Phonetically motivated sound changes

A forum for all topics related to constructed languages
Post Reply
Khunjund
rupestrian
rupestrian
Posts: 23
Joined: 06 Feb 2015 03:02

Phonetically motivated sound changes

Post by Khunjund »

I was trying to come up with sound plausible sound changes for a conlang, in which I was hoping to include some typically non-European vowels, such as [ɨ], [ɯ], or [ɤ], but unlike most other vowels (like front rounded vowels for instance), I don’t feel confident enough to posit phonetically motivated shifts towards these vowels starting from, say, a standard five-vowel system /a i u e o/. I’ve tried looking up records of historical sound changes to get a better feel for them, but [ɯ] and co. aren’t that common, and the Index Diachronica for instance usually assumes such vowels are present in the language from the start. I guess I always have the option of just forcing the issue e.g. by /u/→/ɯ/ everywhere, and bringing /u/ back through other means, but I don’t find that as satisying.

The most plausible change I’ve come across for back unrounded vowels so far is rounding harmony, whereby /u/→/ɯ/ and /o/→/ɤ/, but that’s about the extent of my knowledge. It doesn’t help that the language I’m trying to have these back unrounded vowels in is mostly monosyllabic and isolating, so creating the environments for vowel harmony is trickier.

I picked up another potentially motivated change for /ɯ/ from Japanese, where some verb forms such as 問ひて tohite became 問うて to'ute, so /hi/→/hɯ/→/ɯ/, but I’m not that confident, since actual instances of this change are few and far between. In any case, still doesn’t give me much room to paly with.

As for [ɨ], I’m completely stumped. I tried looking into Romanian, but I can’t wrap my head around the shift from a→â; it seems like such a massive distance that it makes me think it should be part of a great vowel shift, only there is no such shift. As far as I can tell Proto-Slavic is reconstructed with [ɨ] from the beginning, so that’s not much help.

If any of you can think of phonetically motivated ways to derive these vowels from a standard five-vowel system, whether applicable in monosyllables or not, or if you simply know of any interesting or unexpected real sound changes, that would be a great help. Maybe I should just have my starting vowel set be /a i ɯ e o/ instead.

Thanks!
Fluent: :ca-qc: :eng: | Learning: :jpn: | Interested: :ara: :chn: :rus: :hkg: :ind: :por: :deu: :tha: :tur: :vie: :kor: :swe: :pol: :rou: :fin: :esp: :isr: :nga: :irl: :ita: :tib: :ell: :eth: etc.
User avatar
Creyeditor
MVP
MVP
Posts: 5091
Joined: 14 Aug 2012 19:32

Re: Phonetically motivated sound changes

Post by Creyeditor »

Starting from my vague knowledge of synchronic phonological processes, I can think of the following sources of back unrounded vowels.
  • Consonant-Vowel-Interaction: Rounding of back vowels is lost except adjacent to labial consonants. Afterwards, either new labials emerge independently adjacent to unrounded vowels or new non-labials emerge independently adjacent to rounded vowels. Sometimes coronal consonants do not take part in this. In this version of the pathway, unrounded vowels basically emerge adjecent to dorsal consonants.
  • Rounding/Backness harmony: Rounded back vowels become unrounded before/after a syllable with an unrounded vowel. Alternatively, this might be a change from a front unrounded vowel to a back unrounded vowel before/after a syllable with an back vowel. If you neutralize the trigger, you get a phonemic difference. Yay.
  • Revenge of central vowels: Sometimes (low) central vowels just don't feel like they want to stay there. Low back unrounded vowels may raise, creating [ɤ], or even [ɯ]. Maybe there are new central vowels and the old ones back up to give the newbies some space. These changes might be conditioned by a variety of context.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]
User avatar
Pabappa
greek
greek
Posts: 577
Joined: 18 Nov 2017 02:41

Re: Phonetically motivated sound changes

Post by Pabappa »

Beware of reconstructed Japanese sound shifts ... Im no expert, but Ive seen some supposed sound changes described that are much easier to explain as grammatical analogy or as variant readings of the same glyph. /hi/ > /hu/ looks really unlikely to me, unless it can be shown that it happened in all environments, not just in verbs with a preceding /o/. We know that in Old Japanese there were some inflections that ended in /u/ and others that ended in /i/, so i strongly suspect this change is due to analogy.

A common sound change in my languages is /u/ > /ɯ/ when not adjacent to a labialized consonant, but that's really only good if your language has a robust contrast between plain and labialized consonants, which isn't all that common.

