Sound Change Quickie Thread

[Zaarin]

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

ʃ > x is a rather common change. Given that /ç/ is more often an allophone of /x/ or /h/ than an independent phoneme, becoming /x/ doesn't seem implausible.

[احمکي ارش-ھجن]

Could there exist such a concept as "length harmony" where: 
1)CVC + VV = CVVCVV
2)CVC + VC = CVVCVC
Given the diachronic patterning of long vowels in open syllables and short vowels in closed syllables in my conlang.

[Cedh]

Could there exist such a concept as "length harmony" where: 
1)CVC + VV = CVVCVV
2)CVC + VC = CVVCVC
Given the diachronic patterning of long vowels in open syllables and short vowels in closed syllables in my conlang.

I wouldn't call this "length harmony", but if the rule that produces long vowels in open syllables is synchronically productive, the vowels in the CVC portion of your example words would automatically be lengthened when a new vowel is suffixed and the second C gets resyllabified as an onset consonant.

[Nannalu]

It's common for them to change (I would think) but what are some things I can do with /θ/ and /ð/?

[احمکي ارش-ھجن]

What are way to develop /m̥ n̥ ɹ̥ l̥/ that doesn't involve a complex intervocalic chain shift of /m: n: ɹ: l:/ > /m n ɹ l/ > /m̥ n̥ ɹ̥ l̥/?

[sangi39]

What are way to develop /m̥ n̥ ɹ̥ l̥/ that doesn't involve a complex intervocalic chain shift of /m: n: ɹ: l:/ > /m n ɹ l/ > /m̥ n̥ ɹ̥ l̥/?

Simplication of Rh and hR clusters (as I recently mentioned over on the CBB). You could probably also get them from RP clusters where P is a voiceless plosives, which I think happened in Welsh and in Icelandic some of them came from PR clusters, like hnútur from Old Norse knútr.

Also the "intervocalic chain shift of /m: n: ɹ: l:/ > /m n ɹ l/ > /m̥ n̥ ɹ̥ l̥/" isn't that "complex". It's two independent sound changes, first /m n ɹ l/ > /m̥ n̥ ɹ̥ l̥/, followed by /m: n: ɹ: l:/ > /m n ɹ l/.

[WeepingElf]

It's common for them to change (I would think) but what are some things I can do with /θ/ and /ð/?

Affricates. I came up with the idea of shifting them to /ts/ and /dz/, respectively, in a future English.

[sangi39]

It's common for them to change (I would think) but what are some things I can do with /θ/ and /ð/?

Affricates. I came up with the idea of shifting them to /ts/ and /dz/, respectively, in a future English.

That's definitely one of the cooler ideas I've seen. Did you first shift them to dental plosives which then became affricates?

Otherwise something like /t d/, /f v/ and /s z/ are fairly common options. IIRC, /ð/ may have become /ɣ/ at some point in the history of Irish Gaelic before becoming /j/ when slender. I wonder, then if /θ/ could similarly shift to /x/. /ð/ can also become /r/ and /θ/ can become /h/.

[WeepingElf]

It's common for them to change (I would think) but what are some things I can do with /θ/ and /ð/?

Affricates. I came up with the idea of shifting them to /ts/ and /dz/, respectively, in a future English.

That's definitely one of the cooler ideas I've seen. Did you first shift them to dental plosives which then became affricates?

Thank you! To be honest, I did not think of any intermediate steps, but dental stops contrasting against the alveolar ones are a possibility. Also, this way, the two phonemes slip into a potential gap in a language that has postalveolar affricates but no alveolar ones.

Otherwise something like /t d/, /f v/ and /s z/ are fairly common options. IIRC, /ð/ may have become /ɣ/ at some point in the history of Irish Gaelic before becoming /j/ when slender. I wonder, then if /θ/ could similarly shift to /x/. /ð/ can also become /r/ and /θ/ can become /h/.

Sure. /ð/ indeed became /ɣ/ in Irish, so why not /θ/ > /x/?

[vokzhen]

What are way to develop /m̥ n̥ ɹ̥ l̥/ that doesn't involve a complex intervocalic chain shift of /m: n: ɹ: l:/ > /m n ɹ l/ > /m̥ n̥ ɹ̥ l̥/?

