The Correspondence Library

[Xeon]

I would like to put in a request for PIE - Sanskrit sound changes.

[Mecislau]

I would like to put in a request for PIE - Sanskrit sound changes.

Oh, no, not again...

The board just put your post on a nonexistant page. Again.

Ah, well, it should fix itself after this post.

As for the sound changes, well, I have yet to find a good source with those changes, but I'll keep looking.

EDIT: Yep, it fixed itself.

[Åge Kruger]

Anyone know the changes concerning the Northern Germanic languages? Or where to find them? (Preferably starting at Common Germanic.)

Einar Haugen: Scandinavian Language Structures: A Comparative Historical Survey. U of Minnesota Press, 1982.

It's all there, in tabulated form. A good, concentrated read.

Tack ska du ha!

Taking a lookat the University catalogue, Einar Haugen seems to have written quite the number of linguistic treats:

Blessings of Babel : bilingualism and language planning ; problems and pleasures
First grammatical treatise : the earliest Germanic phonology / an edition, translation [from the Old Norse] and commentary
Ibsen's drama : author to audience (Okay, not strictly linguistic, but it caught my eye)
Language conflict and language planning : the case of modern Norwegian
Language as a human problem

And much much more. Total 26 titles under one name, and some more under another.

[Xeon]

JOY! I sucessfully revived this wonderful thread while staying completely on-topic!

[gach]

Good that this got revived. I just got an article (Janhunen, Juha; 1981, Uralilaisen kantakielen sanastoa, [Proto Uralic vocabulary]) in which the writer reconstructs the sound changes from PU to Proto Samoyedic and Proto Finno Permic (an interesting choise but he gave good reasons why to exclude the Ugric branch; anyway Proto Finno Permic and Proto Finno Ugric cannot be very far from each other). I'll use the same othography as the article with slite changes so that:
? = /Q/
? = /7/
y = /1/
@ = /@/
?/' = palatalization
δ = /D/
x = laryngeal, could have been /? h x/ or even a vowel, it's not even present in every reconstructions

PU phonemes

Code: Select all

vowels, stressed (first) syllable
i ? y u
e     o
?     ?

vowels, unstressed syllables
i     y
?     ?

consonants
p t   k
  c
  s ś
m n ń η
  δ δ'
  r
  l
w   j
       x

PU > Proto Samoyedic

stressed syllable:
o > ? / when the following vowel didn't disappear
u > @ / when ?,? in the following syllable, or when a later disappeared y,i in the folowing syllable expect when the original syllable medial C was l or x
y > ? / in developed Proto Samoyed's closed syllables and in special vowel sequences
? > i
i > @ / before original syllable final l
e > i
? > e
plus due to irregular development and new vocabulary new phonemes
?, ? and possibly also ?

unstressed syllable:
? > ? / before original syllable medial l
?,? > @ / due to following (later disappeared) j or w
y,i > @ / after original closed stressed syllable or single δ',ń, or a following (later disappeared) syllable final j or w
y,i > ? / after original open stressed syllable or a x-initial C clusters

consonants:
s > t
ś > s
δ > r
δ' > j
l > j / word initially and syllable finally including when syllable final due to secondary development
w,j > ? / when word final in unstressed syllables
x > @ / _C
x > ? / _V
kt, ks, sk > t
śk > s
ck > c
pp > p
possibly also other C cluster changes including general shortening of geminates but the material doesn't provide any examples; PSam also possibly got secondarily a new ś


PU > Proto Finno Permic

stressed syllable:
o > u / in open syllables if y,i in the unstressed syllable
? > a
y > a
Vx > V: / _C
V: > V / _CC or if a,? in the unstressed syllable
a: > o:
?: > e:

unstressed syllable:
y > i

consonants:
There seems to be no changes petween PU and PFP, only the appearance of the new phonemes ć, l' and ?. The remaining x's must also have vocalized before the end of PFP period.


All data here is according to the cited article. Note that other sources (such as Uralisches etymologisches W?rterbuch) reconstruct the unstressed vowel system rather as /a ? e/. And others like to reconstruct long vowels rather than x into PU (though that doesn't explain the PSam @).

[Nuntar]

x = laryngeal, could have been /? h x/ or even a vowel, it's not even present in every reconstructions

Interesting.... if it was a vowel, what would it be?

[gach]

x = laryngeal, could have been /? h x/ or even a vowel, it's not even present in every reconstructions

Interesting.... if it was a vowel, what would it be?

