Philosophy of Science
Week 1. A Demarcation Problem Page
Popper (falsi ability, testability, refutability) 2
Laudan (demarcation is useless) 3
Pigliucci (empirical understanding and theoretical knowledge) 4
Week 1. B Induction and con rmation problems
Popper (characterization of induction and solutions) 6
Hempel (con rmation problem and the Raven Paradox) 7
Week 2. A Underdetermination & False ndings
Douven (data is not conclusive on one solution) 8
Ioaniddis (most research ndings are false) 8
Week 2. B Structure of Scienti c revolutions
Kuhn (Paradigms, normal science) 10
Week 3. A Scienti c Realism
Klee (Cosmic Coincidence, IBE) 13
Week 3. B The Realism Debate, truth in science
De Regt (D-N, I-S, uni cation, causality, functionalism, new mechanism, 16
plurality, wide implications)
Week 4. A Scienti c explanation and understanding
Ylikoski (micro, macro, and mechanisms) 22
Grimm (Understanding in the social sciences) 24
Week 4. B Natural versus Social Science
Cartwright & Runhardt Measurement in Social Science 25
Week 5. A Values in Social Science
Douglas (epistemic, non-epistemic, 3-phase research) 27
Week 5. B The limits of Science
Douglas (moral responsibility of scientists) 30
Kitcher (Constraints on Free inquiry) 31
Week 6. A Scientism and the limits of science
Rescher (The limits of scienti c inquiry) 33
De Ridder (The danger of scientism in popular research) 34
1
fifi fi fi fi fififi fi
, Week 1. A
Logical Positivism and Veri cationism
In the early 20th century, some scientists called themselves Logical positivists. They wanted only
statements that were properly connected to empirical observation to be deemed science. The
Vienna circle were strict empiricists, they used philosophy and formal logic to analyze language
and clean up the sciences.
Their core ideas were:
1. The di erence between analytic and synthetic statements
Analytic: true/false just based on the meaning of the words used
Synthetic: true/false based on the meaning of the words used and what the world is like
(example: I am home alone). These are the questions science looks at.
2. An ideal language of science
The language would only contain terms and statements based on empirical observation
It needs the veri ability criterion of meaning (to tell what a statement means, you need to be
able to tell what would render it true or false, like an experiment)
The meaning of something is tied to the way it is tested
Only meaningful statements belong to science
Logic and mathematics are analytic statements, and therefore not veri able. This would mean
they are not science. The logical positivists solved this by saying mathematics and logic are
tautologically true and instrumental to other statements, and thus describe meaning within
another idea.
Are all scienti c statements veri able? Not really.
For example, when you want to test “gravity causes all objects to be drawn to a mass”, you are
unable to have an experiment that tests this on all objects. Uni cation is impossible.
Some things, like electrons, can only be seen through their e ects. This means you can never
truly test a theory about electrons, and that deems it non-science. Scientists today do however
believe electrons exist. Thus, empirical testing through e ects rather than the concept itself is
impossible.
Is the veri ability criterion itself veri able? This is a self-referential inconsistency. If we apply the
concept to itself, it will reject itself.
Popper and Falsi cationism
Popper writes how theories that seem to have been con rmed over and over again bother him.
Marx, Freud, and Adler all have theories that cannot, so it seems, be disproven. That apparent
strength is a weakness in Popper's view.
This led him to the following conclusions:
1. It is easy to obtain con rmations, or veri cations, for nearly every theory- if we look for
con rmations.
2. Con rmations should count only if they are the result of risky predictions
3. Every ‘good’ scienti c theory is a prohibition: it forbids certain things to happen. The more a
theory forbids, the better it is.
4. A theory that is not refutable by any conceivable event is nonscienti c. Irrefutability is not a
virtue of a theory, but a vice.
5. Every guide test of a theory is an attempt to falsify it or refute it. Testability is falsi ability.
6. Con rming evidence should not count except when it is the result of a genuine test of the
theory, and this means that it can be presented as a serious but unsuccessful attempt to
falsify the theory.
7. Some testable theories, when found to be false, are still upheld by their admirers. It rescues
the theory at the price of destroying its scienti c status.
This can be summed up by falsi ability, refutability, and testability.
2
fifi ff fi fi fi fi fi fi fifi fi fi
fi fi fiff ff fi fifi fi
, Popper does realize that most scienti c theories have origins in myths. Myths are therefore not
unimportant, but they cannot claim to be backed by empirical evidence in the scienti c sense.
Popper does not want to solve the problems of meaning, signi cance, truth, or acceptability. He
merely treats the problem of drawing a line between science and pseudoscience. This is
demarcation.
