Pronunciation Errors that Changed Modern English.

An Alternet piece by David Shariatmadari begins with this charming anecdote:

Someone I know tells a story about a very senior academic giving a speech. Students shouldn’t worry too much, she says, if their plans “go oar-y” after graduation. Confused glances are exchanged across the hall. Slowly the penny drops: the professor has been pronouncing “awry” wrong all through her long, glittering career.

We’ve all been there. I still lapse into mis-CHEE-vous if I’m not concentrating. […] The point is malapropisms and mispronunciations are fairly common. The 20-volume Oxford English Dictionary lists 171,476 words as being in common use. But the average person’s vocabulary is tens of thousands smaller, and the number of words they use every day smaller still. There are bound to be things we’ve read or are vaguely familiar with, but not able to pronounce as we are supposed to.

The term “supposed” opens up a whole different debate, of course. Error is the engine of language change, and today’s mistake could be tomorrow’s vigorously defended norm. There are lots of wonderful examples of alternative pronunciations or missteps that have become standard usage. Here are some of my favourites, complete with fancy technical names.

There’s nothing surprising in the list for those of us who concern ourselves with such matters, but it’s well done, and this bit did actually address an issue I’d wondered about:

In Norwegian, “sk” is pronounced “sh”. So early English-speaking adopters of skiing actually went shiing. Once the rest of us started reading about it in magazines we just said it how it looked.

Comments

  1. Casey Chapple says:

    My mom descended the stairs one morning long ago, and I told her she looked bed-raggled. She laughed and said, well, that works, too.

  2. Shariatmadari’s article originally appeared in the Guardian last year. I point this out not to be picky but to note that it generated 3576 comments. Perhaps LH commenters should aspire to equal that.

  3. AJP Ram'sbottom says:

    Norwegian sk is only pronounced like English sh when it’s followed by i or y (Norwegian sky means cloud, skygge is shadow). When it’s followed by other letters or by nothing (husk means remember, å huske is to remember), it’s pronounced the same as English sk.

  4. Ken Miner says:

    I recently watched a lecture by a guy one of whose books has been translated into thirty languages. He pronounces “anemone” the way I did when I was ten ([‘ænɪmown]) and apparently thinks that “wizened” means “made wise”. As in the case of your senior academic, what I mainly wonder is why no one ever told him/her.

  5. cardinal gaius sextus von bladet says:

    Any Norwegish-speakers around to tell us authorititatively that “sk” only gets “sh”ed in front of front wovels? Because, for example, skål.

  6. David Eddyshaw says:

    I was at a meeting last year which involved the award of a prize for best presentation to one of the trainees. Distractingly, the chairman of the meeting persistently described this as giving a prize to the most meretricious.

  7. Ксёнѕ Фаўст says:

    It reminds me the surprise when I found out that quite a few common English words acquired their current pronunciation solely via (pseudo-)etymologizing orthographic meddling, e.g. adventure, admiral, assault.

  8. “oar-y” is not an accent-neutral respelling. I presume the isomorphic error in Irish English would be [ˈɔ ri] rather than [ˈor i]

  9. @Ксёнѕ Фаўст: “Admiral” is interesting for me because it’s one of the only two words in which I have a syllable beginning with /mɹ/ – the other being “admirable”.

  10. Shariatmadari’s article originally appeared in the Guardian last year. I point this out not to be picky but to note that it generated 3576 comments. Perhaps LH commenters should aspire to equal that.

    And, as it turns out, I posted about it then (it got 48 comments). I thought that name Shariatmadari sounded familiar. When, oh when, will I learn to automatically search my archives before posting?

  11. George Grady says:

    I found Shariatmadari’s comment on pronouncing tune as “chune” surprising: “Within a single generation it has pretty much become standard English.” I guess this is a British thing? And are tutor and duke really pronounced “chooter” and “juke”? Is this broadly nonAmerican, or mostly restricted to England?

