The last post inspired a brief debate in the comments section about the LOT vowel (i.e. the ‘o’ in ‘not,’ ‘Todd,’ and ‘rot’) in contemporary British English. The question, it seems, is whether this sound has shifted closer to the vowel space more traditional to the vowel in THOUGHT.
Like most questions of this type, it really depends on the dialect. So I’ll turn to a study I occasionally use when comparing individual vowels in different regions: Ferragne and Pellegrino’s Formant Frequencies of Vowels in 13 Accents of the British Isles, which compares acoustic analyses of various accents of the contemporary UK.
I won’t regale you with a lecture about acoustic analysis, other than to say that phoneticians often use what are called ‘formant frequencies’ to make deductions about the relative position of vowels. Long story short, if a vowel has a higher ‘F1‘ (first formant frequency), it roughly suggests a more open vowel, while a lower F1 suggests a more close vowel.
With this rather crude explanation in mind, let’s take a look at the F1 numbers for the LOT vowel in different British accents according to Ferragne and Pellegrino’s study:
Scottish Highlands 439
Glasgow 530
London 552
Cornwall 556
Birmingham 576
Hull 578
East Anglia 580
Newcastle 591
Liverpool 599
Lancaster 615
This list is not terribly surprising. The closest pronunciations of LOT are found in Scotland, which makes sense: many have noted that some Scottish accents feature a LOT vowel fairly close to the pure [o] of Spanish todo. Also unsurprising is that the three most open pronunciations of LOT are in the North of England, where I’ve noticed that this vowel doesn’t seem to have shifted upward much.
The ‘London’ accent (which the researchers label ‘Standard Southern English‘), definitely has a LOT vowel that appears closer to [ɔ] (the vowel more traditionally associated with RP THOUGHT) than [ɒ]. That being said, it’s a little unclear to me who represents the Londoners in this study. ‘Standard Southern English’ seems to be equated with contemporary Received Pronunciation but there are some vagaries in terms of what accents fall under that category.
I’m hesitant to make any conclusions, but roughly speaking, these numbers suggests that the LOT vowel is most open in Northern England, most close in Scotland, and somewhere in between in the South. Although I’m sure there is tremendous variation in the pronunciation of individual speakers.
*Source: Ferragne, E., & Pellegrino, F. (2010). Formant frequencies of 13 accents of the British Isles. Journal of the International Phonetic Association, 40, 1-34.
I missed the debate. Damn, I always leave too early. But anyway, a few things surprise me on the list here. First of all, I’m a bit surprised to see that the Highlands of Scotland have a closer LOT vowel than Glasgow. But it depends on social class too, because I believe working class Glaswegians tend to have a closer LOT than middle class Glaswegians. I was surprised to see East Anglia where it was on the list too, but then I remembered that some East Anglians have an unrounded LOT vowel. According to the book I linked to on your post on Kiwi English a few days ago, lip rounding lowers F1 (and F2 and F3). So it makes since that a place where there can be an unrounded LOT vowel would have a higher F1 for that vowel than a place where LOT is always rounded.
Yeah, lip rounding is a bit of a wild card. Apropos of that, I didn’t include the accent with the highest F1 here: that would be Dublin. I excluded it because of the relatively high frequency of unrounded LOT in that city.
You would also have to retitle this post if you included Dublin.
And I meant “makes sense”. Damn pin-pen merger. That’s also a wonderful example of why spellcheck isn’t that great.
If you always “leave too early”, you could try subscribing to the RSS comments feed using a service like Google Reader. No more need to go back and reload old posts!
If you always “leave too early”, you could try subscribing to the RSS comments feed using a service like Google Reader. No more need to go back and reload old posts!
(A version of this comment with links is being held for moderation — trawicks, please could you turn down the moderation settings? Thanks!)
Just turned them down! (Seems to have reset to the ‘2 links=spam’ setting at some point.)
I don’t know much about British accents, but it does seem there’s been a shift in Received Pronunciation speakers, with /ɒ/ becoming [ɔ̞] and /ɔ/ moving closer to [o] or even diphthongized to [ɒo]/[oʊ]/[ɔʊ].
