We tend to see delusions as the mark of madness. The content of some delusions is so bizarre as to invite scepticism about whether anybody can genuinely believe such things. And yet, people assert sincerely and with great conviction that they are dead (Cotard delusion), that their dear ones have been replaced by impostors (Capgras delusion), or that their neighbours are physically inserting thoughts into their heads (delusion of thought insertion).
My view is that even delusions with bizarre content are beliefs. When we ascribe delusions as beliefs to the people who report them, we are better placed to understand the other things they say and the behaviour they manifest. Even if the delusions themselves are deeply puzzling, the behaviour of the people who report them makes sense in the light of such delusions, and thus we can say that their delusions play the same role as beliefs in our folk-psychological practices of explanation and prediction. Delusions are endorsed, and people are disposed to act on their delusions, even if the motivation to act on delusions fails in some circumstances.
I also argued that delusions are irrational beliefs, but that, when we look closely at what their irrationality consists in, they are not different in kind from many everyday beliefs that we all have. Delusions may conflict with other beliefs; they may be badly supported by evidence; they may be resistant to counter-argument; they may fail to be consistently manifested in behaviour. These are common epistemic failings, and nobody is exempt. We don't necessarily have to look at racist or superstitious beliefs to find analogues of delusions in terms of epistemic failings. I am sure we all have inflated beliefs about our own worth which are badly supported by the evidence available to us, and that we are much more ready to discount evidence against than evidence for our scientific beliefs.
My tentative conclusion is that we shouldn't identify the pathological character of delusions with their irrationality. We are familiar with irrational beliefs (e.g., positive illusions) that not only are common in the non-clinical population, but also effectively contribute to people's confidence and well-being despite epistemic weaknesses. At the same time, we should be theoretically more open to acknowledging that there is no paradox in the idea of irrational beliefs. Mental states can play the folk-psychological role of beliefs without meeting standards of rationality, and they ordinarily do so.
In my future research, I want to explore potential epistemic benefits of delusions and other imperfect cognitions, and this blog is part of that project.