The Correspondence Library
Here's a bit of Latin -> French; might as well, since I've already done most of the work.
Vulgar Latin
h > 0
VV > V:, where VV is a sequence of identical short vowels
n > 0 / _(f,v,s)
rs > ss
m, n > 0 / _# in polysyllables
m > n / _#
u > 0 / CC_V
w > gu (w from Germanic loanwords)
V > "V / "V(stop)r_
V > "V / _C*"(i,e)V
(i,e) > j / _V
Stressed vowels
a, a: > a
e, ae > è
e:, i, oe > é
i: > i
o > ò
o:, u* > ó
u: > u
*u stays u before i:
Judging from their reflexes, è/é and ò/ó had an open/closed distinction, something like SAMPA ɛ/e, ɔ/o.
Initial vowels (first vowel of a word)
a, a: > a
e, e:, i, ae, oe > e
i: > i
o, o:, u > o
u > u
Final vowels
a, a: > a
e, e:, i, ae, (oe) > e
i: > i
o, o:, u, u: > o (u remains u after another vowel)
k > tj / _F (F = front vowel)
g > dj / _F
è > iɛ / free (e.g. when the vowel ends the syllable; I wonder if a following stop + r doesn't also counts as a free syllable (petra > pièrre, palpebra > paupièvre, though that should have a v); also, before a final consonant
è > ɛ / checked (when a consonant ends the syllable)
ò > uo > uɛ / free (but not before a nasal)
ò > ɔ / checked
dj > dʒ / r_
dj > j
j > 0 / V_"F
j > dʒ / #_
j > dʒ / V_V (sometimes, but it usually stays j)
tj > ts / #_
tj > tsj
stj > ssj
gn, nj > ɲ
nkt > ɲt
V > 0 / when V is unstressed and in the penultimate syllable
V(!a) > 0 / when V is intertonic, i.e. immediately precedes the primary stress but is not in the initial syllable (!a = except for a)
a > ə / intertonic
0 > b / m_(r,l)
0 > d / (n,l,J,z)_r
sr > str
k > t / (n,r)_r
g > d / (n,r)_r
n > r / (g,p)_
<!--[αvcd, +obs][βvcd, +obs] > [βvcd][βvcd] (that is, two obstruents in contact with different voicing assimilate to the voicing of the second)-->
C > 0 / C_C(!l,r)
t > s / _(n,m)
gl, kl, lj > ʎ
b > v / V_(V,r)
v > 0 / V_B (B = back vowel)
p > v / V_(V,r)
p > b / _l
t, d > ð / V_(V,r)
ðr > r
ts > dz / V_V
s > z / V_V
ks > jss / V_V
ks > js / _#
k, g > 0 / V_B
k, g > 0 / B_a
k, g > j / _(a,C)
kʷ > v or u / V_F
kʷ > jv or ju / V_a
kkj, kj > ts
gj > j
pj > tʃ
bj, vj > dʒ
mɲ, mj > ndʒ
(note that the above clusters are the only case where a consonant does not receive intervocalic treatment before /j/)
V"e > "Vi
VV (identical vowels merge)
? > e / #_sC
k > tʃ / _a
g > dʒ / _a
tʃ > ʃ
dʒ > ʒ
ts > s
dz > z
ɛ > iɛ / _(Cj, jC)
ɔ > uɛ / _(Cj, jC)
0 > j / (ʃ, ʒ, sj, zj)"(a,é)_ (only in a free syllable)
ssj > jss
zj > jz
rj > jr
sj > s
ɛ > ɛa / _l(C,#)
l > u / _(C,#)
l > 0 / (i,u)_
lle, llo > u / (e,o)_# [this is actually an analogical development, but it applies as regularly as a sound law]
au, ɛau > ɔ
é > ɛi / free
é > ɛ / checked
ó > ou > ɛu / free
ó > ɔ / _[+nas]
ó > ou / checked
e > ə / #C*_..." (free)
e > ɛ / #C*_..." (checked or before another vowel)
o > ou or ɔ (the outcome fluctuates, but ɔ is often the result of analogy rather than strict sound change; always ou before another vowel)
a > ə / #(tS,dZ)_..." (free)
a > ai / _[+nas] (free)
a > ɛ / free (but a following ʎ creates a checked syllable)
kʷ > k
gʷ > g
CC > C (geminates simplify)
t > 0 / V_#
ɛ > i / _CC*i#
V(!a) > 0 / _# (except in monosyllables or after another vowel)
a > ə / _#
V > ə / _(CC(!nt, mp, ng, rt, rd), dʒ, tʃ)
s > 0 / _C
p, b > 0 / _(t, d)
v > 0 / _C, C_
ð > 0
uɛ > ɛu
u > y
ai > e / _#
ai > ɛ
ɛu > œ
iɛi > i
uɛi > yi
ou > u
ei, ɔi > oi / _[-nas]
ɔ > u / _"V
V[-high] > ə > 0 / _V (except that a is kept before o)
ɲ > in / _(C,#)
V(n, m) > V~ / _(C,#)
ɛ~ > a~
ai~, ei~, i~ > ɛ~
y~ > œ~
obs[+vcd] > [-vcd] / _#
t, s > 0 / _#
k > 0 / V~_#
n, m > 0 / C_#
j > 0 / (ʃ,ʒ)_V[-nas]
ʎ > j
r > ʁ
oi > wɛ > wa
oi~ > wɛ~
(I've omitted the loss of ə in various contexts, since it often resurfaces)
V = vowel
C = consonant
F = front vowel
B = back vowel
X* = zero or more X
" = primary stress
Vulgar Latin
h > 0
VV > V:, where VV is a sequence of identical short vowels
n > 0 / _(f,v,s)
rs > ss
m, n > 0 / _# in polysyllables
m > n / _#
u > 0 / CC_V
w > gu (w from Germanic loanwords)
V > "V / "V(stop)r_
V > "V / _C*"(i,e)V
(i,e) > j / _V
Stressed vowels
a, a: > a
e, ae > è
e:, i, oe > é
i: > i
o > ò
o:, u* > ó
u: > u
*u stays u before i:
Judging from their reflexes, è/é and ò/ó had an open/closed distinction, something like SAMPA ɛ/e, ɔ/o.
