Is quantum theory a microscopic theory?

[vanhees71]

The problem here is that you formulate the things in too abstract a way. You cannot simply say, "I measure a beam of light in the photon basis or the field basis". I've no clue what you mean. So I have to guess: The "photon basis" may be the Fock basis, i.e., states of the em. field with a defined total number of photons. I'm a bit at lost how to realize such a measurement. Do you know of any real-world device that measures only photons if they are a prepared in a photon-number eigenstate? I've no clue, how to construct such a device with real-world materials. Also what do you mean by "field basis"? Are these coherent states?

There's a well-developed subfield of relativstic QFT called quantum optics, which clearly defines, what's observed in experiments. They have all kinds of measurement devices. Most of them are based on the photo effect, i.e., an electromagnetic wave (no matter in which state it is prepared, be it a Fock state (not so easy to do, but standard today with parametric down conversion, or a coherent/squeezed state (lasers are your friend), is interacting with electrons which undergo a transition from a bound to a continuum state, and this signal is amplified to make these photoelectrons countable. In this way you can measure correlation functions of the electromagnetic (in the usually sufficient lowest-order dipole approximation the electric) field.

 

[Dr. Courtney]

I agree. But the minimal instrumental version of QM tries to deny it.

I suspect the "denial that every theory is macroscopic" is not unique to the instrumental version of QM.

How would biologists view assertions that cell theory and the germ theory of disease are not really microscopic theories but "microscopic interpretations of macroscopic events"? How would chemists view assertions that atomic theory (from Dalton) and kinetic theory are not really microscopic theories but "microscopic interpretations of macroscopic events"? I expect they can follow the logic, but this is not really how "microscopic" is used in other fields of science.

One could construct a similar logic regarding scientific reconstructions of past events since these are not observed directly but are inferred from modern observations. "The Big Bang theory is not properly a theory of origins, but it is a historical interpretation of modern events." In the same way that one cannot separate microscopic interpretations from the macroscopic things humans actually observe, one cannot separate historical interpretations from the modern events that humans actually observe.

 

[atyy]

I suspect the "denial that every theory is macroscopic" is not unique to the instrumental version of QM.

How would biologists view assertions that cell theory and the germ theory of disease are not really microscopic theories but "microscopic interpretations of macroscopic events"? How would chemists view assertions that atomic theory (from Dalton) and kinetic theory are not really microscopic theories but "microscopic interpretations of macroscopic events"? I expect they can follow the logic, but this is not really how "microscopic" is used in other fields of science.

One could construct a similar logic regarding scientific reconstructions of past events since these are not observed directly but are inferred from modern observations. "The Big Bang theory is not properly a theory of origins, but it is a historical interpretation of modern events." In the same way that one cannot separate microscopic interpretations from the macroscopic things humans actually observe, one cannot separate historical interpretations from the modern events that humans actually observe.

The problem is we would like a theory of reality, but quantum mechanics (as we understand it) is simply not such a theory. Of course it is absurd to say that germs don't really exist or that the measuring apparatus is not made of electrons - but the formalism does not grant us the ability to say these things - it is an open problem of quantum mechanics.

 

[Demystifier]

Conservation of elephants forbids it.

Can this conservation law be derived from a symmetry? Which symmetry?

 

[DarMM]

The problem here is that you formulate the things in too abstract a way

It's the old problem of complimentary between results in different basis. I could make things more precise, but it doesn't really change the point. Why would the details matter? You'll still get complimentary where the statistics in different bases can't be considered as marginals of one set of properties.
It was just an attempt to say why complimentarity makes viewing things in a way that's detached from your device is difficult, it's not intended as a completely accurate rendering of quantum optics.

 

[DarMM]

Can this conservation law be derived from a symmetry? Which symmetry?

Trunkal translations.

 

[Demystifier]

Trunkal translations.

Reference please!

 

[Demystifier]

I suspect the "denial that every theory is macroscopic" is not unique to the instrumental version of QM.

How would biologists view assertions that cell theory and the germ theory of disease are not really microscopic theories but "microscopic interpretations of macroscopic events"? How would chemists view assertions that atomic theory (from Dalton) and kinetic theory are not really microscopic theories but "microscopic interpretations of macroscopic events"? I expect they can follow the logic, but this is not really how "microscopic" is used in other fields of science.

One could construct a similar logic regarding scientific reconstructions of past events since these are not observed directly but are inferred from modern observations. "The Big Bang theory is not properly a theory of origins, but it is a historical interpretation of modern events." In the same way that one cannot separate microscopic interpretations from the macroscopic things humans actually observe, one cannot separate historical interpretations from the modern events that humans actually observe.

In principle one can reason like that in any science, but in reality such reasoning can only be found in quantum foundations. Why is quantum theory different? I think the main reason is that some physicists like to think quantum theory is very close to the final fundamental theory of everything, so they don't like to think that there is some hidden reality on which quantum theory has nothing to say. Instead, they prefer to think that there is no any hidden reality at all. All Copenhagen-like interpretations of QM are nothing but attempts to justify the ideology that QM must somehow be complete.

 

[vanhees71]

It's the old problem of complimentary between results in different basis. I could make things more precise, but it doesn't really change the point. Why would the details matter? You'll still get complimentary where the statistics in different bases can't be considered as marginals of one set of properties.
It was just an attempt to say why complimentarity makes viewing things in a way that's detached from your device is difficult, it's not intended as a completely accurate rendering of quantum optics.

