To: David Y.
From: Geoffrey Klempner
Subject: Kant's Refutation of Idealism
Date: 8 April 2001 09:53

Dear David,

Thank you for e-mail of 28 March, with your Metaphysics essay in response to the question, 'Give a careful account of the argument of Kant's (2nd Edn) Refutation of Idealism.'

You mentioned that you were 'coming out of hospital'. I do hope that you are in a lot better health now then when you went in. It doesn't always happen these days.

This essay is a remarkable piece of work. In a very short space you have managed to give an account of Kant's 2nd edition refutation of idealism which sticks close to what Kant actually says, which also has some persuasive force.

You are right that the crux of Kant's argument is the difference between a 'permanent representation' and a 'representation of a permanent'.

Let's go with this idea for a while, and see if we can construct a suitable thought experiment. Let's suppose that there is a race of alien beings who have viewfinder clock in the top left hand corner of their field of vision (implanted, perhaps, by a benevolent Creator). Rather like the kind of thing one might find in a video camera.

So the aliens have no need of physical clocks. They always know the time. Or do they? It could always happen that an alien gets a bump on its head, which puts its subjective clock out of synch. There are alien doctors who specialize in repairing and resetting these internal clocks.

Now let's suppose that one of these aliens reads Descartes' 'Meditations'. The alien concludes: 'For all I know, my internal clock might be running fast or slow. Or it may be set to the wrong time. But what I do know for sure is that I had a thought about Mars as my clock indicated five minutes past four, and a twinge of pain in my back as my clock was indicated six minutes past four.'

In other words, the temporal order of our subjective experience is a datum. A complete description of subjective experience does not need to refer to objective times at which experiences occur, but merely the order in which they are subjectively perceived to occur.

According to the problematic idealist, this is all I know, or can know - unless some reason can be presented to me why my experience in its subjective time order should correspond to a world of objects arranged in space.

With this in mind, let's look again at step 1. 'I am conscious of my own existence in time.' What I am conscious of, according to our problematic idealist, is a stream of experiences ordered in time. My thought about Mars occurred before the twinge in my back, that is a datum. There is no *measure* of the time that elapsed between the first inner perception and the second other than a purely relative one' (e.g. the aliens' internal video clock).

It now looks as though Kant's argument is not going to get off the ground. The problematic idealist is not going to be so careless as to claim that their subjective impression of the 'speed' at which time passes corresponds to anything outside the subjective stream of experiences!

You are right to mention Kant's 'Analogies'. The key point here is that the idea of an objective time order is bound up with a world of objects which are related to one another causally as well as spatially. Take away causation and there is no way of applying inductive methods to determine, on the basis of perceptual evidence, which objects have moved where. In relation to what Kant is claiming to prove in the Refutation of Idealism, however, this just looks like begging the question.

By contrast, in a purely subjective time order, it seems that the notion of causation is not needed. There is no causal link between my thought of Mars and the twinge in my back. One happened, then the other happened. That's all subjective experience is. Something that just happens. And that, says the problematic idealist, is all I know for sure.

That is why, in the Metaphysics program, the argument I attribute to Kant goes beyond what he actually says. This is always a risky thing to do when interpreting a historical philosopher. We want to make their arguments look as persuasive as possible. It is a methodologically sound, I would say necessary principle, to assume that the great philosophers do not make stupid errors, although they do sometimes express themselves with less care than one would have liked. (By Kant's own admission, the two editions of the Critique of Pure Reason were produced hurriedly.)

Your essay is a timely reminder to me that some times one can go too far along this road, however. Thinking about it now, I am actually doubtful whether Kant would accept my version of the 'Refutation' as an acceptable gloss on what he said, or meant. But no matter. For the purposes in hand, it is important that the philosopher who I term the 'transcendental subjectivist' (or 'transcendental solipsist') has at their disposal an argument (the argument from the difference between 'true' and 'false' memories) which shows that if experience occurs at all, and if there are true judgements of the form, 'I remember that my experience of X occurred before my experience of Y', then experiences must be *interpreted as* perceptions of a world of objects distributed in space which I myself occupy. That is a significant result, because it shows that the theory of transcendental solipsism has a certain degree of resilience.

In other words, I give a stronger argument than the one Kant actually gives for a possibly weaker conclusion. - All of this is blown away, of course, by the private language argument. But that's another essay!

All the best,

Geoffrey