If anything, I think it'd be more likely to go m: > m̥ directly. Long voiced consonants can become devoiced singletons while the standard-length ones remain voiced, probably the clearest example I know of of being Inuit where e.g. /v l ɣ ʁ/ but [f: ɬ: ç: χ:] in Greenlandic. Some Berber languages have something that might be similar: Central Atlas Tamazight has /ʁ dˤ/ and /q: tˤ:/ along with other geminate fortitions like /ʒ dʒ:/ and /j ɣ:/. It's more common just with stops though, and off the top of my head I'm not aware of it occurring with nasals (though Icelandic n: > tn̥, so maybe prestop your geminates, devoice them, and then lose the actual prestopping?).

Another option would be allophonic devoicing. Before voiceless consonants sonorants devoice, then some of the triggering consonants drop out (I'm not aware of such devoicing rules occurring when there's a voice distinction elsewhere though, e.g. Nahuatl's voiceless only and Mayan's voiceless-ejective, and it's usually the liquids and glides, not nasals). Another would be final devoicing that's later made phonemic when derivational suffixes or compounding are added, without re-voicing the consonant, but again most/all examples I know of with final devoicing of sonorants involves no voicing distinction in obstruents.

According to Wikipedia, Vietnamese has a voiceless nasal release to final /p t k/ [p^m̥ t^n̥ k^ŋ̊] rather than being unreleased as usually reported (again, no contrastive voicing though). I also think I've seen a language or two in the Tibet-China-Vietnam region that has final b>m>m̥ but I'm not easily finding which one. Burmese apparently has voiceless nasals from s+nasal clusters and presumably got l̥ the same way. Khmer has "voiceless nasals" (as well as aspirated stops) that are m+h, n+h, p+h, etc that can be broken up by infixes.

It's common for them to change (I would think) but what are some things I can do with /θ/ and /ð/?

From what I can tell, the most common outcomes of /θ/ are /f s t h/, while the most common outcomes of /ð/ are /z d r l v/ null and possibly /j/ (depending on whether it's actually ð>j or ð>null>epenthetic j to break up vowels). But you've got /θ ð/ > /h ɣ/ in Irish, ð~ɮ̪ in Amis (Formosan), ð,z>w in the coda in Catalan, and various Slavey dialects (Athabascan) have [θ ð]~[f v]~[hʷ w]~[f w] along with similar changes in the affricates, including the odd tθ > kʷ. An interesting thing about /θ ð/ is that, as far as I can tell, they often don't or don't quite pattern with each other the way most other voiceless-voiced pairs do: Icelandic, despite being allophones, are laminal versus apical, English dialects have different rules for what PoA they are shifted to and in what positions, they merge differently in various Semitic languages (the voiceless with the plain fricative that becomes modern Hebrew /ʃ/, while the voiced with the voiced part of the voiceless-voiced-ejective triplet that becomes /z/) and in various Romance languages (complicated with prestige dialect influence; Venetian /θ/ merges with one of /ts s/, but /ð/ with /z dz d/, and German shifted ð > d and θ > d but centuries apart with widespread d > t in the middle.

[WeepingElf]

What do you think of /x χ χʷ/ > /h ħ ʕ/?

[Pole, the]

It's common for them to change (I would think) but what are some things I can do with /θ/ and /ð/?

Affricates. I came up with the idea of shifting them to /ts/ and /dz/, respectively, in a future English.

I thought about /θ ð t d/ → /t d ɾʰ ɾ/.

[Zaarin]

What do you think of /x χ χʷ/ > /h ħ ʕ/?

The first two at least are well attested in both directions.

[Pole, the]

What do you think of /x χ χʷ/ > /h ħ ʕ/?

I can smell PIE here.

[Daedolon]

How about this?

P = any plosive

P1P2 > P2P2 > P2ː

Ex: atpa > appa > apːa

[احمکي ارش-ھجن]

How about this?

P = any plosive

P1P2 > P2P2 > P2ː

Ex: atpa > appa > apːa

Yes... that is a normal process of assimilation, and appa is apːa

[GreenBowTie]

What are way to develop /m̥ n̥ ɹ̥ l̥/ that doesn't involve a complex intervocalic chain shift of /m: n: ɹ: l:/ > /m n ɹ l/ > /m̥ n̥ ɹ̥ l̥/?

english has some/all of them as allophones of /əm/, /ən/, etc., so that's another way you could get there

[vokzhen]

What are way to develop /m̥ n̥ ɹ̥ l̥/ that doesn't involve a complex intervocalic chain shift of /m: n: ɹ: l:/ > /m n ɹ l/ > /m̥ n̥ ɹ̥ l̥/?

english has some/all of them as allophones of /əm/, /ən/, etc., so that's another way you could get there

Voiceless n̥, not syllabic n̩. Not that I blame you, at least in this context it's obviously not supposed to be laminal n̻, and at this font size advanced n̟ isn't very distinct either.

