From: Mate Kapovic
Message: 29994
Date: 2004-01-25
----- Original Message -----
From: "Glen Gordon" <glengordon01@...>
To: <cybalist@yahoogroups.com>
Sent: Sunday, January 25, 2004 4:47 AM
Subject: The palatal sham :) (Re: [tied] Re: Albanian (1))
> Jens:
> >That's a statement of rare clarity.
>
> Then why didn't you understand it properly? Logic is about optimal
> probability. That's it. We don't reconstruct theories that fly against
> tendencies without damn good reason. Again, this is Occam's Razor.
>
> I'm not interested in debating the nature of logic. It's self-evident.
But languages are *not* always logic. Sometimes they are, but not always.
The laws of language does not always follow the laws of logic. It would be
more logical that *k' is not more frequent than *k but because of the
historical accident (more front vowels in the past) it is not logic.
Confusing logic with language is a very messy business. If you spoke of
typological plausibility, that would be reasonable. But speaking of logic in
language is not so selfunderstandable.
> In fact, if we can all accept that the traditionally written *swek^s
> is a Semitic loan, then why is there no palatal in the Semitic root?
I wouldn't say we all agree on this. The reconstructions look vaguely
similar but no so much. Besides, just to play guessing, I see no reason why
/s^/ for instance wouldn't become somekind of a /k'/ in a lg which doesn't
have /s^/s.
Mate