Re: Scythian tribal names: Paralatai

--- In cybalist@yahoogroups.com, "stlatos" <stlatos@...> wrote:

>
> --- In cybalist@yahoogroups.com, "david_russell_watson" <liberty@>
> wrote:
> >
> > The only connection possible is between the elements 'trae'
> > and 'tar' alone, if indeed 'tar' meant 'three'.
>
> I could go on for pages about my methods,

No need. I've been a member of cybalist for years now, and
have seen your method in operation many times.

> but instead just consider these few points:
>
> --- In cybalist@yahoogroups.com, george knysh <gknysh@> wrote:
> >
> > First of all let me say that I have no idea about the
> > etymology of this name. Perhaps someone can take a
> > crack at it. But here is why I think Abaev was wrong
> > in trying to understand it via Iranic . There are two
> > similar names found in, respectively, Pontic Indo-Aryan
> > (=TIRGATAO) and Mitannian Indo-Aryan (=TIRGUTAWYIA)
> > {for details see Cyril Babaev's article on Indo-Aryan}.
> > In In/Ar the names are female. So TARGITAUS would be
> > the metathesised and hellenized male version.
>
> There was no tar- in the etymon; it's just metathesis.

But 'tir' doesn't, to my mind, improve your own etymology
in anyway, with the 'g' being the bigger problem. I don't
accept your battery of sound changes that so conveniently
eliminates and creates sounds at will, remember.

> > > The changes involve n > n. > N after a velar which returns
> > > to n much later in most IE (but not if nasal dissimilation
> > > occurred first like Latin -go:(n-) or Iranian *xakYmnixYno-
> > > > *akYm.n.iyn.o- > *akYm.NayNo- > *asagaina- 'of stone').

I don't recognize *asagaina-, and it would have been nice,
and normal practice, to indicate its language, but, if it's
indeed derived from *h2ek^-, it's likely better attributed
to *h2ek^- with some suffix other than *-men-, saving us the
bother of such gyrations as you require to eliminate it in
the end.

Writing of normal practice, I must add that I don't myself
appreciate having to address the likes of "xakYmnixYno-",
"akYm.n.iyn.o-", and "akYm.NayNo-". These are to me no more
than excerpts from an interesting con-lang of your invention
which happens to somewhat resemble Proto-Indo-European.

> If there was no intermediate N stage, why *asagaina- instead of
> *asanín(a)-, etc?

Tell me your basis for *asagaina- and possibly I can answer,
though, as I say, I very much doubt that the suffix *-men-
is involved.

> Balto-Slavic *dwi:-. The difference between i: with level tone
> and ii with falling usually translates to i: in B-S but i in the
> rest if it retains tone. In compounds when the tone moves the
> original length is retained (Skt, G have pítus, pi:tuda:ru-;
> phthísis, phthi:símbrotos).

I don't understand the relevance of this. Do you attribute
the names to Balto-Slavic instead of Indo-Iranian?

> > 'Thrae' here is in fact merely the guna grade, with 'Thraetaona'
> > a patronymic derivative of the name of Thrita.
>
> What kind of ending do you think -auna- would be?

Besides the guna grade of -una- I have no idea.

> > These changes are purely ad hoc and would make hash applied to
> > most other words,
>
> The specific types of metathesis applies to this word only. Why
> would you think met. would apply to others or should?

There's more to your stack of sound changes than metathesis,
and moreover nobody but yourself believes that any of those
sound changes actually took place. Please keep in mind that
you've yet to win acceptance for your theories. I for one
don't lend any credibility to them.

> > and moreover you, as so many other dabblers
> > in comparative linguistics do, mistake the ability to draw a
> > straight line between two words for proof that the words are
> > actually cognate, but which it is not and never could be.
>
> This is ridiculous. I try to see what changes are needed to
> connect possibly related words,

Which is not the way to do it. Rather, see what changes most
simply explain the _total_ array of correspondences, which is
a bigger job than one person can do alone, and a job already
mostly finished by better linguists than you or me. That work
is, as I'm sure you know, at odds with your own reconstruction,
and I've yet to see myself a convincing argument from you to
prefer yours to the standard model.

> and then see if these are found in other words. That is, the
> n>N>g is found elsewhere, not as a random change to make this
> one thing work right.

Can you take ten standard P.I.E. reconstructions through your
battery of changes and end up with recognized words, or do
you have to work with your own idiosyncratic reconstructions
as starting points? You see, if I don't accept the validity
of your method, then I'm certainly not going to accept the
reconstructions based thereupon, and, if I don't accept those
reconstructions, then I'm obviously not going to accept them
as proper points of departure for demonstrating your method.

Do you know what you have to do then? It's not an easy job.
You have to reinvent the wheel, using some shape other than
the circle already in use, and the thing has to roll just as
smoothly as the wheel we already know and love so well.

Oh I forgot to mention that there is in the Vedas mentioned
a character 'Traitana' as well, who must also be taken into
consideration alongside Trita Aptya and Thraetaona Athwiya.
Can your system of sound changes produce his name out of the
same basis you posit for 'Thraetaona', or will you at long
last allow that not every similar word has to be forced into
the same bag?

David