IPA query

Easterner   Monday, October 11, 2004, 09:25 GMT
By the way, I think "r-dropping" is a similar phenomenon to the "s-dropping" in Old French, where e.g. "fenestre" became "fene^tre" in the process, with the resulting lenthening of the vowel. Today, nobody perceives there should be a /s/ in such words.
Tremmert   Monday, October 11, 2004, 10:21 GMT
Also from Afrikaans - 'aadvark'.
'can' and 'kin' are only minimal pairs in certain accents. I pronounce them with completely different vowels.
Mi5 Mick   Monday, October 11, 2004, 12:42 GMT
>>...it would be best to concentrate on vowel position, and disregard length, because that's how it works for the vast majority of English speakers.<<

No, it's a bad idea to disregard it, because vowel length is still key in any form of English. As Paul noted, it helps speakers to distinguish even those words of differing vowel positions: pill/peel, fill/feel, still/steel.
Easterner   Monday, October 11, 2004, 12:49 GMT

That's what minimal pairs are: words with just one phoneme differing (e.g. the vowel), the rest being the same. If two different words sound the same, they are called homophones. So I guess "can" and "kin" are minimal pairs in any dialect, because the only difference between them is the vowel in the middle, in whatever way it is pronounced.
Mi5 Mick   Monday, October 11, 2004, 13:07 GMT
Actually, pill/peel, fill/feel, still/steel do have the same vowel position in many accents, just not mine.
Mxsmanic   Monday, October 11, 2004, 18:33 GMT
Vowels in French are usually all of the same length, except for the last vowel in an utterance, which is slightly stressed and may be longer than the others. The spelling of the words in question has no influence on vowel length, nor does the presence or absence of a circumflex. "Fenêtre," pronounced in isolation, has the last vowel long only because that's the last vowel pronounced.

Furthermore, French, like English, does not make phonemic distinctions based on vowel length. Indeed, vowel length is often ignored in transcriptions and in discussions of French from a phonemic standpoint, just as in English.

Vowel length does not function to distinguish words in any standard English pronunciation. That's why English speakers often have trouble evaluating vowel length--they don't need it for their own language, so they've learned to ignore it.
Mi5 Mick   Tuesday, October 12, 2004, 02:09 GMT
Instinctively, English speakers often only evaluate still/steel, bin/been and pull/pool by length. Anyway, you were clearly wrong about AusE and unwilling to admit to it -- this adversely affects your other reasonings.

Easterner was saying in Old [or Middle] French, the circonflex served to eliminate consonants, which would consequently lengthen vowel sounds. Indeed in some French pronuncations today, this lengthening is preserved, though it doesn't have any phonemic weight. However in Quebecois French, it does to some extent, where not diphthongised, eg. maître (slightly more open but noticeably longer) and mettre/mètre, fête and fait, tête and tette.
Mxsmanic   Tuesday, October 12, 2004, 05:24 GMT
As I've said, English speakers do not instinctively evaluate vowel length; there is a very obvious difference in vowel quality which serves to mark the phonemic distinction.

Standard French ignores vowel length, also, as I have explained. I suppose there are villages somewhere where it matters, just as there may be villages where English vowel length matters, but in teaching students these languages, it's best to concentrate on what they millions of people speaking the standard versions say and understand, rather than on queer exceptions to the rules in isolated enclaves of speakers.
Mi5 Mick   Tuesday, October 12, 2004, 06:29 GMT
Well what you say isn't the be-all and end-all as you have been wrong before. You haven't quoted a single reference, and "As I've said" doesn't count for much either unless you want __ever-cyclic discussions__, so it's all up in the air.

Most natives speakers -- rightly or wrongly -- will often say the first thing the stands out in the above pairs is the length, so this evidently has a significant bearing on the way we speak.

