, Herman Kahn, Robert Jervis, and Colin Gray agree that there are rational reasons why state leaders would attempt to keep nuclear war limited. For instance, Schelling stated “…it’s doubtful whether even a nuclear war that began in some theatre would escalate to a large-scale intercontinental nuclear exchange.”[1] Additionally, prominent U.S. defense officials such as Secretaries of Defense James Schlesinger, Harold Brown, and Caspar Weinberger—although they disagreed on the relative likelihood of limiting nuclear war—agreed that the United States must make such an attempt. As an example, Schlesinger testified that, “… even if there is only a small probability that limited response options would deter an attack or bring a nuclear war to a rapid conclusion, without large-scale damage to cities, it is a probability which, for the sake of our citizens, we should not foreclose.”[2]

Moreover, there are clear indications in public Russian nuclear doctrine that it envisions the possibility of limited nuclear war at a level below general nuclear war; and, while China’s stated nuclear doctrine is less clear, its shifting nuclear posture indicates it may be moving toward a strategy that accounts for the possibility of limited nuclear war.[3] These factors indicate that nuclear war could potentially stay limited, assuming each side has the requisite operational command and control capabilities, and if three conditions are fulfilled: both sides believe nuclear war can stay limited, both sides prefer limited nuclear war to general nuclear war, and one or both sides can clearly demonstrate their beliefs to the other in a credible manner.

In addition to the points above that indicate nuclear war could potentially stay limited, there also appear to be logical gaps in the arguments advanced by critics who believe limited nuclear war is impossible or exceeding unlikely. First, although many critics generally believe states will refrain from nuclear war for fear of its unimaginable consequences, there is no logical reason why those same fears should cease having a restraining effect once a limited nuclear war does begin. In other words, limited nuclear war with all of its attendant horrors may in some cases produce an overwhelming fear of further consequences where there was no fear before, or, it may strengthen a pre-existing fear so that it becomes overwhelming. Second, and on a similar point, critics who claim limited nuclear war is unlikely or impossible often mischaracterize the “survival instinct” that purportedly would drive escalatory pressures. That is, the “fight or flight” response to danger that leaders may experience could in fact cause some pressure to escalate the conflict to a higher level in an attempt to escape or minimize danger, but this neglects the other potentially overwhelming pressure of “flight” (i.e., political accommodation). In short, limited nuclear war need not only produce escalatory pressures.

Potential Restraints in a Nuclear War between Nuclear-Armed States

The question is, therefore, what factors may contribute to restraint if one or both nuclear-armed states conduct limited nuclear strikes against one another? Stated differently, why might a state choose to respond to a limited nuclear strike in a restrained fashion, that is, without significant escalation to a massive response or a disarming first strike attempt? There are several potential reasons for restraint in such a situation, including:

• A state leader seeks limited political aims through limited strikes but believes anything beyond limited nuclear war would unacceptably threaten any political gains.
  • A state leader may believe there are some stakes so high that they are worth entering into limited nuclear war over, yet not so valuable that they warrant going beyond that. At some point, one side’s major escalation of nuclear strikes, beyond limited ones, may begin to look like the start of an unrestricted all-out attack to the other side. Seeing little benefit and too many costs to that course of action, a state leader may then decide that all the coercive bargaining leverage that could be gained at an acceptable cost has been gained, and continuing the conflict only places perceived gains at risk. In short, engaging in limited nuclear war and surviving may be enough to convince a leader that he has reached the end of the acceptable options available to him.
• The military balance beyond a limited nuclear exchange does not appear favorable for gaining a better bargaining position by continuing the conflict.
  • Once both nuclear powers know that the other is willing to employ nuclear weapons to accomplish their objectives, the two remaining primary questions will be which side has the most resolve to continue, and which side might the military balance favor during and after the conflict? As nuclear scholars throughout the Cold War recognized, these two factors may in some situations be related: the side with a perceived advantage in the military balance may have, and be seen as having, the advantage in resolve. This realization may in some situations promote restraint against further conflict.
• A state leader may limit nuclear strikes because the more devastating the prospective defeat the opponent perceives, the more desperate or irrational the opponent may become.
  • A state leader may limit the initial nuclear strikes to minimize the risk of inadvertently promoting the adversary’s belief that an unrestricted general nuclear strike is his only chance for survival. By withholding strikes from particular targets of great value to the adversary (i.e., holding them at risk), or restricting the size of the strikes overall, a state leader may try to minimize the risk that the adversary comes to believe the best remaining course of action is escalation. In short, a state leader may have great incentives to limit initial nuclear strikes in ways that are obvious and of great value to an adversary, to lower the risk the adversary believes he has nothing left to lose.
• A state leader may limit nuclear strikes to achieve a decisive conventional victory or avoid decisive battlefield defeat.
  • A state leader may see the tactical advantages of employing nuclear weapons in a limited fashion on the battlefield as outweighing the risks of additional costs. The purpose of limited nuclear strikes, in these scenarios, would be to gain some advantage at the tactical level or avoid some decisive defeat. A state leader may believe nuclear employment on the battlefield may be primarily for achieving some tactical military benefit with only the secondary (but still important) benefit of sending a strategic deterrent signal to others. Any larger strike beyond those that are militarily necessary might be seen as placing the larger political objective at risk.

