In non-rhotic accents, words like ‘near‘ and ‘fear‘ generally exhibit two pronunciation patterns: either a ‘centering’ diphthong (ɪə), which might crudely be transcribed ‘ih-uh;’ or as a monophthong, which is usually a lengthened version of the vowel in ‘kit‘ (ɪ:).
But which accents use the monophthong, and which the diphthong? In fact, there’s a lot of variability from speaker to speaker.
For example, I’ve heard broad, Tony Soprano-esque impressions of the New York City accent with exaggerations like ‘whaddaya doin’ hih!’ (that last word is hɪ:, like ‘hit’ without the ‘t’). But in reality, I’ve rarely heard authentic sentences like that in the Big Apple. I usually find that New York ‘here’ has the diphthong hɪə.
I’ve heard Australian accents in which ‘beer‘ sounds nearly like ‘bee.’ This results from the tenseness of the Aussie vowel in ‘kit’ combining with the monopthongization of the ‘eer’ vowel. Hence ‘I’ll have another pint of bee!’ (bi:) But there’s quite a bit of variation in Australia, with some speakers more inclined toward iɪ or the more broad iə.
In the aforementioned accents, the monophthong is probably a possibility, but not as widespread as one might think. Yet in Southeast England, monophthongal ‘near’ does seem preferred by many speakers. A piece of anecdotal evidence: the International Dialects of English Archive transcribes the speech of two suburban Londoners who both have a monophthong or greatly weakened diphthong* for the final vowel in ‘idea‘ (usually the same vowel as ‘near’ in non-rhotic accents).
Yet as if in reaction to this trend, some urban Londoners have a strongly diphthongal ‘-ear’ vowel. Two celebrity examples I’ve noticed are Adele and Idris Elba, both with accents located near the Cockney/Estuary border. In words like ‘here,’ both have a very pronounced diphthong: where ‘ee-uh’ is a crude approximation in most cases, it’s actually fairly accurate here (iə).
It also seems possible for an accent to have a mixture of both types of pronunciations. My earlier impressions pertain to situations where the vowel appears in a prominent position. But I can see how a New Yorker might use a monophthong in the phrase ‘I feared the worst’ (i.e. with the vowel occurring in the middle of a phrase and before a consonant), but use a diphthong in the phrase ‘that’s a serious fear‘ (i.e. with the vowel at the end of a phrase). It’s just so variable.
For you non-rhotic speakers out there, do you use a monophthong, a diphthong, or both?
*Both are centralized as well, and the ‘weakened diphthong’ is of the type ɘə.
“For example, I’ve heard broad, Tony Soprano-esque impressions of the New York City accent with exaggerations like ‘whaddaya doin’ hih!’ (that last word is hɪ:, like ‘hit’ without the ‘t’). ”
You mean like this (listen to the word beard)?
Haha, yup. Pretty much EXACTLY like that.
As far as I can tell, I hardly ever give a monophthongal realization of the NEAR vowel. The only time I can imagine it happening is in allegro casual speech before prevocalic /r/, where something like [ˈfɪːrɪŋ] for “fearing” just might be a possibility. (Note that this is distinguished from the KIT vowel (as in “mirror”) by the length of the stressed syllable).
(Grew up in the English midlands speaking near-RP — lived in California for the last 15 years).
I should have added that I probably don’t qualify as a completely “non-rhotic” speaker any more, as a result of living in California, but I think this aspect of my speech has remained pretty consistent since I lived in England.
I think you can arguably still retain it, even if you have more a rhotic accent now. In my mind, there is a subtle difference between pronouncing ‘fear’ [fɪəɹ] and pronouncing it [fɪɚ].
/ˈniɚ/
/ˈfiɚ/
/ˈriɚ/
/ˈmirɚ/
/ˈmirɪkəl/
“mirror” and “miracle” are KIT words, not NEAR.
How would you pronounce “nearer”?
Although they’re arguably part of the NEAR set in most American accents. A vowel more or less can’t appear before /r/ in GenAm without resulting in NEAR, FORCE or SQUARE!
What about word/sir/bird, etc.?
And then there’s words like fire.
I’m with Ben on this one. For most vernacular American English accents (the rhotic ones), you get NEAR, FORCE, SQUARE, NURSE, lettER, poor (traditionally cure and poor and also you and ewe have different vowels).
“Fire” is KITE plus lettER
While “coor” is pronounceable just fine, “courier” slides into NURSE for me, as does “hurry”.
“Tower” is MOUTH plus lettER
Ellen K, he wasn’t exhaustive in his listing and certainly does not advocate “sir” as SQUARE or NEAR or FORCE.
