Lecture 2 Popper

1
Lecture 2Popper
2
Some biographical facts

• Born in Vienna, 1902.
• Interacted with logical positivists, but was
  never a member of the Vienna Circle.
• Before the World War II, emigrated to New
  Zealand.
• After the war became Professor at the London
  School of Economics.
• Main works
• The Open Society and Its Enemies, 1945
• The Logic of Scientific Discovery, 1959
  (1934)
• Conjectures and Refutations, 1963
• Objective Knowledge, 1972
• Died in 1994.

3
The positivists main problem

• The main problem of logical positivists was to
  justify science on their own terms. Here is their
  predicament
• Scientific statements are based on observation.
• The content of scientific statements goes beyond
  what has been observed in the past.
• If the content of a statement goes beyond what
  has been observed in the past, mere observation
  cannot show that this statement is true. (Humes
  inductive skepticism)
• Scientific statements are not based on
  observation. (From 2 and 3)
• It seems that the set consisting of statements 1,
  2 and 3 is inconsistent. Or is it?
• Popper accepted 1, 2 and 3, but he rejected 4!

4
The role of observation

• Take the relation between a given scientific
  statement All swans are white and relevant
  observations. However many observations of white
  swans, it will never be enough to prove that All
  swans are white is true. And even worse, if Hume
  is right, no number of observations of white
  swans will ever be enough to prove that All
  swans are white is true.
• How then can Popper claim that the statements
  like All swans are white are based on
  observation?
• His answer indeed, observations can never verify
  scientific statements (show that they are true),
  but they can falsify them (show that they are
  false)!

5
Falsificationism

• There is a logical asymmetry between theoretical
  statements (T) and observation statements (O).
• There is no road from observation to theoryO1,
  O2, O3 ? T (Wrong! Fallacy!)
• There is a road from theory to observationT ?
  OOT
• We can derive a prediction from a theory,
  something that should be observed if the theory
  is true (T ? O). If we do not observe it (O),
  then it follows by deductive logic that the
  theory must be false (T).
• We cannot prove theories but we can refute them.

6
Falsificationism (2)

• Poppers main idea Scientists are actually not
  trying to prove their theories. They are trying
  to expose their theories to falsification.
• Great scientific theories are those that were
  very risky, that made bold predictions, and that
  were in a big danger of being refuted.
• The example that impressed Popper most was
  Einsteins theory of relativity. Einstein
  predicted that light will bend when passing near
  the sun.
• Moreover, Einstein gave directions about how his
  theory could be tested. He clearly described what
  should be observed during the suns eclipse, and
  suggested that his theory is false if the
  observation was not like that.

7
The bending of light (according to Einstein)
8
Will Einstein be refuted?
9
The demarcation criterion

• Einsteins theory was not refuted, although it
  could easily have been. This is what makes it a
  great science.
• Science is distinguished from non-science by the
  fact that it does not avoid the risk of
  refutation. It is falsifiable.
• Pseudo-science (theory falsely presented as
  science) is characterized by
• Falsifiable theories "prohibit" or "exclude" some
  situations, and it is precisely because of this
  that they say something interesting. Being
  unfalsifiable is a bad thing because if a theory
  doesn't exclude some states of affairs then it
  doesn't say so much about the world, and it is
  empty, boring or trivial.

10
Pseudoscience

• Pseudoscience, according to Popper, is not
  meaningless. It is just unfalsifiable.
  Pseudoscientific statements run no risk of being
  shown to be false. They are true, no matter what
  is observed. This is precisely what makes them
  useless.
• Example Alfred Adlers theory of the inferiority
  complex.
• Popper Once, in 1919, I reported to Adler a case
  which to me did not seem particularly Adlerian,
  but which he found no difficulty in analyzing in
  terms of his theory of inferiority feelings,
  although he had not even seen the child. Slightly
  shocked, I asked him how he could be so sure.
  "Because of my thousandfold experience," he
  replied whereupon I could not help saying "And
  with this new case, I suppose, your experience
  has become thousand-and-one-fold."

11
Prediction true vs. false

• If the prediction is false (we observe the event
  opposite of what the theory predicted) the theory
  is refuted. This negative result is not useless
  because we have now eliminated one wrong theory.
• If the prediction is true, the theory has passed
  the test but it should not be concluded that it
  is true. It should be tested again and again, by
  always deriving some new predictions to see if
  they  will be observed.
• The progress of science consists in trial and
  error, and any theory that is not yet refuted
  should be regarded as a hypothesis that may also
  soon be falsified. A theory that has passed many
  tests is corroborated, which tells us only
  something about its past (and nothing about it
  chances of passing the tests in the future).

12
Degrees of falsifiability

• There are degrees of falsifiability. Theories
  that are more falsifiable are better.
• For example, "All metals expand when heated" is
  more falsifiable than "Copper and iron expand
  when heated" because it is easier to falsify the
  first than the second. Any piece of iron or
  copper that does not expand when heated falsifies
  both hypothesis, but the first hypothesis can
  also be falsified with other non-expanding
  metals.
• More falsifiable theories are better because they
  say more about the world (they are more
  informative).
• On the other hand, if we have two theories that
  are equally falsifiable and only one of them is
  falsified, the unfalsified theory is better
  because it passed the test while the other did
  not.