Piotr contemplates:
>*ek^wos (omitting the laryngeal) is almost generally regarded as
>related to *o:k^u- 'swift', though not without formal problems,
>usually dealt with in an ad-hoc manner. To be frank, I am not sure
>what to make of the long vocalism of *o:k^u-, and I'm not willing
>to multiply laryngeals ad absurdum (*h1oh3k^u- is hardly elegant,
>and the necessity of deriving *ek^wos via a monstrosity like *h1h3k^w-o-
>only adds to the embarrassment).
So true. Hmm, let's see what ol' gLeNny might drum up as an
explanation...
Okay, so we have this root *ek^wo- (I've taken out the initial
glottal stop *? since question marks are just so ugly). The
word is supposed to be related to *o:k^u-. We know that if
this is the case, the word has been thematized with an additional
*-o- (and this is completely normal for IE to do). Thus, we
should obtain **o:k^wo-... but then, this is not the expected
result, is it. So much for that idea. Maybe we need to try
something else.
Let's use penultimate accent theory, MidIE vocalism rules and
the uvular interpretation of *k^/*k to our advantage. A root
*o:k^u- is immediately problematic because of the fact that *k^
(reinterpreted as a non-uvular unmarked *[k]) should only result
next to Mid IE *� (schwa) which later becomes Late IE *e. Instead,
here we find *k^ next to a long *o: and yet, since *o derives
from Mid IE *a, we should have expected *k (a uvular *[q]) due
to this vocalism-dependant allophony of velars. BTW, for the
new people:
The Mid IE velar allophony rule:
*k^ (plain) next to *� and *k (uvular) next to *a
So clearly, something is wrong with *o:k^u-, and *ek^wo- appears
to reflect the original vocalism. Is Piotr correct in suspecting
*ek^wo- as the more original stem?
Hmmm, here's a tantalizing attempt at solving our horsey problem.
First, the evolution of the "horse" and "herd" words:
Mid IE Late IE
*�kw� "horse" > *?�k^w� > *?ek^u > *?ek^w-o-s
*p�k�u "herd" > *p�k^�u > *pek^u
Here we see the typical loss of final vowel by Late IE,
producing Piotr's inanimate *?ek^u, in line with *pek^u which
is of a slightly different original form. The "horse" stem is
then made animate through the very productive thematization of
the stem.
Now observe a now even longer list of juicy examples of a
prothetic Late MidIE *a- which appears to have arisen all
in order to avoid the extreme consonant clustering pattern
of the form *CCC-:
Mid IE Late IE
*�kw�-w� "fast" > *(a-)?k^w�u > *ohk^u- (*o:k^u-)
*kW�tW�x� "8" > *(a-)k^twa: > *ok^to:u
*x�w�i�n� "egg" > *(a-)xwy�n > *oxwy�m (*o:wy�m)
I hope yo'll like that explanation. I thought really hard on
that and now my brain hurts.
- love gLeN
_________________________________________________________________
Get your FREE download of MSN Explorer at
http://explorer.msn.com/intl.asp