I spoke at the Asian Leadership Conference in Seoul a few weeks ago on S Korean/Japanese indigenous nuclearization and then published my basic thinking with Foreign Policy magazine on the topic a few days later.
Both of the venues required a more abbreviated presentation for time/space constraints, so I thought I would put up my full remarks here, at my own site. Here is the 2022 ALC site, and here is my original article for FP.
In brief, my argument is that the US should get out of the way to let Seoul and Tokyo make up their own mind. The US has long opposed ROK/Jpn nuclearization, but increasingly that strikes me as inappropriately hegemonic or strong-arming of them. There is a pretty strong case for SK and Japan to counter-nuclearize against China, Russia, and especially NK. I sketch that in detail after the jump, but the short version is:
1. The US is not going to exchange LA for Seoul/Tokyo. In 1961, de Gaulle asked JFK would he exchange NY for Paris. JFK waffled; de Gaulle was no idiot; he built French nukes shortly afterwards. The logic is the same here. The US is not going to fight a nuclear war solely for non-Americans. This will raise endless, irresolvable credibility debates between the US and its Asian allies. The best way to resolve that is to do what our European allies did – self-insure through indigenous nuclearization.
2. Trump will likely get elected – or ‘elected’ – in 2024, and he will ‘blow up’ the ROK alliance as he promised he would. So ROK nuclearization may happen no matter what we think. And a US retrenchment from SK would probably scare Japan so much that the whole nuclear debate there would shift substantially to the right.
There is a lot of anxiety about this step, and I share it too. So I don’t endorse SK/J nuclearization. But there is SK polling showing high interest in this, and SK is terribly exposed to NK nuclear devastation with few good options as the NKs continue to build relentlessly. (All this I cover below.) So the least we Americans can do is get out of the way and let them debate it themselves.
The original, pre-edited FP essay on this follows below the jump:
A February poll found that 71% of South Koreans wanted their country, the Republic of Korea (ROK), to have nuclear weapons. Another in May found 70% supported indigenous nuclearization, with 64% in support even if that violated the Non-Proliferation Treaty (NPT). The drivers, unsurprisingly, are North Korea’s weapons of mass destruction (WMD) and China’s growing belligerence. These factors impact the Japanese nuclearization debate too, although interest there is noticeably lower. America has long opposed ROK/Japanese counter-nuclearization. But in the light of the Ukraine war, Washington should not hegemonically dictate the outcome of its allies’ WMD debates.
NATO anxiety over Russian WMD in the Ukraine war illustrates potential limits on American counter-escalation when facing a nuclearized opponent. Western pundits have been quite candid that Russian nuclear weapons were the reason for rejecting the no-fly zone sought by Kyiv. Chinese and, especially, North Korean WMD might play a similar blocking or limiting role in East Asian contingencies.
Importantly, US guarantees to South Korea and Japan are formalized as treaty, while NATO is not similarly committed to Ukraine. But during the Cold War, Britain and France were incredulous enough that America would sacrifice ‘New York for Paris’ that they built their own nuclear weapons despite formal US guarantees. That same logic is at work in East Asia today. The US will not sacrifice ‘Los Angeles for Seoul.’
China, with its relatively restrained nuclear rhetoric, is less the issue here than North Korea, which regularly, flamboyantly, and frighteningly invokes its nuclear weapons. Pyongyang is not going to reform, will march relentlessly toward more and better WMD, and is building its doctrine around their use, including tactically. This is the core problem: Kim Jong Un’s North Korea is a terrifying orwellian cult-tyranny, and it will not stop building the world’s most dangerous weapons nor give them up.
Alternatives to direct ROK/Japanese nuclear deterrence of Northern WMD are soft. Extended nuclear deterrence is weakly credible if it means nuked US cities to defend South Korea or Japan. Missile defense does not work well enough to provide a ‘roof’ against as many weapons North Korea appears to be building. China will not take serious action to stop Pyongyang. A negotiated deal – the best solution and hence discussed at length below – might control Pyongyang’s programs somewhat, via missile or warhead limits, or inspector access. But North Korea seems unwilling to negotiate seriously; is an untrustworthy counterparty; is unlikely to cut enough to relieve the existential threat its WMD now pose to South Korea and Japan; and would demand exorbitant counter-concessions as payment.
This poor option-set is already forcing ‘thinking the unthinkable’ discussions in the region. ROK President Yoon Seok-Yeol suggested preemptive strikes on Northern missile sites in a crisis, and former Japanese Prime Minister Abe Shinzo suggested the return of US nuclear weapons to the region. The sheer precarity of South Korean and Japanese exposure to a nuclearized/missilized orwellian tyranny – which will be evident yet again this year if Pyongyang tests another nuclear weapon as predicted – will make it increasingly awkward for America to hegemonically insist that Seoul, and Tokyo even, may not investigate all security options.
