The problem is that it’d be like if matter and energy could just disappear. Black holes would be exclusively tiny, as soon as one formed it’d start vanishing anything that crossed it’s event horizon rather than growing, so galaxies could never have formed as their cores would just shrink away as soon as they got too dense.
Black holes are regions of space where information density hits the upper limits allowed by physics. Add more information to it, and the event horizon expands proportionally to what was added. With that in hindsight, it seems rather obvious that the boundary of the event horizon could encode the information once thought to be lost to the black hole inside.
It could do that but what’s the evidence that it does? Or has someone proved this is already a feature of semi-classical gravity that just wasn’t noticed before? Or is it only a feature of a brand new hypothetical theory?
The other alternative is that the quantum information is somehow converted to some value of the black hole’s measurable properties; charge, mass, and spin. We know that isn’t the case because the values for these that we can infer from observation are consistent rather than growing faster than expected.
I still don’t really understand why the information just can’t be destroyed. It seems like we’re starting from an assumption that it shouldn’t be destroyed despite it being so in semi-classical gravity, and then trying to think of alternative theories which could preserve it such as on the boundary or in its charge/mass/spin. Maybe that’s correct but it seems like speculation, and it’s not speculation based on any actual contradiction between theory and practice, i.e. not because semi-classical gravity has actually made an incorrect prediction in an experiment we can go out and verify, but only because we have certain preconceptions as to how nature should work which aren’t compatible with it. So it doesn’t really come across to me as a scientific “problem” but more of a metaphysical one.
It’s fundamentally a product of one of our most basic assumptions, that the laws of physics don’t change.
When the laws of physics don’t change, symmetries arise in the math used to describe them, and each of these invariant symmetries corresponds to a law of conservation we can observe experimentally and an aspect of the universe it renders un-measurable.
Conservation of Momentum is a space-translation symmetry which makes it so that absolute position is unmeasurable, we can only tell where we are in relation to other things. Conservation of angular momentum is a rotation symmetry that does the same thing for direction. There’s no “center” to the universe and no “up” or “down” without something to stand on for context, and no experiment we could possibly design can prove otherwise.
Conservation of energy (and therefore mass) arises out of time-translation symmetries. There’s no way we can distinguish a particular moment in time from any other without setting a relative “time zero” for comparison, and no possible clock we can build that could be 100% accurate. We have to account for the different rate of time in the atomic clocks in our GPS satellites due to their relative velocity to us on the ground, but the lack of absolute time precision means it can only ever provide an estimate with some range of error.
Exactly how the relativity of spacetime implies a universe with conservation of information would require a lot of math, and a new description of spacetime that breaks these conservation laws would have to explain why it “seems to” adhere to them in all the ways we’ve tested our reality so far.
If I am not mistaken, information loss inside of a black hole comes out of semi-classical gravity. If these symmetries are tied to the assumption that the laws of physics don’t change and the symmetries break down in semi-classical gravity, then does that mean in semi-classical gravity the laws of physics change? Is there a particular example of that in the theory you could provide so I can understand?
I don’t disagree that information is conserved in general relativity and quantum mechanics taken separately, but when you put them together it is not conserved, and my concern is that I don’t understand why we must therefore conclude that this necessarily wrong and it can’t just be that information conservation only holds true for limiting cases when you aren’t considering how gravitational effects and interference effects operate together simultaneously. I mean, energy conservation breaks down when we consider galactic scales as well in the case of cosmic redshift.
Yes, we can experimentally verify these laws of conservation, because in practice we can only ever observe gravitational effects and interference effects separately, as a limiting case, and thus far there hasn’t been an experiment demonstrating the plausibility of viewing them simultaneously and how they act upon each other. In semi-classical gravity these “weird” aspects like information loss in a black hole only arise when we actually consider them together, which is not something we have observed yet in a lab, so I don’t see the basis of thinking it is wrong.
You seem to suggest that thinking it is wrong implies the laws of physics change, but I’m not really sure what is meant by this. Is semi-classical gravity not a self-consistent mathematical framework?
To oversimplify with another example from the theory, assume that planet earth was in superposition between two states with a non-zero separation. Semi-classical gravity says the distribution of the gravity field would be split evenly between the two points, but observing such a state is impossible as it must decohere into 100% of the mass being either in one point or the other. It simply doesn’t make sense when we try to apply quantum maths to gravitationally-significant objects because gravity isn’t a quantum field.
So yes, the predictions made by semi-classical gravity diverge from reality when faced with extreme masses, but that theory was only ever intended to be an approximation. It is useful and consistent with reality under certain ranges of conditions, but we shouldn’t jump to the conclusion that physics breaks from all known fundamentals in the presence of large masses when the simpler answer is that this is a case where the approximation is wrong. A more complete theory will be able to accurately explain physics across a wider range of conditions without requiring the untestable assumption that there are places where the rules don’t apply. We’ve got a good reason to believe that the rules of physics don’t change in the fact that no matter where we look the rules seem to always have been the same and all prior divergences from the model could be explained by better models.
The problem in physics is that we have two models that describe reality with absurd mathematical precision at different scales but which seem to be fundamentally irreconcilable. But we know they must be, because reality has to be assumed to be consistent with itself.
