I spend a lot of my time at the Royal College of Music (RCM) in London teaching British, American and non-English-speaking students how to sing with a particular kind of English accent; this is often referred to as Received Pronunciation (RP), although really it’s a singers’ version of RP. In the UK people who speak like this would probably be labelled as Middle or Upper Class, but what is unusual about this way of speaking is that it has no geographical component: such a speaker may come from North, South, East or West. The great majority of British people have accents that give a clue about where they come from.
As I say, one of my roles at the RCM is to coach students to sing with these standardised RP sounds (specifically vowel sounds – consonant sounds need an article of their own), but it’s important to be aware that nobody actually speaks in this way. The vowels we use require a greater clarity than in speech, and what’s more, they must be sung in a way that is most conducive to best practice in vocal production.
That’s all well and good for most of the repertoire we’d perhaps call Art Song, but sometimes we’re presented with songs in dialect, and I’ll use Ye Banks and Braes as an example. Kathleen Ferrier recorded Roger Quilter’s well-known arrangement, and it was entirely in RP, with no r in birds, heart, thorn, departed etc, but the usual (for her day) massive rolled r on fair and care, and all diphthongs as in standard (again for her day) singers’ English. Here’s the first verse in the version Quilter used:
Ye banks and braes o’ bonnie Doon How can ye bloom sae fresh and fair? How can ye chaunt, ye little birds, And I sae weary, fu’ o’ care?
The sound she makes is gorgeous, of course, but in spite of the obvious Scottishness of Burns’s words, she makes no attempt to take it north of the border. I’ve listened to several other recordings of Quilter’s setting, as well as other versions (Kenneth McKellar sang it beautifully in an arrangement by Robert Sharples) including trad folk-singing style. They all differ in the degree to which they accentuate the Scottish sounds partly, I suppose, because of their own individual accents. But the important fact is that they mostly make the attempt, which I believe is essential.
Here’s a much less well-known song, a setting of the Irish poet Winifred Letts by the Irish composer Stanford – The Bold Unbiddable Child:
Now what is he after below in the street? God save us; he’s terrible wild!) Is it stirrin’ the gutter around with his feet? He’d best be aware when the two of us meet. Come in out o’ that, come in out o’ that, come in, You bold unbiddable child!
Try that in RP or any other non-Irish accent and it doesn’t really work, does it? This and the other verses simply have too many Irish words and idioms. We need to hear the brogue of an ordinary mother who’s at the end of her tether with this ungovernable son of hers; it would entirely miss the point with the polished tones of a smart English person.
The Stanford and Walton song texts above show the Scots and Irish dialects quite clearly, both in their spelling and grammar; a British singer is very likely to be aware of how these texts sound from a native speaker, but they may not be aware of how other dialects sound. We’ll look next at Percy Grainger’s Bold William Taylor. Grainger started collecting folksongs in 1906, and he was the first to use a wax cylinder recorder. With this he could get a complete performance of a song without having to stop the singer at each line in order to write it down. As well as writing the notes of the melody he annotated the words to try and recreate the dialect sounds he had heard:
I’ll sing you ã song abṑut two lōvers, O from Lich-feed-deld tṑwn thã cãme; O the yȯung mȧn’s nãme was Willyum Tãylor, the mã-a-den’s nãme was Sally Grãy.
[The diacritical mark above ‘about’ isn’t quite as Grainger printed it, but this is the nearest thing on my font]. He gives instructions for these special vowel sounds: ã as in the first vowel of German Mädchen; ṑ as a sounds halfway between English note and pull; ō as English note; ȯ as English pull; ȧ as German hat
In 1952 Grainger wrote: ‘To sing a Lincolnshire folksong such as Bold William Taylor without the folksinger’s dialect and without the nonsense syllables and other details of English folksong traditions is … inartistic. The greatest crime against folksong is to ‘middle-class’ it – to sing it with a ‘white collar’ voice production and other townified suggestions.’ That’s as clear a statement as one could get.
This next song, Wapping Old Stairs is written in standard English (except for ‘bacco) both in terms of spelling and grammar. It’s an anonymous text set by Walton:
Your Molly has never been false, she declares,
Since last time we parted at Wapping Old Stairs,
When I swore that I still would continue the same, And gave you the ‘bacco box, marked with your name.
Tom is a sailor and Molly is his girl. Tom threatens to leave her for Susan from Deptford or for some other girl, but Molly pleads with him to stay, saying ‘Still your trousers I’ll wash, and your grog too I’ll make.’
Most performances of this song which I’ve heard have been sung with ‘a white collar voice production and other townified suggestions’. Considering Molly’s position in life that seems all wrong to me, and the best performances of this I’ve heard have her as a poor Londoner with a vocal style much nearer to Music Theatre.
A few years ago something happened which set me thinking more about all this. I was teaching eight amateurs at home over a weekend, and one was a very good baritone from Dundee whom I’ll call Jack. One of the songs Jack had chosen to work on was Whither must I Wander from Vaughan Williams’ Songs of Travel. He sang it through very nicely in the approved RP manner, and I might have carried on for our 30 minutes giving him a normal coaching session. But I thought: here’s a man with a natural and quite strong Scots accent who is singing words by the Scottish poet Robert Louis Stevenson. Add to this what I believe to be the background to the poem, which is all to do with the Highland Clearances, when thousands of crofters were thrown off the land by the Lairds. What could be more appropriate, I thought, than having our Scottish baritone sing with his native accent, even though there aren’t the specific dialect words and phrases such as we heard in Ye Banks and Braes, or The Bold Unbiddable Child, or The Sprig of Thyme? So I asked Jack if he’d sing it with his normal spoken accent. He stood there for quite some time before saying rather shamefacedly that he couldn’t do it. He’d been so schooled in dropping his Scots accent when he sang, that he couldn’t turn the clock back.
I believe Stevenson would have spoken with a cultured Edinburgh accent. However, I’m not suggesting that any poem of his set to music should be sung with his accent unless that is also the singer’s accent. Nor would I suggest, for example, that settings of Thomas Hardy should be sung with a broad Dorset accent by someone who is not actually from Dorset. But what if the singer does speak with an Edinburgh accent, a Dorset accent or any of the other accents from North, South, East or West? Why should they sing in RP rather than with their own familiar sounds?
On TV and Radio we hear a wide range of speech sounds, and that’s very different from the cut-glass accents we heard from the BBC in my childhood. In those days ‘ordinary’ housewives on TV adverts discussing the relative merits of different washing powders spoke as if they’d just been dining with the Queen. Imagine the cast of EastEnders calling each other names in the Queen Vic sounding like that. The late Timothy West was in EastEnders playing Stan Carter, and he wouldn’t have lasted long in the role if he’d given his best King Lear impression, nor would Prunella Scales have been such a convincing Sybil Fawlty if she’d sounded like her portrayal of Queen Victoria. Actors are expected to speak with a range of accents, and these days are often encouraged to use their own natural way of speaking.
However we sing in our own or any other language we modify the spoken sounds to make them grateful for the voice and enable the best kind of resonance throughout our range. Singing with our usual English song pronunciation isn’t the same as the RP from which it springs. Several students I teach have had a go at singing with their own regional accents, and we’ve worked on suitable modifications to accommodate their voice production; the results have been very interesting. Will this ever become part of mainstream performance? Not yet, perhaps, but maybe one day…