There is no Post-focal De-phrasing in English

Danfeng Wu

doi:10.3765/amp.v9i0.4930

Abstract

This paper studies the relationship between prosodic phrasing and prominence by addressing the questions of whether every prosodic phrase must have a head (a most prominent sub-constituent), and if so, how the head is marked. I study these questions by examining the intermediate phrase (iP) in English. If every iP must have a head, and this head must be marked by a pitch accent, then in an environment without any pitch accent, there should be no head/non-head distinction. And if there is no head, there should be no iP in this context either. I conducted a production study in English, and found durational evidence suggesting the presence of iP boundaries in an accent-less context. I also searched for durational evidence for iP-level prominence distinctions in this context, but here my results are mixed. One theoretical possibility that is compatible with my findings is that every phrase must have a head, but the head of an iP can be marked by something other than pitch accent, for example by phrasal stress.

Highlights

Prosodic structure at the sentence level is defined by prominence and phrasing
If the head of an intermediate phrase (iP) must be marked by pitch accent, in a context with no pitch accent, there should be no head and no prominence distinction
If a larger prosodic phrasing break induces more preboundary lengthening, and if as I observed, under broad focus, the boundary immediately following the particle is stronger than the boundary between the verb and the particle, I expect P2 to be lengthened more than V2 in V+Part structures

Summary

Introduction

Prosodic structure at the sentence level is defined by prominence and phrasing. A standard hypothesis posits an intimate relationship between prominence and phrasing such that every phrase has a head (the most prominent sub-constituent). Selkirk (1995) and Ito and Mester (2003), for example, proposed headedness as an inviolable constraint that holds universally. This paper tests this hypothesis by studying the intermediate phrase (iP) in English. If the headedness hypothesis holds, every iP must have a head. If every iP has a head, we may further ask how the head of an iP is marked. If it must be marked by pitch accent (I call this the prominence by pitch accent hypothesis), every iP must contain a pitch accent. If it must be marked by pitch accent (I call this the prominence by pitch accent hypothesis), every iP must contain a pitch accent. Beckman (1996) and theories underlying ToBI transcription conventions (Beckman and Ayers Elam 1997; Silverman et al 1992) adopted this assumption, and claimed that accent-less iPs do not exist

Methods

Results

Discussion

Conclusion