> > The present-stem-as-participle theory is not mine, and it is
> > supposed to hold for Finno-Ugric too [...]
>
> Doesn't work. Since we have the indicative beside a _non-
indicative_
> without *-i, it's clear that it's not as simple as that. Both the
> indicative and the non-indicative must derive from *i-less
**VERBS**.
> Given those facts, we can't simply say that the indicative is a
> locative noun/adjective/whatever while the non-indicative was
always
> a verb and just leave it at that.
Or, as I just did, claim that the primary endings once had a now-
elided -u, which would place primary and secondary forms of verbs in
the same syntactic category, namely that of a noun in the locative,
ie. an adverb. In order to get my parser to parse that I would claim
that sentences were of the form:
'I (am) in-my-V-ing'
(and mutatis mutandis the rest)
which means that this progressive construction differs from the
English etc one by a _superfluous_ possessive suffix. A comment from
Burrow: "The Sanskrit language" (p. 314):
"...The relation of the series -m, -s, -t, -an(t) with the primary -
mi, -si, -ti, -anti can only be explained by the assumptionthat in
the latter series a particle -i indicating present time has been
secondarily added. In the same way in the imperative endings -tu, -
antu, a particle -u is added to the same basic terminations. This is
made clear, among other things, by the fact that these same
elements -i and -u may appear by themselve in formations that have
no personal termination. eg. in Greek 'phérei' ('phere' + 'i'),
Hit. 's´akki' "knows" and in Hittite imperatives of the -h_i
class: 'aku', 'da:u' ('ak-' "to die", 'da:-' "to take""
Now, this seems to indicate that the verb stem and the 'personalised
verb stem' obtained by adding a 'personal possessive suffix' were of
the same syntactic category.
But I have another possible solution also for the secondary endings:
they are gerunds. In other words 'V-m' is "me V-ing" in English.
This matches nicely with their use in "remote" tenses and moods:
they are to be used in originally dependent constructions (not full
clauses). The 'clause'-status of the constructions these
personalised gerunds appear in appears only when those gerunds are
reinterpreted as finite verbal forms.
>
> There is nothing, absolutely nothing, about the nature of the
indicative
> that makes it any less "verb-ey" than the non-indicative. So this
theory
> is really based on nothing but a whim.
>
> As for Finno-Ugric, I'm not aware of any participle turned verb.
Sounds
> just as suspect in FU or Uralic as it does in IE.
>
Ah, nostalgia. I'm reminded of myself your age: young, talented,
dogmatic. I'm glad you havent't entered politics.
>
> > Therefore, and since I don't have a Finnish textbook I looked in
my
> > Hungarian textbook. It does have personal suffixes, they can be
> > attached, apart from nouns, also to participles, [...]
>
> Yes? And? We also see Egyptian verbs marked with what appear to be
> possessive markers. This happens a lot world-wide. It doesn't
apply to
> IE because it doesn't have possessive suffixes. The suffixes *-m,
*-s
> and *-t are exclusively _verbal_. That's it. So you have only your
> imagination to go on.
Interesting that you should say that, since I just discovered at
least one IE language has possesive suffixes, namely Hittite. And
they are in the 1st, 2nd and 3rd sg.:
-mis^
-tis^
-sis^
in the nominative.
My textbook calls them 'enclitic possessive pronouns', but I think
it only does that because everybody and his brother knows IE has no
possessive suffixes. Syntactically they are suffixes to object owned
(check), and semantically they serve to indicate owner to 1st, 2nd
and 3rd sg. (check). So as far as I'm concerned, they are personal
possessive suffixes in an IE language.
Now, the intersting part, can they be inflected? Yes, they can.
Unfortunately Hittite has no locative, it seems, but we could try
the similar dative, which is:
-mi
-ti
-si
Surprise! I don't think the match gets much better than that, except
for the 2nd and 3rd having been swapped. But, as well known, we do
find verbal 2nd sg. -t and 3rd sg. -s in IE languages. Personally I
like to believe that the original sense of *n-(*nW-?), *t- and *s-
is deictic and that the confusion took place at that stage; it's
easier to confound categories of demonstratives ('iste' "the one at
thee" with 'ille' "the one at him") than personal pronouns ('thee'
with 'him').
Torsten