Kevin T. Kelly
Department of Philosophy
Carnegie Mellon University
Every paradigm is refuted (puzzles are counterinstances)
Alternative paradigm is required for crisis (else unresolved puzzle counts against scientist)
Philosophical myth: theories must be rejected when they are refuted.
Plausibility of myth:
Puzzle solving is confused with testing (real goal is to solve puzzle).
Exemplary applications in textbooks are confused with evidence (real goal is to teach paradigm). (!!!!!)
Toleration of anomalies
Large anomalies can be tolerated: advance of moon's perihelion in Newtonian theory solved in 1750.
Puzzles must be scheduled to make progress.
Anomalies are resolved by unforseen means.
Anomalies that become crises
Contradicts central principles (ether drag and Maxwell's theory)
Inhibits technological applications (calendar reform)
New observational techniques (pneumatic chemistry)
Duration of anomaly (Ptolemy)
Onset of crisis
Anomaly attracts best minds
Anomaly becomes central focus of paradigm
Increasingly divergent partial solutions ("ad hoc")
Paradigm rules progressively relaxed
Recognition of crisis is rare.
"it is as though an artist were to gather the hands, feet, head, and other members for his images from diverse models, each part excellently drawn, but not related to a single body, and since they in no way match each other, the result would be monster rather than man" (Copernicus on the Ptolemaic astronomers).
"It was as if the ground had been pulled out from under one, with no firm foundation to be seen anywhere, upon which one could have built." (Einstein)
"At the moment physics is again terribly confused. In any case, it is too difficult for me, and I wish I had been a movie comedian or something of the sort and had never heard of physics." (Wernher Heisenberg, prior to development of matrix mechanics).
Possible outcomes of crisis
Normal science resolves it
Problem is shelved for solution in the future.
New paradigm deals with it before the old one does.
Paradigm replacement is not cumulative.
New paradigm doesn't handle all of the old data.
New goals and emphasis.
Different things are seen (not seen-as). Edmund Husserl and Gestalt psychology.
Extraordinary science: the birth of a new paradigm
Recognition of fundamental anomaly.
Paradigm used to analyze the anomaly as well as possible.
Amplification and refinement of anomaly.
Random experiments to see what will happen.
Generation of speculative explanations.
New interest in philosophical principles and methodology.
Thought experiments.
Proliferation of new discoveries.
New paradigm invention
Ask psychologists how it works.
Best to be young or new to the field (loosely committed to paradigm).
Extraordinary science is not yet revolution.
Scientific revolution = non-cumulative process of paradigm-replacement
Political revolutions:
Growing sense of inadequacy of the status quo in coping with new problems.
No common institutional ground for conservatives and revolutionaries.
Scientific revolutions:
Growing sense of inadequacy in paradigm's ability to confront new problems.
No common ground:
Paradigm choice not normal: paradigm cannot guide its own replacement. Anarchy intervenes.
Circularity: paradigm partisans assume the paradigm in arguing for it. Outside the paradigm, the argument is neither logically nor probabilistically compelling.
Rhetoric as important as logic.
Paradigm is highest court of appeal.
Myth of cumulativity
Cumulativity is suggested by empiricist philosophy and textbook pedagogy (chapters X, XI).
Cumulativity happens, but rarely (conservation of energy)
Novelty is not cumulative
Facts are novel insofar as they violate the paradigm's expectations.
Novel theories are invented only to explain anomalous facts.
Positivistic unity of science thesis: conflicts only occur when scientists draw conclusions beyond what the data properly justified (Newton not contradicted by Einstein).
Can say the same for phlogiston theory.
Restricting applications to observed data amounts to concluding nothing beyond the data.
Paradigm must guide beyond current applications.
Without commitments beyond the data there could be no anomalies.
No progress without commitments beyond the data.
Einstein's mechanics with v << c has different concepts than Newtonian mechanics, so the latter is not derived from the former (e.g., space, time, mass).
Differences among successive paradigms
Substantive
Existence, structure of reality
Normative
methods
importance of problems
standards of solution
Examples:
Descartes demanded an explanation of gravity. Newtonians refused. Einstein again demanded one.
After Lavoisier, chemists explain weights, not qualities. After quantum mechanics, qualitative explanations are again expected.
Etherial explanation of electromagnetism no longer demanded in 20th c.
Talking past one another
Paradigms disagree on importance of problems and on standards of solution, so they disagree on the "score".