A Copenhagen: Restriction on knowledge or restriction on ontology?

[bobob]

This isn't true if retrocausal influences occur only within the light cone.

(1) Huh? If you have spacelike separated events, the events cannot be time ordered. Therefore, the time odering is frame dependent and in particular can be made simultaneous. Neither can be the cause of the other. Neither is in the others light cone.

(2) Events which are within each other's light cone are, by definition, time ordered. If there is a cause and effect, which is the cause and which is the effect are defined by their time ordering. Retrocausal in this case would mean closed timelike world line, since all of the world lines connecting causes and effects are timelike.

To state otherwise is to flat out say relativity is incorrect.

 

[atyy]

The argument is essentially the failure of hidden variable theories and how fine-tuned they have to be. However like most interpretative arguments it's not definitive.

Hidden variables must be in some sense "fine tuned" as proved by Wood and Spekkens. However, they also argue that some forms of fine tuning are more acceptable than others.

https://arxiv.org/abs/1208.4119"Valentini’s version of the deBroglie-Bohm interpretation makes this fact particularly clear. In Refs. [24, 25] he has noted that the wavefunction plays a dual role in the deBroglie-Bohm interpretation. On the one hand, it is part of the ontology, a pilot wave that dictates the dynamics of the system’s configuration (the positions of the particles in thenonrelativistic theory). On the other hand, the wavefunction has a statistical character, specifying the distribution over the system’s configurations. In order to eliminate this dual role, Valentini suggests that the wavefunction is only a pilot wave and that any distribution over the configurations should be allowed as the initial condition. It is argued that one can still recover the standard distribution of configurations on a coarse-grained scale as a result of dynamical evolution [26]. Within this approach, the no-signalling constraint is a feature of a special equilibrium distribution. The tension between Bell inequality violations and no-signalling is resolved by abandoning the latter as a fundamental feature of the world and asserting that it only holds as a contingent feature. The fine-tuning is explained as the consequence of equilibration. (It has also been noted in the causal model literature that equilibration phenomena might account for fine-tuning of causal parameters [27].)

Conversely, the version of the deBroglie-Bohm interpretation espoused by Durr, Goldstein and Zhangi [28] – which takes no-signalling to be a non-contingent feature of the theory – does not seek to provide a dynamical explanation of the fine-tuning. Consequently, it seems fair to say that the fine-tuning required by the deBroglie-Bohm interpretationis less objectionable in Valentini’s version of the theory. On the other hand, the cost of justifying the fine-tuning by adynamical process of equilibration is that, because true equilibrium is an idealization that is never achieved in finite time, one would expect systems to have small deviations from equilibrium and such deviations could in principle be exploited to send signals superluminally. Valentini endorses this consequence of his version of the deBroglie-Bohminterpretation [29] and indeed has made proposals for where the strongest deviations from equilibrium might arise [30].Therefore, anyone who thinks that the absence of superluminal signals is a necessary, rather than a contingent, featureof quantum theory, will not be enthusiastic about Valentini’s approach. "

 

[DarMM]

(1) Huh? If you have spacelike separated events, the events cannot be time ordered. Therefore, the time odering is frame dependent and in particular can be made simultaneous. Neither can be the cause of the other. Neither is in the others light cone.

Yes, but retrocausal theories don't have propogation at spacelike distances. Only at timelike distances. They simply don't have spacelike propogation so there is no point arguing about it. This is exactly the feature that makes them different from nonlocal theories.

(2) Events which are within each other's light cone are, by definition, time ordered. If there is a cause and effect, which is the cause and which is the effect are defined by their time ordering. Retrocausal in this case would mean closed timelike world line, since all of the world lines connecting causes and effects are timelike.

To state otherwise is to flat out say relativity is incorrect.

The part in bold is not necessarily true. Retrocausal theories (see Ruth Kastner's papers and books for an example) have events in both directions of the light cone causing the each other in a sense and yet preserving Relativity.

 

[DarMM]

Hidden variables must be in some sense "fine tuned" as proved by Wood and Spekkens. However, they also argue that some forms of fine tuning are more acceptable than others

I agree with their statements on this. I think the best version of hidden variable theories are ones where the experimental imprecision required to mask signalling is a result of some kind of thermalization or equilibrium process.

Hopefully the occurance of this thermalization can be proven in both cases (or better yet from a scientific perspective just one).

If I remember right Matthew Leifer is looking at retrocausal theories that thermalize into Quantum Mechanics.

On a personal note I think such thermalization/equilibrium hidden variable theories might be the last "hope" for a realist account of subatomic physics. If they are proven to fail (i.e. it is demonstrated such thermalization cannot occur) I think we are driven to Copenhagen or QBism. I get the impression people like Robert Spekkens and Matthew Leifer have similar views, though perhaps not.

 

[atyy]

On a personal note I think such thermalization/equilibrium hidden variable theories might be the last "hope" for a realist account of subatomic physics. If they are proven to fail (i.e. it is demonstrated such thermalization cannot occur) I think we are driven to Copenhagen or QBism. I get the impression people like Robert Spekkens and Matthew Leifer have similar views, though perhaps not.

