Re: Relative sonority of rhotics vs lateral liquids

On Thu, 20 Nov 2008 17:12:23 -0500, Eldin Raigmore
<eldin_raigmore@...> wrote:

>OK, a BTW question.
>
>It's my impression that nearly every rhotic is either at the uvular or the
>retroflex PoA, or is articulated in the tap/flap or trill or non-lateral
approximant
>MoA.  In fact, I'd leave out the "nearly".
>Is this correct?
Let me concentrate on the PoA question.  You may or may not believe that
rhoticity is the property of having a (relatively) low third formant -- I
think Ladefoged doesn't, though he concedes that it characterises a big
chunk of the rhotics well.  But suppose you did.

The third formant is the third resonant frequency of the vocal tract, and
the mode at that frequency would look like the third picture in
  http://www.umanitoba.ca/faculties/arts/linguistics/russell/138/sec4/sh2-15.gif
if the vocal tract were a simple tube.  To lower the third formant you want
to narrow the tube at one of the points where the amplitude of its mode is
maximal, i.e. 1/5 or 3/5 or 5/5 of the way along the vocal tract (from
larynx to lips).  It turns out that in the vocal tract these points of
maximal amplitude are around the uvula, behind the alveolar ridge, and at
the lips.  The first two of these are the canonical PoAs you named, where we
get rhotics.  Labials aren't good rhotics -- I guess because a constriction
at the very end of the tract lowers _every_ formant, not just the third one
relatively? -- but this is what's supposed to be behind [v\] as a rhotic in
some English Englishes.
(I forget the source of this, sorry).

As for your other questions, Google was eager to show me a discussion on the
list from 2004:
  http://archives.conlang.info/bo/bhaerzhu/pirqhuarcoen.html
Perhaps you'll get something of it.

Alex