zompist bboard

They probably don't need any consonants, having such a vowel inventory?

Dzongkha:

Syllables beginning with a voiceless or aspirated plosive, affricate or sibilant, a voiceless liquid or /h/ are pronounced in the high tone ... whereas syllables with a voiced or devoiced plosive, affricate or sibilant initial or initial /r/ are in the low tone. Syllables beginning with a vowel, a nasal or voiced liquid other than /r/ can be either high or low tone.

Devoiced plosives come from earlier voiced plosives: they're pronounced as unvoiced but impart breathy voice and low tone on the following vowel.

The paper is here, and anyone who can read Tibetan script can probably figure out the diachronics.

(Also, the umlauted vowels are inherently long in Dzongkha, and there are labial-palatal affricates /pcʰ pc bɟ bɟ°/. ° is the convention for devoiced plosives, but they could just as well be written /pcɦ/.)

Ukue (f v h), Cacua (ʍ h), several Polynesian languages and Sentani (f h), and Koiari (f ð h) have fricatives but no sibilants.

Nortaneous wrote:Ukue (f v h), Cacua (ʍ h), several Polynesian languages and Sentani (f h), and Koiari (f ð h) have fricatives but no sibilants.

So does Northern Tairora.

gufferdk wrote:

Birdlang wrote:So more like 64 vowels. And Danish is the only language to have a low front rounded vowel.

Well not 64 phonemes due to reduction and rules for the distribution of stød based on vowel length, following consonants, placement of syllabic stress, ... Also stød can be interpreted as a prosodic feature rather than a property of the vowels which leaves one with "only" ~32 vowel phonemes.
With regars to /ɶ/ danish is the only language I know has one and when looking at the WALS page for front rounded vowels it seems plausible that it is very rare or even restricted to danish but do you have any source for that. If yes, then i would really like to see it.

I'm a little late to the party, but... I think Limburgish (my native language) also has a /ɶ œ/ contrast. /ɶː/ occurs, for example, as the umlauted version of /ɒː/.

Here's a minimal pair to illustrate:
bäök (I cry) - /bɶːk/
bök (I stoop) - /bœk/
böök (books) - /bøːk/

Now, /ɶː/ only occurs long, but I don't think it's an allophone of /œ/ (which only occurs short) or /øː/, because you cannot substitute one for the other without it being confusing/unusual sounding.

Note that the actual realization of /ɶː/ is more like [œ‿ɶ] though (a glide from [œ] to a very open [ɶ])

Also note that there's no official orthography, and that there's a lot of dialectal variations. I think the guidelines used most often (though loosely and inconsistently) are too 'Dutch' for the language, which makes a lot of sense from a practical point of view (it's the writing system all speakers are going to be familiar with), but a more German-style writing system makes more sense (though with more consistent ways of writing double vowels, like I did above).

I only just started reading it but holy shit vowel harmony in a dialect of Scots! Blocked by certain intervening consonants.

Nortaneous wrote:Kalaallisut is the only language with a uvular nasal.

(There are four languages with one in PHOIBLE, but the 'uvular nasal' in Japanese [and Burmese, which isn't listed] is a convention of notation, the Kusunda uvular nasal is probably a /ŋʕ/ cluster, and the fourth is Kinyarwanda, which doesn't have one.)
.

Bit old, but earlier in the thread (a year earlier, actually), you posted

Nortaneous wrote:Luobenzhuo Bai apparently has a contrastive uvular nasal, which can also occur as the syllable nucleus.

Do you know if the Bai one is correct or not? (It may have been a new discovery)

The Japanese coda nasal, judging from what spoken Japanese I have heard, varies a lot. Before another consonant, it assimilates to that consonant's POA (as in shimbun 'newspaper'); and in at least some dialects, it doesn't surface as a segment at all but merely nasalizes the vowel.

Maq Lar: Point. The source is here, which gives an almost-minimal pair of ŋa44 'diligent' / ɴa21 'sweat'. I'd take that with a grain of salt, since it's a highly unusual feature reported without comment by an institution that does not consistently produce the highest-quality work -- it's interesting enough to post here, but it's not an ironclad demonstration of phonemicity, or even transcription accuracy. But it's possible that it has a phonemic uvular nasal.

