Analyzing the "chat with you" sentence

[wisnij]

We have this sentence in the corpus:

Tsun oe ngahu nìNa'vi pivängkxo a fì'u oeru prrte' lu
It's a pleasure to be able to chat with you in Na'vi.

Because of its similarity to plltxe, I've been assuming that prrte' means "speak/chat".  However, I can't seem to build a sensible syntax tree out of the sentence in that case.  Now, there's no a priori reason why Na'vi should be built on tree structures like human languages typically are, but since we know the language is learnable by humans it must not violate any human language universals (assuming those exist... but that's another thread ).  It could also be that constructions of this form move the verb out of its normal position in the clause for some reason, somewhat akin to how English pulls relative pronouns out to the front ("The friend who I traveled with" == "The friend [I traveled with him]").

But looking at it again today, I realized it made more sense, at least to me, if was wrong before.  If pivängkxo is "chat" and prrte' is "pleasure/pleasing", then we could analyze it like this:

[[Tsun [oe ngahu nìNa'vi pivängkxo]] a fì'u] oeru prrte' lu
[[Able to [I chat with you in Na'vi]] fact] is pleasing to me

In this case, it seems like "X a fì'u" is effectively nominalizing the whole X clause, much as Japanese koto "thing" can do.  Or to build it up piece by piece:

*Oe ngahu nìNa'vi pivängkxo.
I chat with you in Na'vi.

*Tsun oe ngahu nìNa'vi pivängkxo.
I can chat with you in Na'vi.

*Tsun oe ngahu nìNa'vi pivängkxo.  Fì'u oeru prrte' lu.
I can chat with you in Na'vi.  This fact is pleasing to me.
(Fun fact: apparently the human language Pirahã exclusively expresses things this way, without any direct embedding.  At least according to some linguists.)

[Tsun oe ngahu nìNa'vi pivängkxo] a fì'u oeru prrte' lu
The fact that [I can chat with you in Na'vi] is pleasing to me.

Thoughts?

[Brainiac]

I agree, except for the fact that the prrte'lu really does seem to refer to the speak...
If not, I competely agree.
I've been trying to use the syntax tree as well, but I was... less successful.
Irayo, ma lerteiu

[Taronyu Ayunilyä Alahe]

hey hey look at the bottom of:
http://james-camerons-avatar.wikia.com/wiki/Na%27vi_Language

they said:
"Speak" is listed as pllte'. It is not clear if the word has been changed, or if prrte' is an inflection of pllte'.

and in:
http://www.suburbandestiny.com/

they said:
plltxe is speak


I'm just trying to give some more resources for wisnij since she's not sure about those and I happen to be opening those links (which I do everyday now). I think those sites are pretty reliable, but I'm not sure myself.

[wisnij]

I'm well aware that plltxe means "speak", as that's how Dr. Frommer used it in the Language Log post comments.  What we don't know is whether plltxe and prrte' are actually related despite their apparent similarity.  If there's a directly attested example of prrte' being used to mean "speak", please let me know.  I'm regarding the wikia site as uncertain because their glosses are even shakier than ours and there appear to be some transcription errors in the word list.

[Taronyu Ayunilyä Alahe]

oh right... sorry for the misunderstanding then.
well I'll see if I can get any future references 

[wm.annis]

wisnij, this is an excellent analysis!  That sentence was driving me nuts for days.

I agree that there's no good reason to see plltxe and prrte' as related.

I'm inclined to see the stem of "chat" as *pängkxo (or maybe even just *päng), with <iv> as yet another instance of a dependency marker, which we know can fit into the same slot as the aspect and tense markers.  In the oral examples from the Skxawng! interview, we have the phrase "those things can kill" with tsun once again causing an <iv> infix (into tspang).

[wisnij]

I'm inclined to see the stem of "chat" as *pängkxo (or maybe even just *päng), with <iv> as yet another instance of a dependency marker, which we know can fit into the same slot as the aspect and tense markers.  In the oral examples from the Skxawng! interview, we have the phrase "those things can kill" with tsun once again causing an <iv> infix (into tspang).

