Kokhene Kool Ideas

[VaptuantaDoi]

Inspired by Crey's nifty Fredauon Fun Facts thread, I'm making one of my own (luckily I realised in time that copying his alliteration wouldn't be the best idea), with the caveat that it's gonna mostly be language-related rather than conworld-related, and it's gonna be nebulous ideas rather than interesting titbits.

---

Kokhene Kool Idea №1:

First up, I've got an idea for Vissard re nasal vowels. Old Vissard existed alongside a Celtic language and a pre-Indo-European isolate (possibly distantly related to Vasconic) which had a phonemic inventory something like this:

/tʰ kʰ kʷʰ/
/p t ʈʂ k kʷ/
/b d ɖʐ/
/s ʂ (h)/
/r ɽ/

/i u æ ɒ/
/ĩ ũ æ̃ ɒ̃/

Voiced stops nasalised before nasal vowels or in coda position. This influenced the Celtic language which had been there a bit before the Romans arrived, and Vissard-Celtic ended up allophonically nasalising vowels following nasal consonants. The Romans thought this was just the bee's knees so they started progressively nasalising their vowels too when they got settled in there. In Old Vissard progressive nasalisation was complete:

MARE → /ˈmãr/
NĀSCŌ → /ˈnãskə/
AGNELLUS → /aˈɲɛ̃lt͡s/

Also happening at the time was regressive nasalisation with the loss of coda nasals:

FŪMUM → /ˈfỹ/
CANEM → /ˈkã/
PUGNUM → /ˈpõĩ̯/

Nasalisation reduced the seven-vowel system /i y e o ɛ ɔ a/ to a four-vowel system /ɛ̃ ɔ̃ ã ɑ̃/ a little bit later, giving /ˈmɑ̃r/, /ˈnɑ̃skə/, /aˈɲãlt͡s/, /ˈfɔ̃/, /ˈkɑ̃/, /ˈpɔ̃j/. Later still, influence from the still-mostly-mutually-intelligible Old French/Anglo-Norman meant that progressive nasalisation was stigmatised and was effectively reversed throughout the islands. The denasalised vowels ended up being longer than the nasal ones:

MARE → /ˈmãr/ → /ˈmɑːr/
NĀSCŌ → /ˈnãskə/ → /ˈnɛːk/
AGNELLUS → /aˈɲɛ̃lt͡s/ → /ˈɲaːjs/

However, this also affected words which had had a nasal both before and after the vowel:

MANDUCŌ → /ˈmɑ̃dʒə/ → /ˈmɑːtʃ/
NOMEN → /ˈnɔ̃/ → /ˈnɔː/
NĀNUM → /ˈnɑ̃/ → /ˈnɑː/

Nasal vowels were also denasalised before a nasal consonant, producing marginally phonemic long vowels:

HOMINEM → /ˈɔ̃nə/ → /ˈɔːn/
BONA → /ˈbɔnə/ → /ˈbɔn/

Some recent borrowings have also produced short post-nasal vowels:

English knobhead → /noˈbɛd/
French art nouveau → /ərnuˈvo/

So Modern Vissard has the following vowel inventory:

/i iː y yː u uː/
/e eː øː o oː/
/ɛ ɛ̃ ɛː ɔː/
/a aː ɑː ɒ̃/

/ɔː aː ɑː/ only arise from denasalisation, while /iː yː uː eː oː/ come from loss of /s/ and /ɛː/ comes from either.

[VaptuantaDoi]

Kokhene Kool Idea №2:

For the Click languages, there's still gonna be a major divide of "grasslands" vs. "northern". There's several stages of Grassland Clickness spreading out eastwards.

All GC languages metathesise clicks into the foot onset:

*pɛŋ̊ǁˀɛ → PGC *ɡǁk’ʷɛ → GC1 /ɡǁk’œː/
*dekǃʰo → PGC *ɡǃkxo → GC1 /ɡkǃoː/

The eastern 80% of GC languages also lenide oral C2 s with compensatory length and/or nasalisation:

*ŋǂeβeβɛ → PGC *ŋǂeβeβɛ → GC2 /ǂˀeːː/
*ˈpáβɛ́mù ˈkǀʰɛ́mù → GC2 /pãː kǀʰœ̃/

The eastern 50% also de-click the non-dental clicks:

*pɛŋ̊ǁˀɛ → PGC *ɡǁk’ʷɛ → GC3 /dt͡ɬ’ʷɛ/
*dekǃʰo → PGC *ɡǃkxo → GC3 /ɡk͡xə/

The eastern 25% de-click all of their clicks, making them some of the worlds only Click languages which aren't click languages:

