6

From http://en.wikipedia.org/wiki/Problem_of_induction#Karl_Popper

The rational motivation for choosing a well-corroborated theory is that it is simply easier to falsify: Well-corroborated means that at least one kind of experiment (already conducted at least once) could (but did not) falsify the one theory, while the same kind of experiment, regardless of its outcome, would not falsify the other. So it is rational to choose the well-corroborated theory; it may not be more likely to be true, but it is easier to get rid of it if not.

I'm having trouble understanding the part that says the same kind of experiment, regardless of its outcome, would not falsify the other. If the outcome was false, wouldn't it falsify the other theory?

  • 3
    I don't understand it either. It doesn't make much sense (I mean at a purely linguistic level, not wrt philosophical content). – DBK Apr 21 '13 at 22:16
  • To understand you need to read Miller's book "Critical Rationalism: A Restatement and Defense". Few thinkers today support Popper's corroboration concept. – Annotations Apr 22 '13 at 1:30
  • 1
    I don't have the book so I was hoping someone understood Miller's point. – user3306 Apr 23 '13 at 16:03
2

From my own limited knowledge of philosophy of science I believe this to be a reason to prefer ideas which have some type of experimental backing over ideas which have none.

Experiment A offers data that we can infer supports a Idea A but has no bearing on Idea B.

While our inference may be factually wrong, it would be logically better to support Idea A because it may eventually become falsified, while Idea B can never be falsified because it has no experiments which to reference. Idea B is almost in logical limbo due to a lack of support or falsification

The experiment being false has no bearing on Idea B, while it has bearing on Idea A. Inferences connect the experimental conclusion to the ideas.

My interpretation of the quote almost lends to the idea that it is better to believe in ideas which have been supported by falsification than it is to believe in ideas which have no reference to any experimental data or inferences at all.

1

You're probably going to have to read Karl Popper's paper on Falsification theory. The answer that LostPraxis gives explains it really well but I figured you might like some examples taken from Karl Popper's actual works.

it.http://www.stephenjaygould.org/ctrl/popper_falsification.html

During this time ideas like Communism and Psychoanalysis were all the rage in the world of economics and psychology respectively. Popper made the criticism though that these theories had no way of being disproven and therefore should not be accepted as theories at all.

No matter what you do, you could always look at some injustice in the world and contribute it to class inequality. It's a broad idea. Yet, when Karl Marx was first proposing Communism he had very specific observations that he was going to use to show that Communism was correct. For example, he thought that Communism was going to first happen in very first class economies like England. This didn't happen, Communism first hit Russia a place not too far away from the Dark Ages. Yet when this idea turned out to be not true people revised the theory of Communism to just mean, "Class inequality causes problems." This is too vague to be considered a scientific theory and therefore should be disregarded.

Compare this to Einstein's theory of relativity. Almost all parts of this theory are ultra-specific and have very visible grounds for being proven or disproven. For example, light is a particle and gravity has an effect on light.

To prove this they took pictures of the stars in a certain segment of sky during an eclipse where the sun's gravity should pull the light from the stars closer to the Sun. Then they took a picture of the stars in that segment of the sky when the Sun should have no affect on the starlight. Well, the gravity did affect the light and this helped prove Einstein's theory of relativity. Popper helped with this experiment oddly enough, he even mentions it.

So what Popper is saying is that if you have a theory that can be falsified with experiments then you have an acceptable scientific theory. If you have a theory that cannot be disproven no matter what kind of experiments you do to it, that is not a scientific theory. Like comparing Communism to Quantum Physics.

1

'm having trouble understanding the part that says the same kind of experiment, regardless of its outcome, would not falsify the other.

Let the experiment be X, and we have two theories A and B. If X doesn't affect B regardless of it's result then simply B doesn't take the risk to even can be falsified by X.

If the outcome was false, wouldn't it falsify the other theory?

The point is not X result is true or false. If either result has affect on theory (say A) while has no effect on other theory (say B) then A is more scientific than B because of taking more risk.

1

The most upvoted answer claims that experiments can support ideas. This is not compatible with Popperian epistemology, which explicitly attacks the idea of support: see "Realism and the aim of science" Chapter I. All scientific theories are guesses and there is no no support or justification or whatever that survives critical scrutiny:

Do all epistemologies suffer from the "regress of justifications" problem?.

Now, looking at the specific quote you give:

The rational motivation for choosing a well-corroborated theory is that it is simply easier to falsify: Well-corroborated means that at least one kind of experiment (already conducted at least once) could (but did not) falsify the one theory, while the same kind of experiment, regardless of its outcome, would not falsify the other. So it is rational to choose the well-corroborated theory; it may not be more likely to be true, but it is easier to get rid of it if not.

Corroboration is supposed to be a measure of the severity of a test. One theory A may make a specific prediction about the outcome of a test and so it would be tested severely by an experiment testing that outcome. Another theory B may not make a specific prediction about the outcome of that test. As such, the test is not relevant to judging B, but would refute A if it had some specific outcome. As such, even before doing the test A is more interesting since you can at least get rid of it by a test if it is no good. And if A passes the test you have no reason to discard it.

I should note that Popper did not consider corroboration a particularly important contribution and regarded it as useless in most practical situations for reasons explained in Chapter IV of "Realism and the aim of science".

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.