Buy Print

Buy Print

Roundtable Policy

Policy Roundtable: Nuclear First-Use and Presidential Authority

In this roundtable, we gathered together a group of experts to discuss two important aspects of the nuclear debate: America's nuclear first-use policy and the president's sole authority to launch a nuclear weapon.

1. Introduction: Debating Fundamental Questions in American Nuclear Strategy

By Galen Jackson


Nuclear weapons, and the threat they pose to the world, have reemerged in recent years as a major focus among international security analysts. The election of President Donald Trump, and the attendant concerns many observers have about his judgment as commander-in-chief, only accelerated a preexisting trend toward renewed attention to nuclear security issues. Indeed, scholars and policymakers are now grappling with an array of challenges in this area: a quickening technological arms race, increasing competition among the great powers, and significant proliferation risks, to name just a few. Thus, it was rather unsurprising that during a recent debate among Democratic presidential hopefuls, three candidates rated topics relating to nuclear weapons as the single greatest geopolitical threat currently facing the United States.1

Given this threat environment and the heightened attention being paid to nuclear weapons in American political discourse, it is important for experts to consider carefully the fundamental aspects of U.S. nuclear strategy, which is the goal of this roundtable. The two questions addressed here — whether the United States should adopt a policy of no-first-use (NFU) of nuclear weapons, and whether the president should continue to have sole authority to order the use of the American nuclear arsenal — have been particularly prominent in recent debates in this area. The five contributors — Nina Tannenwald, Jon Wolfsthal, John Harvey, Rachel Whitlark, and Brendan Green — all offer important insights and a range of perspectives that shed light on topics that are enormously important to international security.

The NFU Debate

Tannenwald and Wolfsthal each believe strongly that the United States should adopt a policy of no-first-use. Washington’s reliance on the threat to use nuclear weapons first, they write respectively, is an “outdated legacy of the Cold War” and reflects an “anachronistic understanding of how nuclear strategy may have worked in the past.” The United States, Tannenwald writes, is now pursuing “highly dangerous deterrence policies.” The chance of nuclear weapons being used today, Wolfsthal agrees, has become “alarmingly high.” U.S. adoption of NFU, they believe, would mitigate these dangers in important ways.

What is the logic underpinning this argument? For Tannenwald and Wolfsthal, the main risks of a first-use policy have to do with increasing the likelihood of accidents and miscalculation and, relatedly, undermining crisis stability.2 In a nuclear standoff where a premium is placed on striking first, for example, there is a danger that those involved would feel they have no choice but to preempt.3 This problem may be exacerbated, Tannenwald points out, in an age where counterforce technology is developing at a dizzying pace, strategies for using conventional and nuclear weapons are not easily distinguishable, and the adoption of certain postures may have lowered the threshold for nuclear use.4 And, she adds, the evidence since the end of the Cold War does not support the view that threatening first use has actually enhanced deterrence.

Tannenwald and Wolfsthal, moreover, do not see major risks involved in adopting NFU as official policy because of the United States’ conventional preponderance. Whereas threatening to uphold extended deterrence commitments with nuclear weapons, they argue, is not at all credible, doing so with conventional forces is.5 It is for this reason, Wolfsthal believes, that the United States could safely adopt NFU without compromising its nonproliferation objectives. To be sure, he writes, any change to American policy in this area should be done in consultation with Washington’s allies, but U.S. conventional superiority allows the United States to abandon its existing posture without putting its friends in jeopardy. Indeed, the two authors emphasize, American strategists have every incentive to try to shift competition to the conventional plane.6

With this in mind, Tannenwald and Wolfsthal favor making operational changes to the U.S. nuclear posture, such as de-mating and de-alerting forces, that would make the adoption of an NFU policy credible to states like China and Russia. Otherwise, Tannenwald points out, NFU could simply be dismissed as “cheap talk.” Such changes, according to Wolfsthal, would also lessen the chances of unintended or accidental use. Finally, Tannenwald writes, adopting this sort of policy would strengthen the so-called “nuclear taboo,” a concept she has probably explored more thoroughly than anyone else and that she believes is currently in danger of being eroded.7

Harvey, however, views things very differently. Supporters of NFU, he believes, are no doubt “[w]ell-meaning” and concerned with exhibiting “moral leadership,” but they underestimate the risks that such a policy would run. It is for this reason, he claims, that U.S. presidents have repeatedly decided not to alter American policy in this area. Specifically, Harvey writes, there are three problems with an NFU policy. First, it could weaken deterrence. “Would North Korea,” he asks, “be more willing to contemplate [biological and chemical weapons attacks on the United States and South Korea] if it thought it was immune to a U.S. nuclear response?” Overcoming the second problem, which has to do with the difficulty of maintaining extended deterrence commitments to countries like Japan and South Korea without being able to threaten a nuclear response, might be even more challenging.8 And relatedly, Harvey is far less sanguine than Tannenwald and Wolfsthal about the potential proliferation risks that NFU raises. Some U.S. allies, he observes, are “latent” nuclear powers that could try to develop independent arsenals relatively quickly if they perceived that Washington was weakening its security guarantees to them.9

The supposed benefits of NFU are, moreover, overstated in Harvey’s view. The claim sometimes made by supporters of the policy that it would lead to a less prominent role for nuclear weapons in the international system by convincing other nuclear states to follow Washington’s lead, he writes, is largely unfounded, especially because of American conventional power. “Indeed,” he observes, “several nuclear adversaries have acquired, or are currently seeking, nuclear weapons precisely to offset superior U.S. conventional capabilities.” Nor, he writes, is there any reason to think that NFU would prevent adversaries from launching nuclear weapons against the United States due to miscalculation. In short, “Those who support no-first-use as a way to advance U.S. security must explain what has changed for the better in the international security environment since 2010 that would cause this president, or this Congress, to reverse earlier presidential decisions rejecting it.”

The Question of Sole Presidential Authority

Another major question that has received increased attention since Trump’s election is whether the president should continue to have sole authority to order the use of nuclear weapons. For Green and Whitlark, the advantages of giving the president this power clearly outweigh the disadvantages. In addition, they both believe that the system already includes a number of checks to help ensure that nuclear weapons are not employed unjustifiably.

Both Green and Whitlark point out that it was primarily Trump’s election that has led to an array of proposals for multiple authorization. A number of ideas have been proposed, including involving the Supreme Court, certain members of Congress, or other members of the executive branch like the attorney general and secretary of defense. None of these, they believe, would be a good idea.

To put it succinctly, both authors believe that instituting a system of multiple authorization would tilt American nuclear posture too far toward “never” in the “always/never” divide. “The central paradox of the Cold War,” Whitlark writes, “was that, in order to prevent nuclear use, America had to be prepared to use nuclear weapons.” With that in mind, involving more veto players in the process, both she and Green argue, would dangerously jeopardize deterrence. Adopting multiple authorization, they say, would slow down the decision-making process and could compromise the secrecy of nuclear operations. This, Green writes, could be especially costly given existing technological trends that have made counterforce strikes more feasible, as well as the uncovering of new evidence that suggests that inefficient command systems had important consequences during the Cold War.10 Moreover, Whitlark asserts, introducing new layers of authorization could undermine American extended deterrence commitments and, in turn, exacerbate proliferation risks. If this were not enough, in Green’s view, new checks are not likely to be effective in any case. President Dwight Eisenhower, he points out, found ways to circumvent a number of constraints that Congress had imposed in this area.11

Several procedural checks, both authors also point out, already exist. Military commanders and presidential advisers have a role to play in the system as it is currently structured. The 25th amendment, moreover, serves a useful purpose in this regard. Some reforms might be in order —for example, Whitlark mentions making sure presidents are better educated about nuclear issues and involving Congress more fully in oversight and spending decisions relating to the nuclear arsenal — but adopting a multiple authorization approach simply goes too far.12

In addition, both authors, especially Green, are disturbed by the domestic and constitutional questions that moving to a multiple authorization system would raise. Policy should not be formulated because of the personal tendencies of a single president, and the potential consequences at home should not be underestimated, they agree. A move of this type, Green writes, would be viewed as “highly partisan” and, consequently, would probably “worsen political polarization.” The president, in fact, is “optimally situated to provide domestic legitimacy to nuclear decision-making” given that the office is accountable to a national constituency. In short, the title of Whitlark’s piece nicely captures these two authors’ positions: “Should Presidential Command Over Nuclear Launch Have Limitations? In a Word, No.”


Nuclear weapons pose an enormous danger to international security. Many signs suggest that they could play an increasingly important role in world politics, particularly as great power competition returns to the fore. The questions of NFU and presidential command authority are, in this sense, among the most important topics for scholars and policymakers alike to debate. It is therefore a very good thing to have experts of the caliber of Nina Tannenwald, Jon Wolfsthal, John Harvey, Rachel Whitlark, and Brendan Green debating these issues with each other.


Galen Jackson is an assistant professor of political science at Williams College where he teaches courses in international security, nuclear security, American foreign policy, the Middle East in international politics, and international relations theory. His work has been published in Security Studies, International Security, and the Journal of Cold War Studies. He is currently working on a project that examines the superpower diplomacy of the Arab-Israeli conflict between the June 1967 war and the 1979 Egyptian-Israeli peace treaty. He received his Ph.D. in political science from UCLA in 2016 and became an associate editor at TNSR in 2018.


2. It’s Time for a U.S. No-First-Use Nuclear Policy

By Nina Tannenwald


Beginning in the early days of the Cold War, the United States has relied on the threat to use nuclear weapons first as a way to deter both nuclear and non-nuclear attacks. Yet, the world has changed significantly since then. In the contemporary era, the dangers and risks of a first-strike policy outweigh the hoped-for deterrence benefits. The United States should join China and India in adopting a declared no-first-use policy and should encourage the other nuclear-armed states to do likewise. A no-first-use policy means that the United States would pledge to use nuclear weapons only in retaliation for a nuclear attack. The sole purpose of U.S. nuclear weapons would then be to deter — and, if necessary, respond to — the use of nuclear weapons against the United States and its allies and partners. To be credible, this declaratory pledge would need to be reflected in a retaliatory-strike-only nuclear force posture.

The most important goal for the United States today should be to prevent the use of nuclear weapons. Since the United States dropped atomic bombs on Hiroshima and Nagasaki in 1945 — the only use of nuclear weapons in warfare — it has established a nearly 74-year tradition of not using nuclear weapons. This tradition is the single most important fact of the nuclear age. Today, the risks of nuclear war are increasing. Heightened geopolitical tensions, a more complex calculus of deterrence in a multipolar nuclear world, renewed reliance on nuclear weapons, technological arms races in nuclear and non-nuclear systems, the collapse of arms control, and the return of nuclear brinkmanship have all resulted in highly dangerous deterrence policies that, through miscalculation or accident, could plunge the United States into a nuclear war with North Korea, Russia, or China. The nuclear-armed states urgently need to step back from this dangerous situation by adopting a no-first-use policy that would significantly reduce the risk of nuclear war.

International Relations Theory and No-First-Use

Several theoretical approaches in international relations help to illuminate why states choose to adopt a first-use versus a no-first-use (NFU) policy. A realist approach, which emphasizes the central role of material capabilities, would generally be skeptical of no-first-use pledges, which it would view as “cheap talk” and unenforceable. States that have made such pledges could still launch a nuclear weapon first in a conflict. Thus, NATO leaders and other observers expressed considerable skepticism during the final years of the Cold War that Russia’s declaration of an NFU policy in 1982 had any real substance behind it.13 Today, while India has made an NFU pledge, analysts debate how constraining it really is. In turn, Indian Prime Minister Narendra Modi is sometimes dismissive of China’s NFU policy.14

But some states — India, China, and the Soviet Union for a period — have nevertheless pledged no-first-use and, in the cases of India and China, have attempted to make those pledges credible. What explains these choices? The empirical record suggests that a state’s choice regarding a nuclear first-use policy tends to be strongly influenced by asymmetries in the conventional military balance between nuclear-armed adversaries. Nuclear-armed states that face a conventionally superior military adversary will threaten to use nuclear weapons first because they depend more heavily on nuclear threats to defend themselves. In contrast, nuclear-armed states that possess overwhelming conventional superiority are more likely to declare an NFU policy because it privileges their conventional advantage on the battlefield and might help to keep the conflict non-nuclear.

Thus India, which possesses a much larger conventional military than Pakistan, declared an NFU policy in 1999, following its nuclear test in 1998. Pakistan, which relies heavily on its nuclear deterrent for its defense against India, has rejected Indian calls to adopt a no-first-use pledge.15 This logic also helps explain why, in 1993, Russia dropped its NFU pledge first made in 1982. After the collapse of the Soviet Union in 1989, as Russian conventional military forces deteriorated and the United States declined to reciprocate the NFU pledge, Russian leaders felt they had to rely more heavily on nuclear weapons.

Consistent with this logic, during the Cold War, the United States relied on a first-use threat to offset and counter the overwhelming conventional superiority of the Soviet conventional military threat in Europe. Today, the situation is reversed. The United States possesses overwhelming conventional superiority while Russia’s conventional military has declined. Because U.S. conventional military power now vastly exceeds that of its largest adversaries, Russia and China, many argue that America’s first-use policy is now unnecessary to deter conventional threats.16

China’s NFU policy, on the other hand, while consistent with its small nuclear force, is less well explained by asymmetries in conventional forces. China adopted an NFU policy at the time of its first atomic bomb test in 1964, when its peasant army was still transitioning to a modern military force. Part of the explanation for this decision has to do with Mao’s thinking about the nuclear bomb as a “paper tiger,” but Chinese leaders have primarily seen an NFU policy as an effective way to signal the purely defensive nature of the small Chinese nuclear arsenal and to avoid a U.S.-Soviet-style arms race.17 An NFU policy also conveys the spirit of “peaceful coexistence” to which China is committed.

