zompist bboard

I'm not sure yet. Hathe, Araso and Arve definitely don't have clicks, but I might add them back to any or all of the others.

_________________
Siöö jandeng raiglin zåbei tandiüłåd;
nää džunnfin kukuch vklaivei sivei tåd.
Chei. Chei. Chei. Chei. Chei. Chei. Chei.

Aw fuck it, if everyone else is doing one of these I may as well too.

2013-2150
ɨ > ə EXCEPT _#, henceforth written as <i>
o u > ɛu yu / EXCEPT _l, _${o u}, and in 'both'
hj > ç
yu > y / _C
t > ts / _{ɪ i y(u)}
ɛ > ɪ / _N
æ ɛ > e / _ŋ
a ʌ > ɔ / _l
l > 0 / {ɔ o u}_
l > w / V_C
ə > 0 / #P_$V[+stress]
ar er or > ʌr ir ur, er > ir doesn't apply across morpheme boundaries
æ ɔ > eə oə > iə uə
ai au aiə auə > a æ a æ
ɾ > 0 / V_V EXCEPT V_Vn (where it's realized as a glottal stop anyway)
nt nd > ɾ̃ / V_V
nt nd > ~ʔ ~ / V_{C #}
ʔ > 0 > _C
r > ɾ / θ ð
ə > 0 / V_r
m̩ n̩ l̩ > ə̃ ĩ u
ɪ > i / _{n ŋ}
VN > Ṽ / _C
d > 0 / _#
cluster reduction and glide simplification, todo

All human beings are born free and equal in dignity and rights.They are endowed with reason and conscience and should act towards one another in a spirit of brotherhood.
/uə çymĩ bĩz ɚ burn fri ə̃ɾ̃ ikwu ĩ dɪnəɾi ə̃ rʌi̯ts. ðer ĩdæ wɪθ rizĩ ə̃ kɑ̃ʃĩs ə̃ ʃʊɾ eək turz wʌn ənʌðɚ ɪn ə spirəɾ ʌ brʌðɚhʊ/

2100-2500
ð > v / V_V
θ ð > f d (don't know if these are right)
w > 0 / _{o u}[-stress]
tr dr str ç > tʃ dʒ ʃ ʃ
t > 0 / _{s ʃ}
p k > ʔ / _#
t > 0 / _F#
d > 0 / _ʒ#
ɪ ʊ ɚ > ɛ ɨ ɨ
w > 0 / _u
i iə uə > ʌi i u (nasals aren't affected)
ĩ > ə̃ in unstressed syllables

/u də ʃymə̃ bĩz bʌi burn frʌi ə̃ɾ̃ iku ĩ dɛnəɾi ə̃ rʌis. de gɑʔ ə̃dæ wɛ rizĩ ə̃ kɑ̃ʃĩs ə̃ ʃʊ iʔ tur wə̃ nʌvɨ ɪn ə spirəʔ ʌ brʌvɨhɨ/

finish later, need to get actual work done. should probably plan on short-term hispanic influence and long-term black influence -- or would white dialects absorb hispanic innovations but remain distinct from black dialects? the copout answer is no since this is set in the south, and the copout answer makes things a lot easier so i will go with that

_________________
Siöö jandeng raiglin zåbei tandiüłåd;
nää džunnfin kukuch vklaivei sivei tåd.
Chei. Chei. Chei. Chei. Chei. Chei. Chei.

An attempt at recording the Tower of Babel translation in V'eng: http://vocaroo.com/i/s05wxZLAn3T0

Prosody is hard.

_________________
Siöö jandeng raiglin zåbei tandiüłåd;
nää džunnfin kukuch vklaivei sivei tåd.
Chei. Chei. Chei. Chei. Chei. Chei. Chei.

Idea for a lang related to V'eng: preinitial consonants were generally lost or turned to tone, but preinitial vowels ĕ ŏ conditioned +/-palatal. This would backdate the emergence of the seven-vowel system and then the lowered high vowels merge to schwa. Final vowels are unreconstructable besides height except after /a/ so I'll call them /ɐ ɨ/. Then fricative preinitials condition +high, preinitial glottal nasals turn to creaky voice and then migrate to the end of the word (paralleling the development of /h/ in V'eng), and final nasals drop and turn to nasality.



