The ABCs of Deterring North Korea

September 13, 2017

For special access to experts and other members of the national security community, check out the new War on the Rocks membership.

Editor’s Note: This is the second installment of “The Brush Pass,” a new column by Joshua Rovner (@joshrovner1) on intelligence, strategy, and statecraft.

Kim Jong Un has rekindled a decades-old debate among international relations scholars: Are some leaders too irrational to be deterred? Some in the Trump administration think so. National Security Advisor H.R. McMaster recently suggested that the North Korean regime is so brutal that deterrence theory does not apply. His comments echo previous concerns about rogue states that are impossible to deal with. Others have sought to reassure nervous Americans that deterrence worked against murderous regimes in the past, and we can rely on it today to deter North Korea from nuclear war.

This is an important debate, to be sure, but it is incomplete. The question of deterrence is not simply about preventing war, because a nuclear North Korea can act in several ways that threaten U.S. interests even if it never uses nuclear weapons in anger. Some of these actions are probably beyond U.S. control. But Washington can deter the most dangerous threats, even if it can’t deter everything.

So what, exactly, is the United States trying to deter?

In theory, emerging nuclear powers can act in four ways that threaten U.S. interests. First, they can expand their programs rapidly and in secret. This is not inevitable. Some nuclear powers choose to retain small and unsophisticated arsenals, believing that any nuclear capability is enough to deter potential aggression. But others, like Pakistan, expand quickly and clandestinely. This creates several problems. Rapid expansion undermines efforts to ensure reliable command and control, while simultaneously making leakage and nuclear accidents more likely. This reduces international confidence that the emerging nuclear power is a responsible custodian of its arsenal, which in turn feeds regional insecurity and may fuel arms racing.

Second, emerging nuclear powers can sell technology and materials abroad. This is tempting for states in economic distress and with few alternate sources of revenue. Successful sales boost their own economies, provide incentives for other would-be proliferators, and complicate efforts to track supplier networks.

Third, states can use nuclear arsenals as cover for conventional aggression or to support transnational armed groups. The idea is that acquiring nuclear weapons will “embolden” states and encourage risky behavior in the non-nuclear realm; Iran hawks are particularly concerned about this. Emboldenment may happen because emerging nuclear powers are overconfident about their newfound capabilities, or because they believe great powers will not intervene in regional conflicts if they fear nuclear escalation. States who succeed at going nuclear – no mean technical feat – are also likely flush with nationalism. This encourages hubris.

Finally, of course, emerging nuclear powers can use nuclear weapons in war. They may believe they can integrate escalation into conventional warfighting plans. Alternatively, they can keep nuclear weapons in reserve in the event of conventional losses, and use them to change their adversaries’ calculus and force a settlement. The idea that states can escalate to de-escalate is not unique to new nuclear powers, of course, but they would find the strategy particularly appealing given their conventional disadvantages. Relatively weak powers like France may deliberately choose a posture of asymmetric escalation, implicitly threatening nuclear use to deter conventional attack. In the event of war, they have incentives to strike first.

Different deterrent signals are appropriate for different kinds of threatening actions. Warnings meant to stop one kind of action might not work against another. Threats to destroy cities, or countervalue threats, are inherently incredible against anything less than nuclear use, since no one will believe a country that threatens mass killing in order to punish small transgressions. Similarly, threats to strike the enemy’s nuclear sites, or counterforce threats, may be useful to deter lesser actions in peacetime, but in a crisis they may prove destabilizing. The twofold task for strategists is determining what kinds of action to deter, and then matching them with the appropriate threats and rewards.

There are three ways to deter rapid expansion: threaten diplomatic isolation, use targeted economic sanctions, and dissuade new nuclear powers from racing for parity. The corresponding rewards are diplomatic integration, economic rewards, and the tacit toleration of the target state’s small nuclear arsenal. This approach appeals to both the emerging nuclear power’s material interests and its ego. It earns recognition as a member of the nuclear club, which its leaders may care about greatly, as long as it accepts limits on the size and sophistication of its arsenal.

Deterring the transfer of nuclear materials requires a combination of nuclear forensics and indirect threats against facilitators of illicit deals. Credible displays of forensic skill convince emerging nuclear powers that they cannot anonymously share materials. And third parties, like suppliers and financiers, may be deterred through legal and financial sanctions. Their commitment to the transaction is probably less than that of the parties directly involved in the process, and they may be more sensitive to these kinds of punishments.

The best way to deter nuclear powers from using their arsenals to act more conventionally aggressive is by maintaining local conventional superiority. This enhances deterrence without risking escalation, which in turn reduces questions about credibility and alleviates stress on alliances. In extreme cases, counterforce threats may be useful by changing the emerging nuclear power’s calculation: Is the political objective worth risking its hard-won arsenal? Care is needed, though, to avoid creating a use-it-or-lose-it problem. The ideal threat would be somewhat ambiguous, highlighting advances in intelligence, surveillance, reconnaissance, as well as precision. Such comments might produce doubt within the target state about its ability to control escalation.