The theory that adopting an NFU policy is based on asymmetries in conventional forces is further complicated by the existence of other weapons of mass destruction. During the George W. Bush and Barack Obama years, the strongest argument for the United States to retain the first-use option was that nuclear weapons are necessary to help deter and possibly retaliate against attacks with chemical and especially biological weapons.18 The Trump administration’s 2018 Nuclear Posture Review has expanded the category of non-nuclear attacks that it will seek to deter with nuclear threats to include cyber attacks, a move that previous presidents had ruled out and that most observers view skeptically, given its dangerous escalatory potential.

A second theoretical perspective, “liberal institutionalism,” emphasizes the role of rules and institutions, both domestic and international, in stabilizing expectations and behavior. According to this theory, even if no-first-use pledges are unenforceable, they are not necessarily meaningless. To be meaningful, an NFU pledge must be built into domestic institutions, that is, the structure of operational military capabilities.19 A genuine NFU policy would require that nuclear forces be consistent with an “assured retaliation” posture that eschews counterforce objectives — the ability to destroy an adversary’s nuclear arsenal before it is launched.

This perspective thus emphasizes the value of an NFU pledge in structuring operational forces to make them smaller and less threatening. When Secretary of Defense Robert McNamara, soon after entering office in 1961, sent a directive to the Joint Chiefs of Staff about strategic force requirements, he stated that the first assumption shaping requirements was that “we will not strike first with such weapons.”20 McNamara’s directive was undoubtedly partly an effort to stem Air Force demands for a first-strike capability and the vast procurement of weaponry it would require. This directive, in effect, repudiated the extended deterrent doctrine that the United States would respond to a Soviet conventional attack in Europe with nuclear weapons.

At the international level, liberal institutionalists emphasize the value of rules and institutions to prevent nuclear war. They argue that NFU has become a de facto norm anyway and therefore should be declared publicly and multilaterally. As Morton Halperin, who later became deputy assistant secretary of defense for arms control, wrote as early as 1961, “There now exists a powerful informal rule against the use of nuclear weapons,” and it would be advantageous to the United States to transform this tacit understanding into a formal agreement.21 Indeed, the “negative security assurances” first issued by the United States and the other P5 countries in 1978 and renewed periodically — commitments to non-nuclear states that are members of the Nuclear Nonproliferation Treaty not to use or threaten to use nuclear weapons against them — already constitute a partial NFU regime. Liberal institutionalists would also point out that constantly touting the value of a nuclear threat for security sends signals that nuclear weapons are useful and undermines nonproliferation goals.22

Finally, constructivists, who focus on the role of norms, identity, and discourse, emphasize that a declared NFU policy is an important way to strengthen norms of nuclear restraint and the nearly 74-year tradition of non-use. Strong statements from leaders about the need to avoid using nuclear weapons can help reduce tensions, just as irresponsible tweets can increase them. In the constructivist view, an NFU policy is also a diplomatic tool that can be used to signal that a state is a responsible nuclear power. As Modi recently put it, “India is a very responsible state. We are the only country to have a declared NFU [sic]. It’s not because of world pressure, but because of our own ethos. We will not move away from this, whichever government comes to power.”23 Indeed, India’s NFU pledge has proved useful for portraying Pakistan as a relatively irresponsible custodian of its nuclear arsenal. Likewise, Indian leaders use their NFU pledge as a way to resist pressures to sign any treaties that would restrict India’s nuclear arsenal.

The Weak Case for a First-Use Policy

A first-use policy is based primarily on the belief that the threat of nuclear escalation continues to serve as a deterrent to large-scale conventional war or the use of chemical and biological weapons.24 Critics of NFU argue that the United States should not make any promise that might make it easier for an opponent to plan an effective military action, a strategy known as “calculated ambiguity.” As the Defense Department recently explained,

Retaining a degree of ambiguity and refraining from a no first use policy creates uncertainty in the mind of potential adversaries and reinforces deterrence of aggression by ensuring adversaries cannot predict what specific actions will lead to a U.S. nuclear response. Implementing a no first use policy could undermine the U.S. ability to deter Russian, Chinese, and North Korean aggression, especially with respect to their growing capability to carry out nonnuclear strategic attacks.25

In addition, skeptics believe that an NFU promise would be especially costly for the United States, given its wide-ranging extended deterrence commitments.26

These arguments are not compelling for four reasons. First, a policy of calculated ambiguity is unnecessary. Today, there are very few missions that the United States could not accomplish with conventional weapons. Indeed, U.S. conventional capabilities are more than sufficient to deter and respond to anything but a nuclear attack. None of the United States’ most likely adversaries — Russia, China, North Korea, and Iran — can hope to defeat the United States and its allies in a protracted non-nuclear conflict.

Second, threats of first use are dangerous. As Michael Gerson has argued, they undermine crisis stability in multiple ways.27 The large, highly accurate U.S. nuclear arsenal, along with missile defenses and new dual-use precision-strike weapons, may lead leaders in Russia and China to believe that the United States is capable of conducting a disarming first strike against them. Furthermore, the entanglement of nuclear and conventional weapons in deterrence strategies could inadvertently increase the chance of nuclear war, while new, smaller nuclear warheads, along with doctrines of “escalate to de-escalate” appear to be lowering the threshold for nuclear use.28 In a crisis, Russian or Chinese leaders might come to believe that the United States might attempt a disarming strike, forcing them, in turn, to contemplate acting preemptively.29

Third, although supporters of calculated ambiguity fervently believe it maximizes deterrence, the evidence for such a claim is hardly definitive. Nuclear weapons did not deter the 9/11 attacks; the rise of the Islamic State; Russian interventions in Georgia, Ukraine, or Syria; or North Korean nuclear and ballistic missile tests. Nor have Indian and Pakistani nuclear weapons deterred risky conventional crises between the two countries over Kashmir, most recently in February 2019. The calculated ambiguity argument gained some support from the perception that during the 1991 Gulf War a U.S. nuclear threat had helped deter Iraqi leader Saddam Hussein from using chemical weapons against U.S. and coalition forces or Israel.30 As Scott Sagan has persuasively argued, however, it is highly unlikely that a nuclear threat in fact deterred Saddam from using chemical weapons.31 Indeed, recent research suggests that the threat to use nuclear weapons first against non-nuclear states has little credible coercive power.32

Fourth, even in the very small number of scenarios where nuclear weapons might seem to be necessary — for example, knocking out North Korean mobile missiles or underground command centers — opening the Pandora’s box of nuclear use would likely lead to uncontrolled escalation. There is no scenario in which using nuclear weapons first can make a bad situation better. As James Doyle, a former staffer at Los Alamos National Laboratory, has argued, “It is folly to believe that the use of nuclear weapons could de-escalate a conflict.”33

As for threatening to use nuclear weapons first in support of extended deterrence commitments, such a policy lacks credibility because the costs of starting a nuclear war would vastly outweigh the benefits. As Henry Kissinger once said, “Great powers don’t commit suicide for their allies.”34 Thus, as a number of analysts have persuasively argued, extended deterrence based on a conventional military response to a conventional threat is much more credible. Moreover, constantly arguing that nuclear weapons are necessary reduces the credibility of the United States’ more usable conventional deterrent.35

The Benefits of a No-First-Use Policy 

As Kingston Reif and Daryl Kimball of the Arms Control Association have argued, “a clear U.S. no-first-use policy would reduce the risk of Russian or Chinese nuclear miscalculation during a crisis by alleviating concerns about a devastating U.S. nuclear first-strike.”36 This would mean that the United States would rely on nuclear weapons only to deter nuclear attacks. Adopting this approach would involve more than “cheap talk,” for it would require meaningful doctrinal and operational changes.37 Specifically, it would allow the United States to adopt a less threatening nuclear posture. It would eliminate first-strike postures, preemptive capabilities, and other types of destabilizing warfighting strategies. It would emphasize restraint in targeting, launch-on-warning, alert levels of deployed systems, procurement, and modernization plans. In other words, it would help shape the physical qualities of nuclear forces in a way that renders them unsuitable for missions other than deterrence of nuclear attacks.38

Implementing these steps would significantly reduce the risk of accidental, unauthorized, mistaken, or preemptive use. The removal of threats of a nuclear first strike would also strengthen strategic and crisis stability.39 Of perhaps equal importance, adopting an NFU policy would help address humanitarian concerns and reduce the salience of nuclear weapons.40 Likewise, it would “be more consistent with the long-term goal of global nuclear disarmament and would better contribute to US nuclear non-proliferation objectives.”41

A multilateral NFU pledge would have even more benefits. It would move Russia and Pakistan away from their high-risk doctrines and reduce a source of Russia-NATO tensions. A common NFU policy would help anchor the existing NFU policies of China and India and implicitly acknowledge their leadership in this area, a virtue when middle-power states are feeling disenfranchised from the global nuclear order.

Some analysts have questioned whether, in an asymmetric conflict, an American NFU policy would actually help reduce the risk of nuclear escalation by an adversary. The United States is so conventionally dominant, they argue that, in a crisis, a country like North Korea might employ nuclear weapons preemptively because the United States could take out North Korean targets even with just conventional weapons.42 It is true that an NFU policy might make no difference in such a situation. Still, it might nevertheless remove at least one source of crisis instability. Most importantly, however, in an era of “multi-front” deterrence, North Korea is not the only adversary and a U.S. NFU policy would remain valuable in less asymmetric conflicts.

A second concern is that a real NFU strategy would require a greater commitment to a counter-value targeting strategy — targeting civilians rather than nuclear silos — and thus run up against moral and legal rules prohibiting the direct targeting of civilians.43 This is a legitimate point. However, current U.S. counterforce targeting policy will likely result in massive civilian casualties as “collateral damage,” making the risk to civilians of an NFU strategy little different.44


The United States ought to unilaterally adopt an NFU policy, and ask other nuclear-armed states to do the same. This would constitute the formal adoption of what is already essentially de facto U.S. policy.45 A U.S. NFU policy would create political space for Russia to follow suit: For Russia to consider NFU, its concerns about U.S. ballistic missile defenses, imbalances in conventional forces, and issues of NATO enlargement would need to be addressed. The United States would also need to tackle the issue of extended deterrence with its allies and move toward conventional extended deterrence.46 India and Pakistan would need a modus vivendi on Kashmir, while the United States and North Korea would need to sign a non-aggression pact. In fact, the United States could actually negotiate a mutual NFU agreement with North Korea. The United States is extremely unlikely to use nuclear weapons first on North Korea, therefore an agreement that provided a basis for imposing some restraint on the North Korean arsenal would be in America’s interest.47

Doctrinal and operational changes would need to follow such a declaration. China’s restrained nuclear arsenal provides the best example of an NFU pledge implemented in practice. Unlike the United States and Russia, China keeps its warheads and missiles separated. It has not developed precision-strike nuclear war-fighting capabilities, such as tactical nuclear weapons, and it does not keep its forces on “launch-on-warning” alert. China has also invested heavily in conventional military modernization so that it would not have to consider nuclear escalation in a conventional war.48 India, too, keeps its warheads and missiles separate in support of its NFU pledge, though some analysts argue that India’s NFU policy does not run especially deep and that it “is neither a stable nor a reliable predictor of how the Indian military and political leadership might actually use nuclear weapons.”49 Nevertheless, both countries’ operational postures reflect (to some degree) their NFU policies.50 The United States and the other nuclear powers should move in this direction.


What are the prospects for an NFU policy? On Jan. 30, 2019, Sen. Elizabeth Warren (D-MA) and Rep. Adam Smith (D-WA) introduced legislation that declared, “It is the policy of the United States to not use nuclear weapons first.”51 But Congress is divided on this.52 Skeptics have objected that the geopolitical preconditions are not ripe for an NFU policy at this time. In 2016, the Obama administration seriously considered declaring an NFU policy but then hesitated at the last minute largely because of pushback from European and Asian allies who are under the U.S. nuclear umbrella.53 Donald Trump, for his part, has been busy dismantling arms control agreements, not creating them.54

Adoption of an NFU policy will require close consultation with allies, but the U.S. administration should begin this task. As an initial step on the way to NFU, U.S. leaders should consider the recent proposal by Jeffrey Lewis and Scott Sagan that the United States should declare it will not use nuclear weapons “against any target that could be reliably destroyed by conventional means.”55 This policy would not solve the problem posed by highly asymmetric crises, as noted above. Nevertheless, it would represent an initial important declaratory statement of nuclear restraint.

The most important goal of the United States today is to prevent the use of nuclear weapons. The policy of relying on the threat to use nuclear weapons first is an outdated legacy of the Cold War. As even card-carrying realists such as the “four horsemen” recognized, given U.S. conventional capabilities, there are no circumstances in which the United States ought to start a nuclear war.56 Relying on the pretense that it might do so in order to deter a conventional threat unacceptably increases the chances of nuclear escalation. Moving toward declared NFU policies is the best way to reduce the risks of nuclear war.


