Is Quantum Mechanics Non-Local or Can Local Models Explain the Universe?

[The Dagda]

The way MWI, with a "real" wavefunction, and "deterministic" evolution, nevertheless gets out of Bell's theorem is simply this: in Bell's theorem, you need unique and definite outcomes at Alice and Bob for each experiment, and in MWI, that's not the case: Alice didn't see "up" or "down" ; there is AN alice which saw "up" and ANOTHER alice which saw "down". And the correlation only happens when A Bob compares his results with AN alice. But at that point, there is no distance anymore between them, and they can influence each other (that is to say, the probability to see a specific "alice and bob pair" can depend as well on the alice as on the bob under consideration).
In Bell's proof, you need a single definite outcome at both sides when they are still spacelike separated.

In other words, Bell assumes the "dice are thrown" at Alice and Bob. In MWI, the dice are never definitely thrown.

Yes but this is the same as CI for all practical purposes, if so what's the point of it? I mean I can dream up anything to make QM deterministic does that mean my dreams exist?

If in experiment QM is random, and in MWI which in experiment appears random where's the difference and isn't that just semantics?

 

[The Dagda]

You must understand what "universe" means in MWI: it means "essentially orthogonal term in the wavefunction". So "creating a universe" comes down to "splitting a single term into two others".

If you have something like |psi> = blah ... + |moon>|sun>|earth>|filled-mosquito>|ocean>... +...

then the explicit term is "one universe". Now, if your mosquito evolves into:
|filled-mosquito> ===> |farting-mosquito> + |constipated-mosquito>

and we fill this in the original wavefunction:

|psi> = blah ... + |moon>|sun>|earth>(|farting-mosquito> + |constipated-mosquito>)|ocean>... +...

and we work this out, then:

|psi> = blah ... + |moon>|sun>|earth>|farting-mosquito>|ocean> + |moon>|sun>|earth>|constipated-mosquito>|ocean> ... +...

and lo and behold, where we had 1 term, we now have 2 terms, so we "created a universe".

Yes.

Vanesch is God.

 

[colorSpace]

MWI seems to remove the randomness from the collapse, but isn't the collapse still there? In the double-slit experiment, you still register each photon at a specific location on the screen.

There is still the wave/particle duality which results in seeing a photon register at a specific location, with the wave-like interference only showing up in the statistical distribution.

So isn't MWI saying that there are many non-random collapses, rather than that there would be no collapse?

 

[The Dagda]

MWI seems to remove the randomness from the collapse, but isn't the collapse still there? In the double-slit experiment, you still register each photon at a specific location on the screen.

There is still the wave/particle duality which results seeing a photon register at a specific location, with the wave-like interference only showing up in the statistical distribution.

So isn't MWI saying that there are many non-random collapses, rather than that there would be no collapse?

No because all wave functions are actualised in another universe, all you do is select one to measure, which to all appearances is random, all possible and presumably infinite wave functions except the one you measure are resolved in another universe thus random instead of the true random of CI, and thus deterministic not probabilistic, as said the die has already been rolled. This sounds like hidden variables repackaged to me though so I think its ruled out but then who am I to judge?

 

[Dmitry67]

'Qunatum collapse' in MWI is a mathematical process called quantum decoherence:
http://en.wikipedia.org/wiki/Quantum_decoherence

 

[colorSpace]

No because all wave functions are actualised in another universe, all you do is select one to measure, which to all appearances is random, all possible and presumably infinite wave functions except the one you measure are resolved in another universe thus random instead of the true random of CI, and thus deterministic not probabilistic, as said the die has already been rolled. This sounds like hidden variables repackaged to me though so I think its ruled out but then who am I to judge?

That's just explaining-away the randomness. To me the real paradox of quantum physics is that the probabilities of flying through either slit will interact with each other, but in the end the photon will appear only in one place, not smeared out like butter. And trying to measure the path will strangely make the interference go away. MWI doesn't seem to change that.

 

[WaveJumper]

Vanesch is God.

Wait, wait... I reject decoherence/MWI mainly on the fact that it cannot explain the presence of liquid water.

The molecule of water contains three atoms in the H20 molecule. The H atom has only one electron. The molecule of water relies on this electron to be in multiple places all at once(as waves), so that a covalent bonding can take place between the atoms. Decoherence is an irreversible process, once decohered waves become particles, and that's why according to decoherence we observe a a "physical" universe(how physical is another topic). But...

