Representational versus Derivational approaches to Syntax: Optimality Theory and Chains (1999)
Abstract:
Introduction What do the terms "derivational" and "representational" actually mean for linguistic theory? As formally equivalent, mutually intertranslatable ways of characterizing sets of linguistic expressions, derivationality might appear to be a mathematical topic orthogonal to the core issues of linguistics -- an epiphenomenon of the way linguistic insights are expressed. Such a position insulates linguistic theory from questions about computational complexity at the cost of retreating to a level of abstraction at which a linguistic principle is just a schema for an infinity of equivalent but unspecified proposals. Instead, I take the position that Universal Grammar's derivationality or non-derivationality is an empirical, and open question. This empirical position on derivationality prompts an important question when juxtaposed with considerations of conceptual simplicity: can particular theories in one mode or the other be actually intertranslated? In this essay I attem

