Wednesday, April 25, 2012

Defining truth in terms of human agreement

To: Hamad A.
From: Geoffrey Klempner
Subject: Defining truth in terms of human agreement
Date: 18 July 2007 11:18

Dear Hamad,

Thank you for your email of 7 July, with your first essay for the Metaphysics program, in response to the question: ''A statement is true if, and only if, the majority of persons of sound judgement would assent to its truth.' - How might tone criticise this definition of truth using the reality principle?'

This is a very good essay which states clearly and concisely all that is required to refute the definition of truth offered. However, you don't stop there but also offer an explanation of why one would be tempted to offer such a definition in the first place - as a misunderstanding of the principle that 'truth belongs to the public sphere' - as well as offering some insight into the nature of the reality principle itself.

I am glad that you found this topic difficult, because it should be. You could tried to give a 'knock-down' argument but that would have been unilluminating, as well as question-begging.

What I mean by a knock-down argument would be something like this. According to the reality principle, 'a thought, to have a legitimate claim to truth, must be prone to the possibility of being false'. However, if truth is defined as (is equivalent to) what the majority (etc. etc.) would believe, then provided that the majority believe that P then P cannot be false.

Is there any way to resist this refutation? One thing that a defender of the majority view could say is that it does not, in fact, preclude the possibility of a statement being false because it is always possible that the majority do (or would) not believe the statement. I believe that P. My question whether P is true is equivalent to the question whether the majority (I'll leave out the 'etc. etc.') believe that P. Well, maybe they don't. In believing that P I am claiming that the majority would agree with me, but it turns out that my claim is incorrect.

Your thought is that what is missing, in the majority definition, is recognition of 'a non-personal standard'. It is in relation to such a standard that the beliefs of one, or many, believers are evaluated. The intuition here (which you don't go so far as to state explicitly but which is implicit in your argument) is that, given any P, it is always possible for the majority to be wrong, i.e. for P to be false even though the majority believe that P is true.

Is THAT true?

We are talking about 'any' P, so that includes the case where P is a disjunction of a large number of beliefs which are held in common (such as, human beings live on the surface of the earth, 2+2=4, cows do not hatch from eggs, there is a country called the USA etc. etc.). Logically, we have to allow he possibility that all these beliefs are false - which requires giving a minimal degree of credibility to the Matrix scenario.

The majority in the Matrix believe that P, but in fact they are wrong. In this possible world, P is false, by virtue of an 'external standard'. We have to allow that the Matrix scenario might be true, in some possible world, in order to explain how the majority could be wrong about, e.g. cows, or the USA.

However, we are not done yet. The definition says, 'the majority... would assent'. But no upper bound has been set to defining the number of persons whose 'votes' are to be counted. In 500 BC the majority would have assented to the view that the earth is flat. But they do not constitute 'the' majority. If we take the same vote today, (I guess) the majority would assent to the view that the earth is round. But, again, that leaves out of consideration people who will exist in the future who might conceivably discover that we have been living in the Matrix all along.

In other words, a defender of the definition might say that the intention is not to draw an upper bound on the number of persons whose judgement we are considering, not just in this century or the next but now and forever more.

Would one still want to say that P is false (according to some external standard) even though now and forever more the majority will never cease to believe that P? I think you have to say this, to remain consistent with the position you have laid out. But its truth is not so intuitively obvious.

One other aspect of the question which we have not looked at is the question whether there can be any 'definition' of truth other than the truistic, 'P' is true if and only if P, or, equally truistic, 'P' is true if and only if it is a fact that P. For example, if the 'external standard' is the book of the Recording Angel, then we have to grant the Recording Angel infallibility or else allow that a proposition inscribed in that book can still be false.

When we come to the debate between the realist and the anti-realist about truth, you will see how hard it is to state what the realist 'means' in a non-question begging way.

All the best,

Geoffrey