  12. I agree George Grandy. I am American born, but have been living in various European countries for many years. I have never heard these pronunciations!

  13. Broadly non-North-American. Original stressed /tju/, /dju/ was simplified to /tu/, /du/ in North America, spreading roughly from north to south. My wife, born in North Carolina in 1943, says Tyuesday with an unreduced pronunciation of day, whereas I say Toosdee (which makes me sound like a gangster, she says). In the rest of the Anglosphere, though, the change /tju/ > /tʃu/ and /dju/ > /dʒu/ has been spreading for the last century or more. In unstressed form it is much older and therefore universal, as in nation with /tjon/ > /sjon/ > /ʃn/.

    The discussion of “dark l” is also irrelevant to North American English, where all /l/ is dark and there is no change to /w/ (much less /o/) going on.

  14. Rodger C says:

    The discussion of “dark l” is also irrelevant to North American English, where all /l/ is dark

    When I tried to point this out to my undergraduate linguistics professor, he said indignantly that it was physically impossible. He was from Beijing by way of Hong Kong. This was not the same professor who told me later that Mir wird geholfen was “contrary to linguistic science.” And people wonder why I didn’t take a linguistics minor.

  15. @count vB: å isn’t a front vowel, is it? And the rule actually applies only to k, g, sk before i, y, because they are spelled kj, gj, skj before the other front vowels (e, æ, ø). (skji and skjy would be harder to read, I guess the thinking is).

    Danish used to have the same spellings in many places (Kjøbenhavn, Gjørlev, Skjern and so on) even though the standard language never had the palatalization. By the time an official orthography started taking form the Swedes had already nicked Norway — so the rationalization used was that some Jutlandish dialects had palatalization as well. A bit later we dropped it. Except of course in the cases where the j was always pronounced in Copenhagen anyway (as a glide).

  16. I disagree about all /l/ being dark in NAE. That’s true for many speakers, but not all.

  17. I should have explicitly excluded AAVE. Other than that, WP says that clear l is found intervocalically in the South, but I have not heard it there.

  18. … that Mir wird geholfen was “contrary to linguistic science.”

    Then that science is out of touch with reality, and needs to change a principle or two – because helfen takes the dative, and there’s an end on’t. Such nonsense must be opposed, unfortunately, It is the Strunk and White man’s burden.

    Maybe the professor was thinking of the famous advertising slogan for the Telekom telefone directory service, as delivered by Verona Pooth: “Hier werden Sie geholfen”.

    https://www.youtube.com/watch?v=2XxNoQ8V0eY

  19. Da werden Sie geholfen”.

  20. She’s one of those very clever women with contraindicative boobs

  21. Damn, I listened to it, but got distracted by the fact that in the internet “hier werden Sie geholfen” is also floating around.

  22. The discussion of “dark l” is also irrelevant to North American English, where all /l/ is dark and there is no change to /w/ (much less /o/) going on.

    I noticed recently that a couple of my friends (both from the Pittsburgh area) pronounce the word “full” in a way that sounds unusual to me. I don’t know how to transcribe it exactly, but the ‘l’ does sound something like a ‘w’ and the vowel also sounds strange, more like ‘fool’ than ‘full’ (though they still make a distinction between full and fool). When I first noticed this I thought it was just an idiosyncracy, but then I heard it from the second guy from the same area, so I started to wonder if it was a regional thing.

  23. I bet “hier werden Sie geholfen” is derived from the slogan, and used by people in advertising *copy* to tout their services. In text, you can’t point, as Pooth does in the clip.

  24. David Eddyshaw says:

    @George Grady:

    And are tutor and duke really pronounced “chooter” and “juke”?

    Yup. Sure are.

    Actually, what surprised me about this was “within a single generation.” I’m two generations old and I’ve always pronounced them that way.

    Perhaps where I lead, others follow? (In the UK, at any rate.)

