Sound Change Quickie Thread

[WeepingElf]

This was an old attempt to derive something from PIE, and probably not at all in accordance with the laws of sound change, but what do you think about /p t ḱ k kʷ/ > /kʷ ḱ k h~ʔ k/ when next to /h₂/? (or even with the velars fricativized.)

A velarization effect of *h2 seems plausible to me. The labial stop may retain its labiality as labialization, but why should /kʷ/ lose its labialization? In my Hesperic family, I have a change I have named "Drummond's Law" (named after a fictional linguist), which changes labials and dentals to velars when followed by /x/ and to labiovelars when followed by /xʷ/, with the "laryngeal" disappearing.

[ObsequiousNewt]

This was an old attempt to derive something from PIE, and probably not at all in accordance with the laws of sound change, but what do you think about /p t ḱ k kʷ/ > /kʷ ḱ k h~ʔ k/ when next to /h₂/? (or even with the velars fricativized.)

A velarization effect of *h2 seems plausible to me. The labial stop may retain its labiality as labialization, but why should /kʷ/ lose its labialization? In my Hesperic family, I have a change I have named "Drummond's Law" (named after a fictional linguist), which changes labials and dentals to velars when followed by /x/ and to labiovelars when followed by /xʷ/, with the "laryngeal" disappearing.

I dunno, I guess that the "labio-" part of /kʷ/ was then overshadowed by the "velar". Maybe I'll keep it as /kʷ/.

Actually, I did fricativize them to /xʷ x̂ x h x⁽ʷ⁾/ (and h1 would fricativize without velarizing, to /φ θ x̂ x xʷ/), but I left that out of the original question to simplify matters.

Oh, and, in case you're interested, here's the full set of laryngeal changes:
h1
non-aspirated stops fricativised (as described above; also /(b) d ĝ g gʷ/ > /(β) ð γ̂ γ γʷ/)
aspirated stops de-aspirated
/l m n r/ remain constant and the laryngeal becomes /a/ rather than disappearing
/s hₓ/ debuccalised to /h/ (i.e. laryngeals become /h/ when there are two of them next to each other)
(semi)vowels lengthened: /y w/ > /i u/ > /ī ū/; /e o/ > /ē ō/
h2
non-aspirated stops fricativised and velarised (as described above; also /(b) d ĝ g gʷ/ > /(γʷ) γ̂ γ h γ⁽ʷ⁾/)
aspirated stops de-aspirated and velarised /bh dh ĝh gh gwh/ > /gʷ ĝ g h g⁽ʷ⁾/
/l m n r/ still constant and the laryngeal becomes /a/ rather than disappearing
/s hₓ/ debuccalised
all vowels lengthened as h1
all non-/a/ lowered /i ī ē ō u ū/ > /e ē ā ā o ō/

All laryngeals would then disappear, except when bordering /l m n r/, where they would instead become /a/. h3 had merged earlier into h2.

[WeepingElf]

Given that nobody knows what the laryngeals actually were, one cannot really say to which degree the changes you suggest are plausible. But they look quite sensible to me.

[ObsequiousNewt]

Given that nobody knows what the laryngeals actually were, one cannot really say to which degree the changes you suggest are plausible. But they look quite sensible to me.

Except that they affected only vowels in pretty much every other language. (And I'm going with something like /h/ for h₁ and /ʕ~ʜ/ for h₂.)

[Grunnen]

How realistic do you guys think would it be to change unvoiced consonants into vowels if they're between other consonants (voiced ones I'm assuming are okay), for example:

Lets say that C includes word boundaries here. What consonants it could realistically include is part of my question. I'm thinking of stops and probably fricatives

χ/a/C_C (example: χkønt -> akønt)
k/ɛ/C_C C!=labial (hatkt -> hatɛt)
k/ɔC_C C=labial (hapkt -> hapɔt)

What vowels would such consonants turn into? Labials into , alveolars into , uvulars into [a] and velars [a] as well, possibly slightly higher as in the examples? Or should I just delete these consonants altogether?

