Allophony in Na'vi ('Oma Tireayä guide to translating loanwords)

Started by 'Oma Tirea, June 22, 2010, 12:13:56 AM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

Carborundum

We learn from our mistakes only if we are made aware of them.
If I make a mistake, please bring it to my attention for karma.

'Oma Tirea

Quote from: Carborundum on July 24, 2010, 01:12:30 PM
Quote from: Jameso on July 24, 2010, 01:09:09 PM
Whats this? ???
A necropost.
Sran.  Ma Jameso, did you read the first post?  All the main information is there.
[img]http://swokaikran.skxawng.lu/sigbar/nwotd.php?p=2b[/img]

ÌTXTSTXRR!!

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

'Oma Tirea

Long time no see this thread, kefyak?

I was actually wondering if there's any general rule for the allophony of the plosives.

Here's the loanword cases from both http://masempul.org/ and the current dictionary:

b → p: 5
d → t: 4
g → k: 1
b → px: 1
d → tx: 0
g → kx: 0
g → ng: 1
ng → g: 2
p → px: 1
t → tx: 0
k → kx: 4
p remaining the same: 5
t remaining the same: 6
k remaining the same: 3

From this it is becoming clearer that both voiced and voiceless plosives cross-linguistically will become just p, t, and k.  Occasionally, a plosive, voiced or not, will become ejective, however, k seems to be a split case....

One thing is quite clear: any ng sequence pronounced something like {ŋg} will simply be softened to {ŋ}.

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

ÌTXTSTXRR!!

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

'Oma Tirea

Hmm... now starting to wonder if anything else out there such as the KXWERTY keyboard might have a little bit of influence on why b, d, and g might go to px, tx, and kx....

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

ÌTXTSTXRR!!

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

'Oma Tirea

~ B U M P ! ~

Tse, now for another question regarding loanword/name respelling and case endings.

Sometimes a word will have an ì on the end because it's necessary (e.g. Tsyìräfì).  Does this ì disappear in certain case markings, namely -ä, and sometimes -ur and -it?  If I'm not mistaken we have a canon example of this happening.  However this may require some clarification from Karyu Pawl.

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

ÌTXTSTXRR!!

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

Ikran Ahiyìk

Quote from: 'Oma Tirea on February 12, 2012, 11:49:33 PM
However this may require some clarification from Karyu Pawl.
Definitely.

But one thing: is that canon example really some kind of confirmed text, or just a transcription from the audio? There's a chance for it to be written -ìyä but pronounced nearly or exactly -ä.
Plltxe nìhiyìk na ikran... oe fmeri sìltsan nì'ul slivu, ngaytxoa...


See the new version with fingerings!
Avatar credits to O-l-i-v-i.

Lance R. Casey

Quote from: Ikran Ahiyìk on February 13, 2012, 02:42:36 AM
But one thing: is that canon example really some kind of confirmed text, or just a transcription from the audio? There's a chance for it to be written -ìyä but pronounced nearly or exactly -ä.
Text:

   Meuniltìrantokx Toktor Kìreysä sì Tsyeykä ke lu teng ki steng.


// Lance R. Casey

Ikran Ahiyìk

Irayo.










(and I should make some changes to something mine...)
Plltxe nìhiyìk na ikran... oe fmeri sìltsan nì'ul slivu, ngaytxoa...


See the new version with fingerings!
Avatar credits to O-l-i-v-i.

'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!!

Yawne Zize’ite

More necroposting, but this might be of interest to people trying to guess how some of English's rarer sounds will be borrowed into Naʼvi: http://accent.gmu.edu/index.php. It's an archive of a sample paragraph with most of the tricky sounds read by native speakers of dozens of languages, then transcribed. If you use the "search" function you can search for various common systematic mispronunciations of English. Stopping [θ ð] to [t̪ d̪] was overwhelmingly more common than substituting [s z] or [f v], and speakers who backed [θ] to [s͏] often stopped [ð] to [d̪].

To make it more complicated, though, loanwords don't follow the same rules as people mispronouncing English. Lee (2006) summarizes that Thai, Russian, Hungarian, Serbo-Croatian, Tagalog, Moroccan Arabic, Quebec French, and Xhosa use [t] for /θ/ in borrowed words while Japanese, German, Egyptian Arabic, and European French use [s͏]. I don't see a blindingly obvious correlation between languages that use dental [t̪] (e.g. French, Russian) versus alveolar [t] (e.g. English, German) in the list, or any correlation between loanword adaptations and historical sound changes (/θ/ -> /d/ in German, /θ/ -> /t/ in Moroccan and Cairene Arabic, /θ/ -> /f/ in Russian loanwords from Greek). Park (2007) studies Koreans picking the closest sound to initial /θ/ and they split 40% [s*], 24% [t*], 16% [ph], but /ð/ was matched to [t] by 78%. (Korean has few fricatives and no voiced sounds, so /f/ /v/ /z/ were not options.) However, at the end of a syllable, both /θ/ and /ð/ were matched to labial sounds ([ph] and [p] respectively) by a plurality of Koreans (Park 2007b). (In Korean, /s/ becomes [t] at the end of a syllable.)

So, my thoughts from this research:

[θ] (hard th) can become any of /t/ /s/ /f/ in decreasing order of probability; /t/ and /s/ are both common, but most languages favor one or the other. At the end of a syllable, /t/ and /p/ are both defensible, but I would favor /t/ because the existing loanwords from English have favored a spelling pronunciation.

[ð] (soft th) should be merged with [θ] due to the extreme rarity in words specialized enough to be worth borrowing and established preference for spelling pronunciations, but if needed ("the", etc.) /z/ and /t/ are the best renderings; I would favor /z/ because it matches /ð/ in both manner and voicing and isn't that far off in place. At the end of a syllable, treat as [θ].

Irtaviš Ačankif

Well, Chinglish ;D speakers stereotypically use /s/ for θ and /dz/ for /ð/ rather than stopping them.
Previously Ithisa Kīranem, Uniltìrantokx te Skxawng.

Name from my Sakaš conlang, from Sakasul Ältäbisäl Acarankïp

"First name" is Ačankif, not Eltabiš! In Na'vi, Atsankip.