Na'vi Reference Grammar

Started by wm.annis, August 13, 2010, 08:46:50 PM

Previous topic - Next topic

0 Members and 2 Guests are viewing this topic.

Vawmataw

#220
Section 5.1. Derivational Affixes is a bit strange. It states that the affixes are not freely productive unless stated but it is not CLEARLY indicated that some affix is productive (with the exception of 5.1.4.1.).

Also:

Wllìm Today at 10:28 AM
(unrelated: I think there is a typo in §5.1.2.3: it should be ke-YAWR and e-YAWR instead of key-AWR and ey-AWR, so the y should be underlined in the stress marking?)
Fmawn Ta 'Rrta - News IN NA'VI ONLY (Discord)
Traducteur francophone de Kelutral.org, dict-navi et Reykunyu

Tirea Aean

EDIT: After drafting this post, I think it is exactly what Vawmataw is talking about anyway. I guess it might serve as a clarification or "I agree" sort of post.



Quote from: Vawmataw on August 18, 2018, 09:45:30 AM
Section 5.1. Derivational Affixes is a bit strange. It states that the affixes are not freely productive unless stated but it is not CLEARLY indicated that some affix is productive (with the exception of 5.1.4.1.).

The first introductory paragraph of 5.1 indeed ends with this statement:


Quote
Unless otherwise stated, the affixes below are not freely productive

What that means is, None of them are, unless one of the following subsections declares one to be productive. I suppose what could be done is, for all the subsections that describe a fully productive or semi-productive affix, said subsection could include the word "productive" somehow as and where appropriate. For example,



Quote
5.1.2.3. The ke- prefix may be used with root adjectives and participles, in which case the accent
usually shifts to ke-, as in keteng different from teng same, equal and kerusey dead from rusey
living. However, note keyawr incorrect from eyawr correct.

Is this "may be used with" phrasing to say that it's "fully productive with root adjectives and participles"? or is it that we just see it non-productively used with these?

As noted, 5.1.4.1 already does explicitly specify "fully productive" in the manner suggested here:


Quote
5.1.4.1. Tì- ‹us› creates a gerund. It is fully productive for verb roots and compounds (si-construction
verbs, §5.3.3, cannot be made into a gerund). This is most useful when a simple tìderivation
already has an established meaning, as in rey live, tìrey life, but tìrusey living. In compounds,
tì- comes at the beginning of the word and ‹us› goes into the verbal element of the
compound, yomtìng becomes tìyomtusìng. See also §6.9.2. Forum (31/1/2013)



Quote
Also:

Wllìm Today at 10:28 AM
(unrelated: I think there is a typo in §5.1.2.3: it should be ke-YAWR and e-YAWR instead of key-AWR and ey-AWR, so the y should be underlined in the stress marking?)
Indeed. It's noted as such ([kɛ.'jawr]) in the Official Dictionary PDF

Vawmataw

#222
Quote6.20.7. Tsnì. The conjunction tsnì that introduces some kinds of report clause which cause the
verb to take the subjunctive, ätxäle si tsnì livu oheru Uniltaron I respectfully request the Dream
Hunt, sìlpey oe tsnì fìtìoeyktìng law livu ngaru set I hope that this explanation is clear to you now. F
The verb determines the subjunctive, or the construction?
Horen 6.20.7. is now incorrect.

See this comment by Karyu Pawl:
http://naviteri.org/2020/02/some-words-for-leap-year-day/comment-page-1/

Tsnì doesn't require the subjunctive:
Quote from: Le Karyu PawlWith sìlpey, for example, you do use the subjunctive, because when you hope for something, you don't know it's true, only that it might come true. But when you're pretending, you're claiming that something is in fact true. For that you use the indicative (i.e., the form without -iv-).

There are now two verbs that don't require it, even with the presence of tsnì: leymfe' and la'um.
Quote from: Le Karyu PawlBut when you're pretending, you're claiming that something is in fact true. For that you use the indicative (i.e., the form without -iv-).
To answer the question at the end (originally in maroon, we now have the indirect confirmation that the verb determines it.

