What were Newton's thoughts on non-locality?

In summary, the conversation discusses the concept of non-locality and its acceptance by modern physicists. The participants also mention Newton's misgivings about this concept and discuss the implications of Bell's theorem on causality. They also touch upon the idea of definite events and their impact on statistics in experiments, specifically in relation to the EPR experiment.
  • #36
Derek P said:
The required information is unconditional - namely Alice's actual setting and Alice's actual result.
None of this information is available to Bob, hence he cannot use the conditional information available to him, hence there is no causality problem.

The conditional information and the correlations become actual only when someone has access to the actual data resolving the condition.
Derek P said:
But in a well-designed EPR experiment the observations ARE affected.
No, the observations are simply correlated - you are postulating in addition a notion of being affected which does not exist.

One cannot say that Alice's actions and observations cause (or affect) Bob's observations to be correlated. This is due to the Lorentz invariance of all relativistic arguments. There are always Lorentz frames in which Alice acts later than Bob observes, and others in which Bob acts later than Alice. So who can be said to cause (or affect) what?

Causality is simply inapplicable to an analysis of correlations at spacelike distances because any meaningful use of this notion depends on a notion of before and after, which doesn't exist in this case.
 
  • Like
Likes PeterDonis, Mentz114 and PeroK
Physics news on Phys.org
  • #37
A. Neumaier said:
The point I wanted to make is that Bell's theorem is not about quantum mechanics, but about classical assumptions regarding intuition and rules.
I agree that Bell's theorem is not about QM. But I disagree that it is about classical physics. His theorem does not assume classical physics. His theorem assumes locality and standard probability theory. The standard probability theory is a branch of mathematics and as such it does not depend on physics. The standard probability theory is just a part of the standard scientific method [E.T. Jaynes, Probability Theory The Logic of Science, Cambridge University Press, 2003]. In other words, the Bell theorem says that assumptions of (i) locality and (ii) standard scientific method are in contradiction with QM. Since QM is in agreement with observations, it follows that either (i) locality or (ii) standard scientific method (or both) is wrong.

Those who insist that QM is local actually deny the standard scientific method. For instance, one of the principles of the scientific method is the Reichenbach's common cause principle: a correlation between events A and B indicates either that A causes B, or that B causes A, or that A and B have a common cause (which Bell denotes by ##\lambda##). Those who insist that QM is local usually deny the Reichenbach's common cause principle.
 
  • Like
Likes Boing3000, zonde, Mentz114 and 3 others
  • #38
Demystifier said:
I disagree that it is about classical physics.
Classical physics is implied not in the formal theorem but in its application to the experiments. To apply it to the latter and construct an apparent paradox, one argues that all observables have sharp values, and that the particle travel like classical point particles along straight paths, with finite velocity ##c##, etc.

All this is justified only by invoking the principles of classical physics.

Demystifier said:
Reichenbach's common cause principle: a correlation between events A and B indicates either that A causes B, or that B causes A, or that A and B have a common cause. Those who insist that QM is local often deny the Reichenbach's common cause principle.
Reichenbach's principle is not a fundamental physical law, but a principle with only limited applicability. It is based on classical, deterministic and nonrelativistic reasoning.

It assumes determinism, since in a stochastic process, all noise is uncaused, without a common cause.

It assumes a linearly ordered view of time, hence nonrelativistic physics. Nothing in the formalism of classical relativity forbids uncaused spacelike correlations to exist. For example, one can assume a constant (hence maximally correlated) values of a classical field and its time derivative at $t=0$ and finds a unique solution of the d'Alembert equations, fully consistent with the principles of relativity.

Thus it is completely inapplicable to the discussion of quantum probabilities, or relativistic issues.
 
  • #39
A. Neumaier said:
None of this information is available to Bob, hence he cannot use the conditional information available to him, hence there is no causality problem.

The conditional information and the correlations become actual only when someone has access to the actual data resolving the condition.
The way you defined conditional information, it was available to Bob even before the experiment. The correlations may only be known when someone has access to both sets of data, but the Alice's and Bob's outcomes are known to themselves at the moment of observation.

{quote]No, the observations are simply correlated - you are postulating in addition a notion of being affected which does not exist.[/quote]
I don't think I'm posulating anything. A is set up freely, no common cause with B. B depends on A. Therefore A causes B. What is the alternative?
One cannot say that Alice's actions and observations cause (or affect) Bob's observations to be correlated. This is due to the Lorentz invariance of all relativistic arguments. There are always Lorentz frames in which Alice acts later than Bob observes, and others in which Bob acts later than Alice. So who can be said to cause (or affect) what?.
Both. Which is absurd. Which is precisely why the paradigm used here must be faulty.

