Pavel Adámek wrote:
> There is also an article
> INITIAL LARYNGEALS IN ANATOLIAN
> by FREDERIK KORTLANDT
> www.kortlandt.nl\publications\art202e.pdf
>
> "...initial *H2- and *H3- yielded h- before *-e- and zero before *-o- in
> Armenian and Albanian and ... the same development may be established for
> Hittite, e.g. harp- 'separate' < *H3erbh- versus ark- 'mount' <
> *H3org^h-ey-, Gr. orphanós, órkhis ..."
I thoroughly disagree with Kortlandt's rules for *h2/3 and with many of
his index identifications. To assume a contrast between reflexes of
*h3e- and *h3o- in Proto-Anatolian is, by the way, a serious theoretical
commitment that should be justified using independent evidence. Sure
examples of *h3 are relatively few but give the impression that *h3 was
generally preserved in Hittite, cf. *h3óro:n > haran- 'eagle', where the
expected e-grade of the first syllable rules out either of the other two
laryngeals. I don't exclude the possibility that *h3 was lost
sporadically or in some special environmemnt, but the evidence either
way is scanty.
By the way, Kortlandt claims (or rather makes the prediction) that all
instances of initial <a-> in IE etymologies must go back to *h2e-. Since
I see no reason to question the existence of unconditioned *a in PIE
(marginal as it was), I feel free to derive <alpa:> 'cloud' from
*h1albHo-, for example.
Piotr