*ˈkǀʰɛ́mɛmu → GC4 /tʰɛːm̩/

The eastern 10% also smooth out all the voicing contours (with compensatory tone) and give themselves a vertical vowel system as a reward:

*pɛŋ̊ǁˀɛ → PGC *ɡǁk’ʷɛ → GC5 /lʷɜ̂j/
*dekǃʰo → PGC *ɡǃkxo → GC5 /ɣʷɘw/
*ˈkǀʰɛ́mɛmu → GC5 /sʲɜɲ/

Then finally, the easternmost dialect chain of all does away with the vertical vowel system and turns into an entirely normal-looking language with no clicks, no vertical vowel systems and no trochees:

*pɛŋ̊ǁˀɛ → PGC *ɡǁk’ʷɛ → GC6 /leso/
*dekǃʰo → PGC *ɡǃkxo → GC6 /wovo/
*ˈkǀʰɛ́mɛmu → GC6 /híni/

In fact this language is probably phonologically indistinguishable from the surrounding eastern Nomadic-Bechsukchwan languages.

Northern Click on the other hand just gets lots more oral and click consonants and keeps the feet (weirdo).

[VaptuantaDoi]

Kokhene Kool Idea №3:

There's an island called Ąkuyook off the coast of Kokhene's southern polar continent which has two not-super-closely related languages on it, which I've previously called Ǫneebǫwhítw and Icááwənun. My original idea for proto-Ąkuyookic was something like this:

​ ​ ​ ​ ​​ t ʒ k
​ ​ ​ ​ ​​ 3 s x
​ ​ ​ ​ ​​ n y w

​ ​ ​ ​ ​​ i e a
​ ​ ​ ​ ​​ i꞉ e꞉ a꞉ (o꞉)

(ʒ 3 y were probably /dz s j/ or if we're being less neat then /dʒ l j/.) I do quite like the idea of a miniature family like this where I can do cool sound changes without getting bogged down and giving up (something which has only happened in 100% of my previous conlang family attempts). Proto-Ąkuyookic 2.0 is much more Cadoan-inspired:

​ ​ ​ ​ ​​ t k ʔ
​ ​ ​ ​ ​​ s h
​ ​ ​ ​ ​​ ɾ j w

​ ​ ​ ​ ​​ i o a
​ ​ ​ ​ ​​ iː oː aː

Syllable structure was CCCVC, where consonant clusters had to increase in sonority /t k ʔ/ → /s/ → /h/ → /ɾ/ → /j w/ (e.g. /ksh, khɾ, sɾj, hɾw/ etc.). Consonant clusters were fairly rare however. Gemination of single-consonant onsets was also permitted, although this shortened any preceding long vowel. Initial /ɾ w/ were [d b]. /t k/ were [t͡s ks] before another consonant or /i/; /ɾ/ became /t/ before a consonant other than /j w/; /j w/ were probably /kj kw/ following /s h ʔ/. /ɾi wi/ may have been [di bi] too (haven't decided this bit yet). Stress was initial and caused shortening of long vowels. Also prosodic was high tone, which caused lengthening of long vowels and consonants (i.e. V́ːCV V́CːV → V́ːːCV V́CːːV).

[VaptuantaDoi]

Kokhene Kool Idea №4:

So Proto-Ąkuyookic has the following sonority hierarchy:

​ ​ ​ ​ ​​ t k ʔ → s → h → ɾ → j w

To make a cluster, you just take any one, two or three in order, and you get the following underlying forms:

Code: Select all

t k ʔ s h ɾ j w

ts th tɾ tj tw
ks kh kɾ kj kw
ʔs ʔh ʔɾ ʔj ʔw
sh sɾ sj sw
hɾ hj hw
ɾj ɾw

tsh tsɾ tsj tsw thɾ thj thw tɾj tɾw
ksh ksɾ ksj ksw khɾ khj khw kɾj kɾw
ʔsh ʔsɾ ʔsj ʔsw ʔhɾ ʔhj ʔhw ʔɾj ʔɾw
shɾ shj shw sɾj sɾw
hɾj hɾw

However, a large number of allophonic rules have to apply to get the surface forms. Firstly, we have affrication of t k to ts ks. Specifically, sequences of |t(s)j k(s)j| surface as /ts ks/, |ti ki| surface as /tsi ksi/ and |tC kC| surface as /tsC ksC/ (except |ts ks| → /ts ks/). This means the clusters

​ ​ ​ ​ ​​ |ts th tɾ tj tw| → /ts tsh tsɾ ts tsw/
​ ​ ​ ​ ​​ |tsh tsɾ tsj tsw thɾ thj thw tɾj tɾw| → /tsh tsɾ ts tsw tshɾ tshj tshw tsɾj tsɾw/
​ ​ ​ ​ ​​ |ks kh kɾ kj kw| → /ks ksh ksɾ ks ksw/
​ ​ ​ ​ ​​ |ksh ksɾ ksj ksw khɾ khj khw kɾj kɾw| → /ksh ksɾ ks ksw kshɾ kshj kshw ksɾj ksɾw/

There is also fortition. |ʔs ʔɾ ʔw| become |ʔts t kw|, while |hɾ hw sɾ sw| become |ht hkw st skw|.

