South Korean Army K1A1 and U.S. Army M1A2 main battle tanks fire live rounds during a U.S.-South Korea live-fire exercise in a training area near the demilitarized zone separating the two Koreas in Pocheon, South Korea, on April 21, 2017.  (Kim Hong-Ji/Reuters)

What is the frightening situation between the North Korean regime of Kim Jong Un and the United States about? We have seen an escalation of words and actions suggesting possible war. But the North Koreans say, in so many words, that they just want security against attack by the United States. And members of the Trump administration, just like previous administrations, say they have no program for regime change in North Korea and simply want not to be threatened by its nuclear missiles. In principle, these look like compatible positions. So why is the escalation happening?

International relations scholars often talk about the “security dilemma,” which you may remember if you ever took a college course on international relations. This purports to explain how two states that really would be just fine with peace and the status quo might end up in a war nonetheless.

Given the apparent compatibility of the stated main goals, it is tempting to argue that the current crisis is an example of the “security dilemma” in action. Some analyses implicitly or explicitly invoke the idea.

But this is not what is going on here. The underlying problem — which is further complicated by a variety of psychological and personality issues — is that no U.S. administration, Trump’s or any other, can commit itself not to act to help replace Kim’s government if it were to face major domestic instability.

What is the security dilemma?

The security dilemma refers to two related ideas. The first is very simple: Things that one state does to make itself more secure typically have the effect of making other states less secure. For example, Kim is trying to develop a nuclear intercontinental ballistic missile (ICBM) capability to make his government more secure, but this has the effect of making the United States less secure. The second is about how this fact could lead states whose leaders are both basically interested in maintaining a non-conflict status quo to end up in a very costly war.

How might this work? Imagine two states, State A and State B, that both prefer the status quo to a military conflict. The problem is that while each state knows that it prefers the status quo, it doesn’t know that the other state does, too. So State A might build weapons, or do some other militarily competitive thing, to try to make itself more secure in case State B is in fact an aggressive type. State B, which really prefers the status quo, sees this action, and becomes more worried that State A is aggressive and expansionist, even though State A was really just motivated by fear of State B. So State B increases its arms (or takes some other such action, such as invading a buffer state or establishing a colony). This then confirms or increases State A’s belief that State B must be an aggressive type.

The net effect, according to this story, is a “spiral of hostility” that might lead to war, especially a preventive or preemptive war in which one state thinks it needs to act against the other before it is too late.  Substitute North Korea for State A and the United States for State B for the example of the moment.

The security dilemma story has some puzzling features.  For one thing, it is not clear whether the inferences driving the spiral of hostility make sense. Why shouldn’t State B realize that State A would arm, whether it was an aggressive type or just afraid of B, in which case seeing A arming itself does not provide B with any new information about whether A prefers the status quo or aggression? The same is true for State A’s inferences about B.  Furthermore, if the states’ leaders both just want to preserve the status quo, why can’t they signal this by showing some restraint?

Robert Jervis’s classic statement of the problem in his “Perception and Misperception in International Politics” starts out saying that the “spiral of hostility” might occur between states with rational leaders. But then he pretty quickly shifts into arguments about how psychological biases drive the unnecessary escalation, arguing that state leaders tend to see themselves as nonthreatening and to assume, incorrectly, that others understand that they are not threatening or aggressive also. As a result, when State B sees A arming itself, its leaders think: “They know that we pose no threat to them, so this can only mean that they are aggressive and want to attack us or do something with those weapons that we won’t like.”

However, the North Korea crisis is probably not the result of this security dilemma dynamic

In the security dilemma story, escalation and war might happen even though both sides are in fact just interested in maintaining the status quo. Neither has any underlying ‘revisionist’ aims (i.e. aims to change the larger international situation and their own role in it) that the other side needs to worry about.

That’s not the problem between the United States and North Korea.  U.S. leaders have long hated the North Korean regime and would love to see it gone, just as successive U.S. administrations hated the Gaddafi regime in Libya. And there are truly excellent reasons for the leaders of a democratic country — or really anyone who cares about human rights — to hate the North Korean government and want to see it change.

But from Kim’s perspective, the fact that the United States would ideally like to see him and his gang of thieves and murderers gone makes the United States the revisionist power.  Secretary of State Rex Tillerson can say in public that “we do not seek a regime change” and that “we are not your enemy,” but Kim knows that this is cheap talk. If the opportunity for regime change arose, there is no way that the United States (and South Korea) could make a credible promise not to support opposition to Kim. Just look at Libya. Gaddafi negotiated to end his nuclear program, and then later NATO intervened to help overthrow him when there was an uprising.

So this is not a situation where two states each mistakenly worry that the other is hostile. Rather, it’s what international relations scholars might call a classic problem of anarchy, in the sense that there is no power above states that can enforce agreements they make with each other. Here, the United States can’t credibly commit not to take actions to help depose Kim in certain circumstances. Kim thinks, with some reason, that being able to strike the U.S. mainland would lower this risk. This in turn leads U.S. leaders to think about a preventive war.

Of course, there may be other dynamics in play. In particular, some part of Kim’s nuclear policy is aimed at increasing his legitimacy with domestic audiences, shoring up legitimacy and status, and so trying to lower the odds of a domestic uprising to begin with.

Questions follow about Trump’s approach

What is puzzling is that the Trump administration seems to think that threats and coercion from the United States and China can work to get Kim to agree to stop his missile program in a verifiable way. Defense Secretary Jim Mattis, national security adviser H.R. McMaster and Joint Chiefs of Staff Chairman Gen. Joseph F. Dunford Jr. have said things in recent weeks that suggest a real willingness to try a preventive war rather than allow North Korea to acquire a fully functional capability to hit the U.S. mainland. This isn’t even to mention Trump’s more extreme statements. At the same time, Trump and his advisers also say things that suggest they want to use the threat of war to press the North Koreans to make a deal (including via pressure from China).

But threats and coercion just reinforce Kim’s sense that his safety — from the United States and China — requires a working nuclear weapons capability. Due to the commitment problem just described, he is likely to try to develop this capability whether we make threats and raise nuclear risk or not. So the question is really whether the United States would be willing to undertake a preventive war at this point. Why rattle sabers and draw red lines if you don’t have a clear path to a feasible deal?