Sound Change Quickie Thread

[roninbodhisattva]

/i u e o a/, and yes, that was the reason for the change (avoidance of the palatalization that is)

[Nortaneous]

maybe /s/ is phonetically velarized, although then you'd have to justify that

[Jetboy]

Is there any way to justify /ʃ/ or /sʲ/ > /sʼ/? What about /sʔ/ or /ʔs/?

[Mbwa]

maybe /s/ is phonetically velarized, although then you'd have to justify that

dunno how retarded this is, but maybe there could've been an earlier /s_d s/ contrast, and the (post)alveolar /s/ velarised to sharpen the contrast.

then the /s_d/ could go a few different routes. I'd say it depends on the rest of the inventory.

[sirdanilot]

Is there any way to justify /ʃ/ or /sʲ/ > /sʼ/? What about /sʔ/ or /ʔs/?

The latter seems at least slightly plausible, but might not yield the desired frequency of /s'/.

What seems more plausible to me, is, for example, fortition of geminate *s: . Or perhaps, *s+h > *sh (aspirated s) > /s'/, though I'm really just pulling the latter out of my arse here

Keep in mind /s'/ is a very marked consonant; it's not extremely frequent and it is articulatory difficult to pronounce. You won't find it if the lang doesn't obtain many other ejectives as well.

My conlang has /s'/ , /k'/ and /t'/ for ejectives, though historically also /f'/, /x'/ and /p'/.

[Zaris]

I've reached a dead end in the sound changes I am making for proto-Elezaic, the analogical equivalent to PIE in my conworld Elezai. Syllable roots can contain large clusters; such as: *brlnt "new". My goal is to eliminate these clusters in my current project Shanus. But I'm finding trouble pinpointing the exact changes that I want. I've tried inserting epinthetic vowels. But this results in lexemes like *bralanat which has too many syllables for my tastes. I've considered turning all enclosed sonorants into vowels. But *buēt also does not suit my tastes. I was thinking of combining the two but i believe it is more natural for a language to adopt only one kind of sound change for one class of phonemes, unless I'm wrong. So my question is: Does anyone have any idea of where I could take this language or can anyone show me a counter example to my belief in a single sound change per phoneme class. Any help would be greatly appreciated.

[LinguistCat]

I've reached a dead end in the sound changes I am making for proto-Elezaic, the analogical equivalent to PIE in my conworld Elezai. Syllable roots can contain large clusters; such as: *brlnt "new". My goal is to eliminate these clusters in my current project Shanus. But I'm finding trouble pinpointing the exact changes that I want. I've tried inserting epinthetic vowels. But this results in lexemes like *bralanat which has too many syllables for my tastes. I've considered turning all enclosed sonorants into vowels. But *buēt also does not suit my tastes. I was thinking of combining the two but i believe it is more natural for a language to adopt only one kind of sound change for one class of phonemes, unless I'm wrong. So my question is: Does anyone have any idea of where I could take this language or can anyone show me a counter example to my belief in a single sound change per phoneme class. Any help would be greatly appreciated.

You could vocalize some of the consonants. To use your example, the l in brlnt could vocalize to u yielding brunt. Depending on what you want to go for, you could vocalize different consonants or multiple in some words. If you also had a word brlt you could just vocalize the l in that getting brut, and vocalize both the l and n in brlnt getting bru~t. Then you'd also have nasal vowels, if you like that sort of thing.

[Cedh]

I've reached a dead end in the sound changes I am making for proto-Elezaic, the analogical equivalent to PIE in my conworld Elezai. Syllable roots can contain large clusters; such as: *brlnt "new". My goal is to eliminate these clusters in my current project Shanus. But I'm finding trouble pinpointing the exact changes that I want. I've tried inserting epinthetic vowels. But this results in lexemes like *bralanat which has too many syllables for my tastes. I've considered turning all enclosed sonorants into vowels. But *buēt also does not suit my tastes. I was thinking of combining the two but i believe it is more natural for a language to adopt only one kind of sound change for one class of phonemes, unless I'm wrong. So my question is: Does anyone have any idea of where I could take this language or can anyone show me a counter example to my belief in a single sound change per phoneme class. Any help would be greatly appreciated.

You could decide that only one epenthetic vowel per root can be added; with *brlnt this would give you something like *bralnt. If you feel that the final cluster /lnt/ is still too complicated, you could then just delete the /l/, giving *brant. Note that this is the same outcome that you'd get if you stated that "exactly one sonorant per root may become vocalic". (And it's quite similar to how Indo-Iranian resolved the syllabic sonorants of PIE.)

A sister language could instead vocalise the /n/ and maybe also the /r/, giving something like *bralat or *balat...

[TaylorS]

Don't forget metathesis:

branlt > brantl

[roninbodhisattva]

Don't forget metathesis:

branlt > brantl

Strikes me as an odd metathesization.

[Drydic]

Don't forget metathesis:

branlt > brantl

Strikes me as an odd metathesization.

*metathesis

and why? -ntl- seems more likely than -nlt- to me.

[roninbodhisattva]

Don't forget metathesis:

branlt > brantl

Strikes me as an odd metathesization.

