Are palatals coronal or dorsal?

[PVER•PVERVM•AMAT]

Which is the correct way? I seem to be able to strike my palate with both the tip and the body of my tounge; which is correct?

[Aurora Rossa]

I think it would depend on the language and the diachronic origin of the palatal series. Some languages may treat them like dorsal stops if they derive from palatalized velars while others might treat them more like coronals if they derive from palatalized alveolars.

[ObsequiousNewt]

Palatal consonants are consonants articulated with the body of the tongue raised against the hard palate (the middle part of the roof of the mouth).

They are dorsal. If you articulate it coronally, it's retroflex.

[Nortaneous]

they can be either, eddy is right even though he is doing the usual terrible plato well-gee-if-i-just-think-hard-at-it-enough-i-can-solve-anything thing and not giving examples (they're coronal in albanian and the chinese langs that have them and dorsal in irish gaelic)

Palatal consonants are consonants articulated with the body of the tongue raised against the hard palate (the middle part of the roof of the mouth).

They are dorsal. If you articulate it coronally, it's retroflex.

no, no, this is wrong, you are wrong, alveolopalatals are a thing. they really should add ȶ ȡ ȵ ȴ to the IPA considering that alveolopalatals and velopalatals contrast in at least one natlang

[Jess]

I'm under the impression that it's just an imprecise term, an archi-POA if you will, that is applied to several different phonetic realizations which may be further specified if necessary. That said, if someone is using the term in a more narrow, phonetic sense (e.g., treating "palatal" and "alveolopalatal" as mutually exclusive categories instead of a superset/subset relationship), I think the prototypical/cardinal palatal is dorsal.

they really should add ȶ ȡ ȵ ȴ to the IPA considering that alveolopalatals and velopalatals contrast in at least one natlang

Ooooh really? Which one?

[----]

Lhasa Tibetan distinguishes /tɕ/ from /c/, which is almost the same thing I suppose.

Also Yeli Dnye doesn't have "true palatals" but it does distinguish palatalized dentals, palatalized alveolars, and palatalized velars.

[Nortaneous]

I'm under the impression that it's just an imprecise term, an archi-POA if you will, that is applied to several different phonetic realizations which may be further specified if necessary.

exactly. a palatal is anything between alveolar and velar that isn't apical.

they really should add ȶ ȡ ȵ ȴ to the IPA considering that alveolopalatals and velopalatals contrast in at least one natlang

Ooooh really? Which one?

yanyuwa

[Radius Solis]

Nort is right.

"Palatal" is a fake POA phonetically. It's a very real phonological POA - it's extremely common for there to be a series of sounds between the /t/ and /k/ series - but phonetically, exactly what sort of sounds a language fills that space with is a crapshoot. It can be anything from a slightly-forward version of [k] to a somewhat-backed or laminal alveolar affricate. In describing phonology any such series can loosely be labelled "palatal" if there's only one - English is a good example, occasionally you'll see our /tS dZ S Z j/ called "palatal" - but it doesn't imply much anything about the articulation. Thus in describing phonetics, aside from the approximant, it is often best to avoid the term. There is no single articulation that makes a "true [c]".

[Nortaneous]

The thing between /t/ and /k/ can even be /ts/.

Are there any languages that do odd things with their /j/ like Japanese and Yele do with /w/? Besides Tsou, which just lowers it to a semivowel version of [e].

Then again, those languages would probably not be described as having /j/. /w/ is just any sort of labial usually-coarticulated approximant; there's a lot more room for weird shit there before it stops being /w/. The only variant of /j/ I've seen is the Tsou thing and raising it to a fricative like what Proto-Basque is reconstructed as having; anything weirder than that would probably be called something else, I'm guessing.

[kodé]

Some old-school feature geometry approaches to segments treat palatals as having both a coronal and dorsal specifications.

[Ser]

The thing between /t/ and /k/ can even be /ts/.

Are there any languages that do odd things with their /j/ like Japanese and Yele do with /w/? Besides Tsou, which just lowers it to a semivowel version of [e].

Spanish [ i], [j] and [dʒ] (or their dialectal variants, [ i]~[ʝ]~[ɟʝ] if you will) are often analyzed as the same phoneme: /i/. This explains why Spanish speakers in the US often pronounce English you as [dʒu], and teenager [ti.ˈne.jeɾ], if you were wondering.

