Rare/unusual natlang features

[bradrn]

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

Is that even a phoneme?

Well, it makes about as much sense as any other analysis of Marshallese vowels. From the Wikipedia Article:

Superficially, 12 Marshallese vowel allophones appear in minimal pairs, a common test for phonemicity. For example, [mʲææ̯] (mā, 'breadfruit'), [mʲæ͡ɑɑ̯] (ma, 'but'), and [mʲæ͡ɒɒ̯] (mo̧, 'taboo') are separate Marshallese words. However, the uneven distribution of glide phonemes suggests that they underlyingly end with the glides (thus /mʲaj/, /mʲaɰ/, /mʲaw/). When glides are taken into account, it emerges that there are only 4 vowel phonemes.

Wikipedia analyses these vowels as being /ɨ/, /ɘ/, /ɜ/, /a/, as per MED (1976), but obviously Hale (2000) has chosen a different notation.

[Kuchigakatai]

But yes, since on replaces nous in colloquial French, and the 3pl and 3sg have merged unexpectedly for non-phonological reasons (not unlike Lithuanian)...

The merger of French 3S and 3P in regular verbs (all tenses of -er verbs except the future, and all tenses of -ir verbs except the present indicative and future) is perfectly derived from regular sound changes. Sound change in French is just brutal like that.

Linguoboy: 3pl /-ɔ̃/ a retention of the expected reflex of -ent, or are we looking at something else? (Somewhat reminiscent of 1pl-3pl syncretism in German.)

My guess would be that the -ont ending of ils sont 'they are', ils font 'they do', ils ont 'they have' and ils vont 'they go' probably spread to the rest of verbs, cf. the similar spread of -no in Italian from sum sunt > *[son son] + 1S *-[o] > *[ˈsono son] > sono sono (creating wonderful things like habent > *[an] > hanno, and cantant > *[ˈkanta] > cantano).

[Xwtek]

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

Is that even a phoneme?

Well, it makes about as much sense as any other analysis of Marshallese vowels. From the

No, I mean, I saw "hot beverage", "soccer ball", "black telephone", and "ying yang" emoji as the phoneme inventory of the Marshallese

[Ryusenshi]

//, //, /☎/, /☯/

No, I mean, I saw "a cup of coffee", "soccer ball", "rotary telephone", and "ying-yang" as the phoneme inventory of the Marshallese

Yes, that's the point. Phoneme notation is kinda arbitrary: when there is little allophony, you may as well use the same symbol as the phone, but when allophony is rampant, you could use whatever. KathTheDragon suggested using // (nose emoji) for the syllable-final nasal of Japanese, since people disagree about what it is, exactly.

Marshallese is usually analyzed with four vowel phonemes, but there is so much allophony that describing each phoneme is a nightmare.In a famous paper, in 2000, Hale decided to embrace the crazy and use nonsensical symbols /, , ☎, ☯/ (coffee, football, phone, yin-yang) for the phonemes.

[Xwtek]

//, //, /☎/, /☯/

No, I mean, I saw "a cup of coffee", "soccer ball", "rotary telephone", and "ying-yang" as the phoneme inventory of the Marshallese

Yes, that's the point. Phoneme notation is kinda arbitrary: when there is little allophony, you may as well use the same symbol as the phone, but when allophony is rampant, you could use whatever. KathTheDragon suggested using // (nose emoji) for the syllable-final nasal of Japanese, since people disagree about what it is, exactly.

Marshallese is usually analyzed with four vowel phonemes, but there is so much allophony that describing each phoneme is a nightmare.In a famous paper, in 2000, Hale decided to embrace the crazy and use nonsensical symbols /, , ☎, ☯/ (coffee, football, phone, yin-yang) for the phonemes.

wat.

More: show

Well, I thought there is some problem in the processing of unicode characters, where IPA character is encoded as emoji character and it requires a special font to render the character as IPA instead of emoji

[Pabappa]

I remember seeing that paper and I assumed it was a Unicode error too .... emojis didnt exist yet but there were all sorts of problems with converting from one character set to another. But apparently it was all intentional.

No free access to the original paper online, it seems, but another paper confirms that it was intentional, yes.

[Xwtek]

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).

Thanks for the idea for my signature!

[anteallach]

English distinguishes θ/ð from s/z, and has gendered pronouns but not gendered nouns.

I'd have thought having θ/ð and not contrasting them with s/z, like Turkmen, is more unusual.

