01-09-03 21:59, etherman23 wrote:
>
>> How do you know that Hitt. dai comes from *deh3- in the first
>> place? It means 'take', not 'give'.
>
> Three reasons:
> 1) The semantics are very close.
Possibly, since giving and taking are different aspects of the same
situation, as in the case of 'borrow'/'lend' or 'teach'/'learn'.
However, this kind of closeness is tricky and not far from "lucus a non
lucendo" (is 'left' close to 'right' or 'wake up' to 'go to bed'?).
Where else in Indo-European do we find *doh3- with the meaning 'take'?
> 2) It's what we expect from sound changes laws.
You mean a root equation involving a root of the form "dV" in a language
where reflexes of initial *t, *d and *dH can't even be distinguished.
The likelihood of getting a spurious match is high in such a case. If
Hitt. da- showed any of the idiosyncrasies of *doh3- like the formation
of a characteristic reduplicated present (*di-doh3-mi), you could rest
your case. As it is, there's only a vague possibility that we might be
dealing with the same root. But there are other possibilities, e.g.
*dHeh1-, which is so naturally polysemic ('put, do, make, lay, place,
appoint') that it can also be considered "semantically close" to 'take'.
It has a non-reduplicated o-grade present in Germanic, which might be a
relic of something like Hitt. da-.
> 3) Other Anatolian languages indicate a meaning of give (Lycian B,
> Lydian, and Hittite itself in some of my sources).
Well, I haven't got any reliable Anatolian dictionaries to hand, so I
can't check all the proposed by-meanings, but 'take' is certainly the
usual gloss, and *pai- (Hitt. pai/pijanzi, Luw. pija-, Lyc. pije-) is
the Anatolian verb with the central meaning of 'give'.
Piotr