About the need of agentive (-l, -ìl)

[Tirea Aean]

I stumbled across a cool thread in the Russian forum started by Kemaweyan. Here's how OP roughly translates...

Just got an idea. Always before, I felt that the agentive(-l, ìl ending) is redundant and only a tribute to tradition. Because, in fact, if you leave it off, nothing changes; the meaning will still be uniquely understood. However, only by chance, I came up with a situation when such substitution changes the meaning, and quite strongly. The scenario is like this:

Txantsan lu fwa tslolam oe.

well and, accordingly:

Txantsan lu fwa tslolam oel.

The meaning of these two proposals differ fundamentally. In the first case, it is the fact that I understood something is excellent. In the second case I'm talking about the great thing that I understood. It turns out, the agentive is not a redundancy. I can see why Paul decided that the agent is required.

He adds later, after I don't quite understand his case, the following. And then his point was driven home.

Tsun fko rivawn fa lì'u alu fì'u frastxolì'ut, kefyak? Slä txo fìkem ke sivi, tsakrr fìmelì'fyavi lu:

  Txantsan lu fìtìlen a oe tslam.
  [This event, that I understand, is excellent.]

  Txantsan lu fìsäomum a tslolam oel.
  [This information that I understand, is excellent]

Ral lu law nìwotx. Tì'efumì oeyä, ke tsun fko pivlltxe san Txantsan lu fìsäomum a tslolam oe sìk fu san Txantsan lu fìtìlen a oel tslam sìk. Tsari ke tse'a oel ralit. Txo oel tslolam 'uot, tsakrr tsaw ('uo atslawnam) lu säomum. Nìteng fìtìlen (a oel tslolam 'uot) ke lu säomum - tsaw lu tìlen.

Cool, huh? Thoughts?

[Plumps]

  Txantsan lu fìtìlen a oe tslam.
  [This event, that I understand, is excellent.]

Still ambiguous, tì'efumì oeyä... I would still expect to know 'what' I've understood ... moreover, it could mean "the fact that I am understood is great" as txantsan lu fwa fkol oet tslam in its long form. How would you distinguish that if there was no L or T indicated?

[Tirea Aean]

moreover, it could mean "the fact that I am understood is great" as txantsan lu fwa fkol oet tslam in its long form

How could that be shortened to and logically equivalent to txantsan lu fwa oe tslam? This short version implies the oe is the subject because it's in nominative form(root word untouched by case markers, assumed to be the intransitive subject). I think to have the meaning you're thinking of, you must use oeti, and leave out fko as the understood implied subject.

~beware the edits xD

[Plumps]

This short version implies the oe is the subject because it's in nominative form(root word untouched by case markers, assumed to be the intransitive subject). I think to have the meaning you're thinking of, you must use oeti, and leave out fko as the understood implied subject.

Oh, I see ... I think I misunderstood the intend of the post. Ngaytxoa.

But given your argument, doesn't that imply the subject always has to be present? Then something like txantsan lu fwa oeti tslam wouldn't be possible but it is as far as I know 

[Tirea Aean]

One of them has to be present. (Or even none, if the stuff is established!) Lone nominative form implies it's the subject, like an antipassive form. Leaving out the agentive subject out as implied and understood and only having a patientive direct object in there should be okay and easily understood as such.

That is, I think the following is all good:

1)
A: Palulukanur lu kinam apukap. Thanator has 6 legs.
B: Oe omum. (short for Oel omum tsat) I know.

2)
A. Yom wutsot. Eat the meal.
B. Tsat li yolom. (I) already ate it.
OR
B. Li yolom. Already ate.

3)
A. Tewti! Tìng nari! Tsatskxe lu tsawl fìtxan nang! Wow! Look! That rock is soo huge!
B. Srekrr tsole'a tsat. Saw that before.
OR
B. Oel tsole'a srekrr. I've seen before.

I was pretty sure it's nothing new to leave out stuff if it's been conversationally established in discourse. I think there are a few canon examples of this in the Wiki and perhaps on the Blog.

Of course it will always be correct if the subject is included in the sentences, but I could have sworn I read somewhere that it in possible to exclude it but leave the direct object in. Another obvious example of this is modal syntax where the controlled verb is transitive and has an explicit object:

4)
Oe new tivaron yerikit.
I want to hunt hexapede.

