Re: [tied] Bader's article on *-os(y)o

From: Rob
Message: 32733
Date: 2004-05-18

--- In cybalist@yahoogroups.com, enlil@... wrote:

> Rob, always perfectly analytical:

Speaking of which, could you explain again where the original
distinction between thematic and athematic nouns come from, in your
opinion?

It seems to me that athematic nouns are from words that did not end
in vowels (or perhaps only short vowels), while thematic nouns did
end in vowels (or perhaps long vowels). We know that PIE never had
an accent scheme that was fixed on the first syllable of a word, due
to the abundance of initial clusters in the language. Probably,
then, it had a regular accent scheme on the penultimate syllable of
every word, as you posit. Vowel syncope would then cause a seemingly
free accent system. However, assuming that thematic nouns derived
from roots/stems ending in vowels that were not lost to syncope, how
did this occur?

> Gee, Jens, that almost sounds like a genitive in *-ós that
> has been misanalysed as a thematic nominative, doesn't it.
> Amazing how you can deny what's right in front of your face.

I think it is likely that the genitive was suffix-accented
everywhere, at first.

Here is what I think is *at least* a possibility:

Root: k-w-n or kW-n 'dog'
(Here 'a' stands for a vowel of unknown quality)

Earliest absolutive: kawána / kWána > kWán@ > kWán(@)
Earliest genitive: kWán(@) sa > kWán(@)-sa > kWán-sa > kWáns@ > kWans

Genitive gains ergative (later nominative) force.
New ergative > nominative: kWans
New genitive: kWan ás@ (< *a-sa) > kWan-ás@ > kW@...@ > kW@...:s >
kuná:s

The new genitive was formed from a pronominal element plus the old
genitive marker -- 'the horse its rider' instead of 'the horse's
rider.'

As evidence I offer the compound *dems-potis 'house's lord / lord of
the house' > Gk. dêspotês.

/a:/ > /o/, /a/ > /o/ preceding a nasal.
kWans > kWons, kuná:s > kunós

Other cases:
1. Accusative: kWán(@) ma > kWan-ma > kWánm@ > kWánm. > kWónm.
2. Dative: kWán(@) áy@ > kWan-áy@ > kW@... > kunéi

It seems that a rule can come out of this: If the suffix is
monomoraic, no change in stress-accent occurs; otherwise, the accent
shifts one syllable to the right.

Thoughts?

- Rob