More "ablaut".

[rankin]

Rory writes:
> > One possibility that occurs to
> > me for that -a- is that it might signal something
like active
> > responsibility or intention on the part of the
actor.
John writes:
> In fact, positional-initial-a and a-ablaut are
synchronically different
> things,

Diachronically too, I think.  When I said that the
so-called "ablauting" vowel could be a morpheme, I
didn't mean "now":  I meant something like a couple of
thousand years ago.  Now, the "ablauting" /-a/ is
pretty clearly associated with just 3 or 4
suffixes/enclitics including 'imperative', 'plural' and
'negative' (and may always have been).  I didn't mean
to open the door to bringing back discussions of the
"meaning" of ablaut and segmenting *-api as just -pi.
In Mississippi Valley Siouan I regard such discussions
as "Dakota-centric" and essentially behind us unless
data of the following sort can be clearly isolated
(preferably in volunteered, not elicited, speech).

Instances in which the *same verb* with the *same
plural, imperative or negative enclitic/suffix* can be
seen to have different readings with -e than with -a
between the root and the clitic/suffix.  I think
keeping all other variables constant is essential to
proof because other verbs offer different phonological
environments and, just as important, different
opportunities for analogical restructuring.

Returning to John's discussion of the -a that is used
with positionals, this may be a relatively simple
analogical development using 3rd sg. -api/-ape as a
model.  Or it could also be related to the so-called
'collective' prefix a- used with verbs of motion.
Either source (or both of them together) provide(s) a
good model for using an analogous a- with positional
auxiliaries, although with positionals there may be
other possible sources too.

All the discussions are useful though.  There is so
much we don't know about these languages.  Rory's
examples with the horse eating the food with various
presuppositions were fascinating.

Bob