Narrative and Variable-Centered Analyses of Conflict: Toward a Reconciliation
WHEN: Friday, February 3, 2017, 12:00 - 1:40 p.m.
WHERE: Knox Hall 509, 606 W 122nd Street
Peter Abell put forward a qualitative network-of-events approach to narrative as a counterpoint to the better-established variable-centered form of explanation that dominates empirical social science, and he ultimately proposed to seek an “intellectual rapprochement” between the two.* In this talk, I will review two recent studies that colleagues and I have undertaken that suggest paths toward such a reconciliation.
One is a study of thirty insurgencies as networks of event orderings. An insurgency is viewed as a network of sequenced variables across temporal stages (periods) of a conflict. A network among insurgencies is created based on the degree of overlap of their sequenced activities. This network is shown to be informative for identifying sequences of events that predict outcomes of interest.
The second study begins with a variable-centered analysis of the qualities of 175 events that predict the use vs. pursuit of unconventional weapons (chemical, biological, radiological, nuclear) by non-state actors. The variables are then used to construct a network of profile similarity among the events, and this network among events is shown to have (in an important sense to be discussed) generated the regression outcome.
I will argue that a regression coefficient cannot adequately be viewed as a single number, because it is a sum of effects across events (more generally: cases) drawn from quite different regions of the network among events, and so we must often seek a version of what Abell (in a different context) called “local explanations” according to which some sets of events may be associated with strong positive effects while others exhibit strong negative effects.
Together, these studies suggest prospects for a better understanding of how each approach—networks of events and variable-centered analysis—may be used to enhance interpretation of the other.
Ronald Breiger is a Regents’ Professor and a Professor of Sociology at the University of Arizona, with affiliations in Statistics and in Government and Public Policy. His interests include social networks, adversarial networks, stratification, mathematical models, theory, and measurement issues in cultural and institutional analysis. He currently serves as Editor for Social and Political Science for the Cambridge University Press journal Network Science. His web page can be found here: http://www.u.arizona.edu/~breiger.
*See for example Abell’s The Syntax of Social Life, 1987 and his “Narrative Explanation…,” Ann. Rev. Sociol., 2004.
Funding support for the Networks and Time Seminar Series is provided by the Paul F. Lazarsfeld Lecture Series, administered by INCITE, which features events and programming that embody and honor Lazarsfeld’s commitment to the improvement of methodological approaches that address concerns of vital cultural and social significance.