[WeepingElf]

What do you think of /x χ χʷ/ > /h ħ ʕ/?

I can smell PIE here.

Yes

[احمکي ارش-ھجن]

These sound changes all check out?:

Himoshian Taksheyut Pre-Vrkhazhian Classical Vrkhazhian
Nasal
/tm/ > /tm̥/ > /m̥/ > /m̥/
/km/ > /km̥/ > /m̥/ > /m̥/
/qm/ > /qm̥/ > /m̥/ > /m̥/
/pn/ > /pn̥/ > /n̥/ > /n̥/
/kn/ > /kn̥/ > /n̥/ > /n̥/
/ŋ/ > /ŋ/ > /ɴ/ > /ɴ/
Stops
/ki/[kʲi] > /c/ > /c͡ç/ > /c͡ç/
/gi/[gʲi] > /ɟ/ > /ɟ͡ʝ/ > /ɟ͡ʝ/
/kp/ > /çp/ > /çp/ > /çp/
/kt/ > /çt/ > /çt/ > /çt/ <= attested in Latin descendants
/gb/ > /ʝb/ > /ʝb/ > /ʝb/
/gd/ > /ʝd/ > /ʝd/ > /ʝd/
Fricatives
/xi/[xʲi] > /ç/ > /ç/ > /ç/
/ɣi/[ɣʲi] > /ʝ/ > /ʝ/ > /ʝ/
Liquids
/w:/ > /w:/ > /β/ > /β/
/j:/ > /j:/ > /ʝ/ > /ʝ/
/pɹ/ > /pɹ̥/ > /ɹ̥/ > /ɹ̥/
/tɹ/ > /tɹ̥/ > /ɹ̥/ > /ɹ̥/
/kɹ/ > /kɹ̥/ > /ɹ̥/ > /ɹ̥/
/qɹ/ > /qɹ̥/ > /ɹ̥/ > /ɹ̥/
Laterals
/pl/ > /pl̥/ > /l̥/ > /l̥/
/tl/ > /tl̥/ > /l̥/ > /l̥/
/kl/ > /kl̥/ > /l̥/ > /l̥/
/ql/ > /ql̥/ > /l̥/ > /l̥/
/ɬ/ > /ɬ/ > /ʎ̝̊/ > /ʎ̝̊/ < this is a voiceless palatal lateral fricative (lambda + voiceless diacritic + raising diacritic)
Laryngeals
/ʔp/ > /p/ > /p/ > /p/
/ʔt/ > /t/ > /t/ > /t/
/ʔk/ > /k/ > /k/ > /k/
/ʔq/ > /q/ > /q/ > /q/
/ʡp/ > /Lp/ > /p/ > /p/ <= L = lowered vowel
/ʡt/ > /Lt/ > /t/ > /t/
/ʡk/ > /Lk/ > /k/ > /k/
/ʡq/ > /Lq/ > /q/ > /q/
/hp/ > /V:ʰp/ > /ʰp/ > /ʰp/
/ht/ > /V:ʰt/ > /ʰt/ > /ʰt/
/hk/ > /V:ʰk/ > /ʰk/ > /ʰk/
/hq/ > /V:ʰq/ > /ʰq/ > /ʰq/
/ħp/ > /L:p/ > /p/ > /p/
/ħt/ > /L:t/ > /t/ > /t/
/ħk/ > /L:k/ > /k/ > /k/
/ħq/ > /L:q/ > /q/ > /q/
Additionally, from the ancestor language of Himoshian, known as Proto-Hasjakam, a series of "emphatic" consonants arose from low-tone accented vowels, first becoming creaky voiced in Himoshian, and then into their present state in Classical Vrkhazhian.
Emphatics, All Positions
/p̰/ > /pʶ/ > /pʶ/ > /pʶ/
/t̰/ > /tʶ/ > /tʶ/ > /tʶ/
/k̰/ > /kʶ/ > /kʶ/ > /q/
/r̰/ > /rʶ/ > /rʶ/ > /ʀ/

I still have to find a way to get /t͡s ɸ X/ and more ways to get /β/ and under speculation is:
/ʡ/ > /Lʔ/ > /ʔ/ > /ʔ/
/h:/ > /V:h/ > /h/ > /h/
/ħ:/ > /L:X/ > /X/ > /X/ <= uvular fricative

[Nymrīs]

What do you think of /x χ χʷ/ > /h ħ ʕ/?