Preliminary Application to China and Russia

Given the potential factors for restraint outlined above, there is a basis for making a preliminary application to China and Russia that regional experts could perhaps build on in the future, with more detailed country and scenario specific work. Choosing what appears to be the most likely scenario in which Chinese officials may consider nuclear employment, a conflict against Taiwan, and potentially the United States—there seems to be a number of factors that could promote Chinese restraint in such a case. If the United States, for example, had not yet formally entered an ongoing China-Taiwan conflict on Taiwan’s side, then China’s leaders may wish to keep the conflict non-nuclear for fear that it could raise the stakes of the outcome by employing nuclear weapons and threatening U.S. vital interests. China may also wish to keep any conflict against Taiwan non-nuclear to minimize the chances that U.S. allies in the region pursue their own nuclear weapon programs. If China perceived that nuclear employment was in its national interest, it may choose to limit the size of its strikes or the targets aimed at since a much larger strike may provoke a U.S. response that is costlier than China is willing to incur.

Russia, for its part, may refrain from nuclear employment against Ukraine because that may minimize the chances that the United States and other military powers become direct participants in the conflict on Ukraine’s side. Russian nuclear employment might also severely damage diplomatic relations with states like China, India, or Iran who may not want to be associated with a regime that could then become subject to unprecedented economic sanctions. Moreover, Russia’s leadership may believe that employing nuclear weapons against Ukraine may in fact make a favorable political settlement more difficult to achieve—if in fact, that is an acceptable solution. In a scenario in which Russia and the United States were in direct conventional conflict, there is a chance that the conflict could stay non-nuclear. Moscow may choose not to employ nuclear weapons because the conventional conflict revealed severe deficiencies in its military forces that may extend to its nuclear forces as well. Additionally, Russian leaders may choose not to employ nuclear weapons because if they did, that would open up Russia to symmetrical strikes that could cripple their theory of military victory.

Findings and Conclusion

There are four major findings that result from this analysis overall. First, the most valuable intelligence during a crisis or conflict that may involve nuclear employment will likely be focused on an adversary’s “red lines,” values, decision-making process, centers of power, and a host of other factors. The value of “tailoring” deterrence, while often applied as a label to U.S. nuclear strategy broadly speaking, shows through the examples in this study its value for the narrower focus of escalation control. Thus, even though tailoring deterrence threats, both pre-conflict and during a conflict, is not new to the field per se, this study demonstrates U.S. escalation control strategy is being built on a firm theoretical foundation.

Another major finding of this study begins by acknowledging U.S. officials potentially have two ways of influencing an adversary leadership. First, in the more familiar case, the United States has a variety of deterrence tools that it can threaten to employ so that the adversary would bear a cost greater than the benefits it seeks. Second, in the less familiar case, the United States can gain a better understanding of the factors that are internal to an adversary’s leadership, those most outside U.S. control, relevant to restraint. Whether it is the prospect of a major loss of prestige, domestic unrest, or the fear of losing the support of a major ally, there are many internal factors an adversary leadership may consider that could, in their minds, indicate restraint is the better course of action.

