Sound Change Quickie Thread

[conlangernoob]

Can /ʃ/ be allophonic with /x/ depending on the surrounding vowels? Would this allophony be relatively stable?

[Travis B.]

Can /ʃ/ be allophonic with /x/ depending on the surrounding vowels? Would this allophony be relatively stable?

You mean [ʃ] and [x] - slashes indicate phonemes rather than phones. And why not? IIRC this exact allophony is found in some West Central German varieties.

[conlangernoob]

OK, thanks. Sorry, I knew the slashes thing, I just forgot.

[Darren]

Enggano goes even further with /x/ [s~ç~x].

[Ahzoh]

Can /ʃ/ be allophonic with /x/ depending on the surrounding vowels? Would this allophony be relatively stable?

Depends on the environment, I'd deffo expect /ex ix/ [eʃ iʃ] while /ox ux/ [ox ux], or similar but with consonants before said vowels.

[Moose-tache]

An early form of pre-Slavic probably had a x/sh alternation based on the following vowel, but that's not really how the phonemes work in any attested Balto-Slavic language, afaik.

[Zju]

Cross-posting from the Linguistic Miscellany Thread:

Per A grammar and dictionary of Gayogo̱hó:nǫˀ (Cayuga):

SR – as in węhnihSRí:yo: ‘nice day’ – sounds like the SHR [ʃɹ] in shrink. Some speakers pronounce SR as FR [fɹ] instead, for example in words like ganǫ́hkwasraˀ (ganǫ́hkwaFRaˀ) ‘love’. SR syllabifies as two separate consonants, [ʃ.ɹ] or [f.ɹ].

A fronting of [ʃ] to [f]. I never would’ve figured on that as a direct step. ʃ to x and thence to f, sure, but this may be useful to some.

Is it attested that it was fronting from ʃ to f? A split from earlier *θ/s just seems so much more likely: (θ →) s → ʃ ; (s →) θ → f. Elsewise, it'd have to be one of those rare phonetically unmotivated sound changes, such as *b → nc / V_V.

[Ahzoh]

So I have it that the ancestor of Vrkhazhian doesn't have voiced fricatives, it only has voiceless fricatives. Next I have this set of sound changes:
f s ɬ ʃ x ⭢ v z ɮ ʒ ɣ / V_V
f s ɬ ʃ x ⭢ v z ɮ ʒ ɣ / V_{m n ŋ r l}
{ʡ ʔ}C ⭢ Cː / _
V{ħ h} ⭢ Vː / _C
VC{ʡ ʔ ħ h} ⭢ VːC / _

However given this it means voiced fricatives can't occur word-initially, word-finally, or post consonantally. It also means that voiceless fricatives cannot occur word-medially following a short vowel. Lastly, it means that geminate voiceless fricatives can exist but not geminate voiced fricatives.

So, I need some more ideas to allow voiceless fricatives to exist after short vowels, and to also allow voiced geminate fricatives. I'm not sure if I want voice fricatives in initial, post-consonantal, or final position.

The language is triconsonantal, so analogy may sometimes play a part, but generally allophony doesn't stick. For example, you have a verb rabad- "to guard" which becomes rabad-ni "guard-1sg"and rabad-ma "guard-2sg" but rabat-ta "guard-3sg". It doesn't mean the /d/ in the root will analogize to /t/ for the rest o the paradigm. However, you can have a word like wast-am "mana" which has the construct state form of wasat even though it should be *wazat. One might think the allophony doesn't analogize becuase there are more inflections in the paradigms where the allophony doesn't apply, but I also have a verb wasax- "be loyal" which probably should be *wazax- in most inflections but it's simply not so.

Should also note that the language prefers to anticipatory/regressive assimilation and not progressive, so ns > ss but not zn > zz

[Travis B.]

