Place still matters: The operational geography of jihadist terror attacks against the US Homeland 1990–2012

Abstract

Transnational terrorism in the age of globalization is often viewed as a diffuse phenomenon unbounded by geography. A review of violent Sunni Islamist extremists (i.e. “jihadist”) terrorist attacks against the United States' homeland between 1990 and 2012, however, reveals specific spatial characteristics. Indeed, there has been considerable geographical consistency in the loci of command and control, and target selection. Different cell types also exhibit distinct geographical patterns and preferences posing different levels of threat across space. Understanding the operational geography of transnational terrorist groups may hence increase the ability of governments to predict and assess risks and implement offensive and defensive strategies and countermeasures with greater precision and effectiveness.

Keywords::

• terrorism
• Al-Qaeda
• terrorist geography
• homeland security
• homegrown
• lone wolves

Notes

The views expressed in this paper are the author's and do not represent the views of the US Naval War College, US Navy, or Department of Defense.

^1. The US invasion of Iraq could be viewed as an example of this “spatial dementia” whereby different threats and opponents (i.e. Al-Qaeda and Saddam Hussein) were viewed as necessarily intertwined in a networked and boundless threat while its roots were viewed in terms of an overarching condition (i.e. lack of freedom in the Middle East) rather than considering place-specific contexts and grievances.

^2. This study only discusses only the most important pathways and nodes directly connected to operational activities (training, command and control, cell formation/direct indoctrination, key transit areas). Logistical sites, fundraising areas, sites of occasional meetings, etc., were not considered. This is not to diminish the importance of these types of places, but they were beyond the scope of this paper. They should be considered objects for future research into terrorism geography.

^3. For the purposes of this paper, success refers to the act of actually executing an attack; no judgment is implied as to whether attackers actually achieved their intended tactical or strategic aims.

^4. Bjelopera (2013) provides an excellent one-stop source for concise summaries of the post-911 cases referenced in this paper as well as links to additional data sources for readers seeking more detailed information. Also see Jasparro (2010) for details of Lone Wolf cases through 2010.

^5. This does not preclude exposure to external ideological influence or some contact with other terrorist actors or groups. Indeed, the term “lone wolf” is a misnomer, because in nearly all cases the primary suspects did not operate in complete isolation; many of those who were caught were identified because they attempted to link up or reach out to other extremists and might be better characterized as weakly connected. Some analysts prefer, more accurately, to label such actors “autonomous”, but “lone wolf” will be used in this paper in deference to common usage.

^6. Identifiable locales, for this study, refer to cases where specific city, state, or aircraft are known to have been targeted or considered for targeting. In several cases, plotters considered multiple locations or target types but available information on the cases were too unspecific to be included and hence were categorized as indeterminate (see Appendix A).

^7. For the purposes of this paper, New York City includes the wider New York–Northern New Jersey Metro area and Washington, DC includes the wider National Capital Region.

^8. Of the nine lone wolf attacks which failed to reach the point of execution, all were disrupted by sting or confidential informant operations orchestrated by law enforcement. It is unknowable as to whether these attackers would have ultimately acted or even selected the same target types if left to their own devices. Taking these failures out of the equation would make lone wolf attacks appear to be even more effective yet rarer than commonly supposed. Indeed, many of the successful Lone Wolf attackers (such as Nidal Hassan and Carlos Bledsoe) did pop up on the radar of security services and could have been prevented. That said, the massive volume of threat reporting and intelligence, the fact that some attackers (such as Charles Bishop) appear to be true loners, and more diverse patterns of target selection all suggest that some successful attacks are inevitable.

^9. Similar cell types may exhibit different characteristics in different places. For example, American lone wolves have favored the use of firearms against government, military and Jewish targets, while European lone wolves have largely employed knives against individual journalists and politicians (Jasparro, 2010).

^10. The three exceptions were (1) the Long Island Railroad plot whose primary suspect, Bryant Neal Vinas, is a Long Island native, although his alleged plotting took place overseas; (2) the Brooklyn Bridge plot, also allegedly hatched overseas by key plotter Iman Faris who was an Ohio resident; and (3) the Lodi case, because there is no reliable target information appearing in open sources. In all three of these cases, however, the main homegrown suspects had very strong connections with external actors and had spent time abroad in Pakistan.

^11. The major exception was the Quantico Marine Corps Base plot, where a North Carolina-based cell attempted to launch an attack in Virginia, and the Michael Reynolds case in which the suspect was considering a variety of pipeline and energy-related infrastructure targets across the US. In several cases plotters had not only considered targets close to home but further afield, such as the Liberty 7 plot where a Miami-based cell considered attacking the Sear's Tower in Chicago, the 2005 Washington/Atlanta plot where cell members considered targets in DC as well as in the Atlanta area where they were residing, and the 2011 Aldawsari plot in which the suspect's primary focus appeared to be New York but he allegedly investigated infrastructure such as dams in Colorado and California. The Qazi brothers scouted targets in New York even though they resided in Florida. These plots were all largely aspirational in nature.

^12. Due to the high number of plots by weakly connected cells and lone actors that have been disrupted or even in part precipitated by undercover law enforcement and sting operations, there is a question of whether potentially overzealous law enforcement in particular locales introduces bias into geographic analysis. Upon closer examination, this does not seem to be the case. If only successful attacks by weakly connected and lone wolf cells are examined we see six of the eight places identified as multiple target locations in the general data appear: NY, Seattle, and Ft. Hood/Kileen (two each), DC, LA, Tampa (one each) along with a random selection of other places such as Gasdsen, Little Rock and Chapel Hill. The only locale where aggressive proactive law enforcement could inflate the numbers is New York (with nine disrupted weakly connected or lone wolf plots and attempts) but, given that NY has long been the most targeted US city, any bias would just accentuate a trend rather than distort the geographic pattern. Aggressive policing may have a distorting effect on the overall number of cases and hence perception of general risk; however, it does not seem to have much bearing on general spatial patterns.

^13. For this paper, C2 refers to actions by an organized group or terrorist leader that provide specific operational planning, guidance, instruction, or approval (in form of actual orders or imprimatur) to an operational cell.

^14. This figure is likely to be an underestimate. Only cases where verifiable or strongly plausible evidence exists have been counted. In other cases, lack of open-source evidence does not necessarily mean no command and control was exercised. Information currently emerging in the aftermath of the killing of Osama bin Laden seems to support this view as well.