On China: Forget China.

On Russia: Forget Russia.

On diplomacy: U.S. Senator Lindsey Graham was right we he noted that “if there’s going to be a war to stop [North Korea], it will be over there. If thousands die, they’re going to die over there. They’re not going to die here.” Indeed, the decision to go to war against the North should be made in Seoul, and in Seoul alone. Ultimately, as Catherine Moon succinctly argued in early September, the North Korean threat is first and foremost a South Korean problem. Washington would do well to disengage from seeking any direct diplomatic talks with Pyongyang, and instead let South Korea take the lead in any future negotiations. However, to guarantee that both Koreas are negotiating in good faith and toward an outcome in line with U.S. national-security interests in the region, Washington ought to designate a specific date in 2018 on which Congress will ascertain the strategy’s success or failure and decide whether to launch a preventive war against North Korea.

On sanctions: Given that ‘tougher’ sanctions are not going to work in time and secondary sanctions risk undermining the relations between the Washington and Beijing, the most prudent course of action is to simply freeze international sanctions at the current level. While maintaining economic pressure on the Pyongyang is quintessential for any diplomatic solution to succeed, intra-Korean talks can only flourish in an environment where there is still breathing space for even stronger international sanctions to be implement. While unanimously adopted by the UN Security Council on September 11, the latest round of even “tougher” sanctions has shied away from implementing a full oil-import ban and imposing travel restrictions on Kim Jong-un himself. Russian president Vladimir Putin already noted in late August that “the policy of putting pressure on Pyongyang to stop its nuclear missile program is misguided and futile,” and it is highly doubtful whether Moscow and Beijing can be persuaded to support even tougher sanctions if the last round fails like all the other round before it.

On deterrence: Rather than strengthening deterrence, Washington should lower the U.S. deterrence posture on the peninsula to the extent that it might fail. This approach would entail withdrawing almost all U.S. forces and military assets from South Korea, and stopping all annual U.S.-ROK exercises. While United States and allied military planners will decry this approach as irresponsible and dangerous, the strategy’s underlying motivations are threefold: (1) providing a last goodwill gesture for diplomacy to work, (2) putting pressure on both Koreas to negotiate a feasible agreement on the future of the peninsula, and (3) potentially baiting Pyongyang into crossing the thirty-eighth parallel in search of a military solution.

To prevent Pyongyang from conducting any missile and nuclear tests during future negotiations, the U.S. Department of Defense ought to compile a list of thirty high-value military and political targets inside North Korea. For every missile Pyongyang puts into the air, the U.S. military will strike two listed targets; and for every nuclear test conducted, a North Korean city will be indiscriminately bombed.

If regime preservation and survival is Pyongyang’s overarching aim, and if North Korea is a rational actor capable of cost-benefit calculations, then the strategy of letting deterrence fail is the last and best approach to denuclearize the Korean Peninsula. For far too long has Washington carried a big stick without ever wielding it, and for far too long has tough talk been cheap on the peninsula. It is time to change both.

Stefan Soesanto is a Cybersecurity & Defence Fellow at the European Council of Foreign Relations and a nonresident James A. Kelly fellow at Pacific Forum CSIS.