The first two are fairly common and im assuming they change in different positions as for the pharyngal fricative i could see x^w>ɣ^w>ʕ^w>ʕ through x^w may have to change to xw so it becomes voiced under the influence of w, so its plausible that this shift happens in the environment of the others.

[Nymrīs]

ʔp/ > /p/ > /p/ > /p/
/ʔt/ > /t/ > /t/ > /t/
/ʔk/ > /k/ > /k/ > /k/
/ʔq/ > /q/ > /q/ > /q/
/ʡp/ > /Lp/ > /p/ > /p/ <= L = lowered vowel
/ʡt/ > /Lt/ > /t/ > /t/
/ʡk/ > /Lk/ > /k/ > /k/
/ʡq/ > /Lq/ > /q/ > /q/

If it would lower the vowel in the case of epiglottals it most definetly will in the case of pharyngeals i suggest changing the shift to.
/ʔt/ > /ʁt or /ɣt > /t/ > /t/ > /t/
/ʡt/ > /ʔt > /Lt/ > /t/ > /t/

[احمکي ارش-ھجن]

ʔp/ > /p/ > /p/ > /p/
/ʔt/ > /t/ > /t/ > /t/
/ʔk/ > /k/ > /k/ > /k/
/ʔq/ > /q/ > /q/ > /q/
/ʡp/ > /Lp/ > /p/ > /p/ <= L = lowered vowel
/ʡt/ > /Lt/ > /t/ > /t/
/ʡk/ > /Lk/ > /k/ > /k/
/ʡq/ > /Lq/ > /q/ > /q/

If it would lower the vowel in the case of epiglottals it most definetly will in the case of pharyngeals i suggest changing the shift to:
/ʔt/ > /ʁt or /ɣt > /t/ > /t/ > /t/
/ʡt/ > /ʔt > /Lt/ > /t/ > /t/

Ah, not quite necessary.
The language has a four way contrast of laryngeals /ʡ ʔ ħ h/.
1)The pharyngeal/epiglottal (they are considered the same) consonants lower the vowel while the glottal consonants do not.
2)The glottal fricative (pre)aspirates the stops.
3)The stops keep the vowels short while the fricatives lengthen the vowel.
Thus:
VhC > V:ʰC
VħC > L:C
VʔC > VC
VʡC > LC
The laryngeals simply drop syllable-final.

[vokzhen]

These sound changes all check out?:

Himoshian Taksheyut Pre-Vrkhazhian Classical Vrkhazhian
Nasal
/tm/ > /tm̥/ > /m̥/ > /m̥/
/km/ > /km̥/ > /m̥/ > /m̥/
/qm/ > /qm̥/ > /m̥/ > /m̥/
/pn/ > /pn̥/ > /n̥/ > /n̥/
/kn/ > /kn̥/ > /n̥/ > /n̥/
/ŋ/ > /ŋ/ > /ɴ/ > /ɴ/

Possibly okay, but I'm not sure they're attested. I'd feel better if stops lenited before nasals, and then fricative+nasal clusters devoiced. Also I'm not aware of /ɴ/ being a phoneme in any natlang, the closest being Japanese (uvular pretty much only finally in careful pronunciation), a few Northwest Coast langauges that have a front-velar and back-velar, and Inuit dialects that have [Nɴ] as an allophone of /Nʁ/.

/w:/ > /w:/ > /β/ > /β/
/j:/ > /j:/ > /ʝ/ > /ʝ/
/pɹ/ > /pɹ̥/ > /ɹ̥/ > /ɹ̥/
/tɹ/ > /tɹ̥/ > /ɹ̥/ > /ɹ̥/
/kɹ/ > /kɹ̥/ > /ɹ̥/ > /ɹ̥/
/qɹ/ > /qɹ̥/ > /ɹ̥/ > /ɹ̥/
Laterals
/pl/ > /pl̥/ > /l̥/ > /l̥/
/tl/ > /tl̥/ > /l̥/ > /l̥/
/kl/ > /kl̥/ > /l̥/ > /l̥/
/ql/ > /ql̥/ > /l̥/ > /l̥/
/ɬ/ > /ɬ/ > /ʎ̝̊/ > /ʎ̝̊/ < this is a voiceless palatal lateral fricative (lambda + voiceless diacritic + raising diacritic)

Almost all stop-rhotic clusters I know of change into retroflex affricates if they all change into something (very common in the Himalayan-Southeast Asia region). As with nasals I'd be more comfortable passing it if stop-liquids become fricative-liquids, or possibly all merged into /tɹ̥ tɬ/ and then deaffricated