Yeli Dnye aka Yele or Yeletnye has doubly articulated bilabial-alveolar and bilabial-postalveolar stops as well as the much commoner labial-velars.

[Salmoneus]

Some of the rarest sounds are the linguolabial consonants found in many languages on the island of Espiritu Santo, and some on neihbourhin Malakula (presumably an areal effect, as the Malakula and Santo lanuaes are not particularly closely related within Southern Oceanic). They've also been found in just a handful of other lanuaes in the world.

The weird thin, thouh, is that there's no obvious reason for them to be so rare - they don't involve any awkward or weird tonue movements, and everyone can learn to do them reliably in just seconds - just put your tonue as for /T/, and move the tip forward a little until it touches your upper lip. They're reasonably acoustically distinct - the stops in particular do tend to mere into alveolars, but they're no less distinct than the various dentals, alveolars, postalveolars, all variously laminal or apical, that other lanuaes distinuish. They just... don't seem to occur to people!

[bradrn]

Some of the rarest sounds are the linguolabial consonants found in many languages on the island of Espiritu Santo, and some on neihbourhin Malakula (presumably an areal effect, as the Malakula and Santo lanuaes are not particularly closely related within Southern Oceanic). They've also been found in just a handful of other lanuaes in the world.

The weird thin, thouh, is that there's no obvious reason for them to be so rare - they don't involve any awkward or weird tonue movements, and everyone can learn to do them reliably in just seconds - just put your tonue as for /T/, and move the tip forward a little until it touches your upper lip. They're reasonably acoustically distinct - the stops in particular do tend to mere into alveolars, but they're no less distinct than the various dentals, alveolars, postalveolars, all variously laminal or apical, that other lanuaes distinuish. They just... don't seem to occur to people!

For me personally, pronouncing a linguolabial does feel like quite an awkward tongue movement: for dental, alveolar, postalveolar… consonants the relevant part of the tongue only needs to be moved a short distance, but for linguolabials the tip of the tongue must be moved quite a long distance, from behind the teeth to the top lip. Maybe that might go some way towards explaining why linguolabials are so rare, although it could just as easily have nothing to do with it.

[Xwtek]

Also, the way into linguolabial itself is pretty rare. It's allegedly from labial consonant after a nonrounded vowel. But it should be rather stable for me.

[Salmoneus]

Some of the rarest sounds are the linguolabial consonants found in many languages on the island of Espiritu Santo, and some on neihbourhin Malakula (presumably an areal effect, as the Malakula and Santo lanuaes are not particularly closely related within Southern Oceanic). They've also been found in just a handful of other lanuaes in the world.

The weird thin, thouh, is that there's no obvious reason for them to be so rare - they don't involve any awkward or weird tonue movements, and everyone can learn to do them reliably in just seconds - just put your tonue as for /T/, and move the tip forward a little until it touches your upper lip. They're reasonably acoustically distinct - the stops in particular do tend to mere into alveolars, but they're no less distinct than the various dentals, alveolars, postalveolars, all variously laminal or apical, that other lanuaes distinuish. They just... don't seem to occur to people!

For me personally, pronouncing a linguolabial does feel like quite an awkward tongue movement: for dental, alveolar, postalveolar… consonants the relevant part of the tongue only needs to be moved a short distance, but for linguolabials the tip of the tongue must be moved quite a long distance, from behind the teeth to the top lip. Maybe that might go some way towards explaining why linguolabials are so rare, although it could just as easily have nothing to do with it.

But the actual tongue movement is tiny!

If you have your tongue as for /T/, and just push it a milimetre further, you get a lin'uolabial. If you have it as for dental /t/ and push it a couple of milimetres further, you have a linguolabial.

Compared to the tongue motion involved in, say, a subapical retroflex, it's nothing.

And it involves only the tongue-tip - the most dextrous and consciously controlled bit of the speech apparatus - rather than anythin dorsal and weird. And it doesn't involve multiple motions in a defined order, or simultaneous motions, and it doesn't involve anything in the throat or larynx... so compared to most 'weird' sounds, it's incredibly easy.

Try gettin some schoolchildren and teachin them linguolabials, and then teaching them ejectives, pharyngealised retroflexives, faucal voice, velar laterals, nasalised lateral clicks, coarticulated alveolar-velar implosives, and so forth. Linguolabials will be grasped much sooner, I'm sure! And yet linguolabials are rarer than any of these...

