Nuclear weapons by themselves can say a lot. They may deter aggression, for example, through their simple existence, generating a “threat that leaves something to chance,” as Thomas Schelling famously put it. Sometimes the less said about them the better: This leaves adversaries guessing what may trigger their use. So why do states bother declaring why they have nuclear weapons or when they might use them? In the case of the United States, at least, nuclear weapons do more than deter adversaries — they should also reassure allies about America’s commitment to extending deterrence to them and assure the world that the United States is a responsible steward of nuclear weapons. As such, when U.S. government officials issue statements about the role or employment of the country’s nuclear arsenal — what’s known as nuclear declaratory policy — they are attempting to signal to adversaries, allies, and the rest of the world the role that nuclear weapons play in American security policy, and when they may potentially be employed. Rather than simply relying on an unstated threat that leaves something to chance, the United States broadly outlines when it might consider making such threats, and to what ends, in the first place. Although declaratory policy may sometimes be derided as irrelevant — adversaries care more about what America can do with nuclear weapons than what it says about them — the fact is that allies care a lot about what the United States says about its nuclear weapons, because their very existence may depend on the American pledge to use nuclear weapons in their defense. Given this, it is important to get declaratory policy right.
Become a Member
There is likely to be a lively and contentious debate within and outside of the administration of President Joe Biden on this particular aspect of U.S. nuclear policy in the coming months. Momentum to narrow the declared role of nuclear weapons in American security strategy is high on the agenda. Given significant American conventional capabilities and advantages, there are few, if any, realistic scenarios where the United States would consider using nuclear weapons first in a conflict. Some progressive members of Congress have in fact proposed that this reality become official declaratory policy, and that the United States declare a “no first use” pledge: that it would not be the first to use nuclear weapons in a conflict, no matter what the circumstances, reserving them strictly for retaliating after the United States or its allies had suffered a nuclear attack. Not only has that been met with skepticism from adversaries such as Russia and North Korea, who would doubt the sanctity of any such pledge in a crisis, but it makes certain allies — notably Japan — exceptionally nervous, as they depend on at least the possibility that the United States may use nuclear weapons first to stave off a conventional attack against them.
Biden, both as vice president and as a presidential candidate, proposed an alternative nuclear declaratory formulation known as “sole purpose”: that the sole purpose of U.S. nuclear weapons is to deter nuclear use against it or its allies. Is this the same thing as a no-first-use pledge? Proponents of a no-first-use declaration hope, and allies fear, that it may be.
But a sole purpose declaration need not be exactly or tantamount to a no-first-use pledge. Fundamentally, a no-first-use declaration is an explicit ex ante constraint on the employment of nuclear weapons, whereas sole purpose is statement about why the United States possesses nuclear weapons, without necessarily imposing constraints on their use. As always, however, the devil is in the detail. For instance, there are sole purpose formulations that leave enough room for the United States to use nuclear weapons preemptively or first, in the event of extreme and unforeseen non-nuclear attacks against it or its allies. Because declaratory policy purports to describe what an administration or president — who still retains the sole authority to use those weapons — thinks about the role of nuclear weapons, it can have powerful stabilizing or destabilizing effects in peacetime or crises.
After President Donald Trump’s cavalier rhetoric questioning why the United States “couldn’t use its nukes,” the Biden administration has an opportunity to re-establish sobriety in American declaratory policy. We argue that a new declaratory policy that simply states that the “sole purpose of the U.S. nuclear arsenal is to deter nuclear attacks against the United States and its allies” can meaningfully de-emphasize the role of nuclear weapons in American security strategy — reflecting the reality that they are weapons of extreme last resort — without undermining the robustness of extended deterrence commitments. Instead of adopting a no-first-use policy, which may lack credibility absent broader force structure changes that are not feasible and may not be desirable in the near term, the president should follow his instincts and adopt a sole purpose declaration. Precisely what wording he uses, however, will make all the difference.
