Re: [tied] Re: Pre-Germanic speculation

30-10-03 14:19, Marco Moretti wrote:

> Perhaps the word "parochial" isn't nice. Sorry, I won't cause offence
> to anyone. It wasn't direct to any member of the forum. These IEists
> lived in the last century, and reconstructed IE protoforms
> for "elephant", "monkey", "leopard", "lion". They derived
> *(s)teuros / *tauros from the IE root meaning "swell", they derived
> Greek (w)oinos and Latin vinum from another IE root (these items are
> Semitic loanwords!), etc...
> They are the old Pan-Sanskritists and others. I simply disagree with
> their theories because it is reductive to deny every contribute of
> pre-IE substratum.

You mean Gamkrelidze and Ivanov? They reconstruct PIE elephants, monkeys
and leopards because they need evidence for their favoured (but
unlikely) homeland in the Armenian Plateau. Other folks want to prove
that IE came from India, so they like the idea of IE elephants. I agree
that much of that is nonsense. But the 'wine' word may very well be PIE
(borrowed into Semitic) as Miguel has already explained, and there is a
slight chance that a PIE term for 'lion' is also reconstructible.

> I'm rather dubious. If cognates of *dHelbH- > 'delve, chisel' occur
> in Balto-Slavic, we can admit that Balto-Slavic was contiguous to
> Germanic. IN Balto-Salvic area there is the same word for "silver"
> that we found in Germanic. So the presence of a Balto-Slavic-Germanic
> isogloss is not decisive.

No, it isn't decisive, but the limited distribution is not sufficient
reason for proposing non-IE origin. Note that the root is an ablauting
one in both Germanic and Balto-Slavic, and that it produces a normal
range of derivatives. As opposed to 'silver', it doesn't _look_
borrowed. You can see the difference, can't you?

> Skt cha:ga- is isolated. We cannot even safely reconstruct its
> protoform. For example your reconstruction posit a labiovelar /-gW-/
> in order to fit better the Germanic protoform, that you think to be
> ralated. But /g/ of cha:ga can be from /g/. We aren't sure that proto-
> Germanic *skæ:pa- < *ske:gWo-.

No, but it's formally possible.

> Initial aspirated sound in Skt. is
> difficult to account for.

It isn't in the least. *sk > c(c)h before a front vowel is a regular
sound change in Sanskrit! I'd say that <cha:-> is almost certainly a
reflex of *sk(^)e:-.

> So your matchup (that would have pleased
> Dumézil) is inconsistent. In Vedic Skt. there is a lot of substratum
> of Dravidian, Munda or simply unknown origin. Nothing supports the
> projection of an highly hypothetic protoform of cha:ga- into common
> IE. Proves nothing at all, sorry.

It wasn't my intention to prove that <cha:ga-> and <sheep> _must_ be
related, but the point is that they _can_ be related, and if they are,
they can only be related via PIE. In what way is your assumption of an
otherwise unknown non-IE word in Germanic superior?

> "House" may be a derivative of *(s)keuh- "hide", but why should a
> similar form displace older IE words? It is suspect.

Lexical replacement can happen for so many reasons that one is not
obliged to reason why. Why should "meat", "dog", "they" and "uncle" have
replaced "flesh", "hund", "hy" and "eom" in English? So a Germanic
etymology is suspect but a non-IE one is fine and dandy?

> Germanic *maro:n- has cognates in Slavic, but the central meaning
> is "ghost, demon", Skt. item has a different etymon (cfr. Latin
> morior, mors, etc...).

Why can't the etymon be the same? Goblins can harm or kill you.

Piotr