I want to see more about the starting phonology .... you say it's monosyllabic, but that can still mean a lot of things .... how heavy are your syllables? If you regularly have syllables with clusters in both the onset and the coda, you can get your vowel alternations just from that.
Kavunupupis, šiŋuputata.
When I see you pointing at me, I know I'm in trouble. (Play)
Salmoneus
MVP
MVP
Posts: 3033
Joined: 19 Sep 2011 19:37

Re: Phonetically motivated sound changes

Post by Salmoneus »

Some options...

- unconditioned unrounding of (some) back vowels
- unrounding of back vowels conditioned by nearby vowels (vowel harmony/umlaut)
- unrounding of back vowels conditioned by nearby consonants. Either a distinction is lost (labiolisation is lost, for instance) or a whole consonant is lost (loss of first consonant in a cluster, for instance)
- unrounding of back vowels conditioned by absence of stress

- unconditioned backing of (some) front or central vowels
- backing of front or central vowels conditioned by nearby vowels
- backing of front or central vowels conditioned by nearby consonants
- backing of front or central vowels conditioned by absence of stress (unlikely in a single step, but, eg, unstressed front vowels could become central, could then become backed)

It's possibly for multiple conditionings (eg stress and nearby vowels)

It doesn't matter that your conlang is currently monosyllabic - it could have been polysyllabic in the past.


For high central vowels, your main options are:
- fronting of back vowels (this can result in derounding eventually, as central vowels are less likely to be rounded)
- backing of front vowels
- raising of low central vowels
Khunjund
rupestrian
rupestrian
Posts: 23
Joined: 06 Feb 2015 03:02

Re: Phonetically motivated sound changes

Post by Khunjund »

Creyeditor wrote: 21 Sep 2020 20:12 Consonant-Vowel-Interaction: Rounding of back vowels is lost except adjacent to labial consonants. Afterwards, either new labials emerge independently adjacent to unrounded vowels or new non-labials emerge independently adjacent to rounded vowels. Sometimes coronal consonants do not take part in this. In this version of the pathway, unrounded vowels basically emerge adjecent to dorsal consonants.
About that, wouldn’t it be the opposite in this case? As in, kept rounded except when adjacent to labial consonants? The reasoning is that labial consonants themselves tend to have the lips flat against each other, which works against rounding, and I feel like I’ve seen changes which involve clusters in [labial]+[w] simplifying to just [labial] because of it (e.g. /pwV/→/pV/).
Creyeditor wrote: 21 Sep 2020 20:12 Revenge of central vowels: Sometimes (low) central vowels just don't feel like they want to stay there. Low back unrounded vowels may raise, creating [ɤ], or even [ɯ]. Maybe there are new central vowels and the old ones back up to give the newbies some space. These changes might be conditioned by a variety of context.
Could you perhaps give me some specific examples of these changes in context? Either in a natlang or just how they would make sense to you if you were to use them in a conlang.
Pabappa wrote: 21 Sep 2020 20:14 I want to see more about the starting phonology .... you say it's monosyllabic, but that can still mean a lot of things .... how heavy are your syllables? If you regularly have syllables with clusters in both the onset and the coda, you can get your vowel alternations just from that.
The current inventory I’m on is:

<m n ŋ>
<p pʰ b t tʰ d k kʰ g ʔ>
<t͜s t͜sʰ d͜z t͜ʃ t͜ʃʰ d͜ʒ>
<ɸ s z ʃ x h>
<w j ɾ l>
<i u e o a>

To which I’m debating adding /ə/ as a sixth main vowel.

Syllables would take the shape (C)(r/l/w/j)V(C)(C), with final consonants restricted to a smaller subset. (Affricates are treated as single phonemes.) In the event of polysyllabic roots, I wouldn’t allow for more than two consonants intervocalically.
Salmoneus wrote: 21 Sep 2020 21:26 Some options...
Vowel environments I understand, and I suppose they could be brought on by function words. Unconditional changes I’m not so much a fan of, unless they’re brought on by larger vowel shifts. As for the consonant environments, that’s what’s troubling me: on my own I don’t really see any consonant environments causing that kind of change. Could you maybe give a few concrete examples of each, either from a natlang or just off the top of your head?
Fluent: :ca-qc: :eng: | Learning: :jpn: | Interested: :ara: :chn: :rus: :hkg: :ind: :por: :deu: :tha: :tur: :vie: :kor: :swe: :pol: :rou: :fin: :esp: :isr: :nga: :irl: :ita: :tib: :ell: :eth: etc.
User avatar
Pabappa
greek
greek
Posts: 577
Joined: 18 Nov 2017 02:41

Re: Phonetically motivated sound changes

Post by Pabappa »

Okay you could unround your /o u/ and then get new ones from other sources. I know you said you didnt want to do that, but I dont really see another path forward since your new central vowels have to come from somewhere and /e i/ > /ɤ ɯ/ would be difficult to explain.

