 |
ESCA Workshop on ProsodyLund, Sweden |
 |
|
ESCA Workshop on ProsodyLund, Sweden |
|
Prosodic grouping can be defined as the way in which phonological units are assumed to join together to form larger units. Such grouping has, at various times and by various authors, been accounted for by appealing to one or more of three distinct but inter-related characteristics which I refer to here as peaks (= prominence, headedness), boundaries (= disjunctive, discontinuity) and cohesion. (= juncture, continuity).
There is a certain redundancy in these three characteristics. Thus boundaries and cohesion, for example, are mutually exclusive, together defining prosodic goups (=units, constituents or domains). The relationship between domains and groups is more subtle and also more theory dependent. Halle & Vergnaud (1987) argue that the pure representation of heads (= peaks) and the pure representation of domains (= groups) constitute "conjugate representations" which together define the notion of government: "the property of being a head is the same as that of being a governing element in a constituent, and the property of being in some domain is the same as that of being governed by some head." (p. 16)
Despite this redundancy, I argue that these three characteristics are worth distinguishing when we look at the way prosodic constituents are defined. Before looking at the way they apply to spoken language, as an exercise it may be interesting to see how these concepts can be applied to written language where the notion of constituents seems more easily defined.
Bibliographic reference. Hirst, Daniel (1993): "Peak, boundary and cohesion characteristics of prosodic grouping", In Prosody-1993, 32-37.