I hardly call non-rhotic speaking areas enclaves, and I don't know how villages found their way into the discussion. Maybe you are originally from a village and this has some influence on your mentality.
Jim   Tuesday, October 12, 2004, 08:10 GMT
"I've read up on it, extensively," writes Mxsmanic "and vowel length is not important in English." Either he hasn't read extensively enough or he's just ignored anything that he didn't like. Vowel length is important in English. But what is he really saying? It seems that he has some fancy ideas of what it means to be "important in English."

By "important" he means "phonemic". What a sly misuse of words. There are things which are important in English which are not phonemic. If someone mumbles, what good is it to make all the right phonemic distinctions? Do I need to stress the point any further?

By "English" he means "standard English". He's also got his own exclusive ideas on what can count as standard. It should kept in mind that the narrower your definition of "standard English" the less synonymous it is with "English". Mxsmanic pushes for quite a narrow definition.

It would seem that Mxsmanic has talked himself into a rather snug little corner. "Vowel length is not phonemic in standard English," he insists "and pretending otherwise is also very misleading to ESL students." It seems that he's quite comfortable there and won't be talked out of it. However, I would not be happy to accept the consequences of this line of his.

What consequences are these of which I write? If vowel length is not phonemic in standard English then we have three options. These options are:

a) vowel length is not phonemic in Australian English,
b) Australian English is not standard or
c) both.

Mick and I have argued that vowel length is phonemic in Australian English. We have provided example after example of minimal pairs which show this. What's Mxsmanic's response? In short it is to brush such considerations aside and stick fast to his original shaky assertion. However, this must mean that

a) Mxsmanic is saying that we're wrong about the phonemic status of vowel length in AusE,
b) Mxsmanic is saying that AusE isn't standard or
c) both.

So what is he saying? This is a bit of a difficult question to answer. "There are many types of non-rhotic English," Mxsmanic writes "but only one or two (e.g., RP or Estuary English) can be considered standard." Why? I wonder. Why can only one or two of the non-rhotic dialects be considered standard? How many of the rhotic dialects is he willing to label "standard"?

It certainly seemed to me as if he'd been implying that AusE is not standard. What else would one conclude from a statement like this? So, in reply, I wrote "Thus Aus.E ain't `standard,' ay, Mxs? O.K." here's how he responded "No, it's just that vowel length isn't phonemic in AusE, either."

Okay, then Mxsmanic is not accepting that vowel lenght is phonemic in Australian English. He's saying that we're wrong about the phonemic status of vowel length in AusE. Right? But what about all those nice minimal pairs we've mentioned? Let's take another them. I'm going to divide Mick's list into four:

"DEAD" verses "DARED"
stead, stared
fed, faired
bed, bared

"BID" verses "BEARD"
biz/bus., beers
his, hears
piss, pierce
quiz, queers

"HUT" verses "HEART"
bark, buck
bars, buzz
barn, bun
charm, chum
darl, dull

"SOME" verses "PSALM"
come, calm
bum, balm
sum, pslam

Now, I only mentioned "dead" vs. "dared" and "some"/sum" vs. "psalm" but I backed my claim up with an AusE vowel chart that I found on a Macquarie University website. How does Mxsmanic respond? "As for university linguists, they are welcome to come here and discuss the matter if they so choose." we all know well enough that they've probably got better things to do than come here and join in in this debate. That's why we're bringing them to you, Mxsmanic, that's why we're quoting them. Their observations are no less worthwhile if they don't happen to write up a post on this thread.

Although I'd only mentioned "dead" vs. "dared" and "some"/sum" vs. "psalm" the rest of Mick's list rings true for me. What does Mxsmanic say to this? "That's not what 'phonemic' means. A phonemic difference is one that serves to distinguish between utterances with different meanings." he writes "In the examples you give, there are other distinctions among the words besides vowel length; therefore vowel length is not phonemic."

Other distinctions, ay, Mxsmanic? Tell us what they are. I tell you that in the pronunciation of many Australians there are no such distinctions of any siginificance. Pronouncing the "DARED" words with the long monophthong, [e:], is common in Australian English. It is also not uncommon for the "BEARD" words to be pronounced with the long monophthong, [I:].