Sooryan’s point is that for most dialects [iɚ] and [iɹ] and [ɪɚ] and [ɪɹ] are all allophones of NEAR /ɪər/ depending on whether it is followed by a vowel or not. “Nearer” and “mirror” are both analyzed as NEAR plus lettER.
Also, pre-rhotic vowels lose precision. NEAR can be [iɚ] or [ɪɚ]. “Poor” can be [ʊɚ] or [uɚ] (ignore the FORCE merger). SQUARE can be [ɛɚ] or [eɚ]. FORCE can be [oɚ] or [ɔɚ]. Of course, there are the losses of distinction which come from the marry-merry-Mary as SQUARE merger, but this is due to a re-syllabification, not due to a random TRAP-DRESS mergers as the “great” British phoneticians believe.
A similar thing happens with the pre-/ŋ/ vowels. I think most Americans analyze words like “bang” as /b/ plus “ang” and not as /b/ plus TRAP plus /ŋ/. Again, precision is lost (this time without a loss of distinction). Hence, /bæŋ/ as [bæŋ] or [bɛŋ] or [beɪŋ]. The only word Likewise “king” can be [kɪŋ] or [kiŋ].
I’d say that while GenAm style EFL should be taught with “R”-colored vowels, descriptions of the vernaculars should think of the “NG”-colored vowels as well. I never thought of “ban” and “bang” as having the same vowel for the same reason I never thought of “whet” and “where” as having the same vowel. This syllabification of vowel-ng as one unit is widespread among humans. |ŋ| as an isolated phoneme, a syllabic phoneme, and as a syllable onset is rare outside of notable exceptions like Vietnamese and Cantonese.
Gaelsano: I definitely agree with you about the r-colored vowels, but applying the same analysis to “NG” starts causing problems. You’ve started down a slippery slope. How about “G” (which affects some vowels before it—’bag’ and ‘dog’—in my mother’s midwestern accent). How about “N” with the pin/pen merger. How about “L”, which changes all sorts of vowels before it.
Gaelsano:
“A similar thing happens with the pre-/ŋ/ vowels. I think most Americans analyze words like “bang” as /b/ plus “ang” and not as /b/ plus TRAP plus /ŋ/. Again, precision is lost (this time without a loss of distinction).”
I don’t know about anyone else, but that’s not true for my own Midwestern American accent. I can’t make any distinction between /æ/ and /ɛ/ before /ŋ/. So for me strength [stɹeɪŋkθ] has the same vowel as bang [beɪŋ]. But this is all quite off topic anyway.
Gaelsano, you seem to have a different understanding of “can’t” than me. His statement sounds pretty absolute. And he leaves off NURSE/lettER, which you include. One might argue that’s a syllabic R. But, if one does, then one runs into words like “fire” as words with a vowel before R. In another syllable.
And it now occurs to me, you and him both also left out car/bar/star, etc.
Yes, there’s a limited amount of vowels with R. But 5, not 3. The four you list (I don’t see a difference between NURSE and lettER besides stress, 5 if you count those as different), plus CAR (or whatever the proper word to mark it is).
Gaelsano:
I never thought of “ban” and “bang” as having the same vowel for the same reason I never thought of “whet” and “where” as having the same vowel.
I never thought of them as having the same vowel for the same reason I don’t think of ban and bane as having the same vowel. I do think of bane and bang as having the same vowel. As does bank.
Yeah, there’s a loss of precision, as with the pre-R vowels. But it’s a different situation than with the pre-R vowels.
I personally pronounce “mirror” with a KIT vowel, but most people back at home don’t (they use NEAR). I also definitely pronounce “miracle” with the NEAR vowel.
“Mirror” is particularly interesting because a number of people I grew up with scrunched the second syllable to the point that the word is nearly homophonous with “mere.” In the Quest for Glory video game (old Sierra game, back when the business was in rural California), it has a rhyme that says “mirror with the leer.”
Or, of course, in the opening theme song to Fresh Prince: “I whistled for a cab and when it came near/The license plate said ‘Fresh’ and had a dice in the mirror.” I remember that rhyme always puzzling me until someone explained it.
(PS As an ardent fan of Sierra Online in its late-80’s heyday, I’m glad someone has referenced them here!)
Americans can compress syllable-final-/r/ with lettER as as in mirror to mere since “mirror” is /m/ plus NEAR plus lettER. Brits don’t have syllable final /r/ as [ɹ] or as a rhotic element to a vowel. They compress two syllable-start-/r/ to just one. “Library” for them /l/ plus KITE plus /b/ plus /r/ plus commA plus /r/ plus commA.
