Specifications of Meaning -- Who Does NOT Need them?

--- I was going to title this, "Who needs them?", but found that rude. So there.

----- By JLS.
-------- for the GC.

I am discussing Kramer's apt points about the platonism logicians engage, 'with a straight face':

He writes in the uglily titled (by me), "Vehicle for meaning" thread:

"I think the x, X thing is a logical hammer in search of a nail."

This is excellent. Oddly, an important Gricean (of good sorts) is Stephen Neale, so I should be able to provide a bad pun on it.

Kramer continues:

"The phrase "an utterance" may refer either to a type of behavior or to an instantiation (token) of that behavior."

And then we need to qualify 'behaviour'. This Grice does in 1991 -- actually that's vintage 1975, "Method in philosophical psychology" but repr. --. For why should WE limit '... means ...' to matters of (never mind 'informational impart') but 'behaviour'. Recall Porter's early hit, "Let's MISbehave".

I think 'be-haviour' was the invention of, ... er, behaviourists. So we needn't go there. Matter of fact, I know a linguist who was going to be a philosopher, but felt disappointed by what this linguist refers to as 'behaviourist' tendencies in, of all people, people like Grice, I think -- and turned to mentalism instead -- of the MIT type! Massachusets Institute of Technology forya! --.

So we don't NEED 'behaviour'. Grice knew because he had to endure -- as the younger generation for 1946 -- the pedantic atic of Gilbert Ryle. You read about Ryle in modern histories of philosophy (all wrong, of course) and it's 'analytic behavourism'. A Brit won't have an American label -- like Watson's behaviourism -- used with a straight face: they need to anglo-qualify: 'analytic' behaviourism at most. But it's not even that! Grice is an 'intentionalist', not a behaviourist.

The debate was hot in MIT -- because Chomsky needed some press. So as any linguistic school boy (I'm not one) knows, Chomsky's claim to fame was to 'rebuff' (unsuccessfully) Skinner's latter-day Morrisianism. "Verbal Behaviour", Skinner had said (and filled unnecessarily 358 pages with it -- in a most unreadable prose, if you haven't -- I mean, how can you have read it if it IS unreadable?).

When Searle (who loved Grice, if perhaps unrequitedly, in parts) reprinted Grice 1967:lecture 6 in "Philosophy of Language", Searle made the occasion to give us his 'anti-Chomskyan tirade': For Grice is being described as a 'behaviourist' who is defining lingo (such a Cartesian thing) into 'procedures to' say this or that, or worse, to utter, on occasion, this or that token. Anti-Sknnerian won't have that! But they will!

----

So back to Kramer:

"The type and the token are two different logical devices, so a logician would want, I think, to have in his toolbox a way, should the need arise, to refer to them by different names, a way that would work for any type and its tokens."

Yes. I think a logician (I don't mean, necessarily the same one all the time) will have one tool too many in his toolbox, if you ask me -- and even if you don't -- ask me, that is.

Kramer is being too charitable when he goes on:

"The symbolization thus arises out of the logical distinction, and not out of any practical need to resolve it in actual utterance other than by context."

which is -- I admire Kramer's way to go so Gricean in most ways! -- precisely Grice's points when he is discussing:

"If I am helping the grass to grow" (i.e. 'assisting marijuana to mature'), I shall have no time for reading ('for I will be travelling to South America often', or something.

As it happens, Grice wants to say that 'to help the grass to grow' means (on occasion) 'to be dead'. In which case, it is a truism (cfr. "Do not be more informative than is required", or rather, "use your brains when you talk or converse") to add, "I shall have no time for reading". For how can a corpse find time for anything?

---

The other example (the other of Grice's 'points' referred above) is usually to be elucidated by 'context':

"Palmer gave Nicklaus quite a beating".

One would need to be 'very much in the know' to decipher THAT as an ironic way to state that what is meant is that Nicklaus, say, administered vigorous corporal punishemnt to Palmer' (WoW:120).

The problem here is 'metaphorical'.

To 'beat', indeed, cannot BUT mean, 'to administer vigorous corporal punishment', provided the thing does have a body (and qualifies as feeling 'corporal' things). The idea that it can be extended (via utterer's meaning, only) to mean, "vanquish with ease" is neither here nor there. I.e. nowhere (Vide Kilgariff, "I do not believe in word senses").

