Re: Polysynth Question

Jeff wrote:
<<
I forgot to ask how you came up with all the combination pronominal
prefixes.
 >>

Umm...that's a good question.  I'm so far removed from it, that
I only have a vague idea; I'd have to actually analyze it.  In looking
at the chart, though, there are a couple things to keep in mind:

-The sounds [e] and [o] are old sequences of the phonemes /@i/
and /@u/, respectively.

-The sounds [M] and [y] are old sequences of the phonemes /ui/
and /iu/, respectively.

-/u/ > [w] / _# (in coda position); /i/ > [j] / _# (in coda position).

-/n/ is the default epenthetic consonant (when a consonant is
needed, for some reason, it's /n/).

Keeping that in mind, there are a couple of observations I can
make:

-[m] is associated with the first person.

-[k] is associated with the second person.

-Null or [l] is associated with the third person.

-A [u] is associated with the dual (this holds for the cases, as well).

-A [i] is associated with the plural (this, again, holds for the cases).

Okay.  Bearing that in mind, the reflexive columns should look
an awful lot like the class I column without an indirect object
(the sole exception being class IV verbs, which have a thematic
/l/ for the third person).  Next, take a look at the nominative
case for each verb class:

-Class I: Nominative = subject
-Class II: Nominative = subject
-Class III: Nominative = object
-Class IV: Nominative = subject

This should explain why class III verbs (which are verbs whose
semantic subjects are experiencers) look odd.  When it comes to
the thematic consonant, a verb will agree with the nominative
argument--whether it's the semantic subject or not.  In fact, you
can switch up the order of elements depending on the agreement
used:

anty ma-kano-masa.
/boy-DAT. 3sg.sbj.1sg.obj.-TMA-see/
"The boy saw me."

anty my-kano-masa.
/boy-DAT. 1sg.pass.-TMA-see/
"The boy saw me."

In the second one, there's a notion that, perhaps, the boy saw
me by accident, or without trying.  The difference isn't in the
cases, but with the agreement (the latter takes passive morphology).
You can see the word order switch if both the subject and object
are third person, but the verb form will be the same:

anty m@kw@ kanomasa.
/boy-DAT. fish-NOM. saw/
"The boy saw the fish."

m@kw@ anty kanomasa.
/fish-NOM. boy-DAT. saw(passive)/
"The boy saw the fish."

Undoubtedly, examples like this (where, again, the second one
has a slightly different interpretation) suggest that the distinction,
were this a real language, would die out, like the subjunctive in
English.

Oh, to finish off the class III table, there's also an animacy thing
going on.  So if you have a 3sg. subject and a 1st dual object, you
get 3rd person agreement morphology because the dual is as
low down on the totem pole as it gets.  Of course, such an apparent
irregularity would probably get ironed out over time, and you'd
probably get /maw-/ there.  For now, though, I like it.

Anyway, all that should be able to account for the distribution
of agreement prefixes.  (Should.)

-David
*******************************************************************
"sunly eleSkarez ygralleryf ydZZixelje je ox2mejze."
"No eternal reward will forgive us now for wasting the dawn."

-Jim Morrison

http://dedalvs.free.fr/