Sound Change Quickie Thread

[Whimemsz]

Just have different resolutions for some of the different clusters, like Nort said. So, for example, you can have Sl > tɬ, lS > ɬS, at least some of which > ɬ, as you planned. You can also have things like fricative+ɬ > ɬ. For example (IIRC this is a future English thing, so I'm using English examples, but even if I'm misremembering, you get the idea [also I'm using my own dialect, but again, you get the idea]):

atlas > ætɬɪs
apply > ətɬaʃ
incline > ɪtɬ:ain
alter > ɑɬtɚ > ɑɬɚ
alpine > æɬpain > ?
also > ɑɬsox > ɑɬox
afflict > əfɬɪk > ? (əɬɪk, ətɬɪk, and əfɬɪk would all be plausible outcomes IMO)

etc.

[Knit Tie]

It is actually future English! Though none of the modern-day /l/ survives to see the devoicing begin - that was the reason why I asked about the feasibility of losing /l/ twice earlier in the thread - so the /l/ that devoices actually comes from loanwords.

Though there's supposed to be extensive borrowing from Modern AmE during the early period of the language, as the AmE is the official language then and the language in question, Suraic, is the everyday one, so yeah, your examples work!

Perhaps I can have plosives+/ɬ/ remain as they are, while fricatives before it fortition to plosives and /ɬS/ > /ɬ/ invariably? This would give the following, with adjustments to vowels to fit the vowel changes:

atlas > ɛtɬaʃ
apply > apɬaʃ
incline > akʰɬ:ãj̃
alter > aɬteʃ
alpine > ɛɬpãj̃
also > ɔɬsox > ɔɬox
afflict > afɬak > apɬak

Also, the language will borrow the Arabic definite article al, which will become either /ar/ or /aɬ/ depending on what it's preceding, so the name of the language itself will be <aɬurajeʃu> from al+Surai+plural suffix+nominalizing suffix.

[Whimemsz]

It is actually future English! Though none of the modern-day /l/ survives to see the devoicing begin - that was the reason why I asked about the feasibility of losing /l/ twice earlier in the thread - so the /l/ that devoices actually comes from loanwords.

Ah, gotcha.

Perhaps I can have plosives+/ɬ/ remain as they are, while fricatives before it fortition to plosives and /ɬS/ > /ɬ/ invariably? This would give the following, with adjustments to vowels to fit the vowel changes:

Sounds and looks good to me (I'm assuming from your examples that you mean ɬ+fric. > ɬ invariably).

Also, the language will borrow the Arabic definite article al, which will become either /ar/ or /aɬ/ depending on what it's preceding, so the name of the language itself will be <aɬurajeʃu> from al+Surai+plural suffix+nominalizing suffix.

Will this be borrowed in the process of borrowing lots of words directly from spoken Arabic, or will they be borrowed through written Arabic? Because in spoken Arabic the |l| of the article assimilates to a following coronal sound ("sun letters") already, so al-Surai would be /assuraj/ (or /assure:/) in MSA, and with the same assimilation in all vernacular varieties.

[Knit Tie]

Also, the language will borrow the Arabic definite article al, which will become either /ar/ or /aɬ/ depending on what it's preceding, so the name of the language itself will be <aɬurajeʃu> from al+Surai+plural suffix+nominalizing suffix.

Will this be borrowed in the process of borrowing lots of words directly from spoken Arabic, or will they be borrowed through written Arabic? Because in spoken Arabic the |l| of the article assimilates to a following coronal sound ("sun letters") already, so al-Surai would be /assuraj/ (or /assure:/) in MSA, and with the same assimilation in all vernacular varieties.

Definitely written Arabic. The only actual Arabic speakers who ended up on the planet of Sura (hence the name of the language) are a small Egyptian minority that quickly switches to the day-to-day lingua franca of Middle Suraic, or Ynghysh /ɨ̃ɣɨʃ/, while the rest of the peoples who write in Arabic are the Berbers, the Central Asians and the African Muslims. Same with the English, actually - by the time the borrowing happens, plenty of people write it, but very very few speak it, which leads to a lot of instances of spelling pronunciation in the borrowings, like <-tion> being pronounced [-tjõ].

