Karl Popper
1902–1994, Vienna and London
Philosophy of science; political liberalism
A theory is scientific to the extent that it could be proved wrong.
“It must be possible for an empirical scientific system to be refuted by experience.”
The life
Born in Vienna to assimilated Jewish parents who had converted to Lutheranism. As a teenager he was briefly a Marxist, and was present at a 1919 demonstration where police shot several young socialists; the episode, and the ease with which the doctrine explained it, turned him against the theory permanently.
He trained as a teacher and a cabinetmaker, took a doctorate in psychology, and published Logik der Forschung in 1934. Seeing what was coming, he took a post in New Zealand in 1937 and spent the war there writing The Open Society and Its Enemies, which he called his war work. Sixteen of his relatives died in the Holocaust.
He moved to the London School of Economics in 1946, where he taught until 1969. He was famously combative: the story of his 1946 encounter with Wittgenstein and a poker in a Cambridge room is disputed in every detail and has its own book. He was knighted in 1965.
What shaped the ideas
Vienna in the 1920s contained both the logical positivists, who wanted a criterion of meaningfulness that would eliminate metaphysics, and the intellectual movements Popper distrusted most (Marxist historical theory, Adlerian psychology, Freudian psychoanalysis) all of which struck him as able to explain anything whatever.
Against that, Einstein's general relativity had made a risky prediction about starlight bending near the sun, and the 1919 eclipse expedition could have destroyed it. That contrast, between a theory that stuck its neck out and theories that could absorb any outcome, is the origin of everything he wrote about science.
Core arguments
- 01
Falsifiability as demarcation
What makes a theory scientific is that some possible observation would refute it.
Confirmations are cheap: almost any theory can find some. What distinguishes a scientific claim is that it forbids something. The more it forbids, the more it says, and the more it risks. Note that this is a criterion of scientific status, not of truth or of value: a point almost universally misreported.
- 02
Science without induction
Hume was right that induction cannot justify universal laws, and science never needed it.
No number of white swans entails that all swans are white; one black swan refutes it. The logic runs one way only. So science proceeds by bold conjecture and attempted refutation: what survives severe testing is corroborated, meaning it has not yet failed, which is a report on the past rather than a promise about the future.
- 03
The open society
The right political question is how to remove bad rulers without violence, not who should rule.
Asking who should rule invites an answer (the wise, the workers, the nation) that then justifies entrenching them. Asking instead how institutions can be arranged so that bad government can be dismissed without bloodshed produces democracy as a mechanism rather than an ideal. It is the falsification principle applied to politics.
- 04
Against historicism
There can be no scientific prediction of the course of history.
The course of history depends on the growth of human knowledge; we cannot predict today what we will only know tomorrow; therefore we cannot predict the course of history. Theories claiming to reveal inevitable historical laws are not merely wrong but dangerous, because inevitability licenses any means.
Commonly got wrong
- Calling a theory unfalsifiable is not calling it false or worthless. Popper explicitly allowed that unfalsifiable ideas can be true and can later become testable.
- Naive falsificationism (one failed prediction kills a theory) is not his view. He knew that any test involves auxiliary assumptions, which is the Duhem-Quine problem; his answer is methodological rather than logical.
- Corroboration is not confirmation. It reports survival under testing and says nothing about probability of truth, and he was insistent on the difference.
- The Open Society is not a defence of laissez-faire. He argued for interventionist, piecemeal social reform and against utopian blueprints.
Works
Logik der Forschung 1934
Published in English as The Logic of Scientific Discovery in 1959.
The Open Society and Its Enemies 1945
Two volumes attacking Plato, Hegel and Marx as ancestors of totalitarianism.
The Poverty of Historicism 1957
The argument against predicting history, dedicated to those who died under a belief in inexorable laws.
Conjectures and Refutations 1963
The most readable entry point, and the source of the Freud-and-Einstein comparison.
'Science: Conjectures and Refutations', the opening essay of Conjectures and Refutations. Twenty pages, and it contains the whole argument.
Why they still get taught
The most widely held view of science among scientists themselves, a working test for pseudoscience that non-specialists can actually apply, and a defence of democracy that grounds it in error-correction rather than in the wisdom of majorities.
Chapters
- Falsification · Knowing, and Saying