[hist-analytic] A posteriori knowledge of necessary truths

Roger Bishop Jones rbj at rbjones.com
Sat Dec 18 11:48:13 EST 2010

Here are some observations on your recent message on the 
above topic.

FIrst, I doubt the soundness of your method.
In order to reason soundly you must do so in a context which 
is sufficiently well understood for its principles and rules 
to be definite and their consistency strength establishable.

I know no way of doing this other than though the 
construction of one or more formal models in a well 
established logical system.

By this means you would ensure the soundness of your 
reasoning, and can then debate how well your models 
correspond to ordinary or philosophical usage of natural 
languages on the relevant topics.

Second (if I might now disregard my own advice are reason in 
an informal context), when making judgements about whether 
some knowledge is a priori or a posteriori it is essential 
to differentiate between information obtained by experience 
according to whether it contributes to establishing the 
meaning or the fact.
Our knowledge of the meaning of a natural language is of 
course a posteriori, and if the language contains rigid 
designators then (in any model I would be likely to work 
with) the designatum of a rigid designator will be fixed by 
its meaning (if it is not so fixed then it cannot be rigid).
In that case the use of experience in discovering the 
designata in an equation of rigid designators will be a case 
of knowing the language rather than of knowing the fact.
The fact will be an instance of the reflexivity of identity 
and will be knowable a priori.

I would also myself not accept your opinion that the 
introduction of an existential quantifier makes a de re 
proposition from one which is not de re.
This sounds like the supposition that all existence is 
contingent, which for anyone sympathetic to Carnap's stance 
on abstract ontology will not be the case.

Sorry not to have undertaken a more thorough critique, but
my dominant sentiment is my first, that formality is 
essential in dealing with logical problems of even this 
level of complexity.

Roger Jones

More information about the hist-analytic mailing list