Na'vi Linguistics: Free Word Order

Started by wm.annis, April 27, 2010, 04:42:54 PM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

wm.annis

There's a big problem with the phrase "free word order" — it's linguistic code.  It doesn't mean the same thing to linguists that it means to everyone else, which leads to some confusion and misunderstanding.  So here's a quick interpretation of Linguistickese for people, which I hope can clear up a few confusions.


Free Constituents!
When a linguist says "free word order" they don't actually mean "words," they mean "free constituent order."  The problem is "constituent" is just a pain to say, and very often the constituents of any given phrase will be single words.  Just not always.  Here are two sentences where I have marked the subject constituent in purple and the direct object constituent in orange:

The dog bit the man.
The dangerous dog bit the man I saw yesterday.

Notice that for the purposes of syntax, "constituents" can be quite simple ("the man") or more complex ("the man I saw yesterday").  This matters because even if your constituent order is free the word order within constituents may be quite constrained.  For example, many languages don't let you break up constituents willy-nilly.  Na'vi appears to be the same.  If you want to use the constituent "the dangerous demon," lehrrapa vrrtep you can't really shove the verb of the sentence into the middle of them — they are an indivisible whole (even if you can swap the words vrrtep lehrrap).  A language that has grammatical gender (like Latin or Ancient Greek) has tricks that allows it a little freedom to break up even constituents.


Free Order
Next, when a linguist says the word order is "free" that does not mean word order has no significance.  It simply means that it does not matter for syntax.  In English, word order determines syntactic role:

 The man bit the dog.
 The dog bit the man.

The two phrases above describe completely different states of affairs.  What we can call the "propositional content" of the sentence, to borrow a term from logic, is radically changed by a change in word order.  But in Na'vi we use case endings to mark out the syntactic roles of constituents, which lets us shuffle things around without changing the propositional content:

 Nantangìl frolìp tutet.
 Tutet frolìp nantangìl.
 Frolìp tutet nantangìl, etc.

All these different word orders encode identical propositional content, thanks to the case endings.  However, it is a little misleading to say that they mean the same thing.  Talking is a social act.  We don't just stand around uttering propositional content, but we do things to make sure the people we're talking to understand which parts we think are important, we make sure they can follow the narration, and we often make clear how it is we feel about the information we are conveying.  In fixed word order languages we have to use intonation or special words to do these additional jobs, but in a free word order language, since word order no longer matters for encoding propositional content clearly, you can use the word order to do some of these other jobs.

In a longer conversation anything we say will have both old information (which we've just mentioned) and new information (which we're adding to the full scenario we're trying to communicate).  We have to keep mentioning old information because we usually need to explain how the new information fits into the big picture.  In free word order languages, old information and new information tend to land in particular places.  In ancient Greek, for example, a common word order is:

 Old Information - New Information - Verb - Everything Else

Based on what Frommer has said elsewhere, it sure looks like new information (or possibly contrastive or especially important information) goes at the end of a clause:

Quote from: Frommer(The end of the sentence is where the "punch" comes.)

Canon

But apart from that we don't yet have all the details.  But from now on, when you hear the phrase "free word order" know that it does not mean the word order has no significance at all, but that the word order is used to guide our listeners' understanding of what we're saying.  Everything from basic discourse matters I've discussed above to style and politeness matters could come into play with Na'vi word order.  We just have to wait for Frommer to fine tune the details.

kewnya txamew'itan

This ought to be a very useful post, free word order is a very misleading term.
Internet Acronyms Nìna'vi

hamletä tìralpuseng lena'vi sngolä'eiyi. tìkangkem si awngahu ro
http://bit.ly/53GnAB
The translation of Hamlet into Na'vi has started! Join with us at http://bit.ly/53GnAB

txo nga new oehu pivlltxe nìna'vi, nga oer 'eylan si mì fayspuk (http://bit.ly/bp9fwf)
If you want to speak na'vi to me, friend me on facebook (http://bit.ly/bp9fwf)

numena'viyä hapxì amezamkivohinve
learnnavi's

Kì'eyawn

Ngaru irayo seiyi oe nìtxan, ma William.  I'm sure this will help people—it has me =)
eo Eywa oe 'ia

Fra'uri tìyawnur oe täpivìng nìwotx...