If the electrons of the H atom have decohered 4.5 billion years ago into single electrons, why do we observe liquid water instead of gaseous H and oxygen? If there is no covalent bonding between the atoms in the molecule of water, the molecule H20 would fall apart and we wouldn't see homogenous liquid water. And I've just opened a beer, and it's liquid and doesn't turn into H and O(luckily).

 

[Dmitry67]

colorSpace, check the wiki article. It explains why you detect photon in only one point.

 

[Dmitry67]

WaveJumper, please read the articel again.
"If the electrons of the H atom have decohered 4.5 billion years ago into single electrons" - electrons never decohere until you entangle electorns with a thermodinamically irreversible system with a huge number of states.

 

[colorSpace]

colorSpace, check the wiki article. It explains why you detect photon in only one point.

That's a rather long article. As far as I can tell, it explains what happens to the other parts of the wavefunction after measurement. But the wave function still remains an odd thing of complex-interacting probabilities. The fact that the photon appears only in one place is an expression of that the wavefunction is still expressing a probability, rather than a continuous physical property. Or I haven't found the place in the article where this is explained otherwise.

 

[WaveJumper]

WaveJumper, please read the articel again.
"If the electrons of the H atom have decohered 4.5 billion years ago into single electrons" - electrons never decohere until you entangle electorns with a thermodinamically irreversible system with a huge number of states.

So? What are you saying? That electrons in the atoms of water have not decohered?? Then may i ask how do you see water? Do you think you see wavefunctions?

 

[Dmitry67]

2 colorSpace

Decoherence shows how a macroscopic system interacting with a lot of microscopic systems (e.g. collisions with air molecules or photons) moves from being in a pure quantum state—which in general will be a coherent superposition (see Schrödinger's cat)—to being in an incoherent mixture of these states. The weighting of each outcome in the mixture in case of measurement is exactly that which gives the probabilities of the different results of such a measurement.

So after you detect a photon in a matrix of your camera, the interference is lost and you get for 1 megapixel matrix something like :

1/1000000 * (photon detected by pixel at (0,0)) + ...

 

[Dmitry67]

So? What are you saying? That electrons in the atoms of water have not decohered?? Then may i ask how do you see water? Do you think you see wavefunctions?

In order to decohere some electrons it is not enough to 'look' at water, you need to measure *individual* propeties of some electrons.

Until then they are not decohered.

As an example, you can make lens of water and perform all sorts of interference experiments using such lens and reflection from the surface of water. Interference is not lost in all these cases.

 

[WaveJumper]

In order to decohere some electrons it is not enough to 'look' at water, you need to measure *individual* propeties of some electrons.

Until then they are not decohered.

As an example, you can make lens of water and perform all sorts of interference experiments using such lens and reflection from the surface of water. Interference is not lost in all these cases.

And very definitely i see liquid water because all the particles have already decohered. Otherwise, very definitely, i wouldn't be able to see water. No one has ever seen a wavefunction. You see physical objects because their wavefunctions have "collapsed" to a single state.

The case with water is special, because H has only 1 electron, which means that if it's decohered, water molecules would fall apart, which is very definitely not the case.

 

[colorSpace]

2 colorSpace

So after you detect a photon in a matrixof your camera, the interferenc eis lost and you get for 1 megapixel matrix something like :

1/1000000 * (photon detected by pixel at (0,0)) + ...

Right. Once the wavefunction is decoherent, the probabilities can't cancel each other out (interact) anymore, as they do in an interference pattern.

Let me approach it from a slightly different side:

In CI, the selection of a possible result has no mechanics which explains why that result happens, and not one of the other possible ones. The missing mechanics are called "randomness".

Now, MWI assumes that all possibilities remain real (unless they cancel each other out, which is a very strange thing), and so removes the name "randomness". But the mechanics of why things happen this way in one world, and that way in another, are still missing. How is it possible that a photon which could have been seen by one observer (if there had been one) and a photon which could have seen by an alternate observer, cancel each other out, and are not seen by anyone? In MWI, that's just two complex numbers adding up to zero, but how can two photons disappear by nothing else than mathematical addition? The mechanics for that are still missing, they just don't have the name "randomness" anymore.

Now an MWi proponent could say: that's simply what the wavefunction says will happen. However, the same thing could be said by the CI proponent: it's simply random according to the wavefunction. That's not a scientific difference or even Occam's razor, it just seems a personal preference: Would you rather believe in randomness, or in trillions alternate versions of yourself and everyone else coexisting in trillions of alternate universes.