Could have been /@/ or such. Yet another possible phonetic value to it is /G/ which I remembered only after posting the changes. It would pattern with the /D/ and /D_j/. But then /?/ would pattern perfectly with the stops /p t k/. But really, nobody has an idea what it really was.

I looked through an article on early IE loans in Uralic (Koivulehto, Jorma; 1999, Varhaiset indoeurooppalaiskontaktit: aika ja paikka lainasanojen valossa) and found one loan from PIE to PFUgr. in which a PIE laryngeal has been loaned as the Uralic /x/:

PFUgr. toxi- (or toxe- as it gave it) "to bring, to give" < PIE doH3- (sorry, no PIE meaning is given)

[hwhatting]

I looked through an article on early IE loans in Uralic (Koivulehto, Jorma; 1999, Varhaiset indoeurooppalaiskontaktit: aika ja paikka lainasanojen valossa) and found one loan from PIE to PFUgr. in which a PIE laryngeal has been loaned as the Uralic /x/:

PFUgr. toxi- (or toxe- as it gave it) "to bring, to give" < PIE doH3- (sorry, no PIE meaning is given)

That must be the root meaning "to give", which we have in Latin dare "give", do:num "gift"; Vedic dada:ti "gives", Greek dido:mi "I give", do:ron "gift", Russian dat' "give", dar "gift", etc.
Best regards,
Hans-Werner

[jmcd]

PFUgr. toxi- (or toxe- as it gave it) "to bring, to give" < PIE doH3- (sorry, no PIE meaning is given)

Or, given is no PIE meaning.

[civman2000]

Request: Proto-Germanic to Old English and also the effects of /x/ on preceding vowels in Middle English.

GOD

[Xephyr]

Does anyone have the PIE --> Classical Greek sound changes?

[boomajoom]

From IE to common Germanic:
Consonants
bh, dh, gh > B, D, G
b, d, g > p, t, k
p, t, k > f, T, x > B, D, G (except initially or following IE stress, in which case they remained f, T, x)

According to some, this happened in a certain order.
p, t, k > f, T, x left no p, t, or k, which is awkward. So b, d, g > p, t, k to fill the missing spots. Then it seemed weird to have aspirated voiced stops, but no plain voiced stops so b_h, d_h, g_h > b, d, g. Also, k_w, g_w, g_wh > hw, gw (or w), gw in that order.

[Jaaaaaa]

From IE to common Germanic:
Consonants
bh, dh, gh > B, D, G
b, d, g > p, t, k
p, t, k > f, T, x > B, D, G (except initially or following IE stress, in which case they remained f, T, x)

According to some, this happened in a certain order.
p, t, k > f, T, x left no p, t, or k, which is awkward. So b, d, g > p, t, k to fill the missing spots. Then it seemed weird to have aspirated voiced stops, but no plain voiced stops so b_h, d_h, g_h > b, d, g. Also, k_w, g_w, g_wh > hw, gw (or w), gw in that order.

Didn't they sometimes become full labials, though? I can't remember exactly how, but I do remember something like that.

[NTiOzymandias]

Then it seemed weird to have aspirated voiced stops, but no plain voiced stops so b_h, d_h, g_h > b, d, g.

Not quite. First they went to v, D, G, etc., and reverted to voiced stops inconsistently in the various daughter langs.

[Zhen Lin]

Japanese, Japonic languages

Intervocalically:
pa ? wa
pi ? i
pu ? u
pe ? o
po ? o

Monophthongisation: (via [p] ? [p\] ? [w])
apu, au ? o:
epu, eu ? jo:
ipu, iu ? ju:
opu, owo ? o:
opo ? o: (orthographical reflex, <oo?)
upu ? uu

The most obvious reflex of this change is the Kansai adverbial form:

(haya.i "early") haya + u ? hayou
(utsukushi.i "beautiful") utsukushi + u ? utsukushuu

And in the volitional / probable form:

(da(.ru) copula) da.ra + u ? darou
(de.su copula) de.se + u ? deshou
(ima.su polite aux.) ima.se + u ? imashou
(i.ku "to go") i.ka + u ? ikou

Realisation of p-series syllables in word-initial position:
[pa] ? [p\a] ? [ha]
[pi] ? [p\i] ? [hi]
[pu] ? [p\u]
[pe] ? [p\e] ? [he]
[po] ? [p\o] ? [ho]

The [p] realisation is retained in Ry?kyu languages.

Some other older sound changes:
? [w]
[d] ? [y]

This is seen in some Ry?kyu languages (specifically, Miyako, Yaeyama and Yoniguni): [ba], [da], [du] and [dama] are cognate with Japanese [wa] 1st person [ya] "house" [yu] "hot water" [yama] "mountain".