So Grice is putting forward a caveat for the OLP ordinary language philosopher -- NOT to fall in the traps of the Wittgenstein, who couldn't care less about matters of English usage (Recall that Cambridge (University) paid him -- i.e. endowed him with a salary) to torture enrolled students to bear with his "Germanisms" instead -- this was before the Second World War, of course).

Way of Words, indeed!

A word means what the utterer who utters it (the word or bunch of them) means by it (or them, if a bunch).

Dictionaries usually, and wrongly, define, 'types', not 'tokens'. When I wrote my PhD dissertation I mainly focused on one utterance:

'the cat is on the mat'.

I was then consulting very poor dictionaries -- such as the Longman Dictionary of Contemporary English -- edited by that Gricean of good sorts, Geoffrey Neal (or "Nail" if you must) Leech (born in Gloucester in the 1930s). I looked for 'cat' and 'mat'. It read, respectively, 'prostitute', and 'to be on the mat': to be punished.

So I provided the paraphrase:

The prostitute is being punished -- by the pimp, obviously.

Somehow it all turned very clear to me. Toulmin was using that example, but he (i.e Toulmin) was a Wittgensteinian.

In fact, when Toulmin died last year, I was re-reading his obituaries, and Wittgenstein typically resented him: "Toulmin is stealing my ideas", he (Wittgenstein) wrote. Of course he wasn't. Or he was, I forget now. Wittgenstein was confounded that he was not credited generously enough in Toulmin's claim to fame, "An examination of reason in ethics". "That's precisely what I was lecturing about at Cambridge, for all those years", Wittgenstein typically complained.

----

So we have

'cat' -- felis domesticus.

If the thing is used to mean 'prostitute', that's neither here nor there. "Cat" never, really, and strictly, means 'prostitute'. Grice is ALLOWING for some looser talkers to talk like that, i.e. loosely -- some can't help it, and you can't help THEM either!

"On the mat" may mean, 'to be punished' because that's where you end if someone gives administers on you vigorous corporal punishment, but surely if the cat REMAINS on the mat, and more importantly, PURRS on the mat, what has the pimp got to do with it?

But a linguist is sometimes confused. For they need, first, to study phonetics, and then phonology. -- Here they ARE dealing with x versus X in some odd ways -- a phoneme, for example, is defined as an 'abstract' type to which the phones (which are the only things that 'exist') belong. Odd! And untrue!

---- Having swallowed the x-X distinction at THAT level (a very UNempirical level, if you ask me), it's no wonder he'll go on to appeal to semes and sememes, and graphs and graphemes, and gnostons and gnostemes. (The thing exploded when Pike had the cheek to speak of 'etics' and 'emics'!

So we distinguish between 'cat' and the cats, and The Cat (the idea of a cat), the meaning of 'cat'. And similarly for 'mat' and 'on the mat' and "On the Mat". On-the-matness, etc.

----

But at the level of the behaviour, in this realm (what Austin calls 'phatic' and 'phemic', never 'rhetic' which is the 'that'-clause of the specification in Grice) we only have the sounds:

/k/ and /ae/ and /t/ -- first articulation. Yielding, 'cat'.

And "cat" means 'felis domesticus'.

It may be argued that a tiger is a cat -- but J. L. Borges taught me it's not ("Dreamtigers" -- similarly, while some American 'cats' are called 'tigers' they are not: the only real tiger is the royal Asiatic one of the Brit explorers -- as per Kipling's stories that fascinated him, i.e. Borges).

----

So, who cares for what a logician needs? The only thing that does matter in predicate calculus. So we DO need a symbol for a predicate here: "C" for 'cat' and "M" for mat. And we may need a non-primitive iota-operator, for 'the' (as in 'the cat', 'the mat'):

ix. Cx & Mx.

There is a cat such that she is on the mat.

No need to go onto an otiose higher order where we need to specify 'types' for this.

Now, it may be argued that a dictionary is a good thing. I am always reminded of Grice on this:

I don't give a hoot what the dictionary says

-- cited by Chapman, archival material. Can she edit stuff like that?

