Sound Change Quickie Thread

Substantial postings about constructed languages and constructed worlds in general. Good place to mention your own or evaluate someone else's. Put quick questions in C&C Quickies instead.
User avatar
Soap
Smeric
Smeric
Posts: 1228
Joined: Sun Feb 16, 2003 2:57 pm
Location: Scattered disc
Contact:

Re: Sound Change Quickie Thread

Post by Soap »

I bumped your post, so I'm quoting the entire post to save people the trouble of paging back and forth.
Das Public Viewing wrote:So, I'm trying to make a language family along the lines of Indo-European or Afroasiatic, and I was trying to think of a couple mega-unstable phonemes for the proto-lang. Could y'all tell me which of these are plausibly unstable?
I'd first say that I think any of these could be perfectly stable, in case you decide you want to have a conservative branch of the family that preserves most or all of the distinctions intact.
/ɬ/ the Magnificent: It is of course alongside a /l/ and a /ʎ/, but I was thinking of a satemization-type distinction between evolving it into /θ/ or /ʃ/, which then can become /f/t/s/ and /s/ʂ/x/, respectively, plus maybe others if I think of it.
I like this idea.
PIE Dorsals Plus: This consists of the theory I've often seen lurking on the PIE threads here, of a uvular-velar-labiovelar distinction, but with an added labiouvular series/consonant/haven't decided. This set of /k/kʷ/q/qʷ/ could then become any of /c/cʷ/k/kʷ/, /k/p/kˤ/pˤ/, etc.
I use this setup in some of my conlangs, and a fourfold contrast of k/kʷ/ḳ/ḳʷ in others (here the ones with dots are ejectives). In my languages, non-ejective voiceless stops are always aspirated. One side of the family also has p/pʷ/ṗ/ṗʷ, the other just has a single p.

I've done a lot of things with them, which I can share. Plus, other ideas from natlangs you might like:

If you have no voicing distinction and want to grow one, you might want changes like these:

/k/ > /k/, but /kʷ/ > /gʷ/ and possibly later to /g/ (stops become voiced before [w] in Mingo, a language of West Virginia)

/k kʷ/ unchanged, but /q qʷ/ > /g gʷ/ (/q/ > /g/ is found in some dialects of Arabic)

Other ideas:
/k kʷ/ unchanged, but /q qʷ/ > /x xʷ/. European languages seem resistant to /q/, but tolerate postvelar fricatives such as /χ/, which could then either remain or shift forward to /x/.

Labialization affects surrounding vowels, then drops. Especially useful if your parent language has only a few vowels, especially if there is an incomplete set of rounded vowels or none at all. Note, though, languages like these would probably have more labialized consonants besides just velar and uvular stops.

/k kʷ q qʷ/ > /c kʷ k kʷ/. I generally avoid labialized palatals in my conlangs, and although there are quite a few languages that do have them, there's definitely also a precedent for labialized consonants resisting palatalization (as they did in satem languages in IE). The presence of /q/ could help the labialized consonants hang on to their distinctive labialization even after the original plain [k] slips away.

/c p k p/. Combination of the above shift and the /kʷ/>/p/ type shift. Will produce a language with relatively sparse use of dorsal consonants.

/k kʷ q qʷ/ > /k kʷ 0 w/. No natlang precedent that I know of, but I've used this in a language where the consonant inventory is minimal and the language is nearly CV, so I expect consonant weakening to be common and nearly unconditional. I used a glottal stop as an intermediary,


One other thing to keep in mind is that most sound changes are conditional, so these are likely to be merely the most common outcomes. Even in PIE the satemization rule had quite a few exceptions. I was actually about to type out an example of a shift Ive done right now, but I think it's not one of my best ideas so I'll just link to https://chridd.nfshost.com/diachronica/ if youre not familiar with it already, where there are ample examples of attested natlang shifts in both directions.


Gemination Constipation: There are so many things this could become! It could rhinoglottophilia! It could affricates! It could failure to lenit! etc.
Those sound like good ideas. I think ejectivization of voiceless geminates is also attested, and I think aspiration of voiced ones also is, but the latter change may have first involved devoicing.

