Monday, 4 June 2012

diamond

It won’t have escaped your notice that we are in the midst of celebrating the Queen’s Diamond Jubilee.

I am struck by the number of commentators on British television whom I’ve noticed pronouncing diamond as ˈdaɪmənd rather than ˈdaɪəmənd (or the smoothed variants of the latter ˈdaəmənd, ˈdaːmənd).

In LPD I mark ˈdaɪmənd with the sign §, indicating that as far as BrE is concerned I consider this ‘non-RP’. (In current terminology, you could say that it is ‘deprecated’.) For AmE, on the other hand, I perhaps go too far in giving only this possibility.

I was wondering, has ˈdaɪmənd always been around, though presumably always considered incorrect? Or is it a recent import from AmE? Anyhow, why do Americans tend to use this pronunciation, given that it does not accord with the spelling?

On consulting the OED, I find that ˈdaɪmənd, contrary to my expectation, has a considerable history in BrE. The word diamond appears to be etymologically cognate with adamant, via late Latin diamas, diamant- from the Greek ἀδάμας, ἀδάμαντ- adámas, adámant- ‘untamed, unalterable, unbreakable’.

The a of the middle syllable has tended to disappear since the 16th cent., as shown by the spelling di'mond, dimond. Sheridan and other early orthoepists recognize the dissyllabic pronunciation, but most recent authorities reckon three syllables. In Shakespeare the word is more frequently a trisyllable; but it is very generally dissyllabic in Pope, Thomson, Young, Cowper, Keats, and Tennyson.

So perhaps I’d better remove the § mark. And this word will be a good candidate for inclusion in the next pronunciation preference poll I conduct.

The only other dia- word I can find with a similar pronunciation is diaper. In AmE it seems usually to be ˈdaɪpɚ, to rhyme with wiper, though Webster’s Collegiate adds ˈdaɪəpɚ as an ‘also’ pronunciation. Since this word is not used in modern BrE (though it is found in Shakespeare), it is difficult to say whether there is any established BrE pronunciation of it.

Other candidates? As far as I know, no one says *ˈdaɪdem for diadem or *ˈdaɪgræm for diagram. Does anyone say ˈdaɪlekt for dialect? I wonder.

34 comments:

  1. Are diaper the repeating pattern of small diamonds and diaper the fabric woven in this pattern also disused in BrE? OED3 has not reached the word; the "World English" version of ODO gives them.

    I certainly do say /daɪlekt/ or even /daɪlek/ for dialect. Because the /l/ is dark, there may be an ultrashort schwa-like sound before it, but that does not amount to a /ə/.

    ReplyDelete
  2. Dia- words might be immune, maybe also because people are more aware of the etymological morpheme boundary or an independence of the second part. -dem and -lect (and -cese in diocese) aren't reduced to schwa, -mond and -per are.

    Anyway, just blame the Beatles, filthy hippies, and Lucy in the skaaaɪ with daaaɪmənz.

    ReplyDelete
  3. It seems schwa is optional in 'diary' too, though it's hard to tell before 'r'. In Canada here 'diaper' famously shows Canadian Raising if disyllabic (as for me) but not if trisyllabic (as for my mother).

    ReplyDelete
  4. In Appalatcha they's a-sayin [dɑːlɛk] fer "dialect". (In writin it might could be "dialeck" though.)

    ReplyDelete
  5. Early Philly influence on me gives ˈdʌɪpɚ learnt young alongside ˈdaɪəgræm learnt later.

    ReplyDelete
  6. I've never heard diamond pronounced with three syllables, and I would teach it with two syllables in TEFL. Diaper is the same, but as I'm not American, I only use it under their influence. As for the other ones, they're all /daɪ.ə/ for me.

    You might be interested to know that I've seen this word on a poster in a school where I work here in Japan; but it's misspelt as 'daimond' – presumably because they've assumed that we use the Japanesesque spelling 'ai' for the /aɪ/ sound. I can only assume the poster is about the Jubilee, as I can't think of any other reason for it, but the context escapes me; only this word was interesting.

