This thesis investigates how prosodic phrasing influences listeners' interpretation of scopally ambiguous wh-questions in Japanese. It focuses on sentences such as that in (1), in which the wh-phrase can take either embedded or matrix scope: (1) [CP[IPJohn-wa [CP Mary-ga nani-o katta-ka] kikimasita]-ka]?John-TOP Mary-NOM what-ACC bought-Q askedQ‘Did John ask what Mary bought?’ (Embedded scope)or ‘What did John ask whether Mary bought?’ (Matrix scope) I argue that the comprehension of the wh-phrase is guided by the prosodic phrasing of the sentence, as specified in the Scope Prosody Correspondence in (2): (2) Scope Prosody Correspondence (SPC). The scope of a term X should not extend beyond the Major (phonological) Phrase (MaP) containing X. The SPC predicts that there is a strong bias for an embedded scope interpretation when a MaP boundary appears after the embedded Q-marker in (1). Without such a prosodic boundary, the SPC predicts that both embedded and matrix scope interpretations are equally available. The results of off-line comprehension experiments supported these predictions of the SPC. They also indicated that that prosodic phrasing rather than pitch compression is the primary determinant of listeners' scope assignments. In addition, it was shown that an embedded scope interpretation was induced by the prosodic boundary at the embedded Q-marker, not by the focus interpretation of the matrix verb. Further experiments showed that the SPC also applies to unambiguous wh-questions and to other sentences containing negative polarity items or quantifiers (e.g., interaction between negation and negative polarity sika, relative scope of quantifiers), but not to sentences lacking scope-relevant items. These results suggest that the SPC is not a construction specific principle effective only in wh-questions but rather a general principle that listeners use when they process sentences containing all and only scope relevant items. Finally, speakers of Tokyo Japanese sometimes inserted a MaP boundary after the embedded Q-marker, and sometimes didn't, for both scope interpretations of the wh-phrase. In contrast to the comprehension results, in production, the presence of a prosodic boundary after the embedded Q-marker was not reserved for embedded questions except when the two structures were explicitly compared.
| Identifer | oai:union.ndltd.org:UMASS/oai:scholarworks.umass.edu:dissertations-2443 |
| Date | 01 January 2005 |
| Creators | Hirotani, Masako |
| Publisher | ScholarWorks@UMass Amherst |
| Source Sets | University of Massachusetts, Amherst |
| Language | English |
| Detected Language | English |
| Type | text |
| Source | Doctoral Dissertations Available from Proquest |
Page generated in 0.0024 seconds