This link posted in the comments to the previous post is worth reposting here: http://englishspeechservices.com/blog/?p=1795
It’s a blog post about the vowels in what it calls Standard British English. And it has LOT as [ɔ].
Are you referring to the link I mentioned? I don’t really see how that blog post relates to it. It actually didn’t say anything about the LOT vowel. It was about the NURSE vowel in the Southern Hemisphere. But I was applying what it said about lip-rounding and formants to the LOT vowel in the UK. You can go read it if Google Books still lets you. I don’t know how that works.
(continued) Sorry if I misunderstood you. It’s possible that I didn’t read the passage I linked well enough. Maybe you were referring to the part where Trudgill said, “…the RP-type vowel [ɜː] is actually a rather uncommon vowel in varieties of English…” That might be more relevant to that topic.
I don’t remember who posted the link, if it was you or someone else. I’m referring to the blog post I posted a link to in my comment. Follow the link. It talks about the whole British vowel system, including LOT. It appears to be written by Geoff Lindsey. And it does not have the passage you quote.
I was thinking you were talking about me. I think I’m becoming vain.
Nearly all Scottish English has a LOT-THOUGHT merger, so the closeness of Scottish LOT is not particularly surprising.
I’ve noticed this among Edinburgh natives especially. There seems to be a small shift there in which LOT/THOUGHT moves toward GOAT, and GOAT moves toward cardinal /u/.
I think it’s fair to say that the British LOT vowel has oscillated around the [ɔ~ɒ] area over the centuries. Descriptions of Middle English, for example, often give [ɔ] for the “short O” vowel (i.e. the vowel of LOT).
This is made possible by the fact that English (at least those varieties of English that distinguish between long and short vowels) does not distinguish more than 3 tongue heights for short vowels.
A very similar thing has happened with the the British TRAP vowel, which has oscillated around the [æ~a] area for the same reasons.
The opposite is also true, many times I heard RP/ɒ/ as [ɑ], especially in words like LONG, WRONG, SONG (this is the norm in Dublin English).
As you can see:
TOP
[tʰɒp] is Shakespearean, still used in S.Wales, Devon, and many parts of Northern England, as well as in Canada. RP of today has [tʰɔ ̞p]. [tʰɑp] is Irish, and Standard American. [tʰa ̠p ̚] is used in the NCVS region and in traditional Newfoundland English.
DAUGHTER
[ˈdɑːtəɻ] is Shakespearean, [ˈdɑ ̟ˑɾɚ] is Standard C/C merged American English; Standard Canadian and conservative Chicagoan have [ˈdɔ ̞ˑɾɹ ̩], RP and much of England (as well as Australia and NZ) have [ˈdɔˑtə] (+ t flapping for AU/NZ); Longtown in the North of England has [ˈdɒːtʰɜ]; Buckie in Scotland has [ˈdo ̞θɐ ̠ɾ ].
SALT
Shakespearean English has [saɫt], standard Canadian has [sɒɫt], standard C/C merged American English has [sɑ ̟ˑɫtʰ], Boston has [sɒ̜ːɫt], New York has [sɔːɫt ̟ ̚], Alabama has [sɑːɫt]. (So, you can have an unrounded A in SALT in c/c unmerged accents [Alabama], as well as a rounded A in c/c merged accents (Canada, Boston)). The unrouded SALT vowel can still be heard in a number of English, and Scottish accents, as well in Ireland. RP has [sɔˑɫtʰ].
ALL
Shakespearean English has [ɑːɫ] which is used in NCVS English. Also, in standard US c/c merged English (at least phonemically: ALL OF = OLIVE).
