--- Piotr Gasiorowski <
gpiotr@...> wrote:
> On 2007-03-17 16:44, stlatos wrote:
> > --- In cybalist@yahoogroups.com, Sean Whalen
> <stlatos@...> wrote:
> >
> >> *sidhus sidhw+ > Goth sidus 'custom', L *+a:lis
> >> 'customary' >> *swedwa:lis > soda:les
> 'companions'
> >
> > That should be *swedhxYus > *swidhus swidhw+ >
> Goth sidus
>
> Cf. Ved. svadHá:- (no trace of a *w). I'd analyse
> these forms as
> independent derivatives of the same base:
> *swe-dHh1-ah2 and *swe-dHh1-u-.
I'd say:
*swedhexY+ 'custom / choice / one's own' noun
*swedhxYu+ 'customary / chosen / one's own' adj.
Since Latin can have derivatives in -a:lis
regardless of the stem there wasn't necessarily an *a:
in the noun.
I would never base my idea solely on this word,
especially considering the possibility of analogy with
forms without *swe+, mixing with forms of a word
followed by u or w, or mixing between these words
derived from the same stem (depending on which words
Latin retained at the time of these changes and the
exact order of the changes). It's only because all
the others show evidence of a following glide that I
formed this theory and so derived it from *swedhxYu +
suffix.
That is, "all the others show evidence of a
following glide" unless you don't believe final *-ya:x
> -e: (if so I can try to provide more evidence).
____________________________________________________________________________________
8:00? 8:25? 8:40? Find a flick in no time
with the Yahoo! Search movie showtime shortcut.
http://tools.search.yahoo.com/shortcuts/#news