This is particularly striking for urban Argentinian Spanish, where [ i], [ʃ]~[ʒ] and [dʒ] can effectively be analyzed as the same phoneme /i/. Niño /ˈni.ɲo/ [ˈni.ɲo], allá /a.ˈia/ [a.ˈʃa]~[a.ˈʒa], en llave /en.ˈia.be/ [en.ˈdʒa.βe]. It seems that the strong phonetic differences between [ʃ] and [ i]/[dʒ] are already destabilizing the system though, and there are Argentinians who distinguish the two as different phonemes: [ˈʃeɾ.βa]~[ˈʒeɾ.βa] 'weed' vs. [ˈjeɾ.βa]~[ˈdʒeɾ.βa] 'grass' (both words are the same word hierba /ˈieɾ.ba/ in other dialects).

Besides this, there are also speakers who believe that hiV- and yV-, as in hielo 'ice' and yendo 'going', should be pronounced differently (as [j-] and [ˈdʒ-], or [j-] and [ʝ-]~[ɟʝ-], respectively) because they're spelled differently. (The Academies have been insisting for decades these hiV-'s and yV-'s are pronounced the same way though.) Naturally, many find this awkward, and the hypercorrected pronounciation hielo [i.ˈe.lo] is heard from people attempting to pronounce [ˈje.lo].

Lowering intervocalic /i/ [j] to a semivowel [e] is a thing in Mexican/Central American Spanish too.

There is some Romance language or particular dialect I can't remember where word-final /j/ can be devoiced to something phonetically similar to [ʃ].


Going back to /w/, [ u], [ɣw] and [gw] are traditionally analyzed as the same phoneme in Spanish too, /u/, with the same allophone distribution you find in /i/ [ i]~[j]~[dʒ] (or [ i]~[ʝ]~[ɟʝ]). English is having a hand in phonologizing a /gw/ vs. /u/ distinction though: pronouncing borrowings such as whisky with [gwi]~[ɣwi] is stigmatized among speakers better familiar with English, who insist should start with [w]: [ˈwis.ki]. Naturally, many find this awkward, and the hypercorrection [u.ˈis.ki] is also heard from people who try too hard.

[Nortaneous]

so what's the minimum number of consonants any dialect of spanish can be analyzed as having? as far as i can tell the minimum is 14 (p b t d k g m n J f s x 4 l). any way to get rid of two more?

[vec]

so what's the minimum number of consonants any dialect of spanish can be analyzed as having? as far as i can tell the minimum is 14 (p b t d k g m n J f s x 4 l). any way to get rid of two more?

Are you analysing ñ as /ni/?

[Hallow XIII]

no that is what X-SAMPA <J> stands for

[KathTheDragon]

(p b t d k g m n J f s x 4 l)

= /p b t d k g m n ɲ f s x ɾ l/, for those who don't read X-SAMPA

[vec]

Ah! Gotcha. My X-SAMPA is getting very rusty.

[Drydic]

Going back to /w/, [ u], [ɣw] and [gw] are traditionally analyzed as the same phoneme in Spanish too, /u/, with the same allophone distribution you find in /i/ [ i]~[j]~[dʒ] (or [ i]~[ʝ]~[ɟʝ]). English is having a hand in phonologizing a /gw/ vs. /u/ distinction though: pronouncing borrowings such as whisky with [gwi]~[ɣwi] is stigmatized among speakers better familiar with English, who insist should start with [w]: [ˈwis.ki]. Naturally, many find this awkward, and the hypercorrection [u.ˈis.ki] is also heard from people who try too hard.

Are you seriously saying that agua is analyzed as /aua/?

[Ser]

Ah! Gotcha. My X-SAMPA is getting very rusty.

He did forget /tS/ though.

Are you seriously saying that agua is analyzed as /aua/?

Yes, /ˈa.ua/ [ˈa.ɣwa], since [ˈawa] (or any sort of intervocalic [w]) doesn't exist (unless you're one of those speakers severely influenced by English).

[Thry]

Serafín you make me cringe everyday.

chihuahua, cacahuete, microhuerta?

I'd surely read microhuerta and microgüerta if it existed differently?!?!

I must be one of those speakers severely influenced by English.

PS: also it wouldn't be a sin to even analyze them as /"a.gwa/ [a.Gwa] and /mi.k4o."we4.ta/ [mi.k4o."Gwe4.ta], I've never seen this supposed trend of analyzing them merged phonemically [though who cares about phonemes nowadays].