Re: [tied] IE *-su and the Nostratic "equational" marker *-n :)

From: Glen Gordon
Message: 4767
Date: 2000-11-17

Marc V comments on my take on laryngeals:
>> H1 = /?/
>> H2 = /x/ or /h./
>> H3 = /xW/ or /h.W/
>That's what Beekes said, isn't it?

I'm not sure who said it first. I like it because it's the simplest solution
available... well, aside from the alien conspiracy theory :)

>Of course I'm aware of [Burrow's solution for *-su]. I just don't >agree.
>The loc. sg. has *-i, so the should also have *-i. >There's
>nothing "plural" about the deictic *-u as opposed to *-i.

One must be aware that analogy and native perception of one's own grammar
does play a strong part in linguistic change. From time to time it shakes
things up a bit. Note for instance in English where we have "worse" versus
"worser" or "brid" becoming "bird" or "a nadder" making an unexpected change
to "an adder" by misinterpretation. The primary meaning of *-u here is
locative certainly, but its specific use in the plural may be explained best
by association with collective nouns in *-u which do convey a sense of
plurality afterall. Your pleading for Greek /-si/ out of **-sw-i as opposed
to a simpler and more Occam-compatible solution like *-s-[i/u], such as we
even find in Sanskrit /s'va-su/, is emotional and non-reality-based. You
fail to justify the existence of *w (or is it *W?) in your reconstruction.
Why exactly must Armenian /k`/ derive from *sw per se?

Next, can we honestly say that this locative *-su is opposed by *-u in the
singular? Or is it opposed by *-i? What does the declension of Sanskrit
/s'van-/ "dog" tell us? How should we interpret /s'un-i/ [loc sg] versus
/s'va-su/ [loc pl] where *-i in the singular clearly opposes *-u in the
plural? The locative *-i here is not in both the singular and plural as you
would like to fantasize. The ending *-u appears to even show favouritism for
the plural as though underlyingly it is conveying some sense of plurality or
collectiveness by the analogy I bring up. Thoughts?

The fact that both Greek and Sanskrit share these locative endings *-i and
*-u shows that the suffixes were at least established around the time of the
postAnatolian stage of IE. This could not have been so easily acheived
unless the locative was first null and followed by a stage where both *-i
AND *-u together were suddenly used as locatives by grammatical analogies.
We don't find **-i-su, do we? Instead we find *-(NULL-)su. I can't see how
we might feasibly reconstruct a stage of common IE where only *-i is found
in the singular and plural throughout all types of declensions. There is no
*-i without *-u. They were adopted as locatives at the same time. I think
that you're simplifying where simplification is undue and failing to
simplify when there is no logical need for added complexity.

But I could be wrong... You say, "The loc. sg. has *-i, so the
should also have *-i." Please elaborate and justify somehow with examples
within the IE family (and restrict yourself to ACTUAL IE reconstructions

- gLeN
I sleep now (only had 4 hours last night)...
feelin' drowzy... goddag och adjo".

Get Your Private, Free E-mail from MSN Hotmail at

Share information about yourself, create your own public profile at