Sound Change Quickie Thread

[Dē Graut Bʉr]

Is there any way to conditionally lose breathy voice on vowels?

[Sexendèƚo]

What are some interesting things to do with geminate consonants?
Currently I have /bː,dː,gː,pː,tː,kː,zː,sː,vː,fː,mː,nː,lː,jː,ɲː,ʎː/
I'd like to do more than just degeminate them, especially if they could develop into something less Romance-y.

[hwhatting]

What are some interesting things to do with geminate consonants?
Currently I have /bː,dː,gː,pː,tː,kː,zː,sː,vː,fː,mː,nː,lː,jː,ɲː,ʎː/
I'd like to do more than just degeminate them, especially if they could develop into something less Romance-y.

Affrication. That happened to geminate unvoiced stops in the development from West Germanic to Old High German, i.e. /p:/ > /p_f/, /t:/ > /t_s/; but /k:/ > /x/. That was part of a bigger shift, though, where also single unvoiced stops turned into fricatives or affricates, depending on the environment.

[Zaarin]

What are some interesting things to do with geminate consonants?
Currently I have /bː,dː,gː,pː,tː,kː,zː,sː,vː,fː,mː,nː,lː,jː,ɲː,ʎː/
I'd like to do more than just degeminate them, especially if they could develop into something less Romance-y.

Geminate unvoiced stops could become ejectives. If I recall correctly geminate stops can also become implosives.

[sangi39]

What are some interesting things to do with geminate consonants?
Currently I have /bː,dː,gː,pː,tː,kː,zː,sː,vː,fː,mː,nː,lː,jː,ɲː,ʎː/
I'd like to do more than just degeminate them, especially if they could develop into something less Romance-y.

Geminate unvoiced stops could become ejectives. If I recall correctly geminate stops can also become implosives.

On top of that, /l:/ and /n:/ shifted to [tl] and [tn] in Icelandic (through [dl dn]?), which I assume you could extend to /v: m: jːɲː ʎː/ (> [w: m: j: ɲː ʎː] > [gw bm ɟj ɟɲ ɟʎ]).

You could go in the opposite direction with the nasals, i.e. /m: n: ɲ:/ > [mb nd ɲɟ]. I'm not sure if that's attested anywhere, but it doesn't seem unreasonable.

/l:/ became /ɖ(:)/ in Sardinian and Sicilian, IIRC.

The geminate voiced plosives could shorten, with originally short voiced plosives shifting to voiced fricatives.

Did Welsh /ɬ/ come from an older /l:/ or is that just an orthographic convention?

[Birdlang]

What are some interesting things to do with geminate consonants?
Currently I have /bː,dː,gː,pː,tː,kː,zː,sː,vː,fː,mː,nː,lː,jː,ɲː,ʎː/
I'd like to do more than just degeminate them, especially if they could develop into something less Romance-y.

I made a protolang where it developed from another protolang with gemination, and that developed into pharyngealization.

[vokzhen]

Did Welsh /ɬ/ come from an older /l:/ or is that just an orthographic convention?

Orthographic conversion, but both Inuit and Berber languages have devoicing of geminates: Greenlandic /v l ɣ ʁ/ but geminate [ff ɬɬ çç χχ], Tamazight and Tuareg both have [qq ttˤ] as the most common geminate of /ɣ dˤ/, though Inuit languages mostly don't have voiceless fricatives to begin with, and Berber languages have other singleton-geminate alternations like /j ~ ɣɣ/ /ʃ ~ ttʃ/ /w ~ bb/.

It's pretty common, afaik, to devoice geminate voiced stops just because it's harder to maintain the voicing with the longer hold, and if gemination is later lost you may have created a plain series from an old aspirated-voiced system. I believe this is how geminate > implosive is posited to work too, implosion allows voicing to be held for that duration.

The Northeast Caucasian languages often have distinct "fortis" consonants (geminated, higher tension, voiceless and generally unaspirated, where singleton stops may be lightly aspirated) that originate, at least in part, from geminates.

Preaspiration I think is also a possibility. NEC fortis consonants are sometimes preaspirated, and long consonants became preaspirated in one or some of the Uralic languages I believe.

Instead of going tt > ʔt > t' for the voiceless set (or the voiced too, if they're devoiced), you could have the glottal stop shift to vocalic. Creaky-voiced or high-pitched vowels, or simply introducing a length contrast (ata > aata, at:a > aʔta > ata), and/or causing vowel breaking (a ʔ-coda or creakiness correlates with a fronter offglide). Glottalization can shift to pharyngealization, as Birdlang said. These, especially the last two, are probably operating on a longer time scale than the other suggestions, though.

For the sonorants and fricatives, fortition of /ll/ and /nn/ has been mentioned (to /tɬ tn/ in Icelandic, /ɖ:/ in Asturian, Sicilian, and Sardinian), there are languages that also have /s tt/ alternate, but the ordering of things might be important. That is, it may have been /ts tts/ and one half lenited and one half fortified, rather than /s:/ becoming /tt/ directly. Given that the sonorants can do it, though, I wouldn't think it's too far a stretch, which would allow geminate /s:/ to act like a voiceless stop for further changes. It's also very common for /jj/ for fortify into a different geminate [gg ggʲ dd ddʒ], which then allows it to undergo the same processes that the voiced stops do. And I believe another possibility is just to leave the sonorants alone, if I'm not mistaken it's not uncommon to have /mm nn ll rr/ as the only consonants that allow doubling.

[sangi39]

I completely forgot about pre-aspiration in Icelandic. How'd I manage that

[Qwynegold]

Is there any way to conditionally lose breathy voice on vowels?

Maybe before or after or between voiceless consonants? Idk.

