Please use this identifier to cite or link to this item:
Title: A comparison of Japanese and Blackfoot vowel devoicing
Authors: St. Goddard, Rosalind
Keywords: Linguistics;Algonquian languages;Blackfoot;Siksika language;Japanese language;Phonology;Optimality theory (Linguistics);Constraints (Linguistics);Distinctive features (Linguistics)
Issue Date: Sep-2014
Publisher: University of Calgary
Citation: St. Goddard, R. (2014). A comparison of Japanese and Blackfoot vowel devoicing. Calgary Working Papers in Linguistics, 28(Fall), 81-96.
Abstract: This paper compares and contrasts the factors that contribute to devoicing in Japanese and Blackfoot. Japanese vowel devoicing has received rigorous discussion in linguistic literature. Tsuchida (2001) provides a particularly persuasive argument for Japanese vowel devoicing using the Optimality Theory Framework (Prince and Smolensky 2004); she argues that all Japanese voiceless fricatives are specified for [SG] and devoicing occurs when this [SG] feature is shared within a syllable. The notion that voiceless vowels carry the feature [SG] can also be extended to instances of Blackfoot vowel devoicing. Blackfoot voiceless vowels generally occur in two contexts: They occur word finally, and word-medially when they are followed by the palatal/dorsal sounds [x]/[ç], which are orthographically represented as <h>. In contrast to Japanese voiceless fricatives, it appears that not all Blackfoot voiceless fricatives distribute the [SG] feature. The Blackfoot palatal fricative [ç] and the dorsal fricative [x] both trigger devoicing, whereas the fricative [s] does not. To explain this patterning of [x] and [ç], Reis Silva (2008) argues that [x] and [ç] are not fricatives, but rather preaspiration ([SG]) specified on certain obstruents. In this paper, I will discuss the constraints proposed in Tsuchida (2001), and extend/adapt those constraints to Blackfoot word final vowel devoicing. Additionally, In my analysis of Blackfoot word-medial vowel devoicing, I will adopt Reis Silva’s (2008) analysis that [x]/[ç] are not fricatives, but preaspiration specified on obstruents. Lastly, I argue that the word-medial vowel devoicing that occurs with [x] and [ç] is phonological rather than phonetic.
ISSN: 2371-2643
Appears in Collections:Volume 28. Fall 2014

Files in This Item:
File Description SizeFormat 
028_fall_14_stgoddard.pdf334.17 kBAdobe PDFView/Open

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.