Sound Change Quickie Thread

[Xwtek]

I need to make vowel split based on historical final consonant on syllable. Any suggestion?

[Vijay]

Look at nasalization?

[Vilike]

Make velar segments to /j/, (bi-labial segments to /w/: instant diphthongs.

You could also do /C$/ > /ʔ/ > creaky voice on the vowel; and then break it, as in Khmer.

[Xwtek]

Look at nasalization?

Other than that? There is already nasalization in proto-Asent''oan

[Xwtek]

Make velar segments to /j/, (bi-labial segments to /w/: instant diphthongs.

You could also do /C$/ > /ʔ/ > creaky voice on the vowel; and then break it, as in Khmer.

Your first ideas look like what my original sound change did, but I ditch it because it wasn't realistic.

Final consonant in syllable causes both consonant mutation on the next syllable and the vowel mutation on the same syllable as the deleted consonant. The point is to make the free-form and the compounded form more different. For example: *-so~:ka (to go) and *-poka (to eat) (-a is the indicative suffix), is compounded into *asamp'oka (for example, but depending on the sound change it may turn into something different). The idea is to make a proto-Asent'oan language that have -CVC- root pattern, but the children have -CVXVX- root pattern (where X is cluster).

Proto Asent'oan has only three manner of articulation (nasal, plosive, fricative) at these place of articulation (labial, dental, postalveolar, lateral alveolar, palatal, velar, labiovelar, and glottal). (not all combination are possible). However, the child language has split the consonants, For example, there is voiceless, voiced and ejective stop, fricative, approximant, and nasal in Asent'o)

[Vilike]

Make velar segments to /j/, (bi-labial segments to /w/: instant diphthongs.

You could also do /C$/ > /ʔ/ > creaky voice on the vowel; and then break it, as in Khmer.

Your first ideas look like what my original sound change did, but I ditch it because it wasn't realistic.

Velar to /j/ happened at least to Latin on the way to French, so it's "vanilla" enough to me to be realistic.
Ex: DIRECTA > dreite > droite [dʁwat] 'right'; FACTVM > fait [fɛ] 'fact'

[Ahzoh]

I need ideas for sound changes that can condense the following numerals into tri- or quadraliteral roots:
/ʔasθeːvaː xaːŋuɮaː/
/ʔasθeːvaː ħwe.ħaː/
/ʔasθeːvaː ɮaː.meː.naː/
/ʔasθeːvaː saː.ɬu.raː/
/ʔasθeːvaː tiːb.saː/
Given:
*/θ/ > /s/ > /h/ (chain shift)
*all long vowels become short and /eː oː/ become /i u/
*/ħ/ becomes /X/ which becomes /x/
*/ah/ tends to become an /aʕ/ diphthong
*stress is placed on the heaviest syllables before the last vowel and immediately pretonic syllables are deleted.

[mèþru]

Velar to /j/ happened through [x] or [χ] in French:
/k/ > [x] or [χ] > [ç] or [ʝ]? > /j/

[Zju]

Just how improbable is V[+creaky] → ɹV, maybe when conditional? I'm trying to get from strictly CV to CRV without vowel drops, onomatopoeias or borrowings.

[Tropylium]

I need ideas for sound changes that can condense the following numerals into tri- or quadraliteral roots

If you want them to be different from each other, trying to leverage the stress-based reduction would probably go far. (Some umlauts are also always a good option.)

After applying your stated changes aside from the stress rule, as individual words, you'd get:
/ʔaʕsiva_xaŋuɮa/
/ʔaʕsiva_xwexa/
/ʔaʕsiva_ɮamena/
/ʔaʕsiva_haɬura/
/ʔaʕsiva_tibha/

You could start with minor sandi: a_ha also contracts to *ah > /aʕ/, apocope of /a/ within compounds before single onsets otherwise, and the stress rule is then applied within each compound. This gives (periods for elided vowels):
/.ʕsív.xaŋuɮa/
/ʔaʕs.váxwexa/
/.ʕsív.ɮamena/
/ʔaʕs.váʕɬura/
/ʔaʕs.v.tábha/

Now some i-umlaut, cluster reductions/assimilations, and loss of post-posttonic vowels in open syllables:
/sˤífxandɮa/
/ʔaʕfáxʷex/
/sˤívɮanna/
/ʔaʕfáʕɬur/
/ʔaʕftábba/

Still a lot of the /aʕ/ dipthong around. Let's apply an unstressed chainshift where a aʕ > ə a (and also e > ə, u > ʷə); further cluster simplification (including fx > xxʷ, vɮ > ɮɮʷ > ddɮʷ); and schwas then recolored as ˤə ˤʷə ʷə > a o u;
/sˤíxxudɮə/
/ʔafáxux/
/sˤíddɮunə/
/ʔafáɬɬor/
/ʔattábə/

These are obviously still not going to become "roots" purely through sound changes (roots are by definition morphological elements, not words), but they probably start looking close enough to however your usual root derivatives work for analogy to kick in.