Initial vowels (first vowel of a word)
a, a: > a
e, e:, i, ae, oe > e
i: > i
o, o:, u > o
u > u
Final vowels
a, a: > a
e, e:, i, ae, (oe) > e
i: > i
o, o:, u, u: > o (u remains u after another vowel)
k > tj / _F (F = front vowel)
g > dj / _F
è > iɛ / free (e.g. when the vowel ends the syllable; I wonder if a following stop + r doesn't also counts as a free syllable (petra > pièrre, palpebra > paupièvre, though that should have a v); also, before a final consonant
è > ɛ / checked (when a consonant ends the syllable)
ò > uo > uɛ / free (but not before a nasal)
ò > ɔ / checked
dj > dʒ / r_
dj > j
j > 0 / V_"F
j > dʒ / #_
j > dʒ / V_V (sometimes, but it usually stays j)
tj > ts / #_
tj > tsj
stj > ssj
gn, nj > ɲ
nkt > ɲt
V > 0 / when V is unstressed and in the penultimate syllable
V(!a) > 0 / when V is intertonic, i.e. immediately precedes the primary stress but is not in the initial syllable (!a = except for a)
a > ə / intertonic
0 > b / m_(r,l)
0 > d / (n,l,J,z)_r
sr > str
k > t / (n,r)_r
g > d / (n,r)_r
n > r / (g,p)_
<!--[αvcd, +obs][βvcd, +obs] > [βvcd][βvcd] (that is, two obstruents in contact with different voicing assimilate to the voicing of the second)-->
C > 0 / C_C(!l,r)
t > s / _(n,m)
gl, kl, lj > ʎ
b > v / V_(V,r)
v > 0 / V_B (B = back vowel)
p > v / V_(V,r)
p > b / _l
t, d > ð / V_(V,r)
ðr > r
ts > dz / V_V
s > z / V_V
ks > jss / V_V
ks > js / _#
k, g > 0 / V_B
k, g > 0 / B_a
k, g > j / _(a,C)
kʷ > v or u / V_F
kʷ > jv or ju / V_a
kkj, kj > ts
gj > j
pj > tʃ
bj, vj > dʒ
mɲ, mj > ndʒ
(note that the above clusters are the only case where a consonant does not receive intervocalic treatment before /j/)
V"e > "Vi
VV (identical vowels merge)
? > e / #_sC
k > tʃ / _a
g > dʒ / _a
tʃ > ʃ
dʒ > ʒ
ts > s
dz > z
ɛ > iɛ / _(Cj, jC)
ɔ > uɛ / _(Cj, jC)
0 > j / (ʃ, ʒ, sj, zj)"(a,é)_ (only in a free syllable)
ssj > jss
zj > jz
rj > jr
sj > s
ɛ > ɛa / _l(C,#)
l > u / _(C,#)
l > 0 / (i,u)_
lle, llo > u / (e,o)_# [this is actually an analogical development, but it applies as regularly as a sound law]
au, ɛau > ɔ
é > ɛi / free
é > ɛ / checked
ó > ou > ɛu / free
ó > ɔ / _[+nas]
ó > ou / checked
e > ə / #C*_..." (free)
e > ɛ / #C*_..." (checked or before another vowel)
o > ou or ɔ (the outcome fluctuates, but ɔ is often the result of analogy rather than strict sound change; always ou before another vowel)
a > ə / #(tS,dZ)_..." (free)
a > ai / _[+nas] (free)
a > ɛ / free (but a following ʎ creates a checked syllable)
kʷ > k
gʷ > g
CC > C (geminates simplify)
t > 0 / V_#
ɛ > i / _CC*i#
V(!a) > 0 / _# (except in monosyllables or after another vowel)
a > ə / _#
V > ə / _(CC(!nt, mp, ng, rt, rd), dʒ, tʃ)
s > 0 / _C
p, b > 0 / _(t, d)
v > 0 / _C, C_
ð > 0
uɛ > ɛu
u > y
ai > e / _#
ai > ɛ
ɛu > œ
iɛi > i
uɛi > yi
ou > u
ei, ɔi > oi / _[-nas]
ɔ > u / _"V
V[-high] > ə > 0 / _V (except that a is kept before o)
ɲ > in / _(C,#)
V(n, m) > V~ / _(C,#)
ɛ~ > a~
ai~, ei~, i~ > ɛ~
y~ > œ~
obs[+vcd] > [-vcd] / _#
t, s > 0 / _#
k > 0 / V~_#
n, m > 0 / C_#
j > 0 / (ʃ,ʒ)_V[-nas]
ʎ > j
r > ʁ
oi > wɛ > wa
oi~ > wɛ~
(I've omitted the loss of ə in various contexts, since it often resurfaces)
V = vowel
C = consonant
F = front vowel
B = back vowel
X* = zero or more X
" = primary stress
Last edited by pharazon on Mon Jun 21, 2004 6:04 pm, edited 1 time in total.
Ooh, that's not what I meant... The Latin geminate r.r becomes the Spanish trilled /r/.pharazon wrote:Maknas wrote:rr > r / _rr doesn't merge with r: caro /ka4o/ vs. carro /karo/, pero vs perro, etc.Maknas wrote:Yeah, you missed a rule given a bit further down: OO > O / _ (if O = O)
Latin ipse > esse > ese
Latin capt?re > cattar > catar
I'll add it, but if I may ask, why is that a pronounced [aj]? /lajtu:ka/?pharazon wrote:A missing one that came to mind is aj -> e, which comes sometime after kt -> jt: lactu:ca > lechuga.
Mind giving some of the evolution of the open/close distinction?pharazon wrote:You've skipped a few notable steps in the vowel development (e.g. evolution of an open/close distinction), but it's not a big problem in a sketch like this. One important thing, though, is that VL open e and o (i.e. CL ae/e, o) don't diphthongize to ie/ue before a palatal cluster; that is, CiV or a consonant cluster that yields jC: lectus > lecho, not *liecho, folia > hoja, etc.
