>
> As I theorised before:
> the verb stem is of type gerund, thus <NP>
> the suffixes -m, -t, -s are personal suffixes, of type <NP -> NP>
> therefore
> the verb stem exended with -m, -t, -s is of type gerund, thus <NP>
> those extensions, extended with locative -i, are gerunds in the
> locative, thus of type <adv>.
> And
> since the non-indicative forms occur mainly in subordinate
clauses,
> they should be understood as originally constructions with gerunds.
>
And that would explain the odd construction with double reference to
the subject "I (am) in-my-V-ing": First the person-inflected
gerund/participle was used in subordinate constructions, then, in
the locative, it was used to form a progressive tense (of the
locative type), then that progressive tense ousted whatever present
tense was used before it.
BTW, Sanskrit has aa out-of-place 3rd sg passive aorist in -i.
Burrow:
"Neglecting the augment, which was a secondary and optional addition
to the preterite formations in Indo-European, it is clear that these
forms are nothing more than old neuter i-stems, without any
termination, which have been adapted to the verbal conjugation."
but:
"The oldest form, the locative [singular] without ending, appears
in ... the vr.ddhied forms of the the i- and u-stems."
Obviously, the -i of the passive aorist in -i can just as well be
seen as a locative -i, in which case the root of the verb could
serve as a verbal noun, as I assumed earlier.
But there is an odd fact I can't explain, and which I haven't seen
mentioned:
1) the personal suffixes in sg.:
-m
-s
-t
2) the case and gender suffixes of the Hittite pronominal enclitics:
-m
-s
-t
Huh? The two systems should be completely independent; why are they
using the same set of suffixes?
Torsten