The Correspondence Library

The best topics from Languages & Linguistics, kept on a permanent basis.
User avatar
Nortaneous
Sumerul
Sumerul
Posts: 4544
Joined: Mon Apr 13, 2009 1:52 am
Location: the Imperial Corridor

Re: The Correspondence Library

Post by Nortaneous »

Theta wrote:p pʷ k > β w ɣ
bʷ g > kpʷ k
Wait, what? Voiceless peripheral stops lenite to fricatives and voiced peripheral stops fortite to voiceless stops?
Siöö jandeng raiglin zåbei tandiüłåd;
nää džunnfin kukuch vklaivei sivei tåd.
Chei. Chei. Chei. Chei. Chei. Chei. Chei.

----
Smeric
Smeric
Posts: 1418
Joined: Tue Feb 15, 2011 11:15 pm

Re: The Correspondence Library

Post by ---- »

yes, really! variants of that set of sound changes happened all throughout north and central Vanuatu.

EDIT: here's two more:

Proto-Oceanic to Lemerig

p pʷ k q > β w ɣ ∅
VV > V
ndr R > d r
c j ñ > s s n
t > ʔ often
b bʷ d g > p kpʷ t k
mʷ > ŋmʷ

iCV[-high] > aC
eCV[+mid] > ɛC
eCV[+low] > aC
aCV[+high] > ɛC~œC
aCV[+mid] > aC
aCa > ɒC sometimes, otherwise > aC
oCV[+high] > øC
oCV[-high] > ɔC (oCo sometimes > œC)
uCV[-high] > oC
e o > ɪ ʊ
ia > ɪ (only sometimes?)
[intervening consonants optional in the previous sound changes]

#VCV[+stress] > CV[+stress]
CVCVCV[+stress] > CVCCV[+stress]
CV1CV2[+stress] > CV2CV2[+stress]


Proto-Oceanic to Vera'a

p pʷ k q > β w ɣ ∅
VV > V
ndr R > d r
c j ñ > s s n
t > ʔ often
β > f/#_ (sometimes doesn't occur, β > f in other environments rarely)
bʷ g > kpʷ k
mʷ > ŋmʷ

iCV[+high] > iC
iCV[-high] > iCɪ
eCV[-low] > eC
eCV[+low] > ɛCɛ
aCi aCu > ɛC or aC ; ɛC ɔC or aC
aCe aCo > aC
oCa > ɔCɔ
uCV[+high] > iC sometimes
uCV[-high] > uCʊ
ea ae oa > iɛ iɛ uɔ
e o > ɪ ʊ
(intervening consonants optional except in the specific diphthong changes specified)

#VCV[+stress] > CV[+stress]
CVCVCV[+stress] > CVCCV[+stress]
CV1CV2[+stress] > CV2CV2[+stress]
b d > m n/_C ; _#

User avatar
Tropylium
Avisaru
Avisaru
Posts: 512
Joined: Sun Aug 07, 2005 1:13 pm
Location: Halfway to Hyperborea

Re: The Correspondence Library

Post by Tropylium »

Nortaneous wrote:
Theta wrote:p pʷ k > β w ɣ
bʷ g > kpʷ k
Wait, what? Voiceless peripheral stops lenite to fricatives and voiced peripheral stops fortite to voiceless stops?
IIRC it is more common yet across Oceanic for the stops to have a chainshift along the lines of *p *t *k > *f *θ *x, *b *d *g > *p *t *k. It looks like here *t stayed put, but the peripheral fricatives further became voiced.
[ˌʔaɪsəˈpʰɻ̊ʷoʊpɪɫ ˈʔæɫkəɦɔɫ]

User avatar
Pabappa
Lebom
Lebom
Posts: 210
Joined: Tue Sep 14, 2004 5:51 pm
Location: the Peyron Apartments
Contact:

Re: The Correspondence Library

Post by Pabappa »

At first I was thinking that there was no way that could be real, but with the intermediates I realize its basically just Grimm's Law.
And now Sunàqʷa the Sea Lamprey with our weather report:
Image

User avatar
Pogostick Man
Avisaru
Avisaru
Posts: 894
Joined: Sat Jul 25, 2009 8:21 pm
Location: Ohio

Re: The Correspondence Library

Post by Pogostick Man »

Information from this post is from Wares, Alan C. (1954?), "Three Pai Dialects of Lower California". Summer Institute of Linguistics Bartholomew Collection of Unpublished Materials (CC BY-NC-SA 3.0—I'm not sure if Wares or his estate hold the copyright or if SIL does, given the Creative Commons license)

Proto-Pai to Paipai

b → β
xʷ → w / tʃ_
kʷ xʷ → k x / _#
ʃ → ʂ
ɬ → l
i → ə / unstressed
n → Ø / _t
ʔ → Ø / _ɲ
aw aj → o e

Proto-Pai to Tipai

kʷ xʷ → q χ / _# (the paper calls these "back velar")
b → p
i → ə / unstressed
u → o / _K
t → Ø / n_
tʃ → Ø / _xʷ
ɲ j → n Ø / ʔ_
Vowel length partially lost?
Contrastive stress lost?