Ftiafpi

Good write-up, unfortunately I didn't learn anything new but I imagine this will be a big help to the beginners who are trying to wrap their heads around the linguistic terms involved in "free word order". I especially liked how you defined "free constituent order" as the true meaning of "free word order".

Can't wait to see what other little nuances word order has for Na'vi when it gets fleshed out some more.

Tsuksìm atsawl (KaPTan)

yes, thanks alot, it is a real help.  i've been "studying" for like 4 months now, too bad i have had school, and i won't be here for the summer :(
Also a member of the podcast known as Avatar Nation.  Come listen to our stuff! like us on facebook and chech out our blog at Avatar-Nation.net.

Kì’onga Vul

Annis mentioned that the manipulation of word order can do some of the job that intonation might in more restricted-order languages.  I'm familiar with Finnish, which has relatively free constituent order, and also has fairly predictable inflection: over the course of a sentence, pitch gets lower (even in questions) and rises slightly for constituents.  Were you to greatly violate this, by rising in intonation for a question perhaps, it would sound very strange or even incorrect.


Because of freer word order, is Na'vi similarly restricted as far as intonation?  Or is this something that's less a rule and more an unconscious decision by the speaking community?
學而時習之!
Did I make an error you just can't stand to let survive?  Please, correct me!  I'll give you candy or something.

Tirea Aean

Quote from: Kì'onga Vul on December 23, 2010, 12:17:55 AM
Annis mentioned that the manipulation of word order can do some of the job that intonation might in more restricted-order languages.  I'm familiar with Finnish, which has relatively free constituent order, and also has fairly predictable inflection: over the course of a sentence, pitch gets lower (even in questions) and rises slightly for constituents.  Were you to greatly violate this, by rising in intonation for a question perhaps, it would sound very strange or even incorrect.


Because of freer word order, is Na'vi similarly restricted as far as intonation?  Or is this something that's less a rule and more an unconscious decision by the speaking community?

sorry this is crazy late.

we have little to no information on intonation as far as i know. if i understand correctly, intonation really is not a big deal. tho if you listen to all pawl's stuff, it seems to very much mimic english in that respect. its not like chinese at all where the intonation completely and utterly changes the entire meaning of what you're saying. the "why is this night" stuff is a bunch of questions but isnt like a conversation, its more like a literary work being read. every time i hear him say ngaru lu fpom srak it seems to go low and the fpom srak rises considerably. like english. there is english influence everywhere in Na'vi. probably because Paul is a native speaker of English. tho he is making efforts to try to lean less on English.

'Oma Tirea

[img]http://swokaikran.skxawng.lu/sigbar/nwotd.php?p=2b[/img]

ÌTXTSTXRR!!

Srake serar le'Ìnglìsìa lì'fyayä aylì'ut?  Nari si älofoniru rutxe!!

Tirea Aean


Kì’onga Vul

Actually, I posted here before Sxkxawng made the Prosody thread. :P

But thanks for the responses.  Lucky me, I guess, that I don't have to worry much about rhythm, which I think is the most elusive part of a foreign language.
學而時習之!
Did I make an error you just can't stand to let survive?  Please, correct me!  I'll give you candy or something.

Tirea Aean

Quote from: Kì'onga Vul on January 03, 2011, 08:59:51 AM
Actually, I posted here before Sxkxawng made the Prosody thread. :P

But thanks for the responses.  Lucky me, I guess, that I don't have to worry much about rhythm, which I think is the most elusive part of a foreign language.

just stress of a word. otherwise you would put the wrong emPHAsis on the wrong syLABble and it sounds weird and is hard to understand. ;P

Kì’onga Vul

Quote from: Tirea Aean on January 03, 2011, 06:26:46 PM
Quote from: Kì'onga Vul on January 03, 2011, 08:59:51 AM
Actually, I posted here before Sxkxawng made the Prosody thread. :P

But thanks for the responses.  Lucky me, I guess, that I don't have to worry much about rhythm, which I think is the most elusive part of a foreign language.

just stress of a word. otherwise you would put the wrong emPHAsis on the wrong syLABble and it sounds weird and is hard to understand. ;P

I know, it's strange how the stress from a word in one language can carry over to another.
For example, in Chinese, I used to pronounce the word for fifty as "shi," retaining the first-syllable stress of fifty.  Really, the word should read more like "wŭshí."  (Those tones strike again!)  Fortunately, I don't think I've had this type of trouble with Na'vi so far, but then again I haven't spoken it aloud very much.
學而時習之!
Did I make an error you just can't stand to let survive?  Please, correct me!  I'll give you candy or something.

