Concepts and Meaning

It certainly appears that there should be a relationship between concepts and meaning, but it is not entirely clear what this relation is. We shall assume that concepts are people's psychological representations of categories (e.g., apple, chair); whereas meanings are people's understandings of words and other linguistic expressions (e.g., "apple", "large chair"). 1 Currently, many cognitive scientists, especially psychologists, believe that concepts and meanings are at least roughly equivalent, with the meaning of an expression being its conceptual representation in human knowledge. From identifying the content of a concept, the meaning of the associated expression follows. Malt (1991) and Murphy (1991) review this position and various alternatives. In our paper, we shall argue that concepts and meanings differ substantially. Although they are related in important ways, the relationship is one of complementarity, not equivalence. To reach our conclusion, we shall review standard assumptions about concepts and meaning, challenge these assumptions, and present alternatives. The assumptions that we challenge, and that organize the paper, are: (1) propositional expressions represent concepts; (2) concepts are prototypes of exemplars; (3) concepts are decontextualized and universal in scope; (4) the meanings of words are concepts. We shall argue instead that perceptual symbols represent concepts; concepts are models for types of individuals in world models; concepts are contextualized and local in scope to situations; word meanings use concepts but are not concepts. Our tack in exploring these issues is to develop a theory of concepts and memory in the first three sections of the paper. In the spirit of cognitive linguistics, this theory utilizes perceptual representations and situational knowledge extensively. The final section compares concepts, as defined in our theory, with relatively well-accepted notions about meaning. We shall then assess whether concepts are equivalent to meaning, or whether they exhibit some other relationship. Please note that the bulk of this 'working paper' represents a theory of concepts and memory in the early stages of development. We are the first to acknowledge that the majority of our claims lack strong empirical support and that increased precision is necessary at the theoretical level. Although evidence exists for some aspects of our theory, many aspects rest on more of a rationalist analysis of how a cognitive system might compute concepts and meaning. This paper outlines our theory in its current form so that we can begin to examine its claims empirically and implement it computationally. A variety of experimental …