Proto-Pai to Chapai

tʃ → ʃ / _{w,i}
tʃ → Ø / _xʷ
tʃ → s
t → tʃ / ! n_
n → Ø / _t
kʷ → k / _#
b → p
s → ʂ
o → u
ʔ → Ø / _{ɲ,j}
{w,j} → Ø / a_
Vowel length lost?
Stress lost?
(Avatar via Happy Wheels Wiki)
Index Diachronica PDF v.10.2
Conworld megathread

AVDIO · VIDEO · DISCO

----
Smeric
Smeric
Posts: 1418
Joined: Tue Feb 15, 2011 11:15 pm

Re: The Correspondence Library

Post by ---- »

Tolomako is a language with a very simple phoneme inventory but it has a couple neat changes in its phonological history. Information is taken from The Apicolabial Shift in Nese by John Lynch and various lexical data from this website: http://language.psy.auckland.ac.nz/austronesian/ . I'd like to do other languages from the Santo-Malekula area but data is fairly sparse so we'll see how it goes. Sakao is kinda batshit crazy so I'd like to do that one especially.

Proto-Oceanic to Tolomako:

C > 0/_#
q > 0
R > 0 often, otherwise > r
ñ j c > n z s
m b > n t /_{i, e, a}
p pʷ k > v vʷ ɣ
mʷ b bʷ vʷ > m p p v
d g > r k
{z, dr} > ts

+Vowels pretty much stay the same, except for the sporadic change u > i characteristic of the area, and the word tsae- 'blood' suggests that in some cases a > e (tsae < *draRaq).
+The labial > alveolar change had a linguolabial intermediate stage but I left it out here because it's entirely superfluous.

----
Smeric
Smeric
Posts: 1418
Joined: Tue Feb 15, 2011 11:15 pm

Re: The Correspondence Library

Post by ---- »

I remarked on this language in the Bizarre Sound Changes thread because of an entertaining change that it had: what amounts to a reversal of the fairly commonly seen change θ > f. But here's a more complete elaboration on its phonological history, from Proto-North New Hebridean Reconstructions, by Jacques Guy.

Proto-Oceanic to Shark Bay

C > 0/_#
q > 0
R > 0; otherwise > r
j c ñ > z s n
p pʷ k > v vʷ ɣ
b bʷ g > p pʷ k
V[+high] > 0/_# except when the preceding consonant is d, z, dr, or p.
v t l r > p dr n w/_#
t > ts/_V[+high]
a[+stress] > i/_CV[-high]
a[+stress] > é/_CV[+high]
ŋ p > 0 f/V[+stress]_V
ɣ s d > 0 j dr/_# or before a post-tonic vowel
V > e/C_# (C does not include j)
V > 0/V[+stress]_ or /j_
p v m > t θ n/_{i, e, a} (here's that change)
pʷ vʷ mʷ > p f m
z > s
0 > h/#_V (it isn't clear if this happens unconditionally)
Note: it seems that Shark Bay develops a seven-vowel system through unclear methods: é = [e] and e = [ɛ]. There is also a corresponding set ó o [o ɔ] but they don't appear in these changes. It's likely that this developed in a way similar to the complex vowel systems of the Torres-Banks Islands.

hwhatting
Smeric
Smeric
Posts: 2315
Joined: Fri Sep 13, 2002 2:49 am
Location: Bonn, Germany

Re: The Correspondence Library

Post by hwhatting »

Theta wrote:Proto-Oceanic to Shark Bay
Next: Shark Bay to Manta Ray Reef
(Sorry, couldn't help it. Thanks to you and Pogostick Man for posting all these sound changes.)

CatDoom
Avisaru
Avisaru
Posts: 739
Joined: Fri Sep 20, 2013 1:12 am

Re: The Correspondence Library

Post by CatDoom »

Here's some more California correspondences, this time for Chumashan. My source is Topics in Historical Chumash Grammar, a 1977 dissertation by Kathryn Klar, available online at http://linguistics.berkeley.edu/~survey ... r-1977.pdf.