Assuming these clusters recently arose due to loss of reduced vowels.

[Cedh]

Turning plosives into vowels seems a bit of a stretch at first glance, but it should be doable if you lenite them to fricatives in the first step. All kinds of continuants (fricatives, nasals, liquids, semivowels) can rather easily become vocalised when between two other consonants. For instance:

p t c tʃ k q ɸ θ ç ʃ x χ / C_C
b d ɟ dʒ g ɢ β ð ʝ ʒ ɣ ʁ / C_C
ɸ~f β~v ʊ~ɔ / C_C
ç~ʃ ʝ~ʒ ɪ~ɛ / C_C
x~χ ɣ~ʁ ɐ / C_C
other fricatives ɨ~ə / C_C
followed by mergers of the newly-formed vowels with existing vowel phonemes, e.g. ɪ ʊ ɐ ɨ > i u a e

Most likely, (some of) these changes would only occur in specific situations, e.g. only between two consonants of equal or lower sonority than the to-be-vocalised segment, or only if none of the two surrounding consonants is a fricative or liquid, or something like that.

Alternatively, you could simply invoke vowel epenthesis in all clusters of three or more consonants, followed by simplification of the remaining clusters. For instance:

CCC CəCC
ə ʊ~ɔ / _[+labial]
ə ɪ~ɛ / _[+palatal]
ə a / elsewhere
C Ø / V_C (possibly with further conditions)

What vowels would such consonants turn into? Labials into , alveolars into , uvulars into [a] and velars [a] as well, possibly slightly higher as in the examples? Or should I just delete these consonants altogether?

Initially, the result of vocalisation would probably be a fairly lax and central vowel. Vowels resulting from labial consonants would tend to preserve lip rounding. Vowels resulting from uvulars and pharyngeals (and to a lesser extent velars) would tend to be more open. Vowels resulting from palatals (and to a lesser extent coronals) would tend to be pronounced further to the front of the mouth, and vowels resulting from retroflexes would tend to be rhotacised or at least pronounced further to the back of the mouth. Coronal consonants might also carry some velarisation or pharyngealisation in coda position (such as English "dark L"); in this case their vocalised reflex would also be backed.

It's up to you how much of these tendencies would become manifest. It's equally possible that all of the above groups merge into a single reduced/central vowel, or that some or all of them stay distinct from each other, possibly merging with various pre-existing vowel phonemes.

[Grunnen]

Turning plosives into vowels ... pre-existing vowel phonemes.

Thanks, that's a lot of interesting info!

[ObsequiousNewt]

How about /p ͡t/ > /t/ > /k/, in a situation where no velars already exist? (Or, if it works better, /p ͡t/ > /p/ > /k/)

[WeepingElf]

How about /p ͡t/ > /t/ > /k/, in a situation where no velars already exist? (Or, if it works better, /p ͡t/ > /p/ > /k/)

What other stops do you have? /p/ and /t/?

[ObsequiousNewt]

How about /p ͡t/ > /t/ > /k/, in a situation where no velars already exist? (Or, if it works better, /p ͡t/ > /p/ > /k/)

What other stops do you have? /p/ and /t/?

That's basically it. (This is Elmin I'm working with.)

[WeepingElf]

How about /p ͡t/ > /t/ > /k/, in a situation where no velars already exist? (Or, if it works better, /p ͡t/ > /p/ > /k/)

What other stops do you have? /p/ and /t/?

That's basically it. (This is Elmin I'm working with.)

In that case, I'd have /t/ go to /k/, and /pt/ to /t/, while /p/ stays put. Perhaps, in some environments (next to front vowels?), /t/ stays /t/, too, just to increase the frequency of occurrence of this common phoneme.

[ObsequiousNewt]

How about /p ͡t/ > /t/ > /k/, in a situation where no velars already exist? (Or, if it works better, /p ͡t/ > /p/ > /k/)

What other stops do you have? /p/ and /t/?

