What scientists mean by "consensus"

The word consensus gets used as a weapon by both sides of every health argument: one side wields it to end the conversation, the other dismisses it as groupthink. Both misunderstand what it is.

Not a vote

A scientific consensus is not reached by a show of hands, and it does not require unanimity. It is what emerges when many independent lines of evidence, gathered by people who often dislike each other's conclusions, keep pointing the same way. Different methods, different labs, different countries, different decades, converging. That convergence is hard to fake and hard to ignore.

This is why consensus carries weight that no single study can. One paper can be a fluke, a fraud, or a fashion. A conclusion that survives thousands of attempts to break it, from people motivated to break it, is a different kind of thing.

Not censorship

A strong consensus does not mean dissent is forbidden. It means dissent now carries the burden of proof. Anyone is free to overturn an established view, that is precisely how science advances, but to do it they need evidence at least as strong as the mountain they are trying to move. "Just asking questions" is not that evidence.

The honest position sits between two lazy ones. Treating every settled question as still open is not open-mindedness; it is paralysis. Treating every consensus as permanent is not rigour; it is dogma. Real consensus is strong enough to act on, and never so fixed that new evidence cannot move it.

How we use it

When we say the science is settled on something, we mean the convergence is broad and old and has withstood serious challenge, not that no one, anywhere, disagrees. And when a topic is genuinely contested among serious researchers, we say that too, rather than manufacturing a consensus that isn't there.

The goal is calibration: to match how firmly we state a thing to how firmly the evidence actually holds it.