Klar reconstructs the phonology of Proto-Chumash as follows:

Plain Stops: p t̪ k q ʔ (there appears to have been an irregular alternation between *q and *x; q > x will therefore not be represented as a sound change below)
Ejective Stops: p' t̪' k' q'
Plain Fricatives: s̪ ʃ h
Ejective Fricatives: (s̪' ʃ') (no cognate sets; included on the basis of symmetry and the appearance of these phonemes in the attested Chumash languages)
Plain Affricates: t̪s̪ tʃ
Ejective Affricates: t̪s̪' tʃ'

(note: Klar states that it's unclear whether the distinction between dental and palato-alveolar fricatives and affricates was phonemic or merely phonetic, but posits that the same "sibilant harmony" observed in all of the attested Chumashan languages probably originated in the proto-language. Sibilant correspondences are irregular, and will not be included in the sound changes listed below.)

Plain Nasal: m n
Glottalized Nasal: ˀm ˀn
Plain Approximant: w l j
Glottalized Approximant: ˀw ˀl ˀj

Proto-Chumash obstruents, including both stops and fricatives, were aspirated under specific conditions, particularly when geminated and in clusters with *h.

Vowels: i (ɨ) e a o u (*ɨ does not seem to participate in vowel harmony, and may be a secondary development influenced by language contact)

Proto-Chumash seems to have had some unusual ablaut, whose conditioning factors are unclear; *i and *u seem to have alternated in prefixes, while *e an *o alternated in stems. There was also some kind of vowel harmony, but it's apparently difficult to characterize. Due to these unpredictable vowel alternations, it's more or less impossible to establish regular vowel correspondences between the Chumashan languages.

--

Note: Some Chumashan languages are better attested than others, and some sound changes may not be represented here due to a lack of data.

Proto-Chumash to Ineseño

t' → t
q' → q
ˀm → m
ˀn → n
ˀw → w


Proto-Chumash to Barbareño

Glottalized stem-final consonants (including ejectives) seem to lose their glottalization when a plain sonorant appears in the onset of the stem-final syllable, with the glottalization shifting to the onset consonant.


Proto-Chumash to Ventureño

ʔ → ∅ /_#
p' → p
k' → k
q' → q
ˀm → m
ˀn → n
ˀl → l
ˀj → j


Proto-Chumash to Purisimeño

ˀj → ∅
q' → q


Proto-Chumash to Obispeño (Northern Chumash)

k → tʲ~kʃ~k (allophonically)
q → q~k (allophonically)
ʔ → ∅ /_#
p' → ʔ
t' → ʔ
k' → ʔ
q' → ʔ
ˀm → ∅ (possibly ʔ)
ˀn → ∅ (possibly ʔ)
ˀw → w (possibly ˀw)
ˀj → ∅


Proto-Chumash to Cruzeño (Island Chumash)

k → tʃ ("in certain cases")

User avatar
Pogostick Man
Avisaru
Avisaru
Posts: 894
Joined: Sat Jul 25, 2009 8:21 pm
Location: Ohio

Re: The Correspondence Library

Post by Pogostick Man »

Information from this post comes from Gudschinsky, Sarah C. (1971), "Ofaié-Xavante, a Jê Language". It deals mostly with consonants because the vowels are a mess.

Proto-Ofaié-Jê to Ofaié-Xavante

m → {w,p} (the former was common finally)
c → {ʃ,h,j}
ɲV → jṼ
k(ʷ) → ʔ / _#
kʷ → k
ŋ → n / V_V
ŋ → j̃ / #_ (not sure if this nasalizes the following vowel or not)
ŋʷ hʷ → j̃ h

Proto-Ofaié-Jê to Proto-Jê

Vm → Ṽ / _#
VS → r / C_V
c → {c,z}
ŋʷ {kʷ,hʷ} → m p
ə → Ø / C_CV (not sure if this happened all the time or not)
(Avatar via Happy Wheels Wiki)
Index Diachronica PDF v.10.2
Conworld megathread

AVDIO · VIDEO · DISCO

User avatar
Pogostick Man
Avisaru
Avisaru
Posts: 894
Joined: Sat Jul 25, 2009 8:21 pm
Location: Ohio

Re: The Correspondence Library

Post by Pogostick Man »

Information in this post is from Meira, Sérgio (2005), "Reconstructing Pre-Bakairi Segmental Phonology". Anthropological Linguistics 47(3):261 – 291

Pre-Bakairi to Eastern Bakairi

VNV → ṼṼ / ! _(C)#
ʔ → Ø / _C[+ fricative - voiced]
ɾ → Ø / V_V, when neither vowel is stressed

Pre-Bakairi to Western Bakairi

VNV → ṼṼ / when the first vowel is not stressed
z → h / V_a
z → Ø / V_V
C[+ fricative - voiced] → Ø / ʔ_
ʔ → Ø / ʔV_V ?
ɨ → ə / P_
ɨ → i
ʒ → Ø
ɾ → Ø / V_V, where at least one of the vowels is nasalized
(Avatar via Happy Wheels Wiki)
Index Diachronica PDF v.10.2
Conworld megathread

AVDIO · VIDEO · DISCO

User avatar
Pogostick Man
Avisaru
Avisaru
Posts: 894
Joined: Sat Jul 25, 2009 8:21 pm
Location: Ohio

Re: The Correspondence Library

Post by Pogostick Man »

Information from this post is from here, citing (I presume) Constenla Umaña, Adolfo (2005), "¿Existe relación genealógica entre las lenguas misumalpas y las chibchenses?". Estudios de Lingüística Chibcha 24: 7–85.

