Introduction

The issue this book addresses, the relationship of linguistic and conceptual representation, appears on the surface to be a straightforward matter. Insofar as people have understandings of their language, we must assume that there are mental representations of language. Likewise, insofar as people can think about the world around them, they must have higher cognitive representations. The question is how these two forms of representation are related. A clear example of this approach to language and cognition, and an attempt to answer this question, can be seen in Jackendoff's work. For instance, Jackendoff (1983, 1992) proposes a number of representational levels including phonetic, syntactic, semantic, and conceptual levels. The core level of representation, for Jackendoff (1983: 17), is a conceptual level, “a single level of mental representation … at which linguistic, sensory, and motor information are compatible”. Thus, linguistic representation involves levels of representation separate from, although translatable into, conceptual structure. Jackendoff's theory clearly provides an answer to the book's question.

Inherent in this question, however, is a fundamental epistemological assumption that linguistic and conceptual structures, whether they are viewed from functionalist, formalist, or cognitivist perspectives, are matters of mental representation. Within most scientific branches, it is assumed that objective observers impartially study and describe externally observed phenomena. While this is necessary for science, it is common practice to take these objectively defined structures and to turn them into mental structures.