In Radical, Religious, and Violent, Eli Berman approaches the question using the economics of organizations. He first dispels some myths: radical religious terrorists are not generally motivated by the promise of rewards in the afterlife (including the infamous seventy-two virgins) or even by religious ideas in general. He argues that these terrorists (even suicide terrorists) are best understood as rational altruists seeking to help their own communities. Yet despite the vast pool of potential recruits—young altruists who feel their communities are repressed or endangered—there are less than a dozen highly lethal terrorist organizations in the world capable of sustained and coordinated violence that threatens governments and makes hundreds of millions of civilians hesitate before boarding an airplane. What's special about these organizations, and why are most of their followers religious radicals?
Drawing on parallel research on radical religious Jews, Christians, and Muslims, Berman shows that the most lethal terrorist groups have a common characteristic: their leaders have found a way to control defection. Hezbollah, Hamas, and the Taliban, for example, built loyalty and cohesion by means of mutual aid, weeding out "free riders" and producing a cadre of members they could rely on. The secret of their deadly effectiveness lies in their resilience and cohesion when incentives to defect are strong.
These insights suggest that provision of basic social services by competent governments adds a critical, nonviolent component to counterterrorism strategies. It undermines the violent potential of radical religious organizations without disturbing free religious practice, being drawn into theological debates with Jihadists, or endangering civilians.
This fits with a notion of mine that there are proto-governments (street gangs, organized crime, unions-- now I'm adding terrorist groups to the list): organizations which could grow up to be governments if that niche weren't already filled. The distinctive feature isn't the amount of violence (though I believe there's always at least a threat), it's the combination of some control of territory and offering services.
I've heard that failed revolutionary organizations are a primary source of organized crime. Anyone know?
This suggests that terrorist groups might be easier to form among Muslims because of the very strong tradition of charity. If I'm right, it's more likely for a Muslim terrorist leader to think of offering social services than it would be for terrorist leaders from other religions.