Use this URL to cite or link to this record in EThOS:
Title: The voicing contrast in Serbian stops
Author: Sokolovic-Perovic, Mirjana
Awarding Body: University of Newcastle Upon Tyne
Current Institution: University of Newcastle upon Tyne
Date of Award: 2012
Availability of Full Text:
Access from EThOS:
Access from Institution:
This study investigates aspects of the phonetic realisation of the voicing contrast in Serbian stops. It determines the basic set of acoustic correlates of the voicing contrast, and examines the effect of several linguistic and speaker factors on these correlates. The thesis explores fine details of the phonetic realisation of the voicing contrast that are specific to Serbian, and evaluates the existing theoretical models of the voicing contrast in relation to Serbian data. Twelve native Serbian speakers produced stops in a range of positions in isolated words and in a sentence frame. Acoustic analysis revealed that the voicing contrast is robust in Serbian in all word positions and for each speaker. Utterance-initially Serbian contrasts prevoiced stops and stops with short to intermediate positive VOT values. In word-initial intervocalic position the relevant correlates are duration of voicing in the closure and closure duration; in word-medial and final position the correlates are duration of voicing in the closure, closure duration, and preceding vowel duration. The following factors affect the realisation of the voicing contrast: the place of stop articulation, the vowel environment, gender, age, and place of birth of speakers. This variability is only partly attributable to universal constraints, and is mostly specific to Serbian. The results suggest that the existing models cannot account for the type of realisation of the voicing contrast found in Serbian, in particular for the status of intermediate VOTs and the role of closure duration and preceding vowel duration. Some of the main assumptions of these models should be re-assessed in order to include these findings. Further, these models are unable to account for non-universal and non-contrastive variability found in Serbian and other languages. Advantages and difficulties associated with the integration of the existing models with elements of an exemplar-based model of phonological knowledge are discussed.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available