IE "Urheimat" and evidence from Uralic linguistics

[Stefan Georg]

>>Georg at home.ivm.de writes:

><< Everything can be borrowed, and there are examples for everything actually
>>having been borrowed at some point in space and time. >>

>-- true, although of course some things are more _likely_ to be borrowed than
>others. Numerals, body parts, family relationships, and so forth, are less
>likely to be loan-words.

But in view that they are subject to borrowing after all, this
"unlikeliness" cannot be used as a heuristic for demonstrating relatedness.
Likely or un-, a case has to be made for every single instance of shared
(or merely similar) vocabulary being not due to borrowing, before a claim
of relatedness can expect the attention of the more skeptical parts of the
audience. The "unliklely"-approach is a cul-de-sac. Given a sufficient
degree of language-contact, the borrowing of every lexical item with every
conceivable meaning is equally likely. If it weren't so, we had to watch
out for a theoretical justification of the alleged resistence of some words
to borrowing. What exactly makes them so "hard-wired" into the brains of
language-users that they'd hardly give thought to using a different term
for, say, "water", than all their ancestors did. What, for that matter,
makes them "know" in the first place that, speaking of fully bilingual
individuals, this term is "ours" and that one is "theirs". Lest this sounds
naive, I'll add that frequency of use may be one firewall against borrowing
(though not, as we see from empirical observation, an impermeable one).
Structural differences, I have in mind drastic differences of phonological
inventories, between languages in contact may also counteract large-scale
borrowing, but, then, these tend to dwindle under a lonmg-standing areal
pressure as well.

Of course, I don't deny a general difference of borrowability between names
for cultural items and every-day expressions. New technology changes hands
together with terminology, more often than not. But "hand", "eye", "I",
"water", "brother" aso. are not safe. They simply don't bear a label
"Attention ! Native word! Don't replace by foreign gobbledeegook !" on
them. Since it happened at some time, somewhere, it can happen anywhere.
"Basicness" of vocabulary may be one factor slowing down large-scale
borrowing processes. A different  factor, one which may speed up the
process, is intimateness and longevity of contact (and there are different
kinds of language contact, which equally have to be taken into account).
And the latter may overrule the former.

The reason why I'm polemicising so determinedly against the "unlikely to be
borrowed" mantra is that I see here the danger of a shortcut to the
detection of genetic relationships being advocated. I don't maintain that
anybody on this list actually thinks that, but I know people who do, hence
my zeal.

I think this started with the question whether Uralic *wete is borrowing
from IE or common Indo-Uralic inheritance. I hope I'll be forgiven for
being imprecise, but as far as I know the main argument for the borrowing
scenario (general skepticism against Indo-Uralic can of course not play the
role of a major argument here) builds on the fact that there seems to be
another "water"-term in Uralic, shared only (??) by Saami and one (or both
??) Ob'-Ugric language, which gives the impression that *wete was a
secondary intruder from outside, gradually replacing this original term in
most of Uralic, but not reaching its extreme fringes. This is not "proof"
of borrowing, but it is a state-of-affairs which squares neatly with such a
scenario, so it shouldn't be brushed away, certainly not by saying
"unlikely and that's that". I forgot the term itself, shoot, but I read
about it a few days ago in a source I cannot pin down at the moment. If
this source will  turn out to be one of the postings on this list during
the last week or so, you'll have evidence of my slightly deranged state of
mind these days (and shoot again) ...

St.G.

Dr. Stefan Georg
Heerstraße 7
D-53111 Bonn
FRG
Tel./Fax +49-228-691332