Merger:
/Zi/, /dZi/ ? [Zi] ~ [dZi] (varies by speaker)
/zu/, /dzu/ ? [zu] ~ [dzu] (varies by speaker)

Elision:
/i/ and /u/ are occasionally lost:
"to embrace" /idaku/ ? /daku/
"to leave" /ideru/ ? /deru/

Some dialects completely lose unvoiced /i/ and /u/.

/ki/ turns into /i/ in some places:
Adjectives:
"early" /hayaki/ ? /hayai/
"beautiful" /utsukushiki/ ? /utsukushii/

Verbs:
"to open (perfect past)" /akita/ ? /aita/
"to embrace (-te)" /dakite/ ? /daite/
"to swim (-te)" /oyogite/ ? /oyoide/

/ku/ turns into /u/ in the Kansai adverbial form, which leads to further sound changes, noted above.

/ri/, /ru/ is completely lost in a few places:
perfect past aux. /tari/ ? /ta/
copula (rentaikei) /naru/ ? /na/ (/naru/ is preserved is a few places)

There's a further set of sound changes for the only known sister languages of Japanese, Ry?kyu - but I'm too lazy to type them up now.

[Trebor]

cool thread; i would like to place a request for pie > albanian sound changes, and whatever comes in between.

[Drydic]

cool thread; i would like to place a request for pie > albanian sound changes, and whatever comes in between.

here they are.

1. all stops change to something else, including the original stops.
2. The vowels go batshit.
3. The Romans conquer Illyria, and introduce the GameBoy.
4. Bill Gates is born.

Thank you.

[Mecislau]

1. all stops change to something else, including the original stops.
2. The vowels go batshit.

He's right


Albanian has a bunch of really weird sound changes, and I don't think all of the conditions have been determined yet (or at least according to my books). I don't have any references of Albanian sound changes with me at the moment, but one I do remember is PIE *s > Alb. *gj (IPA /J\/), in some situations, for example, PIE *sweks > Alb. gjasht? "six"

[hwhatting]

1. all stops change to something else, including the original stops.
2. The vowels go batshit.

He's right


Albanian has a bunch of really weird sound changes, and I don't think all of the conditions have been determined yet (or at least according to my books). I don't have any references of Albanian sound changes with me at the moment, but one I do remember is PIE *s > Alb. *gj (IPA /J\/), in some situations, for example, PIE *sweks > Alb. gjasht? "six"

One "fun fact" I remember from my university times is that PIE */s/ has half a dozen different reflexes in Albanian, none of which is /s/.

[Drydic]

It would be much easier for us to figure Albanian out if we had a fair portion of Old Illyrian, which is in most scenarios a large part of what became Albanian.

/me dreams

[Whimemsz]

In other news, I've been spending the last few days going through the list of Proto-Algonquian roots (and the Ojibwe reflexes I wrote down for them) again, and this time I'm going to set up all of the correspondences so that I can know all of the changes (or basically all of them). And since I'm also in the midst of my fourth week of learning Ojibwe, I at least have a slightly better feel for how the language works, although obviously in the scheme of things I know jack shit.

Nonetheless, the point is, expect updated changes no later than, I don't know, September at the latest. Or, if I'm actually not lazy, then maybe by April.

[garrett]

thought this would be a good thread to revive.

*not really sure if my question belongs on this topic though*

Where can i find (or who knows) what phonetic restraints and consonant clusters are allowed in the algic and algonquian languages, specifically proto-algonquian?

[Mecislau]

Well, now that I have a nice new book from the library about Proto-Afro-Asiatic, I can add a bunch of AA changes here.

This is all according to Ehret.


First of all, the posited PAA phonology:

Stops: /p b t d k g k_w g_w X\_r ?/ *
Fricatives: /f s z S x G x_w G_w X\ h/
Ejectives: /p' t' ts' tS' k' k'_w/
Affricates: /ts dz tS dZ/
Nasals: /m n J N N_w/
Laterals: /l K tl' dl/
Rhotics: /4/
Glides: /j w/
Vowels: /a e i o u a: e: i: o: u:/

* /X\_r/ is a pharyngeal stop. For ease of reading, I'm just going to write it as "/?\/".


First, Proto-Afro-Asiatic (PAA) to Proto-Omotic (PO) and North Omotic (NOm). There are three fricative consonants here of unknown value, labelled s1, s2, and s3. A dash here ("-") represents the edge of a root (ie, "_C-" means "before a root-final consonant"). I'm using the Z-SAMPA tiebar ")" to avoid confusion with the underscore marking position.

