On Wed, 28 Feb 2007 00:16:58 +0100 (CET), Mate Kapović
<
mkapovic@...> wrote:
>You didn't consider MAS theory about dominant (+) and recessive (-)
>morphemes in PIE?
To expand a bit on this...
Part of it is simply a question of terminology. Saying that
the PIE suffix *-ikós was dominant (+) or saying that it was
stressed amounts to the same thing. The question is: does
postulating (+) and (-) roots/morphemes in PIE offer any
advantages over the traditional approach? I would say that
it doesn't, at least not for root morphemes.
In the case of suffixes the MAS-approach is interesting,
because it offers 4 different possibilities:
1. + - => /-
2. + + => /-
3. - + => -/
4. - - => /- (?)
A dominant suffix added to a dominant root does not take the
stress. This appears to hold true for PIE, where "stressed"
suffixes are rather "predominantly stressed": sometimes the
stress is unexpectedly on the root. "Unstressed" suffixes
are unstressed under both theories. I see no evidence for
PIE "enclinomena" (a possible resolution of item 4.)
In vowel roots, the MAS approach also makes sense.
Balto-Slavic (+) nominals come from PIE barytones (stressed
on the root), while Balto-Slavic mobiles (-) come from PIE
oxytones (not stressed on the root). The PIE situation would
seem to be closer to a pure (+) or (-) distribution (which
is blurred by mobility in Balto-Slavic).
But when it comes to C-stems, having only two options (+ or
-) becomes severely limiting. PIE C-stems could be
acrostatic, proterodynamic, amphidynamic or hysterodynamic.
Should we assign (+) to the acrostatic words or to the
proterodynamics? Are the HD words (-)? Balto-Slavic is of
little assistance: PD, AD and HD are all mobile (-),
acrostatic nouns are barely attested (but should be (+)).
A more traditional approach makes much more sense here. The
PIE situation depends not only on factors such as stressed
root / unstressed suffix (the PD type: *h2ák^-mo:n) or
unstressed root / stressed suffix (the HD type: *poiH-mé:n),
which are equivalent to MAS "dominant" and "recessive", but
other factors also play a role. In my interpretation, these
are mainly phonological factors, such as the vocalism of the
root (a long **a: implies acrostatic) or the structure of
the root (if it's PD and ends in -CC, it becomes AD).
The MAS-approach also fails for verbal roots, as is clear in
Balto-Slavic itself: be^gnoNti is a.p. a, be^z^e^ti is a.p.
c. So what is the root *be^g- (*bhegW-): dominant or
recessive?
=======================
Miguel Carrasquer Vidal
miguelc@...