But in any case, the same experiments can be done under timelike separation. You can't then rule out the loophole of secret collusion between the particles but you have unambiguous causality violation, under the same paradigm.
 
  • #40
Demystifier said:
I agree that Bell's theorem is not about QM. But I disagree that it is about classical physics. His theorem does not assume classical physics. His theorem assumes locality and standard probability theory. The standard probability theory is a branch of mathematics and as such it does not depend on physics. The standard probability theory is just a part of the standard scientific method [E.T. Jaynes, Probability Theory The Logic of Science, Cambridge University Press, 2003]. In other words, the Bell theorem says that assumptions of (i) locality and (ii) standard scientific method are in contradiction with QM. Since QM is in agreement with observations, it follows that either (i) locality or (ii) standard scientific method (or both) is wrong.

Those who insist that QM is local actually deny the standard scientific method. For instance, one of the principles of the scientific method is the Reichenbach's common cause principle: a correlation between events A and B indicates either that A causes B, or that B causes A, or that A and B have a common cause (which Bell denotes by ##\lambda##). Those who insist that QM is local usually deny the Reichenbach's common cause principle.
Or they point out that probability theory as applied to observations requires a map between the countable events and the observations. Classically the map is one-to-one. The quantum map is different and, with a good theory of measurement, it is quite sufficient to decompose the complete state into superposed separable (local) states. Whether Many Worlds is local or not thus depends on the precise meaning of local - the word was not designed to cope with superposition. Anything other interpretation, though...!
 
Last edited:
  • #41
Demystifier said:
... For instance, one of the principles of the scientific method is the Reichenbach's common cause principle: a correlation between events A and B indicates either that i) A causes B, or that ii) B causes A, or that iii) A and B have a common cause (which Bell denotes by ##\lambda##). Those who insist that QM is local usually deny the Reichenbach's common cause principle.

Referencing this experiment by Hensen et al, 2015: https://arxiv.org/abs/1508.05949

A Bell test is performed on between distant entangled electron spins. They are entangled via swapping, so the measured particles have no local connection to each other. Therefore there is no (prior, local) common cause, leaving just the first 2 options..

If A causes B, or B causes A, then that must occur non-locally. Or we deny Reichenbach's common cause principle as suggested. Or we could modify it to remove the implied requirement that a cause precede the effect.
 
  • Like
Likes Demystifier
  • #42
Demystifier said:
The standard probability theory is a branch of mathematics and as such it does not depend on physics. The standard probability theory is just a part of the standard scientific method [E.T. Jaynes, Probability Theory The Logic of Science, Cambridge University Press, 2003]. In other words, the Bell theorem says that assumptions of (i) locality and (ii) standard scientific method are in contradiction with QM. Since QM is in agreement with observations, it follows that either (i) locality or (ii) standard scientific method (or both) is wrong.
That's not correct. There is no strictly defined "scientific method". Science does not depend on particular mathematical theories. They may be applicable in certain situations, but they needn't be applicable in principle. Science is not dogmatic in any way.

Those who insist that QM is local actually deny the standard scientific method. For instance, one of the principles of the scientific method is the Reichenbach's common cause principle: a correlation between events A and B indicates either that A causes B, or that B causes A, or that A and B have a common cause (which Bell denotes by ##\lambda##). Those who insist that QM is local usually deny the Reichenbach's common cause principle.
Reichenbach's common cause principle is not a principle of the scientific method. On the contrary, probabilistic causation doesn't even work classically and is generally rejected as a conclusive theory of causality. One doesn't need quantum mechanics for that.
 
  • #43
A. Neumaier said:
Reichenbach's principle is not a fundamental physical law,
I agree. It is a part of the standard scientific method, which is much more general that physics.

A. Neumaier said:
but a principle with only limited applicability.
I agree. The scientific method is very powerful, but it cannot solve all problems.

A. Neumaier said:
It is based on classical,
Not in the sense of classical physics, but yes in the sense of standard ("classical") scientific method.

A. Neumaier said:
deterministic
The Reichenbach principle just says that if there is a correlation between something, then there must be a cause for the correlation. It does not insist that everything is deterministic in the Laplace's sense.

A. Neumaier said:
and nonrelativistic reasoning.
The Reichenbach principle is not in contradiction with relativity. For instance, the Reichenbach principle is valid in classical relativistic physics.
 
