Re: Noun cases in Tirelat

On Thu, Sep 18, 2008 at 11:13 PM, Herman Miller <hmiller@...> wrote:
[...]
> So some of what I've been calling "cases" might just be prepositions. The
> "vocative case" is a good example. I was thinking that prepositions in
> Tirełat governed specific cases as they do in Minza, but the newly created
> preposition "ki" is the only example I can find. The main differences with
> the "case/number particles" are that they encode number (which from these
> examples is absent in prepositions like "za"), and that most of them relate
> to specific arguments of verbs.
I've got something similar going on in Gogido. I started out with the
idea of theta-particles, which are sort of like case markers in that
they can attach to any noun and they tell how that noun functions as
an argument to the verb, and prepositions. But, since there are lots
of places where the semantics of verb-arguments vs. noun-modifiers
overlap, like locatives. One could separate things by having one class
of words (theta-particles) that can only be used for
verb-relationships (basically, adverbial prepositions), and another
class of words (prepositions) that can only be used for
noun-relationships (adjectival prepositions), but that seems inelegant
since Gogido doesn't distinguish adjectives and adverbs otherwise, and
it makes more sense to use just one word for the same relation whether
it's supposed to mark a verb argument or a noun modifier

My current analysis just says that everything is a preposition, and
pragmatics tells you that there are a few of them that simply don't
make any sense when applied the wrong way. But I do not like this
solution very much, because it leads to the "I ate the fruit on the
table" problem.

-l.

Reply