Wednesday, 4 July 2012

ideas

I heard a financial expert being interviewed on the radio. She was French, but her English was fluent and generally excellent.

Then she started talking about various aɪˈdiːzthe aɪˈdiː of this, the aɪˈdiː of that. She meant ideas, of course, as was obvious from the context.

Unfortunately, this mispronunciation can sometimes lead to confusion. Have you any aɪˈdiː? will be heard as Have you any ID? (any identification), which is something quite different from Have you any idea? (aɪˈdɪə)

I have heard this error from French people several times. It shows that good pronunciation in EFL depends not just on being able to make the right sounds and phonemic contrasts, and to master syllable structure (clusters, final consonants etc), but also to know the right pronunciation for every word in your vocabulary.

...And not to be misled by the spelling. (Given sea siː and flea fliː, you can see the problem.)

67 comments:

  1. I have heard the same from German speakers. This pronunciation might be influenced by German Idee /iˈdeː/ and French idée /ide/, their equivalent of idea.

    ReplyDelete
  2. This comment has been removed by the author.

    ReplyDelete
  3. Yes, I've also heard French people make this mistake often. I've always assumed it was influenced by the French idée, as teardrop suggests. Of course the spelling doesn't help either.

    Idea is also one of those words that's commonly hyperrhotacised in Ireland and Scotland, i.e. pronounced as if spelt ideer. Perhaps one day it will be spelt with an R and French students will get the pronunciation right (although the Scots & Irish will then be getting it wrong).

    ReplyDelete
    Replies
    1. French r is silent, so I don't think it'll be the solution :).

      Delete
    2. What dialect of French is that?

      In standard Parisian French, orthographic R corresponds to [ʁ], with possible devoicing to [χ]. (that is supposed to be a chi, but for some reason the font makes it resemble a Romand lowercase X).

      Delete
    3. French final r is I think what Kilian refers to, as in aider.

      Delete
    4. But, as I'm sure you're aware, most French final rs are not silent either. The exceptions consist mainly of infinitives and agent nouns which, although ending in -er are pronounced as if spelled (e.g. your aider).

      Fer (iron), for example, is /fɛ(ː)ʁ/ not /*fe/.

      Delete
  4. I went to Kansas City on a Frid'y,
    By Saturd'y I learned a thing or two,
    But up 'till then I didn't have an idea
    Of what the modern world was comin' to!

    Rodgers and Hammerstein

    ReplyDelete
  5. Speaking of that word, JoJo Mayer, the famous Swiss drummer always pronounces it as /aɪˈdiːɚ/

    He moved to New York as a young man, I think, and has spent the last quarter of a century living there. His accent is fully rhotic, and he never uses intrusive /r/s. And yet, this one word...

    ReplyDelete
    Replies
    1. 'Idea' is funny because it belongs to a small group of words (variably including 'real' and 'ideal' and their derivatives, and 'theater') which in non-rhotic speech use a /ɪə/ that has no rhotic counterpart - thus, disregarding spelling, 'idear' may seem to be the logical mapping of the phonemes to rhotic speech. Here in New England, I hear lots of rhotic speakers who use 'idear'. I used to, but I trained myself to use [ɪə] instead, treating it as a sort of 'xenophoneme' in my fully rhotic idiolect.

      Delete
    2. Out here on the West coast I never hear anything other than FLEECE + commA.

      Delete
    3. @vp: I considered adopting a trisyllabic "idea", but having grown up with "idear", I just couldn't make the meter of it work. Saying a phrase like "the idea of it" with a trisyllabic "idea" was just impossible for me, so I opted for the derhoticized [ɪə] instead.

      Delete
    4. (long-time lurker, first-time commenter …)
      This is very interesting, as I have been wondering to what extent central vowel phonemes in RP correspond to a rhotic element in e.g. GenAm. Clearly there are a handful of exceptions in the case of /ɪə/. At the risk of going off-topic, can anyone tell me if there are any examples of words which in RP have /ɛə/ /ʊə/ or /ɜː/ and which are not pronounced rhotically in rhotic accents? Thanks!

      Delete
    5. Not many.

      For /ɜː/ there are naturalized German and French loan-words such as Goethe, meuniere.

