If a Na'vi speaker were to learn another language...

['Oma Tirea]

What are some of the keyey that could happen?  Starting with English....

Phonetically I can see the met-mate merger, as well as fronting of the "ch" and "j" sounds.  There could also be r-tapping, syllabic r-rolling, th-stopping, and various vowel shifts, such as {ɒ}/{ɑ} to {a} and maybe {ɔ} to {o}.  {aɪ} and {aʊ} as well as {u} may fluctuate from tense to lax, rhyming good with food.  {u} may also be more monophthongal than what most English speakers have.  There may even be epethentic insertions of {ə}/{ɪ̈} in some words due to Na'vi phonotactics.  If you like to consider some extreme cases, some stranger consonant mergers such as the let-net merger, zip-gyp merger, or the jet-yet merger may be employed.

Gramatically I can see double-negatives being used and "be"/"is"/"am"/"are" being omitted, as well as word order having a similarity to that in Na'vi.  Not sure what else might occur gramatically....

[Duma Vadamee {Aungia Tsawkeyä}]

honestly the speaker would start out really bad like jake, but would excel way faster then a human would.

[kewnya txamew'itan]

I disagree with a lot of those phonological processes. In general people tend to 'round' foreign sounds to the nearest sounding native sound they have, mate is pronounced remarkably similarly to the na'vi *meyt so I imagine the na'vi would pronounce it like so (although that's still a change in the primary vowel).

I think that ch would probably become a na'vi ts although I'm not sure about a hard (affricated) j, I think it could go to a na'vi tsy, ts, z or y all seem plausible.

I agree with the r-tapping and that r-rolling will probably occur (possibly instead of what folk phonology in non-rhotic dialects would tell us are 'rhotic' sounding vowels e.g. [ɜ], [ɑ], [ɔ], [ʉ], [ɨ], [ɪ̈] and [ʊ̈].

We have implicit confirmation from doctor -> toktor that [ɒ] -> [o ] and maybe even that [ɑ] shifts similarly.

[aɪ] and [aʊ] are almost certain to stay the same, both occur in both languages  identically, there is no reason for a shift to occur and so I would be very much surprised if one did. As for not distinguishing tense/lax [u ], I think that would be very common as I don't think a hypothetical native speaker woud be able to distinguish the two sounds (just like many monolingual Spaniards can't distinguish [i ] and [ɪ]).

I would be extremely suprised by any insertions of [ə]/[ɪ̈] as they are foreign sounds, I would expect (given other uses of <ì> as a na'vi schwa) both to be replaced with [ɪ] (which may well be inserted to try and manage consonant clusters (like many English speakers might insert schwas into a hypothetical consonant cluster thgskldmnswrt).

There is no reason for a let-net merger as your assertion about allophonic variation is almost certainly false as l/n are phonemic in that location (minimal pairs lew/new and le-/ne generally) and there is no evidence that they is any such allophonic variation.

I agree about "to be" being omitted often although I think it might also be replaced with "to occupy" as a translation of tok, otherwise we might see a lack of the preposition "at" when stating your location e.g. I am at the shop -> I am the shop.

Articles (the/a/some etc.) would be confusing and may be dropped.

I'd expect an underuse (if it were used at all) of the passive and instead I'd expect the arguments to be re-arranged (as you said) rendering the utterance ambiguous at best.

I'd expect confusion over the in/on distinction.

I'd also expect an odd use of tenses due to the way na'vi treats tense and aspect separately (and assumes it once it has been stated).

honestly the speaker would start out really bad like jake, but would excel way faster then a human would.

And why would they pick it up so much faster? We know nothing about na'vi developmental speed or skills acquisition, just making them glorified, blue Mary-Sues and Gary-Stues doesn't really help and lessens the impact of their victory over a vastly more technologically advanced army.

[Duma Vadamee {Aungia Tsawkeyä}]

you did read the script didn't you? Grace says that the kids she taught excelled faster then she could keep up.  that must of s***ed.

[kewnya txamew'itan]

I'd forgotten that (I only skimmed the script once when it was first online). That might not just be na'vi developmental pace though, children always pick up new skills faster than adults (the difference widening as the adults get older), Grace certainly isn't young and I doubt that all those cigarettes helped prevent her brain from solidifying (not sure of the technical term), that could just be a sign that she's a bit slow now. It does seem more likely though, that they learn faster than humans, irayo for the correction, I now address my comment about Mary-Sues to Cameron.

[Muzer]

That line is in the special edition FWIW, so it's definitely still the case.

[Duma Vadamee {Aungia Tsawkeyä}]

eh.

['Oma Tirea]

What an analyst you are sometimes, ma kewnya txamew'itan

I disagree with a lot of those phonological processes. In general people tend to 'round' foreign sounds to the nearest sounding native sound they have, mate is pronounced remarkably similarly to the na'vi *meyt so I imagine the na'vi would pronounce it like so (although that's still a change in the primary vowel).

