Vipin Narang, a political science professor at MIT, thinks many moves ahead in the international chess of modern deterrence.
As an undergraduate, I didn’t spend a lot of time in the library. In part, as you probably guessed, because I was a fairly indifferent student. But there was also the problem of the loopy barking man in the corner. None of us had any idea who he was or what he was doing making such a racket in the reading room – possibly someone off the street who had cadged a student ID?
Well, everybody knows now, and not just because he won a Nobel Prize in economics or (far more significantly) was played in a movie by Russell Crowe. Between those canine outbursts, John Nash’s beautiful mind was pushing forward on game theory, as were other economists including John Harsanyi, who shared the Nobel with Nash, and Thomas Schelling, who won it a decade later. Nash’s work is largely theoretical – although vital if you’re willing to spend your idle hours contemplating the prisoner’s dilemma (sorry, it’s no help with Fortnite). Schelling’s oeuvre is more practical, especially if you spend your idle hours contemplating nuclear Armageddon like yours truly.
So, what’s the connection, exactly? I had no clue, until I talked to Vipin Narang, a political science professor at MIT. While not a game theorist himself, Narang’s beautiful mind is pushing forward on how traditional nuclear deterrence strategy can be modernized for the new era of great-power conflict. Here’s a lightly edited version of the first half of our conversation (the second part will be published next week):
Tobin Harshaw: Before we look at the challenges facing the U.S., can you briefly describe game theory and how it relates to nuclear deterrence?
Vipin Narang: Most of the original work on deterrence was done by the Nobel-winning economist Tom Shelling in his very influential book "Arms and Influence," which came out in the mid-’60s. Back then it was rudimentary game. But it's really about a strategic logic -- how your adversary behaves based on your moves and how you react to their reaction to your moves. So it can get formalized in game theory. But I would say that most deterrence theorists now use a loose form of a kind of strategic game in the back of their minds when it comes to deploying particular capabilities or signaling about when we might threaten to use nuclear weapons.
TH: How does that play out on the ground?
VN: It's not just what we do and deploy; it's what the adversary fears we can do. Its perception of what we are doing may not map to what we're actually doing. And that can drive outcomes in ways that we don't necessarily expect from formal game theory. Take the case of North Korea. Kim Jong Un has a very small, rudimentary arsenal. We may not want to overtly threaten the survivability of that arsenal when we run these drills up around North Korea, but Kim may fear that. We have to internalize that the fear of a disarming attack leads Kim Jong Un to do certain things, such as potentially devolving command of his nuclear weapons very early in a crisis, out of fear of a use-them-or-lose them situation. And so things that we may think are innocuous may not be innocuous to Kim. That has real implications for stability. Those are the kinds of dilemmas and twists on classic deterrence theory that game theory doesn't necessarily count for particularly well, but are very real when it comes to the practice of deterrence.
TH: What's meant by Kim "devolving" control the arsenal? Would generals on the ground take over? Would it all be on a sort of autopilot to Armageddon?
VN: We don't know a lot about North Korean command and control. In the management of nuclear weapons, at some point states envision giving firing ability to the units that possess the nuclear weapons. In order to launch, at some point you have to actually transfer the nuclear weapons to the military end users if they do not possess them during the normal state of affairs. The U.S. and U.K., for example, delegate and devolve this authority and ability during peacetime. We're so confident in our command and control that we can do that. Countries like China maintain very assertive, centralized control of the nuclear force during peacetime. But in the North Korean case, the concern would be that in a crisis or a conflict, he might lose the ability to communicate with his military and the people who would have to actually fire the weapon. So while Kim may maintain centralized control during peacetime, as soon as he feared a crisis or war was about to unfold, he may have to quickly devolve use ability to his military so that that his nuclear deterrent doesn’t become impotent.
TH: Do you think then that for the U.S. it’s actually trickier to deal with a North Korea or potentially Iran than it is with China or Russia?
VN: Absolutely. One of the biggest problems in deterring new nuclear states with small arsenals is they tend to have itchy trigger fingers. They haven't practiced, they haven't gone through the pressure of crises. They don't have a force size that they're confident enough can survive a conflict, especially against the U.S. There is a real concern that new nuclear states may behave in ways that classical deterrence theory may not predict. The U.S. and Russia have spent 75 years developing a very sophisticated command and control system. And even we have problems. So you can only imagine what problems that a North Korea or a future Iran or a Pakistan has if they were faced with a crisis or conflict today.
TH: In that sense, it parallels conventional war. We have a lot more trouble with a Cuba than we do with China.
VN: Exactly. It's the small, scrappy ones that have nothing to lose that may resort to asymmetric means early. That's very true with nuclear weapons also.
TH: The leaders of the so-called intelligence community were up before Congress the other day, and a lot of the things they said seemed to contradict what's come out of the White House. The North Korea situation was one of those. Do you think that discussions like the upcoming summit, if it happens, are kind of pointless?
VN: No, there's a world in which the White House and the intelligence community are on the same page. President Trump has pointed to the fact that North Korea has not tested ballistic missiles or nuclear weapons since 2017. And that is visibly and demonstrably true. What the White House has been quiet about is the fact that North Korea has continued to produce ballistic missiles, nuclear weapons material and probably nuclear weapons themselves, and has deployed them and is maintaining missile operating bases because it never said it wouldn't. And in fact, Kim's New Year's Day speech in 2018 ordered the mass production of the nuclear weapons and ballistic missiles they tested in 2017.
