Sound Change Quickie Thread

[Nortaneous]

It seems that languages with one postalveolar PoA generally have /ʒ/, whereas those with two tend to have /ʐ ʑ/, because */ʐ ʒ/ or */ʒ ʑ/ would simply be too close together. An inventory with all of /ʐ ʒ ʑ/ is extremely rare outside Sino-Tibetan.

this is probably an artifact of notational conventions more than anything else

[Ahzoh]

How would I go about creating the distinction of /l l̃ r r̃ ɦ ɦ̃/ and them also having syllabic counterparts /l̩ l̩̃ r̩ r̩̃ ɦ̩ ɦ̩̃/?

the /ɦ̩ ɦ̩̃/ may seem weird but i think a word like /dʱɦ̩s/ is possible. Maybe.

[Rounin Ryuuji]

Coalescence with nasal consonants that were present, or for the syllabic ones, reduction of nasalised vowels before those consonants to schwa, with the schwa disappearing and the nasality transferring to the new syllabic sound?

[bradrn]

the /ɦ̩ ɦ̩̃/ may seem weird but i think a word like /dʱɦ̩s/ is possible. Maybe.

Well, the problem here is that /ɦ/ is really just a breathy-voiced version of an adjacent vowel, so you’ll probably get it vocalising to /d̤ɑ̤s/ or something. I don’t even see how you could get syllabic glottals in the first place. Nasalised /h̃/ actually is pretty-well attested (cf rhinoglottophilia), but it usually ends up realised as nasalisation on the adjacent vowels; here it would vocalise to /d̤ɑ̤̃s/

[StrangerCoug]

I'm working on a phonology for a conlang and I'd like there to be a dialect which has doubly-articulated labial–velar consonants as allophones of labialized velars. (i.e. I want [k͡p k͡pʰ ɡ͡b ŋ͡m] to occur allophonically for /kʷ kʷʰ ɡʷ ŋʷ/) What conditions can I place on the allophony?

For the same conlang, how common are doubly-articulated labial–palatal consonants, even as allophones? The velar articulation becomes palatal before front vowels in this lang and I want the speakers of this language to go to war with one of the other people in my conworld—if [c͡p c͡pʰ ɟ͡b ɲ͡m] are super-rare, I could use those in some interesting shibboleths to use against their enemies.

[bradrn]

I'm working on a phonology for a conlang and I'd like there to be a dialect which has doubly-articulated labial–velar consonants as allophones of labialized velars. (i.e. I want [k͡p k͡pʰ ɡ͡b ŋ͡m] to occur allophonically for /kʷ kʷʰ ɡʷ ŋʷ/) What conditions can I place on the allophony?

Apparently Saigon dialect of Vietnamese has [k͡p] as an allophone of final [k] after [u oː ɔ] (Pham 2006). (The other language which comes to mind here is Sougb, but that’s unhelpful as its orthographic ⟨gb⟩ is actually just [gʷ̚].)

For the same conlang, how common are doubly-articulated labial–palatal consonants, even as allophones? The velar articulation becomes palatal before front vowels in this lang and I want the speakers of this language to go to war with one of the other people in my conworld—if [c͡p c͡pʰ ɟ͡b ɲ͡m] are super-rare, I could use those in some interesting shibboleths to use against their enemies.

Such sounds are extremely rare, if they exist at all, but Yélî Dnye supposedly has labial–postalveolar obstruents in words such as /ṭ͡pənə ṇ͡mḍ͡boo ṇ͡mo/ (where an underdot represents apical retroflexion).

[aliensdrinktea]

I know evolution from voicing contrast to aspiration contrast is a thing, but what about the reverse? And how would such a change realistically occur? Would it just be a case of unaspirated consonants becoming voiced and aspirated ones losing their aspiration, or would the acquisition of voicing be more context-dependent? (I considered applying voicing intervocalically, but then I'm not sure how to introduce word-initial voicing.)

[bradrn]

Would it just be a case of unaspirated consonants becoming voiced and aspirated ones losing their aspiration, or would the acquisition of voicing be more context-dependent?

Both of these sound fine to me. An unconditional shift would basically be just a case of shifting stops along in VOT.

[Nortaneous]

I know evolution from voicing contrast to aspiration contrast is a thing, but what about the reverse?

This happened in Dutch, at least if you take the aspiration contrast of most Germanic languages as original. Probably influence from Romance.

[aliensdrinktea]

Good to know. Thanks!

[Emily]

(let me know if this is too open-ended of a question for this thread)

what are sound changes that happen to tones? that is, not where do they come from, but once a language has them how do they end up changing later? do they stick around? split? merge? go away? etc.

[Richard W]

I'm not sure we have enough information to say much on how tones change, but change and merge they do. We've been recording Siamese for long enough to see changes in the contour shapes of tones. They seem to be driven by strategies for telling tones apart - which I suppose is a fairly universal feature of sound change in general. Tai-Kadai tones show conditioning effects by the phonation of initials and vowel length, and contrasts in these features can be transferred to the tones. If Tai languages with newly developed voiced consonants lost this voice contrast, there would be yet another tone split preserving a former voicing contrast.

[Creyeditor]

(let me know if this is too open-ended of a question for this thread)

what are sound changes that happen to tones? that is, not where do they come from, but once a language has them how do they end up changing later? do they stick around? split? merge? go away? etc.