Tone: Okay, this may not have any far-reaching distinction, but could tone produce differing stress patterns depending on the duaghters? Or does it produce mostly the same effect? Also, this could maybe affect vowels?
Tone generally has no effect on vowel quality, if that's what you're saying, but there are a few exceptions.
Sunàqʷa the Sea Lamprey says:
Image

User avatar
Das Public Viewing
Sanci
Sanci
Posts: 20
Joined: Fri Jul 10, 2015 5:56 pm
Location: Right Behind You

Re: Sound Change Quickie Thread

Post by Das Public Viewing »

Soap wrote:One other thing to keep in mind is that most sound changes are conditional, so these are likely to be merely the most common outcomes. Even in PIE the satemization rule had quite a few exceptions. I was actually about to type out an example of a shift Ive done right now, but I think it's not one of my best ideas so I'll just link to https://chridd.nfshost.com/diachronica/ if youre not familiar with it already, where there are ample examples of attested natlang shifts in both directions.
O_O
Now i'm actually kinda interested in what it was. Also, yes, I was thinking of doing conditional changes in a couple branches.
Tone: Okay, this may not have any far-reaching distinction, but could tone produce differing stress patterns depending on the duaghters? Or does it produce mostly the same effect? Also, this could maybe affect vowels?
Tone generally has no effect on vowel quality, if that's what you're saying, but there are a few exceptions.
I mostly meant that the proto-language would be tonal, and I was wondering how stress patterns could differ from branch to branch.
[meta comment]

User avatar
Soap
Smeric
Smeric
Posts: 1228
Joined: Sun Feb 16, 2003 2:57 pm
Location: Scattered disc
Contact:

Re: Sound Change Quickie Thread

Post by Soap »

Das Public Viewing wrote:
Soap wrote:One other thing to keep in mind is that most sound changes are conditional, so these are likely to be merely the most common outcomes. Even in PIE the satemization rule had quite a few exceptions. I was actually about to type out an example of a shift Ive done right now, but I think it's not one of my best ideas so I'll just link to https://chridd.nfshost.com/diachronica/ if youre not familiar with it already, where there are ample examples of attested natlang shifts in both directions.
O_O
Now i'm actually kinda interested in what it was.
Okay.
Note that this language starts out without an /s/.

1.) /k/ > /tʃ/ before .
2.) /tʃi/ > /tʃ/ before another vowel

*thousands of years pass*

57.) /tʃ/ > /ts/ > /s/ unconditionally.
58.) /k/ > /c/ > /tʃ/ unconditionally.
59.) /q/ > /k/ unconditionally.

I simplified it, but basically why I dont like this setup is that it ties the existence of the /s/ phoneme to places where an /i/ follows it, or previously followed it.

I've decided in most of the daughter languages to take out rule #57 (it's an areal phenomenon, hence why it's in more than one branch) and just let the first /k/ > /tʃ/ merge with the much later second one), making the first shift nearly useless. However, sometimes natlangs have "useless" shifts like this that just end up merging with similar shifts later on. I dont think the shift itself is bad, but without leading to /s/ it doesnt really shine, and is nothing really to be proud of.

Also, yes, I was thinking of doing conditional changes in a couple branches.
Can only repeat myself. Especially in a consonant-heavy language, sound changes are likely to be heavily conditional, so try to use unconditional changes sparingly. Im no linguistics PhD, but I honestly believe even the scholars working on PIE reconstruction, and especially pre-PIE construction (glottalic theory; Nostratic, Indo-Uralic, etc) fall into this trap by reconstructing unconditional shifts like t'/t/d > d/t/dʰ lasting thousands of years. One counterexample, though, is the Semitic family, which has a morphology such that a lot of sound changes really are unconditional, because the same consonant appears in many different environments in the same root, and for the sake of comprehension a sound change that might start out conditional seems to spread throughout the rest of the paradigm.
Sunàqʷa the Sea Lamprey says:
Image

User avatar
Nortaneous
Sumerul
Sumerul
Posts: 4544
Joined: Mon Apr 13, 2009 1:52 am
Location: the Imperial Corridor

Re: Sound Change Quickie Thread

Post by Nortaneous »

I'd like to ask this: what is the quickest way to collapse a SAE phoneme inventory inventory (so 5-7 vowels, nasals, liquids, voiced vs. voiceless fricatives and voiced vs. voiceless plosives) into something with as few distinct consonant series as possible (so something like Greenlandic, with nasals vs. fricatives vs. plosives) and a minimal vowel system?
Look at the IE languages with the smallest consonant inventories: Danish and Spanish.

