Uralic and IE

[Glen Gordon]

Alright, let's see if we got it this time. You're saying:

  Pre-IE     IE      Anat      >        CS        Greek
   *-t       *-t     *-t                *-H1       -
   *-k       *-k    (*-h)               *-H2       -
   *-p       *-H3   (*-h)               *-H3       -

MIGUEL:
  Not quite.  I mentioned no -k in Greek, merely a possible
  alternation -H2 ~ -k-.

Whoops, of course. Slip on my part. I think I get it now <cross
fingers>

ME (GLEN) concerning **-t > *-H1:
  Besides the fact that this sound change itself lacks more than one
  or two examples, *H1 could really be any consonant or even a long
  vowel according to your "evidence".

MIGUEL:
  One good example is all it takes.

I'm afraid not. It's not a good example, it's not even an example. It
simply should be thrown out. Beekes may reconstruct a *-(e)H1 if he
wishes but anyone with any sense at all has to question the validity
of this. So what, there's a lengthened vowel. It still stands as I've
said above that there's no clear indication that the lengthening
derives from such a change. In fact, I don't even think mediofinal *H1
existed since that in itself is hard to justify.

MIGUEL:
  The instrumental (where it exists at all and isn't made with
  *bhi/*mi) shows a lengthened vowel in all roots (-a: for a:-stems,
  -i: for i-stems, -o: or -e: for o-stems), which can only mean -H1.

Which can mean either a lengthened vowel or *H1, really. Besides
probably Latin?, what evidence of this *-(e)H1 exists?

MIGUEL:
  Hittite has -it (< *-et).  If we postulate a development **-t >
  *-H1, the Hittite form can be connected to the others [as well as to
  other, extra-IE, instrumentals in -t, if we so wish.  I give you
  Georgian -it, Sumerian -ta].

Yes, that's right. I'm agreement with you but with vastly different
reasons. And don't forget Uralic *-ta :) which is most evidently more
relatable to the ablative *-ed based on those regular sound changes I
mentioned. In fact, my sound changes show more examples than your **-t
> *-H1 will ever be able to. If you want to talk external connections,
it makes no sense to compare *-eH1 of all things to foreign forms with
/t/ when the more comparable (and fully reconstructable, mind you)
ablative *-ed is awaiting our discovery. I've outlined how *-ed can be
related to Uralic *-ta due to syllable loss that has apparently
affected other endings as well. You have not.

What I'm saying, I think, is further validated by IE's quirky
accentuation that can be explained by the sound changes I've outlined.
It seems that Pre-IE **-VCV becomes *-V'C where the ending actually
becomes accented from the syllable loss. This explains the
accentuation of the 3rd person plural since **-ene > **-e'n and later
*-e'nt/*-e'r. Being that *-men and *-ten are patterned on **-e'n (note
that *-n does not become *-r in *-men as Miguel might want to think),
it should come to no one's surprise that the plural conjugation ended
up with accented suffixes. Nor should it be shocking that the genitive
*-es (**-ese) is also accented (cf. Etruscan -isa [genitive] and ?
Sumerian -se [dative]).

MIGUEL:
  The supposition is merely that if there are masc/fem. roots in -k
  or -t (nom. -ks, -ts), we might expect some neuters too, and
  there aren't any.  This may be due to an Auslautgesetz, as
  suggested by the few clues we have (ins. sg. -t ~ -H1, fem. -H2 ~
  -k-).

Supposition, yes. The fact is there's nothing you've found that shows
what you assert. Greek -k- could, in these cases be caused by a reflex
of *H2/*H3, in which case, there's enough doubt to call into question
the modest clues you give attention to as evidence.

ME (GLEN):
  That kind of logic in itself is deplorable and if Greek -k- does
  point to a laryngeal somehow we cannot, as in the first sound
  change, nail this down to anything more specific than this:

                  **-k ?> *-(H) (?)
MIGUEL:
  Surely -H2, if anything.

Judging by Greek we might have the following sketchy hypothesis:

                  *-H1-/-H2-/*-H3- >? Greek -k-

I of course severely doubt *-H1- as a possibility but lacking evidence
that supports Greek's peculiar whims as archaic, you are unable to
narrow these choices down.

MIGUEL:
  Symmetry is aesthetics, aesthetics is symmetry.

And sometimes, symmetry isn't science nor does it give us the simplest
and likeliest solution based on evidence at hand. Pi isn't very
symmetric but it's still pretty to some... 3.14159265358979323... :)

ME (GLEN):
  In summary, this is what your very uncertain idea amounts too:
                  **-t ?> IE *-(H) (?)
                  **-k ?> IE *-(H) (?)

MIGUEL:
  I agree the whole thing is uncertain, but one question mark
  suffices:

                  **-t > *-H1 (?)
                  **-k > *-H2 (?)

I'm afraid I have to keep both question marks. A correlation between
the *-t in Anatolian and forms without hasn't been properly
established with only one example to show! The second is to show the
uncertainty of the later posited form. You're lucky I didn't add a
third.

ME (GLEN):
  First, whether the heteroclitic stems end in *-t or *-d changes
  nothing since I've been saying that there was no pronunciation
  contrasts in IE between *-t and *-d (or *-dh).
MIGUEL:
  Which is obviously false.
ME (GLEN):
  Obviously how?
MIGUEL:
  Sanskrit, for instance, has -d for the ablative, -t for the
  3rd.p. sg.   Reason enough.

Right, because speakers of IE right down to Sanskrit would have been
aware of the relationship of the 3p secondary *[-t] (*-t ~ *-d) to
primary *[-t<h>i]. Thus, although *-t derives from **-to and should
have become *-d, it only devolved back into *-t due to its primary
counterpart. Aside from the ablative and neuter on the one hand and
the 3rd person on the other, there really is much to show for final
*t/*d contrasts and there's isn't much reason to assume that such
distinctions were actively maintained in PIE. Can you find a valid
example of IE *-t that doesn't involve the 3rd person? I rest my case.

--------------------------------------------
Glen Gordon
glengordon01 at hotmail.com