Agreed.  For that matter, -kxo might even be a suffix that explicitly nominalizes the verb phrase, and then a fì'u just binds to it.

[wm.annis]

Oh, man.  We have three examples of tsun causing that infix.  See Frommer's email to Prrton (lu tute a tsun nì-Na'vi set fìfya pivlltxe).

[Txur’Itan]

wisnij, this is an excellent analysis!  That sentence was driving me nuts for days.

I agree that there's no good reason to see plltxe and prrte' as related.

I'm inclined to see the stem of "chat" as *pängkxo (or maybe even just *päng), with <iv> as yet another instance of a dependency marker, which we know can fit into the same slot as the aspect and tense markers.  In the oral examples from the Skxawng! interview, we have the phrase "those things can kill" with tsun once again causing an <iv> infix (into tspang).

I am thinking that this is undocumented lenition.

That coupled with the plot idea, that the human speakers in the cast were supposed to have a flawed library of phrase construction.

[wm.annis]

What a child I've become!  We have three new sentences from Frommer today — it's like Christmas presents came early!  Relevant to this thread, we have yet another <iv> infix, and I think we can safely add a new rule of syntax to our knowledge.

Txo	new	nga	r<iv>ey,	oe-hu!
if	want	you	live<INF>	me-with

So.  Here Frommer spells "want" as new, rather than the neu in the word list we have now.  I'm not sure which is correct — they will sound very similar, so it would be easy to confuse them.

It looks like modal verbs (tsun can, be able to, new want to, and I'd bet zene must), when they take verbal complement, that complement must have the <iv> infix.

[Prrton]

It looks like modal verbs (tsun can, be able to, new want to, and I'd bet zene must), when they take verbal complement, that complement must have the <iv> infix.

I'm dubious that the "iv" infix would also be required.

Ke fparmìl oel futa lu tute a tsun nì-Na'vi set fìfya pivlltxe!

I didn't think there was anyone who could speak Na'vi like that at this point!

tsun(helping) ... pivlltxe(primary (plltxe))

I simply think that Na'vi marks the conditional on the primary verb and not the "helping" verb.

If English worked a bit differently the Dr. may have commented.

I didn't think that there was anyone who *[can spoke]-(SUBJ) Na'vi like that at this point.

(Of course, "I didn't think that there was anyone who spoke-SUB Na'vi like that at this point" is completely valid, srak?

I'm dubious that "iv" is required per se. At least in all cases.

I just bought the Survival Guide and it includes the Hunt Song. Someone has probably already posted it here somewhere, so I won't type the whole thing out at this point unless requested, but here is an example of "iv" in conjunction with "zene".

Livu win sì txur oe zene
be(iv) fast and strong I must.

But the context of the song includes an <if-(only if)/then> dependency.

I must be fast and strong (so that I can hunt well and be the one worthy to take your life so that you can feed the People).

Dependency. Conditionality. Possibility that the outcome is not assured.

Ha, pxan livu txo nì'aw oe ngari, tsakrr nga Na'viru yomtìyìng
so, worthy be(iv) if only I you-TOP, that-time (=then) you The People-DAT feed.
So, only if I am worthy of you will you feed the people.

I do feel that the Na'vi would be much more likely to pervasively use the subjunctive. They are very religious (in the sense of "bound to Eywa" - literally. physically.) and can never know exactly how Eywa might affect/effect outcomes.

[edmoreira]

I totally agree with wisnij's first analysis. That's how I interpreted and assumed that pivängkxo had to be something along the lines of talk or chat. If we assume that prrte' is an inflected form of pllte' and therefore refers to the speak part, the whole sentence doesn't make any sense. We know that according to Frommer, Na'vi does follow human languages' grammar.
It's frustrating. In any case I made a youtube video saying the phrase and I interpreted as pivängkxo being to chat, and fi'u oeru prrte'lu being this is pleasing to me.
Cheers
Ed

[wisnij]

What a child I've become!  We have three new sentences from Frommer today — it's like Christmas presents came early!  Relevant to this thread, we have yet another <iv> infix, and I think we can safely add a new rule of syntax to our knowledge.