They both start with the consonant inventory of PIE, of course, and then both lose the laryngeals and centumize.

Germanic has Grimm's and Verner's laws, which don't add or remove any phonemes except *z. ɣʷ from gʷʰ merges into w. The voiced fricatives from Verner's law merge into the reflexes of the PIE voiced aspirate series, but the contrasts still exist word-initially. At this stage, there are the following consonants: /p b f t d θ k g x kʷ gʷ xʷ s z m n l r j w/

Then *z merges into /r/, *θ merges into /d/, the labiovelars become KW sequences, and some other things happen that aren't relevant here, leaving /p b f v t d k g h s m n l r j/. From here, you could have /h/ be dropped to lose another consonant, or even do what some varieties of German did and lose the fortis-lenis contrast except in /k/ vs. /g/. (More likely for Danish would probably be p > b, k > g, and t > ts, leaving a stop system of /b d g ts/.)

Spanish is descended from Latin, of course. Latin eliminated the PIE voiced aspirates in a complex series of changes that produced two new consonants, /f/ and /h/, both of which became zero in Spanish. (/f/ is retained in borrowings from Latin and in a few specific environments, but that's not important here -- you could have unconditional *f *h > 0.) Old Spanish expanded the phoneme inventory of Latin with innovated sibilants and affricates; the affricates deaffricated except for /tʃ/, then the voiced sibilants devoiced, then ʃ > x, then the laminal sibilants (from the affricates) merged into OS *s *z as /s/. Also some other things happened that aren't important here. You could probably do without the allophonic variation between voiced plosives and fricatives, and merge *b into *w and lose *d and *g entirely. Whatever.

It's important to remember here that massive mergers can sometimes just happen. Polynesian started out with *p *b *mp *mb series and folded them all into *p. Tocharian collapsed nine PIE velar stops into /k/.
Siöö jandeng raiglin zåbei tandiüłåd;
nää džunnfin kukuch vklaivei sivei tåd.
Chei. Chei. Chei. Chei. Chei. Chei. Chei.

User avatar
Das Public Viewing
Sanci
Sanci
Posts: 20
Joined: Fri Jul 10, 2015 5:56 pm
Location: Right Behind You

Re: Sound Change Quickie Thread

Post by Das Public Viewing »

Soap wrote:Okay.
Note that this language starts out without an /s/.

1.) /k/ > /tʃ/ before .
2.) /tʃi/ > /tʃ/ before another vowel

*thousands of years pass*

57.) /tʃ/ > /ts/ > /s/ unconditionally.
58.) /k/ > /c/ > /tʃ/ unconditionally.
59.) /q/ > /k/ unconditionally.
So...Albanian?
(but seriously, I might steal this)

Also, yes, I was thinking of doing conditional changes in a couple branches.
Can only repeat myself. Especially in a consonant-heavy language, sound changes are likely to be heavily conditional, so try to use unconditional changes sparingly. Im no linguistics PhD, but I honestly believe even the scholars working on PIE reconstruction, and especially pre-PIE construction (glottalic theory; Nostratic, Indo-Uralic, etc) fall into this trap by reconstructing unconditional shifts like t'/t/d > d/t/dʰ lasting thousands of years. One counterexample, though, is the Semitic family, which has a morphology such that a lot of sound changes really are unconditional, because the same consonant appears in many different environments in the same root, and for the sake of comprehension a sound change that might start out conditional seems to spread throughout the rest of the paradigm.[/quote]Point taken. I meant more that I was thinking about /k/kʷ/q/qʷ/>>/c/cʷ/k/kʷ/, where k before back vowels merges with the second two, but yeah, I see what you mean. However, I'm also working on a Semitic-type lang. Would it be unwise to have conditional soundchanges (because I was going to do that)?

Also:
Das Public Viewing wrote:
Tone: Okay, this may not have any far-reaching distinction, but could tone produce differing stress patterns depending on the daughters? Or does it produce mostly the same effect? Also, this could maybe affect vowels?
Tone generally has no effect on vowel quality, if that's what you're saying, but there are a few exceptions.
I mostly meant that the proto-language would be tonal, and I was wondering how stress patterns could differ from branch to branch.
?
[meta comment]

User avatar
äreo
Avisaru
Avisaru
Posts: 326
Joined: Sun Jul 01, 2007 10:40 pm
Location: Texas

Re: Sound Change Quickie Thread

Post by äreo »

Is this outlandish, or?