----

Tagdal is a Songhay language with extensive Berber influence. Its verbal derivational morphology consists of three prefixes -- a causative, a passive, and a reciprocal -- all of which are loaned from Berber. The causative and passive can only be applied to verbs of Berber origin. Verbs of Songhay origin are causativized by suppletion with Berber loans that are otherwise not present in the Tagdal lexicon -- so the causative of the Berber loan ənfər 'push aside' is sənfər, but the causative of the native verb koy 'go' is səglu, and there is no Tagdal word əglu. The same thing applies to passives: the passive of ənfər is tuwənfər, but the passive of bay 'know' is tuwasən, derived from a Berber root sən that doesn't exist in that form in Tagdal. There is only one Songhay verb with a nonsuppletive causative: the causative of kan 'fall' is kanda (from kan nda 'make fall') rather than səggədəl, probably because səggədəl means 'limp'. (source)

Cf. Algerian Arabic, which is apparently in the process of borrowing the French articles for use with French loanwords.

Nortaneous wrote:Dzongkha:

Syllables beginning with a voiceless or aspirated plosive, affricate or sibilant, a voiceless liquid or /h/ are pronounced in the high tone ... whereas syllables with a voiced or devoiced plosive, affricate or sibilant initial or initial /r/ are in the low tone. Syllables beginning with a vowel, a nasal or voiced liquid other than /r/ can be either high or low tone.

Isn't something similar true of Tibetan as well?

Nortaneous wrote:Dzongkha:

Syllables beginning with a voiceless or aspirated plosive, affricate or sibilant, a voiceless liquid or /h/ are pronounced in the high tone ... whereas syllables with a voiced or devoiced plosive, affricate or sibilant initial or initial /r/ are in the low tone. Syllables beginning with a vowel, a nasal or voiced liquid other than /r/ can be either high or low tone.

Devoiced plosives come from earlier voiced plosives: they're pronounced as unvoiced but impart breathy voice and low tone on the following vowel.

The paper is here, and anyone who can read Tibetan script can probably figure out the diachronics.

I thought voiced initial > low tone was a pretty well-established phonological process cross-linguistically at this point.

(Also, the umlauted vowels are inherently long in Dzongkha

Were they perhaps diphthongs at some point?

and there are labial-palatal affricates /pcʰ pc bɟ bɟ°/. ° is the convention for devoiced plosives, but they could just as well be written /pcɦ/.)

That's pretty cool. Reminds me of Hmong's "bilabial lateral" series. I also remember some Bantu langugage had /pʃ bʒ/ onsets with few or no clusters permitted, although I forget which.

Chengjiang wrote:

and there are labial-palatal affricates /pcʰ pc bɟ bɟ°/. ° is the convention for devoiced plosives, but they could just as well be written /pcɦ/.)

That's pretty cool. Reminds me of Hmong's "bilabial lateral" series. I also remember some Bantu langugage had /pʃ bʒ/ onsets with few or no clusters permitted, although I forget which.

Sotho. Actually seems pretty common for palatalized labials to end up as labial+palatal clusters of some kind, or even just plain coronals. Some Polish dialects have /pj bj mj/ [pɕ bʑ mɲ], there's Proto-Greek /pj pʰj mj/ > modern /pt pt in/, there's cases like Latin /rabie:s/ to French /raʒ/ and Romansch /ravʒa~rabʒa/, and there's Standard Thai /pla: pla:w/ versus Nung /pja: pja:w/ versus Bouyei /tɕa: tɕu:/ (vowels might be off, old transcription).

vokzhen wrote:

Chengjiang wrote:

and there are labial-palatal affricates /pcʰ pc bɟ bɟ°/. ° is the convention for devoiced plosives, but they could just as well be written /pcɦ/.)

That's pretty cool. Reminds me of Hmong's "bilabial lateral" series. I also remember some Bantu langugage had /pʃ bʒ/ onsets with few or no clusters permitted, although I forget which.

Sotho. Actually seems pretty common for palatalized labials to end up as labial+palatal clusters of some kind, or even just plain coronals. Some Polish dialects have /pj bj mj/ [pɕ bʑ mɲ], there's Proto-Greek /pj pʰj mj/ > modern /pt pt in/, there's cases like Latin /rabie:s/ to French /raʒ/ and Romansch /ravʒa~rabʒa/, and there's Standard Thai /pla: pla:w/ versus Nung /pja: pja:w/ versus Bouyei /tɕa: tɕu:/ (vowels might be off, old transcription).

Yes, indeed. I'm also quite fond of Russian's [mlʲ plʲ blʲ].

vokzhen wrote:

Chengjiang wrote:

and there are labial-palatal affricates /pcʰ pc bɟ bɟ°/. ° is the convention for devoiced plosives, but they could just as well be written /pcɦ/.)