The Canadian English outside Newfoundland has (an obligate?) rounding: [ɒːɫ]. The Alabama and the North Carolina accents have no c/c merger but both have [ɑːɫ]. [ɑːl] is found in Ireland (the unroundness sounds clearer since the L is light there). In Australia, NZ and RP it’s [ɔːɫ]. In the popular London accent it’s: [oʊ]
LONG
West Saxon had [lɑŋɡ], Shakespearean English had [lɒŋ], Middle Scots had
[laŋ]. Standard Canadian has [ɫɒːŋ]; Standard c/c merged American English has [ɫɑ ̟ːŋ]. [ɫɑ ̟ːŋ] is also found in central Ohio (partially c/c merged), northern Ohio and Illinois (c/c unmerged but with NCVS), and in Alabama (no c/c merger, no NCVS). [lɒ ̝ŋ] in Australia, [lɔˑŋ] in NZ, [loːŋ] in the popular London accent, [lɔ ̞ŋ] in RP. Many English and Scottish accents have [ lɒŋ], some traditional dialects/accents even keep the older not-rounded ɑ (or even ä). [ɑ] is also frequent in Irish English, especially in Dublin.
source: Index of /research/gsound/Eng/Database/Phonetics/Englishes/ByWord
University of Edinburgh
Correct.
I also think Canadians might also use [ɑ], [ɑʷ] or (less commonly) [ɔ̞], and [ɔɫ]/[ɔ̞ɫ] is not that uncommon either.
Also, [ɑ̈], [ɑ̟] and even [ä] might be perceived as “Standard” American.
P.S. Are you using:
http://soundcomparisons.com/
as your reference by any chance?
Ugh, that is the website you’re using! Didn’t catch the source. I find it to be generally good, though their transcriptions aren’t always accurate.
Thank you for this. I am a big fan of unifying EFL around the Gimson system of RP since it’s the closest we have to a “Standard English” but I am thoroughly pissed off by chauvinistic Brits who say their Mockney, Estuary, Yorkshire accents are more “real” than my native accent.
Every dialect and accent preserves certain features, drops others, invents new ones. That’s what makes them what they are. Rather than England being Shakespearean it’s a grandson just as America is a grandson. We are cousins.
Overall, though, it seems that the General American accent is the closest you will get to Shakespeare’s English. Especially since it has a long vowel version of “daughter” that is lower than the NORTH vowel (why dictionaries write them the same I’ll never understand, even Kenyon&Knott did this). The FACE diphthong starts closer (!) than DRESS. GOAT is fully rounded. American is rhotic. Some preserve the NORTH-FORCE and TRHEW-THROUGH distinctions. Scottish’s LOT is way too tight and the put-boot merger further disqualifies it. Irish’s rhythms are definitely off. The cadence of Americans is based on older Bristol so it’s pretty close. Bristolian English is a close second for the Bard, but not if they have to say area-l or mania-l or have to talk to Amanda-l.
I find RP (true RP, not the varieties of near-RP) to be great as a teaching tool, but God Almighty is it irritating to see movies and hear RP spoken by Shakespeare characters, or medieval people, or colonists for that matter. It’s like having ancient Egyptians speak in Arabic. Hell, most of Shakespeare is actually taking place in Italy or some other non-English-speaking area anyway. David Tennant has quite possibly the most annoying RP in his theatrical work. His STRUT and GOOSE and his Esses (S) are so overstressed and pretentious.
Overall, though, it seems that the General American accent is the closest you will get to Shakespeare’s English.
I think most people would award that accolade to a conservative Irish English accent. General American has several distinctive innovations, such as:
* the pre-R mergers (nearer-mirror, Mary-marry-merry, etc.)
* the father-bother merger
* voicing of intervocalic T
In my opinion, conservative General American and conservative RP are roughly equidistant from Shakespearean English, but conservative Irish English is far closer — it has even resisted (to some extent) the FIR-FUR-FERN merger, which is found nearly everywhere else.
There are also some northern English and Scottish accents that are very conservative. But they are too conservative in some ways, e.g., no STRUT-FOOT split (northern England), “night” as [niːt] and “mouth” as [muːθ]. The latter two are actually the pre-GVS pronunciations.
Wow, I really didn’t put much thought into my last comment. If those features exist today then they would have been even more widespread in Shakespeare’s time and the [uː] and [iː] probably would have occurred in a larger number of words than they do today. So they aren’t “too conservative” for Early Modern English. It just depends on what accent of Early Modern English you’re looking at.
Nice post. I used to be checking constantly this weblog and I am inspired!
Extremely helpful info particularly the final part 🙂
I take care of such information much. I was seeking this particular info for a long time.
Thank you and best of luck.
Review my site; couples counseling sacramento