Right.  How did I miss that....

Although who knows?  It could even be the met-mate merger only comes in on Na'vi that are learning a dialect of English that has a purely monophthongal version of the FACE vowel.

I think that ch would probably become a na'vi ts although I'm not sure about a hard (affricated) j, I think it could go to a na'vi tsy, ts, z or y all seem plausible.

Funny how that works, kefyak?

I agree with the r-tapping and that r-rolling will probably occur (possibly instead of what folk phonology in non-rhotic dialects would tell us are 'rhotic' sounding vowels e.g. [ɜ], [ɑ], [ɔ], [ʉ], [ɨ], [ɪ̈] and [ʊ̈].

How are those last 6 considered rhotic?  For {ɑ} and {ɔ} it looks as if the r is simply dropped.  How can we expect a hypercorrection here?

As for the last 4 vowels, aren't those simply reduced vowels elsewhere aside from rhoticity?

We have implicit confirmation from doctor -> toktor that [ɒ] -> [o ] and maybe even that [ɑ] shifts similarly.

I would still have to admit that it is a spelling pronunciation, and perhaps {ɒ} may shift to {o}.  {ɑ} wouldn't really shift to {o} at all, but instead to {a}, and besides, the degree of lip rounding is the same in both vowels, as is the tongue height.

[aɪ] and [aʊ] are almost certain to stay the same, both occur in both languages  identically, there is no reason for a shift to occur and so I would be very much surprised if one did.

I didn't say they would shift, but maybe fluctuate between tense and lax endings.

I would be extremely suprised by any insertions of [ə]/[ɪ̈] as they are foreign sounds, I would expect (given other uses of <ì> as a na'vi schwa) both to be replaced with [ɪ] (which may well be inserted to try and manage consonant clusters (like many English speakers might insert schwas into a hypothetical consonant cluster thgskldmnswrt)).

{ɪ} might be a possibility, but take for example how Italians learn English.  They may sometimes insert epethentic schwas at the ends of words, and yet schwa doesn't exist in Italian.

There is no reason for a let-net merger as your assertion about allophonic variation is almost certainly false as l/n are phonemic in that location (minimal pairs lew/new and le-/ne generally) and there is no evidence that they is any such allophonic variation.

Like I said, very unlikely to occur, if at all.  Apparently you don't like to consider a few of the extreme cases (although I have to admit, I probably was being a bit extreme on this one )

I agree about "to be" being omitted often although I think it might also be replaced with "to occupy" as a translation of tok, otherwise we might see a lack of the preposition "at" when stating your location e.g. I am at the shop -> I am the shop.

Articles (the/a/some etc.) would be confusing and may be dropped.

I'd expect an underuse (if it were used at all) of the passive and instead I'd expect the arguments to be re-arranged (as you said) rendering the utterance ambiguous at best.

I'd expect confusion over the in/on distinction.

I'd also expect an odd use of tenses due to the way na'vi treats tense and aspect separately (and assumes it once it has been stated).

Not a bad speculation.  An extreme case: <ei> or <äng> or even <ats> or <uy> may be inserted into verbs.

...and who knows what could happen with the intonation.

Also, more for the phonology segment: weird things hapen to {æ} and {ɛ} before {ŋ} and {g}, and {ɪ} before {k} and {g} in English.  Sometimes {ɪ} may be raised to {i} before {ŋ} in English, but in English as a Na'vi would speak it, these weird phenomena wouldn't really happen (except for maybe the last one since <ìng> and <ing> don't (yet?) contrast in Na'vi).

[Muzer]

I can't say how it would be for someone learning it natively, but <iv> seems to be the most "natural" infix to me, in that I don't have to think about and it just appears - so I would expect this one to be slipped in the most. Hell, I've done it in the past (mostly when talking/thinking to myself) in English (eg I have to d<iv>o this!) completely by accident, so I can easily see a native doing it.

[Payä Tìrol]

I can see omitting the a/the articles, because that always happens when learning English from a language that doesn't have those.

['Oma Tirea]

Here's some more I though up:

Instead of "you have", "they have," etc.  it might be said as "be to you", "be to they," etc.

In addition, since Na'vi doesn't distinguish "be", "is", "am", or "are", I can see these are quite likely to become the same word: be.  E.g. "I be...", "He be...", "they be..." instead of "I am...", "He is...", "they are...".

One notable thing the Na'vi can only wonder about is "What reason English have no equivalent for what is Na'vi po?" XD

[kewnya txamew'itan]

I agree with the r-tapping and that r-rolling will probably occur (possibly instead of what folk phonology in non-rhotic dialects would tell us are 'rhotic' sounding vowels e.g. [ɜ], [ɑ], [ɔ], [ʉ], [ɨ], [ɪ̈] and [ʊ̈].

