SINGAPORE — Northeast Asia is on the cusp of a major strategic shift. Sanctions have not prevented North Korea from acquiring nuclear weapons. It is already a de facto nuclear state. Sooner or later, Pyongyang will acquire nuclear-capable intercontinental ballistic missiles that can directly threaten the continental United States.

The time for preemptive kinetic action has passed. Since North Korea already has nuclear devices, if not yet fully operational nuclear weapons, all it has to do is detonate those devices close to its South Korean or Chinese borders to raise the stakes of preemptive action to unacceptable levels. The regime of North Korean leader Kim Jong Un is ruthless enough to do so.

Pyongyang has thousands of conventional artillery and missiles trained on Seoul, which is only about 35 miles from the demilitarized zone that separates the two countries. The U.S. does not have the capability to locate and simultaneously neutralize all of these missiles and thus cannot prevent a devastating attack, which would cause hundreds of thousands of civilian casualties in South Korea and perhaps Japan too.

China cannot stop North Korea. To do so, it must change the regime. But even if it has the ability to do so, the Chinese Communist Party is not willing to effect regime change in a fellow Leninist state. The preservation of its own rule is the most vital of all the CCP’s interests. Intervening to change the regime in Pyongyang could give Chinese citizens inconvenient thoughts about their own system. No matter how angry the CCP is with North Korea, that is too great a risk. Beijing has taken symbolic actions against Pyongyang but tolerating a nuclear North Korea is the least bad option.

Pyongyang, therefore, will have to be dealt with by deterrence. And it can be deterred: the regime is bad but not mad. It is coldly rational, calculating exactly how far it can go in any set of circumstances. For example, Pyongyang didn’t pay any significant cost after blowing up a South Korean passenger aircraft in 1987 or sinking a South Korean navy ship in 2010, killing 46 sailors. Once Pyongyang has the capability it believes it needs to ensure the survival of its regime, it has no reason to risk annihilation.

The Obama administration’s approach of so-called “strategic patience” was a failure. But the Trump administration is doing the right thing. Trump’s statements are extreme but they are in accordance with the essential logic of deterrence.

Once North Korea can directly threaten the continental U.S., the question is bound to be asked: Will the U.S. sacrifice San Francisco in order to save Tokyo? Of course not. Still, North Korea is a catalyst, not a cause; Pyongyang’s quest for a nuclear capability may cause this awkward question to be asked sooner, but it will eventually be asked anyway. China is modernizing its own nuclear forces and will eventually acquire a more credible second strike capability vis-à-vis the U.S. One way or another, American extended deterrence in Northeast Asia will be eroded, as it was decades ago in Europe.

Northeast Asia will respond as France and the U.K. did in the 1950s and 1960s respectively. Japan has the ability to quickly develop an independent nuclear deterrent. It is now only a matter of when, not if, Japan does so. Tokyo has been preparing for this eventuality — with American acquiescence and perhaps assistance — for decades.

Where Japan goes, South Korea must follow. I don’t think Japan and South Korea are eager to become nuclear-armed states, nor is Washington eager for that to happen. But for all three, this is also the least bad option. Japan and South Korea will remain within the U.S.-led Northeast Asian alliance, just as France and the U.K. remained within NATO. But a six-way balance of mutually assured destruction — among the U.S., China, Russia, Japan, South Korea and North Korea — will eventually be established in Northeast Asia.

Getting to this new situation will be fraught with serious tension. China will pull out all stops short of war to prevent Japan going nuclear, raising the shibboleth of Japan’s remilitarization to try and rally Americans, Japanese, Koreans and others in East Asia against Tokyo. But it will fail. And the U.S.-Japan alliance will deter China from preemptive military action against Japan. War with the U.S. cannot end well for China; it would jeopardize the CCP’s rule.

