At 7:25 PM 12/9/96, Jonathan Robie wrote:
>At 06:55 PM 12/9/96 -0500, Don Wilkins wrote:
>>That is a good summary, Jonathan (I won't repeat it here due to length), so
>>I'll throw some fuel on the fire.
>I knew this was coming. This puts me in the slightly ridiculous position of
>trying to defend Mari, who is as bright as they come and knows her stuff,
>against someone who has much more experience with Greek than I do, and for
>whom I have a lot of respect, but here goes...
Thanks for the compliment, and you are absolutely right about Mari. She is
going to send me a couple of her papers to help us find some common ground.
>>Thus when Mari argues that the aorist
>>and present are not true tenses, I assume she is referring to the
>>indicative (Mari, if you are reading this, please correct me if I misstate
>As I understand her, and I don't think she says this explicitly, her table
>is independent of mood. The real tenses - imperfect, perfect, pluperfect,
>and future - are tenses regardless of mood, and neither present nor aorist
>are tenses, they both indicate only aspect regardless of mood.
Well, as Carl also pointed out, the only issue concerns the indicative. I
don't think any of us (certainly not I) would try to put time in the other
moods. On the contrary, this is a bad temptation for which I often have to
correct my students. Also, as a matter of fact, the imperfect and
pluperfect are not tenses regardless of mood--they are uniquely past-time
tenses and, not surprisingly, only exist in the indicative.
>Her paper indicates that the augment probably had meaning as a past-time
>indicator in all "tenses", but lost this meaning over time for the aorist.
>She does not establish the period over which this occured.
Again, this is at the core of the problem and it would be most helpful if
Mari would clarify when and how she thinks this happened.
>>The only way the theory of a meaningless augment can be objectively
>>evaluated is by a careful reading of all the relevant texts
>How would you define "all the relevant texts" for this particular question?
>My guess is that it would include most of the Greek corpus. Certainly any
>theory has to correspond with the data, i.e. the Greek corpus, but there
>*are* other ways to prove things.
All the relevant texts would indeed, I think, include most of the Greek
corpus, because the problem is the basic meaning of the aorist indicative.
Mari told me just a little while ago that she has mainly worked with the NT
and some patristic texts. To exclude other texts one would have to argue
that NT grammar is vastly different at its roots. As to other ways to
approach the problem, Mari also said that she is working mainly from
comparative linguistics, so I'll leave it to her to establish whether her
approach can substitute for an extensive study of the Greek corpus.
>Mari seems to focus on the concept of "cancellability". By examining how the
>"tenses" combine with other time indicators, we can determine which are the
>true tenses, the ones which retain their time reference even contrary to
>other indicators in the context. Consider the difference between these uses
>of the Aorist and the Imperfect in combination with NUN:
This is also at the core of the problem: the matter of whether our
interpretation of the context legitimately allows us to redefine the
essential meaning of the tense. Note that I previously remarked that Mari's
approach will always be the easier way, in effect cutting the gordian knot.
But the easiest approach is not necessarily correct. An analogous situation
exists in textual criticism, where most scholars agree that the more
difficult reading is probably correct.
>I've argued in a separate message that the Aorist examples, taken in
>context, do not refer to the past, but the imperfect clearly does. Why? Mari
>says this is because the imperfect is a true tense, and the aorist is not.
This is a matter of definition, not grammar. If by "tense" one exclusively
means time (i.e. not aspect), then Mari is correct, but must we then say
that the imperfect has (despite its name) no aspect? Obviously my view is
that the indicative is the mood of time, and that the aorist examples may
in fact essentially refer to the past, whether or not the context imposes
>In private correspondence I told her that I always do Gramcord searches when
>I'm trying to learn grammar, since I no longer know who to believe. She
> "Exactly. I do the same on other corpora for other things I work on.
> This is why I'm in computational lingusitics at all...verifiability
> for one's theories..."
>So I doubt that you'll find much disagreement with her that searching a
>corpus makes sense.
Amen. The problem is how to carry out the search.
And BTW Jonathan, keep up the good work.