    Also, you've talked a bit about this 'smoothing' of things like [aɪə] to [aə] in your RP dialect. This sounds weird to me when I try and say it, almost like you're trying to be from the American south. It's not something I'd associate with a British accent. Mind you, I am from Scotland, where people are a bit more trigger-happy about splitting things into multiple syllables (I just remember being taught about syllables in primary school by clapping for every syllable in our names, and two kids called Neil and Claire clapped twice for their names, niː.əl and kleː.əɹ . The teacher, who was from England, was at a bit of a loss to explain why he thought this was incorrect...).

    ReplyDelete
    Replies
    1. Towards the end of this video clip you can hear David Cameron pronouncing science as [sa:ns].
      Hannisdal found smoothing of aɪə to be "very frequent in broadcast RP". See my report of her findings in my blog for 27 April 2007.
      I'm sorry, reuog, that you find this "weird". But it's a fact.

      Delete
    2. Neil (and all words with FLEECE + ə) are a problem across England as well. People do not agree on whether they are one syllable or two. I don't think that the same problem exists for Claire. It might've done when diphthongal pronunciations of SQUARE were common, but most English people use a monophthong now.

      Delete
    3. I just want to point out that I find it an interesting fact. I've just never really heard it before – perhaps young folk don't use it, or perhaps I've just never paid attention. That's the only reason I think it's "weird".

      I do know a guy from Kent or somewhere who says he pronounces "Ireland" and "island" the same. That could be related.

      Delete
  7. In the context of dymond: am I imagining it, or is there a similar tendency to pronounce pirate with rather than aɪə//? This way, the syllable structure is the same as in pilot.

    (Geoff will probably claim so.)

    ReplyDelete
    Replies
    1. I don't see how Frederic's nurse could have apprenticed him to a pirate rather than a pilot, were it not so. (The lyrics throughout are clearly meant to have two syllables in Frederic, though three are possible extrametrically.)

      Delete
  8. I have always assumed the surname Dymond/Dimond was derived from diamond, so perhaps the disylabic form really has a long history

    ReplyDelete
  9. I don't think we should consider 'non-RP' as 'deprecated'. Has anyone ever written to the Times to moan that people don't pronounce 'diamond' correctly?

    There's already a symbol in LPD for contentious pronunciations. I prefer to take 'non-RP' as a completely neutral term.

    ReplyDelete
    Replies
    1. "Depracated" has very negative connotations to me. Professor Wells tags this as "In current terminology", so I'm not sure if there is some difference from the layperson's use of the word.

      I think that most non-RP speakers are unaware that anyone would demand for "diamond" to be pronounced ˈdaɪəmənd. I never paid much attention to this until I read about it in Accents of English. Many of the words beginning ir- have a raised schwa in LPD, which might be an option for "diamond" in the next edition. I am yet to hear anybody say "Irish" as ˈaɪərɪʃ or "irate" as ˈaɪəreɪt. Are there any public figures that use these forms?

      I suppose this brings us on to the old chestnut of whether RP is changing or not.

      Delete
    2. I'd think many people say, less conspicuously, ˈaːrɪʃ. (I'm not talking of accents that have a monophthong PRICE vowel in general, of course.)

      Delete
    3. I looked up Tony Benn, as I thought it likely that there'd be clips of his discussing Ireland online. I watched this video. If you listen from 0:40, he says "Irish" as ˈaɪərɪʃ twice and then he switches to ˈaɪrɪʃ for the third time. Even aristocrats vary.

      So now I have heard ˈaɪərɪʃ. How long until I hear ˈaɪəreɪt?

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

      Delete
    5. Irate is stressed on the final syllable.

      Incidentally, even NED gave (əirēⁱ·t) as its first pronunciation. (Mind you, it also gave (əiræ·sĭb’l) as the first pronunciation for irascible).

      Delete
    6. You're right. I copied and pasted ˈaɪərɪʃ, and forgot to change the stress mark when altering it for "irate".

      Is that NED as New English Dictionary? Which edition is that?