Yay!pharazpn wrote:I think I'll give French a shot.
http://www.veche.net/
http://www.veche.net/novegradian - Grammar of Novegradian
http://www.veche.net/alashian - Grammar of Alashian
http://www.veche.net/novegradian - Grammar of Novegradian
http://www.veche.net/alashian - Grammar of Alashian
I thought you might've meant that, but you didn't have something like r > 4.Maknas wrote:Ooh, that's not what I meant... The Latin geminate r.r becomes the Spanish trilled /r/.pharazon wrote:Maknas wrote:rr > r / _rr doesn't merge with r: caro /ka4o/ vs. carro /karo/, pero vs perro, etc.Maknas wrote:Yeah, you missed a rule given a bit further down: OO > O / _ (if O = O)
Latin ipse > esse > ese
Latin capt?re > cattar > catar
Oh, I see; you wroteMaknas wrote:I'll add it, but if I may ask, why is that a pronounced [aj]? /lajtu:ka/?pharazon wrote:A missing one that came to mind is aj -> e, which comes sometime after kt -> jt: lactu:ca > lechuga.
aj > e happens after the intermediate jt, but before tS. Same with things like ks > js > S.Maknas wrote:kt > jt > tS / V_V
Vulgar Latin replaced the CL long/short vowel contrast with an open/closed one (these are for stressed vowels):Maknas wrote:Mind giving some of the evolution of the open/close distinction?
a, a: > a
e, ae > ? (open)
e:, i, oe > ? (closed)
i: > i
o > ?
o:, u > ?
u: > u
But Sardinian and Romanian used different systems: Sardinian simply merged long and short pairs; Romanian followed the above system except that o and o: both went to o and u, u: to u.
I'll give it a try; it may take a few days though.Allie wrote:Has anyone got the Latin to Romanian sound changes? They would be extremely useful.
I'll need the Latin > Romanian changes too...
Anyways, I found a really useful book at my local college library, and I made a bunch of photocopies of some sound changes listed in it. I'll post 'em here whence I type them all up. But for now, here's PIE > Armenian:
C = consonants
V = vowels
N = nasals
L = liquids
S = sonorants
? = nasalized vowels
Proto-Indo-European > Armenian
[u:, u, o:] > u
[ej, oj] > Ej
Ew > ow
o > a (rare)
[a:, a] > a
e > a (rare)
e > E
[e:, i, i:] > i
E > i / _N
o > u / _N
ej > e / _
ia > Ea
[i, u] > @ (in some unstressed syllables)
e > i (in unstressed syllables)
oy > u (in unstressed syllables)
Ea > E (in unstressed syllables)
b > p? > p / _
d > t? > t / _
g_j > k_j? > tS? > c / _
g > k? > k / _
g_w > k_w? > k / _
bh > b / #_
bh > w, v / _
dh > d / _
g_jh > g_j > Z > j / #_
g_jh > g_j > Z > z / _
gh > g > g / _
g_wh > g_w > dZ / #_[e, i]
g_wh > g_w > g / #_
g_wh > g_w > Z / _
p > h, ? / #_
p > w, v / _
t > t? / #_
t > t? / [aw, ow]_
t > d / [N, L]_
y > w / _o
y > ? / _
k_j > tS > s / _
k > g / [N, L]_
k > k? / _
k_w > tS? / _[e, i]
k_w > g / [N, L]_
k_w > k? / _
[sk, ks] > c? / _
kj > tS? / _
[sr, rs] > r: / _
r > r: / _N
l > 5 / C_#, C_, V_V, lV_
Ns > s / _
sN > N / _
N > w / S_S
VN > ? > V > ? / _# (in polysyllable words)
N > n / _# (in monosyllable words)
N= > n / _#
m= > am / _
n= > an / _
r= > ar / _
l= > a5 / _
V > ? / _(C)#
(I'm using _j for the palatal stops here)
I don't guarantee that these are all perfectly acurate, though... Although the book had very good info, the organization was crappy...
When I get around to it, I'll post PIE > Latin, PIE > Common Celtic, PIE > Albanian, and PIE > Tocharian.
Anyways, I found a really useful book at my local college library, and I made a bunch of photocopies of some sound changes listed in it. I'll post 'em here whence I type them all up. But for now, here's PIE > Armenian:
C = consonants
V = vowels
N = nasals
L = liquids
S = sonorants
? = nasalized vowels
Proto-Indo-European > Armenian
[u:, u, o:] > u
[ej, oj] > Ej
Ew > ow
o > a (rare)
[a:, a] > a
e > a (rare)
e > E
[e:, i, i:] > i
E > i / _N
o > u / _N
ej > e / _
ia > Ea
[i, u] > @ (in some unstressed syllables)
e > i (in unstressed syllables)
oy > u (in unstressed syllables)
Ea > E (in unstressed syllables)
b > p? > p / _
d > t? > t / _
g_j > k_j? > tS? > c / _
g > k? > k / _
g_w > k_w? > k / _
bh > b / #_
bh > w, v / _
dh > d / _
g_jh > g_j > Z > j / #_
g_jh > g_j > Z > z / _
gh > g > g / _
g_wh > g_w > dZ / #_[e, i]
g_wh > g_w > g / #_
g_wh > g_w > Z / _
p > h, ? / #_
p > w, v / _
t > t? / #_
t > t? / [aw, ow]_
t > d / [N, L]_
y > w / _o
y > ? / _
k_j > tS > s / _
k > g / [N, L]_
k > k? / _
k_w > tS? / _[e, i]
k_w > g / [N, L]_
k_w > k? / _
[sk, ks] > c? / _
kj > tS? / _
[sr, rs] > r: / _
r > r: / _N
l > 5 / C_#, C_, V_V, lV_
Ns > s / _
sN > N / _
N > w / S_S
VN > ? > V > ? / _# (in polysyllable words)
N > n / _# (in monosyllable words)
N= > n / _#
m= > am / _
n= > an / _
r= > ar / _
l= > a5 / _
V > ? / _(C)#
(I'm using _j for the palatal stops here)
I don't guarantee that these are all perfectly acurate, though... Although the book had very good info, the organization was crappy...