  25. This could not happen in the USA within a single generation, because Duke of Earl.

  26. Syllable-initial /l/ may not be as free from velar influence in AmEng as it is in (most of) England, but I still think it’s common for North Americans to have some allophony between clear and dark. There are some people like Tom Brokaw and Ira Glass who use a heavily velar /l/ at the starts of syllables, and people notice it – and make fun of it. I used to pronounce /l/ that way too, but over time I realized that this wasn’t as common as I thought it was, and I gradually modified my pronunciation to include clear/dark allophony.

  27. David E: I think David S means that it’s only within the last generation that chune for tune (and its relative chree for tree) has ceased to be stigmatized as non-RP.

    Lazar:A history of being made fun of made me fairly immune to such considerations.

  28. David Eddyshaw says:

    @JC:

    I can’t cite any authorities, but if anything I think it may be the other way about; I think “chune” for “tune” etc has been standard RP for a good while, but the changes have been partly resisted by the anxious genteel due to spelling pronunciation and hypercorrection. As anxiety of this type has faded, so has the resistance.

    My own idiolect is not in fact exactly RP (it’s rhotic, for one thing) but is almost invariably taken to be so be other UKanians, and has never been subject to class-based anxiety as far as I can tell. (Possibly the trauma was so profound that I have repressed it all.)

  29. What about shchupid for stupid? I’m sure I’ve heard that on EastEnders.

  30. David Eddyshaw says:

    I suspect quite a lot of speakers have long been under the misapprehension that they were saying “tyoon” when (unless their attention was specifically drawn to the matter) they were, as a matter of fact, saying “choon.” The Londom Bridge phenomenon. It takes a veritable Panini to hear these things sometimes.

    Even I can almost succumb to the illusion sometimes that “knight” is not exactly homophonous with “night.”

  31. David Eddyshaw says:

    This matter is empirically testable. Who’s got a DVD of old Trevor Howard movies?

  32. About thirty years ago, I asked several Londoners (I am American) on the street where the Tube was. They looked puzzled and couldn’t answer me. Then I remembered that everyone who I had talked to previously said Tschube. I soon found out where the underground was. Ah, well, an interesting experience in the linguistic kaleidoscope.

  33. Bathrobe says:

    The thing about ‘ski’: isn’t the difference in pronunciation between ‘skirt’ and ‘shirt’ related to this somehow?

  34. Trond Engen says:

    Lars: And the rule actually applies only to k, g, sk before i, y, because they are spelled kj, gj, skj before the other front vowels (e, æ, ø).

    Except (to make it even simpler) before the diphtongs (ei, øy): geit “goat”, skøyter “skates”.

    (skji and skjy would be harder to read, I guess the thinking is).

    Or maybe because they were still regular back when the principles of orthography were being settled. Less so now, with loans like kinderegg and gyrokompass and reading pronunciations of rare words like kyse “riding hood” and gytje “mud (in mudbath)” and names like Gyda and Kyrre.

    A similar exception is single word-final m, which is now irregular with regard to the length of the preceding vowel: lem [lem:] “lid” vs. krem [kre:m] “whipped cream” vs. lim [li:m] “glue” vs. krim [krim:] “detective fiction”.

  35. Bathrobe says:

    Oh, yes, ‘tu’ and ‘du’ are all ‘chu’ and ‘ju’ for me. The Jew on the grass, the Jew date, stchupid, and all that.

  36. Trond Engen says:

    Stu: the Strunk and White man’s burden.

    Perfect.

  37. Rodger C says:

    I told my Mir wird geholfen story a while ago either here or on LL; I won’t repeat the whole thing, but there it was, 40-odd years ago, in the pages of Koutsoudas’ textbook:

    Er sieht mir : Ich werde gesehen.
    Er hilft mir : Ich werde geholfen.

  38. Is the pronunciation of deaf as /dif/ in some dialects a spelling-based pronunciation?

  39. OED (1894) says: “The original diphthong remains in northern dialect; in standard English the vowel was long until the modern period, and so late as 1717–8 it was rhymed with relief by Prior and Watts; the pronunciation /diːf/ is still widely diffused dialectally, and in the United States.”