      Delete
  10. Diamond -- usually schwaless: possible schwa in formal style
    Diaper, Pirate -- always schwaless
    Diary, Diagram -- always with schwa

    Near-RP exile living in California for the last 15 years

    ReplyDelete
  11. In my experience (I was born and raised in Seattle and have lived in several other parts of the US), many if not most of the same Americans who pronounce "diaper" and "diamond" with simple daɪ for the first syllable—by which I mean the vast majority of GenAm speakers—also pronounce "dilate" as daɪəleɪt, as if it were analogous with "diagram" and "dialect"! At least, I hear that more often than I hear daɪleɪt.

    To Lipman's question: I do not think that I have ever heard anyone pronounce the word "pirate" with aɪə in the first syllable.

    ReplyDelete
    Replies
    1. Yes, but in America diphthongs never regularly had a special pronunciation before R. Not only is "pirate" not paɪərɪt, but the pronunciations dʒaɪəroʊ, dʒaɪəreɪt, aɪərɪs, aɪərɪʃ, saɪərən also don't exist (although as phonetic variants, I think paɪərɪt, paərɪt, paːrɪt are all very common in AmE, too).

      Contrary to what most dictionaries say, even the other tense vowels remain unchanged before an R in AmE (no pre-R-breaking), especially if the R is followed by a V, e.g. hiːroʊ, seɪrə, stoʊrɪ (if not horse-hoarse merged), tuːrɪst (if not to(ʊ)rɪst, tɔːrɪst).

      Delete
    2. Sorry, I meant: MAY remain unchanged; and on the other hand lax /ɪ e ʊ/ can also be of course realized as /iː e(ɪ) uː/ or /ɪə, eə, ʊə/ in the same environment.

      Delete
    3. What dictionary of AmE shows tense vowels before R? Your first example, 'hero', is an exception; as Prof. Wells first noticed in his 1982 volume (2?), a few words like 'zero' act as if they have an inner word boundary (#), like compounds (compare tense [i] in 'pennyroyal')

      Delete
    4. I agree with Darin. hiroʊ, ziroʊ, and perhaps also zirɑks, are rare and exceptional in AmE. To me they sound like marks of Southern speech. stoʊrɪ and seɪrə are, I think, found only in the speech of a small and aging minority of Southern speakers. (I used to have an Aunt Sarah in Montgomery, Alabama who pronounced her name that way.)

      Also, while perhaps it was merely by inadvertence that Levente referred, immediately after discussing , to i, eɪ, oʊ as "the other tense vowels," I will note that the diphthong is not a tense vowel.

      Delete
    5. "dilate" as comparable to "dialect", yes, I can see. But I don't think it compares to "diagram" since that doesn't have an L sound.

      And I think "dilate" and "dialect" may actually have syllabic L rather than a schwa, at least for some speakers. Seems like an L can't follow a close vowel ([u], [i]), including in diphthongs, in the same syllable. So sometimes, instead of pushing the L to the last syllable of the word, a schwa is added before it, or else the L becomes syllabic.

      Delete
    6. Levente & MKR, another great Alabama example is Arab. As described here, "this city's name is pronounced as if it were two separate words – /ˈeɪræb/ ("AY-rab")." This wiki-description confirms Prof. Wells' intuition that some words in AmE exceptionally have inner #

      Delete
    7. The ˈeɪræb pronunciation was used by British soldiers in the Middle East. Not surprisingly, it was often used in a deprecating way — and by extension as a derogatory word which could be used against non-Arabs.

      When I was a boy, there were no large bodies of British soldiers in Palestine, Mesopotamia, Sudan or (with a brief exception) Egypt. But ˈeɪræb survived in the working-class speech of Nottingham — perhaps heard from fathers, uncles and grandfathers.

      I say working-class because I'm pretty sure it was an other ranks thing. My father served in North Africa but as an officer. He never said ˈeɪræb —nor, come to think of it, did he ever say wɒp or even ˈataɪ.

      Anyway, a wag in my predominately working-class scout group once exclaimed:

      You stupid ˈeɪræb! You're not even an ˈeɪræb, you're a ˈbi:ræb!

      Delete
    8. How does Canadian raising generally interact with r's? I (an American) have one diphthong in pirate, tyrant, tight, and a slightly different one in Irish, highroad, fiery, tied.

