Combining Our Efforts II

[omängum fra'uti]

Not to mention every case where there is an ambiguous distinction between diphthong and VCV.  I've actually got a little perl script that splits out the syllables of words, it would be trivial to modify it to scour the word list and split out any word where such a thing occurs.  Perhaps even clarify the syllabification of any occurrence of "VtsV" as "V.tsV" or "Vt.sV", as well as words that are VCV where C can be either a coda or onset.  (Presently my script prefers diphthongs over consonants, and onsets over codas, as that seems to be Na'vi's preferences.  But as I said it would be trivial to modify it to spit out all ambiguous cases from the current vocabulary.)

[roger]

Since F follows morphology, and underlines entire stressed syllables, this would only be an issue where unstressed syllables meet within roots. Since roots are generally fairly short, there are few instances where this would even come up. AFAIK, so far this has only been relevant for w, y, ', and g, and my suspicion is that the latter two are an influence of English. And if F notes that there is little distinction in stressed examples, there's likely to be even less in unstressed examples where it's not marked in the vocab.

[omängum fra'uti]

You are most likely correct there, but it's good to get an understanding.  Of all the diphthong examples, only pizayu had an unstressed syllabic pair...  But if the -yu is the same -yu as elsewhere, it's likely not a diphthong.  (I'm inclined to believe it is CVCV there.)  There were no words where VtsC?V occurred in any unstressed syllable.

Words where it's VCV on an unstressed syllable, and it is not clear from morphology...

atokirina'
kìte'e
meoauniaea
snumìna
tsap'alute

So it's a pretty short list.  I'm inclined to believe that these are all regular.

If anyone wants to look at this on there own, this is the perl regex I'm using to split the syllables out - though this one is not the one I modified to find contested consonants.

Spoiler

Code Select Expand

       /
           \G
           (?| # C?C/VC? syllables
               ( # C, CC or (null)
                   (?:[csf]|ts)?
                   (?:[mgrlwy]|[ptk]x?|ng?)?
               | # Additional C only forms
                   [hvz']
               )
               (?| # Normal vowels
                   ([iìuoä])() # Empty capture because the ?| doesn't seem to work right
               | # Possible diphthongs
                   ([ea])([wy](?!rr|ll))?
               )
               ( # Coda - don't be greedy
                 # Next syllable gets any contended consonants
                   (?:['mgrl]|[pk]x?|ng?|tx)
                   (?![iìuoäeax]|rr|ll)
               | # Special case coda for T, it is given up to a s for ts
                 # Removing this part and adding t to pk above will give the T to the coda
                   t(?![siìuoäeax]|rr|ll)
               )?
           | # CV psuedovowel syllables only
               ([csfmgrlwyhvz']|[pk]x?|t[sx]?|ng?)
               (rr|ll)
           )
       /gcix
This returns the onset in $1, the vowel or psuedovowel (But not the second half of a diphthong) in $2, the y or w from a diphthong in $3 and the coda in $4.  When there are disputes, it prefers diphthongs over consonants and onsets over codas.  Theoretically to be the most flexible it should give a disputed T to the coda and let the program decide what to do with it, but there are so few cases where the T wouldn't go to a TS that it isn't a huge deal.  For pure scientific notation input, however, it should always give the T to the coda in a ts case.  Note that words like "'e.wll" are not considered disputed because the consonant MUST go to the CV syllable.

[roger]

Yeah, worth presenting that list for a quick yes/no; we might even get an explanation of what piza- is.

[omängum fra'uti]

Added to the list.  I made a new section for the PN stress question because it will involve compiling the chart, and sending it.  It's probably best done on it's own not as part of a group.

['eylan na'viyä]

Adjectives & Adverbs

• is it possible to use more than two adjectives with one substantive? like Aa S aB aC || Aa Ba S aC || or could sì be used in a specific way?
• are derived adjectives with "-le" in combination with noun-suffixes or appended adpositions possible ?

i hope that i used the linguistic terms correctly and that this has not been asked already

[roger]

oeyä ean-a txìm a-tsawl "my big blue butt"

2 adj may also occur on the same side of the N, so the combo should be legal. But yes, don't we have "sì" linking adj?

More general question: can the N be dropped out of an NP? Does the adj then take the adp/case?

It would also be nice to fill out the answer-word chart: ?kawtseng "nowhere", ?kawpo "no-one", no reason, no-how, no way, no (action), no (word), and similarly for which (?pefo "which one"), proximals & distals, some-, every-, other-, same-. And the plurals: does pesu "who?" have a plural fesu when you know multiple people are involved? The plural of tsa'u is sa'u; is the pl. of fì'u therefore ayfì'u, or fayu? Is there a contrast or parallel between sa'u and ?tsayu? Etc.

Also, confirm that all adp, including things like teri, na, and pxel, can occur as suffixes/enclitics as well as as preps. It seems from what he's written that teri is the prep. equivalent of the case -ìri. We should ask that explicitly. (Who knows? It might be possible to get an acc. prep. ??teti, or an ergative ??tel.)

