3 ms·
> you make the experiments to confirm the theory The scientific method has usually worked by either coming up with theories and testing them; or by measuring t
by tsimionescu 6y ago
> you make the experiments to confirm the theory
The scientific method has usually worked by either coming up with theories and testing them; or by measuring things we thought we knew and finding new explanations when those turn out to be false.
What has not really worked too often has been coming up with mathematical models out of thin air, fitting them to all of the current data, extrapolating them to things we haven't been able to observe yet, and then building those things to check if they also worked.
That is why research into astronomical objects to ever greater precision is a promising candidate: we know what we expect to find if our current theories are correct, we know of real problems with our theories, and we can come up with new theories that predict such and such structures that should be visible in precisely such and such conditions.
This is the exact same model of how physics advanced from Tycho Brahe to Newton - people had models of the motions of the stars, they collected data to confirm those models, and found problems that they couldn't explain. Then Newton came along and found a mathematical model that, for the first time, explained those observations, and even reconciled them with other observations on the ground. We then spent a lot of time testing Newton's theories and didn't find any holes for a few hundred years.
Along came electromagnetism, the study of its dynamics, and the discovery of the speed of light problem, which couldn't be explained - requiring a new model that could actually explain everything we knew + the new data, and so on.
- johncolanduoni 6y ago> What has not really worked too often has been coming up with mathematical models out of thin air, fitting them to all of the current data, extrapolating them to things we haven't been able to observe yet, and then building those things to check if they also worked. But this is the story of early to mid 20th century physics! General relativity, the initial quantum mechanical theories, Dirac’s equation and anti-particles, neutrinos were all the product of this process of adding mathematical conveniences and extrapolating to things that turned out to be real. You can argue that more recent efforts in this direction have been much less successful than those were, but I think you have to turn a completely blind eye to the history of modern physics to claim that method “has not really worked too often”.
- tsimionescu 6y agoI don't think general relativity at least fits into this model: general relativity explained phenomena that special relativity just couldn't, and that were already well known to exist (gravity, objects of equal mass falling at equal speeds, time dilation in relation to acceleration). I'm less well versed about the other subjects you mention, but there were several unsolved problems in physics at that time that absolutely required new theories (the photoelectric effect VS the wave behavior of light, black body radiation, the need to reconcile quantum mechanics with special relativity). In all of these cases, we had unexplained data or inconsistencies between theories. By contrast, most of the theories that could be tested with a new collider are not solving anything - they are just theories about things that might exist, but could very well not exist at all. There is no hope of gaining insights into the measurement problem or quantum gravity at the next collider. Even the theories of dark matter that could be tested are still too early to be worth a 20b bet. There are too many equally plausible theories of dark matter to say any one of them is promising enough to be worth that amount of money. And even ruling one of them out will not have a major effect on the field. It's not like it would rule out string theory, for example.