Understanding the Survival Mechanisms of Global Salafi Jihad

Understanding the temporal dynamics of networks is what many researchers in the field consider the “holy grail” of the science. As people in national security realm began to understand the structure of terrorist organizations as networks, the ability to track and predict their growth and decay became the center piece of much their research. Unfortunately, at present we lack the requisite mathematics needed to truly understand the trajectory of these systems. At best, thorough case studies of good data can help inform our understanding of these dynamics, and this is the approach taken in a new study entitled, “The Dynamics of Terrorist Networks: Understanding the Survival Mechanisms of Global Salafi Jihad.”

The paper attempts to analyzes the topology, growth pattern, and robustness to attack of the Global Salafi Jihad (GSJ) data set, collected by Marc Sageman for his 2004 book. The authors use the GSJ data for its robustness to missing data, which is a welcome position given that most research in this area tends to ignore the inherent fallibility of the data, as the authors rightly note:

The structural mechanisms responsible for the survival of terrorist networks remain unknown for two major reasons. First, nearly all theoretical and practical studies on terrorist networks suffer from the lack of empirical data. As terrorist networks are clandestine organizations that operate covertly, data about the individual members and their social ties are extremely difficult to gather. Anecdotal evidence from news stories and media sources is highly unreliable. Second, the dynamic nature of terrorist networks is largely ignored. Terrorist organizations are dynamic systems and undergo constant changes over time.

Sageman’s data suffers less for these problems because it is mostly drawn from an extensive review of court documents and detainee interviews, though I suspect Sageman himself would be willing to concede a certain level of error in his data. That aside, the topological analysis finds that the network follows a power-law structure, which is surprising given that terrorist organization are closed systems, where membership highly selective. Power-law structures, on the other hand, are good models of open systems, such as the world wide web, where membership is free and attachment is based on actor preference.

This finding, however, may be an artifact of the relatively long time period (15 years) from which the GSJ data is based. The authors do not control for the fact that many of the ties in their data likely decay over time, which would have a large affect on the topology. A better approach would have been to examine slices of the data at various time intervals, applying a decay function from the literature. These slices could be taken at even or random intervals, or better yet, informed by historical events affecting the network. This alternative approach would have allowed the researchers to note any inconsistency in growth pattern or topology over time.

Finally, their analysis on the robustness of the network to attack is consistent with the literature on scale-free networks. The analysis, therefore, would have benefitted greatly from a more thorough review of this literature, though it is interesting to perform these tests on real data rather than simulated. In all, however, the research is noteworthy and I recommend it to anyone studying network dynamics and/or terrorist organizations.

Photo: David Yates

Automatically Generated Related posts:

1. Detecting Terrorist Activity by Monitoring Technology Usage
2. Network Analysis for International Relations
3. Survival Analysis of Critical Events on chatroulette.com
4. Structural Implications of Watts’s Foreign Fighter Recruitment Model
5. The Royal Society’s Take on ‘Stochastic Networks’