V'eng: mka, ʊng, theng, s vɪng
Proto-Vengic: *mĕka, *ongɨ, *hŏtengɐ, *sŏvengɨ
???: ngkæ, ə̃, tɨẽ, vɨ̃

Assuming verb conjugation remains the same:


etc.

So, the vowel inventory is:

Which isn't really so bad. Not sure how to write it. This is a bit weird, but how about:
<ea a e eo o ie io oa uo ei eu au i iu ou u ua ue ui ae ai oe ia oi>
Or I could just
<ae a ɛ oe ɔ e ø ə o ɪ ʏ ʊ i y ɨ u ɨa ɨe ɨi ai əi aə eə iə>

Presumably V'eng <z> would be reconstructed as */j/ then. I might do away with final vowel-semivowel coalescence; that must have happened late in V'eng, later than final vowel loss.

_________________
Siöö jandeng raiglin zåbei tandiüłåd;
nää džunnfin kukuch vklaivei sivei tåd.
Chei. Chei. Chei. Chei. Chei. Chei. Chei.

Looks pretty good so far. Any idea where clicks in your conlangs originated in the first place?

_________________

"There was a particular car I soon came to think of as distinctly St. Louis-ish: a gigantic white S.U.V. with a W. bumper sticker on it for George W. Bush."

Not going anywhere near that. Might develop more clicks in the Hathic languages that don't lose them -- from implosives, at least.

_________________
Siöö jandeng raiglin zåbei tandiüłåd;
nää džunnfin kukuch vklaivei sivei tåd.
Chei. Chei. Chei. Chei. Chei. Chei. Chei.

Time scale for conlangs:

Kett and Arve have to be spoken at the same time, so Gehui must be at least 3000 years before then, and probably more. Insular and Continental Kett are very close together; anywhere from 500 to 1000 years apart would do it. Kannow goes through a complete phonological restructuring, but it shouldn't take as long as it may appear, and Miar lost prominence sometime after Kannow rose. Vʼëng should be at about the same time as Cherun.

I'll set year zero to the Kett conquest of northern Harue, so:
3500 BH: Gehui, Proto-Vengic
2000 BH: Cherun, Ngmwragh, Vʼëng, Voangʼ
500 BH: Proto-Kett, Miar
500 AH: Kett, Arve, Kannow
1500 AH: Enzielu
2500 AH: Renzell

_________________
Siöö jandeng raiglin zåbei tandiüłåd;
nää džunnfin kukuch vklaivei sivei tåd.
Chei. Chei. Chei. Chei. Chei. Chei. Chei.

Proto-Vengic phonology

Root structure: (C₁{ĕ ŏ ĭ₁ ŭ₁})(C₂){a e o i u}(C₃({ă ĭ₂ ŭ₂}))

C₁ = p t c k ʔ f s x h m n
C₂ = any C or br dr gr fr sr xr my py by vy
C₃ = any C except unvoiced non-glottal plosives

/p t tʃ k ʔ/ <p t c k ʔ>
/b d dʒ g/ <b d z g>
/β r j ɣ/ <v r y ğ>
/f s x h/ <f s x h>
/m n ŋ/ <m n ŋ>
/l/ <l>

--------

Proto-Vengic to Vʼëng:

1. Lenition of finals
z v > y w / _#
e i o u + {y w} > ɪ i ʊ u
ay aw > e o
m n > ŋ / _#

2. Loss of final vowels
a > e / _Cĭ₂
a > o / _Cŭ₂
e o > ɪ ʊ / _C{ĭ₂ ŭ₂}
i u > ɪ ʊ / _Că
ă ĭ₂ ŭ₂ > 0

3. Loss of preinitial vowels
ĭ₁ ŭ₁ > ĕj ŏv
ĕ ŏ > 0
{j v h}C₂ C₂{j v h}

4. Preinitial cluster collapse
Fr > ʂ
i u > ɪ ʊ / ʂ_
Pʂ > c
ʂ > s
by and vy merge into a cluster written <by> that varies freely between either
py > tʃ

5. Velar shit
l > gʟ
gʟ > ɣ / _# gʟV_
g > k / #_
g > ŋ

6. Final stop lenition
b d z > v r y / _#

7. Preinitial mergers
p t c k > ʔ / _C except _{j v}
n > m / _C[lab]
y > 0 / C[pal]_
v > 0 / C[lab]_

_________________
Siöö jandeng raiglin zåbei tandiüłåd;
nää džunnfin kukuch vklaivei sivei tåd.
Chei. Chei. Chei. Chei. Chei. Chei. Chei.

