Re: Sketch: Tatari Faran

On Wed, Oct 13, 2004 at 09:10:05AM -0500, James W wrote:
> >>>> H. S. Teoh<hsteoh@...> 10/12/2004 6:01:22 PM >>>
> >On Tue, Oct 12, 2004 at 11:57:05PM +0200, Steven Williams wrote:
[...]
> >> What are the function of these cases?
> >
> >The core cases essentially function like Ebisédian, which takes a bit
> >of an explanation. Basically, the core case of an NP is chosen
> >semantically, depending on the role of the noun relative to the verb.
> >The originative is used for source, origin, or active entity; the
> >receptive for destination or receiving entity. The conveyant is for
> >the transported, or conveyed, entity. This is a bit abstract, so I'll
> >use some examples:
> >
> [snip great examples]
>
> I've looked at your grammar for Ebisédian, and now the grammar for
> Tatari Faran, and I think they are great!
Thanks :-)

> Such a refreshing change from Accusativity, Ergativity, etc.
It was more or less a reaction to what I considered ugly in accusative
systems: passives and indirect objects.

To me, passives essentially conveyed the same factual content as their
active counterparts (X performed some action A on Y), and so I saw no
reason why it should be a different syntactical construct. In fact, I
saw no reason why passives should exist at all, except to conform to
an essentially arbitrary system that requires a subject to be always
present. In English, if we don't know X, we have to use a passive and
subjectivize Y. Yet if X is known, then regardless of whether Y is
known X remains the subject. This asymmetry to me was undesirable. Why
not a system where all verb arguments are optional? So if I didn't
know X before, I could just state the verb A and the object Y, and
once I learn what X is, I simply add X to the sentence without
needing to change the verb form. There is no reason to, since it is
stating the same facts, just in different degrees of completeness.

Indirect objects to me were an afterthought grafted onto an inadequate
system---besides playing the "wrong" role (eg. in such verbs as
"give", the object should be the recipient, not the thing given, since
without a recipient, giving makes no sense). Verbs like "to give" were
essentially trivalent; the reason for the ugliness of the indirect
object was because it was trying to express a trivalent verb in a
divalent system (subject-verb-object). Hence, I concluded that a
non-ugly system would have to be trivalent.

This led me to consider if it was possible to express all verbs in a
trivalent system, unambiguously. After looking at several sample
verbs, I realized that the trivalent system I made, which uses "give"
as a paradigm for the three verb arguments, had a fixed directionality
from the first argument to the third. (I.e., the first argument must
be the source, and the third the recipient. I couldn't allow both
directions, because that would resurrect the evil passive.) In order
for other verbs to fit into such a paradigm, then, the cases must be
semantically derived. Thus Ebisédian's case system was born.

Of course, this line of thought reflects my personal sense of
grammatical aesthetics; it doesn't mean that I'm out to expunge the
Evil Passive from language.  :-) (You've got to be glad I'm not of the
auxlanger persuasion... :-P)

The Ebisédian case system acquired 2 more cases afterwards, which in
retrospect was an unnecessary complication. Tatari Faran's system is
essentially both a return to the original conception of the trivalent
system, and a refinement of it, in recognition of the fact that in
normal language, the concept of subject is necessary. The solution
that presented itself to me, then, was just to make the subject
orthogonal to the semantic roles of the nouns.


> I hope you don't mind if I borrow some ideas for my emindahken.  :))
[...]

Of course I don't mind. As they say on the 'Net, "imitation is the
best form of flattery".


T

--
Everybody talks about it, but nobody does anything about it! -- Mark Twain