A nice half-hour episode of the BBC’s Word of Mouth show:
To take us back to Shakespeare’s own time Michael Rosen and Dr Laura Wright hear Shakespeare as he himself would have spoken. The original, unvarnished version from linguist David Crystal and actor Ben Crystal. They look at the fashion for Original Pronunciation and ask what it can tell us about how we speak now.
Michael and Laura perform some of Shakespeare’s best known work in the original accent and attempt to bring new meaning and wit to language coated by centuries of veneer.
David Crystal explains how, for instance, we know that love and prove had the former sound, not the latter (because Ben Jonson said so); puns are explained (reason/raisin, hour/whore, lines/loins); and several passages are read in original pronunciation. Well worth your time. Thanks, PK!
Interesting suggestion that the r-lessness of RP is comparatively recent; though the fact that John Lewis spoke an intermittently rhotic RP scarcely seems hard evidence given his West Country background. I too speak a sort of rhotic RP, with the substrate in my case being Scots. Very few people notice my r’s (or “date”, as they say in Australia.)
(I am younger than John Spedan Lewis.)
To me, a much more noticeable and significant feature of John Lewis’s Rs was that he trilled at least some of them. That’s what died out with that generation of RP speakers – my grandmother (1903-64) trilled them too – and it seems odd that they did it only intermittently. Conan Doyle’s accent was I believe upper-class Edinburgh, and so he trilled all the time. I think that trilling may also be on the decline nowadays, perhaps the accent too.
Michael Rosen’s BBC show was very honest about the difficulties. Some of the readings sounded like The Archers.
We don’t know that “love” and “prove” had the former sound, and there are many reasons to think that they had the latter sound, and both verbs were pronounced with rounded vowels in Shakespeare’s time. The issue is uncertain.
Briefly, the following is relatively settled:
1. “love” had the short rounded [u] sound in the 16th century, together with words like “mother”, “brother”, and the myriad words spelled with u that had the short sound: sun, much, cut etc. All these had about the same sound in the 16th century as “put” retains today: [u].
2. During the 17th century, this short [u] sound in closed syllables migrated to [ʌ], the sound of today’s sun, much, cut, mother, love, etc. The migration was possibly through an intermediate variant [ɤ], which is unrounded [u] – a sound much further back than [ʌ]. By the end of the 17h century, the unrounded pronunciation was the norm.
3. Words like move, prove, remove escaped this process because they had the long vowel [u:] (and retain it today), which was not affected by it. Thus prove was certainly pronounced [pru:v] in the 16th century, and at the end of the 17th century, and today. It’s possible that there was a variant pronunciation [pruv] with the short vowel, which then would’ve migrated into [prɤv]/[prʌv] – which then would have had to die out, because we have no evidence of [prʌv] later on.
It’s not clear when the process of unrounding [u] in the 17th century got started, and it’s not clear that it was ongoing during Shakespeare’s life (it certainly wasn’t complete, if it was ongoing). There’re a few bits of conflicting evidence that’s vague and has been interpreted differently. One expert opinion is that it started after Shakespeare; thus Otto Jespersen quotes a 1633 grammar which has the same sound for “sun, soon, too, much, etc.”, and a 1653 source which possibly indicates an unrounded pronunciation, and concludes that around mid-century is the likely threshold. Charles Barber thinks it possible that there was an unrounded variant already in the early 17th century; but he also says that there was a “poetic” variant [lu:v], maintained for some time precisely for the benefit of convenient rhymes.
David Crystal has been claiming that it was [lɤv]/[prɤv] for some years (and Ben Crystal has been pronouncing for some years) in various TV/radio programs, in a booklet he wrote about the OP, etc. One source that he’s quoted consistently in support of that view is Ben Johnson’s grammar, which, however, is ambiguous evidence, and I believe it doesn’t established an unrounded vowel at all. What Johnson says is that “cosen, dosen, brother, mother, love, prove” all have the same vowel. He also says that the letter O is pronounced “with a round mouth, the tongue drawn back to the root”; he says that before distinguishing the short and the long versions of O. I take it as (weak) evidence that in fact all these words had [u] for Johnson, rather than the unrounded [ɤ]. The evidence is weak because Johnson’s phonetic observations are not especially reliable, and he may have had the “long O” in mind when he was talking about the rounding. But either way, I don’t think it works well as evidence for [lɤv]/[prɤv], and certainly not conclusive evidence of any kind.
I had an email exchange with Prof. Crystal on this issue about a year ago, where he:
– acknowledged that [lɤv]/[prɤv] is not a settled issue, and there’re authorities which think (as I do), that the vowel was rounded throughout Shakespeare’s time.
– reminded me correctly that other experts think that [lɤv]/[prɤv] was possible, and said that he felt justified in the use of this pronunciation for guiding contemporary actors in using the OP. In fact, since it’s difficult to get [ɤ] right, he feels it’s OK to use [lʌv]/[prʌv] as well
– told me that he acknowledges the rounded variant in his forthcoming The Oxford Dictionary of the OP (I just checked and it’s due to be published later this year)
I should’ve also said in my long comment above that I’m not an expert on the OP, nor a trained linguist. David Crystal is both. He’s much more knowledgeable in everything that has to do with Shakespeare and pronunciation than I could ever hope to be. My knowledge of these issues is very superficial.
