The current issue of International Security carries an article by Jacob N. Shapiro and Christine Fair, "Understanding Support for Islamist Militancy in Pakistan," which is based on a survey of urban Pakistanis. According to the abstract, they find that poverty, personal religiosity, and support for sharia law, among other things, are poor predictors of support for militant groups, as is faith in the country's democratic process. The last line of the abstract ventures to suggest, on the basis of the findings, that economic development, democratization and the like "may be irrelevant at best and might even be counterproductive" in the effort to reduce support for militancy.
The absence of a link in the urban setting between poverty and militancy is not, I think, too surprising, but the statement that those "who support core democratic principles...are not less supportive of militancy" sounds so counterintuitive that one suspects the abstract has been consciously written as a tease, to entice people to read the whole article. It will be interesting to see if people do read it and if it stirs debate.