This is the the second post in my Theories of Knowledge and Reality series. Follow the link for the introduction to the series that explains what it's all about and the list of posts in the series, which I will update as they appear.
Perhaps the most famous line in the history of philosophy is Descartes' "I think therefore I am". What's often misunderstood is what he meant and why he was saying it. He wasn't saying that only things that think exist, as if thinking is what makes us exist, though many jokes rely on that mistake. [Descartes walks into a bar. The bartender asks him if he wants a drink. He says, "I think not" and promptly disappears.] What Descartes was up to with that line, often called the Cogito (from the Latin "cogito ergo sum"), was a response to a particular skeptical claim, and it was the beginning of an extended response to skepticism that most philosophers today believe was completely ineffective. They do, however, tend to think that the skeptical questions he raised are very difficult to get around, and they attribute the persistence of skepticism to him, even if his goal was to refute it. This is how these things go sometimes.
What I want to do in this post is motivate the kind of skepticism Descartes raised. I'll proceed to other kinds of skepticism and responses to skepticism in further posts.
The ancient skeptics had argued that we shouldn't believe anything. Most of their examples had to do with trusting our senses or philosophical views. They said that any view can have a counterargument, so we shouldn't believe anything, and they argued that any perception can have a contrary one from another person, from the same person at a different time or with a different sense, or from how animals perceive from how we do. We have perceptions, and you don't deny your perceptions, but you don't affirm them either.
The Cartesian skeptic is more radical in one sense, however. Descartes, when he starts off his Meditations by portraying the doubts of the skeptical frame of mind, raises doubts not just about particular things, with the particular arguments the ancients had used. He gives these comprehensive skeptical scenarios, which if true would mean most of our beliefs are false. If everything I perceive right now is just a dream, then I don't know if any of it is real. If it could have been presented to me by an evil deceiving spirit, then I don't know if any of it is real. The contemporary example is the Matrix. If I don't know that I'm not in the Matrix, then how do I know there's really a computer here that I'm typing on. How do I even know I have hands? Everything that seems to be true could simply be a delusion created by a mad scientist or evil demon.
The structure of the argument is extremely important, because the responses to the argument have to pay attention to the details of what's being said and what's concluded on what basis. The crucial issue here is that, if X implies Y, then our inability to know Y means we cannot know X.
If we can't know that we're not in the Matrix or dreaming, or deceived by a neuroscientists, then we can't know that our beliefs about our own hands are about real hands we have. Yet it seems obvious that we can't know that we're not in the Matrix. The Matrix scenario is just as consistent with my evidence as the more common belief of what's going on around us -- that there really are these objects out there in the world, and I'm perceiving them. In both cases things seem as they seem. So how can I know which one is true? But then if I can't know I'm not in the Matrix, then I can't know what wouldn't be true unless I weren't in the Matrix -- that the hands I seem to be seeing and feeling are really there.
This works with any ordinary belief that wouldn't be true if some skeptical scenario that you can't absolutely rule out were true. If the skeptical scenario that you can't rule out would have some belief of yours turn out false, then you don't know that thing that you believe. This is why Descartes' skeptical arguments lead to a truly radical conclusion. We know hardly anything. In the next post, I'll sketch out how similar arguments can lead to some serious skepticism even without radical skeptical scenarios, and then I'll examine various responses to skepticism.
Update: If you want to read Descartes for yourself, I suggest using Jonathan Bennett's translations of the Meditations. It's designed for people who aren't familiar with reading early modern philosophical texts. Those who are familiar with this sort of thing should of course look at one of the most recent scholarly translations (not that Bennett isn't a scholar, but these texts were prepared for introductory courses), but those just exploring this sort of thing for the first time can probably benefit most from Bennett's version, which is more readable than any other translation you could possibly find and sometimes contains some commentary set off in brackets. For more information on what sorts of things he's done in his translations, see this page. If you want a more traditional translation, Bennett recommends Cottingham. If you want one online, see here.