By JLS
-- for the GC.
Jones, "Reductionist", THIS BLOG:
"I am tempted to assign to pragmatics rather than semantics, but I am a pluralist like Carnap, so if you want to define or model(analyse) a language in terms of speaker's meaning then go right ahead and we will see how successful you are."
The problem is that Grice loved a television programme (He was called "Trekkie" because he would never miss an episode of Star Trek). A programme is a programme is a programme.
Grice speaks of 'programme'. He never works it out. That's NOT the task of the philosopher. The philosopher designs the programmes. The technicians work it out.
It took me to own a channel almost to realise that. I would write to zillion of people, "I am engaged in some research work in the Gricean programme of conversational pragmatics..." etc. This usually got me good stuff from people -- till I met Schiffer.
He had said, "Gricean programme, kaput". He signed my copy of PGRICE though, "That he may SAVE the Gricean programme". I stay tuned...
-----
The Gricean channel is a channel that only transmits Gricean programmes. There's the commercial for the Gricean formula, which reduces the semantic to the psychological in such a way that you won't see the difference.
So, the point about utterer's meaning -- or rather locutions of the form: "By uttering x, U meant that p" iff U intended that... -- is seen as a 'programme' i.e. what a philosopher says about things that matter. Surely we cannot go and define this for each word, e.g. 'pig', or 'runt' (Grice's favourite).
This works case by case, and on request. "Runt" for example, his Aunt Matilda uses only LITERALLY, but she'd rather seen dead than using "runt" to apply to, say, Mr. Smith. (i.e. 'runt' as 'undersized person' as Grice has the gloss of this metaphor. WoW:vi).
So, the philosopher's task is to clarify issues or presudo-problems by people who use '... mean ...' loosely. Here fits Quine and his inability to distinguish clearly between ' ... mean ...' and ' ... is used ...'. The conditional, for example, as branded by Strawson and Quine, and Carnap, -- 'if' -- what does it _mean_. How is 'if' used?
Surely the horseshoe of the logician is harder to pronounce, but even in Logiclandish there ARE idiolecta divergences. Recall C. I. Lewis who was so infuriated by Philo of Megara that he had to COIN an fishhook to do the duty of the horseshoe. Etc.
This is Griceian casuistics. We work on a case-by-case basis. It's programmatic, said smugly.
Subscribe to:
Post Comments (Atom)
Posts
I'm inclined to disagree with myself, as here quoted, and suspect that I have confused some matters which I should now like to distinguish.
ReplyDeleteI am certainly inclined to stick with Frege and Carnap in keeping psychology out of semantics (but not with Frege's recklessness in trying to use the word "thought" for something non-psychological, assuming that in German that was as bad as it sounds in English).
However, if we must try to understand natural languages, and I suppose that is a good idea, then I do think semantics has to come to terms in some way with the fact that not all of us use the language in the same way, that some of that divergence is in what we suppose our language to mean, and that if we want to understand what someone has said, it is his language not ours that counts.
Still, I am reluctant to admit psychology into the semantics of even this kind of language, except insofar as psychology is actually the subject matter.
Two possibilities come to mind for addressing "speakers meaning" (or writers for that matter).
Either we should regard each speaker as having his own particular variant of the language, so that we don't actually speak the same language, or else we absorb all this particularity into the general sensitivity of meaning to context (or at least of the proposition expressed to context, one might still have a single "meaning").
This is horrible, but there may be nothing better, natural languages are a mess, however wonderful.
This however does not necessarily get psychology into the semantics.
The issue at stake is not what the speaker intends, it is how he understands the language, and if we construe the semantics as indexed by a complex context which includes the speaker, it is the identity of the speaker which is in there, not his state of mind at the time of speaking.
"Speaker meaning", insofar as it is relevant to the meaning of the language, is not what he intended, it is what his sentence meant in the language as he understands it, it is still possible for someone in his own idiolect to make a mistake and say something which he immediately (or later) recognises as not what he intended to say.
Now I will make yet another concession to psychology and say that in some kinds of language you really can't expect a good account of how the language works without getting into psychology. Body language might be one of these kinds of language, or, perhaps the language with which a great tenor moves our hearts (if we count these as languages).
However, this is my preferred dividing line betweem pragmatics and semantics.
As language evolves it moves from being some kind of "game" (though not often that trifling) into being a vehicle for objective content. This transition is very important, because it is this transition which makes knowledge communicable and recordable, and it is this transition which makes deductive reasoning possible (before this the notion of sound deduction makes no sense).
The point of separating semantics from pragmatics (terminologically) is that the terminology marks the distinction between this objective propositional use of language from the more diverse (and sometimes even more important) ways in which people may play with words.
So how does this place me viz a viz Grice?
RBJ