Skip to main content

Two friends with very different lives carried the debate between Popper and Kuhn into the 1970s. Imre Lakatos was born in Hungary in 1922; his mother and grandmother were murdered at Auschwitz, he became a communist official after the war, spent years in a Stalinist prison camp, and fled to Britain after the 1956 uprising, ending up at the London School of Economics. Paul Feyerabend, born in Vienna in 1924, was wounded as a German soldier in the Second World War and walked with a cane for the rest of his life. Lakatos defended the rationality of science; Feyerabend doubted any method could capture it. They planned a joint book, for and against method, but Lakatos died in 1974.

Research programmes

Lakatos accepted Kuhn's historical point: scientists do not abandon a theory at the first failed test, and they are often right not to. The unit to judge, he argued, is not a single theory but a research programme, a series of theories developing over time.

Every programme has a hard core: the central claims its followers refuse to give up. For Newtonian physics these were the laws of motion and the law of gravity. Around the core sits a protective belt of auxiliary hypotheses that absorbs anomalies: assumptions about planetary masses, friction or instruments. When a prediction fails, scientists adjust the belt and leave the core alone. The negative heuristic is the programme's rule not to aim refutations at the hard core. The positive heuristic is its plan for which problems to tackle next.

Progress and degeneration

What separates good tenacity from stubbornness? Lakatos answered with a contrast. A programme is progressive when its changes predict novel facts and some of those predictions are confirmed. It is degenerating when it only patches itself after the fact, accommodating each surprise without anticipating any. Scientists may rationally stick with a struggling programme, but one that keeps degenerating while a rival progresses should eventually lose its following.

Feyerabend's objection

Feyerabend admired the account and then turned it inside out. Since a degenerating programme might recover, and history shows that some did, Lakatos could never say when scientists ought to abandon one. The methodology of research programmes gives no deadline for giving up a degenerating programme, so in practice it permits anything a scientist decides to do. Feyerabend concluded that his friend was an anarchist in disguise.

Against Method

Feyerabend's own book, Against Method, appeared in 1975 and was dedicated to Lakatos. Its central thesis is historical: every proposed rule of scientific method has been broken in some episode of successful science. His chief example was Galileo. When Galileo argued that the Earth moves, the evidence was against him: a stone dropped from a tower lands at its foot, which seemed impossible if the Earth turned beneath it. Galileo won, Feyerabend argued, partly through rhetoric and an untested telescope, and only later did physics supply the theory of motion that made his case good.

From this Feyerabend drew provocative advice. Instead of only following the evidence, scientists should sometimes practise counterinduction: developing theories that clash with well established theories or apparent facts, because an alternative can expose hidden assumptions that the dominant view keeps out of sight. He summed up his position in a deliberately outrageous slogan to the effect that, if one insists on a universal rule, the only one left standing is that anything is allowed.

Feyerabend's politics followed. He argued that science is one tradition among many, with no automatic right to rule the others, and that science and the state should be separated, as church and state had been. Citizens, not only experts, should decide what is taught and funded.

Critics replied that Feyerabend confused a true observation, that method is flexible and depends on judgment, with a false one, that there is no difference between good and bad reasoning. Separating science from the state, they noted, would leave citizens unprotected from fraud in medicine and food. Feyerabend, for his part, often said he was a provocateur and did not wish to be read as a simple relativist. His lasting challenge is a humane one: who decides how people should live, and should expertise ever override their own judgment about their lives?

Worked example: Prout's hypothesis

In 1815 the English physician William Prout suggested that the atomic weights of all elements are whole multiples of the weight of hydrogen, so that every element might be built from hydrogen. Chlorine, measured carefully, came out at about 35.5, which is no whole number. The Belgian chemist Jean Servais Stas later measured carefully and concluded Prout was wrong. Lakatos used the case as a model of a research programme. Its followers kept the hard core, that elements are built from one basic unit, and blamed the protective belt: perhaps the chemical samples were impure, or the methods of measurement flawed. For decades the programme looked degenerating. Then in the 1910s the discovery of isotopes showed that ordinary chlorine is a mixture of atoms of weight about 35 and about 37, and that the individual atoms come close to whole numbers. The programme had recovered. The case supports Lakatos on the value of tenacity, and it supports Feyerabend on the danger of telling scientists when to give up.

Quiz

5 questions · Practice: scored, not saved

1. What is the hard core of a research programme?
2. What is the central thesis of Feyerabend's Against Method?
3. What did Feyerabend want separated, as church and state had been?
4. What was Feyerabend's main objection to Lakatos's methodology?
5. What does the protective belt of a research programme do?
6. What is the negative heuristic of a research programme?
7. What did Feyerabend mean by counterinduction?
8. When does Lakatos call a research programme progressive?

CivilSimian.com created by AxiomaticPanic, CivilSimian, Kalokagathia