This study of restraints also suggests another finding: that U.S. leaders will likely benefit from a range of military capabilities, non-nuclear and nuclear, whose diverse characteristics can be applied in a tailored manner. That is, weapons that leaders can employ selectively, both in number and physical effects, will be more likely to support deterrence signaling in crisis and limited conflict scenarios than weapons that are less discriminate. If the goal is to identify an adversary’s potential reasons for restraint and send deterrence signals tailored to reinforce those reasons for restraint, then U.S. leaders will be better served by a more diverse set of weapon types and characteristics.

As a final finding, it is interesting to note how many factors outlined above that might promote restraint that include states not directly involved in the conflict. That is, state leaders must not only contemplate the potential effects nuclear employment may have on the adversary, but also the effects it may have on that adversary’s allies and partners, or the state leader’s allies and partners—all of whom could in some combination change the nature and direction of the conflict. This finding thus confirms the prudence of the long-standing U.S. practice of forming alliances and partnerships around the world.

The prospects for success in keeping war limited are, admittedly, uncertain at best. The cost of failure is potentially existential, yet so too is the cost of not trying to limit destruction. Leaving U.S. leaders with only two possible choices in the face of nuclear conflict, surrender or suicide, is an invitation for adversary coercion and even outright aggression. As long-standing U.S. policy has recognized, U.S. officials are obligated to find ways to deter aggression, and should deterrence fail, end the conflict with the lowest level of damage possible consistent with achieving U.S. objectives. No matter how difficult it may seem, the stakes are simply too high to ignore any opportunity to influence an adversary toward restraint at the nuclear brink.

[1] Thomas C. Schelling, as quoted in, “Interview with Thomas Schelling, 1986,” GBH Archives, March 04, 1986, available at https://openvault.wgbh.org/catalog/V_5293F77426B84C68A360BD628 3ACF4FC.

[2] James R. Schlesinger, Annual Defense Department Report, FY 1976 and FY 197T (Washington, D.C.: Department of Defense, February 5, 1975), pp. II-6-II-7, available at https://history.defense.gov/Portals/70/Documents/annual_reports/1 976-77_DoD_AR.pdf?ver=5Yhnnc5giX2RjfQtSjD-Vw%3d%3d.

[3] On Russia, see, Michael Kofman, Anya Fink, and Jeffrey Edmonds, Russian Strategy for Escalation Management: Evolution of Key Concepts (Arlington, VA: CNA, April 2020), available at https://www.cna.org/reports/2020/04/DRM-2019-U-022455- 1Rev.pdf.; and, on China, see, Christopher P. Twomey, “China’s Nuclear Doctrine and Deterrence Concept,” chapter in, James M. Smith and Paul J. Bolt, eds., China’s Strategic Arsenal: Worldview, Doctrine, and Systems (Washington, D.C.: Georgetown University Press, 2021).; and, U.S. Department of Defense, Military and Security Developments Involving the People’s Republic of China (Washington, D.C.: Department of Defense, November 29, 2022), pp. 98-99, available at https://media.defense.gov/2022/Nov/29/2003122279/-1/-1/1/2022- MILITARY-AND-SECURITY-DEVELOPMENTS-INVOLVING-THEPEOPLES-REPUBLIC-OF-CHINA.PDF.

This Information Series is adapted from, Matthew R. Costlow, Restraints at the Nuclear Brink: Factors in Keeping War Limited, Occasional Paper, Vol. 3, No. 7 (July 2023).

The National Institute for Public Policy’s Information Series is a periodic publication focusing on contemporary strategic issues affecting U.S. foreign and defense policy. It is a forum for promoting critical thinking on the evolving international security environment and how the dynamic geostrategic landscape affects U.S. national security. Contributors are recognized experts in the field of national security. National Institute for Public Policy would like to thank the Sarah Scaife and Smith Richardson Foundations for the generous support that made this Information Series possible.

The views in this Information Series are those of the author(s) and should not be construed as official U.S. Government policy, the official policy of the National Institute for Public Policy or any of its sponsors. For additional information about this publication or other publications by the National Institute Press, contact: Editor, National Institute Press, 9302 Lee Highway, Suite 750 |Fairfax, VA 22031 | (703) 293- 9181 |www.nipp.org. For access to previous issues of the National Institute Press Information Series, please visit http://www.nipp.org/national-institutepress/informationseries/.

© National Institute Press, 2024