A Unified Account of Two Vowel Devoicing Phenomena: the Case of Cheyenne

This paper investigates two vowel devoicing processes in Cheyenne, which appear on the surface to be fundamentally different, occurring in distinct segmental and prosodic environments. One process occurs in phrase-final vowels in any segmental environment, while the other occurs only before voiceless consonants in the surface penultimate vowels of some words. The first is consistent with typological expectations and is phonetically grounded, whereas the second is at first glance, neither typologically expected nor phonetically motivated. I provide a unified Stratal Optimality Theory account of these processes, demonstrating that both can, in fact, be treated as cases of domain-final devoicing, and attributed to the same family of positional markedness constraints. Different rankings of the markedness constraints relative to a faithfulness constraint result in different segmental conditions for the two processes. Moreover, I suggest that the two processes may be related via Domain Generalization, whereby a phonetically motivated utterance-final effect phonologizes and extends to smaller prosodic domains. In this way, while the word-level process is not itself phonetically motivated, it can be understood as an extension of another phonetically motivated process in the same language.

Reconciling CV phonotactics and high vowel deletion in Japanese

High vowel devoicing is a productive process in Japanese, where /i, u/ become unphonated between voiceless obstruents. Recent studies have shown that the vowels can completely delete as a result of the process, resulting in surface consonant clusters. This seemingly conflicts with the strong CV phonotactic preference that has repeatedly been shown in both phonological and psycholinguistic studies of Japanese. This paper proposes that the apparent conflict can be resolved by having phonotactic repairs and high vowel devoicing apply at different phonological levels, adopting a more sophisticated phonological representation than simple /underlying/ vs. [surface] forms. The proposed framework also makes an empirically testable prediction regarding syllabification of clusters that result from high vowel deletion.

Assessing surface phonological specification through simulation and classification of phonetic trajectories

Many previous studies have argued that phonology may leave some phonetic dimensions unspecified in surface representations. We introduce computational tools for assessing this possibility though simulation and classification of phonetic trajectories. The empirical material used to demonstrate the approach comes from electromagnetic articulography recordings of high-vowel devoicing in Japanese. Using Discrete Cosine Transform, tongue-dorsum movement trajectories are decomposed into a small number of frequency components (cosines differing in frequency and amplitude) that correspond to linguistically meaningful signal modulations, i.e. articulatory gestures. Stochastic generators of competing phonological hypotheses operate in this frequency space. Distributions over frequency components are used to simulate (i) the vowel-present trajectories and (ii) the vowel-absent trajectories. A Bayesian classifier trained on simulations assigns posterior probabilities to unseen data. Results indicate that /u/ is optionally produced without a vowel-height specification in Tokyo Japanese and that the frequency of such targetlessness varies systematically across phonological environments.

Onset Cy and High Vowel Devoicing in Japanese

Japanese has contrasts between onsets romanized as singletons and Cy clusters (e.g., kaku ‘rank’ vs. kyaku ‘guest’). It is uncertain whether Cy should be treated as a distinctively palatalized consonant /Cy/ or as a /Cy/ cluster; the phonetic realization is compatible with either. A distributional argument favoring the /Cy/ analysis is the neutralization of the contrast before front vowels: Ci, Ce, *Cyi, *Cye; the absence of */Cyi/ and */Cye/ follows automatically from */yi/ and */ye/. In this paper, high vowel devoicing is used as a diagnostic to investigate this phonemicization issue. A production experiment was conducted, comparing the rates of vowel devoicing in words containing Cu/iC and CyuC sequences (e.g., maputa vs. mapyuta). The devoicing rate was significantly lower in CyuC than in Cu/iC. This supports the cluster analysis /Cy/ (with the /y/ inhibiting devoicing) over the singleton analysis /Cy/ for the onset palatalized consonants of Japanese.

Sociolinguistic Factors Affecting Vowel Devoicing in Spontaneous Japanese: A Preliminary Corpus-based Analysis

Maekawa and Kikuchi (2005) used the Corpus of Spontaneous Japanese (CSJ) to analyse the frequency of vowel devoicing in different phonological environments. According to their analysis, the devoicing rate is highest when a fricative is followed by a stop and lowest when an affricate is followed by a fricative. Moreover, the results of their study suggest that devoicing also occurs in atypical environments, as in non-close vowels and in contexts where a vowel is followed by a voiced consonant. However, the frequency of devoicing is conditioned not only by phonological factors but, at a certain extent, also by extra-linguistic and sociological factors such as age, gender and speech style. This paper aims to analyse the sociolinguistic variation of vowel devoicing in spontaneous Japanese using the CSJ-Core consisting of about 45 hours of speech, all of which have been (sub-)phonemically segmented. Age, gender and speech style variation has been analysed for different phonological environments. Particular attention will be given to atypical environments which showed a higher rate of variability.