Abstract

Haspelmath argues that linguists who conduct comparative research and try to explain patterns that are general across languages can only consider two sources of these patterns: convergent cultural evolution of languages, which provides functional explanations of these phenomena, or innate building blocks for syntactic structure, specified in the human cognitive system. This paper claims that convergent cultural evolution and functional-adaptive explanations are not sufficient to explain the existence of certain crosslinguistic phenomena. The argument is based on comparative evidence of generalizations based on Rizzi and Cinque's theories of cartographic syntax, which imply the existence of finely ordered and complex innate categories. I argue that these patterns cannot be explained in functional-adaptive terms alone.