Rare/unusual natlang features

[Nortaneous]

Bringing this thread back from the old board.

Mapos Buang has a full set of uvulars: /q ɢ ʁ ɴ/. /ɴ/ is otherwise unattested except in a few Inuit languages, where it probably isn't phonemic, and in some dialects of Bai.

Nias Selatan contrasts all of /ʙ b v ʋ w/, although /v/ is marginal. Some dialects also have a shift of t d > pf bv before /u/.

Iau has six consonants, eight vowels, and eight tones, but two tones can appear on a syllable. Foau, another Lakes Plain language, has 17 consonants and no nasals.

[mae]

-

[Frislander]

Northwest Caucasian languages, in addition to all the nonsense of their phonologies, also feature in grammatical terms not only ergative person marking on the verb, but also polypersonalism that extends to incorporating inflected prepositions into the verb to increase the number of arguments marked.

q-aː-ɕə-f-aː-ʂ’ə-ʁaː-ʁ
DIR-3plI-LOC-BEN-3plE-do-PST-PST
"they did it for them there"

qʷaːʂʷa-m r-aː-r-jə-ʁa-ʁa-wət͡sʷaː-ʁ
boat-OBL LOC-3plI-DAT-3sgE-CAUS-CAUS-stand-PST
"He asked them to put it into the boat"

qə-zə-sħaː-tja-z-ʁa-t’əs-haː-ʁ
DIR-RFLI-head-LOC-1sgE-CAUS-sit.down-go.in-PST
"I put it on my head"

[Birdlang]

Ngadha contrasts /ɗ/ and /ᶑ/. It also has no prefixes or suffixes, or affixes of any kind. Which makes its grammar very streamlined.
Fijian has /ⁿdʳ/. Ndrumbea and Nyindrou have the same phoneme.
Sundanese has a vowel system of /i u e ɤ o ə a/ written as i u é eu o e a.
Acehnese also contrasts /ɤ/ and /ə/. The vowel system is /i u e ɤ o ə ɛ ʌ ɔ a/ written as i u é eu ô e/ë (second only in diphthongs) è ö o a.

[Kuchigakatai]

Northwest Caucasian languages, in addition to all the nonsense of their phonologies, also feature in grammatical terms not only ergative person marking on the verb, but also polypersonalism that extends to incorporating inflected prepositions into the verb to increase the number of arguments marked.

q-aː-ɕə-f-aː-ʂ’ə-ʁaː-ʁ
DIR-3plI-LOC-BEN-3plE-do-PST-PST
"they did it for them there"

qʷaːʂʷa-m r-aː-r-jə-ʁa-ʁa-wət͡sʷaː-ʁ
boat-OBL LOC-3plI-DAT-3sgE-CAUS-CAUS-stand-PST
"He asked them to put it into the boat"

qə-zə-sħaː-tja-z-ʁa-t’əs-haː-ʁ
DIR-RFLI-head-LOC-1sgE-CAUS-sit.down-go.in-PST
"I put it on my head"

What specific language is that? Are you familiar with enough with that language, so that you know the answer if I ask you how one says "I put them on my head" in it?

The order of the morphemes looks pretty amusing. I imagine that the reason why -aː-r- -3PL.I-DAT- appears after LOC in the 2nd example is because datives are a core argument in a way that the LOC morpheme is not, hence also the expulsion of the -aː- -3PL.I- morpheme in the 1st example to a position before the LOC (benefactives are not core). But then, why doesn't the qə- DIR- morpheme appear next to -r- DAT? Is the so-called "DIR-" morpheme a marker of transitivity instead? If "(I put) them (on my head)" turns out to be rendered by something that starts with 3PL-DIR-, I'm going to be even more amused.

