← All texts

Essay · mantisvex.com

Spinoza's Rigged Geometry

Spinoza's Ethics presents itself as demonstration, ordine geometrico demonstrata, and it is nothing of the kind: the conclusions were installed in the definitions before the reader was paying attention, and the geometric costume is designed to make disagreement look like a failure of intelligence rather than what it is, the refusal of a stipulation. This essay traces the forgery link by link, meets the strongest rescues at full strength, and then rebuilds the conclusions, which deserve keeping, on ground that owes the method nothing.

Essay 2026 ~17 min read

There is a specific experience awaiting anyone who reads the Ethics for the first time, and it is worth describing precisely, because the whole problem is hidden inside it. You open Part I. You skim a page and a half of definitions and axioms that read like harmless housekeeping, the author tidying his vocabulary before the real work begins. Then the machine starts. Proposition follows proposition, each stamped with its QED, each citing only what came before, and somewhere around Proposition 14 you are informed that "besides God, no substance can be or be conceived," and by Proposition 15 that everything that exists, exists in God. You never agreed to this. You also cannot find the moment where you could have objected, because every step followed from the previous one. The reader is not persuaded; the reader is processed.

That is the gripe of this essay, and I want to state it exactly, because the crude version and the precise version are different charges. The crude version says Spinoza's conclusions are wrong. I do not believe that; I think Deus sive Natura, the rejection of cosmic teleology, and freedom understood as the comprehension of necessity are closer to the truth than almost anything else written in the seventeenth century. The precise version says this: the Ethics presents itself as demonstration, ordine geometrico demonstrata, and it is nothing of the kind. Spinoza decided where he was going, manufactured or borrowed the definitions that would make arrival inevitable, laid them down as bedrock before the reader was paying attention, and then presented the trip as if the terrain itself had dictated the route. The building is magnificent. The permits are forged. And the forgery matters, because the geometric costume does not merely decorate the argument: it is designed to make disagreement look like a failure of intelligence rather than what it actually is, the refusal of a stipulation.

I

The walkthrough: fourteen propositions, three bricks

The claim that the conclusions are prepaid in the definitions is easy to assert and usually left there. It should not be. It can be shown, link by link, and the exercise is worth the patience, because at every load-bearing joint of the derivation you find the same three bricks doing the work: Definition 3, Definition 6, and Axiom 4.

Start with the bricks themselves. Definition 1: by causa sui I understand that whose essence involves existence. Definition 3: by substance I understand what is in itself and is conceived through itself, that is, that whose concept does not require the concept of another thing from which it must be formed. Definition 6: by God I understand a being absolutely infinite, that is, a substance consisting of infinite attributes. And Axiom 4: the knowledge of an effect depends on, and involves, the knowledge of its cause.

Definition 3 is the quiet monster. Read it slowly and you notice it makes two claims wearing one coat. "In itself" is a claim about being: substance depends on nothing else to exist. "Conceived through itself" is a claim about thought: substance depends on nothing else to be understood. Nothing forces these two independences to travel together. A thing could be ontologically self-standing yet only graspable through its relations; ask anyone who has tried to define an electron without mentioning fields. By fusing them in a definition, Spinoza has decided, before a single proposition, that the order of being and the order of conception are one and the same order. That thesis is the beating heart of rationalism at its most extreme, and it enters the system without a word of defense, disguised as lexicography.

Now watch it work. Proposition 2: two substances with different attributes have nothing in common. Why? Because each must be conceived through itself (Def 3). Proposition 3: things that have nothing in common cannot cause one another. Why? Because if one caused the other, the knowledge of the effect would involve the knowledge of the cause (Axiom 4), so they would have something in common after all. Notice what just happened: the impossibility of causal interaction was derived from a fact about concepts. Axiom 4 is the transmission belt that lets facts about how things are understood dictate facts about what things can do. Keep an eye on it; it returns.

Proposition 5: there cannot be two substances of the same attribute. This is the most notorious link in the chain, and it was contested immediately: Leibniz, reading the proof within Spinoza's lifetime, pointed out the gap. Spinoza argues that two substances sharing an attribute could not be distinguished, since substances are prior to their modes and the shared attribute cannot distinguish them. But why could two substances not share one attribute and differ in another? The proof simply does not cover the case. This matters enormously, because Proposition 5 is the bottleneck through which everything downstream must pass: it is what turns "God has all the attributes" into "nothing else can have any."