  • Like
Likes Derek P
  • #44
rubi said:
Science does not depend on particular mathematical theories.
All science, from psychology and medicine to experimental QM, uses the same principles of statistics, which is based on the same theory of probability.
 
  • Like
Likes Derek P, Mentz114 and PeroK
  • #45
rubi said:
Reichenbach's common cause principle is not a principle of the scientific method. On the contrary, probabilistic causation doesn't even work classically and is generally rejected as a conclusive theory of causality. One doesn't need quantum mechanics for that.
Can you give a non-quantum example of science that rejects the Reichenbach's common cause principle?
 
  • #46
Demystifier said:
I agree. It is a part of the standard scientific method, which is much more general that physics.
Don't you see how terribly poor and unscientific it is to dogmatically claim that some principle is part of the scientific method? There is no authority that defines what the scientific method is. The success of science critically depends on the ability of scientists to eventually reject wrong principles if they happen to disagree with our current best theories. If you need a dogma to justify your point of view, then you leave the realm of scienctific discussion.

Demystifier said:
All science, from psychology and medicine to experimental QM, uses the same principles of statistics, which is based on the same theory of probability.
If some field uses probability theory, then it is because probability theory just happens to suffice in the particular situation. If the data seems to be in conflict with with probability theory, people would explore different approaches. So far, this hasn't been necessary in psychology or medicine. However, experimental QM of course does not use probability theory. For example in quantum tomography, the statistics specifically reconstructs a quantum state rather than a classical probability distribution. If classical probability theory is used in some particular quantum experiment, then it is because in that situation, classical probability theory and quantum theory are the same.

Demystifier said:
Can you give a non-quantum example of science that rejects the Reichenbach's common cause principle?
Here's one example: https://www.journals.uchicago.edu/doi/abs/10.1086/psaprocbienmeetp.1986.1.193140
You will find many more if you look for them. (Edit: For example, this page has a few of them listed: https://plato.stanford.edu/entries/physics-Rpcc/#2)
 
Last edited:
  • Like
Likes Boing3000 and Demystifier
  • #47
A. Neumaier said:
There are always Lorentz frames in which Alice acts later than Bob observes, and others in which Bob acts later than Alice. So who can be said to cause (or affect) what?

Causality is simply inapplicable to an analysis of correlations at spacelike distances because any meaningful use of this notion depends on a notion of before and after, which doesn't exist in this case.
Given that there is no 'before/after' available then the instantaneous communication between the pair (they always have the same/opposite spin) has the same affect as the non-local correlations. There appears to be no way to decide experimentally which is a better hypothesis.
 
  • #48
Derek P said:
I don't think I'm posulating anything. A is set up freely, no common cause with B. B depends on A. Therefore A causes B. What is the alternative?

Both. Which is absurd. Which is precisely why the paradigm used here must be faulty.

Let's assume that experiments went the other way and we have hidden variables. Now, that does not mean that each electron has a definite spin (about all three axes simultaneously) as would be expected classically and would be, let us say, a robust paradigm. Instead, we know from basic QM (without entanglement) that an electron has to decide when measured what spin measurement to return. Where is this information held? And how does the electron process this information? Given the nature of what happens when we measure about different axes that is a lot of information per electron - potentially an infinite amount of information.

It seems to me this is just as faulty a paradigm. Just as physically unjustifiable.

But, also, let's take a look at something apparently more robust: a particle following a geodesic or moving under a potential. How does the particle know which way to move? If we were to simulate this, we would have to do a certain number of measurements or calculations on behalf of the particle to determine the local spacetime or local change in potential. It can't logically be done without any measurement or calculation; yet nature does it! That something tests the surrounding locality in every direction and informs the particle which way to move - what is the alternative to that?

It seems to me if you really ask how nature achieves things, then all the paradigms are faulty. Perhaps, therefore, nature managing to correlate two remote measurements is not as absurd as you may think. And it doesn't imply that nature is sending information from A to B any more than information is being gathered and processed about local changes in spacetime or a potential field.
 
  • #49
Derek P said:
Alice's and Bob's outcomes are known to themselves at the moment of observation.

Yes, but Bob doesn't know Alice's outcome when he knows his own (and vice versa).

Derek P said:
B depends on A.

No, A and B are correlated. Correlation has no ordering.

Derek P said:
the same experiments can be done under timelike separation. You can't then rule out the loophole of secret collusion between the particles but you have unambiguous causality violation

This is simply false. There is no causality violation in these experiments.
 