[Pabappa]

Is it possible it just looks too silly in person and people avoid developing those sounds because it's hard to say anything dark and serious with linguolabials? i see it as a bit like Pabappa with all its regular labials ...i get a kick out of translating physics terms, fishing and farming terminology etc into Poswa & Pabappa but i think that if those languages were real even the native speakers would blush when it was their turn with the mic at a multilingual confrefence.

[Salmoneus]

Other languages aren't English, so the cultural associations some English speakers might have with a given sound wouldn't apply. [although I can't say i'm familiar with those associations myself - if anything, I'd see labials as particularly suited to 'dark' concepts, but that might just be because I grew up with Tolkien (Moria, Mordor, Balrog, etc) and Irish mythology (Balor, the Morrigan/the Badb, etc), and the various villainesses of English legends (Morgan, Morgause, Mab, etc)]

[bradrn]

Some of the rarest sounds are the linguolabial consonants found in many languages on the island of Espiritu Santo, and some on neihbourhin Malakula (presumably an areal effect, as the Malakula and Santo lanuaes are not particularly closely related within Southern Oceanic). They've also been found in just a handful of other lanuaes in the world.

The weird thin, thouh, is that there's no obvious reason for them to be so rare - they don't involve any awkward or weird tonue movements, and everyone can learn to do them reliably in just seconds - just put your tonue as for /T/, and move the tip forward a little until it touches your upper lip. They're reasonably acoustically distinct - the stops in particular do tend to mere into alveolars, but they're no less distinct than the various dentals, alveolars, postalveolars, all variously laminal or apical, that other lanuaes distinuish. They just... don't seem to occur to people!

For me personally, pronouncing a linguolabial does feel like quite an awkward tongue movement: for dental, alveolar, postalveolar… consonants the relevant part of the tongue only needs to be moved a short distance, but for linguolabials the tip of the tongue must be moved quite a long distance, from behind the teeth to the top lip. Maybe that might go some way towards explaining why linguolabials are so rare, although it could just as easily have nothing to do with it.

But the actual tongue movement is tiny!

If you have your tongue as for /T/, and just push it a milimetre further, you get a lin'uolabial. If you have it as for dental /t/ and push it a couple of milimetres further, you have a linguolabial.

Compared to the tongue motion involved in, say, a subapical retroflex, it's nothing.

And it involves only the tongue-tip - the most dextrous and consciously controlled bit of the speech apparatus - rather than anythin dorsal and weird. And it doesn't involve multiple motions in a defined order, or simultaneous motions, and it doesn't involve anything in the throat or larynx... so compared to most 'weird' sounds, it's incredibly easy.

Try gettin some schoolchildren and teachin them linguolabials, and then teaching them ejectives, pharyngealised retroflexives, faucal voice, velar laterals, nasalised lateral clicks, coarticulated alveolar-velar implosives, and so forth. Linguolabials will be grasped much sooner, I'm sure! And yet linguolabials are rarer than any of these...

Good points, especially the last paragraph.

[bradrn]

I’ve just discovered that Tlingit has /ɬ/, /ɬʼ/, /t͡ɬ/, /t͡ɬʰ/ and /t͡ɬʼ/, but no /l/. (Except for some older speakers, who have [l] as an allophone of /n/.) Admittedly, I’m not sure how reliable Wikipedia is for these things, but Omniglot seems to concur.

[Vijay]

Having /ɬ/ but not /l/ is pretty common. See: Mongolian, Amis, Avar, Chukchi, Muscogee/Mvskoke/Creek, Dogrib, Saaroa...

[bradrn]

Having /ɬ/ but not /l/ is pretty common. See: Mongolian, Amis, Avar, Chukchi, Muscogee/Mvskoke/Creek, Dogrib, Saaroa...

According to Wikipedia at least, Avar and Muscogee have /l/. But this does appear to be more common than I thought!

[Nortaneous]

Having /ɬ/ but not /l/ is pretty common. See: Mongolian, Amis, Avar, Chukchi, Muscogee/Mvskoke/Creek, Dogrib, Saaroa...

According to Wikipedia at least, Avar and Muscogee have /l/. But this does appear to be more common than I thought!

According to PHOIBLE: Kabardian, Nootka, Chukchi, Ahtna, Lushootseed, Tseshaht, Tigak, Tlingit, Shona, Bana, Rigwe, Chulupí, Korubo, Jurúna, Ket, Mongolian

[Vijay]

Lushootseed has both /l/ and /ɬ/, Shona has neither...