Sound Change Quickie Thread

[StrangerCoug]

I find DVN DṼND DṼD more plausible than DVN DṼD directly, but I'm hard-pressed to see how they are implausible on their own. Voiced plosives can become nasals and vice-versa without a lot of difficulty if memory serves.

[GreenBowTie]

what might cause an unvoiced consonant/cluster at the beginning of a word to voice?

[hwhatting]

what might cause an unvoiced consonant/cluster at the beginning of a word to voice?

Sandhi (Assimilation to voicedness at the end of the previous word, intervocalic position due to previous word ending in a vowel.
Unconditioned voicing of intitial stops has happened in Turkish (/k/ > /g/, /t/ > /d/).

[vokzhen]

You've also got initial /s/ in German voicing, and dialectically for fricatives in English (no longer productive? fox > vixen). I'm sure I've heard of it elsewhere as well, but examples aren't coming to me. Clusters I'd think would resist voicing unless it includes a sonorant.

[Pogostick Man]

Initial voicing basically seems to be plausible.

[Grunnen]

Initial voicing basically seems to be plausible.

Yeah, and in Dutch it's not just the initial /s/ that got voiced, but initial /f/ did so as well. Compare English fare, Germen fahren, Dutch varen. In contemporary Dutch word initial fricatives have a tendency to devoice, so that <varen> may be pronounced with a [f], but that's a whole nother story.

[hwhatting]

You've also got initial /s/ in German voicing,

Not saying that this is no valid example, but it's part of a bigger picture, as intervocalic /s/ was voiced as well.

[Tropylium]

How plausible is vowel metastasis across consonants? Things like uCa -> Cua, or uCa -> aCu? What kinds of environment could trigger it?

Things like aCi > aCʲi > ajC(i) or aCu > aCʷu > auC(u) are attested (e.g. in Avestan, Livonian, and I've seen one proposal that Germanic umlaut also sort of went like this). Quite probably this could also happen in the other direction i.e. iCa >> Cia, uCa >> Cua.

If you want to limit the extent to which this happens, then e.g. "poorly palatalizable" or "poorly labializable" consonants can block the change:
ali > alʲi > aili > ail, but ari > ar (without **rʲ)
aku > akʷu > auku > auk, but abu > ab (without **bʷ)

I don't know about full metathesis, that seems unlikely to happen as a systematic change.

[KathTheDragon]

Isn't there -Cj- > -iC- for some consonants in some Greek dialects (the others typically having -Cj- > -CC-)?

[Tropylium]

Isn't there -Cj- > -iC- for some consonants in some Greek dialects (the others typically having -Cj- > -CC-)?

Yep, Greek too.

[8Deer]

uCV iCV > CuV CiV is quite common initially in Paman apparently. In some languages, even /a/ metathesises when followed by /i u/, which then become the corresponding glides, so /ati/ > /tay/.

[KathTheDragon]

How plausible is t,d > tʰ,t > t:,t ?

[StrangerCoug]

I find the tʰ t: change hard to buy unconditionally. Intervocalically, I can see it.

[احمکي ارش-ھجن]

I got a few questions about vowel mutation.
Does it (the vowel that causes mutation) only affect the previous vowel? Or can it affect a vowel that comes after it?
Does it only make the affected vowel the same quality as itself, or does it just cause, say, fronting and unrounding or backing and rounding, or perhaps it can do either/or?
Could it affect a vowel before it as well as after it?

[StrangerCoug]

Does it (the vowel that causes mutation) only affect the previous vowel? Or can it affect a vowel that comes after it?

It can affect following vowels as well. If it affects previous vowels, it's called regressive metaphony; if it affects following vowels, you have progressive metaphony.

Does it only make the affected vowel the same quality as itself, or does it just cause, say, fronting and unrounding or backing and rounding, or perhaps it can do either/or?

All of those are quite plausible. Turkish has two types of vowel harmony, you may want to look at how this affects suffixes in the language.

Could it affect a vowel before it as well as after it?

I'm not aware of any attestations thereto.

[vokzhen]

Does it (the vowel that causes mutation) only affect the previous vowel? Or can it affect a vowel that comes after it?

It can affect following vowels as well. If it affects previous vowels, it's called regressive metaphony; if it affects following vowels, you have progressive metaphony.

I do believe regressive is far more common, though. I wouldn't be surprised if i-mutation alone makes up the majority of known instances of metaphony, though I myself don't know.

Could it affect a vowel before it as well as after it?

I'm not aware of any attestations thereto.

From a synchronic point of view some vowel harmony works like this, where the presence of one set of vowels present anywhere in the word affects a change in all other vowels in the word (though afaict basing it off the root is [far] more common). Diachronically, though, I'm not aware of any instances off the top of my head.

[Nortaneous]

re: aspiration -- no, but if you want something to do with it, look up Monguor. it went #CVCʰV > CʰVCV, #VCʰV > xVCV, but i don't know the details

[KathTheDragon]

re: aspiration -- no, but if you want something to do with it, look up Monguor. it went #CVCʰV > CʰVCV, #VCʰV > xVCV, but i don't know the details

Is that in response to me?

[احمکي ارش-ھجن]

Does it (the vowel that causes mutation) only affect the previous vowel? Or can it affect a vowel that comes after it?

It can affect following vowels as well. If it affects previous vowels, it's called regressive metaphony; if it affects following vowels, you have progressive metaphony.

Does it only make the affected vowel the same quality as itself, or does it just cause, say, fronting and unrounding or backing and rounding, or perhaps it can do either/or?

All of those are quite plausible. Turkish has two types of vowel harmony, you may want to look at how this affects suffixes in the language.

Could it affect a vowel before it as well as after it?

I'm not aware of any attestations thereto.

Ah, no, I'm not looking for vowel harmony. I'm doing the mutations for the ablaut that leads to a triconsonsonantal root system.
Something like:
sudan (active present) + e (object pronoun) sudane (passive present) esudane (passive present) suden (passive present), but also nuesudane nisuden (passive past)

[StrangerCoug]

If you're not interested in vowel harmony, then regressive metaphony would probably be your better bet, since I believe vowel harmony usually operates on progressive metaphony.

[احمکي ارش-ھجن]

If you're not interested in vowel harmony, then regressive metaphony would probably be your better bet, since I believe vowel harmony usually operates on progressive metaphony.

Ah, but, I have to use progressive metaphony to get the unmarked active stem CuCaC

sodan usodan usudan sudan
menit umenit umunit munit

It has to be this way.

Also, I never asked, but could there be e-mutation and o-mutation? even schwa-mutation?

[Zaarin]

How plausible is xl, hl ɬ, noting that several languages that the language is in intimate contact with have /ɬ/.

[vokzhen]

Extremely.

[Zaarin]

Extremely.

Thanks.

[8Deer]

Would you say that the shifts ç > x and ʃ > x are about equally plausible?