That's pretty cool. Reminds me of Hmong's "bilabial lateral" series. I also remember some Bantu langugage had /pʃ bʒ/ onsets with few or no clusters permitted, although I forget which.

Sotho. Actually seems pretty common for palatalized labials to end up as labial+palatal clusters of some kind, or even just plain coronals. Some Polish dialects have /pj bj mj/ [pɕ bʑ mɲ], there's Proto-Greek /pj pʰj mj/ > modern /pt pt in/, there's cases like Latin /rabie:s/ to French /raʒ/ and Romansch /ravʒa~rabʒa/, and there's Standard Thai /pla: pla:w/ versus Nung /pja: pja:w/ versus Bouyei /tɕa: tɕu:/ (vowels might be off, old transcription).

So it seems. The argument to analyse these as distinct phonemes is also simply distributional. I haven't encountered any reason to think that these would be released differently from clusters.

Chengjiang wrote:Were they perhaps diphthongs at some point?

The regular Tibetic path is the loss of syllable final dentals. So in any case, the originals were bimoraic.

vokzhen wrote:there's cases like Latin /rabie:s/ to French /raʒ/

Didn't /b d g/ 0 / V_V in Western Romance? So the source of /ʒ/ is less mysterious, the change was more like rabie:s raje raʒe raʒ.

Αυτοβοτα wrote:

vokzhen wrote:there's cases like Latin /rabie:s/ to French /raʒ/

Didn't /b d g/ 0 / V_V in Western Romance? So the source of /ʒ/ is less mysterious, the change was more like rabie:s raje raʒe raʒ.

The paper also points out <sapiat> yielding French <sâche> and Romansch <sapcha> where I believe we'd expect sapiat > sabia > sa(v)ge if there wasn't odd palatalization of labials going on. Going to wiktionary it even says on the page for sache that Latin pj > French ch regularly, giving sepia > seiche, appropiāre > approcher, Frankish happja > hache.

And in Old French <g> before <e, i, y> was pronounced an affrcate, not an approximant.

And AFAICT those voiced plosives turned into voiced fricatives rather than disappearing, at least at first.

Kashmiri has a phonemic contrast between palatalized and non-palatalized consonants. OK, that's not cross-linguistically weird, but that does seem pretty weird for India. Its basic word order is either SVO or V2 from what I've heard. I don't think I know of any other Indian language that has that word order. It sounds more like German.

It also apparently has a phonemic contrast between /ɨ/, /ɨː/, /ə/, and /əː/, in addition to the /a aː i iː u uː e eː o oː/ vowels that are typical of e.g. Dravidian languages.

Vijay wrote:It also apparently has a phonemic contrast between /ɨ/, /ɨː/, /ə/, and /əː/, in addition to the /a aː i iː u uː e eː o oː/ vowels that are typical of e.g. Dravidian languages.

Not particularly rare to contrast /ɨ ə/, and it's really common in SEA and the eastern Himalayan area, so that's not really as much of an odd natlang feature so much as a place you wouldn't particularly expect it (but then, you probably wouldn't expect it in Great Britain [Northern Welsh] or in a Romance Language [Romanian] either).

Just wanted to say thanks!

<Ser> in Spanish, it's funny
<Ser> when you say "if he comes", that is, when you're not sure somebody's gonna come in the future, then you use the indicative present: si viene
<Ser> but when you say "when he comes", that is, when you are sure somebody's gonna come in the future, then you use the subjunctive present: cuando venga
<Ser> it's like, the opposite of the stereotypical uses of the indicative and subjunctive

Manambu has a "substitutive" case meaning "instead of X". It also has two cases marking mode of transportation, and may also attach case affixes to verb roots, with the locative being a completive and the dative a purposive.

In Hungarian, the distal demonstrative pronoun is az. Before a vowel, the definite article is also az (it's a before a consonant. The copula is dropped in 3rd person when linking nouns with nouns or nouns with adjectives ... so ...

az alma = the apple
Az alma. = That is an apple.
Az az alma. = That is the apple.

To use demonstratives as determiners ... they must precede the definite article, so ...

az az alma = that apple
Az az az alma. = That is that apple.

Maniq and Mam having no word nicely equivalent to "to eat", but rather a group of words chosen depending on the food, as mentioned in:

viewtopic.php?f=7&t=44462

I'm actually not sure that's so odd. At least, I know anecdotally that some languages have specialized words for 'to eat'...like 'to eat (anything)' vs. 'to eat a meal', and this distinction is found across a pretty wide variety of languages, too.