      For /ɛə/ there is "yeah".

      For /ʊə/ there is, perhaps, "skua"?

      Delete
    6. Most helpful, thank you.

      Delete
  6. @gassalascajape - we must have been typing our comments simultaneously. That's funny, that the same word is hyperrhotacised in New York as well.

    New York is a city of mixed rhoticity, which must cause some confusion. But that this one word is treated the same way on both sides of the Atlantic suggests the rhotic pronunciation may have a long history.

    ReplyDelete
    Replies
    1. The thing is, I'm not sure I've ever heard that hyperrhotacised form from a native New Yorker.
      I'm not sure how JoJo, as a NNS, ended up with that pronunciation.

      Delete
  7. I have heard the same pronunciation in many speakers of various mother-tongues. I'd suppose it is spelling-influenced: sea, flea, tea, plea... . Also, many NNSs say 'reel' for 'real', I feel. Spelling again, I'd supppose.

    I once heard an Italian professor explaining St. Augustine's image of a lion fighting with a bear, but he kept saying 'beer', lion against beer, which did not fail to produce mirth.

    ReplyDelete
    Replies
    1. Many (possibly most) NSs say 'reel' for 'real', too. As best I can tell, it's near-universal here in the US, and gaining ground in Britain too. Oxford (which seems to favor progressive styles of transcription) gives /riːl/ as the only attested pronunciation in its online BrEng dictionary, although Cambridge still only attests the conservative /rɪəl/.

      Delete
    2. Definitely not universal: I have a diphthong in real because of the dark l, which always needs a glide to separate it from a preceding non-back vowel (including even /u/). We can argue about whether this glide is phonemic or sub-phonemic, though. In any case, rear has a monophthong in my accent.

      Delete
  8. aɪm ˈpɒzɪtɪv ðət aɪv ˈhɜd "aɪˈdiər" ən ðə ˈlaɪk frəm ˈmenɪ əˈmerɪkənz frəm ðɪ ˈistən jʊˈnaɪtɪd ˈsteɪts, ˈivn ˈʌpsteɪt njuˈjɔk.

    ən ˈjes, ɪts ɪkˈstrimlɪ ɪmˈpɔtnt tə ˈnoʊ ðə fəˈnetɪk ˈʃeɪp əv ˈevrɪ ˈwɜd ən ˈnɒt bi ˈmɪzld baɪ ðə ˈspelɪŋ :)

    ReplyDelete
  9. Can't say I've noticed rhotic "idea" in Ireland, other than from a few members of that minority who are generally nonrhotic.

    It was easier to forgive French Euro 2012 soccer commentators who called John O'Shea [oʃi] rather than [oʃe], but it shows the same -ea = [i] heuristic.

    ReplyDelete
  10. Wojciech

    Also, many NNSs say 'reel' for 'real', I feel. Spelling again, I'd supppose.


    This native speaker has never made a distinction between real and reel. I've always been puzzled by writers who used the spelling reely to hint at a non-standard pronunciation of really.

    In the LPD, John includes ri:ᵊl in his entry for real and lists it as the only RP pronunciation of reel.

    For me, they're both in the NEAR set. So, similarly, I make no distinction between career and Korea.

    ReplyDelete
    Replies
    1. used the spelling "reely"

      As distinct from /rɪli/, I always thought.

      Delete
  11. David,

    I too have been taught to say 'ri:ᵊl', for 'real' and 'ri:l' for 'reel'. But I must say I most often (seem to) hear 'reely' for 'realy' even from NSs.

    I say 'kori:ᵊ' for 'Korea' and 'kᵊri:ᵊ' for 'career'. Is that wrong?

    I also tended to imagine, perhaps quite wrong(ly) that the schwa in 'idea' (ajdi:ᵊ) was sort of lower or more open, as if the word were spelt 'idear' pronounced non-rhotically, than a normal schwa, in say 'mᵊnotᵊnᵊs' (monotonous). Wrong?

    As Dr. Johnson once said ('London'):

    Some frolic drunkard, reeling (not: realing) from a feast//
    Provokes a broil, and stabs you for a jest.

