Res: Res: Res: [tied] Re: 'dyeus'

--- In cybalist@yahoogroups.com, Piotr Gasiorowski <gpiotr@...> wrote:

>
> W dniu 2010-07-12 06:35, stlatos pisze:
>
> > Is this other 'yesterday'?
>
> Yes.
>
> > What about the odd cognates in other
> > languages that sometimes have -t- even when the word had no -sr- and
> > they don't insert -t- anyway? It looks to me like no more than, say, *
> > gYHd.es i '(at) yesterday', * gYHd.es ti '(to) yesterday', with adv. >>
> > adj. * gYHd.esin.os, * gYHd.estin.os (with, as always, opt. n. > r. and
> > i > ï > ë > 0). The dif. in meaning is difficult to render in idiomatic
> > English, perhaps something like: "I was hit yesterday" and "I've been
> > hurt since yesterday".
>
> I'm not sure which particular cognates you have in mind.

L hesternus, S hyastana-, possibly G khthizós if < * khthizdós < * khthistRós < * khthestir.ós / * khthesin.ós since khthesinós and serós (Elean; < * khthestRós ?) exist.

> My solution
> accounts for some otherwise problematic aspects of the Germanic term,
> including the long vowel in the alleged root locative.

I have no problem with loss of -s- causing lengthening; it's known to happen all the time (-Vzd-), and I made the same rec. years ago, but Vzd > VVd is opt. and not based on whether accent caused s>z or not (sim. to Goth. mimz, mammo:). That doesn't prove your solution accounts for _r / str alt. in these cases.

My criticisms are more specific, some specific to these words, and have to do with you having no good ev. about what the original form was, whether s>z should occur in one but not the other, whether there was opt. -t- in the proto-form anyway (for various reasons), or whether the languages you're looking at had opt. -sr>str- (perhaps Baltic) before any s>z in Gmc, etc.