Help me analyse the clauses in this sentence (please)!

[zaba]

Here's the sentence:

nde fidmo ngnieu-sibi-rip, naprafdi zwapa-nyo-o    on-htsi     kiwbasa-o
the ghost anger-self-subordinator really eat-NMZ-ACC 3-want sausage-ACC
'the ghost, having angered himself, really wanted to eat sausage'

What sort of clauses are here? Would I be correct in assuming that there's a complement clause somewhere? How would you identify it?

Thanks gang, you're the best (collectively, possibly individually)

[Daniel]

Your gloss presupposes an analysis. Based on that--

"want to eat" looks like it involves a nominalized predicate, as in "wants the eating of X" or "desires consumption". You might reconsider the gloss and call it an "infinitive", but either way you pretty clearly have some sort of partly nominal, partly verbal predicate/clause that is a complement of 'want'.

As for "rip"/'subordinator', that one is less clear. That's probably what is considered a converb, which is likely not a complement but either an adjunct or a coordinated phrase. (A converb can sometimes act like coordination.) You might consider a translation as "The ghost angered himself and really wanted to eat sausage." Exactly what to do about converbs is a tricky topic.

[zaba]

can any part of this be considered as complementation? I think the part with NMZ may be insofar as the nmz verb is the direct complement of the main verb, right?

[Daniel]

Yes, based on your gloss, that does look like complementation.

[zaba]

How so? Sorry to press the issue, but this can be so tricky sometimes...

[Daniel]

What else would it be? What is "complementation"? It looks like the verb "want" selects an argument, which happens to be an nominalized predicate. Fairly straightforward. Is there something I might be missing?

(Again, your gloss might be entirely wrong, then it's unclear where to start from. But I have no reason to assume that it is wrong.)

[zaba]

So, it is correct to summarise it thus:

nde fidmo ngnieu-sibi-rip, naprafdi zwapa-nyo-o    on-htsi     kiwbasa-o
the ghost anger-self-subordinator really eat-NMZ-ACC 3-want sausage-ACC
'the ghost, having angered himself, really wants to eat sausage'

The complement clause here is `want to eat sausage'. Complementation with the NMZ -zwapa is used when the action is the direct complement of the main verb. The resultant noun (i.e. the ``infinitival clause'') is in the accusative.

>> Assuming the correct gloss, do you agree?

[Daniel]

Yes. Assuming that gloss is right, that's what I'd expect.

[zaba]

Sure, I'd expect that too -- but on one hand, I'm worried because upon reflection it looks discontinuous. What do you think?

[Daniel]

What kind of word order flexibility does the language have in general?
That's an interesting observation but not in itself a reason to question that analysis.

It might turn out something else is going on. For example, it could be parallel to the perfective aspect in European languages:

"I have 5 read books" => "I have read 5 books"

In some languages, the participle still agrees with the noun. That might be where you get the accusative on each of them.

So roughly it could meaning something like:
"He wants sausages for eating"


But that's just an example of what might be an analysis given a lot more data. At the moment as a working hypothesis I think it's fine to consider it a complement clause.

Displacement/discontinuity can happen. But it rarely must happen-- can you change the order to make it continuous? Is that still acceptable to speakers?

[zaba]

Thanks a lot for your help. I am learning a lot with your expertise.

What kind of word order flexibility does the language have in general?

A lot of flexibility, though SOV is the most common by far, though since there are so many cases, anything is possible.

In some languages, the participle still agrees with the noun. That might be where you get the accusative on each of them.

Excuses for my ignorance, but how could I conclusively determine that? Can you give me an example or two? If I can determine that there is no required agreement, then I can more comfortable claim that it is a real CC and not discontinuous, correct?

Thanks!

[Daniel]

Excuses for my ignorance, but how could I conclusively determine that?

More data. The patterns should become clear and either fit the analysis or not. Check different verbs and different word orders.

Can you give me an example or two?

You can look into Italian participle agreement with object clitics. I don't have a paper in mind and can't think of an example off the top of my head, but it's an interesting case where there's a small remnant of an older system still in the modern language. This also means it might be harder to be certain about what's going on in this language, if it's potentially grammaticalizing/changing.

If I can determine that there is no required agreement, then I can more comfortable claim that it is a real CC and not discontinuous, correct?

"real CC" and "not discontinuous" are the same thing, right? The word order may be due to something else.
That's an interesting question. Probably, I guess.

Can you elicit an example with a dative or other case? That would be interesting. Hard to say now whether it would be conclusive, but it would be helpful. I imagine something like this:

help-NMLZR-ACC want-3SG man-DAT
'He wants to help [to] the man.'
[I'm assuming dative case for 'help' as in German, just as an example. You can choose whatever verb you want.]

[zaba]

Yes.

town-DAT go-NMZ-ACC want-3
'he wants to go to town'

The word order may be due to something else.

What do you think?

[Daniel]

town-DAT go-NMZ-ACC want-3
'he wants to go to town'

Seems like the right analysis to me. I don't (at the moment) see how that could be anything other than a complement.

What do you think?

Not sure... Pragmatics? Do you have a speaker available who you can ask about emphasis and context? ("But what if you wanted to emphasize the WANTING part?")


It is possible, but less likely, that there is some kind of topicalization or second-position effect here, which is causing the words to be in this order. Finding out what permutations are possible will help answer this. Try also getting a sentence with an explicit subject and perhaps some adjuncts and see where everything can move.

[zaba]

Thanks!