 

[Dmitry67]

1
We start with what you call "back-references" but it is going back to the true fundamentals of science...the experiment. Get too far from this and you begin arguing about things no-one can observe and that is a theological debate, not science.

2
No of course not. At some point the mathematical terms must be related to the physical...that by the way is the true interpretation of the theory. How a ket or Hermitian operator relates to an actual experimental device.

3
How can we define anything without "we"? Science is what scientists do.

I think I found why our views are so different.

3
Take the Classical mechanics. let's take F=ma. Do you see any 'we', 'our knowledge about'? For that reason Hilbert wanted to find the axiomatisation of physics.

1
No, the fundamental of science is a THEORY. An experiment is just a tool to prove or disapprove it. Without theories, the science would be just a heap of recepies... like alchemistry... This is what science about: the underlying formulas!

Let me ask some questions again

What do you think about the max Tegmark program 'physics from scratch'
We define TOE is a pure mathematical form, TOE(f)=0. So there are only equations, no words. Then we derive everything from there. We ask 'what a complicated system would percieve?" building frog's view from the equations?

Or do you believe that there are some 'physical' axioms which can not be expressed in forms of equations?

Do you agree that MWI is the best to be expressed in the TOE(f)=0 form?

Do you agree that MWI (when we pay a contre-intuitive price of accepting parralel realities) saves not only determinism, but also realism?

 

[Dmitry67]

1
Now, MWI assumes that all possibilities remain real (unless they cancel each other out, which is a very strange thing), and so removes the name "randomness". But the mechanics of why things happen this way in one world, and that way in another, are still missing.

2
it just seems a personal preference: Would you rather believe in randomness, or in trillions alternate versions of yourself and everyone else coexisting in trillions of alternate universes.

1
This is a very good question, I remember I was thinking about it when I learned the MWI...
I would say this is "no-issue": let's take dead/alive cat.
So what MWI predicts is that BOTH observers would say "I understand that there are alternative branches of reality where cat is alive/dead, but why *I* get this result, not another one? Why *my* consiousness is in THAT branch of reality?"
This is exactly what happens!
So MWI predicts BOTH (any) observer to be surprised and believe in randomness!

2
True
But again, waht is wrong with trillions alternative universes when our universe is INFINITE (in space at least)? If you multiply infinity by billion, the result is not bigger.

 

[Dmitry67]

And very definitely i see liquid water because all the particles have already decohered. Otherwise, very definitely, i wouldn't be able to see water. No one has ever seen a wavefunction. You see physical objects because their wavefunctions have "collapsed" to a single state.

The case with water is special, because H has only 1 electron, which means that if it's decohered, water molecules would fall apart, which is very definitely not the case.

Sorry, this is a nonsense
Please check how QED explains why light moves slower then C from water.

Hint: photons are slowed down because they are absorbed/reemitted by atoms, putting these atoms for a very short peiod of time into an excited state. That causes a delay. However, the process is absolutely reversible, so after photon had passed thru the water, there is no way to tell which atom had actually slowed the light down.

 

[colorSpace]

1
This is a very good question, I remember I was thinking about it when I learned the MWI...
I would say this is "no-issue": let's take dead/alive cat.
So what MWI predicts is that BOTH observers would say "I understand that there are alternative branches of reality where cat is alive/dead, but why *I* get this result, not another one? Why *my* consiousness is in THAT branch of reality?"
This is exactly what happens!
So MWI predicts BOTH (any) observer to be surprised and believe in randomness!

2
True
But again, waht is wrong with trillions alternative universes when our universe is INFINITE (in space at least)? If you multiply infinity by billion, the result is not bigger.

To 1) No problem with the surprise factor, the appearance of randomness is a given.
However unless there is decoherence (which is the trivial case in both MWI and CI), MWI says the probabilities interact simply by mathematical operation, the physical process, the mechanics are still missing.

To 2) The problem is not the largeness of the number, but the belief that there would be (many) multiple alternate versions of each living being and non-living thing coexisting at the same time. Why would they not collide? Just because they are different terms in a wave function which can't be added mathematically? Doesn't that require a "how", a physical explanation?

 

[WaveJumper]

Sorry, this is a nonsense
Please check how QED explains why light moves slower then C from water.

Why should I? What the heck does this have to do with anything that i was talking about?

