Does a measurement setup determine the reality of spin measurement outcomes?

[A. Neumaier]

Yes, but q-expectations of what? Of all observables. In particular, there is an observable (essentially Hamiltonian times inverse number of particles) the q-expectation of which is the temperature.

No. Your observable $H/N$ is the mass-based energy density, which is quite independent from the temperature.

The fundamental observable quantities are the q-expectations of quantities. The latter are defined as the densely defined operators. The temperature is not such a quantity. It is instead a parameter in the grand canonical definition of the density operator.

 

[PeterDonis]

in Bohmian Mechanics the wave function isn't considered just a tool for calculating predictions

in Bohmian Mechanics the wave function is a "law of physics" thus part of the metaphysics, but not a "thing" of the theory and thus non-ontic

Don't these two statements contradict each other? I'm really confused by your terminology.

 

[DarMM]

Don't these two statements contradict each other?

What is the contradiction? My understanding is that there can be elements of a theory that are not "things" but are also not just epistemic quantities. For example the Euler Langrange equations are not "things" of classical theory, i.e. objects posited to exist thus not part of the ontology. They are not however epistemic, they are objective relations between the objects.

Thus they are part of the metaphysics but not elements of the ontology.

As far as I am aware this is not "my terminology" but reasonably common terminology, this article might help:
https://plato.stanford.edu/entries/metaphysics/

 

[PeterDonis]

As far as I am aware this is not "my terminology" but reasonably common terminology, this article might help

I'll take a look. I am generally not a fan of using terminology in physics that is made up by philosophers instead of physics.

 

[DarMM]

I'll take a look. I am generally not a fan of using terminology in physics that is made up by philosophers instead of physics.

Doesn't "ontology" and "epistemic" come from philosophy? Is the issue with using the word "metaphysics" in the sense of seperating "things" from "relations"?

If we don't use any philosophical terminology I guess we could say in Bohmian Mechanics the wave function is a relation obeyed by particles like the Euler Lagrange equations, an equation of motion. Not an actual propagating field like the EM field.

 

[A. Neumaier]

For example the Euler Langrange equations are not "things" of classical theory, i.e. objects posited to exist thus not part of the ontology. They are not however epistemic, they are objective relations between the objects.

Thus they are part of the metaphysics but not elements of the ontology.

As far as I am aware this is not "my terminology" but reasonably common terminology, this article might help:
https://plato.stanford.edu/entries/metaphysics/

You are the very first I know who calls Newton's laws or the Euler-Langrange equations for classical multiparticle systems metaphysics rather than physics.

 

[PeterDonis]

Doesn't "ontology" and "epistemic" come from philosophy?

Yes, and I'm not a fan of those terms. IMO they lead to lots of questions which do not have well-defined answers and do not add anything useful to the physics.

If we don't use any philosophical terminology I guess we could say in Bohmian Mechanics the wave function is a relation obeyed by particles like the Euler Lagrange equations, an equation of motion. Not an actual propagating field like the EM field.

We could just as well say that the EM field is "a relation obeyed by particles", since the only observations we have of the EM field are observations of particles whose motions are affected by the field. So what's the difference between the EM field and the wave function in Bohmian mechanics?

 

[microsansfil]

That's correct, but why does that matter for the wavefunction being non ontological?

Can an entity that is not measurable be considered as being ontological?

/Patrick

 

[DarMM]

You are the very first I know who calls Newton's laws or the Euler-Langrange equations for classical multiparticle systems metaphysics rather than physics.

Why do you think I also don't call them physics?

 

[A. Neumaier]

What is the contradiction? My understanding is that there can be elements of a theory that are not "things" but are also not just epistemic quantities. For example the Euler Langrange equations are not "things" of classical theory, i.e. objects posited to exist thus not part of the ontology. They are not however epistemic, they are objective relations between the objects.

Thus they are part of the metaphysics but not elements of the ontology.

As far as I am aware this is not "my terminology" but reasonably common terminology, this article might help:
https://plato.stanford.edu/entries/metaphysics/

The same encyclopedia says here that
https://plato.stanford.edu/entries/einstein-philscience/ says:

Spacetime coincidences play this privileged ontic role because they are invariant and, thus, univocally determined.

and thus attests that there is an ontic role for relations.

Maybe your usage is specific to recent quantum foundation discussions? I'd like to repeat my request for a reference where I can find a critical discussion of the term.

 

[DarMM]

Can an entity that is not measurable be considered as being ontological?

/Patrick

Yes. Whether that's sensible is another question.