When I get around to it, I'll post PIE > Latin, PIE > Common Celtic, PIE > Albanian, and PIE > Tocharian.
http://www.veche.net/
http://www.veche.net/novegradian - Grammar of Novegradian
http://www.veche.net/alashian - Grammar of Alashian
http://www.veche.net/novegradian - Grammar of Novegradian
http://www.veche.net/alashian - Grammar of Alashian
Latin > Romanian
Vulgar Latin
h > 0
VV > V:, where VV is a sequence of identical short vowels
n > 0 / _(f,v,s)
rs > ss
m, n > 0 / _# in polysyllables
m > n / _#
u > 0 / CC_V
s > 0 / _# in polysyllables
s > i / _#
V > "V / "V(stop)r_
V > "V / _C*"(i,e)V
(i,e) > j / _V
Stressed vowels (note the difference from the French development)
a, a: > a
e, ae > E
e:, i, oe > e
i: > i
o, o: > o
u, u: > u
Initial vowels
a, a: > a
e, e:, i, ae, oe > e
i: > i
o, o:, u > o
u > u
Final vowels
a, a: > a
e, e:, i, ae, (oe) > e
i: > i
o, o:, u, u: > o (u remains u after another vowel)
s > i / #C*V_#
s > 0 / _#
t > 0 / _#
ks > s / _#
kw > p / V_a
gw > b / V_a
gn > mn
kw > k
gw > g
k > tS / _F
g > dZ / _F
E > ie
t > ts / _i
d > dz / _i
s > S / _i
o > u / _[+nas]
e > i / _[+nas]
a > 1 / _[+nas](!nn,mn)
ie > e / [+obs][+liq]_
ie > ia / _C*(a,e)#
E > ea / "_C*(a,e)#
o > oa / "_C*(a,e)#
i > 0 / (ts,dz,S)_V
li > lj / _V
l > lj / _i
ll > 0 / _i
"ella > e"a / _#
b, v > 0 / V_(V,t)
l > r / V_V
mn > un / a_V
b > u / V_(l,r)
ps > s
c > p / _(s,t)
sc, st > St / _F
skl > Skl
nct > nt
kj > tS
gj > j (sometimes 0)
[+lab]j > j[+lab]
stj, sj, ssj > S
tj > ts
dj > dz / r_
dj > Z / _"B
dj > z / V_V
nj, lj > j
rja > re / _#
rj > r
dz > z
kl > kj
gl > gj
a > e / [+pal]_#
o > u / "..._#
a > @ / "..._#
e > @ / "...[+lab]_C*V(!i)#
oe > oue
e > @ / ou_#
u > 0 / (![+obs][+liq], "V)_#
a > e / #C*V...[+pal]_..."
a > @ / #C*V..._..."
o > u / #C*V..._..."
e > i / #C*V..._n..."
e > @ / #C*V...(t,d,n)_..."
e > @ / [+lab]"_C*B
ea > a / [+lab]_
e > @ / #(r, [+lab])_..."
o > u / #C*_..."
a > @ / #CC*_..."
au > @u / #CC*_..."
nt, nd > n / _#
i > _j / _# (unless preceded by a C(sonorant) cluster)
This is probably missing some minor changes, and maybe some are out of order, because I didn't spend nearly as much time on these as the French one. They should be pretty accurate though.
If people are wondering, my (main) source for these is Jensen's A Comparative Study of Romance.
Vulgar Latin
h > 0
VV > V:, where VV is a sequence of identical short vowels
n > 0 / _(f,v,s)
rs > ss
m, n > 0 / _# in polysyllables
m > n / _#
u > 0 / CC_V
s > 0 / _# in polysyllables
s > i / _#
V > "V / "V(stop)r_
V > "V / _C*"(i,e)V
(i,e) > j / _V
Stressed vowels (note the difference from the French development)
a, a: > a
e, ae > E
e:, i, oe > e
i: > i
o, o: > o
u, u: > u
Initial vowels
a, a: > a
e, e:, i, ae, oe > e
i: > i
o, o:, u > o
u > u
Final vowels
a, a: > a
e, e:, i, ae, (oe) > e
i: > i
o, o:, u, u: > o (u remains u after another vowel)
s > i / #C*V_#
s > 0 / _#
t > 0 / _#
ks > s / _#
kw > p / V_a
gw > b / V_a
gn > mn
kw > k
gw > g
k > tS / _F
g > dZ / _F
E > ie
t > ts / _i
d > dz / _i
s > S / _i
o > u / _[+nas]
e > i / _[+nas]
a > 1 / _[+nas](!nn,mn)
ie > e / [+obs][+liq]_
ie > ia / _C*(a,e)#
E > ea / "_C*(a,e)#
o > oa / "_C*(a,e)#
i > 0 / (ts,dz,S)_V
li > lj / _V
l > lj / _i
ll > 0 / _i
"ella > e"a / _#
b, v > 0 / V_(V,t)
l > r / V_V
mn > un / a_V
b > u / V_(l,r)
ps > s
c > p / _(s,t)
sc, st > St / _F
skl > Skl
nct > nt
kj > tS
gj > j (sometimes 0)
[+lab]j > j[+lab]
stj, sj, ssj > S
tj > ts
dj > dz / r_
dj > Z / _"B
dj > z / V_V
nj, lj > j
rja > re / _#
rj > r
dz > z
kl > kj
gl > gj
a > e / [+pal]_#
o > u / "..._#
a > @ / "..._#
e > @ / "...[+lab]_C*V(!i)#
oe > oue
e > @ / ou_#
u > 0 / (![+obs][+liq], "V)_#
a > e / #C*V...[+pal]_..."
a > @ / #C*V..._..."
o > u / #C*V..._..."
e > i / #C*V..._n..."
e > @ / #C*V...(t,d,n)_..."
e > @ / [+lab]"_C*B
ea > a / [+lab]_
e > @ / #(r, [+lab])_..."
o > u / #C*_..."
a > @ / #CC*_..."
au > @u / #CC*_..."
nt, nd > n / _#
i > _j / _# (unless preceded by a C(sonorant) cluster)
This is probably missing some minor changes, and maybe some are out of order, because I didn't spend nearly as much time on these as the French one. They should be pretty accurate though.
If people are wondering, my (main) source for these is Jensen's A Comparative Study of Romance.
Last edited by pharazon on Mon Jun 05, 2006 8:42 am, edited 2 times in total.
Um, Phar, quick questions about your Latin > French changes:
(mabye it's your notation here that's confusing me)
Sorry for my lack of understanding...
EDIT: Another:
EDIT2:
EDIT3: (last one, I promise! )
EDIT4: (this is about Romanian, not French, so hah! )
EDIT5:
Mind explaining what you mean by the last two?pharazon wrote:V > ? / "..._...V (i.e. unstressed vowels in the penultimate syllable are dropped)
V(!a) > ? / #C*V..._CC*" (!a = except for a)
a > @ / #C*V..._CC*"
?pharazon wrote:[αvcd, +obs][βvcd, +obs] > [βvcd][βvcd] (that is, two obstruents in contact with different voicing assimilate regressively)
(mabye it's your notation here that's confusing me)
What do you mean by "ll(e,o)"? Do you mean "lle"/ "llo" become u after e or o?pharazon wrote:ll(e,o) > u / (e,o)_ [this is actually an analogical development, but it applies as regularly as a sound law]
Sorry for my lack of understanding...