Worse, US resistance to allied nuclearization assumes a traditional American internationalism which is no longer assured. One of America’s two parties increasingly disdains alliances and admires authoritarianism. If Donald Trump – or a similar ‘Trumpist’ – retakes the US presidency in 2024, American opposition to east Asian allies’ nuclearization will decline dramatically – if only because the US will no longer care what they do. As president, Trump was more interested in personal relationships with regional autocrats like Xi or Kim than with traditional US partners. He notoriously “fell in love” with Kim and expressed particular loathing for South Korea. Trump signaled a desire to “blow up” the US-ROK alliance if re-elected elected and hinted at breaking the 1951 US-Japan Security Treaty too. So, there is a reasonable chance South Korea will nuclearize after 2024 regardless of what the Americans think. US abandonment of South Korea would also push the Japanese nuclearization discussion to the right.
This would not be the first time the United States has tacitly accepted another country’s nuclearization. Ostensibly, the US has supported the NPT for decades. In practice though, Washington tolerates at least six states’ unwillingness to build down, per the reduction exhortation of the NPT’s Article VI: itself, Britain, France, Israel, India, and Pakistan. By the vague standard implied by these examples – friendship with America, reasonable state capacity, passably liberal and democratic – South Korea and Japan more than clear the bar for a US NPT ‘exemption.’
Judged by US behavior toward NPT contravention, the NPT is better understood as a US effort to prevent unfriendly or hostile states from nuclearizing, rather than as a blanket, ‘Global Zero’ commitment to fewer nuclear weapons in the world. The US does not pressure friendly nuclear weapons-states, including itself, to meet NPT requirements. It gave up sanctioning India and Pakistan’s violation after just three years. Applying this more honest standard of US interests to arms control, the NPT is of questionable utility in East Asia.
Most obviously, the nonproliferation upside of the NPT is now passé in the region. China, Russia, and North Korea already possess nuclear weapons and show no signs of building down. So there is no regional nuclearization ‘cascade’ for the NPT to inhibit, by blocking ROK/Japanese nuclearization, because it has already happened. And Taiwanese nuclearization is unlikely, as South Korea and Japan are friendly states whose WMD would not impact Taipei’s security. Taiwanese elites are quite aware their nuclearization would provoke China. More generally, recent nuclear trends – modernization (in America, Britain, China, and Russia), arsenal growth (China, North Korea), the breakdown of the Intermediate-Range Nuclear Forces and Open Skies Treaties, the growth of low-yield options – undercut the NPT.
Next, there is an under-discussed NPT downside: it provokes the alliance-debilitating, ‘New York for Paris’ debates mentioned above. If US allies may not nuclearize and must rely on US nuclear weapons for nuclear deterrence, then they will inevitably question whether the US will use those weapons in their defense if that might incur a retaliatory nuclear strike on the US homeland. The answer to that question is almost certainly no, as French President Charles de Gaulle realized sixty years ago. Because this debate is irresolvable and endemic, it will permanently bedevil US alliances in East Asia as North Korea continues to build up. US Senator Lindsey Graham pointedly stated this quandary in 2017:
If there’s going to be a war to stop [Kim Jong Un], it will be over there. If thousands die, they’re going to die over there. They’re not going to die here. And [Trump’s] told me that to my face…That may be provocative, but not really. When you’re president of the United States, where does your allegiance lie? To the people of the United States.
The easiest way to reduce this bitter, alliance-undermining dissension is to let US allies ‘self-insure’ via indigenous nuclearization.
Finally, ROK/Japanese nuclearization could serve shared regional interests by providing supplemental, local deterrence (as British and French nukes did during the Cold War) and by improving alliance burden-sharing. Further, the threat of ROK/Japanese nuclearization might finally prompt Pyongyang and Beijing to take North Korean denuclearization negotiations seriously. Should South Korea and Japan respect the NPT and Global Zero, while China, Russia, and North Korea do as they will, the effective outcome is unilateral disarmament. This is politically and strategically infeasible; we regrettably live in a world of persistent nuclear armament.
Global Zero advocates such as Scott Sagan worry about the transactional issues of WMD possession, because they are uniquely dangerous weapons. Indeed, theft, loss, rogue scientists, and so on are legitimate fears. But they are no more resonant with South Korea or Japan than with any other nuclear weapons-state. Indeed, as liberal democracies with robust state capacity and pre-existing, well-managed nuclear energy programs, they will likely be quite responsible, as Britain and France have been.
No one seriously believes Seoul or Tokyo will launch an out-of-the-blue nuclear first strike on an opponent; set up something like the A.Q. Khan proliferation network; sell WMD to terrorists or other rogues; put Homer Simpson in charge of nuclear safety; or be so sloppy as to require something like the Nunn-Lugar program. Britain and France never did such things; Pakistan and India have been better with their arsenals than the panic of the late 1990s suggested. Even dictatorships have been cautious about these issues. And as democracies with a history of foreign policy restraint, democratic peace theory suggests they would be good stewards, certainly better than east Asia’s autocratic nuclear powers.
There is a generalized anxiety about a regional arms race, which ROK/Japanese nuclearization might exacerbate. Perhaps, but as noted above, there is no local cascade to be sparked, because it has already occurred. China, Russia, and North Korea have all moved first. China and Russia have established nuclear arsenals and no intention of complying with the NPT Article VI build-down imperative. Russia’s growing rhetorical invocation of its nuclear weapons is a disturbing evolution. North Korea repeatedly agreed, non-bindingly since 1992, to avoid nuclear weapons – only to exit the NPT and keep building. It now has intercontinental ballistic missiles and several dozen nuclear warheads.
