sUMMARY - Philosophy of Science

Idea of Science vs Non-science, The “demarcation” problem
We studied the scientific point of view in the readings of Feymann, and on Astrology and the rationalist point of view in
the readings on scientific temper and the ones by Narendra Dabholkar. The discussion proceeded to the distinction
between science and non-science. A bottom-up approach was used where in we first came up with examples for the
two classes and then tried to develop a definition for the two. Examples for science - Physics, Laws of genetics, Darwin
evolution; non-science - Religion, Astrology, Art, palmistry, Reiki. Mathematics lies in the grey area of science and non-s
An interesting point to note was that science supersedes and theories are proven wrong but the previous theories still
remain in the science basket. Scientific doesn’t necessarily mean true. We discussed science vs non-science in order to
understand why, if at all, science is of high value. IS it that non-science means not true?
The “demarcation problem” is the problem for the basis to draw a line between science and non-science. This is
important for social justice (explaining cast hierarchy as non-science), for devising superior ways to represent
world. So, we further sub-divided non-science and pseudo-science. We want to demarcate pseudo-science as it is the
part of non-science that aspires the status of science. Eg- Astrology, Witchcraft, Marxism. Non-science, which has
things like art and literature are not the problem and concern of demarcation problem.
The Sceptic's Challenge (Conversation between Alice and Thomas), Basics of Epistemology and Metaphysics.
Alice is serious about the demarcation problem and firmly believes in popular scientific theories. Thomas is a sceptic
who thinks that all science is propaganda. Thomas brings up four arguments against scientific beliefs. 1. Self-interest
drives what scientists say. 2. Belief in science is essentially based on authority, just like belief in witchcraft. 3. Scientists
often disagree with each other on scientific theories. Therefore, how can we trust the scientific community if they cannot
reach a consensus? 4. Trial and error is the only way to arrive at results(experiments cannot lead to theories)
The Scientific Method often produces results far from common sense or what appears to be or is intuitive. This is ironic
because the scientific method often involves observation. Scientists are also people, they can be shady themselves and
produce flawed theories and research. These cases also lead us to question what science is. Is the scientific method
what scientists say it to be?
Epistemology asks the question, "How do we justify our beliefs about the world?" Metaphysics asks the question,
"What exists?" In a way, the philosophy of science can be posed as both questions.
The Problem of Induction, the distinction between a priori and a posteriori knowledge, and Naive Inductivism.
The Problem of Induction is framed as the question: "Can Experience be the sole source of the
understanding/knowledge of the world?" Consider that we have a set of observations and a conclusion L which is
arrived at using the observations. {O1,O2,..,Ok}→L. Now, L is something general about the world. We can use L to
know things about the future. The problem of induction asks, how can we use past observations (Oi's) to know
something about the future(L)? The problem of induction focuses on the "→" in the above statement.
a priori - knowledge that comes before (is prior to) experience. Definitions, Mathematical statements, and Logical
Tautologies are examples of this. Example: A bachelor is an unmarried male. 2 + 2 = 4. Either it is raining, or it is not.
a posteriori - knowledge that comes after experience/observation. Example: It rained today.
Naive Inductivism: "When a large number of observations of Xs under a wide variety of conditions have been made,
and when all Xs have been found to possess property Y, and when no instance has been found to contradict the
universal generalisation ‘all Xs possess property Y'."
We talked briefly about the example of "How do you know the sun will rise tomorrow?" and how all answers to this
seemingly simple question have an inductive argument as a foundation.
Presentation of Induction by Hume, Solutions, and dissolutions to the problem of induction.
Russell: Looking at inference in a single instance. Ex. Sun will rise tomorrow. 2. Making use of the notion of law of
nature. Ladyman (Hume): 1. Looking at inference regarding a general law. Ex. All A's are B's. Emphasis on
causation. Central Thought: Inference A->B is done because we have underlying causation to base it on.
Hume said that the only way we can justify the jump from an observed instance to an unobserved generalization is
through an underlying causal mechanism.
Constant Conjunction: Relationship between two events where one event is invariably followed by another.
As we don't see the cause when we look at a piece of evidence, just one thing following another. Hume's claim: All we
observe are thus, correlations and not causations.
Some important characteristics of causal relations are: 1. Contiguity: Relation of being connected in space and time. 2.
Causes precede effects in time.
In the case of induction, the process of justifying a belief and explaining why we are doing it are two different things. For
example, slavery is not justified but if you had asked someone a few decades back why are they practicing it they would
probably give you an answer.
Solutions and dissolutions: It seems to be the most rational by definition as perceived by people; The argument
Hume makes is too general. In a lot of cases, only one instance is enough to infer something but in some, it is not.
There is a need to make our inductive process detailed. This, however, doesn't give a reason to doubt our practices just
because they are inductive; We look at it because it is the best explanation which is in accordance with our
observations. (Gilbert Harman's account); Frequentist idea: The more evidence in favor of something, the more
probable it is and vice-versa.
Discussion on Bayesianism.
Context of Discovery v/s Context of Justification & Duhem Problem

