Short O: Separated by an Uncommon Vowel

Letter oBrits and Americans are perplexed by each other’s short-o‘s.

The short-o, of course, it the curious little vowel sound in words like lot, rod and top. It’s is one of the English Language’s more inconsequential phonemes (you can write entire paragraphs without using a short-o word); yet, paradoxically, it is one of the more noticeable separators between British and American accents.

In Received Pronunciation (Standard British), short-o is typically a open back rounded vowel (IPA ɒ), whereas in American English (General American) it is a open back unrounded vowel (IPA ɑ). Crudely speaking, hot is “hawt” in British English, but “haht” in American English.

Well, not exactly. There are many ways of pronouncing short-o in both the UK and North America. In Canada, the Western US and Eastern New England, IPA ɒ (i.e. the British vowel) is fairly common. Likewise, the flat ɑ (or “ah”) vowel can often be heard in the West Country of England.  So there is some overlap.

That being said, I’m often amused by how befuddled Brits are by the “flat” American short-o (i.e. the unrounded vowel).  Did this come from Ireland, they ask?  Or some other foreign influence?

Actually, I’m pretty sure it came from England. Below are three dialect recordings from the classic University of Leeds Survey of English study from the 1950s.  All three feature the unrounded “American” pronunciation of Short-o (i.e. ɑ):

This one from Kent.
This one from Norfolk.
This one from Berkshire.

Notably, these recordings are all from the Eastern half of England, with the last one deriving from a small village just outside Reading.  In fact, as per the Handbook of the Varieties of English, Reading still features the “unrounded” short-o among some speakers. It certainly appears to have been more widespread in England before World War II. This is hardly an “American” vowel at all!

We Americans are a bit confused by British short-o as well.  Since most American accents don’t have any real short vs. long distinctions, we have a hard time finding an American vowel to approximate the standard British short-o.  Hence this now notorious video of an American speech coach trying to teach people the British short-o.

Anybody found a pronunciation of this vowel particularly odd? Or have some short-o tidbit to share? Or simply want to express displeasure at all the uses of the phrase “short-o” in one post?

Share

About Ben

Ben Trawick-Smith launched his dialect fascination while working in theatre. He has worked as an actor, playwright, director, critic and dialect coach. Other passions include linguistics, urban development, philosophy and film.
This entry was posted in English Phonetics and tagged . Bookmark the permalink.

25 Responses to Short O: Separated by an Uncommon Vowel

  1. 'enry 'iggins says:

    No, I don’t want to express displeasure at all the uses of the phrase “short-o”. I’m okay with that. But I do want to say that I also think that the unrounded LOT vowel probably came from England too after doing some research. I like that you think that too, because I haven’t heard anyone else say that. Most linguists seem to think it was an American innovation. Although it may have come from Ireland too I suppose, i.e., it could have come from both places. We did have a lot of Irish settlers in America. It’s interesting to me how the unrounded LOT vowel was (is?) found in areas of England that are known to be linguistically conservative in other ways (rhoticity, etc.).

    • trawicks says:

      The unrounded short-o actually used to be heard throughout a pretty large percentage of England, geographically-speaking. That being said, we do tend to view British England through London-centric eyes, which is why we don’t think of this feature, or rhoticity, as “English.”

    • dw says:

      It’s interesting to me how the unrounded LOT vowel was (is?) found in areas of England that are known to be linguistically conservative in other ways (rhoticity, etc.).

      But not only in such areas. It was also found in East Anglia, whose traditional speech is in many ways rather innovative.

  2. Nick says:

    That YouTube clip is crazy! It’s like she’s completely unaware of her own cot-caught merger, all the short Os come out as ᴐ – the continental value! Is she attempting a Northern accent or is she just all mixed up?

    I know you said “crudely speaking”, but you make it sound like she’s right when you say that “hot” is pronounced “hawt”. Unfortunately, there’s no other way to describe it with Latin letters – that’s why we have IPA!

    • Nick says:

      Actually, after listening to her advice on long-Os I think she’s been listening to recordings of the Queen (or someone else with a very old RP) and took just one side of her lot-cloth split.

    • trawicks says:

      I didn’t excoriate her more than she already has been (JC Wells took her to task on his blog a while back). But yeah, pretty much everything is wrong with her tutorial: the vowel is prononced too high, is often too long, and most problematically, she actually throws “caught” words into the mix!