So I have it that the ancestor of Vrkhazhian doesn't have voiced fricatives, it only has voiceless fricatives. Next I have this set of sound changes:
f s ɬ ʃ x ⭢ v z ɮ ʒ ɣ / V_V
f s ɬ ʃ x ⭢ v z ɮ ʒ ɣ / V_{m n ŋ r l}
{ʡ ʔ}C ⭢ Cː / _
V{ħ h} ⭢ Vː / _C
VC{ʡ ʔ ħ h} ⭢ VːC / _

However given this it means voiced fricatives can't occur word-initially, word-finally, or post consonantally. It also means that voiceless fricatives cannot occur word-medially following a short vowel. Lastly, it means that geminate voiceless fricatives can exist but not geminate voiced fricatives.

So, I need some more ideas to allow voiceless fricatives to exist after short vowels, and to also allow voiced geminate fricatives. I'm not sure if I want voice fricatives in initial, post-consonantal, or final position.

The language is triconsonantal, so analogy may sometimes play a part, but generally allophony doesn't stick. For example, you have a verb rabad- "to guard" which becomes rabad-ni "guard-1sg"and rabad-ma "guard-2sg" but rabat-ta "guard-3sg". It doesn't mean the /d/ in the root will analogize to /t/ for the rest o the paradigm. However, you can have a word like wast-am "mana" which has the construct state form of wasat even though it should be *wazat. One might think the allophony doesn't analogize becuase there are more inflections in the paradigms where the allophony doesn't apply, but I also have a verb wasax- "be loyal" which probably should be *wazax- in most inflections but it's simply not so.

Should also note that the language prefers to anticipatory/regressive assimilation and not progressive, so ns > ss but not zn > zz

Have you thought of having your proto-lang have long fricatives, and then (unconditionally or conditionally) shortening your long fricatives after the intervocalic voicing of short (but not long) fricatives, leaving both voiced and voiceless fricatives intervocalically even after short vowels?

[linguistcat]

Is there anything "fun"* I could do with phonemes like /m^j m^w n^j n^w ŋ^j ŋ^w/, especially at the ends of words? I do have ideas for word initial instances but I don't think they'd work well for medial or final contexts. I'd like to keep a high number of nasal and nasalized phones for this specific language, but I feel like they might not be as easily differentiated word finally especially with some being coarticulations that might make things fuzzier. I want to keep at least plain /m n ŋ/ separate, and I don't want the coarticulated forms to just become plain forms of any of those phonemes, but other than that I'm open to a lot.

* I know fun is subjective. I like less common sound changes that make sense together. I checked for sound changes for nasals like these and most of them collapse to the plain forms, or plain forms of other nasals. If we could avoid that without losing the nasal aspect completely I think I'd be happiest.

[Zju]

Just decompose them to a semivowel and a nasal? Nʲ Nʷ → jN wN / V_#
Or if you're feeling rather more rebellious: Nʲ Nʷ → Ni Nu / _#

[Ahzoh]

Just decompose them to a semivowel and a nasal? Nʲ Nʷ → jN wN / V_#
Or if you're feeling rather more rebellious: Nʲ Nʷ → Ni Nu / _#

Or turn them into nasal diphthongs/nasal vowels. Or turn them into nasal-stop clusters

[linguistcat]

Just decompose them to a semivowel and a nasal? Nʲ Nʷ → jN wN / V_#
Or if you're feeling rather more rebellious: Nʲ Nʷ → Ni Nu / _#

Or turn them into nasal diphthongs/nasal vowels. Or turn them into nasal-stop clusters

Since they were mostly Ni Nu > Nʲ Nʷ; _# in the first place, I'm not going to reverse that but jN and wN could definitely work. And nasal stop clusters could work well intervocalically! Thank you both for the suggestions. I can also do some fun things with pitch accent that way.

[Ahzoh]

These sound and stress changes make sense?

Active Voice:

Code: Select all

ìt-ta-ynád-ma        > ìt-tī-nád-ma
it-tà-ya-ynád-ma     > it-tà-yī-nád-ma

Negative Active Voice:

Code: Select all

it-tà-ynad-máh-sa    > it-tī̀-nad-mā́sa
ìt-ta-yà-ynad-máh-sa > ìt-ta-yī̀-nad-mā́sa

Applicative voice:

Code: Select all

it-tà-yannád-ma        > it-tà-yannád-ma
ìt-ta-yà-yannád-ma     > it-tī̀-yannád-ma

Negative Applicative Voice:

Code: Select all

ìt-ta-yànnad-máh-sa    > ìt-ta-yànnad-mā́sa
it-tà-ya-yànnad-máh-sa > ìt-tī-yànnad-mā́sa

Graves indicate secondary stress while acute indicate primary stress.