Have we ever gotten an answer on those mysterious verbal infixes? tovaron, tevaron, telaron, tairon? They may be spurious, but we should get them out of the way if they are.

Also, those odd adverbs that look like adjectival forms of stative verbs:

lìm "be far" --> alìm "in the distance". Perhaps stative verbs cannot take the participle? Is the form ?lìma with the opposite word order? Is this a subord. form lexicalized as an adverb, or is there s.t. more productive going on? Do all stative verbs behave like this?

['eylan na'viyä]

Phonology and Phonotactics

• what happens when adpositions are added to words that end with the same vowl or consonant as the adposition begins?
  eg: oe + eo http://forum.learnnavi.org/intermediate/question-about-adp-eo/
  
• can a adposition at the end of a word cause lenation on the following word?
  

[roger]

• can a adposition at the end of a word cause lenation on the following word?
  

I very strongly suspect not. There's nothing inherent to the sounds of those adp. which cause lenition, but rather s.t. historical, so that lenition is part of the prep+N paradigm. Think of the retention of lost C's during liaison in French. But it should be asked, just in case.

[NeotrekkerZ]

Did you want to include for the sake of completeness
*Ikran zolup ne kllte na/pxel fwa rìk zup ta utral and
*Ikran zolup ne kllte na/pxel rìk a zup ta utral

Sure, let's clear it up completely.

[omängum fra'uti]

First set of questions have been asked.  Asked questions are marked in green, I'll move them to the answered section as they come in.  No promise everything will get answered, last time we did a group he just answered a few.

['eylan na'viyä]

Verb usage / forms

• can kä be attatched generically to all verbs where it sounds senseful?
  Word meanings or Phonology and Phonotactics
  
• is "ois" a word?
  
  I looked through some threads and came across these.
  maybe some exportable questions could be found in there:
  
  http://forum.learnnavi.org/advanced-grammar/genitive/
  http://forum.learnnavi.org/advanced-grammar/pragmatics/
  http://forum.learnnavi.org/advanced-grammar/negation/
  http://forum.learnnavi.org/advanced-grammar/si-compounds-and-word-order/
  http://forum.learnnavi.org/advanced-grammar/txotsakrr/msg153784/#msg153784

[Kì'eyawn]

First set of questions have been asked.  Asked questions are marked in green, I'll move them to the answered section as they come in.  No promise everything will get answered, last time we did a group he just answered a few.

Ma omängum fra'uti, may i ask why you didn't include the fayu/ayfì'u question?  That's one i really would like answered.

[omängum fra'uti]

Because I was going off HTML I'd grabbed a little before the server went down, and somehow missed that one when compiling the final list to send.  That post was a little hard to pick through to find the questions without the full forum formatting.

[Kì'eyawn]

Because I was going off HTML I'd grabbed a little before the server went down, and somehow missed that one when compiling the final list to send.  That post was a little hard to pick through to find the questions without the full forum formatting.

Well, apparently it's a moot point, anyway; looks like we got an answer.

[omängum fra'uti]

I call it a part answer...  The question of plural question words is still open.  Certainly following the fì + ay = fay doesn't work because "pay" seems like it would be too easily confused with water in some cases.  Pey would be a lenited number 3...  Peay is a moutful, so fe does seem the most promising there.  So I think that should still get asked.  IMO anyway.

[Plumps]

Don't know if that has been answered somewhere else...

Does the interrogative pronouns take case endings in their pe+N form when asked with a transitive verb?
Is it, e.g.:

• pesu tsaye'a samsiyut(i)? or
• pesul tsaye'a samsiyut(i)?

What about the answer? Does it suffice to just say, e.g.:

• 'ite or rather
• 'itel (tsaye'a fot(i).)

[wm.annis]

Aahhh!  I should have remembered this earlier.

Phonotactics: We are told that there is no "-eng-" infix.  What's the conditioning factor in "sengi" (presumed from *sängi)?

[roger]

Don't know if that has been answered somewhere else...

Does the interrogative pronouns take case endings in their pe+N form when asked with a transitive verb?
Is it, e.g.:

• pesu tsaye'a samsiyut(i)? or
• pesul tsaye'a samsiyut(i)?

What about the answer? Does it suffice to just say, e.g.:

• 'ite or rather
• 'itel (tsaye'a fot(i).)

You would need case, or I wouldn't know who you were asking about, the doer or the doee.

[Plumps]

If you're sure ... thanks
txo law lu ngar ... irayo

I was reminded of Irish Yes/No constructions where the verb is repeated (positive/negative) for that ... but it's not necessarily conjugated ... thought this could happen as well in Na'vi. If you ask already for a "Who" then the meaning of whom you are talking about (with just e.g., 'ite in my example) should be clear - only if you repeat the whole sentence ergative and accusative need to be there... But I don't want to force it I'm rather on the safe side