This is something I’ve been wondering about for some time. Is it possible that some laws of physics are straight-up wrong, and can be disproved as our understanding/technology improves? How concrete are the laws of physics? Is it possible for us to be absolutely certain about anything?
In: Physics
“Laws” and “Theories” in science are written in such a way that they are [Falsifiable](https://en.wikipedia.org/wiki/Falsifiability) which means that it would be possible to prove them wrong
There are statements that are impossible to prove wrong like if I were to claim that there were an invisible teapot in orbit around Mars that is completely invisible to all detection methods ever, you have no way to even approach that to disprove it, the statement itself is impossible to prove false, it is “unfalsifiable”
But laws and theories in science and math come with hard rules that you could prove don’t actually align with reality. If you show up with good data showing that gravity *actually* falls off with Radius^2.005 and not Radius^2 then you will have proved Newton’s Law of Universal Gravitation to be incorrect. There’ll be a lot of push back and people will rerun your experiment to double check but if it holds up we’d go through and update the equation in every textbook.
In general we try not to call something a “Law” unless it has sooooo much existing experimental evidence behind it that its exceptionally unlikely to be far off of reality
There are also a lot of poorly represented theories in science that may be competing to be the leading theory with one ahead at any given time and that leads to a lot of public confusion, but providing the particle theory of light wrong didn’t mean that the physics was straight up wrong, just that our modeling of it needed adjustment because light sometimes does act like a particle.
Yes – it is absolutely possible to prove something held as a scientific law to be wrong. Today’s scientific laws are the product of doing exactly that for thousands of years.
Back in the 4th century BC, the Greek philosopher Aristotle stated that falling objects accelerated at a rate directionally proportional to their weight. He “proved” this assertion by using a thought experiment – the idea of dropping a cannonball and a feather from the top of a tower. Everyone knew how differently feathers and cannonballs fell, so his statement went unchallenged for hundreds of years, and everyone just accepted it to be true.
In the 6th century AD, the Egyptian philsopher Philoponus put this thought experiment to the test, but instead of using a feather and a cannonball, he used two dense weights, one twice as heavy as the other. If Aristotle’s assertion was correct, the lighter of two would take twice as long to hit the ground. When he actually did the experiment, the two weights hit the ground at almost the same time – the lighter one did take longer, but only slightly. Philoponus later made a quite profound comment in regards to his disproving Aristotle:
>Our view may be completely corroborated by actual observation more effectively than by any sort of verbal argument
And that’s pretty much how science has worked ever since – it’s not enough to simply have an idea that makes sense, you have to be able to prove it experimentally.
As others have commented, at this point the laws of physics have been (mostly) tested and found to be “correct” in that their results have been found to be reproducible.
A current example of this has to do with something you might have heard about in the general news of late – the “g-2 muon anomaly.” [Here](https://www.youtube.com/watch?v=kBzn4o4z5Bk) is a detailed explanation, but to summarize: the theoretical calculation of the muon dipole moment (for this ELI, what that means is not important) is not matching up with the experimental measurement (though the uncertainties in the measurement are not *quite* good enough to be certain). How much “not matching up”? The difference is in the ~~eighth~~ **ninth** decimal point – 2.00233318362 vs 2.00233318412! What does this difference mean? That there is (potentially) physics involving elementary particles that we do not understand. The theory would be incomplete.
Similarly, one of the first hints that something was wrong with the physics governing how the planets move was the realization that Mercury was not *quite* where it was predicted to be. “Fixing” the theory ultimately required relativity to explain (or, conversely, the error spurred others, especially Einstein, to “discover” the ideas behind relativity).
This iteration – compare the theoretical prediction to the experimental value, and then “fix” the theory – is at the heart of the scientific method.
I think maths and physics are both sciences made to explain what surrounds us. Maths enables us to go one step further by trying to not only rely on what we witness. For example it is hard for us to see above 3 dimensions. With maths you dont really care, you use a variable and think how you can generalize.
But remember the whole “You need a referential to make such calculus”, why? We need to state that we are here to calculate let’s say how things move. Dont you think it’s a very human thing to do?
If we use our imagination, what the world would look like if we were omniscient or being at two place at the same time or even being at two different point in time. How would it affect our conception and axioms for both maths and physics?
So do we model what’s really happening or have we just found an optimal solution to explain most of the things that happens based on how we perceive the world?
I feel like there’s a problem here with definitions so I’ll take a shot at it.
**Hypothesis**: “I have an idea that cows do not eat meat because they lack the proper teeth to chew it.” Basically this is someone saying that they have some **data** and have an idea on some sort of general case based on that data. Datum: Cows do not eat meat. Hypothesis: they don’t eat meat because they can’t
Note that hypothesis here is very close to the popular definition of **theory**.
**Theory**: this is a hypothesis that has been tested many times and been found to be true (so far). It should be able to explain many different sets of data using one coherent set of rules and also make predictions about what we will observe if we were to uncover new data.
**Laws**: these are rules that apply because all of the evidence we’ve found points to them being true. They are separate from theories because they are simple observations about the world and the universe around us. While there may be places where the laws of physics or chemistry or biology break down, we haven’t found any yet. Laws do not make predictions about future data, they simply describe all data to date.
So a hypothesis can become a theory but a theory never becomes a law because theories and laws are different things.
> Is it possible for us to be absolutely certain about anything?
Outside of mathematics, no. Even then, you can’t be 100% certain of everything in mathematics because of something called Goedel’s Incompleteness Theorem which notes that a mathematical system cannot both be 100% error free and cover all cases. A simple way to look at this is 1/0 (1 divided by zero) which cannot be defined in our ordinary mathematics but does have solutions under different axioms.
So, the real answer is ‘No, we cannot be 100% certain of anything.” But we can be pretty damn sure we’re right.
The word “law” in science is mostly just used to describe simple statements that appear to be generally true, or that were thought to be in the past. Some “laws” are routinely violated – for example Kepler’s laws of planetary motion are not precisely correct as they fail to take into account general relativity and N-body effects. At the other extreme, the laws of thermodynamics seem to be pretty watertight, and any violation of them (at least at large scales) would be extremely surprising. Not because there is any a priori reason to believe that they *must* be true, but because countless experiments have found that all manner of systems obey them and nobody has found any that don’t.
> Is it possible that some laws of physics are straight-up wrong
Yes. It’s perfectly conceivable that somebody does an experiment under some kind of exotic conditions that nobody has looked at before and finds a violation of (say) conservation of energy. However, the law of conservation of energy (aka the 1st law of thermodynamics) would not be thrown out overnight, as there are many contexts in which we know it does work and has proved to be useful. Similarly, Kepler’s laws of planetary motion do actually give a very good approximation of planetary motion in many circumstances with very little effort, so they’re still useful even if we know they aren’t 100% correct.
Latest Answers