    ReplyDelete
    Replies
    1. Wojciech

      I too have been taught to say 'ri:ᵊl', for 'real' and 'ri:l' for 'reel'.

      I'm not sure where that 'too' came from. I for one would never teach anybody to say ri:l for reel — or for anything else.

      I suppose it's tied up with my 'dark l' allophone. But it's exactly the same for me with other NEAR words. And I don't think I have 'dark r' and 'lights' allophones.

      As for the different schwas, I have no opinion. In my accent, and many others, there's a a commA lexical set and a NEAR lexical set. The two are distinct — so it doesn't matter to me whether the ə symbol represents exactly the same sound in both.

      Delete
    2. 'dark r' and 'lights' allophones.

      That should, of course, be

      'dark r' and 'light r' allophones.

      Delete
    3. I for one would never teach anybody to say ri:l for reel — or for anything else.

      Goodness, why?!

      Delete
    4. I suppose because of my 'dark l' allophone.

      A following front vowel would change it. For me steel and steal are homophones — but you could possibly hear a difference between my steeling and my stealing. Even so, I wouldn't teach any student to copy me.

      I might well transcribereel as ri:l as a simplification, but not if I had to choose between ri:ᵊl and ri:l — the choice that Wojciech presented in his post.

      Delete
    5. Wojciech

      I say 'kori:ᵊ' for 'Korea' and 'kᵊri:ᵊ' for 'career'. Is that wrong?

      In the LPD John gives:
      RP kəˈrɪə, kɒr-
      (=British non-RP) -ˈri:‿ə
      Gen Am kəˈri:ə.

      Some frolic drunkard, reeling

      John gives ri:ᵊlɪŋ.
      cf ri:ᵊlɪŋ which he gives for both pealing and peeling.

      Delete
    6. Ah yes, so I am not quite wrong.

      Delete
  12. The second chorus of The very though of you begins with
    The mere idea of you, the longing here for you....

    I've just listened to a bunch of performances on YouTube by American singers (and Al Bowly, who tried to sound American). Nearly all had a full blown R or R-colouring at the end of idea. Some had a glottal stop or simple interruption at the start of of but even then had an R-colouring at the end of idea. Only one soloist (Carmen Macrae) seemed to be singing aɪˈdiəʔəv — possibly because that's what the female backing chorus was singing.

    ReplyDelete
  13. The fat that "idea" should rhyme with "sea" is not that far fatched, even for native speakers. Anyone remember the I'm a PC and Windows 7 was my idea Microsoft Windows 7 campaign?

    ReplyDelete
  14. Historically reel was monosyllabic and real disyllabic. In my accent (Northern Ireland) the distinction is preserved (because I have no dark L and no automatic schwa before L) so these words are not homophones for me: ril and 'riəl.

    It's also a rhotic accent so it doesn't make sense to analyse real as having the NEAR vowel. For me it has FLEECE + schwa.

    In RP, FLEECE before L (except when the L is followed by another vowel) has merged with NEAR, so reel has merged with real. But you can see that real has always been pronounced this way (i.e. before this merger) by looking at related words: realise has three syllables.

    I seem to remember typing some of this before...is this the second time we've had this discussion?

    ReplyDelete
  15. 'In RP, FLEECE before L (except when the L is followed by another vowel) has merged with NEAR, so reel has merged with real. '

    I learnt my English 'in th'olden dayes of that Kynge Arthoure', so I haven't reached that stage yet and still pronounce 'reel' and 'real' as does Pete, i.e. as ril and 'riəl, respectively.

    Ad David

    'too' was supposed to mean: 'like a host of those who have been taught a false doctrine/a no longer true doctrine'

    ReplyDelete
    Replies
    1. What is it that John C. Wells said? Oh, that he wouldn't be caught dead prouncing real as a disyllabic word. There exists only rɪəl.

      Delete
    2. Maybe I am being mizzled by the (phonetic) spelling but 'rɪəl' does look like disyllabic to me. Actually, if 'reel' be pronounced with a schwa-like off-glide by some (due to the dark l) I still (stiᵊl) hear it monosyllabic, as I do not hear 'real'.