Hint: photons are slowed down because they are absorbed/reemitted by atoms, putting these atoms for a very short peiod of time into an excited state. That causes a delay. However, the process is absolutely reversible, so after photon had passed thru the water, there is no way to tell which atom had actually slowed the light down.

No, what you say is completely not what i was talking about. I never spoke of light or c. You are talking about a subject of your choice, that i had no intention of discussing. Hint: Decoherence does not happen only when waves of matter hit photon waves.

 

[Dmitry67]

1. Mechanics is described in the Wiki article. It actually begins from the chapter "Mechanisms" :) There are all formulas you might need.

2. Why would they not collide? - because of the loss of coherence. Non-diagonal elements of the density matrix vanish, and branches lose the ability to influence each other. Check the "Density matrix" chapter.

 

[Dmitry67]

Why should I? What the heck does this have to do with anything that i was talking about?

Because it is directly related to the subject we discuss:
OPTICAL effects (refraction, reflection, etc) are reversible and collapse-free. Put an aquarium behind 2 slits and still you will be able to see an interference pattern. This is an absolute proof that when photons pass thru water, they do not leave and 'which-path' traces, so there is no 'collapse' at all. So you can not say that 'photon had been slowed down by this and this hydrogen atom'. If you can say it then you know thepath and there would be no interference.

 

[Hurkyl]

Just because they are different terms in a wave function which can't be added mathematically? Doesn't that require a "how", a physical explanation?

If the wave function represents the physical state, then that is a physical explanation.

Why would they not collide?

MWI worlds are not Sci-Fi parallel universes; they do not consist of distinct collections of matter living in parallel dimensions. Worlds are just another wave phenomenon, a product of evolution according to the Schrödinger equation. In principle, they could interfere, but that's incredibly unlikely for large systems, and essentially impossible for nonisolated systems.

 

[The Dagda]

That's just explaining-away the randomness. To me the real paradox of quantum physics is that the probabilities of flying through either slit will interact with each other, but in the end the photon will appear only in one place, not smeared out like butter. And trying to measure the path will strangely make the interference go away. MWI doesn't seem to change that.

Yeah people seem to get all goose pimply when we say it's absolutely random, not chaotic but there is no way to know what state photons are in without a measure. MWI is I suppose just a semantic/philosophical issue, it makes no difference to scientists experiments or what actually is, except they have the notion that the wavefunction an abstract mathematical object is actually representing the pictorially and really the nature and image of photons, which of course we can't know. Strange times.

 

[vanesch]

Yes but this is the same as CI for all practical purposes, if so what's the point of it? I mean I can dream up anything to make QM deterministic does that mean my dreams exist?

First of all, the fundamental difficulty with QM in the CI is not its random character, but rather two other things:
1) the distinction in physical description of what is "observation" and what is "physical process". In other words, there are magical things out there which are called observers, and whenever they enter the picture, the way to solve the problem is different and mathematically incompatible with whatever are the rules when we consider physical processes. In other words, the way quantum theory is usually done (as in CI), it is impossible to analyze, from within the theory, what is the physical process of observation.
You cannot analyze detectors quantum-mechanically, in principle. You cannot write the Schroedinger equation of a detector. Of course, you can, but then your detector is no detector anymore.

2) The problem with Bell, if we insist on locality.

If in experiment QM is random, and in MWI which in experiment appears random where's the difference and isn't that just semantics?

Again, the problem is not the randomness per se. It is the fact that in CI, it is in principle impossible to describe the detection process. This is the problem that MWI tries to solve, and you get as two bonuses, that Bell isn't a problem anymore, and that on top of that, the objective reality (contrary to the subjectively perceived reality) became deterministic.

In how much this is actually *true* is a totally different (and in my opinion even irrelevant) matter. What you do get from it is a much clearer picture of how things are behaving *according to quantum theory*, because there's a whole lot of fuzzy handwaving contradiction that is gone now. It's why I like it.

Yes, I'm god, but don't tell anyone, I'm incognito

 

[vanesch]

To 2) The problem is not the largeness of the number, but the belief that there would be (many) multiple alternate versions of each living being and non-living thing coexisting at the same time. Why would they not collide? Just because they are different terms in a wave function which can't be added mathematically? Doesn't that require a "how", a physical explanation?

The answer to that is simply the fact that the time evolution operator is a linear operator. You cannot influence the result of a linear operator acting on something, by adding something to the argument.

U (a + b) = U(a) + U(b).

U(a) is independent of whether there was a b or not.