The decision to go nuclear will be extremely difficult for any Japanese government, far worse than the backlash against the U.S.-Japan security treaty in the 1960s and early 1970s. But when America’s extended deterrence is eroded, a Japan without an independent nuclear deterrent would be subordinate to China. This is an existential issue for Japan. Ever since Toyotomi Hideyoshi invaded Korea in the 16th century in explicit defiance of the Chinese world order of the time, refusal to accept subordination to China has been an integral part of the Japanese sense of identity. To accept subordination would require a wrenchingly painful redefinition of what it means to be Japanese, which I do not think the Japanese will accept.

However difficult the process of getting to a six-way balance of mutually assured destruction may be, once established, it will be stabilizing. All six countries are rational and are functioning polities. The North Korean regime is brutal, but it works. Despite numerous predictions of its imminent demise, it is still here after more than 70 years. North Korea is certainly more coherent than Pakistan, which is also a nuclear-armed state but constantly teetering on the brink of failure.

A Northeast Asian balance of mutually assured destruction will freeze the status quo. It will be an absolute obstacle to the revanchist ambitions that are embedded in the narrative of the “great rejuvenation” of China by which the CCP now legitimates itself and which are manifest in the East China Sea and the South China Sea. China’s reclamation activities in the South China Sea cannot be reversed, nor will China give up its claims. But forcing China to at least suspend its ambitions at their current level will make for more stable Sino-American and Sino-Japanese relations and a more stable East Asia.

Freezing the status quo will put an end to the chimera of Korean reunification and make for healthier relations between the North and the South. Reunification is an aspiration that neither has really been in a hurry to achieve. For the North, unless entirely on its own terms, reunification means the end of the regime. For the South, the incorporation of 25 million North Koreans who have no experience of a modern market economy into a population of 51 million will irrevocably change South Korea and not for the better. Better to end all pretense.

The new balance could, however, tempt Taiwan to consider its own nuclear options. Taipei harbored such ambitions in the 1960s until the U.S. quashed them the following decade. It would be extremely dangerous for such ambitions to return. The one issue over which China must risk war with the U.S. is Taiwan — the CCP’s rule cannot survive if it allows Taiwan to go independent. Taiwan cannot acquire a nuclear capability without U.S. acquiescence. Thus, it is crucial that the U.S. reassure China that it will not tolerate a nuclear-armed Taiwan.

As Northeast Asia moves inexorably toward a new nuclear balance, it is important that the U.S. couples pressures and deterrence with serious exploration of a peace treaty with North Korea. A peace treaty will enable deterrence to be maintained at a less intense level. It could de-incentivize North Korea from taking provocative actions. At present, Pyongyang has no other card to play.

Despite Trump’s strong rhetoric, there have been hints that his administration has not ruled out a more accommodating approach. Although denuclearization is a pipe dream, a peace treaty in return for capping North Korean missile and nuclear capabilities and ending its proliferation activities will be complicated but not impossible. There are two main difficulties, neither of which are insurmountable.

First, at what level of capability will Pyongyang feel secure enough to agree to a verifiable freeze, and will such a deal be politically palatable in Washington, Tokyo and Seoul? The simple answer to this is that we will never know until the possibility is explored. Second, a peace treaty must include a package of continuing economic assistance to give Pyongyang an alternative source of revenue. North Korea has exported nuclear and missile technology to the Middle East — not in pursuit of any grand strategic or ideological design but for money.

Purists will regard this as rewarding bad behavior, and indeed it is. But we should not pretend that there are no precedents for doing so. This is how the Korean Peninsula Energy Development Organization came about. The agreement on KEDO would not have been reached in 1995 if Pyongyang had behaved with sweet reason in 1993 and 1994. The same is true of the Joint Comprehensive Plan of Action with Iran.

North Korean ambitions are limited — it only wants to survive. Who knows the limits to Tehran’s ambitions? The U.S. has been willing, against the objections of some of its closest Middle Eastern friends and allies, to tolerate Iran as a threshold nuclear state with no guarantee that it will not, in a relatively short time, cross the threshold when the JCPOA expires. By comparison, a peace treaty with a de facto nuclear-armed state is a small price for stability in the most economically dynamic part of the world. After all, everything else has failed.