Proto-Vengic to Voangh:
1. Lenition of finals
z v > y w / _#
e i o u + {y w} > ɪ i ʊ u
ay aw > e o
m n > ŋ / _#

2. Loss of final vowels
a > e / _Cĭ2
a > o / _Cŭ2
e o > ɪ ʊ / _C{ĭ2 ŭ2}
i u > ɪ ʊ / _Că
ă ĭ2 ŭ2 > 0
ɪ ʊ > ə

3. Loss of preinitial vowels
ĭ₁ ŭ₁ > ĕj ŏv
{j v}C₂ > C₂{j v}
a e ə o i u / ĕ$_ > ä e ə̈ ö i ü
a e ə o i u / ŏ$_ > a ë ə o ï u

4. Cluster collapse
P[-voice]r P[+voice] Fr > ʈ ɖ ʂ

5. Preinitial loss
P_p > h / _P _CC _{ʈ ɖ ʂ}
ä e ə̈ ö i ü / F₁$_ > ɨɑ ɨe ɨ o ɨi u
a ë ə o ï u / F₁$_ > eə e jə jo i ju
ä e ə̈ ö i ü > ɑ ai ə ɔ əi ʊ
a ë ə o ï u > æ ɛ æə jɔ ɪ jʊ
jɔ jo jʊ ju > œ ø ʏ y
F₁ > 0
m₁ > n₁

6. Final loss
L / _# > :
b d z g / _# > v y y v
m n / _

7. Preinitial glottal metathesis
ʔ₁ / _C > ʔ₃

_________________
Siöö jandeng raiglin zåbei tandiüłåd;
nää džunnfin kukuch vklaivei sivei tåd.
Chei. Chei. Chei. Chei. Chei. Chei. Chei.

Insular Kett:
Veruan haue lehang lat ḩengg ḩi lekter kaumaru.
[ˈveɾwɔn χawɪ lɪˈχɑŋ lɐt heʰŋ̥ çi ˈlektɪɾ ˈkɔwmaɾʊw]
véru=an háu-ue leháng lat ḩeng-ng ḩi létak-er káu-m-aru
heaven=TOP thing-PL many INT give-IPFV.ACT REL man-COLL nurture-GER-do

Lekteran sey aleih ḩi Veru apeummaru.
[ˈlektɪɾɐn sɵɥ ɐˈlɪjç çi ˈveɾʊw aˈpɵwʰm̥ɐɾʊw]
létak-er=an sey aléih ḩi véru apeung-m-aru
man-COLL=COP nothing good REL recompense-GER-do

Nakkau. Nakkau. Nakkau. Nakkau. Nakkau. Nakkau. Nakkau.
[ˈnɑʰkɔw ˈnɑʰkɔw ˈnɑʰkɔw ˈnɑʰkɔw ˈnɑʰkɔw ˈnɑʰkɔw ˈnɑʰkɔw]
nakk-au
kill-IMP.ACT

or, Cerrais dialect:
Veuron haur lehannat ḩengg lekter kaumar.
[vʊwɳ̩ χɔɽ lɪˈχaʰnɐt heʰŋ̥ ˈlektɐɽ ˈkɔmɐɽ]
véru=n háu-r leháng=lat ḩeng-ng létak-er káu-m-ar
heaven=TOP thing-COLL many=INT give-IPFV.ACT man-COLL nurture-GER-do

Lekteron sey aleih vay Veru apeummaur.
[ˈleʰʈɳ̩ sɵɥ ɐˈlɪjç vœɥ ˈveɽʊw aˈɸʊwʰm̥ɔɽ]
létak-er=an sey aléih vay véru apeung-m-aur
man-COLL=COP nothing good NEG recompense-GER-do

Nakkau. Nakkau. Nakkau. Nakkau. Nakkau. Nakkau. Nakkau.
[ˈnaʰkɔ ˈnaʰkɔ ˈnaʰkɔ ˈnaʰkɔ ˈnaʰkɔ ˈnaʰkɔ ˈnaʰkɔ]
nakk-au
kill-IMP.ACT

_________________
Siöö jandeng raiglin zåbei tandiüłåd;
nää džunnfin kukuch vklaivei sivei tåd.
Chei. Chei. Chei. Chei. Chei. Chei. Chei.

