Re: Development of Silindion verbal inflection

Okay, here's that response I promised, but late.

I think the split you're describing, with the archaic endings actually being a
compound of person markers for agent and patient, is perfectly plausible. There
are numerous native american languages that compound the endings like that, to
the point that the compounded form changed phonetically, _as a unit_, to become
a new morpheme indicating both members. It's not a large leap, or even a hop,
to see that compound morpheme moving in other directions as the grammar
changes.

The real issue is how well the prototypical verbs conform to this empty 3s/agent
or patient 3rd compound. Certainly there will be some analogy (how else for the
form to spread?), but if there's not a good basic correspondence in the
original verbs, then it becomes an unrealistic change.

If it helps any, one of my first thoughts was that it was some sort of consonant
harmony system, and that the verbs took -n (< -m) unless there was a nasal in
the preceding syllable. Of course, that fails to explain 'nen' entirely. I like
this compound system you're proposing immensely more, providing that the
correspondence to the original verbs holds up (or can be forced with a big fat
linguistic hammer).

Aidan

Elliott Lash <erelion12@...> wrote:  As I have shown before many times, in Silindion there
are two main endings for the 3rd person singular [in
the present tense], namely, -n and -r.

Example:

   tek��n  "he praises X"
   an��r   "he promises"
   mir    "he falls"
   nen    "he gives X"

The main difference between them is often that -n is
for transitive verbs (not universal however) and -r is
for intransitive verbs (not universal as well).

What I'm beginning to understand is that, in some time
in Silindion's past it must have had a SPLIT
inflection system of some sort. This can account for
the less than straightforward distribution of -r and
-n. (I'm not sure exactly how, but it's a start of a
theory).

Anyway, beyond this assumption of an ancient split
system, I also think that the basic divisions in
inflection for Silinestic (Silindion's immediate
predecessor) must have been the following:

predicate with two arguments:
  *tekk-Vm       nistad       lawando-m
   praise-3A/3P  king[AGENT]  hunter-[PATIENT]
 "the king praises the hunter"

predicate with one argument:
  *an-Vr       nistad         gelet��-di
   promise-3S  king[SUBJECT]  gold-[GENITIVE]
  "The king promises gold"

[this is intransitive, really "makes a promise of"]

predicate with no arguments:
   tekkn-V                     lawando-di
   praise.stative-[empty 3s]   hunter-[GENITIVE]
   "The hunter is being praised"
   "There is praising of the hunter"

(the verb here is a n-stative derivative of the root
TEKK "praise". N-statives as derivatives are very
archaic, and mostly have become grammaticalised as
passive 3rd singular presents in Silindion. One
n-stative which is still derivational is
be ready"> as opposed to )

So, this means that the suffixes -Vm and -Vr may be
compound suffixes,  -V-m and -V-r, with -V being an
empty 3rd singular, -m relating that 3rd singular to
an object, -r relating it to a subject. Found by
itself, the verb remains neutral as to argument count.


Is this at all plausible?

 -Elliott


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around
http://mail.yahoo.com


		
---------------------------------
Yahoo! Mail
Bring photos to life! New PhotoMail  makes sharing a breeze.