Proto-Lenmichian to Proto-Chibchan

l → ɾ
w → Ø

Proto-Lenmichian to Proto-Lencan

b d → m n / _V[+ nasal] (I'm inferring this from the statement that "[t]here are also a series. . .of nasal vowels")
b d → p l
{ʔ,h} → Ø
{s,ts} → tsʼ
o a → {u,o} {a,e}

Proto-Lenmichian to Proto-Misumalpan

d b → n m / _V[+ nasal] (I'm inferring this from the statement that "[t]here are also a series. . .of nasal vowels")
d b → d {b,p}
ʔ → Ø
ts → s
h → Ø
(Avatar via Happy Wheels Wiki)
Index Diachronica PDF v.10.2
Conworld megathread

AVDIO · VIDEO · DISCO

----
Smeric
Smeric
Posts: 1418
Joined: Tue Feb 15, 2011 11:15 pm

Re: The Correspondence Library

Post by ---- »

I'll be going through all the languages traditionally classified as 'Southern Vanuatu languages'. Lynch reconstructs a medial stage, Proto Southern Vanuatu (PSV):

POc to PSV

m b > mʷ bʷ/_u
p > bʷ or b (sometimes)
p > vʷ/_u **
p > v
k > ɣ (frequently)
R > r (frequently)
R > 0
dr > d or r
ñ > y
n > ŋ if an adjacent syllable contained *q and the intervening vowel was unstressed; i.e. qn nq > ŋ (?)
c > s
t > c/_i, e
q > v ***

vowel changes:
a > e/_(C)i
a > ə/_Ca

PSV's phoneme inventory is thus reconstructed as:
/p pʷ t c k q/
/b bʷ d g/
/v vʷ s j ɣ/
/m mʷ n ŋ/
/l r w y/

/i e a o u ə/



If you were really invested in my writeups for the Northern/Central Vanuatu languages, none of this should seem very strange.
**Lynch does not use the symbol vʷ in the book or reconstruct such a phoneme himself, but I don't think I'm lacking in any prudence by doing it myself. The reasons will become clear in later posts.
***(apparently happened in some isolated cases; what may have actually happened is q > kw as an irregular change)

This and further posts on this language group, are of course taken from Lynch, John. The Linguistic History of Southern Vanuatu (2001).

----
Smeric
Smeric
Posts: 1418
Joined: Tue Feb 15, 2011 11:15 pm

Re: The Correspondence Library

Post by ---- »

Here's what I think is the most fun of them all :)

Proto-Southern Vanuatu to Anejom

C > 0/_# unless C = *t
t > s/_#
{v, vʷ} > h
k > 0/V_V sometimes
b bʷ g > p pʷ k
s > h rarely
s > θ (sometimes doesn't occur when s/_i)
d j c > tʃ s s
{n, ŋ} > ɲ/_{i, e}
w > v in the majority of words
l > tʃ/_{i, e, o}
q > 0 **
V > 0/_# almost all of the time
{h, r} > 0/_# (final r is preserved in one word: inhar from *na-paRi 'stingray', and in the suffix -r from *-dra '3PL POSS')
[some funny business goes on here, a lot of word medial vowels get elided, sometimes even when they should be stressed]

Vowel Changes

these are quite difficult to get down formally and Lynch says the same thing; vowels in Anejom are enormously inconsistent with the reflexes they report and this trend continues somewhat with other SV languages. However, some trends can be noticed. The unconditioned, typical changes are as follows:
i o u > e e o
others:

i > o/{u, w}_
i > i morpheme finally but word medially (isn't always preserved even here)
i also preserved morpheme initially in some cases
u > e/_θ,θ_,ɣ_ (a few counterexamples)
u > u morpheme initially or /_V
u > w/_V sometimes
u > i in a few cases
ai > i/_C
ei > i
ua > ou
au > o, u sometimes
a > e/_C{i, u} almost always
one example of ue > ou
one example of eo > a
e > i adjacent to palatals [tendency]
a > o adjacent to labials [tendency]
and numerous other bizarrities

Lynch proposed an intermediate vowel *ə in PSV but didn't really go much into its results in any of the languages. it seems to > e in Anejom quite often but I can't figure out any real trends (Neither could Lynch, as far as I can tell)