That's basically it. (This is Elmin I'm working with.)

In that case, I'd have /t/ go to /k/, and /pt/ to /t/, while /p/ stays put. Perhaps, in some environments (next to front vowels?), /t/ stays /t/, too, just to increase the frequency of occurrence of this common phoneme.

That's probably a good idea. Thanks!

[Hallow XIII]

I just remembered the labial-to-labial-velar conversation we had and I thought of

pʼ → kʼʷ

wouldn't that make sense?

[Click]

Is unconditional [l] → [j] possible?

[ObsequiousNewt]

Is unconditional [l] → [j] possible?

I don't see why not. Spansh palatalized /l/ in about half of its environments, and the step from /ʎ/ to /j/ is an easy one to make.

[8Deer]

Is unconditional [l] → [j] possible?

I think it should be. I'm fairly certain r > j happened in some parts of Mayan, so l > j makes sense.

[Pole, the]

Is unconditional [l] → [j] possible?

If the inventory is small, why not. I have done it once myself.

[ObsequiousNewt]

Is it at all plausible to have a series of palatals but no velars (except allophonically)? If so, how might it come about? Fronting of velars?

I in fact have one right now. It came about from a system with only labials and alveolars, and the alveolars became palatalized. Look at my Proto-Elmin thread if you'd like more details

Relatedly, does it make any sense to turn a nasalised vowel into a nasal consonant?

[8Deer]

Is it at all plausible to have a series of palatals but no velars (except allophonically)? If so, how might it come about? Fronting of velars?

I in fact have one right now. It came about from a system with only labials and alveolars, and the alveolars became palatalized. Look at my Proto-Elmin thread if you'd like more details

Relatedly, does it make any sense to turn a nasalised vowel into a nasal consonant?

õ > on and then deleting the vowel.

[ObsequiousNewt]

Is it at all plausible to have a series of palatals but no velars (except allophonically)? If so, how might it come about? Fronting of velars?

I in fact have one right now. It came about from a system with only labials and alveolars, and the alveolars became palatalized. Look at my Proto-Elmin thread if you'd like more details

Relatedly, does it make any sense to turn a nasalised vowel into a nasal consonant?

õ > on and then deleting the vowel.

Thanks; I wasn't sure if it was plausible to do... well, anything with a nasalized vowel, besides denasalizing it.

[8Deer]

Is it at all plausible to have a series of palatals but no velars (except allophonically)? If so, how might it come about? Fronting of velars?

I in fact have one right now. It came about from a system with only labials and alveolars, and the alveolars became palatalized. Look at my Proto-Elmin thread if you'd like more details

Relatedly, does it make any sense to turn a nasalised vowel into a nasal consonant?

õ > on and then deleting the vowel.

Thanks; I wasn't sure if it was plausible to do... well, anything with a nasalized vowel, besides denasalizing it.

TBH, I don't know if this has happened diachronically in any language, but it happens all the time with borrowings.

Another thing that often happens with nasal vowels is mergers.

[Znex]

Realistically speaking, how many sound changes would occur on average every 100 years or so? Is there some way of reckoning the temporal distance between an ancestor and its descendant generally?

[Herr Dunkel]

I kinda remember someone legitimate on the forum saying it's about one major change (such as the GVS) per century, and I've cited that figure several times over. Haven't seen actual literature cite it, though.

[WeepingElf]

I kinda remember someone legitimate on the forum saying it's about one major change (such as the GVS) per century, and I've cited that figure several times over. Haven't seen actual literature cite it, though.

I think the rate of sound change in languages is not constant, and periods of social upheaval seem to have an accelerating effect (at least, it looks like that in the history of English), but one major change per century isn't too far off the mark, I think.

[Herr Dunkel]

Yeah, I'd agree on that. Would certainly explain the sound system mutability difference between 1300s and modern English and the equivalent Icelandic, for example.