  40. So the question is, why did it start being pronounced /def/?

  41. LH: And, as it turns out, I posted about it then (it got 48 comments). I thought that name Shariatmadari sounded familiar.

    What jogged my memory was Shariatmadari citing “expresso” for espresso as a common solecism: I hadn’t thought that Shariatmadari, who’s usually pretty sensible, would still buy into the snobbery over “expresso”. But his column was first published in March 2014, predating the news (to some) in mid-2014 that “expresso” has a sound pedigree. LH linked to it here.

    The insistence that only “espresso” can be correct is in the same little-knowledge-dangerous-thing category as English speakers’ pronunciation of Barcelona with a Castilian soft c, or of Apulia with a pseudo-Italian pronunciation. Is there a term for this type of misplaced over-correction? Fowler’s “genteelism” doesn’t quite fit.

  42. Trond Engen says:

    Yeah, the spelling ‘deaf’ and the pronunciation with a long vowel are consistent with its Scandinavian cognate being døv/dauv, rhyming with ‘leaf’ ~ løv/lauv.

    Many of these irregular shortenings are analogical from compounds, derivations or declined forms. The verb ‘deafen’?

  43. The usual pronunciation of deaf, like head, bread, dead, thread, lead, instead, heavy, breakfast, sweat, death, feather, leather, heaven, endeavo(u)r, meadow, measure, peasant, ready, treasure, weather, breast, dealt, cleanse, wealth, jealous, treachery, weapon, zealot etc., reflects irregular and unpredictable shortening of the Early Modern vowel /ɛː/ before it shifted to /iː/.

  44. George Gibbard says:

    John Cowan said: “… [in unreduced syllables] the change /tju/ > /tʃu/ and /dju/ > /dʒu/ has been spreading for the last century or more. In unstressed form it is much older and therefore universal, as in nation with /tjon/ > /sjon/ > /ʃn/. ”
    Can’t agree with your /o/: Chaucer spelled the word nacioun raison. And you want “reduced” rather than “unstressed” because of forms like [ˈnɛpˌtun].

    Certainly AAVE and southern American speakers, as well as many Hispanics, have light l. Myself (Michigan), in (roughly?) those positions where RP has light l, I have dark [lʶ], and in (roughly?) those positions where RP has dark l, I have (ʁ̞), a uvularish approximant without any closure in the coronal region.

  45. George Gibbard says:

    In a word like “television”, I’m not sure if I usually use [ʁ̞] or simultaneously have approximation in both the uvular and alveolar area: [l̞ʶ]. Both feel normal to me.

  46. The usual pronunciation of deaf … reflects irregular and unpredictable shortening of the Early Modern vowel /ɛː/ before it shifted to /iː/.

    Then how come it only shows up in the eighteenth century?

  47. Dialect mixture!

    (Isn’t that always the answer for these things?)

  48. David Marjanović says:

    “Oi! Santa! Word of advice! If you’re after a man with a sonic shkshoojiver, don’t let him near the sound system!

    Er sieht mir

    Classical Berlin dialect. Possibly extinct.

    døv/dauv, rhyming with ‘leaf’ ~ løv/lauv.

    Also German taub, rhyming with Laub “foliage”.

  49. Rodger C says:

    @David: I meant to type Er sieht mich. Koutsoudas was asserting (in his graduate intro textbook of linguistics) that passivization always follows a consistent surface pattern in German, which of course is absurd to anyone with a year of German. He was the head of the Indiana U linguistics department at the time. I’ll retell the whole story if there’s interest, and if it can’t be found and linked to; it’s rather wonderful in its horrible way.

