Just as further proof that I'm not hostile to expertise, let me quote from Assistant Professor at the Patterson School of Diplomacy and International Commerce Robert Farley's review of Negotiating Change:
Negotiating Change, by Jeremy Jones, is about democratization and political change in the Middle East. Jones, a Research Fellow at the Kennedy School of Government and a Senior Research Associate at the Oxford Centre for Islamic Studies, is extremely, if often implicitly, critical of US policy in the Middle East and in particular the process through which policy is made. In short, I think Jones would say, American policymaking has made cultural illiteracy a virtue, with disastrous effects.
Jones point is that context is important. Readings of Middle Eastern politics that don't understand the local meaning of party politics and civil society inevitably fail to capture a reliable picture of what's going on. For example, Jones argues that the success of the Muslim Brotherhood in Egypt doesn't necessarily indicate that the movement is politically popular, or that it has achieved success on its own merits. Rather, the repressive Egyptian state has limited the capacity of civil society to develop. The state, however, is reluctant to invade the mosque, meaning that Islamic groups have a freedom to organize and assemble that other societal groups lack. The result of political oppression, then, is the production of a movement that may be more dangerous to the survival of the Egyptian state than the forces that the state is trying to repress. Although Jones recognizes that their may be cross-national similarities, he doesn't apply the same lens to every country; again, context matters, and superficially similar events may have entirely different political meanings in different countries. [...]
Given his approach, Negotiating Change is necessarily fragmented and episodic. The main theme that comes through, though is that the statements of US policymakers on democracy in the Middle East are almost universally myopic and ill-informed. Without understanding Middle Eastern societies, it's impossible to craft a policy likely to promote, rather than foreclose, democratization. I would add that this insight is particularly unfortunate for a foreign policy group that purports to believe that a) democratization should be the primary goal of US Middle Eastern policy, and b) virtually all experts on the Middle East are ideological poison. Indeed, it's hard to imagine a world in which the combination of those two traits could lead to any success at all.
Food for thought. Over the years, I've come to believe that the central takeaway lesson of this sort of critique is, on some level, that we simply have to make our policies more robust against the possibility that we don't understand what's happening. In other words, if your pet scheme for American policy toward the Middle East crucially depends on your particular interpretation of the Muslim Brotherhood's rise and significant, then it looks very, very risky to bet the farm on that interpretation. To actually understand what's happening in these different countries is hard and requires a great deal of specific knowledge. We shouldn't overestimate the capacity of the government to obtain that knowledge, disseminate it to the right people, and then effectively micromanage outcomes halfway around the world based on up-to-date fine-grained understandings of Yemen, Jordan, Kuwait, Syria, etc.
Obviously, one does one's best with these difficulties, but the main goal should be to do the best one can to outline policies that work okay one way or the other. We should want political actors in the Arab world to believe that killing Americans is not necessary to achieving their domestic political goals (whatever those goals may be) and try to promote a general climate of peace and prosperity since those would be good things even if they didn't promote democratization. The alternative path of trying to figure out the best possible way to effectively determine political outcomes on the other side of the world seems doomed to failure.