On Wed, 21 Jul 2004 14:36:09 +0200, Andreas Johansson <[log in to unmask]> wrote:

>Quoting "J. 'Mach' Wust" <[log in to unmask]>:
>> Andreas Johansson wrote:
>> >Quoting "J. 'Mach' Wust" <[log in to unmask]>:
>> >>German [2] merges with [Y], being rather [2_r] than plain [2].
>> [snip]
>> >do you mean that German has merged /Y/ and /2/? As in, _möchte_ and
>> >_Früchte_ rhymes? That would certainly not conform to my experience, nor
>> >has a such phenomenon been mentioned in any of the phonological texts on
>> >German I've read.
>> _möchte_ has /9/ and is clearly distinct from /Y/. Between /2/ and /Y/,
>> however, there's no significant difference of quality. Compare _rüsten_
>> [rYstn] 'to set up' and _rösten_ [r2_r:stn] 'to roast': the distinction
>> is in the quantity.
>Quoting Philip Newton <[log in to unmask]>:
>> On Tue, 20 Jul 2004 23:03:18 +0200, Andreas Johansson <[log in to unmask]>
>> > Quoting Philip Newton <[log in to unmask]>:
>> >
>> > > "möchte" has /9/, not /2/ - perhaps "mögen" and "Mücken" is a closer
>> > > pair to exemplify /2/ vs. /Y/ in the standard language.
>> >
>> > We would seem to be using different phonematization schemes -
>> Quite possibly. Phonemic transcription, as opposed to phonetic, is
>> always language-specific. I never learned a specific one for German so
>> I use an ad-hoc notation.
>I think we all perfectly well know that the symbols chosen to designate
>phonemes are largely arbitrary.
>What I still do not understand is J. 'Mach' Wust's original claim that
>[O2] and [OY] are the same. This is not an issue of phonemics; the
>difference is objectively physically measurable.
>Strictly speaking, his subsequent claim that [2] and [Y] have merged in
>German is equally unintelligible; based on his later post, he meant that
>the chief distinction between the phonemes I'd indicate as /y/ and /2/ are
>one of length, not one of quality.

I'd go as far as claiming that their quality's identical.

>I therefore suspect that the claim re:
>the diphthongs too was supposed to refer phonemes (presumably, then,
>prompted by a misunderstanding of my earlier post about phonetic
>realizations of the 'eu' phoneme), but I'd like to have it cleared up.

You've affirmed that you've seen different phonetic transcriptions of a
German diphthong: [O2] and [OY]. I've replied that they are meant to
represent exactly the same sound. Phonemic/phonologic representations are
language specific. So are phonetic representations, to a certain degree,
even though they shouldn't, of course. German scholars will think of the
IPA cardinal vowel [2] as being identical to the vowel of standard
German /2:/. At least in broad phonetic transcription, that vowel is
normally represented by [2:] for the sake of simplicity, even though in
narrower transcription, it'd be rather something like [2_r:].

In traditional broad phonetic transcription of German, there are two IPA
signs that represent one and the same vowel quality: [Y] and [2]. I know
this shouldn't occur. I'm not responsible of it. I suspect it's an
influence of phonology/phonemics and of traditional spelling on phonetics.
In most cases, they're not mixed up, since they correspond to specific
traditional spelling letters: [Y] to <ü> as in <hübsch> [hYpS] 'nice' and
[2] to <ö>, as in <schön phönizisch> [S2:n f2'ni:tsIS] 'nicely phenician'.
They're mixed up, however, in positions where they don't correspond to a
traditional spelling sign, as in the mentioned diphthong. That is the
reason why you find both ways of transcribing the same diphthongs: [ao, ae,
O2] and [aU, aI, OY].

I remember the first time I heard an 'official' recording of the IPA
sounds: [i, I, e, E] seemed all too opened to me (even though [i] should by
definition be the closest possible), whereas [&] seemed too close. If I
remember correctly, the theoretical definition is: take the extreme points
of tongue position and divide the distance between them into steps of the
same dimension to get the cardinal vowels. This definition leaves a lot of
space for interpretation. Does anybody know if there's a more exact
definition of the cardinal vowels? The authority of John Wells?

j. 'mach' wust