Response to Edwin William's Review of Indices and Identity

Abstract

Our response to Edwin Williams's review of our book Indices and Identity (II before there can be meaningful debate it must be clear to both participants what the issues are. In I&I, our concern is with the nature and representation of syntactic identity. The central issue we address is a foundational one: to determine the conditions under which two syntactic inscriptions are occurrences of the same syntactic object. Any answer to this question will have implications for a broad range of topics, including, we argue, for the theory of anaphora. It is a central consequence of our approach to syntactic identity that if two syntactic representations, or inscriptions, have the same index, they are thereby indicated to be occurrences of the same syntactic expression. If two syntactic inscriptions do not have the same index, they are thereby indicated to be occurrences of different syntactic expressions. Since the identity conditions given are SYNTACTIC identity conditions, it is to be expected that two inscriptions might be syntactically identical but lexically distinct. This circumstance holds when, for example, an occurrence of Cicero and an occurrence of he have the same index. We refer to the relationship between syntactically identical but lexically distinct inscriptions as VEHICLE CHANGE. This relation is not arbitrary, and a good part of I&I is devoted to investigating the constraints on Vehicle Change. (See especially the discussion on pp. 218-27 and 275-88.) We will consider this relationship in more detail below. Our conception in I&I raises a question as to the relationship between the indication of syntactic identity and the theory of anaphora. What we hold on this point is that the representation of syntactic identity, expressed as a theory of indexing, is sufficient to express the syntactic relation of anaphora: expressions that have the same index are syntactically indicated to be anaphoric. The intuition behind this is easy to grasp: if an expression is repeated, it follows that the two occurrences have the same semantic value, in the cases at hand, its reference. If two inscriptions are coindexed, and hence repetitions in the appropriate sense, they are coreferential. On the other hand, if two inscriptions are not coindexed, whether they are coreferential or not depends on the structure of context, formally defined. The initial chapters of I&I formally develop the syntax and semantics of our treatment of anaphora, at the heart of which lies the intuition mentioned, and the theory is applied to a variety of empirical puzzles.