NUCLEAR WEAPONS LAW AND THE KOREAN NUCLEAR CONUNDRUM

PETER HAYES
OCTOBER 25, 2026
I. INTRODUCTION
Peter Hayes analyses the applicability of international law to nuclear threat-making, nuclear use, and nuclear proliferation to the Korean peninsula. He concludes that international law cannot by itself stop nuclear threat-making and nuclear use but that international law “supplies the vocabulary and instruments from which a settlement could be built, and it defines the limits on the use of force that every party must observe while that settlement is pursued.” Consequently, he argues, it “behooves nuclear weaponeers, nuclear decision-makers, parliamentarians and their aides, journalists and scholars, civil society organizations, and interested individual citizens—in Seoul, Pyongyang, Washington, Tokyo, Beijing and Moscow—to understand how international law applies to and constrains the use of nuclear weapons.”
This essay was published in an abridged version by Korea Times here and the Asia Pacific Leadership Network here
This essay is published under a 4.0 International Creative Commons License the terms of which are found here
The views expressed in this report do not necessarily reflect the official policy or position of the Nautilus Institute. Readers should note that Nautilus seeks a diversity of views and opinions on significant topics in order to identify common ground.
Acknowledgement: AI was used to fact-check this essay but it is wholly the work of the author.
Image: Peter Hayes generated by Perplexity AI Computer model.
II. NAPSNET BLUE PETER ESSAY BY PETER HAYES
NUCLEAR WEAPONS LAW AND THE KOREAN NUCLEAR CONUNDRUM
OCTOBER 25, 2026
War is politics by other means; and international law is the codification of the practices and legal doctrines of great powers that define the limits on the use of “other means” by states pursuing political ends in endless, ruthless competition. Nowhere is this more apparent than on the Korean Peninsula, where a war suspended by the 1953 Armistice has never formally ended, and where the forces and treaty commitments of four nuclear-armed states—the United States, China, Russia and the DPRK—converge.
“Other means” in today’s world refers above all to nuclear weapons, the ultimate form of coercive force due to the absolute, annihilative ability they confer on a possessor to project an existential threat against an adversary. In Korea, the prime adversaries—the United States and the DPRK– describe their deterrence in precisely these terms.
The DPRK’s 2022 law on its nuclear forces policy declares that their mission is to make hostile forces understand that military confrontation with the DPRK “brings about ruin.” The United States’ 2022 Nuclear Posture Review states that “there is no scenario in which the Kim regime could employ nuclear weapons and survive,” and the 2023 Washington Declaration promises that any DPRK nuclear attack on the Republic of Korea (ROK) “will be met with a swift, overwhelming and decisive response.”
Since the end of World War II, thousands of lawyers working for states and as independent scholars, together with the UN General Assembly, the UN Security Council, the International Court of Justice and dozens of international agencies with legal remits, have developed a corpus of international law that addresses a vast range of circumstances in which military force may be used, including nuclear weapons—including in Korea.
Much of this law has customary status and applies to all states, whether or not they have joined the relevant treaties. This matters in Korea. The DPRK announced its withdrawal from the Nuclear Non-Proliferation Treaty in 2003 and has been under UN Security Council sanctions since its first nuclear test in 2006, but it remains bound by the UN Charter and by the customary laws of armed conflict—as are the ROK, the United States and every other state that might fight in Korea. Both Koreas are also parties to Additional Protocol I to the Geneva Conventions; the United States is not, one of many legal asymmetries within the alliance and across the Demilitarized Zone (DMZ).
The vast bulk of the world’s nuclear weapons are in the arsenals of the United States (~5,040) and Russia (~5,420), followed by China (~620) a distant third, and six smaller nuclear-armed states—the United Kingdom, France, India, Pakistan, Israel and the DPRK—that account for the rest (~1,100). Four of the nine nuclear-armed states are tied to the peninsula by forces or treaty: the DPRK itself; the United States through its 1953 Mutual Defense Treaty with the ROK; China through its 1961 treaty of alliance with the DPRK; and Russia through its 2024 Comprehensive Strategic Partnership treaty with Pyongyang, which commits each party to provide immediate military assistance if the other is attacked.