      I won’t quite defend her, but I will say that she is actually a public speaking coach. As I’ve said before, a lot of people assume, bizarrely, that a pedigree in voice/speech/speaking coaching immediately qualifies you to be a dialect coach. What’s upsetting is that the woman above may very well be an excellent public speaking coach, yet posting this kind of thing can only hurt her reputation.

  3. Dw says:

    Since LOT remains distinct from TRAP in AmE, LOT-unrounding must have been preceded by the fronting of TRAP from [a] to [&].

    And because CLOTH and NORTH remain rounded in AmE, it is likely that LOT-unrounding came later than the lengthening before voiceless fricatives and /r/.

    These two developments would, I think, give us an earliest possible date of the late 17th century for AmE LOT-unrounding, which is probably still consistent with its being an inherited rather than an indigenous development.

    On the other hand, it’s difficult to rule out the possibility that it’s an independent development.

    • trawicks says:

      This is slightly speculatory, but I think in North America you find LOT pronunciations that are both inherited and/or organic innovations. For example, Northeastern New England has a rounded vowel for LOT words, and it’s possible this feature was indeed inhereted from certain types of British English (Eastern New England inherits a number of later British innovations that are rarely found elsewhere in the States).

      But then there’s the peculiar case of Canadian English. Is the LOT vowel rounded because it inherited features from accents of the British Isles (Scottish English is often believed to influence Canadian speech, although I’m not 100% sold on that notion). Or is the rounded LOT vowel merely a product of the Canadian Vowel Shift?

      • dw says:

        Northeastern New England seems to have been the region of the US most prone to imitation of prestige English speech. I would put LOT-rounding into this category along with (historically) R-dropping .

        It remains to be explained why LOT in Boston is long, while in England it’s phonetically short. This is probably a consequence of the cot-caught merger there.

        • dw says:

          Actually, LOT-rounding in Eastern New England could simply be a consequence of the cot-caught merger, with the former “caught” vowel (rounded and long) generalized for the LOT lexical set.

        • trawicks says:

          Just for the record, not all Eastern New England is merged–Rhode Island and some patches of Southeastern Massachusetts have a split more along the lines of unmerged GenAm or (in some parts of Rhode Island) New York City. The question being why Boston and points Northward are merged.

        • 'enry 'iggins says:

          One thing to consider is the Scots-Irish settlements in America. Scots-Irish areas of Ulster have the COT-CAUGHT merger and pronounce the merged vowel as rounded, as in Scotland. Here’s a map showing Scots-Irish settlements in the U.S. It’s kind of hard to see, but you’ll notice that both western Pennsylvania and the coast of Maine have a high percentage of people with Scots-Irish ancestry. Those also both happen to be areas with a rounded LOT vowel (and the COT-CAUGHT merger of course). But why doesn’t Appalachia have the rounded vowel or the merger?

        • trawicks says:

          Some areas of Appalachia definitely have the rounded vowel, such as rural Western PA and West Virginia. That being said, I haven’t seen much evidence that the Scots-Irish impacted the phonology of Appalachian English. Whereas Northern Irish English and Scottish English are obvious cousins, I can’t find any clear similarities in Appalachian accents besides of the fronted GOOSE vowel. It’s a bit of a mystery, really!

        • 'enry 'iggins says:

          When I wrote “Appalachia” I meant the dark red area on the map around western North Carolina and eastern Tennessee. I was also thinking about the dark red area to the northeast of that one in Virginia. As far as I know, those areas do not have a rounded LOT. Even though western PA might technically be part of Appalachia, I don’t think of it as being part of Appalachia for some reason. Not that I expect you or anyone else to be able to read my mind.

          I’ve found a few similarities between Ulster accents and Appalachian accents upon close inspection. To start with the obvious and the part that actually didn’t take much close inspection: Appalachian speech, like Ulster speech, has always been firmly rhotic as far as I know, even when many (especially coastal) areas of the South had non-rhotic accents. Maybe the Scots-Irish influence sort of reinforced this.

          Another pretty obvious thing to mention is the way /t/ is pronounced when it comes between vowels in both Appalachia and Ulster (and in North American accents of English in general). It is of course often pronounced as a voiced alveolar tap in both places.

          Also some southern Appalachian and Ulster accents have a lax [ɪ]-like (or perhaps more open) happY vowel. In my experience, this is an older feature that isn’t heard so much today in white Southern accents, but it still seems quite common to me in AAVE.