If we are driven to "pure" Copenhagen that will be just fine with me. I have always hated how Bohmian Mechanics takes all the mystery out of QM

 

[A. Neumaier]

It is argued [by Valentini] that one can still recover the standard distribution of configurations on a coarse-grained scale as a result of dynamical evolution

I think such thermalization/equilibrium hidden variable theories might be the last "hope" for a realist account of subatomic physics. If they are proven to fail (i.e. it is demonstrated such thermalization cannot occur)

Valentini's argument is faulty; see this post. Much improved arguments would be needed to prove thermalization.

 

[PAllen]

Unfortunately, that reasoning is flawed. If events are spacelike seperated, then a lorentz transformation can make them simultaneous or make either event occur before the other. So, if you assume one event causes the other, you can make a lorentz transform to a frame in which that is false, which requires the event you assumed was the cause to act retrocausally. That is what non-local means.

I assume there there is no causal relationship between E2 and E3 as I defined them, and this was explicitly stated in my post.

The whole point of my post was to argue against lumping together retrocausal models with pure nonlocal models. Others have clarified here some further distinctions and terminology I hadn’t been familiar with. Thus, I learn that my mental model of a pure nonlocal model seems best described by what is called acausal, which emphasizes the lack of causal relation between E2 and E3.

This was all purely in response to @stevendaryl’s classification. I really like his whole argument through this thread, and just was arguing not lumping together acausal and retrocausal.

 

[DarMM]

Valentini's argument is faulty; see this post. Much improved arguments would be needed to prove thermalization.

I fully agree, I don't think the required thermalization/equilibrium process has been demonstrated either in nonlocal or retrocausal theories.

 

[jocarren]

For me, this has to do with determinism and the very definition of "observation".

In the broadest possible meaning, observation is an interaction, an exchange of information between the phenomena and the observer. In that sense, there are some phenomena which cannot be observed, for instance because it doesn't interact, because it is beyond a causal horizon or because it's in the future.

Events that are observed acquire some value (or set of values) and become part of a deterministic space, for those which are not observed (or not yet observed) you can assign a probability, they are part of a phase space. For instance, a prediction of some future event exists only as a probability until the event occur and it's measured.

Furthermore, every observable and observed event exists in the past. Then the past can be viewed as deterministic and the future (plus regions behind horizons) is stochastic.

 

[A. Neumaier]

Then the past can be viewed as deterministic

But the past cannot be observed either, and becomes more and more uncertain as one goes back in time.

 

[jocarren]

But the past cannot be observed either,

Well, the measurement is made in the present, but the information transferred in that interaction always comes from the past.

and becomes more and more uncertain as one goes back in time.

I believe the uncertainty of a distant past event should be the same as a closer one. The information when measured might be more ofuscated though (entropy increased more between event and measurment)

 

[Pleonasm]

In case one gives up the concept of 'physical realism' - a viewpoint according to which an external reality exists independent of observation – and doesn’t insist on thinking about quantum phenomena with classical ideas, quantum mechanics doesn’t unsettle anymore.

Why wouldn't anyone do that, though? To deny the existence of a fixed external reality independent of observation/measurement would amount to madness, in the classical sense of the word, or simply radical relativism.

 

[Lord Jestocost]

To deny the existence of a fixed external reality independent of observation/measurement would amount to madness, in the classical sense of the word, or simply radical relativism.

Modern physics has no need of the hypothesis of a mind-independent reality.

 

[A. Neumaier]

Modern physics has no need of the hypothesis of a mind-independent reality.

Modern physics has no need of the hypothesis of a mind-dependent reality.

 

[Demystifier]

I fully agree, I don't think the required thermalization/equilibrium process has been demonstrated either in nonlocal or retrocausal theories.

Even if there is no general mathematically rigorous proof, it has been demonstrated in various numerical simulations. For a review see https://www.mdpi.com/1099-4300/20/6/422

 

[Pleonasm]

Modern physics has no need of the hypothesis of a mind-independent reality.

Classical mechanics is an axiomatically mind-independent field of inquiry. QM could very well be too. We don't know the correct interpretation, and the Copenhagen interpretation is not as widely accepted as it once was. Sean Carroll conciders it a "kind of a scandal" that it's still the "default" interpretation taught in the QM mechanics textbooks (source: William Lane Craig- Sean Carroll debate).

 

[DarMM]

Even if there is no general mathematically rigorous proof, it has been demonstrated in various numerical simulations. For a review see https://www.mdpi.com/1099-4300/20/6/422

I've read it before, as well as this interesting review/development which I enjoyed:
https://arxiv.org/abs/0911.2823
The numerical results are suggestive and I'm not overly skeptical, I wouldn't be surprised if Bohmian Mechanics has a rigorous equilibrium theorem, but they are a long way off being convincing. Most simulations involve 2D finite volume cases with particular potentials.

