précis Interview: Vipin Narang
Vipin Narang is assistant professor of political science at MIT and a faculty member of the MIT Security Studies Program.
VN: That is my undergraduate class. The Security Studies Program already has classes on the technology of nuclear weapons, especially at the graduate level. I wanted to teach a class on what states do with nuclear weapons once they have them. For example, how does deterrence work? A lot of classes like this focus on the Cold War. I focus equally on the regional powers. There are seven regional nuclear powers, but the division of labor in the field is overly focused on the super powers. The regional powers have chosen different nuclear strategies, so I give the students a different view of the proliferation landscape than other similar courses offered across the country. Any future nuclear states will look more like this distinct class of states than the superpowers.
I think the students enjoy it. I get a lot of students from nuclear engineering and political science. I start with theory, then discuss the super powers and regional powers. The students really like the discussion of modern topics such as nuclear black markets.
VN: I am just starting that this semester. There are not many classes anywhere on this subject, so it is trial by fire. I wanted to teach about South Asia as a regional system together. This class is in the foreign policy analysis tradition—focusing mostly on India and Pakistan. There are issues on the subcontinent that travel to other regions, but there is increasing focus on security issues in South Asia itself.
There are quite a few students from MIT and from outside that are taking it. We focus on internal politics, relations in the region, and external relations with the United States. For the graduate students, it is about a research paper which will hopefully be something they can try to publish or incorporate into their dissertations. I try to have them play with the substance of the theories as applied to particular puzzles and problems in South Asia.
This has been a more difficult class to prepare because the state of the literature on South Asian security is not as well developed as on Chinese security, for example. In India, they don't write doctrine openly—you need clearances, which can take years to get—so this limits the state of the literature. Usually, sources are media reports, so you have to be careful to distinguish between popular conception and what is actually happening.
VN: The traditional quantitative methods sequence does not always have applied examples. In most schools, the sequence is about learning to do the methods more than looking at how they are applied, though at MIT there is a fair amount of the latter. Nevertheless, there are problems unique to international relations and security studies that motivated a separate class to look at how these methods are applied well or poorly. In general, the methods are more easily applied in the American and comparative politics sub-fields. The international relations subfield has lagged a bit behind because we focus on macro questions. So, the class is trying to show where the state of the field is. We look at specific methodological issues and assign articles that try to address those problems. We then ask how successful those pieces are. The class complements the strong security curriculum by exposing students to the methodological challenges of doing large-N work in security studies.
The final paper asks the students to assess an existing article. They had fun dissecting data sets. The students found that a lot of results are unstable and that some scholars sometimes overstate the robustness of results. The take home is that we need to be transparent about how strong results are. All results break at some point—in international relations, maybe a little quicker. None of these findings, on their own will stand. So, we always want some theory and some actual cases where the correlation exists. In the battle between methods and research design, good data and design always beats good methods. The problem in international relations tends to be that the data isn’t very good.
VN: Many expected the subcontinent to be stable after India and Pakistan acquired nuclear weapons. But, Pakistan has a declared first use policy. It threatens to use nuclear weapons to deter conventional power. That policy has allowed Pakistan to support sub-state groups that can then perpetrate mass casualty attacks on Indian metropoles.
India, on the other hand, is trying to primarily deter nuclear use. That is more of a second strike strategy and posture. This has created a discrepancy where Pakistan’s posture is really relevant to the situation on the subcontinent and India’s is not as relevant. This has created the conditions for periodic attack and some conventional crises. There is an assumption that all attacks can be deterred against nuclear states. But, Israel had nuclear weapons but it was still attacked—Saddam still launched missiles. So, deterrence may depend on how a state operationalizes that capability.
Part of the choice of nuclear posture has to do with the state’s goals. These are probably net optimal strategies. India has strong and assertive civilian military relations. I think that explains its retaliatory strategy. China is a similar case. India, because of its size, can survive some attacks. Pakistan cannot afford the same tradeoffs because of its geographic disadvantages.
VN: The large-N part of my dissertation is forthcoming. It is complementary to my earlier article on India and Pakistan that argued that Pakistan’s strategy is deterrence optimal and that India’s isn’t. This looks at other states and conflicts and shows that the argument has wider applicability.
VN: Absolutely. I had my book workshop last month. Having SSP faculty there—they have thought about these issues for a long time—has really sharpened the theory, arguments, and structure. They pointed out how my work relates to old Cold War debates. It has really improved the manuscript and the arguments in general.