He also, while critiquing Wittgenstein, made a statement coinciding veri ability, meaningfulness,
and scienti c character.
It said: ‘The statements which may fall within the province of science are those which may be
veri ed by observation statements; and these statements, again, coincide with the class of all
genuine of meaningful statements.’
Wittgenstein’s criteria of demarcation are veri ability or deductibility from observation statements.
It is, according to Popper, too narrow and too wide. Besides, due to the induction problem, you
cannot reason in a logically valid way from observations to universal theories.
Popper thinks the problem of meaning is less important than the problem of demarcation.
Problems for Popper:
1. Is falsi cation too liberal? (Astrology)
2. Is falsi ability too strict? Probabilistic statements, existential statements, unfalsi able
scienti c principles
3. How do we falsify theories and what counts as a falsi cation?
4. Should falsi cation always be the end of a theory?
Popper Logical positivism
Fallability and nessecary risk taking. Emphasizing Always want to con rm theories, certainty
mistakes
Starting point of science is hypotheses, conjectures Start with observations, then create general
theories (objection to realism)
Weed out all the bad stu as you go Make sure you don’t let any bad stu in, everything
needs to be done through good observation
Process Product
Laudan
Laudan believes that the philosopher must explain the di erence between science and non-
science. Philosophers, in his opinion, have failed in doing so.
Aristotle’s demarcation criterion were dealing with causes, demonstrations, and most of the
apomictic certainty. This is also certainty and deductibility.
He also distinguished know-how from know-why. The know-why question was relevant to
science, unlike the know-how.
Most seventeenth-century thinkers accepted his rst criterion but rejected the second. Most 17th
and 18th-century thinkers believed that science and infallible knowledge were the same things.
In the mid-19th-century, people found out science could be fallible. Certainty was no longer a
demarcation tool.
Thinkers said the methodology was the mark that set science apart from all other
(semi-)knowledge. The new instance in this period is on a fallible method which, for all its fallibility,
is nonetheless superior to its non-scienti c rivals.
To mark science di erently from other things, one needs to prove two things:
1. Various ‘sciences’ use the same methodology
2. The epistemic credentials of this method had to be established
No one could however agree on a method, therefore the scienti c method was rejected.
3
fi fi fifi fiff ff ff fi fi fi fi fi ff fifi fi fi fi
, Laudan engages in meta-philosophical preliminaries:
1. What conditions of adequacy should a demarcation criterion satisfy
2. Is the criterion o ering necessary and/or su cient conditions for scienti c status
3. What actions are implied when calling something ‘science’ or ‘non-science’
There was a shift, and a new demarcation tradition came to be, This one does not say anything
about the epistemological side of science, only about the label. This is the moment when Popper
says the research for meaning is a pseudo-science, and he only cares for demarcation in the
sense of the label.
Laudan thinks demarcation isn’t all that important, and useless. If the label ‘science’ does not say
anything about the quality of the knowledge, what does it matter? Only knowing which
knowledge is good matters. The new tradition of demarcation is pointless to Laudan.
Laudan says science is also too diverse to be characterized as one thing.
Pigliucci
Pigliucci disagrees with Laudan on what the history of demarcation tells us. To Laudan, it means
that philosophers have failed at their task because the task itself is hopeless. To Pigliucci, it
means that history proves philosophers are making progress.
Laudan argues that the onset of fallibilism in epistemology during the nineteenth century meant
the end of any meaningful distinction
between knowledge and opinion.
Pigliucci gravely disagrees, since even a
scienti c ‘opinion’ would be more
probable than a regular one.
Pigliucci describes how he disagrees
with Laudan’s notion that the label
science is always value-laden. He
names the example of conspiracies that
cause many people to disagree with
science. This proves his point, namely
that not all science has value (to the
public).
Pigliucci’s thoughts on a solution to the
demarcation problem start with Dupre
(1993). He states that science is a
cluster concept, to Wittgensteinian family resemblance. This means that a characteristic may t a
concept, yet this characteristic doesn't need to belong to every concept in a certain cluster. He
names the example of games and winning.
At a very minimum, Pigliucci argues that two threads run throughout any meaningful treatment of
the di erences between science and
pseudoscience, as weak as further
distinctions within science itself. He
labels those threads as theoretical
understanding and empirical
knowledge.
Fuzzy logic was developed to deal
with situations that contain degrees of
membership or degrees of truth, as in
the standard problems posed by
notions like being old or young. One
can make sense of science/
pseudoscience as cluster concepts,
4
fffi ff ffi fi fi