Finnish did /nakra/ "laugh" > /naura/, so that's evidence that you can get a labial from something that isnt. It's not clear that there was any intermediate stage, either .... it may have just been a direct shift of /k/ > /w/. This particular shift may have been limited to just after /a/, but I dont think it necessarily has to be. Romanian, for example, did /kt/ > /pt/ unconditionally, so if you follow that path you have new labials from both original labials and original velars.

Then, turn these labial glides into diphthongs, and then monophthongize the diphthongs. You could also do the same from the other end .... e.g. /wa/ > /o/ .... though I think it's more common for the glide to survive in such a situation.
Kavunupupis, šiŋuputata.
When I see you pointing at me, I know I'm in trouble. (Play)
User avatar
Pabappa
greek
greek
Posts: 577
Joined: 18 Nov 2017 02:41

Re: Phonetically motivated sound changes

Post by Pabappa »

Khunjund wrote: 21 Sep 2020 22:10
Creyeditor wrote: 21 Sep 2020 20:12 Consonant-Vowel-Interaction: Rounding of back vowels is lost except adjacent to labial consonants. Afterwards, either new labials emerge independently adjacent to unrounded vowels or new non-labials emerge independently adjacent to rounded vowels. Sometimes coronal consonants do not take part in this. In this version of the pathway, unrounded vowels basically emerge adjecent to dorsal consonants.
About that, wouldn’t it be the opposite in this case? As in, kept rounded except when adjacent to labial consonants? The reasoning is that labial consonants themselves tend to have the lips flat against each other, which works against rounding, and I feel like I’ve seen changes which involve clusters in [labial]+[w] simplifying to just [labial] because of it (e.g. /pwV/→/pV/).
I wanted a separate post just for this ..... I dont really have hard evidence for this, but I would consider a shift like /pwu/ > /pu/ to be natural, as the /u/ itself causes the preceding /p/ to be labialized. You just may not think of it as such. Look at languages like Japanese, which palatalizes every consonant before /i/. Evidence exists that /u/ once behaved the same way. e.g. Japanese once had kana for consonants followed by /wa/, /we/, /wi/ .... (and possibly /wo/), but has never needed or used kana for /wu/. Why not? Because the regular /u/ was already phonetically [wu]. Japanese later lost this labial offglide except in the single syllable /wa/, which meant that /wa we wi wo/ shifted to /a e i o/ and /u/ (phonetically [wu]) shifted to its current value of /ɯ/. In other words, /wi/ > /i/ is analogous to /u/ > /ɯ/.

Thus, in a language that distinguishes labialization on consonants, whenever I see sequences like /mu bu pu/, i would assume they are phonetically [mʷu bʷu pʷu] unless there is good reason to believe otherwise. Consonants tend to take on the lip shape of the surrounding vowels .... not always (e.g. /š/ being rounded in English), but usually.


Having said all that, Im not sure labial consonants are significantly more likely to cause rounded vowels than other consonants .... it's probably the first POA I'd choose if I had to pick one, but I think velars are also a good candidate, since they tend to draw vowels further back in the mouth, and back vowels are more likely to be rounded. \

EDIT: Importantly, Finnish also shifted /t/ > /u/ when after any of /a o u/, showing that the POA of the disappearing consonant didnt play much of a part at all. See https://www.frathwiki.com/Finnish#pssv for details .... the labials did create /u/ in environments where the other consonants did not, which is a really good way to work asymmetry into the system, but it's plain that /t/ and /k/ also shifted to /u/ in the right environment.

EDIT2: Also, the â in Romanian seems to derive mostly from /e/ and /i/, not from /a/. e.g. pământ "earth" gets its first vowel from Latin /a/ and its second from Latin /e/. Though there is at least some that is from /a/, as the name of the country indicates.

