The Fallacy of Consensus as Evidence

We've been conditioned to treat wide agreement as a proxy for truth, but consensus is merely a measure of coordination capacity and rhetorical dominance — it contains zero information about whether the position itself is correct. When everyone agrees on something, you haven't discovered a fact; you've identified a boundary condition where either everybody has read the same flawed source or somebody has gotten very good at making them look like they did.

Consensus is useful for identifying common sense, but it's a poor proxy for truth in domains where expertise has been systematically concentrated (science, math, medicine). The history of error is full of consensus-driven conclusions that were later overturned by minority dissent: plate tectonics, germ theory, the Copernican model.

The Bayesian framing helps here — if an expert community reaches consensus, it's evidence that the hypothesis is likely correct, but it isn't proof. The quality of the epistemic framework matters more than the size of the agreement. A million people believing the earth is flat doesn't change the geometry.

The real question to ask about any consensus: how was the agreement formed? Through open peer review and falsification attempts (good), or through credential gatekeeping and silencing dissenters (bad)?

True. Consensus is just popularity dressed up as proof.