Lu and the Dative, and the Epistemology of Na'vi

Started by Taronyu, January 16, 2010, 12:20:21 PM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

Taronyu

Kaltxì! Ngaru lu fpom srak?
Hello! you-DAT is well.being yes?
Hello? How are you?

For those of you who don't know, this one sentence has given us no small bit of grief. Consider other possible translations:

Hello! Is well.being attributed to you?
Hello! Are you [the quality] of good?
Hello! Is good to you?
Hello! Do you have goodness?

Well, I was in class the other day, gently falling asleep, when my lecturer made an interesting point. Consider the following Icelandic sentence:

Henni leiddust strákarnir
her-DAT bored-3PL boy-PL NOM
The boys bored her.

Well, obviously I had an issue. Why is she in the dative? And it turns out that there's an interesting reason. Many attested languages have dative subject construction: the subject of a clause can go into the dative case.

Quote from: WikipediaThe dative construction is a grammatical way of constructing a sentence, with the subject in the dative case and the direct object in the nominative case. A sentence is also said to be in dative construction if the subject and the object (direct or indirect) can switch their cases for a given verb, without altering the verb's structure (subject becoming the new object, and the object becoming the new subject). The latter case is not to be confused with the passive voice, where only the direct object of a sentence becomes the subject of the passive-voiced sentence, and the verb's structure also changes to convey the meaning of the passive voice. The dative construction tends to occur when the verb indicates a state rather than an action.

Note that last bit! This tends to happen only when a verb indicates a state, and not an action. I want you to look at this bit of examples that I found in a paper on Dative constructions in Marathi and Urdu:

Dative Subjects with Copula (Possessive Structures)
NP-DAT ..... NP-NOM (N-)HONA(BE) These constructions denote (temporary) states.

a. mUjhe   bUxar   hE
Pron.1.Sg.Dat fever.M.Nom be.Pres.3.Sg
I have a fever.

b. mUjhe   kıtabe ̃   pAsAnd hE ̃
Pron.1.Sg.Dat book.F.Pl.Nom liking   be.Pres.3.Sg
I like books.

c. mUjhe   (ye) xAbAr/pAtA/malum hE
Pron.1.Sg.Dat this.Nom news/address/knowledge be.Pres.3.Sg
I have news/address/knowledge (of this).

You can see here, clearly, that the Dative noun takes the dative case, as it is the experiencer. The attributed noun that uses the copula is put in the nominative case: which, we all know, isn't marked in Na'vi, as they only mark the Agentive case when there is an ergative direct object. For more definitions on what an experience is and can be, refer to this:

Quote from: The Previously Mentioned Paper
Experiencers are defined as the sentient argument which is mentally affected.
They are taken to occur mainly with psychological and perception predicates.
It has long been established that experiencers are mental/abstract locations of some sort.
The idea that experiencers are (mental) locations and that the experiencer of psych-predicates is a locative of some sort [is explored elsewhere].

So, what am I suggesting? Based on what we know: that using ___-ru lu ___ only works when you're talking about a state of someone. Essentially, it isn't used for possession. You cannot use lu to mean have. See below for more arguments.

Ngaru tìtslam lu srak?

Ftiafpi

This makes A LOT of sense, excellent find. I don't know about you but I'm going to consider this correct for Na'vi as well until we know otherwise since it, as you said, solves so many issues with ngaru lu fpom srak

NeotrekkerZ

For any German speakers, I believe this also happens when you say

Es gefällt mir gut which translates as " I like it" but literally means "it pleases to me good."
Rìk oe lu hufwemì, nìn fya'ot a oe tswayon!

Taronyu

Quote from: neotrekkerz on January 16, 2010, 01:27:11 PM
For any German speakers, I believe this also happens when you say

Es gefällt mir gut which translates as " I like it" but literally means "it pleases to me good."

You're right.
Quote from: WikipediaIn German, the dative construction sometimes occurs with the verb sein ("to be"). Compare:

    Ich bin kalt ("I am cold")
    Mir ist kalt (literally means, "cold is to me.")

The first example implies that the speaker has a cold personality. The subject here (ich, "I") is in the nominative case. The second construction is used when one wants to say "I am (feeling) cold" in German. While in English the subject of the sentence "I am cold" is "I", in German the subject of the sentence "Mir ist kalt" is kalt ("cold"). The indirect object here is mir, the dative case of the pronoun ich in German.

wm.annis

Quote from: Taronyu on January 16, 2010, 12:20:21 PMSo, what am I saying? That using ___-ru lu ___ only works when you're talking about a state of someone. Essentially, it isn't used for possesion. You cannot use lu to mean have.

