Re: USAGE: convenient symbol for Swedish long _u_

Quoting Isaac Penzev <isaacp@...>:

> On Mon, 12 Jan 2004 13:03:49 +0100 Benct Philip Jonsson wrote:
>
> > Which of of the notations 2\ Y\ 8\ would
> > you find most convenient/intuitive for the
> > Swedish "long _u_", which in strict
> > X-SAMPA is [2_w]?
>
> [2_w] is contradictory. [2] itself is a rounded vowel. If you need to
> distinguish between less rounded and more rounded vowel, you should use [_c]
> for
> less r. and [_O] for more r. In IPA they are left half and right half of a
> ring
> below respectively.
Thing is, it isn't just _more_ rounding, but a different sort thereof. The
lips are thrust forward a bit. Describing it as labialization and contrasting
it with mere rounding isn't perhaps ideal terminology, but it seems to be what
we've got.

> Especially taking into account that it may be a dialectal feature:
> - quoting Andreas Johansson :
> > In case anyone's forgotten, my 'lect does not have [2_w:] for this
> phoneme,
> > but [u\:]; a phone sensible enough to have its own IPA sign! :)
Isn't pretty much everything a dialectal feature? But my limited reading on
this suggests that BP's pronunciation is the commoner one.

In any case, while it lacks [2_w:], my 'lect does have a bunch of
these "labialized" vowels: [O_w], [o_w:], [U_w] and [u_w:]. There isn't any
unlabialized series [O], [o:], [U], [u:] for them to contrast with, so the
phenomenon is subphonemic (unlike BP's dialect, which distinguishes [2:] vs
[2_w:]), but we still ought to be able to note it in close transcription.

                                                         Andreas