[Tropylium]

I'm trying to get from strictly CV to CRV without vowel drops, onomatopoeias or borrowings.

Doesn't seem like the most probable thing to happen (clusters almost always come from vowel drops). Dipthongizations could readily give you CjV CwV though, and these could then dissimilate etc. to yield some further variety. E.g. for labials + j, Slavic has a precedent for turning the glide into lʲ, while Greek turns it into a voicing-assimilated dental stop (pj bj > pt bd; don't recall if there are examples with pʰ), which could be subsequently flapped.

[Ahzoh]

I need ideas for sound changes that can condense the following numerals into tri- or quadraliteral roots

If you want them to be different from each other, trying to leverage the stress-based reduction would probably go far. (Some umlauts are also always a good option.)

I wanted each numeral be be modified to such a degree that you couldn't analyze them as components of the numeral six and the other numeral component. I didn't think umlaut could happen since there aren't any affixes or modifiers to trigger them.

These are obviously still not going to become "roots" purely through sound changes (roots are by definition morphological elements, not words), but they probably start looking close enough to however your usual root derivatives work for analogy to kick in.

Thanks, though, I had made a mistake, the first two words were supposed to be /tiːb.θaː/ and /θaː.ɬu.raː/, but actually I don't like some of numbers I came up with.
Also, /ɦ/ is actually supposed to be in free variation with /ʕ/.

I should provide more information as to what I intend (the last vowel -ā is the feminine gender marker and is not part of the root):

/tíːb.θaː/ > /tib.sa/
/ðáː.su.raː/ > /z.ɦu.ra/ > /zuɦ.ra/ ~ /zuʕ.ra/ ~ /zaʕ.ra/
/ɮáː.meː.naː/ > /ɮ.mi.na/ > /ɮim.na/ (by analogy)
/ħwè.ħaː/ > /xwe.xa/ > /xaw.xa/ [xau̯.xa] (also by analogy)
/xé.or.vaː/ > /çe.or.va/ > /çər.va/
/ʔásθeːvaː/ > /ʔaɦ.si.va/ ~ /ʔaʕ.si.va/

So given the original words and the intended outcome above, what sound changes could lead to the below being condensed into 4-literal or 3-literal roots (and I say this mostly as opposed to 5-literal+ "short" words that could possibly result):

/ʔásθeːvaː tíːb.θaː/ > ???
/ʔásθeːvaː ðáː.su.raː/ > ???
/ʔásθeːvaː ɮáː.meː.naː/ > ???
/ʔásθeːvaː ħwè.ħaː/ > ???
/ʔásθeːvaː xé.or.vaː/ > ???

[Xwtek]

Another tips, you can shift the meaning of the word of "taken" to nine. The shift is "one taken (from ten)" > "taken". Also "one added (from five)" > "added" for 6.

[Qwynegold]

But if original /u/ fronts, it seems appropriate for original [u̯] to front

It doesn't have to. You can just state that u fronts when not V_.

[bbbosborne]

is /anː/ -> /ãn/ plausible? and can it be generalized to any vowel/nasal?

[Xwtek]

is /anː/ -> /ãn/ plausible? and can it be generalized to any vowel/nasal?

By the way, I have similiar but at the exact opposite direction.

[Das Public Viewing]

is /anː/ -> /ãn/ plausible? and can it be generalized to any vowel/nasal?

If you specifically want to keep it distinct from /Vn/, perhaps this?
V>>Ṽ/_N
VCː>>VːC
Ṽ>>V/_N (short vowels only)
Ṽː>>Ṽ

[Raholeun]

In my language Sataw there are plenty of Vʔ.CV-series. Normally, I have no problem maintaining the syllable split. Only when the second consonant is /w/, the glottal stop turns into [gʷ]. Are there other alternations that could think of in this case?

[malloc]

In my language Sataw there are plenty of Vʔ.CV-series. Normally, I have no problem maintaining the syllable split. Only when the second consonant is /w/, the glottal stop turns into [gʷ]. Are there other alternations that could think of in this case?

Given that change, I would expect the glottal stop to become [dZ] or [J\] before /j/ at least, and perhaps also /t/ before /r/ or something.

[Xwtek]

Is grammatical change included here?

Is it possible that an S-Aux-O-V isolating language turns into S O V polysynthetic language? The PAR language is fusional-isolating language that have V2 word order, except that object pronoun (but not regular noun) is put before auxiliary verb. I it is possible that massive fronting occurs in that language so the remaining word fuses into a single verb?