DEEP READ · 50
Meditationes de prima philosophia · René Descartes · 1641
A man shuts himself in a room, resolves to throw out everything that admits even a sliver of doubt, and checks whether any ground is left to stand on. All he salvages is a single sentence—I am, I exist—and from that one point he tries to buy the entire world back. He succeeds, but the world he gets back has been cleaved in two: a mind that takes up no space, and a matter that obeys nothing but mechanics. Modern philosophy, modern science, and every headache we still have about what consciousness is grew out of that crack.
Descartes (1596–1650) was first of all a mathematician and physicist—analytic geometry, the coordinate grid you drew in school, is his. He wrote this slim work of metaphysics not out of speculative appetite but to pour a foundation under his new physics, in which the universe is nothing but matter in motion. When the Latin first edition appeared in Paris in 1641, it was printed together with six sets of objections from Hobbes, Gassendi, Arnauld and others, plus Descartes' point-by-point replies: the book was a public argument from day one. It is called the opening of modern philosophy for one reason—after it, philosophy's first question is no longer "what exists" but "what entitles me to claim I know".
Descartes offers a fine image of his own in the Replies: if a basket of apples has a few rotten ones in it and you fear they will spoil the rest, the efficient move is not to inspect them one at a time but to tip the whole basket out and put back only what you have checked. What he wants to clear away is not a few mistaken opinions but the ground opinions stand on. He has believed many things that turned out false, so the source of his beliefs is untrustworthy; better to assume that whatever can be doubted is flatly false and see what is left.
Two qualifications are routinely forgotten, and they decide how the whole book reads. First, this is hyperbolic doubt: treating "slightly doubtful" as "already false" is absurd in daily life but ferociously effective as a filter—it guarantees that any survivor is solid gold. Second, Descartes says plainly that the procedure belongs in the study. For living he keeps a provisional morality (set out in the Discourse on Method): follow the laws and customs of your country, act resolutely once you have decided, and sooner change your own desires than demand that the world rearrange itself. He is not asking you to doubt whether to eat breakfast. He is running a controlled experiment.
The First Meditation is a beautifully engineered machine that raises the pressure in stages, each one taking away what the last let through.
Stage one: sensory illusion. A square tower looks round from far off; an oar dipped in water looks bent. The senses have deceived me, and by the rule "doubtful counts as false" they forfeit my trust. But this stage is weak: it undermines the distant and the blurry, not "I am sitting here by the fire holding this piece of paper."
Stage two: the dream argument. I have dreamt of sitting by the fire often enough, and no dream ever carries a reliable mark saying "this is a dream." If waking and dreaming cannot be told apart from the inside, then "I am sitting here" is doubtful too. That knocks down the entire external world—but it leaves a loose end: even asleep, 2+3 is still 5. Dreams can dismantle and recombine sensory material (as a painter can paint a centaur, using parts of real horses and real people), but they cannot touch the simplest and most general things.
Stage three: the malicious demon (genius malignus). To bring mathematics down as well, Descartes calls in the ultimate hypothesis: an "extremely cunning and extremely powerful" demon bending all its effort to deceive me—sky, earth and my own body are phantoms it plants in me, and every time I count 2+3 it tampers with the result. The elegance of the hypothesis is that it does not ask you to believe such a demon exists; it only requires that you cannot prove one doesn't. Since it cannot be ruled out, the rule says treat it as false. At this point everything is gone: world, body, mathematics, memory. (Every brain-in-a-vat and simulation argument you have heard is this demon in electronic dress.)
| Stage of doubt | What makes it doubtful | What is still standing |
| 1. Sensory illusion | The senses have deceived me before | Near, vivid things; my body; mathematics |
| 2. The dream argument | No reliable mark divides waking from dreaming | Mathematics and the simplest general truths |
| 3. The demon | An omnipotent deceiver cannot be ruled out | Only the fact that I am being deceived |
Each stage eats the survivors of the last, until a single point is left.
The stronger the deceiver, the more it exposes one hole: to deceive me, there must first be a me to deceive. "Let him deceive me all he can, he will never bring it about that I am nothing, so long as I think I am something." Hence the pivot of the book: "I am, I exist—this proposition is necessarily true whenever I put it forward or conceive it in my mind."
