@MISC{Explanation_markbaker, author = {Morphosyntactic Explanation}, title = {Mark Baker The Mirror Principle and}, year = {} }
Share
OpenURL
Abstract
Pretheoretically, there are processes in languages of the world that have both a syntactic component and a morphological component. An example is the English passive, illustrated in (1): (1) a. The cats chase the mouse every day. b. The mouse is chased by the cats every day. (lb) differs from (la) in two ways. First, the NP that bears the patient or "logical object" semantic role appears as the surface direct object in (la) but as the surface subject in (lb). Second, the main verb in (lb) is morphologically derived from the (stem of the) verb in (la) by suffixing the-ed morpheme. Any complete account of the passive construction will have to encompass both of these aspects, the syntactic and the morphological. On this, all are agreed. How to integrate the two components into a unified account is another matter, however, and differing viewpoints abound regarding which component is primary and which is derived, at what level(s) of representation the two are explicitly related, and so on (for a cross section, see Chomsky (1981), Bresnan (1982c), Perlmutter and Postal (1977), Marantz (1981)). Part of the reason for this diversity is that the phenomena in and of themselves do not supply a wide enough range of evidence to guide theoretical decisions in this area. This article will shed new light on these issues by considering interactions of these processes in morphologically complex languages. In particular, it will argue that the morphology and the syntax in this class of cases must be two aspects of a single process. This result in turn will be shown to place strong, substantive constraints on the kind of syntactic framework that should be adopted.