weqatal in Jdg 3:23

Rolf Furuli furuli at online.no
Tue Oct 6 03:02:07 EDT 1998

Paqul Zellmer wrote

>This is the second posting you have sent on this subject.  I was one of
>the addressees
>on the first.
>I can't speak for the others, but I, for one, am not trying to "force upon the
>language modern theories."  Rather, I am doing what the scientific method
>namely, taking a theory and bouncing it against the data.  When the
>results don't
>match the predictions, or the data seems abberant, I don't blame the data.
>I blame
>and seek to adjust the theory.

>And different people try to study a language in different ways.  Some of
>us are
>currently concentrating on discourse analysis.  You, at least in the past,
>checked the raw data from the viewpoint that all yiqtols are imperfective
>and all
>qatals are perfective.  Are you trying to suggest that you did not use
>your theory to
>"outrule both the brute data and the institutional data of Hebrew."
>working in a two "tense" language in the Philippines, I have no problem
>with your
>theory.  But that doesn't mean that the function in an English translation
>reflect the underlying mindset of the Hebrew.  Only colloquial English and
>certain few
>novelist would generally use the present tense to relate historical
>Just because we have different theories, and those theories cause us to
>focus on
>different "problem areas" does not mean that the adherents to a theory
>cannot allow for other forces to change the data.  Idiom frequently overrides
>grammar.  That doesn't bother me in the least, nor does it make me try to
>force my
>grammar on the idiom.

>But we will never know if we *have* minute differences if we don't see the
>from the model predictions and then investigate to see if they are
>significant.  I see
>perfectly logical cases to be made for the weqatal to be off-line
>information in the
>relating of the dream.  And, if you read the note to which your first note
>was in
>response, you will see that I very clearly noted that the data base was
>very small.
>The implication, which I didn't feel I needed to express, was that the
>were very weak.

Dear Paul,

Let me start to say that my two posts were not intended to be ad hominem.
I am not accusing you or Bryan or anybody else to be stupid, or denying you
the right to test your theories against the material. To the contrary, what
you describe in the paragraph above is exactly what I recommend. I am
concerned , however, with the application of Harald Winrich`s model
(expressed by Niccacci) to biblical Hebrew in a way which prevents what you
say above. To state it technically: Discourse analysis is used as an
explanation for the Hebrew verbal system in a deductive-nomologic way, and
that is bad methodology from a linguistic point of view, because it
prevents what you describe above. Let me explain.

The deductic-nomologic explanation (modus ponens) is typical for the
natural sciences, and the explanation (explanandum) is reached because a
natural law is used as premise (explanans). Example:

Water expands when freezing   (explanans)
A is water and freezes             (explanans)
A expands		               (explanandum)

A deductictic-nomologic explanation used in linguistics (i.e. using
something comparable to a law of nature as explanans) would prevent us from
"taking a theory and bouncing it against the data". A deductic-nomologic
explanation has no place in linguistics! What I have criticozed is the
following explanation applied to Hebrew:

All verbs used in narrative texts are perfective
V is used in a narrative text
V is perfective

I have no objection to the formation of the HYPOTHESIS ( not law): "All
verbs used in narrative texts are perfective.", and then test this
hypothesis against the data. The important thing, however, is that such a
test cannot be done by help of discourse analysis because discourse
analysis is concerned with chunks of text and not with singular words (a
test by discourse analysis would be hopelessly circular). One important
question to answer in the test of this hypothesis is: How can both prefix
forms and suffix forms, participles and infinitives be used in narratives
if narrative verbs are perfective? Do the forms have one value in one
context and another in another context? This can only be answered by
starting with the brute facts (the fundamental lexemes) and this is outside
of discourse analysis.

I value discourse analysis as a useful tool covering a particular area in
linguistic study. What clearly is wrong, is to use it, or parts of it as
laws in explanans. The only things that can be used in explanans in a sound
linguistic study are rules, i.e. deductive systems expressing institutional
relations. And these rules are open to modification, when it is discovered
that they do not fit the data. In this way can we use the hermeneutic
circle, bouncing the rule against the data.  This is not the way discourse
analysis use to be applied to Hebrew, and I feel it is important to point
this out. To have different theories and use different approaches is fine
indeed, but the methodology should be sound.


Rolf Furuli
Lecturer in Semitic languages
University of Oslo

More information about the b-hebrew mailing list