Each of these states declares that it observes international law, especially the laws of armed conflict, which apply directly to nuclear weapons. At the same time, international law is conflicted when it comes to the threat or use of nuclear weapons if and when a state is threatened existentially. In its 1996 advisory opinion, the International Court of Justice could not “conclude definitively whether the threat or use of nuclear weapons would be lawful or unlawful in an extreme circumstance of self-defence, in which the very survival of a State would be at stake.” The DPRK’s nuclear law occupies exactly this space: its catch-all condition for nuclear use is a “catastrophic crisis to the existence of the state and safety of the people.”
Each of these words is critically important in Korea. What exactly is a threat, when is a threat illegal under international law, and when do nuclear threats cross that line? Are the DPRK’s drills simulating tactical nuclear strikes on South Korean military targets unlawful threats? Are allied exercises that Pyongyang denounces as rehearsals for invasion and “decapitation,” or the visible deployment of US nuclear-capable bombers and ballistic missile submarines to the peninsula, lawful signals of deterrence—or threats in their own right?
When exactly is a state threatened “existentially,” and who decides? Is a threat to the leadership of a state enough for it to be under existential threat? In Korea this question is not academic. The ROK’s “Korea Massive Punishment and Retaliation” concept contemplates striking the DPRK leadership, while the DPRK’s 2022 law authorizes nuclear use when a nuclear or non-nuclear attack on “the state leadership and the command organization of the state’s nuclear forces” is launched or judged imminent. Does existential threat refer to the state’s capacity to exist and to defend itself with its own military means, or does it encompass the population of a state and its ability to recover? And may a regime lawfully equate its own or its supreme commander’s survival with the survival of the state?
Does a threat that is specific as to time and likely place and is made by a senior figure exercising control authority over nuclear weapons combined with observed and non-routine movement of nuclear weapons constitute a pending nuclear attack under international law? What confidence is required that such an attack is about to occur or is actually underway given that North Korean nuclear-capable missiles can reach Seoul in minutes and American missiles can hit Pyongyang in less than half an hour?? Does the size of a nuclear attack relative to the targets that must be destroyed based on military necessity have to reach a threshold in order to threaten the existence of a state and thereby be a legally justified basis for pre-emptive strikes or a retaliatory nuclear counter-attack? The DPRK’s law permits nuclear use once an attack is judged to have “drawn near,” while the ROK’s “Kill Chain” concept rests on striking DPRK missiles before they are launched. Each side’s doctrine of anticipatory action potentially creates the other side’s evidence of an impending attack. Does international law countenance such postures?
Must such attacks be conducted with nuclear weapons for a nuclear response to be legal? Or can nuclear weapons be used to counter a conventional military threat or attack? In Korea the question cuts both ways. The DPRK reserves the right to use nuclear weapons against a “fatal military attack against important strategic objects of the state” and to take “the initiative in the war”—in effect, to offset the alliance’s conventional superiority. The United States, for its part, has declined to adopt a no-first-use policy, and retains the option of nuclear use in extreme circumstances to defend the vital interests of its allies, including the ROK.
What about non-military forms of attack, such as cyber-attacks or AI-enabled attacks aimed at crippling military forces, sowing terror among civilians or crippling critical infrastructure such as water, communications or public health systems with resultant mass casualties—domains in which the DPRK has demonstrated sophisticated capabilities?
What about attacks aimed at mass destruction by disavowable but state-backed autonomous non-state actors, whose attacks may be aimed at catalysing wars between nuclear-armed states, or may simply be nihilistic or apocalyptic in motivation? In Korea, attribution of an attack may not be simple or fast enough to stop a war.
