The Reliability of Theories 04/14/2014

I'm a graduate student in physics, so I'll reply with a physics perspective on this. Concerning modern physics (the quirky stuff like relativity and quantum mechanics), I would not say that any part of it has "disproven" any of the old concepts (which we collectively refer to as classical physics) since Newton. For example, special relativity does not "disprove" classical, non-relativistic mechanics. Newton's three laws are still valid, it's just that the old theory is not as general as we once thought it was (once you move close to the speed of light, things look different to you vs. how it looks for others).

Of course, it goes without saying, what a theory fucking is is experiment. Special relativity, general relativity, quantum mechanics, QED, all these theories are theories because they gave predictions which were confirmed by experiment. Newton's laws, universal gravitation, these too have so much experimental backing so they are theory too, it's just that they are valid for non-relativistic, non-quantum systems. Just because you found that light bends around stars doesn't mean that all the basic experiments you did on the earth that didn't seem to show that bending 50 years ago were all a lie. Your old experiments were still valid! You just weren't looking at a scale where such effects were important: those mirrors and lenses didn't have enough mass to make the light bend enough for you to measure it (the deviations were but small fractions of the markings on your ruler) so you simply didn't see such effects. When you do look at that scales*, however, you'll see divergence from theory, and thus the cycle repeats, and the boundaries of understanding are pushed forward.

We never forgot the old theory, we just find where it fails and how it really was only correct for a smaller scale than we thought it was. Think about science as the old explorers of Columbus' day; they had a very small view of the world, but as they explored, they found that they could add new features to their maps. Nothing about their old maps of Europe had changed though! May be new technology had made the maps more refined, but discovering the West Indies didn't make Germany south of France! It's similar for science. Nothing about modern physics makes us doubt a lot of Newton's results, we just know he wasn't 100% correct.

> Evolution is a scientific fact. Climate change is a scientific fact.

This kind of talk can only hurt science. The single most important principle in science is that any theory must in principle be falsifiable by new evidence. No falsifiability, no science. Therefore there are no scientific facts, there are only scientific theories, some very good, but all potentially falsifiable by new evidence.

If you want unchallengeable facts, ideas cast in concrete, join a religion.

http://en.wikipedia.org/wiki/Falsifiability

"The concern with falsifiability gained attention by way of philosopher of science Karl Popper's scientific epistemology 'falsificationism'. Popper stresses the problem of demarcation—distinguishing the scientific from the unscientific—and makes falsifiability the demarcation criterion, such that what is unfalsifiable is classified as unscientific, and the practice of declaring an unfalsifiable theory to be scientifically true is pseudoscience."