Occam's Razor Is Not What You Think It Is

In 2003, a man in rural Pennsylvania started losing weight. His doctor ran blood tests. Thyroid was normal. No diabetes. No cancer markers. The doctor's diagnosis: stress and diet. Eat more, worry less. Simple explanation for a simple symptom.

The man kept losing weight. He went to a second doctor, who noted mild jaundice and ordered an ultrasound. Nothing remarkable. Diagnosis: probably a gallstone. Simple explanation. The man was referred to a gastroenterologist — in three months, when the next appointment was available.

By the time he got the appointment, the jaundice had deepened to the colour of old parchment. A CT scan revealed pancreatic cancer, stage IV, metastasised to the liver. He died eleven weeks later.

Every doctor along the way had applied "Occam's Razor" — or what they believed Occam's Razor to be. The simplest explanation is probably correct. Weight loss? Stress. Jaundice? Gallstone. Each diagnosis was simpler, more common, and more comforting than the alternative. Each diagnosis was also wrong. And the cumulative cost of preferring simplicity over accuracy was a man's life.

Occam's Razor didn't kill him. A misunderstanding of Occam's Razor did. And that misunderstanding is so widespread that it's become one of the most dangerous thinking errors hiding in plain sight.

What Occam Actually Said

William of Ockham was a 14th-century Franciscan friar and logician. He never actually said "the simplest explanation is usually correct." What he wrote, in various forms across his work, was closer to this:

"Entities should not be multiplied beyond necessity."

(Entia non sunt multiplicanda praeter necessitatem.)

Read that carefully. It doesn't say "prefer the simple explanation." It says "don't add unnecessary components to your explanation." The difference is enormous.

Occam wasn't making a claim about which explanation is true. He was making a claim about how to construct explanations efficiently. If two explanations account for all the observed evidence equally well, prefer the one with fewer assumptions — not because it's more likely to be true, but because the extra assumptions are doing no explanatory work. They're dead weight.

The Shaving Metaphor

The "razor" metaphor is instructive. A razor doesn't judge what's true. It shaves off excess. It removes the unnecessary parts of an explanation while leaving the necessary parts intact. If the necessary parts require complexity — if the evidence demands a complicated explanation — the razor doesn't touch them.

Occam's Razor says: don't make your explanation more complex than the evidence requires. It does not say: prefer simple explanations over complex ones. When the evidence is complex, the correct explanation will be complex. When multiple unlikely factors are genuinely at play, invoking "simplicity" to deny them isn't wisdom. It's laziness wearing a philosopher's mask.

The Simplicity Bias

So why does everyone misuse it? Because the human brain has a built-in preference for simplicity that predates William of Ockham by about 200,000 years.

Your brain is an energy-conservation machine. Processing complex, multi-causal explanations is metabolically expensive. Processing simple, single-cause explanations is cheap. Given a choice, your brain will reach for the simple explanation the way your hand reaches for the nearest glass of water — not because it's the best option, but because it's the easiest to grasp.

Psychologist Daniel Kahneman describes this as the difference between System 1 (fast, automatic, effortless) and System 2 (slow, deliberate, effortful). Simple explanations are System 1 friendly. They feel right intuitively. They click into place. Complex explanations require System 2 engagement — sustained attention, tolerance for ambiguity, willingness to hold multiple factors in mind simultaneously.

The "simplicity" that people attribute to Occam's Razor is often not a philosophical principle at all. It's cognitive laziness justified by a Latin phrase.

The Conspiracy Theory Paradox

Here's the strange part. The same people who invoke Occam's Razor to dismiss complexity sometimes embrace conspiracy theories that are spectacularly complex — involving dozens of coordinating actors, secret communications, suppressed evidence, and implausible levels of competence.

This seems contradictory. It isn't. Both moves serve the same psychological function: they replace a messy, multi-causal, unsatisfying explanation with a neat, single-cause, emotionally satisfying one. "The economy is bad because of dozens of interacting global factors" is complex and unsatisfying. "The economy is bad because a secret group is manipulating it" is simple — one cause, one villain, one story.

The simplicity bias doesn't actually prefer simplicity. It prefers narrative coherence — explanations that feel like stories, with characters and causation and a beginning, middle, and end. A conspiracy theory is structurally simpler than reality because it has a plot. Reality doesn't.

When Simplicity Kills

The misapplication of Occam's Razor has real consequences beyond misdiagnosed patients.

The Single Cause Fallacy

When a plane crashes, the investigation typically reveals a chain of contributing factors — a maintenance error, a weather condition, a communication breakdown, a design flaw, a fatigued crew. The crash was caused by the intersection of multiple independent failures, none of which alone would have been catastrophic.

But the public wants a single cause. "Pilot error." "Mechanical failure." "The airline cut corners." The media, understanding the simplicity bias, delivers a single-cause narrative. The real lesson — that complex systems fail through the interaction of multiple small failures — is lost. And because it's lost, the systemic factors that produced the failure remain unaddressed.

The same pattern plays out in business. A startup fails. The post-mortem identifies a single cause: "the market wasn't ready" or "the founder couldn't execute" or "they ran out of money." In reality, the failure was overdetermined — five or six things went wrong simultaneously, and fixing any one of them might not have changed the outcome. But the single-cause narrative is publishable, tweetable, and satisfying. The multi-cause reality is none of those things.

Medical Hickam's Dictum

Medicine has a formal counterpoint to the misuse of Occam's Razor. It's called Hickam's Dictum, attributed to physician John Hickam: "A patient can have as many diseases as they damn well please."

