Leiden, a.d. VIII id. Dec. MMDCCLXVIII A.U.C.,
In the modern philosophy of science there are two giants, Kuhn and Popper. Popper argued that we can only gain knowledge by refuting theories, and not through confirming them. The most famous example is the black swan. You have a theory that all swans are white, and you keep seeing new swans, and they are all white. But, you can never be sure all swans are white based on all the swans you have seen already. Then suddenly you find a black swan, so your theory that all swans are white is false, it has been falsified. Popper argued that science progresses, because the theories it uses are falsifiable. Popper termed this the demarcation criterion: in order for a theory to be scientific, it must be falsifiable.
Kuhn disagreed, he argued that if scientists used falsification all the time, they would not get anywhere. All the theories we have are false, they never explain all cases. Scientists have paradigms, exemplary scholarly works, that show them how to look for which phenomena. These paradigms so fundamentally impact how scientists interpret the world, that paradigms can be incommensurable, which means that two people in different paradigms see fundamentally different things. Therfore scientists are not some kind of falsifying gods, but humans, who are cognitively limited and inherently biased. They try to use variants of the same theory over and over again to explain phenomena, rather than multiple completely different theories, as Popper seems to demand. These cognitive biases, and the tendency to cling on to variants of the same theory is paradoxically what makes science so effective, because it allows scientists to work on ‘puzzles’ they expect to have a solution. Moreover, it gives scientists a shared language and worldview, so they can easily communicate, and do not have to discuss the basics of their discipline over and over again. Then, once in a while, when too many things cannot be explained using the paradigm, science comes into a period of crisis, in which scientists try to come up with increasingly creative solutions to the problems, until one potential solution seems to be very promising, and a new paradigm gets established.
I think Kuhn certainly has a point that scientists do not aim to falsify all the time, science does indeed progress because scientists often try to look for the things their theory can solve, and simply ignore the things they fundamentally do not understand in a slightly embarrassed way. However, I think it would be wrong to see paradigms as theories themselves, rather, they are tools through which we see the world. This is necessary, because any model of the universe needs to leave out a lot of information, and paradigms allow us to filter out ‘irrelevant’ data. So, paradigms are tools, they do not have a truth value, so they cannot be falsified, just like a hammer cannot be falsified. Paradigms precede theory, they allow theories to exist, by providing an ontology (what exists?) and an epistemology (how can we know?).
This means that while Kuhn is right that scientists do not aim to falsify all the time, and science progresses exactly because it leaves aside those problems it cannot solve at the time, but theories can and should still be falsifiable. The very fact that scientists leave aside some problems in a slightly embarrassed way, shows that they have tried to look at the world through a certain paradigm, have based theories on that paradigm, tried to solve puzzles with those theories, failed, and put the problem aside. However, they do know they failed, they have indeed falsified their theory, before they went on to the next problem. In a pseudo-science such failure is impossible, because the pseudo-science can explain everything. No matter what the outcome, you basically do everything because of your mother (Freud), because of stars (astrology), because of particles which are no longer in measurable quantities in the substance (homeopathy), or because of class (Marx).
So, while disagreeing with Popper that science progresses because of falsification, making it a sufficient factor in Popper’s view, I think science progresses because scientists cling on to their paradigm, but while trying to solve puzzles looking through the lenses of that paradigm, try different theories based on that paradigm. Without falsifiable theories, the environment would never give feedback to the theory, so it would never be adapted. It would be a completely dead theory, incapable of ever making any progress. We would never know when we are wrong. So, falsification is a necessary factor for scientific progress.
Bottom Line: It seems like Popper and Kuhn disagree, which would mean the end of falsification as a demarcation criterion. However, given a certain reading of Kuhn, in which paradigms precede theory, we both gain a better understanding of science, while retaining falsification.
P.s. Given the importance of falsification, it is really strange that most journals do not publish many well-executed studies which show statistically insignificant results.