Decline in Diversity of English Dialects.

The University of Cambridge has produced an English Dialects App that is allowing them to track how dialects are doing:

Regional diversity in dialect words and pronunciations could be diminishing as much of England falls more in line with how English is spoken in London and the south-east, according to the first results from a free app developed by Cambridge researchers.

The English Dialects App (free for Android and iOS) was launched in January 2016 and has been downloaded more than 70,000 times. To date, more than 30,000 people from over 4,000 locations around the UK have provided results on how certain words and colloquialisms are pronounced. A new, updated version of the app – which attempts to guess where you’re from at the end of the quiz – is available for download from this week.

Based on the huge new dataset of results, researchers at Cambridge, along with colleagues at the universities of Bern and Zurich, have been able to map the spread, evolution or decline of certain words and colloquialisms compared to results from the original survey of dialect speakers in 313 localities carried out in the 1950s. […]

Dialect words are even more likely to have disappeared than regional accents, according to this research. Once, the word ‘backend’ instead of ‘autumn’ was common in much of England, but today very few people report using this word (see map slideshow).

However, the research has shown some areas of resistance to the patterns of overall levelling in dialect. Newcastle and Sunderland stood out from the rest of England with the majority of people from those areas continuing to use local words and pronunciations which are declining elsewhere. For example, many people in the North-East still use a traditional dialect word for ‘a small piece of wood stuck under the skin’, ‘spelk’ instead of Standard English ‘splinter’.

On the controversial word “scone”:

Adrian Leemann said: “Everyone has strong views about how this word is pronounced but until we launched the app in January, we knew rather little about who uses which pronunciation and where. Our data shows that for the North and Scotland, ‘scone’ rhymes with ‘gone’, for Cornwall and the area around Sheffield it rhymes with ‘cone’ – while for the rest of England, there seems to be a lot of community-internal variation. In the future we will further unpick how this distribution is conditioned socially.”

Fascinating stuff; thanks, Pat!


  1. This is more or less what’s happened in the U.S.: the old dialect boundaries defined by classical dialectology (mostly based on variation between words, like New England spider for general frying pan) has been leveled out, but the exact same boundaries are now maintained by accent differences, and new accent differences have arisen.

  2. Bathrobe says

    I was surprised when listening to U.S. speech samples on line (I can’t remember which site) to find that Southern speakers did not have what I would (probably incorrectly) regard as stereotypical Southern accents. Most of the Southern speakers sounded to me only a little different from General American. Is it fair to say that Southern accents are dying out (in their more extreme form, at least), or is it just a product of the methods used to collect samples (which is to have people read out a sample of written English)?

  3. @Bathrobe: Well, the typical Southern features today are a lowered second element in PRICE, a breaking of most of GA’s true monophthongs (though rarely as extreme as what you hear in imitations), a fronting of GOAT, and the pen-pin merger. My impression is that these traits are all pretty healthy, and the last one even seems to be spreading. What you may be noticing is the decline of non-rhoticity, which doesn’t represent de-Southernization but rather the increasing dominance of upland over lowland Southern varieties.

  4. Of my wife’s three cousins, who were all born in the 1940s in North Carolina and lived in the South all their lives, two were biaccentual: they could readily shift between their native accent and an approximation to broadcast American. The third one was not, and sounded like a good ole girl all the time. (My wife’s own situation is inevitably more complex.)

  5. John Cowan’s mention of frying pan made me realize I don’t use that term or hear it much anymore. It seems that skillet is taking over.

  6. Well, I will venture to argue with the data. Skillet to me is a higher-register word and more likely to appear in formal sources like books. I imagine a historical corpus of cookbooks properly tagged would show sauté steadily dominating pan fry or just fry over time, but people still speak of frying their eggs (scrambled or otherwise); sautéed eggs would be pretentious.

    Now it’s true that does show skillet dominating frying pan 200 to 65; however, this reflects not just raw ghits but also search volume, and more people search for single words than for phrases by a long chalk.

    As further anecdata, my daughter and her ex-boyfriend, both native speakers and native New Yorkers of about 30, responded with frying pan when shown the article in question.

  7. Could be. I’ve never thought of skillet as high-register, but maybe I’m just too much of an urbanite now.

  8. Matthew Roth says

    Downton Abbey is perhaps the best evidence in the decline of accents. Some accents were very authentic, but it seems that no one was from Yorkshire, nor was the register anywhere near that of 1912.

    The University of Texas, Austin blog had some comparisions of speech patterns between various politicians, including Trump and Sanders, but of importance here are the ones on Texan politicians & Texan accents. Their research shows that the pen–pin merger is younger and more urban for Texan speakers, but it seems to be stereotypical Southern (or perhaps Appalachian, more specifically) English to me.

  9. Nah, pin-pen is all over the South. It’s one of the few things that marks off the New York dialect (non-AAVE-influenced), which doesn’t have it, from Yat dialect in New Orleans, which does. My friend from NOLA sounds like he was born in Brooklyn (where he now lives) until you hear him say pen, which stands out like a sore thumb.

  10. @Matthew Roth: Huh. I saw a paper (can’t find it at the moment) that found that the pen-pin merger was less common in big Southern cities than in surrounding areas, largely because of non-Southern migrants.

