Abstract

We glide easily from thought and talk about worldly objects to thought and talk about the contents of our beliefs about such worldly objects all the time. Smith ask Jones about the whereabouts of their pet cat and on the basis of Jones’s assertion that the cat is on the mat, Smith comes to believe that the cat is on the mat. Black in turn may ascribe to Smith the belief that the cat is on the mat. Such transitions from thought and talk about worldly objects to thought and talk about states of mind are so familiar to us as to seem second nature. But there is a long-standing philosophical tradition, originating with Frege, but endorsed by philosophers with otherwise varying philosophical outlooks, which makes the very possibility of such transitions puzzling. That tradition assumes that in making at least certain attitude ascriptions – so-called de dicto or “notionally sensitive” ascriptions -- speakers refer to, describe, quantify over, or somehow pragmatically implicate the notions, representations, or modes of presentations that plausibly figure as constituents of our mental contents -- either to the exclusion of the worldly objects themselves or in addition to those objects. Such attitude ascriptions are widely taken to be the primary or unmarked case of an attitude ascription. But it is seldom acknowledged that twin facts that on this approach worldly objects will relate to the representational items that supposedly serve as ingredients of thought content in a one-many fashion and there is no automatic way “back-up” from worldly objects to modes of presentation thereof together generate a mystery about how possibly we are able execute transitions from thought and talk about worldly objects to thought and talk about representational states of mind. It is argued in this essay that the way around this mystery is to see that de re, rather than de dicto ascriptions are the unmarked form of attitude ascription and that our representations of mental contents are parasitic on our representations of worldly objects. That is, we talk about the contents of our states of mind not by adverting, in the first instance, to talk about peculiarly mental or representational entities like notions or modes of presentations, but primarily by talking about worldly entities themselves. That is, to attribute to another the belief that the cat is on the mat, one need not refer to modes of presentations, or their ilk, of said cat or said mat, but only to the relevant cat and the relevant mat.