About
Janet Fletcher
The University of Melbourne

Main areas of research
Phonetics theory, prosodic phonology, laboratory phonology, and articulatory and acoustic modelling of prosodic effects in various languages, particularly indigenous Australian languages and French.
Janet Fletcher is Professor of Phonetics in the School of Languages and Linguistics. She has held previous appointments at the University of Edinburgh, the Ohio State University, and Macquarie University. Her research interests include phonetic theory, laboratory phonology, prosodic phonology, articulatory and acoustic modelling of prosodic effects in various languages. She is currently working on phonetic variation in French Polynesian languages, and has worked extensively on coarticulation, prosody, and intonation in Indigenous Australian languages. She is a member of the Research Unit for Indigenous Language in the School of Languages and Linguistics and an elected Fellow to the Australian Academy of Humanities.
Exploring the phonetics and phoology of Tahitian
Outline
In the last ten years or more, we have seen expanded interest in quantitative phonetic analyses and documentation of under-resourced languages of the Pacific. In this talk I will present recent work on the Eastern Polynesian language, Tahitian. The Eastern Polynesian languages, including Hawaiian, Tahitian, and Marquesan (North and South) are often described as having extremely simple phonological properties. They have relatively few consonant contrasts and have simple syllable phonotactics compared to many well-resourced languages like English. Tahitian is also claimed to have a predictable lexical stress system, however this has never been fully investigated experimentally. Nor has the consonant or vowel system or intonational phonology of the language been analysed fully. I will argue that the phonetic exploration of so-called simple languages can contribute to various aspects of phonetic and phonological typology. Results of recent experimental investigations of Tahitian accentual prominence and stop consonant variation will be presented. I will also focus on glottal stop variation in light of recent research on a close-related neighbouring language, Hawaiian, and cross-linguistic surveys of glottal articulations (e.g. Davidson 2021, Davidson and Parker-Jones 2023, Garellek et al 2021, Keating et al. 2023). It is well accepted that in many languages, phonemic /ʔ/ has multiple variants. Glottalization can also signal higher level prosodic boundaries for example and Tahitian presents no exception in this regard.
Our studies to date have shown that previous impressionistic descriptions of accentual prominence are more or less confirmed with Tahitian behaving like a classic “head prominence” language with late-rising or early-rising post-lexical pitch prominences docking on to phonologically stressed syllables, and there are key tonal alignment differences depending on whether the vowel is phonologically long or short. Accentually prominent syllables are also longer and louder than unstressed syllables although the effect is clearest in long vowels.
Close inspection of the consonant system also shows that similar to many other languages of the world with contrastive glottal stops, Tahitian has glottal variants ranging from a fully or partially-occluded “canonical” glottal stop to creaky voiced, breathy, and modal voiced allophones. However, compared to the closely-related Eastern Polynesian language Hawaiian, the proportion of canonical glottal stops is somewhat higher at 68% compared with 7% (Fletcher and Gregory 2024, Davidson 2021). In general, glottal stops in Tahitian tend to be more “consonantal” rather than “vocalic”. Moreover, as one would expect there is also stress-conditioned vowel glottalization in the vicinity of phonemic glottal stops. The data clearly show evidence of glottal coarticulation in either direction, with many standard acoustic parameters of voice quality including spectral tilt and harmonics-to-noise ratio (HNR) signaling more vowel glottalization in unstressed versus stressed syllables suggesting localized hyperarticulation due to prosodic prominence. Languages with so-called “simple” phonetics and phonology like Tahitian show similar patterns of phonetic variation to those with far more complex phonetic and phonological structures and are therefore worthy of experimental investigation.