GROUPLANG: cases (was: noun and verb roots)

Mathias M. Lassailly wrote:
>
>Thanks :-) We understood each other although we don't use the same names for cases
>(my 'absolutive'
>is actually 'agentive' and your 'undergoer' is 'absolutive' :-) But I let down
>case issues now : they
>will pop out again later re. negative causative and factitive we don't have discussed yet.
>
Wait, don't go yet! :-)
I wouldn't like to face trouble with cases again, a month from now
or so. Let's make a uniform terminology.

I'll mention all the cases we have been dealing with so far,
and you tell us what you call them:

Case 1:
THE DOG bit me
(agent)

Case 2:
The dog bit ME
(patient)

Case 3:
HE sees us
(undergoer?)

Case 4:
The dog BIT me
(predicate)

Case 5:
THE RED dog bit me
(determinant) (also serves as genitive/possessive)

Case 6:
A RED dog bit me
(modifier) (also serves as genitive)


Now, there are some other possibilities:

Case 7:
HE runs.
(factitive?) -> this could be merged with agent.

Case 8:
The dog is red
(attributive?)

Case 9:
A dog is an animal
(equative?)

These two could be shown by making the dog the theme.
--> dog, pred-red  "The dog, it is red"
    dog, pred-animal  "The dog, it's an animal"

The lack of case inflection and the pause would mark
the theme; you can't possibly misinterpret something like
"The dog was reddened" (which would be [patient-dog pred-red]).

I mean, you could, but context would tell you it's not probable.

 dog, caus-I pred-red

would probably mean "I made the dog red".

What I'm trying to do here is to show you that there's
a way to reduce the complexity of the case system by merging
some related functions into a single case, and letting word
order and context do the rest. I find the 10-case system very
interesting but too difficult to use... I don't know if the
rest of us feel the same.


--Pablo Flores