This is undoubtedly on the core problems of case syntax. I
understand that when you say "persons" you include other beings capable of
"grammatical" personification (QEW/=, A)GGE/LW/, ktl.) as well.
The kind of Indirect Complements (IC) you can find in a given text
depends to certain extent of the kind of text: v. gr. if you look for IC in
the Batrachomyomachia, most of the ICs will be mice or frogs, and so on. In
the same way, "keraunoi=s" is not a dative that you expect to be used with
instrumental force unless you are reading about Zeus (or B. Franklin, for
that matter).
In my opinion, it is preferable to speak of "positions typically
occupied by a human", a locution that you can simplify as "human" only when
everybody agrees that the term covers the mentioned "typical positions".
Then it is easy to find non-humans in position typically human:
ei)pe\ tw=| li/qw| tou/tw| Eu.Luc.4.3
tai=s e(te/rais po/lesin eu)aggeli/sasqai Eu.Luc.4.43 (not local,
IMO)
the usage is classical:
th=| po/lei (...) le/gein X.HG 2.3.35, cf. 2.3.15
And now it's when the real problems arises: a) how you define the
"typical position"? b) is it the CI a grammatical function to be defined in
terms of the lexical meaning of the noun and the verb? mmm. I wait for the
list's commentaries.
Besides, I would like to emphasise a point that seems really
important to me (but maybe because I am the only one who doesn't seem to
realise the obvious explanation): The frequency of the dative in Lucas is
*much* greater than the frequency in Classical literature (my estimations
are: Eu.Luc 11: Thuc. and. Plb. 8: Xenopho 9): in a time when the dative is
clearly disappearing. It's uses, true, are very limited and the personal
dative takes the part of the lion: it is however, a use that I think hasn't
been fully explained.
___________________________________________________________________
Daniel Rian~o Rufilanchas
c. Santa Engracia 52, 7 dcha.
28010-Madrid
Espan~a
e-mail: danielrr@mad.servicom.es