From: Abdullah Konushevci
Message: 42516
Date: 2005-12-15
> wrote:
I.
----------------
pn > m (gjumë < PIE *sup-no)
np > mp > mb (mbesë < PIE *nepo:tya:) (-> or a Latin loan?) (Rom np<-
>mp)
bn > m (lumë < PIE *lubh-no -> PIE *leubh-)
nb > mb (similar with np) (Rom np<->mp)
mn > m (dëm < Lat.damnum)
nm > m (emër < PAlb *enmen- < PIE *h1nh3men-)
II.
----------------
gn > nj (njoh < PAlb gna:-ska < PIE *g'neh1?-sko)
NOTE: the transformation was only gn > nj (and not also g'n>nj)
because the depalatisation of palatal velars in an {n,m,l,r} context
happened long before -> Timeframe: Dialectal IE.
ng -- ng
kn > nj (similar with gn)
NOTE: majë is more probable from mal-ja: (< mal 'mountain, shore').
nk > ng (nga 'out' < PAlb *en-ka)
III.
----------------
nc^ > s (pesë < PIE *penkWe -> in fact it was nc^>nc>(ns)>s )
c^n > n? (I suspect n -> based on c^>s & sn>n ) (Rom. c^n > s^n)
ng^ > z (I suspect similar with n3>nz>z see mëz)
g^n > n? (I suspect n -> similar with c^n) (Rom. g^n > z^n)
IV.
----------------
nc > s (I don't have example here but nc > s similar with ns>s)
cn > n? (any help is welcome)
n3 > z (mëz <-> Rom. mânz < PAlb *mandya)
3n > n? (any help is welcome)
sn > n (nus 'thread' < PAlb *snutja)
ns > s (maybe pishë < *pin-so?)
Note: Abdullah, ns, ms > s not /m/ -> PIE *mem-so > mish
What I have to note, when I explained more than one etmymolgy, tesftifying that cluster -ms- and -ns- yielded /sh/.
V.
----------------
-V'ntV(V) > V'nVV (man~mënd 'mulberry') (Dacian
manteia 'blackberry')
(V' = any vowell but not e => -entV(V) -- entV(V)
nt > nd (lend 'acorn' < PAlb *lenta)
tn > n (zotnë (Old acusative) > zonë 'lord, master', motnë >
monë 'weather, year')
nd -- nd
dn > n (ënj 'to swell, inflame' < PAlb *aidnja < PIE *Hoid-njo ->
Arm. aytnowm 'to swell')
Alb. <ânj/ënj> are derived from PIE *anH- 'to breath' and probably this is most funny etymology I ever seen.
NOTE: Abdullah, I prefer this example as a more sure one, due to
Armenian cognate, than lej or gjej.
VI.
----------------
ln > ll (we need to find a good example)
NOTE: sjell is not from PAlb c^el-na because the output would have
been *sjall not sjell...So Alb. sjell < PAlb *c^e-la (for the same
reason vjell < *we-la)
nl > ? (sorry for my previous post, I'm not sure here)
rn > rr (zorrë < PAlb *g^a:r-na: < PIE *g^wer-no (with unexplained
long vowel?))
nr > ? (sorry for my previous post, I'm not sure here)
VII Other Rules
----------------
Tosk V-n-V > r (gjër~gjanë 'soup' < PAlb *jausna:)
n > nj (in some contexts) => I suspect: if the previous syllable was
accented...(any feedback here?)
Next step is to can fix the right timeframes....and I will update the
Excel accordingly
Marius