Fictionalism about Folk Psychology

Abstract

For nothing is actually removed from existence by being labeled ... appearance . . . What appears is there, and must be dealt with; but materialism has no rational way of dealing with appearance.-BradleyAppearance and Reality, 1930, 12.1. IntroductionFictionalists avoid commitment-full-fledged ontological commitment, that is. To be a fictionalist about a given domain is to adopt a strategy of taking discourse about the domain seriously while at the same time stopping short of accepting that such discourse entails any, potentially embarrassing, ontological commitments. Endorsing fictionalism about some domain is thus meant to provide a way of taking a less-thanfully realist stance towards the domain's entities without rejecting the value or usefulness of discourse about the domain altogether. Sainsbury highlights this as the main motivation of fictionalism:the starting point for fictionalism is some kind of ontological scruple: one cannot bring oneself to believe in moral values, nonactual things, unobservable things, or abstract things. But one has somehow to do justice to the fact that one cannot simply throw away the related regions of discourse: morality, modality, elementary physics, or mathematics. Fictionalism to the rescue. (2010, 2)This paper argues that fictionalism about folk psychology, FaF, is ill motivated in any domain. It is not a successful way of dealing with eliminativist threats aimed at classical cognitivist theories-theories which predict that folk-psychological constructs, understood subpersonally, will be vindicated by a mature cognitive science. Nor would FaF be an appropriate response to worries about the ontological status of mental states and the determinacy of their content, where such mental states are understood as featuring in reason explanations about the actions of people. The paper's moral? Friends of folk are best advised to expose such threats as empty rather than by making fictionalist ontological concessions.The case against FaF is made in the following steps. It is argued that there is no advantage in trying to vindicate folk by treating the constructs of classical cognitivism-viz. subpersonal mental representations-as useful fictions as opposed to serious scientific posits or as serving as the basis for philosophical explanations. Both scientific and philosophical considerations point to the conclusion that subpersonal representations of the sort that classical cognitivism posits should be eliminated, not preserved by our best science of mind (Section 1). Yet there is no need to assume that folk-psychological explanations are subpersonally based. It is possible and plausible that such explanations are based, just as they appear to be, in nonscientific interpretative, narrative practices. A recent attempt to motivate FaF based on this assumption is examined and rejected (Section 2). Then a more compelling, Dennettstyle rationale for adopting a FaFish line based on worries about the indeterminacy of folk-psychological attributions is considered. Dennett endorses FaF, broadly construed, in arguing that while folk-psychological phenomena exhibit objective patterns they are nevertheless, at best, mildly real (Section 3). The final section offers three considerations that should encourage the reader to resist such FaFish conclusions (Section 4).1. Fictionalism about Folk Psychology: Take OneClassical cognitivism characterizes subpersonal cognitive states in a way that is deeply indebted to our everyday, folk-psychological conception of mind. Ramsey (2007) exposes classical cognitivism's commitment on this front, showing how it gives rise to an outlook on representation that amounts to a sort of merger between classical computational theory and folk psychology (Ramsey 2007, 38). Why so? What's the connection? On the one hand, cognitivism embraces the view that cognition is computation, which [is] . . . understood as a form of quasi-linguisticsymbol manipulation done in accordance with specifiable rules (39). …