Re: [tied] PIE laryngeals?

27-08-03 22:56, etherman23 wrote:

> Unfortunately the Anatolian h(h) doesn't always appear where
> predicted. So that's not really spectacular. Kurlyowitz (sp) felt
> that there would have to be a fourth laryngeal to account for
> Anatolian words where H2 didn't appear. This H4 would have the same
> vowel coloring properties as H2 but wasn't preserved in Anatolian.

And Eric Hamp used Albanian data to distinguish *h2 from *h4. I don't
accept any of that. There's no need to reconstruct *h4 nowadays; there
has been some progress since Kurylowicz [sp: sic!]. In those cases where
*h2 does not seem to be reflected as <h> there are reasons for that
(e.g. predictable loss in certain positions, or the vocalisation of
syllabic *h2 [a.k.a. *&2]).

> Likewise it seems to me that you'd have to postulate an H5 to account
> for those H3's that don't appear in Anatolian (e.g. PIE *deH3 appears
> as Anatolian da). So I can't say I'm impressed by the Anatolian
> evidence. Even Lehmann said the Hittite evidence wasn't very strong.

How do you know that Hitt. dai comes from *deh3- in the first place? It
means 'take', not 'give'.

Piotr