Lecture 4 Lakatos

1
Lecture 4Lakatos
2
Some biographical facts

• Born Imre Lipsitz in Debrecen (Hungary), 1922.
• During the occupation joined a Marxist resistance
  group.
• 1945 adopted the name Lakatos (locksmith).
• After the war, finished his studies of science
  and philosophy.
• 1947 became a secretary in the Ministry of
  Education but was arrested in 1950 and sentenced
  to 4 years.
• Emigrated after the Hungarian revolution (1956).
• 1969 became Professor of Logic at the London
  School of Economics.
• Main works
• Proofs and Refutations, 1976
• Philosophical Papers (2 vols), 1978
• Died in 1974.

3
Lakatoss criticism of Popper (1)

• Popper said that good science is when you specify
  in advance under what conditions you would be
  ready to abandon your theory.
• He was aware that no observational prediction can
  be derived from a given theory alone, but only
  together with some auxiliary hypotheses. (For
  example, All metals expanded when heated
  implies This thing will expand when heated only
  if we also assume This thing is made of metal.)
• But Popper assumed that at least in some
  situations the truth of auxiliary hypotheses is
  taken for granted, and that it is a sign of bad
  science if the false prediction is blamed on them.

4
Lakatoss criticism of Popper (2)

• Lakatos disagreed. He said that real scientists
  never specify conditions under which they would
  reject a given theory.
• For example, when Newtons theory had problems
  and gave wrong predictions about the motions of
  planets, they did not reject it (what Popper
  would advise) but simply blamed an auxiliary
  hypothesis (about the number of planets).
• Again, when the Copernican theory had problems
  and gave wrong predictions about the star
  parallax, Galileo did not reject it (what Popper
  would advise) but simply blamed an auxiliary
  hypothesis (about the distance of stars).
• Sometimes, if an observation refutes a theory,
  it can even be doubted whether we really observed
  that thing.

5
(No Transcript)
6
All theories are born refuted!

• Lakatos says that all scientific theories (even
  in their happiest moments) are contradicted by
  some observations. All theories are falsified!
• Obviously, then, Poppers approach is wrong. He
  said that good scientific theories are those that
  are falsifiable and corroborated, i.e. (the
  theories that passed all the tests, and were not
  falsified).
• If all theories are falsified, how can we
  distinguish between good science and bad science?
• Lakatos said that for an answer we shouldnt look
  at individual scientific theories but at larger
  units research programs.

7
Research programs

• Research programs consist of two parts the hard
  core and protective belt.
• The hard core is the basic theory T, which plays
  a central role in a given scientific orientation
  and which scientists cannot give up so easily.
• The protective belt consists of a number of
  auxiliary hypotheses (AH), from which, together
  with T, observational predictions can be derived
  (T AH) ? O
• Now if scientists happen to observe O, this
  means (logically) that the conjunction (T AH)
  is false. But scientists will usually blame AH,
  and not T!
• Popper would say This avoiding of falsification
  is pseudoscientific! But Lakatos says this
  behavior is rational.

8
Hard core is unfalsifiable
Protective belt
Hard core
9
Newtons theory hard core and protective belt
Protective beltPlanet positions, optics,
telescope reliability
Hard coreLaws of motion, and gravity
10
Progress and degeneration

• If deflecting the arrow of modus tollens away
  from the hard core (and toward hypotheses in the
  protective belt) is what all scientists are
  doing, how can we distinguish good science from
  bad science?
• Lakatoss answer we have to look at how research
  programs develop over time.
• A research program is theoretically progressive
  if it constantly predicts new facts, and it is
  empirically progressive if these predictions turn
  out to be true.
• On the other hand, a research program is
  degenerating if it can accommodate newly
  discovered facts only after the event (if it lags
  behind empirical discoveries).
• So, anticipation is the key, not falsification.

11
What is rational behavior in science?

• Lakatos says that no theory should be abandoned
  if there is no better theory.
• But he also admits that sometimes a research
  program that was degenerating can make a turn and
  become progressive.
• His evaluations are about the past record, and
  cannot be used as an advice about what scientists
  should do. (Like Popper, Lakatos also rejects
  inductive inference.)
• If a scientist clings to a degenerating program,
  hoping that it will stage comeback and empirical
  success later, Lakatos has no way to condemn this
  behavior as bad science.
• Some philosophers wondered whether his approach
  is useful at all.
• Paul Feyerabend took Lakatoss philosophy of
  science as a methodological anarchism with a fig
  leaf.