Descartes' Cogito and Methodical Doubt

Strip away every belief that can be questioned, and watch what is left standing. Descartes' radical experiment in 1641 gave modern philosophy its first foothold — and its first dilemma.

In the winter of 1639, in a warm room somewhere in the Dutch Republic, a forty-three-year-old former soldier sat down to take everything he had ever believed and burn it to the ground. Not literally. He was going to do it on the page. He had decided that the only way to build knowledge on something solid was to first see whether anything was solid. The result, published two years later as the Meditations on First Philosophy, is a slim book of six chapters. It is one of the strangest, most influential things ever written in philosophy.

René Descartes was not a sceptic by temperament. He was a mathematician who had just helped invent analytic geometry, and he believed deeply that knowledge could be made as certain as a proof in Euclid. But you cannot prove anything from premises that are themselves doubtful. So his method was, in a sense, surgical: doubt everything you possibly can, exhaustively and on purpose, and see what cannot be doubted no matter how hard you try. Whatever survives that is bedrock. From the bedrock you rebuild.

The method of doubt

Descartes begins with the senses. They sometimes deceive — a stick in water looks bent, a tower at a distance looks round when it is square. A prudent person, he says, never trusts entirely what has deceived them even once. But he does not stop with optical illusions; that would be too cheap. He pushes further.

Right now, as I sit by the fire holding this paper, surely this cannot be doubted? Except that I have dreamed, vividly, of sitting by fires holding papers, and inside the dream I had no idea I was dreaming. There is no internal mark, Descartes notices, that reliably tells a waking experience from a dream. So the entire physical scene around me — the fire, the paper, the body holding it — might be a kind of hallucination. The whole sensory world goes onto the pile of doubtful things.

Mathematics, though, ought to be safe. Two plus three is five whether I am awake or asleep. But Descartes goes further still. Suppose, he says, there is an evil demon — an immensely powerful and supremely cunning deceiver — whose entire purpose is to fool me. The demon makes the external world appear, generates the dream of a body, and even arranges things so that when I add two and three my mind ticks over to five, although in fact the answer is something else. If such a being exists, even arithmetic cannot be trusted. There is no belief, no matter how innocent-looking, that the demon could not corrupt.

This is the maximum dose. After it, almost nothing is left.

The thing that cannot be doubted

And yet. The demon is busy deceiving me. He is filling my mind with false beliefs about the world, false sensations, false arithmetic. But for him to deceive me, there must be a me to deceive. There must be something doing the being-deceived. The demon can falsify the content of my thoughts, but he cannot falsify the bare fact that thoughts are occurring and that someone is having them. The act of doubting itself is a kind of thinking. And thinking, occurring right now, is undeniable. The harder I doubt, the more certain it becomes that there is something doing the doubting.

“I think, therefore I am.”

This is the Cogito — from Descartes' Latin phrasing, cogito, ergo sum. (He had used the formulation already, in French, in the Discourse on Method four years earlier: je pense, donc je suis.) It is not a syllogism, despite the therefore. It is not the conclusion of an argument with hidden premises. It is closer to a flash of self-recognition: an act that, in being performed, guarantees that there is a performer. To doubt that I exist, I have to think; and if I am thinking, I cannot help but exist. The Cogito is the survivor, the one piece of furniture left standing after the demon has flung the rest out of the window.

Notice what Descartes has and has not won. He has won the certainty that some thinking thing exists. He has not won the certainty that he has a body, or that the world around him is real, or that other minds exist, or that two plus three really is five. All of those remain, for now, on the rubble heap. The Cogito is a foothold, not a house.

The thinking thing and its body

What kind of thing, then, am I? I am, at minimum, a thing that thinks — res cogitans. Whatever else is uncertain, it is certain that there is here some activity of doubting, perceiving, affirming, denying, willing, refusing, imagining, sensing. That is what I am, at the minimum I can be sure of.

The body is different. The body, with its hands and fire and paper, is exactly the sort of thing the dream argument and the demon together have just put in doubt. Descartes concludes that mind and body are, at this point in the argument, two distinct sorts of stuff: one whose essence is thought and which can be known directly, the other whose essence is being extended in space and whose existence still has to be argued for. This is Cartesian dualism, the metaphysical inheritance that haunts every philosophy-of-mind course since.

This dualism is also where Descartes' system starts to wobble. If mind and body are genuinely different kinds of stuff, how do they touch? When I decide to lift my hand, how does a thinking thing move an extended one? Descartes' own answer — that the interaction happens in the pineal gland — satisfied almost nobody. The mind–body problem has been with us ever since.

What the Cogito starts, and what it doesn't finish

Descartes himself thought he could rebuild the rest. The bulk of the Meditations goes on to argue, from the existence of the thinking self, to the existence of a perfect God, to the conclusion that such a God would not allow us to be systematically deceived about clear and distinct ideas, and therefore that mathematics and a chastened version of the external world can be trusted again. This larger argument convinced few outside his own circle. The bridge from the Cogito back to the world rests on a proof of God most readers find unpersuasive, and on a circularity that critics have flagged since the seventeenth century: he uses clear and distinct ideas to prove God, then uses God to underwrite clear and distinct ideas.

But the move at the heart of the book has had a much longer life. Three habits of thought, in particular, descend from it. The first is the idea that the foundations of knowledge are to be found by stripping away, not by piling on. The second is that the most secure thing we know is something about our own minds — the inward turn that runs through Locke, Berkeley, Kant, Husserl, the phenomenologists, and most of cognitive science. The third is dualism, which we have spent four hundred years trying either to vindicate or to dissolve, with no settled verdict.

What is striking is how light the original move is. Descartes is alone in a heated room. He performs an act of attention. He notices that the very performance of doubting cannot doubt itself away. From that one point he believes he can build everything else. He could not, in the end. But that small luminous point — the thinking thing catching itself in the act of thinking — has been there at the start of modern philosophy ever since, like a pilot light that nobody has managed to put out.