According to Popper, a methodology of science is one of falsification in the sense, rather than doing inductive inference to get a general argument, one needs to state the general (bold) argument first then try to falsify them. By doing this, one can avoid a problem of induction and change the process into a deductive process. And as stated before, the conclusion of inductive inference even tough can be legitimately accepted if one can provide a premise that contains large numbers of data and a regularity of the claim still cannot be justified due to the problem of induction. In other hands, in deductive inference, the conclusion always entails to the premises. If the premises are true, then the conclusion must be true. A conclusion that is obtained by this procedure is said to be a valid argument.
We can use the previous example to get more grip on Popper method of falsification. Remember in the case of inductive inference we have a case where a claim that based on a large number and diverse data about planetary bodies and stars, scientists always find their shape to be round so it must be a likely argument to state that in general all planetary bodies and stars have a round shape. Now, instead start from such premise, we start from the conclusion as the premise.
Premise 1: All planetary bodies and stars have a round shape.
Now, according to Popper in order to justify this premise, one needs to find whether there is a way to falsify it or not. The obvious answer is one needs to find if there exists even one of the planetary bodies or stars that does not have a round shape. Suppose this is the case.
Premise 2: There has been observed by a group astrophysicists that there exists a planetary body in a solar system in certain part of Milky Way galaxy, that based on its physical behavior and data from satellites do not have a round shape.
Premise 2 falsify premise 1, it gives us a valid conclusion: Conclusion: Not all planetary bodies and stars have a round shape.
This conclusion is not just merely likely but must be right, since it entails to the falsification of the bold argument that is given in the premise 1. Like explanation before, the line of thought that has been presented is a deductive inference.
Popper method of falsification is indeed an interesting way to live with the problem of induction that plagues knowledge in science. Especially in my specialty, which is theoretical (particle) physics. Rather than proving the truth of a theory, it is easier to prove its falsehood. Because no matter how many number experiments that have been done to prove one’s theory, the theory will never be assumed to be right. It is due to the infinite number of possibility to come up with different theories that can explain the same results. In general in my opinion, a good theory is the one that can give a prediction that can be tested its falsehood through experiment. If a theory does not have this trait, then that theory cannot be tested hence it is not a good candidate to represent the natural processes.
There was a famous case of falsification in Physics made by Einstein when he first released his bold claim that due to the gravity (spacetime curvature) of stars, the trajectory of light will be curved in the vicinity of the stars due to the gravity field of the star. His theory was so complicated to understand but he gave a prediction that can be used to implement Popper’s falsification method. A couple years later, Eddington (British experimental physicist) took action and put Einstein’s theory to a test. He set an experiment to measure the trajectory of light from far away star around the gravitational field of the Sun when the Solar eclipse happened. If the trajectory is curved around the Sun then Einstein’s theory is survived but if it is not, Einstein’s claim and his theory must be false. The result of the experiment proved Einstein’s theory of relativity not yet wrong.
It is widely used in Physics community that to justify a claim of a theory is to find a way to falsify it. And I personally agree with this. Even if we extend Popper falsification method in further to distinguish science and non-science subject, I will be on board with this. Because in my opinion, the purpose of science is to gather knowledge. And we know from the start that knowledge is justified true beliefs. The most important point in identifying beliefs as knowledge, therefore, is a justification method. We can argue and have a discussion in different perspectives and point of views about how to do a justification of beliefs and may not find the right way to do it. But for me, at the end of the day, I will argue that the justification method for science is the one that can clear the doubt (not completely of course) as good as possible and as simple as possible. And this is the Popper falsification method. It gives us a way to deal with the disaster of inductive inference and at the same time gives us a most practical way to actually justify claims in science. Any claim that passes this test will be considered as a knowledge. Any subject that cannot give a clear way to justify mere beliefs and knowledge, is not a science.
Although I personally admit that, falsification has some issues like mentioned by (Pritchard, 105), such as the valid knowledge that we actually know is the one that has been falsified. And the bold argument used as premise remains likely to be true because of the inductive inference, i.e. all the experiments that cannot falsify the shape of planetary bodies and stars to make the argument all planetary bodies and stars have a round shape is likely true. But there is a difference here. While in inductive inference arguments, we generalize the conclusion from representable observations, and in case of the bold arguments that pass the falsification process also seems like inductive inference, I can argue that the later has a strong sense of justification than the previous one. It is because the former doesn’t seem to represent our best inquiry than the later. So, in my opinion any bold argument that has not been falsified (provide there is a way to falsify it) must be considered as our best opinion that we got from “the best ground available for believing what we do” (Pritchard, 184) -which refers to the falsification itself-, hence this is an anti-realist point of view.
In anti-realist point of view, any truth (thus knowledge) is just our best opinion. It is derived from considering weak objectivism that “we believe about the world right now could be false” (Pritchard, 183) but once we got a way to form our best opinion about that false belief, it cannot be false anymore.
The other problem with falsification explained by (Pritchard, 106) is even if we do get a falsification of the bold claim, not immediately everyone will abandon it. In our case, if someone truly finds a planetary body or star that does not have a round shape, we will not immediately abandon our argument that all planetary bodies and stars have a round shape. Instead, we will do more experiments to duplicate the result with different way until it is proven to be true or not. To be honest, I did not see any problem in doing this, since even if falsification does work and we have doubts about it, it does not mean we do not follow it. We just need to do the same to such claim (Not all planetary bodies and stars have a round shape), we have to check it by falsification again. And when we get our best opinion about it, let’s say we agree that it is true not all planetary bodies and stars have a round shape, we have to consider it as a truth (anti-realist) until a new claim appears again.
Science, in my opinion, is like a fluid, changing shape and moving, not a hard-rock stone. If we insist that knowledge (gathered by science) has to be justified true beliefs (in the sense of it cannot be true in one time and be false in other time), then we might not be knowing anything at all. We need to relax our claim, find a best way to look at the world (not too skeptical, strong objectivism, etc.) judge our beliefs in the best way we can, and in the end, whatever remains uneliminated by all the possibilities of our justification method (falsification in science-at least for me), is our best opinion, (the closest thing to) the truth (knowledge).