It is a well·known fact that. although function words in English can reduce in most environments, the reduction is blocked at certain syntactic junctures, whose characterization has proved elusive in the past. In this paper we offer a unified account of those environments in which reduction is blocked. Our approach is to define the relevant contexts in prosodic rather than in syntactic terms, stating reduction conditions on the domain of the phonological phrase. In this we rely on past work in prosodic phonology. in particular that of Selkirk (1978. 1984, 1986) and Nespor and Vogel (1982, 1986). Rather than describe the environments where reduction is blocked directly in syntactic terms, we instead focus on the syntactic conditions on phonological phrasing. We show that there are two kinds of such conditions: those coming from the general phrasing algorithm, and those imposed by a well-defined class of syntactic constructions. The interaction of these mechanisms provides phonological phrase breaks precisely at the points where no reduction is allowed. The data we account for subsume cases that have been used in the past to argue for the influence on phonology of extraction and deletion sites. We show that reference to empty nodes is not necessary and, moreover, would lead in some cases to incorrect predictions. We first give an account of the so-called auxiliary reduction, demonstrating the relevance of the domain of the phonological phrase for this phenomenon. After developing a phrasing algorithm which properly generates phonological phrases in English. we then proceed to discuss a small subclass of cases which draw upon an additional source of phrasing. We conclude with a discussion of the relevance, or rather the irrelevance, of empty categories for phonological phrasing.

This paper is copyrighted, and is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0) - see https://creativecommons.org/licenses/by-nc-nd/4.0/