To: Julian P.
From: Geoffrey Klempner
Subject: Two essays on Truth
Date: 30 January 2004 11:03

Dear Julian,

Thank you for your e-mail of 23 January with your (corrected) University of London BA essay, 'What is Falsity and Why Does it Exist?', and your e-mail of 29 January with your essay, 'Does it Mean Anything to Say that a Statement is True?'

I have just read the first essay, and am responding to it before I look at the second essay, while the thoughts which it has provoked are fresh in my mind.

Essay on Falsity

This is an excellent piece of work. Reading this I had a powerful sense of deja vu. I made a similar 'discovery' about falsity while I was writing my B.Phil (later to become my D.Phil) thesis 'The Metaphysics of Meaning'.

I began my thesis with the question, 'How is false belief possible?' There are two questions we have to answer here: how can there be such a thing as a false proposition? and how can there be such a thing as a false judgement?

As Wittgenstein argues in the Tractatus, the primitive T/F polarity is the essence of a proposition. The capacity for truth-or-falsity is what distinguishes a proposition from a name which either refers to an object or is meaningless. In the first chapter of my thesis I imagined two omniscient Gods conversing with one another. Just like the prisoners who have heard every joke many times over, the Gods have a number for each proposition and just say, 'number 349210!... wow!... number 2091!... yeah!...'

As you remark (following Wittgenstein), what a proposition has over a name is the possibility of making a new thought out of elements which we are familiar with. So the possibility of conveying new thoughts and the possibility of a false proposition are intimately connected.

But there is a further question - illustrated by your caterpillar example. There cannot be such a thing as a belief/ judgement which is incapable of falsity - contrary to the assumption made by sense datum theorists (including Russell). Of course, caterpillars do not make judgements. The sense datum theorist's thought is that, if my mind is in direct contact with its object (e.g. my private impression of what the colour of this tomato looks like to me at this moment) then I can't be wrong. My judgement must be true. But this is like 'shooting' an arrow at a target which is already attached to it. Because the judgement can't be false, it can't be true either. Objects in the genuine sense - targets to shoot our judgements at - only appear when there is a possibility of falsity as well as truth.

Your observation that false beliefs, but not true beliefs are revealed in behaviour to a 'third person' observer raises the question whether it would not be logically possible that no-one has ever had a false belief. Suppose a molly-coddling deity decided to set things up so that things never went wrong in the belief-forming process. Would that mean that we could never understand one another's behaviour? On the face of it, people would still do different things because they have different desires. The only use for a concept of belief would be to contrast knowledge with ignorance.

Or perhaps the problem of ignorance could be fixed up too - with each person being given a mental almanac answering any question about the world they might ever think of asking.

However, the idea that 'false beliefs show up in behaviour' is important precisely because it emphasizes the 'third person' aspect. This is the upshot of Wittgenstein's private language argument in the Philosophical Investigations. Kant, in the 'Refutation of Idealism' argued that the recognition of 'objects' requires space as well as time in order to make the right kind of 'gap' for judgement to aim at something. But this is all couched in terms of the first person. A more sophisticated idealist could still hold that all that ultimately exists is a world of 'objects' seen from my point of view. That is why the *possibility* of a 'third person' perspective is crucial.

Essay on Truth

Another excellent essay. However, by contrast with the first essay, I have a few criticisms here relating to some of your arguments. I am also not persuaded that your statement, 'To say that a statement is true is the same as saying that it ought to be believed' is in fact, as you claim, a theory of truth.

Let's deal with this first. Belief/ judgement and desire show a profound asymmetry. Roughly speaking, an unsatisfied desire means that there is something wrong with the world; a false belief means that there is something wrong with you.

I want an apple from the fridge, but there is no apple there. That is not my fault, but the world's. I believe that there is an apple in the fridge, but there is no apple there. That is my fault, not the world's.

Arguably (cf. my comments on the first essay) explanation of behaviour necessarily requires attributions of belief and desire in tandem, assuming norms of rationality such as the ones you describe. (This is familiar Donald Davidson territory.)

Now, of course, you are free to say that my example of the apple just shows the distinctive nature of *truth*. However, as the deflationary account shows, there is another way to describe this. It is precisely because the concept of truth allows us to generalize over propositions that it figures in your 'theory' of truth. To say that snow is white is to say that one ought to believe that snow is white. To say that grass is red is to say that one ought to believe that grass is red. What is different between the two cases? Snow *is* white but grass *is not* red. So what one ought to believe is only that snow is white and not the second proposition.

'True' figures here simply as a device for generalizing over propositions, just as the deflationary theorist claimed. Your theory, in other words, is not a theory of truth but a theory of judgement, one version of which is the claim I made about the asymmetry of belief and desire.

Just for the record, I share Dummett's view that there is a substantial *issue* about truth, and this shows that there is more to say than the deflationary theory. What this 'more to say' is, is whether we should be realists or anti-realists about truth (although I do not accept Dummett's formulation of this dispute - but that's another story).

Some of your arguments left me thinking, 'Yes, but...'.

The difference between the infinite axioms and a table giving the positions of stars at all times is that you need to see the whole table to know where Betelgeuse will appear on 30 January at 9.44 am. However, I don't need to have the infinite set of axioms to know that 'Betelgeuse is visible in the Northern Hemisphere on 30 January at 9.44 am' is true if and only if Betelgeuse is visible in the Northern Hemisphere on 30 January at 9.44 am. I just apply the schema. So the schema does tell me something. It tells me that 'is true' behaves unlike any other predicate which can be applied to a proposition/sentence.

I would never say to anyone, on any occasion that 'Whatever JP says is true.' That is rather a big commitment. But it is perfectly possible to say, 'Everything that JP says on page 2 of his essay is true.'

Your criticism of Horwich's statement that ''Snow is white' is true because snow is white' wrongly assumes that the place holders in a 'because' statement must be transparent, not opaque contexts. Substituting snow is white for 'snow is white' is true shows that this is not the case. That is what Horwich would say.

Here is another example:

1. There are eight planets, because the ninth planet suffered a cataclysmic explosion.

2. I had eight chocolate fingers with my tea yesterday.

Therefore,

3. The number of planets is the same as the number of chocolate fingers I had with my tea yesterday because the ninth planet suffered a cataclysmic explosion.

(The second premis is not true, of course, I would never be that greedy. Five's my limit.)

All the best,

Geoffrey