Talking about the Role of U.S. Nuclear Weapons
What is the stated role of nuclear weapons in U.S. national security strategy today? For much of the Cold War, the purpose of American nuclear weapons was broad and central to the country’s grand strategy: to deter nuclear and conventional attacks against the United States and its allies in Europe and East Asia. This required that the United States leave open the possibility — and, indeed, generate the risk — that it would use nuclear weapons first if faced with an overwhelming conventional attack. But even by the late Cold War, massive, survivable strategic nuclear arsenals, where any nuclear use could lead to mutual suicide, had already spurred thinking on the practical purpose of the U.S. nuclear arsenal, at least with respect to the Soviet Union. Writing in 1983, Robert McNamara observed, for instance, that when it came to U.S. nuclear weapons, their “sole purpose, at present, is to deter the other side’s first use of its strategic forces.”
At the end of the Cold War, American nuclear forces adapted to new challenges, including a focus on so-called “rogue nations” such as Iraq and North Korea, but retained an expansive view of the types of threats — both nuclear and non-nuclear — that ought to be deterred by nuclear weapons. In the 2010 Nuclear Posture Review, the Obama administration laudably attempted to de-emphasize the role of nuclear weapons in American security strategy as part of a broader effort to reduce the centrality of nuclear weapons in crises and to lessen risks of miscalculation, declaring that the “fundamental role of U.S. nuclear weapons … is to deter nuclear attack on the United States, our allies, and partners.” This language was aspirational and, insofar as declaratory policy was concerned, it narrowed relevant contingencies to little more than deterring enemy nuclear attack — coming close to but stopping short of making this mission the “sole purpose” of U.S. nuclear weapons. And it had a limited effect on actual nuclear employment options.
At the very end of the Obama administration, the notion of further narrowing the declared role of American nuclear weapons was resurrected. In January 2017, Biden gave a wide-ranging speech on nuclear security, a topic he has worked on for decades and is seen as holding relatively progressive views on. In that speech, he declared: “Seven years after the Nuclear Posture Review charge — the President and I strongly believe we have made enough progress that deterring — and if necessary, retaliating against — a nuclear attack should be the sole purpose of the U.S. nuclear arsenal.” Biden reiterated his interest in such a declaration in Foreign Affairs as a presidential candidate, pledging to “work to put that belief into practice, in consultation with the U.S. military and U.S. allies.” This language was adopted in the Democratic Party’s official 2020 platform.
The Trump administration, however, had different ideas and the 2018 Nuclear Posture Review declared that the role of American nuclear weapons explicitly included the deterrence of “significant non-nuclear strategic attacks,” leaving the definition of what constitutes a “significant” or a “non-nuclear strategic attack” intentionally ambiguous — carving out a role for nuclear weapons more expansive than in the 2010 review. The Biden administration now has an opportunity to revisit various facets of U.S. nuclear policy. In addition to the nuclear modernization program and questions about presidential sole authority, we should now expect serious consideration of a sole purpose declaration.
An Enduring Debate: Words That Matter
Many arms control advocates are attracted to the sole purpose declaration because it is close to a no-first-use declaration, and some have argued the two are effectively the same. Some Democrats, such as Sen. Elizabeth Warren and Rep. Adam Smith, have even introduced legislation proposing the adoption of a no-first-use policy. Advocates of an American no-first-use policy argue that such a declaration would set a model for the world, reducing the role of nuclear weapons in global security and easing pressures on nuclear postures during peacetime, while lowering the risk of inadvertent nuclear use in a crisis.
A no-first-use declaration is unambiguous: A country pledging no first use conveys to its adversaries that it will not under any circumstances use any nuclear weapons before it has suffered a nuclear attack itself. Whether a no-first-use declaration allows for a state to launch nuclear weapons once the other side has launched (launch-under-attack) or is imminently judged to be launching nuclear weapons (preemption) is ambiguous, but an absolute no-first-use policy restricts the role of nuclear weapons to one, and only one, scenario: retaliating against a nuclear attack.
Few countries maintain a no-first-use policy because of the difficulty of making such a pledge credible. China is the only country to maintain such a policy without any explicit reservations or qualifications. It has done so since 1964, and for decades undertook costly restraining measures — such as keeping nuclear warheads de-mated from most of its missile force — to make its pledge about as credible as possible (and yet the United States is still skeptical of its commitment). India, meanwhile, also includes a no-first-use declaration in the public version of its 2003 nuclear doctrine, but carves out a major exception: “in the event of a major attack against India, or Indian forces anywhere, by biological or chemical weapons, India will retain the option of retaliating with nuclear weapons.” This is not and has never been a real no-first-use policy. India has also steadily eroded its no-first-use declaration over conventional attacks and preempting nuclear attack over the past two decades as security dynamics, with Pakistan in particular, have evolved.
