Sound Change Quickie Thread

[Knit Tie]

So if /ɫ/ cam become /ɰ/ pretty much everywhere, why not say that it becomes /ɣ/ later on throuɡh /ɫ/ → /ɰ/ → /ɣ/?

That is a possibility, but I would limit it to the most stressed positions, i.e. initial consonants in stressed onsets and geminates.

Just there?

[StrangerCoug]

Also, I'm wondering if there's a dialect of English where both the rhotics and the labials are transitioning to something else - that could potentially make for a cool proto-language with no liquids.

There are English English dialects where /r/ has become [ʋ].

This sometimes happens as a result of a speech impediment as well. I had it myself as a kid, and I'd broadly transcribe my case as /ɹ/ → /w/ in prevocalic position. (I speak a rhotic dialect and usually got the rhotic vowels right.) You could combine Knit Tie's ideas and mine and go /ɹ/ → /ʋ/ → /w/ or something like that.

[linguistcat]

If I have a sound change that velarizes the previous consonant, but said consonant was already palatalized, would it be more likely it would just velarize, remain palatalized, become "plain" (neither velarized nor palatalized), or would some other sound change be likely?

And likewise for the reverse, a palatalizing sound change on velarized consonant?

[Pabappa]

For my conlangs palatalized always wins unless the consonant is a labial.

[Xwtek]

For my conlangs palatalized always wins unless the consonant is a labial.

Why different rule for labial?

[linguistcat]

For my conlangs palatalized always wins unless the consonant is a labial.

But is this based on personal preference backed by phonological plausibility? or because palatalization is "stronger" than velarization or other secondary articulations besides labial?

[Travis B.]

So if /ɫ/ cam become /ɰ/ pretty much everywhere, why not say that it becomes /ɣ/ later on throuɡh /ɫ/ → /ɰ/ → /ɣ/?

That is a possibility, but I would limit it to the most stressed positions, i.e. initial consonants in stressed onsets and geminates.

Just there?

Sorry for the delay - was on a trip and did not have access to my laptop, and typing on my phone is a pain.

You could have unconditional fortition of /ɰ/ - it just is to me that it is a rather "weak" consonant to begin with, which makes it seem like that if it were to undergo fortition at all it would undergo fortition in the positions most prone to this, i.e. at the starts of stressed onsets and when geminate.

[Pabappa]

For my conlangs palatalized always wins unless the consonant is a labial.

But is this based on personal preference backed by phonological plausibility? or because palatalization is "stronger" than velarization or other secondary articulations besides labial?

The first one, i think .... personal preference, because I dont like palatalized labials very much....but i do believe that when palatalization drops, it often drops from labials before other series. e.g. Slavic did /pʲ bʲ/ > /pl bl/ i think, and when Slavic languages evolved a new set of palatalized labials, some languages dropped those as well at least in the coda. (e.g. the word for dove ends in /bʲ/ in Russian but /b/ in most other Slavic languages.)

I think the second idea might be true even so, though ..... instinctively it seems to me that a "velarized /c/" would still be fairly close in sound to /c/, and I cant think of any examples offhand of /c/ shifting to /k/ except in some languages before a back vowel (and even that is fairly rare). I wouldnt say so much that the labial articulation makes a sound stronger, but that simply because it doesnt use the tongue, the palatalization has no means to cling on like it does with the coronals and velars.

[Nortaneous]

If I have a sound change that velarizes the previous consonant, but said consonant was already palatalized, would it be more likely it would just velarize, remain palatalized, become "plain" (neither velarized nor palatalized), or would some other sound change be likely?

And likewise for the reverse, a palatalizing sound change on velarized consonant?

velarization is sort of fake, and palatalized vs. velarized and "plain" vs. emphatic are sort of the same thing. i don't think a three-way palatalized/plain/velarized system is attested. the closest i can think of is /l ɫ ʎ/ in some dialects of albanian, and this is limited to the laterals and could potentially be analyzed as parallel to /ts t c/ or something. (standard albanian lost /ʎ/.)

[bradrn]

velarization is sort of fake

What exactly do you mean by this?

palatalized vs. velarized and "plain" vs. emphatic are sort of the same thing.

This I can agree with. I assume that by “plain” you mean “without any secondary articulation”, and by “emphatic” you mean “pharyngealised”.

[Nortaneous]

velarization is sort of fake

What exactly do you mean by this?

I don't know of anything that would be lost aside from the narrowest articulatory detail if the difference between velarization, uvularization, and pharyngealization were removed

[Travis B.]

velarization is sort of fake

What exactly do you mean by this?

I don't know of anything that would be lost aside from the narrowest articulatory detail if the difference between velarization, uvularization, and pharyngealization were removed

But what if the narrowest articulatory detail between "plain" (i.e. no secondary articulation) and velarization was lost?

[Travis B.]