**q sometimes behaves like v, which is quite difficult to explain but it seems to only happen before *u, so perhaps there was some kind of labialized intermediate form

One thing I'd like to point out is that these changes produce some drastically eroded reflexes of some Proto-Oceanic vocabulary. A couple of my favorites:
*a-Ropok > a-e
*a-maqurip > u-mu
and the best of them all: *a-matakut-akini > e-mtit-aɲ

----
Smeric
Smeric
Posts: 1418
Joined: Tue Feb 15, 2011 11:15 pm

Re: The Correspondence Library

Post by ---- »

Here I'll be going over the languages of Tanna. Like Anejom (and all the languages of Vanuatu to some degree), they're notable in that words can become highly eroded, and more uniquely, they developed a six vowel system /i e a o u ə/ in which /ə/ is fully phonemic, appearing even in stressed syllables. **Vowel changes will be done separately at the end because they are really confusing and inconsistent.**

PSV to Proto Tanna

{vʷ, w} > kʷ
s c j > h s z (s and c also sometimes become z)
g > k
q > 0
l > r

Proto Tanna to North Tanna

kʷ > p/_#
kʷ > 0/_u, (sometimes _a as well)
kʷ > w~u
v > 0/_i
v > u~w sporadically
ɣ > 0/_{i, e} and word initially
ɣ > ŋ
r > {i, l} ( > l tends to happen _{i, e, o} )
d > t in many words
d > k/_ŋ
z > r
{s, c} > h or s with extreme irregularity (this is the case with the rest of the Tanna languages as well)

Proto-Tanna to Whitesands

r > i, l
d > r/_ŋ and in other places sometimes
b bʷ d > p pʷ t
kʷ > 0/_u,u_
kʷ > w~u
ɣ > 0/#_ and next to {i, e}
ɣ > ŋ
c j > s z~s
s > h often
z > r

Proto-Tanna to Lenakel

r > i, l
t > r
b bʷ d > p pʷ t
kʷ > u~w
ɣ > 0 next to {i, e}
ɣ > k
r > l/_Vl
c j > s z~s
z > t
s > h irregularly
{p, pʷ, v}Vh > fV

----
Smeric
Smeric
Posts: 1418
Joined: Tue Feb 15, 2011 11:15 pm

Re: The Correspondence Library

Post by ---- »

Proto Tanna to Southwest Tanna

t > r
b bʷ d > p pʷ t
ɣ > 0/#_
ɣ > k
r > l
{c, j} > s
s > h irregularly
{p, pʷ, v}Vh > fV

Proto Tanna to Kwamera

t > r
b bʷ d > p pʷ t
ɣ > 0
{c, j} > s
s > h irregularly
{p, pʷ, v}Vh > fV
a marginal phoneme fʷ develops (by a similar process?)

Vowels:
Okay, so from POc to Proto Tanna, it goes like this:

e > i
o > u~ə
a > o/_Cu or next to a labial
a > e/_Ci
a > ə/_Ca

And then it all goes wild.

+Kwamera tends to change ə > e except in final syllables, where, like Southwest Tanna, it tends to change it to a.
+vowels tend to lower near h: *taci > taha (N. Tanna)
+S Tanna and Kwamera tend to change u > e, i/_Cu
+any vowel is subject to reduction to ə in any position, it seems to happen fairly randomly. (this does, however, happen after ə > e in Kwamera)
+other unexplained vowel shifts also occur with relative frequency
+vowels are often lost in unstressed position

ə is actually often [ɨ], and /v/ in all languages of Tanna is actually a semivowel version of this, rather than a fricative. This means the languages of Tanna have a phonetic distinction between all of [i̯ ɨ̯ u̯] as well as their syllabic counterparts.

----
Smeric
Smeric
Posts: 1418
Joined: Tue Feb 15, 2011 11:15 pm

Re: The Correspondence Library

Post by ---- »

-Erromango languages-

PSV to Proto Erromango

mʷ pʷ bʷ vʷ > m p b v
v > p/_#
v > f when the adjacent syllable contained a sibilant
r > L sometimes (L is a formal symbol, not IPA or anything)
s c j > h s s
o > a
i u > e o sometimes
a > i/_C{i, u}
many word medial vowels lost

Proto Erromango to Sye

L > r
b d g > p t k/#_, C_
b d g > m n ŋ/_#
b d g > mp nt ŋk
f > p/#_
f > v/V_V
k > ɣ
ɣ > k/_i
s > h often
s > 0/_C (not always)

i > e when next to a labial obstruent
e > o when next to a velar
a > o/{w, m, ŋ}_
a > e/_#

Proto Erromango to Ura

h > 0
L > l
nr > d
b d g > m n ŋ/_C
b d g > p t k/_#
p > b/V_V
ɣ > 0/_#
k > 0 (happens a lot in every position)
s, t > h/_{n, r, l}
t > r except word finally and initially

u > e/_# when there was a ɣ

Apparently there was also an *ə phoneme in Proto Erromango reflected as e or o in Sye, and reflected as i in Ura.