ʔp/ > /p/ > /p/ > /p/
/ʔt/ > /t/ > /t/ > /t/
/ʔk/ > /k/ > /k/ > /k/
/ʔq/ > /q/ > /q/ > /q/
/ʡp/ > /Lp/ > /p/ > /p/ <= L = lowered vowel
/ʡt/ > /Lt/ > /t/ > /t/
/ʡk/ > /Lk/ > /k/ > /k/
/ʡq/ > /Lq/ > /q/ > /q/

If it would lower the vowel in the case of epiglottals it most definetly will in the case of pharyngeals i suggest changing the shift to.
/ʔt/ > /ʁt or /ɣt > /t/ > /t/ > /t/
/ʡt/ > /ʔt > /Lt/ > /t/ > /t/

Well, first of all that's not pharyngealization versus epiglottalization. The pharyngeals he lists *do* lower the vowels. But also I'm not sure it's true both will lower: I remember seeing diagrams of tongue position in some Khoisan language of /i/ versus /iˁ/ versus /i^ʢ/ and there were some pretty noticeable differences between the latter two (iirc the pharyngealization was all in the back, with the front constriction actually being almost identical to /i/, whereas epiglottalization pulled the entire tongue back). Every time it comes up it kills them that I can't find that paper again. Though I still think it's *likely* both will lower.

[احمکي ارش-ھجن]

These sound changes all check out?:

Himoshian Taksheyut Pre-Vrkhazhian Classical Vrkhazhian
Nasal
/tm/ > /tm̥/ > /m̥/ > /m̥/
/km/ > /km̥/ > /m̥/ > /m̥/
/qm/ > /qm̥/ > /m̥/ > /m̥/
/pn/ > /pn̥/ > /n̥/ > /n̥/
/kn/ > /kn̥/ > /n̥/ > /n̥/
/ŋ/ > /ŋ/ > /ɴ/ > /ɴ/

Possibly okay, but I'm not sure they're attested. I'd feel better if stops lenited before nasals, and then fricative+nasal clusters devoiced. Also I'm not aware of /ɴ/ being a phoneme in any natlang, the closest being Japanese (uvular pretty much only finally in careful pronunciation), a few Northwest Coast langauges that have a front-velar and back-velar, and Inuit dialects that have [Nɴ] as an allophone of /Nʁ/.

Right... or maybe just start with voiceless fricatives to begin with and keep the clusters in the descendant language. Well... a phonemic /ɴ/ may be unattested, but nothing says it couldn't and I think it would be articulatorily easier to produce than a voiced uvular stop, cause all the air goes through the nose instead of the small closed space before the uvula.
EDIT:Oh look, ANADEW: https://en.wikipedia.org/wiki/Klallam_language

How's this sound?
/xm/ > /xm̥/ > /m̥/ > /m̥/
/xn/ > /xn̥/ > /n̥/ > /n̥/
/sɹ/ > /sɹ̥/ > /ɹ̥/ > /ɹ̥
/xɹ/ > /xɹ̥/ > /ɹ̥/ > /ɹ̥/
/sl/ > /sl̥/ > /l̥/ > /l̥/
/xl/ > /xl̥/ > /l̥/ > /l̥/

ʔp/ > /p/ > /p/ > /p/
/ʔt/ > /t/ > /t/ > /t/
/ʔk/ > /k/ > /k/ > /k/
/ʔq/ > /q/ > /q/ > /q/
/ʡp/ > /Lp/ > /p/ > /p/ <= L = lowered vowel
/ʡt/ > /Lt/ > /t/ > /t/
/ʡk/ > /Lk/ > /k/ > /k/
/ʡq/ > /Lq/ > /q/ > /q/

If it would lower the vowel in the case of epiglottals it most definetly will in the case of pharyngeals i suggest changing the shift to.
/ʔt/ > /ʁt or /ɣt > /t/ > /t/ > /t/
/ʡt/ > /ʔt > /Lt/ > /t/ > /t/

Well, first of all that's not pharyngealization versus epiglottalization. The pharyngeals he lists *do* lower the vowels. But also I'm not sure it's true both will lower: I remember seeing diagrams of tongue position in some Khoisan language of /i/ versus /iˁ/ versus /i^ʢ/ and there were some pretty noticeable differences between the latter two (iirc the pharyngealization was all in the back, with the front constriction actually being almost identical to /i/, whereas epiglottalization pulled the entire tongue back). Every time it comes up it kills them that I can't find that paper again. Though I still think it's *likely* both will lower.

The consonants /ʡ ħ/ are pharyngeal, and wikipedia lists them as such. No epiglottalization here.