Now, there is a "non-linear" part somewhere, which is the amplitude-> probability change. Indeed, if U(a) and U(b) have a common component, then the amplitude -> probability change will be influenced by the presence or not of b and hence U(b).

However, if a is a vector in many many dimensions and so is b, and they are essentially orthogonal, then (because U is not only linear, but also unitary), U(a) will also be orthogonal to U(b). As such, in the amplitude -> probability transition, you will not have any effect of the presence or not of b (and U(b) ).

That's what decoherence essentially tells you. Whenever you get hopelessly entangled, every "world" is essentially orthogonal to any other, and this will remain so. So the presence of another world or not will not influence whatever happens to one.

As long as a and b are not hopelessly orthogonal, you do have effects in the amplitude -> probability transition of the presence or not of b, and that is exactly what we call quantum interference.

 

[The Dagda]

First of all, the fundamental difficulty with QM in the CI is not its random character, but rather two other things:
1) the distinction in physical description of what is "observation" and what is "physical process". In other words, there are magical things out there which are called observers, and whenever they enter the picture, the way to solve the problem is different and mathematically incompatible with whatever are the rules when we consider physical processes. In other words, the way quantum theory is usually done (as in CI), it is impossible to analyze, from within the theory, what is the physical process of observation.
You cannot analyze detectors quantum-mechanically, in principle. You cannot write the Schroedinger equation of a detector. Of course, you can, but then your detector is no detector anymore.

2) The problem with Bell, if we insist on locality.
Again, the problem is not the randomness per se. It is the fact that in CI, it is in principle impossible to describe the detection process. This is the problem that MWI tries to solve, and you get as two bonuses, that Bell isn't a problem anymore, and that on top of that, the objective reality (contrary to the subjectively perceived reality) became deterministic.

In how much this is actually *true* is a totally different (and in my opinion even irrelevant) matter. What you do get from it is a much clearer picture of how things are behaving *according to quantum theory*, because there's a whole lot of fuzzy handwaving contradiction that is gone now. It's why I like it.

Yes, I'm god, but don't tell anyone, I'm incognito

See that's why I don't like it, it's too convenient. It's like string theory: what physics would be like if God was a mathematician. But I suppose it's all hand waving at the end of the day to try and solve handwaving issues.

As physicists say the universe has no regard for what you expect to be true only what is.

 

[vanesch]

See that's why I don't like it, it's too convenient. It's like string theory: what physics would be like if God was a mathematician.

I'm not a mathematician (but I play one on TV...)

The way I view MWI is not as some "ultimate truth", but rather as the bare bones logical consequence of the theory of quantum mechanics if you want to keep to the math and the logic all the way down. The price to pay is that it doesn't fit at all with any preconceived ideas of what could be reality, but what you win from it is a crystal-clear view on the wheels and gears of the quantum-mechanical formalism, and that all so-called paradoxes disappear in a puff of logic. There are no difficulties anymore in viewing any EPR experiment, or any quantum eraser experiment or anything. It all comes out very clear.

(ok, what becomes incomprehensible in this view is general relativity of course...)

In other words, to me, MWI is the picture I try to keep in my mind when doing quantum mechanics, in order to "understand" and "get a feeling" for how the theory behaves. You can't do that if you are having a machinery which produces you apparently paradoxical situations, and so the fact of removing those, and even removing all the ambiguity of "do I measure this here now or not ? " or "is the information still available or not ? " or "did this polarizer actually measure the polarisation or not" and all the handwaving that comes with CI-style views on some more subtle experiments disappears completely from an MWI viewpoint.

Questions raised by CI, such as: if the detector clicks, but I don't look, is there a measurement or not ? And if I destroy the record ? And if I throw it in a black hole ? and over which one can have heated philosophical debates become a trivial issue from the MWI viewpoint.

 

[The Dagda]

I'm not a mathematician (but I play one on TV...)

God is on TV playing a mathematician! :D I must notify The Vatican?!??

 

[Dmitry67]

but what you win from it is a crystal-clear view on the wheels and gears of the quantum-mechanical formalism, and that all so-called paradoxes disappear in a puff of logic. There are no difficulties anymore in viewing any EPR experiment, or any quantum eraser experiment or anything. It all comes out very clear.

Very good description of the 'enlightment' I felt whan I realized the beauty of MWI!

The second enlightment I experienced when I read the Max Tegmark article about the Mathematical Universe... what do you think about it?