That's all I've got on Southern Vanuatu.

User avatar
Nortaneous
Sumerul
Sumerul
Posts: 4544
Joined: Mon Apr 13, 2009 1:52 am
Location: the Imperial Corridor

Re: The Correspondence Library

Post by Nortaneous »

Proto-Indo-European to Proto-Tocharian
Ḱ > K (centum)
Ch > C / _(V)Ch (Grassmann's Law)

d > 0 / _N
d > dz
dz > 0 / V[back]_
dz > ts

Kʷ > K / _C _o _a
Kʷ > Kʷ / _i _u _Ṛ
Kʷ > ś / _e _e:
Kʷy > ś

C > C[+pal] / _y (also _e _e: and sometimes _i _i: -- see below)
The palatalized pairs are:
p > py
t dʰ > c if palatalized by vowels
t dʰ > ts if palatalized by y
ts k kʷ > ś
s > ṣ
m n > my ñ
l r y > ly ry wy

D Dʰ > T

ṇ > ä > 0 / C_#
Ṛ > uR / C_[C #] (some people say CäR)
Ṛ > æR / #_C
Ḥ > ā / C_[C #]
Hṇ > ān / #_C
HṚ > æR / #_C
H > 0 / V_V (this may have happened after umlaut, but the rest happened before)

h2e > ä / _# (?)
h2e > a
a > ā

eh2 aH > a:
a: > å

h3e > o
o > æ

eh3 oH > o:
o:(s y) > u / _#
o:n > either u or õ / _#
õ causes umlaut?
õ > o
o: > a

u > wä / #_
u > ä u ("in some sense both [are] equally likely")

uH > u:
u: > wā (but: uh1 > wä?)

Ci > Cä / if C is bilabial, dorsal, labiovelar, or s
Ci > C[+pal]ä / otherwise
s > ś / _c (s > sʲ / _tʲ)

ih1 > yä
ih2 ih3 > yā

(h1)e > yä / #_
Ce > C[+pal]ä (unlike i, e always palatalizes velars)

eH > e:
(h1)e: > yæ / #_
Ce: > C[+pal]æ

ey ay oy > äy ai æi
ew aw ow > äw au æu

æ > o / _$[o u]
æ > ā / _$ā
(this may have applied to other vowels)
(å triggers o-umlaut in TB)

ā[+stress] ä[+stress] ā[-stress] ä[-stress] > ā a a ä

Proto-Tocharian to Tocharian A
st > ṣt
k > p / _[s ṣ]
y > : / C[+pal]_
n > y / V[-front]_sV (i.e. ansV > ainsV -- then > ensV, see below)
V[-front] > y / V_ñc#
wy > w

ai æi äy > e e i
au æu äw > o o u
æ å > a a

labiovelars trigger rounding of an adjacent vowel (only ä?)
kʷ kw > k

ä > 0 / _. (i.e. in open syllables)
V > 0 / _#
ä epenthesis to break up 'difficult' consonant clusters (mostly in the coda?)

Proto-Tocharian to Tocharian B
ai æi äy > ai ai i
au æu äw > au au u
æ å > e o

wy > y
labiovelars are usually but not always retained
w > : / C[coronal]_
mn > nm / V_V
s > 0 / n_#
0 > t / N_S L_S

ä[-stress] > 0 / _. (i.e. ä is lost in open syllables when unstressed)
ä epenthesis to break up 'difficult' consonant clusters (mostly in the coda?)
Siöö jandeng raiglin zåbei tandiüłåd;
nää džunnfin kukuch vklaivei sivei tåd.
Chei. Chei. Chei. Chei. Chei. Chei. Chei.

User avatar
KathTheDragon
Smeric
Smeric
Posts: 2139
Joined: Thu Apr 25, 2013 4:48 am
Location: Brittania

Re: The Correspondence Library

Post by KathTheDragon »

Source?

User avatar
Nortaneous
Sumerul
Sumerul
Posts: 4544
Joined: Mon Apr 13, 2009 1:52 am
Location: the Imperial Corridor

Re: The Correspondence Library

Post by Nortaneous »

put together from http://www.utexas.edu/cola/centers/lrc/ ... -TC-X.html, except for the treatment of syllabic resonants, which is from https://azargoshnasp.net/history/Tochar ... harian.pdf -- he says *all* R > uR, but it looks like initial resonants developed differently
Siöö jandeng raiglin zåbei tandiüłåd;
nää džunnfin kukuch vklaivei sivei tåd.
Chei. Chei. Chei. Chei. Chei. Chei. Chei.