Txo	new	nga	r<iv>ey,	oe-hu!
if	want	you	live<INF>	me-with

So.  Here Frommer spells "want" as new, rather than the neu in the word list we have now.  I'm not sure which is correct — they will sound very similar, so it would be easy to confuse them.

It looks like modal verbs (tsun can, be able to, new want to, and I'd bet zene must), when they take verbal complement, that complement must have the <iv> infix.

I wonder if it's a general infinitive, or something specifically used for making subordinate clauses.

Also, is it just me or are all the examples we've seen so far with <iv> have the verb marked with it come at the end of its subclause?  It might be early to generalize, but I know German and Japanese do basically the same thing, so that might end up being a rule in Na'vi too.

I'm dubious that the "iv" infix would also be required.

Ke fparmìl oel futa lu tute a tsun nì-Na'vi set fìfya pivlltxe!

I didn't think there was anyone who could speak Na'vi like that at this point!

tsun(helping) ... pivlltxe(primary (plltxe))

I simply think that Na'vi marks the conditional on the primary verb and not the "helping" verb.

I would gloss that sentence as

Ke	fparmìl	oe-l	futa	lu	tute	a	tsun	nì-Na'vi	set	fì-fya	p<iv>lltxe!
not	think(?)	1-A	(?)	be	person	that	able	ADV-Na'vi	now	like this	speak<iv>

Even with the gaps (which might be things we actually know, I'm just tired -_-) it still seems like the end would syntactically be broken out like:

...tute a tsun [nì-Na'vi set fì-fya plltxe]
...person who is able [speak in Na'vi like this now]

So plltxe is the subordinate to tsun, not the primary.  (In a larger sense, I suppose lu is really the primary overall; tsun just heads a relative clause that modifies tute, so it's sort of in the middle of three levels total.  (Or is it four with fparmìl?  (Argh, it's too late to deal with this much recursion.)))

[omängum fra'uti]

I just realized you can hear Frommer himself recite nearly exactly this phrase.  From the ABC clip referenced in this other thread at about 3:55 he says "Tsun oe ngahu pivängkxo a fì'u oeru prrte' lu".  Of course following his nice calm recital of that is a clip of Neytiri shouting at Jake.

Of course now I'm trying to figure out what she is shouting.

[wm.annis]

Let me start with this:

I do feel that the Na'vi would be much more likely to pervasively use the subjunctive. They are very religious (in the sense of "bound to Eywa" - literally. physically.) and can never know exactly how Eywa might affect/effect outcomes.

In his UGO Movie Blog interview, Frommer explicitly denies that he uses grammar to encode cultural values:

Jordan Hoffman: Well, this leads to an actual question – the Na'vi philosophy is a very natural, holistic way of life. Those blue suckers are very green. Did this in any way inform the construction of the language?

Paul Frommer: No. Only with certain concepts that I knew needed to be in there like "Tree of Souls" or "Hometree" and some of the religious concepts of Eowah. Otherwise, no, I just wanted exotic sounds to the Western ear – the ejectives the [proceeds clicking and making beat box sounds] and something fun for the audience. There is really no connection between the grammatical structure or aural quality of a language with the culture of a people.

So, I see no reason to expect Na'vi to be unexpectedly full of subjunctives.

This morning I ran out and got the Activist Survival Guide.  Now that I've had a chance to see the two songs in there, I agree with you — there's no doubt that <iv> can be used for subjunctive functions.  But I still have a very hard time fitting all uses of it into that scheme.  No matter how I squint, tilt my head or move my glasses around, I can find no sense of irrealis in the (audio from NY Times Magazine) tsun tutet tspivang (-ko) they can kill a person, you know.  Also, in no language I have ever studied, or even just read a grammar on, do modal verbs like "can, must, want" operate in the way you've suggested.  If you know of an example, please let me know.