​ ​ ​ ​ ​​ |ʔs ʔh ʔɾ ʔj ʔw| → /ʔts ʔh t ʔj kw/
​ ​ ​ ​ ​​ |sh sɾ sj sw| → /sh st sj skw/
​ ​ ​ ​ ​​ |hɾ hj hw| → /ht hj hkw/
​ ​ ​ ​ ​​ |ʔsh ʔsɾ ʔsj ʔsw ʔhɾ ʔhj ʔhw ʔɾj ʔɾw| → /ʔtsh ʔtst ʔtsj ʔtskw ʔht ʔhj ʔhkw tj tw/
​ ​ ​ ​ ​​ |shɾ shj shw sɾj sɾw| → /sht shj shkw stj skw/
​ ​ ​ ​ ​​ |hɾj hɾw| → /htj htw/
​ ​ ​ ​ ​​ /tsɾ tsw/ → /tst tskw/
​ ​ ​ ​ ​​ /tshɾ tshw tsɾj tsɾw/ → /tsht tshkw tstj tstw/
​ ​ ​ ​ ​​ /ksɾ ksw/ → /kst kskw/
​ ​ ​ ​ ​​ /ksɾ ksw kshɾ kshw ksɾw/ → /kst kskw ksht kshkw kstw/

Finally we have a rule which deletes /h/ between two obstruents:

​ ​ ​ ​ ​​ /ʔht ʔhkw sht shkw tshkw kshkw/ → /ʔt ʔkw st skw tskw kskw/

To see the surface clusters, initial /ɾ b/ surface as [d b], written d b – obviously in word-internal clusters they'd still be r w though. This gives us:

Code: Select all

t k ʔ s h d y b

ts tsh tst ts tskw
ks ksh kst ks kskw
ˀts ˀh t ˀy tw
sh st sy skw
ht hy hkw
dy dw

tsh tst tsy tskw tst tshy tskw tsty tstw
ksh kst ksy kskw kst kshy kskw ksty kstw
ˀtsh ˀtst ˀtsy ˀtskw ˀt ˀhy ˀkw ty tw
st shy skw sty stw
hty htw

Or in a more surface-form arrangement:

Code: Select all

t k ˀ s h d y b

ts ty tw
ks
ˀt ˀh ˀy
sh st sy 
ht hy
dy dw

tst tsh tsy
kst ksh ksy
ˀts ˀkw ˀhy
sty stw skw shy
hty htw hkw

tsty tstw tskw tshy
ksty kstw kskw kshy
ˀtst ˀtsh ˀtsy

ˀtskw

Maybe not all 42 clusters appear in monomorphemic forms. All of these rules were productive morphologically, so you might have something like:

*kasó꞉ˀs- + *bánnniskót → *kasó꞉ˀtskwanniskót

or in a modern language

kasə́wʔ- + máantskəs → kasə́wt’panskás

[VaptuantaDoi]

Kokhene Kool Idea №5:

Wåpwopwåhisi

Like fifty years ago I posted a little Micronesian inventory:

/pʲ pˠ tʲ k/
/bʲ bˠ dʲ ɡ/
/mʲ mˠ/
/ɸʲ ɸˠ ɕ x/
/βʲ~ɥ βˠ~w ʑ~j ɣ~ɰ/

/ə/ [i~ə]
/ɐ/ [a~ɤ]

I randomly found it the other day and decided to beat some life back into it. Here's the revised version:

Code: Select all

 pʲ   pʷ   tʲ   k    ʔ
 bʲ   bʷ
 mʲ   mʷ   nʲ   ŋ
 βʲ   w    j         h
 
 ɨ
 ə
 ä

The second labial series should really be /pˠʷ bˠʷ mˠʷ/ but code doesn't do ˠs. Basically the diachronics would be from *pʲ pˠ tʲ t kʲ k → pʲ pˠʷ tʲ k tʲ ʔ etc. I don't know what happened to the voiced coronal stops; maybe there never were any, *vʲ vˠ hardened to /bʲ bˠʷ/ and /βʲ w/ arise from old glides.

