Relative frequency of /oi/ vs /eu/
Relative frequency of /oi/ vs /eu/
In my experience, /oi/ seems to be a more common diphthong cross-linguistically than /eu/. Every language I know of with /eu/ also has /oi/, but the reverse does not apply. Somali has /ai/, /au/, /ei/, /ou/, and /oi/, but not /eu/, and English has a similar situation to Somali. Basque shows /au/, /ai/, /ei/, /eu/, and /oi/, supporting sentence 2. To my knowledge, Japanese allows /oi/ but not /eu/.
Is there any reason for this distribution of diphthongs? Or do I only see this as a result of limited knowledge?
(I'd be interested in a dedicated treatment of diphthongs cross-linguistically, if any of you have 'em.)
Is there any reason for this distribution of diphthongs? Or do I only see this as a result of limited knowledge?
(I'd be interested in a dedicated treatment of diphthongs cross-linguistically, if any of you have 'em.)
-_-_Aftovota_-_-
- marconatrix
- Lebom
- Posts: 234
- Joined: Sun Apr 09, 2006 4:29 pm
- Location: Kernow
- Contact:
Re: Relative frequency of /oi/ vs /eu/
Some sounds are more 'volatile' (tend to change into other sounds faster) than others. Could it be that /ew/ has a shorter half-life than /oj/, so taking a snapshot over a number of languages you'd be likely to catch more processes passing through /oj/ than through /ew/ ?
Kyn nag ov den skentel pur ...
Re: Relative frequency of /oi/ vs /eu/
I have the same impression, and often hesitate to include /eu/ in a conlang unless it also has /oi/.Αυτοβοτα wrote:In my experience, /oi/ seems to be a more common diphthong cross-linguistically than /eu/.
A possible counterexample: Proto-Germanic had */eu/ but no */oi/. It did have */ōi/, but this diphthong was rare and it occurred only word-finally in some grammatical endings like fem. singular dative, while */eu/ was relatively common (class 2 strong verbs, words like *deupaz, *leuhtą, *þeubaz, etc.).
- KathTheDragon
- Smeric
- Posts: 2139
- Joined: Thu Apr 25, 2013 4:48 am
- Location: Brittania
Re: Relative frequency of /oi/ vs /eu/
Note that Proto-Germanic also had no /o/, so you'd expect an absence of /oi/.Benturi wrote:I have the same impression, and often hesitate to include /eu/ in a conlang unless it also has /oi/.Αυτοβοτα wrote:In my experience, /oi/ seems to be a more common diphthong cross-linguistically than /eu/.
A possible counterexample: Proto-Germanic had */eu/ but no */oi/. It did have */ōi/, but this diphthong was rare and it occurred only word-finally in some grammatical endings like fem. singular dative, while */eu/ was relatively common (class 2 strong verbs, words like *deupaz, *leuhtą, *þeubaz, etc.).
Re: Relative frequency of /oi/ vs /eu/
IIRC Japanese had /eu/ → /juː/.Αυτοβοτα wrote:In my experience, /oi/ seems to be a more common diphthong cross-linguistically than /eu/. Every language I know of with /eu/ also has /oi/, but the reverse does not apply. Somali has /ai/, /au/, /ei/, /ou/, and /oi/, but not /eu/, and English has a similar situation to Somali. Basque shows /au/, /ai/, /ei/, /eu/, and /oi/, supporting sentence 2. To my knowledge, Japanese allows /oi/ but not /eu/.
Maybe it has something to do with the fact that /i̯/ is more common than /u̯/?
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.
If we don't study the mistakes of the future we're doomed to repeat them for the first time.
Re: Relative frequency of /oi/ vs /eu/
Of course...KathAveara wrote:Note that Proto-Germanic also had no /o/, so you'd expect an absence of /oi/.
Maybe in some stage after a-mutation created /o/ there was /eu/ but no /oi/?
- KathTheDragon
- Smeric
- Posts: 2139
- Joined: Thu Apr 25, 2013 4:48 am
- Location: Brittania
Re: Relative frequency of /oi/ vs /eu/
Probably, but I don't think it lasted very long.Benturi wrote:Of course...KathAveara wrote:Note that Proto-Germanic also had no /o/, so you'd expect an absence of /oi/.
Maybe in some stage after a-mutation created /o/ there was /eu/ but no /oi/?