EDIT: Another:
Where do the tS and dZ in the last change come from, if only a few stages before, tS and dZ disappeared?pharazon wrote:tS > S
dZ > Z
[...]
? > j / (tS, dZ, sj, zj)"(a,?)_ (free)
EDIT2:
Where did the w come from? W's were already taken care of, and the 'qu' (Latin /k_w/, right?) already changed.pharazon wrote:kw > k
gw > g
EDIT3: (last one, I promise! )
What do you mean by 'high'?pharazon wrote:V[-high] > @ > ? / _V (except that a is kept before o)
EDIT4: (this is about Romanian, not French, so hah! )
What is ?? Do you mean ?? /1/pharazon wrote:a > ? / _[+nas](!nn,mn)
EDIT5:
Since when is /i/ a consonant? I'm just asking to make sure you didn't forget a V or something there...pharazon wrote:e > @ / "...[+lab]_C*(!i)#
http://www.veche.net/
http://www.veche.net/novegradian - Grammar of Novegradian
http://www.veche.net/alashian - Grammar of Alashian
http://www.veche.net/novegradian - Grammar of Novegradian
http://www.veche.net/alashian - Grammar of Alashian
*sigh*
I've finally finished typing up the PIE > Latin sound changes. Here you go:
C = consonants
V = vowels
P = labials
A = diphthongs
T = stops
F = fricatives
N = nasals
Proto-Indo-European > Latin
e > i / _N (as in -nk, -ng, -kn)
e > o / _5
o > u / _N
o > u / _mb, _mk, _5
o > e / w_[r, s, t]
o: > u: / _r
aj > ai > ae / _ (or > e: in rustic dialects)
ej > e: > i: / _ (but stays at e: in rustic dialects)
oj > oi > oe > u: / _
aw > au / _ (or > o: in rustic dialects)
[ew, ow] > ou > u: / _
V > i / $(C)(C)_$ (short vowels to i in unstressed medial syllables)
V > o / $(C)(C)V_$ (short vowels to o in unstressed medial syllables when preceded by a vowel)
V > e / $(C)(C)_$r (this change has some exceptions; syllables are always unstressed)
V > i, u / $(C)(C)_$P (unstressed)
a > e / $(C)(C)_C(C)$ (unstressed)
o > u / $(C)(C)_C(C)$ (unstressed)
a > e > i / $(C)(C)_N (unstressed)
a > e > u / $(C)(C)_5 (unstressed)
e > u / $(C)(C)_5 (unstressed)
ai > ei > i: / $(C)(C)_ (unstressed)
ei > i: / $(C)(C)_ (unstressed)
oi > e: / $(C)(C)_ (unstressed)
ou > u: / $(C)(C)_ (unstressed)
i > e / _#
o > e / _#
[i, e] > ? / _# (sometimes)
a > e / _C(C)#
e > i / _[s, t]#
o > u / _C(C)# (unless preceded by u or w)
[ai, ei, oi] > ei > i: / _(C)(C)#
V: > V / _[m, t, nt, l, r]#
V: > V / _#
j > i / C_
w > u / t_
e > o / _w, w_
w > ? / s_o
m= > em / _
n= > en / _
n:= > n / _ (eg, PIE gn:tos > Latin gn?tus)
l= > ol / _
l:= > l / _ (eg, PIE wl:na > Old Latin wl?na > Latin l?na)
r= > or / _
r:= > r / _ (eg, PIE grnom > Latin gr?num)
bh > f / #_, (> h in rustic dialects)
bh > b / _
tl > kl / _
t > ? / C_#
t > d / V_
dw > b / _
d > ? / V:_#
d > ? C_
d > l / _ (in many dialects) (eg, PIE dnghum > Old Latin dingua > Latin lingua)
dh > f / #_
dh > b / rV_, _Vr, _l, u:_
dh > d / _
k_j > k / _
g_j > g / _
gh > g / N_
gh > h / _
kw > ? / C_C
kw > k / _[o, i, C]
[gw, gwh] > gu / N_
gw > g / _V[l, r]
gw > v / _
gwh > f / #_
gwh > f / _r
gwh > v / V_V
s > z > r / V_V
sr > Tr > fr / #_ (/T/ in Latin!?)
sr > Tr > br / _
s > z / _C[+vcd]
V > V: / _zC[+vcd]
z > ? / _C[+vcd]
p > kw / _ ? kw
V1 > V2 / _ ? V2 (sometimes)
V > V: / _T[+vcd][T[-vcd], F[-vcd]] (i, e, and o sometimes bypass this)
T[+vcd] > [-vcd] / _[T[-vcd], F[-vcd]]
T[-vcd] > [+vcd] / _N
s > z / _N
s > z / _[l, r]
V > V: / _z[l, r]
z > ? / _[l, r]
T > f / _f
[t, d] > s / _s
[p, b] > m / _[n, m]
[t, d] > n / _[n, m]
[k, g] > N / _n
mm > n / [A, V:]_
[d, n, r] > l / _l
n > r / _r
s > z > l / l_
s > z > r / r_
n > l / l_
V > V: / _[dt, tt]
[dt, tt] > tst > ss / _
ssr > sstr / _
ml > mpl / _
ms > mps / _
mt > mpt / _
ss > s / _#
ss > s / [A, V:]_
l > r / _Vl
l > r / lV_ (in suffixes with l if root already has l) (eg, *milit-alis > militaris)
r > ? / r ? _
n > r / _ ? r
d > r / _ ? r
V: > V / _C(C)# (irregular: often before -m, -t, -nt, but never before ?s)
V > V:, V~ / _[nf, ns]
n > ? / V~
Nasalized vowels in Latin! Apparently, they were treated as the same as long vowels. This also explains early changes in Latin such as OLat. *lupons > Lat. lup?s, or OLat. puppins > pupp?s. It also explains early spelling variations, such as c?sul ~ c?nsul, or c?sor ~ c?nsor. Apparently the orthographic N was re-introduced in order to "standardize and maintain... analogical and etymological regularity." The preceding vowel, though, seems to have kept its new length.
