Hacker News Reader

Comment by danbruc | Hacker News Reader

This does not help at all, quantum systems - even messy complex ones, boxes containing cats and air - evolve unitarily and do not destroy a superposition. The box and the air and the cat can interact all they want, you will still have a superposition of cat dead - and box and air in some state - and cat alive - and box and air in some other state. And no point are we addressing how we get rid of one half of the entangled state. And even if I see the box wiggle a bit, we only declare that me watching the box somehow was an measurement and that collapsed the wave function without answering the question why I did not simply get entangled with the box, too.

View on HN

> quantum systems - even messy complex ones, boxes containing cats and air - evolve unitarily and do not destroy a superposition

This is wrong, they do collapse, you can test this using the double slit experiment. No human has to be there for the wavefunction to collapse and thus break entanglement.

If what you said was true then this wouldn't be a part of physics, physics is only about stuff we can measure not nonsense that has no meaning.

It's not at all clear how, when, and why wave function collapse happens. The only known quantity there, decoherence, is related to interaction between a coherent quantum system and a noisy decohered environment, which causes the original system to decohere itself.

Theoretically, as far as it is known today, it is possible for a quantum system of any size, and at any temperature, to remain in a coherent state, if it could be perfectly isolated from the rest of the world.

Any limits to this process, where a perfectly isolated system would undergo wave function collapse without any outside interference, would be a Nobel-prize worthy discovery and would constitute a new theory of physics: quantum mechanics and QFT allow no such process.

Edit: and the double slit experiment is unrelated - the screen and the slit are noisy classical systems that are not in any way isolated from the rest of the classical world. In principle, if you performed the experiment using a screen and slits made entirely of atoms entangled with the emitter, and isolated perfectly from the rest of the world and the CMB and everything, you'd get a different result: this is what QM predicts.

> quantum mechanics and QFT allow no such process

Double slit experiments shows collapse happens without any outside interference, nothing outside of the room has to do anything for the resulting measurements to show collapse happened. You can then afterwards enter the room, look at the recorded data and see that collapses did happen.

So you must have misunderstood what those theories says about collapses. They don't say exactly when collapses happens or what causes them, that is the unsolved problem, but we can know that they do happen.

In the normal experiments we perform, the screen is already a classical system, collapsed / entangled with the rest of the "classical world" (depending on the interpretation). When the particles interact with the screen, of course their wave function collapses as well.

But what if instead you had a screen that was itself a quantum system? You'd then use the Schrodinger equation without any collapse (without the Born rule) to describe the results. You'd have to perform other measurements on the quantum screen to obtain a classical result that can be interpreted, of course. But nothing in QM says that there is some size limit, so in principle it should be possible to prepare a quantum system of any size, including the size of a typical double slit experiment screen, that doesn't collapse.

Quantum mechanics, i.e. the Schrödinger equation, and wave function collapse are incompatible. While not measured a quantum system evolves unitarily, when measured the quantum system ends up in the measured Eigenstate which is a non-unitary change. But any measurement device is also a quantum system and therefore the combined system should also evolve unitarily. And now we have a serious problem to which nobody has a generally accepted answer.

What is deficient about the original explanation that the wave function is not real but rather a scientific theory? The issue has been unresolved for about a century now, it's relatively unsatisfying but hasn't held the science back in any way.

The problem with that idea is that we don't have any other model, and little room for it. We already know from Bell's inequality and the experiments that verified it that there can't be underlying variables that we haven't detected yet (or if they exist, we have to rethink special relativity and by extension QFT). It's also quite hard to explain why the wave function is so good at predicting the results of experiments if it's not a part of the underlying reality.

There is a thing we call the universe and it does what universes do. What physics does is searching for models - mostly described in the language of mathematics - that describe the universe or parts of it. Some parts of the universe are well modelled by the Navier-Stokes equations, we also call those parts of the universe water [1]. But upon closer inspection one notices that description is not perfect, one can obtain a more accurate model if one treats the water bits of the universe as a collection of an enormous number of particles, water molecules. And we can keep going, treat molecules as collections of atom, break atoms into electrons and nucleons, split the nucleons into quarks and gluons, who knows how deep the rabbit hole goes.

Maybe we will even eventually find the ultimate model, a model that describes every aspect of the world perfectly, even though it is not obviously clear how we would determine that we did not miss some aspect of the universe in our models. But even then the map would not be the territory, the model would not be the universe or a piece of it. Imagine a wave function would perfectly describe an electron, would we be justified to say that an electron is a wave function? No, because for them to be the same thing, it is not sufficient that the wave function has all the properties of the electron, it would also have to have only the properties of the electron. But the wave function is an equation, I can write it down or square it, things I can most certainly not do with the electron.

So the wave function is of course only a mathematical description of a quantum system and not a real thing out there in the universe. But that is not the problem, the problem is that we have two seemingly contradictory descriptions of quantum systems. We have the Schrödinger equation describing the unitary evolution of isolated quantum systems and it works really well. But we also have the Born rule and the collapse postulate which tell us the probability distribution of measurement outcomes and how measurement changes the state into the measured Eigenstate, a non-unitary change, and this also works really well. But there is the problem, on the one hand a measurement - whatever that exactly is - seems to change the state non-unitarily but on the other hand a quantum system together with a measurement device is just a bigger quantum system and should therefore evolve unitarily. Something has to give.

As you say, that is very unsatisfying, but it was also not a complete roadblock, shut up and calculate worked pretty well. But I would not confidently declare that this does not hold us back at all, who knows what progress can be unlocked by solving the issue? It might of course also be mostly inconsequential but I guess that is something we will only find out after the fact.

[1] Or fluids if you want to be pedantic.

Hey thanks for your thoughtful response.

In terms of what we may/not be missing out on, I say that because it seems few of the possibilities remaining after a century of investigation (many worlds, relational/information based approaches, etc) contain any new physics or scientific insight.

Eg if we could somehow tell that MW or pilot wave theory is "correct", we would still have nothing to do but shut up and calculate the same things, no?

If unitary evolution continues in some sense, we would still need to throw factors away in order for the theory to remain predictive to us.

Particles are only observed to be in a single place at a single time. My understanding is that it may be that we lose the perspective necessary to see them elsewhere, since we are embedded in the same eigenstate (or something close to it). But the task of science is confined to observables. I question, what are the scientific questions that collapse postulate holds back from being solved?

> evolve unitarily and do not destroy a superposition

Unitary evolution can destroy superposition of the small subsystem by introducing negligible amount of superposition into large system.

Not sure if that is true, but assuming it is, does this really count as destruction instead of moving it around and spreading it? After all unitary evolution is reversible and I could therefore reconcentrate the superposition into the small system. It would however seemingly be a good explanation how to get from the decayed non-decayed superposition of a few atoms to an almost alive or almost dead cat. This is however also what makes me doubt your statement, if true, why are we still puzzled about Schrödinger's cat?

Well the naming is a different question, but it is consistent with observation of what we call wavefunction collapse.

In Schroedinger cat experiment, the evolution is not thermalization but rather entangling the whole cat with atom state. I don't think we're really puzzled about it now, since we know that macroscopic objects can be in superposition. So cat, theoretically, can be in superposition of alive and dead until measured. I don't think there is any contradiction or confusion here physics-wise.

Reply on news.ycombinator.com