The remaining seven nuclear states, including the United States, do not maintain “first use” policies per se, but simply reserve the right to use nuclear weapons first under certain circumstances, which are described with varying degrees of ambiguity in national doctrines and statements. Per the 2018 Nuclear Posture Review issued by the Trump administration, the United States reserves the right to use nuclear weapons “in extreme circumstances to defend the vital interests of the United States, its allies, and partners” (language nearly identical to the 2010 Obama administration review). This is consistent with the long-running U.S. policy of calculated ambiguity: Instead of spelling out exactly what “extreme circumstances” mean in practice, the matter is left to the imagination. The reasoning behind this is ultimately to minimize the possibility of commitment traps while simultaneously maximizing the deterrent potential of U.S. declaratory policy: In a crisis, as an adversary chooses to escalate, U.S. planners hope that they would remain uncertain whether their action might amount to what the United States considers “extreme” and, subsequently, whether the United States might respond with nuclear weapons. In short, the United States explicitly reserves the right to presently use nuclear weapons first.
Detractors of a no-first-use policy in the United States argue that without drastic changes to American nuclear posture or alert levels to preclude at least rapid first use to make such a pledge credible — such as separating warheads from intercontinental ballistic missiles, as China is believed to do, or eliminating them altogether because they can be launched so promptly — the costs significantly outweigh the benefits. The military strongly opposes those changes, however, since they may reduce the overall survivability of U.S. nuclear forces. Furthermore, absent these practical changes to force posture — or even with them, for that matter — U.S. adversaries would never believe a no-first-use pledge because nothing would physically prevent the United States from violating it in a crisis or conflict.
Meanwhile, American allies might find such a declaration too credible. For allies, such as Japan, the fear that the United States may abandon them at the most crucial moment and fail to use nuclear weapons to defend them may drive them to develop their own nuclear weapons, as detractors often note. Allied fears have intensified in recent years in Northeast Asia in particular, where rapid advances in North Korea’s nuclear capabilities have introduced the Cold War specter of “decoupling” to the U.S. alliances with both South Korea and Japan. Furthermore, for Japan if not South Korea, the prospect of U.S. nuclear use to deter large-scale Chinese conventional aggression is a crucial pillar of its security strategy. They have therefore strongly opposed any movement toward an American no-first-use declaration. A no-first-use pledge may similarly cause concern in European capitals that Moscow would have a free hand to use overwhelming conventional force against them. A final argument against a strict no-first-use policy is that it would require the United States or its allies to suffer a nuclear attack — and countless fatalities — before retaliation. The risk of a no-first-use declaration, opponents therefore argue, is that not only would it fail to generate crisis stability against adversaries and incentivize allies to seek their own nuclear weapons, but that it is also immoral — even if U.S. retaliation against any aggressor were assured. Even if there are very few scenarios at present where the United States would ever contemplate using nuclear weapons first, skeptics of no-first-use argue that some scenarios do exist even today, and that others may arise in the future.
In short, U.S. adoption of a no-first-use pledge in the near term would likely be highly controversial.
Sole Purpose Is Not No First Use
The constituencies in allied states that have vociferously objected to a no-first-use U.S. policy view a sole purpose declaration as effectively tantamount to one. It was these precise concerns that kept the Obama administration from ultimately adopting a sole purpose declaration. The administration, at the eleventh hour, deemed that the “conditions” for a sole purpose declaration were not present in 2016. Furthermore, in the present environment in East Asia, the challenge of sustaining extended deterrence is more — not less — difficult than it was then, largely because of the Trump administration’s behavior toward allies. The allies may profess an even stronger allergy to sole purpose today than they did in 2016.
But is sole purpose equivalent to a no-first-use declaration, as so many have argued? Not quite. Even in its most stringent formulation, a sole purpose declaration is not equivalent to a no-first-use pledge — it comes close, but is not the same thing. No first use is a statement about when the United States would (and would not) use nuclear weapons. It is an explicit employment constraint: It commits a state to not use nuclear weapons except in retaliation for nuclear attacks. Sole purpose, in contrast, is as its name implies a statement about why the United States possesses the nuclear arsenal that it does, not how it will use it. It does not, in extremis, impose employment constraints as a no-first-use policy might. Rather, it explicitly de-emphasizes the role of nuclear weapons in overall U.S. national security strategy.