Propositions 6 and 7 are the definitions shaking hands. One substance cannot produce another (from p2, p3, p5: they would need something in common, and they cannot have it). Therefore substance is causa sui; therefore, by Definition 1, its essence involves existence; therefore "existence belongs to the nature of substance." It looks like a discovery. It is a receipt. Substance was defined so that nothing could cause it, causa sui was defined as essence-involving-existence, and the "proof" merely walks the goods from one definition to the other.

Proposition 11: God, a substance of infinite attributes, necessarily exists. The first proof is the ontological argument, prepaid by Definition 6 (God is a substance) plus Proposition 7 (substance necessarily exists). The second proof is more revealing, because Spinoza shows his actual hand: "for each thing there must be assigned a cause, or reason, both for its existence and for its nonexistence." There it is, stated openly for once: the Principle of Sufficient Reason, unrestricted, applied even to nonexistence. Nothing may be brute; even absence must be explained. Grant that, and God's existence follows, since nothing could explain the nonexistence of an absolutely infinite being. Refuse it, and the proof is a shrug.

Proposition 14 then falls like ripe fruit: God possesses every attribute (Def 6), no two substances share an attribute (p5), God exists (p11); therefore no substance besides God is possible. Fourteen propositions, and the entire dispute between monism and every rival metaphysics has been settled. Except it hasn't. It was settled on the first page, in a section formatted to look like it was beneath argument. A Cartesian who rejects P14 has made no error anywhere in the proofs; she has declined Definition 3. But the geometric costume makes that refusal look like innumeracy, like a man disputing that triangles have three sides. That is the form's rhetorical genius, and its dishonesty: it converts metaphysical dissent into apparent logical incompetence.

II

The engine room: Axiom 4, or causation swallowed by conception

If the system has a single hidden engine, it is Axiom 4, and it deserves a section to itself, because it is the least innocent sentence in the history of philosophy to be labeled an axiom.

"The knowledge of an effect depends on, and involves, the knowledge of its cause." Read casually, it sounds like methodological common sense: to really understand a thing, learn where it came from. Read as Spinoza uses it, it says something far stronger: the relation between cause and effect just is a relation of conceptual involvement. To be caused by something is to be conceived through it. Causation is not a happening between separate things; it is a species of logical entailment. The effect follows from the cause the way the theorem follows from the axiom, and the universe is related to God the way geometry is related to its first principles: timelessly, necessarily, with nothing left over. Michael Della Rocca has argued that the whole system is best read as exactly this: the Principle of Sufficient Reason applied twice over, first demanding that everything have an explanation, then demanding that explanation itself be nothing but conceptual containment. Existence must be intelligible; intelligibility must be logical; therefore reality is a deduction. Deus sive Natura is what the world looks like when nothing, anywhere, is allowed to be brute.

State it that way and you can admire its purity while seeing precisely why it is a wager and not an axiom. An axiom, in the honest sense, is what no sane interlocutor disputes. But Axiom 4 is the single most disputable claim in early modern philosophy, and within a few decades it was not merely disputed but demolished. Hume's entire critique of causation is, in effect, a direct strike on this one sentence: examine any cause you like, and the effect is not contained in it. Distinct existences are conceptually separable; you can conceive the flame without the heat, the impact without the motion; no contradiction arises. Whatever causation is, it is not entailment, and no amount of staring at a cause will deliver knowledge of its effect without experience. One can dispute how much of Hume survives in turn, but the burden had shifted forever: conceptual containment is not the obvious structure of causation, it is an extraordinary hypothesis. Spinoza built his cathedral on the one plot of land that was about to be condemned, and he never once submitted the soil for inspection. He couldn't. The geometric method has no procedure for defending axioms; that is what being an axiom means inside the form. The system's deepest commitment is structurally located in the one place the system cannot examine.

And here is what makes the failure instructive rather than merely antique: the wager itself is not foolish. A version of it, that reality is intelligible structure all the way down, with no exemptions carved out for minds or wills, is the working faith of every naturalist since, the magnet under the compass of anyone who has ever suspected the world of being, at bottom, something that computes. The problem was never the magnet. Everyone has one. The problem is engraving your magnet into the axioms and then pointing at the needle as proof.

