[hist-analytic] Bootstrap

Roger Bishop Jones rbj at rbjones.com
Mon Jan 31 15:24:06 EST 2011


On Monday 31 Jan 2011 15:45, jlsperanza at aol.com wrote:

> The example of "analytic" in Grice & Strawson, in that
> paper is:
> 
> "My neighbour´s three-year old child is an adult"

Analytically false of course, if you allow that usage, more 
usually, self contradictory (if you follow Carnap in a broad 
notion of logical truth).
 
> as opposed to "and he understands Russell´s theory of
> types".
> 
> G & S argue that there is a meaning postulate here:
> 
>       (x)3x --> -Ax
> 
> If x is a three-year old thing, x is NOT an adult.

I'm not fond of meaning postulates, they are one of the 
conspicuous features of an account of semantics which helps 
to make the notion of analyticity seem more arbitrary than 
it need appear.
Carnap, mind you, didn't think you could do it for natural 
languages.
 
> In any case, G & S seem to be arguing that our "actual
> conceptual scheme" has items like "three-year old",
> "adult", "understands the theory of types", etc., and
> that there ARE behavioural output in terms of the
> creatures posessing those languages. So there IS a way
> out.
> 
> When faced with an analytically false statement, Stawson
> goes:
> 
>        ------ "I don´t understand that!"

Which, I would say, is because of an implicature, without 
which he would simply say "nonesense", meaning by that 
"patent falsehood".
There is of course no difficulty in understanding what it 
means, the problem is in understanding why anyone would 
assert something so obviously false.

> whereas when faced with a synthetically false statement,
> he rather goes:
> 
>       ------- "I don´t believe that!"

There is perhaps just a slightly better chance of grasping 
why someone would present his child as a genius.

> Grice and possibly Strawson held two projects here. One
> is more of an eschatological (or transcendental) one: to
> consider the profitability of a creature possessing the
> notion of "analytic truth".

Of course "analytic truth" is a technical term which most 
people don't have.  But they do have closely related things, 
in particular an understanding of the meanings sentences in 
the English language (or some other natural language).
In reflection on my earlier exchange with Danny I came to 
doubt that in ordinary English there is any conception of 
logical or metaphysical necessity, and I would say of 
course, even less of analytic truth since that particular 
use of "analytic" is purely philosophical.

My favourite example of something close enough to 
analyticity to make that notion meaningful even for natural 
language is indirect reported speech.
If I say something, and you subsequently report that I said 
that ..., then you are offering a statement which is not 
required to be exactly what I said, but must mean the same, 
and if your report is accurate then the claim that tne two 
have the same meaning will be analytically true.

> Another seems to be more
> local: to review for this or that concept if it can
> yield some analytic statement ("three-year old"). Note
> that if you are a cat, and 3 year old, you are possibly
> an adult. So one has to be careful here. There seems to
> be a further predicate in need here: HUMAN.

that one is entailed by "neighbours child"'.

> The idea developed in 1960s Oxford into the idea of a
> language for a Population (P) for which some ideas merge
> in analytic statements. Bennett in Linguistic Behaviour
> is an example. Also my favourite: Loar´s DPhil thesis on
> "Sentence meaning", supervised by Warnock, and Peacocke
> in the Evans/McDowell collection, Meaning and Truth. It
> would seem that these authors are indeed claiming that
> there is a "naturalistic" (in terms of behavioural
> output -- "I don´t understand what you mean") which
> determines which truths are held to be analytic for a
> population, or an individual (vide Grice, on idiolect
> meaning, idiosyncratic, in WoW:6).

So I would be suspicious about how specific the indicators 
are.  How can you tell the analytic falsehoods from other 
kinds of incomprehensibility?

> Grice in any case, as his "Retrospective Epilogue"
> (lastest section) shows, was leaning towards a
> "pragmaticist" (in terms of utility) of the
> justification of "analyticity". In this conception, he
> would argue that it is not necessary that the statements
> of descriptive metaphysics ARE analytic. Only that are
> "deemed" as valid in terms of their justification power,
> or so. This pragmatism may have more than a point of
> contact with the leader at Carnap Corner.

It seems to me that Strawson and Grice did a good job of one 
half of a response to Quine.
They are really saying that Quine's arguments are irrelevant 
because in fact we can tell the difference ("draw the line") 
often enough for the concept to be useful.
The other half which they completely dip out on is a 
critique of the arguments which Quine deploys against the 
possibility of defining semantics (and hence analyticity).

Did anyone do that?

Roger Jones














More information about the hist-analytic mailing list