In general it seems like things can be palatalized, velarized, uvularized, or pharyngealized, but not more than one of the four. And it seems that if one attempts to combine palatalization and velarization in the same phone, palatalization wins.

[akam chinjir]

Suppose a language has rhythmic stress, and unfooted syllables are fairly common. Is there any tendency for vowels in the weak syllable of a foot to reduce more than do vowels in unfooted syllables? To reduce less? Could foot-type make a difference? (My thought is that with iambic feet you relatively often get lengthening of stressed syllables, maybe this spills over into reduction of the unstressed vowel---another way to give the strong foot greater weight than the weak foot. But I don't know if this is right.)

[mae]

-

[Nila_MadhaVa]

I've got a lang which starts off with five ejective stops, of which three are lost in pull chains and the other two simply loose ejectivity after that. I'm wondering , is it at all plausible for stressed syllables to become allophonically ejective as part of those changes?

[linguistcat]

I've got a lang which starts off with five ejective stops, of which three are lost in pull chains and the other two simply loose ejectivity after that. I'm wondering , is it at all plausible for stressed syllables to become allophonically ejective as part of those changes?

Off hand it seems pretty reasonable to me. But it would probably depend on what the pull chains were and when ejectivity became a feature of stress more than of the stops themselves.

And to sum things up for my own question, 1) velarization is a weak coarticulation so it would probably be dropped if a sound change applied other coarticulations, and a sound change to make something velarized would probably be blocked by other articulations being present; 2) Labials have a tendency to block or drop palatalization, or change it into something else like /l/; 3) Other than that, it's pretty much up to my preferences phonetically.

[dɮ the phoneme]

Suppose a language has rhythmic stress, and unfooted syllables are fairly common. Is there any tendency for vowels in the weak syllable of a foot to reduce more than do vowels in unfooted syllables? To reduce less? Could foot-type make a difference? (My thought is that with iambic feet you relatively often get lengthening of stressed syllables, maybe this spills over into reduction of the unstressed vowel---another way to give the strong foot greater weight than the weak foot. But I don't know if this is right.)

The foot type will make a difference. If a language allows for "moraic trochees", i.e., it allows heavy syllables to form trochaic feet in-and-of themselves, then you can get reduction of vowels in the weak syllable: thus (LL) > (H). This is primarily driven by a preference for the foot to be headed by a heavy syllable, and reducing the second vowel in an (LL) foot accomplishes this. Of course, this is only "legal" if the language allows for moraic trochees, since otherwise (H) isn't a valid foot. There is no parallel phenomenon for iambic feet IIRC. There is no moraic iamb. Also IIRC whether or not unfooted syllables undergo reduction is an independent parameter to whether the weak syllable of a trochee reduces.

But in truth I haven't done enough reading to say whether this account is actually borne out by the data, or is just the result of people trying to cram a bunch of vowel reduction phenomena into the framework of optimality theory.

[akam chinjir]

My recent reading suggests that (H) is available in all footing schemes (iambic, syllabic trochaic, moraic trochaic)---though I think you've got to be right that it's most at home in a system with moraic trochees, and it makes sense that in an iambic language there'd be resistance to losing the weak syllable entirely.

One thing that I think is supposed to be specific to languages with moraic trochees is shortening of the stressed vowel, (H́ L) → (Ĺ L).

...And it certainly is true that it's hard to read up on this stuff without getting lost in a sea of Optimality.

(Though come to think of it, one of the few bits of OT that's really appealed to me is Maria Gouskova's dissertation Deriving Economy: Syncopy in Optimality Theory, which náʼoolkiłí linked to back in April and which I really ought to look back at.)

Anyway thanks!

[Nila_MadhaVa]

I've got a lang which starts off with five ejective stops, of which three are lost in pull chains and the other two simply loose ejectivity after that. I'm wondering , is it at all plausible for stressed syllables to become allophonically ejective as part of those changes?

Off hand it seems pretty reasonable to me. But it would probably depend on what the pull chains were and when ejectivity became a feature of stress more than of the stops themselves.

While having lunch and typing out a reply, I realized that my reasoning for only including 3 out of 5 stops in the shifts was completely arbitrary since changes I'd already decided would kick in later will get me to where I want anyway. So I've adjusted my notes accordingly. As for the pull chains, they are ejective → voiceless → voiced → voiced non-sibilant fricative / V_V and for singletons. It might also be helpful to know that before these kick in, ejectives as the first element in clusters allophonically either transfer ejectivity to the following voiceless stops/fricatives or loose it, so phonemic ejectivity is on it's way out over a couple of successive stages and allophonic ejectives are already a thing.

With all that in mind, my thinking on stressed ejectives was that the speakers were still familiar with them, so while they were being lost phonemically the more "emphatic" environment of stressed syllables attracted ejectivity to the voiceless onset. Which leads me to another question: if this change is plausible, would it be reasonable for voiced stops to become allophonically implosive in the same environment by analogy?