We discussed Descartes’s method of doubt and how it leads to skepticism about knowledge based on the senses.
By “skepticism” I mean the view that we have no knowledge in a given area. In this case, Descartes is saying that we do not know anything on the basis of the senses. Suppose I see what looks like a table. My belief that there really is a table that is causing me to think I see a table is no better than any one of a number of alternative explanations such as “I’m dreaming that I see a table” or “an evil demon is causing me to think I see a table.” They’re all tied for last. That’s what I mean by skepticism about knowledge based on the senses: the senses give you no better reason for believing one thing rather than another.
The dream argument does two basic things:
So, for instance,
When we considered the obvious test to determine whether you are dreaming, we came across the same argument: you can’t know that the test is satisfied (and hence that you are not dreaming) unless you know that you aren’t dreaming that the test is satisfied.
We generalized Descartes’s condition on knowledge thusly: you can’t know something unless you also know that incompatible alternative possibilities are not true.
Ian, defender of common sense, noted that this is a very strong requirement with apparently absurd consequences.
He’s right. At the same time, it isn’t hard to come by the requirement even in normal reasoning. That was the point of my example of the canary and the goldfinch: you don’t know that the yellow bird you saw was a canary unless you know that it was not a goldfinch.
More broadly, we seem to work with an everyday sense of terms like ‘knowledge’ and a stricter one. That’s why you can get people in the lunch room to admit that they don’t ‘really’ know what’s on their plate even though they will go on to act as if they know it is food (they don’t feed a bite to a squirrel just in case it’s poisonous, and so on).