  • Like
Likes PeroK
  • #50
PeroK said:
Let's assume that experiments went the other way and we have hidden variables. Now, that does not mean that each electron has a definite spin (about all three axes simultaneously) as would be expected classically and would be, let us say, a robust paradigm. Instead, we know from basic QM (without entanglement) that an electron has to decide when measured what spin measurement to return. Where is this information held? And how does the electron process this information? Given the nature of what happens when we measure about different axes that is a lot of information per electron - potentially an infinite amount of information.

It seems to me this is just as faulty a paradigm. Just as physically unjustifiable.

But, also, let's take a look at something apparently more robust: a particle following a geodesic or moving under a potential. How does the particle know which way to move? If we were to simulate this, we would have to do a certain number of measurements or calculations on behalf of the particle to determine the local spacetime or local change in potential. It can't logically be done without any measurement or calculation; yet nature does it! That something tests the surrounding locality in every direction and informs the particle which way to move - what is the alternative to that?

It seems to me if you really ask how nature achieves things, then all the paradigms are faulty. Perhaps, therefore, nature managing to correlate two remote measurements is not as absurd as you may think. And it doesn't imply that nature is sending information from A to B any more than information is being gathered and processed about local changes in spacetime or a potential field.
I have never said that nature sends information from A to B. I have said that under the assumptions of the pardigm used here, nature must do so. nd you are using the same paradigm when you say "nature decides".
Incidentally "no hidden variable theory" includes "potentially an infinite amount of information."
 
  • #51
Derek P said:
under the assumptions of the pardigm used here

As far as I can tell, @PeroK is not assuming anything or advocating any paradigm; he's basically saying we don't have a paradigm that works. The question is, given that we have a theory that makes correct predictions, why do we need a paradigm?

[Moderator's note: the rest of this post was off topic and has been deleted.]
 
Last edited:
  • Like
Likes PeroK
  • #52
Thread closed for moderation.
 
  • Like
Likes Derek P
<h2>1. What is non-locality according to Newton?</h2><p>Non-locality, also known as action at a distance, is the concept that an object can affect another object without any physical contact or intermediary medium.</p><h2>2. Did Newton believe in non-locality?</h2><p>Yes, Newton believed in non-locality as he proposed the theory of universal gravitation, which states that all objects in the universe are attracted to each other without any physical contact.</p><h2>3. How did Newton explain non-locality?</h2><p>Newton explained non-locality through his theory of gravity, which states that every object with mass in the universe exerts a gravitational force on every other object with mass, regardless of the distance between them.</p><h2>4. Was Newton's concept of non-locality accepted by other scientists?</h2><p>Initially, Newton's concept of non-locality was met with skepticism and criticism from other scientists. However, as his theory of gravitation was proven to be accurate through experiments and observations, his concept of non-locality gained more acceptance.</p><h2>5. How does Newton's concept of non-locality relate to modern physics?</h2><p>Newton's concept of non-locality is still relevant in modern physics, especially in the field of quantum mechanics. It has also been incorporated into Einstein's theory of relativity, which explains the relationship between space and time.</p>

1. What is non-locality according to Newton?

Non-locality, also known as action at a distance, is the concept that an object can affect another object without any physical contact or intermediary medium.

2. Did Newton believe in non-locality?

Yes, Newton believed in non-locality as he proposed the theory of universal gravitation, which states that all objects in the universe are attracted to each other without any physical contact.

3. How did Newton explain non-locality?

Newton explained non-locality through his theory of gravity, which states that every object with mass in the universe exerts a gravitational force on every other object with mass, regardless of the distance between them.

4. Was Newton's concept of non-locality accepted by other scientists?

Initially, Newton's concept of non-locality was met with skepticism and criticism from other scientists. However, as his theory of gravitation was proven to be accurate through experiments and observations, his concept of non-locality gained more acceptance.

5. How does Newton's concept of non-locality relate to modern physics?

Newton's concept of non-locality is still relevant in modern physics, especially in the field of quantum mechanics. It has also been incorporated into Einstein's theory of relativity, which explains the relationship between space and time.

Similar threads

Replies
14
Views
2K
  • Quantum Interpretations and Foundations
4
Replies
138
Views
5K
  • Quantum Physics
Replies
5
Views
1K
  • Quantum Interpretations and Foundations
6
Replies
175
Views
6K
  • Quantum Physics
Replies
23
Views
3K
  • Quantum Physics
2
Replies
47
Views
3K
Replies
6
Views
1K
Replies
242
Views
21K
  • Quantum Interpretations and Foundations
2
Replies
37
Views
4K
Replies
17
Views
2K
Back
Top