Any travelers waiting for the few flights out of Pyongyang International Airport early on August 29 were treated to the spectacle of a North Korean intermediate-range missile blasting off only a few miles beyond the runways. Just before six in the morning, a Hwasong-12 missile, also known as the KN-17, with a purported range of nearly four thousand miles, arced northeastward over North Korea and the Sea of Japan. Eight minutes later, it passed over Hokkaido, the northernmost of Japan’s four home islands. Roughly six minutes after that, and approximately 730 miles east of Hokkaido, it broke apart and fell into the Pacific Ocean.
If the trajectory of the KN-17 had been a little more northerly, and had it broken up a few minutes earlier, it could have rained rocket debris down on Sapporo, Japan’s fifth-largest city, with a population of two million. Like many North Korean rocket tests, this one ended in structural failure, a reminder that Pyongyang has not yet perfected its missile technology. While that may give temporary solace to those worrying about North Korea’s nuclear capability, it serves as a warning about perhaps the most serious threat posed by Kim Jong-un’s nuclear and ballistic missile arsenal: its safety.
Perhaps the world should worry less about the threat of a North Korean-instigated nuclear war and more about the risk of a nuclear accident. The most frightening question raised by Kim Jong-un’s pursuit of the ultimate weapon is also the simplest: Can he control his nukes?
Unlike a conventional military, where tanks, trucks, even planes are relatively simple instruments of war, owning nuclear weapons is a huge, expensive, and complex responsibility. Warheads must be maintained, as must the missiles that deliver them. Launch procedures are—or should be—complicated enough that no weapon can be fired on a whim, yet reliable enough that a national leader has confidence his nukes are ready when he is. Given the terrible responsibilities involved, nuclear personnel need to be carefully chosen and trained, since the most mundane procedures have the potential to turn into unimaginable catastrophes.
The warhead that detonates over a target is but one part of a complex system. The US government describes the warheads, missiles, launchers, communications networks, satellites, production and maintenance facilities, trucks, guards, bunkers, and the like as the “nuclear enterprise.” Having confidence in that enterprise—that the weapons are safe, are in place when needed, will work as desired, and (as important) will not work when not desired, and that crews are fully trained—is known as “nuclear surety.”
Even if Pyongyang’s laboratories and factories are safe, weapons systems break down, age, and suffer untold problems. Even the nations that have been working with nuclear weapons the longest—the United States and Russia—still make mistakes, and struggle to ensure that their nuclear operators are competent and honest. The history of the cold war is littered with accidents involving nuclear weapons, known as “broken arrows,” and incidents that could have sparked a global thermonuclear war. Despite multiple safety systems and rigorous training and maintenance, the US military alone suffered hundreds of accidents such as the 1980 Damascus incident, when an armed Titan II missile blew up in its Arkansas silo after a technician dropped a socket that punctured the missile’s fuel tank. Three years later, Soviet radar posts mistakenly reported sunlight glinting off clouds as American ICBM launches; only the doubts of the duty officer at the time, Stanislav Petrov, prevented nuclear retaliation.
As detailed in Eric Schlosser’s history, Command and Control (2013), there were thirty-two broken arrows between 1950 and 1980 alone, including no fewer than six hydrogen bombs dropped on American soil that have never been recovered, some sinking into the swamps or coastal shallows, lurking forever. In 1961, a dropped megaton bomb was one safety switch away from detonating in North Carolina.
North Korea does not have nuclear bombs or squadrons of bombers, but we cannot take for granted that the North will invest in the safest designs for its warheads or missiles. It is unlikely, but conceivable, that a warhead, jettisoned from a missile that explodes due to a fueling mistake, could detonate. A nuclear detonation on North Korean soil would be hard to cover up, and Kim Jong-un would undoubtedly deflect blame by accusing the Americans, South Koreans, or Japanese of sabotage or an attack, sparking a military crisis that could be uncontrollable.