And once nuclear weapons have been used, may the normal rules of the laws of armed conflict—that attacks, including retaliation, may not target civilian populations or civilian objects such as hospitals, and must be proportionate when they strike military facilities located in cities—be set aside on grounds of military necessity? In Korea, command centres in both Seoul and Pyongyang sit amid millions of civilians, and major US and ROK bases adjoin South Korean cities. Additional Protocol I, to which both Koreas are party, prohibits reprisals against civilians; the United States does not accept that this prohibition binds it.
In planning for nuclear wars, how proportionate should nuclear strikes be, and what metrics should be used to determine the scale of the likely impacts of nuclear detonations? The United States, for example, uses direct blast overpressure and direct radiation effects to estimate the casualties caused by specific nuclear weapons at a given yield, height of burst and so on, but ignores the effects of firestorms ignited by nuclear explosions in cities. Yet by our calculations in our Northeast Asia nuclear use study with colleagues at Nagasaki University, we found that firestorms can increase casualties by at least 10–30 percent in many instances, possibly more.
What about the other states? Is ignoring firestorms legal in determining the proportionality of a planned nuclear strike? In Korea, where about half of South Korea’s population lives in the Seoul Capital Area and Pyongyang concentrates the DPRK’s elite and its command, ignoring firestorms could grossly understate civilian harm or lead to manifestly disproportionate attacks by using more and bigger nuclear weapons than could be justified.
What about nuclear decision-making? Ignorance is no defense under international law. Yet there are no international rules (yet) on how many officials, and who, should be involved in a decision to use nuclear weapons, nor any norms as to what checks and balances should be in place before nuclear strike orders are issued. In Korea, the contrasts are stark. The DPRK’s 2022 law vests “all decisive powers concerning nuclear weapons” in the President of the State Affairs Commission. Its revised March 2026 constitution permits delegation of nuclear command to a “State Nuclear Forces Command Organization.” The US president retains sole authority to order the use of US nuclear weapons, including in defence of the ROK. For its part, the ROK has no nuclear weapons of its own, now joins in nuclear planning through the US–ROK Nuclear Consultative Group in planning how ROK forces may support US nuclear forces in targeting the DPRK—but it has no veto and no decision-making authority whatsoever over the disposition or use of nuclear weapons in Korea.
Mistakes happen. Organizations make errors, sometimes grotesque, even incredible errors, due to cybernetic pathologies. Systems fail, sometimes catastrophically at the worst possible moment, and often when multiple, individually improbable failures occur at the same time in ways that cannot be anticipated in advance. Nuclear forces are no exception—least of all on a peninsula where opposing forces operate at short range, at high readiness and with only minutes of warning.
Because such accidents are normal, redundant backup systems should be in place, and crisis communications systems between adversaries should be in place to manage such failures before they become perceived threats of pending or actual attack. International law strongly suggests that all states armed with or dependent on nuclear weapons should implement backup and recovery systems to avoid needless risk-taking and potential escalation. In Korea, the DPRK has not answered the inter-Korean liaison and military hotlines since April 2023. The main routine channel that remains is the UN Command line to the Korean People’s Army at Panmunjom. On October 7, South Korea’s defence ministry said it would seek inter-Korean military talks to prevent accidental clashes and to reopen the hotlines. Are the parties ready and should a dedicated nuclear hotline between the DPRK and the United States be established?
International law also requires nuclear commanders to be well-informed, and their respective states to make every effort to avoid the use of nuclear weapons—first by avoiding conflict in the first place, and then by using non-nuclear forces to achieve military goals in all but the most extreme of circumstances. In Korea, where the ROK–US alliance commands formidable conventional forces, that obligation weighs heavily on any decision to cross the nuclear threshold. It also sits uneasily with a DPRK doctrine that contemplates early nuclear use to seize the initiative in a war.
Nuclear commanders and nuclear weaponeers at every level are accountable under international law to ensure that nuclear weapons are never used inadvertently, and, if they are used deliberately, that doing so complies with international law. This imperative applies equally to US and ROK officers in the Combined Forces Command; and to DPRK officers charged with executing a law that provides for a nuclear strike to be launched “automatically and immediately” if the DPRK’s nuclear command and control system is endangered by attack.
