I did a Bloggingheads debate with Felix Salmon last week, and the New York Times excerpted the “only political science relevant”:http://video.nytimes.com/video/2011/07/14/opinion/100000000946909/bloggingheads-murdoch-under-scrutiny.html bit of our discussion, where we compare the post-Murdoch debate to Ireland after the collapse of the Catholic Church’s moral authority in the early 1990s (me) and the Arab Spring revolutions (Felix). Obviously, these are very different empirical phenomena. But they arguably involve a quite similar mechanism, under which people are unwilling to state their actual preferences (most Irish Catholics had a nuanced relationship to their religion; most Tunisians wanted a different government; most UK Labour politicians, and probably some Conservatives loathed the Murdoch press) because they fear that they will be punished for them. This in turn generates uncertainty among others who share these beliefs, as to whether they are in a minority or a majority – what Timur Kuran calls preference falsification. Such uncertainty can (in the absence of alternative mechanisms for people to signal their beliefs) allow unpopular social orders to persist for quite a long time after their legitimacy has collapsed – but they can also mean that change can happen very rapidly after people have been convinced that the emperor has no clothes (they do not have to change their preferences; only their estimate of the cost associated with expressing those preferences).