[Kuchigakatai]

People who write in Arabic but actually speak Ynghysh... I hope I'm not the only one getting the joke.

[foxcatdog]

Any attested examples of certain tones weakening sounds say voiced stops to approximates or voiceless stops to fricatives.

[Knit Tie]

In sequences of two heterorganic fricatives, could one of them, the one that has a plosive counterpart fortition into said plosive?

To elaborate: I have a lot of sequences a-la /xʃ/, like in the word for tower, /taxʃa/. Would it ve unrealistic to have it become /takʃa/?

[bradrn]

In sequences of two heterorganic fricatives, could one of them, the one that has a plosive counterpart fortition into said plosive?

To elaborate: I have a lot of sequences a-la /xʃ/, like in the word for tower, /taxʃa/. Would it ve unrealistic to have it become /takʃa/?

Related question: can the second fricative fortite as well? So e.g. /xʃ/ → /xt͡ʃ/. I came up with this change independently but have no idea if it’s plausible…

[Richard W]

In sequences of two heterorganic fricatives, could one of them, the one that has a plosive counterpart fortition into said plosive?

To elaborate: I have a lot of sequences a-la /xʃ/, like in the word for tower, /taxʃa/. Would it ve unrealistic to have it become /takʃa/?

Related question: can the second fricative fortite as well? So e.g. /xʃ/ → /xt͡ʃ/. I came up with this change independently but have no idea if it’s plausible…

The general answer is that they are quite plausible - Greek exhibits /euθ/ > /eft/ (in Demotic /lepteria/ʹfreedom') and /eus/ > /eps/ (in the aorist of a fair few verbs). English has next and dialect heckfore ʹheifer'. _Diphtheria_ with /p/ is a related example.

However, these are all changes of fricative to *plosive*. I can't think of a change of fricative to affricate in these environments.

[Knit Tie]

Unrelated question, but still something that's bugging me:

Can a language with /tʃ/ and /dʒ/ [*]sequences[*] somehow not start to treat them like affricates and keep them as sequences? As in, disallow them in initial positions and stick a prothetic vowel in front of them in this environment?

[bradrn]

In sequences of two heterorganic fricatives, could one of them, the one that has a plosive counterpart fortition into said plosive?

To elaborate: I have a lot of sequences a-la /xʃ/, like in the word for tower, /taxʃa/. Would it ve unrealistic to have it become /takʃa/?

Related question: can the second fricative fortite as well? So e.g. /xʃ/ → /xt͡ʃ/. I came up with this change independently but have no idea if it’s plausible…

The general answer is that they are quite plausible - Greek exhibits /euθ/ > /eft/ (in Demotic /lepteria/ʹfreedom') and /eus/ > /eps/ (in the aorist of a fair few verbs). English has next and dialect heckfore ʹheifer'. _Diphtheria_ with /p/ is a related example.

However, these are all changes of fricative to *plosive*. I can't think of a change of fricative to affricate in these environments.

Well, affricates — especially /tʃ dʒ/ — are well known to commonly pattern with plosives. But that wasn’t my question; I was asking whether this is plausible for all fricative sequences. In other words, [+fric] [+fric] → [+fric] [+stop] / _ (or → [+fric] [+affricate], as the case may be).

[Richard W]

Well, affricates — especially /tʃ dʒ/ — are well known to commonly pattern with plosives. But that wasn’t my question; I was asking whether this is plausible for all fricative sequences. In other words, [+fric] [+fric] → [+fric] [+stop] / _ (or → [+fric] [+affricate], as the case may be).

Can you find an example of fricative dissimilation turning /ss/ to /ts/, /st/ or [sts]?