C1C2C3C4 > C3C4, C4 / _
C1C2C3 > C1C3 / _
Cluster simplifications. For example, *?rds? > ?rsi, *fulgmen > fulmen, *lowksna > l?na, *stl?tis > l?s
There are also some haplology, analogy, and metathesis changes that I can't really explain...
If you're curious, my sources are mostly from "The Indo-European Languages", edited by Anna Giacalore Ramat and Paolo Ramat.
I've finally finished typing up the PIE > Latin sound changes. Here you go:
C = consonants
V = vowels
P = labials
A = diphthongs
T = stops
F = fricatives
N = nasals
Proto-Indo-European > Latin
e > i / _N (as in -nk, -ng, -kn)
e > o / _5
o > u / _N
o > u / _mb, _mk, _5
o > e / w_[r, s, t]
o: > u: / _r
aj > ai > ae / _ (or > e: in rustic dialects)
ej > e: > i: / _ (but stays at e: in rustic dialects)
oj > oi > oe > u: / _
aw > au / _ (or > o: in rustic dialects)
[ew, ow] > ou > u: / _
V > i / $(C)(C)_$ (short vowels to i in unstressed medial syllables)
V > o / $(C)(C)V_$ (short vowels to o in unstressed medial syllables when preceded by a vowel)
V > e / $(C)(C)_$r (this change has some exceptions; syllables are always unstressed)
V > i, u / $(C)(C)_$P (unstressed)
a > e / $(C)(C)_C(C)$ (unstressed)
o > u / $(C)(C)_C(C)$ (unstressed)
a > e > i / $(C)(C)_N (unstressed)
a > e > u / $(C)(C)_5 (unstressed)
e > u / $(C)(C)_5 (unstressed)
ai > ei > i: / $(C)(C)_ (unstressed)
ei > i: / $(C)(C)_ (unstressed)
oi > e: / $(C)(C)_ (unstressed)
ou > u: / $(C)(C)_ (unstressed)
i > e / _#
o > e / _#
[i, e] > ? / _# (sometimes)
a > e / _C(C)#
e > i / _[s, t]#
o > u / _C(C)# (unless preceded by u or w)
[ai, ei, oi] > ei > i: / _(C)(C)#
V: > V / _[m, t, nt, l, r]#
V: > V / _#
j > i / C_
w > u / t_
e > o / _w, w_
w > ? / s_o
m= > em / _
n= > en / _
n:= > n / _ (eg, PIE gn:tos > Latin gn?tus)
l= > ol / _
l:= > l / _ (eg, PIE wl:na > Old Latin wl?na > Latin l?na)
r= > or / _
r:= > r / _ (eg, PIE grnom > Latin gr?num)
bh > f / #_, (> h in rustic dialects)
bh > b / _
tl > kl / _
t > ? / C_#
t > d / V_
dw > b / _
d > ? / V:_#
d > ? C_
d > l / _ (in many dialects) (eg, PIE dnghum > Old Latin dingua > Latin lingua)
dh > f / #_
dh > b / rV_, _Vr, _l, u:_
dh > d / _
k_j > k / _
g_j > g / _
gh > g / N_
gh > h / _
kw > ? / C_C
kw > k / _[o, i, C]
[gw, gwh] > gu / N_
gw > g / _V[l, r]
gw > v / _
gwh > f / #_
gwh > f / _r
gwh > v / V_V
s > z > r / V_V
sr > Tr > fr / #_ (/T/ in Latin!?)
sr > Tr > br / _
s > z / _C[+vcd]
V > V: / _zC[+vcd]
z > ? / _C[+vcd]
p > kw / _ ? kw
V1 > V2 / _ ? V2 (sometimes)
V > V: / _T[+vcd][T[-vcd], F[-vcd]] (i, e, and o sometimes bypass this)
T[+vcd] > [-vcd] / _[T[-vcd], F[-vcd]]
T[-vcd] > [+vcd] / _N
s > z / _N
s > z / _[l, r]
V > V: / _z[l, r]
z > ? / _[l, r]
T > f / _f
[t, d] > s / _s
[p, b] > m / _[n, m]
[t, d] > n / _[n, m]
[k, g] > N / _n
mm > n / [A, V:]_
[d, n, r] > l / _l
n > r / _r
s > z > l / l_
s > z > r / r_
n > l / l_
V > V: / _[dt, tt]
[dt, tt] > tst > ss / _
ssr > sstr / _
ml > mpl / _
ms > mps / _
mt > mpt / _
ss > s / _#
ss > s / [A, V:]_
l > r / _Vl
l > r / lV_ (in suffixes with l if root already has l) (eg, *milit-alis > militaris)
r > ? / r ? _
n > r / _ ? r
d > r / _ ? r
V: > V / _C(C)# (irregular: often before -m, -t, -nt, but never before ?s)
V > V:, V~ / _[nf, ns]
n > ? / V~
Nasalized vowels in Latin! Apparently, they were treated as the same as long vowels. This also explains early changes in Latin such as OLat. *lupons > Lat. lup?s, or OLat. puppins > pupp?s. It also explains early spelling variations, such as c?sul ~ c?nsul, or c?sor ~ c?nsor. Apparently the orthographic N was re-introduced in order to "standardize and maintain... analogical and etymological regularity." The preceding vowel, though, seems to have kept its new length.
C1C2C3C4 > C3C4, C4 / _
C1C2C3 > C1C3 / _
Cluster simplifications. For example, *?rds? > ?rsi, *fulgmen > fulmen, *lowksna > l?na, *stl?tis > l?s
There are also some haplology, analogy, and metathesis changes that I can't really explain...
