Abstract

Certain belief-ascription statements suggest that people can think about the same object, even if that object doesn't exist. Peter Geach, who in 1967 first discussed these statements, gave them the name of statements of intentional identity. In an especially puzzling form of these statements, a pronoun in one belief clause has its antecedent in another: "Hob believes a witch is F, and Nob believes she is G." In this form, the statements raise three interesting problems. They generate puzzles for a program in philosophy of mind, according to which intentional properties of thought are to be analyzed as theoretical analogs of something independently explicable. They are beyond the scope of standard intensional logic , and therefore jeopardize its explanatory value in certain respects. For instance, they jeopardize the received theory of the de dicto/de re distinction in terms of the scope distinctions of SIL. They pose problems for certain theories of anaphoric pronouns, in the context of SIL. ;At first it seems possible to modify SIL so as to yield a quantificational interpretation of the recalcitrant intentional identity statements. In the modified theory, the statements are analyzed so that a wide-scope quantifier binds variables in separate belief contexts. The problem with such quantificational theories is that they are unable to account for another puzzle about intentional identity. Intentional identity appears not to be a symmetric relation: it can be argued that the inference from "Hob thinks a witch is F and Nob thinks she is G" to "Nob thinks a witch is G and Hob thinks she is F" is not a plausible one. A better formal approach than the quantificational ones contains two key components. The first is a framework for anaphoric pronouns combining certain features of quantificational, discourse referential, and demonstratival interpretations of pronouns. The second is the use of structured domains in the formal models. Various interpretations of the new formalism are possible, each drawing on a different analogical theory of the nature of thoughts and their intentional properties. The new framework explains in a unified fashion not only the apparent asymmetry of intentional identity, but a number of other perplexing features of belief.