>From: "Piotr Gasiorowski" <gpiotr@...>
> This is precisely what Greenberg does:
[snip]
> If the first pair to strike him as possible cognates were e.g. Etruscan
>apa : Hungarian apa, he'd pounce on it and start looking for more
>lookalikes in the two languages. The end of that would very likely be the
>grouping of Etruscan and Hungarian together and leaving Finnish aside, as
>GENUINE Finno-Ugric cognates are of course more difficult to spot and could
>easily escape a mass comparativist.
I should've mentioned Greenberg. There is that conincidence that the
Dyirbal (an Australian language) word for "dog" is indeed _dog_! But the
Proto-Pama-Nyungan word is _gudaga_, which is not quite the same as our word
for the mutt. Greenberg obviously ignores history (and the tried-and-true
concepts of sound correspondence in historical or comparative linguistics)
and grasps at straws.
You may have also seen how a common word _maliq'a_ was proposed for "milk".
It's on a webpage run by Mark Rosenfeld (I think). Someone also posted this
list on conlang, which was borrowed from some article:
AJA -- mother, older female relative
BUNKA -- knee, to bend
BUR -- ashes, dust
CHUNGA -- nose, to smell
KAMA -- hold (in the hand)
KANO -- arm
KATI -- bone
K'OLO -- hole
KUAN -- dog
KUN -- who?
KUNA -- woman
MAKO -- child
MALIQ'A -- to suckle, breast
MANA -- to stay in a place
MANO -- man
MENA -- to think
MIN -- what?
PAL -- the number 2
PAR -- to fly
POKO -- arm
PUTI -- vulva
TEKU -- leg, foot
TIKA -- earth
TSAKU -- leg, foot
TSUMA -- hair
?AQ'WA -- water
Don't laugh too hard. Remember that this is supposed to be from the
language our caveman ancestors spoke 100k years ago. If this was true, then
Lojban is derived from Proto-World, since it was designed by common words in
the top six languages of the world, including "portmanteau" reconstructions.
(The latter case is where you concatenate two roots together -- for
example, Russian _bog_, English "god" and Latin _deus_ should form a common
Indo-European word for "god/deity": *bogodeus.)
But this isn't quite as bad as the fact that Greenberg rejected the
old-school theories of comparative linguistics and sound correspondences in
the oldest, not the most modern, languages or language roots. (Well, you
can with Lithuanian and Icelandic, to a point.) That's like throwing the
baby out with the bathwater and replacing the baby with a rattlesnake.
Today I e-mailed Serget Starostin, who has more than a few language families
compared and reconstructed (though I still can't accept Japanese being an
Altaic language, Korean maybe but with considerable doubt), and asked him
his thoughts on Proto-Uralic (which indeed would be very difficult to
accurately reconstruct) and Proto-Afro-Asiatic (which would be even more
challenging), and ultimately Nostratic and the spurious classifications of
Greenberg (Dene-Caucasian and Amerind being the most infamous). I also
don't accept Etruscan as a likely Nostratic offshoot. Mainly becuase the
corpus of recovered Etruscan lexical items is way too small, and you also
have a possible (though again not very possible) Basque-Aquitanian
affiliation. Same case with Sumerian, which should just be considered an
isolate until we really find out differently.
Daniel A. Wier ����
Lufkin, Texas USA
________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at
http://www.hotmail.com