S/A pivot?!

[kamine]

Here is a sentence from NeotrekkerZ's guide - the sentence is illustrating mood/tense (I added the gloss).  Also, in this post I'm referring to S as the subject of an intransitive sentence, A as the subject of a transitive, and O as the object of a transitive.

    Trram oe t<am>aron, oel ts<am>e'a palulukanit ulte t<am>ätxaw ne Kelutral.
    Yesterday I hunted, I saw a thanator and returned to Hometree.
    Day.past I-Null hunt.past, I.Erg See.Past Thanator.Acc and Return.past to Hometree.

From what I understand, Na'vi is morphologically tripartite (so S, A, and O are all marked differently).  However, when conjoining transitive and intransitive clauses, it looks like you can drop the subject of the second clause (if they are the same subject).  That is what happens in the second part of the sentence: "I saw a thanator and returned to Hometree".  Since the subject is dropped in "returned to Hometree", does that mean that S and A are coreferential?

Here's another example, from the Wikibooks article on Na'vi/Verbs:

    trram    kä    na'rìng-ur    fte    tsun    t‹iv›aron    yerik-it.
    day-past    go    forest-dat    so.that    be.able    hunt‹sjv›    hexapede-acc
    "Yesterday (we) went to the forest so we could hunt a hexapede"

Here, I REALLY don't get how the subject is omitted - how is the semantic role of "we" as an agent conveyed at all.    Assuming that the subject is added in the first clause, It looks like it's optional (or not present at all) in the second.  This also suggests that S and A are coreferential.

Anyway, the conclusion that I'm reaching is that although Na'vi is morphologically tripartite, syntactically it behaves like a Nominative-Accusative language (with the S/A pivot).  Does that make sense?

I am an introductory-level linguistics student at a small liberal arts college.  This is my first post, so I'm not sure if it belongs on this board (maybe it should be in a less advanced board).

Help me please

-Ilay

PS:
Here's a link to that second example:
http://en.wikibooks.org/wiki/Na%27vi/Verbs

[omängum fra'uti]

This is a perfectly fine place for this.  The beginner boards tend to be about learning and questions related to that, while the intermediate has more advanced learning, in addition to some amount of analysis.

That said, I wish I could offer up some wisdom, but you've flown right past my linguistic knowledge, and I only have one language besides Na'vi to draw on from experience, and that's English.

That said, Na'vi is VERY heavy on context.  And it is my understanding that when it comes to contextual subjects, both the nominative and ergative can be treated equally in that regard.  (That said, that sentence looks wrong to me for other reasons, but that's a different matter.)

[NeotrekkerZ]

I hadn't really thought about the implications myself of dropping a subject in a subordinate clause when the verb in the main clause is transitive and the verb in the subordinate clause is intransitive (or vice versa).  You might want to check out wm.annis' translation of Coyote Tale in Na'vi and see how the subject dropping plays out in it as it is one of the few documents that has been looked at by Frommer, though I don't know for sure how critically he looked at it in regards to subject dropping.

[wm.annis]

Anyway, the conclusion that I'm reaching is that although Na'vi is morphologically tripartite, syntactically it behaves like a Nominative-Accusative language (with the S/A pivot).  Does that make sense?

Most ergative languages generally have some amount of N/A in their systems.  Pure ergative languages are quite rare.  There are two places where Na'vi shows N/A tendencies.  First, in the participles: there's a subject participle — the active — and an object participle — the passive.  Second, in terms of pro-drop behavior (as in your examples), which I've been calling pragmatic split-ergativity for myself.  I'm not sure calling it "syntactic" is quite right.

That said, I agree with omängum fra'uti — the wikibooks example has some serious problems (the dative is not used to indicate location; after fte it should be tsivun, not tsun).

[roger]

That would seem to be a draft example that was never verified. It is marked with a comment tag to run it by Paul. I just checked the rest of the book, apart from the N-E dict, and that is the only instance of such a tag. There is one other claim marked as 'dubious' in the visible text. I've asked Paul to comment on both.