^ in Harue script

ve·r·w·n ha·we le·ha·ng la·t ḩe·ng·n ḩ·i le·k·te·r ka·w·ma·r·w
le·k·te·r·n se·y a·le·j·h ḩ·i ve·r·w a·pe·w·ng·ma·r·w
na·ka:·w na·ka:·w na·ka:·w na·ka:·w na·ka:·w na·ka:·w na·ka:·w

_________________
Siöö jandeng raiglin zåbei tandiüłåd;
nää džunnfin kukuch vklaivei sivei tåd.
Chei. Chei. Chei. Chei. Chei. Chei. Chei.

bump, working on scorpion/fox translation in kannow but too lazy to finish

A scorpion was walking along the bank of a river, wondering how to get to the other side.
Qeeŗemtʼand itkʼynitu iłipnu; heltìkdind—Murqdutʼàw kìjindija?
ģ-he-ŗem-tʼa-nd itkʼynit-u iłipn-u; he-ltìk-d-nd—m-r-qdu-tʼa-`ow kì-je-nd-ija
along-I.RP-walk-C5.D-C5.A scorpion-C5.INDEF riverbank-INDEF; I.RP-think-C6.P-C5.A—Q-S.PR-go-C5.D-1S.A M.NF-other-C5.A-C5.DEF

Suddenly, he saw a fox.
Tpèenastʼand kandaŗ.
tp`-he-nys-tʼ-nd kand-ŗ
suddenly-I.RP-see-C2.D-C5.A fox-C2.INDEF

He asked the fox to take him on his back across the river.
Helettʼend—Plʼuqdundyntʼar kìjindija ŗakèest sycisʼís?
he-li-d-tʼ-nd—plʼ-qdu-ndon-tʼa-r kì-je-nd-ija ŗa-kì-e-st syci-sʼ-s`
I.RP-ask-C6.P-C2.D-C5.A—CAUS.OPT.PR-go-C5.P-C5.D-C2.A M.NF-other-C5.A-C5.DEF INSTR-M.NF-do-C7.A back-C7.DEF-C2.POSS

The fox said, "No. If I do that, you'll sting me, and I'll drown."
Hedaqʼatr—Kneņèw. Ņenrèw, getkʼomg, ņeclʼamkʼ.
he-daqʼ-d-r—k-neņ-e-`ow. ņenr-e-`ow ge-tkʼo-m-g, ņec-lʼam-kʼ
I.RP-say-C6.P-C2.A—NEG-S.NRF-do-1S.A. COND.I.NRF-do-1S.A, IRR-sting-1S.P-2S.A, IRR.M-drown-2S.P

The scorpion assured him, "If I do that, we'll both drown."
Hekàŗdand—Ņenrèw, ņeclʼammatʼ magèpʼ.
he-kàŗ-d-nd—ņenr-e-`ow, ņec-lʼam-matʼ ma`-gèpʼ.
I.RP-reply-C6.P-C2.A—COND.I.NRF-do-1S.A, IRR.M-drown-1PL.P 1.POSS-two

The fox thought about it and finally agreed.
Heltìkkar, gàatʼaackar.
he-ltìk-hka-r, gà-he-tʼaj-hka-r
I.RP-think_about-C6.D-C2.A, finally-I.RP-agree-C6.D-C2.A

So the scorpion climbed up on his back, and the fox began to swim.
Mykypgìndonsand sycisʼís i ģaarʼamr.
my-kʼ-hap-gìn-(n)don-sa-nd syci-sʼ-ís i ģa-he-rʼam-r
REFL-on-CAUS.I.RP-get_up-C5.P-C7.D-C5.A back-C7.DEF-C2.POSS and begin-I.RP-swim-C2.A

But halfway across the river, the scorpion stung him.
As poison filled his veins, the fox turned to the scorpion and said, "Why did you do that? Now you'll drown, too."
"I couldn't help it," said the scorpion. "It's my nature."

_________________
Siöö jandeng raiglin zåbei tandiüłåd;
nää džunnfin kukuch vklaivei sivei tåd.
Chei. Chei. Chei. Chei. Chei. Chei. Chei.

