Grice in defense of a dogma: analyticity and substantive inquiry

Speranza

We are discussing Grice's seminal defense of a 'dogma' (or "under-dogma", as Richard Grandy prefers) back in the day (actually back in the year of "grice": 1956). (True: the essay is by Grice AND Strawson, or Strawson and Grice, but I'm disimplicating here), vis-à-vis L. P. Halpin's essay, "Analyticity and substantive inquiry", available at:

http://www.amazon.com/gp/aw/d/1478328738

As Halpin comments,

"Frege built endorsement into the syntax, by way of the judgement stroke." -- and by the way, check Halpin's page for a nice photograph of Frege at:

https://sites.google.com/site/lucashalpin/

-- and cfr. L. R. Horn's F-Implicatures.
Halpin goes on:

"I [on the other hand, i.e. as differently from Frege] build endorsement and rejection into the syntax by way of 'T' and 'F', which are semantically insignificant bits of the object language syntax with novel valence (They attach to sentences in the object language)."

This is v. good.

In fact, I prefer to use 0 and 1 rather than "T" and "F" (if we are not going to attach any 'Griceian' "mean"(ing) to them. In this way, we can more or less (insignificantly) co-relate:

1  with CERTAINTY

and

0 with UNcertainty -- as in Grice's seminal lecture, "Intention and UNcertainty".

In between we have, since he recently passed away, Urmson's scales. In "Parentheetical Verbs", Urmson discusses the parenthetical uses of

"I know"

versus

"I believe"

and

"I disbelieve"

and

"I ignore".

----- "know" correlates with "certainty", i.e., 1. "ignore", with 0. But there's a gamut, indeed infinite, qua continuum, in between -- and we can refer to this continuum by way of 0.5, say, or 0.75, or 0.25 -- Ian Cargan taught me how to handle these subdivisions.

Hamblin goes on:

"[In the essay "Analyticity and substantive inquiry"] 'True' and 'False' are treated as expressions of the meta-language that indicate correspondence between the intrepretation/content of a sentence and the world."

Grice would have loved that, and rename it "Bootstrap!". He thought that indeed one's metalanguage can be as rich as one wants -- the only problem "for the morrow" is when you 'try to pull up yourself by your own bootstraps" ("Reply to Richards", originally, "Prejudices and Predilections; which become the Life and Opinions of Paul Grice", by Paul Grice). Grice respected the object-language/meta-language distinction which he inherited from Russell.

Halpin goes on:

"Where S is a sentence, S-T and S-F receive numerically distinct interpretations/contents. A statement is analytic iff S-T is interpreted and S-F is un-interpreted."

This is good. R. B. Jones and I have elsewhere elaborated on this -- i.e. on how to deal with 'interpretation'. We have played with what Grice calls a "System" (his System Q, in honour of Quine). Jones has provided symbolisations of this, and gone to mention variants of Grice's System G: -- Myro's System G (in honour of Grice) and System C (among others) -- "C" for Carnap.

The root for this, in Ancient Griceian Philosophy, lies in Grice's "Vacuous Names", which should be compiled in toto in a book (or two) -- the latter part of it is repr. in a MIT book collection on Definite descriptions.

Halpin goes on:

"A speaker who rejected an analytic statement, i.e., assigned it 'F', would thereby fail to express anything. A speaker who endorsed it, i.e., assigned it T, would succeed in expressing something. (This aspect strikes me as very Grice/Strawson)."

Yes. We could revise all that.

Recall that we can do with just one subject-item:

"My neighbour's three-year old son"

either he "understands Russell's theory of types" (synthetically false)

or

"is an adult" (analytically false).

Pre-judging, as it were, what would become Grice's full-blown theory of conversational sequences and implicatures, they are wanting to say that the conversational countermove to "SF" is

"Well, I don't BELIEVE it".

--- while a conversational countermove to the more interesting "AF" is:

"I don't UNDERSTAND it".

-- where "it" is possibly apocryphal!

--

Halpin goes on:

"Thus, if S is analytic in L, the L-speaker is forced to endorse it. If she rejects it, she is reinterpreting the language, and hence, not speaking L."

Very true.

I don't know what the good a festschrift is if you're going to republish your contribution elsewhere, but this is what DAVIDSON did in his beautiful, "A nice derangement of epitaphs". This was meant as a contribution to a festschrift, but I believe he did go and publish it elsewhere. In any case, in that beautiful essay, Davidson (who loved Grice) sheds doubts on the expressions,

"to know a language"

---- He would go on to deny the idiom, "language", and Grice would not feel any sort of despair about that. Note the troubles Grice goes to define 'utterance occasion meaning' to convoluted abstract notions like "meaning in a language" (for a population P) and so on.

---- Since Grice's earliest published essay was on "Personal identity" (Mind, 1941), where "I" is defined in terms of "my memories", it may do to check this with the scenario where, say, 'three-year old' can CHANGE a meaning.