If you're curious, my sources are mostly from "The Indo-European Languages", edited by Anna Giacalore Ramat and Paolo Ramat.
http://www.veche.net/
http://www.veche.net/novegradian - Grammar of Novegradian
http://www.veche.net/alashian - Grammar of Alashian
http://www.veche.net/novegradian - Grammar of Novegradian
http://www.veche.net/alashian - Grammar of Alashian
Oh, I forget to mention " indicates the stress. Thus, all vowels except /a/ drop when between the first vowel and the stress: quaesi"t?rem > *quest?re > qu?teur.Maknas wrote:Um, Phar, quick questions about your Latin > French changes:
Mind explaining what you mean by the last two?pharazon wrote:V > ? / "..._...V (i.e. unstressed vowels in the penultimate syllable are dropped)
V(!a) > ? / #C*V..._CC*" (!a = except for a)
a > @ / #C*V..._CC*"
Well, the rule is just like the voicing assimilation in Russian: sdjelat' > /zd'elat'/. People should probably just ignore that one though; I've never found an example for it, and I have no clue now why I even put it in . Probably just wanted to use some fancy symbols.Maknas wrote:?pharazon wrote:[?vcd, +obs][?vcd, +obs] > [?vcd][?vcd] (that is, two obstruents in contact with different voicing assimilate regressively)
(mabye it's your notation here that's confusing me)
Yes: pellem > peau (analogized from the phonological development pellos > peau).Maknas wrote:What do you mean by "ll(e,o)"? Do you mean "lle"/ "llo" become u after e or o?pharazon wrote:ll(e,o) > u / (e,o)_ [this is actually an analogical development, but it applies as regularly as a sound law]
Sorry, should be (S, Z, sj, zj)Maknas wrote:Where do the tS and dZ in the last change come from, if only a few stages before, tS and dZ disappeared?pharazon wrote:tS > S
dZ > Z
[...]
? > j / (tS, dZ, sj, zj)"(a,?)_ (free)
I miswrote the earlier rules about /k_w/; they're supposed to be intervocalic. /k_w/ stays until then in other positions. Plus I wrote w > gu, which should really be w > g_w; I was thinking of Latin orthography (g_w so that the rule u > ? / CC_V doesn't eliminate it in cases like sangue). I'll edit it.Maknas wrote:Where did the w come from? W's were already taken care of, and the 'qu' (Latin /k_w/, right?) already changed.pharazon wrote:kw > k
gw > g
V[-high] means vowels that aren't high.Maknas wrote:What do you mean by 'high'?pharazon wrote:V[-high] > @ > ? / _V (except that a is kept before o)
Damn, I got used to my conlang's use of ? for /1/.Maknas wrote:What is ?? Do you mean ?? /1/pharazon wrote:a > ? / _[+nas](!nn,mn)
I suppose I should've had a V there so my notation stays consistent: e > @ / "...[+lab]_C*V(!i)# ; that is, e > @ when it's in the penultimate syllable, is preceded by a labial, and the final syllable doesn't have i.Maknas wrote:Since when is /i/ a consonant? I'm just asking to make sure you didn't forget a V or something there...pharazon wrote:e > @ / "...[+lab]_C*(!i)#
Also, I forgot a fairly recent change in Romanian, which will be edited in.
I've got that one. What's the university around there?Maknas wrote:If you're curious, my sources are mostly from "The Indo-European Languages", edited by Anna Giacalore Ramat and Paolo Ramat.
But what is "high"? I don't recognize the term...pharazon wrote:V[-high] means vowels that aren't high.Maknas wrote:What do you mean by 'high'?pharazon wrote:V[-high] > @ > ? / _V (except that a is kept before o)
Um, mind saying what it is here? I've been copying down all of the changes in this thread (in my own notation), and I don't want to parse through the Romanian changes to find the difference...pharazon wrote:Also, I forgot a fairly recent change in Romanian, which will be edited in.
Maryland.pharazon wrote:I've got that one. What's the university around there?Maknas wrote:If you're curious, my sources are mostly from "The Indo-European Languages", edited by Anna Giacalore Ramat and Paolo Ramat.
Speaking of Romanian, I can give you some modern changes, if you wish (which Metaforik taught me when he heard my Romanian pronounciation):
1 > u / ? (I don't know the rules here... for example, older s?nt 'I am' is now sunt /sunt/, and [s1nt] is considered improper)
l > ? / _# (such as in the definite ending -ul: ceasul 'the watch' [tSasu])
I'll get back to you when I think of more...
Also, question: You say that Romanian /1/ came from: a > 1 / _N (if N ≠ nn, mm). But, Metaforik says the Romanian orthography represents the etymology: < ? > came from Latin /a/, while < ? > came from Latin /i/. You give the rules for a > ?, but then where did the ? come from?
http://www.veche.net/
http://www.veche.net/novegradian - Grammar of Novegradian
http://www.veche.net/alashian - Grammar of Alashian
http://www.veche.net/novegradian - Grammar of Novegradian
http://www.veche.net/alashian - Grammar of Alashian
high vowels = close vowelsMaknas wrote:But what is "high"? I don't recognize the term...pharazon wrote:V[-high] means vowels that aren't high.Maknas wrote:What do you mean by 'high'?pharazon wrote:V[-high] > @ > ? / _V (except that a is kept before o)
i > _j / _# (unless preceded by a C(sonorant) cluster)Maknas wrote:Um, mind saying what it is here? I've been copying down all of the changes in this thread (in my own notation), and I don't want to parse through the Romanian changes to find the difference...pharazon wrote:Also, I forgot a fairly recent change in Romanian, which will be edited in.
That's what I get for not mentioning irregular changes. I have e > i / _[+nas] right next to that change, but e also went to 1 in that same context, without a pattern (e.g. pl?nus > plin, but f?nus > f?n).Maknas wrote:Also, question: You say that Romanian /1/ came from: a > 1 / _N (if N ≠ nn, mm). But, Metaforik says the Romanian orthography represents the etymology: < ? > came from Latin /a/, while < ? > came from Latin /i/. You give the rules for a > ?, but then where did the ? come from?
If anyone wants these in a listed / not roundabout form (some of those links reference the same change, which, at least, confused me at first), I've decided to list them out in regular notation:pharazon wrote:If it's English sound changes you want, there's some here.