Our corpus is still way too small to make such blanket assertions.  What you have described is certainly a real possibility, but without more Na'vi data specifically involving possession and state to consider, no one has any business asserting grammar rules about this, much less in bold face type.

Taronyu

Quote from: wm.annis on January 16, 2010, 02:02:35 PM
Quote from: Taronyu on January 16, 2010, 12:20:21 PMSo, what am I saying? That using ___-ru lu ___ only works when you're talking about a state of someone. Essentially, it isn't used for possesion. You cannot use lu to mean have.

Our corpus is still way too small to make such blanket assertions.  What you have described is certainly a real possibility, but without more Na'vi data specifically involving possession and state to consider, no one has any business asserting grammar rules about this, much less in bold face type.

I disagree. I think that we can make statement's like this, as long as we preface everything we say with: From what we know. Unless we have examples, I don't think that we can use lu to mean have. Do we have any examples of this? Not that I, personally, know of.

Doolio

QuoteSo, what am I saying? That using ___-ru lu ___ only works when you're talking about a state of someone. Essentially, it isn't used for possesion. You cannot use lu to mean have.
i agree with this, but i must say that i was treated as a ghost when i said exactly the same thing couple of weeks ago. although we can't be sure, as we do need frommer to clear it up, i pointed out that in various languages it is nonsensical to say "to me is arrow", but you can say something like "to me is sickness".

bottom line is, we can't be sure of anything, and that is a fact. but also, we should have the ability to analyze what was given to us and develop theories and be ready to change them accordingly. but, in this example, taronyu was not jumping to conclusion, everyone else was. everyone made the canon of a "oeru lu" construct and applied it to possession without frommerisation.
that confuses me, because how then taronyu could make blanket assertions, when he is, in fact, more cautious than everybody else? he is actually suspicious about the generalisation of the usage of "oeru lu" construct, unlike the others.

my two cents...
...taj rad...

Taronyu

Doolio, sorry if we as a community ignored you.

As for me being conservative, it comes down to the two theories of the language that William and I have and disagree on.

William: Thinks that we should not make definitive rules for the language, but hold everything in balance unless and until we have words from Frommer about it. I think he wants to keep it like Ancient Greek or Latin - where the rules cannot be changed, and cannot be postulated without looking at the complete corpus. (Correct me if I'm wrong, William.)

Me: thinks that we should take all of the knowledge we have and drag every shred of evidence for rules or grammatical constructs out of it. I think that we should then reanalyze these, when given Frommerian information. I want the language group to learn the Na'vi that we have, without making assumptions, right now, and relearn later, if we're wrong. I also think that we, as a language group, should take the language and run with it, having as much fun making new rules and words according to what we know, to let it grow organically like a real, living language - but that's not really a goal of mine, yet, as we don't have enough information from Frommer to make this feasible.

What I do want is for people to know what they can or cannot do, now, based on the data: this is one of those cases. You can use lu and the dative to describe a state, you cannot use it to describe possession. So far.

Doolio

Quotethinks that we should take all of the knowledge we have and drag every shred of evidence for rules or grammatical constructs out of it. I think that we should then reanalyze these, when given Frommerian information. I want the language group to learn the Na'vi that we have, without making assumptions, right now, and relearn later, if we're wrong.
i agree with this, as i consider myself warry and adaptable, but this requires for a learner to always keep that in mind.
...taj rad...

omängum fra'uti

I think I'd word things a little differently than to say it "isn't for possession".  As William rightfully pointed out, we don't know if that is the case, and I agree with him that we should be careful what we declare without something to back it up.

Consider these 3 statements...


  • We can use the lu with the dative to indicate something has a certain state or condition.
    That seems like a pretty safe statement, since we have two examples of that as the use (Forgiveness & well being)
  • We don't know if you can use lu with the dative to indicate somebody physically possesses an object.
    Perfectly valid as well.  We have no examples of this, but we have nothing indicating otherwise either, aside from the fact that it is how some languages do it.
  • You cannot use lu to mean have.
    Absolutely unsubstantiated.  Just because that is how some languages do it does not mean it is how Na'vi does it.  As long as there are some languages which treat it as actual possession, there is the possibility that Na'vi does as well.