A person can possess a car for what she declares to be the sole purpose of driving to work, but if one day she has to drive to the emergency room, nothing will stop her from using the car for that purpose. A no-first-use pledge, by contrast, explicitly declares ex ante that the car will never be used to drive to the emergency room. Sole purpose stops well short of that.
As such, a meaningful sole purpose declaration can be constructed that is not, in fact, tantamount to a no-first-use declaration — one that simultaneously de-emphasizes the role of nuclear weapons in American security strategy without eroding the robustness of extended deterrence. The search for an alternative formulation to a no-first-use declaration is itself informative — if the administration wanted to declare a no-first-use policy it could simply attempt to do so. We argue that, instead, an appropriately crafted sole purpose declaration could help to realize the president’s stated vision on nuclear weapons without unduly jeopardizing U.S. alliances. Allies, too, once fully consulted, should be ready to avoid a knee-jerk response to policy shifts — especially if U.S. declaratory policy continues to account for their interests.
Whereas a no-first-use declaration is relatively straightforward, sole purpose maintains some of the traditional ambiguity in U.S. nuclear declaratory policy. How it does so depends on the precise formulation. The amount of daylight between these various formulations and “we will not use nuclear weapons first” varies from a sliver to a bay window. Despite this, some formulations do come very close to a no-first-use pledge and might understandably cause concern in allied capitals.
Consider the three following alternative sole purpose declarations:
Sole Purpose 1: The sole purpose of the U.S. nuclear arsenal is to deter — and, if necessary, to retaliate against — a nuclear attack against the United States and its allies.
Sole Purpose 2: The sole purpose of the U.S. nuclear arsenal is to deter nuclear attacks against the United States and its allies.
Sole Purpose 3: The sole purpose of the U.S. nuclear arsenal is to deter significant strategic attacks, including nuclear attacks, against the United States and its allies.
Each of these formulations stipulates what the “sole purpose” of the U.S. nuclear arsenal might be, but the practical consequences of the adoption of each declaration by the United States are very different. The closest to a no-first-use policy would be option 1 — the phrasing largely adopted by Biden before entering office. The addition of the embedded clause — “and, if necessary, retaliating against” — clearly stipulates that nuclear weapons are not envisioned for first use, perhaps not even to preempt imminent nuclear use against the United States or its allies. The first statement may, however, still allow for the possibility of launching-under-attack — a scenario where the United States would release nuclear weapons before any incoming warheads detonated, but upon confirmation with a high degree of confidence that an adversary had initiated a nuclear launch against American or allied soil. (While practicable with China and Russia, short missile flight times to South Korean and Japanese soil from North Korea may render this impractical in Northeast Asia.) But by including a clause about possessing nuclear weapons for the sole purpose of retaliating following nuclear use against the United States or its allies, option 1 turns entirely on litigating what point in the process of an adversary launching nuclear weapons constitutes “use” — a very narrow set of circumstances to begin with.
The second sole purpose statement, by omitting this embedded clause about retaliation however, leaves substantially more room for U.S. allies to be reassured, leaving the “chance” of first use on the table.
Take a realistic example to illustrate the distinction between the first and second statements, relevant especially to U.S. allies in Seoul and Tokyo. Imagine that North Korea has begun to disperse nuclear warheads to ballistic missile operating areas. U.S. and allied intelligence believe North Korean leader Kim Jong Un is in the early stages of a preemptive nuclear attack. This attack may be the result of a miscalculation by North Korea, or it may be a response to an allied attempt to attack or invade. The specifics of why Kim is readying nuclear missiles are unimportant. Under existing U.S. nuclear policy, the fact of North Korean launch preparations may not trigger the use of nuclear weapons, but U.S. allies could reasonably expect that a U.S. president would at the very least contemplate the preemptive use of nuclear weapons to maximize the probability that North Korean missiles are successfully destroyed. The United States could consider employing precision-guided conventional munitions, but the task of finding, fixing, and finishing North Korean missile launchers may be too difficult and risky. An American president consulting with military advisors may be told that the prospects for disarming North Korea are significantly higher if he or she uses nuclear weapons to do so, rather than conventional weapons.
