I Can anti-realism really save nonlocality?

  • I
  • Thread starter Thread starter sahashmi
  • Start date Start date
sahashmi
Messages
96
Reaction score
18
TL;DR Summary
Can anti realism really save non locality?
Anton Zeilinger, an experimentalist who proved that QM seems to be non local, doesn’t seem to actually believe in non locality himself. In a conference in Dresden, he stated that if one simply abandons the notion that objects have well defined properties before measurement (i.e. if one doesn’t adopt realism), one does not need to posit any sort of non locality or non local/faster than light influences in quantum entanglement.

Tim Maudlin, a prominent proponent of non locality, responds to him stating, as detailed in the book Spooky Action At A Distance by George Musser,

> “When Zeilinger sat down, Maudlin stood up. “You’ll hear something different in my account of these things,” he began. Zeilinger, he said, was missing Bell’s point. Bell did take down local realism, but that was only the second half of his argument for nonlocality. The first half was Einstein’s original dilemma. By his logic, realism is the fork of the dilemma you’re forced to take if you want to avoid nonlocality. “Einstein did not assume realism,” Maudlin said. “He derived it.” Put simply, Einstein ruled out local antirealism, Bell ruled out local realism, so whether or not physics is realist, it must be nonlocal.

> The beauty of this reasoning, Maudlin said, is that it makes the contentious subject of realism a red herring. As authority, Maudlin cited Bell himself, who bemoaned a tendency to see his work as a verdict on realism and eventually felt compelled to rederive his theorem without ever mentioning the word “realism” or one of its synonyms. It doesn’t matter whether experiments create reality or merely capture it, whether quantum mechanics is the final word in physics or merely the prelude to a deeper theory, or whether reality is composed of particles or something else entirely. Just do the experiment, note the pattern, and ask yourself whether there’s any way to explain it locally. Under the appropriate circumstances, there isn’t. Nonlocality is an empirical fact, full stop, Maudlin said.”

Let’s suppose Zeilinger is right. Before any of the entangled particles are measured, none of their properties exist. But as soon as one of them is measured (say positive spin), must the other particle not be forced to come up as a negative spin? Note that the other particle does not have a defined spin before the first one is measured. So how can this be explained without a non locality, perhaps faster than light, or perhaps even an instantaneous influence?

A common retort to this is that according to relativity, we don’t know which measurement occurs first. But then change my example to a **particular** frame of reference. In that frame, one does occur first. And in that frame, the second particle’s measurement outcome is not constrained until the first one is measured. How is this not some form of causation? Note that if there is superluminal causation, relativity would be false anyways, so it makes no sense to use relativity to rule out superluminal causation (that’s a circular argument)

Let’s assume that the many worlds interpretation or the superdeterminism intepretation is false for the purpose of this question, since I know that gets around these issues
 
Physics news on Phys.org
Quantum mechanics is local, in the sense that it doesn't allow for superluminal interactions. This does not contradict results such as Bell inequalities or anything allowed by entanglement. The point is that quantum mechanics allows for correlations that cannot be explained by any local classical theory.
 
  • Like
Likes Drgnslr and bhobba
I don't understand the desire to make QM realistic (in the technical sense of the word). From the very beginning it has been non-realistic and all experiments for over a hundred years have confirmed its predictions.

sahashmi said:
“Einstein did not assume realism,” Maudlin said. “He derived it.” Put simply, Einstein ruled out local antirealism, Bell ruled out local realism, so whether or not physics is realist, it must be nonlocal.
That is not true. This is just some peoples' interpretation.

sahashmi said:
Let’s suppose Zeilinger is right. Before any of the entangled particles are measured, none of their properties exist. But as soon as one of them is measured (say positive spin), must the other particle not be forced to come up as a negative spin? Note that the other particle does not have a defined spin before the first one is measured. So how can this be explained without a non locality, perhaps faster than light, or perhaps even an instantaneous influence?

You are making an assumption here. You assume that right after the measurement on particle one, particle to acquires a property too. Why should that be? Why not the spin of particle two is undetermined until it is measured?
 
