Re: /S/ and /tS/ in natlangs (and now conlangs!)

[Padraic Brown <pbrown@POLARIS.UMUC.EDU>]

On Wed, 8 Jul 1998, Steg Belsky wrote:

>
> I guess maybe i should have added in about conlangs, too :) .

That'd be swell!

> Rokbeigalmki has [s], [S], [ts], and [tS], all as different sounds.
> It also has [z], [Z], and a J-like [dZ], but no true [dZ] (voiced [tS]).
> I distinguish between the "one sound" versions of [tS] and [dZ], in words
> like "check" and "jacket", and the "one and a half sounds" versions in
> words like the name Tshernikhovsky.  I'm not sure if anyone reading this
> has a clue what i'm blabbering about, though :) .

Not entirely; but blabber away!

> Back to Rokbeigalmki itself, the rules of consonant-softening shift this
> way:
> [s] <s> -> [S] <sh>
> [ts] <tz> -> [tS] <ch>
> [z]  <z> -> [Z] <zh> (the normal form of the sound)
> [dZ] <j> -> [Z] <jh> (only used when the sound is actually derived from
> <j>)

Ah, mutations?  The only mutation that can occur with these sounds in
Kernu is the nasal:
s -> nz,  la sulys (nom.), la nzul (obl.)
z -> nz,  il Zawzu (nom.), le nZawzen (obl.)

I have no clue how to describe those sounds in IPA!  The nasal element is
there, it's just not really a part of either word.

>
> Btw, <> are the correct ways of distinguishing written forms, right?

Don't know.

Padraic.

>
> -Stephen (Steg)
>
> _____________________________________________________________________
> You don't need to buy Internet access to use free Internet e-mail.
> Get completely free e-mail from Juno at http://www.juno.com
> Or call Juno at (800) 654-JUNO [654-5866]
>