How are those last 6 considered rhotic?  For {ɑ} and {ɔ} it looks as if the r is simply dropped.  How can we expect a hypercorrection here?

As for the last 4 vowels, aren't those simply reduced vowels elsewhere aside from rhoticity?

We have implicit confirmation from doctor -> toktor that [ɒ] -> [o ] and maybe even that [ɑ] shifts similarly.

I would still have to admit that it is a spelling pronunciation, and perhaps {ɒ} may shift to {o}.  {ɑ} wouldn't really shift to {o} at all, but instead to {a}, and besides, the degree of lip rounding is the same in both vowels, as is the tongue height.

[aɪ] and [aʊ] are almost certain to stay the same, both occur in both languages  identically, there is no reason for a shift to occur and so I would be very much surprised if one did.

I didn't say they would shift, but maybe fluctuate between tense and lax endings.

I would be extremely suprised by any insertions of [ə]/[ɪ̈] as they are foreign sounds, I would expect (given other uses of <ì> as a na'vi schwa) both to be replaced with [ɪ] (which may well be inserted to try and manage consonant clusters (like many English speakers might insert schwas into a hypothetical consonant cluster thgskldmnswrt)).

{ɪ} might be a possibility, but take for example how Italians learn English.  They may sometimes insert epethentic schwas at the ends of words, and yet schwa doesn't exist in Italian.

1. None of them are true rhotics, the first three are all just a dropped r, but in non-rhotic dialects, the vowel shift that occurs before the r e.g. [a]-> [ɑ] is the only thing that happens as a result of the r. The last four certainly are just reduced vowels, but, to my non-rhotic-ly tuned ear they sound like either the uncrossed or un-umlaut-ed version of IPA symbol modified as if by a following r (although they'd normally be followed by a non-syllabic schwa as well in that case). All these sounds occur almost always around an r and so might be more likely to become "rr" than other vowels.

2. But the RDA people who taught the na'vi "toktor" would almost certainly be speaking GA and pronounce it [ˈdɑk.tɚ] (or a slowly spoken and emphasised [dɑktɔɻ]) implying a shift (in at least one instance) from [ɑ] to [ o]

3. Sorry, I got completely the wrong end of the stick here.

4. I didn't know that about Italian, that's very interesting.

['Oma Tirea]

ok, was just thinking of a mistake that could happen:

A na'vi that have learned the inglesh  language pretty whell is on a cafe:
the na'vi: I want a glass of whater and a burger and 3 aples
The tawtute: yes here it is, but you need to pay the money?
the na'vi: pay the money? ok..
(the na'vi take the glass of whater and por the water on the 50$ sedle)
Man: uh.....

This from a Na'vi who hasn't brished up enough on vocab hiyìk

None of them are true rhotics, the first three are all just a dropped r, but in non-rhotic dialects, the vowel shift that occurs before the r e.g. [a]-> [ɑ] is the only thing that happens as a result of the r. The last four certainly are just reduced vowels, but, to my non-rhotic-ly tuned ear they sound like either the uncrossed or un-umlaut-ed version of IPA symbol modified as if by a following r (although they'd normally be followed by a non-syllabic schwa as well in that case). All these sounds occur almost always around an r and so might be more likely to become "rr" than other vowels.

Hmm... maybe that would be so if a Na'vi is trying to learn a non-rhotic dialect such as yours.

[kewnya txamew'itan]

Sorry, I didn't make that quite clear, I only expected them to correct those folk-rhotic-vowels to rr in a non-rhotic dialect, in a rhotic one a similar effect could occur (probably with a much lower chance) if the r dominated the vowel in their perception.

['Oma Tirea]

More missed insight: plosives...

They could be unaspirated in all positions, and unreleased in word-final position.  This may be confusing to the native English speakers because they may be mistaken for the voiced plosives (although Na'vi isn't the only language to have this somewhat-minute issue).

['Oma Tirea]

Next language to investigate: French

I imagine a Na'vi might have trouble with the front rounded and nasalized vowels: {y}, {ø}, {œ}, {œ̃}, {ɛ̃}, {ɑ̃}, {ɔ̃}.  Similarly, they would have trouble with the approximant based on {y}: {ɥ}.  There may also be trouble distinguishing {e} from {ɛ}, {ɔ} from {o}, and {ɑ} from {a}.

I also imagine the palatal nasal being pronounced in a {nj} sequence rather than a straight-up {ɲ}.  There may also be alterations to {ʃ} and {ʒ} as well, similar to how they are treated in English.  There can also be trouble with the guttural rhotic {ʁ}, and it can tend to go to {ɾ} instead.

The treatment of the six plosives {p}, {b}, {t}, {d}, {k}, and {g}, would be about the same as in English.  There may also be a schwa/schwi offset to words ending in a fricative.

That's really all as far as I can see for phonetics.

As far as grammatical treatment goes, I cannot say because I haven't really studied French