All this raises the question of accountability. As retired Air Force Lieutenant General James Kowalski, a B-52 pilot and the former deputy commander of US Strategic Command, put it to me in an interview: “Who is going to be the guy who goes to Kim Jong-un and tells him he has a problem with his nukes?” Absolute trust is required between leaders and those charged with maintaining and operating nuclear weapons. It is hard to imagine that existing among Kim’s circle of terrified sycophants.
During the cold war, the most harrowing specter of error hung over the command and control of nuclear weapons. The use of nuclear weapons, whether authorized or unauthorized, begins with arming a missile and making it ready for launch. Except for the small number of weapons on alert, US missiles are kept separate from their nuclear warheads, and an order must given to move the warheads out of secured storage bunkers to be matched with the delivery systems. Given that a good part of Kim’s arsenal comprises road-mobile missiles, the time required to move them to safe launching locations may push the North Koreans to keep more warheads outside secured storage, either mated to missiles or quickly accessible.
At the heart of the nuclear enterprise is the turning of the launch key. No one yet knows what North Korea’s nuclear release procedures will be. While Kim Jong-un will likely keep all control over nuclear weapons in his hands, he won’t physically fire the missile, and so he must delegate that authority in some way. Will Kim have the equivalent of the American president’s nuclear “football,” with its menu of launch options? Once Kim has decided what he wants to do, will the order go from him solely to the commander of the Strategic Rocket Forces, the military unit that presumably controls the North’s nuclear ballistic missiles? Or will Kim want to give orders directly to the field units, which, in addition to the mobile launchers, comprises launch pads and possibly a silo complex, not unlike the US Minuteman III force?
Even more opaque is the question of who will have ultimate launch authority at individual sites. It is hard to imagine the dictator of one of the world’s most ruthless and hierarchical states allowing subordinate officers the autonomy to launch nuclear missiles. Yet if Kim fears a “decapitation strike” by US or South Korean forces, he might issue orders that delegates launch authority to dispersed units. One can envision a scenario during a crisis in which a panicky junior officer loses communications with upper-level commanders and decides that he needs to launch before he is attacked, or because a first strike has taken out Kim.
If Pyongyang does not have reliable communications with its nuclear launch systems and personnel, then the uncertainty in nuclear operations increases dramatically. This, in turn, will put pressure on US commanders who are trying to decide how the North may respond to any American action. “If they did something and we responded,” Kowalski says, describing one of his greatest concerns, “we have to be careful to understand their C2 command and control system and not interfere. You don’t want to put into motion an automatic delegation system where a junior officer has launch authority.”
Living in this new environment will demand some new, perhaps radical thinking on the part of the United States government. This may sound bizarre, but it is in America’s interests to make North Korea’s nukes safer. While some US strategists must draw up plans to deter and, if necessary, defeat a nuclear-armed North Korea, others should consider how to ensure a safe North Korean nuclear arsenal.
Squaring that aim with credible deterrence will be difficult. For example, to make sure that Kim has constant communications with his nuclear units, so that he does not fear losing contact with them, would Washington assure him that it will not sabotage North Korea’s command and control capabilities, whether through cyber warfare or other means? That might reduce pressure on Kim and his senior officers in a crisis. Could the United States ever propose nuclear stability steps like establishing a hotline between Washington and Pyongyang? Given the North’s closed and hostile system, such cooperation may seem impossible, but the alternative—constant suspicion and hair-trigger reactions—is more daunting.
If Donald Trump decides not to attack North Korea in the next few months, and if Kim Jong-un refrains from giving him cause to do so, the world will settle down to the long-term challenge of learning to live with Pyongyang’s nuclear weapons. Washington may decide never to acknowledge that North Korea is a nuclear weapons state, in a bid to keep nonproliferation aims alive, but the US will need to figure out how to ensure that the accidents and miscalculations of the cold war are not repeated in North Korea, with catastrophic consequences.
Michael Auslin is the Williams-Griffis fellow in contemporary Asia at the Hoover Institution, Stanford University.