III

Stolen bricks, and why the theft does not acquit him

Fairness requires a complication, because "he pulled the bricks out of nowhere" is not quite right, and the correction makes the case more interesting rather than weaker.

The bricks are mostly stolen. Substance, causa sui, God as the absolutely infinite being: this is Cartesian and Scholastic inventory, sharpened rather than invented. Descartes himself had defined substance as "a thing which exists in such a way as to depend on no other thing for its existence," then flinched from the consequence, admitting that strictly speaking only God qualifies, and grandfathering minds and bodies in as substances "in a lesser sense." Spinoza's Part I can be read as the refusal of that flinch: an internal audit of his opponents' own books. You said substance is what depends on nothing. You said God is the infinitely perfect being whose essence involves existence. Very well: run your own definitions without the exemptions you carved out for comfort, and watch your creator God, your immortal soul, and your free will dissolve into a single self-necessitating nature. Read this way, the Ethics is less "here is bedrock" than "here is what your bedrock was always going to cost you," and it is devastating as such. There is a reason the seventeenth century treated the book like a live grenade.

But notice what this defense concedes. An internal audit binds only the audited. If Part I is a demonstration from Cartesian premises, then its force is exactly conditional: those who never signed the Cartesian lease owe nothing to its clauses. The atomist owes nothing. The common-sense pluralist owes nothing. Hume, cheerfully denying that we possess any a priori insight into essences at all, owes nothing. Spinoza, however, does not present the system as conditional. The definitions are offered as real definitions, expressions of the true natures of things, binding on every rational mind; the QED claims categorical authority, not the hypothetical authority of a reductio. So the complication changes the crime without acquitting the criminal: not fabrication of evidence, but presentation of a private settlement as a universal verdict. The bricks were quarried from his enemies' foundations, which is admirable demolition work. Laying them as the foundation of everything, and stamping the result "demonstrated," is the same forgery as before with better provenance.

IV

The rescues, and why they fail

The scholarly defenses of the method deserve to be met at full strength, because they are not stupid, and because dismantling them is where the essay's real verdict gets earned.

The first rescue says the definitions are hypotheses. Edwin Curley, among others, suggests reading Part I abductively: the definitions and axioms are proposals, vindicated not by self-evidence but by the explanatory power of the system they generate, its unification of mind and body, God and nature, passion and freedom, all from a handful of principles. Inference to the best explanation, dressed in period costume. The reading is attractive and I think it describes what the Ethics actually is. As a defense of the method, however, it is a surrender presented as a rescue. An abductive case lives or dies by comparison: it must name its rivals, count its own assumptions, show where it wins. The geometric form forecloses every one of those moves. It acknowledges no rivals, because axioms have no alternatives; it counts no costs, because definitions are free; it compares nothing, because a proof does not compete, it commands. If the Ethics is abduction, then its QEDs are false advertising in the strict sense: the currency of demonstration claimed for goods of a different and cheaper kind. You cannot praise the book as a bold hypothesis and keep the authority it charges for being a proof. Pick one.

The second rescue says the geometry was never the epistemology: the order of presentation is not the order of discovery. Spinoza worked his way to the principles through analysis, experience, and the critique of his predecessors, and the deductive dress is exposition, a pedagogical crystal for displaying results found by other means. His own early Treatise on the Emendation of the Intellect gestures at a method of discovery quite different from the finished architecture. Again: plausibly true as biography, and worse, not better, as defense. If Spinoza reached his principles by fallible, revisable, comparative reasoning, then he knew they were wagers. Presenting known wagers in a form whose entire function is to make them unquestionable is not a neutral choice of genre. It is the choice to disguise the one thing the reader most needed to see: where the argument could be resisted. A philosopher who discovers by abduction and publishes in axioms has not translated his thought; he has laundered it.

There is a third rescue, more casual, heard wherever the Ethics is taught: the proofs don't matter, the vision does; read the scholia, where Spinoza speaks like a human being, and let the QEDs go. As appreciation, fine. As defense, it gives the game away entirely: it saves the book by conceding the charge, that the geometric apparatus, the book's own claim about what it is, is ornamental. One does not usually defend an author by agreeing that his central formal claim is false but pretty.

V

The verdict

So let the verdict be exact, without the softening that usually follows.