  • Like
Likes Drgnslr, bhobba and Lord Jestocost
martinbn said:
I don't understand the desire to make QM realistic (in the technical sense of the word). From the very beginning it has been non-realistic and all experiments for over a hundred years have confirmed its predictions.


That is not true. This is just some peoples' interpretation.



You are making an assumption here. You assume that right after the measurement on particle one, particle to acquires a property too. Why should that be? Why not the spin of particle two is undetermined until it is measured?
Because quantum mechanics says that the spin of particle two is fixed as soon as either particle one or two is measured. This isn’t an assumption
 
  • Skeptical
  • Haha
Likes Drgnslr, bhobba and weirdoguy
sahashmi said:
Because quantum mechanics says that the spin of particle two is fixed as soon as either particle one or two is measured. This isn’t an assumption
No, QM says that if you measure the spin of particle two, along the same axis, you will get the opposite result with 100% certainty. But it doesn't say that it has value before the measurment. That is interpretation dependent. Assuming it is a form of realism.
 
  • Like
Likes PeterDonis, bhobba, dextercioby and 1 other person
martinbn said:
No, QM says that if you measure the spin of particle two, along the same axis, you will get the opposite result with 100% certainty.
But this is precisely Einstein's definition of realism. If you can predict the outcome with certainty, then there exists a physical property corresponding to the measurement outcome -- almost verbatim from the EPR paper. Using Einstein's "definition" of realism, would you say there is realism in QM?
 
  • Like
Likes bhobba, dextercioby and sahashmi
martinbn said:
No, QM says that if you measure the spin of particle two, along the same axis, you will get the opposite result with 100% certainty. But it doesn't say that it has value before the measurment. That is interpretation dependent. Assuming it is a form of realism.
If you get a particular result with probability 1, then it automatically becomes real. The whole point of something not existing before measurement is if its state cannot be determined with certainty.

Also, the math in QM disagrees with you. One measurement collapses the entire joint wave function. Thus, the other measurement becomes determined as soon as the first one is. Exactly “when” the other particle’s outcome is revealed is irrelevant.
 
lodbrok said:
But this is precisely Einstein's definition of realism. If you can predict the outcome with certainty, then there exists a physical property corresponding to the measurement outcome -- almost verbatim from the EPR paper. Using Einstein's "definition" of realism, would you say there is realism in QM?
Yes, that is his definition, but it is not somthing QM has.
 
Last edited:
  • Like
Likes PeterDonis and Lord Jestocost
sahashmi said:
If you get a particular result with probability 1, then it automatically becomes real. The whole point of something not existing before measurement is if its state cannot be determined with certainty.
That is an interpretational statement. It is not true in all interpretations.
sahashmi said:
Also, the math in QM disagrees with you. One measurement collapses the entire joint wave function. Thus, the other measurement becomes determined as soon as the first one is. Exactly “when” the other particle’s outcome is revealed is irrelevant.
This doesnt contradict what i said.
 
  • #10
sahashmi said:
Also, the math in QM disagrees with you. One measurement collapses the entire joint wave function. Thus, the other measurement becomes determined as soon as the first one is. Exactly “when” the other particle’s outcome is revealed is irrelevant.
That is one conclusion that might drawn from the math of QM, but it is not the only one and it is not free from conceptual difficulties:
Thinking in terms of “collapses the entire joint wave function” only makes sense if we’ve chosen an interpretation that includes collapse.
“First [measurement]” and “other measurement” are not clearly defined when the two measurement events are spacelike-separated.
 
  • Like
  • Love
Likes PeterDonis, bhobba and dextercioby
  • #11
sahashmi said:
Because quantum mechanics says that the spin of particle two is fixed as soon as either particle one or two is measured. This isn’t an assumption
You're confusing a known spin state with a fixed spin. There is no such thing as fixed spin.

In this respect an entangled state is no different from other quantum superpositions. A measurement determines the state relative to the measured quantity by producing an eigenstate of the observable. Thereafter, you know the result of a repeated measurement of that observable. But, the system remains indeterminate in terms of other incompatible observables.