Ostensibly South Korea and Japan are not competing in this race – but only because they ‘outsource’ their nuclear deterrence to America. Extended deterrence does not remove US-Japan-ROK alliance WMD from the East Asian security discussion. It only means they are not located in-theater. That may have value in keeping China from building yet more (although it is already doing so) or Russia from playing the nuclear card in the region as it does in Europe. But it is not stopping North Korea. And that is the core issue – always and again.
North Korea will not sign a deal which reduces its arsenal enough to reduce the strategic threat which brought Yoon to float preemption earlier this year. Even if Pyongyang signed a deal – and did not cheat – it would never cut deeply enough to obviate the existential threat it now poses to Japan and South Korea. Nuclear weapons are an excellent deterrent for North Korea, and tactically, they help equalize the conventional military competition with the South and US where Pyongyang lags behind. Complete, verifiable, irreversible denuclearization is fantasy.
The negotiations between Kim, Trump, and previous ROK President Moon Jae-In strongly suggest this. From 2018 to 2020, North Korea had the best chance in its history to capture a balance-positive deal with the ROK and US. Revealingly, Kim passed it up, even though the constellation of forces was nearly ideal for Pyongyang in two, overlapping dovish presidencies in the North’s primary opponents:
In Trump, Pyongyang had the best American president ever for its interests. Trump loathed South Korea. He knew little about Korean history, nuclear weapons, or ballistic missiles; according to former Trump National Security Advisor John Bolton, Trump did not even prepare for his summits with Kim. Trump did not care about the US position in East Asia and disliked US allies generally. He desperately wanted to sign a peace deal with Kim to win a Nobel Peace Prize and help his re-election bid. Moon too was about the best North Korea could hope for in a South Korean president. His roots were in the South Korean left which genuinely believes that North Korea is brother Korean state which would never nuke South Korea.
It is hard to imagine better counterparties to whom Kim might make some genuine concessions in order to receive large counter-concessions. Instead, Kim’s one serious offer to Trump, at Hanoi in 2019, was very unbalanced. Kim offered to shutter one aging nuclear plant for full sanctions removal. Even Trump, desperate for a deal for a Nobel and the consequent prestige, realized this was balance-negative and rejected it. Kim never again proposed anything serious to Trump or Secretary of State Mike Pompeo, and talks collapsed. North Korean nukes are here to stay.
Finally, a ROK/Japanese nuclearization discussion indicates a seriousness about their own security which is long overdue. Cheap-riding and strategy immaturity among US allies are long-established problems. This is glaringly obvious in Europe now, where local US allies, much more impacted by the Ukraine war, are nonetheless buck-passing leadership of the response to America. The US should discourage this if it is to finally achieve a more restrained, less sprawling foreign policy, a less gargantuan defense budget, greater focus on China, fewer ‘forever war’ interventions, and so on.
If allied democracies want nuclear weapons, if their foreign policy elites and voters decide to take this step, then the US should accept that that is their choice. As a liberal alliance leader, the US should not tell its partners what to do, nor what they may even debate. ROK/Japanese interest in WMD is defensive, in good faith, and follows decades of restraint; it is obviously not offensively intended. America should want US allies to take greater responsibility, develop deep national security doctrines, spend more, stop turning to America for foreign policy direction, and so on. Indeed, Yoon recognizes that, in the very title of his Foreign Affairs article for the 2022 South Korean presidential election: “South Korea Needs to Step-Up.” Precisely.
Allied cheap-riding is bad for America at home too. Militarized hegemony is deeply toxic to US domestic politics. The American national security state is too large and intrusive. American policing has become militarized, and the culture fetishes soldiers and military violence in a manner unique and disturbing for a republic. Greater allied burden-sharing has long been a goal of US foreign policy, and it would be good for US republican values at home if America did less abroad. There is no reason why greater allied strategic responsibility should not include WMD if well-governed democratic allies so choose.
No one wants more nuclearization if avoidable. The decision is momentous, and I do not endorse it. Ideally, arms control with North Korea would alleviate some risk. And missile defense could provide a bit of ‘roof.’ And extended deterrence and Chinese resistance could encourage North Korea to slow down.
But these options are all poor and getting worse. The Trump-Moon-Kim talks – the most concentrated effort at a deal ever – failed. China is not helping. Missile defense is costly and does not work well. Yoon’s preemption could provoke precisely the war it intends to prevent. Trump’s possible re-election is a huge wildcard. The Ukraine war illustrates the risks of extended deterrence, and the better/faster/more powerful North Korean WMD become, the less credible the US commitment becomes. America will not fight a nuclear war solely for allies, a point about which America analysts should be honest even if US officials dance around it. Direct ROK/Japanese deterrence is increasingly a better option than these alternatives – which point another North Korean nuclear test this year will likely throw into sharp relief. The US should at least allow its allies to debate the issue without strong-arming them.