Nina Tannenwald is director of the International Relations Program at Brown University’s Watson Institute for International Studies 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 targeted killing, the future of the nuclear normative order, and the effectiveness of the laws of war. 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.


3. Nuclear First-Use Is Dangerous and Unnecessary

By Jon B. Wolfsthal


The greatest military risk facing the United States and its allies is the use of nuclear weapons, whether via an accident, escalation, or deliberate use. As such, it is essential that the United States and its allies use all available tools to reduce and, if possible, eliminate this danger. American nuclear strategy continues to reserve the right to use nuclear weapons first to deter or defeat both nuclear and non-nuclear attacks. However, maintaining this posture today unnecessarily increases the risk of a nuclear strike against the United States and its friends without providing any demonstrable security benefits, despite conventional wisdom and commentary to the contrary.57 America and its allies should adopt a nuclear no-first-use (NFU) posture that would enhance deterrence and reduce the risk of nuclear use.

In today’s nuclear landscape, the United Should and its allies need to objectively consider the costs and benefits of maintaining a first-use nuclear posture. An evidence-based assessment heavily favors the United States adopting NFU and should motivate a broader effort, led by the United States, to convince other nuclear states to do the same, following India and China’s example. American nuclear weapons should only be assigned the mission they can credibly carry out: deterring other states from using nuclear weapons against America and its allies through the threat of retaliation. Going beyond this strains credibility, decreases crisis stability, and undermines the believability of America’s other defense commitments.58 It also has little, if any, proven deterrent effect on potential adversaries.

Of course, any significant change to American nuclear policy should not be imposed by fiat, but must come through sustained consultation with American allies. As the United States moves over the coming years to repair its damaged alliances and reinforce deterrence, NFU advocates must be aware that such a move could be seen by some allies — as well as adversaries — as a weakening of America’s commitment to their security. An NFU decision, therefore, needs to be met with a concerted set of other adjustments to reaffirm U.S. defense commitments and deterrence threats.

Because there is a significant security benefit to be gained — in the form of reduced risk of a nuclear accident or exchange — and because the United States and its allies can take political and conventional military actions to counter any such perception, America ought to adopt a policy of no-first-use and work with its allies to make it a credible reality. Such a change can be effectively pursued if done right.

The State of Play

Some experts believe the risk of nuclear use is as high today as during the worst days of the Cold War.59 This danger, however, remains widely underappreciated and misunderstood by both the public and national leaders. Sadly, too many nuclear practitioners, officials, and experts similarly discount the dangers of accidental use and other risks associated with nuclear weapons generally, and with first-use strategy specifically. For nuclear first-use proponents, nuclear weapons are seen as affordable and low-risk assets,60 despite repeated accidents, incidents, and spiraling costs. Even the officials and officers assigned the task of managing American nuclear operations often overestimate their own ability to avoid mistakes and control nuclear dangers.61

Convinced America can safely manage these risks, the Trump administration has sought to develop new, more diverse nuclear delivery options and weapons, and has expanded the conditions under which it would consider using nuclear weapons first, to include massive conventional attacks and other “non-nuclear strategic attacks.”62  The Defense Department has recently gone out of its way to refute arguments in favor of no-first-use,63 likely in response to multiple bills in Congress calling for America to adopt an NFU policy.64

Despite statements from the Defense Department to the contrary, adopting an NFU posture would improve stability and reduce the risks of nuclear conflict through deliberate action, accident, or miscalculation. When states maintain first-use postures and position their forces to carry out such threats, the pressure to preempt grows, as does the risk of accidental use. Changes in declaratory policy that rejects first-use, backed by operational changes designed to make first-use options less credible — such as de-alerting and de-mating — would increase decision time and reduce the risks of accidental or unintended nuclear escalation. Of course, neither America nor its adversaries can be totally certain that the first use of nuclear weapons is off the table based on policy statements alone, which is why such a statement needs to be matched by changes to nuclear operations that make first-use less plausible. Thus, any move to reconsider nuclear first-use would be observable, raising the nuclear threshold, reducing the dangers of sudden escalation, and giving decision-makers more time to de-escalate a potential nuclear crisis. Given today’s dangers, such changes are urgently needed.

We are seeing this play out already. Russian experts have recently begun raising the possibility that the demise of the Intermediate Range Nuclear Forces Treaty could lead Russia to adopt a policy of preemption.65 Ignoring for a moment that Russia violated that treaty and bears the bulk of the blame for its demise, that Russian officials are talking about a nuclear preemption strategy further demonstrates the growing instability in Europe and the increased nuclear risks that come with it.  Recent moves by the Trump administration and NATO to address the loss of the treaty — including increasing overflights of nuclear-capable aircraft and building a new generation of ground-based, nuclear-capable missile systems — do nothing to reduce these risks and in some ways make them worse.66

The United States — as the world’s foremost conventional military power — can safely adopt NFU without compromising its security or that of its allies. While it will be harder for conventionally inferior states — like Russia — to adopt NFU, the last thing Washington should do is make it easier for Moscow to use or threaten to use nuclear weapons first. A U.S. NFU stance would make it harder for others to maintain first-use doctrines and would enable the United States to take more effective collective action — militarily, politically, and economically — should a state ever cross the nuclear threshold. It would also put the United States and NATO, as well as America’s East Asian allies, in a stronger position to politically challenge states that maintain first-use postures, and to seek engagement in order to reduce the risks of nuclear use.

There is, of course, much discussion about whether America’s conventional advantage can be maintained.67 However, if America, with a defense budget more than four times as large as that of China,68 cannot maintain a conventional military advantage, then there is a larger problem with America’s defense and security strategy that no amount of nuclear weapons or threats can overcome. The core of deterrence and reassurance among allies rests not in making threats to initiate nuclear disaster, but in harnessing a collective political will to ensure that America can collectively maintain the political, military, and economic capabilities needed to defend its interests and security.

Is First-Use Credible?

If a first-use posture were safe and made America more secure, it might be worth keeping. However, it is neither of these things. Nor is there much to suggest that U.S. threats to use nuclear weapons first against other nuclear-weapon states — a threat that would be certain to bring about nuclear retaliation — are seen as credible by America’s main nuclear adversaries, including Russia and China.

Just as America is right to assume that its large, diverse, and highly-capable nuclear forces are able to deter nuclear attacks against itself and its allies, so too is Russia right to assume that its forces are able to deter a nuclear attack by the United States in the absence of a nuclear provocation. It is not clear why the United States believes now, let alone in the past, that it can credibly threaten a nuclear first-strike on Russia or its forces and then control or prevent a response or escalation. Russia plainly is aware that its conventional capabilities are no match for those of the United States and NATO, which is why it relies on a first-strike posture and has invested in and is expanding its hybrid and disinformation capabilities. The United States has no such need, and yet it maintains a first-use posture for anachronistic political reasons, which carry with them real and observable political and financial costs, as well as costs to stability. In other words, the United States believes that it is the one who will and can control nuclear escalation, but that Russia will not and cannot. A dangerous bet, with no real facts on which to base it.

In the case of smaller nuclear states like North Korea, U.S. threats to “go nuclear” first might be seen as credible because North Korea lacks the ability to destroy all of the United States in retaliation. But threats to “go nuclear” first may make Kim Jong Un more, not less, likely to rely on rapid nuclear launch decisions because of the possible vulnerability of his country’s nuclear forces and leadership to nuclear strike. First-use threats in the Korean context, in which America continues to have massive conventional advantages, actually increase the likelihood of North Korea launching a nuclear weapon first. Most informed analysts believe Kim sees nuclear weapons as an insurance policy, to be used only to prevent his destruction. If his destruction seems increasingly imminent, so too would his own willingness to “go nuclear.” As the United States adjusts to the reality of a nuclear standoff with North Korea, this reality has to be taken more seriously, and the risks of nuclear-crisis instability on the Korean peninsula more carefully considered.

U.S. allies and adversaries know that, in the absence of a threat to America’s national existence or that of an ally, an American decision to initiate nuclear war against a nuclear adversary is highly doubtful. It is far more credible, and thus effective, to promise to respond to conventional aggression with America’s and its allies’ combined conventional capabilities, and to reserve the role of nuclear weapons to deter, and if necessary retaliate against, a nuclear attack against an American ally. This as much as anything else argues in favor of adopting a no-first-use policy — to make a nuclear retaliatory threat all the more credible.

Proliferation Risks

Advocates for maintaining a first-use posture also claim, with little evidence, that adopting an NFU posture would lead allies to reconsider their own non-nuclear status.69 In theory, this is a risk. Decades ago, the extension of the U.S. nuclear umbrella influenced the decision of states to forgo independent nuclear options during the Cold War and commit instead to an international nonproliferation norm. Any proliferation now, even by a close U.S. ally, would be met with global concern and a collective response.70

In reality, it is hard to see which U.S. ally would respond to a coordinated move by the United States to adopt NFU by breaking with the U.S. alliance structure, withdrawing from the Nuclear Nonproliferation Treaty, and building nuclear weapons. To be sure, countries in Europe and Asia are facing complex security and political challenges. Yet, it remains unlikely that Germany, South Korea, Poland, or other “front line” states are likely to be driven to proliferate by a U.S. decision to maintain a strong alliance and a credible retaliatory nuclear posture while adopting an NFU pledge. Indeed, the United States adopting NFU would likely lead to steps to increase the credibility of U.S. conventional and political commitments to allied security.

Proponents of maintaining first-use options also need to consider the possibility that the very steps needed to enhance conventional deterrence of Russia and China and to reassure U.S. allies are not being pursued out of a false belief that America’s nuclear first-use policy — and investment in nuclear capabilities to back it up — will achieve the same goal.

Reassuring allies, as with deterring adversaries, requires a balanced combination of capability and intent. Above all, to feel secure, America’s allies need strong political commitments from the United States that it will ensure their security, and they need to feel confident that America can and will act upon those commitments. If the will of America to act in support of its defense and alliance commitments are questioned — as they are now in a way not seen in a generation — then no amount of nuclear posturing will make up for it. Sadly, the Trump administration has done a poor job of reassuring U.S. allies, despite making massive increases in defense spending and pursuing new nuclear weapon systems.71

Likewise, the Trump administration has done a poor job of maintaining strong and clear deterrent statements and postures toward Russia, China, and North Korea. Trump believes that unpredictability is a virtue, but in deterrence and alliance management it is the opposite. While America’s formidable nuclear and conventional capabilities remain fully able to deter nuclear attacks on the United States and its allies, the current state of America’s political leadership undermines the credible use of those capabilities. Washington needs to pursue improved political engagement and ensure its efforts to deter aggression are clear and credible. As these steps are taken, there is an opportunity to put America’s nuclear strategy on a more credible footing without damaging its defense capabilities or its alliance commitments.

Does America Need to Threaten First-Use?

Does the United States need to rely on nuclear first-use to respond to non-nuclear threats from a nuclear state? If not, then the United States can adopt a policy where the role of nuclear weapons is limited solely to nuclear deterrence.

It is far from certain that America must rely on nuclear weapons in such situations. Indeed, the scenarios identified by first-use advocates, while plausible, do not make a compelling case for a nuclear first-use policy and generally ignore its risks.72 Among these scenarios, the most prominent are the risks of a conventional attack by a state like Russia or a chemical or biological weapons attack by a nuclear state such as North Korea. A careful step-by-step review of these scenarios suggests that it is very hard, if not impossible, to imagine that the conditions would come about that would lead an American president to initiate a nuclear conflict, while it is easy to see how threatening first-use does more to increase the danger to America and its allies than to decrease it.

Before looking at these specific scenarios, it is useful to note that the United States — under both Presidents Barack Obama and Donald Trump — has clearly stated that it will not use or threaten to use nuclear weapons against non-nuclear weapon states. That means that there are a range of conventional, cyber, and chemical and biological threats that the U.S. military and political leadership agree do not require a nuclear response. Thus, it is reasonable to assume that the same goes for nuclear-armed adversaries that possess those same capabilities.

No one should believe that America can deter all conflict — with or without nuclear first-use threats — or that it can prevail quickly or cleanly in any conventional war.  No amount of defense or nuclear spending can achieve such a goal. To be sure, conventional war scenarios in Korea or the Baltics take months to play out and result in disastrous consequences for the country in which the fighting occurs, for global stability, and for the global economy.73 But in all of these scenarios, America and its allies would likely prevail if the full force of its allied conventional capabilities were brought to bear. If a conflict remains non-nuclear, the United States and its allies should be confident that they can confront, stop, and eventually repel any threat to the United States or its allies.


In the oft-cited scenario of Russian aggression against a NATO ally, it is not clear that American nuclear first-use is needed or would be credible. In the case of a conflict in a Baltic state stoked by Russian agents and actions, nuclear weapons would play no credible role. If a conflict escalates to a full conventional conflict between Russia and NATO, the consequences are unpredictable to a point. There is, of course, real concern about how long it would take NATO to stop and repel a Russian conventional attack. But in a purely conventional scenario, NATO forces over time would be able to defeat any Russian attack on NATO territory. Russia appears to believe this as well, which is why until now it has avoided a direct conventional attack against NATO (this rationale is at least as credible, if not more, as the idea that Russia is deterred from such attacks by the threat of a NATO nuclear first-strike). It is also why Russia has put forward a nuclear escalation strategy that would counter conventional losses with a first nuclear strike to protect the existence of the Russian state.

