Writing in the pages of the authoritative journal Foreign Affairs (Jan/Feb 2025), Robert E Kelly and Min-hyung Kim make a compelling case for why South Korea should get a nuclear weapon. At first glance, the arguments they make seem plausible. Accordingly, rambunctious North Korea is rapidly expanding its nuclear capabilities, hurling nuclear threats at South Korea with increasing vigour. While South Korea could in times past rely on the United States (US) nuclear umbrella, the North’s development of an intercontinental missile capability now puts the United States within range, thus casting doubt on the deterrent effects of the US nuclear arsenal. In short, with Pyongyang able to strike US cities, doubts arise about the credibility of the US nuclear umbrella. Such doubts are compounded, the authors contend, by fears in Seoul about the Trump administration, which has formally criticised the US alliance commitment to South Korea (and, one should add, several other alliance commitments), thus stoking fears that the US would be unwilling to expend American treasure and shed American blood in defence of South Korea if war erupted on the Korean Peninsula. The net effect of this set of arguments is that South Korea finds itself in an increasingly perilous security situation, one that can and should best be alleviated by arming with nuclear weapons. To strengthen their case further about US unwillingness to aid its alliance partners in the face of an attack by a nuclear adversary, Kelly and Kim ask us to consider the war in Ukraine. In this war, they note, Putin’s nuclear threats have constrained the US from putting boots on the ground in defence of Ukraine’s cause. While nuclear weapons can and do act to ensure the survival of those who possess them, there are, I contend, sufficient grounds for believing that South Korea does not need them.
The US – an unreliable alliance partner?
Kelly and Kim are correct in pointing out that fears about the reliability of the US as an alliance partner are rife in South Korea. Incidentally, such fears are not restricted to South Korea, with several US alliance partners over the last decade or so casting doubt on the reliability of US defence commitments. Although the US has no formal alliance commitment to Taiwan, concern has been brewing there for years about whether the US can be trusted to come to the island nation’s defence in the event of a conflagration with China. Japan has also at various times questioned the US security commitment. It is worth pointing out that concerns about the reliability of the US as an alliance partner are not confined to the present, but were often (though surely less than today) voiced during the Cold War. To assuage South Korea’s fears, as Kelly and Kim point out, the Biden administration often publicly voiced their commitment to the US-South Korean alliance. Kelly and Kim are less optimistic about whether the Trump administration can and will keep the alliance afloat. While Trump has complained about the costs of US security commitments to South Korea (and elsewhere), this hardly spells the end of the US defence commitment to Seoul. In the event of the termination of all US security commitments to South Korea, especially those relating to extended deterrence, South Korea would surely have grounds for going nuclear. Importantly, a drastic reduction in US defence spending towards South Korea does not inevitably imply an open door for North Korea to attack its neighbour. Where extended nuclear deterrence remains in place, South Korea can rest assured that its nuclear neighbour will be constrained from launching a large-scale attack.
Misunderstanding the nature and requirements of deterrence
Kelly and Kim’s case for a nuclear-armed South Korea rests on a fundamental misunderstanding of the nature and requirements of nuclear deterrence. The fundamental problem that Kelly and Kim identify is not that North Korea no longer sees the US security commitment to South Korea as valid or that the US no longer views its interests in that part of the world as strategically vital. Instead, their central argument is that Pyongyang’s ability to strike US cities undermines the stability and credibility of the US nuclear umbrella. They thus conclude that “North Korea’s nuclear capabilities undermine US deterrence”, believing that North Korea’s ability to strike US cities raises the costs of US intervention to excessively high levels. That the projected costs of any US involvement in a war on the Korean Peninsula increases in light of the North’s ability to strike US cities is, of course, true, but striking US assets and shedding US blood have been on the table for Pyongyang even prior to developing long-range missiles – US military bases and assets in South Korea have been in striking distance for a long time. More fundamentally, emphasising the costs to the US in the face of North Korea’s ability to strike American cities surely puts the problem the wrong way round. In Kelly and Kim’s depiction of deterrence logic, the US would be self-deterred from responding to North Korean nuclear action, owing to fears about the destruction of American cities. This misconstrues the nature of deterrence. Where the possible use of nuclear force exists, as the late Kenneth Waltz noted some time ago, we should not overemphasise the retaliator’s conceivable inhibitions while downplaying the aggressor’s obvious risks. Put differently, the clear and present risks confronting a North Korea bent on unleashing unacceptable destruction surely weighs heavier than possible US inhibitions about using nuclear force. In a conventional world, running risks based on the possible inaction of a state to use force is acceptable; in a nuclear world, such risks could mean national suicide.
Where a formal US-South Korea alliance commitment remains in place, North Korean bellicosity towards South Korea should not be feared. As is often the case, fears about the credibility of US deterrent threats, especially those relating to extended deterrence, abound, leading some to speculate that North Korea might not perceive US deterrent threats as all that credible. Fretting about the credibility of nuclear deterrence and deterrent threats is hardly new. Throughout the Cold War, and through to today, the credibility of nuclear deterrence was and is attached to all manner and sorts of things: the personalities of leaders, their rationality, the regime type, and whether leaders believe in the doctrine of nuclear deterrence. In a conventionally armed world, such considerations do greatly matter in affecting the likelihood of war and peace. A nuclear world, however, operates by a different logic. The credibility of nuclear threats – and deterrent threats in a nuclear as against a conventional world – derives from the particular qualities of nuclear weapons and the effects they produce. The nuclear weapon, as Bernard Brodie suggested several decades ago, constitutes the absolute weapon. They are able to work their deterrent effects irrespective of what other countries do. If North Korea believed that it could wipe out all US strategic warheads or construct a leak-proof missile defence, then Brodie’s depiction of the nuclear weapon as the absolute weapon would be seriously misleading. However, North Korea is hardly capable of destroying the US strategic missile arsenal, and no leak-proof missile defence system exists. Absent this reality, that deeply constraining force at the heart of nuclear deterrence – fear of retaliation – will continue to clarify North Korea’s calculations of war and constrain a possible course of aggression.
