Readings › Severity and error

The Methodology of Scientific Research Programmes: Philosophical Papers

Lakatos, 1978. §1, §2.

Card published · 64 underlinings · 9 claims drawn · last read 2026-11-15
Publisher
Cambridge University Press
ISBN
9780521216449
DOI
10.1017/cbo9780511621123
Shelf
Severity and error, also Language models and knowledge
Concepts
Falsification, Claim type

Why it is hereThe programme answers to Popper through Lakatos: falsification survives only as a decision, and a decision has to say what is under test and what is taken for granted. That is where the typing of a claim begins.

The underlinings across the pages

Copy 1 · 30 marks

78910111213141516171819

pp. 7–19 · 30 marks

Copy 2 · 35 marks

1819202122232425262728293031323334353637383940414243

pp. 18–43 · 35 marks

structure (red in Zotero) definition (green) key passage (yellow) book pages · pages not read are folded

1 Science: reason or religion?

pp. 8–10

Kuhn has shown that neither proof nor refutation explains how science grows, and seems to conclude that theory change is a conversion. Lakatos wants a rational account instead: intellectual honesty means stating in advance the conditions under which you would give up your position.

2 Fallibilism versus falsificationism

pp. 10–42

Three retreats from certainty, each more honest than the last: from proven knowledge to probable knowledge, from probable knowledge to refutable theories, and from refutation by facts to refutation by decision.

  1. Justificationism collapses Knowledge as proven knowledge was the tradition; intellectualists and empiricists both lost it. Probabilism retreats to highly probable theories, but on any evidence all theories turn out equally unprovable and equally improbable.C044 →Corrective evidence
  2. Dogmatic falsificationism All theories are fallible, but the empirical basis is not: the theoretician proposes, the experimenter disposes in Nature’s name, and science grows by overthrowing theories with hard facts.C045 →Conceptual tool
  3. Its two false assumptions There is no sharp line between observational and theoretical statements, and facts cannot prove the statements that report them. Add the ceteris paribus clause most theories carry, and a failed prediction never refutes the theory alone.C049 →Conceptual toolC051 →Conceptual tool
  4. Equally undisprovable Theories turn out not only equally unprovable and improbable but equally undisprovable: proof, probability and refutation fail together, and on its own criterion most of the history of science would be irrational.C050 →Corrective evidence
  5. Methodological falsificationism The activist answer: conceptual frameworks are made and can be rebuilt. With Duhem and Popper, conventionalism turns revolutionary; Popper’s version keeps experiments powerful even in mature science by accepting singular statements by agreement.C046 →Conceptual tool
  6. Basic statements by decision A statement counts as observational when a technique exists that anyone trained in it can use to decide it. Our best theories become extensions of the senses.C052 →Working method
  7. What is under test Every test has to decide what is being tested and what goes into unproblematic background knowledge, and relegating a ceteris paribus clause there is a risky decision with grave consequences.C053 →Working method
  8. Criticism with fallibilism The gain: hard criticism without an infallible basis, over a much wider range than the dogmatist allowed.C054 →Conceptual tool
  9. Sophisticated falsificationism Naive and methodological versions share two faults: a test as a duel between one theory and one experiment, and refutation as the only interesting outcome. Any theory can be saved by auxiliary hypotheses; the question is which savings are progress and which are ad hoc. A problemshift is progressive when it predicts new facts and some are corroborated, degenerating when not.p. 35“There is no falsification before the emergence of a better theory.”
  10. Slower, possibly safer Fewer conventions remain. Even a metaphysical theory is dropped not because it clashes with a corroborated one, but when it degenerates over time and a better rival is there to replace it.

Claims drawn from this card

Each claimA sentence from a source, kept verbatim, that the programme stands on, uses or has to answer to. shows its ENTIAThe notation this site uses on itself: what each piece is for, read through five layers. functionWhat a claim does for the programme: a conceptual tool, a working method, corrective evidence, a technical reference.; open + for its stanceHow the programme holds the claim: assumed, used, or contested., claim typeWhat kind of statement it is, so that the right test can be applied. Proposed until checked by hand. and relationsHow one claim bears on another, in the CiTO vocabulary: refines, extends, presupposes, confirms, contests..

Paraphrase CC BY-NC 4.0, Claudio Cammarano. Quoted passages belong to their author and are kept short; the text itself is not distributed. Notes from the Zotero library, updated weekly.