It remains unclear under what scenario it would make sense for NATO to escalate to the nuclear level when such a move would lead to a Russian nuclear response either in Europe, against the United States, or both. This remains true even given concerns about Russian cyber and unconventional attacks on U.S. command-and-control and early warning assets. By using nuclear weapons first, NATO would make it harder to prevail conventionally and, thus, such action should be avoided.  Of course, should Russia initiate a nuclear attack to avoid conventional defeat, all options would be legitimately on the table and no-first-use constraints would be lifted.

Furthermore, if Russia were the aggressor against NATO, U.S. nuclear use would limit if not eliminate the ability of Washington to rally the global community to condemn and punish Russia for its actions. The world would not be concerned with Russia’s aggression and would instead focus on the United States crossing the nuclear threshold for the first time in more than 70 years. Moreover, some would see Russia’s inevitable retaliation as legitimate.

North Korea

A second possible first-use scenario is that of a North Korean invasion of South Korea. In this hypothetical situation, South Korea and America would respond to a conventional invasion by North Korea with conventional forces. Yet even here the United States continues to hold out nuclear first-use threats.

In such a scenario, and as the United States works to degrade North Korea’s early warning and command-and-control network, nuclear threats are likely to greatly influence North Korean actions. North Korean officials could interpret U.S. and South Korean moves as a prelude to a nuclear decapitation strike, especially when the United States has emphasized so clearly and over such a long time its nuclear options on the Korean peninsula. Given that America and South Korea, with allied support, would likely defeat North Korean forces, it is unclear why threatening nuclear first-use would be an advantage to military planning or operations — yet it is quite clear how it could increase the likelihood of North Korea launching a nuclear weapon first. As horrible as a conventional war in Korea would be, it would be preferable to a nuclear war.

The only plausible scenarios in which nuclear first-use might influence Pyongyang’s behavior is on the issue of chemical or biological weapons. But here too the evidence is thin and the dangers likely too great to justify the risks. North Korea is believed to have both chemical and biological weapons capabilities and might seek to use them in an open conventional war on the peninsula. The United States and South Korea rightly want to deter North Korean leaders from even considering such use. Some current and former security and military officials believe the best way to do this is to keep open the option of nuclear escalation.74

Yet, to be effective, an American nuclear strike would need to be able to stop the use of chemical and biological weapons or be credible enough to convince North Korea’s leaders not to cross that line. However, the United States does not have anything close to perfect intelligence about North Korea’s capabilities or the location of its chemical and biological weapons production, storage, and employment sites. Even if it did, the use of multiple nuclear weapons to neutralize North Korea’s use of chemical or biological weapons remains a scenario that is hard to envision. Moreover, North Korea knows the United States has and continues to threaten nuclear use to deter or retaliate for the use of chemical and biological weapons. Should Pyongyang be in a situation where it would consider using such weapons, then it would automatically also have to consider its own preemptive use of nuclear weapons, ratcheting up the nuclear cycle of escalation and making it more, not less, likely that a smaller conflict would escalate into a broader nuclear exchange.

In these two different scenarios, it remains hard to identify the benefits of nuclear first-use, while it is clear that the posture carries with it significant risks. Thus, when it comes to the question of whether America needs to threaten first-use, it seems that not only are its conventional capabilities sufficient, but that threatening first-use lacks military or political utility or credibility, and actually risks significant escalation in most scenarios that do not benefit American or allied interests.


No-first-use is not a new idea.75 In the past, it has been considered as a way to reduce nuclear instability, protect American conventional superiority, pursue nuclear reductions and disarmament, and enhance non-proliferation goals. More recently, however, it has come back to the forefront because of concerns about the ability of any one leader, including in the United States, to be trusted with the awesome responsibility of whether and when to initiate a nuclear war.

The risk of nuclear conflict is a growing part of the global security landscape. The slow growth in nuclear states, the deterioration of traditional alliances and deterrent relationships, and broader instability on the global stage has forced both nuclear strategists and the public to wrestle anew with the appropriate role of nuclear weapons and how best the United States can achieve one of its top security objectives: how to avoid the use of nuclear weapons against itself or its allies. For some, the answer is to double down on longstanding nuclear threats and to enhance the range of options for America to initiate the use of nuclear weapons. However, there is now growing political support for reversing course and adopting a more restrained view of nuclear weapons in order to diminish the chance of them being used, while retaining the ability to use them to deter and respond to nuclear weapons use. This renewed debate requires objective thinking grounded in realistic scenarios and backed up by evidence. The adoption of a clear no-first-use posture by the United States, combined with strong efforts to repair America’s alliances, reinforce the credibility of deterrence, and reduce the risks of nuclear accident or escalation would make America and its allies more secure.


Jon B. Wolfsthal has worked on nuclear strategy, stability and nonproliferation for more than 35 years. He is a senior Advisor to Global Zero and the Director of the Nuclear Crisis Group. He is the former Senior Director at the National Security Council for Nonproliferation and Arms Control. He is a member of the Bulletin of Atomic Scientists Science and Security Board that sets the Doomsday Clock and the author of Deadly Arsenals: Tracking Weapons of Mass Destruction. He is the author of hundreds of articles and opinion pieces on nuclear security and stability. He has worked on nuclear projects on the ground in Russia, North Korea and all over the United States.  He worked in the 1990s in the U.S. Department of Energy.


4. A Considered “No” on “No First Use”

By John R. Harvey


Over the past few decades, the United States has weighed the risks and benefits to both its nuclear deterrence posture and its non-proliferation policy goals of renouncing first-use of nuclear weapons in a conflict. In President Barack Obama’s 2010 Nuclear Posture Review and, later, near the end of Obama’s second term as part of a mini-nuclear review, the adoption of a so-called “no-first-use” pledge was considered.76 Both times, Obama rejected adopting such a policy. The 2018 Nuclear Posture Review carried out by the Trump administration reviewed the policy and reaffirmed Obama’s decision.77

Recently, Rep. Adam Smith, the new chair of the House Armed Services Committee, and Sen. Elizabeth Warren have called for a U.S. no-first-use policy.78 Well-meaning supporters of no-first-use are taken with the simplicity of the idea and its potential for bolstering U.S. “moral leadership” in the world. After all, they argue, the United States has no intention of starting a nuclear war so why not just say so? Given the recent revival of this topic, it is appropriate to review some of the considerations that caused both Obama and Trump, as well as Presidents Bill Clinton and George W. Bush, to reject adopting a policy of no-first-use.

There are three major risks in adopting a nuclear declaratory policy of no-first-use. The first risk is to deterrence: Adversaries, absent a fear of reprisal, could be emboldened to act against U.S. interests. The second risk is to U.S. assurances to its allies: If America adopts no-first-use, then allies could lose confidence in America’s extended deterrence commitments. The third risk is to the goal of non-proliferation: Such lost confidence among America’s allies could spur them to develop and field their own nuclear weapons. The purported benefits of adopting a no-first-use policy, which I discuss below, are insufficient to offset these inherent risks.

Deterrence Risks

Every president since Dwight Eisenhower has viewed nuclear weapons not just as another weapon of war augmenting conventional arms, but as a special kind of weapon to be used only in the direst circumstances when vital U.S. security interests are at stake. The main concern in adopting a policy of no-first-use is that it could lead an enemy to believe that it could launch a catastrophic, non-nuclear strike against the United States, its allies, or U.S. overseas forces without fear of nuclear reprisal. Consider, for example, a North Korean biological attack on an American city that kills hundreds of thousands, or an artillery bombardment of Seoul with chemical weapons, resulting in the deaths of tens of thousands of Korean and U.S. forces and citizens. Would North Korea be more willing to contemplate such attacks if it thought it was immune to a U.S. nuclear response? Recent presidents have been unwilling to accept the risk to deterrence that would accompany a pledge of no-first-use.

Two factors might mitigate such risks to deterrence were a no-first-use policy adopted. First, a no-first-use pledge is unlikely to appear credible to an adversary contemplating major aggression. For example, North Korea is unlikely to base any military planning to reunify the Korean Peninsula by force, or plans for its regime survival after an unsuccessful effort to achieve that objective, on a U.S. promise of no-first-use. Consider China’s existing no-first-use pledge, which has not caused the United States to moderate its own nuclear posture one iota. Few states will risk their national security based on a declaratory policy that can be reversed overnight. Dominic Tierney, an academic who supports a no-first-use policy, eloquently addresses this point:

Viewed through a strategic — and perhaps more cynical — lens, the no-first-use doctrine also has a huge credibility problem. For the U.S. pledge to truly matter, a president who otherwise favors a nuclear first strike would have to decide not to press the button because of this policy. But in an extreme national crisis — one involving, say, North Korean nuclear missiles — a president is unlikely to feel bound by America’s former assurance. After all, if a country is willing to use nuclear weapons, it’s also willing to break a promise.79

Second, it’s not at all clear that an adversary could count on U.S. public opinion to act as a “brake” on an American president contemplating first use in response to a catastrophic non-nuclear attack. Several surveys conducted by Scott Sagan and Ben Valentino look at the American public’s willingness to support first-use under such circumstances. The results reveal a surprising level of support. Sagan and Valentino thus argue:

Would we drop the bomb again? Our surveys can’t say how future presidents and their top advisers would weigh their options. But they do reveal something unsettling about the instincts of the U.S. public: When provoked, we don’t seem to consider the use of nuclear weapons a taboo, and our commitment to the immunity of civilians from deliberate attack in wartime, even with vast casualties, is shallow. Today, as in 1945, the U.S. public is unlikely to hold back a president who might consider using nuclear weapons in the crucible of war.80

In other words, the American public might well demand, rather than oppose or simply tolerate, a nuclear response to a catastrophic non-nuclear attack — no-first-use pledge or not.

Thus, an adversary’s doubts about a no-first-use pledge and its belief that the U.S. public may well support breaking such a pledge in response to a horrific attack could mitigate some of the deterrence risks of adopting a no-first-use policy. However, the degree to which those risks would be mitigated remains uncertain and, so far, no president has been willing to find out.

Assurance and Nonproliferation Risks

Building and maintaining strong alliances has been a centerpiece of America’s effort to produce and sustain a more peaceful world. Critical to this is assuring U.S. allies of America’s commitment to their defense by extending to them the full range of U.S. military power.

Many countries, including those that share a border with an adversary that presents a threat to their very existence, see no-first-use as a weakening, symbolic or otherwise, of U.S. extended deterrence. In response to Chinese provocations in the western Pacific and North Korea’s nuclear tests and missile launches, Japan regularly seeks, both in official consultations and ongoing military cooperation, assurances that America will continue to fulfill its security commitments to protect the island nation. Some in South Korea have already pressed to explore an increased U.S. nuclear presence in their country to further deter regional threats.81 Loss of confidence in U.S. security commitments could cause some allies to seek accommodation with regional adversaries in ways that run counter to U.S. interests.

Moreover, both South Korea and Japan, similar to many NATO allies, have latent nuclear weapons capabilities characteristic of advanced industrial economies with commercial nuclear power. Any perceived wavering of U.S. security commitments could cause allies to develop and field their own nuclear weapons.

Further, America’s allies have made their feelings about America adopting a no-first-use policy known. U.S. officials consulted America’s allies extensively in the lead up to the 2010 and 2018 nuclear posture reviews. This dialogue has been rich and productive and, in some ways, surprising in its candor. For example, in 2009, Japanese officials briefed the Perry-Schlesinger Commission, established by Congress to seek a bipartisan approach to the U.S. nuclear posture, on specific features and capabilities of the U.S. nuclear deterrent that Japan viewed as critical to its security.82 In related dialogue, many foreign counterparts to U.S. officials, including those of Japan, have urged the United States not to adopt a no-first-use policy.83

Supposed Benefits of a U.S. No-First-Use pledge

In light of these risks, what are the benefits of a U.S. no-first-use pledge that could offset them? Would it, as Sen. Warren claims, “[reduce] the risk of a nuclear miscalculation by an adversary in a crisis … ”?84 If an adversary launches a nuclear weapon because it has misinterpreted America’s actions or intentions, or even if it launches a nuclear weapon by accident, the consequences would, of course, be tragic. Such actions must be assiduously avoided with clear crisis communications, transparency, and strong negative control of nuclear weapons. But, a U.S. no-first-use pledge, by itself, is unlikely to have any effect at all in preventing such a situation from arising in the first place.

Some argue that adopting such a policy would set an example and cause nuclear adversaries to follow America’s lead. If promises were kept, this would allow the U.S. conventional juggernaut to win wars absent the threat of nuclear use. But this outcome is unlikely. Indeed, several nuclear adversaries have acquired, or are currently seeking, nuclear weapons precisely to offset superior U.S. conventional capabilities. Again, quoting Tierney:

If [a President] made a dramatic announcement of no-first-use, it would probably have less impact than people think because other countries wouldn’t follow suit, especially if they’re weak.

Would U.S. adoption of no-first-use cause other countries to be more inclined to cooperate with the United States to work toward a strengthened nonproliferation regime and less likely to acquire nuclear weapons of their own? No evidence exists to support such a contention and, as noted above, allied perceptions of weakened extended deterrence could actually spur proliferation.

Another purported benefit of adopting a no-first-use policy is that it might silence criticism from Non-Aligned Movement countries that periodically denounce the United States for, among other things, not having disarmed unilaterally. This is unlikely. Indeed, the enormous progress made in the decades leading up to the end of the Cold War and beyond in ending the nuclear arms race, reducing nuclear stockpiles, and eliminating other global nuclear threats has done little to moderate such rhetoric.

