Samenvatting

Radicalization and terrorism remain areas of special interest in terms of security policies. It’s common knowledge that most of the activities related to radicalization and terrorism heavily rely on the involvement of multiple actors. Therefore, an increased understanding of the underlying social structures is considered to offer important leads for the development of effective countermeasures (in particular when related to demographic, cultural, psychological and other social factors). Yet, the number of empirical network studies in this domain (at least those openly available) that incorporate arithmetic tools known as Social Network Analysis (SNA) remain extremely scarce. In this paper the author presents an exploratory Social Network Analysis of the Hofstad network based on publicly available data. Members of the Hofstad network, a radical Islamist network in the Netherlands, were active recruiters for the violent jihad, spreaded radical propaganda, some attended training camps in Pakistan, and the network was suspected of planning several terrorist attacks on strategic objects and prominent people in the Netherlands. One of the members, Mohammed B., was sentenced to life in prison for murdering the Dutch filmmaker Theo van Gogh in November 2004. Although the Hofstad network was considered by trial as a terrorist organization in the first instance in 2006, the judgment was reversed on appeal in 2008 when most members were acquitted. As is characteristic of home-grown networks, our analysis indicated that the Hofstad network (N=67) was relatively sparse and decentralized and evolved around a more cohesive core of key players (N=13). The key players were identified based on their central network position and a hierarchical clique analysis. Mohammed B., who had been considered a marginal player by the secret service, turns out to be the most central actor of the network. Although the analysis clearly suggests that quantifying network structures provides actionable intelligence, more research is needed to validate the results.