So, right now I don't have a way to fit <iv> into all the uses we've seen.  Frommer spent two years in Malaysia in the Peace Corps.  I may try to hunt down a precis of Malay grammar, to see if there is any clue there about this matter.  It doesn't come from Persian.

[Prrton]

Let me start with this:

I do feel that the Na'vi would be much more likely to pervasively use the subjunctive. They are very religious (in the sense of "bound to Eywa" - literally. physically.) and can never know exactly how Eywa might affect/effect outcomes.

In his UGO Movie Blog interview, Frommer explicitly denies that he uses grammar to encode cultural values:

Jordan Hoffman: Well, this leads to an actual question – the Na'vi philosophy is a very natural, holistic way of life. Those blue suckers are very green. Did this in any way inform the construction of the language?

Paul Frommer: No. Only with certain concepts that I knew needed to be in there like "Tree of Souls" or "Hometree" and some of the religious concepts of Eowah. Otherwise, no, I just wanted exotic sounds to the Western ear – the ejectives the [proceeds clicking and making beat box sounds] and something fun for the audience. There is really no connection between the grammatical structure or aural quality of a language with the culture of a people.

So, I see no reason to expect Na'vi to be unexpectedly full of subjunctives.

Oeru txoa livu, ma wm.annis. Ngengaru ohel zoplo sìmi a txele livu futa lerängam oheru.  

I should have clarified my feelings regarding "respect" (for Eywa and therfore almost everything on Pandora) in the Na'vi culture for all things as my core assertion. In my experience with languages of cultures that have complex social structures or religions that engender a sense of showing respect being better than not showing respect, this "attitude" (which I will not go so far as to assert as full on "culturally encoded") tends to show up in the verbs.

Spanish:

Si usted me pueda (present 3rd.p sing. subjuntive of «poder» "to be able to") ayudar ("to help"), es seguro que tendrá las bendiciones de Eywa siempre.

If you (could possibly) help me, it is certain that you will have the blessings of Eywa forever.

No se si usted me pueda ayudar con mi Na'vi, o si conoce a alguien que me pueda instruir...

I don't know if you (might possibly) help me with my Na'vi, o if you know anyone who (perhaps may be able) to instruct me...

Dudo que mienta acerca de los Na'vi.

I doubt that you (should/might/would lie (present 3rd.p sing. subjuntive of «mentir» "to lie") about the Na'vi (people).

All of these verbs could be in regular old present 3rd.p sing. in common speech without changing the core meanings of the sentences. They show more deference this way. (Although, a language instructor would probably not accept something other than mienta after "dudo que" as a function of an examination. Sorry not to be 100% confident. My Spanish grammar was a LONG time ago...)

Japanese:

"Wait a moment"

to family or close friend or coworker:
ちょっと待って
chotto matte
brief-period wait-GERND (used as informal requet/demand)

to customer by department store employee or telephone receptionist
少々お待ちになって頂けませんでしょうか
shōshō o-machi-ni-natte itada-ke-masen-deshō-ka
brief-period-POL HON-wait-HON receive[intrinsically-humble]-CAPBL-NEG-PROB-COPULA-INTRROG
LIT: "few-few (you)become-an-honorable-sate-of-waiting not-posible-for-me-to-humbly-receive is-it?"
"I humbly ask if you (might or) might not be willing to do me the favor of waiting briefly."

It is, of course, an extreme, (and on the decline) but not out of the realm of possibility.

All of the "politeness" hangs on the complexity and conditional/subjunctive atmosphere in the verbs.

"May peace be with you" is (a tad) more complex and more "conditional/subjunctive" than simply "Peace be with you."

