Topics Knowledge

Question I have a question about Cartesian skepticism. One of the premises of the argument is something to the effect of: (1) I don't know that I'm not dreaming. My question is: What justifies this proposition? My intuition is that the evidence for (1) cannot possibly be empirical; for the upshot of the skeptical argument is precisely that all empirical claims are dubious. (Maybe it's helpful to rephrase (1) as "It's possible that I'm dreaming," if that is legitimate.)

Accepted:
September 14, 2008

Comments

Descartes' arguments suggest that he believe d that, for any empirical test that you might devise to determine whether you are awake or dreaming, it might be the case that anytime you appeal to putative test results you have merely dreamed that you have performed the test. So, I think you are right that, whatever arguments Descartes developed to respond to his own skeptical doubts, those arguments were not straightforward empirical ones appealing to things like tests of that sort. Similarly, when he addressed issues related to skepticism Descartes' eighteenth-century successor developed a style of non-empirical transcendental argumentation.

Another way to approach the issues that your question engages is to focus not on Descartes' specific argument but rather on the general skeptical doubt that it can raise so hauntingly. The contemporary philosophy Barry Stroud puts this doubt this way in his wonderful book, _The Significance of Philosophical Skepticism:"Could not the external world be completely different from what we perceive it and believe it to be?"

One reason why I find this "bare doubt" interesting is because it can seem so powerful: even without pondering sophisticated skeptical arguments, one can find arresting the thought that we might be experiencing systematic and general error: one need not read Descartes to worry about whether or not one might be dreaming right now.

One might wonder what the strength of this doubt tells us about strategies for investigating skepticism philosophically. How seriously should philosophers take this doubt? For example, should they strive to prove that the global skeptical possibility does not obtain? If not, why not? If so, how might one do this? Can any of Descartes' arguments do this? Do the transcendental arguments offered by Kant or by the contemporary philosopher Donald Davidson achieve this?

One can also use this doubt to explore specific characteristics of knowledge. For example, many philosophers have written about an epistemic principle that seems intuitively true to many of us: knowledge is "closed under known deduction" in the sense that, if I know one thing, and I know that a second thing is logically entailed by that first thing, then I also know the second thing (schematically: If S knows that p, and knows that p entails q, then S knows that q). This relates to the skeptical doubt in two important ways. First, if knowledge is deductively closed in this way then, if we have empirical knowledge, we must know that we aren't in massive error. Second, we can have empirical knowledge only if we know that the global skeptical possibility does not obtain.

This gives us one exegetical hook for exploring the dream argument: Does something like the closure argument lurk behind Descartes' concern with whether or not he can prove that he is not dreaming? Did he believe in a principle like this, or, as we attempt to reconstruct his reasoning, we conclude that he should have?

Likewise, one could pursue questions like: Does Descartes' argument provide any reason for thinking that the closure principle is true? Are there reasons for rejecting the closure principle that might also cause us to reject Descartes' argument?

There are many other ways to interpret--and to criticize--Descartes' dream argument, and may other ways to approach issues related to skepticism. Finding various ways to summarize and re-phrasing Descartes' argumentation is a powerful method of exploring this philosophical abundance that his texts provide us, and so I am pleased that you began that work in your question.

You write "One of the premises of [the skeptic's] argument is something to the effect of:I don't know that I'm not dreaming." And yes, as you imply, it would be rather odd for a skeptic to start by being too dogmatic about what he can or can't know!

But perhaps he doesn't need to be. Perhaps it is better to think of the argumentative situation like this. You are cheerfully going about your business, thinking that you know perfectly well that you are seeing a computer screen right now (and the like). Your friendly neighbourhood skeptic then issues a challenge: how do you know that? You appeal to the evidence of your senses. Your friendly skeptic chides you: how do you know they don't lead you astray all the time? Perhaps it is all just a vivid dream. Or perhaps an evil demon is making it seem as if you are seeing a computer screen when you aren't. In modern dress: perhaps you are a brain in a vat, being stimulated by a mad scientist so you still think you are embodied and seeing a computer. How do you rule that out?

Here, then, the skeptic does his work by challenging you to rule out an alternative story to the commonsensical one, an alternative story which apparently fits the evidential facts equally well. The skeptic doesn't himself have to dogmatically assert that you don't know you aren't dreaming. It is enough -- to sow the seeds of doubt -- to challenge you to show that you know that you aren't!