--- In
cybalist@yahoogroups.com, "david_russell_watson" <liberty@...>
wrote:
>
> --- In cybalist@yahoogroups.com, "stlatos" <stlatos@> wrote:
> >
> > --- In cybalist@yahoogroups.com, "david_russell_watson" <liberty@>
> > wrote:
> >
> > Since A:ptyá- and A:thwiya- are obviously related, how can you
> > use your preferred method to relate them? There's no other case
> > of p becoming w and moving so any rule would apply to this word
> > only, just like with Targitaos.
>
> 'Aptya'/'Athwiya' is indeed an irregular correspondence, but
> one on obviously much more solid ground than 'Thraetaona'/
> 'Targitaus'.
Yes, of course, but the principle that changes might occur in only
one word in a language and still be valid and identifiable must be
used for both. You can object to the particulars of my
reconstruction, but not the theory behind it.
Since you think the two examples are on the ends of the difficulty
spectrum, what about, say:
G hálme:, Lith só:ly:mas 'brine' ?
The fem. vs m. endings aren't important, but what about 0 vs i: and
a vs a:? Are these close enough to be from the same etymon with
unique changes or so dif. they must be from dif. forms (and, if so,
where does a>a: fit in?)?
I'd say that the main difference comes from confusion with
*sxaLixYn.os (L sali:nus) or similar words in -i:nos / -i:mos in
Lith., then either more conf. with the nominative of the noun *sxa:L
or metathesis of xa > ax > a:. Whatever the case, it's not clear
enough for one or the other to be rejected out of hand.
What of *sxaLixYn.os itself? For:
? > *sali:no- > L sali:nus, etc.
? > *sala()no- > OIr salann 'salt'
? > *saleyno- > W halen, C haloin
? > *saliyen- > TB sa:lyiye
Are these all from slightly different words? Did PIE have a
plethora of endings with -n- that all happen to only be found in forms
of 'salty' for dif. languages? I'd say they're all from the same source.
Just like the correspondence -i:nus vs -in(a)-, Indic first turned
ixY to iy before n, then iy > i (like ii > i). In Celtic, only when i
was turned to a in a specific environment would xY > y. In this case
it's the same as in *asagaina-, a velar became KY shifting ixY > axY > ay:
*sxaLixYn.os ... *xakYm.n.ixYn.os
*sxaLixYn.os ... *xakYm.NixYn.os
*sxaLYaxYn.os ... *xakYm.NYaxYn.os
*sxaLYayn.os ... *xakYm.NYayn.os
In Irish the y just disappeared (that is, ay was dif. than ai), but
P-Celtic changed a>e between two Y (LYay > LYey), sim. to a-i > e-i.
Toch. was sim. to Indic, but instead of iy > i a vowel was inserted
(or there was met. after o>e) to allow y to remain.
And concerning variant forms, what about Sanskrit words for types of
beans:
mapas.t.a(ka)-
mapus.t.(h)aka-
makus.t.(h)a(ka)-
mukus.t.a-
mayas.t.a(ka)-
mayus.t.aka-
maris.t.aka- ?
Is there any explanation that involves only regular changes that
affect all words? I'd say that they're from ma:s.apis.t.a- 'ground
beans' with *mas.pis.t.a(ka)- often having s.-s. dissim., vowel
assimilation, etc.
> I cited it to demonstrate how closely related
> Vedic Trita and Avestan Thrita and Thraetaona are, to make
> a case for 'Thraetaona' being based on 'thrita-' "third", not
> as an example of perfect regularity.
But Thrita and Thraetaona aren't the same person. Even if they were
once father and son, that doesn't mean one name came from another.