Sean Carroll podcast on many worlds interpretation

[DarMM]

MWI, of course, satisfies Bell's theorem, just like other interpretations. I never remember which way John Bell wrote the inequality. I assume QM breaks the inequality? And hidden variables would uphold the inequality?

Local Classical theories satisfy Bell's inequality, all interpretations of QM break it.

Bell's theorem has nothing to do with discarding or keeping parts of the wavefunction.

 

[OCR]

My spoiler post was half in jest. . .

Yeah, mine was too. . . .

it's not like Minnesota Joe gave away the end of House season 4 or anything.

Yes, that would be unforgivable. . . .
Wait, what!?

Sean Bean dies.

Oh, my. . .! .

Lol. . . carry on.
.

 

[Heikki Tuuri]

Bell's theorem has nothing to do with discarding or keeping parts of the wavefunction.

Let me explain. To calculate correct results, you have to use a QM superposition state for the EPR two particles. That is a "wave function".

If there is some mechanism which codes in hidden variables the spin state already right at the preparation, then the system is classical and does not have a wave function in the QM sense.

In the double slit experiment we need the wave function to calculate the pattern on the screen. The Bohm model utilizes both the wave function and hidden variables. It would not work without the wave function.

Are there wave phenomena where one could do with just a few, real number, hidden variables and would not need any kind of a wave function? Yes! A free planar wave is an example. We do not need complex information in the planar wave pattern, because the wave pattern is trivially simple.

 

[PeterDonis]

I assume QM breaks the inequality?

Yes. Which means that saying Bell's theorem is "true" and the MWI "satisfies" Bell's theorem is a rather confusing way of speaking.

you cannot discard parts of the wave function if you want to calculate correctly

You can if decoherence has happened and you are only concerned with one decohered branch (the one corresponding to the measurement result you observed). We've been through this before.

 

[DarMM]

Let me explain. To calculate correct results, you have to use a QM superposition state for the EPR two particles. That is a "wave function".

If there is some mechanism which codes in hidden variables the spin state already right at the preparation, then the system is classical and does not have a wave function in the QM sense.

In the double slit experiment we need the wave function to calculate the pattern on the screen. The Bohm model utilizes both the wave function and hidden variables. It would not work without the wave function.

Are there wave phenomena where one could do with just a few, real number, hidden variables and would not need any kind of a wave function? Yes! A free planar wave is an example. We do not need complex information in the planar wave pattern, because the wave pattern is trivially simple.

Bell's inequality is proven in a general framework called the ontological models framework. This framework essentially encodes the idea of a "classical" theory, i.e. causal and with one world. If such a classical theory is local we get the inequality. Observed correlations in for example the famous Aspect experiments break this inequality.

None of this is really related to discarding or keeping parts of the wavefunction, it just says what facets of a local classical theory must be lost to avoid contradicting the predictions of QM. The alternate theory may include the wave function (Bohmian Mechanics) or it may not (various alternate causality models). Thus the theorem isn't really about retaining parts of the wave function.

 

[Minnesota Joe]

Bell's theorem says that we cannot assume local hidden variables which would determine the measurement result at the ends of the EPR experiment setup.

Quantum mechanics says that Bell's theorem is true. It has been empirically demonstrated many times.

MWI, of course, satisfies Bell's theorem, just like other interpretations. I never remember which way John Bell wrote the inequality. I assume QM breaks the inequality? And hidden variables would uphold the inequality?

Yes, sorry, my phrasing was terrible. I am not saying that for MWI there are no Bell violations.I was taking a point of view where all outcomes occur, not the point of view of any single world. So forget about that please. Hopefully I explain what I was on about better below.

My original question is what sense is the MWI considered a local theory. And--this should go without saying but bears repeating--I'm not claiming I'm correct in my understanding, just discussing it in hopes I will be corrected by all you smart people here if I'm wrong.

Einstein protested the "spooky action at a distance" in the EPR experiment. In MWI, there is no spooky action at a distance for the simple reason that all data processing happens in the head of a single scientist, and there are no great distances inside a single head.

I don't understand your explanation in terms of the head of a single scientist. But "spooky action at a distance" is the sense of local I'm getting at.

In EPR, Alice becomes entangled locally with electron 1 and on the other end Bob becomes entangled locally with electron 2. (Electrons 1 and 2 were themselves entangled locally at the beginning of the experiment of course.) But on MWI entanglement means that multiple worlds result. Yet a person in any particular world only ever sees a subset of all the outcomes and this happens in such a way as to explain the correlations or the "spooky action at a distance". And all the physics for Alice, say, occurred locally. Nothing but the Schrodinger equation is involved.

