Carnap and Grice on Morris

By J. L. Speranza


In his note on Chapman on Carnap, Jones takes issue with Chapman's rather simplified view of syntactics vs. semantics.

We are discussing

pirots karulize elatically
A is a pirot
-----
A karulizes elatically

-- tr. from the German original
by A. Smeaton, Countess of Zeppelin, 1937

---

Chapman notes that Carnap is sticking to 'syntax'. A point she and indeed Carnap fails to give enough attention to is that some sem- (call it semantics) is present in the above: nothing is said about the morphogrammatical words like the -s marker for singular, the -ly marker for adverb, and the totally meaningful formation of "is a" (in "A is a pirot").

So, what Carnap seems to be concerned here -- but I wouldn't know what form he gives to universally quantified sentences:

(x) Px --> KEx
Pa & KEa

i.e. he _is_ using "A" as a singular term ("A is a pirot"). In Quine's scheme, and indeed Grice, in his talk "How pirots karulize elatically", this "A" comes out as "a" as per above.

Then we have to turn onto a semantic bloc the idea of karulising-elatically (KE) because the logical form of adverbial modifiers was unknown till Davidson's first shots at this in "The logical form of action sentences".

But if we see the above, we see that it's only with the "extension" of predicates:

"P" and "K" (or "KE" if you wish) that he is concerned as leaving 'meaning-free', or as I prefer, "interpretation" free. This may be, as Jones suggests, a bit of an influence from Hilbert (rather than Frege) and that's a very good thing to have, if we are going to deem Carnap a die-hard formalist, for nobody can be more of a formalist than Hilbert was.

I _love_ the label 'formalist' because that's precisely the one that Grice uses to oppose these people (say Hilbertians -- here we may need to see the Hilbert/Russell-Whitehead interface) and, ... Strawson! (Who Grice, amusingly, calls an 'INformalist'). Later, Grice learned the lesson and changed this into modernists versus neo-traditionalists!

---

Roger Bishop Jones writes of Carnap's oevure where, to use Jones's apt wording, he

"he syntacticised all the relevant semantic concepts" (emphasis mine. JLS).

For it's precisely that what we want. Let's propose four words here:

to semiotise: to turn into a semiotic system.

This has three subbranches:

The first is indeed:

to syntactise: to turn syntactic. Where syntactics is that branch of semiotics that deals with formation and transformation rules -- Carnap was enamoured with syntax, and so was Grice. Carnap wrote of "the logical syntax of language"; Grice has an unpublished thing on "The syntax of illusion" (for an investigation of phenomenalist talk _sans_ physicalist backing: 'that stick is bent').

Then we have

(b) to semanticise. This is done via interpretations. It's the truth-table thing. BUT it can be done more 'formally' via 'syntactising'. I.e. no interpretation required, or truth-table indeed. The system remains purely a syntactic one.

(c) to pragmaticise. This is done via leaving room for things like 'assertion', which is Carnap's paradigmatical pragmatic notion. One can have a syntactic string ("(x)Px --> Kx") and we may go on to semanticise it ("P" -- meaning postulate, "K", meaning postulate") and we may go on to pragmaticise it by examining what belief the utterer is endorsing when he asserts to pirots karulising elatically.

Etc.