In the examples of Australian English I've consulted, the transcriptions were not my own, so any difficulty I might have in distinguishing vowel length would not be a factor. Additionally, these were narrow phonetic transcriptions, in which vowel length was marked even though it is not phonemic.
There may well be variations of English in Australia in which vowel length has some phonemic importance, but none of the standard accents I've heard seems to use vowel length at all for this purpose. Indeed, Australian English is often closer to American English than British English; many portions of utterances in Australian English sound entirely American.
There are a few regional accents that seem to be exceptions to this; I find them terribly unpleasant to the ear, and so do a number of Australians with whom I've discussed this. I suppose they are the Australian equivalents of New York or Cockney accents (indeed, some of them sound like drunken Cockney, as I recall). I certainly would never suggest that an ESL student come within a kilometre of such an accent, nor would I consider such regional deviations to be standard in any way. Perhaps these variants also make use of vowel length phonemically.
Keep this in mind: Any variant of English that makes vowel length phonemic will be largely unintelligible to the rest of the English-speaking world, because standard versions of English ignore vowel length and depend on vowel quality instead. It's quite a dramatic difference and it will make it difficult or impossible for different English speakers to communicate to the extent it is used. Failing to distinguish the vowel position in "pin" and "pen" will make these words unintelligible to speakers of standard English, no matter what vowel lengths are used or what phonemic value is given to those vowel lengths by the speaker.
This afternoon I made a few phone calls to discuss these issues with two linguists of The University of Queensland and two of Macquarie Uni., NSW. They were very helpful and went into great detail to explain the importance of vowel length for intelligibility in non-rhotic English and its phonemic nature in AusE -- hence their papers which discuss phonemic vowel length. If you like, I can email you their contact details and you can personally tell them how they've got it all wrong, with your accomplished knowledge of AusE.
As for variations in AusE, Australian pronunciation is very uniform; regional accents have relatively little overall influence on common pronunciations.
You're likely to found more hideous, drunken sounding accents among American Midwesterns than you are in Australian accents. Anyway, New Yorker sounds more eloquent than any generic Midwesterner accent I've heard. But that could be because Midwesterner gives an impression of blandness.
Keep "merry" and "Mary" in mind... Are you merry today or Mary?
*found-> find
As for your impressions of it sounding closer to American English, well actually, Australian English is far closer to Cockney.
I've done some research of my own, and the overwhelming majority of sources I consulted confirmed that vowel length is not phonemic in English; indeed, one of the key differences between modern English and older forms of English is that the older forms made vowel length phonemic, whereas it is not today.
I've also noticed that many people who should know better continue to confuse vowel length and vowel position or quality in their work. This is probably encouraged by the widespread but incorrect use of certain IPA symbols as I have previously described (/i/ and /i:/ incorrectly used for /I/ and /i/, respectively, etc.). Monolingual English speakers in particular often are firmly convinced that vowel length is phonemic, because they were taught in school that the vowel in "seen" is a "long" vowel and that the vowel in "sit" is a "short" vowel, when in fact the key difference between these vowels is position, not length. I remember being taught that myself in second or third grade, and it wasn't until I studied phonetics much later that I discovered that it was a serious misrepresentation of reality.
That's interesting how so many sources get phonemic vowel length so wrong, but not so strange, because this was the case when I researched the (supposed) presence of diphthongs in French. The majority of the sources written in English, falsely refer to French monphthong-semivowel segments as "diphthongs", along with examples. Conversely, most of the French sources get it right! (ie. diphthongisation doesn't exist in French.) It's best to go straight to the horse's mouth.
Mxsmanic,
You're asking us to keep something in mind. Sure, I'll keep it in mind as long as this debate continues but when it's done it'll be put where it belongs. I mean, it's best not to keep things in mind if they are just untrue.
"Any variant of English that makes vowel length phonemic will be largely unintelligible to the rest of the English-speaking world," this is a false conclusion. Here's the false premiss upon which it is based "because standard versions of English ignore vowel length and depend on vowel quality instead."