/æ e i ɑ o u/ > /ɛ ʲe i a o u/
/æː eː iː ɑː oː uː ai oi/ > /ɛː ʲeː iː ɔː oː uː aː wɛ/

Now long vowels rarely appear in unstressed syllables in the motherlang, but for morphological reasons /ai oi/ do. Would /ai oi/ > [aː wɛ] when stressed and > [ɛ ø~y] when unstressed be implausible?

Ascima mresa óscsma sáca psta numar cemea.
Cemea tae neasc ctá ms co ísbas Ascima.
Carho. Carho. Carho. Carho. Carho. Carho. Carho.

User avatar
Nortaneous
Sumerul
Sumerul
Posts: 4544
Joined: Mon Apr 13, 2009 1:52 am
Location: the Imperial Corridor

Re: Sound Change Quickie Thread

Post by Nortaneous »

Looks fine to me.
Siöö jandeng raiglin zåbei tandiüłåd;
nää džunnfin kukuch vklaivei sivei tåd.
Chei. Chei. Chei. Chei. Chei. Chei. Chei.

User avatar
ˈd̪ʲɛ.gɔ kɾuˑl̪
Avisaru
Avisaru
Posts: 255
Joined: Wed May 18, 2016 11:11 pm
Location: Łódź

Re: Sound Change Quickie Thread

Post by ˈd̪ʲɛ.gɔ kɾuˑl̪ »

äreo wrote:Is this outlandish, or?

/æ e i ɑ o u/ > /ɛ ʲe i a o u/
/æː eː iː ɑː oː uː ai oi/ > /ɛː ʲeː iː ɔː oː uː aː wɛ/

Now long vowels rarely appear in unstressed syllables in the motherlang, but for morphological reasons /ai oi/ do. Would /ai oi/ > [aː wɛ] when stressed and > [ɛ ø~y] when unstressed be implausible?
The first part is nice and feasible for me.
The second feels worse: I understand and sometimes apply these things with /oi/ unconsciously in my mother language, but I'd do a reverse thing with /ai/; for this, I'd wait for others' opinions.
In Budapest:
- Hey mate, are you hung-a-ry?

User avatar
Pole, the
Smeric
Smeric
Posts: 1606
Joined: Sat Feb 11, 2012 9:50 am

Re: Sound Change Quickie Thread

Post by Pole, the »

You could do this in two stages.

(1) Unstressed /ai oi/ shorten and monophthongize to /æ ø/. Other vowels remain unchanged.

(2) The proper part of the vowel shift. /æ ɑ/ → /ɛ a/, /æː ɑː/ → /ɛː ɔ:/, /ai oi/ → /aː wɛ/.

(You could also make /oi/ → /wɛ/ later still, as it looks like something that could have happened independently.)
The conlanger formerly known as “the conlanger formerly known as Pole, the”.

If we don't study the mistakes of the future we're doomed to repeat them for the first time.

Knit Tie
Lebom
Lebom
Posts: 101
Joined: Thu Oct 13, 2016 2:43 am
Location: Russia, Saint-Petersburg

Re: Sound Change Quickie Thread

Post by Knit Tie »

So with regards to my previously-mentioned phonemic inventory collapse idea, what do you say about this:

1)b d dʒ g → m n ɲ ŋ/_Ṽ

1.5)g → ɣ

2)w r j ɣ → m n ɲ ŋ/_Ṽ

3)b d dʒ → w r j/!_Ṽ
m n ɲ ŋ → w r j ɣ/!_Ṽ

Does this sound plausible to you? Wouldn"t it create too many morphemes of one series?

Also, I'd like to ask you about this fricative chain shift:

ts → s → ʃ → x

The /ʃ/ to /x/ change doesn't occur before /i/ but the rest of the changes do - is this feasible? And is it at all possible for a language to have /ʃ/ as its "default" sibilant, like what this sound shift is going to result in? What if then /z/ devoices unconditionally? Also, could somebody suggest something equally diachronically radical to do with the voiced fricatives?

