Re: Ergativity Reference Done

On Nov 26, 2004, at 10:26 PM, Thomas R. Wier wrote:

> From:    Dirk Elzinga <dirk_elzinga@...>
>> On Nov 22, 2004, at 7:24 PM, Thomas R. Wier wrote:
>>> Andreas wrote [concerning A and O marked one way, S another]:
>>>> I cannot off-hand think of any
>>>> examples for clairvoyant ones, but they're out there, or so I'm
>>>> told.
>>>
>>> Nahuatl might count as one.  Most nouns in Nahuatl have
>>> a so-called "absolutive" suffix (occuring in all three Dixonian
>>> roles), which must be removed to add possessive or plural morphology:
>>>
>>>   n-inekwisti-s xonaka-tl
>>>   1-smell-fut   flower-abs
>>>   'I will smell the flower'
>>>
>>>   n-inekwisti-s xonaka-meh
>>>   'I will smell the flowers'
>>>
>>>   n-inekwisti-s mo-xonaka
>>>   'I will smell your flower'
>>>
>>> The absolutive suffixes (either -tli, -li, or -tl) was IIRC
>>> originally an article, which over time lost its deictic sense and is
>>> now just frozen nominal morphology.
>>
>> Do you have a reference for this? I was not aware that any Uto-Aztecan
>> language had articles (with the exception of Tepiman), or that anyone
>> had proposed that the absolutive came from an article. It would had to
>> have been a Pre-Proto-Uto-Aztecan development, since the absolutive as
>> such is found in all branches of the family.
>
> I don't remember a precise citation, unfortunately.  I vaguely
> remember Langacker saying something along these lines in an article
> in _Language_ in the 1970s.  I could be misremembering this, or
> confusing it with a different article on Nahuatl dialectology.
I'll have to reread the relevant bits from his overview of Uto-Aztecan.
I don't remember seeing there before, and the idea intrigues me.

>> (BTW, my Nahuatl
>> dictionary gives _xonoca-tl_ as 'onion' -- _xochi-tl_ is 'flower'; I
>> once had a student named Xochitl, and so the word has stuck with me.)
>
> I should probably note that I am currently studying the modern
> dialects spoken in Oapan and Ameyaltepec, not Classical Nahuatl.
> Amith's (unfortunately yet to be published) book clearly says
> "flower".  His lexicon lists lots of forms referring to various
> parts of the onion plant, though, so clearly you're right about
> the general jist.
Okay. I thought you might have been looking at a modern dialect based
on the spelling. (I used to rebel at the archaizing spelling
established by the Spaniards, but it has its charms -- the
juxtaposition of the civilizing force of Western Romance with what must
have been to them the utterly foreign nature of Nahuatl -- which I've
gotten used to.)

>>> Since it is no longer an article,
>>> one could just as well call it case -- except that there is no
>>> opposition  defining it as such. Such systems are obviously
>>> dysfunctional.
>>
>> One would expect a dysfunctional system to change, but Luiseño, in the
>> Takic branch of Uto-Aztecan, has absolutives with about the same
>> distribution as Nahuatl (i.e., deleted in possessive constructions,
>> with postpositions, and in compounds; though unlike Nahuatl, they are
>> retained in the plural). If absolutives were feature of PUA, then the
>> retention of absolutives in Luiseño (and indeed in modern Nahuatl)
>> would be strange if the feature is dysfunctional.
>
> This is true.  I was proposing the analysis rather more in the
> attempt to show how something so weird could arise.  I'm sure you
> will recall that definiteness is sometimes associated with case
> marking (e.g. in Persian). I will certainly bow to your greater
> experience with this family on this.
No genuflection necessary :-). I don't even think that it's an
unreasonable analysis; it's just odd that if it were dysfunctional that
it should have persisted for so long.

>> The more so if it were considered a case, since in Luiseño the
>> absolutive can cooccur with case inflection:
>
> True enough.  But we were talking about Nahuatl.  Just because the
> languages are related doesn't mean they work the same! :)
Right you are. But if the absolutive shows the same (or similar)
distribution in Luiseño as in Nahuatl, the presence of additional case
marking would seem to me to weaken the case for considering the Nahuatl
absolutive to be moribund case marking. (Or would it strengthen it? If
the L absolutive lost its case function, then the language would be
free to innovate new case marking. Since that hasn't happened in N,
perhaps it's because it could still be considered case ... Nah. That's
just weird.)

Dirk
--
Dirk Elzinga
Dirk_Elzinga@byu.edu

"I believe that phonology is superior to music. It is more variable and
its pecuniary possibilities are far greater." - Erik Satie