>>The most important objection is that the beginning often is included in the
>>imperfective aspect.
>I would suggest that this confuses semantics and pragmatics. All my def.
>requires is that any emph. on the "beginning" is _not grammaticalized_, but
>comes from the context.
Dear Rod,
Methodologically speaking you are correct, and the reason why I defend this
"confusion" is that I believe we will achieve more with the traditional
hermeneutic circle-approach than by trying to differentiate between
semantics and pragmatics in a dead language. Let me tell why.
THE ROOT FALLACY
Barr«s 1961-book truly was "a trumpet blast against the monstrous regiment
of shoddy linguistics, as Moises Silva (1983:18, Biblical Words and their
Meaning) expressed it. While the modern "Semantic domain" approach builds
on a better methodology, I also see serious problems with this method. The
basic problem is that it relates to a situation of communication different
from the original one. When for instance John wrote his gospel, he and his
audience (broadly speaking) had similar presupposition pools and the words
he wrote served as semantic signals for concepts with fuzzy edges in their
minds. In this original situation of communication, one word, say KOSMOS,
could be used with different nuances, but based on their common knowledge
and the context, they would understand it. (Based on psycholinguistics I
reject the idea that words have no meaning without a context. The context
makes visible the side of the concept intended, but adds no meaning to the
word/concept).
Our situation of communication is that of translation. We have a completely
different presupposition pool, and by help of this and of the languages we
understand, we want to understand and translate the original message of the
Bible. In this situation the "semantic domain" approach tries to outline
some of the meaning that the original audience automatically got because of
their presupposition pool. This task is not an easy one, and in the
lexicons based on this method, much theological material is read into the
enties from the presupposition pools of the authors. I see two extra
problems, namely the lack of means to control the authors and that to much
power is in the hands of the authors at the expense of their readers. One
alternative would be to take the best of the root fallacy and of the
semantic domain fallacy (my words) and and try as much as possible to
mimick the original situation of communication and let the word serve as a
semantic signal for its concept rather than to try to find a semantic
domain (I am thinking of Bible translation).
My point is that even in word studies and case studies, which are much
easier than the study of aspects, the new methods are not so unimpeachable
as viewed by most, and a differentiation between semantics and
pragmatices, while methodologically desired may have problematic side
effects.
PRAGMATICS VERSUS SEMANTICS IN ASPECT STUDIES
Does the aorist both represent the perfective aspect and past time? Porter
says no, Fanning says yes (I look forward to reading Mari«s book which I
have ordered but not yet revceived). This illustrates the problem of
differentiating between pragmatics and semantics. We are working with a
dead language without living informants. Before us are different planes of
meaning such as time, aspect, mood, lexical meaning and linguistic
convention. These play together and contribute to meaning, but what is the
role of each one and what is inherent in each one? When most instances of
aorist have past meaning, is past time an intrinsic part of aorist, or is
it just the convention that because of the nature of the aspect it is used
in past contexts? I cannot see how we can answer the question without
studying all the uses of aorist. So how can we isolate meaning (semantics)
without studying the whole clauses in context (pragmatics)?
I started my study of Hebrew aspects using the model of Carlota Smith and
Carl Backe. Testing the model against the corpus I soon realized that I had
to change it (Hebrew is easier than Greek because we need not in the first
run account for time), and after much work, the model based on
distance/scope was made. This must of course be further tested. Applying
the model to the LXX and the NT I found that the fit was just as good as
for Hebrew. I fully agree that we should try to differentiate between
semantics and pragmatics in all our work, and while working with modern
languages with living informants this is possible, but I cannot see that it
is really possible, with dead languages. So I think the best approach is to
try to isolate as small units as possible and test our model against all
the ways each unit is used, and then udjust our model when necessary.
The question we are discussing is very important and I greatly appreciate
viewpoints from others on the list. I am also working with a Ph.D thesis
and sound methodology is mandatory for a good result.
Regards
Rolf
Rolf Furuli
University of Oslo