Falsifiability
A theory is scientific if and only if it could be refuted by experience. A theory no observation could contradict is not science but metaphysics or pseudo-science.
The Twentieth Century to Today · 1902 – 1994
波普尔 · Karl Raimund Popper
Falsifiability demarcates science from non-science, knowledge grows by conjecture and refutation, and the open society advances by piecemeal social engineering rather than utopian blueprints.
I may be wrong and you may be right, and by an effort, we may get nearer to the truth.
Source: The Open Society and Its Enemies, vol. 2, ch. 24
Popper was born in Vienna to a prosperous lawyer of Jewish descent whose library held over ten thousand books. After the First World War he was briefly a socialist; in 1919 he saw police shoot workers dead at a communist demonstration and realised he had accepted such sacrifices for the sake of a 'scientific' theory, and abandoned Marxism for good. The same year he heard Einstein lecture in Vienna and was struck by a remark: if the predicted red shift failed, relativity would fall. He contrasted this with psychoanalysis and Marxism, which could explain anything.
He apprenticed as a cabinet-maker, taught primary school, studied mathematics, physics and psychology at the University of Vienna, took his doctorate in 1928, taught in secondary school and mixed with members of the Vienna Circle on the side. The Logic of Scientific Discovery (1934) argued that theories can never be verified, only falsified; a theory is scientific insofar as it risks refutation by experience, and good science is bold conjecture, severe testing and the elimination of error. In 1937 he escaped the coming Nazism to Canterbury University College in New Zealand, where he wrote The Open Society and Its Enemies and The Poverty of Historicism as his 'war effort'.
The Open Society (1945) attacked Plato, Hegel and Marx as the intellectual sources of totalitarianism: all believed history has laws and a goal, and all wanted to remake society by a blueprint. The open society seeks no perfection, admits that no one knows the final truth, replaces revolution with criticism and gradual reform, and replaces 'who should rule' with 'how can we limit the damage bad rulers do'. In 1946 he joined the London School of Economics, and that October a ten-minute quarrel with Wittgenstein at Cambridge became legend, Wittgenstein reportedly brandishing a poker.
He held the chair of logic and scientific method at the LSE until 1969 and was knighted in 1965. Late in life he proposed 'three worlds': the physical world, the world of consciousness, and the world of objective knowledge, in which theories, problems and arguments, once created, have a reality of their own. He argued for years with Kuhn over whether science is rational criticism or the succession of paradigms. Stubborn and combative, he ran, his students said, 'the open society by one of its enemies', but his criterion of falsifiability and his warning against utopia changed the twentieth century's understanding of science and politics.
A theory is scientific if and only if it could be refuted by experience. A theory no observation could contradict is not science but metaphysics or pseudo-science.
Science does not induce theories from observation; it proposes bold conjectures and tries hard to refute them. Knowledge grows by eliminating error and is always provisional.
A society that admits no one holds final truth, replacing blueprints and revolution with criticism, discussion and piecemeal reform. The question is not who should rule but how rulers can be removed without bloodshed.
History has no predictable laws, because its course depends on the growth of knowledge, and we cannot foresee what we will come to know. Those who prophesy history are trying to make it.
Physical objects, subjective experiences and objective contents of knowledge are three real domains. Once expressed, theories and problems have a life independent of any mind.
He accepted Hume's conclusion that induction cannot be justified but rejected his psychological account: science never relied on induction, it works by conjecture and refutation.
Often counted among the Vienna Circle, he was its 'official opposition': verifiability cannot demarcate science from non-science, only falsifiability can.
He respected Marx's moral motives but treated the prophecy of historical laws as the classic historicist error: the growth of human knowledge makes the future unpredictable.
The first volume of The Open Society and Its Enemies reads Plato's Republic as a blueprint for totalitarianism, a reading still fiercely disputed.
His debate with Popper defined the philosophy of science in the 1960s: the daily life of science is not refutation but puzzle-solving within a paradigm, and refutations are taken seriously only in crisis.