Here's a question in applied political philosophy. Would the U.S. government be justified in "cognitively infiltrating" 9-11 conspiracy theory circles in order to undermine the credibility of those theories? Cass Sunstein and Adrian Vermeule think so, arguing in the Journal of Political Philosophy last year for "developing and disseminating arguments against conspiracy theories, governments hiring others to develop and disseminate arguments against conspiracy theories and governments encouraging others informally to develop and disseminate arguments against conspiracy theories." This might involve entering "chat rooms and online social networks to raise doubts about conspiracy theories and generally introduce ‘cognitive diversity’ into those chat rooms and social networks."
Steve Clarke is right to point out that these measures are unlikely to have the intended result, and may even backfire. But that strikes me as besides the point. Even if we knew it would have great results, it still seems like we ought to be suspicious of this kind of government action. For one thing, while I have no sympathy for any of these conspiracy theories, it's not as if it's a crime, or even particularly subversive with respect to law and order, to believe some absurd theory about 9-11. So there's clearly not a national security justification here, like there would be if we were talking about spying on an Al Qaeda cell, for instance. What other good justification could there be here?
Also, a good way to think about this is to consider other possible targets for cognitive infiltration. For example, suppose it came to light that in the run-up to the U.S. invasion of Iraq the Bush administration was cognitively infiltrating anti-war groups; or suppose the Obama administration was secretly hiring people to write blogs aimed at destroying (what's left of) the reputation of Sarah Palin, or some other GOP figure. While liberals might be more upset at the former, and conservatives at the latter, clearly almost everyone would be upset about one or the other. What principled basis could we have for thinking that targeting conspiracy theories is fine, but that it's not fine to target other kinds of social or political beliefs? I think the obvious answer is: none.
 These quotes are from Steve Clarke's summary at Practical Ethics.