Australian English speakers’ attitudes to fricated coda /t/

Abstract

The fricated allophone of coda /t/ is a variant in which full occlusion of the alveolar stop is not achieved, resulting in the consonant instead being produced via frication. Fricated /t/ is attested in several varieties of English from the British Isles and Southern Hemisphere. While awareness of the variant can be found in Australian popular culture, it has been the focus of few sociophonetic studies. Here we report on an experiment which investigated the social meanings that Australian English (AusE) speakers ascribe to fricated /t/. We used an online matched guise paradigm in which listeners were presented with short utterances from six speakers that had been acoustically manipulated to differ only in the variant of phrase-final /t/. Using a series of sliding scales, 100 listeners recorded their impressions of the speakers, both in terms of the speakers’ social identity and favourability. We hypothesized that AusE listeners would associate fricated /t/ with the descriptors ‘urban’ and ‘educated’, and that, for the three male speakers, fricated and released /t/ would be associated with the description ‘gay’. Partially consistent with the first hypothesis, results revealed that tokens were rated significantly more educated in the fricated guise than the unreleased guise, but this effect was driven by one male speaker who was also rated ‘straight’ and ‘rural’. Guise did not significantly predict ratings of ruralness, nor were male speakers rated significantly more gay in any specific guise. Additionally, straight males rated the accent of the gayest rated male speaker least like their own, unlike gay males or females and others. It is posited that the articulation associated with fricated /t/ situates it within an indexical field pointing to education, but that the effect of this is modulated by the presence of other indicators or markers.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 4 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.