An investigation of the effect of semantic and contrastive pitch accents of prosodic phrasing | Posted on:2005-09-28 | Degree:Ph.D | Type:Dissertation | University:University of Pittsburgh | Candidate:Vance, Janice E | Full Text:PDF | GTID:1455390011951377 | Subject:Health Sciences | Abstract/Summary: | PDF Full Text Request | This study examined the relationship between semantic and contrastive accents and prosodic phrasing in short declarative sentences. Specific predictions derived from Selkirk's (2000) proposed hierarchy of constraints for prosodic phrasing in English (Align Focus, Wrap XP, DEP Accent, MiP Accent) were examined. Utterances were elicited in read dialogue contexts from fifteen male speakers ranging in age from 21 to 25 years. Six focus conditions were created to elicit sentences with single or double semantic accents (SA) and contrastive accents (CA): (1) single SA; (2) single CA; (3) two SAs; (4) one CA followed by one SA; (5) two CAs; (6) two CAs, one within an appositional noun phrase. Separate one-way ANOVAs (p < .016) were conducted for each of three dependent measures: phrase-final lengthening (PFL), silent inter-word interval (IWI), and fundamental frequency shift (FoDelta). For PFL, there were significant differences between Condition 5 and all other conditions, and between Condition 6 and all other conditions. No significant differences existed between Conditions 1 and 3, nor between Conditions 2 and 4. For IWI, the only significant differences were between Condition 6 and all other conditions except Condition 5. For FoDelta, only Condition 6 was significantly different from all other conditions. These results suggest that in these contexts, a constraint requiring the alignment of the right edge of a focus constituent with the right edge of a prosodic phrase (Align Focus) did not over-ride constraints requiring the presence of at least one pitch accent in each prosodic phrase and a faithful correspondence between output and interface representations. However, even in contexts with more than one pitch accent, speakers did not produce prosodic breaks following words in semantic focus or words in contrastive focus when there was another word in semantic focus in the utterance. This suggests a constraint requiring the maintenance of prosodic cohesion (Wrap XP) also out-ranked an Align Focus constraint. It is only in the context of two separate points of contrastive focus that the Align Focus constraint appeared to out-rank this cohesional constraint. A modification of Selkirk's constraint hierarchy is suggested. | Keywords/Search Tags: | Prosodic, Semantic, Contrastive, Accents, Focus, Constraint, Pitch | PDF Full Text Request | Related items |
| |
|