> Another thing about Bomhard: how come there are no context-sensitive
> palatalisations in his rules over several millenia, when Slavic alone
> had three or four?
In defense of Bomhard, I still think did a better job than
anything else I've come across from the older works of IS or
I don't expect him to find all the rules at once but I do
think that proper analysis of each individual language family
is lacking in Nostratic. It's a large task to get very
familiar and educated on more than one proto-language afterall.
There's just so much to read. I'm still learning about crazy
things in IE that throw me for a theoretical loop.
> BTW Holger Pedersen has the same idea of breaking up all PIE
> consonant clusters in a short remark about future trends (using
> <deus> and <divus>, supposedly from *däjäwä-) in his History of
> Linguistics in the 19th century. Cuny uses it too. (Written as
But not for IE where *deiwo- or *dyeu- is reconstructed. These
syllabics can only be reconstructed into a pre-IE stage.