Re: The Language Code

On Wednesday, May 28, 2003, at 01:30  AM, Rob Nierse wrote:

>>> Dirk, can you include ' i inverse ' á la Kiowa? I happen to have that
>>> in my conlang.
>
> <snip explanation>
>> If the default for a word like 'finger' were plural,
>> the marker would indicate singular.
>> Does your system work like that?
>
> Yes, it does.
>
>> If so, here's the problem that I see. For the morphology section of
>> the
>> Language Code I've been tacitly assuming that the attributes and their
>> values refer to morpho-syntactic categories and not to the formal
>> realizations of these categories. While 'inverse' might be a novel way
>> of realizing number categories, it isn't a category itself.
>
> I understand. So I guess my lang has singualr and plural, but just
> expresses in an 'inverse way'.
Right.

>> The question is whether the Language Code should include
>> realizational properties of morphological categories beyond the
>> general
>> agglutinating/isolating/inflecting cast of the language as a whole.
>> I'm
>> inclined to not include them, unless someone has good arguments for
>> doing so. This also points out a weakness of schemes like the Language
>> Code; you can't put in everything, and many interesting and even
>> important features will go unmentioned.
>
> I think you are right in this. A scheme like the Language Code should
> just mention some rough characteristics, a grammar should mention
> the interesting features.
Exactly. I have a similar problem with Tepa/Miapimoquitch. Number is
expressed by 3 different kinds of reduplication, but there's no way to
capture that fact in the Code, nor do I think there should be;
interested readers should consult the grammar (or ask me since the
updated grammar isn't available yet).

> So, Tlapóa is :
> Tlapóa :Tp Pt*p++16,4s(c)v(v/c) Ws Ma+ h+t2a2c3g2nsp Sbso.argn
> Lc+++d+300
About 'W'; is the script constructed or an existing natural script? I'm
guessing a constructed script, so this would be 'Wcs'. You also have
'Sbso.'; does this mean that there are no verbs, but that whatever
takes their place is final in the clause?

> Just nitpicking about tones: what is the difference between 'register'
> and 'level' tones?
Register seems to be a feature of SE Asian tonal systems, which are
predominantly contour systems. Register usually refers to phonation
type -- creaky/breathy/normal (though I've only ever seen 2-way
distinction). Level tone systems are common in Africa and southern
North America; register doesn't seem to play a role in those systems.
I'm not a tonologist, so I am happy to be corrected on any of this.

> How to indoacte 'pitch accent'? Is that just 'register'?
Pitch accent plays the same role in languages which have it that stress
does; it's a way of indicating syllable/mora prominence within a
phonological word.

Dirk
--
Dirk Elzinga
Dirk_Elzinga@byu.edu

"I believe that phonology is superior to music. It is more variable and
its pecuniary possibilities are far greater." - Erik Satie