Hard choices, soft targets: Terror proscription and strategic targeting decisions of FTO

ABSTRACT

Proscription lists are common counter-terror tools, yet their impact on terrorist violence is unclear. We find that proscription can be effective at constraining the violence of some types of groups, especially those that are young, secular, and without institutional support. However, proscription also can backfire from a counter-terrorist prospective, especially when applied to groups that are well-established, religious, and/or sponsored by states. Our analysis evaluates 534 terrorist groups, including sixty-six that were ultimately proscribed under the United States’ Foreign Terror Organization list. Unsurprisingly, we find that terrorist groups that attract proscription are more violent and better equipped to tap into international terror networks. While younger groups and nationalist groups are more vulnerable to proscription, older religious groups and those with state sponsors seem to be more violent after proscription. Proscription can be an effective tool for reducing terrorist attacks and lethality, but it is most effective against younger terror groups and states should exercise caution in its use as it may not have the desired effects on all types of groups.

Las listas de proscripción son herramientas antiterroristas comunes; no obstante, su impacto sobre la violencia terrorista es poco claro. Observamos que la proscripción puede ser efectiva en la limitación de la violencia de algunos tipos de grupos, en especial de los grupos nuevos, seculares y que no reciben apoyo institucional. Sin embargo, la proscripción también puede resultar contraproducente desde una perspectiva antiterrorista, especialmente cuando se aplica a grupos sólidamente establecidos, religiosos o que cuentan con el patrocinio del Estado. Nuestro análisis evalúa 534 grupos terroristas, incluidos 66 que se proscribieron finalmente conforme a la lista de Organizaciones Terroristas Extranjeras de los Estados Unidos. Como era de esperarse, observamos que los grupos terroristas que suscitan la proscripción son más violentos y están mejor equipados para acceder a redes internacionales de terrorismo. Mientras que los grupos más nuevos y los nacionalistas son más vulnerables a la proscripción, los grupos religiosos más antiguos y aquellos que reciben el patrocinio del Estado parecen ser más violentos después de la proscripción. La proscripción puede ser una herramienta efectiva para reducir la letalidad y los ataques terroristas, pero es más efectiva contra los grupos terroristas más nuevos, y los Estados deberían aplicarla con precaución, ya que es posible que no tenga los efectos deseados en todos los tipos de grupos.

Les listes de proscriptions sont des outils courants de lutte contre le terrorisme, mais leur impact sur la violence terroriste n’est pas clair. Nous constatons que la proscription peut être efficace pour limiter la violence de certains types de groupes, en particulier ceux qui sont jeunes, laïques et sans soutien institutionnel. Cependant, la proscription peut également se retourner contre elle dans une perspective antiterroriste, en particulier lorsqu’elle est appliquée à des groupes bien établis, religieux et/ou soutenus par des États. Notre analyze évalue 534 groupes terroristes, dont soixante-six qui ont en définitive été proscrits en vertu de la liste des organizations terroristes étrangères des États-Unis. Sans surprise, nous constatons que les groupes terroristes qui attirent la proscription sont plus violents et mieux équipés pour puiser dans les réseaux terroristes internationaux. Tandis que les groupes plus jeunes et les groupes nationalistes sont plus vulnérables à la proscription, les groupes religieux plus anciens et les groupes soutenus par un État semblent devenir plus violents suite à la proscription. La proscription peut être un outil efficace pour réduire les attaques terroristes et leur létalité, mais c’est contre les groupes terroristes plus jeunes qu’elle est la plus efficace et les États devraient faire preuve de prudence dans son utilization car elle peut ne pas avoir les effets souhaités sur tous les types de groupes.

KEYWORDS:

• Proscription
• terrorist lethality
• state sponsorship
• counter-terrorism
• terrorist ideology
• foreign terror organization

Notes

¹ Passed in the wake of the previous year’s Oklahoma City bombing and the 1993 attack on the World Trade Center.

² https://www.treasury.gov/resource-center/sanctions/Documents/facei.txt

³ At least three groups have lobbied to get off the list, the Liberation Tigers of Tamil Eelam and the Kurdistan Workers Party, which both failed. The Mujahedin-e-Khalq was taken off the FTO through an Act of Congress.

