Re: Sidaan Verbal System

On Sun, 18 Dec 2005 18:11:02 +0100, Henrik Theiling <theiling@...>
> wrote:
> [snip]
>"David J. Peterson" <dedalvs@...> writes:
>> [snip]
>> Okay, now for the verbs.  I've always found the idea of dependent
>> marking rather fascinating.  So instead of marking case, Sidaan
>> arguments agree with the other present arguments.  Each non-direct
>> object is marked with /-s/ if a direct object is present, and each
>> non-indirect object is marked with /-t/ if an indirect object is
>> present.  Here are some examples:
>>
>> (1)
>> (a) tox cLeho.
>> /I cry-PERF./
>> "I cried."
>>
>> (b) lEmba cLeho.
>> /girl cry-PERF./
>> "The girl cried."
>>
>> (c) toxs maza loN\quno.
>> /I-DO apple eat-PERF./
>> "I ate the apple."
>
>You have a very interesting system here.
>
>However, I had always understood 'dependent marking' differently,
>namely that the dependents (here: arguments) are marked for agreement
>instead of the head (here: verb).  So case would be dependent marking
>while a verb marking for transitivity would be head marking.
I thought what Henrik thought, David.
I still like your system.
(BTW I kind of toyed with a slightly-similar idea;
Comrie says some Tense systems have only Past and Non-Past,
others have only Future and Non-Future.
As far as he knew there isn't any that has only a Present and Non-Present,
but, what if a language had morphemes (short, single-segment morphemes) for
Non-Past, Non-Future, and if necessary, Non-Present?

Similarly Siewierska says some Person systems have First and Non-First, or
Second and Non-Second, or Third and Non-Third, person markers.  Why not
have a conlang whose "person" morphemes basically told you what person the
marked nominal "_wasn't_", instead of what it "_was_"?)

>You system definitely is dependent marking in this sense, and seems to
>use a 'reverse' agreement system similar to Classical Nahuatl (which
>is head-marking).  There was a thread on this list:
>
>http://listserv.brown.edu/archives/cgi-bin/wa?A2=ind0503D&L=CONLANG&D=0&I=-
3&P=2220
>
The system explained in that thread isn't all that unusual, actually, for
polypersonal verb-agreement.
If a verb with more than two participants is going to agree with only two
of them, the subject and another one, it will usually choose to agree with
a human other one, or, if there is no human other one, an animate other
one.
Transitive verbs with beneficiaries usually have human beneficiaries, or if
not, then animate beneficiaries, even when the direct object is not human
or even not animate.
Similarly, ditransitive verbs usually have human indirect objects, or at
least animate indirect objects, even when the direct object is not human,
or even not animate.
So it makes sense that if a Nahuatl tri-or-more-valent verb is going to
agree with its subject and just one other participant, it will more often
agree with a beneficiary or indirect object than with a direct object.

The fact that there is a hierarchy that this is contrary to, and that
Nahuatl is consistently contrary to it even when, say, both the direct and
indirect objects, or both the indirect object and the beneficiary, are
equally human or equally inhuman or equally animate or equally inanimate,
may be unusual, if indeed that is the case.

>> [snip interesting stuff from David]...
> [snip interesting stuff from Henrik]...
Thanks, David and Henrik and others.

Tom H.C. in MI

Reply