And he was right. Two hoots at least.

'cat': felis domesticus.

You find in the dico (or ditionary). Grice was irriated that, for him, 'the' dictionary, was the 21-volume edition of OED2, and surely it would be stupid to think that 'the' dictionary says this or that. A dictionary does NOT say: it's the otiose staff that WORKS for the Corporation (usually Oxford University Press -- one of the richest on earth) who say this (or that) -- usually uncreditedly so.

The idea would run:

"We look up a word in the dictionary to find out the TYPE of meaning: and then we work out of this TYPE of meaning (e.g. "cat", 'felis domesticus') to infer that, on this particular occasion, what the utterer meant was, rather, 'a whore'. How come, or cum, if you must (Dutch English, "Wie Kommst?")

-- "Well, a whore is like a cat; or vice versa, if you look at her". But she is not!

This is NOT your common or garden transcategorial mistake of your common or garden metaphor ("You are the cream in my coffee"). For if 'you' are 'Cream', and cream does not have ears, why am I telling cream tht you are the cream in my coffee? The idea is that you are LIKE the cream in my coffee. So the idea is perhaps that a prostitute is LIKE a cat. Cat people, with Natassia Kinski?

---- A (human) prostitute and a cat belong in the same order: mammals -- unlike 'you' and 'cream'. (Why, cream does not belong in an order at all.

So the idea is that the lexicographer (who is, we allege, into 'utterance types' and thier meanings or what they mean) will provide a statistical analysis (this is what Kilgariff, who worked for Longman -- implicating he left - objects to, or rather what he thinks the 'theoretical lexicographer' does not take seriously enough into account).

And out of that 'sampling' he concludes: every token of the string of phones /kaet/ is associated with the idea (in the brain of the utterer, if he has one -- cfr. "Tin Man" in Wizard of Oz) of 'felis domesticus' -- cfr. Saussure on 'arbor' and "tree" in Cours de linguistique generale.

But the probelm here is the rabbit.

Quine found out that in some Amerindian language, 'gavagai'. I.e. that gavagai can mean this or that (rabbit). But it is IMPOSSIBLE, he claimed, to get to KNOW what a word means (i.e. what 'object' the word refers to). He boringly argued for this in some 400 pages of his book, unimaginatively entitled, "Word and object" which is sadly described in some fora as 'American philosophy at her best'. The vengeance came from ... Grice. When the editors gathered (Davidson and Hintikka) to provide a collective response to Quine they came up with "Words and ObjectIONS" to which Grice contributed (with his vacuous names).

---

Kramer concludes his note with:

"Whether such a need will arise remains to be seen, but that's not the logician's concern. The hammer is in the box, because logic, not language, requires it to be there."

Yes, and one wonders. When Grice got (unintentionally) his thing typed and distributed (against his will), he found himself being approached (friendlily he (Davidson) later said) by Davidson and Harman. His "Logic and Conversation" (i.e. Grice's) eventually got published (by request) in Davidson/Harman, "Logic and Grammar" -- so he knew (Grice) what he was talking about.

He wants to SIMPLIFY logic -- by appealing to 'grammatical' misconceptions that he needs not make. The misconception of 'sense' or senseS in plural for example. Just because some TOKENS of 'kaet' are uttered to 'mean' 'prostitute', are we to grant that 'prostitute' is one of the senses of 'cat' (other than the only one it has, to wit -- and cats are witty critters -- 'felis domesticus'? No way!

He also wants to simplify logic by sticking to 'entailment' and DISALLOW anything remotely RESEMBLING an implicature into the realm of logical discourse. We do NOT want 'pragmatic' inferences which are defeasible, indeterminate (alla "Gavagai") and stuff. We want our logic to be clean and tidy. An entailment IS necessary because it is the good ole "meaning postulate" of good ole Carnap. But who cares if 'seem' is USED to mean something which can be problematically used to elucidate the already problematic philosophical notion of 'sense datum'?

The problem for Grice is NOT that philosophers, some of them, are loose talkers. The problem is that a loose-philosopher reader needs NOT be loose! Or something