User avatar
Pole, the
Smeric
Smeric
Posts: 1606
Joined: Sat Feb 11, 2012 9:50 am

Re: Sound Change Quickie Thread

Post by Pole, the »

[del]
Last edited by Pole, the on Thu Jun 15, 2017 3:44 pm, edited 1 time in total.
The conlanger formerly known as “the conlanger formerly known as Pole, the”.

If we don't study the mistakes of the future we're doomed to repeat them for the first time.

User avatar
Frislander
Avisaru
Avisaru
Posts: 836
Joined: Mon Feb 29, 2016 6:34 am
Location: The North

Re: Sound Change Quickie Thread

Post by Frislander »

Knit Tie wrote:ts → s → ʃ → x

The /ʃ/ to /x/ change doesn't occur before /i/ but the rest of the changes do - is this feasible? And is it at all possible for a language to have /ʃ/ as its "default" sibilant, like what this sound shift is going to result in?
It's perfectly reasonable as a series, and there are quite a few languages with /ʃ/ as their only sibilant, I say go for it.
https://frislander.tumblr.com/

First known on here as Karero

----
Smeric
Smeric
Posts: 1418
Joined: Tue Feb 15, 2011 11:15 pm

Re: Sound Change Quickie Thread

Post by ---- »

Could retroflexion become velarization/pharyngealization? I'm doing an African Romance conlang and I'm wondering whether I could go the Sardinian route and have retroflexes arise in certain places; like ll > ɖɖ etc. but then the tongue retraction could be emphasized into a retracted tongue root type thing too. But I don't know if this is attested or the association is as phonetically reasonable as it seems to me.

Vijay
Smeric
Smeric
Posts: 2244
Joined: Sat Feb 06, 2016 3:25 pm
Location: Austin, TX, USA

Re: Sound Change Quickie Thread

Post by Vijay »

thetha wrote:Could retroflexion become velarization/pharyngealization? I'm doing an African Romance conlang and I'm wondering whether I could go the Sardinian route and have retroflexes arise in certain places; like ll > ɖɖ etc. but then the tongue retraction could be emphasized into a retracted tongue root type thing too. But I don't know if this is attested or the association is as phonetically reasonable as it seems to me.
Idk about "retracted tongue root," but I can certainly think of a language where retroflexes merged with velars if that helps: Pashto. In [ˈpexəwəɾ], it's Pa[x]to, not Pa[ʂ]to. :) /ʐ/ also merged with /g/ in this variety of Pashto.

Lots of Hindi speakers have also at least partially merged /ʂ/ with /kʰ/.

User avatar
احمکي ارش-ھجن
Avisaru
Avisaru
Posts: 516
Joined: Mon Dec 02, 2013 12:45 pm

Re: Sound Change Quickie Thread

Post by احمکي ارش-ھجن »

How could /j/ could become /cç/ before back vowels?
ʾAšol ḵavad pulqam ʾifbižen lav ʾifšimeḻ lit maseḡrad lav lit n͛ubad. ʾUpulasim ṗal sa-panžun lav sa-ḥadṇ lav ṗal šarmaḵeš lit ʾaẏṭ waẏyadanun wižqanam.
- Article 1 of the Universal Declaration of Human Rights.

User avatar
Zaarin
Smeric
Smeric
Posts: 1136
Joined: Sun Aug 15, 2010 5:00 pm

Re: Sound Change Quickie Thread

Post by Zaarin »

احمکي ارش-ھجن wrote:How could /j/ could become /cç/ before back vowels?
/j/ > /cç/ isn't hard to imagine at all (simple chain of /j/ > /ʝ/ > /[c]ç/), but before back vowels is a little tricker--honestly I'd expect /kx/ in such an environment rather than simply devoicing and affricating.
"But if of ships I now should sing, what ship would come to me,
What ship would bear me ever back across so wide a Sea?”

User avatar
ˈd̪ʲɛ.gɔ kɾuˑl̪
Avisaru
Avisaru
Posts: 255
Joined: Wed May 18, 2016 11:11 pm
Location: Łódź

Re: Sound Change Quickie Thread

Post by ˈd̪ʲɛ.gɔ kɾuˑl̪ »

[j] :> [ɟʝ] :> [cç]
1. Fortition, which seems to me ok especially before back vowels.
2. Devoicing, which could happen in e.g. "Great Consonant Shift", but making it alone isn't that unusual, I think.
In Budapest:
- Hey mate, are you hung-a-ry?