Vowels are coloured by consonants according to the following rules:

- All labial(ised) consonants (/pʲ pʷ bʲ bʷ mʲ mʷ βʲ w/) are [+round]
- All palatal(ised) consonants (/pʲ tʲ bʲ mʲ nʲ βʲ j/) are [+front]
- All other consonants (/k ŋ ʔ h/) and word boundaries are neutral
- Features spread through vowel sequences indefinitely

Surface vowels are determined as below (pʲ = /pʲ bʲ mʲ βʲ/ etc.)

- V /ɨ ə ä/ → V[+round][+front] [y ø œ] / pʲ_, _pʲ, tʲ_pʷ, pʷ_tʲ
- V /ɨ ə ä/ → V[+round][–front] [​u o ɔ] / pʷ_pʷ, k_pʷ, pʷ_k
- V /ɨ ə ä/ → V[–round][+front] [i e ɛ] / tʲ_tʲ, k_tʲ, tʲ_k
- V /ɨ ə ä/ → V[–round][–front] [ɨ ə ä] / k_k

Word-initially however, lingual consonants (/tʲ k nʲ ŋ j/) are debuccalised:

- tʲ k → ʔ / #_
- j → h / #_
- nʲV ŋV → Vː / #_

Hence [i y ɨ~u e ø ə~o ɛ œ ä~ɔ] appear phonemic long word-initally, or word-initially following [ʔ] or [h]. [u o ɔ] only appear before /pʷ bʷ mʷ/ where they'd be realised as such anyway. Also coda /j w ŋ/ are often elided with compensatory length and nasalisation, which also makes extralong vowels, although not extralong nasal vowels (i.e. /ŋəŋ/ → [ə̃ː] not *[ə̃ːː], but /ŋəw nʲəj/ → [oːː iːː])

Orthography:

/pʲ pʷ tʲ [​s] k ʔ/ ⟨p pw t s k ʻ⟩
/bʲ bʷ/ ⟨b bw⟩
/mʲ mʷ nʲ ŋ/ ⟨m mw n g⟩
/βʲ w j h/ ⟨v w y h⟩

[​i y ɨ u] ⟨i i u u⟩
[​e ø ə o] ⟨é é o o⟩
[ɛ œ ä ɔ] ⟨e e a å⟩

[​iː yː ɨː uː] ⟨ii iu ui uu⟩
[​eː øː əː oː] ⟨ée éo oe oo⟩
[ɛː œː äː ɔː] ⟨ee eo aa åå⟩
(these digraphs used for short vowels following word-initial [ʔ h])

[​iːː yːː ɨːː uːː] ⟨iii iiu uii uuu⟩
[​eːː øːː əːː oːː] ⟨éee éeo oee ooo⟩
[ɛːː œːː äːː ɔːː] ⟨eee eeo aaa ååå⟩
(these trigraphs used for long vowels following word-initial [ʔ h])

[Ṽː] ⟨iiñ⟩ etc.


Syllable structure is probably CVC; maybe codas are only allowed word-finally or in compounds.

/wɨmʲəŋäw/ → [wymʲiŋɔː] wimigåå
/wɨmʲəŋäw-əwɨʔə/ → [wymʲiŋɔwowuʔə] wimigåwowuʻo
/wäpʷäʔəbʲ/ → [wɔpʷɔʔøbʲ] wåpwåʻéb
/pʷäbʲəwɨpʷəbʲ/ → [pʷœbʲœwupʷøbʲ] pwebewupéb
/wäʔämʲä/ → [wɔʔœmʲœ] wåʻeme
/ŋɨwɨ/ → [uːwu] uuwu
/wä-ŋɨwɨ/ → [wɔŋuwu] wåguwu
/ʔəŋ-ʔəmʲ/ → [ʔə̃ːʔømʲ] ʻoeñgʻém
/pʷäbʲəŋ/ → [pʷœbʲø̃ː] pwebéoñ
/wəpʷəpʷɨj/ → [wopʷopʷyː] wopwopwiu
/käbʷə/ → [ʔɔbʷɔ] ʻååbwå
/tʲäbʷə/ → [ʔœbʷɔ] ʻeobwå
/wä-käbʷə/ → [wɔkɔbʷɔ] wåkåbwå
/wä-tʲäbʷə/ → [wœtʲœbʷɔ] wetebwå
/pʷəpʷähɨtʲɨ/ → [pʷopʷɔhisi] pwopwåhisi

Here's a challenge, see if you can work out the phonemic transcription of this text:

Wåwu obomiwu wo mepwå. Uuuve ʻo ʻååpwå åpwåå wébpwu wuwåwåh uuwuåʻuwåm. Bépwuʻ pwupwååñ benéʻa wobwuʻa ui ʻåbwåmibe. Wåwopw muiñ méwu ʻeopw pwuo uiʻ. Owowu hiiʻiyepw bwébi béwåpw oeu bwenepw bewå.