"In RP, his is pronounced /hIz/ and hears is pronounced /hi@z/; vowel length is unimportant." Mxsmanic writes "Dead is pronounced /dEd/ and dared is pronounced /dE@d/."

I certainly accept that in RP these phonemes would be centring diphthongs not monophthongs but it seems that these centring diphthongs are fading away from Australian English. There may be some Australians who still use one or even both of those centring diphthongs but even for them I'm sure that [bI:d] and [de:d] would be interpreted as "beard" and "dared" not "bid" nor "dead".

Is this unique to Australian English? I don't think so. I get the feeling that this is just part of a general trend in non-rhotic accents. I think I've read something or other to this effect but I can't seem to put my finger on it now. I'll do some digging around.

Anyway, the "BEARD" and "DARED" words aside, we still have "HUT" verses "HEART" and "SOME" verses "PSALM". Australians pronounce the "HUT" and the "SOME" words with the short vowel, [â]. We also pronounce the "HEART" and the "PSALM" words with long monophthong, [â:]. Any difference in position of these vowels in Australian English is negligible. Mxsmanic wants to talk about "other distinctions". What are they? There are none of siginificance in the Aussie accent for these minimal pairs. Note: because we can't use the IPA here, I'm using the symbols I introduced on page one.



Where symbols appear in pairs, the one to the right
represents a rounded vowel

"I rather doubt that all 20 million make vowel length phonemic regularly, or even at all." writes Mxsmanic. Let me assure you: we do. There may be some variation in the "DARED" and the "BEARD" words but when it comes to "SOME"/"HUT" verses "PSALM"/"HEART" it's a very different story. Yes, I'd definitely say that all 20 million of us make vowel length phonemic regularly. Many Aussies might use a centring diphthong for the "DARED" words. Perhaps most still use one for the "BEARD" words. But for all of us the distinction between the "SOME"/"HUT" words and the "PSALM"/"HEART" words is almost entirely one of length.

Now for a rather more uncommon distinction. "Many southern English make a longer vowel in 'bad' than in 'lad' and may even have minimal pairs between the name 'Sally' and the verb 'to sally', or between 'shandy' the long drink and 'brandy' the short drink;"

J.C. Wells

Let me be clear about what I mean by "uncommon" here. This distinction is uncommon in terms of the number of minimal pairs it forms but, acccording to Wells, it occurs in southern English, it also occurs in Australian English and, I believe, in some dialects of American English. I can only think of two minimal pairs which exhibit this distinction in my accent. Let's consider this sentence: "Anne can put an egg in a can." Here's how I pronounce it: [æ:nkænpUtæneginêkæ:n]. I don't think that it's worth bothering ESL beginners with this one though.

Anyway, I believe that any reasonable person would concede that Mick and I had shown that vowel length is phonemic in Australian English inspite of what Mxsmanic might have to say. By "had" I mean even before Mick quoted the article by Felicity Cox, a linguist at Macquarie University. It seems that even Mxsmanic was somewhat convinced by this. "I'll keep your information in mind should I ever wish to learn to talk like an Australian." he wrote. Yeah, well thanks for the condescending tone, Mxsmanic, but thanks even more for showing us what you really think. He continues "In the meantime, I'll stick to standard pronunciations of English."

So there you go: according to Mxsmanic, Australian English is not "standard". Therefore we can simply ignore any noise from Down-Under if it rocks our little boat. Presumably Mxsmanic would not be willing to admit that his accent is not standard.

I might pronounce "hair" and "hear" as /h/ plus a monophthong. Mxsmanic says "hay" and "hoe" can be pronounced as /h/ plus a monophthong. He'll insist that I'd be using a non-standard pronunciation. Will he admit that he'd be using a non-standard pronunciation to?

