Two rather recent papers by Christine C. Fair and Jacob N. Shapiro aimed at investigating the foundation for militant and violence support in Pakistan, after in recent years studies in this direction have become numerous, but most were not so clear when it came to data aquiring and whether this would be representative for Pakistan. The most important problem was that ‘conventional wisdom’ on Pakistan is often taken as granted and used as a basis for such studies without further testing. Mainly conducted in urban areas and without considering the huge differences between the provinces, these studies become next to worthless – but they shape international policies. Fair and Shapiro have attacked these conventional wisdoms and I believe do give some scholarly backing to observations that are rather obvious to people with experience in Pakistan.
The Roots of Militancy: Explaining Support for Political Violence in Pakistan; C. Christine Fair, Neil Malhotra, Jacob N. Shapiro, December 2009
Shapiro, Jacob N. and C. Christine Fair. “Why Pakistanis Support Islamist Militancy.” Policy Brief, Harvard University, February 10, 2010.
These conventional widoms revolve around
(1) religiosity, with more religious people expected to be less supportive of destructive political attitudes in general but more supportive of groups using religious justifications for violence (Putnam 2000; Jefferis 2010); (2) education, with more educated individuals hypothesized to be more tolerant and therefore less supportive of groups employing violent tactics (Stouffer 1955; McCloskey and Brill 1983); (3) income, with poorer people expected to be more supportive of violent politics or more likely to participate in violent groups (see Sambanis 2004 for a thorough review); (4) democratic values, with those valuing democracy expected to be less supportive of militant groups (e.g. Almond and Verba 1963; Kirwin and Cho 2009); and (5) anti-Americanism, with people holding negative views of the United States hypothesized to be more supportive of certain forms of violence (Tessler and Robbins 2007)
(of which the last I would not really take serious from the start).
Our approach yields a number of interesting findings. Most generally, broad-brush theories about the impact of individual-level characteristics on political attitudes perform poorly at predicting support for specific organizations.
None of these conventional wisdoms rests on a firm evidentiary basis, yet they dominate in varying degrees popular media accounts of Pakistan’s political woes, debates in the U.S. Congress, and policies adopted by Western states to help stabilize Pakistan since 2001.
and more specifically on religion
Taken together, these results strongly suggest that if there is some common factor driving support for all these militant organizations, it is not religion.
It is not adherence to Islam per se that drives support, but rather a specific fundamentalist, textual reading of the Koran.
This is a significant policy point: support for Islam as a governing principle in Pakistani politics does not predict support for any of the militant groups of concern to Western policymakers.