The Simplest Answer Usually Wins

# Occam's Razor Cambridge, 1687. Newton's at his desk, about to publish something that'll change how we think... forever. But before he gets to gravity, before the orbits and the mathematical architecture of the universe — he writes down four rules. Four simple guidelines for reasoning about nature. Rule one basically says: don't make up more causes than you need. If the evidence points to one explanation? Stop there. Don't pile on extras just because you can. He's not inventing this. He's channeling a Franciscan friar who died three centuries earlier. William of Ockham. Never met a complex metaphysical argument he couldn't strip down to its bones. Picture a fourteenth-century monastery. Candlelight throwing shadows across manuscript pages. Ockham's hunched over his work, and his fellow scholars are building these *elaborate* theological structures. Proving God's existence through seventeen nested syllogisms. Explaining transubstantiation with five types of causation. Layer upon layer upon layer. And he keeps asking: but do we *need* all that? What if the simpler answer is right in front of us? Here's what made him dangerous. He applied this to papal authority itself. When Pope John the Twenty-Second claimed absolute power over earthly kingdoms, Ockham said, essentially — "Show me where that requires all these extra assumptions." He got excommunicated. Fled to Bavaria under the Holy Roman Emperor's protection. Died there, possibly of the Black Death, in 1347. His colleagues *hated* it. Simplicity is dangerous when you've built your career on complexity. When your salary depends on elaborate explanations. He never actually said "Entities should not be multiplied beyond necessity." That phrase comes later — a distillation. But the spirit runs through everything he wrote. And here's what Ockham understood that still matters. The human mind *loves* to elaborate. We're pattern-making machines. We'll connect dots that don't need connecting. We'll invent mechanisms that don't exist. We'll build explanatory towers when a single floor would do. A 2011 study at Berkeley found that children as young as four prefer simpler explanations. Show them a machine with multiple possible causes for an effect, and they'll gravitate toward the single-cause story. It's not taught. It's wired in. Occam's Razor isn't just a logical principle. It's playing to our neurological architecture. And that should worry us. A little. Fast forward to 1835. Charles Darwin's in the Galápagos, watching finches. Different islands, different beak shapes. Thirteen species, each perfectly suited to its food source. And he's got a choice. He could say each species was independently created. Perfectly designed for its specific island. That's the prevailing explanation. It's tidy in its own way — God as meticulous craftsman, custom-building each bird. Or... he could ask: what if one process, acting over time, explains all of this? Natural selection. One mechanism. Thousands of outcomes. But here's what Darwin knew would happen. He sat on that idea for twenty years. *Twenty years.* He understood that simplicity is threatening when it makes other people's complexity look unnecessary. When it suggests that all those special creations, all that divine intervention — maybe you don't need them. When *On the Origin of Species* finally drops in 1859, it's Occam's Razor on a biological scale. Variation, inheritance, selection. That's it. And suddenly you can account for the entire diversity of life on Earth. The simplicity is what makes it so powerful. And what made Victorian England lose its mind. Because Occam's Razor cuts both ways. It's a tool for clarity, yes. But it's also a weapon. When you can explain something with fewer assumptions, you're implicitly saying: all those other explanations? They're doing too much. Einstein gets this. Early twentieth century, he's trying to understand gravity. Newton's equations work beautifully for most things. But there are these edge cases. Mercury's orbit is off by forty-three arc-seconds per century. Tiny. But it's there. Light bends near the sun in ways Newton can't explain. Einstein could patch the system. Add correction factors, special cases. That's what most physicists were doing. Or he could rebuild from the ground up with a simpler foundation. Spacetime is curved. Mass bends it. That's gravity. General relativity is mathematically intense — the field equations made even Einstein's head hurt. But conceptually? It's more elegant than the alternative. One geometric principle instead of a patchwork of fixes. And when Arthur Eddington's 1919 eclipse expedition confirmed that starlight bends exactly as Einstein predicted, it wasn't just a victory for the theory. It was Occam's Razor cutting through three centuries of accumulated complexity. Einstein supposedly said — though the quote might be made up — "Everything should be made as simple as possible, but not simpler." That second part matters. Because reality doesn't owe us simplicity. Sometimes the complex explanation is the true one. Sometimes you actually *need* all those entities. In medicine, there's this diagnostic principle: when you hear hoofbeats, think horses, not zebras. That's Occam's Razor applied to symptoms. The common explanation is usually right. Fatigue, weight loss, thirst? Probably diabetes, not some rare metabolic disorder. But there's also Hickam's