The speakers of most versions of English (whether you've chosen to grace them with the "standard" label or not) systematically pronounce certain vowels long and others short. There may be a degree of variation as to which vowels are long monophthongs, short monophthongs and diphthongs but there is also a significant degree of agreement across the board. (Tom's) /a:/ tends to be long whilst (Tom's) /^/ tends to be short in almost every accent (including RP & midwestern US).
Certain dialects may not use vowel length phonemically (in the strict sense) but this doesn't imply that vowel length is ignored. You might not believe that you're able to distinguish between long and short monophthongs. You might think that most English speakers are in the same boat. However, if you were right, why then would almost all English speakers use a long monophthong for (Tom's) /a:, u:, e:/ and /i:/ whilst they use a short one for (Tom's) /^, u, e/ and /i/?
I propose that even for those who do not use vowel length phonemically, they'd be able to understand someone who does. I suggest that if I said [sâm] (using the symbols on page one), you'd know it was either "some" or "sum", and if I said [sâ:m] you'd know that it was "psalm". You might even think you percieve a difference in timbre between the vowel (even though it wouldn't exist).
"It's quite a dramatic difference" you write. My point is that it is not a dramatic difference at all. Indeed it is so far from being dramatic that you've not even noticed it. The truth of the matter is that it is both duration and timbre that are used to distinguish one vowel from the next. In many accents you can get by just on the position of the vowel ... as long as you stick to that accent but go into a different accent and you'll find that the position has changed.
"There may well be variations of English in Australia in which vowel length has some phonemic importance, but none of the standard accents I've heard seems to use vowel length at all for this purpose." you write.
"There are a few regional accents that seem to be exceptions to this;" you write "Perhaps these variants also make use of vowel length phonemically."
The study that F. Cox has done, to which Mick posted the link, was done around the area where I grew up (in the north-east of Sydney). Her study confirms that vowel length is phonemic in AusE. A regional accent? Sure, it's a region of the largest city in the country.
<<Are you merry today or Mary?>>
Speaking personally on this very dark, chilly, not yet light gloomy Scottish October morning...the former....sort of.....full consciousness not yet arrived but a mega strong tea and a quick peep into Antimoon is a remedy for that.
The latter....erm....nope, not at all....I just looked in the mirror. Definitely more of a Damian.
You know.... With that blonde hair and all. You could be mistaken for a ...
The auxillary verb 'can' and the noun 'can' certainly form a minimal pair where vowel length is phonemic in my pronunciation...
"fifty notes" vs. "fifteen notes" -- it's not just about length, there is a difference in stress ("teen" is stressed in the second phrase).
But how about:
"I was late twice" vs. "I was laid twice"?
http://www.antimoon.com/temp/late_laid.mp3
Isn't vowel length (/ei/) phonemic in this case (if you pronounce it fast as in the recording)?
LOL, late twice/laid twice.... True.
How about this mister Mxsmanic,
as you might know /@/ become /e/ before /n/ (of course i'm talking about american English) this has something to do with the fact that when you pronounce the nasal /n/ no air comes out your mouth, it comes out thru your nose instead, even american themselves will say that this is false, well ,
this is not what's being discussed anyway, but if you ask those concious that their pronounce pan and pen the same , they will tell you that there's a tiny difference, vowel lenght,
The is /e/ sound in pen is shorter than the /e/ sound in pan, just like the /e/ in den is shorter than the e in /dan/.
>>The /e/ sound in pen is shorter than the /e/ sound in pan, just like the /e/ in den is shorter than the e in /dan/.<<
A southern US accent?
On the subject of late/laid, the only way I distinguish "lust" from "last" is length.
Me too.
"lust" = /l^st/ = [lâst]
"last" = /la:st/ = [lâ:st]
(Phonetic transcription mine, phonemic transcription Tom's)
This is much the same as the "some" verses "psalm" bunch of word pairs. As for their being differentiated by position in other dialects, I'd been blissfully unaware of this until quite recently. I didn't even notice it when I lived in Canada. I guess I just kept relying on vowel length to distinguish this pair of vowels like I'd done back home. Nor do I recall ever being misunderstood because of my pronouncing these two vowels with the same timbre.
