--- by JLS
----- for the GC. (Feel free to distribute in the Carnap Corner, :))
In this online site with Carnap's "Concepts of Pragmatics" (his appendix in "Meaning and Necessity") he writes of an 'assertion' as being (p. 249)
A(X, t, S, L)
i.e.
t: time
S: sentence
L: language
with
X: utterer
Carnap glosses the above as: "X deliberately wills [Carnapian for Gricean 'intend' -- or better 'mean-N', WoW: 'Meaning' (1948) -- 'mean to' as 'natural-meaning'] to utter a token of S [utterance type or 'sentence'] as a sentence of the language L in the sense of an assertion" (emphasis on 'assertion' Carnap's).
He adds in a way that confirms my reading of 'deliberately willing' (I mean, I cannot be THAT obtuse):
"Since [this] concept [of assertion] involves the purpose or intention, it is clearly a theoretical concept."
----
But Carnap, like Grice, wants to play 'rough' -- we need a rougher concept, extensional, or 'observational' if we can: (when we can):
U(X, t, R)
with
R: "a series of audible sounds" produced by X with the aid of "his speaking organs".
Even if we were to replace "R" (sounds) by "S" as in the above definition of "Assertion" (A) -- to mean a TOKEN of a sentence, this would not turn the replacement,
U(X, t, S)
----
into a 'theoretical' construct -- 'utterance' is thus, unlike 'assertion' and 'belief' merely observational.
----
At this point I want to be more Gricean than Grice. For Carnap wants to say that there is a 'fact': viz. that X uttered S as a token of a "sentence in a language". Since it IS a fact, Carnap assumes, we must explain it -- theorise it even, since it cannot be 'observed directly' but has to be inferred.
Personally (as Jones and Kramer are well aware) I have never totally internalised a grammar (of L) so I cannot see how I can be said to have uttered a sentence of, say, English. I just rely on some observations, usually wrong -- "at a later date" vs. "at a later time", say. --- But then I have Carnap telling me that I DO have an inner-language, only that it is "difficult" to see. Etc.
--- I would think that was one of the reasons Grice stuck with utterer's meaning and gave a hoot, most of the time, as to what sentences and sentence-parts meant (""marijuana" means 'grass' in Southern California"" and "'grass' means "marijuana" in Boston" -- or vice versa -- I mean, who cares? A dealer is a dealer is a dealer).
---- (The example is from Grice, "marijuana" = grass, cannabis, WoW:v -- section ii).
Carnap continues in somewhat anti-Gricean ways (when I'm in an ultra-Gricean day), to speak of the 'normality' of the situation, when X, an utterer can be taken, qua utterer of S -- a token of a sentence in a language L -- to have ASSERTED the sentence. "It rains".
-----
He ultimately wants to reach the stage when we can 'allege' he believes (asserts, alleges -- as Jones notes) that: it rains. Or that snow is white.
To assert that snow is white when you believe that snow is white, by uttering 'snow is white'.
To assert that grass is green when you believe that grass is green, by uttering 'grass is green'.
BUT NOT:
(unless you are joking at the Grice Club):
To assert grass is white, or snow is green, unless you've had too much of the former and think an Eskimo is Daltonian.
---
So it was mainly the closing paragraph of this two-page thing where Carnap concludes that a system of 'theoretical pragmatics' which seems to be much in need 'for analytic philosophy' (think Grice's and Strawson's reply to Quine's confusion of 'meaning' and 'use' for example -- as per Strawson 1968, for example -- Introduction to his "Philosophical Logic" collection, or better, Grice's Prolegomena to WoW:i --).
"Such an outline [of theoretical pragmatics] may
first be restricted to small groups of concepts
(e.g. those of belief, assertion, and utterance)" to be later
extended to cover all of analytic philosphy, almost. Good approach!
---
Carnap can write like that because he thinks 'theoretical semantics' is pretty much developed as it is. The problem is: as with Grice: these two were geniuses. But with EACH generation, it's starting from Square 1 again --. Imagine: you have to teach the zombies Frege's Sense and Reference and totallly fail in the proceeding!
---
=======
Subscribe to:
Post Comments (Atom)
Posts
As you probably guessed, I have not myself paid any attention to pragmatics, and I am still in the dark about why Carnap considers pragmatics important (in fact, up until about 5 minutes ago it had not even occurred to me that he might think it important).
ReplyDeleteClearly Carnap was mistaken about pure semantics being in the bag, the full significance of the Quinean dragon evidently not yet apparent.
It is possible that the concerns which Carnap thought to address by pragmatics are related to those which I aim to address by "X-Logic" and epistemic retreat.
There is a little muddle of concepts (is that the correct collective?) in this area that I should do something to untangle in my head. Pragmatics, propositional attitudes, forms of judgement.
I have thought of X-Logic as involving epistemic retreat managed through relatively complex forms of judgement, but began to wonder whether it might be better to keep the forms of judgement simple and do the epistemic retreat using languages with a nice range of propositional attitudes.
Then I might ask, can we dispense with pragmatics (or some of it) by using propositional attitudes to move the relevant material into "pure semantics"?
Is this even faintly intelligible?
Probably not! I need to come up with some examples, and a stab at X-Logic (there are some formal models up there, but nothing which addresses my present aspirations).
Yes. I will have to re-read the topic: that appendix and what Carnap thought he was addressing. As you mentioned in your first comment on this, it was Carnap's explicit reply to Chisholm, and it strictly involves a 'pragmatic' approach to 'intension', that it was NOT Carnap's intenTion (or it was, I forget) to left unanalysed, as it were.
ReplyDelete----
The symbolisms he uses to define 'assertion' versus 'utterance' (his "U") are interesting per se. The role of 's' qua sentence of the L makes all of the difference, it seems. I was wondering about completeness there -- but I suppose he could retreat to the idea that in a formal language, for any string, we CAN decide algorithmically if it IS a sentence of L or not -- I suppose this trades on 'grammar' but it need not -- but at least indeed 'syntax' CUM some sort of 'semantic'. For what sort of an utterance is "Pirots karulise elatically". How can we assert it, or allege that pirots karulise elatically? It's the pirot's revenge as it were.
So keep doing the good thoughts and report -- when you can! --.
His use of Church on 'sense' rings a good Gricean bell, too. That's of course the 'intension' simpliciter. And Carnap is suggesting the compositionality: i.e. that the intension of a sentence s (of Language L) is defined by the intension of its components -- what Grice would have as the referential and predicational sub-components -- which both Carnap and Grice agree -- can be provided 'extensional' definitions for.
----
That sort of accounts for 'assertion' and 'utterance'. And your idea of the gamut of propositional attidudes is a good one. Carnap suggests to stick with 'belief'. Grice considered this, and suggested, too, an alternative where we stick with 'desire' -- it doesn't matter which one one takes as primitive, Grice says in the appendix to Grice, Conception of Value.
-- (I personally think it DOES matter, and would only stick with 'desire').
-- A further Carnap/Grice overlap is then the idea of the 'saying' (what is said by the utterance of sentence s -- qua beliefs or desires literally expressed by them. Think, "Close the door!") versus what is just 'empirically' connected, according to Carnap, and via an appeal to rationality -- Carnap's 'normality' of the situation --. This for Grice could just as well be expressed ("I am cold" -- and when there is a divergence between a logical constant and its vulgar counterpart, he suggest that the alleged divergence thus does not pertain to the 'sense' but to the 'use' of the constant. Etc.