Neyn'ite Te Tsahìk Txeptsyìp'ite

basically what you're saying is that Na'vi sentences are free word order but to a certain point, that some things have rules to where they go. as in, tute kawng- evil person- person being first because the noun goes before the describing word, srane?
old gallery link?id=2025[/img]


oel ayngati kameie, ma aysmukan sì aysmuke, Eywa ayngahu.
oeyä tsmukan, ma Nick, oeru ngaytxoa livu. nìmwey tsurokx. nga yawne lu oer.

Blue Elf

Quote from: Neyn'ite Te Tsahìk Txeptsyìp'ite on September 07, 2011, 08:13:41 PM
basically what you're saying is that Na'vi sentences are free word order but to a certain point, that some things have rules to where they go. as in, tute kawng- evil person- person being first because the noun goes before the describing word, srane?
this is not good example, as you can use both kawnga tute and tute akawng (don't forget attributive -a-)
Oe lu skxawng skxakep. Slä oe nerume mi.
"Oe tasyätxaw ulte koren za'u oehu" (Limonádový Joe)


Tirea Aean

#14
current restrictions against "FREE" word order:

1. ke must always go directly before the verb, kea must go before its noun.

2. Attributive/non-predicative adjectives must either go directly before or directly after the noun they describe:
Quote from: Blue Elf on September 08, 2011, 01:00:53 AM
you can use both kawnga tute and tute akawng (don't forget attributive -a-)

3. Possessives/genitives must either go before directly before or directly after their possessed noun, except that an adjective describing the noun may come in between if necessary:

oeyä 'eylan = 'eylan oeyä,
oeyä sìltsana 'eylan = 'eylan asìltsan oeyä

4. Ma must go directly before the name who is being addressed.

5. Srak(e) may (as far as I know) only go either at the very beginning or very end of a yes/no question.

6. The words ko and nang must only go at the end of the sentence.

7. Modal verbs can only come somewhere before the secondary verb; the secondary verb cannot happen somewhere before the modal:

nga yivom tsun = wrong
tsun nga yivom / nga tsun yivom / tsun yivom nga = correct forms.

As far as I can remember off the top of my head, these are the known word order restrictions.

Blue Elf

Maybe better said: free order of clause elements. Some example:
Hì'ia fwampop lu oeyä 'eylan asìltsan.
You can change order of all colored parts and order of words inside of colored parts. But you can't put word of one color into part with different color.

'eylan asìltsan oeyä lu fwampop ahì'i.
Hì'ia fwampop 'eylan asìltsan oeyä lu. All these are Ok

Oeyä hì'ia fwampop lu 'eylan asìltsan. This is grammatically still correct, but meaning changed (!). Well, this is not very good example, I'll try to find better, where changing word order breaks grammar.
Of course rules as written by TA must be applied
Oe lu skxawng skxakep. Slä oe nerume mi.
"Oe tasyätxaw ulte koren za'u oehu" (Limonádový Joe)


Tirea Aean


Kemaweyan

Also. I'm not sure, but it seems that words in topical case must go at the beginning of the clause. We have not an official rule, but in all Pawl's examples it's so... I've never seen that he uses it otherwise.
Nìrangal frapo tsirvun pivlltxe nìNa'vi :D

Tirea Aean

Yeah, you're right, it does seem to be generally true that Paul has a tendency to put the topic at the beginning of a clause It's a safe assumption to make and possible guideline to follow until further instruction :)

Kemaweyan

Nìrangal frapo tsirvun pivlltxe nìNa'vi :D