Now I ask this because normally glides like /j w/ elide intervocalically, in coda, and word-finally. But if I do that then a lot of important gramamatical distinctions will be lost. I generally solve this by making the exception that they don't elide when they follow a stressed or long vowel.

[spindlestar]

I'm working on mangling some morphological boundaries and currently have the following system for getting rid of /h/:

• Ph > Pʰ | stops become aspirated
• Nh > Nː | nasals become geminated (complete assimilation)
• hh > ːʔ | second h becomes a glottal stop; first h vanishes, vowel before takes compensatory lengthening
• Vh > Vː | h vanishes, vowel before takes compensatory lengthening

...all of which are unremarkable. the ones I'm wavering on are the liquids and glides:

• my instinct is to have a lateral fricativize, lh > ɬ
• and I would like to have that be a pattern, so that /r/, /j/, and /w/ also fricativize. currently, the closest representations I can determine for what my tongue wants to do with that is rh > ʐ — jh > ç — wh > ɸ

...but i'd love a reality check on how plausible that feels, particularly with regard to what feels like somewhat inconsistent voicing shifts.

[StrangerCoug]

At least conditional /r/ > /ʐ/ is attested, and the rest I find plausible (the last one especially with /ʍ/ as an intermediate step).

[bradrn]

I'm working on mangling some morphological boundaries and currently have the following system for getting rid of /h/:

• Ph > Pʰ | stops become aspirated
• Nh > Nː | nasals become geminated (complete assimilation)
• hh > ːʔ | second h becomes a glottal stop; first h vanishes, vowel before takes compensatory lengthening
• Vh > Vː | h vanishes, vowel before takes compensatory lengthening

...all of which are unremarkable. the ones I'm wavering on are the liquids and glides:

• my instinct is to have a lateral fricativize, lh > ɬ
• and I would like to have that be a pattern, so that /r/, /j/, and /w/ also fricativize. currently, the closest representations I can determine for what my tongue wants to do with that is rh > ʐ — jh > ç — wh > ɸ

...but i'd love a reality check on how plausible that feels, particularly with regard to what feels like somewhat inconsistent voicing shifts.

All of this seems reasonable (with the possible exception of hh→ːʔ). For /rh/, some voiceless options could be /θ/ or /ʂ/.

[spindlestar]

All of this seems reasonable (with the possible exception of hh→ːʔ).

The current phonotactics are such that if an /hh/ dropped entirely it would create a vowel hiatus—the sample word I was working with is tih-hik "close to the speaker" + (stative) > tiːʔik "this"—which the stop interrupts.

For /rh/, some voiceless options could be /θ/ or /ʂ/.

I like /rh/ > /ʂ/, thank you bradrn!

[Darren]

/ʃ/ is the normal fricative outcome of /r̥/, e.g. Tsakonian. (and /ʂ/ is basically /ʃ/ anyway)

[Travis B.]

I'm working on mangling some morphological boundaries and currently have the following system for getting rid of /h/:

• Ph > Pʰ | stops become aspirated
• Nh > Nː | nasals become geminated (complete assimilation)
• hh > ːʔ | second h becomes a glottal stop; first h vanishes, vowel before takes compensatory lengthening
• Vh > Vː | h vanishes, vowel before takes compensatory lengthening

...all of which are unremarkable. the ones I'm wavering on are the liquids and glides:

• my instinct is to have a lateral fricativize, lh > ɬ
• and I would like to have that be a pattern, so that /r/, /j/, and /w/ also fricativize. currently, the closest representations I can determine for what my tongue wants to do with that is rh > ʐ — jh > ç — wh > ɸ

...but i'd love a reality check on how plausible that feels, particularly with regard to what feels like somewhat inconsistent voicing shifts.

All of this seems reasonable (with the possible exception of hh→ːʔ). For /rh/, some voiceless options could be /θ/ or /ʂ/.

I agree. The only one of these that seems off to me is hh > ːʔ.