Three details are worth nailing down, because almost everyone gets them wrong:
So what is this "I"? A thinking thing (res cogitans)—where "thinking" is stretched very wide: doubting, understanding, affirming, denying, willing, refusing, imagining, and sensing. Note that last item. Even if the fire I see is entirely hallucinated, it remains true that I seem to see a fire. This is the watershed of modern philosophy: it slices the contents of consciousness away from the way the world is, and thereby creates an inner realm that is independent and absolutely reliable. The gain is a foundation. The cost is a wall—I have direct access only to my own ideas, and everything outside has to be reached by proof. Three hundred years of epistemology were spent pacing in front of that wall.
The wax passage in the Second Meditation is the best concrete reasoning in the book. Fresh from the comb, the wax has the taste of honey, the scent of flowers, a visible colour and shape; it is hard, cool, and makes a sound when rapped. Hold it to the fire: the taste goes, the scent disperses, the colour changes, the shape collapses, it softens, heats, and no longer sounds. Yet everyone will say: it is the same wax.
So what do I know it by? Not the senses—everything they delivered has just changed. Not imagination either (the power of picturing): the shapes this wax could take are infinite, and imagination can run through only a few. Only the intellect is left. The wax is nothing but "something extended, flexible and changeable"—and that is grasped by an inspection of the mind, not by the eyes.
Two enormous consequences follow. First, real knowledge of bodies is mathematical and intellectual: sweetness and scent are subjective add-ons, and the body itself retains only extension and motion. That single move clears the ground for two centuries of mechanistic physics—if a body is just "an extended thing," then of course geometry and mechanics can handle it. Second, I know my mind more securely than I know any body. Descartes adds a lovely observation here: looking out of the window at people crossing the square, I actually see only hats and coats—underneath there could be automata—and it is my judgement that supplies "those are men." If even seeing a lump of wax involves a judgement, then every perception is also evidence that I am perceiving. The conclusion is counter-intuitive but sound: the body is the last thing I confirm; the mind is the first.
Why is "I am" trustworthy? Because, Descartes says on looking back, my perception of it is clear and distinct—so he promotes that into a general criterion of truth. The two words are not synonyms. Clear means present and open to an attentive mind, nothing vague about it. Distinct means so sharply cut off from everything else that it contains nothing but what is clear. A toothache is clear (nobody doubts they are in pain) but not distinct—people take the pain to be in the tooth, when pain is a state of the mind.
The trouble arrives at once: why should "it seems clear and distinct to me" amount to "it is true"? What if the demon deceives me precisely where I am clearest? Descartes sees this exactly, which is why he must first prove a God who does not deceive before the ruler can be used. From here the book is not really about God; it is about finding collateral for reason. God's job in these pages is to make "what I see clearly is true" hold—and thereby make physics possible.
The Third Meditation runs on a scholastic principle: an effect cannot contain more reality than its cause—nothing comes from nothing, and more cannot come from less. He then draws a distinction that is unreadable unless it is spelled out. Considered as a mental event, every idea is equally real; that is its formal reality. But considered as representing something, ideas come in grades, and that grade is called objective reality—where "objective" is the old scholastic usage meaning "as an object presented to thought," which is nearly the reverse of what the word means today. The idea of a stone has the objective reality of a finite substance; the idea of God, that of an infinite one.
Then the strike: could I—finite, fallible, doubting—have manufactured the idea of an infinitely perfect being? Descartes says no. I recognise myself as finite only by contrast with the infinite; the whole comes first, and only then do I see myself as a gap in it. So the idea must have a cause at least as real as what it represents, namely God—"the mark the craftsman has set on his work."
The Fifth Meditation adds an ontological argument: existence belongs to the essence of a supremely perfect being as having angles summing to two right angles belongs to the essence of a triangle; a supremely perfect being that lacked existence would be a contradiction. Kant later cut this argument in half: existence is not a predicate. His point is homely and devastating—the concept of a hundred thalers and a hundred actual thalers in my pocket have identical conceptual content; "exists" adds nothing to a concept, so it cannot be "contained" in an essence the way ordinary properties are. Almost nobody accepts either proof today. But see what Descartes was really after: not a religious conclusion but an axiom that the universe is not a cheat—that the world can be read by reason. Modern science still assumes that axiom; it has only stopped using God's name for it.