Finally, deterring nuclear use in war requires unambiguous countervalue threats. I mean countervalue literally: threatening what the target state values most. This is often regime survival. Emerging nuclear powers considering nuclear escalation must have no doubt that taking the leap means the end of their regime. The corresponding reassurances in this case include promises of regime security and possible rewards for military leaders who disobey orders to launch.

What does this tell us about deterring North Korea? First, the United States should have low confidence in its ability to limit the scope and pace of North Korea’s expansion. Pyongyang has been willing to tolerate both diplomatic isolation and economic sanctions. Further measures to isolate and punish North Korea are possible – expelling it from the United Nations, perhaps, or accelerating efforts to target its network of international front companies — but so far it has been insensitive to these measures.

The only tactic that hasn’t been tried is dissuasion plus toleration of its small arsenal. The Trump administration is not likely to publicly accept the fact of North Korea as a nuclear power, but quiet diplomacy might eventually allow U.S. officials to tolerate its nuclear weapons. There is a precedent for this kind of shift. The United States was horrified at the prospect of a Chinese nuclear capability in the early 1960s, especially given Mao Zedong’s breathtakingly cavalier talk about nuclear war. But U.S. leaders soon learned to accept the arsenal as a fact on the ground, and within a decade Nixon went to China.

Some observers worry that North Korea will look to sell nuclear materials and technology. According to this argument, the regime counts on peddling its wares abroad to generate hard currency. It has a long history of illicit military sales, and it has few other options. Deterring transfer will be difficult, but it is not impossible. It is one thing to sell conventional weapons abroad, and quite another to deliver nuclear weapons. The United States may be able to send credible warnings that Pyongyang will not be able to do so clandestinely and that it will face punishment if one of its buyers uses North Korean-supplied materials in an attack. The United States may also benefit by deterring the intermediaries that North Korea may enlist to facilitate such deals – bankers, shipping companies, front companies, and so on.

In either case, this would require the United States to reveal some intelligence capabilities as a way of signaling its ability to monitor not just North Korea’s activities, but also those of third parties. The use of intelligence as a signaling device is counterintuitive, as it means revealing clandestine capabilities. But doing so may inject a dose of caution into the calculations of those who might take part in the North Korean nuclear trade.

Deterring North Korea from using its nuclear weapons as cover for conventional escalation is more straightforward. The United States and South Korea already enjoy overwhelming conventional superiority, and Washington can conspicuously surge more forces if North Korea ramps up activity below the nuclear threshold. It is possible that a newly emboldened North Korea will harass South Korean and U.S. forces, but it is unclear that these will be substantially different from prior attacks and forays across disputed territory. Many of these were carried out long before the North went nuclear, suggesting that acquiring nuclear weapons does not explain patterns in Pyongyang’s conventional behavior. Indeed, it is not clear that North Korea has become more aggressive since its first nuclear test in 2006.

There are other reasons to be confident about deterring North Korea from waging war under the cover of its own nuclear umbrella. If local conventional dominance is not enough, the United States can inspire caution in Pyongyang by threatening its nuclear arsenal. Again, this will require revealing some of intelligence capabilities, but Washington need not be specific. It must only broadcast enough awareness to create doubt that Pyongyang cannot count on reliable nuclear cover if it chooses a more ambitious conventional campaign. Such a warning may disabuse North Korea of the notion that it can control escalation. Scholars have argued that convincing nuclear powers that they can’t control escalation may prevent them from initiating conflict in the first place.

Finally, deterring wartime use will require unambiguous threats to the regime. This is the most important goal of deterrence, not only because of North Korea’s new capabilities, but also because of the growing belief that escalation is at the heart of North Korean strategic doctrine. The Kim family must be aware that using nuclear weapons against the United States or its friends will spell the end of its dynasty. If Korea watchers are correct that Kim Jong Un values regime security above all, he will be unlikely to shrug off such warnings. These threats also have the virtue of credibility, given both the historical U.S. concern about nuclear proliferation and its recent enthusiasm for regime change.

Achieving a lasting peace on the peninsula will require patient diplomacy, because deterrence is as much about unresolved political disputes as it is about the technical details of the military balance. Stability, meaning fewer and less intense crises, requires successfully resolving those disputes. By way of analogy, the most intense nuclear crises of the Cold War occurred because the political status of West Berlin was contested. The danger of war fell away after the issue was largely settled by the mid-1960s, even though the nuclear balance between the United States and the Soviet Union was changing dramatically. A similar breakthrough on the Korean peninsula is not in the foreseeable future.

For now, there is bad news and good news. Unfortunately, it will be very hard to deter North Korea from expanding its arsenal, as the last decade has shown. It will also be difficult, though not impossible, to deter North Korea from selling its nuclear knowledge. The good news is that it is much easier to deter the things we fear most: the use of nuclear weapons as cover for conventional aggression and the use of nuclear weapons in war.


Joshua Rovner is Associate Professor in the School of International Service at American University. He is the author of Fixing the Facts: National Security and the Politics of Intelligence (Cornell, 2011), and writes widely about intelligence and strategy.  

Image: U.S. Army/Sgt. Christopher Dennis

We have retired our comments section, but if you want to talk to other members of the natsec community about War on the Rocks articles, the War Hall is the place for you. Check out our membership at!