Biblical Hebrew Forum

[Rolf Furuli <furuli AT online.no>]

Dear Clay and Dave,

I agree that words (sounds and letters) just are semantic signals with no
intrinsic meaning (de Saussure). However, each word signals a concept in
the minds of those having the same presupposition pool (the common
understanding of the society). I discuss this at length in my book on Bible
translation, and argue that this realization is a very good foundation for
a *concordant* translation (provided that the target group wants such a
translation). This is so because the consequence of this view is that
*meaning* is found in the minds of living people, and neither in a lexicon
nor in the context of the words. The context does not generate new meaning,
it only makes visible which part of the concept (in the minds of people)
that the author wants to stress. True, new meaning can be conveyed by help
of combinations of words in a context, but this meaning is still dependent
upon the combination of the concepts signaled by the words.

Regarding Bible translation I have coined the concept "The contextual
fallacy" which has the following definition: The view that words (the
concepts they signal) do not have a meaning without a context and its
application to two different situations (the original situation and the
modern one) as if they were just one situation of communication." Polysemy
need some extra treatment.

The basis for my conclusion is Psycholinguistics and particular experiments
inside this discipline. Methodologically speaking, it means that it is
possible to find the fundamental linguistic units of a language and that it
is meaningful to work with such entities. Just as each basic word signals
one single concept (or two having been fused to one in some situations), so
the basic morphosyntactic forms signal one concept in the minds of those
having the same presupposition pool. For instance, several experiments
clearly suggest that word classes are real entities in the minds of people,
because one way to store words in the mind, is on the basis of word classes
(this should not be difficult for you Dave, who rely on Chomsky). And now I
come to my point.

To find lexical meaning in a dead language is much easier that to find the
meaning of morphosyntactic forms. Polysemy is in many cases easy to find.
However, the verbal system consists of just a few fundamental parts, and
from a synchronic point of view, it is likely that each part signals a
particular concept in the minds og those having the same presupposition
pool. This may of course change with time, and that is something that can
be studied. Because we do not have the original presupposition pool, we
have to study function, use, and context to try to find the meaning of a
form such as YIQTOL. I try to avoid assumptions as far as possible. But if
we do not assume that the YIQTOLs of the Tanach do not signal one concept
(until the opposite is shown), we will start our study with a great
handicap. Then we have no point of reference, and it is difficult to
evaluate our results. We should not confuse the functions of a form with
its meaning(s), but should find a generalisation (meaning) that covers all
its functions. I therefore assume that a difference in morphology means a
difference in meaning, but I am open for the possibility that there are
exceptions.


Regards Rolf


Rolf Furuli
University of Oslo







>> Dave Washburn wrote:
>>
>> Agreed! Actually I would apply this principle to all linguistic categories:
>> semantics, pragmatics and also syntax. In the more completely developed
>> version of my theory I call it "social convention." Basically it says that
>> words mean what they mean because a society chooses to use them that way.
>> I've been accused of doing a Humpty-Dumpty, but this is why I say "a
>> society" rather than "a person." In this regard, words have no inherent
>> meaning; the word "dog" means a canine, but it also means one who lags
>> behind, one who has questionable ethics, as a verb it means to follow
>> closely behind, and in Hebrew it means "fish!" Words are strings of sounds,
>> and it is only the convention of a society that determines what those
>> strings of sounds signify. That's why English "prevent" used to mean
>> "precede" but now doesn't, and why no word has an intrinsic uncancellable
>> meaning. In English, "bad" used to have a negative connotation that was
>> though to be uncancellable; then a segment of American society started 
>> using
>> it with positive overtones, and now it can either mean "bad" or "good"
>> depending on context. It is, however, much harder for a social convention 
>> to
>> change the force of a grammatical construction than it is to change the
>> meaning of an individual word; e.g. it took the Norman conquest to change
>> English word order from VSO to SVO.
>
>
>Dave,
>
>Back in 1997 (b-greek) I was branded a  Humpty-Dumpty for suggesting that
>any generalized theory of semantics needed to include both lexical and
>syntactical features of the language and that the principles that governed
>lexical semantics were equally applicable to formal language features used
>to mark syntactical relationships,  like verb inflection and word order and
>so on.
>
>This suggestion drew fire from certain folks who claimed to be doing
>functional linguistics. It was suggested that viewing aspect marking of the
>verb as semantically and functionally indeterminate was a slippery slope
>toward grammatical chaos.
>
>I would agree with Dave (quoted above) that the semantic function of some
>syntactical features of a natural language may be more stable than the
>semantic value of some lexical item. However,  I am not sure that language
>change is really the most important aspect of this question.  If we take a
>snapshot of a language state I think we will discover that there is some
>polysemy present at any given time in the grammatical system of the
>language.
>
>What boggles my mind is the lack of willingness on the part of folks who
>take polysemy for granted in lexical semantics to recognized and accept
>polysemy in the relationships between formal features of the grammar and the
>syntactical functions marked by those formal features.  People who would
>never dream of doing a concordant translation in terms of lexical semantics
>will turn right around and defend to the death a concordant translation in
>terms of verb aspect marking or some other feature of syntax.
>
>Once you accept a model for doing semantics, the relationship between formal
>grammatical features  and syntactical functions should IMHO be treated in
>the same manner as the relationships between a lexeme and meaning. This
>would include the acceptance of polysemy, the mapping of formal features to
>functions in a complex domain network, all of the methods which are common
>place in lexical semantics.
>
>Clay