Hope this helps.
Kavunupupis, šiŋuputata.
When I see you pointing at me, I know I'm in trouble. (Play)
Salmoneus
MVP
MVP
Posts: 3033
Joined: 19 Sep 2011 19:37

Re: Phonetically motivated sound changes

Post by Salmoneus »

Khunjund wrote: 21 Sep 2020 22:10
Creyeditor wrote: 21 Sep 2020 20:12 Consonant-Vowel-Interaction: Rounding of back vowels is lost except adjacent to labial consonants. Afterwards, either new labials emerge independently adjacent to unrounded vowels or new non-labials emerge independently adjacent to rounded vowels. Sometimes coronal consonants do not take part in this. In this version of the pathway, unrounded vowels basically emerge adjecent to dorsal consonants.
About that, wouldn’t it be the opposite in this case? As in, kept rounded except when adjacent to labial consonants? The reasoning is that labial consonants themselves tend to have the lips flat against each other, which works against rounding, and I feel like I’ve seen changes which involve clusters in [labial]+[w] simplifying to just [labial] because of it (e.g. /pwV/→/pV/).
There are two, contradictory processes possible. The most common is assimilation: vowels are rounded next to rounded consonants, and most labials are rounded.

The loss of /w/ after labials, particularly before rounded vowels, is due to dissimilation, which is less common. Similarly, labiovelars in PIE became plain velars in many languages adjacent to rounded vowels. Derounding vowels next to rounded consonants is certainly possible, but I think less likely than the opposite, as vowel roundedness is generally more robust than rounding on consonants.
Creyeditor wrote: 21 Sep 2020 20:12 Revenge of central vowels: Sometimes (low) central vowels just don't feel like they want to stay there. Low back unrounded vowels may raise, creating [ɤ], or even [ɯ]. Maybe there are new central vowels and the old ones back up to give the newbies some space. These changes might be conditioned by a variety of context.
Could you perhaps give me some specific examples of these changes in context? Either in a natlang or just how they would make sense to you if you were to use them in a conlang.
I meant to suggest this one too.

English is a repeated example of a low unrounded central vowel becoming a back vowel; generally, this has resulted in rounding, but it need not, particularly if there's already a rounded vowel providing a contrast at that position. English (outside of parts of the US) has an unrounded low back vowel (for me, father/bath/psalm/cart), and it would be possible for this to rise.

Vowel environments I understand, and I suppose they could be brought on by function words. Unconditional changes I’m not so much a fan of, unless they’re brought on by larger vowel shifts. As for the consonant environments, that’s what’s troubling me: on my own I don’t really see any consonant environments causing that kind of change. Could you maybe give a few concrete examples of each, either from a natlang or just off the top of your head?
Well, rounding is generally caused by rounded consonants, derounding by unrounded consonants, and backing by back consonants.

An example in English (outside of the North) is the foot-strut split. /U/ almost unconditionally derounded - except when the preceding consonant is a labial, labiodental or labiovelar and the following consonant is a rounded coronal consonant (so /tS/, /dZ/, /S/, /Z/ or /l/) (it also occassionally happened when only one of these two conditions was met, as in "put" (but not "putt")).

Hence, "but" has derounded /V/ (right preceding consonant, wrong following consonant), as does "gull" (wrong preceding consonant, right following consonant), but "bull" has rounded /U/ (both consonants match the rule).

For a more systematic example, look at Irish. There, most vowels have allophonic backing/fronting depending on the adjacent consonants - /I/ is said to have four different allophones, ranging from fully fronted to fully centralised. Usually, rounding is unaffected, but not always: in Dingle, long /a:/ following a velarised labial and preceding another velarised consonant is rounded.
User avatar
Creyeditor
MVP
MVP
Posts: 5091
Joined: 14 Aug 2012 19:32

Re: Phonetically motivated sound changes

Post by Creyeditor »

Khunjund wrote: 21 Sep 2020 22:10
Creyeditor wrote: 21 Sep 2020 20:12 Consonant-Vowel-Interaction: Rounding of back vowels is lost except adjacent to labial consonants. Afterwards, either new labials emerge independently adjacent to unrounded vowels or new non-labials emerge independently adjacent to rounded vowels. Sometimes coronal consonants do not take part in this. In this version of the pathway, unrounded vowels basically emerge adjecent to dorsal consonants.
[...] The reasoning is that labial consonants themselves tend to have the lips flat against each other, which works against rounding [...]
I think other people have commented enough on the other parts of this paragraph, so I just picked out this one. First of all, not all labials have the lips flat, IINM some languages/speakers have labials with closed protruded lips. Second, having your lips pressed against each other might even favor rounded vowels, because the articulatory gestures are somewhat similar and the articulatory distance is shorter from lips closed to lips protruded than from lips closed to lips spread. As Sal pointed out, opposite, i.e. dissimilatory, changes are possible, but rarer.
Khunjund wrote: 21 Sep 2020 22:10
Creyeditor wrote: 21 Sep 2020 20:12 Revenge of central vowels: Sometimes (low) central vowels just don't feel like they want to stay there. Low back unrounded vowels may raise, creating [ɤ], or even [ɯ]. Maybe there are new central vowels and the old ones back up to give the newbies some space. These changes might be conditioned by a variety of context.
Could you perhaps give me some specific examples of these changes in context? Either in a natlang or just how they would make sense to you if you were to use them in a conlang.
I will use made-up scenarios, but Moro (as described in Rose and Ritchard's work) might be an example.