I have been linking it to longer term debates. The US and Soviet Union believed that postures and strategy mattered. Then, at the end of the Cold War, with the growing number of nuclear powers, scholars and policymakers adopted a new outlook that even small or 'existential' nuclear forces were believed to be game changers for states, and that strategy and posture didn't matter. But, I argue that issues of strategy and posture can still really matter as we learned earlier in the Cold War.
Also, one of the best things about MIT and SSP is that the graduate students are really smart. You don't get this anywhere else especially in security studies. It's something special.
VN: The way I look at it is that there is a popular belief that Iran will look more like Pakistan than India as a nuclear power—that it will be aggressive, that it will use nuclear weapons as a shield behind which it will unleash its proxies. But, Iran and Israel don't share a border. So, ground power is not as relevant to Iran as India's is to Pakistan. And, the emboldening aspect of nuclear weapons may not be as relevant in Iran's case. We don’t know a lot about Iran's civil-military organs. They have these stove piped organizations. Will it have assertive or delegative command and control structures? If the structures are centralized because of regime paranoia, Iran's nuclear posture will probably look more like India's. In Pakistan, the military runs the program from cradle to grave. It remains to be seen, but I have an open mind as to whether Iran will go one way or another. There are lots of indicators that it will go like India, using nuclear weapons to deter nuclear use and an existential threat to the state. Pakistan, about 10 years after it acquired nuclear weapons, underwent a lot of organizational cooption of the nuclear program to adopt a first use posture with delegative control.
In the case of Iran, I would tend to focus more on the unit level, domestic variables. As I said, we don't know a lot about civil military relationship in Iran. It is not monolithic in any country. There are trusted and less trusted organizations. It really depends on center-military relationship. The more we know about that the more we can get a fix on what the Iranian nuclear strategy might look like. But if I were to guess, I'd expect that Iran will have highly assertive control with an assured retaliation capability like China or India.
VN: That is a tough question. Part of the problem in Afghanistan is that Pakistan's paranoia about India has always led Pakistan to want strategic depth in Afghanistan even without knowing what that means other than Afghanistan not having an alliance with India. The Pakistanis likely have the upper hand once we draw down given their relationships in Afghanistan. It will be difficult for India to compete without a US presence. India lived with Taliban in the 1990s, and, though there were airplane hijackings, there was no existential threat. India will have to come to terms with Pakistan having more influence there. As long as India doesn't overreact, the situation will be stable. I think India will focus on internal growth and likely won't overreact. To the extent that India can monkey around in southern Afghanistan, it may, but it isn't going to put its neck on the line to do that. India has not traditionally been an overreaching power. It has enough internal problems to deal with and their covert capabilities aren't as capable as Pakistan's.
VN: In a crisis, things can get a little hairy. The nightmare scenario is that Lashkar-e-Taiba gets a hold of the weapons. They are as well or better best positioned to acquire a nuclear weapon as any group in the world. Right now, it is believed that the weapons are pretty well controlled by the security services during peacetime. If that changes, the easiest way to get a weapon is to engineer a crisis that makes the Pakistanis disperse their nuclear weapons, making them more vulnerable. The LET could try to precipitate a crisis by attacking India and then making a run at nuclear weapons. It could also pose an insider threat based on its relationship with the army. In general the stewardship procedures are probably pretty secure. But, if the Pakistanis get concerned about survivability of the arsenal, things could change. It also depends on how they move the weapons. If they move them on the road, they will want to reduce signatures, so they could have less convoy security. These aren’t high probability scenarios, but they aren’t impossible.
VN: The US probably should not talk so much about render safe options through [special operations force] neutralization, which would be a very difficult operation. On the other hand, Pakistan is going to assume we are working the problem anyway, so there is not much we can do. They are paranoid about Indian and American threats to the survivability of their arsenal. Quantity has a quality all of its own, so they may feel less insecure as they get more weapons. After the Abbottabad raid, they are definitely more paranoid about what the US can do. Whether we have that capability or not, they will probably want to move things around a lot so real time intelligence becomes impossible. The US would only probably consider render safe options in the most extreme of scenarios. And, the probability of success without a nuclear weapon going off.
There was a recent Atlantic article that reported the Pakistanis may do this in peacetime after Abbottabad, but it could just as easily be one big fake out. If the paranoia is so high and they want to keep this shell game moving, you could imagine the dangers. The US has reiterated the security of Pakistani arsenal in peacetime, so they may not be doing this now.
VN: I am trying to get my book manuscript out. Paul Staniland and I are working on a project on Indian security—the content of Indian strategy, drivers, the ideological landscape—so we are trying to spend some time on that as well.