However there are some interesting developments. I liked this one:
https://arxiv.org/abs/1108.5496

 

[Pleonasm]

As for ontology vs expistemology which the threadmakers asks about, you would have to answer this question:

Suppose I am a quantum particle walking in the dark. You could not, no matter what instrument, see me and my exact future positioning in the dark, yet you know I am in the dark, so you attempt to measure where I am going. You can via that process derive deterministic equations for the probability of me arrising at a place in time, but that is as good as it gets. If you don't look at me, the equations for my future positions are classically deterministic.

Is the hiddeness of my (exact) motions during measurement ontological or expistemological in such a world, or is the question simply irrelevant?

 

[DarMM]

@Demystifier was asking a specific question about how the Copenhagen (and similar) interpretation avoids nonlocality and what Copenhagen says about the existence of quantities like position and momentum, i.e. are they simply not known outside of measurement or nonexistent.

The answer to both is:

1. It avoids non-locality via multiple sample spaces/counterfactual indefiniteness/contextuality. Different words for the same thing.
2. According to most Copenhagenish views, no. Momentum and position of quantum systems do not exist outside of measurement. Usually it is assumed quantum systems have properties that are not momenta and positions etc, but whatever they are they cannot be modeled mathematically, i.e. no hidden variables.

 

[Pleonasm]

@Demystifier was asking a specific question about how the Copenhagen (and similar) interpretation avoids nonlocality and what Copenhagen says about the existence of quantities like position and momentum, i.e. are they simply not known outside of measurement or nonexistent.

The answer to both is:

1. It avoids non-locality via multiple sample spaces/counterfactual indefiniteness/contextuality. Different words for the same thing.
2. According to most Copenhagenish views, no. Momentum and position of quantum systems do not exist outside of measurement. Usually it is assumed quantum systems have properties that are not momenta and positions etc, but whatever they are they cannot be modeled mathematically, i.e. no hidden variables.

And Schrodingers cat demonstrate how that leads to reductio ad absurdum, while other interpretations such as the Many Worlds interpretations of QM can account for it.

 

[Pleonasm]

"The Copenhagen interpretation is basically nonsense". "No thoughtful person still holds to it"

Sean Carroll: at 1:50:30

 

[DarMM]

And Schrodingers cat demonstrate how that leads to reductio ad absurdum, while other interpretations such as the Many Worlds interpretations of QM can account for it.

I don't see the connection between what I wrote and Schrodinger's cat.

"The Copenhagen interpretation is basically nonsense". "No thoughtful person still holds to it"

That's just rhetoric of little value.

 

[Pleonasm]

That's just rhetoric of little value.

Rhetoric? Another user in here claimed that once external reality assumptions of the copenhagen interpretation is accepted, everything else makes sense. This is a truism. If you accept the unacceptable, everything else will follow naturally in your theory. Carroll is of the opinion that no thoughtful person would accept the assumptions of the Copenhagen interpretation.

 

[Pleonasm]

I don't see the connection between what I wrote and Schrodinger's cat.

Not you but the other user arguing that Copenhagen interpretation is perfectably reasonable accepting an assumption that is completely unacceptable.

 

[DarMM]

Rhetoric? You claimed that once external reality assumptions of the copenhagen interpretation is accepted, everything else makes sense.

I don't recall saying that once the Copenhagen Interpretation is accepted everything will make sense. I was discussing how it treats Classical physical quantities and what it says about locality in Bell's theorem.

Carroll is of the opinion that no thoughtful person would accept the assumptions of the Copenhagen interpretation.

And this is nonsense, because there are several experts in Quantum Foundations, Quantum Information and Quantum Probability theory and several other areas that accept the the Copenhagen view. To dismiss them all as not "thoughtful" is just a rhetorical move. These people have clearly thought deeply about QM.

 

[DarMM]

Not you but the other user arguing that Copenhagen interpretation is perfectably reasonable accepting an assumption that is completely unacceptable.

And that assumption is?

 

[Pleonasm]

And that assumption is?

The proposition below, which is pure rubbish.

Modern physics has no need of the hypothesis of a mind-independent reality.

 

[DarMM]

The proposition below, which is pure rubbish.

Well I agree that @Lord Jestocost said it, but it's not an assumption in most Copenhagen views I have read, e.g. Bub, Healey, Fuchs, Brukner, Zeilinger, etc

 

[Pleonasm]

And this is nonsense, because there are several experts in Quantum Foundations, Quantum Information and Quantum Probability theory and several other areas that accept the the Copenhagen view.

So? There are nutcase physicists who question the bing bang theory. They might score high on an IQ-test and be thoughtful in other areas, but not in cosmology.

 

[Pleonasm]

Well I agree that @Lord Jestocost said it, but it's not an assumption in most Copenhagen views I have read, e.g. Bub, Healey, Fuchs, Brukner, Zeilinger, etc

Not in those exact terms but crazy enough that it doesn't matter how you phrase it.