If you ask me, suggesting that the dialect spoken by several millions of educated native English speakers is not standard is laughably absurd. If you're telling me that the mid-Western US newscaster's accent is "standard" whilst the Aussie accent is not, it's just as absurd and somewhat insulting at the same time. It's a sorry thing that there are Americans who think this way. It's folk like those who are giving the USA its bad name.

Do they believe that they've got hundreds of millions on their side? What if they did? I doubt it though. There is a great deal of dialect variation across North America. Anyone who doesn't realise this probably shouldn't be teaching English. The variation across AusE is small by comparison. Not all Americans speak this so-called "General American".

To single out one dialect or even a well-defined few and label it as "standard" is nothing but elitism. I don't believe in such notions of "standard English". Anyway enough of this.

Even if vowel length were not phonemic in English, does this mean it is unimportant? Have a look at this "fifty" verses "fifteen". Get the lenght wrong and ... "But the /n/" you cry "the /n/!". Sure they are not a minimal pair but they're close. If the you're in a situation where that /n/ could easily go astray, these can easily be confused unless you make the length distinction.

If Mxsmainc tells his students to pronounce "fifteen" as [fiftin], I feel sorry for them. If he says to pronounce "fool" as [ful] (IPA), I feel sorry for them. It's very risky to tell students not to worry about vowel length. Even in dialects where it might not be technically "phonemic" (according to Mxsmanic's defintion) it is still important. Get the length wrong and you might just strike it lucky and be understood. But you might not.

It is students who are told not to worry about vowel length I feel sorry for. I don't accept Mxsmanic's nonsense that it's better not to mention length. Length is one of the things which distinguishes one phoneme from another. Position is not the be all and end all. Indeed phonemes can move around as you go from dialect to dialect (which many ESL students do). When you're faced with shifting vowel positions, you can rely on lenght ... unless Mxsmanic was your teacher.

Vowel length may not be phoemic in many pronunciation of English but it is important. Students who refuse to recognise this or who've been told to ignore it risk having trouble making themselves understood and understanding others.
Mxsmanic   Tuesday, October 12, 2004, 18:58 GMT
I have recordings and transcripts of Australian speakers in front of me, and vowel length is not phonemic in their speech.

In any case, ESL students want to learn English that can be understood everywhere, and any pronunciation of English that attempts to make vowel length phonemic is not in this category. I teach them American English; my British colleagues teach them clean standard varieties of British English, such as RP.

Most native speakers of English are incapable of evaluating vowel length on demand; they must be taught to listen for it. This is because vowel length is not phonemic.
Smith   Tuesday, October 12, 2004, 21:53 GMT
What about the Australian ''steak'' vs. ''stake'' distinction? Is that difference only in vowel length. I've heard by Dave (an Australian on the other thread) that it is.

Here's a link to that thread,

Jim   Tuesday, October 12, 2004, 23:31 GMT
It's a difference only in the imagination of Dave.
Mi5 Mick   Tuesday, October 12, 2004, 23:38 GMT
Contradiction: You have accepted in the past that vowel length alone distinguishes "merry" from "Mary" including American pronuncations. Work your way up from that idea; it's not hard... you can do it!

As Jim rightly wrote:
"Vowel length may not be phonemic in many pronunciations of English but it is important."
"Position is not the be all and end all. Indeed phonemes can move around as you go from dialect to dialect (which many ESL students do). When you're faced with shifting vowel positions, you can rely on length..." -- excellent point! It's misleading to claim that vowel length is trivial.

PS: Smith: It is.
Smith   Tuesday, October 12, 2004, 23:46 GMT
''It's a difference only in the imagination of Dave.''

Jim, If it's a difference only in the imagination of Dave then why does Steve say that he makes the same distinction?

Quote-''I'm an Australian aswell, and like Dave, I pronouce those words above with different distictions. I also say pure and fewer as if they rhyme.''

Is that ''fewer'' and ''pure'' rhyme in Australia only in the imagination of Dave too?