This isn't a joke. It's a correction to the diagnostic habit of seeking one explanation for all symptoms. An elderly patient presenting with fatigue, joint pain, and depression could have a single underlying condition — or they could have arthritis, anaemia, and depression as three separate, co-occurring problems. Forcing all symptoms into a single diagnosis because "simplicity" demands it can delay the correct treatment for each.

Hickam's Dictum is Occam's Razor properly understood. It doesn't say "always assume complexity." It says "don't force simplicity when the evidence doesn't support it." The razor shaves off unnecessary entities. If the entities are necessary — if the evidence requires multiple independent explanations — the razor leaves them alone.

When Occam's Razor Drew Blood

The pattern repeats across domains:

Challenger disaster (1986). Engineers warned about O-ring failure in cold weather. Management applied Occam's Razor: "The simplest explanation is that the O-rings will perform as they always have." They hadn't launched in weather that cold before.
Theranos (2014-2018). "The simplest explanation is that Elizabeth Holmes has invented revolutionary blood-testing technology." The simpler explanation — that she was lying — required assuming fraud, which felt like an unnecessary entity. Until it wasn't.
Subprime mortgage crisis (2008). "The simplest explanation is that housing prices will continue rising, as they always have." The simpler explanation required a complexity the model didn't contain.

The correct application of Occam's Razor isn't "choose the simplest explanation." It's "choose the explanation that fits all the evidence with the fewest unnecessary assumptions." Sometimes that explanation is simple. Sometimes it isn't. The evidence decides, not your preference.

The Real Power of the Razor

Strip away the misuse and Occam's Razor is genuinely one of the most powerful thinking tools available. Used correctly, it does three things.

It Kills Pet Theories

Everyone has explanations they're attached to — theories about why their business is struggling, why their relationship is difficult, why the world works a certain way. These explanations often include assumptions that aren't supported by evidence but feel necessary because they complete the narrative.

The razor asks: "Is this assumption doing explanatory work, or is it just making me feel better?" If you can remove an assumption and the explanation still accounts for all the observed evidence, the assumption was unnecessary. Cut it.

A manager believes their team is underperforming because of "low morale caused by the company's strategic direction." The razor asks: is "strategic direction" doing explanatory work here, or is it an extra entity? Maybe the team is underperforming because their two best engineers left last quarter and the remaining workload is unsustainable. That explanation is simpler, more testable, and requires fewer assumptions about abstract concepts like "morale."

It Generates Testable Predictions

The beauty of a shaved-down explanation is that it's easier to test. An explanation with twelve assumptions is almost impossible to falsify — if one assumption fails, you can adjust another to compensate. An explanation with two assumptions generates clear predictions: if A and B are the causes, then removing A should produce effect X and removing B should produce effect Y.

This is why Occam's Razor is central to scientific method. It's not a truth claim. It's a methodological tool — a way of selecting which hypothesis to test first. You start with the explanation that has the fewest assumptions, test it, and add complexity only when the evidence forces you to.

It Exposes Hidden Assumptions

The most valuable thing the razor does is make you list your assumptions explicitly. When you try to shave an explanation down to its minimum, you discover assumptions you didn't know you were making.

"Our product isn't selling because the market doesn't understand it." How many assumptions are hidden in that sentence? That the market has been exposed to the product. That the product's value proposition is clear. That "understanding" is the bottleneck rather than pricing, distribution, competition, or timing. That the market's failure to understand is the market's fault rather than the product's communication failure.

Each hidden assumption is an untested hypothesis. The razor forces you to surface them, examine them, and decide which ones the evidence actually supports. Most of them, you'll find, are doing no work at all. They're decorative. Cut them.

Simplicity as an Outcome, Not a Starting Point

The deepest misunderstanding of Occam's Razor is treating simplicity as the goal rather than the result.

Einstein — who understood this better than anyone — is often quoted as saying "Everything should be made as simple as possible, but not simpler." The second half of that sentence is the one that gets dropped. "But not simpler." Simplification that removes necessary complexity isn't elegance. It's distortion.

The best explanations in science are often stunningly simple — but they arrived at that simplicity after decades of working through extraordinary complexity. Einstein's E=mc² looks simple. The mathematics that produced it fills volumes. The simplicity is the compressed output of enormous intellectual work, not the starting point.

Simplicity, when genuine, is an achievement. When premature, it's a trap. The doctor who diagnoses "stress" on the first visit has achieved the appearance of simplicity by avoiding the work of investigation. The doctor who runs every reasonable test and arrives at "stress" after ruling out serious alternatives has achieved genuine simplicity — the kind the razor actually endorses.

The difference between these two doctors isn't intelligence. It's willingness to sit with complexity long enough for the evidence to tell you how simple the answer actually is — rather than deciding how simple you want it to be and forcing the evidence to comply.

William of Ockham was a medieval logician trying to clean up sloppy theological arguments. He noticed that his contemporaries kept adding unnecessary metaphysical entities to their explanations — extra angels, extra causal layers, extra dimensions of divine intention — when simpler explanations accounted for the same observations.

His razor was a tool for intellectual hygiene. Shave off what the evidence doesn't require. Keep what it does. Don't confuse your preference for simplicity with the universe's obligation to be simple.

Seven hundred years later, his razor is the most quoted and least understood principle in popular thinking. It's been flattened into "the simplest answer is usually right" — a claim that Ockham never made, that evidence doesn't support, and that actively harms clear thinking when applied without discipline.

The next time you reach for the simple explanation — the single cause, the clean narrative, the satisfying diagnosis — ask yourself one question: "Am I choosing this because the evidence supports it, or because my brain prefers it?"

If you can't tell the difference, you're not using the razor. You're being cut by it.