Along these lines, some view no-first-use as a means to delegitimize nuclear weapons in general, and, more specifically, as a first step to removing from alert and eventually getting rid of the inter-continental ballistic missile (ICBM) leg of the Triad.85 After all, if ICBMs are not survivable unless used first, and if America’s policy becomes one of no-first-use, then why does the United States need them at all, much less on alert? This claim misrepresents both the role of America’s ICBMs and the obligations that America would be under as part of a no-first-use pledge. Thus, such arguments are unlikely to sway any president who views a nuclear Triad as an essential element of U.S. security for managing risk in a dangerous world.

Many who favor a U.S. no-first-use pledge see it as a way to signal to the world a reduced role for nuclear weapons in U.S. national security.86 Reducing that role, and hence the likelihood that the United States would ever have to resort to nuclear use, is a laudable goal advanced in the nuclear posture reviews of the three previous presidents. But, in regard to its foreign impact, the actual security benefits that could justify accepting the risks of this policy are not well understood, nor are they quantifiable, and so far they have not tipped the scales toward the adoption of no-first-use.

Those who support no-first-use as a way to advance U.S. security must explain what has changed for the better in the international security environment since 2010 that would cause this president, or this Congress, to reverse earlier presidential decisions rejecting it.


It has been a precept of U.S. policy for decades that deterrence is strengthened when an adversary is unsure of the precise conditions under which the United States would employ nuclear weapons — essentially, that uncertainty breeds caution. America has made exceptions, however, in certain cases to advance concrete security interests — for example, in regard to nuclear negative security assurances provided to non-nuclear weapons states that are parties in good standing with the Nonproliferation Treaty. If the United States were to adopt a policy of no-first-use, it would present clear risks for deterrence, for regional security more broadly, and to the non-proliferation regime, while the supposed benefits of such a policy that could offset such risks are largely illusory. It is thus no surprise that since the dawn of the nuclear age presidents across party lines have rejected no-first-use. The United States should continue to do so.


Dr. John R. Harvey is a physicist who has spent his career working to advance U.S. nuclear weapons programs and policies including in senior posts in the Departments of Energy and Defense. He retired from government service in 2013 as principal deputy assistant secretary of defense for nuclear, chemical, and biological defense programs.


5. Should Presidential Command Over Nuclear Launch Have Limitations? In a Word, No.

By Rachel Elizabeth Whitlark


In the United States, the president has sole authority to order the launch of nuclear weapons. This feature of the American foreign policy apparatus is unique, especially relative to other war powers, which are shared by the executive and legislative branches. While there are checks in place for procedural verification to be sure that orders from the president are carried out appropriately, there are few institutional regulations to certify that justification exists for a nuclear attack. This means that one individual — the president — has near total autonomy over what might be the most important element in national security. Though congressional calls to limit presidential authority over this power are not new,87 they have grown increasingly frequent since President Donald Trump’s election. Today, multiple proposals from politicians and scholars alike recommend imposing limitations on presidential authority to mitigate against potentially dangerous impulses.88

While passing a law to require congressional or military-legal approval before a nuclear launch could take place may seek to address fears of an unjustified attack, taking such steps would be misguided. Specifically, it would introduce complexity and dangerous time delays that would, as an unintended consequence, undermine deterrence, the quintessential purpose of nuclear weapons in the United States. That said, there are ways to build more oversight of the president’s nuclear authority while still maintaining the critical deterrence mission.

Presidential Authority

From the moment consideration begins, a president can launch nuclear weapons in mere minutes. To initiate the process, the president discusses the situation with key members of the defense establishment, including the secretary of defense, the head of U.S. Strategic Command (responsible for strategic nuclear weapons in the U.S. arsenal), and the combatant commanders whose geographical jurisdictions might be relevant to the mission at hand. As a group, these individuals are critical for discussing attack plans and targeting, as well as for offering advice and counsel to the president, who alone must make the eventual determination for how to proceed. Embedded in the deliberations are legal considerations, as the military personnel are bound by the Law of Armed Conflict,89 which demands necessity, distinction (between civilian and military targets), and proportionality for any use of force. As such, legal expertise is woven into military activities, including nuclear missions, from the planning stage to the execution. From the initial deliberation, if an order is given, it is verified by the Department of Defense and communicated to the relevant launch crews, who carry it out.90

Two sets of actors seem to be missing from or lacking formal roles in this process. The first is Congress. For foreign and military matters, the U.S. Constitution deliberately enshrined a system of shared powers between the executive and legislative branches.91 Article I gives Congress the power to declare war, raise and support armies, and provide and maintain navies. Article II reserves the role of commander-in-chief of the Army and the Navy for the office of the president. While only Congress can declare war, presidents have repeatedly ordered forces into action without congressional approval. Likewise, although the Constitution is silent on nuclear matters for obvious reasons, the president’s commander-in-chief authority has extended to control over nuclear use. One key motivation for this policy follows from the founders’ desire to enshrine civilian control over the military. Nuclear authority specifically derives from World War II, when the president’s commander-in-chief authority extended, by default, to Harry Truman’s nuclear launch decision in 1945.92 Since then, when issues have arisen regarding which war powers of the president are beyond congressional control, little has been resolved, perhaps because the judiciary has been wary of wading too far into this debate.93

In light of this lack of a formal role for Congress in nuclear command authority, on Jan. 29, 2019, Sen. Ed Markey and Rep. Ted W. Lieu reintroduced a bill first brought forward during the Obama administration that seeks to prevent the president from launching a nuclear first strike without congressional approval.94 There are 82 House co-sponsors and 13 in the Senate. Functionally, the bills seek to legally prohibit the president from using nuclear weapons without first determining that an enemy has launched a nuclear attack against the United States. Absent such a determination, the launch of nuclear weapons must be preceded by a congressional declaration of war that explicitly authorizes nuclear use.

The second set of actors without a formal role in the process is the secretary of defense and the attorney general. In light of this, a second proposal, authored by Columbia University professors Richard Betts and Matthew Waxman, supports requiring additional authentication of a presidential order to use nuclear weapons,95 and does so by formalizing a role for the defense and legal leadership. Betts and Waxman advocate two added layers of verification. First, the defense secretary, or her/his designee, would certify that the order to launch nuclear weapons was valid, i.e., that it was actually from the commander-in-chief. Second, the attorney general, or her/his designee, would certify the order was legal. Through these measures, the defense and legal authorities would have a formal role beyond their current advisory capacity.

Dangerous Limitations

Beyond constitutional concerns, there are substantive reasons to be skeptical of limiting presidential authority in this arena. Specifically, from a national security perspective, it is useful to have the ability to conduct war in the hands of a single person because of the relative speed with which one actor can mobilize when compared to the speed of 535 people. When threats manifest, it is often the case that speed and secrecy are paramount considerations for a state deciding how to respond. To that end, national security decisions like mobilizing for war could suffer — through leaks and lengthy discussion and debate — if they must occur within the halls of Congress. Speed, stealth, and nimble deliberations can be incredibly important for executing foreign policy and military operations. This remains the case both for consideration of nuclear strikes as well as for conventional scenarios, as these features are central to deterrence. Any changes to the existing system that could undermine deterrence should be avoided.

Indeed, perhaps the most critical consideration is the need to ensure and promote the deterrence and assurance operations that are the principal goals of the nuclear mission. Deterrence functions by convincing a state’s adversary that the costs and risks of its threatened reaction to an attack outweigh any benefits an aggressor might stand to gain.96 The central paradox of the Cold War was that, in order to prevent nuclear use, America had to be prepared to use nuclear weapons.97 This paradox remains in place today and must stay front and center.

Critical to this paradox is what academics describe as the always/never dilemma.98 This refers to the notion that the nuclear arsenal must always work as intended, but simultaneously never work by accident or via unauthorized use. For deterrence to work, both the possessor of the arsenal and its adversaries must believe this to be the case. If not, a commander-in-chief cannot be confident that the arsenal will work when needed. Such doubts could cause an adversary to launch a preemptive attack. Steps to undermine this balance paradoxically increase the risk of nuclear use and make the United States less safe. If, for example, the president orders a nuclear launch, but congressional dawdling fails to authorize the command (or fails to do so in a timely manner), then the “always” is undermined. As a consequence, adversaries will begin to doubt the readiness of the United States to fight to defend its interests and might seek to capitalize on this vulnerability.

Beyond this core dilemma, there are many indirect ways in which nuclear deterrence is fundamental to U.S. national security. Protecting the American homeland is only the most obvious. The nuclear arsenal also protects against enemy coercion. Furthermore, it protects allies and interests beyond the homeland. Nuclear weapons, because of the immense dangers they pose in the context of escalation, also prevent conflicts spiraling out of control.99 The U.S. nuclear arsenal, moreover, prevents nuclear proliferation to new states, as the American arsenal is extended to allies through guarantees that the United States will use it to protect them from enemy (nuclear) attack. As candidate and later President Trump discovered,100 making statements perceived to undermine these guarantees can make allies very nervous and prod them into revisiting their own nuclear capabilities.101 Any steps taken to undermine the American deterrent will, therefore, have significant direct and indirect consequences. In short, limiting presidential authority could jeopardize U.S. security by exposing it to blackmail, raising the risk of escalation, undermining alliance commitments, and endangering nonproliferation goals.

A System for the Future

This is not to suggest that the existing system is perfect and should be left entirely as is. Rather, there are modifications to consider to improve launch authority for today’s environment.

First, the country should take steps to ensure that its current and future leaders — who are increasingly less likely to have grown up during the Cold War or to have served in the military, and thus are less likely be knowledgeable about nuclear weapons — are educated about the strategic and specifically nuclear choices made during and after the Cold War. John F. Kennedy — himself a foreign affairs junkie, former service member during World War II, and deeply thoughtful individual regarding nuclear weapons102 — still balked when he learned that casualty estimates for one Soviet missile striking near an American city stood at 600,000.103 As new generations of leaders with different formative experiences and far less nuclear knowledge come into office, such information needs emphasizing.

Second, the U.S. government should examine existing procedures to explore avenues to improve the decision-making process itself. This could include training and contingency planning so leaders are more aware of the scenarios they might encounter. Decision-making under stress is far from ideal,104 and, especially for leaders with limited foreign policy exposure and experience, there may be a bias toward issuing a nuclear attack since the system was designed to launch before an enemy attack struck the American homeland.105 Particularly worrisome are leaders with limited understanding of nuclear dynamics. That said, conducting such training exercises might be especially complicated in today’s polarized news environment, where the risk of leaks is omnipresent. Politicians will not be inclined to practice decision-making scenarios if poor performance or missteps might cause personal embarrassment or political punishment. Nevertheless, it is worth exploring if practicing making the tough decisions a leader might encounter can give additional confidence to both leaders who might face such situations in the future as well as to observers assessing how the leader might respond.

Taking demonstrable and behind-the-scenes steps to improve both nuclear education and training can be productive: enhancing deterrence by convincing adversaries that clarity and deliberate actions remain features of the system, despite changes that have taken place in the international environment since the height of the Cold War.106 Such measures can also help assuage any concerns inside the force itself, in Congress, and among American citizens that these critical issues are not being taken seriously.107

It is also worth mentioning that there are some relevant policies already in place, including specifications of the circumstances under which the United States will use nuclear weapons. According to the Defense Department’s 2018 Nuclear Posture Review,108 nuclear use is circumscribed for all but the most extreme circumstances to defend U.S. vital interests. It articulates that deterrence is the sole purpose of nuclear weapons, and pledges to augment the conventional capabilities the United States will use in combat. Moreover, the document articulates negative security assurances, commitments not to use nuclear weapons against non-nuclear weapons states in good standing with their Nuclear Non-Proliferation Treaty obligations. President Barack Obama pledged this commitment and Trump has reaffirmed it.

A Role for Congress?

Legislating congressional oversight of presidential authority to launch nuclear weapons or building additional limitations into the command authority to fetter the president’s unilateral discretion is a mistake insofar as it undermines deterrence. But this is not to say that there is no role for Congress. Rather, as with most matters of foreign affairs, there are constructive ways for Congress to participate.

First, Congress should recognize that there are already some checks that exist to limit rogue presidential behavior. Specifically, the 25th Amendment provides procedures for replacing the president (or vice president) in the event of death, resignation, removal, or incapacitation. It has been used three times previously, though not because of a president’s mental state.109 When Nixon’s mental capacity was in question at the end of his presidency, key individuals, including a senator and Nixon’s own secretaries of defense and state stepped in (though with questionable legal authority) to ensure that Nixon did not hastily order a nuclear attack.110 At present, while there is an almost infinite list of policy disagreements one might have with Trump, few seem to believe that this is currently an appropriate consideration.111 Should the situation change, or if a clearly illegal order is issued by the president, then the 25th Amendment would be an appropriate recourse.

Second, Congress can and should step up in its oversight and consultative capacities more generally. The first means of doing so would be to consider a new Authorization for Use of Military Force, which is needed to govern current and future military action. The current authorization was passed in the aftermath of the Sept. 11, 2001, attacks and authorizes the use of force against those responsible. Given that the strategic environment has changed dramatically since then and the United States is still involved in at least two wars tangentially related to 9/11, it is high time for Congress to consider reasserting its authority over military action. Even though it is not related to the nuclear arsenal specifically, a new authorization would send an important signal demonstrating the seriousness with which Congress is taking its Article I responsibilities.

