Re: Locative constructions in a:seka`eni (long)

Hi!

James W. writes:
> I'm currently working on the locative constructions in
> a:seka`eni. ...
Hehe, that's something I constantly think about in my engelangs,
too. :-)

> I found a really interesting paper by Marcus
> Kracht, "On the Semantics of Locatives (2000)" describing
> them as consisting of a configuration and a mode ...
Hmm, a three-dimensional system of distinction is not too uncommon,
the third dimension being touching vs. non-touching.  I think this
would split the 'modes', and ony locative, allative, ablative, and
perlative would remain, making the 'approximative' mode a non-touching
allative.  I think I like this better from the engelang point of view,
since it is more systematic.

It would also have been nice by the author if he had mentioned the
relationship between spatial, temporal, and notional cases in some
languages.  E.g. spatial case forms in Finnish show quite a clear link
to some of the notational cases.  This would make the system even more
systematic and link spatial relations with notional ones.  Putting the
Finnish cases into a grid:

               'in'       'at'       notational
   location:   inessive   adessive   essive
   towards:    illative   allative   translative
   from:       elative    ablative   partitive

If you compare the endings, still the Modern language shows hints that
'in' is expressed with -s- and 'at' with -l- and that the 'mode' part
of the case is from the notational case endings:

               'in'    'at'    notational
   infix:      -s-     -l-     -

   location:   -ssa    -lla    -na
   towards:    -aan    -lle    -ksi
   from:       -sta    -lta    -(t)a

Apart from the 'towards' row, the relations between the endings is
quite striking.  Maybe there are people on the list who can give proto
forms to make it even more visible?

In the paper, this would have made some of the case stacking examples
easier to explain (e.g. the behaviour of the ablative ending).

Starting with Qþyn|gài, I used what the author calls 'modes' to mark
temporal, spatial, and notional relations all alike.

>... The exception to this is in the above and below configurations,
> which don't appear in the recessive mode, because they seemed to me
> to be semantically identical to each other in the approximative mode
> ...
Hmm, I don't see why they are special:

    towards the above  vs.  away from the above
vs. towards the below  vs.  away from the below

I think this can all be distinguished.  Using only ascending
vs. descending seems to ignore the point of reference, i.e., you might
be
    ascending from the earth  (away from the above of the earth)
vs. ascending towards the sky (away from the below of the sky).

>...
> It seems to me that the approximative and recessive (and
> maybe transitory?) are kind of like a continuous aspect,
> where the others are more like a perfective aspect.
>...
Ah, maybe, yes.  Or touching/non-touching, but maybe that's only a
matter of view (aspect vs. result view).

>...
> I've laid out all the possible combinations in the chart below. What
> I need is for you to check my interpretation of the config + mode
> combinations, especially for the approximative and recesisve
> modes. Do they make sense? Are there any natlangs with systems that
> work like this? I'm not too concerned if there aren't, but hopefully
> the system is plausible, because it is supposed to be humanly
> useable.
Tabasaran is usually cited in this context, a Dagestan Caucasian
language.  It has 48 cases: 8 'configurations', 3 'modes' (at, to,
from), and touching vs. non-touching (to vs. towards).  I think John
Quijada posted an overview recently.  ...searches...  It was about
Lezgian, not Tabasaran, and the tables are here:

   http://republika.pl/lezgi/paradigms.html

Wrt. conlangs, Jim Henry's GZB also has a large and very regular
system (and very concise tables making reading easy :-)):

   http://bellsouthpwp.net/j/i/jimhenry1973/gzb/grammar.htm#postp

**Henrik