Hi folks,
I'm back. Yet again, Hotmail had managed to spam me over quota, bouncing
further emails back to the sender. That made YahooGroups think that I'm
dead. As a result I had to re-initiate my connection to the List. VERY
VERY annoying.
Anyways, the pause gave me some time to think about Miguel's idea on the
origin of some *a's. I still don't agree with what he has to say but I've
modified the concept a bit. So here it goes.
Let's assume that there are other causes of *a in IE and let's accept
that *m is somehow the cause of an apparent resistance of *a > *o in
some pre-IE stage. Nasalization cannot be the cause for the simple
reason that we don't see *na- nearly as much as we see *ma-.
On the other hand, it seems to me that there is an unusual rarity of
word initial *bHo- or *bo- along with the reasonably common enough
occurences of word initial *ma-. So here's a cool theory I drummed up...
I'm tempted to say that the trigger truly causing *a in these occurences
are VOICED BILABIAL PHONEMES. These phonemes would be *m, *bH and *b
where total closure of the mouth might be the key to all of this. In early
Late IE, I figure the vowel system looked like this phonetically:
*i *u
*E
*A *&
*a
Now *[a] would eventually be backed and raised to *o, while the always
unstressed *[&] would become *e/*o (depending on voiced segments coming
after that vowel). The vowel *[A] was a newly lowered version of *[E] next
to uvulars that would eventually become IE *a. Perhaps then, initial
voiced bilabials helped to cause raising of a following *[a] to *[A] at
this point in time. On the surface, we would have *e/*a ablaut alternation
(becoming *e/*o) but following voiced bilabials the precise phonetics of
the ablaut might have been *[E]/*[A] rather than the usual *[E]/*[a].
By the time of IE proper, verbal paradigmatic alternations would have
reestablished the normal *e/*o ablaut after *m, *bH and *b but *a would
be preserved in stems surviving from early Late IE that were now
disassociated from their origins.
The process causing the resistance of *a>*o then is not nasalization at
all. It is the closure of the mouth which caused raising of the eLIE
vowel *a to *A, followed by a vowel shift of *A > *a. The following vowel
would be raised (or rather "closed") because of the mechanics of
vocalizing a vowel immediately after a complete bilabial closure. The
change would only affect voiced phonemes because voiceless aspiration
would only accompany *p, buffering and thereby protecting the following
vowel from raising. The buffering would allow more time for the mouth to
adjust to the required vowel position after the bilabial closure. This is
not so for sequences involving a voiced bilabial phoneme where voicing
is uninterrupted up to the vowel.
Interestingly we don't see *pa very often either, so this idea of
bilabiality, voicing and aspiration as the interdependent causes of *a
in words such as *markos might just work afterall.
PS: I'm starting to question if *b actually did exist as a very rare phoneme
in IE afterall, especially if the above is correct. This theory would
imply that instances of *b date to at least the early Late IE period!
- gLeN
_________________________________________________________________
Protect your PC - get McAfee.com VirusScan Online
http://clinic.mcafee.com/clinic/ibuy/campaign.asp?cid=3963