To: Simon Y.
From: Geoffrey Klempner
Subject: Descartes and the malicious demon
Date: 12 November 2007 10:46

Dear Simon,

Thank you for your email of 1 October, with your University of London essay in response to the question, 'I shall suppose that… some malicious demon of the utmost power and cunning has employed all his energies in order to deceive me.' What leads Descartes to make this supposition?

The concerns that you express in your email betray a misapprehension about the way that philosophers work. When looking at a philosophical question, it is impossible to go right back to the basics and question everything: you have to start somewhere, by 'bracketing' (assume answers to) quite a lot of questions in order to concentrate on other questions.

In particular, when you are writing an essay in response to an exam question, your main concern is evaluating arguments - as in the above, in Descartes's argument in the First Meditation. You can be interested in an argument because it is a good argument, even if you have already concluded (for other reasons) that the conclusion which the argument purports to establish is, in fact, false.

This is, in fact, a very good essay. Given what I have said about 'bracketing', it is worth mentioning that what Descartes is seeking to do in the First Meditation does not contradict the idea that one can put aside certain questions in order to concentrate on other questions. Descartes stated aim here is to consider 'what may be doubted', in order to determine whether there might be, after all, a way to resist this doubt and thus establish secure foundations for knowledge.

What examiners are looking for is an exposition of Descartes' argument which either makes it sufficiently credible that one would seriously consider the hypothesis of an evil demon, or explains where Descartes goes wrong in thinking that he needs to address this hypothesis. In other words, 'what leads Descartes to make this supposition' can either be a valid argument (in your view) or an invalid argument. If it is invalid, then the question is where the fallacious step arises.

You steer a middle course between these alternatives, endorsing Descartes' reasons up to a point, but also suggesting difficulties with some of the claims that he makes: which is a perfectly acceptable strategy.

I liked your example, or rather 'counter-example' to the claim that if a belief relies on other beliefs one or more of which has been found to be false, then that belief must itself be rejected. Let me see if I have got this right. If I believe that team A beat team B 2-1, and as a result of this believe that team A have gone up one position in the table, my belief that team A have gone up one position is unaffected by doubt whether the scoreline was in fact 2-1 or 3-2, so long as A won. (This of course depends on my prior belief that A is ahead of the next team in the table on points and not merely on goal average, as there are possible circumstances under which A would go up one place based on a score of 3-2 but not if the score is 2-1.)

Suppose we put this point to Descartes. His response would be that even though I am less than certain about my belief that the score was 2-1 rather than 3-2, what I am certain of, and what my belief that A has gone up one place depends on is the truth of the disjunctive belief 'Either the score was 2-1 or 3-2', or maybe just the more general belief that A won the game. If I was uncertain whether, in fact, A won or lost the match then that would destroy my confidence in the belief that they went up one place. Nevertheless, it is still a good point to make: in the real world, we often tolerate a degree of uncertainty in the grounds for our beliefs, where this can be kept within acceptable limits. In the witness stand, under cross-examination I stick firmly to my claim that the person I saw running out of the bank is the accused, even though I can't remember whether that person was wearing a grey hoodie or a black one.

You are right to raise the sensitive issue, which Descartes dismisses far too readily, concerning whether he is in fact sufficiently sane to be able to form rational beliefs. Why isn't this something concerning which one might consider doubt? Descartes is not forthcoming about this point, and therefore you are justified in pressing him and pointing this out as a possible weakness in his case. Suppose Descartes convinces the reader that the possibility of a malicious demon has to be considered, in order to be rejected. That rejection will not count as a refutation of scepticism, so long as the even more radical possibility that Descartes is mad or irrational is left unanswered.

All the best,

Geoffrey