To: Emelie G.
From: Geoffrey Klempner
Subject: Can you know that you're not dreaming?
Date: 4th April 2011 12:10

Dear Emelie,

Thank you for your email of 22 March, with your essay for the University of London BA Epistemology module, in response to the question, 'Can you know that you're not dreaming?'

I agree with a lot of this essay, and in particular your criticisms of Nozick and Moore.

Nozick offers a definition of 'know' in which the closure principle is denied (possibly, you could have been more explicit about this, and talked about Nozick's notion of 'tracking truth'). But the question is why we should be interested in this (re-)definition, given the intuitive force of Descartes' evil demon scenario.

Moore asserts that he knows that he has two hands, and therefore (by the closure principle) he knows that he is not dreaming. But how is this claim of knowledge to be distinguished from a mere confession of subjective certainty? (which Descartes fully admits when he says that he is sitting by the fire etc.). You can be certain, to the maximum degree, and still be wrong. Arguably, we have all experienced this at some time or other.

Wittgenstein, in his 'On Certainty' (which I think is up there with 'Philosophical Investigations' in terms of quality and originality) offers a more subtle diagnosis, based on a critique of Moore's argument. The question of whether or not I am dreaming is not a 'proposition' susceptible of 'proof' or 'disproof', but more like the 'river bed' along which language flows, or the 'hinge' of a door. This has a very Kantian sound to it: the idea that certain fundamental propositions are the 'condition for the possibility of meaning', and therefore not susceptible to disproof (just as, for Kant, the principle of determinism or the belief in the existence of an external world are not susceptible of disproof).

I'm not totally satisfied with this, however.

I'm pleased that you *didn't* pursue the suggestion in paragraph 1, 'As a rationalist [Descartes] aimed to find some certain starting point from his own private consciousness. But certainty as such is a psychological state one is in and about whether one is right or wrong.' The next step would be to call up Wittgenstein's argument against a private language (where one cannot be 'wrong' one cannot be 'right' either). But all that shows is that Descartes should have been *more* of a sceptic, and doubted everything, including his knowledge of his own psychological states. So that doesn't help.

But what is Wittgenstein's *argument* in 'On Certainty' against the sceptic? He makes claims about 'meaning', but why can't the sceptic just reject all these claims. 'I don't know whether I'm not dreaming; I don't know whether my words have meaning.' All you can do is wag your finger.

I think there is an argument here, which can be extracted from 'On Certainty'. It's a dialectical argument, directed against the sceptic who says, 'I doubt whether or not I am dreaming.' The question is, what the sceptic means by 'doubt'. Let's say, I doubt whether the chair which I accidentally knocked over and damaged yesterday will hold my weight. So I sit down gingerly, rocking backwards and forwards, listening intently to any signs of the wood cracking. Finally, I'm satisfied, the chair is safe. Doubt is connected to action. That isn't to say that there is any specific action which one should take when one doubts; only that, along with my beliefs and desires, my doubts are 'explained' by the things I do, my physical actions in the world. There doesn't need to be a direct connection, an indirect one will do. Wittgenstein's argument against the sceptic is that there is no possible connection, however indirect, between the putative doubt whether I am dreaming and my actions in the world.

Unfortunately, in your essay, you allude to a fatal flaw in this argument. We've seen The Matrix. We can believe that this is a possible world, a possible experience that you or I might have. So let it happen. We wake up in our pods and join Neo and the other members of the resistance. Now, the very real possibility of 'doubt whether one is dreaming' arises. One morning, I wake up in my bunk in the Logos hovercraft, but everyone seems to be behaving oddly. Then it occurs to me that I'm being tricked, this isn't real, someone has inserted me into the Matrix. What *actions* would be appropriate in those circumstances?

You can extend this thought experiment various ways. I could kill Morpheus, Trinity, Neo etc. but the consequences will be just as they would be if I killed the real Morpheus etc. (which maybe they are). I can try various ways of 'waking up', but maybe I fail and this experience goes on indefinitely. How long would my 'doubts' survive? There doesn't seem to be any rational basis on which to decide. At some point I shrug off my original feeling, or maybe I don't. But the *question* is still real, whether I continue to dwell on it or not.

Although the makers of The Matrix trilogy don't pursue this plotline, it does feature in another movie eXistenz. I won't say anything more in case you haven't seen it!

All the best,

Geoffrey