So after the Singapore summit, when president Trump said he had solved the problem, there was a concern that he didn't realize that shuttering the test sites doesn't mean that they're not producing nuclear weapons and ballistic missiles and actually deploying them and trying to improve survivability. But over time as these leaks came out, it became apparent to a lot of us who study this that it's not that Trump didn't know they weren't disarming, he just doesn't care that they're not disarming.
TH: So what's the aim right now?
VN: I think for the summit process to be success right now, the administration has quietly given up on the goal of immediate North Korean disarmament in one shot. It's just unrealistic. Kim is not going to give up these nuclear weapons. He worked so hard to get them. They have bought him so much international political legitimacy, and two summits with the president of the United States. And he has a deterrent against a U.S. invasion. It is good to be a nuclear weapons power — and the problem is other countries like Iran are almost certainly taking note of that.
So a realistic immediate aim in this next round would be something that talks about slowing the program down, to slow down the production of fissile material at particular facilities so he can't build more nuclear weapons. We know they're producing uranium not only at the Yongbyon plant but at suspected covert facilities. The reactor at Yongbyon produces plutonium for nuclear weapons and tritium for thermonuclear weapons (H-bombs). If you can get them to slow down production by shutting down even the reactor verifiably and permanently, that's the first step in slowing down the growth of the program. You can maintain a long-term goal of dismantling and disarming. Even if it's not going to happen in our lifetime in all likelihood.
TH: What do we concede to get to that first step?
VN: Right, what's the price? And whether the ask is that the U.S. remove nuclear weapons from the region and reduce the footprint of our forces, along with sanctions relief. And you know, Kim gets Trump alone in a room, and Trump has long been skeptical of the value of military alliances. And maybe you can talk him into giving some sanctions relief. The reason why the working-level groups have seemingly not made a lot of headway is because the North Koreans realize if they slow walk that and get Trump in a room, they have a better shot of getting concessions.
TH: Last time Trump promised to cancel military exercises with the South Koreans, and that seemed to hit even Defense Secretary James Mattis by surprise.
VN: Trump thinks these things are a waste of money. I don't think he sees the value in readiness and what these exercises do. And I think this is kind of the North Korean objective. In a lot of ways, for the summits to be useful for North Korea, the agreements need to be as vague as possible so they don’t commit North Korea to much and they can exploit ambiguity to continue to produce nuclear weapons. The reason the statement that came out of the Singapore summit was as vague as possible is because the North Koreans are experts at exploiting loopholes. To this day we don't have a common definition, for example, of denuclearization. And that was by design. That text in all likelihood I think was provided by the North Koreans. The worry would be that a kind of sequel happens in Hanoi this month, and you get vague North Korean language with loopholes large enough for them to drive their ICBMs through. For it to be worthwhile for us, though, we need something very concrete to hold North Korea to its commitments. Otherwise, I don't know how many more rounds we can go without anything concrete in writing from the North Koreans.
TH: One last question kind of related to North Korea. And perhaps it’s a silly question because it's not going to happen. But there's talk of Japan and South Korea going nuclear.
VN: Never say never.
TH: I should know better. So would that be good or bad?
VN: From our perspective, allied proliferation is a problem for two reasons. One, the more nuclear weapons in the system, the more risk of actual use. The second is from an alliance management perspective, the U.S. has always opposed our formal allies acquiring nuclear weapons. We prefer extended deterrents where we maintain control of the proverbial button in order for Washington to maintain use-initiative and escalation-control initiative. We do not want allies being able to start a nuclear war that the U.S. has to finish. We were even opposed to the French acquiring nuclear weapons. I don't think we're anywhere close to the point yet where Japanese public opinion would shift that far. South Korean public opinion, actually, has long supported the acquisition nuclear weapons at surprisingly high numbers. But they are much further away than Japan and, presumably, China would oppose a South Korean nuclear weapons program and probably do something to stop it.
TH: Prime Minister Shinzo Abe has talked about it.
VN: That's right. And he is in a position to elevate the status of the self-defense forces in conventional readiness. And there have been official legal findings in Japan even since the 1950s about the pursuit of and possession of nuclear weapons, which the Japanese will classify as defensive, that may actually be consistent with Article IX of the postwar treaty that forbids offensive warfare. The problem from an alliance-management perspective would be that Japan would have the ability to use nuclear weapons on its own initiative and to escalate a conflict in ways that Washington may not want. This actually gives our allies some leverage against us, and Japan has laid the groundwork over 50 years to do this, and sometimes implicitly used the threat of developing its own nuclear weapons to elicit greater reassurance commitments from the U.S.
Ultimately, though, these are sovereign decisions. So if Japan decides that the U.S. is no longer capable of meeting – or willing to meet -- its extended deterrence commitments, it could move ahead provided the government overcame public resistance. The problem with extended deterrence is there's an infinite appetite for reassurance, and the allies always want more reassurance than we're able to provide. So far it's been enough. Our commitment to Japan is pretty rock solid because of the significant conventional deployments we have on Japanese territory. But if there comes a day a where a Japanese government decides that the U.S. no longer meets its security needs and they decided to acquire an independent arsenal, that would really have reverberations across the East Asia security architecture because China would be strongly opposed. And then South Korea would have a big decision to make. Even under the current circumstances in which Japan faces a North Korean threat and there are broader concerns about America’s commitment to its allies, Japan developing its own nuclear weapons is highly unlikely. But never say never.
This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.
To contact the editor responsible for this story:
Philip Gray at firstname.lastname@example.org
Tobin Harshaw is an editor and writer on national security and military affairs for Bloomberg Opinion. He was an editor with the op-ed page of the New York Times and the paper’s letters editor.