This confusion caused you problems understanding polarised light in a previous thread. Polarisation is equivalent to spin in this context.

QM is probabilistic, but sometimes for some measurements the probability can be 0 or 1. That's not something to get excited about, IMO.

PS and by that I mean saying "Aha, now we have an element of physical realiry".
 
Last edited:
  • #12
PeroK said:
You're confusing a known spin state with a fixed spin. There is no such thing as fixed spin.

In this respect an entangled state is no different from other quantum superpositions. A measurement determines the state relative to the measured quantity by producing an eigenstate of the observable. Thereafter, you know the result of a repeated measurement of that observable. But, the system remains indeterminate in terms of other incompatible observables.

This confusion caused you problems understanding polarised light in a previous thread. Polarisation is equivalent to spin in this context.

QM is probabilistic, but sometimes for some measurements the probability can be 0 or 1. That's not something to get excited about, IMO.

PS and by that I mean saying "Aha, now we have an element of physical realiry".
Again, there is no difference between “the probability of the chair existing behind this wall is 1” and “the chair exists behind the wall in reality.”

Even if you think the particle’s state only exists upon measurement once the other particle is measured, it makes no difference, since it is now constrained to only be spin up if the first particle is measured spin down. This constraint did not happen before the first particle was measured
 
  • Like
Likes lodbrok
  • #13
martinbn said:
Yes, that is his definition, but it is not somthing QM has.
In that case, QM has the prediction with 100% certainty, so I'm not sure what you mean by it's not something QM has. So by definition, at least per EPR, QM has realism. You could provide a different definition of realism, which QM doesn't have, but according to the EPR definition, QM has realism in that case.
 
  • Like
Likes sahashmi
  • #14
sahashmi said:
Again, there is no difference between “the probability of the chair existing behind this wall is 1” and “the chair exists behind the wall in reality.”
Yes, but you are making a different sort of claim: that there is no difference between "The probability of an 'up' result when and if we measure the spin on the up-down axis is 1" and "the spin on the up-down axis is 'up'".

There is a fundamental difference between the mathematical treatment of macroscopic objects like chairs and quantum particles, so the equivalence you are asserting here requires additional assumptions and adopting a particular interpretational stance.
 
  • Like
Likes PeterDonis, javisot and PeroK
  • #15
PeroK said:
You're confusing a known spin state with a fixed spin. There is no such thing as fixed spin.
Spin of what?
PeroK said:
Thereafter, you know the result of a repeated measurement of that observable. But, the system remains indeterminate in terms of other incompatible observables.
Does the system exist before the measurement is performed?
 
  • #16
sahashmi said:
Again, there is no difference between “the probability of the chair existing behind this wall is 1” and “the chair exists behind the wall in reality.”

Even if you think the particle’s state only exists upon measurement once the other particle is measured, it makes no difference, since it is now constrained to only be spin up if the first particle is measured spin down. This constraint did not happen before the first particle was measured
You are hopelessly confusing classical and quantum mechanical concepts. A quantum state can be known and it can be an eigenstate of a given observable. This has nothing to do with existence.

An electron always has a spin state. If that is spin-up about a given axis it does not mean that the electron is spinning about that axis. The spin about that axis accounts for only a third of the total spin. This is elementary QM and is why you must learn QM before you start dissecting it.

Electron spin is never real in the way the Earth's spin is real. Or, in the way that a chair behind a curtain is real.

You're just blundering around in the dark here with little understanding of what QM really says.
 
  • #17
lodbrok said:
Spin of what?

Does the system exist before the measurement is performed?
Spin of elementary particles.

Yes, systems exist whether or not you measure them. It's the dynamic quantities of a system that do not in general have fixed values before a measurement.
 