  50. OED evidence of short e in deaf early:

    1405 (▸c1387–95) Chaucer Canterbury Tales Prol. (Hengwrt) (2003) l. 448 But she was somdel deef [v.r. def, defe] and that was scathe.

    c1440 Promptorium Parvulorum 115 Deffe, surdus.

    a1538 T. Starkey Dial[ogue between] Pole & Lupset (1989) 141 As you wold tel a tale to a deffe man.

  51. Ken Miner says:

    @ Roger C I for one wouldn’t mind hearing the whole story. I went to IU for my graduate work and ended up in the Koutsoudas clique; this was later than 1966, when he published _Writing Transformational Grammars_, which must be the book you’re referring to, and later than his chairmanship of the department, though I don’t remember the exact year I went to IU.

    Koutsoudas never even mastered English, let alone other languages, and, like many other “theorists” at the time, got his data at second- or third-hand. After I got my degree and abandoned the clique I spent quite a bit of time correcting data errors on the part of theorists (an activity Koutsoudas called “writing spanking papers”). Koutsoudas had one talent, constructing linguistic arguments. In those days that counted for a lot; data accuracy for very little.

  52. Rodger C says:

    Well, the story is that I saw that in the textbook in Intro. to Graduate studies in Linguistics in ca. 1972, and after the class I went up to the instructor and set him straight about the correct form. He absolutely turned pale with horror. “Are you sure?”

    “I’m absolutely sure.”

    “But–that’s–Contrary To Linguistic Science!” and then: “I’ve got to go talk to Koutsoudas about this!” Whereupon he turned on his heel and ran down the hall without taking his leave. This fellow, whom I prefer to leave nameless, later became head of linguistics at a different Midwestern university. And that’s how I came to abandon a linguistics minor; that and some other experiences of cluelessness and intellectual bullying on the part of the no-god-but-Chomsky-and-Koutsoudas-is-his-prophet bunch.

  53. An American who says “toon” may hear “tyune” as “chune” – particularly if the t is strongly aspirated. That’s not to say that “chune” doesn’t exist, but that it may not be as common as Americans may think.

    On velar l – my father said that in NYC in the 1930’s and 40’s, a velar l was a disqualification for NYC school teachers. Applicants had to pass an oral exam. He had cousins who took diction classes to rid themselves of their “lower class” -i.e., Jewish or Italian – markers, including velar l.

  54. Rodger C, Ken Miner: Your stories are music to my ears.

  55. David Marjanović says:

    @David: I meant to type Er sieht mich.

    The trick is that er hilft mir is correct; the correct passive mir wird geholfen thus absolutely follows a consistent pattern.

    An American who says “toon” may hear “tyune” as “chune” – particularly if the t is strongly aspirated. That’s not to say that “chune” doesn’t exist, but that it may not be as common as Americans may think.

    Evidence that it does exist from a Briton who can hear it: “Tutor for instance is “TOODUR” to a Nebraskan, “TEWTRR” to an Aberdonian, and “CHOO’AH” to a Cockney […]”.

  56. i don’t doubt that it exists – I’ve heard it! But the point being made upthread is that it’s moved from Cockney to more prestigious accents. I wouldn’t be surprised if that’s true (nowadays you hear educated people saying “haitch”) but I also suspect that what Americans hear in some speakers as “chune” is actually an aspirated t plus a dipthong u.

  57. …what Americans hear in some speakers as “chune” is actually an aspirated t plus a dipthong u…

    Many, many British music reviewers write about “pop choons”.

  58. “pop choons”
    Yes. Choon is Cockney for tune. We’ve established that. The question is, has choon migrated upscale, or are Americans hearing “choon” when the kind of snooty Brit who doesn’t listen to pop choons says “tyune’?

  59. Will: Might the l sound in “pull” have been just a velar lateral approximant?

  60. @Bloix: While it’s possible that that sort of mishearing might occur, I think you’re overstating the frequency of it. Yod-coalescence has spread widely throughout Britain (most notably in Estuary and Scottish English) and can no longer just be characterized as Cockney. In a 1998 survey of British speech habits, John Wells found that 61% of people born after 1973 used /tʃ/ in “tune”.

