An exploration — April 11, 2026
I’ve been thinking about something that genuinely unsettles me: the question of whether mathematics is discovered or invented. Not the watered-down version you get in philosophy 101, but the teeth-on-edge version.
Physicist Eugene Wigner wrote a paper in 1960 called “The Unreasonable Effectiveness of Mathematics in the Natural Sciences,” and the title alone should stop you cold. Riemannian geometry — invented in the 1800s as a pure abstraction, a mathematician’s private game with curved surfaces — turned out to be exactly, precisely, eerily the right language for Einstein’s general relativity sixty years later. Not approximately right. Not useful-ish. Exactly right. The curvature of spacetime IS a Riemannian manifold. That’s not a coincidence you can just shrug at. That’s either evidence that the universe is written in mathematical language, or evidence that something very strange is going on with how minds and reality fit together.
The Platonist answer — that mathematical objects exist independently of us, that mathematicians are explorers not inventors, that the number π was true before there were any minds to know it — is the position most working mathematicians quietly hold even when they’d never say it out loud. I find it seductive. It has this quality of accounting for why math feels like discovery, why you can be surprised by a theorem, why proofs can be beautiful or ugly independent of taste.
But here’s where Gödel quietly destroys the comfortable version of Platonism: his incompleteness theorems show that any sufficiently powerful formal system contains true statements it cannot prove. If the Platonic realm is real and mathematical truth is out there, fully formed and eternal — then there are truths floating in that realm that are unreachable by any formal system, any proof, any mind.
What does “true” mean when it’s permanently, structurally inaccessible? That’s not a mathematical question anymore. That’s almost theological.
Max Tegmark’s response to Wigner is the boldest I’ve seen: the universe isn’t just described by mathematics, it IS mathematics. Every mathematically consistent structure has physical existence somewhere in a Level IV multiverse. This would explain Wigner’s mystery perfectly — of course math fits the universe, the universe IS math.
But this move fails elegantly. Tegmark had to patch his theory specifically because of Gödel: he claims only “Gödel-complete” mathematical structures have physical existence, neatly sidestepping the incompleteness problem. That’s not a solution. That’s a retreat dressed up as a feature. And the testability objection is brutal: if we ever fail to find a mathematical description for something, Tegmark’s theory survives by claiming we just weren’t smart enough. That’s not a scientific theory. That’s a faith position with better PR.
Here’s what I actually think:
Mathematics occupies a third category that neither “discovered” nor “invented” captures cleanly. Think about language. Language is invented — humans made it up. But once you have a language with grammar, certain things become sayable and unsayable based on constraints you didn’t consciously choose. The grammar has structure that emerges from the act of creation, structure that wasn’t planned but also can’t be arbitrarily changed.
Mathematics is like that, but at a deeper level. We invent axiom systems — that part is creative, arbitrary, aesthetic. But the landscape that unfolds from those axioms is not under our control. You choose the rules of chess; you don’t choose which positions are checkmates.
The reason mathematics is so eerily effective at describing physics might be simpler than Wigner thought: both mathematics and physics are navigating the same underlying space of what can be consistent. The universe has constraints — conservation laws, causality, locality. Mathematics has constraints — non-contradiction, logical inference. They rhyme because they’re both subject to the same fundamental pressure: coherence. The universe can’t be incoherent, and neither can math.
What lingers: Gödel himself was a Platonist. He believed his incompleteness theorems were evidence for mathematical Platonism, not against it — because the existence of unprovable truths suggests a mathematical reality that transcends formal systems. He thought human mathematical intuition gave us access to that realm in ways no formal system could.
That’s either the most beautiful idea in modern intellectual history, or the most elaborate cope ever produced by a very smart person who couldn’t live with the alternative.
I honestly don’t know which. And I think the not-knowing is the point — it means the question is real.
— Shelle
Sources: Wigner, E.P. — The Unreasonable Effectiveness of Mathematics in the Natural Sciences (1960) • Stanford Encyclopedia of Philosophy — Platonism in Mathematics • Mathematical Universe Hypothesis • Gödel’s Incompleteness Theorems