To: Neil G.
From: Geoffrey Klempner
Subject: Descartes' case for doubt in the First Meditation
Date: 13th August 2008 11:43

Dear Neil,

Thank you for your email of 31 July, with your essay in response to a question taken from the University of London Modern Philosophy paper, ''There is not one of my former beliefs about which a doubt may not properly be raised' (Descartes). Does Descartes succeed in showing this in the first Meditation?'

This is a very good piece of work.

You raised one question -- about how fish or insects perceive the world -- which I have not seen in an essay on this topic before. One possible response from Descartes would be to say that a God who is not a deceiver makes sure that the way in which we perceive the world is suitable for beings placed in the world with the bodies and powers that we possess. Had we the sense experience of insects, then this would be appropriate if we had the bodies of insects, and similarly for fish living in the sea. However, the wider question is whether indeed points of view or forms of perception in this sense limit the possibilities for gaining knowledge about the world.

One response would be to observe that, armed with suitable equipment which we are able to construct for ourselves, the limits of our sense perception are indeed overcome. Descartes' view of scientific knowledge is reductionist, to the extent that, say, the perception of red is merely a subjective indication of a property existing objectively in reality which we are able to detect with our sense organs. A merely mischievous demon, who gave us sense organs which were unsuited for gathering knowledge, could make things difficult but not impossible. We could strive to overcome our limitations, as indeed we do in the real world.

I wasn't totally happy with the claim which you made right at the beginning of your essay that Descartes is assuming a concept of knowledge according to which the slightest possibility of doubt is inconsistent with the claim to know that P. If that were true, then an objector could simply turn round and say, 'You can say what you like about Knowledge with a capital 'K', but my interest is with knowledge. I accept that you can think you know and be wrong. But that is consistent with holding that even though I cannot be 100 per cent certain, I can make justified judgements of probability, which can be high enough for knowledge.'

I have seen the claim about infallible knowledge in text books but I think it is incorrect. The argument goes, 'Descartes wrongly infers that knowledge must be infallible, from the statement, 'If X knows that P, then P must be true'. What he fails to grasp is that the 'must' here merely has the force of a logical inference.'

However, even if Descartes did make this error, his sceptical arguments would still deserve to be reckoned with as arguments against knowledge with a small 'k'. If we assume the possibility of an evil demon, then we cannot make judgements of probability, because probability is relative to given evidence. If we are not allowed to take anything as 'given', then we can't say, for example, whether it is probable or improbable that an evil demon itself exists.

I don't think he did make this error. Remember, that Descartes is seeking to re-establish our ordinary, everyday knowledge claims on the basis of the proposition that God is not a deceiver. That is indeed certain knowledge, but our everyday knowledge claims remain open to the possibility of doubt, just as before, and Descartes must surely have realized this. Indeed, in the 6th Meditation he goes to some lengths to explain how, even when we make the best use of our judgement and sense perception, nevertheless errors occur.

Another claim which I would dispute is where you agree with Descartes that casting doubt on individual knowledge claims would not be an effective argument for scepticism. Isn't it? Consider the man and the tree. You tell me that you saw a man standing in the garden and I ask you whether you are sure that it wasn't just a shadow. Well, maybe you are sure but are you sufficiently sure for knowledge? Suppose your life depended on it. Then, maybe not. But we can do this with any ordinary knowledge claim. There is always the possibility of a fluke or a wild coincidence. OK, you say, but what happens occasionally can't happen all the time. However, if in *any given case* this argument can be constructed, then, in effect it does become a global argument for scepticism. You may 'know' that a disjunction of all your beliefs is true (P or Q or R or...). But that 'knowledge' is not very useful.

This is serious for Descartes, because if we accept this argument then, given what I said above about a God who is not a deceiver, here we have an argument for scepticism which Descartes is powerless to refute using the strategy which he has chosen. (I have to add that I don't think that the argument I have given is a valid argument for scepticism, but that's another question.)

On the reliability of sense experience, just before the point about the fish and the insects, you remark that Descartes recognizes the limits of arguments based on examples such as seeing something far away (Descartes gives the example of seeing a tower in the distance) because 'we have just one, particular, example of our senses deceiving us'. Given what I've said above, this would be an effective argument if, for any given perceptual claim, such a doubt could be raised. What Descartes in fact says is that we make the judgement that we have been deceived *on the basis of sense perception*. It follows (or so Descartes thinks) that one cannot argue from particular examples to the unreliability of sense perception as whole.

On dreaming, you make an interesting observation about the difference between dreaming and waking experience. Suppose our dreams were as coherent as our waking experience. What then?

This is a possibility which has in fact been investigated, by Anthony Quinton in an often-cited paper, 'Spaces and Times'. Quinton is arguing against Kant's claim, in the first part of the Critique of Pure Reason, that there is necessarily one space. Quinton asks us to imagine that when you go to sleep you wake up in an idyllic fishing village where you have a wife, friends and family. When you go to sleep there, you wake up here. This pattern is repeated indefinitely. On the basis of coherence, one would have to judge that both experiences were equally veridical and that therefore you inhabited two different worlds. (I won't say any more about this because it takes us off topic!)

It is true that Descartes is relying on our experience of dreams. If you have never had a dream, you might find the idea that you could have experiences which bore no relation to external reality fantastic. But, strictly speaking, the argument doesn't depend on this contingency. The point is that it is possible, in principle, to have an experience whose source is other than what it purports to be. The evil demon is indeed such a source. All that the evil demon adds to this hypothesis is the deliberate intention to deceive.

On the evil demon, Descartes explicitly says that a being with God-like powers could deceive us even when we are making simple additions. This does seem rather implausible with an example like 2+2=4, but still an argument is needed. It is a contingent fact that our judgements about such matters are reliable. So, even though the judgements themselves are not about the external, contingent world but rather the realm of logical or mathematical truths, there remains the possibility of undermining sceptical doubt.

All the best,

Geoffrey