Re: Re : new language; langmaker

Dans un courrier dat=E9 du 21/07/99 00:39:54  , Jennifer a =E9crit :

> 25 cases? 25 cases? This is not what I had mind. Again, sigh.
>  Yeah, I did try to base which cases on Asiteya and that quote. Thanks for=20
>  responding. All right, I'll try it, but 25 cases? I wish I could find tha=
t=20
>  link to Kelen...
> =20
>  Jennifer
Sylvia's Kelen is on
http://pw2.netcom.com/~sylvia1/

of course you don't need 25 cases.
you need as few cases as you think is ok. i was only
referring to the range of cases you find
in different systems to cope with different "mappings" :
x makes y into z
x thinks y is z
x works on y as a Z
etc.
but you can skip all that ok. Some conlangs may
be helpful to you although they are not allnoun :
Kristian's Boreanesian "trigger system" is 3.
Joshua's Danov=EBn is excellent to try get rid of PoS (tags).
Christophe's Azak has an interesting
system whereby nouns can be tagged as "active".
Brad's Nova is very different : try its final "smell" list, i'm
sure you will like it (guessing from your color names).
=20
Asiteya's 10 or 12 cases are enough too, don't you think so ?
the only thing is :
you may want to use as predicates
nouns referring to result (like "wound") and others
referring to agent (like "hammer") or to process ("container").
so i think you may need a terminative case "into"
(like sumerians had it too ;-) so that you can deal with result,
agent and process noun-predicates alike.
here is a very stupid example - i'm no masochist ! -
where wound is result, hammer is agent and pain a process.
it's all c-AP (case-Argument Predicate) and 1 and 2 refer to
a pair of attribute-attributee :

usedby-Him on-Me into1-Wound into2-Pain Hammer(ing).
madeby-Him on-Me with-Hammer into-Pain Wound.
madeby1-Him to-Me with-Hammer madeby2-Wound Pain(ing).

mathias