Notes on development of Kannow into earliest ancestor of all Enzielu dialects + Thalian. [which was supposed to split from Western Enzielu but dunno I need to completely reorganize everything here unless the planet is much smaller than Earth which it can't be]

Noun class 9 folds into noun class 7; already mostly the same in Kannow. Class paradigms simplify. Class 6 -ndon affix simplifies to -don.



This simplifies further in different ways depending on dialect.

Remote future turns into irrealis; sensory remote future is lost. Inferential loses remoteness distinction in past. Sensory NRF turns into... jussive? Evidentiality distinction is lost completely in present/future tenses: 1S 2S 3E pres; 1S 2E 3E past; 1E 2E 3E fut. Some regularization occurs. Optative lost in present, generalized into -lʼ- for past tense, merges with jussive. So:


Need to develop a conditional jussive/optative; probably just conditional + lʼ.

_________________
Siöö jandeng raiglin zåbei tandiüłåd;
nää džunnfin kukuch vklaivei sivei tåd.
Chei. Chei. Chei. Chei. Chei. Chei. Chei.

General sound correspondences:

p b p' > h b p
t d t' > z d t
s s' > s ts
ř r r' > s? ? ts? or x ? tx? or l > j, r > l?
ł l l' > z l tz
c j c' > x j tx or merges with k after affecting vowel?
k g k' > h g? k
q ģ q' > ? ʕ? ? or merges with k after affecting vowel?
m n ņ > m n ɰ̃
w ŗ h > v r h
ɲ j > ? ?

Problem: many consonants affect vowels in as yet undescribed ways; keeping all/most consonant series distinct may be a problem. Try to describe:
p + a e ə i ɨ u = a e ə i ɨ u
t + a e ə i ɨ u = æ e ə əi ɨ y
s + a e ə i ɨ u = a e ə əi ɨ u
r + a e ə i ɨ u = a æ ə əi ɨ u
l + a e ə i ɨ u = a e ə i ɨ u
c + a e ə i ɨ u = æ e jə i jɨ y
k + a e ə i ɨ u = a æ ə əi ɨ u
q + a e ə i ɨ u = a æ a? ai aɨ au
still need to work out effects of rounding. a o merger after q? or a e?

Secondary developments:
b d j g > v r i~x ? (r in standard, what in dialects that don't have derhotacized r?)
ʕ ɰ̃ > a i/u
could have ɰ̃ > m or v in standard to parallel g > ɣ > ð

_________________
Siöö jandeng raiglin zåbei tandiüłåd;
nää džunnfin kukuch vklaivei sivei tåd.
Chei. Chei. Chei. Chei. Chei. Chei. Chei.

need to step back and do pre-enzielu. obv some of these are gonna be cover symbols for things that go different ways in different dialects/langs, like i did with gehui (which has maximum six vowel POAs but is written as if it had nine to make the correspondences work)

so, basic correspondences: voiceless plosives turn to fricatives. voiced plosives don't in pre- but do in some lects (incl std) and develop through borrowing or clusters or w/e in others, may be absent in native vocabulary for others idk. voiced *affricates* do (so r l) tho r might deaffricate for some. gonna push kannow ahead 500 years to 2000AH, that's about the difference bn. old + modern chinese. it'll help that standard will come from a relatively innovative dialect

need to work out differences bn. lects influenced heavily by continental kett substrate vs. not; c kett preserves some proto-kett rounded consonants (tw kw)

absolute first thing that happens is cluster breakup, once that's out of the way everything becomes much easier

p b p' > ɸ b p
t d t' > θ d t
s s' > s ts
ř r r' > ʂ (ɖ)ʐ ʈʂ
ł l l' > ɬ l tɬ
c j c' > ç* j* k [these vary, j usually devoices + merges with ç, which goes to something palatal in std.]
k g k' > x g k
q ģ q' > x 0 k
m n ņ > m n ɰ̃
w ŗ h > v r h
ɲ j > ? ? [this depends on c kett influence -- merge with n j if not]

lack of non-palatal coronals before i is a *feature* not a bug. so probably derive palatals from l series and r series merges with...? s?

p + a e ə i ɨ u = a e ə i ɨ u
t + a e ə i ɨ u = æ e ə əi ɨ y
s + a e ə i ɨ u = a e ə əi ɨ u
r + a e ə i ɨ u = a æ ə əi ɨ u
l + a e ə i ɨ u = a e ə i ɨ u
c + a e ə i ɨ u = æ e jə i jɨ y
k + a e ə i ɨ u = a æ ə əi ɨ u
q + a e ə i ɨ u = a æ a? ai aɨ au

_________________
Siöö jandeng raiglin zåbei tandiüłåd;
nää džunnfin kukuch vklaivei sivei tåd.
Chei. Chei. Chei. Chei. Chei. Chei. Chei.

