Re[3]: [tied] PIE *a -- a preliminary checklist

At 12:05:37 PM on Thursday, February 14, 2008,
fournet.arnaud wrote:

>> There is a Finnish <ahku> that I've seen glossed as
>> 'Asche' and as 'ash ball' that looks like a borrowing of
>> PGmc. *asko:n, and an <ahjo> 'forge' from PGmc. *asjo:n
>> 'forge' (ON <esja> 'Schieferart, Seifenstein zum
>> Schmieden benutzt'), both from *h2as- 'to burn, to glow'.
>> If <kaski> belongs here at all, it seems to me that it
>> must be from a substantially older layer of borrowing.

> On account of Finnish regular -s- > h
> and irregular maintenance of -s- > -s-

> I would say that kaski must be
> a Finnish *recent* loanword.
> not an older layer.

I'm not convinced that it belongs here at all, but if it
does, it can't be recent: there's no source for the initial
/k/.

> It does not look like a pre-Saami word
> listed by Ante Aikio.

> Maybe *kaw "burn" could be a better source
> that *H2as-

I presume that this is Watkins' *ka:u-/*ka:w- < *kah2u-.

> *kaw-s-k- => kask-
> Cf. Greek kau-s-tos

> Is there Baltic reflexes for *kaw ?

Pokorny thought so: Lith. <ku:léti> 'brandig werden, vom
Getreide', <ku:le~> 'Getreidebrand'.

Brian