Thread by @E_JMarshall

Do choirs have accents?
Now I am 3 years into the PhD I have some thoughts I would like to share with you…
Here’s a wee thread🧵#AcademicChatter #AcademicTwitter #SingingResearch #SpheresOfSinging
1\

Why?
Over 2 million people sing in choirs every week in the UK.
Everyone has an accent - and accents can differ by many different social factors (incl. region, age, gender, education and more).
\2

Choir directors anecdotally report differences in choir sound by region. And, in musicology, Classical singing technique™ and choral singing has been associated with Received Pronunciation AKA “the Queen’s English”.
\3

Do choirs have accents?
- Is there evidence of change over time in historical choir recordings?
- Is there evidence of regional differences in choir recordings?
- And, possibly, is there evidence of a standard emerging in historical choir recordings?
\4

In this thread, I will be talking about evidence for change over time in front vowel height from historical recordings of choral singing.
\5

In order to answer the research questions I have been working with historical choir recordings from the Glasgow Orpheus and Glasgow Phoenix choirs (recordings 1925-2017) and the choir of King’s College, Cambridge (recordings 1945-2019).
\6

I created corpora containing these recordings and their texts, storing them in @LaBBCAT. The soundfiles and texts were forced aligned using the #HTK aligner trained on speech data (and hand corrected a lot!).
\7

So far, I have investigated monophthong vowels - “pure” vowels where there is usually limited tongue movement. Formants were automatically extracted in LaBB-CAT using #Praat and modelled with Bayesian linear mixed models using #brms in #R.
\8

Glasgow front vowel height has lowered over time. Vowel by time/director interaction plot for Glasgow F1: \9

King’s front vowel height raised by David Willcocks (1959-1974) and subsequently lowered under Philip Ledger (1976-1982) and Stephen Cleobury (1984-2019). Perhaps, front vowel lowering was complete already under Boris Ord in the 1950s. Vowel by time interaction for King’s F1: \10

Day (2018) writes that: ‘Inevitably, Willcocks cultivated certain sounds which reflected his own style of spoken English, perhaps more the received pronunciation of English he heard as a chorister at Westminster Abbey in the 1930s than that of the 1960s…’
\11

‘...So alleluia became e-lleluia. “I know thett my Redeemer liveth, ent thett he shell stent…”.’ (Day, 2018: 261) In phonetic terms, [ˌaleˈlujə] became [ˌɛleˈlujə] as in conservative-RP.
\12

TL;DR Choir recordings from Glasgow and King’s show KIT, DRESS and TRAP lowering over time potentially following a pattern of front vowel lowering of Received Pronunciation over the 20th Century.
\13

Further modelling and post hoc comparisons show differences between RP vowel pairs TRAP/BATH, LOT/THOUGHT and FOOT/GOOSE in both King's and Glasgow data
\14

Choral singing in English is based on the phonology of Standard Southern British English. Choral singers in Glasgow also produce these vowel contrasts in singing - contrasts which for most Scots will not be present in their speech!
\15

What does this mean?
For choral singing, in English at least, singing ability is likely conflated with the ability to sing in a standard accent which may be very different to singers' spoken accents e.g. here in Scotland...
\16

I have also looked at the F2 and duration for front vowels; F1, F2 and duration for back vowels; and a combined analysis of the King’s and Glasgow datasets which I’m writing up atm. I hope to share this work in future threads.
\17

I’m currently planning consonant variables to investigate over the next 6 months so if you’re more interested in consonants, watch this space!
\18

If you would like to hear more about the work in this thread, Linguists I am excited to have been accepted to present a poster at #LabPhon18 in June. Singers/choir directors I have submitted an abstract to present at @SingingSpheres #SpheresOfSinging2022
\end