Falsifiability
Popper's criterion that a theory is scientific only if some possible observation could show it to be false.
Explanation
No number of white swans proves that all swans are white, but a single black swan refutes it. Karl Popper built his account of science on this asymmetry. A genuine scientific theory forbids certain things from happening, and so takes a risk. The more it forbids, the more it says.
Einstein's general relativity predicted that light would bend near the sun by a definite amount, and could have been refuted by the eclipse observations of 1919. By contrast, Popper argued, Freudian psychoanalysis and Marxist theories of history could accommodate any outcome, which made them unfalsifiable and therefore unscientific. Good scientists seek refutations and discard theories that fail.
Critics
Following Pierre Duhem, W.V.O. Quine pointed out that a failed prediction never refutes a single hypothesis, since auxiliary assumptions can be blamed. Thomas Kuhn argued that scientists do not, and should not, abandon a theory at the first anomaly. Imre Lakatos tried to combine the two views.
Originator
Related ideas
Problem of induction, Verification principle, Paradigm shift
Further reading
Stanford Encyclopedia of Philosophy
Open Falsifiability in the interactive Atlas, with lessons, quizzes, the map and the timeline.