      I now recall having heard, in days long past, sort of Brian-Sewellesque British accents ('whose lenguid hend...') with a very distinct schwa in 'really', which could not help striking me as exaggerated.

      Delete
    3. This comment has been removed by the author.

      Delete
    4. Wow, Wojciech, I thought you were taught English by the people who wanted their students to speak like the Navy officers from the 1950s, but that then you've managed to neutralize your accent and now speak with a fairly neutral and standard one. But now you tell me you do not say stɪl (‘still’). How did that happen?

      How can rɪəl be disyllabic, when ɪə is a diphthong (though, according to Geoff Lyndsey, a nonexistent one nowadays)?

      Thank you, vp, for the correct quote! I believe that what he said applies to both pronunciations.

      Delete
    5. ʍen did misled become mizzled (ˈmɪzɫ̩d)?

      Delete
    6. Most of the time I still do say [stɪl], only if I suspect that the formidable David Crosbie is eavesdropping on what I'm saying do I flexibly align myself with his well-trained pupils... .

      Re the diphthongicity of 'ɪə' I dunno. Maybe you're right... . But then, 'real' would be, if not one syllable, then perhaps one mora longer than 'reel', to my ear.

      Delete
    7. 'ʍen did misled become mizzled (ˈmɪzɫ̩d)?'

      On the 4th of July 2012, 17:37. I was misled by the spelling, of course, it should have been [maɪsɫ̩d], of course...

      Delete
    8. Duchesse de Guermantes

      What is it that John C. Wells said? Oh, that he wouldn't be caught dead prouncing real as a disyllabic word. There exists only rɪəl.

      Nevertheless he lists ri:ᵊl along with rɪəl for RP. For Gen Am he gives ˈri:‿əl — the stress mark implying dissyllablicity.

      Surely it doesn't matter whether the FEAR sound in any one person's speech is a monophthong or a diphthong? What does matter is that most of us distinguish FEAR from FLEECE.

      Delete
    9. What does matter is that most of us distinguish FEAR from FLEECE.

      Do most Americans?

      Delete
    10. Regarding if American distinguish FEAR from FLEECE: Yes. One's a R-vowel, one's not. Well, that's how this American see it, anyway.

      As for real and reel, personally, I can't see a one syllable pronunciation with FLEECE. Reel is two syllables; not sure if the second, though, is a syllabic L or has a schwa. Real, however, rhymes with ill, rɪl. Except in real estate, where it's pronounced like reel. Am I odd in this, or do other Americans do that?

      Delete
    11. Wojciech

      What does matter is that most of us distinguish FEAR from FLEECE.

      Do most Americans?


      John describes the vowel of the FLEECE set as i in General American and the sound of the NEAR set as ɪ(r.

      Delete
    12. @Ellen K:

      I haven't heard "real" with KIT (I only live here: I'm not a native AmE speaker) -- but do you perhaps have the fill-feel merger?

      Delete
    13. Nope. Didn't know there was such a merger, and definitely don't have it. Those words sound different. Just like, as I said, real and reel do. Real rhymes with fill, reel rhymes with feel, and they don't sound alike.

      Delete
    14. This comment has been removed by the author.

      Delete
    15. This comment has been removed by the author.

      Delete
    16. good gracious... I used to think the most difficult European language vowel-wise was Danish---now I think it's English, after all. And John requires that we poor NNSs should 'know the right pronunciation for every word in [our] vocabulary'...

      Ad Ellen: FLEECE and FEAR. Aside the r-colo(u)ring of the latter, in your speech, are they different? E.g., is the second lower, more open, than the first (in your speech)?

      I too have heard 'rɪl' for real in the US quite often but I thought it was a development analogous to that of 'bɪn' for 'been' except that first the schwa got deleted, so: ree-schwa-l, then reel, then rill. Lazar Taxon in his:

      Many (possibly most) NSs say 'reel' for 'real', too. As best I can tell, it's near-universal here in the US, and gaining ground in Britain too.

      seems to be recording this last-but-one stage, real-->reel.

      Delete
    17. Jason Reid, yes, I'm from the Midwest. Grew up in the St. Louis suburbs, with parents from Chicago.

