Smarter About Terrorism: Using the tools of complex adaptive systems research to attack the complexities of terrorist networks

Professor Philip Vos Fellman is the perfect man to formulate a new approach to making America safe from terrorists. [“Less Is More in the Fight Against Terrorism”] As his middle name and mine are the same, I should disclose that he is my cousin.  Hurt me, and he’ll come after you.  The networks of genealogy are complex, and that’s also true of the networks of terrorism.  And that’s their greatest weakness.

Philip’s mathematical analysis of terrorism, published this month in the International Journal of Networking and Virtual Organisations, suggests that an optimum way to fight networks of terror is to isolate the hubs within the network rather than trying to destroy the network as a whole through short-term battles.

According to Philip Vos Fellman (currently a Lecturer at Suffolk University, Boston; member of the New England Complex Systems Institute, USA; and an expert in mathematical modeling and strategy) the same mathematical and computer science tools used to analyze complex systems in physics or biology can also be used to study terrorist networks in order to defeat them.  He explained that terrorist networks are “typical of the structures encountered in the study of conflict, in that they possess multiple, irreducible levels of complexity and ambiguity.”

“This complexity,” he adds, “is compounded by the covert activities of terrorist networks where key elements may remain hidden for extended periods of time and the network itself is dynamic.” The inherent nature of a dynamic network is more like the robust Internet than it is like the structure of the armed forces or homeland security systems, which are primarily centralized and hierarchical.

Philip Vos Fellman used network analysis, agent-based simulation (“intelligent” agents in software interacting with each other in ways that emerge from simple rules), and dynamic NK Boolean fitness landscapes (theories initiated by Stuart A. Kauffman) [“Kauffman’s NK Boolean networks”] to understand the complexities of terrorist networks.  Specifically, he has focused on how long-term operational and strategic planning could be implemented so that tactics that appear to offer immediate impact are avoided, if they cause little long-term damage to the terrorist network.

His computer simulations of terrorist networks indicate that isolation, rather than removal, might be the key to successfully defeating them.  “The results which these simulation and dynamical systems modeling tools present suggest that quite literally sometimes less is more,” he says, “and that operational objectives might be better directed at isolation rather than removal.” Philip also claims that the simulations show that soft (easy) targets of small cells within a network are, mostly, not worth pursuing.  Rather, efforts should be focused on the hubs around which the network is built.  “If you are not focused on the top problems, then considerations of opportunity cost suggest that it may be better to do nothing rather than to waste valuable resources on exercises which are doomed to fail.”

In joint work published in 2006 [“Disrupting Terrorist Networks, a dynamic fitness landscape approach”], this author, Philip Vos Fellman, and colleagues analyzed terrorist networks in terms of adaptation and self-organizing systems.

The researchers write:

Over a period of approximately five years, Pankaj Ghemawat of Harvard Business School and Daniel Levinthal of the Wharton School have been working on a detailed simulation (producing approximately a million fitness landscape graphs) in order to determine optimal patterns of decision-making for corporations.  In 2006, we adapted this study, combining it with our own work on terrorism to examine what would happen if we inverted Ghemawat and Levinthal’s findings and sought to provide disinformation or otherwise interfere with the communications and decision processes of terrorist organizations in order to optimize poor decision making and inefficiencies in organizational coordination, command and control.

The bulk of this study was then presented at the 2006 annual meeting of the North American Association for Computation in the Social and Organizational Sciences.  The authors concluded: “We present here an updated version of that study, emphasizing the rather counter-intuitive finding that ‘soft’ targets have almost no value and that unless one can influence key factors, an effort directed at the easy to reach elements of terrorist organizations may actually be worse than mounting no effort at all.  We conclude with the recommendation that some fundamental rethinking may be required if the United States is to effectively defend itself from future terrorist attacks.”


Philip Vos Fellman.  The complexity of terrorist networks.  Int. J. Networking and Virtual Organisations, 2010, 8, 4-14.

Disrupting Terrorist Networks, a dynamic fitness landscape approach
Authors: Philip V. Fellman, Jonathan P. Clemens, Roxana Wright, Jonathan Vos Post, Matthew Dadmun
Comments: 12 pages, 8 figures.  Proceedings of the 2006 annual meeting of the North American Association for Computation in the Social and Organizational Sciences
Journal-ref: InterJournal Complex Systems, 2060

Subjects: Adaptation and Self-Organizing Systems (nlin.AO)  Download the PDF

Stuart A. Kauffman.  Scientific American August, 1991. P. 64.

Stuart A. Kauffman. The Origins of Order: Self-Organization and Selection in Evolution, Oxford University Press, New York, 1993.

Stuart A. Kauffman. At Home in the Universe: The Search for Laws of Self-Organization and Complexity, Oxford University Press, Oxford, 1995.

See Also

Not So Quiet on the Cyber Front

Sousveillance: Wearable Computing and Citizen “Undersight”

Leave a Reply