I think you can google tonexodus for sound changes that lead to tone loss. Another great place is to look at is Bantu. Tone in Proto-Bantu is well reconstructed. Crazy things happen sometimes, like tones shifting or becoming their polar opposite. Simple dissimilation (known as Meussens rule) is more common, as well as the emergence of floating tones when vowels are deleted.

[geltreo]

Aromanian seems to have repurposed the accusative as the nominative.

I have language resources/sources on Aromanian that are written in Greek (although they have their own Greek propaganda trash shoved in there) and I am paternally Aromanian (and possibly half-maternally) so you can ask me anything.

[Pabappa]

Aromanian seems to have repurposed the accusative as the nominative.

I have language resources/sources on Aromanian that are written in Greek (although they have their own Greek propaganda trash shoved in there) and I am paternally Aromanian (and possibly half-maternally) so you can ask me anything.

Nice to meet you, and thanks for the reply ..... i actually dont remember writing out that post and i think i was just blushing past different things at the time to answer the original question. But you could answer another simple question that I think I know but have not been able to confirm, since resources arent always easy to find ....

Does Aromanian have final /ʲ/ in the same positions as Romanian, that is, wherever it is spelled with a single final -i?
Does Aromanian have /ʲ/ in any position where it is *not* spelled with a single final -i? i.e. i see a word spelled cochilj on Wiktionary, though it is a redlink.
Does Aromanian retain phonemic labialization in any position, whether spelled with u, with some other letter, or not spelled at all?

This is really three questions instead of one, but they're all tied together..... thank you for your help.

[geltreo]

Aromanian seems to have repurposed the accusative as the nominative.

I have language resources/sources on Aromanian that are written in Greek (although they have their own Greek propaganda trash shoved in there) and I am paternally Aromanian (and possibly half-maternally) so you can ask me anything.

Nice to meet you, and thanks for the reply ..... i actually dont remember writing out that post and i think i was just blushing past different things at the time to answer the original question. But you could answer another simple question that I think I know but have not been able to confirm, since resources arent always easy to find ....

Does Aromanian have final /ʲ/ in the same positions as Romanian, that is, wherever it is spelled with a single final -i?
Does Aromanian have /ʲ/ in any position where it is *not* spelled with a single final -i? i.e. i see a word spelled cochilj on Wiktionary, though it is a redlink.
Does Aromanian retain phonemic labialization in any position, whether spelled with u, with some other letter, or not spelled at all?

This is really three questions instead of one, but they're all tied together..... thank you for your help.

First of all I have to discern that I am not a linguist and my linguistic knowledge of Aromanian is lacking in deepth. Since my father is essentially assimilated into Greek society/Greeks it's hard to learn Aromanian by yourself without anyone around and through pdfs alone.

1. I think Aromanian does have the /j/ sound in the same positions as Romanian has, yes, wherever it's spelled with single final -i.
2. Yes. Usually words ending in -lj don't have a final -i sound or letter. You can spell -lj at the end of the word without the -i.
3. I am not knowledgable enough to answer this question.

[Ahzoh]

How could I take the following declension paradigms (acutes mark stress)...

Masculine (sarbi "raven"):

	Singular	Plural
Nominative	sárb-i	sarb-ī́
Accusative	sarb-ít	sarb-ī́t
Instrumental	sarb-ís	sarb-ī́s
Locative	sarb-ík	sarb-ī́k
Vocative	sárb-i	sarb-ī́

Feminine (ḫuzu "horse"):

	Singular	Plural
Nominative	ḫúz-u	ḫuz-ū́
Accusative	ḫuz-út	ḫuz-ū́t
Instrumental	ḫuz-ús	ḫuz-ū́s
Locative	ḫuz-úk	ḫuz-ū́k
Vocative	ḫúz-u	ḫuz-ū́

...and diachronically/synchronically reducing it to this?...

Masculine (sarbi "raven"):

	Singular	Plural
Nominative	sárb-i	sarb-ī́
Oblique	sárb-e	sarb-ḗ

Feminine (ḫuzu "horse"):

	Singular	Plural
Nominative	ḫúz-u	ḫuz-ū́
Oblique	ḫúz-a	ḫuz-ā́

[Zju]

Vowel breaking and then monophthongisation? Would ofc affect other parts of the lang, too.

[Ahzoh]

well I don't know if there's anything special about the environments of the vowels that might cause changes. I think I remember reading somewhere that an alveolar obstruent like /t/ *could* lower high vowels, but then what of /s/ and /k/? I'm also not sure what role stress would play in combination with the endings.

Well, I think these two paradigm sets are supposed to develop parallel to each other since the five-case one is preserved in the written variety while the two-case one is found in the spoken variety. So maybe they are both the result of some unknown older paradigm, which I lack the imagination to reconstruct =/

[Zju]

I was thinking that cases other than nom and acc could simply drop off, with acc becoming oblique case. AFAIK coronal consonants can trigger vowel breaking by centering the second element (in a way assimilating preceding vowel's PoA to the following consonant), and after that, pretty much all bets are off as to what happens to individual Və diphthongs.

so I don't see why you couldn't have
iT uT → iəT uəT → ieT uoT → ieT uaT → eT aT / _{C #}
where T is any coronal, and of course other vowels will probably also be affected in some way. Finally, you could just drop word-final plosives or something. Or just have the oblique endings drop their consonants, simplification of individual inflectional morphemes happens.