On Fri, Nov 6, 2009 at 2:44 AM, Alex Fink <[log in to unmask]> wrote:
> On Thu, 5 Nov 2009 13:50:17 -0500, Daniel Demski <[log in to unmask]> wrote:
>
>>On Wed, Nov 4, 2009 at 4:48 AM, John Vertical <[log in to unmask]> wrote:
>>
>>>
>>> Whistling would seem to require labialization as the immediate precedessor.
>>> If you have a palatalization contrast to begin with, enhancing it from
>>> plain<>palatalized to labialized<>palatalized could be a way to work
>>> labialization in (at least that makes more sense to me than spontaneous
>>> labialization of palatalized consonants).
>>
>>Yes, I wasn't thinking it would just suddenly happen. I suppose
>>turning a palatal distinction labio-palatalization could be... fairly
>>plausible?
>
> No, not like that.
>
> Labialisation and palatalisation are relatively antithetical gestures.
> Something could probably be said about their having opposite effects on
> certain formants, but I never learned that stuff -- so I'll just say that,
> as you noted, although some languages have labiopalatals, they're a fairly
> rare breed; also observe that front rounded vowels are by far the rarest of
> all combinations of [+-front][+-round].  So you can imagine that palatalised
> and labialised "stand on opposite sides" of plain.  Then from an original
> contrast
>  plain <> palatalised
> (where "<>" means 'contrasting with'!), rather than having palatalised "jump
> over" plain to labialised, it makes more sense for the contrast to be
> increased by sliding plain over to labialised and keeping palatalised in
> place, to
>  labialised <> palatalised.
> This is what JV was suggesting.

OH! Wow, thanks. That is much more workable.

(snip)
> -- I imagine you don't need to produce labialisation on all of your Cs,
> though, only the ones that are going to turn whistled, which is probably
> just one or a few fricatives or affricates.  So my first idea to get the
> labialisation would be a smaller shift.  Labialisation and retroflexion and
> velarisation/pharyngealisation all have similar acoustic effects (lowering
> F3?) so you could just shift one of those to labialisation, [s`] > [s_w] or
> something of the sort, which you could pull off without having the effects
> of the change be too wide-ranging.

What I wanted was whistled k and r (and r_0).

>
> Why would trilled vowels cause voicing?

OK, they wouldn't.

A trilled vowel could become a voiced fricative or stop, though.

>
> Alex
>