Re: Relative frequency of /oi/ vs /eu/
/eu/ → /joː/ actually.Pole, the wrote:IIRC Japanese had /eu/ → /juː/.Αυτοβοτα wrote:In my experience, /oi/ seems to be a more common diphthong cross-linguistically than /eu/. Every language I know of with /eu/ also has /oi/, but the reverse does not apply. Somali has /ai/, /au/, /ei/, /ou/, and /oi/, but not /eu/, and English has a similar situation to Somali. Basque shows /au/, /ai/, /ei/, /eu/, and /oi/, supporting sentence 2. To my knowledge, Japanese allows /oi/ but not /eu/.
Maybe it has something to do with the fact that /i̯/ is more common than /u̯/?
…and that lead me to Middle Chinese, where the bulk of /eu/ were from:
http://en.wikipedia.org/wiki/Middle_Chinese_finals
And did I happen to hit the bull's eye? Every reconstruction contains at least one /eu/ or /ɛu/, but only in Baxter's reconstruction can /oj/ be found, and it's missing in everyone else's works.
Re: Relative frequency of /oi/ vs /eu/
I've actually heard both of these sound changes attributed to Japanese in different places, so I'm confused and wondering if someone familiar with the phonological history can explain.M Mira wrote:/eu/ → /joː/ actually.Pole, the wrote:IIRC Japanese had /eu/ → /juː/.Αυτοβοτα wrote:In my experience, /oi/ seems to be a more common diphthong cross-linguistically than /eu/. Every language I know of with /eu/ also has /oi/, but the reverse does not apply. Somali has /ai/, /au/, /ei/, /ou/, and /oi/, but not /eu/, and English has a similar situation to Somali. Basque shows /au/, /ai/, /ei/, /eu/, and /oi/, supporting sentence 2. To my knowledge, Japanese allows /oi/ but not /eu/.
Maybe it has something to do with the fact that /i̯/ is more common than /u̯/?
Over on the CBB, clawgrip, who seems to know a lot about Japanese, says:
On the other hand, Wikipedia apparently has a source for the statement that:clawgrip wrote:Some facts about Japanese vowels:
1. The sequences iu, eu, io, eo, resulting from loss of /w/ (from earlier /f/) become in /ju: ju: jo: jo:/ respectively. The front vowel is reduced to a semivowel, and the back vowel is lengthened so that the sequence retains a length of two morae. The two vowels remain distinct and identifiable elements, so it's not exactly a vowel merger. The sequences au ou and usually ei, on the other hand, simply become /o: o: e:/, but only within a root, not in grammatical endings (so the verb 会う au from afu does not become ō, for example.
Is one of these sources simply wrong, or are both changes attested (perhaps in different contexts, such as sequences within a root vs. sequences derived from the addition of grammatical endings)?The vowel sequence /au/ contracted into [ɔː], while /ou/ and /eu/ contracted into [oː] and [joː], respectively.
- Nortaneous
- Sumerul
- Posts: 4544
- Joined: Mon Apr 13, 2009 1:52 am
- Location: the Imperial Corridor
Re: Relative frequency of /oi/ vs /eu/
Dananshan Hmong. Only counterexample I can think of. Might be more in that area.
Siöö jandeng raiglin zåbei tandiüłåd;
nää džunnfin kukuch vklaivei sivei tåd.
Chei. Chei. Chei. Chei. Chei. Chei. Chei.
nää džunnfin kukuch vklaivei sivei tåd.
Chei. Chei. Chei. Chei. Chei. Chei. Chei.
Re: Relative frequency of /oi/ vs /eu/
Wikipedia is correct. Look at historical orthography, e.g. 〜ませう for 〜ましょう.Sumelic wrote:Is one of these sources simply wrong, or are both changes attested (perhaps in different contexts, such as sequences within a root vs. sequences derived from the addition of grammatical endings)?
書不盡言、言不盡意
-
- Sanci
- Posts: 27
- Joined: Sun Sep 25, 2011 12:14 pm
Re: Relative frequency of /oi/ vs /eu/
My (likely wrong) intuition is that diphthongs that go from back → front are more common and stable than front → back. So [au] is more common than [æu].
Both are correct. Clawgrip is talking about where some /au/, /eu/, and /ou/ came from, while WIkipedia just notes the later soundchanges.Sumelic wrote:Is one of these sources simply wrong, or are both changes attested (perhaps in different contexts, such as sequences within a root vs. sequences derived from the addition of grammatical endings)?
Re: Relative frequency of /oi/ vs /eu/
Peculiarly, Wikipedia describes the diphthong in question as [eβ] rather than [eu]. If there's any reality to this analysis, it would seem to provide further evidence of [eu] being unstable or otherwise unusual.suoenatroN wrote:Dananshan Hmong. Only counterexample I can think of. Might be more in that area.