[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Event contours and ZAhO tcita



la xorxes. cusku di'e

> This is fine, but then what is the definition of broda' for a
> PU ZAhO tcita, say "puza'o"?
>
> "which takes place in the past of the za'o phase of x4"?
> (This splits "puza'o")
>
> or "which takes place in the za'o phase of x4, which is in the past"?
> (This keeps puza'o as one tense, but is different from the use of
> "pu" as tcita.) This (more or less) is the interpretation lojbab gave.

I believe the right answer is that nobody knows the right answer yet.
We attempt to define semantics to the extent needed to get the language
bootstrapped: thus (e.g.) the le/lo distinction discussed here recently.
There are many constructs that are grammatical but as yet have no known
use:

        mi klama le zarci ve'i le nu do dansu le bisli

Somebody may someday mean something by this, but I at least do not understand
it.  (I know several things it probably doesn't mean, though.)

> > Specifying a sumti has generally the following consequencies:
> >
> >   (1) it NEVER does give a contour to the outer bridi -- all
> >       the sumti must be specifiable at the same time and
> >       different sumti might imply different contours, even
> >       mutually contradictory.
>
> I'm not sure why this is a GENERAL consequence of specifying a sumti.
> It can only apply to ZAhO tcita sumti, or am I missing something?
> Of course it is possible to say things that are contradictory, with
> any interpretation.

I think the answer is that specifying a sumti (modal or standard) does not
contour the bridi, because contouring is achieved with selbri tcita.

> Compare:
>
>         __mi dunda le cakla do__
>         I give you the chocolate.
> and
>         __mi dunda le cakla do bai le mi mamta__
>         I give you the chocolate compelled by my mother.
>
> Is the relationship between "mi", "le cakla" and "do" unchanged?

I would be inclined to say no: it is still the case that I give you the
chocolate.  The "bai" place has added new information, not overridden old
information.

> I agree that the meaning that the ZAhO have is reasonable for a
> sumti tcita. What I'm saying is that that meaning does not fit well
> with their meaning as tenses. This is most clear in the case of
> the ZAhOs other than "pu'o" and "ba'o". For instance:
>
> mi zutse le srasu za'o le nu co'a carvi
> *I keep sitting on the grass as it starts to rain.
>
> Under the current interpretation, I'd have to say:
>
> mi za'o zutse le srasu ca le nu co'a carvi

No, that doesn't work, because "ca" is aorist.  You can say:

co'a carvi za'o le nu mi zutse le srasu

or equivalently:

mi zutse le srasu .iza'obo co'a carvi

or:

za'ogi co'a carvi gi mi zutse le srasu

with differences in what is claimed.

One possible way around this problem would be a new particle, say "xa'o",
such that xa'o+ZAhO is grammatically equivalent to a brivla with place
structure:

        x1 (event) is the <whichever> phase of x2 (event).

That would allow sentences like:

        le nu co'a carvi xa'o za'o le nu mi zutse le srasu

Comments?

--
John Cowan      cowan@snark.thyrsus.com         ...!uunet!lock60!snark!cowan
                        e'osai ko sarji la lojban.