The normative dimension of commonsense psychology

Each of us displays a remarkable facility with the concepts of our commonsense psychology (CSP), the conceptual framework in terms of which we understand ourselves as beings who know, think, believe, desire, and perform intentional actions. Some philosophers have suggested that CSP is a theory of human behavior, no different in kind from scientific theories in general. Although it is now deeply entrenched in contemporary debates about the nature of mind, this so-called 'theory' theory (TT) has received surprisingly little critical attention. Its influence is especially apparent in the emergence of eliminativism, according to which CSP is a radically incorrect theory (like phlogiston theory). Those who believe, as I do, however, that eliminativism is self-defeating, conclude that TT is fundamentally misconceived and in need of displacement by an alternative account of the nature of CSP.;The Simulation Theory (ST), as it is known, presents itself as such an alternative. It has its roots in the Verstehen tradition, in particular in the idea that understanding human action requires that one revive, re-enact, rethink, or re-experience the mental states of the subject. On this view, CSP is an essentially dramatic idiom involving an ineliminable projective element. I argue that significant conceptual difficulties threaten the coherent development of this view. More critically, ST--even if coherent--simply fails to offer a genuine alternative to TT.;The Verstehen tradition, however, admits of a different interpretation, namely that the uniqueness of CSP lies not in its method but in its explanatory aim. According to this view, CSP is an interpretive enterprise aimed at rendering behavior rationally intelligible. Dennett and Davidson are perhaps the most prominent devotees of this view, but both develop it within the general framework of TT rather than as a clear alternative to it. Borrowing an analogy (exploited for very different ends by both Lewis and Brandom), I argue that psychological concepts are like scorekeeping concepts (e.g., a strike in baseball); to attribute a belief is not to point to a causally efficacious underlying internal state of an agent, but to locate the agent in a normative structure.