On Nominal Extraposition and "Empty" Zt: A Reply to Michaelis & Lambrecht (1994)

Abstract

Particles have always been a source of embarrassment for most linguistic theories. The reason for this embarrassment, more than anything else, has to do with the semantics of these elements. Such is the case of phrasal-verb constructions, subjects of atmospheric verbs, cleft-sentences, and various types of extraposition. Some theories have no problem with assigning panicles no semantic value (e.g., most Chomrkyan and post-Chomkyan theories); others on the other hand, rake pains to show that all particles must have some semantic content (e.g., cognitive grammar). Here we would like lo concentrate on just one of these cases, namely the nominal-extraposition construction. It has recently been claimed by Michaelis & Lambrechr (1994) that the subject of this construction has no semantic content. We will show that this is not the case, and, in fact, the problem can be easily solved if a frame-semantics approach is incorporated into the description. The purpose of our paper, then, is not to challenge the it analysis, but rather to refine it.