----
Smeric
Smeric
Posts: 1418
Joined: Tue Feb 15, 2011 11:15 pm

Re: The Correspondence Library

Post by ---- »

This time I'll actually be using a variety of sources but they'll all be by the author Françoise Ozanne-Rivierre. I'll specify if any other authors are credited in the sources I use, and as before I'll mark any notational differences and such that I put in myself with some kind of footnote.

First thing's first:

Proto Oceanic to Proto New Caledonia (PNC)

c > s
{l ñ} > n
R > 0 (only a few exceptions, where in the Loyalty islands it is preserved in a handful of forms and apparently merges with *r)
r > ʈ *
V > 0/_CV[+stress] **
NS > S[+prenasal] (that is, a nasal consonant+a stop produces a prenasalized stop as a singular phone)
CC > C: (the symbol <:> here isn't being used for gemination per se, more of a fortis-lenis distinction)


Some explanation is needed here. First of all (*) in New Caledonian languages the prenasalized sets of stops act much more parallel to the plain stops, which is actually atypical for Oceanic languages. To expediate sound change transcription, I'm going to be writing them as sequences from now on; that is, instead of <b bw d dr j g> I'll be writing <mp mpw nt nʈ ns nk>. In addition, this is not only more similar to how they were pronounced but probably how they were conceived of as well.

Secondly, (**) pre-tonic vowel loss and the production of prenasalized stops and fortis stops seems to have been a process that occurred over a long period of time, because not only are there many mismatches (one language shows a fortis stop reflex while another shows a lenis reflex, one shows a prenasalized consonant, etc.) but it also produced fossilized grammatical distinctions. There are many verb-noun doublets where the noun has a prenasalized initial and the verb has a plain one, e.g. Nyelâyu cep 'to poison' vs. njep 'D. trifoliata' both from the POc noun *tupa. That being said, let's go to yet another medial reconstruction:

Proto New Caledonia to Proto Northern

s s: k k: q q: > t̪ t̪: c c: k k:
C: > Cʰ
-palatalization of original velars was incomplete at this stage, it only occurred initially in the whole Northern family, but in one cluster of languages the palatalization occurred everywhere. Nevertheless, *k and *q don't seem to have merged anywhere.
-final consonants irregularly lost, but always lost before possessive suffixes.

Next time I'll start doing the Northern New Caledonian Languages. By Ozanne-Rivierre's classification this includes everything north of and including Paicî and Cemuhî.

Information in this post is taken from The Proto-Oceanic Consonantal System and the Languages of New Caledonia and Structural Changes in the Languages of Northern New Caledonia, both by Ozanne-Rivierre.

----
Smeric
Smeric
Posts: 1418
Joined: Tue Feb 15, 2011 11:15 pm

Re: The Correspondence Library

Post by ---- »

One thing I forgot: most languages of the North got a y- on a-initial words. This changes further in several other languages, which I'll definitely describe in the relevant sections of the following posts. Also ŋ > n in every language except those of the Loyalty Islands (e.g. Iaai, Drehu, and Nengone).

Proto-Northern to Nixumwak-Nêlêmwa
V > 0/_# in many cases
k > c intervocally and finally (i.e. k palatalization is completed)
t > k
t̪ > t
k > 0/_{a, o}
cʰ > ʃ~yʰ
kʰ > h/_a
pw p t ʈ c k > vw~w v r l y ɣ /V_V
pw > w
ʈ > t
n > l/V_V (adjacent vowels are then nasalized)
prenasalized stops became the associated plain nasal word finally, e.g. *-gu > -ɲ
SN > Nʰ i.e. stop-nasal clusters became voiceless nasals

In Nelemwa, the remaining aspirated stops, pwʰ pʰ tʰ kʰ, became fricatives, except in the alveolar case: > fw f rʰ x

Vowel changes:
vowels are pretty stable, but some pretty consistent ones are:
uCu > iCi
e o > a a (unknown conditions, but probably influenced by nasals: *qenop > kãlãp)
i u > e o in monosyllabic forms almost always, and sometimes elsewhere
stressed vowels lengthen in many cases

Also two vowels, ɨ ə, developed, but their origins are rather obscure. I can only find one example in the index: *paRu > pɨɨ-c
there's a consonant lʰ according to something else I read but now I can't remember, but I don't know how this developed either. There's not really anywhere it could have come from.

----
Smeric
Smeric
Posts: 1418
Joined: Tue Feb 15, 2011 11:15 pm

Re: The Correspondence Library

Post by ---- »

Okay Proto-Northern is too difficult to work with so I'll just be deriving the rest of the stuff straight from Proto-New Caledonia. Sorry for the mess that's occurred so far, but I'll try to clean things up after I've finished with all the Northern languages.