  • #18
PeroK said:
Electron spin is never real in the way the Earth's spin is real. Or, in the way that a chair behind a curtain is real.
The EPR definition of realism doesn't say the observed property is real. It says "there exists a physical property corresponding to the observed property"
 
  • Like
Likes sahashmi
  • #19
PeroK said:
Yes, systems exist whether or not you measure them. It's the dynamic quantities of a system that do not in general have fixed values before a measurement.
Absolutely, and this is true not just of QM but also of classical mechanics. Thus it's not a revolutionary statement to say in some cases there is realism in QM.
 
  • #20
lodbrok said:
The EPR definition of realism doesn't say the observed property is real. It says "there exists a physical property corresponding to the observed property"
I know. What I'm saying is that EPR were misguided.
 
  • #21
lodbrok said:
Absolutely, and this is true not just of QM but also of classical mechanics. Thus it's not a revolutionary statement to say in some cases there is realism in QM.
In classical mechanics, quantities are assumed to have real values at all times. Whether we know what they are is a different question.

Also, if thus were the case and QM were not revolutionary, why all the fuss?
 
  • #22
PeroK said:
I know. What I'm saying is that EPR were misguided.
EPR provided an operational definition of "realism" that has been used ever since. I'm not aware that anyone has provided a different one. If you think that definition is misguided, please suggest a new one. But if you are going to claim that "realism" is present or absent somewhere, it better be based on consistent application of the operational definition.
PeroK said:
In classical mechanics, quantities are assumed to have real values at all times. Whether we know what they are is a different question.
Take "velocity" as a very simple example. A particle does not "have" a real value for velocity at all times, irrespective of measurement. You need a reference frame, and that frame is provided by the observer.
PeroK said:
Also, if thus were the case and QM were not revolutionary, why all the fuss?
The point is not that QM is not revolutionary. It is that the often repeated blanket claim that QM doesn't have realism is misguided. There is realism in some situations and no realism in others, and the acknowledgement that particles/systems exist independent of measurement confirms this.

Indeed, it is wrong to say a value is real if that value can be predicted with 100% certainty -- because this was not the EPR definition. They only said, "there exists a physical property corresponding to" the value.

On the other hand, it is also wrong to handwavingly say, "there's no realism in QM" when you mean the value itself that's predicted is not real. EPR didn't say that either; they said "there exists a physical property corresponding to" it.

This is my point.
 
  • Skeptical
Likes weirdoguy and PeroK
  • #23
The consensus, AFAIK, is that the EPR paper was misguided in terms of demonstrating that QM was an incomplete theory.

In any case, the issue here is that the OP is confused between the concept of a definite state in QM. And the notion of "reality" in terms of definite values for dynamic quantities.
 
  • #24
PeroK said:
You are hopelessly confusing classical and quantum mechanical concepts. A quantum state can be known and it can be an eigenstate of a given observable. This has nothing to do with existence.

An electron always has a spin state. If that is spin-up about a given axis it does not mean that the electron is spinning about that axis. The spin about that axis accounts for only a third of the total spin. This is elementary QM and is why you must learn QM before you start dissecting it.

Electron spin is never real in the way the Earth's spin is real. Or, in the way that a chair behind a curtain is real.

You're just blundering around in the dark here with little understanding of what QM really says.
You seem to be using vague language and relying on semantics to avoid the point I’m making. Again, once one particle is measured, the other particle now has a definite value at measurement. It doesn’t matter whether it has the value before or during measurement. The point is that the probability is now 1 for the second particle to be spin up if the first particle is measured spin down, whenever the measurement of the second particle occurs.

Given that the probability was not 1 before the first particle was measured, it then stands to reason the first particle influences the second measurement outcome.
 
  • Skeptical
Likes weirdoguy and PeroK
  • #25
PeroK said:
The consensus, AFAIK, is that the EPR paper was misguided in terms of demonstrating that QM was an incomplete theory.

In any case, the issue here is that the OP is confused between the concept of a definite state in QM. And the notion of "reality" in terms of definite values for dynamic quantities.
There is no such consensus. The EPR paper was misguided in that it assumed local hidden variables to be under the hood. The fork established in the EPR paper: either there are local hidden variables or there is nonlocality was correct. Local hidden variables were ruled out. Non-locality thus must be correct.

