Re: And who needs vowels?

On Dec 22, 2006, at 10:02 PM, H. S. Teoh wrote:

> Having recently learned a little Russian and stumbling across some
> Czech
> in, of all places, a little puzzle game, and having heard of
> Georgian on
> this list on a few occasions, I started to wonder where one may end up
> should vowels remain in such poverty and consonants continue to
> proliferate. And so, I dreamed of a conlang where all vowels have been
> elided and substituted with sonorant consonants, and some words
> consist
> of nothing but stops. For example:
What a cool idea! I've toyed with making a conlang where all the
underlying segments are consonants, but where some of them surface as
vowels. The only problem (as I see it, anyway) is that I think having
(underlying) vowels is a universal of some kind. I may be wrong,
though (if so I'd love to hear about it).

I've also played a little bit with making a language which does have
underlying vowels, but also has long initial clusters. This was
influenced more by Old Tibetan than by Georgian, however.

>
> plb [pl=b] (I suspect this one may happen to be an actual Czech word)
> bg [bg]
> pg [pg] (yes there is an audible difference between /bg/ and /pg/)
> pglbz ["pgl=bz=]
> mnvpl [mnf="pl=]
> bxtm ["bx=tm=]
> strvtz [strv="tz=]
Before I thought about/read about syllabic sibilants, I always
considered e.g. <street> to have one syllable; but since then it
actually feels to me like it has two: [s=tr\i:t_}]. But then I'm odd :)

> 'b [?b=] (perhaps an implosive [b]? There is no audible release)
Is there a reason you put [b=] but didn't put an = after the other
stops?

Also, is it possible for a stop to be "truly" syllabic? I seem to
remember reading about a language where some stops *act* as if
they're syllabic, but phonetically they have a schwa next to them.

> pstng [ps"tN=]
>
> The difference between /bg/ and /pg/ is that /bg/ has a short
> prevoicing
> before the release of the lips, whereas /pg/ doesn't. The [g] is not
> audibly released.
>
> Now, [l=] alone isn't all *that* interesting. The interesting part is
> that it is (relatively) easy to pronounce two different kinds of [l=],
> one high (palatised?), and one low (velarized? maybe retroflex?).
> Similarly, palatised and non-palatised [x] may well be two different
> phonemes. (Which, in retrospective con-history, came from consonants
> preceding [i] and [M], which have elided.)
>
> Of course, to make the prospective language more mellifluous
> (*cough*),
> it seemed good to me that it should be tonal.
I wonder if tone can really be distinguished on syllabic consonants.
I am working on a conlang that uses some voiceless vowels and has two
tones, and at first I was going to allow voiceless vowels to
distinguish tone just like the voiced ones, but when I would try to
pronounce them differently I found it very hard. Then I found out
Japanese does not distinguish pitch in its voiceless vowels, so I
decided I wouldn't either. Of course, I suppose it is possible to
distinguish them, but Japanese speakers and I have trouble doing so;
maybe some people don't have that problem.

[snip other interesting features]

Reply