          There is also a lengthening of /ɛ/ and some other vowels in Ulster in many environments. From what I’ve read it happens in monosyllables which are closed by a consonant other than /p, t, tʃ, k/. This lengthened vowel also becomes a diphthong in some parts of Ulster. So head can be [hɛˑəd], Jeff can be [ʤɛˑəf], egg can be [ɛˑiɡ], etc. This is similar to what happens in southern Appalachia and in the South in general (and also something that doesn’t happen in Scotland).

          I’ve read about Scotch-Irish accents in Ulster that have the diphthong [aːe] in words like mine (adjective) and lie (recline). To me this diphthong, with its lengthened first element and more open second element, already seems to be approaching the southern Appalachian (and Southern) pronunciation of PRICE words. In fact many Ulster accents actually pronounce PRICE words as monophthongs before /r/, e.g., tire is often [tæːɻ]. Maybe what happened over time was a “takeover” by the pre-/r/ allophone of this vowel, i.e., it spread from this environment to all others and eventually became the “default realization” of PRICE in southern Appalachia (although I’m well aware that not all Southerners have /aɪ/ monophthongization before voiceless consonants). This is like J.C. Wells’s theory on where GOOSE fronting came from. I’ll admit that this might be a stretch though.

          Monophthongization of MOUTH also happens before /r/ in Ulster. Maybe a (near) takeover of this allophone happened in western Pennsylvania (specifically Pittsburgh) and eventually lead to monophthongs in out and downtown (they also have monophthongization of PRICE there before resonants). Once again, this may be a bit of a stretch, but it’s fun to think about these things.

          However, I don’t know how people in Ulster would have spoken back when they started immigrating to America. I’m just comparing modern Ulster accents to modern Appalachian ones. Also please note that I’m not actually a linguist. Thank you for reading this far.

        • trawicks says:

          It’s not crazy; I’ve done quite a bit of my own thinking about it over the years!

          Southern English has “glide deletion,” which can sometimes resemble the long KITE dipthong in Northern Irish English. That being said, I find it odd, then, that many parts of Appalachia don’t have any distinction between this dipthong before voiced and voiceless consonants as in Northern Ireland. For example, in Belfast “time” would probably be pronounced something like [ta:ɛm] where as “right” would be more like [ɹɛɪt] (rough approximations since I don’t have any texts in front of me). In Appalachia, they would usually both be pronounced with [a:]

          You mention vowel-breaking as well, which indeed does happen in both Northern Ireland and Appalachian English in DRESS words. But in Appalachia, this extends to the KIT and TRAP vowels as well. So again, there is some similarity there, but it’s not quite the same.

          My theory is a bit different. The Scots-Irish initially settled in their largest numbers in Pennsylvania, then filtered down into Appalachia. I’m a little iffy on the history here, but what I think may have happened was that they brought Pennsylvania English down with them and it mixed with some features of American Southern English, while maintaining some Scotch-Irishisms. That to me reflects the way the accent sounds today, although we will probably never know exactly.

        • 'enry 'iggins says:

          “Southern English has “glide deletion,” which can sometimes resemble the long KITE dipthong in Northern Irish English. That being said, I find it odd, then, that many parts of Appalachia don’t have any distinction between this dipthong before voiced and voiceless consonants as in Northern Ireland. For example, in Belfast “time” would probably be pronounced something like [ta:ɛm] where as “right” would be more like [ɹɛɪt] (rough approximations since I don’t have any texts in front of me).”

          Actually many parts of the South and Appalachia do have that distinction. Western Pennsylvanian accents and many Southern accents have raising of /aɪ/ before voiceless consonants (often known as Canadian raising). In Belfast, time and right would both have the same vowel from what I’ve read about that accent. Both would have [ɛɪ] (or even [eɪ], making them sound very much like tame and rate, respectively) because /aɪ/ is in a closed syllable in both words. There would of course be “pre-fortis clipping” (thanks John Wells) in right though, so the vowel in it would be shorter than the one in time.

          “You mention vowel-breaking as well, which indeed does happen in both Northern Ireland and Appalachian English in DRESS words. But in Appalachia, this extends to the KIT and TRAP vowels as well.”

          It extends to TRAP/BATH/PALM (which are of course all merged) in Northern Ireland too. But KIT is always short there. So the lengthening and diphthongization in KIT could have been a Southern American innovation. But my point is, there may have already been short vowel lengthening going on in Ulster a while ago and that lengthening may have extended to other traditionally short vowels after those people immigrated to the South.