Third, in terms of congressional consultation, steps akin to the Betts-Waxman proposal to more formally include the secretary of defense and attorney general in the deliberative process could be appropriate. Instead of being inserted into the authorization process for a nuclear launch order, however, they would be mandated to consult with Congress far earlier in the process. Specifically, officials from the defense and legal communities, including the top cabinet leadership, should engage with Congress during peacetime to be sure that use of force scenarios are well explored and time is allotted for discussion and consultation. Doing so might offer a more reasonable solution than formally inserting these officials into the launch process. While, admittedly, getting their buy-in would hypothetically take less time than securing the approval of the whole of Congress, even this inserts time-delaying complications into the launch command that may be ill-advised for the reasons described above. It may also be unnecessary since these agencies and their perspectives are already built into the process via the strategic elements, combatant commands, and the legal authorities that oversee all military actions. Additional advanced consultation with congressional leadership, and especially leaders of key committees relevant to defense and foreign affairs, offers a useful middle ground.

Fourth, and related, Congress should draw upon its authority over the power of the purse to amplify its voice in nuclear discussions.112 Through the purse, Congress has the ability to control the modernization of the American nuclear arsenal. This authority gives the legislative branch enormous power and influence. If Congress feels that the president has cut it out of key deliberations with respect to the modernization process, then the White House’s requests and directives in this area should be taken seriously and not given carte blanche approval. Instead, agreeing to fund new weapons could be tied to increased oversight or consultation procedures. In this way, the president would be mandated to discuss war plans and scenarios where nuclear weapons might be employed. Only then would he or she get the new funding requested.

Finally, Congress can institute regular review of command and control systems, which seems especially necessary given the increased complexity of the international arena in the post-Cold War period. Today, there are more adversaries armed with nuclear weapons, cyber concerns, and more non-state challenges, all of which result in a more complicated strategic landscape than when all of America’s energies were focused on the Soviet threat. Time and money should be spent making sure that hardware, software, and wetware (human) systems are ready to face the evolving threat environment. By having hearings, seeking expert testimony, and conducting conversations with key administration officials, Congress can catalyze important discussions and build support for modifications deemed necessary.


As Congress considers ways it can involve itself in the question of presidential nuclear authority, it should pause and take caution before legislating oversight of presidential launch authority. Not only might such legislation be unachievable given the constitutional questions surrounding Articles I and II, but it might also be practically impossible as presidents have often found work-arounds in the face of such congressional assertions of power. For example, as securing congressional approval for war has proven difficult since World War II, successive presidents have deemed all subsequent military actions “extended military engagements” or not sought authorization in the first place. Indeed, presidents have circumnavigated Congress repeatedly since 1945. Moreover, it remains unclear if in today’s volatile partisan environment any president could get congressional authorization to launch nuclear weapons even with an infinite timeline. This fact would weaken the United States tremendously in the eyes of adversaries looking to make offensive or coercive gains. The ability to exploit the domestic political environment within the United States would play into their hands.

In today’s complicated and evolving strategic environment, changes are necessary to continuously improve one of the United States’ most important strategic procedures. Such improvements should not include legislating congressional authorization or other unnecessary fettering of the president’s sole launch authority. These sorts of steps would undermine the credibility of the nuclear deterrent, the lynchpin of U.S. national security policy. Instead, more concerted attention should be given to nuclear matters, including steps like mandating deliberation with key legislative leaders and improving education of national leaders regarding the use of nuclear weapons during the Cold War and now. More cross-pollination and discussion between the two branches of government that share war powers in this critical arena during peacetime is essential. Additionally, nuclear deterrence should not be made into a political wedge issue. Leaders may disagree about related issues, like arsenal size or how to confront proliferation challenges,113 but launch authority is not a political matter. Instead of altering policy out of concerns about a single president’s tendencies, the U.S. government must design a policy that puts the best interests and national security of the United States front and center today and into the future. Launch authority is critical for nuclear deterrence and that deterrent has myriad purposes. We should all be wary of taking steps that could potentially undermine those important goals.


Rachel Elizabeth Whitlark is a political scientist and assistant professor of international affairs at the Georgia Institute of Technology.


6. Somewhere Between “Never” and “Always”

By Brendan Rittenhouse Green


Many brave barrels of ink and intrepid pixels have gone to their fate to advance a better understanding of nuclear weapons. However, prior to 2016, few of these seriously grappled with the implications of a figure like Donald Trump in the Oval Office. His arrival has precipitated a reexamination of long-settled questions in nuclear analysis — most importantly the president having the sole authority to launch a nuclear weapon. Although such reconsideration of common wisdom is always welcome, the increasingly popular answer to the question of presidential nuclear authority is not: that the United States should adopt a policy requiring authorization from multiple actors to fire nuclear weapons. In fact, multiple authorization remains a bad idea, even if the alternative places formal control of America’s fate in hands of uncertain stability.

The debate over how to authorize the use of nuclear weapons is one facet of the “always/never” problem that was identified by Peter Feaver: States want high assurance that nuclear weapons will always be used when directed and that they will never be used otherwise.114 The problem is that there tends to be a tradeoff between “always” and “never.” Measures that add safeguards against accidental or unauthorized use tend to reduce operational efficacy in situations where speed and tight control are crucial, while command-and-control arrangements that prioritize flexibility and speed tend to be more vulnerable to abuse and/or mishaps.115

The balance between “always” and “never” can be set in a number of different ways, and states can reasonably make adjustments depending on the circumstances. For instance, warfighting doctrines (either for a damage-limiting first strike or for battlefield use) are more operationally challenging, often depending on speed, coordination, and adaptability. By contrast, doctrines emphasizing retaliatory punishment do not necessarily depend on a quick response, though credibility may be more fundamental for deterrence. Likewise, in peacetime, the need to prioritize successful nuclear operations would appear to be less than during a crisis or war, while the threat of unauthorized use is also lower. The key policy question is, under what circumstances, if any, is making changes to the authority to launch nuclear weapons a good way to push the balance toward “never?”116

Several recent articles argue that such circumstances were created with Trump’s election. Whether referring to the president explicitly or not, they all express the fear of an unstable or irrational actor launching a nuclear war. These arguments for multiple authorization come in several flavors. More restrictive arguments aim to require the explicit approval of other constitutional actors — either the Supreme Court or a majority of a specially constituted congressional committee — or the uniformed military.117 Less restrictive arguments suggest that the president should be compelled to consult with other civilian and military actors, time and circumstances permitting, without actually giving them veto power.118

How Effective Are Legal Restrictions?

One problem with arguments like these is that the most effective restrictions on an irrational president’s nuclear authority are informal, not legal, and by and large already exist. In the case of a bolt-from-the-blue peacetime attack, resistance from the armed forces — perhaps even vigorous resistance — can probably be expected regardless of the legal situation. Jeffrey Lewis and Bruno Tertrais note that “captains of US SSBNs [ballistic missile submarines] are expected to make communications contact in the event of [an] unexpected launch order that seems out of place or character,” citing one former captain who is on record saying “that, in the event of a peacetime launch, he would insist on confirmation and a justification.”119 Similarly, at the height of the Watergate scandal, when President Richard Nixon was drinking heavily and talking bombastically about nuclear weapons, Secretary of Defense James Schlesinger famously gave orders that military commanders needed to double check with him or Secretary of State Henry Kissinger before executing any order for a nuclear launch.120 Personally, I think a bolt-from-the-blue attack order is more likely to precipitate a soft coup than it is actual nuclear use. In such a case, the constitutional fallout is likely to be significant and thus it ought to be the focus of our attention.

At the same time, during a crisis or conventional war, legal requirements are not likely to stop a determined president, rational or otherwise. For instance, President Dwight Eisenhower noted in public that, regardless of Congress’ formal power to declare war, any president who didn’t act with alacrity to order a nuclear attack when needed to protect the American people “should be hanged.”121 He privately reassured members of Congress that “in the event of a real emergency,” he “would not come to Congress, but” would “go ahead” and act on his own.122 Even when America was at peace, Eisenhower circumvented the legal restrictions of the Atomic Energy Act, which forbade sharing custody of nuclear weapons with American allies. He set up a system where American controls over nuclear weapons in Europe were extremely loose. As Eisenhower put it, “we are willing to give, to all intents and purposes, control of the weapons. We retain titular possession only.”123 Congressional reports howled with outrage, but the policy was not changed until a new administration decided to change it. Eisenhower was an exceptional president in many ways, but not with regard to legal restrictions and foreign policy.

In any event, requiring a president to consult with other actors prior to launching a nuclear weapon, as envisioned by less restrictive proposals, will probably take place anyway in the form of the chain of command pushing back during peacetime, as described above. More restrictive proposals for checking presidential nuclear authority would need to be aimed at very difficult, and rare, cases: situations where a determined but irrational president wants to launch a nuclear strike and is unsuccessfully opposed by military and perhaps other civilian authorities, likely during a war or crisis. The only way to stop nuclear use under these circumstances is to require multiple authorizations, with each party possessing an effective veto. This is the kind of command arrangement that was pursued by the Soviet Union during the Cold War and is still in place in Russia today.124

Tilting Too Far Toward “Never”

However, this leads to a second problem, which is that proposals for multiple authorizations tilt the dial much too far toward “never” at the expense of “always.” The costs of such an approach should not be dismissed as fanciful, as there are many signs that the world could be headed for increased nuclear competition. Keir Lieber and Daryl Press point out that technological change has made nuclear arsenals more vulnerable than ever before.125 I argue that these technical trends, along with unstable beliefs about mutually assured destruction and domestic hurdles to arms control, have heightened the global risks of peacetime nuclear competition.126 Similarly, the 2018 Nuclear Posture Review highlights the increasing emphasis that American adversaries are placing on their nuclear forces. Analysts like Brad Roberts have made similar observations.127 Even states with historically stable nuclear force postures, like India, are moving toward counter-force competition.128 If these trends persist, requiring multiple levels of authorization could be costly during war and crises, as well as peacetime.

In the event of a war, the time needed to surmount a congressional or Supreme Court veto could spell doom in instances where a first strike might be justified, such as an attempt to preempt a North Korean nuclear force about to fire. Moreover, additional veto concerns of any sort make retaliatory missions much more challenging, since they increase the demands on the secure communications capabilities and continuity of government procedures necessary to ride out an attack. During a crisis, adding veto power beyond the president could encourage American adversaries to take more nuclear risks in the belief that Washington will have a relatively lower risk tolerance. The result could be more nuclear crises with worse outcomes. In the end, almost the entire value of nuclear weapons comes from the (at least implicit) threat to use them. Standing up before the world and ostentatiously making them more difficult to use during a crisis sends exactly the wrong signal.129

A complex system requiring multiple authorizations could impose peacetime costs as well. The experience of the Soviets is instructive here. In the later part of the Cold War, Soviet military officers and civilian analysts alike were deeply pessimistic about the survivability of the Soviet nuclear command system. In part, this was a technical problem, but these worries were also caused by political decisions about Soviet command and control, including the multiple actors needed to authorize a launch. The command system necessitated buying heavily redundant nuclear forces and early warning capabilities to compensate for decapitation risks.130 It also caused a ferocious civil-military upheaval, with the General Staff of the Armed Forces of the U.S.S.R. pressing strongly for greater ability to launch on warning. Chief of the General Staff Marshal Nikolai Ogarkov was eventually removed in response, at least in part, to this controversy.131

Importantly, Soviet command arrangements did not go unnoticed in Washington. When the CIA learned of the Soviet multiple authorization system in 1977, National Security Adviser Zbignew Brzezinski flagged the issue for President Jimmy Carter. Brzezinski noted that a Soviet decision to retaliate was not “the choice…of a single individual.” Instead, it was “probably a three-man collegial decision.” Brzezinski concluded that consequently “it might be very difficult for the USSR to ‘launch from under attack’ unless the three men are kept constantly in touch — no separate vacations.”132 The Carter administration’s interest in Soviet command and control was part of a broader learning process that indicated that Soviet nuclear forces were much more vulnerable than policymakers had initially anticipated. In conjunction with other changes in the intelligence picture, these discoveries led to a more aggressive American nuclear posture during Carter’s last three years in office.133

Domestic Considerations

A third problem with command and control based on multiple authorizations is its effects on domestic politics.134 Though some analysts claim to use Trump only as a foil to highlight structural risks that would exist in any administration, one cannot help but notice that there was a conspicuous dearth of interest in presidential control of nuclear weapons during the Barack Obama years.135 Whatever one thinks of Trump, it seems like a bad idea to change long-standing defense structures in response to displeasure with the current Oval Office occupant, however justified. Such a move would set a bad precedent, worsen political polarization, and be perceived as highly partisan.

Some would defend the motives behind instituting a multiple authorization protocol as a reconstitution of congressional authority in national security matters. Congress is, after all, granted the sole power to declare war under Article I of the Constitution.136 One would think that Obama’s robust defense of presidential prerogatives over the use of force and Congress’ craven abdication on most national security issues of significance would have cured analysts of this particular anachronism. It has the air of special pleading about it — one doubts that any analysts calling to restore congressional dignity are really prepared to embrace a thorough-going originalist approach to the constitution, or even the kind of foreign policy that a legislature could manage.

