Talks by Bob Ladd and John Esling this week

This week was a busy and exciting P-side week in the department, with visits and guest talks by both Bob Ladd (University of Edinburgh) and John Esling (University of Victoria, emeritus). On Tuesday, Bob Ladd presented "Lexical Allophones", a discussion of marginal phonological contrasts, subject to both phonological conditioning and lexical effects. On Thursday, John Esling presented "The Effect of the Laryngeal Articulator on Vowel Quality", showing some excellent MRI images of different laryngeal states and their effects on tongue shape and cavity sizes. The abstracts for both talks are below.

The Phonetics Brigade

(L-R: Alexei Kochetov, John Esling, Jessamyn Schertz)

Lexical Allophones
(Robert Ladd, Edinburgh)

In this talk I will discuss a type of marginal phonological contrast that I propose to call lexical allophony.  Lexical allophones are pairs (or perhaps occasionally triplets) of similar sounds that are categorically distinct phonetically and generally recognizable as different by native speakers, but which distinguish few minimal pairs and may be felt as related variants. Their distribution is often subject to ordinary phonological conditioning, which may however be irregular or limited; where phonological conditioning fails or does not apply, individual lexical items may be variable; nevertheless, there is often broad agreement about which items contain which sound.  This definition organizes a number of superficially diverse cases – from German ‘ich-Laut’ and ‘ach-Laut’ through higher and lower mid vowels in French and Italian to the apparent three-way laryngeal distinction in Swiss German stops - along a continuum according to the relative importance of phonological and lexical conditioning in the distribution of the marginally contrastive sounds.

By proposing that lexical allophony is a coherent phenomenon that varies along a continuum, I attempt to distinguish it from other cases that might be relevant to any discussion of 'marginal contrast'.  In particular, lexical allophony differs from 'near-merger', in which clear phonetic distinctiveness is lost, and from various other semi-systematic relations between sound types, such as the cases covered by Wells’s English lexical sets bath and cloth, or the variation between /i/ and /ε/ in English words like economics, evolution, methane, premature, where there is no sense that the sounds involved are mere variants of each other.  Classifying such marginal contrasts into a few broad types should help clarify how they can be integrated into our phonological theories.


The Effect of the Laryngeal Articulator on Vowel Quality
(John Esling, University of Victoria)

This talk addresses how vowel quality interacts with larynx state. Larynx height varies in relation to vowel quality, but inconsistently so (Ewan & Krones, 1974; Ladefoged, DeClerk, Lindau, & Papçun, 1972). Some have suggested that the intrinsic F0 of vowels arises from lingual-laryngeal interaction in producing vowels (Ohala, 1987; Whalen, Gick, Kumada, & Honda, 1998). Lingual-laryngeal interactions form an important part of Esling’s (2005) Laryngeal Articulator Model, which advocates a reshaping of the vowel space to reflect the laryngeal component of vowel articulation, and Moisik (2013) proposes that lingual-laryngeal interaction emerges in numerous phonological phenomena around vowel quality in laryngeal and pharyngeal consonants. But uncertainty still remains about the nature of the articulatory changes in the larynx and pharynx in response to vowel articulation.

Magnetic resonance imaging (MRI) lends itself well to studying lingual-laryngeal state. An MRI data set will be presented featuring key vowel qualities produced by two trained phoneticians in a number of different laryngeal phonetic contexts captured using 2D (axial, coronal, and sagittal) multi-slice sequences and 2D midsagittal static and real-time sequences. Variation in laryngeal structures as a function of vowel quality and the different contexts (modal and creaky phonation, glottal and epiglottal stop, and raised larynx voice) will be examined. Figure 1 shows a sample data set. The focus will be on laryngeal constriction, larynx height, and epilaryngeal cavity dimensions, and their relation to tongue shape.

Figure 1: Midsagittal MR images of a phonetic modal vowel series produced with controlled phonatory quality and pitch (left of each image = anterior).