ABSTRACT
This chapter deals with boundary strength phenomena, particularly with final lengthening. The elongation of segments in the vicinity of certain linguistic boundaries in comparison with their duration earlier in the utterance is a widely attested cross-linguistic phenomenon (cf. Vaissiere 1983). Such preboundary lengthening is usually interpreted as a correlate of some sort of disjuncture present in the organization of speech, or in other words as a boundary marker of some constituent edge. Other wellknown boundary signals are pauses and pitch movements. These temporal and melodic means of boundary marking have been shown to be positively correlated with the rank of the boundary they signal, i.e. higher boundaries are stronger than lower boundaries and these differences in boundary strength are reflected in the realization of the temporal and melodic edge markers (cf. Goldman-Eisler 1968, and Cooper and Pacia-Cooper 1980). In a study of boundary strength, it is thus important to consider the following questions: (i) the nature of the constituents involved and their hierarchical level; (ii) the physical correlates of boundaries and their varying realization according to boundary type and strength; (iii) the domains over which the boundary marking phenomena may extend. This introductory section is concerned with these questions, together with the issue of evidence for boundary strength in EP.