The second clause has an implied Oel excluded, if I understand correctly.

5)
Po tsun ayramit apxa tsyivìl.
He can climb large mountains

Again.

And even more obvious:

6)
Sa'nokìl new futa tivìng sneyä eviru syuvet.
The mother wants to give her kids food.

[`Eylan Ayfalulukanä]

Very interesting-- and instructive. This is something really good to keep in mind when trying to teach the whole idea of tripartate grammar, and why we need to use case markers, sometimes in unintuitive spots. (It is helpful in understanding pro-drop and other sentence contractions) And although you could theoretically drop both subject and object in the contest of a conversation, I don't think you would want to ever drop both.

A couple of your examples are interesting:

Tsat li yolom. (I) already ate it.

This is ambiguous, because it kind of saying 'that thing already ate', where 'thing(object)' could be just about anything, but (I suspect) less likely a person. Although I can see where this might be gramatically correct, it is really 'clipping the speech'. li yolom seems to me to actually be a bit clearer, but one would, I think, more likely reply here li yolom oel.

Sa'nokìl new futa tivìng sneyä eviru syuvet.
The mother wants to give her kids food.

This looks like it should be a modal construction, but it really isn't. Or, is it? By using ìl on sa'nok, you are implying that new is being used non-modally. I have also never seen a modal construction with a futa in the middle of it. But the use of iv in tivìng is interesting, as the need for the subjunctive is very weak, due to the presence of new (for its meaning, not its modality). So, is the use of iv in tivìng due more to the wishful nature of the sentence, or more due to the fact that new is really being used modally here?

[Tirea Aean]

Very interesting-- and instructive. This is something really good to keep in mind when trying to teach the whole idea of tripartate grammar, and why we need to use case markers, sometimes in unintuitive spots. (It is helpful in understanding pro-drop and other sentence contractions)

And although you could theoretically drop both subject and object in the contest of a conversation, I don't think you would want to ever drop both.

Why?

A couple of your examples are interesting:

Tsat li yolom. (I) already ate it.

This is ambiguous, because it kind of saying 'that thing already ate', where 'thing(object)' could be just about anything, but (I suspect) less likely a person.

No, it cannot be saying "that thing already ate" because tsat is a direct object, not a subject. it explicitly has the t marker. and this sentence here is a direct response to an imperative to Eat the meal. obviously, tsat = the meal.

Although I can see where this might be gramatically correct, it is really 'clipping the speech'. li yolom seems to me to actually be a bit clearer,

How is it necessarily clearer than the one above if both subject and object are dropped?

but one would, I think, more likely reply here li yolom oel.

Sure. This is also plausible.

Sa'nokìl new futa tivìng sneyä eviru syuvet.
The mother wants to give her kids food.

This looks like it should be a modal construction, but it really isn't. Or, is it?

It's the long form. It CAN be condensed down to regular modal form. But I chose not to do that, in order to make it totally obvious that this is two clauses, where the second one is missing its agentive subject because it is implied by it being in the first clause.

By using ìl on sa'nok, you are implying that new is being used non-modally. I have also never seen a modal construction with a futa in the middle of it.

NiaN 5.3

But the use of iv in tivìng is interesting, as the need for the subjunctive is very weak, due to the presence of new (for its meaning, not its modality). So, is the use of iv in tivìng due more to the wishful nature of the sentence, or more due to the fact that new is really being used modally here?

It can PROBABLY be left out, but I think maybe due to the influence of one of those or both, I left it in.

[`Eylan Ayfalulukanä]

And although you could theoretically drop both subject and object in the contest of a conversation, I don't think you would want to ever drop both.

Why?

Because it seems weird, even in Na'vi, to drop both subject and object. Essentially, you are down to just a verb.

A couple of your examples are interesting:

Tsat li yolom. (I) already ate it.

This is ambiguous, because it kind of saying 'that thing already ate', where 'thing(object)' could be just about anything, but (I suspect) less likely a person.

No, it cannot be saying "that thing already ate" because tsat is a direct object, not a subject. it explicitly has the t marker. and this sentence here is a direct response to an imperative to Eat the meal. obviously, tsat = the meal.

