[hist-analytic] Aune and the a/s distinction

Bruce Aune aune at philos.umass.edu
Sun Feb 8 17:03:57 EST 2009

These remarks are in response to Steve’s email, “Aune and the a/s  

First, Steve quotes my remark, “"Quine thought that, for philosophical  
purpose, a conception of analytic truth as truth by virtue of meaning  
was too vague to be taken seriously." If we wish to speak loosely,  
“truth by virtue of meaning” may be all right as a way of describing  
an analytic truth, but it doesn’t give us the information we need to  
decide whether a statement a given statement is analytic.  How could  
considerations limited to meaning show us whether a given statement is  
true?  Kant’s description of a true analytic judgment offered some  
help in the matter, but it applied only to a limited class of judgments 
—namely, universally affirmative judgments of a subject/predicate  
form.  It provided help because if a predicate “is an A that is a B”  
is applicable to any individual i (so that i is an A that is a B) then  
the predicate “is a B” is also applicable to i (so that i is a B).   
Thus, with just a little logical tinkering, we can conclude that “All  
i’s that are As that are a B are I’s that are a B” or, more simply,  
“All ABs are Bs.”  But not all statements are of the “All Bs are B”  
form.  What about them?

As I pointed out, Frege in his Foundations of Arithmetic tried to  
improve on Kant’s definition of an analytic truth; he tried to bring  
it up to date in a way we are all familiar with.  But Frege did  
nothing to show that why logical truths should be considered  
analytically true, nor did he, as Quine objected, give us a means of  
recognizing synonymous expressions, which we must have if we can  
successfully apply Frege’s definition.  If Quine was right in  
recognizing Frege’s means of drawing the distinction, How do we  
proceed?  Well, I have offered a means of drawing the distinction, one  
that follows Carnap’s lead, in my new book, so anyone who is  
interested in what I think need only read my chapter 3.  I could  
accept the vague characterization “true by virtue of meaning” for very  
rough and ready purposes, but this characterization cannot be deemed  
helpful in an informed philosophical discussion.

I said that Carnap said that no useful a/s distinctioncan be drawn for  
the sentences of a natural language.  He held this, as I explained,  
because he thought the sentences of a natural language are  
insufficiently determinate in meaning to admit of such a distinction.  
He did think, though, that we can draw such a distinction if we make  
our meaning clear and determinate for purposes of discussion. I  
explain all of this in my chapter 3, though I add qualifications in my  
appendix Best, Bruce
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://rbjones.com/pipermail/hist-analytic_rbjones.com/attachments/20090208/2c06dac6/attachment.html>

More information about the hist-analytic mailing list