Destroying Pyongyang’s nuclear arsenal is still in America’s national interest.

Nothing can be known about this week’s talks between North and South Korea other than their likely outcome. As in every previous encounter, South Korea will almost certainly reward North Korea’s outrageous misconduct by handing over substantial sums of money, thus negating long-overdue sanctions recently imposed by the United Nations Security Council. Meanwhile, the North will continue to make progress toward its goal of deploying several nuclear-armed, mobile intercontinental ballistic missiles, having already tested nuclear-explosive devices in October 2006, May 2009, February 2013, January 2016, September 2016, and September 2017

Each test would have been an excellent occasion for the United States to finally decide to do to North Korea what Israel did to Iraq in 1981, and to Syria in 2007 — namely, use well-aimed conventional weapons to deny nuclear weapons to regimes that shouldn’t have firearms, let alone weapons of mass destruction. Fortunately, there is still time for Washington to launch such an attack to destroy North Korea’s nuclear arsenal. It should be earnestly considered rather than rejected out of hand.

Of course, there are reasons not to act against North Korea. But the most commonly cited ones are far weaker than generally acknowledged.

One mistaken reason to avoid attacking North Korea is the fear of direct retaliation. The U.S. intelligence community has reportedly claimed that North Korea already has ballistic missiles with nuclear warheads that can reach as far as the United States. But this is almost certainly an exaggeration, or rather an anticipation of a future that could still be averted by prompt action. The first North Korean nuclear device that could potentially be miniaturized into a warhead for a long-range ballistic missile was tested on September 3, 2017, while its first full-scale ICBM was only tested on November 28, 2017. If the North Koreans have managed to complete the full-scale engineering development and initial production of operational ballistic missiles with nuclear warheads in the short time since then — and on their tiny total budget — then their mastery of science and engineering would be entirely unprecedented and utterly phenomenal. It is altogether more likely that they have yet to match warheads and missiles into an operational weapon.

It’s true that North Korea could retaliate for any attack by using its conventional rocket artillery against the South Korean capital of Seoul and its surroundings, where almost 20 million inhabitants live within 35 miles of the armistice line. U.S. military officers have cited the fear of a “sea of fire” to justify inaction. But this vulnerability should not paralyze U.S. policy for one simple reason: It is very largely self-inflicted.

When then-U.S. President Jimmy Carter decided to withdraw all U.S. Army troops from South Korea 40 years ago (ultimately a division was left behind), the defense advisors brought in to help — including myself — urged the Korean government to move its ministries and bureaucrats well away from the country’s northern border and to give strong relocation incentives to private companies. South Korea was also told to mandate proper shelters, as in Zurich for example, where every new building must have its own (under bombardment, casualties increase dramatically if people leave their homes to seek shelter). In recent years, moreover, South Korea has had the option of importing, at moderate cost, Iron Dome batteries, which are produced by both Israel and the United States, that would be capable of intercepting 95 percent of North Korean rockets headed to inhabited structures.

But over these past four decades, South Korean governments have done practically nothing along these lines. The 3,257 officially listed “shelters” in the Seoul area are nothing more than underground shopping malls, subway stations, and hotel parking lots without any stocks of food or water, medical kits or gas masks. As for importing Iron Dome batteries, the South Koreans have preferred to spend their money on developing a bomber aimed at Japan.

Even now, casualties could still be drastically reduced by a crash resilience program. This should involve clearing out and hardening with jacks, props, and steel beams the basements of buildings of all sizes; promptly stocking necessities in the 3,257 official shelters and sign-posting them more visibly; and, of course, evacuating as many as possible beforehand (most of the 20 million or so at risk would be quite safe even just 20 miles further to the south). The United States, for its part, should consider adding vigorous counterbattery attacks to any airstrike on North Korea.

Nonetheless, given South Korea’s deliberate inaction over many years, any damage ultimately done to Seoul cannot be allowed to paralyze the United States in the face of immense danger to its own national interests, and to those of its other allies elsewhere in the world. North Korea is already unique in selling its ballistic missiles, to Iran most notably; it’s not difficult to imagine it selling nuclear weapons, too.

Another frequently cited reason for the United States to abstain from an attack — that it would be very difficult to pull off — is even less convincing. The claim is that destroying North Korean nuclear facilities would require many thousands of bombing sorties. But all North Korean nuclear facilities — the known, the probable, and the possible — almost certainly add up to less than fewer dozen installations, most of them quite small. Under no reasonable military plan would destroying those facilities demand thousands of airstrikes.

Unfortunately, this would not be the first time that U.S. military planning proved unreasonable. The United States Air Force habitually rejects one-time strikes, insisting instead on the total “Suppression of Enemy Air Defenses.” This is a peculiar conceit whereby every single air-defense radar, surface-to-air missile, airstrip, and combat aircraft in a given country must be bombed to destruction to safeguard U.S. pilots from any danger, instead of just bombing the targets that actually matter. Given that North Korea’s radars, missiles, and aircraft are badly outdated, with their antique electronics long since countermeasured, the Air Force’s requirements are nothing but an excuse for inaction. Yes, a more limited air attack might miss a wheelbarrow or two, but North Korea has no nuclear-warhead mobile missile launchers to miss — not yet.

Perhaps the only good reason to hesitate before ordering an attack on North Korea is China. But that’s not because Beijing would intervene against the United States. The notion that China is North Korea’s all-around protector is badly out of date. Yes, the Chinese do not want to see North Korea disappear with U.S. troops moving up to the Yalu River and China’s border. But President Xi Jinping’s support for maximum economic sanctions, including a de facto blockade of oil imports — a classic act of war — amounts to a change of sides when it comes to North Korean nuclear weapons. Anybody who believes China would act on North Korea’s behalf in the event of an American attack against its nuclear installations has not been paying attention.

But China’s shift has surfaced a quite different reason for the United States not to bomb: While North Korea’s acquisition of nuclear weapons is of course very dangerous, it does ensure its independence from Chinese influence. In a post-strike scenario, the Pyongyang regime might well crumble, with the country becoming a Chinese ward. That could give Beijing dominant influence over South Korea as well, given the preference of some South Koreans — including President Moon Jae-in, according to reports — for Chinese as opposed to American patronage. A China-dominated Korean Peninsula would make Japan less secure and the United States much less of a Pacific power.

In theory, a post-attack North Korea in chaos could be rescued by the political unification of the peninsula, with the United States assuaging Chinese concerns by promptly moving its troops further south, instead of moving them north. In practice, however, this would be a difficult plan to carry out, not least because South Korea’s government and its population are generally unwilling to share their prosperity with the miserably poor northerners, as the West Germans once did with their East German compatriots.

For now, it seems clear that U.S. military authorities have foreclosed a pre-emptive military option. But the United States could still spare the world the vast dangers of a North Korea with nuclear-armed long-range missiles if it acts in the remaining months before they become operational.

It’s true that India, Israel, and Pakistan all have those weapons, with no catastrophic consequences so far. But each has proven its reliability in ways that North Korea has not. Their embassies, for instance, don’t sell hard drugs or traffic in forged banknotes. More pertinently, those other countries have gone through severe crises, and even fought wars, without ever mentioning nuclear weapons, let alone threatening their use as Kim Jong Un already has. North Korea is different, and U.S. policy should recognize that reality before it is too late.