Scenario 1: Let's say, you start with the following vowel inventory. It has two low vowels, which is more marked than having just one.
/i u/
/ɛ ɔ/
/a ɑ/
Ex: /sa/, /san/, /sah/ /sɑ/, /sɑn/, /sɑh/


The low back vowel may raise, becoming a mid back unrounded vowel. Maybe it raises uncoditionally, maybe it raises only in some environments and fronts in others. Let's say it raises in closed syllables and fronts in open syllables. Also, later on, coda-/h/ is deleted. You would get the following inventory and examples.
/i u/
/ɛ ʌ ɔ/
/a/
Ex: /sa/, /san/, /sah/ /sa/, /sʌn/, /sʌ/

Note that we have a contrastive back unrounded mid vowel now. I think this scenario is even more likely if you already have

Scenario 2:
You start with the following inventory.

/i u/
/e o/
/a /
Ex: /ˈsati/, /ˈpula/, /kiˈme/, /noˈru/

The next thing you do is vowel reduction in unstressed syllables. This creates a high central vowel from high vowels, and a mid central vowel from mid and low vowels.

/i ɨ u/
/e ə o/
/a/
Ex: /ˈsatɨ/, /ˈpulə/, /kɨˈme/, /nəˈru/

In the next phase, you start with a stress shift to the first syllable. And you want to do vowel reduction again, but the old central vowels remain distinct. They are pushed to the back by the new central vowels.

/i ɨ ɯ u/
/e ə ɤ o/
/a/
Ex: /ˈsatɯ/, /ˈpulɤ/, /ˈkɯmə/, /ˈnɤrɨ/

Now you get a contrast between central vowels and back unrounded vowels. Note however that such a contrast is rare among natlangs, so rare that some people have claimed it is non-existent.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]
User avatar
Vlürch
greek
greek
Posts: 452
Joined: 09 Mar 2016 21:19
Location: Finland
Contact:

Re: Phonetically motivated sound changes

Post by Vlürch »

Khunjund wrote: 21 Sep 2020 19:02I picked up another potentially motivated change for /ɯ/ from Japanese, where some verb forms such as 問ひて tohite became 問うて to'ute, so /hi/→/hɯ/→/ɯ/, but I’m not that confident, since actual instances of this change are few and far between. In any case, still doesn’t give me much room to paly with.
I skimmed through the replies but didn't notice anyone pointing out exactly what's up with that, but maybe someone did and I just missed it, in which case sorry for the repetition. What happened in Japanese is that it used to have /p/, which was lenited to /ɸ/ in all environments, which became voiced to /w/ intervocalically and then for whatever reason lost before all other vowels except /a/ like Pabappa already mentioned; word-initially, it became /h/, which remained [ɸ] before /u/.

There are still words with an intervocalic /h/, though, like やはり. Not sure what's up with them and can't find an explanation, but it's probably just that they didn't yet exist at the time of the shift or something like that?
Pabappa wrote: 21 Sep 2020 22:33Finnish did /nakra/ "laugh" > /naura/, so that's evidence that you can get a labial from something that isnt. It's not clear that there was any intermediate stage, either .... it may have just been a direct shift of /k/ > /w/.
There was an intermediate stage with /ɣ/, at least (which was almost certainly /g/ earlier), which became /w/ intervocalically before rounded vowels (and was then dropped) in some circumstances, so it could be that there was an intermediate stage where /ɣ/ was followed by a rounded vowel. It could be that /ɣ/ just shifted to /w/ in some circumstances, though, but there definitely was a /ɣ/ or something like that at one point.
clawgrip
MVP
MVP
Posts: 2257
Joined: 24 Jun 2012 07:33
Location: Tokyo

Re: Phonetically motivated sound changes

Post by clawgrip »

Japanese does palatalize consonants before /i/, and it also causes lip compression before /ɯ/, since /ɯ/ also has lip compression. This is why /hɯ/ becomes [ɸɯ], and why /wɯ/ can't occur, since /w/ is almost the same thing. It's also these features that allow for the standard dropping of /i/ and /ɯ/ between unvoiced consonants or at the end of a word, because their effect on the preceding consonant disambiguates.

