Auxiliary Hypotheses Problem

When a theory clashes with evidence or a refutation, one can rescue it by tacking on an extra assumption — an auxiliary hypothesis — engineered solely to absorb the contradiction. A flat-Earther meeting a horizon photo can posit a lens distortion; a young-Earth creationist meeting a fossil can claim God planted it to deceive us. Each move restores consistency with the data while teaching nothing. This threatens falsifiability: if any theory can be patched indefinitely, refutation seems toothless, and bad ideas can be propagated endlessly in bad faith.

Popper treated such stratagems as illegitimate and tied his answer to content — a modification is acceptable only if it increases the theory’s testable content rather than merely shielding it. CF judges Popper’s handling reasonable but incomplete, and rejects his degree-based machinery (corroboration, degrees of content) on independent grounds.

CF’s distinctive move is to separate topical criticism (about the subject matter — fossils, teeth, hides) from epistemological criticism (about method, explanation quality, and arbitrariness). Ad hoc patches typically wreck the explanation they defend, converting a principled account into an arbitrary claim — and arbitrariness is itself criticizable, independent of the topic. So even an unfalsifiable add-on can be killed by a decisive criticism of its bad methodology. As one criticizes whole categories of bad thinking, novel arbitrary dodges get harder to invent; any genuinely new non-refuted idea is then either worth taking seriously or a useful specimen for sharpening the next criticism. This keeps refutation effective within CF’s binary epistemology.