Re-Examining the Local Roots of Tunisia’s Foreign Fighter Mobilization

On November 28, Aaron Zelin, Richard Borow Fellow at the Washington Institute for Near East Policy, published an essential new report on the mobilization of Tunisia’s foreign fighters to fight in Syria, drawing on ISIS fighter entry records among other data. In addition to its own analysis, Zelin’s report, thanks to its transparency regarding its data samples, provides an opportunity to re-examine and check the conclusions of New America’s All Jihad is Local Volume II report that also addressed Tunisian foreign fighters using ISIS entry records. Overall Zelin's data support the main conclusions of the All Jihad is Local report regarding the distribution of key regional and local hotspots of per capita recruitment in Tunisia even though there is a variance in the total numbers that he analyzed. This alignment of key hotspots based on the two samples emphasizes the importance of jihadist organizing in the aftermath of the Arab Spring, prior histories of protest and jihadist network building, and economic marginalization.

Zelin’s sample of ISIS records includes 639 records (636 of which include provincial level information on residence), slightly more than the 589 examined in All Jihad is Local Volume II (578 of which include provincial level information on residence). In addition the records’ provincial origins are distributed somewhat differently than those examined in All Jihad Is Local. As Zelin notes, these differences emphasize that “anyone looking at such information [the ISIS entry records] should recognize that different authors (and their publications) draw from different parts of the data set. Therefore, no particular source should be taken as gospel.” However, Zelin’s expanded and somewhat distinct set of Tunisian ISIS entry records also presents an opportunity to examine the replicability of key findings of the All Jihad is Local report with a distinct though likely overlapping sample.

Comparing the two samples using the per capita method of measurement presented in All Jihad is Local suggests that while there is flux in some of the numbers – emphasizing the need for caution regarding the impact of using different samples of the archive whose full nature is not well known - key findings from All Jihad is Local are replicated using the Zelin sample.

At a regional level the Zelin sample provides a very similar breakdown of the regional hotspots. The order of the regions in terms of per capita rates remains the same with Bizerte having the highest rate followed by Grand Tunis followed by the Southern Border region followed by the Interior Periphery followed by the Eastern Coast in both samples. Bizerte’s rate increased the most (3.4 points) followed by the Southern Border (1.0 points) while the Interior Periphery and Eastern Coast both increased by .4 points and Grand Tunis maintained a similar rate dropping by .1 point.

As noted in All Jihad is Local, this places the wealthiest Tunisian region and the one least roiled by protest – the Eastern Coast – as having the lowest rate of fighter production while the highest rates of fighter production exist in Bizerte, a site of significant Ansar al Sharia organizing, the economically struggling suburbs of Grand Tunis, and the economically struggling Southern Border region.

This similarity exists despite differences in the regional per capita rates produced by the two samples. The per capita rate of the Zelin sample can be compared to the rate that would be expected were the Zelin sample distributed the same way as the All Jihad is Local sample just with a larger number of total fighters. This is done by multiplying the number of fighters in the All Jihad is Local sample for a particular region by 1.1 (the total number of fighters with their province of origin known in the Zelin sample divided by the same number in the All Jihad is Local sample) before calculating the rate.

Table 1: Comparison of All Jihad is Local Volume II and Zelin Sample by Tunisian Region

The Zelin sample also suggests a relatively similar pattern at the provincial level in per capita terms (grouping the components of Grand Tunis into one grouping - the reasoning for this is discussed in All Jihad is Local). Of the ten provinces with the highest fighter production rates in the Zelin sample, all but one were in the top ten of the All Jihad is Local sample. Likewise, of the top five fighter producing provinces in the Zelin sample, all but one were in the top five of the All Jihad is Local sample.

Moreover, the shift at the top in the Zelin sample supports the finding of the regional examination. Of the top five provinces in the Zelin sample, one is the Grand Tunis region, one is Bizerte, and three are in the Southern Periphery, the three top hotspots at a regional level.

Table 2: Provincial Level Comparison of All Jihad is Local and Zelin Sample with Combined Grand Tunis

Finally, Zelin’s sample appears to confirm the importance of several economically struggling areas noted in the All Jihad is Local report. First, the governorate of Kebili in the Southern Border region has the highest rate of fighter production in both the Zelin and All Jihad is Local samples. Zelin’s sample also finds a large number of fighters (25) from the economically struggling Grand Tunis suburb of Ettadhamen – also a hub of protests - in the governorate of Ariana constituting 12 percent of the fighters from Grand Tunis in Zelin’s sample compared to 11% in the All Jihad is Local sample.

Zelin provides an extremely useful breakdown of the specific neighborhoods referred to by fighters, which would allow for a broader evaluation of the neighborhood-level differences in recruitment in the Grand Tunis area. Though that is beyond the scope of this paper, the high number of fighters from Ettadhamen supports the finding in All Jihad is Local that fighters from Grand Tunis may come disproportionately from struggling areas.

Finally, Zelin’s sample also appears to find an overrepresentation of fighters from the town of Ben Guerdane within Medenine – also noted in All Jihad is Local with about 37% of the fighters coming from the area home to about 17% of the governorate’s population.

The similarity in distribution of per capita hotspots of recruitment in the Zelin data compared to the All Jihad is Local data and the prominence of economically marginalized areas provides a cautionary note regarding Zelin’s conclusion that “economic deprivation is not a leading cause for joining with jihadist groups or fighting abroad.” It also emphasizes the analytical pitfall that can occur when correctly noting, as Zelin does, that “Tunisia’s foreign fighter mobilization was a national phenomenon and not specific to a particular city or region” without continuing on by addressing the differences in per capita recruitment and how such differences align with economic and political patterns

Zelin’s report reveals the substantial flux that complicates efforts to analyze and compare fighter production rates at a local level from captured ISIS documents. However, the Zelin sample also aligns with the key findings of All Jihad is Local.