Do choirs have accents? A sociophonetic investigation of choral sound

Marshall, Edward Joseph (2024) Do choirs have accents? A sociophonetic investigation of choral sound. PhD thesis, University of Glasgow.

Full text available as:
[thumbnail of 2023MarshallPhD.pdf] PDF
Download (29MB)


Many UK cities, including Glasgow, have a long history of choral singing, with recordings dating back to the 1920s (for example, the Glasgow Orpheus Choir). Choirs have a ‘sound’ which is both musical and linguistic. Speakers from different localities can be said to have an accent. Is there such as thing as a regional choral ‘accent’? Neither phoneticians, singers, nor choir directors have a clear understanding of how such choral sound–accents are achieved, how they arise, and are maintained. The main research questions for this thesis are:

1. Is there evidence of regional differences between Glasgow and Cambridge, in the phonology of vowels, rhoticity, and word-final /d/?
2. Is there evidence of a common choral accent uniting Glasgow and Cambridge in the phonology of vowels, rhoticity, and word-final /d/?
3. What changes have taken place in the phonology of choral singing over time?
(a) Are the changes linked with changes in spoken phonology over the relevant time period?
(b) Are the changes linked with changes in aesthetic conventions of choral singing?
(c) Are the changes linked with individual properties of the choir directors?

To answer these research questions, two time-aligned electronic corpora were constructed in LaBB-CAT containing 26 hours of commercially-released recordings of British classical choral singing of choirs from two different regions, Glasgow and Cambridge. 1. Recordings of the Choir of King’s College, Cambridge (1949–2019). 2. Recordings of the Glasgow Orpheus and Phoenix choirs (1925–2016).

This thesis presents the analysis of three different variables. Analysing the front vowels I found a shared front vowel phonology and realisation. The consonant variable rhoticity (e.g. car) was selected to investigate impact of spoken dialect. Word-final /d/ (e.g. lord) was selected to investigate aesthetic–stylistic differences between the two corpora.

In a Bayesian analysis of acoustic measures F1 and F2, I found that the vowels KIT, DRESS and TRAP (over 14,000 tokens) demonstrate a pattern of lowering over time consistent with a change in a spoken prestige accent e.g. from Received Pronunciation to Southern Standard British English. The analyses also support separate TRAP and BATH phonemes in both Glasgow and Cambridge, which we would not expect based on spoken vowel phonology. These findings suggest an emerging standard ‘accent’ of choral singing that has changed over time, following the pattern presented by Received Pronunciation. However, the realisation of the GOOSE vowel differs between Glasgow and King’s, perhaps relating to the sociolinguistic salience of GOOSE in Scottish English.

Rhoticity (auditory coding of 8,407 tokens) differs between the Glasgow choirs and King’s, as we might expect, based on regional accent phonology. The Glasgow choirs produce postvocalic /r/ in all contexts, though there is a reduction over time; they also produce alveolar trill realisations in initial position 50% of the time in the Orpheus Choir early recordings directed by Hugh S. Roberton (1925–1945), perhaps indicating that the variable was enregistered as part of a distinctly Scottish choral sound.

The realisation of word-final /d/ (auditory coding of 3,213 tokens) also differs between the two corpora, with King’s producing more affricated variants and more shadow vowel (epenthetic schwa) variants in pre-pausal contexts. The /d/ findings confirm a change in style suggested by musicological literature (Day, 2000). Phonetic affrication at King’s increases over time as the choir sings more frequently with orchestral accompaniment, likely to improve audibility, and this was carried over into the choir’s unaccompanied singing (Day, 2018).

This thesis is the first to provide a quantitative acoustic analysis of choral sound and explore the sociolinguistics of classical choral singing in a UK context. I have found evidence that supports a non-regional standard British classical choral singing vowel phonology and regional differences based on the phonology of the spoken accent of the singers and their choir directors. Future research is needed to explore the perceptual salience of the findings reported in this thesis and whether recordings of other regional choirs support the pattern of non-regional standard vowel phonology.

Item Type: Thesis (PhD)
Qualification Level: Doctoral
Additional Information: This work was supported by the AHRC [grant number 2284740].
Subjects: M Music and Books on Music > M Music
P Language and Literature > P Philology. Linguistics
Colleges/Schools: College of Arts & Humanities > School of Critical Studies > English Language and Linguistics
Funder's Name: Arts & Humanities Research Council (AHRC)
Supervisor's Name: Stuart-Smith, Professor Jane, Butt, Professor John and Dean, Professor Tim
Date of Award: 2024
Depositing User: Theses Team
Unique ID: glathesis:2024-84372
Copyright: Copyright of this thesis is held by the author.
Date Deposited: 20 Jun 2024 07:35
Last Modified: 20 Jun 2024 07:41
Thesis DOI: 10.5525/gla.thesis.84372

Actions (login required)

View Item View Item


Downloads per month over past year