And please forgive my feeble attempts to follow valid glossing conventions. I'm an "armchair linguist" at best.

I saw (both in the sense of "kame" and "tse'a") this attitude of respect in Neytiri's heartfelt euthanizing of the viperwolves (nantang) that Jake had injured out of his own inexperience on/with Pandora. I'm sorry not to have her "apology" here as an example, but it's at least partially represented (conceptually) in the Hunt Song (and I may again be mistaken, but I believe in the ABC interview she comes up with an ad hoc version. I have not had the time to aurally parse that yet).

Also, in no language I have ever studied, or even just read a grammar on, do modal verbs like "can, must, want" operate in the way you've suggested.

In my imprecise (naïve?) explanation of my suggestion/assertion, I'm afraid that I may have conveyed some sense or proposal that I did not intend.

Nìmun, oeru ngengayä txoa livu.

Without the grammar documentation on these "et alii" we really cannot know about the different registers and their relationship to particles like -kxo (friendly/casual?... passive...) and contractions and the like. The "abbreviated" greeting response he gave NPR's "ma Renée" was "Kamengat" (which she struggled with—briefly). I see that as a clear contraction of "(oel) kame+ngat(i)" with no infix. The songs in the Survival Guide also give "-it" as an apparently accusative noun suffix. Is that some as yet mysterious -i-t(i) that is "standard"? Contextual metathesis based on some kind of poetic license? Who knows...? Oeru ke law nang!

Eywa ngengahu

[Taronyu]

This is just a footnote, but I wouldn't be surprised if Frommer made the word for pleasure as prrte'. He knew he was writing this for feline humanoids, and it would make sense to make a prr sound to mean pleasure. Basic onomatopoeia.

[Prrton]

This is just a footnote, but I wouldn't be surprised if Frommer made the word for pleasure as prrte'. He knew he was writing this for feline humanoids, and it would make sense to make a prr sound to mean pleasure. Basic onomatopoeia.

I completely agree. And I see/feel (kame?) the same thing with plltxe. And I might even be temped to extend it to ŋ in initial position and the vowel ɛ. I have two cats. They are brothers from the same litter. The less intelligent one has the better personality and is the more aggressive hunter (and his hunting call could easily be mistaken to include kitty ejectives. ) This is not my taronyu (Dex) in the clip, but her vocalizations are similar to the ones he produces. Tyra on YouTube . His call (c.f. beginning at about second :28 of Tyra's) has a much less nasal vocalic quality and a lot more of the "popping/crackling" kxìkx kxä kxìkxìkxi kxìkx). The more intelligent one of my two (Sy, the alpha mammal in the house) has a much larger "vocabulary" than the other when communicating his needs to the "staff" (e.g "me"). He commonly uses something along the feline vocal tract lines of [ŋɹ̃æ˦k] for "Now that you know what I want, let's hurry." (The vocabulary of human words (adapted for them including pitch/tone) that they recognize is a different story altogether. For another place and day...)

I'm also very keen to find out how rutxe ("please") works (or perhaps *will* work?). <Perhaps "rutxe" is no longer spelled that way? Or was some kind of cut-and-paste-o when the Survival Guide was laid out?? <And what is that mysterious apostrophe on the end of [pllte'] in [tsun oe nga-hu nì-Na'vi pivängkxo a fì-'u oe-ru pllte' lu]?> *Almost* anything is still possible. Opt. A: Ma tsmuk, rutxe!, oeru txoa livu. Opt. B: Ma Neytiri, oeru lora ikran ngeyä rutxe. (And let's not even think about how taboo "Oeru sevina ikran ngeyä rutxeie nìwotx!" might be. )  

[nue] vs. [new]? If I were he and working on the script I would spell it "neu" in order to help the native English-speaking (and literate) actors to avoid pronouncing it "ɲuː" in the context of the filming. But then when codifying "my baby" properly for posterity (i.e. as a potentially commercial product or public précis that should be "correct"), I'd likely move to my proper diphthong convention corresponding to [aw] and use "new".

