Re: THEORY: [i:]=[ij]? (was Re: Pronouncing "Boreanesia")

Adrian Morgan wrote:

>Kristian Jensen wrote, quoting myself:
-----<snip>-----
>Your post leaves me as confused as ever. Sorry. Three sounds:
>
>(1) The sound in school/cool/fool/tool is essentially [w]. Therefore, if
>    I take your description literally, it is [u].
Yes. At least in Brisbane.

>(2) The sound in could/book/wool/woman/ is similar to (1) and [w], but
>    looser. Some dialects in Britain (in rural England I think) are
>    caracaturised for their use of (AFAICT) this sound in words like
>    _mud_. Does that make it clear?
Yes again. At least in Brisbane. Note though that the difference between
(1) and (2) is that (1) is longer than (2). Similar but different.

>(3) The sound in moon/you/true/do/universe/suitcase is completely
>    different in every respect from (1), (2) and [w]. AFAICT it is
>    frequent in all dialects that I know of, including British, American
>    and Australian dialects. (I don't listen to American dialects
>    regularly.)
Yes its different from both (1) and (2) in ALL dialects. They would have
to be since all dialects basically have the same phonemic system, but the
system is phonetically realized differently across dialects. For instance;
the Australian vowel in (1) sounds a lot like the corresponding American
phonetic vowel for /u/. The phonetic vowel in (2), however, does not exist
in American. In American the [u] quality exists only in the quality heard
in American /u/, which is always long, not short.

-----<snip>-----
>> seems to be a shift of some sort such that the semivowel element of
>> long rounded vowels (/u/ and /o/) gets fronted while maintaining the
>> rounding. So /u/ and /o/ get rendered as [u-y] and [o-y] respectively
>> (compared with American [uw] and [ow]). For instance, words like
>> "no/know" get rendered as [no-y].
>
>I've heard visiting Americans claim that we tend to say 'noi' for 'no',
>but that's obviously just an overactive imagination on their part :-)
>although I'm willing to believe a Queenslander might say anything. Sounds
>like something a Queenslander might do :-) Best to ignore any such
>phenomena, though, when talking to a South Australian like me. Otherwise
>I'll only be confused.
Well, Queenslanders don't say *[noi]~[noj], but it may sound like that to
most Americans who would not be able recognize front-rounded vowels. What
I heard in Brisbane was [no-y].

>I'm quite certain my speech does not contain [o], although my singing
>voice does. When I sing, no/know are [no:]. When I speak, they contain a
>diphthong that begins with a neutralish vowel (perhaps [rounded-V]?) and
>ends with vowel #3 in my list above.
YES YES!! This neutralish vowel is [o-], a mid central rounded vowel. Its
like a rounded schwa but a bit more closed. And YES, it then ends with the
same vowel in (3) above indeed. Seems like Queenslanders aren't as foreign
to the rest of Australia as most would like them to be. ;)

-----<snip>-----
>> >2. The vowel in good/book/wool/woman is similar to [w] but a little
>> >   wider.
>>
>> In Brisbane, /U/ is articulated much more closed than in the US.
>> Phonetically, American /U/ is more mid-centralized (laxed) and less
>> rounded compared to Australian /U/. In fact, it seems to me that
>> Australian /U/ is quite the opposite of American /U/ -- more closed
>> and more rounded. Indeed, its articulation is quite a lot like [u].
>> Obviously, this makes Australian /U/ much closer to /w/ than
>> Australian /u/. This does not mean that Australian /U/ is tense and
>> long like /u/ however. Although Australian /U/ is more closed than
>> American /U/, it is by no means long like /u/. It functions just like
>> the /U/ found in other dialects of English.
>
>Doesn't help me a lot, because I can't picture an American voice (nor a
>Queenslander, for that matter). What are the phonetic symbols for the
>phonemes you've mentioned? Also, what do you mean by 'quite the opposite
>of'?
'Quite the opposite of' meant that while Australian /U/ is more closed and
more rounded, American /U/ is less closed and less rounded. Its difficult
to post IPA symbols via email. Nevertheless I have tried to list the
phonetic symbols for the phonemes I have mentioned for Queensland and
(Western) American English below:

     Phoneme   Australian    American
       /u/       [u-y]       [Uw]~[u:]
       /U/        [u]           [U]
       /o/       [o-y]         [ow]
     (where: [u-] and [o-] are centralized vowels)

As you can see, the American realizations of these phonemes resemble the
phonemic symbols a lot better than the Australian realizations.

>I guess I could do some research - email someone in the speech pathology
>department at the university and ask what the IPA is for the vowels in
>the words we've mentioned (I don't believe there are any phonetics
>experts in the language dept). Last time I sent an email that way I
>didn't get a response, but I may have more luck with a very specific
>question like that.
What you need is a phonetician. Also, all universities with a linguistics
department ought to have recordings of Daniel Jones reciting the cardinal
vowels. He was the one who invented the schematic representation of
cardinal vowels around 1917. Those recordings helped me a lot.

-----<snip>-----
>> I think more precisely, the tongue is more *retracted* in the latter.
>
>I really don't know about this at all. The back of the tongue is
>definately higher in the former _moon_, and the tongue is definately more
>relaxed in the latter _could_. Doesn't that make the former 'retracted'
>(i.e. the back of the tongue moves up to make way for the retraction)?
Certainly not. If there was retraction, the tongue would move back, not
up. There is no difference in height per se, at least in Brisbane. Both
of these vowels are closed vowels in Australian. The difference is that
_moon_ [u-y] has the tongue fronted while _could_ [u] has the tongue in
a non-fronted position.

-----<snip>-----
>> >The third I just don't get. I'm quite certain I've never heard _anyone_
>> >pronounce "woo" such that the {w} even remotely resembles the {oo}.
>>
>> Assuming that your Australian English dialect is not that different
>> from that I know from Brisbane, then it should be obvious.
>
>_Woo_ is [w] followed by the vowel in _moon_.
>
>Not in the least obvious.
Sure it is. I see you have cut out what I have written. Once again, the
vowel in _moon_ is fronted in Australian. And, once again, front-rounded
vowels do not resemble [w] at all that well. I don't know how much simpler
I can make that.

-kristian- 8)