To: Simon K.
From: Geoffrey Klempner
Subject: Kant's refutation of idealism in Critique of Pure Reason 2nd edition
Date: 29th October 2009 13:56

Dear Simon,

Thank you for your email of 24 October, with your second essay for the Metaphysics program, in response to the question, 'Give a careful account of the argument of Kant's Refutation of Idealism.'

This is a very good piece of exposition, which has the merit of keeping close to the actual text of the Refutation of Idealism from the 2nd edition of Kant's Critique of Pure Reason.

Taken simply as it stands, however, I don't find Kant's argument fully convincing. Right at the start, as you indicate, Kant states that he is not concerned with the 'dogmatic' idealism of Berkeley according to which there are no material objects. His target is the 'problematic' idealism of Descartes which adopts an agnostic position, looking for proof that material objects exist, in the face of the possible hypothesis that my experiences have a different nature or cause from what I take them to have.

According to Descartes' sceptical hypothesis, my experiences are not caused by perception of material objects in space. (Kant does not actually mention Descartes' 'evil demon'.) What Descartes never once considers, however, is that my ideas are all that exist, lacking any kind of 'external' cause.

The problem with Kant's account of what he is setting out to prove is that we already have, in Berkeley's theory of 'subjective ideas which correspond objective ideas existing in the mind of God', or what almost amounts to the same thing, 'ideas produced in my mind according to the whim of an evil demon', a model of something which exists externally to me and my ideas. By hypothesis, both God and the evil demon 'persist outside' the 'representations' which I find in me.

Now, of course, Kant can say (although he doesn't say) that the very same problem of time determination applies to the consciousness of God or the evil demon. If all they are given is their ideas, then they are no better of than me. The philosopher isn't really in a position to say what the consciousness of God 'must be like'.

Despite these reservations, I would accept that Kant successfully argues for the necessity of an idea of space. That is to say, whatever its ultimate source, experience must be represented 'as being of' external material objects.

The problem with this interpretation is that both Berkeley and Descartes can happily agree to this. Yes, my experience must have a 'spatio-temporal' structure. But that says nothing about how things are, in reality, in particular it says nothing about whether or not 'material objects' in the sense attacked by Berkeley, exist.

Indeed, what we find in the Critique is the view that material objects are 'empirically real but transcendentally ideal'. The world of my experience is a 'phenomenal world', a world of appearances. But as there 'cannot be an appearance without something that appears', we must also posit a noumenal world, concerning which no knowledge can be had (because such knowledge would transcend the bounds of experience).

It has been said that Kant' added the Refutation of Idealism to the 2nd edition of the Critique of Pure Reason because of charges levelled at the 1st edition that Kant theory was just a version of idealism. As such, one can hardly say that he succeeded.

However, I do think that something is achieved by the Refutation of Idealism, if we interpret this along the lines suggested by P.F. Strawson in 'The Bounds of Sense'. Kant's real target is not the Berkeleian idealist or Cartesian sceptic but rather a 'subjective idealist' or solipsist who claims that 'my ideas are the only things that can conceive of as existing'.

As you state, there is nothing to 'identify' an idea that I find in me at one time with an idea that I find in me at another time. To posit an 'I' which is aware of both ideas, one after the other, begs the question because we haven't so far established an 'I' with an identity over time.

In fact, if one is looking for a historical antecedent for this view, it would be neither Berkeley nor Descartes but Hume, in his 'bundle theory of the self'.

As I argue in unit 4 of the Metaphysics program, what is missing in the subjective idealist picture is simply the theory of a spatial or quasi-spatial array. This is a structure which places perceptions at specific points on the array, which the subject repeatedly tests against the actual course of experience. In other words, what the Refutation of Idealism actually establishes is that 'space is a necessary theory'. What it doesn't establish is what Kant seems to claim for it, namely that we exist as material entities in a material world.

Or, rather, according to Kant, the world is 'material' or 'real' but only in the 'empirical' sense, in terms of what is needed to make coherent sense of experience. With this, Berkeley would have no disagreement.

All the best,

Geoffrey