Antoine Arnauld, the theologian behind the Fourth Objections, hit the sore point on the spot, and the blow has never quite been parried: you prove God by clear and distinct perception, then guarantee clear and distinct perception by God. Is that not a circle?
Descartes replies with a distinction. While you are actually perceiving something clearly and distinctly, you simply cannot doubt it—stare at 2+3=5 and doubt is not available to you. What needs underwriting is memory: propositions I once proved but now merely recall need "God is no deceiver" to hold them up, or the demon could have tampered with them the moment I turned away. So, he argues, there is no circle: every step in the proof of God is a present intuition, and what God secures is the chain of remembered results. Most readers judge that this eases the problem without dissolving it—because proving God is itself a sequence of steps, and by step three, step one is already a memory.
The Fourth Meditation handles an awkward question: if God does not deceive, why do I get things wrong? The answer is elegant and entirely usable today. A judgement is made of two parts: the intellect supplies the content, and the will says yes or no. The intellect is finite—only so much is ever in clear view—while the will is unlimited: you can assent to anything at any time. Error happens where the will oversteps: you nod at what the intellect has not yet made out.
So error is not God's fault (both faculties are good in themselves) and not the fault of a small intellect; it is a misuse of freedom. From which follows a hard rule of intellectual conduct: withhold assent until you see clearly and distinctly. That is not a slogan—it is the metaphysical ancestor of "no conclusion without sufficient evidence." Descartes turns epistemology into an ethic: believing something on inadequate evidence is not a misfortune but a fault.
The Sixth Meditation closes the net. I can clearly and distinctly conceive myself as a thinking, unextended thing, and body as an extended, unthinking thing; and whatever God can let me conceive apart in this way, he can make exist apart. So mind and body are two genuinely different substances: res cogitans and res extensa. He adds an argument from properties: bodies are always divisible, minds are not—you cannot conceive of half a mind; in willing, sensing and understanding, it is always the whole of me.
| res cogitans (thinking) | res extensa (extended) | |
| Essential attribute | Thought (incl. willing, sensing) | Extension (occupying space) |
| Divisibility | Indivisible, wholly present | Infinitely divisible |
| How it is known | Directly; certain first | By inference; certain last |
| What it bequeathed | The inner world; the problem of consciousness | A mathematisable nature; mechanistic physics |
One cut: it set physics free and manufactured three centuries of the mind-body problem.
Descartes is far subtler than the label later pinned on him. He explicitly denies that the mind is merely lodged in the body "as a pilot in a ship." Were that the relation, an injury would be something I noticed intellectually, the way a pilot notices a hole in the hull; in fact I feel pain, hunger and thirst directly—which shows that mind and body are "very closely joined and, as it were, intermingled." He even discusses how the system misfires: a man with dropsy feels thirst although drinking will harm him. The office of sensation is not to report the truth about the world but to report what is good for this body—so it can give false signals. The thought that perception is survival-oriented rather than truth-oriented is startlingly modern.
What he could not answer is the fatal question. In 1643 Princess Elisabeth of Bohemia wrote to ask it directly: how can something unextended, with no surface to make contact, move a mass of flesh? Transmitting motion requires contact and push, and the mind by definition has neither. Descartes tried several answers and ended close to an admission: the union of mind and body is not to be understood by pure intellect, only felt in ordinary living. In The Passions of the Soul he located the interface at the pineal gland (one of the few unpaired organs in the brain)—which names a place, not a mechanism. Much of the rest of modern philosophy is housekeeping after this mess: Spinoza declared mind and matter two aspects of one substance, Leibniz likened them to two clocks synchronised in advance, Malebranche had God intervene at every apparent interaction—all of them detours around an impossible seam.
The six Meditations are a single demolish-then-repurchase movement, and Descartes insists he follows the order of reasons, not the order of topics—so the sequence cannot be shuffled; each step stands only on the one before:
In a line: what the book establishes is neither that God exists nor that I exist, but that human reason is entitled to decide on its own what is true. It doubts reason all the way down, then lets the doubting itself issue reason a licence.