I don't really know the reason that morpheme-medial /h/ is retained in certain words, but it is fairly rare: ahiru "domestic duck", afureru "overflow", yahari "as expected", haha "mother" are the only ones I can think of off the top of my head.

Of these four, haha is probably retained due to its reduplication, and yahari has a more informal version, yappari, that fits into the common (C)VC:Vri mimetic pattern, so yahari could likely be retention of an archaic form for more formal speech. The other two, I have no idea.
Khemehekis
mongolian
mongolian
Posts: 3885
Joined: 14 Aug 2010 09:36
Location: California über alles

Re: Phonetically motivated sound changes

Post by Khemehekis »

clawgrip wrote: 06 Oct 2020 15:08 I don't really know the reason that morpheme-medial /h/ is retained in certain words, but it is fairly rare: ahiru "domestic duck", afureru "overflow", yahari "as expected", haha "mother" are the only ones I can think of off the top of my head.
Does Mizuho count? Or maybe the name Iha? Ahegao?
♂♥♂♀

Squirrels chase koi . . . chase squirrels

My Kankonian-English dictionary: 87,413 words and counting

31,416: The number of the conlanging beast!
clawgrip
MVP
MVP
Posts: 2257
Joined: 24 Jun 2012 07:33
Location: Tokyo

Re: Phonetically motivated sound changes

Post by clawgrip »

Mizuho is actually mizu.ho (and the name Iha is surely i.ha), so the /h/ is morpheme initial, meaning it is not (necessarily) subject to the p > w/Ø sound change (but cf. Fujiwara < fuji.hara). This allows words like Yokohama, Chihiro, kaihatsu, tsunahiki, shifuku, taihen, ehon etc. to exist. And since this is a sound change rule, not a phonotactic rule, borrowed words, newly coined words, and memetic words are freely able to take morpheme-medial h, e.g. sofuto < "soft", Bauhausu < "Bauhaus", rihāsaru < "rehearsal", mohīto < "mojito" etc.

I am not familiar with the term uhegao, but a search for the word seems to turn up only anime results; it sounds to me like some sort of relatively newly coined mimetic word, meaning it postdates the completion of the sound change in question.
Khemehekis
mongolian
mongolian
Posts: 3885
Joined: 14 Aug 2010 09:36
Location: California über alles

Re: Phonetically motivated sound changes

Post by Khemehekis »

clawgrip wrote: 07 Oct 2020 07:17 Mizuho is actually mizu.ho (and the name Iha is surely i.ha), so the /h/ is morpheme initial, meaning it is not (necessarily) subject to the p > w/Ø sound change (but cf. Fujiwara < fuji.hara). This allows words like Yokohama, Chihiro, kaihatsu, tsunahiki, shifuku, taihen, ehon etc. to exist. And since this is a sound change rule, not a phonotactic rule, borrowed words, newly coined words, and memetic words are freely able to take morpheme-medial h, e.g. sofuto < "soft", Bauhausu < "Bauhaus", rihāsaru < "rehearsal", mohīto < "mojito" etc.
I see. I kind of suspected that.
I am not familiar with the term uhegao, but a search for the word seems to turn up only anime results; it sounds to me like some sort of relatively newly coined mimetic word, meaning it postdates the completion of the sound change in question.
Not uhegao, ahegao.

NSFW Warning: The following article is about a slightly pornographic cosplayer.

https://www.scmp.com/magazines/style/ce ... influencer
♂♥♂♀

Squirrels chase koi . . . chase squirrels

My Kankonian-English dictionary: 87,413 words and counting

31,416: The number of the conlanging beast!
clawgrip
MVP
MVP
Posts: 2257
Joined: 24 Jun 2012 07:33
Location: Tokyo

Re: Phonetically motivated sound changes

Post by clawgrip »

Sorry, uhegao was a typo; I did search for ahegao. It sounds like ahe is probably onomatopoeic for the sound the person makes when making whatever face that is (sorry, I really know very little about this anime subculture stuff).
Post Reply