[Qwynegold]

What are some interesting things to do with geminate consonants?
Currently I have /bː,dː,gː,pː,tː,kː,zː,sː,vː,fː,mː,nː,lː,jː,ɲː,ʎː/
I'd like to do more than just degeminate them, especially if they could develop into something less Romance-y.

Idk if you think this is boring, but they could affect preceding vowels before degemination. Either they could dump their length on the preceding vowel, or they could first shorten a preceding vowel and then degeminate. Ooh, you could do some kind of Slavic thingy if you get extra short vowels.

[Nortaneous]

V>C length spreading happens in Seri (apparently even V>C>V, hóopatj used to be written hóoppaatj) so C>V could happen. Is it attested?

Fortition of geminate approximants is common.

[Nortaneous]

If I have the pharyngealized consonants /pˤ kˤ bˤ gˤ fˤ sˤ (vˤ zˤ ʃˤ)/ and a contrast between /t̪ d̪ t d/ and /tʂ dʐ tʃ dʒ ʂ ʃ/, could the dental and retroflex consonants develop pharyngealization to heighten the contrast, instead of merging into the alveolars and postalveolars?

That is, instead of t̪ d̪ tʂ dʐ ʂ > t d tʃ dʒ ʃ, could I have t̪ d̪ tʂ dʐ ʂ > tˤ dˤ tʃˤ dʒˤ ʃˤ?

[8Deer]

I think something similar happened in Chilcotin, so you could look into that.

[StrangerCoug]

Given /a ɛ i o p t k ʔ m n s h r/ and the fact that all syllables are CV, is there any plausible way to split off /u/ as a fifth vowel phonemically distinct from /o/? (This is a pitch-accent language—the "stress" would be on the high-pitched syllable.)

[Zaarin]

Given /a ɛ i o p t k ʔ m n s h r/ and the fact that all syllables are CV, is there any plausible way to split off /u/ as a fifth vowel phonemically distinct from /o/? (This is a pitch-accent language—the "stress" would be on the high-pitched syllable.)

Many Native American languages have allophonic variation between /o/ and /u/ (or similar consonants). Offhand, in the environment of /p m/ might be a good place to start. Once you have the allophonic variation, other changes could make the phonetic distinction phonemic.

[gach]

You could try height harmony where the presence of /a/ prefers [o] over for /o/ or the presence of /i/ over [o] for it. Then contract some of the syllables containing /a/ or /i/ away for example through Vh/ri > Vi > V or Vh/ra > Va > V and the [o ~ u] allophony turns phonetic.

[Porphyrogenitos]

Given /a ɛ i o p t k ʔ m n s h r/ and the fact that all syllables are CV, is there any plausible way to split off /u/ as a fifth vowel phonemically distinct from /o/? (This is a pitch-accent language—the "stress" would be on the high-pitched syllable.)

You could have /o/ shift allophonically to in all cases except before /h/ and /ʔ/, and then cause something else to merge (conditionally or unconditionally) with /h/ and/or /ʔ/ in order to create contrast. Unconditional /r/ > /ʔ/ happened in some Polynesian languages, or perhaps you could fortite /h/ to /k/ in high syllables (assuming high syllables have any of the same acoustic properties as stressed syllables in English and other languages). Or you could debuccalize /p/ or /s/ to /h/, or back /k/ to /ʔ/. And so on.

[Nortaneous]

Souletin Basque developed ü from u in some specific conditions, and then turned high vowels near ü to ü. You could have something like:

o > u / C[-voice]_[-accent]C[-voice] (o > u when unstressed and surrounded by voiceless consonants)

and then [+high] spreads to adjacent mid vowels, blocked only by /a/, and then /r/ lowers following unstressed /i/, so:

*ritoséro > rituséro > ritusíro > retusíro
*kasotári > kasutári > kasutáre
*kótoke > kótuke > kútuki
*satokesáti > satukesáti > satukisáti

If you want to mix things up more, you could have s > š / _, t > š / _i, s > h, š > s before the spreading, so *kasotári > kasutári > kašutári > kasutari, *satokesáti > satukesáti > hatukeháši > hatukihási.

Alternatively, o > u / C[+voice]_, r > h, t > s / _i, V1hV2 > V2:, V: > V so
*ritoséro > ritoséru > hitoséru
*kasotári > kasotáhi
*kótoke > kótoke
*satokesáti > satokesási
*marónotiro > marúnutiru > mahúnusihu > múnusu
*tohíno > tohínu > tíínu > tinu
*nohotí > nuhoti > noosi > nosi
You could use this to develop tonal pitch accent: (A = any vowel) ÁhA AhÁ Á > Â Ǎ Á

[Bristel]

kʷʰ(aspirated)/V_V → xʷ?
gʷʱ(breathy)/_V → C[+voice -breathy]V[+breathy]?

What can voiceless consonants, especially approximants, turn into? If they are also aspirated, can this turn them into fricatives?

What can /c͡ç/ turn into?

[Shemtov]

Can Palatalazation become Velarazation?

[Daedolon]

Coud this work?

C = Any consonant

Cˤ > Cʕ

[Nortaneous]

Bristel: Plausible, plausible if generalized to all breathy-voiced consonants although they could also become voiceless, they can voice or turn into fricatives, it can front to /ts/ or /t/ or lenite to /C/ or /S/.

Shemtov: No.

Daedolon: Probably. Chechen vs. Ingush -- one of them pushed pharyngealization from consonants to vowels or the other way around, I can't remember which.

[StrangerCoug]

What can I do to /tʷ/?

[Bristel]

What can I do to /tʷ/?

Potentially it can become /p/.

[StrangerCoug]

I have /kʷ/ /p/, so I think that's consistent with what I'm doing.