There is much to be said against the new understanding of the constitution that arose, largely without amendment, in the 20th century. But one of its great virtues is that is has provided a viable political order under modern conditions, including a global foreign policy, as well as a massive technological revolution in warfare. Together, these conditions increase the need for making speedy decisions about nuclear war while making those decisions largely immune to legislative oversight. For good or for ill, both Congress and the executive recognized this fact during the Cold War. Attempting to close our eyes to it now seems unwise. The romantic in me would like to enhance congressional authority in foreign policy, and much else besides. The realist in me, however, has learned to stop worrying and to love leviathan, who, as Hobbes promised us, at least provides the most promising route to survival.

Moreover, I would argue that the president is optimally situated to provide domestic legitimacy to nuclear decision-making, insofar as such legitimacy is possible at all. The president owes his authority to victory in a nation-wide election, conducted according to the procedures mandated by the Constitution. No other elected official who might be included in the chain of command can claim as much, while the mandate of civilian appointments like the secretary of defense or the Supreme Court justices is even more tenuous. And, of course, military officers have no democratic mandate at all, their many other virtues notwithstanding. If democracy is to have the high national value it is generally accorded, surely decisions of national survival should be made by the one office elected by the majority of the entire nation.

In the end, nuclear weapons force policymakers to face uncomfortable truths. No sane person wants to see these weapons used. But for America to be able to effectively deter its enemies, it must be prepared to use its arsenal in extreme circumstances and use it with speed and certainty. Eisenhower, again, put it best: In nuclear operations, “the United States would be applying a force so terrible that one simply could not be meticulous as to the methods by which this force was brought to bear.”137 America’s national interests and, indeed, its survival, require a clear chain of command headed by the only individual who has a democratic and constitutional claim to speak for the entire nation: the president. Even if that president is Donald Trump.


Brendan Rittenhouse Green is assistant professor of political science at the University of Cincinnati.


1 Sigal Samuel, “The Biggest Threat Facing the US, According to the First Democratic Debate,” Vox, June 27, 2019, Trump, for his part, has made similar comments. During his run for the presidency in 2016, he said, “Biggest problem, to me, in the world, is nuclear, and proliferation.” He added, “When people talk global warming, I say the global warming that we have to be careful of is the nuclear global warming. Single biggest problem that the world has.” See, “Transcript: Donald Trump Expounds on His Foreign Policy Views,” New York Times, March 26, 2016,

2 On nuclear safety and accidents, see Scott D. Sagan, The Limits of Safety: Organizations, Accidents, and Nuclear Weapons (Princeton, NJ: Princeton University Press, 1995); Eric Schlosser, Command and Control: Nuclear Weapons, the Damascus Accident, and the Illusion of Safety (New York: Penguin, 2014).

3 This is what Thomas Schelling referred to as “the dynamics of mutual alarm.” See Thomas C. Schelling, Arms and Influence (New Haven, CT: Yale University Press, 1966), 221.

4 On emerging developments in counterforce technology, see for example Keir A. Lieber and Daryl G. Press, “The New Era of Counterforce: Technological Change and the Future of Nuclear Deterrence,” International Security 41, no. 4 (Spring 2017): 9-49, On escalation dynamics in this area, see Barry R. Posen, Inadvertent Escalation: Conventional War and Nuclear Risks (Ithaca, NY: Cornell University Press, 1991). On nuclear postures, see Vipin Narang, Nuclear Strategy in the Modern Era: Regional Powers and International Conflict (Princeton, NJ: Princeton University Press, 2014).

5 On extended deterrence dilemmas, see for example Schelling, Arms and Influence.

6 This, in fact, is one reason why the United States tends to resist nuclear proliferation. On this point, see for example, Francis J. Gavin, Nuclear Statecraft: History and Strategy in America’s Atomic Age (Ithaca, NY: Cornell University Press, 2012), 27.

7 Nina Tannenwald, “The Vanishing Nuclear Taboo? How Disarmament Fell Apart,” Foreign Affairs 97, no. 6 (November/December 2018): 16–24,; Nina Tannenwald, “The Nuclear Taboo: The United States and the Normative Basis of Nuclear Non-Use,” International Organization 53, no. 3 (Summer 1999): 433–68,; Nina Tannenwald, “Stigmatizing the Bomb: Origins of the Nuclear Taboo,” International Security 29, no. 4 (Spring 2005): 5-49,

8 The problem would especially acute in cases where a conventional response might not be feasible. This was the situation that existed with respect to Europe during the Cold War, particularly during the Berlin crises that occurred between 1958 and 1962. As President John Kennedy said at one point, “I suppose if we get involved in a war in Europe we will have no choice but to use nuclear weapons.” See John C. Ausland, “A Nuclear War to Keep Berlin Open?” International Herald Tribune, June 19, 1991. A crisis over Taiwan today might generate a similar set of dynamics.

9 Such a development would by no means be unprecedented. For example, see Rebecca K.C. Hersman and Robert Peters, “Nuclear U-Turns: Learning from South Korean and Taiwanese Rollback,” Nonproliferation Review 13, no. 3 (November 2006): 539–53,

10 Green, it is worth noting, has done remarkable work in this area. For example, see Austin Long and Brendan Rittenhouse Green, “Stalking the Secure Second Strike: Intelligence, Counterforce, and Nuclear Strategy,” Journal of Strategic Studies 38, no. 1-2 (2015): 38–73,; Brendan Rittenhouse Green and Austin Long, “The MAD Who Wasn’t There: Soviet Reactions to the Late Cold War Nuclear Balance,” Security Studies 26, no. 4 (2017): 606–41,

11 For the seminal work related to this issue, see Marc Trachtenberg, A Constructed Peace: The Making of the European Settlement, 1945-1963 (Princeton, NJ: Princeton University Press, 1999), 146–200.

12 Whitlark has addressed the role of leaders in nuclear decision-making in her other work. See Rachel Elizabeth Whitlark, “Nuclear Beliefs: A Leader-Focused Theory of Counter-Proliferation,” Security Studies 26, no. 4 (2017): 545–74,

13 Ankit Panda, “’No First Use’ and Nuclear Weapons,” Council on Foreign Relations, July 17, 2018, In fact, there is some evidence that it was real and affected Soviet war planning. See Soviet Intentions, 1965-1985, (BDM Federal, Inc., 1995), chap. 3, 41–43,

14 Kumar Sundaram and M.V. Ramana, “India and the Policy of No First Use of Nuclear Weapons,” Journal for Peace and Nuclear Disarmament 1, no. 1 (February 2018): 152–68,

15 Sadia Tasleem, “Pakistan’s Nuclear Use Doctrine,” Carnegie Endowment for International Peace, June 30, 2016,

16Although the Trump administration rejected an NFU pledge in its 2018 Nuclear Posture Review, increasing calls in recent years for the United States to adopt an NFU policy from former government officials, members of Congress, and civilian analysts, as well as serious consideration by the Obama administration itself in 2016, draw on this logic.

17 Zhenqing Pan, “A Study of China’s No-First-Use Policy on Nuclear Weapons,” Journal for Peace and Nuclear Disarmament 1, no. 1 (May 2018): 124,

18 For its part, India has a carve-out in its NFU pledge for chemical and biological weapons attacks.

19 Ronald Mitchell, “Regime Design Matters: Intentional Oil Pollution and Treaty Compliance,” International Organization 48, no. 3 (Summer 1994): 425–58,

20 Memo, McNamara to Chairman, Joint Chiefs of Staff, “Task Force Reports,” Feb. 20, 1961, US Nuclear History,00307, National Security Archive, 1.

21 Morton Halperin, “Proposal for a Ban on the Use of Nuclear Weapons,” Institute for Defense Analyses, Special Studies Group, Study Memorandum No. 4 (Washington DC: IDA, 1961), 12, iv.

22 Scott D. Sagan, “The Case for No First Use,” Survival 51, no. 3, (2009): 163–82,

23The interview can be seen here: “PM Shri Narendra Modi’s Interview to India TV: 04.05.2019,” YouTube video, May 4, 2019,

24 Amy F. Woolf, “U.S. Nuclear Weapons Policy: Considering ‘No First Use,’” Congressional Research Service, March 1, 2019.

25 “Dangers of a Nuclear No First Use Policy,” U.S. Department of Defense, April 2019,

26 Parris H. Chang, “No-First Use Would Only Embolden China,” Bulletin of the Atomic Scientists, Sept. 21, 2016,

27 Michael S. Gerson, “No First Use: The Next Step for U.S. Nuclear Policy,” International Security 35, no. 2 (Fall 2010): 9,

28 James M. Acton, “Escalation Through Entanglement: How the Vulnerability of Command-and-Control Systems Raises the Risks of Nuclear War,” International Security 43, no. 1 (Summer 2018): 56–99,; Fiona S. Cunningham and Taylor M. Fravel, “Why China Won’t Abandon Its Nuclear Strategy of Assured Retaliation,” Policy Brief, U.S.-China Project (Institute for Security and Conflict Studies, Elliott School of International Affairs, February 2016).

29 Gerson, “No First Use,” 9.

30 Keith B. Payne, “Strategic Hubris,” in “Forum: The Case for No First Use: An Exchange,” Survival 51, no. 5 (October/November 2009): 27–32,

31 Scott Sagan, “Reply: Evidence, Logic and Nuclear Doctrine” in “Forum: The Case for No First Use,” 30–41.

32 Todd S. Sechser and Matthew Fuhrmann, Nuclear Weapons and Coercive Diplomacy (Cambridge, UK: Cambridge University Press, 2017).

33 James E. Doyle, “Nuclear No First Use (NFU) Is Right for America,” Real Clear Defense, July 12, 2016,

34 Quoted in Elbridge Colby, “If You Want Peace, Prepare for Nuclear War,” Foreign Affairs 97, no. 6 (November/December 2018): 30, China can afford to declare an NFU policy more easily than the United States because it lacks the kind of wide-ranging extended deterrence commitments of the United States.

35 Sagan, “No First Use”; Steve Fetter and Jon Wolfsthal, “No First Use and Credible Deterrence,” Journal for Peace and Nuclear Disarmament 1, no. 1 (April 2018): 102–14,

36 Kingston Reif and Daryl Kimball, “Rethink Old Think on No First Use,” Bulletin of the Atomic Scientists, Aug. 29, 2016,

37 Sagan, “No First Use.”

38 Alberto Perez Vadillo, “Beyond the Ban: The Humanitarian Initiative and Advocacy of No First Use Nuclear Doctrines,” BASIC (British-American Security Information Council), May 10, 2016, 11,

39 Bruce Blair, “How Obama Could Revolutionize Nuclear Weapons Strategy Before He Goes,” Politico Magazine, June 22, 2016,

40 Vadillo, “Beyond the Ban,” 12.

41 Sagan, “No First Use.”

42 Alexander Lanoszka and Thomas Leo Sherer, “Nuclear Ambiguity, No-First-Use and Crisis Stability in Asymmetric Crises,” Nonproliferation Review 24, no. 3-4 (2017): 343–55,

43 I thank Vipin Narang for this comment on Twitter.

44 Jeffrey G. Lewis and Scott D. Sagan, “The Nuclear Necessity Principle: Making U.S. Targeting Conform with Ethics and the Laws of War," Daedalus 145, no. 4 (Fall 2016): 62–74,

45 Blair, “How Obama Could Revolutionize Nuclear Weapons Strategy.”

46 Sagan, “No First Use.”

47 Abigail Stowe-Thurston, “Why It’s Time to Negotiate a Mutual No First Use Policy with North Korea,” NK News, April 15, 2019,

48 Pan, “China and No First Use,” 117.

49 Sundaram and Ramana, “India and No First Use,” 152.

50 Cunningham and Fravel, “Why China Won’t Abandon Its Nuclear Strategy of Assured Retaliation.”

51 “A Bill to Establish the Policy of the United States Regarding the No-First-Use of Nuclear Weapons,” S.B. 272, 116th Congress (2019–2020),; “To Establish the Policy of the United States Regarding the No-First-Use of Nuclear Weapons,” H.B. 921, 116th Congress (2019–2020),

52 “Fischer: ‘A No-First-Use Policy Erodes Deterrence,” Office of Deb Fischer, United States Senator for Nebraska, Jan. 30, 2019,

53 Nina Tannenwald, “The Vanishing Nuclear Taboo? How Disarmament Fell Apart,” Foreign Affairs (November/December 2018),

54 Paul R. Pillar, “Trump’s Demolition of Arms Control,” National Interest, May 1, 2019,

55 Lewis and Sagan, “The Nuclear Necessity Principle.”

56 George P. Shultz, William J. Perry, Henry A. Kissinger, and Sam Nunn, “A World Free of Nuclear Weapons,” Wall Street Journal, Jan. 4, 2007,

57 Brad Roberts, “Debating Nuclear No-First-Use, Again,” Survival 61, no. 3 (June–July 2019): 39-56,

58 Scott D. Sagan, “The Commitment Trap: Why the United States Should Not Use Nuclear Threats to Deter Biological and Chemical Weapons Attacks,” International Security 24, no. 4 (Spring 2000): 85–115,

59 “The Doomsday Clock,” Bulletin of the Atomic Scientists, accessed July 12, 2019,; “Urgent Steps to De-Escalate Nuclear Flashpoints,” Nuclear Crisis Group, Global Zero (June 2017),

60 “Ground Truth: Nuclear Weapons: Essential, Affordable, Cornerstone of Our Defense,” House Armed Services Committee, March 5, 2019,

61 “Report of the Secretary of Defense Task Force on DoD Nuclear Weapons Management: Phase I: The Air Force’s Nuclear Mission,” Department of Defense (September 2008),

62 Nuclear Posture Review, Department of Defense (February 2018),

63 “Dangers of a Nuclear No First Use Policy,” Department of Defense, April 1, 2019,

64 “Chairman Smith, Senator Warren Introduce Bill Establishing ‘No First Use’ Policy for Nuclear Weapons,” Office of Adam Smith, Jan. 30, 2019,

65 Private discussions with author, Moscow, April 2019.

66 Robin Emmott, “NATO Calls on Russia to Destroy New Missile, Warns of Response,” Reuters, June 25, 2019,

67 Summary of the National Defense Strategy, Department of Defense, January 2018,; “Providing for the Common Defense, The Assessment and Recommendations of the National Defense Strategy Commission,” U.S. Institute for Peace, November 2018,

68 “U.S. Military Spending vs. the World,” National Priorities Project, last accessed July 12, 2019,

69 Richard C. Bush and Jonathan D. Pollack, “Before Moving to ‘No First Use,’ Think About Northeast Asia,” Brookings Institute, July 20, 2016,; and, Franklin C. Miller and Keither B. Payne, “The Dangers of No-First-Use,” Bulletin of the Atomic Scientists, Aug. 22, 2016,

70 This assumes a response in keeping with American security interests. The author has serous doubts about how Trump would respond to a Saudi Arabian decision to pursue nuclear weapons.