Horen 6.8.5 is also impacted. The maroon text could be turned black because it's correct.
Fmawn Ta 'Rrta - News IN NA'VI ONLY (Discord)
Traducteur francophone de Kelutral.org, dict-navi et Reykunyu

Tirea Aean

#223


Quote from: Vawmataw on March 01, 2020, 07:38:35 PM
Quote6.20.7. Tsnì. The conjunction tsnì that introduces some kinds of report clause which cause the
verb to take the subjunctive, ätxäle si tsnì livu oheru Uniltaron I respectfully request the Dream
Hunt, sìlpey oe tsnì fìtìoeyktìng law livu ngaru set I hope that this explanation is clear to you now. F
The verb determines the subjunctive, or the construction?
Horen 6.20.7. is now incorrect.

I'm wondering.. what change should we make to that text?  :-\



Quote
Horen 6.8.5 is also impacted. The maroon text could be turned black because it's correct.

Easy fix. :)

EDIT: interestingly, there was supposed to be a 6.20.7.2 section which lists all the verbs known to take tsnì, but it has been commented out. not sure how long ago. Probably quite some time ago, because ätxäle si, rangal, and sìlpey were the only ones mentioned in the comment. What a shame, because this section, though it would require continual updates, it would be a very useful reference. o.o

Quote from: Horen Source Code
Code (LaTeX) Select
%\subsubsection{} Verbs known to take \N{tsnì}: \N{ätxäle si},
%\N{rangal} (a marginal use), \N{sìlpey}.
% http://forum.learnnavi.org/language-updates/confirmation-on-use-of-rangal/


Vawmataw

#224
Quote from: Tirea Aean on March 05, 2020, 11:24:13 AM


Quote from: Vawmataw on March 01, 2020, 07:38:35 PM
Quote6.20.7. Tsnì. The conjunction tsnì that introduces some kinds of report clause which cause the
verb to take the subjunctive, ätxäle si tsnì livu oheru Uniltaron I respectfully request the Dream
Hunt, sìlpey oe tsnì fìtìoeyktìng law livu ngaru set I hope that this explanation is clear to you now. F
The verb determines the subjunctive, or the construction?
Horen 6.20.7. is now incorrect.

I'm wondering.. what change should we make to that text?  :-\
It's a good question. Maybe just The conjunction tsnì that introduces some kinds of report clause? :-\

Quote from: Tirea Aean on March 05, 2020, 11:24:13 AM
EDIT: interestingly, there was supposed to be a 6.20.7.2 section which lists all the verbs known to take tsnì, but it has been commented out. not sure how long ago. Probably quite some time ago, because ätxäle si, rangal, and sìlpey were the only ones mentioned in the comment. What a shame, because this section, though it would require continual updates, it would be a very useful reference. o.o
Related to my comment above, this section could be more pertinent now that the subjunctive is edited out from the main section.

Fmawn Ta 'Rrta - News IN NA'VI ONLY (Discord)
Traducteur francophone de Kelutral.org, dict-navi et Reykunyu

wm.annis

A long overdue update with a bunch of minor tweaks and updates, just in time for 2021!  You can get past the file caching by using this link to download it. Otherwise, it'll probably be a few hours before the usual link gets the new version.

  • many small style and formatting changes
  • more examples, citations, and some more indexes
  • a visual layout of the legal Na'vi syllable
  • full set of tsa- conjunction contractions
  • productivity of -yu with si-verbs
  • clarifications on omission of san or sìk
  • refinements on use of pum
  • numbers after zam
  • -tsim compound element
  • modals that take futa section redone (new rules on subjunctive)
  • sequential verbs with modal
  • syntax section for sno redone (but we are waiting on more detail from K Pawl on this)
  • dative in -ru with word-final glottal stop
I missed this until I had already produced the new version:

Quote from: Vawmataw on May 13, 2018, 12:58:55 PM
After discussion on Discord, Plumps, Kawnu, Tirea and I found out that a passage is missing in the Horen. The rule 5.1.1.2 doesn't mention that nì- is freely productive on adjectives. Here is the source of the rule:

QuoteFinally, some affixes are midway on the productivity scale. The adverb-former nì- is productive when used with adjectives: nìngay 'truly,' nìwin 'fast,' nìsti 'angrily,' nìftue 'easily,' etc. But it's sometimes also used with other parts of speech—nìtut'continually,' nì'eyng 'in response,' nì'awtu'alone'—and these words have to be learned as separate lexical items; you can't take them as patterns on which to base new forms.
http://naviteri.org/2010/07/diminutives-conversational-expressions/

I've already incorporated this into the word creation chapter for the next version.

