Can the Humanities and Social Sciences Improve National Security?

We are very pleased to have a guest post from Jacob N. Shapiro, an assistant Professor at Princeton University, on the contributions quantitative social science and the humanities can make to important questions of (national) security. Jake is one of the leading young scholars in international security. We have blogged before on his research on support for Islamic militants in Pakistan and financing of terrorists.


Can the Humanities and Social Sciences enhance National Security? This seems like a relevant question with NSF funding to Political Science under Senator Tom Coburn’s microscope once again, not to mention the coming budget crunch coming that’s going to squeeze federal support for all manner of research. I recently had the chance to do some thinking on it for a May 19thbriefing the National Humanities Alliance and Association of American Universities (AAU) put on for Congressional staffers. So when my kind hosts at the Monkey Cage asked me to do a guest post on the subject, I readily agreed. (Full disclosure, as a former Naval Officer with close friends still deployed this is a subject I am passionate about.)

My co-panelists offered up a lot of wisdom that was nicely covered by Inside Higher Ed (“A National Defense”). Their comments mainly focused on the role the Humanities can play in advancing the kinds of cultural understanding that form a   necessary backdrop to formulating sensible policy in most any setting. One novel argument did come out of the panel and it was embedded in Debra Hess Norris’ description of her work providing training on photo-restoration in Iraqi Kurdistan. As she described it the work built a series of trusting ties between her team and archivists from all over Iraq, amounting to a kind of bottom-up Track II diplomatic effort that may pay dividends for years to come.

Nice as the role of academic research in building deeper understandings and closer ties is, however, there are some underappreciated ways in which regional expertise and area studies can interact with hard-core social science to inform national security policy. So, in a fit of shameless self-promotion I want to mention some of the survey research I’ve done in Pakistan with Graeme Blair (Princeton), Christine Fair (Georgetown), Kosuke Imai (Princeton) and Neil Malhotra (Penn). I like to think this work shows how pairing deep area expertise (Christine) with thoughtful application of social science method (the rest of us) can help provide insight on some big national security questions.

This is a line of work and that’s motivated by a series of big questions relating to policy in Pakistan including: “How do local perceptions of the United States and its actions relate to support for violent groups?”; “Can we take actions that make terrorist tactics in Pakistan and against the United States less legitimate?”; and “Which Pakistanis actually support militant organizations?” There are questions of great importance well beyond U.S. concerns. By their government’s estimates Pakistan has suffered more than 30,000 military and civilian casualties from terrorism since 2001. So, if we can understand what drives support for militant groups in Pakistan, that’s a big deal.

Where the social-science to humanities link comes into this research agenda is that doing good work on this issue requires understanding more than the concepts we think Pakistanis are going to work with, or how we think they should think about terrorism, but understanding how they actually do.  This is a point Chairman Leach of the NEH made in his opening remarks at the May 19th briefing and it bears elaboration so I want to give a very specific example.

In the spring of 2009, we did a large, 6,000 person survey in the four main provinces of Pakistan. One of the questions we had on the survey was designed to measure attitudes towards al-Qa’ida. This turned out to be tricky as we learned during enumerator training when most of our enumerators mostly didn’t know what we were talking about. They would say, “So, what is that?” This was striking as these folks were an educated bunch, all had some college and most had some graduate education and 4-5 years of survey research experience.

What we learned thanks to Christine’s expert intuition and lengthy conversations with the team is that they understood Osama bin Laden qi tanzeem, which loosely translates as ‘Osama bin Laden’s militia.’ Basically, the average Pakistani understood that there was this guy out there named Osama bin Laden, that he had a militia, and that the U.S. really hated him. But they didn’t link that with this transnational terrorist organization that had committed the attacks on 9/11. The valance of the organization to us was totally different than what is was to them.

What this meant was that surveys went out and asked people questions such as, “Is al-Qa’ida a threat?” or “Do you approve of al-Qa’ida’s actions?”, they weren’t actually measuring the thing we care about. In this case the area studies knowledge, and the deep, years-long interaction with Pakistanis that was required for us to understand this nuance (again, Christine), played a big role in our ability to measure support for al-Qa’ida and other militant organizations in Pakistan.

So that’s the role for the humanities. Social science’s role in enabling this research was really in figuring out a way to measure support for groups in an environment when asking about them directly is: (a) downright unsafe in many parts of the country; and (b) so sensitive that one gets very worried that respondents are saying what they think enumerators want to hear, not what they really think. To get around these issues we used an ‘endorsement experiment’ in which we measure differences in support for various policies unrelated to militancy between two experimental groups—those told only about the policy and those told a militant organization supports the policy. Because we randomly assign respondents into these groups, the difference between the two conditions reveals how policy support increases or decreases as a consequence of being associated with a militant group, and is thus an indirect measure of support for the group. Unlike a direct measure, safety, non-response and social desirability are less prominent since respondents are reacting to the policy and not to the group itself. Relative to list experiments, the other prominent way of studying such sensitive attitudes, endorsement experiments do more to minimize safety concerns as they do not involve asking people how they feel about sensitive groups.

What did we find when we assessed support for militant organizations using the combination of this subtle measurement approach and a deep appreciation for the particulars of the local context? Four findings stand out. First, we found that most Pakistanis dislike militant groups. Second, people living in violent parts of the country strongly disliked these groups relative to their countrymen in more pacific regions (Bullock, Imai, and Shapiro 2011). Third, we found that poor Pakistanis disliked the militant groups more than their middle-class countrymen, an effect driven by the intense dislike of militants among the urban poor (Blair, Fair, Malhotra, and Shapiro 2011). Fourth, among Pakistanis who believe the basic factual claims behind the narrative which says militants in Pakistan are fighting for freedom and democracy, support for core liberal democratic values positively correlated with support for militancy (Fair, Malhotra, and Shapiro 2011).

At the end of the day as we’re debating what our strategies should be towards reducing the threat of terrorism in and from Pakistan and other countries, insights like these seem like they might be useful in crafting policy. These kinds of insights are only possible though, when we have a combination of two things: (1) generous grant funding which enables us to do these large projects – surveying 6,000 people in all parts of Pakistan takes a lot of money; and (2) close collaboration between the humanities – to understand the details of the situation – and the social sciences – to know how to analyze these big-picture social trends. That combination is supported today by NEH, NSF, and the other granting agencies in the federal government and is making a real contribution to national security.

[Cross-posted at The Monkey Cage]

Erik Voeten

Erik Voeten is the Peter F. Krogh associate professor of geopolitics and global justice at Georgetown University.