A map connecting the September 11 hijackers by Valdis Krebs.
While ideally, government officials and researchers would like to use social network analysis to help stop terrorist attacks, one of the most well known analyses was conducted retroactively, on the 9/11 terrorist network.
Valdis Krebs conducted the first open-source social network analysis of the 9/11 hijackers, using news reports to piece together the relationships among the various members. His article, “Uncloaking Terrorist Networks,” is now a classic of the field and likely the most cited public analysis of the 9/11 network.
Krebs found there were a number of unique attributes of the 9/11 network, which helps explain why covert networks are so hard to detect. At the heart of Krebs’ analysis is the supposition that the 9/11 hijackers’ network was not highly centralized, which made it difficult to subvert because taking out one person, or “node,” would have had a minimal impact on the rest of the network.
One important attribute of the 9/11 network is that its members, or “nodes,” of the network avoided contacts with those outside of their group, or “cluster.” This helped keep the network under the radar, and minimized opportunities for information to escape beyond the core members.
But even within the network, it would have been difficult for the intelligence community to “connect the dots” between and among the various members because they appeared not to be directly connected. For example, while there were only 19 hijackers, there was an average of 4.75 degrees of separation between them, which makes it appear as if they weren’t highly connected.
The network worked hard to keep their ties hidden while they were in the United States preparing for the 9/11 attacks – a strategy that made them more difficult to track. That lack of contact made their network ties appear to be weak, when in fact members of different cells had strong connections from prior years.
While the secrecy required of a covert network meant making strong ties appear weak, that didn’t mean the network itself was weak. “The hijacker’s network had a hidden strength – massive redundancy through trusted prior contacts,” Krebs wrote. “These ties made the network very resilient.”
While this sort of retrospective study may not have stopped a terrorist attack, it helped provide a window into understanding why the 9/11 hijackers proved so hard to detect ahead of time, and may have paved the way for later social network analysis work, such as the capture of Saddam Hussein in 2003.