-eer vowel (accent/dialect variation?)




hero, cheer, fear, searing, here, ...

Most dictionaries will cough up /ɪə/ or even short I for this vowel.

Unfortunately I cannot hear such a thing. Instead I hear /i/.

I am a native speaker (North American), and I also asked two other natives on what they hear. They said pretty much the same thing. I also compared that /ɪə/ sound to what I heard from other natives, and what they said either did not match up, or that my ears suck.

Is this an accent or dialect variation? Or that my ears suck at picking it up and replicating it?


Posted 2015-10-29T01:14:53.677

Reputation: 4 502

4No, you’re right: you hear /ir/. You have to read pronunciations for rhotic speakers. – tchrist – 2015-10-29T01:29:52.417

Cambridge, Merriam-Webster and Oxford Learners cough up short I for AmE. Wiktionary is all over the place, with some AmE pronunciations with /i/ and some with short I and then short I + schwa. Dictionary.com gives the short I + schwa diphthong, even though its pronunciation audio actually uses /i/! – Nihilist_Frost – 2015-10-29T01:39:16.970

1To me, I have the same vowel in the word he as I have in the word hear, and that is not a short i, but an /i/. – tchrist – 2015-10-29T01:46:45.500

Most dictionaries for native speakers of AmE use schemes other than IPA. I recommend Macmillan Dictionary, which is freely available online and has fairly good IPA transcriptions for both AmE and BrE along with audio. hero, cheer, fear, searing, here, . . .

– snailplane – 2015-10-29T01:46:51.250

1However Macmillan's transcribes it as short I... and the audio sounds like /i/. – Nihilist_Frost – 2015-10-29T02:02:07.197

Your ear is playing tricks with you! Imagine that there's a "sea-ring" or an "e-ring" then go to the entry for earring or searing. You'll hear that the vowel there is quite different. it's more open. Or compare the name of the letter, "e", with the vowel you get in "ear"!

– Araucaria - Not here any more. – 2015-10-29T09:08:31.493

@Araucaria: that adds syllabification effects, which can exist in many other cases: I can also also differentiate "spa-ring" and "sparring" but that doesn't mean they have to be analyzed with different vowel qualities. – sumelic – 2015-10-29T23:50:56.920

@sumelic No but them being differentiable doesn't mean that we somehow become completely deaf and unable to differentiate between the quality of the two vowels presumably? My elementary students can do it, so I'm supposing that most other people can too ... ; ) – Araucaria - Not here any more. – 2015-10-30T00:00:14.547

@sumelic However, Snailboat makes good points below. I'd definitely say that the audio for the disyllabic words is significantly nearer to /ɪ/ than /i/, but then that may be my ear playing tricks on me. Nonetheless, it's clearly different in quality to /i/, even if it's not kit. – Araucaria - Not here any more. – 2015-10-30T00:06:51.543

1@Araucaria: it's clearly different in your dialect, but not for me (or apparently for tchrist, or Nihilist_Frost...) – sumelic – 2015-10-30T00:08:02.300

@sumelic I'm wondering now if perhaps your /i/ is slightly lower than ours, which is a possibility that hadn't occurred to me. That might be why it clearly sounds like kit to me while it seems to be the same vowel for you guys that you use in words like (presumably stressed) he and be (thinking of hear and beer) ... Because I'm making my judgements off the Gen Am dictionary audio provided by Nihilistfrost ... – Araucaria - Not here any more. – 2015-10-30T00:13:17.770

@sumelic Sorry, I mean provided by Snail above. – Araucaria - Not here any more. – 2015-10-30T00:33:28.050

2@Araucaria Judging by the charts on p.90 of A Course in Phonetics (Ladefoged & Johnson 2011), the average AmE /i/ is only subtly lower than BrE /i/, while the AmE /ɪ/ is rather noticeably lower than BrE /ɪ/. So perhaps the dividing line between the two is a bit lower in AmE, and perhaps BrE speakers would tend to judge some sounds in the middle as /ɪ/ when AmE speakers would judge the same sounds as /i/. (The varieties the charts are supposed to represent are "Standard American Newscaster English typical of many Midwestern speakers" and "British English as spoken by BBC newscasters".) – snailplane – 2015-10-30T00:35:10.190

@snailboat That would explain a lot a lot (sic)! Was trying, unsuccessfully to find some info about fleece, hadn't occurred to me that kit might be different too! Thanks. (oh and plus 1 of course for the very good answer ..) – Araucaria - Not here any more. – 2015-10-30T00:40:12.637



There's no phonemic contrast between /i/ and /ɪ/ in this context, and in fact the phonetic realization varies between the two and tends to lie somewhere in between:

In a syllable closed by /r/, there is no contrast in quality between a tense vowel and the lax vowel nearest to it. Consequently, as often happens in contexts in which there is no opposition between two sounds, the actual sound produced is somewhere between the two.

(A Course in Phonetics, 6th ed., Ladefoged & Johnson, p.99)

Since the contrast is neutralized and the vowel tends to lie between the two, we could reasonably indicate this sequence with either /i/ or /ɪ/. The traditional choice is /ɪ/, and that's what you'll find in most dictionaries that give AmE pronunciation in IPA.

But is that the best choice? Clark and Hillenbrand published a study in the Journal of the IPA in 2003 called Quality of American English front vowels before /r/. Although they did find that the NEAR vowel tends to lie somewhere between /i/ and /ɪ/, they found that overall it tended to be closer to /i/:

One major finding of the study is the stronger similarity of the high front vowels of hear and beer to /i/ than to /ɪ/. This similarity was evident in the acoustic measurements shown in the F1–F2 plots, in the discriminant analyses, and in the listening tests. If transcription is to accurately capture phonetic quality, [i] is a better choice than [ɪ] for representing the initial part of this diphthong, at least for this dialect most of the time.

(p.13, emphasis added)

But they also wrote:

On the other hand, the tendency of these vowels to fall in the /i/ territory is not absolute. Figures 1 and 2 show that many of the tokens lie intermediate between /i/ and /ɪ/ (consistent with the notion that the phonetic quality of the initial part of the diphthong is intermediate) and, as noted in the description of the listening results, several of the tokens excised from the diphthongs ending in [ɚ] were more strongly heard as /ɪ/, more so for hear than for beer, and more so for the men than the women talkers.

(p.13, emphasis added)

So although the convention is to transcribe this sequence with /ɪ/, you could make a reasonable argument that it's better transcribed /i/. Either way, there isn't a clear-cut choice between the two, so the choice of transcription comes down to a matter of convention.


Posted 2015-10-29T01:14:53.677

Reputation: 30 097