In this scenario, the first statement leaves little wiggle room. Under a strict interpretation, only after the launch of the first North Korean nuclear warhead — or even more strictly, after the first detonation — on allied soil could the United States then contemplate nuclear use. Such a scenario, along with concerns about China’s growing conventional missile arsenal, is largely why Tokyo has strongly opposed sole purpose in the past. The second sole purpose formulation, however, leaves the matter open. If the sole purpose of U.S. nuclear weapons is to deter nuclear attacks, then it follows that the president could threaten nuclear first use in a crisis to deter an adversary from following through on preparations to launch nuclear weapons. Critically, unlike a no-first-use pledge and sole purpose option 1, the second sole purpose formulation does not make any statement about the conditions under which the United States would use nuclear weapons. Instead, it merely describes the purpose of the U.S. nuclear arsenal, leaving a wider scope of employment, including first use, if it were in the service of “deterring nuclear attacks” against the United States or its allies. Instead of constraining use, option 2 explains to allies and adversaries alike that the role of nuclear weapons and nuclear threats is limited to nuclear deterrence, which may include threats of preemption or even broader use options that seek to deter nuclear use. It is these latter possibilities that distinguish the second sole purpose statement from the first.
The third sole purpose formulation, meanwhile, expands the bounds further, allowing a U.S. president to deploy nuclear weapons against “significant strategic attacks,” which may include the use of biological or chemical weapons, or even a major conventional attack. This option is, essentially, status quo U.S. nuclear policy reformulated in the language of sole purpose and, indeed, includes multiple expansive purposes for American nuclear weapons.
No sole purpose declaration — nor a no-first-use declaration — by itself is likely to convince adversaries such as Russia, China, or North Korea that the United States would not or could not use nuclear weapons first in a crisis or conflict. The simple fact is that no matter what the United States says about restricting the role of nuclear weapons, adversaries will remain skeptical that Washington will adhere to any such strictures because nothing physically prevents the United States from violating them. The concerns they have today will thus likely persist irrespective of American declaratory policy absent drastic force structure and procedural changes that are unrealistic in the foreseeable future.
The sole purpose debate primarily concerns U.S. allies, and this is where the precise formulation matters the most. Even if the Biden administration chooses to adopt the narrowest possible sole purpose declaration (option 1) — the one proposed by Biden himself for several years now — there may be just enough of a sliver of daylight between it and a no-first-use declaration to assure allies that even if the envisioned purpose of American nuclear weapons is to retaliate against nuclear use against the United States or its allies in extremis, it does not explicitly commit the United States to not using nuclear weapons first in extreme unforeseen circumstances. This may be too clever by half, however. And it is unlikely to alleviate abandonment concerns in Seoul and Tokyo.
As such, the second formulation may square the circle. By declaring simply that “the sole purpose of U.S. nuclear weapons is to deter nuclear attack on the United States, our allies, and partners,” the United States can meaningfully de-emphasize the role of nuclear weapons in its security strategy without undermining the robustness of its extended deterrence commitments. It is not a no-first-use declaration — and allies would have a harder time believing or arguing that it is — but it declares and states the reality that the United States currently possesses nuclear weapons solely — not primarily or fundamentally, but solely — to broadly deter nuclear attack on itself and its allies. And it leaves just enough ambiguity about how the United States may do so and against what threats to avoid eroding primary or extended deterrence. This formulation does not constrain U.S. nuclear employment options, but it assures the world — adversaries and allies alike — that the United States would only ever use nuclear weapons in the most extreme of circumstances. Advocates of American alliances, extended deterrence, and a more restrained U.S. nuclear posture alike should welcome such a declaration. And now-President Biden should deliver on the vision he sketched out first as vice president, and later as a presidential candidate: The sole purpose for American possession of nuclear weapons is to deter nuclear attack against the United States and its allies.
Become a Member
Ankit Panda is the Stanton Senior Fellow in the Nuclear Policy Program at the Carnegie Endowment for International Peace. Vipin Narang is associate professor of political science and member of the Security Studies Program at the Massachusetts Institute of Technology.