I think it's very important for us to keep in mind that this is a fictional language (artlang) and that even for Paul Frommer it's incomplete. "A work in progress."

@Wm Annis: Your point about free word order is very well taken. You're absolutely right: by "free" I meant, as you said, "not syntactic"; discourse issues will affect the choice of word order. Na'vi is still a work in progress, and up to now I've been making some word order decisions on the basis of "feel"—no doubt influenced by the languages I know, given that I don't have native speaker intuition. (Wish I could find someone who does.) Codifying the rules of discourse is something I'm working on.

From: Some highlights of Na'vi

I've fiddled with conlanging myself enough to know that it is very easy to lose track of one's own rules and conventions in the process. Before one *knows* one's own vocabulary and grammar instinctively (à la a native English speaker's effortless manipulation of "a/an" & "the [ði]/the [ðə]") and how the words behave in all situations, one must write it all down (or document it in a database).  I can ESPECIALLY see losing track of things over the course of 2 to 4 years. I wonder if the final [-it] accusative we see in the Weaving and Hunting Songs is not just some relic of an earlier age/stage (like 9 and a half months sooner than the creator would have liked to have had to finalize things for which someone else in the production ecosystem had a publishing deadline). I also predict that "something" will have to be done (at a minimum clarification of pronunciation rules) for the agent marker [-l] in Oel ngati kameie. The leonopteryx, who undoubtedly will be briefly satiated by his direhorse snack in [Torukl pa'liti yereiom nìwotx.] has a phonological problem (at least a small one). And that's NOTHING compared to his apparent existential situation etymologically speaking. His namesake (purportedly "last shadow") is currently composed of an infrequently attested conjuction [to] ("than") and a missing-in-action [ruk]. Or perhaps [tor] ("final"?/"shadow"?) + ['uk] ("final"?/"shadow"?) . <Quick! Decide which is which; throw in the glottal stop on [(')uk]; and sneakretively secure their places in the dictionary before anyone notices!  >

The "Survival Guide" was clearly thrown together from a variety of sources. The "Leaf Plate" is attested as [sumin jiit'luy] or [ulu'tah inib'sey mulsi]. <Well, maybe that's what those weirdo seaside ig'gram-riding Na'bi who can't pronounce things properly call a [su'min ziitxlui], but their dialect is so "hick", don'cha'gree?> Honestly, I don't think Dr. Frommer had anything to do with the final compilation or editorial supervision on that book portion of the project. Cameron might have even come up with [Eywa k'sey nivi'bri'sta] ("Eywa cradles everyone" >>> "hammock") & [sumin'sey hulleh] ("sleeping outside the group") himself. But, clearly Paul worked on the songs (in isolation? based on some simple conceptual treatment of what it means to the Na'vi to hunt and a melody?) and without a doubt the vocabulary for the lyrics came from him.

The "new community" has clearly staked the "future" of the language on the Survival Guide's word list. And, Eywaru (ay?)(tì?)irayo livu that the list came from Frommer—even if ke muiä livu nìwotx  

PS: RE: from my own brain...

"9 and a half months sooner than the creator would have liked to have had to finalize things someone else in the production ecosystem had a publishing deadline"...

Would someone please translate even just the part "sooner than the creator would have liked to have had to finalize things" into Na'vi for me. I REALLY want to see that...

[Txur’Itan]

Interesting...

[ritx]

It is too bad he didn't say ...

Tsun oe ngahu nìNa'vi tspivangkrr a fì'u oeru prrte' lu
It is a (cat-quality) pleasure to me that I can kill time with you in Na'vi.
That would solve the -kxo suffix problem.

Or based on my hearing of the skxawng interview, tspivängkxo = kill (inf.), so
It is a (cat-quality) pleasure to me that I can kill you in Na'vi.
That makes even more sense, depending on your cat.