I was just reading a sci-fi novel where physics "breaks down". While of course fiction is fiction and I don't expect this to happen in real life, when I tired to contemplate the concept I find that I cannot even imagine what it would mean for physics to break down. Is my imagination too limited or is such breakdown indeed impossible/undetectable? Apologies if this is outside of the scope of this SE, feel free to close if appropriate.

Let me try to narrow down the problem. I would like the everyday physics to remain the same - no planes falling down from the sky and bridges collapsing, etc. However, as we keep poking into more and more subtle subtle properties of our universe, the fundamental rule saying that repeating the same experiment should give the same results stops to be valid. In the story, this happened to particle physics: we build a few huge particle accelerators and it turns out that running the same experiment on different accelerators or at the same accelerators at different times gives different results. For simplicity, say we measure some value X (one real number) and depending on the time and location we get drastically different results.

What I don't understand is what possible measurement could make us think that physics no longer works. If X differs randomly, then we could classify this as random behaviour, measure average value, etc. - if I understand correctly, randomness is already is a fundamental feature of quantum mechanics and appears a lot in thermodynamics and other branches. If X varies in a structured way, then we could describe this structure and obtain a law of nature like any other. I don't think I could make a rigorous mathematical statement out of this, but it seems that any signal can be decomposed into a combination of something that looks random and structured for the purposes of physics. The worst I can think of is if X was supposed to be a constant, but I've heard that there are theories (maybe somewhat niche, but not ridiculous if I'm not mistaken) where the usual constants are not constant but vary slowly over time.

TL;DR Is there a conceivable results of a physics experiment that would make us conclude that physics has broken down?

  • 2
    $\begingroup$ When someone says that physics "breaks down" during some phenomenon, it means that the currently formulated physical theories are insufficient to explain the said event. This signals a gap in our knowledge and provides the next step towards unraveling undiscovered physics. $\endgroup$ – Avantgarde Mar 26 '19 at 1:01
  • $\begingroup$ I understand, but I don't think that's the kind of physicas breaking down that is meant. Perhaps I should be more precise: is there a possible experimental result that would violate time- and translation-invariance of physics in a way that is consistent with known physics (up to the precision with which it has been tested) and gives us a fundamental worry that our usual approach to physics no longer works? $\endgroup$ – Jakub Konieczny Mar 26 '19 at 1:05
  • $\begingroup$ You are essentially wondering if there is any process where we can't ever make any useful predictions about that process. Am I understanding you correctly? $\endgroup$ – BioPhysicist Mar 26 '19 at 1:17
  • $\begingroup$ Aaron - I think in effect that's what I'm asking. $\endgroup$ – Jakub Konieczny Mar 26 '19 at 1:21
  • 1
    $\begingroup$ This seems like somewhat of a philosophical question. I think a good answer would depend on what we really mean by "physics", "predictions", etc. For example, would having a "law" about this contrived example that states "this process cannot have a predicted outcome" be a sufficient enough law to say it is still physics? This is a hard question, and I think it could have many different answers. I'll have to think more, but I am interested to see what answers might arise. $\endgroup$ – BioPhysicist Mar 26 '19 at 1:25

Physics is not a solid object. It's constantly being fitted to explain current observations and /hopefully/ predict the new ones. So, the whole discipline is a kind of our best approximation to reality. This way, it can't break by definition - as long as the universe is alive. If the law which we stated is not working, we simply modify it.

Oftentimes, we are using different models for different regimes. So, each model has its own limits of applicability, outside of which you can say that it "breaks down", which is not very surprising.

These ideas have very concrete realization in quantum field theory, where most theories are considered "effective", being simplified versions of some "more fundamental" ones.

So, the short answer - even if physics breaks down, physicists will never admit it.

  • $\begingroup$ I don't think it's about not admitting it. It's about redefining, experimenting more, etc. so that nothing is broken anymore. $\endgroup$ – BioPhysicist Mar 26 '19 at 1:16
  • 1
    $\begingroup$ My 2 cents: physics is about modeling physical "reality" as best we can. Mathematical models are not actually "reality" (which I include in parentheses due to the current understanding of quantum mechanics), but an approximation to what we think reality is. In this context, it probably doesn't make much sense to describe physics as "broken". $\endgroup$ – David White Mar 26 '19 at 1:42
  • $\begingroup$ @DavidWhite This is one way to look at it. This is why we need to define what we mean by "physics" to answer this question. Certainly mathematical models are useful in understanding reality, but some would argue that "doing physics" means preforming experiments to understand how use, create, etc. these models. I could then see someone arguing that broken physics would occur when we can no longer use experiments to make and validate testable predictions with repeatable excperiments. If there existed some process like this, I might view physics as "broken" for that process. $\endgroup$ – BioPhysicist Mar 26 '19 at 2:51
  • $\begingroup$ @AaronStevens, you can't separate the mathematical models from the experiments, because the models (in principle) are developed to represent the result of all of the various experiments that are done. $\endgroup$ – David White Mar 26 '19 at 3:11
  • $\begingroup$ @DavidWhite I agree. I'm not suggesting they should be separated. I'm saying if there existed a process that could not be modeled because you cannot form any repeatable experiments with predicable results then you could argue physics is broken in this case. It wouldn't fit in any existing models, and you wouldn't be able to make any new ones to explain the process. $\endgroup$ – BioPhysicist Mar 26 '19 at 11:53

For experimentation it is necessary that an effect be both observable and repeatable. If an effect is not, it does not mean that physics has broken down, but rather that we can't apply the scientific method.

Non-repeatable phenomena routinely occur in behavioural 'science'. I might have a hypothesis that says 'If I punch Mike Tyson, he will say "ow"'. This might be observable but I can assure it is not repeatable, yet Physics will chug on quite well during my resultant period of hospitalisation.


In the novel The Three-body Problem the author gave a possible explanation: If there is absolutely no description we can give to the phenomenon. This can be realized through objects like, say, Chatin's constant: We can only give its definition, and infer what is not computable, there is no program that can compute any digit of the constant in any amount of time, there is no way to give the distribution of the digits, etc.

Since it's just sci-fi, it is most likely that the idea cannot be realized as real physics. Just read and enjoy ;)


Not the answer you're looking for? Browse other questions tagged or ask your own question.