Undergrad Getting rid of nonlocality from quantum physics

[martinbn]

What's obvious? What's the "territory" in the case of the electron?

You are missing the point. The point is that there is a territory, which is different from the model. So, something exists, we call it an electron, we study it and model it. If the model is constructed in such a way that some observables like position don't have a value at all times, unlike the analog in a different model (classical physics), it doesn't in any way imply that the electron doesn't exist. Of course the model may be inaccurate and with better understanding will make us revise our ontology, say that a field exists, but that is not what we are argueing about.

 

[martinbn]

But it answers the specific question that you asked.

No, it doesn't. The quotes are missing the context, and they don't say that things don't exist. You choose to interpret them that way. For example Heisenberg's quote only says that the world as described by QT is very different from the world as described by Classical Physics.

 

[Demystifier]

What does an electron have? As any particle, it's classified by its intrinsic properties, i.e., its mass, spin, and its charges wrt. the interactions described in the standard model.

How can you know that? All you can know is that the state in the Hilbert space has those properties (by being an eigenstate of the corresponding operators). But the real electron is not the same as its state in the Hilbert space. So if you consistently stick to the minimal interpretation, you have no right to say that the "electron itself" (if that notion makes sense at all) has all those properties.

 

[martinbn]

Indeed, there is a contradiction(in my understanding). What is the correct answer?

The answer is that the terminology isn't perfect. Quantum fields refer both to the things that exist in space-time and to some of the objects in the mathematical model.

 

[vanhees71]

How can you know that? All you can know is that the state in the Hilbert space has those properties (by being an eigenstate of the corresponding operators). But the real electron is not the same as its state in the Hilbert space. So if you consistently stick to the minimal interpretation, you have no right to say that the "electron itself" (if that notion makes sense at all) has all those properties.

It's known, because all observations are in accordance with this description at a very high level of accuracy.

I think again, this philosophical approach to physics rather mystifies things than clarifying them. A physical theory is about a description of observational facts, no more no less.

 

[Demystifier]

For example Heisenberg's quote only says that the world as described by QT is very different from the world as described by Classical Physics.

Let me repeat what Heisenberg said:
"But the atoms or the elementary particles themselves are not as real; they form a world of potentialities or possibilities rather than one of things or facts."
Sorry, but I don't see how to interpret this in a way you suggest. He talks about "particles themselves", not about our descriptions of particles.

 

[vanhees71]

That's the typical Heisenbergian gibberish. I never cared much...

 

[martinbn]

Let me repeat what Heisenberg said:
"But the atoms or the elementary particles themselves are not as real; they form a world of potentialities or possibilities rather than one of things or facts."
Sorry, but I don't see how to interpret this in a way you suggest. He talks about "particles themselves", not about our descriptions of particles.

Yes, and he says that they are not as real. He doesn't say that they don't exist, does he? He could mean that that they don't have coordinates if not measured.

 

[Demystifier]

You are missing the point. The point is that there is a territory, which is different from the model. So, something exists, we call it an electron, we study it and model it. If the model is constructed in such a way that some observables like position don't have a value at all times, unlike the analog in a different model (classical physics), it doesn't in any way imply that the electron doesn't exist. Of course the model may be inaccurate and with better understanding will make us revise our ontology, say that a field exists, but that is not what we are argueing about.

I agree with all the above, but it seems to me that you are missing my point. The simplest model we have (QM with the collapse postulate) is nonlocal. The simplest model without the collapse (many worlds) is nonlocal. The simplest model without the collapse and without the many worlds (Bohmian mechanics) is nonlocal. The very general class of models (defined by the Bell theorem) are nonlocal.

So is there a local model at all? No there isn't. There is only a local interpretation (Copenhagen), but such an interpretation is not a model. (By a model I mean a set of equations, by an interpretation I mean a set of words in plain English that explain what those equations mean.) In particular, any model containing a nonlocal wave function such as $\psi({\bf x}_1,{\bf x}_2)$ is a nonlocal model. A local interpretation of such a nonlocal model may contain a statement such as: "But this $\psi$ is not real, it's only our description.", but it doesn't change the fact that the model is nonlocal.

Now you may argue that the reality itself is local, but how then would you interpret the Bell theorem that no model of reality (under very general assumptions) can be local?

 

[Demystifier]

The answer is that the terminology isn't perfect. Quantum fields refer both to the things that exist in space-time and to some of the objects in the mathematical model.

In Bohmian mechanics there is no such confusion in terminology.

 

[Demystifier]

It's known, because all observations are in accordance with this description at a very high level of accuracy.

That's nonsense. All the observations are also in accordance with Bohmian mechanics (because it makes the same measurable predictions as standard quantum theory, including QFT), and yet you will not say that "it's known" that nature obeys nonlocal laws just because it does according to Bohmian mechanics.

