[hist-analytic] Carnap and Grice on (alleged) "necessary truth"

Jlsperanza at aol.com Jlsperanza at aol.com
Tue Mar 2 18:24:17 EST 2010

In a message dated 3/2/2010 rbj at rbjones.com writes:
"Well this sounds interesting stuff, but it veers at the end too close  to 
treating the analyticity as being something we decide upon by fiat, which is 
 how Quine would have us think of it, but not how Carnap does and even less 
how  he should! (You have to decide meaning and accept the extension of 
analyticity  which flows from it, and its best to present the matter so as to 
make that  clear, which Carnap often does not 
Necessary truths? Or rather at most 'empirical generalisations over  
functional states' that a philosopher may reflect on? Carnap and Grice on the  
Yerkes-Dodson Law. So, this is Carnap and Grice on empirical psychology versus  
rational psychology, say, and the role of 'necessary truth' or analytic 
truth,  in things involving 'beliefs', etc. Re the 'form' of a psychological 
law, I  think I was referring to things like this by Grice in "Conception of 
Value" Gr91  -- google.books, "But whether or not the Yerkes- Dodson Law is 
properly  so-called, it is certainly 'law-allusive'; and the feature of being 
law-allusive  is one which I would ..." This goes back to p. 124 of that 
book, where Grice  wonders about such 'laws', i.e. postulates from a Theory -- 
Symbolise it by  Theta. Some relevant quotes by Grice:

"Are we go give them contingent or non-contingent status?" Consider "He who 
 wills the end, wills the means" Is this NOT REFUTABLE? So it seems. We 
observe  Toby, he wills the end, but does not will the means. Therefore, we 
conclude,  Grice says, "Toby did not REALLY will the end". This, Grice says, 
means to treat  the law as a "necessary truth"  -- his collocation. But if 
it's "a  psychological law" is SHOULD be 'contingent'. This is what he 
formulates as a  problem. In general terms, "How (without a blanket rejection of the 
 analytic/synthetic distinction) we [are] to account for, and if possible,  
resolve, the ambivalence concerning their status with which we seem to look 
upon  certain principles involving psychological concepts."
      There are major issues here: the idea is  indeed that '... 
judges...', "... assserts...", "...believes..." (to use the  ones Carnap was more 
involved with) are _psychological THEORETICAL concepts".  Thus their meaning is 
given by Theta, a theory. A theory comprises only  'contingent truths'. So 
what role is A-truth supposed to play there. If we  formalise the language of 
psychology (empirical pyschology) no such truths seem  required. It's only 
when it comes to what Grice calls "rational psychology" (or  'philosophical 
psychology' simpliciter). Because as such, it's a game (applied  game) to 
see generalisations philosophers want to consider regarding desiderata,  
self-entrenched, if you wish, as to want count as having a 'rational belief'.  
Since Carnap did expand on 'rational choice' theory in his probability work on 
 subjective probability that occupied his mind, they say, for most of the 
last  decade or so of his life, if not more -- to the detriment of other 
topics that  Carnpians and neo-Carnapians would have him rather treat) he may 
have said  something on this, too, I hope.  I woud think the compromise 
Carnap/Grice  at this point would be to concede Carnap that NOTHING in empirical 
psychology is  a necessary truth and that something MAY be a necessary truth 
(if pressed, but  why would WE be pressed, since it's all so 'caeteris 
paribus' anyway) in  _rational_ psychology. In fact, my leader in this is B. Loar 
in his "Meaning and  Mind" where he has the lovely cheek to say that a 
Gricean maxim such as "Say the  truth" boils down to merely (if you allow me the 
split) an "empirical  generalisation over functional states"! (and right he 
is -- the point is that a  philosopher -- or some of them, call them 
"Grice" in his best moments (?) may  want to say it is _more_ than that? (The 
problem is so constructivist it hurts,  but in a nice way!)
Grice would, I would think, be happy with 'necessary truth' being 'alleged' 
 in a case like "he who wills the end, wills the end'. It seems NOT a 
necessary  truth. To add, 'necessary' brings a 'conceptual' ring to it that makes 
it immune  to refutation. We don't want that. Rather, we want to elaborate 
on 'alleged'. If  that's read as "rationally _deemed_" then it is surely 
compatible with the thing  _being_ an "empirical generalisation over functional 
states" which "passes  rational muster" (to use G/W's phrase in PGRICE). 
J. L. Speranza


More information about the hist-analytic mailing list