I just now listened to the V'eng recording, and it doesn't sound as ugly as your 'langs normally look. I even quite liked the sound of it.

bump

_________________
Siöö jandeng raiglin zåbei tandiüłåd;
nää džunnfin kukuch vklaivei sivei tåd.
Chei. Chei. Chei. Chei. Chei. Chei. Chei.

should probably fuck up nonfinite verb system in enzielu

expand kannow noun incorporation system, allow into nonfinites but person marking becomes prohibited in nonfinites -- null conjunction grammaticalizes into serial verbs with mood marking shit, nonfinites develop more toward c kett gerunds

wait this already exists with some verbs in kannow... hm

some translations i've used nonfinites for adjectives, maybe preserve person marking there, same prefixes used for nominalizations and adjectives... or develop new way of nominalizing and have that grammaticalize

calque postposition constructions from c kett; kannow does all that with verbs right now. then those can grammaticalize into cases, which fits the fenno-basque thing nicely... try to come up w regular tsez-like case matrix, that'd be cool

_________________
Siöö jandeng raiglin zåbei tandiüłåd;
nää džunnfin kukuch vklaivei sivei tåd.
Chei. Chei. Chei. Chei. Chei. Chei. Chei.

Aaaand redoing the time scale, since I decided writing originates with Miar.

Here's what's set in stone:
* Enzielu is spoken at the same time as Hathe.
* Kett is spoken at roughly the same time as Arve.
* Hathe descends from Gehui.
* Enzielu descends from Kannow.
* Miar is the first language to develop writing. Gehui may or may not have writing, but Ngmwragh definitely does.
* Year zero is 600 years after the Kett expansion of northern Harue -- explain this later.

Enze can play Rome to Harue's Greece, except for the part where it's actually modeled on China. Arve is probably going to have to go extinct. :/

1000BE: Beginning of Kett expansion? (cf. Rome + Roman Greece)
600BE: Kett conquest of northern Harue (this is about equivalent in tech to year zero in our timeline)
0E: Beginning of Enze expansion / unification? (~= Qin)
1000E: Old Enzielu
1400E: presumably something here would happen involving more expansion (~= Ming)
1600E: empire breaks down, western split -- Thalia
2000E: Enzielu proper
2500E: Renzell, presumably -- bunch of shit happens in the interim

so, linguistically:

2000BE: Miar, Gehui
1000BE: Cherun, Ngmwragh
500BE: Kett, Arve
0E: Kannow, work out what Kett would be like at this point
1000E: Old Enzielu
2000E: Enzielu, Hathe

The Vengic languages can go pretty much anywhere since they're spoken on a shitty little peninsula no one cares about.

_________________
Siöö jandeng raiglin zåbei tandiüłåd;
nää džunnfin kukuch vklaivei sivei tåd.
Chei. Chei. Chei. Chei. Chei. Chei. Chei.

fixed

_________________


Read all about my excellent conlangs
Basic Conlanging Advice

double fixed, wrong coast

_________________
Siöö jandeng raiglin zåbei tandiüłåd;
nää džunnfin kukuch vklaivei sivei tåd.
Chei. Chei. Chei. Chei. Chei. Chei. Chei.

Insular Kett diachronics

Inventory:
/pʰ p t tʙ̥ ts k/
/f v s x h/
/m n ŋ/
/l r j ɥ w/
/a e/

Insular Kett to Cerrais Kett: (this happens in a pretty short time, or at least appears to, but probably covered up by maintenance of standard)
w > v / #_ (in standard)
P1P2 > P2:
VC*w{V[-stress] #} > VwC*
pʰ tʙ̥ > ɸ > f
some VN sequences, esp. word-finally > N̩
jC[-labial] Cj[-labial] > C[+palatal]
rC[+alveolar] Cr[+alveolar] > C[+retroflex]
rʲ > ʑ
ɭ ɭː r: > r ʂ: ʂ:
NC[-long] > C[+voice]
vC > wC[+voice]
l vocalization?
C: > ʰC[-voice] though this already happens most places, incl. standard
palatovelars > alveolopalatals and stops affricate
dz dʐ > z ʐ