 

[martinbn]

I agree with all the above, but it seems to me that you are missing my point. The simplest model we have (QM with the collapse postulate) is nonlocal. The simplest model without the collapse (many worlds) is nonlocal. The simplest model without the collapse and without the many worlds (Bohmian mechanics) is nonlocal. The very general class of models (defined by the Bell theorem) are nonlocal.

So is there a local model at all? No there isn't. There is only a local interpretation (Copenhagen), but such an interpretation is not a model. (By a model I mean a set of equations, by an interpretation I mean a set of words in plain English that explain what those equations mean.) In particular, any model containing a nonlocal wave function such as $\psi({\bf x}_1,{\bf x}_2)$ is a nonlocal model. A local interpretation of such a nonlocal model may contain a statement such as: "But this $\psi$ is not real, it's only our description.", but it doesn't change the fact that the model is nonlocal.

Now you may argue that the reality itself is local, but how then would you interpret the Bell theorem that no model of reality (under very general assumptions) can be local?

I am not talking about nonlocality at all in this thread. My original qustion and the following discussion was only about the existence. There is no disagreement about the nonlocality except about the terminology.

 

[Demystifier]

Yes, and he says that they are not as real. He doesn't say that they don't exist, does he? He could mean that that they don't have coordinates if not measured.

That's typical for Copenhagenians to use such a vague language.

But I think you still didn't tell me what I really want to here from you. What is your interpretation of the Bell theorem? Does it prove nonlocality of QM?

 

[martinbn]

That's typical for Copenhagenians to use such a vague language.

But I think you still didn't tell me what I really want to here from you. What is your interpretation of the Bell theorem? Does it prove nonlocality of QM?

Yes, it does prove that. I only dislike the choice of terminology.

 

[Demystifier]

I am not talking about nonlocality at all in this thread.

Well, I am. See the title of the thread.

Concerning "realism", I only talk about it to the extent it is relevant to nonlocality.

 

[Demystifier]

Yes, it does prove that. I only dislike the choice of terminology.

Good! Now do you agree that some physicists don't agree that the Bell theorem proves "nonlocality" (or whatever you want to call it)? And if you do, how do they justify it? It seems to me that they justify it by denying the existence of things in the absence of measurement, but maybe I'm wrong, maybe they justify it by denying something else.

 

[martinbn]

Well, I am. See the title of the thread.

Concerning "realism", I only talk about it to the extent it is relevant to nonlocality.

I know, but the OP doesn't seem interested in any discussion. I started in post #10.

 

[martinbn]

Good! Now do you agree that some physicists don't agree that the Bell theorem proves "nonlocality" (or whatever you want to call it)? And if you do, how do they justify it? It seems to me that they justify it by denying the existence of things in the absence of measurement, but maybe I'm wrong, maybe they justify it by denying something else.

I don't think that they disagree with the theorem. They (and I) disagree with the conclusions some physicists jump to.

 

[Demystifier]

I don't think that they disagree with the theorem. They (and I) disagree with the conclusions some physicists jump to.

So what's your conclusion regarding the Bell theorem?

 

[Lord Jestocost]

No, it doesn't. The quotes are missing the context, and they don't say that things don't exist.

To merely claim the existence, or reality, of something is no great challenge. A realist has to provide more: a specific set of claims concerning what exists and, especially, how it exists, viz. the character of its existence.

“The separate images that we form of the quantum world (wave, particle, for example) from different experimental viewpoints do not combine into one comprehensive whole” (Nick Herbert). Thus, we cannot simply think of an “electron” or an “atom” as an object which can be given a description in its own right: There exists no single image that corresponds to an electron or an atom: The character of their existence is missing. What is within this context the meaning of the term "things"? It's an empty notion. That’s the reasoning behind Heisenberg’s "Atoms are not things".

 

[vanhees71]

I agree with all the above, but it seems to me that you are missing my point. The simplest model we have (QM with the collapse postulate) is nonlocal. The simplest model without the collapse (many worlds) is nonlocal. The simplest model without the collapse and without the many worlds (Bohmian mechanics) is nonlocal. The very general class of models (defined by the Bell theorem) are nonlocal.

So is there a local model at all? No there isn't. There is only a local interpretation (Copenhagen), but such an interpretation is not a model. (By a model I mean a set of equations, by an interpretation I mean a set of words in plain English that explain what those equations mean.) In particular, any model containing a nonlocal wave function such as $\psi({\bf x}_1,{\bf x}_2)$ is a nonlocal model. A local interpretation of such a nonlocal model may contain a statement such as: "But this $\psi$ is not real, it's only our description.", but it doesn't change the fact that the model is nonlocal.

Now you may argue that the reality itself is local, but how then would you interpret the Bell theorem that no model of reality (under very general assumptions) can be local?