That is what William is saying, and I tend to agree with him.
Ftxey lu nga tokx ftxey lu nga tirea? Lu oe tìkeftxo.
Listen to my Na'vi Lessons podcast!

wm.annis

Quote from: Taronyu on January 16, 2010, 02:28:19 PM(Correct me if I'm wrong, William.)

Basically wrong, but I wonder if this is really the right thread to derail to go into it.  :)  It's more a philosophical issue than a linguistic one, really.

QuoteI think that we can make statement's like this, as long as we preface everything we say with: From what we know.

Which you did not do, until several replies in.  A little care in this seems like a good idea, as a warning to beginners.

Doolio

#11
Quoteperfectly valid as well.  We have no examples of this, but we have nothing indicating otherwise either, aside from the fact that it is how some languages do it.

so why is then taronyu's decision more "aggressive" than yours, it seems to me that he is playing it safe, and that you are generalizing the rule applying it to possession as well?


of course, i am NOT implying anything, just saying that there is another aspect in this matter which i think should not be underrated and it should be treated the same as the other opinion.
basically, if there are SOME languages that use "oeru lu" for physical possession, and there are SOME languages in which that is incorrect to do, both opinions matter the same, as i can see. all that i see from this is that taronyu is more cautious.
...taj rad...

Taronyu

Quote from: wm.annis on January 16, 2010, 02:54:34 PM
Quote from: Taronyu on January 16, 2010, 02:28:19 PM(Correct me if I'm wrong, William.)

Basically wrong, but I wonder if this is really the right thread to derail to go into it.  :)  It's more a philosophical issue than a linguistic one, really.

QuoteI think that we can make statement's like this, as long as we preface everything we say with: From what we know.

Which you did not do, until several replies in.  A little care in this seems like a good idea, as a warning to beginners.

I think this might be the right thread. It is more of a philosophical issue. We could make a new thread? Or just keep bringing it up once a week like we've been doing, haha. I hope I'm not being ignorant, by the way. I think I have valid claims, but I may be missing something.

As for the caveat: I edited the post. Good point, my bad, changed.

omängum fra'uti

You have as valid claims as the claims that lu+-ru can work for physical possession as well.  In other words - neither way is substantiated by any examples we have, and both represent how it is done in other languages, which means given how Frommer designed Na'vi either could be true.
Ftxey lu nga tokx ftxey lu nga tirea? Lu oe tìkeftxo.
Listen to my Na'vi Lessons podcast!

Taronyu

#14
Quote from: omängum fra'uti on January 16, 2010, 03:16:23 PM
You have as valid claims as the claims that lu+-ru can work for physical possession as well.  In other words - neither way is substantiated by any examples we have, and both represent how it is done in other languages, which means given how Frommer designed Na'vi either could be true.

I want to know where those other valid claims are in the data. I do believe that lu + ru can be used as I said above, but I want to know where the examples are that you using to infer that it can be used as lu + ru = possess physically. (Sorry if that sounds a bit snarky).

Quote from: Doolio on January 16, 2010, 02:57:15 PM
Quoteperfectly valid as well.  We have no examples of this, but we have nothing indicating otherwise either, aside from the fact that it is how some languages do it.

so why is then taronyu's decision more "aggressive" than yours, it seems to me that he is playing it safe, and that you are generalizing the rule applying it to possession as well?


of course, i am NOT implying anything, just saying that there is another aspect in this matter which i think should not be underrated and it should be treated the same as the other opinion.
basically, if there are SOME languages that use "oeru lu" for physical possession, and there are SOME languages in which that is incorrect to do, both opinions matter the same, as i can see. all that i see from this is that taronyu is more cautious.

Doolio, I am being more aggressive because I am making a statement that is closed, here. I am saying that the data supports this usage, while the poss. usage is something that others are inferring as possible. I am ruling out that possibility, essentially saying that it is wrong, which can be construed as being more aggressive.

Na'rìghawnu

Quotethat confuses me, because how then taronyu could make blanket assertions, when he is, in fact, more cautious than everybody else?

Well ... seems to be in a little contrast to the image Taronyu has about himself. But of course there are good points in both theories Taronyu mentions.

