Leaderless Covert Networks: A Quantitative Approach
Tilburg University - Department of Econometrics & Operations Research
Roy H. A. Lindelauf
Military Operational Science
Tilburg University - Center for Economic Research (CentER); Tilburg University - Department of Econometrics & Operations Research
July 12, 2012
CentER Discusion Paper Series No. 2012-057
Lindelauf et al. (2009a) introduced a quantitative approach to investigate optimal structures of covert networks. This approach used an objective function which is based on the secrecy versus information trade-off these organizations face. Sageman (2008) hypothesized that covert networks organize according to leaderless principles, i.e., clear leaders can not be identified. This flat organizational structure is quite robust to destabilization tactics which target the most important persons in a network. There exist several centrality measures to express the importance of persons in a network. The most recent one introduced in the field of covert networks is a game theoretical centrality measure which takes into account both the structure of the covert network, which usually reflects a communication structure, as well as non-network features, which represent individual parameters like financial means or bomb building skills, see Lindelauf et al. (2011). The question we try to answer in this chapter is whether there is a relationship between the quality of a covert network based on their optimality with respect to the trade-off between secrecy and information and the variance of the game theoretic centrality measures of the respective individuals in the network. The leaderless hypothesis seems to suggest that good covert networks do not have a high distinction between centrality of the individuals, i.e., they are leaderless. We investigate this by looking at homogeneous networks and heterogeneous networks in which the links between individuals are weighted. We find that (approximate) optimal networks have low variance in game-theoretic centrality, i.e., we find evidence that supports the leaderless hypothesis. However, if the networks are heterogeneous in the sense that, for instance, certain individuals communicate much more often than others, we find that the variance increases significantly. Finally, we look at the Jemaah Islamiyah 2002 Bali bombing. We find that the operational network used to conduct and to coordinate the bombing not only facilitated both secrecy and efficiency but also adhered to the leaderless principle.
Number of Pages in PDF File: 14
Keywords: terrorism, network analysis, centrality, game theory
JEL Classification: C71working papers series
Date posted: July 12, 2012
© 2014 Social Science Electronic Publishing, Inc. All Rights Reserved.
This page was processed by apollo3 in 0.484 seconds