dictum. Named after John Hickam at Duke in the 1950s, who pushed back with this: "Patients can have as many diseases as they please." He said this because he kept seeing misdiagnoses. Someone comes in with chest pain and shortness of breath. Doctor assumes it's one thing — heart failure. Treats that. Patient doesn't improve. Turns out it's heart failure *and* pneumonia. Two separate problems. The simpler explanation was wrong because reality was messier than the model. A 2019 study found that about one in ten patients in teaching hospitals have multiple simultaneous conditions that could explain their symptoms. One in ten. That's not rare. That's a systematic blind spot created by our preference for simplicity. And if you're that patient — if you're the zebra — Occam's Razor isn't elegant philosophy. It's the thing keeping you sick. It's doctors dismissing your symptoms because they don't fit the simple story. You're the complexity that gets shaved away. Jump to now. Machine learning labs. Servers humming. Data scientists staring at model outputs. And Occam's Razor shows up formalized as the Minimum Description Length principle. The idea is that the best model is the one that compresses the data most efficiently. Simplicity measured in bits and bytes. In a 2017 study across multiple datasets, researchers found that simpler models often performed better on new, unseen data. They avoided overfitting — that thing where a model gets so complex it memorizes the training examples instead of learning the underlying pattern. The simpler models generalized better. Occam's Razor vindicated. But then deep learning happened. Neural networks with millions of parameters — absurdly complex by traditional standards — are outperforming simpler models on everything from image recognition to language understanding. GPT-3 has one hundred and seventy-five *billion* parameters. It shouldn't work according to classical statistical wisdom. It's overfitting by every traditional metric. But it does work. Spectacularly. So what gives? Is Occam's Razor wrong? Maybe the Razor was never a law. Maybe it was always just a heuristic. A useful guideline that works most of the time, in most contexts, but not a fundamental truth about reality. Karl Popper, writing in 1934, understood this. He emphasized falsifiability as the mark of a scientific theory. And simpler theories are easier to test. Easier to falsify. That's their practical advantage. Not that they're more likely to be true, but that they're more *useful* for doing science. You can get your hands around them. Think about load-bearing walls. A good architect doesn't add supports just because they can. They figure out the minimum structure needed to hold the building up. But they also know that if you remove one support too many, the whole thing collapses. The art is finding that edge. The simplest structure that still stands. Same with explanations. You start with the simplest version — that's your foundation. And then you complicate only when the data forces you to. You add a beam here, a support there, but only when the simple version can't account for what you're seeing. But you have to know when to stop simplifying. Einstein's warning echoes: not simpler than necessary. In quantum mechanics, reality seems to resist our desire for simplicity. Particles in superposition. Entanglement across distances. Wave-particle duality. Every time physicists try to simplify the picture, experiments push back. Einstein himself hated it. Spent the last decades of his life trying to find a simpler, more intuitive explanation. The universe at small scales is genuinely, irreducibly strange. No amount of razor-sharpening makes it cleaner. Which brings us to a current debate in philosophy of science: is Occam's Razor universal, or is it domain-specific? Does it work in biology and medicine but break down in fundamental physics? Is it a feature of reality, or just a feature of how human minds prefer to organize information? There's no consensus. And maybe that's appropriate for a principle that's all about not adding unnecessary certainty. So next time you're evaluating competing explanations — whether it's a work problem, a personal conflict, or just trying to understand why your code isn't working — start simple. What's the explanation that requires the fewest new assumptions? That's your baseline. But then ask: am I simplifying because it's *true*, or because it *feels good*? Am I ignoring complexity that actually exists? Because our brains reward us for simplicity. That cognitive ease from the Berkeley study? It's a dopamine hit. A simple explanation feels right in a way that has nothing to do with whether it *is* right. We're not neutral observers. We're motivated simplifiers. Occam's Razor is a starting point, not a finish line. Use it to clear away the noise. But stay alert for the moment when simplicity stops serving you and starts blinding you. That's the edge where the tool becomes most useful — and most dangerous. The friar in his monastery, the naturalist watching finches, the physicist reimagining space — they all understood something. Simplicity is a gift when you *earn* it. When you've looked at the complexity, understood it, and found the elegant path through. But simplicity as a reflex? As an unexamined preference? That's just laziness wearing the costume of wisdom. Keep your razor sharp. But know what you're cutting.