[Richard W]

Can a language with /tʃ/ and /dʒ/ [*]sequences[*] somehow not start to treat them like affricates and keep them as sequences? As in, disallow them in initial positions and stick a prothetic vowel in front of them in this environment?

Well, the Hebrew hithpael has metathesis of /t/ and root-initial sibilant, despite the existence of affricates seemingly going back to Proto-Semitic.

[Kuchigakatai]

Well, affricates — especially /tʃ dʒ/ — are well known to commonly pattern with plosives. But that wasn’t my question; I was asking whether this is plausible for all fricative sequences. In other words, [+fric] [+fric] → [+fric] [+stop] / _ (or → [+fric] [+affricate], as the case may be).

Can you find an example of fricative dissimilation turning /ss/ to /ts/, /st/ or [sts]?

Some medieval Hispano-Romance dialects occasionally show <ns> and <ls> for <ss>, so that assi 'this way' (< Lat. ad 'to' + sīc 'thus') or siesso 'anus' (< Lat. sessus 'a sitting') show up as ansi or sielso. Judging by the fact some modern Spanish dialects retain ansí with the pronunciation [anˈsi], the medieval variant spellings may represent old dissimilations. From [s:] > [ns ls] it's not hard to introduce an epenthetic [t] in the middle, and from there drop the first consonant: [anˈsi] > *[anˈtsi] > *[ãˈtsi] > *[aˈtsi].

[Nortaneous]

Well, affricates — especially /tʃ dʒ/ — are well known to commonly pattern with plosives. But that wasn’t my question; I was asking whether this is plausible for all fricative sequences. In other words, [+fric] [+fric] → [+fric] [+stop] / _ (or → [+fric] [+affricate], as the case may be).

Can you find an example of fricative dissimilation turning /ss/ to /ts/, /st/ or [sts]?

In some Papuan languages, plain /s/ turns to [ts] word-initially - but in some cases that's known to be an artifact of /s/ being from earlier *t.

[bradrn]

Well, affricates — especially /tʃ dʒ/ — are well known to commonly pattern with plosives. But that wasn’t my question; I was asking whether this is plausible for all fricative sequences. In other words, [+fric] [+fric] → [+fric] [+stop] / _ (or → [+fric] [+affricate], as the case may be).

Can you find an example of fricative dissimilation turning /ss/ to /ts/, /st/ or [sts]?

Sorry — I meant all fricative sequences except geminates! Not sure how I forgot to write that, but that’s how I implemented it in my SCA.

[Pabappa]

well Greek did phth khth > /ft xt/, but I cant be sure what the path was. It could be that the second element of the series never lenited in the first place ... this would be similar to what happened in Germanic a little earlier on.

[dɮ the phoneme]

I'm starting with an inventory of sibilants (fricatives and affricates) that has a three way contrast between alveolar, retroflex, and alveolo-palatal series. What are some interesting things I can do with the alveolo-palatals to get rid of them, without just merging them unilaterally into one of the other series? I was thinking maybe ɕ ʑ > x j, but that doesn't leave an obvious path for the affricates. Any suggestions?

[dhok]

There's no reason the affricates and fricatives have to undergo the same changes--in many languages' histories, they haven't. If you have newly orphaned /tɕ dʑ/, you could deaffricate them to get new fricatives, or you could e.g. depalatalize them to give /ts dz/. Weirder paths, for example to lateral affricates, are also possible.

[Richard W]

well Greek did phth khth > /ft xt/, but I cant be sure what the path was. It could be that the second element of the series never lenited in the first place ... this would be similar to what happened in Germanic a little earlier on.

Which is why I quoted the Greek developments from /euθ/ and /eus/ - the change from voiceless aspirates to fricative seems to be older than the conversion of the second element of the diphthong from vowel to fricative.

Germanic may have fricativised both voiceless stops in a cluster. That makes it easier to lose the dental in the development of seven from *septm.