wm.annis

Version 1.29, released 2 days ago, with a few updates:

  • declension of foreign nouns in
  • use of tìk as a conjunction
  • reflexive of causative, ‹äpeyk›
  • as always, some minor style refinements, some more links to sources, and a few new Frommerian examples in older sections

wm.annis

Version 1.30 has a few bigger changes in it —


  • some examples now use interlinear glosses
  • a new font (I needed small caps)
  • -tu clarifications
  • formal 3rd person pronouns
  • some verb dropping in colloquial speech
  • a new section, Semantics, which has copula and perception expression notes
  • more links for how we know things
The point of the new Semantics section is to cover ground that is a mix of both grammatical and lexical material which normally would be spread out over both the grammar and a dictionary, but which it is useful and informative to see all in one place. I don't know how much larger that will get in the future (Paul suggested moving some Topical (-(ì)ri) discussion into that might make sense), but it has a good start for now.

When I first started assembling this grammar I was most concerned about accuracy and keeping track of how we know what we know, especially for trickier corners of grammar. One result of this is that the grammar can be very terse, and while all the examples are translated, the grammar does assume you know a fair bit of Na'vi already. This can make the Horen a bit of a slog for beginners. There's not a good way to resolve that problem completely, but I have started doing two things to help a bit.

First, I have started to expand explanations in some places where before I just included a reference to another section of the grammar. A little redundancy is worth it if the resulting document is easier to use.

Second, I have started putting interlinearized examples in. This is a format used in linguistics texts to help people who don't already know a language follow the grammar and vocabulary of an example. This is what they look like:



The first line has the unaltered Na'vi text. The second shows how the prefixes, suffixes, and infixes are broken apart. The third line explains the parts, and the last line is a natural English translation. To make life easier, the explanation on the third line does use some abbreviations — otherwise the text explodes, and is hard to read. I give a summary of the abbreviations in the intro, and also include an explanation of interlinears there now, too (this example is from that section).

Adding more detail and turning more examples into interlinears will take some time, but the next few releases should see more of this. I won't be turning every Na'vi example into an interlinear, but the more complex the example, the more likely I'll convert it. Typesetting them is a bit fussy, so, as I say, it may take a while.

Tekre

Just a very minor detail: I was wondering about your formatting of infixes in the glossing. Is it on purpose that the < doesn't close with > in the glossing line? Just asking because I learned a different standart for glossing infixes, and of course I know not everyone follows the Leipzig glossing standard (in which I also think the infix is before the word, not after, so "<poss.aff>see"), but I am just curious how you got to that formatting /what your thoughts behind that were :D

wm.annis

#229
Quote from: Tekre on October 18, 2022, 02:20:49 AM
Is it on purpose that the < doesn't close with > in the glossing line?

It is, because I've seen that used in some linguistics papers I've read recently. It seems a bit less cluttered. However, I've checked the Leipzig rules, and they use the brackets on both sides of the gloss, which would probably be easier for me to use in the future. I'll think about it.

QuoteJust asking because I learned a different standart for glossing infixes, and of course I know not everyone follows the Leipzig glossing standard (in which I also think the infix is before the word, not after, so "<poss.aff>see"), but I am just curious how you got to that formatting /what your thoughts behind that were :D

I am aiming in the Leipzig direction. Because Na'vi has two infix positions, I'm trying to have first and "pre-first" infixes before the glossed word, and the second position infixes glossed after. Even the Leipzig document recognizes that infixes might be glossed before or after a word. I added an example yesterday which had säpoli which I glossed "REFL›do‹PFV", for example, which breaks that convention, but I suspect is easier to read.

Thanks for the questions! The glossing conventions for the Horen are still settling a bit, so feedback is useful. Ease of use for beginners is a higher priority than perfect compliance with a convention.

Tekre

I thank you for the answers! Like I said, I was just curious and have not much experience with glossing yet, so I'm using every chance to learn more, and of course readability and fitting the target group always is more important than following conventions :)

`Eylan Ayfalulukanä

Your hard work keeping horen up to date is really appreciated! Na'vi is one of the very few constructed languages I have seen that has such a comprehensive reference grammar. Its a key component in the overall success of the language, putting it in a place that few other conlangs have achieved.

