[Cogling-L] Proposition query

[Tahir Wood]

Chris, thanks, a couple of brief ripostes below.

>>> "Chris Sinha" <Chris.Sinha at port.ac.uk> 01/27/09 1:52 PM >>>
Dear Tahir,

Taking a psychological or cognitive stance does not solve the problem,
since there is no definition of a proposition that is theory
independent.

Tahir: No, I wouldn't have expected it. I am looking at a wide variety
of approaches and different notions of proposition. I must say I have
come to doubt whether there is anywhere a really good cognitivist
definition that is even theory-dependent, which is why I put out this
request for information. I am working on a theory of the proposition,
and I have not been able to find something like this, but of course I'm
happy to be proved wrong.

 Most generally (and I guess theory - neutrally) a
proposition is a single attribute predicated to a (not necessarily
singular) argument. 

Tahir: This is a grammatical definition of the kind I'm not satisfied
with. It implies that there is lexical selection as part of the make-up
of the proposition. If you take that as your starting point you end up
saying that a proposition is not different from a predication, and,
moreover, it is therefore tied to expression in one specific language
only. No way do I think that a proposition (if it really is a content,
meaning or thought of some kind) can be this.


Classical Cognitivism in the style of Fodor
considers propositions to be logically structured entities of this
sort
couched in the Language of Thought, or (secondarily)  the natural
language translations of these. If however you consider (as many do)
natural languages to BE the language(s) of thought, then propositions
are linguistic entities and propositional reasoning is
language-dependent. Different theories have different psychological
consequences, amongst others that in classical cognitivism infants
(possessing an innate LoT) can be legitimately ascribed propositional
attitudes such as belief, whereas if propositions are natural language
dependent only language users are able to entertain propositional
attitudes.

Tahir: "Language-dependent" is very vague. Does it mean the strong
version of the Whorf hypothesis? "Propositional attitudes" implies that
you already know what a proposition is. A proposition cannot be a belief
for me, because it belongs to ideation, not to truth, logic or
reference. If 'proposition' is going to just mean the same as things
like 'belief', 'judgment', 'reference', 'declarative sentence' or
anything similar then we don't need it; i.e. then it doesn't exist, as
Quine and others have said. But I think it does exist as a mental
operation, or as a semantic form, if you like.

Fodor takes propositional attitudes to be directed to propositions in
LoT, bracketing off the issue of the referential relation of these to
the  world (traditionally this is supposed to be dealt with by truth
in
relation to "states of affairs" or facts). A qualified realist but
language-relative approach consistent with Cognitive Linguistics would
say that the attitudes are directed to linguistically conceptualized
or
construed referential situations; thus the referential scope of Eng
"the
cat is on the mat" is different from that of Dutch "de kat ligt op de
mat" or of utterances in more elaborate positional languages such as
Tzeltal, or body part locative languages such as Mixtec or Zapotec;
and
there is neither a LoT into which each of these are translated nor an
unmediated relation to "states of affairs" (though we can assume that
there is a world independent of our linguistic concpetualizations of
it).

Tahir: But translation is possible in many cases. I don't accept a
strong version of the language-relative argument, because it means that
all the different languages are totally incommensurable and so therefore
are all the thought processes of the groups who speak them. I think
languages only differ radically in terms of their possible
meanings/contents to the degree that the experiences of their speakers
are different. And then this also applies to speakers of the same
language who have very different experiences of life! So there you go.

 The language-relative argument really doesn't take us very far and it
can just become a dogma. Why not go all the way and say that
propositions only pertain to some languages and not to others?
Anthropolgy also shows that very farflung people with very different
languages often come up with the same things, bow and arrow, incest
taboo, etc. etc. There is one very compelling idea here that is
motivating my research and that is onomasiology: Meanings arise before
and independently of their precise mode of expression. That is why
identical meanings can appear in different cultures; in fact that is why
they can appear at all. The semasiological view, on the other hand, says
that first there is a sentence (or expression or whatever) and THEN it
has a meaning. We all know that this is the dominant approach to
linguistics, but it is implausible. There could never have been any
linguistic innovation (nor any original thought) in history if that were
true. 



-------------- next part --------------
All Email originating from UWC is covered by disclaimer 
http://www.uwc.ac.za/portal/public/portal_services/disclaimer.htm