Read only this book and you will picture a metaphysician who spent his life doubting—badly out of proportion. He drew a truer self-portrait in the preface to the French Principles: philosophy is a tree whose roots are metaphysics, whose trunk is physics, and whose branches are medicine, mechanics and morals. The Meditations is only the root; the fruit he actually wanted grows on the branches.
The Discourse on Method (1637) gives four rules that sit closer to how he really worked: accept only what is clear, distinct and beyond doubt; divide each difficulty into as many parts as it will take; proceed in order from the simplest objects to the complex; and enumerate and review so completely that nothing is left out. Modern problem-solving—decompose, then build up from the simple—is a direct descendant of those four lines. The book's three appendices were the real event at the time: the Geometry welded algebra to geometry by handling curves with equations (analytic geometry), and the Optics stated the law of refraction now called the Snell–Descartes law.
The suppressed World. He had already written a cosmology incorporating heliocentrism when word reached him of Galileo's condemnation in 1633; he withdrew it at once. That episode explains why God stands so conspicuously in the Meditations: Descartes needed the philosophical collateral and the political amulet alike.
The Principles of Philosophy (1644) is where the ambition really lies—a physics that explains the universe purely as extended matter in motion. He states that a body left alone continues in a straight line at constant speed (an early formulation of inertia), holds that the "quantity of motion" is conserved (Leibniz later showed he had picked the wrong quantity, though the idea that something is conserved opened the road to later conservation laws), and explains planetary motion by vortices in a subtle medium—a cosmology that was for a time Newtonian gravitation's chief rival before being discarded entirely. His physics was thoroughly wrong; his programme for physics—nature is matter and motion, and can be handled mathematically—won thoroughly.
The Passions of the Soul (1649), his last book, is also his most underrated. He treats emotion as a psycho-physiological mechanism, sorts it into six primitive passions—wonder, love, hatred, desire, joy, sadness—and names the highest virtue générosité: recognising that the only thing truly one's own is the free disposition of one's will, and therefore neither despising nor inflating oneself, and dealing generously with others. Add that face back and you get the whole man: he never wanted people to become emotionless reasoning engines, only to know which things are theirs to govern. On that point he stands nearer to the Stoics than to the cartoon posterity drew of him.
1. To find certain knowledge, tip the whole basket out: whatever admits a sliver of doubt counts as false for now, and goes back only after inspection.
2. Doubt comes in three stages: the senses deceive; waking and dreaming have no reliable boundary; and an omnipotent deceiver tampering with your arithmetic cannot be ruled out. By stage three, world, body and mathematics are all gone.
3. But the stronger the deceiver, the clearer the hole: to deceive me there must be a me. "I am, I exist—necessarily true whenever I put it forward." (That is the actual line; "I think, therefore I am" is from two other books—and it is an intuition, not an inference.)
4. This "I" is first of all a thinking thing—doubting, willing, imagining and sensing all count. Even if the fire is hallucinated, "I seem to see a fire" is true. The contents of consciousness part company with the way the world is, and the inner world is born.
5. Melt the wax and every sensory property changes, yet you still call it the same wax—so real knowledge of bodies is intellectual, not sensory; and since every seeing is also evidence that I am seeing, the mind is established earlier and more firmly than the body.
6. The ruler of truth is clear and distinct perception—but the ruler itself needs underwriting, which is why God appears. God's job in this book is to make "what I see clearly is true" hold, and so to make science possible.
7. Both proofs of God have since failed (the ontological one died of Kant's "existence is not a predicate"), and Arnauld's question still hangs: clear perception proves God, God secures clear perception, and the circle was never fully escaped.
8. Error comes neither from God nor from a small intellect but from a will larger than the intellect—nodding where you have not yet seen. Hence a discipline still worth keeping: no clear and distinct perception, no assent.
9. Mind is unextended and indivisible, matter extended and endlessly divisible. That cut let physics treat nature mathematically and left the seam nobody has closed: how something unextended moves flesh (Elisabeth of Bohemia asked in 1643; the pineal gland is a location, not a mechanism).
10. What the book really establishes is worth remembering: not that God exists, but that reason is entitled to judge on its own what is true. As for whether "thinking is occurring" yields "a persisting I"—from Lichtenberg's "it thinks" through Hume to Nietzsche, the answer has drifted steadily towards no.