Re: [tied] Why is PIE more centum than satem?

----- Original Message -----

From: george knysh

To: cybalist@yahoogroups.com

Sent: Monday, February 04, 2002 4:56 PM

Subject: Re: [tied] Why is PIE more centum than satem?

*****GK: Here also one should look at the whole
picture, and not be swayed by the value judgement
contrasting the "fundamental" and "significant" to the
"deceptive" and "superficial". Any biologist who
ignores the fundamental fact that dolphins are marine
creatures would be something of an oddball.*****

 

He would be. But note how long it took biologists to understand and appreciate the difference between homology and analogy.

 

*****GK: There is an important fallacy here, since
time is clearly not the only and perhaps not the major
factor determining difference. On this analogy,
currently existing snakes, humans, bees, are equally
close relatives as the "lineal descendants" of some
primeval slime microbe...******

 

In a way, they are equally close (or equally distant) relatives of the microbe in question (ignoring one fact that should not be ignored -- the average generation lifetimes are different for different lineages, which is less of a problem in linguistics). They are not equally close relatives of each other, since snakes and humans had a common ancestor as recently as some 300 million years ago. There is no real fallacy here, just a preference for "phylogenetic distance" over "difference", whatever the latter means (I don't doubt that snakes and bees belong together as "stingers" in some popular classifications).

> You may have heard of
> lexicostatistics, which uses the statistical study
> of vocabulary to "measure" linguistic proximity.
> Unfortunately, the "hard figures" it produces are of
> no real use in reconstructing linguistic history:
> vocabulary diffuses easily and at unpredictable
> rates, introducing errors that cannot be controlled.

*****GK: I think this is a very promising area
actually, since it deals to some extent with real
differences in real languages. The discovery (or claim
not yet verified to everyone's satisfaction) that some
60% of Ancient Greek's lexicon could be non-IE (with
similarly large percentages for Hittite, and somewhat
less for Germanic)I find extremely significant as
potenetially relevant historical data. This sort of
thing should be done for every IE language.*******

I hope you don't mean that only things that can be measured in terms of scalar values are "real". That would be a reductionist fallacy. I agree that lexicostatistics has its legitimnate uses, provided that the analyst remembers that word statistics don't work as a measure of "genetic distance" (still, a number of people have tried to substitute such a trick for painstaking comparative analysis). The percentage of shared vocabulary can suggest a lot about past language contacts, for example. No problem with that.

 

Piotr