To: Jason I.
From: Geoffrey Klempner
Subject: Truth, realism and anti-realism
Date: 13 November 2003 12:32

Dear Jason,

Thank you for your e-mail of 5 November, with your first essay for the Metaphysics program, in response to the question. 'Explore one example, base don any reading you have done, of a problem that you see as raising a question about the nature of truth, or a question about the nature of existence.'

I like the way you start off by developing a plausible argument for preferring anti-realism over realism, and then give an equally plausible argument for preferring realism over anti-realism.

Admittedly, the first argument depends heavily on 'intuitions'. 'A city will never be built here' presents a particularly strong challenge to realism because of all the areas of discourse, we have least resistance to accepting anti-realism about the future. This was in fact Aristotle's view in 'De Interpretatione' (reprinted in Gale Ed. 'Philosophy of Time' Macmillan). In response to the question, 'Will there or will there not be a sea battle tomorrow?' Aristotle rejects the idea that it is in fact true - a truth which we cannot know before tomorrow - that there is a sea battle, or, alternatively, that there is not a sea battle.

So this would be one case of 'an example...of a problem that [raises] a question about the nature of truth'.

When we consider the progress of science, on the other hand, our intuitions go the other way. Surely, the heliocentric theory propounded by Copernicus was true, even during the times when under the Church's influence it was believed to be false. Interestingly, in his more recent writings on anti-realism Dummett has not pursued the 'relativist' implications of the anti-realist view. (I did see a TV interview once where Dummett speculated that the impossibility of accepting 'relativism' about truth might be used as the premiss in a Berkeley-style proof of the existence of God.)

But *is* Dummett's anti-realist committed to saying the heliocentric theory 'was' false and later 'became' true?

I would question your statement that 'the anti-realist takes the view that [a] statement can only be true if we have evidence that such a state of affairs actually exists.' To my knowledge, this was never Dummett's position. I don't at the present moment have evidence that the post has arrived. It is possible that the sound of the letter box clattering was drowned out by my printer. So, on the stated view, it cannot be true that the post has arrived! Surely, the statement about the post belongs to the class of statements for which there is, in Dummett's words, 'an effective decision procedure' (i.e. go downstairs and look). In simpler terms, the class of truths, for the anti-realist is larger than the class of propositions known to be true.

If the only true propositions were those which were known to be true, not only would there be a problem about the progress of science, we couldn't even make sense of the idea of 'discovering' a truth.

To see this from another angle, the process of logical inference involves the idea of the transmission of truth from premisses to conclusion. Dummett accepts that the anti-realist is capable of making logical inferences. In his essay 'Truth', Dummett states, as a test for whether we view a proposition as capable of having a truth value, whether we would be prepared to have that proposition as the antecedent of an indicative conditional. So, I can meaningfully say, 'If the post has arrived then my parcel didn't come' (because if it had come the postman would have rung the bell). But on Dummett's view, I can't meaningfully say, 'If a city will never be built here then the beauty of this spot will always be preserved.'

There is a dialectical point to make about the statement, 'The heliocentric theory was false but now is true'. 'The heliocentric theory was false' logically implies that the heliocentric theory is false. This is a simple fact about the logic of attributions of truth and falsity. (There is a fuller discussion of 'truth value linkage' in Dummett's paper, 'On the Reality of the Past'.) So the attempt to *state* relativism involves a straightforward logical contradiction. However, the second point is that we can only speak for ourselves, we cannot speak for people who lived in the time of Copernicus. We can say, 'The statement, 'The heliocentric theory is false' met the criteria for truth that existed at the time', but that is not the same as stating that it 'was true then'.

The difference in viewpoint comes into sharp relief when one considers 1984-type cases, where someone destroys evidence (as in the story of Simon and Natalie). Here again, the crucial point is that when speaking for others, we can only say what they believe on the basis of such and such evidence or 'criteria'; we cannot meaningfully state that such-and-such 'was true for them', except as an ellipsis for, 'they believed that such-and-such'.

All the best,

Geoffrey