Porphyrogenitos
Lebom
Lebom
Posts: 168
Joined: Thu Dec 22, 2011 1:13 pm
Location: Ohio

Re: Sound Change Quickie Thread

Post by Porphyrogenitos »

So I like the syllable stress and weight system of Wiyot, which is a bit unusual but has a certain appeal, and am trying to use it (or something similar) in a conlang:
Wiyot syllables always begin with consonants or consonant clusters, which are followed by a vowel. This vowel may be long or short. If the vowel is short, the syllable must end in the same consonant that begins the next syllable. Therefore, all non-final syllables are heavy, acquiring either a CVV or CVC structure. Word final syllables may or may not be heavy.

These syllable-final consonants are lengthened in speech, but do not appear as doubled letters in transcription. For example, in the word palógih, meaning 'flounder', the 'l' is lengthened. Thus, the first syllable ends with 'l', and the second begins with 'l', and both syllables are considered heavy.
My question is: What previous system could this have developed from?

I have an idea in mind - there's a sound change in Bolognese that lengthened all consonants after stressed short vowels, so combining that with word-internal deletion of short vowels, you'd get the rule that all non-final syllables have to be heavy. And that takes you most of the way there.

But does anyone have their own thoughts on developing a system like this?

User avatar
gach
Avisaru
Avisaru
Posts: 472
Joined: Mon Feb 17, 2003 11:03 am
Location: displaced from Helsinki

Re: Sound Change Quickie Thread

Post by gach »

You can simply invoke universal intervocalic gemination that's blocked by a preceding heavy syllable. No need for additional vowel deletion. The motivation for the change can be a simple pressure for levelling the lengths of the syllables. You don't necessarily even need special treatment for the last syllable. If the lengthening rules don't apply to it, it's automatically dropped out of the process.

On its own such a length levelling process could also lead to the lengthening of vowels on open syllables. It's thus a good idea to think what would be the reason for preferring intervocalic gemination over vowel lengthening. My view is that such a process is most likely to happen where it affects intelligibility the least. If the language already uses phonemic vowel length a fair bit, doing the extra lengthening on consonants will be less disruptive overall.

desmond
Lebom
Lebom
Posts: 94
Joined: Tue Sep 28, 2004 4:40 am
Location: Australia, Hong Kong
Contact:

Re: Sound Change Quickie Thread

Post by desmond »

A group of us from 人造語言迷 Conlinguaphiles are making a separate branch out of PIE. Are these rules too complicated for ProtoX which is on par with Proto-Italic, Proto-Germanic, Proto-Celtic? And are these rules plausible?

PIE/ProtoX/Environment

k/g/V_V
kʷ/ɡ͡b/V_V
ḱ/gʲ/V_V

ḱ/s/_E
gʲ/z/_E

k/q/#_O
k/k/#_
kʷ/k͡p/#_
ḱ/k/#_

k/x/_#
kʷ/xʷ/_#
ḱ/xʲ/_#

E=ie
O=uo
[b]Desmond[/b] - [url=http://conlang.wikia.com/Rangyayo][b]Rangyayo[/b][/url], spoken on the islands of elixir of life

User avatar
StrangerCoug
Avisaru
Avisaru
Posts: 269
Joined: Sat Aug 02, 2014 8:56 pm
Location: El Paso, TX

Re: Sound Change Quickie Thread

Post by StrangerCoug »

Other than the line where you have a sound changing to itself (k/k/#_), they're changes I buy.
Current avatar by malibupup of FurAffinity.

My conlangs on this site:
Proto-Wideriver

User avatar
Chengjiang
Avisaru
Avisaru
Posts: 437
Joined: Sun Feb 26, 2006 4:41 am
Location: Davis, CA

Re: Sound Change Quickie Thread

Post by Chengjiang »

What about fricatives undergoing fortition before [r], specifically [fr sr hr] > [pr tr kr]? I have fricative + rhotic clusters and I'd kind of like to get rid of them this way. [hr] > [kr] seems pretty natural to me, as does [sr] > [tr], but I'm less sure about [fr] > [pr] when [f] is unchanged in most environments.
[ʈʂʰɤŋtɕjɑŋ], or whatever you can comfortably pronounce that's close to that

Formerly known as Primordial Soup

Supporter of use of [ȶ ȡ ȵ ȴ] in transcription

It is pitch black. You are likely to be eaten by a 青.