Contrast that with de Broglie-Bohm where there really is spooky action at a distance. Non-local hidden variables.

So sometimes I hear people say that MWI goes through the horns of Bell's dilemma and provides a local hidden variable theory or something like that. It is both local and realist. At the cost of many worlds of course.

Note that Bell's theorem is just a special case of a general rule: you cannot discard parts of the wave function if you want to calculate correctly. Hidden variables would mean that we discard the relevant wave function immediately after we have prepared the two particles in the EPR experiment.

I'm not sure what you mean...it almost sounds like you are saying you won't get the correct statistics if you use the "collapsed" state after the first measurement (the preparation I guess). I think that is true. You have to use the full entangled state if you are measuring a different property of electron 2.

Is that what you mean? So "discard" is like "collapse"?

A classical analogue: if you want to calculate the route of a toy boat on waves of water, you need to know the full wave pattern. You cannot discard the information about the waves and calculate from the location of the boat alone.

I like the analogy, so I'll be sure to steal it. ;)

 

[PeterDonis]

on MWI entanglement means that multiple worlds result

Yes, and this is how MWI evades the conclusion of Bell's Theorem, since the proof of the theorem assumes that measurements have single outcomes, but in the MWI measurements have multiple outcomes (every possible outcome is realized).

 

[Minnesota Joe]

If the many worlds interpretation is local and realist that is an attractive feature it must be admitted. And since it is just the Schrodinger equation it can be extended to relativity (Klein-Gordon is an example) so it is local in that sense too.

The cost is only a very, very, very large number of worlds, possibly an uncountable infinity of worlds. Come on, you know you want to buy it. Perhaps on credit.

 

[PeterDonis]

If the many worlds interpretation is local and realist

That depends on what you mean by "local" and "realist". Measurements having multiple outcomes violates many people's definition of "realist". And having wave functions that entangle spatially separated systems violates at least some people's definition of "local".

 

[Michael Price]

@Minnesota Joe ,
. I never remember which way John Bell wrote the inequality.

Nah, I can never remember that, either!

Well Copenhagen views takes the formalism "as is" as well. It's just that they read it differently, i.e. that $\psi$ is a generalized probability distribution rather than a physical degree of freedom.

Except that your 'generalized' probability distribution is nothing like a probability distribution.

 

[DarMM]

Except that your 'generalized' probability distribution is nothing like a probability distribution

That's contrary to the last thirty years of Quantum Information. The wave-function behaves as a generalized probability distribution, up to obeying things like a de Finetti theorem.

https://arxiv.org/abs/quant-ph/0601158
For an example that makes it easier, one can see that Classical probability with an epistemic limit behaves very like QM.

 

[Minnesota Joe]

That depends on what you mean by "local" and "realist". Measurements having multiple outcomes violates many people's definition of "realist". And having wave functions that entangle spatially separated systems violates at least some people's definition of "local".

Both are good points. Encountering some of these different definitions is why I was discussing locality in the first place. I haven't even gotten around to considering the "realist" angle and only lumped the multiple outcome problem into MWI's probability problem, so the comment about multiple outcomes is appreciated. I'll keep in mind that there is disagreement on the matter.

 

[vanhees71]

That depends on what you mean by "local" and "realist". Measurements having multiple outcomes violates many people's definition of "realist". And having wave functions that entangle spatially separated systems violates at least some people's definition of "local".

Indeed, a measurement device which doesn't show a clear outcome when measuring something usually has to be sent to the shop to be repaired. It's not considered to prove the existence of many worlds in a experimentalist's lab. SCNR.

 

[Minnesota Joe]

Indeed, a measurement device which doesn't show a clear outcome when measuring something usually has to be sent to the shop to be repaired. It's not considered to prove the existence of many worlds in a experimentalist's lab. SCNR.

But you wouldn't complain about several measurement devices that all reliably measured what they were supposed to, right?

I can't even make out an experimentalist's lab from this thread. If there is one, it is much too far away.

 

[vanhees71]

Well, that's a bug rather than a feature in all these debates about "interpretation"

 

[Michael Price]

That depends on what you mean by "local" and "realist". Measurements having multiple outcomes violates many people's definition of "realist". And having wave functions that entangle spatially separated systems violates at least some people's definition of "local".

Local means the relativistic Lagrangian dynamics are just functions of X and not X, Y etc. QFT fulfils this requirement and as that is all there is to MWI, MWI is local. No non-local FTL collapse.