Again one should really be specific what's meant by "nonlocal"! Relativistic QFT is local by construction, i.e., any local observables commute at space-like-separated arguments. This implies the cluster-decomposition principle and thus there are no nonlocal causal influences. There's inseparability described by entanglement, but that's nothing that contradicts relativistic causality as the collapse postulate implies, and I still have not seen any convincing argument, what the collapse is good for. I also don't understand in how far "many worlds" helps to solve any physical problem. Bohmian mechanics adds some mathematical formalism on top of standard QM and describes "trajectories". That's of course in a sense non-local, but it's just an addition in the math which doesn't add anything to what's observable according to standard QT, i.e., probablities to find a particle at some position.

I do not know, what you mean by "local Copenhagen interpretation". I also have no idea, why you consider a wave function $\psi(x_1,x_2)$ as non-local. Usually it's a wave function describing a two-particle system. There's nothing more or less non-local than with a single-particle wave function. I've also no idea what it means a "wave function is non-local". What can be non-local are interactions, and this they indeed are in Newtonian physics (no matter whether they are classical or quantum) since there the standard are instantaneous interactions at a distance as in Newtons law of gravity.

It seems that gets again a thread, where phase space is filled with equal probability, i.e., it's leading to maximum entropy (the state of least information content). SCNR.

 

[vanhees71]

Good! Now do you agree that some physicists don't agree that the Bell theorem proves "nonlocality" (or whatever you want to call it)? And if you do, how do they justify it? It seems to me that they justify it by denying the existence of things in the absence of measurement, but maybe I'm wrong, maybe they justify it by denying something else.

No Bell's theorem does not prove "nonlocality" it only proves that there's no local deterministic theory that reproduces the outcome of quantum theory and the corresponding results of their real-lab tests, which always agree with QT rather than local deterministic hidden-variable theories. Relativistic QFT has only local interactions and describes the Bell tests correctly. So there is a local model, namely simply relativistic QFT! It's of course not deterministic. Some people say it's "non-realistic" and they say that one gives up "realism". But if you simply consider as realistic what's objectively and reproducibly observed, there's nothing non-realistic with QT.

 

[martinbn]

So what's your conclusion regarding the Bell theorem?

That it proves Bell non-locality (which is a misnomer), not non-locality in the usual sense of the word.

 

[martinbn]

To merely claim the existence, or reality, of something is no great challenge. A realist has to provide more: a specific set of claims concerning what exists and, especially, how it exists, viz. the character of its existence.

“The separate images that we form of the quantum world (wave, particle, for example) from different experimental viewpoints do not combine into one comprehensive whole” (Nick Herbert). Thus, we cannot simply think of an “electron” or an “atom” as an object which can be given a description in its own right: There exists no single image that corresponds to an electron or an atom: The character of their existence is missing. What is within this context the meaning of the term "things"? It's an empty notion. That’s the reasoning behind Heisenberg’s "Atoms are not things".

I don't know what to say. This seems to have absolutely no connection to science.

 

[Lord Jestocost]

I don't know what to say. This seems to have absolutely no connection to science.

The connection is simple. Quantum mechanics is in no need for any reality hypothesis.

 

[Demystifier]

That it proves Bell non-locality (which is a misnomer), not non-locality in the usual sense of the word.

Fine. And which name (instead of nonlocality) would be better? And how would you explain (interpret) what it means in plain English?

 

[martinbn]

The connection is simple. Quantum mechanics is in no need for any reality hypothesis.

And how does this relate to what I wrote? I never claimed that QM needs a reality hypothesis.

 

[EPR]

The answer is that the terminology isn't perfect. Quantum fields refer both to the things that exist in space-time and to some of the objects in the mathematical model.

The mathematical formalism says that quantum fields can extend indefinitely(mathematically, this isn't an issue, right?). There is always some non-zero probability that particles with definite momentum may be completely delocalized(and found on Mars, theoretically). What about conservation laws in the scenario that involved my conundrum( quantum fields which permeate all of space-time)?

This point never became clear to me.

 

[martinbn]

Fine. And which name (instead of nonlocality) would be better? And how would you explain (interpret) what it means in plain English?

Any name that is not already in use. I am sure you agree that it is confusing to call it red, and claim that Bell proved that QM is red! Plain English is not often suitable for these things.

 

[martinbn]

The mathematical formalism says that quantum fields can extend indefinitely(mathematically, this isn't an issue, right?). There is always some non-zero probability that particles with definite momentum may be completely delocalized(and found on Mars, theoretically). What about conservation laws in the scenario that involved my conundrum( quantum fields which permeate all of space-time)?

This point never became clear to me.

I don't know, it seems like a valid question about quantum field theory. But it is not related to existence.