C = consonant, V = vowel, N = nasal, all other capitals are X-SAMPA.

* Changes in Pre-Proto-Omotic *

dz > Z
S > s1
tS > s3
dZ > tS > S
ts > s / #_

K > l
f > p

* Depharyngealization *

a(:) > e(:) / _[?\, X\]-
?\ > ?
X\ > h

* Delabiovelarization *

a > o / #Cw)_[l, dl, s3]
Cw) > C / #_V[!i(:)]
S > s2 / [i, j]_
VNC > V:C[-voice]

* Lowering of PAA High Vowels in NOm *

u > o
i > e

* Raising of /o/, /e/, and /a/ in NOm *

e > i / #N_C
e > i / #l_C[+labial or +voice]
e > i / #b
e > i / p_r
e > i / #[s, S, ts']_[k, kw), ?]
e > i / #C_C[+labial]
e > i / #(?)_C
e > i / #[k, x, k']_[t, t', ts']
e > i / #(?)_C- (but not #(?)_CC-)
e > i / #C[+labial]_[ts, tS']
o > u / #C_C[+labial]
o > u / #(?)_C
o > u / #[k, x, k']_[t, t', ts']
o > u / #(?)_C- (but not #(?)_CC-)
o > u / #C[+labial]_[ts, tS']
a > o / #[z, dZ]_C[+labial]

e(:) > i(:) / #C[+sibilant]_[d, n, r]

* Delabialization of Remaining Labiovelars *

Cw) > C

* Shortening of NOm Vowels *

V: > V / #C[+velar, -voice]_C

* Lengthening of NOm u before Labials *

u > u: / #C[+voice, +plosive, -nasal]_C[+labial, -voice]

* Second Shortening of NOm Vowels *

V: > V / #C_C- + -(V)C- suffix

* Loss of Nasals before Other Consonants *

N > 0 / V_C[+sibilant]
N > 0 / V_p


And now changes in individual Omotic Languages:

Wolayta:

b > w / V_V
p > f
x1 > k
x2 > h / #_
x2 > 0 / V_V
x2 > k / _-#

(I'm not quite sure where these x1 and x2 came from... I'll update if I find out)

ts > s / V_
tsts > t / V_
s3 > s / V_(V)
s3 > S / V_-#
s1 > S
s2 > S / V_
J > n
l > n / #_VN
d' > t' / #_


Male:

b > w / V_V
p > f / V_
z > d / V_
z > ts / Vj_
x1 > k
x2 > h / #_
x2 > g / V_V
ts > s / V_
tsts > ts / V_
[s1, s2, s3] > S
ts' > tS' / #_
ts' > s / V_
J > n
h2 > w / #_

(Again, another odd variable consonant...)


Zayse:

b > w / V_V
p' > ?p
z > ts / Vj_
ts' > s'
x1 > k
x2 > h / #_
x2 > g / V_#
x2 > g / n_
x2 > 0 / V_V
tsts > ts / V_
[s1, s3] > S
s2 > tS / V_
ts' > tS', s
J > n
l > n / #_VN


Yem:

b > w / V_V
p > f
p' > b
z > d / V_
x1 > k
x2 > k / #_
x2 > 0 / V_V
k' > k / #_
Z > s
tS > ?j
ts > s / #_
[s1, s2, s3] > S
tS' > tS
J > n
h1 > h, 0 / #_
h2 > w / #_
l > n / #_VC
d' > t
r > r, ll / V_


Ugh, I have six more Omotic languages to do, and then several other AA languages. This'll take awhile.

[garrett]

glad to see people contributing to this thread again

[Mecislau]

More Proto-Omotic to daughterlang changes:

(Remember, most of these PO > whatever changes are very incomplete, which is why a lot of these changes look so specific or fragmented; they're coming off a correspondence chart)

Kafa:

b > w / _-#
p > f / V_
z > j / _
zz > dZdZ
s > S / !V_
ts' > tS'
x1 > k
x2 > k / #_
x2 > 0 / V_V
Z > S / #_
Z > tS / V_
ts > tS / V_
s3 > S / #_
s3 > S / V_V
s3 > s / V_-#
ts' > tS'
J > n
h2 > w / #_
l > d / #_VC
l > n / #_Vb
d' > t'


Mocha:

b > w / V_V
p' > p / V_
p' > b
z > j / #_
z > j / V_
z > dZdZ / Vj_
zz > dZdZ
s > S / !V_
ts' > tS
x1 > k
x2 > k / #_
x2 > 0 / V_V
Z > S / #_
Z > s / _-#
ts > S / V_
ts > s / _-#
s3 > S / #_
s3 > s / V_-#
s1 > S
s2 > S, tStS / V_
h1 > h, 0 / #_
h2 > w / #_
l > d / #_VC
l > n / #_Vb
d' > t'


