Thank you both for responding to me both here and in PIEML. I enjoy
this exchange of ideas. You have both raised many interesting
points, and I'll try to address them all here:
First and foremost, the transitive-intransitive distinction. There
is no doubt a strong connection between IE and Uralic languages. In
both languages, there is a pronominal root e- (one among many). The
Kartvelian languages, who are also connected to IE (and Uralic),
share this pronoun base; it is best attested (to my knowledge) in
Georgian e- 'this'. Many Uralic languages have verbal paradigms
which inflect for direct objects (e.g. Hungarian); although this is
also the case with most (if not all) Kartvelian languages, there is
nowhere near as strong a connection between Kartvelian and IE or
Uralic as there is between the latter two. In light of these facts,
I see no reason why it would be wrong to posit that PIE verbs could
also inflect for direct objects. And to be fair, Miguel, my own
ideas on (Pre-)PIE verbal morphology were inspired and influenced in
part by your essay on the subject.
The origin of the third person plural ending(s) is somewhat
ambiguous. I presented the hypothesis that it derived from the
active participle in -nt (or vice-versa) because of a similar
occurrence in the Uralic languages. For example, Finnish
sanovat 'they say' is clearly related to the active participle
sanova 'saying'; in fact, it can be analyzed as participle + plural
marker -t. It is possible that the 3rd plural ending of (late) PIE
developed in a similar way. Of course, there is also a good
possibility that I am wrong.
What I don't understand is why a plural form -en or -an would be used
in verbs but not in nouns. Unfortunately, I have no satisfactory
explanation for Greek 1pl -men, etc.
I would, however, like to point out something. It's possible that a
unified language that could be called PIE never in fact existed, and
that the irregularities we see from using the comparative method are
due to the "fact" that what we consider to be a unified proto-
language was actually a much looser conglomeration of languages.
That might be a stretch, but I think it's possible.
Glen, I agree with you that the initial enclitic pronouns are very
old: 10,000 BC sounds like a good starting point as any to me.
Finally, I would like to thank you both for pointing out that the
tense-markers were developed later than I had realized, especially
the past-tense augment e-. It would also seem redundant to have an
overtly marked past tense AND an overtly marked non-past tense. And
since many (if not most) languages derive non-past tense inflections
from past tense ones, it seems clear that the -i suffix was developed
first.