[Nortaneous]

Some developments of initials in Zbu Rgyalrong dialects:
*ɬ- > lt- / ltʰ-
*ɮ- > ld-
*sl- > ldz- / zd-
*v- > ʁ-
*m- > mŋ- before a uvularized (velarized?) vowel

[Frislander]

Northwest Caucasian languages, in addition to all the nonsense of their phonologies, also feature in grammatical terms not only ergative person marking on the verb, but also polypersonalism that extends to incorporating inflected prepositions into the verb to increase the number of arguments marked.

q-aː-ɕə-f-aː-ʂ’ə-ʁaː-ʁ
DIR-3plI-LOC-BEN-3plE-do-PST-PST
"they did it for them there"

qʷaːʂʷa-m r-aː-r-jə-ʁa-ʁa-wət͡sʷaː-ʁ
boat-OBL LOC-3plI-DAT-3sgE-CAUS-CAUS-stand-PST
"He asked them to put it into the boat"

qə-zə-sħaː-tja-z-ʁa-t’əs-haː-ʁ
DIR-RFLI-head-LOC-1sgE-CAUS-sit.down-go.in-PST
"I put it on my head"

What specific language is that? Are you familiar with enough with that language, so that you know the answer if I ask you how one says "I put them on my head" in it?

It's Adyghe, I should have said, I think "I put them on my head" would be either the same (as 3rd person arguments tend not to be marked) or /qaːzəsħaːtjazʁat’əshaːʁ/ (it's not given in the article though, cause this is taken from an article not a comprehensive grammar), or maybe even /aːqəzəsħaːtjazʁat’əshaːʁ/, but to be honest I'm not sure, I'll need to track down a full grammar.

The order of the morphemes looks pretty amusing. I imagine that the reason why -aː-r- -3PL.I-DAT- appears after LOC in the 2nd example is because datives are a core argument in a way that the LOC morpheme is not, hence also the expulsion of the -aː- -3PL.I- morpheme in the 1st example to a position before the LOC (benefactives are not core). But then, why doesn't the qə- DIR- morpheme appear next to -r- DAT? Is the so-called "DIR-" morpheme a marker of transitivity instead? If "(I put) them (on my head)" turns out to be rendered by something that starts with 3PL-DIR-, I'm going to be even more amused.

The article gives a gloss for DIR as "directive", which I'm uncertain what that's supposed to mean. Judging from some paradigms, it looks like an optional marker of a direct object, where the absolutive argument is always marked first.

[mèþru]

Doesn't Japanese have /ɴ/?

[Ryusenshi]

I think /ɴ/ for Japanese means "nasal with unspecified point of articulation" (or even "nasalization of the previous vowel") rather than actually "uvular nasal". At any rate, it never contrasts with /n/ (at least not directly).

[mèþru]

I read that it actually has uvular or near uvular pronunciaion when not assimilated

[Ryusenshi]

I have read contradictory evidence about that. Let's just say that phoneticians are conflicted.

[Whimemsz]

.

[mae]

-

[Kuchigakatai]

The problem with these descriptions is that if there is "incomplete contact" then the claim that it's a 'uvular nasal' is simply false. Nasals are stops, and stops do not have 'incomplete contact'.

Nasals are typically but not necessarily stops, can also be nasalized approximants. See Mandarin allophony in China (especially Beijing), where it's common to pronounce pinyin -an [æn] and -en [ən] as [æɨ̯̃] and [əɨ̯̃] too. This kind of thing is presumably what the "[ɴ̞]" of Nogita et al. stands for with that lowered diacritic.

[Nortaneous]

Japanese "ɴ" is placeless, probably glottal at a deep enough level of Theory (like Burmese, where codas are neutralized to the glottals -ʔ -ŋ), but in singing you do get what sounds like a very backed syllabic velar nasal sometimes.

[mae]

-

[KathTheDragon]

I don't think it's reasonable to call it "the word final allophone of /n/"

Phoneme labels are arbitrary, however. We could denote Japanese's nasal phoneme as //, for example. It's merely convention that phoneme labels are similar to their primary allophone(s).

[Ryusenshi]

Are you thinking about the infamous Hale (2000) paper that analyzed Marshallese with four vowel phonemes //, //, /☎/, /☯/?

[bradrn]

Ngadha contrasts /ɗ/ and /ᶑ/. It also has no prefixes or suffixes, or affixes of any kind. Which makes its grammar very streamlined.

I know I’m a bit late to reply to this, but: aren’t there many other languages like this as well?

[KathTheDragon]

Are you thinking about the infamous Hale (2000) paper that analyzed Marshallese with four vowel phonemes //, //, /☎/, /☯/?

Not specifically, no, but that's a good real-world example of what I mean.