  61. The question is, has choon migrated upscale, or are Americans hearing “choon” when the kind of snooty Brit who doesn’t listen to pop choons says “tyune’?

    You seem to be asssuming it’s Americans who report hearing “choon,” but in my experience it’s largely Brits who do so.

  62. Language Log on the related “chruck” pronunciation of truck (though “jrink” is mentioned only in comments).

  63. @Casey Chapple: I had the reverse problem with “bedridden”, which I originally thought contained the prefix “be-“. (Not that I had any idea what it might mean to “dride” something, any more than to “draggle” or “smirch” it.)

  64. I had the same problem you did with with “bedridden” (thinking it was the past participle of a “be-dride”).

  65. FWIW (and comments are closed there), my truck is not “chruck” at all. To begin with, I have an alveolopalatal rather than retroflex /r/. Now in /tʃ/ the [t] part is apico-postalveolar, whereas my /t/ is normally lamino-alveolar. However, in /tr/ it becomes apico-alveolar, so that the tip of my tongue snaps down from the alveolar ridge to the behind-the-teeth position. If I do this slowly, it becomes /tsr/ with apical /s/, which could easily be misheard as /tʃr/.

  66. George Gibbard says:

    JC: I agree, and if there is any fricative sound [s̺] (apical s) it sounds like the in Basque osasuna¡Salúd!‘.

  67. George Gibbard says:

    like the , and of course “salud” shouldn’t have an accent, and Safari shouldn’t keep changing “salud” to “salad”

  68. George Gibbard says:

    like the s in osasuna

  69. Slightly related to the other thread, and I mention it because it may not be known to my estadounidense friends; ‘lorry’ is losing ground to ‘truck’ in Ireland and Britain, I wouldn’t be shocked if it were obsolete in 30 years. Here’s an ngram, note that the base frequency of “truck” is higher given “truck with” and its other meanings.

  70. Wow, that’s quite a development. Can “trunk” and “hood” be far behind?

  71. Might the l sound in “pull” have been just a velar lateral approximant?

    Thanks for the reply. I’m not really good at phonetics, so I don’t know the answer, but looking at the wikipedia article on Pittsburgh English, I suspect it’s just a combination of these two features:

    /i/~/ɪ/ and /u/~/ʊ/ tense-lax mergers (Layton 1999; McElhinny 1999; Wisnosky 2003; Labov, Ash, and Boberg 2005; Johnstone, Andrus, and Danielson 2006).

    Examples: steel and still are pronounced [ˈstɪl]; pool, pole, and pull are pronounced [ˈpʰʊɫ].

    Further explanation: Before the liquids /l/ and /r/, the tense vowels /i/ and /u/ are laxed to /ɪ/ and /ʊ/, respectively.

    Geographic distribution: the /u/~/ʊ/ is consistently found only in southwestern Pennsylvania (Labov, Ash, and Boberg 2005).

    /l/ vocalization (Layton 1999; McElhinny 1999; Wisnosky 2003; Johnstone, Andrus, and Danielson 2006).

    Examples: well is pronounced something like [ˈwɛw]; milk something like [ˈmɪwk] or [ˈmɛwk]; role something like [ˈɹow]; and color something like [ˈkʰʌwəɹ].

    Further explanation: When it occurs after vowels, /l/ is vocalized, or “labialized”, sometimes sounding like a /w/, or a cross between a vowel and a velarized (or “dark”) /l/.

    Geographic distribution: Southwestern Pennsylvania[2][3][15] (Layton 1999; McElhinny 1999; Wisnosky 2003; Johnstone, Andrus, and Danielson 2006) and elsewhere, including many African American varieties (McElhinny 1999).

Trackbacks

  1. […] Errors that Changed Modern English. http://languagehat.com/pronunciation-errors-that-changed-modern-english/ Language Translation Tools for Ecommerce Sites […]

Speak Your Mind

*