Bench:

x1 > k
x2 > k / #_
x2 > 0 / V_V
tS > ts
s3 > S / V_V
[s1, s2, s3] > S
[?, h1, h2] > 0
l > d / #_VC
l > n / #_VN
d' > t'


Maji:

p' > b
z > d / Vj_
ts' > Z / V_
x1 > k
x2 > k / #_
x2 > 0 / V_V
Z > tS ~ ts
ts > tS / _i
[s2, s3] > tS
s1 > S
? > 0
[h1, h2] > h
d' > t'


Dime:

p > f
z > d, z / V_
k' > g' / #_
tS > ts
ts > S / i_
s3 > tS / V_
s2 > tStS / V_
s1 > S


Ari:

p' > b', p'
z > d, z / V_
[x1, x2] > g
k' > q
tS > ts
s3 > tS
s2 > z
s1 > S
h1 > 0




Now, back to descents of PAA.


These changes are for Proto-Afro-Asiatic to Proto-Erythraean (the ancestor of all non-Omotic Afro-Asiatic langs):

[dz, dZ] > dz
[ts, tS] > ts

These changes are from Proto-Erythraean to Proto-Cushitic:

b > m / #_Vn
g > k / #dV
g > k / #wV_
G > g / #_Vx-

These changes are from Proto-Erythraean to Proto-North-Erythraean (the ancestor of the Chadic, Egyptian, Semitic, and Berber languages):

[Vj, Vw] > VV / #C__C
e: > i
o: > u
[o, e, a] > a
[u, i] > @
-VC- > -CV-

(This last rule turned all VC roots into CV)

in > N / #_C



Now, here's the more interesting changes.


Proto-North-Erythraean to Proto-Chadic:

a: > a
?\ > ?
X\ > h
[t', ts'] > s'
tS' > S'
dz > z
ts > s
N > 0 / V_[ts, X\]


Proto-North-Erythraean to Proto-Boreafrasian (Egyptian, Semitic, and Berber):

s' > s
h > X\ / #_Vs
z > d (when another sibilant is in the same word nearby)
[Nw), J] > n
V > 0 / _# (in nominals)
z > d / _-# (noun-stem final)
N > 0 / #_CV (ie, deleting the nasal cluster formed in Proto-North-Erythraean)


Proto-Boreafrasian to Egypto-Berber:

p > f
p' > p


Egypto-Berber to Ancient Egyptian:

@ > i
h > X\ / _Vz
dl > d / #_VC
tl' > t' / #_VC
K > s / #_VC
t' > ts'
tS > ts
tS' > ts'
dZ' > dZ
S > s
ts' > ts
ts > s
dz > z
z > s
f > p / #_VC[+continuant]
f > p / #_VC[-voice]
f > p / #_Vr
p' > p
p > b / #dlV_
xw) > Gw) > X\
k > g / _[w, j]
C[!gw)]V?\ > X\?
gw)VX\_r > ?\ (here I'm using X\_r once again for the pharyngeal plosive, as now we're dealing with the actual fricative ?\)
[g, gw)] > dZ / #_Vd
[k, kw)] > tS / #_Vt
xVh > kh
C[+labiovelar] > [-labial]
X\_r > 0 / _i
X\_r > i / #_VC[+continuant]
X\_r > i / #_VC[+dental]
? > i / #_VC[+sonorant]
X\_ru > w / #_[f, s] (in some cases)
? > ?\ / _V[n, r, g]
X\_r > ? / C[+voice]V_
X\_r > ? / C[+dental]V_
[h, X\] > ? / C[+voice]V_
[h, X\] > ? / C[+velar]V_
X\_r > ? / h_
X\_rh > ?\X\
G > ? / X\_
tl' > dl / #_VX\r)
C[+lateral, +obstruent] > C[+palatal, +obstruent]
r > l / #_V(V)C[+obstruent, +labial]
r > ? / C[!g,m,n,r,w,S,x]_[t,w,j]#
l > j / #iC
l > r / #nV_C
l > n


Ancient Egyptian to Coptic:

n > l / #_Vb_Vb
n > l / #_(V)[s,S,h]V[m,b]#
n > l / #_V[m,b][s,S,h]
n > l / #_Vk
n > l / CV_C (where one of the consonants is /m/)
r > l / #(C)_c(C)#
r > l / #o_#



Semitic changes coming up next time.