I did eventually figure this out, but it was not clear at first (it was clear that tsat was the object here). What made this especially confusing was that tsat was where you would expect the subject to be (in English, anyway).

Although I can see where this might be gramatically correct, it is really 'clipping the speech'. li yolom seems to me to actually be a bit clearer,

How is it necessarily clearer than the one above if both subject and object are dropped?

Because the object being in the wrong sopt, and no subject, at least to me, made this sentence awkward. But we do end up here with a sentence with neither stated subject or object!

but one would, I think, more likely reply here li yolom oel.

Sure. This is also plausible.

And probably common sense.

Sa'nokìl new futa tivìng sneyä eviru syuvet.
The mother wants to give her kids food.

This looks like it should be a modal construction, but it really isn't. Or, is it?

It's the long form. It CAN be condensed down to regular modal form. But I chose not to do that, in order to make it totally obvious that this is two clauses, where the second one is missing its agentive subject because it is implied by it being in the first clause.

It seems to have worked. Perhaps quite well.

By using ìl on sa'nok, you are implying that new is being used non-modally. I have also never seen a modal construction with a futa in the middle of it.

NiaN 5.3

But the use of iv in tivìng is interesting, as the need for the subjunctive is very weak, due to the presence of new (for its meaning, not its modality). So, is the use of iv in tivìng due more to the wishful nature of the sentence, or more due to the fact that new is really being used modally here?

It can PROBABLY be left out, but I think maybe due to the influence of one of those or both, I left it in.

Although the expanded sentence is quite effective, I suspect you are right in that a simpler, modal construction would be fine.

In any case, this is a very refreshing way to look at agentive/patientive and how to effectively use them.

[Tirea Aean]

And although you could theoretically drop both subject and object in the contest of a conversation, I don't think you would want to ever drop both.

Why?

Because it seems weird, even in Na'vi, to drop both subject and object. Essentially, you are down to just a verb.

check out this canon example of dropping down to just a verb: http://wiki.learnnavi.org/index.php/Canon#Auxilary_verb_SI I figure it isn't too much a stretch to go down to just a verb if it's transitive too. But I suppose since I've no solid ground there, it's speculation on my part. As I said before, including every subject and every object in every sentence is obviously also correct every time.

A couple of your examples are interesting:

Tsat li yolom. (I) already ate it.

This is ambiguous, because it kind of saying 'that thing already ate', where 'thing(object)' could be just about anything, but (I suspect) less likely a person.

No, it cannot be saying "that thing already ate" because tsat is a direct object, not a subject. it explicitly has the t marker. and this sentence here is a direct response to an imperative to Eat the meal. obviously, tsat = the meal.

I did eventually figure this out, but it was not clear at first (it was clear that tsat was the object here). What made this especially confusing was that tsat was where you would expect the subject to be (in English, anyway).

Yeh, that's what happens when one relies too much on word order and expecting things to match up to familiar English grammar.

Although I can see where this might be gramatically correct, it is really 'clipping the speech'. li yolom seems to me to actually be a bit clearer,

How is it necessarily clearer than the one above if both subject and object are dropped?

Because the object being in the wrong sopt, and no subject, at least to me, made this sentence awkward. But we do end up here with a sentence with neither stated subject or object!

wrong spot? Na'vi has a wrong spot for objects to go? And my case is, with much conversational discourse at hand and a clear established context, I don't think dropping down to a verb is THAT weird. I mean, I do it in English sometimes. Albeit obviously colloquial:

A: Hey did you do you clean your room?
B: Yes
A: Do your homework?
B: ((Already)) Done.
A: Sweep off the porch?
B: Swept.
A: Laundry done, dishes done, floors clean?
B: Done, done, and cleaned.

but one would, I think, more likely reply here li yolom oel.

Sure. This is also plausible.

And probably common sense.

I don't know about common sense. Just because it seems like common sense to you. I mean, It makes sense to me, obviously. But it's not immediately obvious to me whether it's more common to leave out the subj. or the d.o. or both or neither. All I know is what this community seems to do often. Which is sort of a mix between them all, honestly.

Sa'nokìl new futa tivìng sneyä eviru syuvet.
The mother wants to give her kids food.

This looks like it should be a modal construction, but it really isn't. Or, is it?

