• Lakatos "Methodology of scientific research programmes"

He thought Kuhn was a disaster, because he took Kuhn to say that there is no general methodology of science. He proposed a sophisticated "methodology" of science.

• • Scientific research programmes
    • A research programme is a lot like a broad paradigm, but it has more structure: it has a hard core, which never changes within the research programme and a protective belt.
    • A research programme is progressive if the problem shifts in the protective belt make progress. (Degenerating if not progressive.)
  • Competing research programmes
  • Lakatos's rule: one should drop degenerating research programmes in favor of progressive ones.
  • Though there are reasons to prefer progressive research programmes, they aren't decisive. It is sometimes a good idea to stick to a degenerating programme in hopes that it will recover.
  • Lakatos thought we should give "rational reconstructions" of the history of science, that is (Godfrey-Smith) lie. The point is to see that our present beliefs nearly are, and could actually have been, justified on the basis of scientific norms.

• • Lakatos's theory has a fatal flaw: he wants a methodology, but, because it is reasonable to hang on to a degenerating programme, he doesn't actually have one.

Max Planck said that no scientist ever abandons a theory, the scientist gets replaced by his graduate students.

• Laudan
  • Lakatos without the methodology. "Research traditions" He gives principles, but ones that are not hard and fast, not "rules" of a "method." He also introduced the very important distinction between pursuit and acceptance. To pursue a research tradition is to work in it without necessarily believing it. To accept a research tradition is to, by and large, take it to be correct. He says we should pursue progressive traditions, but we should accept whatever tradition actually accounts for the most.

• Feyerabend His best-known work is

Against Method, and he calls his view methodological anarchism.

Feyerabend thought that the idea that normal science is an important part of science and a good thing for science is dangerous and dehumanizing. He thought it was bad from a human perspective: practitioners of normal science were trained to never realize their full capacities as human beings, and he thought it was bad for the development of science.

He makes several important points:

• Often the best way to understand a theory is to look at it from the perspective of a different theory.
• Sometimes the best way to support and advance a theory is through propaganda and even faked experiments instead of rational argumentation. Galileo is the best example.

That doesn't sound like he could be serious, but he was: Many of the things Galileo did, we can see with hindsight were good and necessary for the advancement of his theory. Nonetheless, when looked at from the perspective of his contemporaries, what he did was to propagandize and fake experiments. Since we, with hindsight, can see he did the right thing, it follows that it is sometimes an appropriate part of scientific procedure to propagandize and fake experiments.

• There can never be any decisive reason not to adopt the theory you prefer, for whatever reason you prefer it.
• Feyerabend is misinterpreted by practically everyone, including Godfrey-Smith. He is taken to think that you should do whatever you want, that there can never be reason for doing any particular thing. That is not what he says at all: he is against "method" not against reason and good judgment. Feyerabend's favorite book was Mill's On Liberty.

---

• One process versus two process philosophies of science.

In two-process views we have

1. Frameworks
2. Work within them

One process views are more like traditional empiricism: We have evidence, it gets used to figure out what is true. It is always important to approach evidence with an open mind.

-- ShaughanLavine - 05 Oct 2005 - 26 Sep 2007