Re: New H/G lang?

Christophe writes:
>>>>>>
Paul Bennett wrote:
>
> The lang has an enormous consonantal phono,
[snip]
> Also, there's only two vowels,
[snip]
>
        I think that with a structure of the word like yours, so many
consonnants don't pose any problem. But if the schwa was to be elided
for example, there would be a problem of consonnant clusters and
reduction of them, as I think there would be no PoA left for that
without loss of information. About the vowels, I'm not sure that i-bar
and lowered-schwa are a good idea. They are much too schwa-like both (in
fact, the schwa can often be a i-bar in some languages). If you want to
have only two vowels, I would vote for two very different vowels, like
/a/ and /@/. To stick with your idea, I think /a/ and i-bar would be
just fine, with a lot of allophonic variation (maybe triggered by the
consonnants) of course.
<<<<<<

It'd help if I explained myself a bit better in the first place, I suppose.  /a/
and /i-/ do sound like likely candidates.  Some rules that I missed are:

Vowels take a degree of fronting that is euphonious with the consonant they
follow.*
A labialised consonant causes rounding of the following vowel.
A palatalised consonant causes raising of the following vowel.

*This rule is currently "under construction", and may be discarded.

There's not going to be any vowel elision.  I do that far too often in my
conlangs.

>>>>>>
[snip]
> consonant table (ie {pitaca}, {xat[ika})  Each of the three syllables "moves"
to
> a different row on the consonant table (ie to Voiced Affricate or Prenasalised
> Voiced Stop) to show a different grammatical function.
[snip]
        I had once this idea too. But I never quite used it. I find sometimes
remnants of this idea in some of my conlangs, like the use of voiced
infixes in Moten for singular cases, whereas the plural cases use the
voiceless versions.
<<<<<<

Having posted this, I also see elements in Celtic lenition, nasalisation, etc.

I'm also thinking (having done the math) that 36^3 is far too many "shades", and
there'd have to be very many "redundant" positions.  Unfortunately, to do much
about it would mean mangling the "deep structure" of the phono, which is
actually a 5D grid of 3x4x2x3x4 structure.  Nice and regular, but not subject to
very easy "elimination" of phones without being broken.

>>>>>>
> How reasonable would it be to postulate this as a late-neolithic/early
> bronze-age hunter/gatherer lang?
[snip]
>
        Such a structure would be good for a proto-form of a language, a
reconstructed form of a language by the linguists of your conculture, as
so many consonnants and so few vowels promise a good deal of phonetical
changes, especially seen the grammar you want to give to it.
<<<<<<

I see this system as a result of a conprotolang that was a lot more analytical,
which has "collapsed" (like the Celtic model) and the collapsed forms have been
(hyper)corrected.  It's going to be a sister-lang to the highly analytical
m"/21aw conlang, which has (btw) become a human lang, with it's own concommitant
phono reorganisation.

I was also inspired in part by an amusing tale told in sci.lang quite a long
time ago about a linguist attempting to find the Sakao word for the genetive
form of "tail", and finding that Sakao inflects (deep structure) to the point of
having unrecognisably different (surface structure) words.  Has anyone got any
more info about Sakao, btw?






*************************************************************
This email and any files transmitted with it are confidential
and intended solely for the use of the individual or entity
to whom they are addressed.
If you have received this email in error please notify the
sender. This footnote also confirms that this email message
has been scanned for the presence of computer viruses.
*************************************************************