      Wojciech, I don't get the point of your question. You're asking me if, aside from the big major difference that makes them completely different things, are they different. I'm not sure if they differ in height, but FEAR is much more rounded. It's also a diphthong (or something like that) with the R that doesn't hold at all at the starting point.

      Delete
    18. 'I'm not sure if they differ in height, but FEAR is much more rounded'

      you mean, your 'ea' in 'fear' is somewhat similar to [y], German 'ü', French 'u'?

      Delete
    19. I supposed that's accurate enough, except, transitioning to R.

      Curiously, though, listening to it on a couple different vowel charts, I hear [y] as FLEECE. Partly because the R isn't there, but it also comes from being a singer (church choir), and hearing and singing FLEECE that way in music.

      Delete
    20. 'I supposed that's accurate enough, except, transitioning to R.

      Curiously, though, listening to it on a couple different vowel charts, I hear [y] as FLEECE.'

      both of the above truly amazeth me!

      Delete
    21. Well, there's no /y/ phoneme in any language I speak, so no reason I should hear [y] as a phoneme of it's own. And I certainly can hear the difference between [y] and [i], however, they both for me map onto the same phoneme, /i/ (FLEECE, happY).

      Delete
    22. I'm somewhat surprised your choir director instructs you to realize FLEECE as a rounded [y].

      I'm familiar with choirs being instructed to front GOOSE to [ʉ] or even [y] in the cause of "projecting the sound" (or one of the other pseudoscientific exhortations beloved of choral conductors), but a [y] in FLEECE would sound, I imagine, very odd indeed.

      Delete
    23. Now you are changing what I said. But I really don't want to get into a long discussion. We are told to open our mouths tall, rather than wide. And /i/ sung the same way it is spoken is pretty harsh sounding. And I hope this contents you, because I really don't want to prolong this discussion further.

      Delete
    24. @Ellen K:

      I don't see how I "changed what you said", but, after your further explanation, I think I understand what you're talking about now.

      In view of your desire to terminate this discussion, I hereby do so :)

      Delete
  16. Duchesse de Guermanted

    What is it that John C. Wells said? Oh, that he wouldn't be caught dead prouncing real as a disyllabic word. There exists only rɪəl.

    Nevertheless he lists ri:ᵊl along with rɪəl for RP and ˈri:‿əl for Gen Am — the stress mark implying dissyllablicity.

    ReplyDelete
    Replies
    1. Of course they do. I was saying that for him there is only that one possibility.

      Delete
    2. Sorry to duplicate the post. The Reply link was disabled forcing me to post outside the sub-thread.

      Then the link re-appeared, and since I'd written Duchesse Guermented I decided to re-post it where it belonged and without the typo. I also thought of another paragraph.

      But before I could delete the our-of-sub-thread posting, the Duchesse had posted her reply. Now I hesitate to delete in case I also delete her post.

      Delete
  17. I do have to say that names such as Abdiel first didn't make sense as ˈæb.dɪəl (CEPD), the US version ˈæb.di.əl made more sense.

    ReplyDelete
  18. Ad David

    'Surely it doesn't matter whether the FEAR sound in any one person's speech is a monophthong or a diphthong? '

    crazy tho' this may seem, I sometimes have the impression that English verse could be pliable to moraic metrics, in which case it would matter (whether a monophthong or a diphthong).

    ReplyDelete
  19. Wojciech

    I'm not sure i understand what moraic metric is, but presumably it would only describe the syllabification choices made by one particular user: the author of the verse.

    Pete tells us above that there are more thingies — if not syllables then something you might call morae — in his Norther Irish pronunciation of real than there are in, for instance, mine. Does that mean that if I wrote a poem (with scansion, of course) which included the word real it would offend Pete's ear as a reader? Or conversely if Pete wrote a line with real?

    ReplyDelete
    Replies
    1. No I am not saying 'offend the ear' or such, mine was just speculation based on impressions... rather, get 'value added' out of... unfortunately I am not up to the task of creating moraic metric myself, in English, I haven't ever written any poetry in that language. Would take a new G. M. Hopkins, I reckon.

      Yet, when I hear

      ... and every fair from fair sometime declines...

      the sequence long-short-long in the middle 'tickles' my ear...

      Delete