| While it's fun to criticize science this way, this line of criticism misses an important fact: namely that open scientific inquiry has been one of the most successful enterprises in the history of humanity. In a period of a few hundred years we've gone from believing that there were a few basic elements and that the sun revolves around the earth, to understanding the deep nature of particle physics and the structure of the Universe. We turned a basic understanding of chemistry into an understanding of subatomic particles, and the ability to create entirely new elements. We did all of this through a process of open and skeptical inquiry, which has been remarkably consistent in its ability to tear down unsupportable theories. The reason the Kuhnian critique exists is not because the scientific process failed, it's because the process worked but just took longer than people expected it to because people are human and imperfect. And the speed of scientific advances over the past decades has been higher than at any point in human existence. The reason the term "scientific consensus" exists is because most fields are vastly too complex for a single human being to be able to evaluate the totality of the evidence by themselves, at least in a reliable way. So the process is necessarily decentralized and broken up among many experts, who share their opinions. This isn't some popularity contest that you should ignore, it's a critically necessary task that has to be performed in order to digest the research contributions of any field, and make progress on solving open problems. You're absolutely right to point out that consensus evaluation can malfunction sometimes. You'd be equally right to point out that sometimes experimenters produce invalid results. You're wrong that the answer to the former is to reflexively ignore the scientific consensus process, just as you'd be wrong to say that "don't do experiments anymore" is the correct response to a few experimental errors. |
We know that a conductor moving in a magnetic field produces a voltage, and knowing the strength of the field and the motion, we know absolutely the voltage produced. Applied to moons of Jupiter and Saturn, we expect forces much larger than surface gravity of the moons, and therefore material leaving the poles. But when we find it occurring, we talk about "volcanoes" and "geysers". We carefully ignore that the volcanoes drift about like rubber ducks in the bath. We carefully ignore collimation that would need for the geysers to be shot from perfect paraboloid-shaped nozzles. It is easiest to just agree not to talk about perfect collimation, because it doesn't lead in a comforting direction.
Socially, people like a consensus. A challenger needs "extraordinary" evidence to displace it. But Nature doesn't play favorites: any alternative that accounts for all the established evidence is on equal footing. A consensus in the absence of compelling evidence, or in the presence of incompatible evidence, should make us suspicious that the consensus is a product not of evidence, but of biased preference. Seeing evidence carefully ignored should make us suspicious.
I am not aware of carefully-ignored evidence in the case of galaxy rotational anomalies, but this paper may be rubbing our noses in examples.
Nature is just as happy for all the leading theories to be wrong, and for us not to have invented the right one yet. The consensus can be wrong without any of the alternatives being right. It is discomforting to find yourself wrong, but science isn't about comfort.