From: Arnaud Fournet
Message: 62708
Date: 2009-02-01
----- Original Message -----
From: "Brian M. Scott" <BMScott@...>
>
> Not quite: he gives the immediate PIE source of the theonyms
> as *deiwos (> PGmc. *ti:waz), but the underlying root is
> *dyeu-. OE <ti:r> and ON <tírr> look to me as if they ought
> to be from *dih2-ro- (> PGmc. *ti:-ra-), where *dih2- is the
> zero grade of *deyh2- 'to shine'. I believe that this root
> is generally thought to be related to *dyew-; you'll note
> that Watkins calls it a variant in part IV of his entry for
> *dyeu-.
>
==============
How do we know it is H2 in *diH2- ?
A.