Do We Over-state the Danger of Illict Networks?

I would like to thank my CTLab colleague Mike Innes for pointing out an article of particular interest to me. “Assessing the Dangers of Illicit Networks: Why al-Qaida May be Less Dangerous Than Many Think“, by Mette Eilstrup-Sangiovanni and Calvert Jones appears in the latest issue of International Security, and asserts that the threat from adversarial networks is over-stated in the literature. While the authors present a valuable alternative assessment of network-centric organizations, they fall well short of dispelling the claims within the literature of the threat from network-centric adversaries.
The authors central claim is that the current literature on illicit networks does goes too far in exposing the threat posed by these organizations. Their contention is that this threat is unfounded due to the inherent weaknesses of network structures. In the introduction, the authors state, “we combine theoretical and empirical evidence to illuminate some important network weaknesses”. To that end, the authors examine a series of claimed weaknesses, buttressed by historical examples that reinforce their claims. I will not attempt to address all of their claims, though I believe each of them is refutable, instead I will mention the general aspects of the authors’ argument I contest:

• Illicit networks do have central leadership – the authors make repeated claims of the weakness of autonomous operational execution in illicit networks. This claim, however, is unsubstantiated by the data. In illicit networks, the so-called “core-periphery” network structure is most often observed, with clear central leadership, and layered levels of authority emanating out from the center.
• Information flows occur slowly in illicit networks, and that is the way the actors like it. – the authors claim that information is hard to retrieve from networks because it requires some number of queries on the network until the appropriate source can be found. In illicit networks, this acts as a built-in alarm function, letting the membership know which members are asking for what. At their core, illicit networks are networks of trust, and if an untrusted actor is making lots of queries, middle and core leadership will want to isolate that actor until a more precise judgment of trust can be rendered.
• Growth in illicit networks is slow, and that is the way the actors like it – the illicit networks that pose a threat of violence are agile and adaptive because they are small (e.g., Hamburg and Madrid). It is true, as networks grow, they become weak, something I have written about in the past. The large and scalable networks the authors discuss, however, are akin to the ideological networks of extremism, which exist as the landscape against which these threats exists, but themselves do not pose a threat.
• Learning is not critical in illicit networks – this dovetails the previous point on information flows, but it bears highlighting in terms of the claims made by the authors. In the autonomous cells at the periphery of most illicit networks, the operational knowledge is almost certainly only known by one or two cell leaders, and in the multiple case, knowledge is often split between members. This is done for two main reasons: first, the less that is known at the periphery, the more protected the center; and second, the expendability of peripheral actors. The second point is critical, as the authors claim knowledge is needed for actors to be adaptable, but this is clearly not the case for most actors, as their minimal connections to the network’s core gives them very little value outside a particular operation.

I encourage everyone to read the article by Eilstrup-Sangiovanni and Jones, but my initial assessment is that the threat from network-centric adversaries has not been over-stated, and the analysis presented by the authors has not shaken this notion.
Image: Ross Mayfield

Automatically Generated Related posts:

1. Networks and ‘Implication for Network Centric Warfare’
2. Optimal Terrorist Network Structures
3. Challenges to the ‘Advocacy Networks’ Theory in Human Rights
4. Networks, Collective Action, and State Formation
5. Networks, Collective Action, and Building State Capacity