To: Geoff C.
From: Geoffrey Klempner
Subject: Kant's refutation of idealism
Date: 5th November 2010 12:48

Dear Geoff,

Thank you for your email of 30 October, with your second essay for the Metaphysics program, in response to the question, 'Give a careful account of Kant's second 'Refutation of Idealism'.'

Your answer demonstrates a very good grasp of the essential core of Kant's argument against the 'idealist'.

We may assume that 'idealism' -- or whatever is the intended target of Kant's argument -- entails the view that experience is nothing but a temporal series of representations, whose content logically need not be linked in any particular way. To say that the content 'need not' be linked is consistent with observing that it is in fact linked in our experience. The point, however, is that to the idealist (or this kind of 'idealist') this is a purely contingent matter.

What Kant aims to disprove is the assumption of contingency. Moreover, what he claims to be necessary is not only 'some' degree of linkage between the representations in the temporal series but a very particular kind of linkage: they must be such as one could meaningfully hypothesize to be the result of *perception* of *objects arranged in space*.

In short, as you explain very clearly, the concept of space is an a priori requirement for experience as such, for without this there is nothing to bind the series of momentary states into something recognizable (and recognizes itself as) a 'self'.

As you state, 'The Egocentric Subjectivist is refuted because if there is nothing outside of us there cannot be an us at all.'

If you look at Kant's argument in its original context in the second edition of the Critique of Pure Reason, however, there questions which this brief account leaves unanswered.

Right at the beginning of his exposition, Kant states that he is not concerned to refute the 'dogmatic' idealist who claims that the very concept of space is 'impossible' but only the 'problematic' idealist, i.e. Descartes, who raises the question how I can know that there exists an external world, on the assumption that I know that I exist.

Descartes seems a million miles away from the 'Egocentric Subjectivist'. He never considers the idea that all there is might just be 'this experience'. For him, the limits of scepticism reach to the point where we are undecided what is the *true cause* of this experience, whether it is, as our senses seem to tell us, the perception of external objects in space, or, alternatively, an evil demon directly feeding the experiences to us.

Nevertheless, the consequence is the same in either case: whether experiences arise in me uncaused, or whether they are directly produced by an evil demon, on either alternative there is no logical necessity that their content be linked in any particular way. Kant's response to this is that without the required linkage (in the absence of the spatial 'form') there can be no self. Even an all-powerful evil demon cannot create a 'self' if the very conditions for the possibility of a self are not realized.

This tells us something about the 'evil demon' idea, or indeed any hypothesis about the 'ultimate ground for experience'. So far as the refutation of idealism is concerned, it makes absolutely no difference whether you assume an ultimate ground or not. The only thing that is required is that experience take a spatio-temporal 'form'.

In other words, so far as this argument is concerned, any reference to noumena or to how things stand in noumenal reality is otiose: irrelevant and besides the point.

Yet, clearly, this was not Kant's view! Like Descartes, he never once considers the possibility that 'this experience', whatever its form, say spatio-temporal, is 'all there is'.

This is a point on which there has been a significant amount of dispute. Did Kant think that the argument from the refutation of idealism, taken as it stands, is somehow sufficient to establish the existence of a noumenal reality? Is our very notion of an object somehow bound up with the idea that objects of perception have two aspects, the way they exist 'for us' (as spatio-temporal entities) and the way they are 'in themselves'? Are we indeed talking merely about two 'aspects' or two entirely separate realms of existence, the realm of our possible experience and the realm of 'things in themselves'?

And what about the 'dogmatic idealist'? Kant has in mind Berkeley, whose theory of immaterialism is almost indistinguishable from Descartes' evil demon hypothesis. I think that Kant could be accused of misunderstanding Berkeley, however, given that Berkeley never denies the meaningfulness of spatial concepts. Rather, what he rejects is the idea that we have a notion of 'matter' independent of our idea of perception. This is Kant's position too. The only difference between Berkeley and Kant seems to be that whereas Berkeley believes that it is possible to describe the ultimate ground of experience ('ideas in the mind of God') for Kant such talk goes beyond the bounds of metaphysical inquiry.

All the best,

Geoffrey