71 Nuclear Posture Review; and, Sandra Erwin, “Trump’s 2020 Budget Increases Defense Spending by 5 Percent, Funds Space Force,” Space News, March 11, 2019,

72 Nuclear Posture Review; and, “Dangers of a Nuclear No First Use Policy.”

73 Dennis P. Halpin, “Millions Dead: You Can’t Imagine How Bad a Second Korean War Could Get,” Natonal Interest, April 26, 2019,; Dave Majumdar, “Europe’s Worst Nightmare: Here’s What a NATO-Russia War Would Look Like,” National Interest, March 31, 2019,

74 The 2018 Nuclear Posture Review cites the threat of “non-nuclear strategic attack,” including chemical and biological threats, as a main motive for maintaining nuclear first-use scenarios.

75 Morton H. Halperin, “A Proposal for a Ban on the Use of Nuclear Weapons,” Institute for Defense Analyses, Oct. 6, 1961,

76 Nuclear Posture Review Report, Department of Defense, April 2010,

Regarding the “mini” review see: Josh Rogin, “Obama Plans Major Nuclear Policy Changes in His Final Months,” Washington Post, July 10, 2016,

77 Nuclear Posture Review, Department of Defense, February 2018,

78 Joe Gould, “Warren, Smith Introduce Bill to Bar US from Using Nuclear Weapons First,” Defense News, Jan. 30, 2019,

79 Dominic Tierney, “Refusing the Nuke First,” Atlantic, Sept. 14, 2016,

80 Scott D. Sagan and Benjamin A. Valentino, “Would the U.S. Drop the Bomb Again?” Wall Street Journal, May 19, 2016,

81 Dan Lamothe, “Pentagon Chief Says He Was Asked About Reintroducing Tactical Nuclear Weapons in South Korea,” Washington Post, Sept. 18, 2017,

82 Masakatsu Ota, “Japan Warned U.S. Over Reductions to Nuclear Arsenal and Sought Flexible Deterrence, 2009 Memo Reveals,” Japan Times, April 4, 2018, - .XMh0ELi1tPY.

83 Josh Rogin, “U.S. Allies United to Block Obama’s Nuclear ‘Legacy,’” Washington Post, Aug. 14, 2016,

84 “Senator Warren, Chairman Smith Unveil Legislation to Establish ‘No-First-Use’ Nuclear Weapons Policy,” Press Release, Office of Sen. Elizabeth Warren, Jan. 30, 2019,

85 James E. Cartwright and Bruce G. Blair, “End the First-Use Policy for Nuclear Weapons,” New York Times, Aug. 14, 2016,

86 “Senator Warren, Chairman Smith Unveil Legislation.”

87 In 1972, Sen. William Fulbright sought to pass legislation prohibiting presidential use of the nuclear armed forces absent a declaration of war by Congress.

88 Owen Daugherty, “Dems Reintroduce Bill to Prevent Nuclear First Strike Without Congressional Approval,” The Hill, Jan. 29, 2019,; Richard K. Betts and Matthew Waxman, “Safeguarding Nuclear Launch Procedures: A Proposal,” Lawfare, Nov. 19, 2017,; Lisbeth Gronlund, David Wright, and Steve Fetter, “How to Limit Presidential Authority to Order the Use of Nuclear Weapons,” Bulletin of the Atomic Scientists, Jan. 23, 2018,

89  “Law of War Manual,” Department of Defense, June 2015 (Updated December 2016),

90 For a high-level overview of the process, see David Merrill, Nafeesa Syeed, and Brittany Harris, “To Launch a Nuclear Strike, President Trump Would Take These Steps,” Bloomberg, Jan. 20, 2017,

91 “Balance of U.S. War Powers,” Council on Foreign Relations, accessed May 30, 2019,

92 Nikolai Sokov and Miles A. Pomper, “It Is Time to Update the President’s Nuclear Command Authority,” National Interest, Oct. 4, 2018,

93 For a discussion of these constitutional and legal issues, see Stephen P. Mulligan, “Legislation Limiting the President’s Power to Use Nuclear Weapons: Separation of Powers Implications,” Congressional Research Service, Nov. 3, 2017,

94 “Restricting First Use of Nuclear Weapons Act of 2017,” H.R. 669, 115th Congress (2017–2018),; “Restricting First Use of Nuclear Weapons Act of 2017,” S.B. 200, 115th Congress (2017–2018),

95 Betts and Waxman, “Safeguarding Nuclear Launch Procedures.”

96 Glenn Herald Snyder, Deterrence and Defense: Toward a Theory of National Security (Princeton, NJ: Princeton University Press, 1961).

97 C. Robert Kehler, “Statement of General C. Robert Kehler United States Air Force (Retired) Before the Senate Foreign Relations Committee,” Nov. 14, 2017,

98 Peter D. Feaver, “Command and Control in Emerging Nuclear Nations,” International Security 17, no. 3 (1992): 160–87,

99 Robert Jervis, The Meaning of the Nuclear Revolution: Statecraft and the Prospect of Armageddon (Ithaca, NY: Cornell University Press, 1989); Charles L. Glaser, Analyzing Strategic Nuclear Policy (Princeton, NJ: Princeton University Press, 1990).

100 “Transcript: Donald Trump Expounds on His Foreign Policy Views,” New York Times, March 26, 2016,

101 Gene Gerzhoy, “Alliance Coercion and Nuclear Restraint: How the United States Thwarted West Germany’s Nuclear Ambitions,” International Security 39, no. 4 (Spring 2015): 91–129,

102 Rachel Elizabeth Whitlark, “Nuclear Beliefs: A Leader-Focused Theory of Counter-Proliferation,” Security Studies 26, no. 4 (2017): 545–74,

103 Richard Reeves, President Kennedy: Profile of Power (New York: Simon & Schuster, 1993), 175.

104 Irving L. Janis and Leon Mann, Decision Making: A Psychological Analysis of Conflict, Choice, and Commitment (New York: Free press, 1977); Margaret G. Hermann, “Indicators of Stress in Policymakers During Foreign Policy Crises,” Political Psychology 1, no. 1 (Spring 1979): 27–46,

105 Albert Wohlstetter, “The Delicate Balance of Terror,” Foreign Affairs, no. 37 (1959): 211,

106 I do not mean to minimize the difficulty of pursuing some of these recommendations, especially given classification and politicization concerns. Rather, I seek to suggest that these are more fruitful avenues for consideration, even if difficult, inasmuch as they may improve the existing system without undermining deterrence.

107 Kehler, “Statement of General C. Robert Kehler United States Air Force (Retired).”

108 The nuclear posture review is the Department of Defense’s process to determine what role nuclear weapons should have in the U.S. arsenal. Reviews generally happen with each new presidential administration or quadrennially. The 2018 document is accessible here: Nuclear Posture Review, Department of Defense, February 2018,

109 David Greenberg and Rebecca Lubot, “Stop Talking About the 25th Amendment. It Won’t Work on Trump,” Politico Magazine, Sept. 8, 2018,

110 Garrett M. Graff, “The Madman and the Bomb,” Politico Magazine, Aug. 11, 2017,

111 Articles of impeachment for a high misdemeanor, felony, or treason are a different matter. This is not to ignore the resistance evidently occurring within the Trump administration itself. Even from within, however, discussions of the 25th Amendment appear to have been set aside at least for the time being. “I am Part of the Resistance Inside the Trump Administration,” New York Times, Sept. 5, 2018,

112 Mulligan, “Legislation Limiting the President’s Power to Use Nuclear Weapons.”

113 Whitlark, “Nuclear Beliefs.”

114 Peter D. Feaver, “Command and Control in Emerging Nuclear Nations,” International Security 17 no. 3 (Winter 1992/93): 163,

115 Discussing many of these issues in the context of U.S. military organizations is Scott D. Sagan, The Limits of Safety (Princeton, NJ: Princeton University Press, 1995).

116 I take no position on whether it is desirable to push the balance of priorities toward “never.”

117 See, e.g., Michael E. O’Hanlon, “Going It Alone? The President and the Risks of a Hair-Trigger Nuclear Button,” Order From Chaos, March 1, 2016,; Ted Lieu, “H.R.669 - Restricting First Use of Nuclear Weapons Act of 2017,” 115th Congress (2017–2018),; Peter D. Zimmerman, “Two to Tango With Nuclear Weapons,” US News & World Report, April 27, 2017,; Jeffrey Bader and Jonathan D. Pollack, “Time to Restrict the President’s Power to Wage Nuclear War,” New York Times, Jan. 20, 2018,; Alex Wellerstein, “No One Can Stop President Trump from Using Nuclear Weapons. That’s by Design,” Washington Post, Dec. 1, 2016,; Garrett M. Graff, “The Madman and the Bomb,” Politico Magazine, Aug. 11, 2017,; Richard K. Betts and Matthew C. Waxman, “The President and the Bomb,” Foreign Affairs, (March/April 2018),

118 Jeffrey Lewis and Bruno Tertrais, “Pressing the Button: How Nuclear-Armed Countries Plan to Launch Armageddon (And What to Do About the U.S.),” War on the Rocks, April 24, 2019,

119 Jeffrey G. Lewis and Bruno Tertrais, “The Finger on the Button: The Authority to Use Nuclear Weapons in Nuclear-Armed States,” CNS Occasional Papers no. 45 (Monterey, CA: Middlebury Institute of International Studies at Monterey, 2019), 39.

120 Graff, “The Madman and the Bomb.”

121 Graff, “The Madman and the Bomb.”

122 Brendan Rittenhouse Green, “The President and Nuclear Weapons, or: How I Learned to Stop Worrying and Love Trump Having the Bomb,” War on the Rocks, Sept. 25, 2017,

123 Marc Trachtenberg, A Constructed Peace: The Making of the European Settlement, 1945-1963 (Princeton, NJ: Princeton University Press, 1999), 200.

124 Lewis and Tertrias, “The Finger on the Button,” 11.

125 Keir A. Lieber and Daryl G. Press, “The New Era of Counterforce: Technological Change and the Future of Nuclear Deterrence,” International Security 41, no. 4 (Spring 2017): 9–49,

126 I make this argument in my forthcoming book: Brendan Rittenhouse Green, The Revolution That Failed: Nuclear Competition, Arms Control, and the Cold War (New York: Cambridge University Press, 2019), chap. 9.

127 Nuclear Posture Review, Department of Defense, February 2018, 5–14,; Brad Roberts, The Case for U.S. Nuclear Weapons in the 21st Century (Stanford, CA: Stanford University Press, 2015).

128 See Christopher Clary and Vipin Narang, “India’s Counterforce Temptations: Strategic Dilemmas, Doctrine, and Capabilities,” International Security 43, no. 3 (Winter 2018/19): 7–52,

129 These examples are covered in Green, “The President and Nuclear Weapons.”

130 The United States was keenly worried about such risks during the Cold War and, as a consequence, pre-delegated authority to use nuclear weapons, as well as the ability to launch, in ways that the Soviet Union was loath to contemplate. See Bruce G. Blair, The Logic of Accidental Nuclear War (Washington, D.C.: Brookings Institution Press, 1993), chap. 3.

131 For the arguments in this paragraph, see Brendan Rittenhouse Green and Austin Long, “The MAD Who Wasn’t There: Soviet Reactions to the Late Cold War Nuclear Balance,” Security Studies 26, no. 4 (October 2, 2017): 631–36,

132 Brzezinski to Carter, December 2, 1977, Declassified Document Reference System CK2349136586, pp. 2–3.

133 See Green, The Revolution That Failed, chap. 8.

134 The next three paragraphs also draw on Green, “The President and Nuclear Weapons.”

135 See, e.g., O’Hanlon, “Going It Alone?”

136 This kind of rhetoric is everywhere, but for a recent book length exposition see: Elaine Scarry, Thermonuclear Monarchy: Choosing Between Democracy and Doom (New York: W. W. Norton & Company, 2014).

137 Trachtenberg, A Constructed Peace, 161.