--- On Tue, 1/27/09, Anders R. Joergensen <ollga_loudec@...> wrote:
"I hope I haven't misrepresented the views of the Leiden school, but
at least one member of it subscribes to this list and will be able to
correct me :-)"
I guess that was aimed at me :)..
Well, your description of the ideas of the "Leiden School" is mostly correct (although opinions vary from person to person, so don't regard us as all following one doctrine).
About the loss of laryngeal after lengthened grade in Balto-Slavic: this is not an ad-hoc-rule in order to only explain the circumflex in sa:`ls, but is also necessary to explain e.g. Lith. duõs 'he will give' < *de:h3-s-t (the original injunctive of the sigmatic aorist). Here you see as well that *e: is coloured by *h3, and that later on the laryngeal is lost after the long vowel, causing the circumflex intonation.
The only thing that is slightly ad hoc in the explanation of sa:`ls is the postulation of a long *e: in its preform *se:h2ls, which is from a PIE morphological point of view a bit awkward (cf. also Kortlandt, Baltistica 21: 119 (1985)).
Btw, Oliver & Arnaud: in my dictionary you can find a more explicit denial of Eichner's Law on page 98 fn. 218, where the example duõs would have been very appropriate as well. And Arnaud, thanks for the praise :).
Best, Alwin K