User avatar
Soap
Smeric
Smeric
Posts: 1228
Joined: Sun Feb 16, 2003 2:57 pm
Location: Scattered disc
Contact:

Re: Sound Change Quickie Thread

Post by Soap »

Chengjiang wrote:What about fricatives undergoing fortition before [r], specifically [fr sr hr] > [pr tr kr]? I have fricative + rhotic clusters and I'd kind of like to get rid of them this way. [hr] > [kr] seems pretty natural to me, as does [sr] > [tr], but I'm less sure about [fr] > [pr] when [f] is unchanged in most environments.
Latin did something similar, turning */sr/ into /br/ in its "month of" ending (though this is not known for certain, it;s the standard assumption). If /sr/ > /br/ can happen, Im sure /fr/ > /br/ can also happen. We wouldnt know for sure from Latin because Latin never had medial /f/ to begin with, but it sounds plausible to me. And if there's /b/, there's /p/, and where there's /p/, there's /t/. Lastly I'd say /hr/ > /kr/ might seem the simplest, but I'd have to say I'm not sure about this one happening without a wider fortition affecting other /h/ sounds. Closest ananlog I can think of is /hʷ/ > /kv/ in Icelandic.
Sunàqʷa the Sea Lamprey says:
Image

User avatar
Nortaneous
Sumerul
Sumerul
Posts: 4544
Joined: Mon Apr 13, 2009 1:52 am
Location: the Imperial Corridor

Re: Sound Change Quickie Thread

Post by Nortaneous »

Latin had *sr > ðr, probably via *zr. ð > b in that position is regular; cf. *-dhrom > Proto-Italic *-ðrom > -brum, e.g. crībrum 'sieve' < *krey-dhrom, dēlūbrum 'temple' < *deh1-lewh3-dhrom (with characteristic Latin vowel reduction: the form of lav- 'wash' with a prefix is -lu-)

(for some reason, PIE has both *-dhrom and *-trom)
Siöö jandeng raiglin zåbei tandiüłåd;
nää džunnfin kukuch vklaivei sivei tåd.
Chei. Chei. Chei. Chei. Chei. Chei. Chei.

User avatar
Soap
Smeric
Smeric
Posts: 1228
Joined: Sun Feb 16, 2003 2:57 pm
Location: Scattered disc
Contact:

Re: Sound Change Quickie Thread

Post by Soap »

Nortaneous wrote:Latin had *sr > ðr, probably via *zr. ð > b in that position is regular; cf. *-dhrom > Proto-Italic *-ðrom > -brum, e.g. crībrum 'sieve' < *krey-dhrom, dēlūbrum 'temple' < *deh1-lewh3-dhrom (with characteristic Latin vowel reduction: the form of lav- 'wash' with a prefix is -lu-)

(for some reason, PIE has both *-dhrom and *-trom)
Wiktionary has -dʰrom, -trom, -dʰlom, -tlom as four variants of a single suffix, appearing with different reflexes in different languages, possibly hinting at an earlier l/r alternation and/or a dʰ/t alternation dependent on the phonetic shape of the root. Germanic retained the affix, yielding English words like lather ("wash-i-tron") and fodder ("feed-i-tron") as well as some words, perhaps vowel stems, where the medial consonant disappeared and gave words like ear, apparently an archaic term for "plow" unrelated to the also agricultural sense of an ear of corn.

I suspect the suffix may have originally only used /l/, but that there was some sort of change that breezed through the outer layer of IE languages which made all coronal + /l/ clusters into coronal + /r/, but somehow Latin retained a few forms with /l/ that had a secondary vowel in the middle.

The thing about medial PIE /dʰ/ > proto-Italic /ð/ is just a theory, I should add, since there are no known Italic or pre-Italic alphabets with a specific symbol for that sound. The initial reflex of that sound in Latin is /f/. Osco-Umbrian seems to have /f/ in medial position too, so perhaps the path was /dʰ/ > /ð/ > /v/ > /b/ in Latin but /dʰ/ > /ð/ > /v/ > /f/ in Osco-Umbrian. On the other hand, maybe the /f/ of Osco-Umbrian was actually [v] in some position, and they just didnt bother with a separate symbol.
Sunàqʷa the Sea Lamprey says:
Image

Post Reply