Phonemic Transcriptions in dictionaries.

Jim   Friday, November 12, 2004, 02:44 GMT

As an Aussie I pronounce "wade" as [wæId] but I'd certainly think twice before suggesting that any ESL student try adopting my pronunciation for it could easily be misinterpreted as "wide". In California they may well pronounce "wade" as [wed] but, similarly, this could easily be misinterpreted as "wed" if used by an ESL student. Perhaps you too would be best thinking twice before you suggest that they try adopting a Californian pronunciation.

You consistantly harp on about "standard pronunciation". On every other post you remind us that AusE doesn't conform to your idea of what this "standard" is. "Rubbing it in our faces" wouldn't even be too strong a phrase. I ask again, though, "Why is it that you'd consider a Californian accent 'standard' whilst you deny the label to AusE and NZE?"

Whilst Aussies and Kiwis might speak of [getINlæId] a Colifornian would speak of [gEtINled]. Both of these pronunciations are on the extremes: most other accents are somewhere in between. I suggest that students be made aware of these extremes but I don't believe that it's advisable to recomend either. If [æI] for /ei/ is a regional varient, then [e] is no less.
Mi5 Mick   Friday, November 12, 2004, 02:46 GMT

This is closer to the English "i":!s.mp3

Do you hear "get" or "gate" in the following?

"As big as a partition as you can get [get]"
This would be transcribed as [get] not [gEt].
Mi5 Mick   Friday, November 12, 2004, 03:05 GMT
I guess the accent gap between American, British and Australian English is greater than we all think.
Mxsmanic   Friday, November 12, 2004, 05:52 GMT
I don't recommend a California pronunciation to my students; I recommend RP or GAE (and since I am American I can only model GAE for them). California pronunciation is essentially GAE, though, and it's not the only place where /e/ is pronounced [e]. A glance at a spectrogram shows that I pronounce it that way myself in my GAE, and I'm not from California.

No ESL student that I've had has misinterpreted [wed] as "wed." However, I've only taught students whose native languages make a phonemic distinction between [E] and [e]. My students hear [wed] as "wade," which is correct.

As I've said, I teach students to at least pronounce [e] for /e/, and preferably [eI] if they can manage it. They'll be understood either way.
Jim   Friday, November 12, 2004, 07:50 GMT
[e] is an allophone of /ei/ (as in "gate")*, Mxsmanic would have us believe, this is true it is an allophone of /ei/ ... on one extreme of the dialect spectrum, that is the detail that Mxsmanic would have us neglect. On the other extreme [e] is an allophone of /e/ (as in "get"). [e] he says "this is a widely-used allophone for /e/". This I very much doubt. In most dialects of English it is definitely a diphthong (especially in AusE and NZE).

"GAE uses [e] for /e/ in many cases," writes Mxsmanic "particularly in California." What the buggery ... ? What kind of fools do you take us for Mxsmanic? GAE is a convienent label for Midwestern US English. The Californian accent is not GAE. This statement of yours would be akin to my saying "RP uses [e] for /e/ (the vowel in "head") in many cases, particularly in New Zealand." A quick glance here

will show you that the Californian and Midwestern US accents are distinct.

Your students, Mxsmanic, hear [wed] as "wade" not because it is "correct" (which is very much isn't for most of us) but because you teach them that it's correct. Or perhaps it's because you actually pronounce it [we:d] or [we(j)d]. Anyway, I do wonder what you'd do with students whose native language made no phonemic distinction between [e] and [E]. What would you teach Japanese students, for example, who use only five different vowel positions?

"The misconception I'm seeing here is in the idea that Anglophones around the world make no distinction between /E/ and /e/." you write. Hold on a moment ... what are you trying to say, Mxsmanic? It seems clear enough to me that Mick has the ear to distinguish the phone [E] from the phone [e]. As for myself, did I not write "These are two different phones."?

Could you be saying that we claim that Anglophones around the world make no distinction between the phoneme /e/ and the phoneme /ei/? Surely none of us would make such an absurd mistake. Any three-year-old native speaker can tell the difference between "get" and "gate".
None of us are confusing [e] and [E] much less /e/ and /ei/.

"One can ask if the distinction reported by Paul is phonemic." One can ask whether this fellow is telling the truth at all. For all we know he may not be Welsh and he may not speak like this but what he writes is not complete phantasy.

"Likewise there are those who consistently distinguish such pairs as 'mane' vs. 'main' (Swansea Valley) or 'wait' vs. 'weight' (parts of the North of England)."

J. C. Wells

Either way you will perhaps notice that this Paul Waters claims to use [e:]; a long monophthong, not just [e]; for the vowel in "vane".

I could more easily recognise [e:] as an allophone of /ei/ than I could [e]. If someone were to pronounce "gate" and "wade" as [ge:t] and [we:d], then I believe I'd have a better chance of understanding them then if they'd said [get] and [wed]. I don't think I'd mistake their [e:] for /e/. This is because the length of the vowel would tell me that it couldn't be /e/. Of course there could be a chance of mistaking this [e:] for /e../.

Oh, but Mxsmanic also tells his students not to worry about vowel length either. "Don't worry about distinguishing long and short vowels. Don't worry about pronouncing /ei/ and /Ou/ as diphthongs. Don't worry about pronouncing anything at all as a diphthong except /au/, /oi/ and /ai/." It would seem to me that Mxsmanic's reliance on vowel position is exceptionally strong.

In the real world vowel position is not set in stone as Mxsmanic likes to believe. The phoneme /e/ ranges from [e] to [E]. The phoneme /ei/ ranges from [æI] to [e] (albeit [e:] in reality). Teaching pronunciations on one extreme is as well justified as teaching pronunciations on the other.

