1 The Self, Self-knowledge, and a Flattened Path to Self-improvement Robert D. Rupert University of Colorado, Boulder November 14, 2020 I. Introduction Philosophy of mind and philosophy of psychology aim to provide a clear and compelling account of the human self (Schechtman 2011); and epistemology hopes to illuminate self-knowledge, revealing its nature and identifying the conditions under which it can be acquired (Gertler 2010). Theories of these phenomena are, of course, not logically independent of each other. If one’s theory of self-knowledge entails that humans acquire self-knowledge via method M, yet one’s theory of the human self entails that the self is not the sort of thing that could be known about via M, then revision of at least one of those theories is in order. This essay explores the connection between theories of the self and theories of self- knowledge, arguing (a) that empirical results strongly support a certain negative thesis about the self, a thesis about what the self isn’t, and (b) that a more promising account of the self makes available unorthodox – but likely apt – ways of characterizing self-knowledge. Regarding (a), I argue that the human self does not appear at a personal level the autonomous (or quasi- autonomous) status of which might provide a natural home for a self that can be investigated reliably from the first-person perspective, independent of the empirical sciences. Regarding (b), I contend that the most promising alternative view of the self is revisionary: the self is to be identified with the cognitive system as a whole, the relatively integrated collection of mechanisms that produces intelligent behavior (Rupert 2009, 2010, 2019). The cognitive system teems with reliable, though not necessarily perfect, indicators (cf. Dretske 1988) of its own properties or of the properties of its proper parts, many of which are available for detection by, or