Measuring Territorial Control in Armed Conflict Using Open-Source Information

• In Event: New Research on Territorial Control and Violence

Abstract

Under what conditions do nonstate actors control territory in defiance of the incumbent government? How does nonstate territorial control and governance affect violent and nonviolent conflict processes, states legitimacy, and political and economic development? This article introduces the "Subnational Nonstate Actor Governance" (SNAG) project, which advances this research agenda by developing a measurement strategy to capture armed rebel nonstate actors' territorial control in conflict zones that is spatially and temporally disaggregated and comparable across countries and conflicts. Existing approaches to measuring territorial control during conflict focus on (primarily violent) conflict and contentious politics events, which represent areas of contested rather than complete control and indicate changes in control without tracking where it remains stable. SNAG addresses these issues by developing a measurement strategy that emphasizes both static and dynamic conditions at the local level in conflict zones. We employ text analysis methods on a corpus of news and other reporting to extract mentions of rebel presence and control, along with location and timing, which we use alongside events data in a measurement model to estimate the areas of rebel control. We then validate SNAG measures of subnational territorial control against “ground truth” data on rebel control in the Syrian Civil War.

Authors

• Michael A. Rubin, University of Connecticut

• Andrew Halterman, Michigan State University

• Nina McMurry, Vanderbilt University

• Benjamin J Radford, University of North Carolina at Charlotte