"These types of regional ambiguities are very dangerous to teach to ESL students." Therefore teach them to pronounce /ei/ as a diphthong rather than as a monophthong in the Californian style. But if they really can't manage a diphthong, do them a favour and stress that it should be a long vowel so as not to risk it's being mistaken for an /e/.

* I'm using Tom's ASCIIbet for phonemic transcription (between slashes).

Mxsmanic's /e/ is Tom's /ei/. Tom's /e/ is Mxsmanic's /E/. Were it up to me I'd want to use /e/ and /A/ but unlike Mxsmanic I don't feel the need to swim against the current. Whilst I'm at Antimoon I'll stick to Tom's alphabet. It helps to be speaking the same language.
Mi5 Mick   Friday, November 12, 2004, 08:48 GMT
Tom wrote:
>>Mxsmanic,....I would be immensely grateful if you would help me out of my ignorance and provide (references to) audio examples of the [e] sound.<<

[e], [E]
Mxsmanic   Friday, November 12, 2004, 12:29 GMT
Tom, see the IPA handbook site, specifically:

Look under Oral-vowels and listen to "greenhouse.wav" for [E] and "his-her-pl.wav" for [e].

Note that these two vowels are quite close, and note also that the exact positions of IPA vowels don't always correspond to the exact positions of vowels in actual languages. Normally one chooses the nearest IPA vowel, and as a general rule the difference is so small that it is not phonemically significant.

English /E/ is slightly closer to /e/ than the IPA position for /E/, but there is still a distinct, phonemic difference between the two, and /E/ is still the closest IPA equivalent. For GAE, the correspondence is nearly perfect, but RP references I've consulted claim a slight difference.

Looking at spectrograms, these two vowels are often hard to distinguish. In the examples you gave me, the first example seems to be a diphthong, and the next two examples, while I'd call them /E/, are not identical, nor are they as open as a classic IPA /E/.
Mxsmanic   Friday, November 12, 2004, 12:34 GMT
Let's see: In

I hear [gEIt], and I understand it to be the word "get." In

I hear [e] followed by [E]. The former sounds possibly a bit of a diphthong but I haven't looked at a spectrogram.

Incidentally, when I pronounce "gate," I get [get] in the spectrogram—no diphthong. The formants are flat across the board.
Mi5 Mick   Friday, November 12, 2004, 13:27 GMT
You're very good at your craft. Don't bother with a spectrogram... I agree with you for the most part; I just couldn't find any American speakers to fit my "Australo"-centric thinking and perception. I thought "get" above was the closest thing I could find but it didn't fool you. I'm not American afterall so I have to say you're right.

But Jim is right about AusE in that [e] and [E] are allophones; this vowel is short and pure in my accent. If [e:] and [E:] are lengthened versions, then this would represent the word "air"; there might be slight diphthongisation, but is negligible.

And for many (if not, most) New Zealands, [e] is typically the American equivalent of [E].

PS: "sick" and "seek" are pronounced differently in AusE, just not to your ears.
Tom   Friday, November 12, 2004, 17:02 GMT

I hear [gIEt], unlike Mxsmanic who hears [gEIt] (perhaps you mistyped Mxsmanic?)


Thanks for the sample, but I hear something that's a bit more open than the vowel in "seat". I hear no "e-ness" in it. In fact, I find it unbelieveble that someone could represent this sound with the [e] symbol.

Mi5 Mick,

What I meant was that the sound in "byk", "byt" and "pyza" is close to the vowel in "sit". The right symbol is probably the "crossed i".
Thanks for this. I can hear the first vowel is a sound between "seat" and "set". I can see how this could be an allophone of [eI].

From now on, this will be my reference recording for [e].
Mxsmanic   Friday, November 12, 2004, 21:33 GMT
Well, his-her-pl.wav comes from the IPA (it's the recording they transcribe in the handbook), so if they don't know what /e/ sounds like, who does?
Mi5 Mick   Friday, November 12, 2004, 23:38 GMT
>>I hear [gIEt], unlike Mxsmanic who hears [gEIt] (perhaps you mistyped Mxsmanic?)<<

Yes that's right Tom.


As Mxsmanic noted, in this is a slightly diphthonged [eI], so "his-her-pl.wav" is a better and purer representation of [e]; this is the one to use. I'd say the [e] in NZ English, is slightly more closed in words like "bet", "fed", etc.

The sound in "byk", "byt" and "pyza" is the closest thing I could find in Polish to [e].
Tom   Saturday, November 13, 2004, 00:14 GMT
This is crazy. Why would the IPA take a sound that sounds basically like [i] and represent it with [e]?

Anyway, I think the first vowel in your recording is very pure and not at all diphthongized.
Mxsmanic   Saturday, November 13, 2004, 06:23 GMT
The IPA is representing [e] as [e]. It's a very different vowel from [i]. (Incidentally, [I] is right between these two vowels.)

Looking at spectrograms, I see that the vowels in are pure. The "get" in is being pronounced [gjEt].
Mi5 Mick   Saturday, November 13, 2004, 13:14 GMT
>>Anyway, I think the first vowel in your recording is very pure and not at all diphthongized.<<

Well that's what I thought but I'm indecisive now that I find out Americans pronounce "gate" as [get]; I always thought it was a diphthong for them. A long [get] is how a Scot would pronounce "gate" and I've never heard (I thought) this from an American, so I've lost my bearings. I guess this is a legacy of the Scots.