Re: THEORY: Ray on ambisyllabicity

Adrian Morgan:
> And Rosta wrote, quoting myself:
>
> > > No non-Eastern-stater would be caught *dead* saying 'skUl' for school or
> > > 'kUl' for cool, both of which are common in Sydney, for example. For
> > > non-easterners, [Ul] can *only* occur if the [l] begins a new syllable. I
> > > don't know the IPA for the vowel the rest of us use.
> >
> > I'm a bit confused. [U] is (close to) the vowel you have in _book_
>
> Is it ???
> Don't tell me I've got [u] and [U] the wrong way around!
> I was absolutely convinced it was [u] in _book_, [U] in _boot_.
> <very confused>
The vowel phoneme in _book_ is conventionally represented with /U/ and
that in _boot_ with /u:/, but there may be different local conventions
prevailing in Australian universities, quite possibly as a symptom of the
rejection of the Cultural Cringe. The vowel in _book_ in Aus E is realized
with [o], i.e. sort of midway between [U] and [u], but I think [u] is
also a possible Aus E realization. The vowel in _boot_ in Aus E is realized
as something that ends in a high frontish vowel that may or may not be
rounded, and can be diphthongal ending on the same phone and starting on
something frontish and lower. I don't hear enough Aus E to have a sense of
what's the norm.

The /U/ and /u:/ symbolizations, btw, reflect conservative RP and General
American realizations fairly well, I think.

> > Changing topic, can you tell me whether _chance_ has the vowel in _hat_
> > or the vowel in _bra/start/grass_?
>
> As a general rule:
> Eastern states --- the former [&]
> Other states -- the latter [a:]
Where is Bob Hawke from? I heard him recently say _chance_ with [&:], and
was wondering whether this is because his accent has /&/ (the HAT vowel
phoneme) in _chance_ or because it has /A:/ (the BRA vowel phoneme) but
with the realization of this phoneme shifting from [a:] to [&:] either
across the board or in certain environments. I hadn't noticed this before
in Australians of my acquaintance (who would have [a:] in _chance_), most
of whom are middle class from NSW.

> And Rosta wrote, quoting myself:
>
> > > I have a hard time telling [Ou] from [Ow]
> >
> > One can safely treat them as equivalent, either because they are, or
> > because the difference between them is empirically insignificant.
>
> Now, hang on.
>
> Near the bottom of your message, which I answered first, you tell me that
> I've had [u]/[U] confused all this time and that my belief that _book_ is
> [buk] and _boot_ is [bUt] is incorrect.
I wonder whether you're confusing phonetic representation with phonemic
representation? /bUk/ and /bu:t/ are the norms I'm aware of for phonemic
representation. For phonetic representation of Aus E, [bUt] seems dead
wrong, but [buk] seems in the right ballpark, tho I'd prefer [bok].

> However, here you tell me that it's perfectly acceptable to confuse [u]
> with [w].
[w] is basically defined as a [u] with intrinsically short duration.

> Now [w] is very similar to the vowel in _book_ (except that the lips are
> slightly tighter) and not even remotely similar to the vowel in _boot_.
I agree.

> If [w] is a close phonetic relation of anything, it would have to be the
> vowel in _book_, which is indeed what I had down as [u].
I agree.

> In other words, I'm getting mixed messages here.
I tentatively -- and with no disrespect and with every desire not to be
thought to sound patronizing -- reckon you've a very good ear, but have
made the common mistake of assuming that standard symbolizations of
phonemes accurately represent the phones that realize the phonemes. IOW,
contrary to what a sensible but not well-informed person would think,
the phonemes commonly represented as /U/ and /u:/ are in many accents not
realized by phones [U] and [u:], and indeed sometimes the phone denoted by
the symbol is *totally* different from the realization of the phoneme
denoted by the same symbol.

If you have made this mistake then you're in good company, since 99% of
undergraduates make it too, and accordingly end up talking total bollocks
(unless one mentally changes their brackets to whatever makes more
sense).

> > > /oU/ as in _ode_, _mode_. The initial vowel could be phonetically
> > > equivalent to a wide range of things - among them [a], [@], [o] - but
> > > I've picked up that it's traditional to use /oU/ for diphthongs of
> > > this type.
>
> Make this, "Neutral vowel, perhaps in the vicinity of rounded-[V] but
> phonemically best transcribed /a/ or /@/, followed by the {oo} in
> _boot_".
I agree completely. I think it would be too confusing to try to represent
this in ascii.

> > I've not seen "/oU/", and it's not very mnemonic for the Aus vowel. I
> > really don't believe it ever begins with [o].
>
> I know it doesn't! However it doesn't quite begin with [a] or [@],
> either. The advantage of using /a/ or /@/ would be that these already
> exist in the phoneme inventory.
In the British tradition, but unlike the American tradition, diphthongs
(and long vowels) are treated as single phonemes, not as sequences of two
phonemes. So /@U/, for example, is a representation of the single phoneme
/@U/ and not a sequence of the phonemes /@/ and /U/. The evidence for which
is the better tradition is mixed. In terms of phonetic realization, diphthongs
(and long vowels) certainly look like independent phonemes and not phoneme
sequences, but systemically they do to some extent lend themselves to being
analysed as sequences of double short vowels or of short vowel + /w/ or /j/
(/y/) or /@/. (NB This is only English we're talking about, here.)

> Do we understand each other yet?
When you describe the actual sound I do understand you and agree with you.
Your use of symbols is mildly flummoxing.

Phew! I don't think I'll be able to keep this delurk mode up for much
longer...

--And.