Moreover, the credibility of US nuclear threats rests on what it can do to another country, not on what it will do. Fears about North Korea believing – or hoping – that the US would sit idly by while the North launched a large-scale invasion of South Korea is wholly misplaced. What is required to deter, and what will ultimately constrain, Pyongyang, is the ability – or, more importantly, the appearance of the ability – to meet a course of unacceptable aggression with a retaliatory blow causing large-scale destruction. It matters little whether you can retaliate, as Waltz noted, as long as the aggressor state believes that you have the ability to do so. The threat of retaliation, Bernard Brodie noted, “does not have to be 100 percent certain; it is sufficient if there is a good chance for it. … The prediction is more important than the fact.” To put it differently, states in a conventional world can initiate war if they believe the possibility of success to be high; in a nuclear world, aggression is stymied where the aggressor believes retaliation is possible. What is required to deter is not certainty but uncertainty of response, because if retaliation occurs, an aggressor stands to lose so much. In a conventional world, uncertainty tempts states to initiate war; in a nuclear one, it produces hesitation, because the consequences of action are too dangerous to bear. Creating the belief that a nuclear state can retaliate against an aggressor’s military adventures is not as daunting as is often believed, with the construction of credible second-strike forces far easier than often thought. In the US-North Korea equation, there is little doubt that Pyongyang believes that the US indeed can retaliate.
Throughout the Cold War, observers often fretted about whether weak nuclear states – those with modest nuclear forces – could deter the strong. As it turned out, this was no problem at all. The weak can deter the strong. Strangely enough, today we are constantly worried not about whether the weak can deter the strong, but about whether the strong can deter the weak. Apparently, US fears about a few North Korean missiles falling on its cities renders it wholly constrained, while North Korea is wholly unconstrained by the US nuclear arsenal. This is foolhardy. Without a doubt, the strong can also deter the weak. Nuclear weapons bring caution and moderation, irrespective of who wields them and the comparative size of respective nuclear arsenals. Today, as before, the blatantly obvious risks confronting a nuclear would-be aggressor decisively outweigh the presumed inhibitions of the retaliator. Accordingly, I see no clear way in which North Korea’s nuclear capabilities “undermine US deterrence”, as Kelly and Kim aver, and I see little warrant for believing that a South Korean nuclear arsenal can deter the North but a secure US one cannot.
What’s Ukraine got to do with it?
Unlike Kelly and Kim, I see little relevance in drawing on the war in Ukraine to build a case for a South Korean nuclear deterrent. The two cases are not comparable. The US has no standing or formal alliance with Ukraine. This is not the case with South Korea, where a longstanding alliance is in place and where the US nuclear umbrella has for decades been a stabilising force in the region. Looking at Ukraine to conclude that the US will not act in defence of South Korea is therefore misplaced. That Washington and other Western states have been loath to put boots on the ground in Ukraine is true, with Putin’s nuclear sabre-rattling undoubtedly an important constraining force. A North Korea-South Korea conflagration would, however, unfold against the backdrop of US security guarantees and, by implication, its nuclear umbrella – an assessment that undoubtedly will weigh heavily on the mind of Kim Jong Un. Also, while there are indeed fears about the Trump administration abandoning decades-old alliance partners, such concerns are exaggerated. The general thrust of US foreign policy over the last decade or so – and the plea for urgent intervention from several US-based China experts – has been to take the fight in the unfolding Sino-US rivalry to China, which in the long run is likely to produce greater emphasis on the US sustaining strong defence relations with its partners in the region. Of course, this might not always mean that US alliance partners in the region get the defence support they deem necessary, but it does suggest that the US has every incentive to ensure the survival of its alliance partners. In a nuclear-armed world, modest but advanced conventional forces paired to the US nuclear umbrella should suffice to stave off would-be aggressors.
Where does this leave South Korea?
In the main, I find a great deal of agreement in Kelly and Kim’s case for a South Korean nuclear deterrent. In particular, where a state’s survival and security are threatened, the case for a nuclear deterrent becomes particularly strong. Nuclear weapons remain a particularly useful guarantee to deter a course of aggression by a rambunctious nuclear neighbour. However, nuclear weapons are a pain in the neck, and states should best steer clear of them if their security does not require them. Currently, and for the foreseeable future, South Korea’s external situation is not as dire as portrayed, with the US-South Korean alliance firmly in place. With the centre of international politics increasingly pivoting to Asia, we have warrant for believing that the US will invest more, not less, of its military and other assets in that part of the world. In investing more in that part of the world, the US will send a clear and powerful message to North Korea that it has clear and strategically vital interests in the region, thus further enhancing the credibility of its extended deterrent threats. The US nuclear umbrella is sufficient to ensure South Korean security. Absent such guarantees, I would be the first to urge South Korea to go nuclear and do it quickly.
Also read:
The future of Taiwan: strategic realities and the nuclear option