Bell’s theorem has two assumptions: locality and statistical independence. It does NOT have a realism assumption. Locality has been ruled out by experiments period. No amount of hand waving around this changes this fact.
 
  • Skeptical
Likes PeroK and weirdoguy
  • #26
The issue is complex.

First, as taught to beginning students, QM is incorrect (I thought it a limiting case of another theory, but as I will explain later, that is not strictly true). Quantum Field Theory (QFT) is the 'correct' theory. In QFT electromagnetism and elections, in fact, everything is a field, and by something called Noether's Theorem, the fields have energy. Now we have E=MC^2 and all that, so mass is a form of energy, and the majority, if not all, physicists consider mass real. Hence, quantum fields would seem to be real. Some dialectic could likely be formulated where it is incorrect, but I think most physicists would disagree. Also, we have what Wienberg calls a folk theorem - that any theory at large enough distances (these days often thought of as above the Plank Scale) will look like a QFT. Distances we can currently probe are well above the Plank Scale.

Beginning QM, however, deals with this thing called the state, and many (I certainly did) thought the state was the limiting case of a quantum field. However, it turns out that is wrong:

https://arxiv.org/abs/1712.06605

It is different from ordinary QM. It must include antiparticles in its formulation, which ordinary QM does not. This means the state is just a mathematical device used in the theory to calculate probabilities. That view corresponds with Gleason's Theorem, which the reader can look into. I have done several posts on it, including the whole proof.

Also, have a look at:
https://faculty1.coloradocollege.edu/~dhilt/hilt44211/AJP_Nine formulations of quantum mechanics.pdf

Interpretation F, including antiparticles, if one wants to be exact, would likely be QM in the classical realm (i.e. at velocities well below the speed of light). At a fundamental level, care needs to be taken in reading too much into basic QM; although it is the best place for the beginner to start, but as one advances, a more nuanced view is required. Even in the classical realm, the quantum field is still real but is not in the quantum state. Are quantum fields local? There seem to be different views on that. They, for sure, are Einstein Local. QFT was constructed, so it must be.

Thanks
Bill
 
Last edited:
  • #27
sahashmi said:
It does NOT have a realism assumption.
Doesn't it? I don't see how to justify the integrations across ##\lambda## without making an assumption about realism.
 
  • Like
Likes javisot and bhobba
  • #28
sahashmi said:
Locality has been ruled out by experiments period. No amount of hand waving around this changes this fact.

Careful here. The theorem rules out bell locality, which is probably better-called factorizability. It is, roughly, the condition that any correlations between distant events be explicable in local terms.

Note the use of correlations. This is NOT the usual definition of locality used by physicists, where actual object A influences actual object B. When entangled; quantum objects, except in some hidden variable interpretations, lose their individuality, so the usual idea of locality makes no sense (except in some hidden variable theories like DBB). In the quantum field view of my previous post, they are considered a single 'excitation' of the field. An observation breaks the entanglement, so you have two separate excitations. Question to ponder - did going from a single excitation to two distinct ones happen non-locally in the usual conception? If we cut a single string to form two strings, did that occur non-locally?

See:
https://plato.stanford.edu/entries/bell-theorem/

Thanks
Bill
 
Last edited:
  • #29
sahashmi said:
Given that the probability was not 1 before the first particle was measured, it then stands to reason the first particle influences the second measurement outcome.
There is no definite ordering of the measurements if they are spacelike separated. You cannot say which measurement influenced the other.

Things that "stood to reason" have been abandoned in light of the experiments to test Bell's theorem.
 
  • Like
Likes bhobba
  • #30
PeroK said:
There is no definite ordering of the measurements if they are spacelike separated. You cannot say which measurement influenced the other.

Things that "stood to reason" have been abandoned in light of the experiments to test Bell's theorem.
That’s why QM breaks relativity. We need a preferred frame
 
  • Skeptical
  • Sad
Likes weirdoguy, bhobba and PeroK
  • #31
sahashmi said:
That’s why QM breaks relativity. We need a preferred frame
I think your problems with QM come from not understanding relativity.
 