 

[A. Neumaier]

Why do think I also don't call them physics?

Well, maybe you do. But then this also contradicts traditional usage, which sees physics and metaphysics (''after physics'') as disjoint, though loosely related disciplines. From Wikipedia:

For instance, Thomas Aquinas understood it to refer to the chronological or pedagogical order among our philosophical studies, so that the "metaphysical sciences" would mean "those that we study after having mastered the sciences that deal with the physical world"

 

[DarMM]

Yes, and I'm not a fan of those terms. IMO they lead to lots of questions which do not have well-defined answers and do not add anything useful to the physics.

How should we refer to the two different views of the wave function you see in quantum foundations?

We could just as well say that the EM field is "a relation obeyed by particles", since the only observations we have of the EM field are observations of particles whose motions are affected by the field. So what's the difference between the EM field and the wave function in Bohmian mechanics?

Isn't the EM field detected in ways that aren't simply particle effects, is it true it's always the motion of particles? The difference would be that between the EM field and the Euler Lagrange equation. How would you phrase this difference?

 

[DarMM]

Well, maybe you do. But then this also contradicts traditional usage, which sees physics and metaphysics (''after physics'') as disjoint, though loosely related disciplines. From Wikipedia:

For instance one can look at statistical mechanics and say the phase space distribution is epistemic, but that doesn't mean one is calling statistical mechanics "epistemology" and not physics since "physics and epistemology are disjoint subjects"

 

[A. Neumaier]

For instance one can look at statistical mechanics and say the phase space distribution is epistemic, but that doesn't mean one is calling statistical mechanics "epistemology" and not physics since "physics and epistemology are disjoint subjects"

Saying ''the phase space distribution is epistemic'' does not make the phase space distribution metaphysics but only this statement about it. The phase space distribution itself is physics and not metaphysics.

Thus "physics and epistemology are disjoint subjects", though they are loosely related in as much as epistemology uses physical concepts as objects of discourse.

 

[PeterDonis]

How should we refer to the two different views of the wave function you see in quantum foundations?

Whatever terms we use, they are terms of philosophy (or perhaps "metaphysics"), not physics. Both views agree on all predictions for actual experiments, so whatever difference there is between them is not a matter of physics.

Isn't the EM field detected in ways that aren't simply particle effects

What ways? Can you give an example?

 

[DarMM]

Whatever terms we use, they are terms of philosophy (or perhaps "metaphysics"), not physics. Both views agree on all predictions for actual experiments, so whatever difference there is between them is not a matter of physics.

So are we to talk about them at all? They are divisions in approaching QM that have led to highly cited results. If you prefer not to use them then how do we deal with the fact that they're commonly used and motivate research?

What ways? Can you give an example?

I think this might go off on another difficult tangent. I think it's better to focus on the basic difference. To you is there a distinction between the electric field and the Euler Langrange equations? If yes, the wave function in Bohmian Mechanics is like the latter.

 

[DarMM]

Saying ''the phase space distribution is epistemic'' does not make the phase space distribution metaphysics but only this statement about it. The phase space distribution itself is physics and not metaphysics.

Thus "physics and epistemology are disjoint subjects", though they are loosely related in as much as epistemology uses physical concepts as objects of discourse.

Yes, it is the same above regarding Euler Lagrange equation. They're physics, but that classification is a metaphysical one.

 

[PeterDonis]

the wave function in Bohmian Mechanics is like the latter.

Why?

 

[DarMM]

Why?

Part 12 of this paper has an exposition:
https://arxiv.org/abs/quant-ph/9512031

 

[A. Neumaier]

Yes, it is the same above regarding Euler Lagrange equation. They're physics, but that classification is a metaphysical one.

The classification of particle positions in Bohmian mechanics as 'real' or 'ontic' - i.e., the common interpretation of Bohmian mechanics - is also metaphysical, though Bohmian mechanics itself is, as an extension of the quantum mechanical formalism, (potentiallyy falsifiable) physics.

 

[PeterDonis]

Part 12 of this paper has an exposition

So basically, the criterion being given is that for something to be "real" it must be capable of being acted on by other "real" things, whereas the wave function is not acted on by anything. But this doesn't seem right, because the wave function is determined by Schrödinger's Equation, which includes the potential energy, and the potential energy is a function of the particle configuration. So I don't think it's correct to say that the wave function is not acted on by anything.

 

[DarMM]

So I don't think it's correct to say that the wave function is not acted on by anything.