*metathesis

and why? -ntl- seems more likely than -nlt- to me.

It just seems to be a very odd word final cluster to me. Unless it went syllabic after the change, which is completely possible. But -lnt- is nice and neat sonority-hierarchy wise, so I don't see the reason why it would metathesize in the first place. Though I suppose weirder things have happened.

[Jetboy]

Yeah, I according to my historical linguistics book, a dental obstruent followed by /l/ is fairly cross-linguistically marked, and often undergoes metathesis to get rid of the cluster; metathesis to form it seems unlikely.

[sirdanilot]

brlnt > brntl (metathesis) > bɹntəl (with a syllabic alveolar approximant)
or: brlnt > brntl > brəntəl.

if you don't like methathesis, go with a lateral flap: bəɿənt (sorry I forgot the actual ipa symbol)
you can then proceed to have fun with nasalization: baɿãt

I liked the other vocalization suggestions as well, though.

[Zaris]

These are all some good suggestions and I thank you all for suggesting them. These suggestions have given me an idea of what to do with this language next. Therefore, I have devised a rule:

"Only one sonorant per root had an epinthetic vowel inserted following the effected sonorant. Sonorants that adopted the nucleic position after this effect took place were then vocalized. However the latter rule did not apply to atonic vocalic sonorants. These specific sonorants had epinhetic vowels inserted preceding them."

Therefore, *brlnt *bralnt *bralet
*rbrlnt *arbralnt *arbralet

This rule seems original, simple, and logical to me. But I am always open for creative criticism. I would like to ask what you think of this rule, considering its naturalness most importantly. Thank you!

(On a side note. I must give credit were credit is due. It is cedh that really inspired this rule and is send him much thanks )

[Cedh]

(On a side note. I must give credit were credit is due. It is cedh that really inspired this rule and is send him much thanks )

You're welcome.

[roninbodhisattva]

Not necessarily a sound change, but I couldn't think of anywhere else to put this:

Would it be implausible to have word/syllable initial stop + nasal clusters without allowing stop + liquid clusters in that same position?

[Nortaneous]

Depends on what liquids you have, I'd say. But you could definitely at least get away with prestopped nasals there.

[roninbodhisattva]

Depends on what liquids you have, I'd say. But you could definitely at least get away with prestopped nasals there.

I'd like to have /km kn tm tn/ and then I have a lateral /l/ that I really don't want to combine with anything.

[Zaris]

Depends on what liquids you have, I'd say. But you could definitely at least get away with prestopped nasals there.

I'd like to have /km kn tm tn/ and then I have a lateral /l/ that I really don't want to combine with anything.

It does seem more likely that other liquids should be allowed to take the place of /mn/. However you could attribute lack of C/l/ clusters to a change in process. maybe something like: CLV > CV or CLV CVV with L being an liquid. However only /l/ would be effected up to that point in you language. It really depends on where you want you language to head, if your concerned about sound change and diachronics.

And as Nort said, you could just use pre/post nasal stops.

[roninbodhisattva]

And as Nort said, you could just use pre/post nasal stops.

I might actually just go with this, at least at the phonetic level. Do the stop parts of prestopped nasals have to be homorganic to the nasal? That is, is something like [^km] reasonable?

[Zaris]

And as Nort said, you could just use pre/post nasal stops.

I might actually just go with this, at least at the phonetic level. Do the stop parts of prestopped nasals have to be homorganic to the nasal? That is, is something like [^km] reasonable?

I think you have it backwards pal

[roninbodhisattva]

And as Nort said, you could just use pre/post nasal stops.

I might actually just go with this, at least at the phonetic level. Do the stop parts of prestopped nasals have to be homorganic to the nasal? That is, is something like [^km] reasonable?

I think you have it backwards pal

How? A prestopped nasal is a nasal consonant with a stop articulation before the nasal (stop) articulation, right? If so- does the stop articulation have to be at the same place of articulation as the nasal?

EDIT: Or is it just completely a difference in timing?

[Zaris]

And as Nort said, you could just use pre/post nasal stops.

I might actually just go with this, at least at the phonetic level. Do the stop parts of prestopped nasals have to be homorganic to the nasal? That is, is something like [^km] reasonable?

I think you have it backwards pal

How? A prestopped nasal is a nasal consonant with a stop articulation before the nasal (stop) articulation, right? If so- does the stop articulation have to be at the same place of articulation as the nasal?

Oh crap I had no idea prestopped nasals existed. So i thought you just got prenasal stops backwards. XD My bad brother

[roninbodhisattva]

I think you have it backwards pal

How? A prestopped nasal is a nasal consonant with a stop articulation before the nasal (stop) articulation, right? If so- does the stop articulation have to be at the same place of articulation as the nasal?

Oh crap I had no idea prestopped nasals existed. So i thought you just got prenasal stops backwards. XD My bad brother

S'all good. Yeah, they're totally something. I think the term is generally applied to a nasal stop that has delayed nasality though, from what I'm reading. So it might just be easier to say that there are word initial stop + nasal clusters and that they don't have to be of the same POA.