How did your education shape your direction in research?
I was fortunate to grow up in Englewood, N.J., a diverse community with a single public high school. I think my earliest research project was a study of seating patterns, by race and neighborhood, in my high school's cafeteria. I found that social relations among my peers easily spanned the neighborhood, but remained segregated by race. I didn't think of this as a network study at the time.
As an undergraduate at Brandeis University in the late 1960s, I focused on social theory and broad questions of values, and I chose my graduate school, Harvard, because it didn't have a methods requirement.
You've both taught and studied at very prestigious institutions. Who has had the most profound impact on your career and the path you took in academia?
I would say Harrison White, who became my graduate advisor in sociology at Harvard, coauthor, and, when I joined the Harvard faculty for six years, senior colleague. I think of him primarily as a social theorist, and he was leading a truly revolutionary shaking up of the field of social network analysis methods. I have been working on network models and theory ever since.
What drew you to the Consortium and its research?
As a sociologist I knew Gary LaFree's work, and first heard about START when he took up the directorship, but my first contact with the Consortium was much more round-about. I had chaired a 2002 workshop at the National Academy of Sciences on social networks and national needs, particularly national security in the wake of 9/11. Six years later I participated in a network seminar at the Defense Threat Reduction Agency (DTRA) and became very interested in thinking about social network applications to counterterrorism.
I wanted to apply for a grant from DTRA, but I knew I didn't know much about terrorism or weapons of mass destruction. On the basis of a Google search I cold-called Victor Asal, who soon led me to Karl Rethemeyer, both START-affiliated researchers at the University at Albany, and to Gary Ackerman, now Director of START Special Projects. In 2010 we together were awarded a DTRA grant along with my Arizona colleague Brint Milward. START Senior Researcher Lauren Pinson has also been crucial in supporting our research. The six of us have enjoyed working with one another ever since, and so my relations with START have correspondingly flourished.
I have always felt that START welcomes researchers with strong analytic abilities. I know I've learned a lot from terrorism experts and people with government experience, and I oppose the idea of boundaries between subject experts and methods experts. In fact, I think it is very important to maintain a conversation, as it enriches in both directions.
Tell us a little about the TANC project. What are your plans for its application?
TANC stands for "Terrorist Attributes to Network Connections." It's not always feasible to obtain good social network data on relations among terrorist groups. We want to explore this concept using attributes and behaviors of such groups -- such as number of previous attacks, membership size, whether territory is controlled, whether state sponsored, and so forth -- to form connections among them based on their degree of profile similarity.
It may be that such networks are good predictors of social network ties. Networks of profile similarity are of great interest in their own right, apart from their degree of correspondence to social networks. Although many researchers are working with profile similarity, my particular interest is to show that such networks among "cases" are fundamental to regression analysis and its generalizations, which are the main work horse of quantitative social science.
How have you been able to incorporate your START research into your classroom teaching at the University of Arizona?
I regularly teach a seminar on network analysis, which recently had an enrollment of 25 graduate students drawn from nine departments across five different colleges at the University of Arizona. My DTRA research carried out with other START colleagues not only appeared methodologically in several topics in the course, but motivated me to discuss many examples of social network research countering terrorism, and also (not incidentally) to include a section on the ethics of social network research that touched on several different aspects of government and military applications.
What are your plans for future research?
I have a longstanding line of interest in cultural sociology, especially in the intersection between networks and culture (defined not in terms of national character and ultimate values, but more locally as the practices people adopt and what they think is meaningful). I'd like to write more in that area, though it turns out that culture, networks and terrorism studies are becoming increasingly intertwined. My work on the DTRA project has spun off some research projects in other areas, and I've been involved in planning a couple of collaborative grants between social and computational scientists.
What have been some of your most profound or interesting findings in the application of your expertise in social networks with terrorism studies?
In an important sense, we are combining aspects of information science and network science to improve our understanding of efforts by violent non-state actors to pursue and / or use CBRN agents. We have applied the new methods we've developed to START databases such as POICN and BAAD, and one payoff is to identify not a single path to the outcome variable, as happens using standard regression modeling, but multiple paths. In an analysis of 395 terrorist groups worldwide, we identified distinctive combinations of variables that led to enhanced prediction of drug trade participation.
Another payoff is now having a procedure to use the cases to discover statistical interactions among the variables that significantly improves prediction of CBRN use or pursuit. We are also conducting several studies of anomalous cases; in particular, groups that were coded as engaging in CBRN activities but were not predicted to do so by our model.
Dr. Ronald Breiger, Professor of Sociology at the University of Arizona, holds joint affiliations with the Interdisciplinary PhD program in Statistics and with the School of Government and Public Policy. He is a leading contributor to theory and methods in social network analysis, and he has substantial strengths in the sociology of culture and and organizations. He served as Editor of the journal Social Networks (1998-2006), was the 2005 recipient of the Simmel Award of the International Network for Social Network Analysis, and was elected Chair of the Section on Mathematical Sociology of the American Sociological Association (2009-10). He chaired a 2002 National Academy of Sciences workshop on dynamic social network modeling and analysis, which was focused on the contributions of that area to national needs and especially to national security. The proceedings (edited by R.L. Breiger, K.M. Carley, and P.E. Pattison) were published in 2003 by National Academies Press. Dr. Breiger has been named a Fellow at the Center for Advanced Study in the Behavioral Sciences and a Fulbright Senior Scholar, and he is currently an editorial board member of the American Journal of Cultural Sociology. He currently holds research grants from the Defense Threat Reduction Agency (PI) and the Air Force Office of Scientific Research (co-PI).