I do think that it’s a mistake to claim that “we know” that “love”-rhymes were unrounded in Shakespeare’s time. My personal opinion is that they were rounded (& I can’t seem to decide on whether short or long is more likely).
An earlier discussion on the Crystals and love / prove.
We don’t know that “love” and “prove” had the former sound, and there are many reasons to think that they had the latter sound, and both verbs were pronounced with rounded vowels in Shakespeare’s time. The issue is uncertain.
Thanks very much for that detailed discussion — it was worth posting the link to get it!
An earlier discussion on the Crystals and love / prove.
Sigh. At least this was a different link!
(When I read “the Crystals,” this is what I think of.)
I thought it was still mocked by the upper class in the 1840s?
Between vowels.
(Including across word boundaries in unstressed cases like for it.)
Let’s be a bit pickier and call it [ʊ], because:
Unrounded [u] is [ɯ]; [ɤ] is unrounded [o]. (Mandarin and Vietnamese, for example, have both.) Unrounded [ʊ], which is what the Turkish ı is, doesn’t have an IPA symbol. Anyway, though, a drift from there over [ɤ] to [ʌ] must have happened; [ɤ] > [ʌ] has also happened in Korean. [ʌ], incidentally, is unrounded [ɔ]; and although it’s not as far back as [u] because the back margin of the vowel diagram isn’t vertical, it is at that margin.
The STRUT vowel of RP doesn’t actually seem to be [ʌ] for most people nowadays. As John Wells found quite some time ago, it’s at most [ɐ]… The vowel I was taught to use seems to be a back vowel between [ɑ] and [ʌ], so around the same height as [ɐ] and [æ].
I wonder if [ɤ] is responsible for the French rendering as [œ] in such 19th-century loanwords as club.
While clicking on “Post Comment” I remembered the lurve vowel. That must be the half-mythical [ɞ], the rounded counterpart to [ɜ]…
David: wonder if [ɤ] is responsible for the French rendering as [œ] in such 19th-century loanwords as club.
No doubt.
The STRUT vowel of RP doesn’t actually seem to be [ʌ] for most people nowadays. As John Wells found quite some time ago, it’s at most [ɐ]… The vowel I was taught to use seems to be a back vowel between [ɑ] and [ʌ], so around the same height as [ɐ] and [æ].
There’s a tendency to retract it, though. There’s an anticlockwise vowel shift in progress (involving mostly short vowels) in mainstream southern British English (the opposite of what’s happening to Australian English), with /æ/ being realised as [a], and /ʌ/ mowing towards [ʌ] (reversing its earlier fronting tendency). RP /e/ is now lower than it used to be (practically = cardinal [ɛ]), while Gimson’s /ɒ/ (once very open and barely rounded) is close to [ɔ]. Long /ɔː/, in turn has moved up and acquired more lip-rouding, becoming [oː], /ʊ/ is strongly centralised, [ɵ] (listen to how people say [gɵːːːd], and /uː/ is at least central, [ʉː] ~ [ɨʉ] if not frontish.
There’s an anticlockwise vowel shift in progress … in mainstream southern British English …. ), with /æ/ being realised as [a],
Listening to Canadian radio a few years ago, a heard a woman who appeared to be British but had this pronunciation, so cat had [a] (perhaps even farther back) not [æ]. Her initial velar also sounded kind of back.
I had noticed most of the individual changes, but not the big picture, thanks!
I love the reference to OP being the speech of an Irish pirate washed up on the shores of Somerset, where “Irish” captures the OP vowels and “pirate” and “Somerset” capture the consonants (postvocalic /r/ and initial /hw/).
anticlockwise vowel shift
I wonder how the convention arose of having the front vowels and consonants on the left in the IPA chart, making it a view of the left side of the speaker’s face. It could equally well have been the other way, in which case this would be a clockwise vowel shift. Funny how these things get fixed in a culture early. In Indic script, the velars come first, so that a table written left-to-right would be a view of the right side of the face.
Does any variety of English actually have [u:]? The GOOSE vowel is at least somewhat fronted and/or diphthongized in every dialect I’m familiar with.
TR: Supposedly parts of New England, the North Central (relic GA) region of the US, and the west of Ireland. What’s unusual about GOOSE-fronting is how universal a shift it is, happening fairly independently in almost all parts of the Anglosphere. Even I, conservative old fart as I am phonologically, have some degree of fronting.
Welsh English has pretty much cardinal [u:].
Welsh English has pretty much cardinal [u:]
That’s perhaps because it preserves the contrast between /uː/ and /ɪʊ/.
Compare blew, flew (5:20-5:25) and noon, Moon (5:43-5:50) as realised by this Welshman.