Re: Greek conditionals

From: Kimmo Huovila (
Date: Fri Apr 07 2000 - 15:44:32 EDT

"TAYLOR, MARK D [FND/1000]" wrote:
> It was interesting for me to read the discusson on 1st class conditionals,
> because I have recently been trying to understand them as well. I read
> through Daniel Wallace's description in "Greek Grammar Beyond the Basics",
> which is the "assumed to be true for the sake of argument" viewpoint, but
> wasn't really satisfied with that.
> My own, uneducated feeling is that, while the 2nd class protasis is
> definitely false, the 1st class protasis is EITHER definitely true or
> definitely false (not just definitely true), and the 3rd class is more
> open/general (although not exclusively) and sometimes gnomic. To me, the 1st
> class is saying something "is" or "is not", used more in regard to specific
> occurrences. The indicative saying that the protasis is in regards to fact,
> but the context telling if it is factually true or factually false.
> Mark
> St. Louis

Thank you, Mark, for your input. Could I paraphrase your suggestion as
'the first class protasis assumes the truth of the premise to be known
by the speaker'? I am trying to figure the sense you give to being
definitely true or definitely false - is it supposed to say that the
truth of the premise is not indefinite or fuzzy or unclear (to whom?) or

I wonder if we can make the kind of conclusion from the indicative that
it is in regards to a fact, as the protasis does not present any fact.
The conditional sentence represents the relationship of two propositions
(if the apodosis is a proposition), and that is what the conditional
sentences says to be the fact. The issue is presupposition versus
statement of fact. The protasis can presuppose something, but in itself
it does not assert anything. Thus an understading of the indicative very
different from its meaning in normal declarative sentences is implied in
the argument. I have seen the argument used by grammars, but it seems
that often they do not stop to consider the basis for their argument.
For it to hold, it requires quite a bit more elaboration and study than
what I have seen so far (though what I have seen so far does not
constitute much of a proof).

The strength of the 'indicative' argument can only be assessed (to a
degree) by a cross-linguistic study of how different languages arrange
their conditionals. I have not done this, nor have I read a study where
such a thing has been done (but I am aware of Palmer's claim that in
many languages the past tense in the protasis is used when the falsity
of the premise is presupposed).

Or perhaps you meant that the indicative indicates that it regards the
factual state of a proposition. In that case I would not disagree, but I
do not see how that would differ with respect to any other type of
conditional, eg. with the subjunctive. In these cases also we are
concerned with the factual state of the proposition presented by the

If I venture to give a more detailed hypothesis (so far I have not made
enough study to assess its viability), I would suggest that 3rd class
conditions would present either neutral future conditions or generic
conditions. The future indicative is not parallel with the rest of 1st
class conditionals, because of the strongly modal nature of the future
(purely temporal cases seem to be a rather small minority with future
indicative). I assume that future indicative protases need to be
analyzed separately, and they involve some expectation, threat,
obligation, or
the like. The first class conditional could be the neutral past and
present conditional. The fourth class condition is stylistically marked
as archaizing (with the same hypothetical character as in Classical

It would be very interesting to make a study on how the conditionals
changes from Classical Attic to Koine with the disappearance of the
optative. Perhaps someone has already studied it. Any comments or


B-Greek home page:
You are currently subscribed to b-greek as: []
To unsubscribe, forward this message to
To subscribe, send a message to

This archive was generated by hypermail 2.1.4 : Sat Apr 20 2002 - 15:41:04 EDT