Corroboration

Corroboration is Karl Popper’s word for the status of a theory that has so far passed our attempts to falsify it. Popper insisted it is not confirmation: surviving severe tests does not raise a theory’s probability or supply positive support, because inductive support is impossible. A well-corroborated theory remains a conjecture that may still be false. So far, so compatible with Critical Fallibilism.

CF’s reconstruction of what is genuinely good about test-survival is sharp: passing tests builds up a stock of known criticisms of potential rivals, so it becomes harder to invent a competing theory that is not already refuted by existing knowledge. The value is purely eliminative — it narrows the field of viable alternatives — never a positive mark added to the theory itself.

CF parts ways with Popper over degrees of corroboration. Popper repeatedly spoke of preferring the theory with the “higher degree of corroboration” and of corroboration as a belief “capable of degrees.” CF reads that as smuggled justificationism: it scores one non-refuted idea above another by accumulated weight, which is exactly the strong-versus-weak-argument thinking CF rejects. CF instead demands binary evaluations — refuted or non-refuted for a goal in a context — reached by decisive criticism, not by tallying past successes. Corroboration therefore cannot rank two surviving theories; it can only register that each is, for now, unrefuted. This also bears on critical preferences, where Popper let comparative corroboration guide choice and CF does not.