  • Like
Likes bhobba, PeroK and weirdoguy
  • #32
martinbn said:
I think your problems with QM come from not understanding relativity.
There is no way to explain QM without breaking relativity.

A occurring before B and B occurring before A simultaneously violates logic. Both cannot be true. Something that is illogical can’t be true. And bell proved non locality to be true.

So either the first measurement influences the second or vice versa. Without an influence, there is no reason for independent stochastic random variables to be correlated to each other
 
  • Skeptical
Likes bhobba and PeroK
  • #33
lodbrok said:
The EPR definition of realism doesn't say the observed property is real. It says "there exists a physical property corresponding to the observed property"
Any examples of "observed property" that are not real by definition?
I mean, if an observed property exists, by definition it must be real, or not?
 
  • #34
sahashmi said:
A occurring before B and B occurring before A simultaneously violates logic. Both cannot be true.
Not understanding the simultaneity of relativity is the first sign that you have no understanding of relativity. That was my point.
 
  • Like
Likes bhobba and PeroK
  • #35
martinbn said:
Not understanding the simultaneity of relativity is the first sign that you have no understanding of relativity. That was my point.
No, this has nothing to do with not understanding the simultaneity of relativity. I’m aware of what it says.

But you cannot explain QM with relativity intact. I’m saying that that relativity is wrong.

Relativity could simply be emergent and not fundamental, the same way Newtonian mechanics is wrong but its predictions still work. This principle has to be wrong because a) the correlations in QM cannot be explained otherwise and b) as a matter of logic, A occurring before B, and B occurring before A cannot both be true.
 
  • Skeptical
  • Sad
Likes weirdoguy, bhobba and PeroK
  • #36
sahashmi said:
No, this has nothing to do with not understanding the simultaneity of relativity. I’m aware of what it says.

But you cannot explain QM with relativity intact. I’m saying that that relativity is wrong.

Relativity could simply be emergent and not fundamental, the same way Newtonian mechanics is wrong but its predictions still work. This principle has to be wrong because a) the correlations in QM cannot be explained otherwise and b) as a matter of logic,
This statement
sahashmi said:
A occurring before B, and B occurring before A cannot both be true.
is your own statement. If it isn't true, fine, but why do you blame it on relativity!
 
  • #37
martinbn said:
This statement

is your own statement. If it isn't true, fine, but why do you blame it on relativity!
Relativity says that in the case of entanglement, either the first measurement occurs before the other or vice versa, depending on the frame. I’m saying that this is impossible given the correlations that occur.
 
  • Skeptical
Likes bhobba, PeroK and weirdoguy
  • #38
sahashmi said:
Relativity says that in the case of entanglement, either the first measurement occurs before the other or vice versa, depending on the frame. I’m saying that this is impossible given the correlations that occur.
QM and relativity are combined in QFT. There's no physical reason to abandon relativity. In that respect you are simply wrong.

One alternative is that nature is nonlocal. In the sense that it can manage correlations without FTL influences. You might not like that idea, but there is no reason to dismiss the idea.

I don't like the idea of MWI, for example. But, that doesn't mean I think it must be wrong.

QFT is a successful theory, so there is no reason to abandon it on philosophical grounds.
 
  • Like
Likes bhobba
  • #39
Thread closed for Moderation...
 
  • Like
Likes bhobba
  • #40
sahashmi said:
That’s why QM breaks relativity. We need a preferred frame

QFT, the most accurately verified theory we have, and basically at large distances, how nature must be (remember Wienberg's folk theorem), not only does not break relativity but is built on it. Special relativity, of course - general relativity is a whole new ball game.

Apologies to Berkeman; I did not notice the thread was closed. In my defence, the fact that QFT is built on special relativity is crucial.

Thanks
Bill
 
Last edited:
  • #41
The thread will stay closed
 
  • Like
Likes bhobba
Back
Top