It's the long form. It CAN be condensed down to regular modal form. But I chose not to do that, in order to make it totally obvious that this is two clauses, where the second one is missing its agentive subject because it is implied by it being in the first clause.

It seems to have worked. Perhaps quite well.

Coolness

By using ìl on sa'nok, you are implying that new is being used non-modally. I have also never seen a modal construction with a futa in the middle of it.

NiaN 5.3

But the use of iv in tivìng is interesting, as the need for the subjunctive is very weak, due to the presence of new (for its meaning, not its modality). So, is the use of iv in tivìng due more to the wishful nature of the sentence, or more due to the fact that new is really being used modally here?

It can PROBABLY be left out, but I think maybe due to the influence of one of those or both, I left it in.

Although the expanded sentence is quite effective, I suspect you are right in that a simpler, modal construction would be fine.

Yeah. It's actually much more common. Hence you apparently haven't seen it the expanded way. And new is probably the only time you'll see it. Now, to be honest, the reason why it's so rare to see it expanded out in taht exact situation is, because teh subject is actually the same in both clauses. The futa expansion is actually required when the second clause has a different subject than the first:

oel new futa po kä. etc.

In any case, this is a very refreshing way to look at agentive/patientive and how to effectively use them.

Oe mllte. Kemaweyan definitely seemed to have a point there that I had never even thought of before: Ambiguity in the F words in such a situation.

[Plumps]

Yeah. It's actually much more common. Hence you apparently haven't seen it the expanded way. And new is probably the only time you'll see it. Now, to be honest, the reason why it's so rare to see it expanded out in taht exact situation is, because teh subject is actually the same in both clauses. The futa expansion is actually required when the second clause has a different subject than the first:

oel new futa po kä. etc.

... and ‹iv› as well, as far as I remember (see here).

I could be wrong but I've never seen an official sentence with the {Subject}ìl new futa construct that didn't include the ‹iv› in the subordinate verb.

[Tirea Aean]

Yeah. It's actually much more common. Hence you apparently haven't seen it the expanded way. And new is probably the only time you'll see it. Now, to be honest, the reason why it's so rare to see it expanded out in taht exact situation is, because teh subject is actually the same in both clauses. The futa expansion is actually required when the second clause has a different subject than the first:

oel new futa po kä. etc.

... and ‹iv› as well, as far as I remember (see here).

I could be wrong but I've never seen an official sentence with the {Subject}ìl new futa construct that didn't include the ‹iv› in the subordinate verb.

Huh. Now I think about it and look around, this is quite true..

[Tirea Aean]

So it seems we've never come to a clear conclusion to the problem in OP.

I introduce a slight variant, this time, it's my opinion that -l cannot be used to circumvent the ambiguity:

Sunu oeru fwa oe tsun tsive'a.

Is this saying I like the very fact that I am able to see? Or is it saying I like the object which I am able to see?

I sent an email to Paul about this on Friday. We shall see what the answer is soon, I hope.

[Blue Elf]

Well, some guidance is needed for such ambiguous sentences.
What is interesting to me is how people still want to drop something to make statement shorter (or "better"  )

my guidance: do you want to tell your idea clearly? Use long (full) form.
do you want to have good discussion? Use shortest form possible. 

Note: one small sign was missing here
Sa'nokìl new futa tivìng sneyä 'eviru syuvet.

[Tanri]

Sunu oeru fwa oe tsun tsive'a.
Is this saying I like the very fact that I am able to see? Or is it saying I like the object which I am able to see?

The first one seems correct. "I like the object which I am able to see" should be "Sunu oeru 'u a (tsat) oe tsun tsive'a"

I hope the full lenght "fi'u a" (and alike) can be used if we want to distinguish between attached subclause and the noun (this thing which...):
Li smon oeru fì'u a new nga wivìntxu ayoer, slä vivar rutxe, sutel alahe ke li ke tsole'a tsat. - I am already familiar with this thing you want to show us, but continue please, the others haven't seen it yet.
Smon oeru fwa new nga wivìntxu 'uoti oer, slä pivey nì'it, pxiset ke lu oeru krr letam. - I am familiar with the fact that you want to show me something, but wait a little, right now I have not enough time.

[Tirea Aean]

The answer to this is come, but I don't have release permission yet. Hang tight, ma frapo.