I'm not sure that I'd see a great difference, if any, between the
functioning of the two constructions; EN + dative here is clearly
instrumental rather than locative and so indicates "means" -- while DIA +
genitive here must indicate the efficacious agency.
>Is DIA here a weaker hUPO? I say this because "faith" is not something
>tangible that can literally be an agent to safeguard one. So it looks to me
>like "intermediate" agency. The "chosen ones" exercise faith and because of
>this, God safeguards them. Correct?
Well, it certainly would be odd to see hUPO here unless "faith" were being
personified in the manner that Paul sometimes personifies Sin and Death,
thereby permitting them to be used in an agent construction. I don't know
that I'd characterize the relationship between the two prepositional
phrases and the verb quite so precisely as you have done (you make it sound
like the safeguarding is a "reward" for having faith, which would seem to
imply that "faith" is a form of "works." I'd rather see the safeguarding
presented here as a process sustained simultaneously by God's power and the
faithfulness of believers.
>If we say that EN here is dative of means, then we would say that God
>(agent) uses power (means) to safeguard.
>
>So why doesn't the author simply use DIA in both places? or EN in both
>places? How do I distinguish the uses of agency/means here?
I'm not confident that it's possible to make any sharp distinction here. If
one were employing an Aristotelian aitiology here, I suppose that one could
say that EN + dative here indicates a "material" cause while DIA + genitive
indicates an "efficient" cause; but I doubt seriously that the author of 1
Peter was thinking in Aristotelian terms. The guess I would hazard is that
he wants to distinguish between the divine and the human factors involved
in the safeguarding. Alternatively it might be that this is simply a matter
of rhetorical variation. In any case, for my part, I wouldn't try to draw a
very sharp distinction between the semantic function of the two
constructions here.
Carl W. Conrad
Department of Classics/Washington University
One Brookings Drive/St. Louis, MO, USA 63130/(314) 935-4018
Summer: 1647 Grindstaff Road/Burnsville, NC 28714/(704) 675-4243
cwconrad@artsci.wustl.edu OR cconrad@yancey.main.nc.us
WWW: http://www.artsci.wustl.edu/~cwconrad/