Under international law, all military and civilian personnel are under an obligation to disobey manifestly illegal nuclear strike orders. How exactly they would do so in the modern era is an open question when supreme commanders on all sides of the Korean conflict often make public statements that suggest they are prone to ignore international law.
The laws of war apply to all parties to the Korean conflict in complicated and often counter-intuitive ways. All the world’s militaries—especially the nuclear-armed militaries—pay careful attention to the constraints of international law. The United States military and the UN Command in particular invoke the laws of armed conflict in their preparations to wage war in the cause of deterrence against the DPRK’s nuclear force. The Armistice Agreement itself remains a legal instrument and various incidents related to landmine explosions south of the Military Demarcation Line this year are active UNC concerns.
North Koreans for their part also study international law and craft their doctrines with great care, emulating the legal manoeuvres and evasions of the nuclear-armed great powers to justify their own nuclear weapons doctrine and practices. The DPRK’s 2022 law, for example, mirrors the negative security assurances of the recognized nuclear-weapon states, pledging not to threaten or use nuclear weapons against non-nuclear-weapon states unless they join aggression against the DPRK “in collusion with other nuclear weapons states”—a carve-out that plainly encompasses the ROK and Japan.
In addition to providing a narrowly circumscribed and legally justified way to use nuclear weapons, international law also offers opportunities it to communicate, coordinate and even collaborate on how best to minimize the risk of nuclear use in Korea. States have shared best practice in other conflict zones, for example, on control technology, and explored common definitions and especially the translation of key legal terms in advance of a crisis.
It therefore behooves nuclear weaponeers, nuclear decision-makers, parliamentarians and their aides, journalists and scholars, civil society organizations, and interested individual citizens—in Seoul, Pyongyang, Washington, Tokyo, Beijing and Moscow—to understand how international law applies to and constrains the use of nuclear weapons.
This is especially the case now that inter-Korean relations are in transition. The DPRK has declared that reunification is a dead issue and that the two Koreas are now separate, “normal” states bound by international law and, by implication, no longer by prior inter-Korean “political” agreements that are not international legal instruments. Its revised constitution of March 2026 deleted its reunification provisions and defined its territory as bordering the ROK to the south. This shift strips away what little standing was left of the 1992 Joint Declaration on the Denuclearization of the Korean Peninsula in which both Koreas pledged not to test, manufacture, produce, receive, possess, store, deploy or use nuclear weapons, and which Pyongyang declared void in January 2013. If denuclearization is to be revived, it will have to rest on legally binding, verifiable and reciprocal international instruments rather than on inter-Korean political pledges.
All this is happening in an era when the legal foundations of world order since World War II are under immense stress. Existing international law is a crucial foundation on which to stand while constructing a new international order over the coming decades. It must also be reformed and updated to accord with new geopolitical circumstances that will demand legally binding agreements and set new constraints on the exercise of military force by states. In Korea, that agenda includes a peace treaty to replace the Armistice, legally binding security assurances to the DPRK as well as to the ROK and Japan, and, ultimately, a verified nuclear-weapon-free Korean Peninsula—perhaps within a regional nuclear-weapon-free zone. With President Trump signalling that he hopes to meet Kim Jong Un before the end of the year, Seoul seeking to revive military talks, and Pyongyang telling the UN General Assembly that its status as a nuclear weapons state “can never be reversed,” international law will not by itself denuclearize the DPRK. But it supplies the vocabulary and instruments from which a settlement could be built, and it defines the limits on the use of force that every party must observe while that settlement is pursued.

III. NAUTILUS INVITES YOUR RESPONSE
The Nautilus Asia Peace and Security Network invites your responses to this report. Please send responses to: nautilus@nautilus.org. Responses will be considered for redistribution to the network only if they include the author’s name, affiliation, and explicit consent