The problem is that it’d be like if matter and energy could just disappear. Black holes would be exclusively tiny, as soon as one formed it’d start vanishing anything that crossed it’s event horizon rather than growing, so galaxies could never have formed as their cores would just shrink away as soon as they got too dense.
Black holes are regions of space where information density hits the upper limits allowed by physics. Add more information to it, and the event horizon expands proportionally to what was added. With that in hindsight, it seems rather obvious that the boundary of the event horizon could encode the information once thought to be lost to the black hole inside.
It could do that but what’s the evidence that it does? Or has someone proved this is already a feature of semi-classical gravity that just wasn’t noticed before? Or is it only a feature of a brand new hypothetical theory?
The other alternative is that the quantum information is somehow converted to some value of the black hole’s measurable properties; charge, mass, and spin. We know that isn’t the case because the values for these that we can infer from observation are consistent rather than growing faster than expected.
I still don’t really understand why the information just can’t be destroyed. It seems like we’re starting from an assumption that it shouldn’t be destroyed despite it being so in semi-classical gravity, and then trying to think of alternative theories which could preserve it such as on the boundary or in its charge/mass/spin. Maybe that’s correct but it seems like speculation, and it’s not speculation based on any actual contradiction between theory and practice, i.e. not because semi-classical gravity has actually made an incorrect prediction in an experiment we can go out and verify, but only because we have certain preconceptions as to how nature should work which aren’t compatible with it. So it doesn’t really come across to me as a scientific “problem” but more of a metaphysical one.
It’s fundamentally a product of one of our most basic assumptions, that the laws of physics don’t change.
When the laws of physics don’t change, symmetries arise in the math used to describe them, and each of these invariant symmetries corresponds to a law of conservation we can observe experimentally and an aspect of the universe it renders un-measurable.
Conservation of Momentum is a space-translation symmetry which makes it so that absolute position is unmeasurable, we can only tell where we are in relation to other things. Conservation of angular momentum is a rotation symmetry that does the same thing for direction. There’s no “center” to the universe and no “up” or “down” without something to stand on for context, and no experiment we could possibly design can prove otherwise.
Conservation of energy (and therefore mass) arises out of time-translation symmetries. There’s no way we can distinguish a particular moment in time from any other without setting a relative “time zero” for comparison, and no possible clock we can build that could be 100% accurate. We have to account for the different rate of time in the atomic clocks in our GPS satellites due to their relative velocity to us on the ground, but the lack of absolute time precision means it can only ever provide an estimate with some range of error.
Exactly how the relativity of spacetime implies a universe with conservation of information would require a lot of math, and a new description of spacetime that breaks these conservation laws would have to explain why it “seems to” adhere to them in all the ways we’ve tested our reality so far.
If I am not mistaken, information loss inside of a black hole comes out of semi-classical gravity. If these symmetries are tied to the assumption that the laws of physics don’t change and the symmetries break down in semi-classical gravity, then does that mean in semi-classical gravity the laws of physics change? Is there a particular example of that in the theory you could provide so I can understand?
I don’t disagree that information is conserved in general relativity and quantum mechanics taken separately, but when you put them together it is not conserved, and my concern is that I don’t understand why we must therefore conclude that this necessarily wrong and it can’t just be that information conservation only holds true for limiting cases when you aren’t considering how gravitational effects and interference effects operate together simultaneously. I mean, energy conservation breaks down when we consider galactic scales as well in the case of cosmic redshift.
Yes, we can experimentally verify these laws of conservation, because in practice we can only ever observe gravitational effects and interference effects separately, as a limiting case, and thus far there hasn’t been an experiment demonstrating the plausibility of viewing them simultaneously and how they act upon each other. In semi-classical gravity these “weird” aspects like information loss in a black hole only arise when we actually consider them together, which is not something we have observed yet in a lab, so I don’t see the basis of thinking it is wrong.
You seem to suggest that thinking it is wrong implies the laws of physics change, but I’m not really sure what is meant by this. Is semi-classical gravity not a self-consistent mathematical framework?
To oversimplify with another example from the theory, assume that planet earth was in superposition between two states with a non-zero separation. Semi-classical gravity says the distribution of the gravity field would be split evenly between the two points, but observing such a state is impossible as it must decohere into 100% of the mass being either in one point or the other. It simply doesn’t make sense when we try to apply quantum maths to gravitationally-significant objects because gravity isn’t a quantum field.
So yes, the predictions made by semi-classical gravity diverge from reality when faced with extreme masses, but that theory was only ever intended to be an approximation. It is useful and consistent with reality under certain ranges of conditions, but we shouldn’t jump to the conclusion that physics breaks from all known fundamentals in the presence of large masses when the simpler answer is that this is a case where the approximation is wrong. A more complete theory will be able to accurately explain physics across a wider range of conditions without requiring the untestable assumption that there are places where the rules don’t apply. We’ve got a good reason to believe that the rules of physics don’t change in the fact that no matter where we look the rules seem to always have been the same and all prior divergences from the model could be explained by better models.
The problem in physics is that we have two models that describe reality with absurd mathematical precision at different scales but which seem to be fundamentally irreconcilable. But we know they must be, because reality has to be assumed to be consistent with itself.