Both cases are caused by a universal distaste for having too many “r”s in a word. Americans may compress “mirror” to “mir” but sometimes “the mirror’s on the table” to “the mirra’s on the table” even in rhotic accents. I’m rhotic, yet I say gove’nor and su’prise.
The only thing to keep “r”s alive in America is the fact that we need it to distinguish homophones. Despite a distaste for the “r” we need to distinguish “hot” and “heart” and, for the cot-caught merged, to distinguish “shot” and “short”. The monophthongal nature of FACE and GOAT also is a road block. Americans do not have centring diphthongs (excepting old non-rhotic NYC). If “feared” were said non-rhotically as [fɛ:d] then many would assume “fed” [fɛd] or “fade” [fe:d]. For the cot-caught and horse-hoarse merged, “court” [kɔ:t] would be either “cot” [kɑ:t] or “coat” [ko:t].
Can’t believe that. I like my r’s just fine, and welcome their use anywhere. Heck, I pronounce the first r in “February.”
Gaelsano:
The only thing to keep “r”s alive in America is the fact that we need it to distinguish homophones.
I disagree. At least for my dialect. R (post-vocalic) can’t simply be dropped. What’s keeping R’s alive, it seems to me, is the lack of anything killing it.
And as far as mirror becoming mere, seems to me it’s not a case of losing an R, but of losing a rather minimal intervening vowel.
@gaelsano:
The only thing to keep “r”s alive in America is the fact that we need it to distinguish homophones
I’m afraid that explains both too much and too little. The COT/CAUGHT merger creates plenty of homophones, but that has not stopped it from gradually spreading in the US.
On the other hand, there are other distinctions, such as ʃ/ʒ, that seem robust even though they distinguish few if any minimal pairs (“Aleutian” vs. “allusion”, perhaps?)
I’m a little confused over your use of the term ‘broad’, which you’re obviously not using in the way it’s normally used in Australian sociophonology. Diphthogonal “beer” is not a broad speech marker in that sense. I would suggest (though this is untested) that pronunciations that tend toward the monothogonal are more likely to be found in the Sydney region than elsewhere.
I’m actually a bit confused over my use of ‘broad’ there as well 😉 Just a misleading, late-night choice of adjective, I’m afraid. Actually, if we’re talking about ‘Broad’ (as opposed to ‘broad’) Australian English, I would hazard to guess the monophthong would be more, rather than less, likely in open syllables. Very impressionistically speaking, that is: as you suggest, I’m not sure NEAR has ever been found to be a strong sociological marker in Australia.
I’ve actually read that monophthongal pronunciations of NEAR are more common in closed syllables in Australian English.
Looking at the list here, particularly in the North England column, it strikes me that it’s more a case of the word having 2 syllables rather than 1 diphthong, the second being of the lettER set.
Geoff Lindsey, a phonetician from UK, describes General British r-colored vowels as long monophthongs. He also claims that centering diphthongs are part of RP, which is obsolete.
In this case, NEAR vowel is: [ɪɪ]
The naturalness of British vowels
That’s not quite what I say, Pedro. NEAR, like WIRE and SOUR, is varisyllabic , an option that often gets neglected. This post is devoted to it.
I’m pretty sure I have both, and a full two-vowel version as well, depending on context. Can’t be sure of course. I’ll listen to people today and see if I can get some better idea.
I’m from Brighton in the south-east of England but have lived in inner south-east London for over twenty years, and my accent is pretty typical urban south-east English. Probably at the end further from RP, I have “innit?” and “weren’t it?” from childhood (though use them a lot less now) and glottalise every T and vocalise every L I can, but my vowels are probably nearer RP than London (and maybe influenced by not-quite extinct Sussex accents as well)
I don’t much like the name “estuary English” partly because the accent isn’t particularly associated with the Thames Estuary, partly because it implies that it is something unusual or novel or wrong, when its been quite normal where I come from since at least the 1930s, and its now the standard English of this part of the country. Most people I hear in London speak some version of it, its more common than either marked London accents or new-fangled MLE (I suspect that both the spread of MLE and its difference from other accents are journalistically exagerrated) – and they are both much more common than RP.
Speaking as an Estuary-speaking American (thank you, years of accent reduction), I normally use ɪ in near and fear, but ɪə in idea and Julia. Personally, I’d say iə is most common in the US, though, among non-rhotic speakers. My grandfather always pronounced here as hiə or hijə.