Many psychopathological disorders – clinical depression, borderline personality disorder, schizophrenia and autistic spectrum disorder (ASD) – are commonly classified as disorders of the self. In an intuitive sense this sort of classification is unproblematic. There can be no doubt that such disorders make a difference to one’s ability to form and maintain a coherent sense of oneself in various ways. However, any theoretically rigourous attempt to show that they relate to underlying problems with say, such things as minimal selves or, even, so-called narrative selves – where these latter constructs are invoked to do genuine explanatory work – would require, inter alia, philosophical clarification of what it is that one is precisely committed to in talking of such things (if things they be). It would also require justification for believing in selves of these various kinds. I have elsewhere put on record some of my worries about proposed justifications for believing in minimal selves (Hutto 2008b). But – lest I be accused of favouritism – it should be noted that I also have concerns about the very idea of narrative selves. Several authors have made strong claims about the role of narratives in self constitution (e.g. Dennett 1991, Flanagan 1996, Schechtman 1996). Under standard interpretations these proposals are ambiguous, underdeveloped in key respects, embed obvious tensions or generate puzzles. For these reasons I think we should be cautious of lax talk of selves that are woven from narrative cloth. This is not to say that I agree with Strawson (2004) that adopting a narrative perspective might not be essential for being a self (or at least being a self of a certain sort – even an ethically interesting sort).1 It is rather that I think that before we get around to assessing such claims we need a better understanding of just what we are committed to in talking of selves in general. This is a major philosophical programme, and not one with which I will attempt to engage in this paper – not even in passing..