Generics and Exceptinos: A Reply to Cohen (2004)

Abstract

It is well known that exceptions are allowed in generics. Concerning this widely-agreed-upon phenomenon of exceptions in generics, Cohen (2004) argues that exceptions are allowed only if "homogeneity" is not violated. That is, the exceptions should not form a salient "chunk" of the domain of the generic. For this, he proposes two ways of mapping of cognitive mental representations, namely, "tree" and "geometric" representations. It will be argued in this paper, however, that choices between these two mental representations claimed to be involved in the interpretation process of generics are quite arbitrary, and that counterexamples also exist for the "homogeneity" requirement. Given this, the main purpose of this paper will be to discuss the problems of Cohen's theory. Its other purpose will be to try to delve into the fundamental issue of the meaning of generics, and to suggest that generics, similar to metaphors, involve cognitive conceptualizations based on the language users' encyclopedic knowledge, world knowledge from experiences, common sense, beliefs, stereotypes, prejudices, etc.