Proto New Caledonia to Yuanga 1 and 2

s k q > t̪ c k
p: pw: t̪: t: ʈ: c: k: > pʰ pwʰ t̪ʰ tʰ ʈʰ yʰ h
all other fortis consonants become aspirated consonants
k > 0/_{a, o}
t tʰ > k kʰ
yʰ > h/_i
p pw t̪ ʈ c k > v w ð l y 0/V_V
V > 0/_#
stop+nasal clusters become voiceless nasals
word final prenasalized stops become nasals

Vowel changes:
The same i u > e o lowering as in Nixumwak-Nelemwa
ai au > ɛ ɔ
o > ɔ sometimes
a > e or ɛ in some words

Further changes to distinguish the dialects:

Yuanga 1:
C > 0/_# (word final nasals nasalize the preceding vowel as well)
yʰ > θ
y > ð or z? unclear whether this change actually occurred, but in some publications Yuanga is referred to as 'Zuanga' "based on the pronunciation of one of its dialects"

Yuanga 2:
t̪ ʈ > t t
ð > l

----
Smeric
Smeric
Posts: 1418
Joined: Tue Feb 15, 2011 11:15 pm

Re: The Correspondence Library

Post by ---- »

Proto New Caledonia to Nyelâyu

0 > n/#_V (only in nouns, probably fusion with Proto Oceanic article *na)
V > 0/_# in disyllabic words
C > 0/_# sometimes
s t k > t̪ c y (k palatalization apparently only happened initially, and there's no evidence of it happening to the prenasalized stop)
p pw t̪ ʈ c > v vw~w r l y/V_V (the dentals and retroflexes lenited finally as well)
pw > w occasionally
t̪ ʈ > t t
nq > nk
p: pw: t: c: q: > pʰ pwʰ tʰ cʰ h
other fortis consonants become aspirated
q > 0
y yʰ > 0 h/_i
w > ɣ in some cases, for unknown reasons
word final prenasals become nasals
stop+nasal clusters become aspirated (voiceless) nasals
n > l/V_V, offloads its nasality onto the vowels
prenasalized stops become nasals before nasalized vowels (including allophonically nasalized ones, such as before other nasal consonants)

Vowel Changes
u > i often
i u > e o in many contexts

Proto New Caledonia to Caaàc

t s ʈ k q > c t t c k
p: pw: t: c: k: > pʰ pwʰ tʰ cʰ/yʰ h (c: > ch when c < *t, and > yh when < *k)
k > 0/_{a, o}
N > 0/_#
V > 0/_# in disyllabic words
yʰ > h/_i
{p pw k} > 0/V_V
t c > l y/V_V
n > l/V_V, offloads nasality onto adjacent vowels
word final prenasalized stops become nasals

basically the same vowel changes

----
Smeric
Smeric
Posts: 1418
Joined: Tue Feb 15, 2011 11:15 pm

Re: The Correspondence Library

Post by ---- »

Proto New Caledonia to Pwapwâ

V > 0/_# sometimes
s ʈ k q > t̪ t̪ c k
p: pw: t: t̪: c: k: > pʰ xʷ tʰ t̪ʰ s x
t t̪ tʰ t̪ʰ > c t cʰ tʰ
C: > Cʰ
p pw t c k > v~0 w l 0 0/V_V
y > z
bw mw > gw ŋw (and further to g ŋ before rounded vowels)

Proto New Caledonia to Pwaamei

V > 0/_# sometimes
s t k q > t c y k
p: pw: t: ʈ: c: k: > f hʷ tʰ lʰ s h
ʈ > l/V_V
c k > y~0 0/V_V
C: > Cʰ
y > z
yʰ > s
s > h/_i
bw mw > g ŋ

Proto New Caledonia to Jawé

s ʈ k q > t̪ t̪ c k
p: pw: t: t̪: c: k: > pʰ hʷ tʰ t̪ʰ yʰ h
t̪ > l/V_V
other stops > 0/V_V
C: > Cʰ
t tʰ > c s
t̪ t̪ʰ > t tʰ
yʰ > h/_i

Proto New Caledonia to Nemi-Pije-Fwâi

s ʈ k q > t̪ t̪ c k
p: pw: t: t̪: c: k: > f hʷ tʰ t̪ʰ yʰ h
t̪ > l/V_V
other stops > 0/V_V
C: > Cʰ
t tʰ > c h/_{i,e}
t̪ t̪ʰ > t tʰ
yʰ > h/_i
bw mw > g ŋ
n nʰ > ɲ ɲʰ/_V[+front]

In Pije and Fwai, obstruent-nasal clusters became aspirated stops followed by nasal vowels. Also, instead of f they have ɸ.

Post Reply