I don't necessarily disagree, I was only stating what the "common" Bohmian position is.

 

[DarMM]

The classification of particle positions in Bohmian mechanics as 'real' or 'ontic' - i.e., the common interpretation of Bohmian mechanics - is also metaphysical, though Bohmian mechanics itself is, as an extension of the quantum mechanical formalism, (potentiallyy falsifiable) physics.

Agreed.

 

[kith]

We could just as well say that the EM field is "a relation obeyed by particles", since the only observations we have of the EM field are observations of particles whose motions are affected by the field. So what's the difference between the EM field and the wave function in Bohmian mechanics?

Without particles, there is no physical system which could be described by a wave function (in non-relativistic QM). There could still be an electric field because the Maxwell equations admit vacuum solutions. To me, this seems like an important difference and it justifies the view of the electromagnetic field as a physical system in its own right instead of a tool to describe particle interactions.

 

[A. Neumaier]

the wave function is determined by Schrödinger's Equation, which includes the potential energy, and the potential energy is a function of the particle configuration. So I don't think it's correct to say that the wave function is not acted on by anything.

But the coordinates are dummy variables, not Bohmian positions.

 

[PeterDonis]

Without particles, there is no physical system which could be described by a wave function (in non-relativistic QM). There could still be an electric field because the Maxwell equations admit vacuum solutions.

This seems to me to be a red herring. "Vacuum solutions" of the Maxwell equations just means there's no charge-current density "source" on the RHS. But there is no analogue to this "source" in Schrödinger's Equation anyway; the only "current density" is $\psi^* \psi$, which is always "there" anyway.

Also, talking about particles as the only "physical system which could be described by a wave function" already presupposes that the wave function itself is not "real". If the wave function itself is "real", there doesn't need to be any other physical system that it describes; it is a physical system all by itself.

 

[kith]

This seems to me to be a red herring. "Vacuum solutions" of the Maxwell equations just means there's no charge-current density "source" on the RHS. But there is no analogue to this "source" in Schrödinger's Equation anyway; the only "current density" is $\psi^* \psi$, which is always "there" anyway.

I need to better understand what you mean by the wave function being a physical system in order to comment on this. Right now it doesn't seem sensible to me to compare the equations like this.

Also, talking about particles as the only "physical system which could be described by a wave function" already presupposes that the wave function itself is not "real".

I just noted that in non-relativistic QM, the notion of particles (or more generally the notion of the physical system) comes before the notion of the wave function.

Similarly, in classical mechanics, the notion of particles comes before the notion of velocity. I wouldn't say that velocity being secondary to particles makes it not "real" but since "real" is notoriously ambiguous (in this thread alone, I count at least three different usages) I'd like to avoid this terminology completely.

If you are saying that I presuppose that the wave function is not a physical system similar to velocity not being a physical system in classical mechanics, I agree that I do this.

If the wave function itself is "real", there doesn't need to be any other physical system that it describes;[...]

Also in dBB -which the discussion seemed to be about- there needs to be a physical system which determines the observables. No such system, no wave function.

[...] it is a physical system all by itself.

I have a hard time wrapping my head around this.

It makes a certain sense to me to regard the electromagnetic field as a tool for accounting for complicated direct particle interactions which gives it the same status as the wave function in Copenhagenish interpretations. That we usually don't do this but imagine the electromagnetic field as an independent entity has clear reasons: I can shoot a light pulse into empty space and have a mathematical description for what happens there in the absence of matter. This motivation is absent for the wave function. I cannot create a wave packet which propagates in the absence of matter (in NRQM). No matter, no wave function.

What does make a certain sense to me is to question the primacy of matter and say that the wave function and matter are intertwined. But I don't see any motivation to treat the wave function as a physical system all by itself.

 

[PeterDonis]

in dBB -which the discussion seemed to be about- there needs to be a physical system which determines the observables.

Yes, agreed.

No such system, no wave function.

I don't agree with this. The wave function is a separate thing from the Bohmian particle positions. It even has its own evolution equation (I know I said earlier that I thought that equation depends on the particle positions through the potential energy, but after @A. Neumaier's response to that I'm not so sure). So it's perfectly possible to look at the wave function in dBB as a separate entity, not dependent on particle positions.

 

[A. Neumaier]

@kith and @PeterDonis points of view can be reconciled by differentiating between the abstract physical system (n scalar particles in 3-space) and a concrete physical system (defined by its trajectory of states) providing an instance of it. The former is prior to the latter and determines the kinematics, but the latter is needed to make predictions.