Cerrais Kett inventory:
/p t ts ʈ ʈʂ tɕ k/
/ʰp ʰt ʰts ʰʈ ʰʈʂ ʰtɕ ʰk/
/b d ɖ dʑ g/
/f s ʂ ɕ x h/
/ʰf ʰs ʰʂ ʰɕ ʰx/
/v z ʐ ʑ h/
/m n ɳ ɲ ŋ/
/ʰm ʰn ʰɳ ʰɲ ʰŋ/
/l ʎ r j ɥ w/

And then everything blows up.

aj aɥ aw ej eɥ ew > æ œ o i y u
a / C[palatal]_ > æ
l ʎ > w j / _$
ʎ > lj
ɥ > j + rounding on following vowel

Tonogenesis:
word-final unstressed vowels drop and transfer falling tone onto the preceding vowel
'V${VR C̩} > falling tone
VC[preasp] > high tone
VC[voiced] > low tone
syllables ending in /w j/ take mid tone

low falling > breathy
high > mid rising
high falling > high
mid falling > high falling

affricates > fricatives / _#

probably need to kill some more syllables

_________________
Siöö jandeng raiglin zåbei tandiüłåd;
nää džunnfin kukuch vklaivei sivei tåd.
Chei. Chei. Chei. Chei. Chei. Chei. Chei.

Nouns:

Two classes: animate and inanimate. Two inflection patterns: vowel-final, consonant-final.

Animate: dqera 'dog', bgul 'man'. Inanimate: andzo 'article of clothing', mtaq 'rock'.


So, the declension pattern is:


The -y in these inflections can only appear after l r n ts dz s z, and in the latter four cases, palatalizes them to ch j sh zh. In all other environments, it drops, as is seen in mtaq.

_________________
Siöö jandeng raiglin zåbei tandiüłåd;
nää džunnfin kukuch vklaivei sivei tåd.
Chei. Chei. Chei. Chei. Chei. Chei. Chei.

Verbal person inflection is ergative-absolutive, but with added topic marking. Ergative is unmarked; absolutive marks person and animacy/gender; topic marks person, animacy/gender, and shape (standing, sitting, lying down). If top = abs, no topic markers applied.



Bgul qaris.
[bguɫ qaris]
The man is eating.

Bgul thams tqari.
[bguɫ thams tqari]
The man is eating stew, standing up.

Bgul thams mqari.
[bguɫ thams mʁari]
The man is eating stew, sitting down.

Bgul thams tuqari.
[bguɫ thams tuqari]
The man is eating stew, lying down.

Cxeme bgul thams chqari.
[tsχeme bguɫ thams tʃqari]
With a spoon, the man is eating stew.

_________________
Siöö jandeng raiglin zåbei tandiüłåd;
nää džunnfin kukuch vklaivei sivei tåd.
Chei. Chei. Chei. Chei. Chei. Chei. Chei.

And here's the phonology.

/p t ts tʃ k q/ <p t c ch k q>
/b d dz dʒ g/ <b d dz j g>
/s ʃ χ/ <s sh x>
/z ʒ/ <z zh>
/m n l r j w/ <m n l r y w>
/a (æ?) e o i u/ <a ä e o i u>
Aspiration (preceding h, <h>) is a quality of the vowel. Two aspirated syllables can't occur in a row. Aspiration doesn't occur after voiced stops/fricatives. /tsh sh/ are spelled <c'h s'h> Probably have <ch'h sh'h> instead of <chh shh> also.

Marginal distinction between l and dark l. Might reflect this in the spelling, either <ly l> or <l ł>.

Onset of last syllable in polysyllabic root is unlikely to be a stop, esp. unvoiced; /ts/ is less rare in that position. /m l r/ are most common.

Initial clusters:
PP, where P1 is fronter than P2
Px, where P1 is unvoiced
P{s ʃ}, where P1 is unvoiced
{P|F}{l r}
{m n}C

Medial clusters:
{r l N}P
FC
rF
ndr
NP[-voice]{x q}

Final clusters:
{m n l r j w}{P{F} F{F}}
FP{F}

_________________
Siöö jandeng raiglin zåbei tandiüłåd;
nää džunnfin kukuch vklaivei sivei tåd.
Chei. Chei. Chei. Chei. Chei. Chei. Chei.