Harvard's Rich Nielsen looks into the reasons why some fundamentalist Muslim clerics advocate violence, and others do not:
I argue that clerics strategically adopt or reject Jihadi ideology because of career incentives generated by the structure of cleric educational networks.Well-connected clerics typically enjoy successful careers within state-run religious institutions. Inexchange for continued support from the state, they assist the political elites by opposing - or atleast not adopting - the ideology of militant Jihad. In contrast, clerics with low-quality educationalnetworks cannot rely on connections to advance through the state-run institutions, so many pursuecareers outside of the system by appealing directly to lay audiences for support. These clerics aremore likely to adopt Jihadi ideology because it credibly signals to potential supporters that theyhave not been theologically coopted by political elites.
For his analysis, Nielsen analyzes "29,430 fatwas, articles, and books written by 91 contemporary clerics" -- a sample consisting of both "Jihad clerics" and "conservative Salafi clerics who share similar beliefs to Jihadis but reject the ideology of militant Jihad." These are mostly in Egypt and Saudi Arabia.
The finding suggests that jailing radical clerics is not a particularly effective deterrent -- it only increases their credibility. Rather, Nielsen wonders whether governments should spend more resources on finding ways to coopt these clerics, getting them to sell out rather than locking them up.
The paper briefly mentions the case of Anwar al-Awlaki, though I'm not sure he really fits the pattern. In the immediate period following 9/11, the middle class, highly-educated Awlaki had led one of the largest mosques in the United States, was frequently consulted by the media, and even invited to speak with Pentagon officials. Whatever involvement with terrorism he had already had, it seems like there were career incentives for him to distance himself from violence.
Of course, this doesn't really conflict with Nielsen's argument, but it would be interesting to explore whether the incentive structure works differently in non-majority Muslim countries.
Hat tip: Chris Blattman