The centum-word

From: Glen Gordon
Message: 6969
Date: 2001-04-04

Miguel:
>I thought you were opposed to **-nC > *-rC. In which case, you >really
>mean *wat:r/*wet:ar + -x ?

Oy veh. What a huge question... Yes, I do remain opposed to **-nC > *-rC
because I only support a *-n>*-r change, but you are missing something big
here that I must have discovered while you were sleeping.

My _current_ view is that final *-n simply became *-r (between 6000 and 5500
BCE) just before the loss of final vowels. This was probably a small
dialectal change starting first in the western edge of the Mid IE language
area gradually spreading eastward (but you can take that idea or leave it).
Another western change would be the uvularisation of velars, possibly caused
by Tyrrhenian areal influence (Piotr mentioned this earlier to explain the
satem dialects better and boy am I lovin' it!).

Here are examples of words and their hypothesized changes during this stage
of Mid IE where final *n became *r:

[*-n > *-r] [vwl loss]
-----------------------------------
*w�t:n "water" > *w�t:r > *w�t:r (*w�dr)
*kW�tWn "four" > *kW�tWr > *kW�tWr (*kWetw�res)
BUT *kew�ne "dog" > *kew�ne > *kew�n (*k^wo:n)

Medial *-n- simply never underwent this change to *r, unless corrupted by
analogical changes. This change occured in absolute _final_ position,
affecting the inanimate suffix which was *-n at one time (accented form:
*-�n-), in order to produce our familiar and productive *-r in slippery
words like "liver", "water", "blood", etc. Any inanimate stem with a
previous *-n such as above was still not affected in the
non-nominoaccusative cases, however. Cases like the genitive where *n was
never final due to the added endings, continued the archaic *n. Example:

*w�t:n "water" > *w�t:r
BUT *wet:�nse "of the water" > *wet:�nse (!!)

I hope yous can see the accents. At any rate, this best demonstrates the
origin of the irregularities one sees in this heteroclitic declension.

Now, in the case of collective suffixes or any other more secondary
suffixing outside of the declensional paradigm, the *n was NOT retained. For
the collective of *wat:n, we are simply adding an optional suffix *-xe
(later *-x or that devilish lengthening) to the bare nomino-accusative stem.
Due to the strict penultimate accent of that time prior to vowel loss, we
may only reconstruct *wet:�n-xe with accent on the _second_ syllable. (Also
note the *a/*e rule: When the accent is off of a stem with accented *-a-, it
becomes *-e- (schwa))

When *wat:n became *wat:r, the collective form followed suit as *wet:�r-xe.
The reason is that the collective suffix, while used occasionally, was never
a systematic and regular part of the declension. You see, the formation of
the collective form for *wat:r depended solely on the singular
nomino-accusative stem form and not via some remembered paradigm as with
*wet:�nse "of the water".

As for "dog", if you're still confused about the development of *k^wo:n,
here's the picture:

*kew�ne(-se) > *k^ew�n(-s) > *k^wan-s > *k^won-s > *k^wo:n

In this case, *n would never have become *r because it was in medial
position at the time of the *n>*r change, and not because of the animate
nominative *-s.

Oh and finally, before you get wise, I must explain the animate *-�r actor
suffix (or whatchamacallit) which I still claim to derive from *-�ne (and
not the expected **-�re). In Early IE, this suffix *-ene was associated with
the corresponding and phonetically similar inanimate suffix *-an (Mid IE
syllabic *-n). When the inanimate suffix *-n became *-r in Mid IE, the
animate suffix *-ene irregularly underwent the same change to *-er(e) to
continue the phonetic association between the two suffixes...

... but only in the western "heteroclitic" dialects of Mid IE.

I suspect now that in some more eastern dialects, these suffixes had
remained as *-�na for animate and the ol' *-n for inanimate for some time
yet. Later, western and eastern IE dialects converged and thus, both sets
from either dialects were used in the same language that we now know and
love as "Common IE".

Gee... and I still have to explain the many uses of early *e/*a ablaut! Any
more good questions, Miguel? :)

- gLeN

_________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com