Falsifiability, on the one hand, helps in demarcation(descriptive) and, on the other hand, identifies aspects of good
scientific practice(normative).
Context of Discovery v/s Context of Justification: What is the object of a good scientific method?
Popper's reply: "Discovery" doesn't matter to the scientific method. The discovery of scientific theories can have
religious, dream, and metaphysical origins. Example: Benzene, Penicillin, Ramanujan's mathematical results.
Ad hominem fallacy: Attacking a theory by attacking the person who believes in the theory/ who came up with it. (Eg.
Hitler was a vegetarian)
Duhem Problem:(Problem of auxiliary hypotheses): T & A entails E, not E; not T or not A.
When do we know which one is falsified? T(the hypothesis in question) or A(auxiliary hypotheses, initial conditions etc.)
Popper's reply: (Intersubjective Agreement) The group of scientists should agree to what conditions the hypothesis will
be rejected/accepted before the experiment is performed.
Problems with falsification

1 Legitimate portions of science are not falsifiable.
Philosophers have two kinds of approaches: 1. vindicate (justifying a practice) (approach taken in the philosophy of
science). 2. Revisionary (questioning a practice) (approach usually taken areas like political/social philosophy)
2. Probabilistic statements (Example: Half-life of uranium); Existential statements (To falsify ∃xPx, we would need to go
through all x in the universe); Simplicity/Unity: Simple hypotheses are preferred over Complicated ones. Similarly,
unifying hypotheses are preferred.
3. Degrees of Falsifiability ("A metal expands when heated" is more falsifiable than "Copper expands when heated")
When trying to falsify a theory, the other auxiliary hypotheses are assumed to be true and not falsifiable. But Popper is
not okay with this (Any scientific hypothesis is accepted until it is falsified).
4. Expectations about the future (Example: Why is going down the stairs more rational than jumping out of the window?)
Kuhn on Popper, Normal Science v/s Revolutionary Science

Popper's proposition: Testability/Falsifiability is the essence of science.
Kuhn's reply:
1. Most science doesn't involve falsifiability.
Usually, the practice of "puzzle-solving" is performed wherein the scientists tries to solve a particular puzzle under the
framework of accepted scientific knowledge(paradigm). If they fail to solve the puzzle, then only their ability is impugned
and not the current scientific framework. Kuhn calls this "normal science" or "normal research". He believes that it is this
aspect of puzzle solving and not falsifiability which draws the demarcation line. For example, astrology is a pseudo-
science because its practitioners never had a puzzle-solving mentality as opposed to astronomy.
2.When scientists throw out theories because of testing, factors are much more complicated than Popper allows.
Kuhn says that when enough anomalies accumulate in the normal research under a particular accepted paradigm, that
is when it might be thrown out the reasons for which might be non-rational. He calls this "Revolutionary science".
Distinction between philosophers of science and their theories: Philosopher Is Is

Science Science
The Received View of Science: 1. Science is cumulative. 2. Science is Inductive? Rational?
unified (Reductionism). 3. There is a distinction between context of
discovery v/s context of justification. Context of justification is of critical Hume Y N
importance. Discovery is irrelevant to the validity of the theory. 4. Underlying
logic of confirmation/falsification is value free. 6. There is distinction Carnap/Hempel Y Y
between theory and observation.
Popper N Y
Kuhn on Cumulative and Rational views of science:
Kuhn N N(?)
"Received View of Science": 1. Science is Cumulative. (Makes Progress). 2. It is rational and there is a 'scientific
method'. 3. Privacy of observation over theory.
We looked at Kuhn's opinion on 1 and 2.
The concept of paradigm is used by Kuhn for meaning two things: - Achievements, Examples. - Disciplinary
Matrix: Skills, Background Values, What counts as a good explanation?
Normal Science - Occurs within a paradigm
Revolutionary Science - Occurs between paradigms (This progress from one paradigm to another is complicated)
Parameters for different paradigms are different. So, finding common parameters to judge two paradigms is difficult.
Hence, there is not a lot of rational reason to choose one paradigm over the other. Kuhn says that this choice of
paradigms is usually influenced by non-rational factors. Science is not as rational as one might think.

sUMMARY - Philosophy of Science

Uploaded by

Copyright:

Available Formats

sUMMARY - Philosophy of Science

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

sUMMARY - Philosophy of Science

Uploaded by

Copyright:

Available Formats

Idea of Science vs Non-science, The “demarcation” problem

Presentation of Induction by Hume, Solutions, and dissolutions to the problem of induction.

Context of Discovery v/s Context of Justification & Duhem Problem

Problems with falsification

Kuhn on Popper, Normal Science v/s Revolutionary Science

Distinction between philosophers of science and their theories: Philosopher Is Is

You might also like