personal responses?

Here's what I got that wasn't posted to the entire list (now that I
looked at the return paths).  

>There are hardly any syntactically analysed texts outside the NT & LXX
>amongst Greek texts so it isn't so easy outside the NT.  If I were
>wishing to know this I would take some or all of Polybius (for Hellenistic
>writing) or Dionysius of Halicarnassus (for early Atticizing) with
>TLG.  I would use Ibycus and ask for every instance of #ei# or #ean#
>and then check the context to see what % of the first  n 100 samples
>were of which tense.
>David M[ealand].

This was in response to my query concerning the relative frequency of
past-referring conditionals.  Since the Porter 1989 analysis is
tenseless, he would say that looking at the verb form would not
suffice, since he claims all forms may (in principle) be used in
connection with deictic indicators to refer to past, present, future
and timeless instances.  (THis is simplifying it somewhat, since he
claims that conditionals are outside temporal reference altogether,
with which I disagree).  SO one would need to look at context in the
broadest sense, to see, for example, which subset of the aorist
conditionals were actually past-referring.

Mari Broman Olsen
Northwestern University
Department of Linguistics
2016 Sheridan Road
Evanston, IL 60208