          As for monophthongal MOUTH and PRICE, another possible place of origin is England. Many settlers in the South and elsewhere came from there. I’ve read that many settlers in the South and some in PA came from the North of England and the West Country in particular. A monophthongal MOUTH can be found in some parts of the North of England . A monophthongal PRICE can also be found in some areas of the North of England. Apparently, some parts of West Country have a monophthongal PRICE too.

          “I’m a little iffy on the history here, but what I think may have happened was that they brought Pennsylvania English down with them and it mixed with some features of American Southern English, while maintaining some Scotch-Irishisms.”

          But where did Pennsylvania English and Southern American English come from? That’s what I’m trying to figure out.

        • trawicks says:

          “In Belfast, time and right would both have the same vowel from what I’ve read about that accent.”

          My personal perception is that it varies. Although after listening to a Belfast speaker last night, his pronunciation of the word “fine” sounded like fæ:en, so it’s definitely not far off from ɛɪ. That supports Wells’ assertion.

          “Actually many parts of the South and Appalachia do have that distinction.”

          Sort of. The distinction is very common in the Coastal South, and is certainly a huge feature of Pennsylvania English (Philadelphians usually pronounces “right” as [ɹəɪt]). But if you take a look at Rick Aschmann’s map here, you’ll see that this distinction notably does not apply for the classic Southern Appalachian region. But mergers happen all the time in American English, so it’s hard to say if that indicates anything.

          “A monophthongal MOUTH can be found in some parts of the North of England.”

          Most definitely. Both monophtongal PRICE and monopthongal MOUTH can be found in many parts of Northern England. Actually, if somebody said Appalachian English phonology came from Northern England, you could make a very convincing argument.

          One problem I want to point out here is that Appalachian English, like much of Southern English, has undergone Labov’s “Southern Vowel Shift,” which obscures the original sources of many of these accents. The basic jist of the shift is similar to that of Australian English: TRAP –> DRESS –> KIT –> FLEECE –> FACE –> PRICE. Sadly, this shift steamrolls over a lot of evidence of original phonetic influences.

          Now as to how Scots-Irish influences Appalachian dialects, well that’s a whole other ballgame. I don’t want to suggest that the Scots-Irish didn’t leave its mark, just that its impact on pronunciation is difficult to discern.

  4. Doris says:

    I’m originally from Essex, England, although my accent veers between RP and an almost Antipodean tendency to vowel shift (not even close to what’s become known as Estuary, by the way). What I can’t do is pronounce “Mom” anywhere close to the pronunciation that my expat Texan friend is attempting to teach to her English born toddler. It either comes out closer to Ma’am or rhyming with bomb…

    • dw says:

      I don’t know exactly how your Texan friend sounds, but you could try “Mum” with an extra-long vowel.

    • Muhammed says:

      I agree! The teachers here are somieemts really mean to the kids, especially if they don’t know the answer right away. I always try to compliment them if I can at least understand what they’re saying. And telling them I feel the same way when I speak Spanish. Hopefully it boosts their confidence!

  5. m.m. says:

    Your mentioning of cot-caught/lot-thought merged areas [canada/western us/eastern new england] reminds me that there are also people who are transitional, and can use [ɒ] or [ɑ] interchangeably within the same sentence, even for the same word. Talk about overlap.

  6. Short-o patterns are a surefire(ish) way to distinguish Philadelphia and New York City, which can be kind of difficult for people outside the Mid-Atlantic. Of course, these aren’t quite law-like patterns, but in general

    Philadelphia:
    — on = Dawn
    — chocolate = chock-late

    NYC:
    — on = Don
    — chocolate = chalk-late

    • trawicks says:

      That’s often true, although there are definitely a few On = Dawn speakers in New York City. And, as you mentioned, there are certainly a number of On = Don speakers as well. Not sure what factors influence which pronunciation, though. Needless to say, there’s a lot of variation in the Big Apple!

  7. Andrej Bjelaković says:

    While it is true of course that [ɑ] in LOT had been present in many English accents since… well, certainly since Shakespeare’s time, the LOT-PALM merger (i.e. father-bother merger etc.) is an American inovation, which had taken place by early 19th century.

  8. You actually make it appear so easy along with your presentation however I to find this matter to be actually something which I believe I’d by no means understand. It seems too complicated and very wide for me. I am looking forward in your next put up, I