    On the topic of old-time British accents, something that I think often goes unnoted is that there seemed to be two somewhat different prestige varieties: an “aristocratic” one and a “bourgeois” one. They shared several traits, but in some respects they were on opposite sides of today’s normative RP: the former had a more fronted GOAT and a higher THOUGHT, for example, while the latter had a more backed GOAT and a lower THOUGHT. “Bourgeois” U-RP (which I find to be the more pleasant of the two) is what classically trained British actors used to use, and still survives in a few old fogies like Patrick Stewart and Ian McKellen.

  11. I’m surprised at the (alleged) disappearance of ‘backend’ instead of ‘autumn’. ‘Backend’ was alive and well when I lived in West Yorkshire ~25 years ago. Particularly in the phrase “it’s getting backendish”.

    Perhaps global warming means no more no warmth no proper time of day November.

  12. David Marjanović says

    two somewhat different prestige varieties: an “aristocratic” one and a “bourgeois” one. They shared several traits, but in some respects they were on opposite sides of today’s normative RP: the former had a more fronted GOAT and a higher THOUGHT

    The former? My preliminary impression is that these two features are spreading, with THOUGHT often reaching cardinal [o] and GOAT being all over the place, having for some people hit the front edge of vowel space.

  13. Yes – back in the day it was the royals and aristocrats who were closer to Cockney. (Chalk it up to countersignaling.) Their “plummy” THOUGHT was basically the same as what Estuary now uses; their fronted GOAT stayed mid, though, whereas Estuary tends to lower it. Bourgeois or theatrical speech treated these two vowels more conservatively, akin to what you’d hear in the North of England or in old-fashioned General American.

  14. Not countersignaling (maybe later on it was) but simple conservatism. It was the middle classes that needed a differentiator and got one in the form of Conservative RP.

  15. Not countersignaling (maybe later on it was) but simple conservatism.

    I don’t catch your meaning? The aristocratic/royal lect was the more innovative one and the closer to Cockney: what I was getting at is that those at the top could afford to share some traits with the poor, because they knew there was no risk of being confused with them.

  16. Matthew Roth says

    Lazar, I think that you and John Cowan agree. It seems that the middle class abandoned earlier spoken features to distinguish themselves from the lower class and the aristocrats above them.

    OK, so for PIN–PEN, I misinterpreted the article. It is a younger feature according to their research, which is still found in large cities. I know that there’s a recent paper by a UT prof on this topic from a conference…

  17. Here are some (very very long) quotes from a University of Pennsylvania-educated linguist that might be relevant here. I really hope you’ll let me post them here. These quotes are from an online conversation he had in his free time, not from a research paper. If you don’t believe me, google part of these quotes and you’ll find where they’re from. It is a response to an American person who said that they thought regional accents were receding in America (emphasis mine):

    About 10 years ago the Atlas of North American English came out, the first full-scale in-depth nationwide survey of regional accents. It was based on data collected in the mid-’90s, and seemed to show that regional accents were becoming more strongly differentiated. So in the mid-2000s, the party line among the linguists who study this kind of thing was “the idea that regional dialect differences are fading away is a myth; they’re actually increasing”, and so that’s what we would all say when we were interviewed by the media about dialects or whatever. But in the past 10 years more and more research has shown that regional dialect features actually are diminishing; this has been found in New York and Philadelphia and New Hampshire and Lansing and Syracuse and Raleigh and Houston. And the accent features that still are increasing are the ones that are shared by a whole lot of regions (this has been found in Columbus and San Francisco and Anchorage and Toronto and….) So now maybe it’s starting to seem like the conventional wisdom was right all along. And maybe there are new regionally distinctive features whose differentiation still is increasing, but if so most of them are still under the radar and we haven’t really detected them yet.

    Here’s what his interlocutor said at this point:

    I think I’ve got it. Local accents are going away, but gross regional accents (northeast/southeast/midwest/west coast) with distinctive features may be developing? Something like that?


    Not even! The Atlas, based on data from the ’90s, found tons of gross regional accents…The Northern Cities accent stretches from Utica NY to Milwaukee WI, basically identical, with only a brief interruption at Erie PA. But the research done in the past several years has seemed to find the distinctive features even of large regions like the Northern Cities and the South fading somewhat, in favor of a relatively unmarked pattern that was originally described as the “Canadian Shift” and the “California Shift” until elements of it were found appearing in other regions as well as they retreated from the dialects described in the Atlas.

  18. Interesting; thanks for the quotes!

  19. Yeah, I’ve been skeptical of that party line. Here in New England, the trajectory definitely seems to be one of General-Americanization, with a decided lack of new regional features to replace those being lost. And I agree that a mild form of the California Shift is the new sound of General American, having thoroughly displaced the conservative Midland way of speaking that used to bear the mantle.

  20. Bathrobe says

    Homogenisation. How depressing.

  21. Here’s the source of the conversation in my last comment, by the way.

    @ Lazar: You’re not alone there. Most non-linguists (including me) have always been skeptical of that party line.

    @ Bathrobe: Yeah, I agree. But at least there are still national accents, at least for the time being. And there are still cultural differences between countries. There are different norms of politeness in different countries, for example (see the work of Lynne Murphy of “separated by a common language” fame). Those can often be more frustrating than interesting, though.


  1. […] Hat notes the apparently declining diversity of English dialects in […]

Speak Your Mind