      Delete
    9. David, I'm very interested to hear about the British origin and use of [ˈeɪɹæb], thanks. As MDK & I mentioned above, this pronunciation is especially common in Southern AmE. It was also natural to Ray Stevens from small-town Georgia when he wrote Ahab the Arab, a popular novelty song in the early sixties. Mitchell & Rawls (2006 [1976]) highlight "AY-rab" in their tongue-in-cheek The complete how to speak Southern. These are legitimate regional and colloquial contexts, but [ˈeɪɹæb] is now widely generally stigmatized. In The Arab-American handbook: A guide to the Arab, Arab-American & Muslim (Shora 2009), we are specifically warned that "mispronouncing Arab as "Ay-rab" is considered very offensive. ... "Ay-rab" has been used in an insulting manner and has become a racially charged insult" (p. 46). Ironically, I suspect that, at least in North America, this negative connotation owes its spread to the derogatory view of Southerners as ignorant and prejudiced.

      Delete
    10. Peter, In his '"Canadian Raising" in some dialects of the Northern United States' (American Speech 62) Timothy Vance (1987:200) cites "examples of [ʌi] in environments other than before a voiceless consonant. One large group of such examples consists of words in which /ay/ is immediately followed by /r/. All three of my speakers agreed that the following words have [ʌi]: fire, firing, fiery, inquire, iris, inspire, Ireland, Irish, iron, pirate, spiral, tired, wire, and wiry. It is clear from these examples that, in the dialects under consideration here, /ay/ before /r/ tends to be [ʌi]. This is true whether or not the /r/ is syllabic. For example, compare iris, with nonsyllabic /r/, and iron, with syllabic /r/. On the other hand, there are also several words in which all three of my speakers had [ai] before /r/. The complete list of such examples is as follows: briar, crier, diary, flier, friar, gyrate, higher, liar. In crier, flier, higher, and liar, of course, /ay/ is followed by a morpheme boundary. In briar and friar, however, there is no boundary unless we posit some sort of folk reanalysis. [Interestingly, in the context of Prof. Wells' entry:] The ordinary pronunciation of diary for my three speakers has only two syllables, but perhaps we can attribute the [ai] to an alternative three-syllable pronunciation in which the second syllable is schwa: di$a$ry. Aside from gyrate, the Vr spellings of these words indicate that the [r] has been in a separate syllable ever since English spelling was codified. Although words like fire and wire in the all-[ʌi] list clearly have two syllables in my own pronunciation, the respellings indicate that they were monosyllables at some point in the past. Notice also that fiery and wiry allow a two-syllable pronunciation, whereas briary does not. I have no suggestions to offer for gyrate."
      In her 'Canadian raising in a Midwestern US city' (Language Variation and Change 9) Jennifer Dailey-O'Cain (1997), found that in speakers in Ann Arbor, Michigan, /aɪ/ raises almost categorically before voiceless consonant and about half of the time before /ɹ/.
      Raising before /ɹ/ is not widely reported in Canada. In a small study of Vancouver-area English, Rachel Klippenstein (a former student) found raising before /ɹ/ to be much more rare, lexically-restricted and variable across speakers, when compared to Vance and Dailey-O'Cain's findings.

      Delete
    11. I'm a Californian with small children, and for me, the first syllable of "diaper" is not the same as the first syllable of "diamond", the former is /dəɪ/, not /daɪ/.

      I used to live in Oakland's "Dimond District", named for one Hugh Dimond.

      Delete
  12. I think /daɪəleɪt/ is just an exaggeration of the "ultrashort schwa" I mentioned as coming before certain instances of dark /l/. I think this is what Ellen Kozisek is describing. In any case, I don't think it's phonemic /ə/.

    I'm not quite sure what you mean by "markers of Southern speech". I agree that Southerners say hero and zero as if they were two words. I don't, but I still have tense /i/ in them; a lax vowel would belong to a different dialect of AmE from mine.

    In Baltimore, street vendors are called ay-rabbers, pronounced as a compound word, though clearly < Arab.

    ReplyDelete