Halpin goes on:

"Thus, S is indefeasibile in L, in the sense that the rejection of S immediately takes one out of L and into another language.(This aspect also strikes me as very Grice/Strawson.)".

Yes. Grice was careful about things, though. At times, he can be TOO pedantic. As when in "Conceptual analysis and philosophy" (or something like that!) in WoW he considers the goal of philosophical analysis: "to elucidate one's problems with concept C". Of course "it need not be MY problem; I can help elucidate OTHER people's problems!" (rephrased). But yes, in Grice (and this is Grice and Strawson, though), it's all about the IDIO-lect -- or as I prefer, since I find the '-lect' TOO abstract, 'idio-syncratic'.

It all starts with what an expression means, when utterer U utters it on occasion O.

Halpin goes on:

"Supposing S is analytic, S-T recieves an interpretation/content. Thus, S may well be semantically false if the content/interpretation fails to correspond to reality. In this case, revision of a theory involving an analytic statement is forced. However, because one cannot assign F to S (reject it) without reinterpreting the language, she is forced to reject the language itself, rather than merely adjusting the T/F (endorsement) assignment."

Yes, that is very good. And hence the title of the 'substantive inquiry'. I would assume that the first-order, common-or-garden inquiries (of perhaps the NON-philosophical type) are of this kind: one _tests_ (as Popper would have it) where 'p' corresponds to the fact that p.

"My neighbour's three-year old son understands Russell's theory of types".

TRUE?
Well, let's go on to _test_ the claim.

-- But beyond these 'inquiry', there is the ESCHATOLOGICAL inquiry. Here we are to make sense of

"My neighbour's three-year old son is an adult".
FALSE. But why?

Grice and Strawson reject a 'metaphorical' interpretant ("he looks mature for his age", or "he is 'adult' in the etymological 'use' of "adult"", say).

And the falsity lies in the categorical mistake, as it were, involved. But there are subtle ways of dealing with 'categorial epithets' and 'transcategorial epithets' that may not be metaphors or tropes in the obvious way. Grice explores this in his "Eschatology" essay in WoW. For example, consider the mediaeval philosophers' ways of conceptualising "God". One read 'mediaeval theology' and one admires the ways these theologians tried to 'cross' categorial barriers as it were. Even if they failed! (Hence the 'negative' theologies where the interpretation of 'negation' is tricky: "When we say what God is NOT, what are we meaning 'not'? -- hypernegation?).

Halpin goes on:

"This is a brief sketch. More detial is presented in the book, which I am shamelessly self-promoting.:

Beautiful. And thanks.

Halpin goes on:

"Quine's objections: The book begins with what I take to be the primary objection to analyticity in Two Dogmas, the rejection of the distinction between the linguistic and factual components in the truth of sentences. In response, I explain how to accomodate holism, the Duhem/Quine thesis, etc, using the materials above."

Very good. Quine, for what I remember, tried to address Grice's and Strawson's point in "Words and Objects". They remained FRIENDS! Of course, Grice's "Vacuous Names" was meant for Quine's festschrift (edited by Hintikka/Davidson; Strawson also collaborated). And in his later, "Part of my life" (or something), Quine goes on to provide an affectionate (if partial) recollection of Grice. In his "Prejudices and predilections" Grice mentions only TWO mentors: Quine and Chomsky ("with whom I never agreed on anything", he hastens to disimplicate -- rephrased).

Halpin goes on:

"The book is short, 25,000 words, constituting a mere sketch of a meta-theory that treats analyticity and logicality descriptively, such that neither entials semantic truth."

Good.

Halpin goes on:

"With respect to exactly how Grice/Strawson the view is, I'm not sure. Some similarities are obvious enough, but I'm not familiar enough with all the ins and outs of their views to comment on that with any confidence."

No worry!

Even Jonathan Bennett (whom I love) got it all wrong! When he wrote his "Linguistic Behaviour" he explained that Grice's "Meaning" (1957) was the obvious response or sequel to "In defense of a dogma": "In "In defense of a dogma", Grice supports analyticity and in "Meaning" he offers a way to handle semantics without recourse to a vicious notion" (rephrased). The only problem is that "Meaning" was actually written in 1948 (and only PUBLISHED, when retyped by Strawson and his wife (Strawson's) and sent, without Grice knowing, to the editors of the "Philosophical Review").

---- I wouldn't know ALL the details of the "Defense of a dogma" thing. I'm not sure what Quine was lecturing back in Oxford in the day. But Strawson will go on to quote from stuff like Quine's "Methods of Logic" and not too sympathetically. For example, Quine -- in Strawson's and Grice's reading -- is never too careful as to distinctions regarding what a connective, for example, says and what it entails, or what it means, and what it is used for (Strawson, "Introduction" to "Philosophical Logic", 1968, Oxford Readings in Philosophy, which includes Grice's "Meaning").

---- And in any case, there is quite an evolution (if that's the word) in Grice's thinking about these questions anyway (or 'anyways', if you are a pluralist).

Cheers, and thanks again for sharing.