Terrorism Research Has Not Stagnated

Tim Foley

Tim Foley

We are puzzled by Marc Sageman’s assertion that terrorism research is stagnant.

We wholeheartedly agree that several issues regularly hinder progress. While both qualitative and quantitative research on terrorism are necessary, scholarship has gone too far in the direction of quantitative methods. A prejudice in favor of elegant models and statistics, sometimes based on poor-quality data, is not unique to terrorism research. It reflects a fashion in social science more broadly.

It is also true that there is no single academic discipline of terrorism research, which is defined by its multidisciplinary nature. Terrorists are influenced by relations among nations and their immediate communities; by the identities they choose, the groups they join in the digital or analog worlds, the costs of the choices they make, their individual psychology. They come from a wide variety of countries, backgrounds, and languages. Expertise from a variety of disciplines is essential to understand the phenomenon.

The scholars who “took the bait” as federal money poured in, as Sageman puts it, were responding to a genuine need. In their zeal to protect the public, government officials may have overreacted to their perceptions of the Al Qaeda threat. But contrary to Sageman’s claim, few serious terrorism experts ever saw Al Qaeda as an existential threat to the West. Our role as scholars is to discover and understand what influences the changing nature of terrorism and the impact it will have.

Sageman’s essay also decries the split between academe and the intelligence agencies that study terrorism and train analysts. But as Thomas Hegghammer, of Stanford University, tweeted in response, the article gets at only part of the problem: “academe’s prejudice against terrorist research. Few uni’s want to do it, so agencies have to train in-house.” Gregory Johnsen, a leading expert on Yemen, responded: “If you research terrorist groups you will likely kill your academic career before it starts.”

Many social scientists do seem to view terrorism as a figment of the imaginations of terrorist researchers. No question, most domestic extremists in the United States have been inept and foolish. But the vast majority of conflicts today involve violent, nonstate actors.

Sageman raises another issue deserving attention: Scholarship deemed “serious” usually involves quantitative studies, which are only as accurate as the data on which they rely. Even poorly conducted quantitative analysis has an air of rigor to it. And ethnographic research takes an extraordinary amount of time, patience, and effort. Nor is it easy to win approval of the boards that protect human subjects: Even if those subjects are wielding Kalashnikovs, we must demonstrate that our interviews will not harm them psychologically or legally.

How can we fix those problems? Certainly not by crudely calling an entire field of study “stagnant.”

Our main problem with Sageman’s article is not that his concerns are unwarranted but that he appears oblivious to the kinds of research actually being done to determine who becomes a terrorist, why, and how many there are in the West. We see a field in ferment, not stagnation, with the entry of more young scholars trained in an ever-wider variety of disciplines.

There are far too many studies to cite, but consider a few. A recent article by Hegghammer in the American Political Science Review shows that Western jihadists are more likely to join jihads in foreign countries than they are to fight at home, as the Boston attackers did—even though Al Qaeda leaders are promoting “stay home” terrorism. When jihadists in the West do remain at home, more often than not they are radicalized by veterans of foreign jihads or by their own trips abroad. According to Hegghammer’s research, if the suspects in the Boston bombings truly radicalized themselves, they would be unusual. Hegghammer developed his own database, which he admits is a first pass at the problem. Were a government agency to provide better data, his findings would presumably be more robust.

A study by Paul Gill and colleagues at the International Center for the Study of Terrorism, at Pennsylvania State University, has traced the behavioral signals that individuals who turn to violence reveal and reliably delineates different behavioral profiles among lone-actor terrorists.

There is also a growing literature about the radicalization of groups. Two studies—one by James W. Pennebaker and another by Margaret G. Hermann and Azamat Sakiev—have shown that, in the period before an attack, violent groups in Yemen changed their rhetoric in measurable ways. Daveed Gartenstein-Ross and Laura Grossman have identified six indicators of jihadist radicalization in the United States and Britain, including a tendency to adopt legalistic interpretations of Islam and to attempt to impose religious beliefs on others.

Also relevant are studies of foreign groups attacking the United States, including work by Gary LaFree, Sue-Ming Yang, and Martha Crenshaw. They examined 16,916 attacks attributed to foreign terrorists deemed dangerous to the United States from 1970 to 2004: Just 3 percent were directed at the United States.

That an academic field has not provided all the answers to fundamental questions about human behavior is not evidence of stagnation. The experience of other disciplines suggests that progress will be made as scholars painstakingly build on one another’s work. The National Institute of Justice has recognized that, and it is now financing studies on domestic radicalization.

It is also important to point out that Start, the center for terrorism research at the University of Maryland that Sageman cites as producing one of the few global databases (what he calls “crude” descriptions of terrorist incidents), also promotes qualitative studies. According to Start, about 30 percent of its projects include a qualitative component. Next on Start’s (and others’) agenda is integrating multiple levels of analysis—world events, community and individual factors, organization activities, small-group dynamics, social network, discourse, content, and more.

Moreover, an increasing number of universities are responding to the change in the security environment. In addition to the two terrorism centers Sageman mentions (Start and Create, at the University of Southern California), Penn State’s center brings together researchers from around the United States and abroad to learn about its research. Northeastern University and the University of Massachusetts at Lowell are setting up major programs to bridge the academic-practitioner divide.

Research is likely to be financed by federal sponsors today only if it can demonstrate its relevance to policy and operational concerns. At a time when political scientists are struggling to retain what little funds they receive from the National Science Foundation, the challenge of effectively “translating” academic research into policy has never been more urgent. Sageman is correct that it can best happen with greater dialogue between academics and those given the task of responding to terrorism.

We also agree that it would be good to include in the mix more ethnographic studies of how, why, and where individuals have turned to terrorism, rather than inferring their motivations from their actions or public statements. We feel strongly that former terrorists can be interviewed safely and effectively, and that students can be taught how to do so. But that requires financing sources to recognize that a multiplicity of methodologies is needed. As the military strategist Bernard Brodie once warned, elegant methods can be seductive, even when they are inappropriate.

Perhaps it is finally time to make terrorism studies a more formal area of scholarship.

John Horgan is director of the International Center for the Study of Terrorism at Pennsylvania State University. Jessica Stern is a former member of the National Security Council staff, a member of the Hoover Institution’s Task Force on National Security and Law, and a fellow at Harvard University.

Return to Top