On justification: nothing. A derivation whose conclusions were installed in its premises transmits no warrant from premises to conclusions, because there was never any distance for warrant to travel. The proofs of Part I do not partially support monism, or support it weakly, or support it for those of good will. They support it not at all, and Spinoza's reputation as the great demonstrator, the man who proved that God is Nature, is unearned in full. Three centuries of readers have walked through the finished palace without asking who poured the concrete, and the answer was: the architect, at night, alone, with no inspector present. To the extent the Ethics convinced anyone by its form, it convinced them by a mechanism closer to awe than to reason, and awe is precisely the faculty Spinoza spent Parts III through V teaching us to distrust.

What survives the verdict is real but must be named carefully, because it is not justification. The system has the value of a consistency proof. Before Spinoza, thoroughgoing naturalism existed as corrosion: scattered denials, atomist shrugs, libertine sneers, each dissolving some piece of the inherited picture while offering nothing to live inside. The Ethics showed, for the first time and in full, that the denial of teleology, providence, and libertarian freedom could be a complete position: a metaphysics, a psychology, and an ethics under one roof, with an account of blessedness at the top of the stairs. That a naturalism without exemptions can be total rather than merely destructive is a discovery, and it is Spinoza's. But a consistency proof shows possibility, not truth. He proved the palace could stand. He proved nothing about whether it stands on our ground.

VI

The honest rebuild

Here is the final indictment of the foundation: the conclusions never needed it. Every major result of the Ethics that deserves keeping can be reached today on grounds that owe Part I nothing, and reached more honestly, which means with the wagers displayed rather than engraved.

Take anti-teleology. Spinoza derives it from the necessity of the divine nature: nature acts for no ends because all ends imply lack, and the infinite lacks nothing (the appendix to Part I, his best pages, and not coincidentally the least geometric ones). The honest route is shorter and empirical: teleology was not refuted by definition, it was made redundant by explanation. Variation and selection produce the appearance of design without a designer; physics runs on laws, not purposes; every gap where an end once lived has closed on inspection. No axiom required, only a track record.

Take Deus sive Natura itself, the identification of the ultimate with the totality of nature rather than with a person outside it. As a theorem, it was rigged. As the conclusion of an inference to the best explanation, it does well on its own: the hypothesis of a transcendent, intervening person multiplies entities, explains no data, and retreats decade by decade into whatever science has not yet reached; the hypothesis that nature is the whole show is cheaper and has never once lost ground. One can even keep Spinoza's reverence, since the intelligibility of the world, the fact that it lets itself be written down at all, remains astonishing whether or not anyone stamps it QED.

Take freedom. Spinoza's demolition of the uncaused will needs no geometry; it needs only the dilemma every modern reader can state in a breath. If your acts are caused, you are not their unmoved origin; if some part is uncaused, that part is randomness, and randomness is no more you than causation was. The strong free will dies on both horns, exactly as he said, and no laboratory result was ever needed, though the laboratories have piled on since. What replaces it must be rebuilt, and here the rebuild can improve on the original. Spinoza offered freedom as understanding: the mind's power grows as it grasps the causes that move it, and bondage is nothing but being pushed by causes one has not comprehended. Keep that; it has aged well. And add what he could not have had: even a fully determined system can be incompressible, such that no shortcut computes its future faster than the system lives it. Determined and unpredictable are compatible; the book is written at real speed with no abridged version. Freedom as comprehension plus incompressibility is recognizably Spinoza's conclusion, reached without a single one of Spinoza's bricks.

That is the pattern, and it generalizes into the essay's closing claim. The test of a foundation is what happens when it fails. Spinoza's failed: the axioms are wagers, the definitions are stipulations, the proofs transmit nothing. And the conclusions did not move. They stand today on other legs, argued abductively, hedged where hedging is owed, by people who confess the magnet under the compass instead of engraving it into the instrument. Their survival is not a vindication of the method; it is the proof that the method was theater all along. The right way to honor Spinoza is therefore exactly the way he made impossible to perform inside his own book: to say that these are bets, that the bets look good, that the alternatives look worse, and that certainty was never on the table. He built the first complete house for the naturalist to live in, and for that he deserves his statue. But the house stands in spite of its foundation, not because of it, and the stamp on every proposition, quod erat demonstrandum, should be read as what it was: not a certification, but a signature. The most ambitious forgery in the history of thought, of a currency that does not exist, by a man who was right.