C = consonants
N = nasals
E = front vowels
U = back vowels
P = labials
K = consonants (not /r/)
Old English > English
n > N / _[k, g]
? > ?@ / _[h, l, r, w]
?: > ?: @ / _[h, l, r, w]
e > e@ / _[h, l, r, w]
i > i@ > e@ / _[h, l, r, w]
i: > i:@ > e:@ / _[h, l, r, w]
[a, e, i, o, u] > [+lng] / _[mb, nd, ld, rd, rs, rT, rl] (if not followed by another C and if not stressed)
[a, e, o, u] > @ / _ (unstressed)
i > I / _ (unstressed)
e > i@ / [dZ, tS, S]_ (stressed)
e: > i:@ / [dZ, tS, S]_ (stressed)
? > ?@ / [dZ, tS, S]_ (stressed)
?: > ?:@ / [dZ, tS, S]_(stressed)
a > ? (in monosyllables, except when followed by N, [w, h, r]C, or when preceded by [g, k, S]
a > ? / _(C)$(C)[e, i] (if C = N)
a > o / _N
[a:, e:, i:, o:, u:] > [-lng] / ?_[mb, nd, ld, rd, rs, rT, rl],?_(C)$(C)V(C)$(C)V
[?, o] > e / ?_(C)$(C)(C)[i, j]
a > e / ?_N(C)$(C)[i, j]
a: > ?: / ?_(C)(C )$(C)(C)[i, j]
o: > e: / ?_(C)(C )$(C)(C)[i, j]
u > y / ?_(C)(C )$(C)(C)[i, j]
u: > y: / ?_(C)(C )$(C)(C)[i, j]
[?@, eo, io] > ie / ?_(C)(C)$(C)(C)[i, j]
[?:@, e:o, i:o] > i:e / ?_(C)(C)$(C)(C)[i, j]
ie > i / _
i:e > i: / _
E > a / _r
O > o / _l
? > b / m_[l, r, n] (irregular)
? > d / n_[l, r, n, #] (irregular)
? > t / s_[l, r, n, #] (irregular)
[a, E, O] > [+lng] / _$ (unless followed by (C)ig)
I > e / _$ (irregular) (unless followed by (C)ig)
U > o / _$ (irregular) (unless followed by (C)ig)
j > I / _
[tS, n] > ? / V_# (unstressed; irregular)
b > ? / m_#
h > ? / _[l, n, r, w]
w > ? / [s, t]_
h > ? (when unstressed)
@ > ? / _# (unstressed)
? > i / ?_h
g > j / E_
?@ > a / _
?:@ > E: / _
? > a / _
? > u / V_h (if V is not i)
j > ? / [i, y]_
g > u / B_
w > u / V_
V: > V / _ (when unstressed)
e@ > E / _
e:@ > e / _
x > w / [l, r]_
x > i / _
? > j / ?V(C)$(C)_
dj > dZ / _
sj > S / _
tj > tS / _
zj > Z / _
V > V: / _h[t, #]
h > ? / _t, _#
? > d / r_, _l
d > ? / ?V_Vr
Great Vowel Shift:
i: > @i / _K
u > @u / _K
e > i / _K
o > u / _K
a: > ?: / _K
E: > e / _K
O: > o / _K
?: > E: / _
e > i / _
E: > e / _
@i > aj / _
@u > aU / _
g > ? / _#
a > ? / _ (unless preceded by l)
a > O / l_
a > O / w_ (unless followed by a velar)
[g, k] > ? / #_n
l > ? / a_, _[P, k, g]
r > ? / _s, U_, _C (in non-rhotic dialects)
w > ? / _r
E > I / _N
u > U / _ (irregular)
?i > e / _
au > ?, a / _P
au > O / _
Eu > iu > ju / _
[Ui, Oi] > Oi / _
Ou > o / _
t > T (in loanwords with <th>)
? > h (in loanwords with <h>)
?U > ?@ / _ (unless P_ or _[l, S, tS]
http://www.veche.net/
http://www.veche.net/novegradian - Grammar of Novegradian
http://www.veche.net/alashian - Grammar of Alashian
http://www.veche.net/novegradian - Grammar of Novegradian
http://www.veche.net/alashian - Grammar of Alashian
Dang, I never realized that sound changes could be so complicated...
Now that I think about it, I should modify some of the sound changes that Kahrhtehlahrhno went through to evolve into its daughter languages. A whole lot of them are generic (e. g. a set of sounds changes in certain environments, like vowels shifting before liquids) instead of specific (e. g. a specific sound shifts in a precise environment, like a changing to e before r).
Does anyone know if there are any good resources (preferably on the internet) on the history of Japanese, Tagalog, Thai, and/or Hindi? I'm looking for lists of changes in particular, like the ones given here. I've tried looking for them on Google, but apparently I haven't found anything real useful.
~Tayanrai
Now that I think about it, I should modify some of the sound changes that Kahrhtehlahrhno went through to evolve into its daughter languages. A whole lot of them are generic (e. g. a set of sounds changes in certain environments, like vowels shifting before liquids) instead of specific (e. g. a specific sound shifts in a precise environment, like a changing to e before r).
Does anyone know if there are any good resources (preferably on the internet) on the history of Japanese, Tagalog, Thai, and/or Hindi? I'm looking for lists of changes in particular, like the ones given here. I've tried looking for them on Google, but apparently I haven't found anything real useful.
~Tayanrai
[url=http://www.spinnoff.com/zbb/viewtopic.php?p=127442#127442]If you want to know where I've been and what I've been doing lately, please click this.[/url]
I'm afraid you probably won't find much of that online; there's not even much in detail about major Western languages like Romance. A book is your best bet (not that I can recommend any).Tayanrai wrote:Dang, I never realized that sound changes could be so complicated...
Now that I think about it, I should modify some of the sound changes that Kahrhtehlahrhno went through to evolve into its daughter languages. A whole lot of them are generic (e. g. a set of sounds changes in certain environments, like vowels shifting before liquids) instead of specific (e. g. a specific sound shifts in a precise environment, like a changing to e before r).
Does anyone know if there are any good resources (preferably on the internet) on the history of Japanese, Tagalog, Thai, and/or Hindi? I'm looking for lists of changes in particular, like the ones given here. I've tried looking for them on Google, but apparently I haven't found anything real useful.
I know a book with some PIE > Sanskrit changes (id est, the book I've been using ). Next time I go to the library I'll photocopy those changes for you.
http://www.veche.net/
http://www.veche.net/novegradian - Grammar of Novegradian
http://www.veche.net/alashian - Grammar of Alashian
http://www.veche.net/novegradian - Grammar of Novegradian
http://www.veche.net/alashian - Grammar of Alashian
I have a book that has an article on PIE > Proto-Celtic (same one that Maknas got his PIE changes from), but typing up some changes for it isn't high on my priority list. Maybe you could figure out Proto-Celtic > Welsh from http://www.aber.ac.uk/~awcwww/s/p5_lexicon.html and get back to us.boomajoom wrote:Also, does anyone have PIE > Proto-Celtic and Proto-Celtic > Welsh?