Authors: Professor Eduardo Lopez, Dr. Annette Idler, Dr. Katerina Tkacova • Resource type: Working Papers
What drives nonlinear change in armed conflict? We identify changes within the conflict in Colombia across time and space which are invisible to descriptive statistics that assume linearity of conflict developments. To do so, we explore existing datasets on conflict events and our novel dataset on armed groups’ presence and behaviour in two Colombian departments. Drawing on Complexity Theory, we treat the conflict in Colombia as a complex system. Among the key characteristics of those systems are complex interaction among many decision-making agents resulting in a phenomenon called emergence, a property of the system but not of the individual parts (Axelrod and Cohen, 2000). In our case, emergence has a form of approximate power-law distribution of the size and frequency of conflict events with the scaling coefficient equal to 2. 5. Building on Bohorquez et al. (2009) and Spagat, Johnson, and Weez (2018), we focus our attention to the variation in the scaling of the power-law distribution across time and space and provide an explanation rooted in the different types of interactions between conflict actors as suggested in Idler (2019).