⁴ An example is the Muslim Brotherhood’s 2017 lobbying to stay off the list. (https://www.memri.org/reports/reports-creation-muslim-brotherhood-lobby-us-prevent-trump-administration-designating#_ednref11)

⁵ Political leaders are constrained by the legal criteria for designation. For example, when President Trump indicated that he would designate Antifa a terrorist organization in May 2020, critics correctly noted that there is no legal provision to designate domestic groups. However, in a departure from precedent and the conventional understanding of which entities could be designated, the Trump administration added the Islamic Revolutionary Guard Corps (IRGC), part of the Iranian military, to the FTO list, sparking controversy in and outside the administration. The IRGC is not in our data however as it was proscribed after December 2017.

⁶ https://www.state.gov/foreign-terrorist-organizations/

⁷ State sponsors of terrorism may become reluctant to continue support for groups due to the threat of economic or military sanction. McLean et al. (2018) find that sanctions imposed against “home bases of terrorist organizations may encourage these states to scale back their support for transnational terrorists” (p. 394).

⁸ We thank Casey Johnson of the US State Department’s Bureau of Conflict and Stabilization for suggesting this final explanation.

⁹ We thank an anonymous reviewer for this suggestion.

¹⁰ US Court of Appeals Case no. 97–1648 “People’s Mojahedin Organization of Iran v.US Department of State” and US Court of Appeals Case no. 97–1670 “Liberation Tigers of Tamil Eelam v. US Department of State.”

¹¹ Ibid

¹² Supreme Court of the United States case no. 08–1498 Holder, Attorney General, ET AL v. The Humanitarian Law Project, ET Al (2010).

¹³ For instance, in United States v. Thavaraja, 740 F.3d 253, 255 (2d Cir. 2014) two members of the Tamil Rehabilitation Organization (TRO) were convicted under the FTO statue for providing material assistance to the LTTE. Also see Taxay (2014).

¹⁴ For example, the LTTE continued to target political figures like Douglas Devananda, leader of The Eelam People’s Democratic Party (EPDP), a Tamil party. By some counts Devananda survived at least 10 and up to 19 assassination attempts by the LTTE. The LTTE also targeted moderate Tamil intellectuals, journalists and other Tamil leaders (based on author interview with EPDP official 2005).

¹⁵ This is an artifact of the list’s initiation in October of 1997.

¹⁶ Additional differences are explored in Table 2.

¹⁷ We use annual fatalities rather than casualties out of concern that casualty counts are more likely to be subject to bias. The GTD source materials are media articles, which means that the data are as much coverage data as they are event data – that is the data are more complete in areas where journalists have better access and where there are fewer events to record (Best, Carpino, and Crescenzi 2013). We expect that non-fatal casualties are less likely to be recorded accurately in areas characterized by higher baseline levels of violence or where government authority is looser. Fatalities should be less affected by this bias.

¹⁸ As a robustness check we used a variable capturing the percentage of a group’s attacks in a given year that used any of these top four types. The results are robust to this different formulation of the dependent variable.

¹⁹ Sponsorship in EDTG is coded using data from San-Acka (2009, 2016).

²⁰ Carter (2012) focuses on sponsors in particular, but the logic extends to hosts that are less willing as well.

²¹ Variables relating to ideology and goals are allowed to change over time, but do not in our sample and are not reported in these models.

²² We use random-effects models to facilitate graphing the interactive effects. Dropping groups that are never on the FTO, mitigates the selection effects associated with proscription. All models with dependent variable fatalities are panel Poisson regression; all models with dependent variable attacks are panel negative binomial regression. Each model used to produce graphs includes only the interactions of age with on list as well as the interaction of on list with the variable of interest in the graphs.

²³ Stata does not allow for fixed effects Tobit models.

²⁴ ISIL has 11,690 fatalities attributed to it in 2016. The second highest annual total of fatalities is attributed to Boko Haram in 2014, which killed 7,118 people. Both groups were already proscribed at the time.

²⁵ Initial mean was .947, suggesting that sponsorship was only coded for groups that were known to have sponsors and not otherwise. Therefore, to facilitate analysis, we recoded all groups with a missing value as 0.