This workshop continues the discussion of best practices in sociophonetics begun at NWAV33. The ever-expanding range of knowledge necessary to do high quality work in the interdisciplinary field of sociophonetics demands that we provide quick access to the best methodological, technical, and procedural information to all researchers.
This year the workshop focuses on the coarticulation effects of nasals and velars on adjacent segments, especially on vowels. While both nasals and velars are implicated in vowel shifts, vowels adjacent to nasals are often excluded from studies because of the measurement difficulties that nasalization poses and the effect of velars are often ignored.
Velar Obstruent-Vowel Coarticulation: Why are velars so special to sociophoneticians?
Matt Bauer (Illinois Institute of Technology)
Tom Purnell (University of Wisconsin-Madison)
The goal of this segment is to describe the effect of velar obstruents on neighboring vowels. As a coherent sound class, velar obstruents may induce regional variants. In VA and SC, for example, velars before /aɹ/ (card, guard) result in an intervening [j] ([kjaː], [gjaː]. (Kurath 1964) Present research on velar obstruents and neighboring vowels reveal a more asymmetrical pattern: /æ/ raises towards /e/ before /g/ but not /k/ in some areas of the US and Canada (Zeller 1996; ANAE). Other asymmetries are due to history, resulting in a paucity of /g/--but not /k/--before /i, u, o/. Moreover, variable regional pronunciation of vowels /ɔ ~ a/ in words such as log, dog, and frog suggest that /g/ is a less stable articulatory environment than /k/. ANAE finds vowel raising before the voiced velar troubling “since most environmental effects are the products of the action of a uniform articulatory apparatus and operate in the same way across dialects” (p. 183). In this presentation, we describe the methodology of identifying velar coarticulation, motivating the importance of informing acoustic analysis with articulatory data.
A Descriptive Approach to the Measurement of Nasalization
Ryan Shosted (University of Illinois-UC)
Velopharyngeal opening modulates the acoustic output of the vocal tract in a number of well-studied ways: introduction of nasal formant(s); increase in formant bandwidths; damping of overall sound energy; and the introduction of anti-resonances. Unfortunately, after many decades of research, it seems unlikely that an invariant acoustic measure of vowel nasalization will be determined. Nevertheless, an acoustic characterization of nasalization is still commonly sought by linguists and clinicians, so a diagnostic approach to the problem is warranted. This talk will review methods that have been proposed for detecting and describing nasalization (including as a function of time), highlighting the challenges involved in acoustic analysis. Practical suggestions will be given for semi-automatic formant extraction in nasalized vowels, based on the presenter's current research on vowel nasality in Hindi and Portuguese. An aerodynamic approach to nasalization will also be outlined, with practical suggestions as to how such a program can be carried out in the field.