As far as the gloss thing, here are my thoughts: The idea of putting the pre-first and first infix functions before the word and the second position infixes after the word makes a lot of sense. And using closed brackets also helps, especially because this is a pretty formal document. If you plan to add a lot of these interlinear glosses, a note somewhere, perhaps as an appendix on how they work and what the abbreviations mean might be very helpful.

A long time ago, when the Bible in Na'vi translation project was slowly advancing, I had the idea of doing all translations as an interlinear gloss. I struggled with the format of such a gloss, as I wanted anyone looking critically at the translation to understand exactly the thinking that went into the translation process. I never came up with something that was simple, yet complete. Your interlinear gloss format used in horen is perfect for this application. (A separate volume of translator's notes, such as those that accompany a complete edition of the NET Bible would also be useful.) Now, for the existence of sufficient vocabulary to revive this project!

Yawey ngahu!
pamrel si ro [email protected]

wm.annis

Quote from: `Eylan Ayfalulukanä on October 28, 2022, 03:09:09 AMIf you plan to add a lot of these interlinear glosses, a note somewhere, perhaps as an appendix on how they work and what the abbreviations mean might be very helpful.

There is one in the intro, though that has been getting tweaks for the next version.

QuoteYour interlinear gloss format used in horen is perfect for this application.

The standard in linguistics these days is the Leipzig Glossing Rules. Lots of languages require additions to the standard abbreviations, but the format and core abbreviations are fairly settled at this point, I think.

QuoteYour hard work keeping horen up to date is really appreciated!

Irayo seiyi!

wm.annis

#233
Version 1.31 is ready. You can find a link to it here, though the local version should be updated soon, too: https://github.com/horen-lenavi/Horen/releases/tag/v1.31 (you only need the PDF).

  • more examples in lenition section (rather than forcing people to track down cross-references)
  • topical word order with srake, conjunctions; and the new word order available in Reef Na'vi
  • when to use subjunctive with txo
  • note on genitives with pesu, tupe
  • adjective pile-ups and adjective + relative word order
  • clarify elision of unstressed, final e
  • ceremonial pronoun updates
  • ye'rìn redundant with proximal future
  • clarify productivity of nì-
  • more interlinearlized examples
  • our first notes on Reef Na'vi, from the blog and a few extras from Paul's OmatiCon 2023 talk; this includes notes in the Intro

As with every release, a few more citations have been tracked down and added, and the occasional style or phrasing oddity fixed.

Toliman


Tekre

We were discussing the examples about txo in the horen (page 58: "Txo nga ngeyn livu, tsurokx" and "Txo nga ngeyn lu, tsurokx") and were wondering about the usage of "lu" instead of "'efu" there. Going to the source of these two sentences, it turns out that Pawl actually send a second mail where he corrected both of them to use "'efu" instead of "lu":
https://forum.learnnavi.org/language-updates/use-of-the-subjunctive-after-txo/

Maybe it makes sense to replace these two sentences with the updated versions with "'efu" :)

wm.annis

Quote from: Tekre on April 14, 2023, 11:20:11 AMMaybe it makes sense to replace these two sentences with the updated versions with "'efu" :)

I've fixed that for the next release. There was also a minor copy-and-paste error in the second example, which I have also fixed.

Irayo!

Tekre

Me again, ngaytxoa hrh

Horen 3.1.1.5. states "Due to the similarity in sound between y and i, the patientive ending -it is simplified
when suffixed to a diphthong ending in y, as in keyeyt errors instead of *keyeyit."

This implies that the same must also happen for any word ending with ay, which is not the case as the source for this information (http://naviteri.org/2013/01/awvea-posti-zisita-amip-first-post-of-the-new-year/) states:

"With nouns ending in ey, the -it ending becomes simply t. Example: keyeyt 'errors' (not *keyeyit). With nouns ending in ay, the –it ending may become t: wayt  or wayit 'song'—both forms are possible."

Maybe this small piece on info should be added to not cause confusion over the question it -ayit is valid or not :) For completions sake, it might be worth to also mention -ewr then (and that -ewur is not possible), because only mentioning -awur/-awr might imply to some that also with -ew both forms are possible together with -ru.

wm.annis

I'll clarify that section a bit for the next release.

Tan Jala

I have a question about 2.3.2 where also was said that monosyllable verbs behave like they are unaccented. And there was vol from v<ol>ll as an example. I just want to be finally sure that it also extends to rr, namely ner from n<er>rr, because all else known verbs with rr are always accented on rr syllable.