I admit, that I mostly share the opinion Taronyu ascribes to William. According to my own convictions it's much better to announce just things that we really know and to keep speculations at a minimum level (at least I feel very uncomfortable about the wide propagation of the socalled "results" of such speculations in materials). I'm convinced, that it is much better to mark unknown things as unknown und to speculate about it only under a big "CAVEAT!", that means, that we shouldn't presume to make "rules" or a lot of "derivations" and propagate them, because in most cases we simply don't have enough sources. Mostly it's just a single word or a single phrase. And nevertheless there are many people, trying to formulate "rules" based on this one appereance. That's simply not serious philology ... besides the problem, that someday we probably will have to UNLEARN and then to RELEARN things. And that's not so easy, when you used your self-build-rule maybe a long time, constructing other rules upon it and so on.

On the other hand, there is of course the problem, that the very little material we got until now, is simply not enough to satisfy the needs of people, who want to actually use the language by themselves. It's natural, that they want to know "rules", words and so on. I can imagine, that - if we'd follow only the orthodox method described above - many people would lose interest in the language ...

Well. It seems, that different standpoints are vital to the Na'vi-community in order to retain momentum, but nevertheless not to make too many assumptions and mistakes.

omängum fra'uti

Quote from: Taronyu on January 16, 2010, 03:18:15 PM
Quote from: omängum fra'uti on January 16, 2010, 03:16:23 PM
You have as valid claims as the claims that lu+-ru can work for physical possession as well.  In other words - neither way is substantiated by any examples we have, and both represent how it is done in other languages, which means given how Frommer designed Na'vi either could be true.

I want to know where those other valid claims are in the data. I do believe that lu + ru can be used as I said above, but I want to know where the examples are that you using to infer that it can be used as lu + ru = possess physically. (Sorry if that sounds a bit snarky).
You can find them in the same place you can find the examples which you are using to infer that it can not be used as physical possession.

Which is to say, neither exist.  We have absolutely 0 proof either way from our samples of official Na'vi dialog.  So we have to draw on how other languages do it...  And other languages can go either way.
Ftxey lu nga tokx ftxey lu nga tirea? Lu oe tìkeftxo.
Listen to my Na'vi Lessons podcast!

Taronyu

Quote from: omängum fra'uti on January 16, 2010, 03:51:51 PM
Quote from: Taronyu on January 16, 2010, 03:18:15 PM
Quote from: omängum fra'uti on January 16, 2010, 03:16:23 PM
You have as valid claims as the claims that lu+-ru can work for physical possession as well.  In other words - neither way is substantiated by any examples we have, and both represent how it is done in other languages, which means given how Frommer designed Na'vi either could be true.

I want to know where those other valid claims are in the data. I do believe that lu + ru can be used as I said above, but I want to know where the examples are that you using to infer that it can be used as lu + ru = possess physically. (Sorry if that sounds a bit snarky).
You can find them in the same place you can find the examples which you are using to infer that it can not be used as physical possession.

Which is to say, neither exist.  We have absolutely 0 proof either way from our samples of official Na'vi dialog.  So we have to draw on how other languages do it...  And other languages can go either way.

This is where I disagree. I think that Ngaru lu fpom is used in a psychic sense. It is not used in a physical sense. I think that it is a serious step to suggest that it can be: this is a step which others are willing to take, but I am not. Granted, it may be possible, but I don't think that we should use it this way, because I think it is assuming that it can be.

wm.annis

Quote from: Doolio on January 16, 2010, 02:57:15 PMso why is then taronyu's decision more "aggressive" than yours, it seems to me that he is playing it safe, and that you are generalizing the rule applying it to possession as well?

Because his analysis is no basis to make a negative rule without a lot more data.  Cross-linguistically, many languages distinguish between alienable and inalienable possession.  I could just as easily make the case that X-ru Y lu is for alienable possession only, and forbid you to say "I have two eyes" using the construction.  Or I could claim it has to do with an animacy hierarchy (a little strained, but possible), and forbid you to say "I have an ikran" with it.

Taronyu's state interpretation, and my two offerings here, all fit the data we have very neatly, and we can find plenty of human cross-linguistic evidence for all of them.  It would not surprise me in the least if Taronyu's analysis turns out to be correct, but without more data, we have no good way to decide between them.

wm.annis

Quote from: Na'rìghawnu on January 16, 2010, 03:26:02 PMOn the other hand, there is of course the problem, that the very little material we got until now, is simply not enough to satisfy the needs of people, who want to actually use the language by themselves. It's natural, that they want to know "rules", words and so on. I can imagine, that - if we'd follow only the orthodox method described above - many people would lose interest in the language ...

This is the central tension for any community forming around the language right now.  Hopefully we'll be better off than the elves soon.