Abstract

This dissertation examines claims made by philosophers and cognitive scientists that cognition does not involve the use of internal, mental representations. Anti-representationalism, the name for the position advocated in such claims, has become rather popular in recent years; indeed, it has become fashionable to simply adopt anti-representationalism. Arguments in favor doing so usually go like this: Here is a model of some cognitive phenomenon. There are no representations in this model. If cognition in general works like this model does, there are no representations in cognition either. Such arguments suffer from a lack of attention to a particular type of detail. Though those who make arguments like this one may spend a good deal of time characterizing cognitive models, they rarely say in any detail what exactly they take representations to be. It is, therefore, difficult to know how seriously to take their arguments. ;The purpose of this dissertation is provide the detail necessary to make as good a case as is possible for anti-representationalism. In order to do that, I begin by sketching and defending a theory of representation. I then use this theory to show that, given three assumptions, none of the arguments for anti-representationalism are sound: all systems for which the assumptions hold will use internal representations. This leaves two options for defending anti-representationalism. First, one might argue that anti-representationalism is a claim about explanations of cognition; that is, it is a claim that our best explanations of cognition will not refer to internal, mental representations, despite the fact that cognitive systems may really use representations. Second, one might reject one of the three assumptions required to show that the arguments for anti-representationalism are unsound; in particular, one might defend anti-representationalism by claiming that the required assumption of metaphysical realism is false