the poop on voiced uvular stops

On Tue, 26 Oct 1999, Nik Taylor wrote:

> Daniel Andreasson wrote:
> > Another thing is that we have /q/ but no voiced equivalent.
>
> That's not very unusual at all.  A lot of langs have /q/ but no voiced
> uvular stop.  Which makes sense, if you think about it.  /G"/ is (at
> least I think so) harder to make than /q/, due to the narrow space
> between the glottis and the uvula for vibrating.
Well, almost. Voicing is the periodic vibration of the vocal
folds. Vocal fold vibration occurs when there is airflow across
the glottis; this airflow occurs when air pressure below the
glottis is greater than air pressure above the glottis. In the
production of a nasal consonant, the pressure differential is
relatively easy to maintain since supraglottal air can escape
through the nasal cavity. However, when the vocal tract is
completely occluded in the production of a stop this pressure
quickly equalizes since the air has no place to escape. This
means that a voiced stop will be of shorter duration than a
voiceless stop, no matter what place of articulation.

If the stop is bilabial, expanding the oral cavity by puffing
the cheeks will help to maintain the pressure differential
necessary for voicing. The further back the closure is, though,
the less volume there is, and the quicker the pressure will
equalize. A uvular stop is about as far back as you can go, so
it's not surprising that voiced uvular stops are relatively
rare. The vocal folds have just as much room to vibrate, it's
just that the pressure essential for voicing equalizes so
quickly (because of the small volume) that a voiced uvular stop
doesn't get much a of a chance to be heard. Lots of articulatory
effort, but not much perceptual payoff.

As far as voicing goes, then, the most unmarked stop inventory
would be something like:

                t       k
        b       d

Notice no voiceless [p], and no voiced [g]. I believe this is
the Furbish stop inventory (Ray, Jeff? Do you remember?).
Perhaps the conlanger responsible for that bit of work knew
something about voicing ...

Dirk

--
Dirk Elzinga
dirk.elzinga@m.cc.utah.edu                 "All grammars leak."
http://www.u.arizona.edu/~elzinga/               -Edward Sapir