I'm not saying you should make things more precise, but you should make a statement about what you mean by "measuring" in the one or the other basis. Particularly to understand complementarity (one of Bohr's enigmatic terms) right, you need to think about many concrete experiments, i.e., descriptions of real-world measurement devices applied to concrete real-world preparations of measured objects. I don't know of a single example, where this has not resolved apparent quibbles with overly abstract formulations like "measuring in a basis".

Quantum optics is just the example, where the real-world descriptions of measurement and preparation processes become most transparent, because they are usually not that complicated after all. You need just a good grasp of classical optics. To understand all kinds of devices like lenses, polarizers, beam splitters even linear electrodynamics as taught in classical electrodynamics is sufficient. For parametric down conversion and preparation of proper Fock states also some nonlinear optics is useful. The quantization part is also not too hard. Usually you just need a good understanding of the quantization of the free electromagnetic field and some perturbation theory for the interactions with matter. The latter almost always can be described with non-relativistical QM.

 

[DarMM]

I'm not saying you should make things more precise, but you should make a statement about what you mean by "measuring" in the one or the other basis. Particularly to understand complementarity (one of Bohr's enigmatic terms) right, you need to think about many concrete experiments, i.e., descriptions of real-world measurement devices applied to concrete real-world preparations of measured objects. I don't know of a single example, where this has not resolved apparent quibbles with overly abstract formulations like "measuring in a basis".

A good understanding of how concrete experiments work resolves the conceptual difficulties with complementarity? Even if I specified the exact devices and what they measure it wouldn't do this. I can't imagine how it would. It would still reduce to the fact that the statistics for two different measurements don't seem to be marginals of a third more complete measurement as is always the case in classical mechanics.

For example an $x$ measurement and a $p$ measurement on a single particle cannot in general be considered as a coarse graining of some $(x,p)$ measurement, preventing you from considering them as measurements of properties already present.

 

[Lord Jestocost]

Perhaps, but the minimal instrumental view of QM says nothing about that.

I don’t see any way out of the instrumentalist minimal interpretation. Our perceptions of events occurring on a macroscopic space-time scene (the “empirical reality”) cannot be traced back to the “behavior” of fundamental microscopic space-time realities (the mysterious “things” behind the space-time scene which we denote as a matter of convenience “electrons”, “atoms” etc. allow no space-time description).

J. Robert Oppenheimer in “Atom and Void: Essays on Science and Community”

“If we ask, for instance, whether the position of the electron remains the same, we must say "no"; if we ask whether the electron's position changes with time, we must say "no"; if we ask whether the electron is at rest, we must say "no"; if we ask whether it is in motion, we must say "no." The Buddha has given such answers when interrogated as to the conditions of a man's self after his death; but they are not familiar answers for the tradition of seventeenth- and eighteenth-century science.”

 

[Demystifier]

I don’t see any way out of the instrumentalist minimal interpretation. Our perceptions ...

A way out is to reject that science must only be based in perceptions.
https://www.physicsforums.com/threads/is-quantum-theory-a-microscopic-theory.974961/#post-6208601

 

[vanhees71]

What's a conceptual difficulty with complementarity? Quantum theory resolves all these difficulties, as far as I know, very well. Where are, in your opinion, issues with the position and momentum observables left?

 

[Lord Jestocost]

A way out is to reject that science must only be based in perceptions.

A way out might be that we at first accept that this what we call 'reality' is just a state of mind.
https://www.theguardian.com/science/blog/2009/mar/17/templeton-quantum-entanglement

 

[DarMM]

What's a conceptual difficulty with complementarity? Quantum theory resolves all these difficulties, as far as I know, very well. Where are, in your opinion, issues with the position and momentum observables left?

There's no issues with these observables and of course QM deals with how these things work very well. It's just odd, I don't think complementarity makes more intuitive sense once one considers the realistic details of the device. Let me try again.

Say we observe two quantities $N$ and $M$ with two separate devices and they can have outcomes $n$ and $m$. In classical mechanics their statistics are basically always modeled as some distribution over the space of pairs $(n,m)$. So even if you don't measure $M$ it can still be reasoned about.
Even if the world was fundamentally random but random in the sense of a classical stochastic theory this would be the case.

In QM however if $N$ and $M$ don't commute then this isn't true. The statistics of $N$ observations are not marginals for statistics of $(n,m)$ pairs, but simply a distribution over $n$ outcomes. This means when you measure $N$ you can't reason about some outcome $M$ had that you don't know. So basically $M$ events can't even be discussed. Only the quantity you measured has an outcome.

Now it seems to be what is true and QM models it perfectly but it's pretty weird and I don't think that weirdness goes away because of an accurate comprehension of devices. It seems that the device is embedded in one's description of the system in a way that isn't true in classical mechanics. Even in a fundamentally random classical theory.

 

[Mentz114]

Conservation laws are not enough. For example, from conservation laws alone you cannot deduce that the Moon is there when nobody observes it. As far as conservation laws are concerned, the Moon could spontaneously turn into a gigantic pink elephant of the same energy-momentum as that of the Moon, whenever it is no longer observed.

We only have conservation laws to constrain physical models. There is no need to 'deduce' the persistent existence of the moon unless you believe it is not there when your eyes are closed.

Earlier you deny that perceptions are relevant to physics and here you contradict this.

 

[vanhees71]

There's no issues with these observables and of course QM deals with how these things work very well. It's just odd, I don't think complementarity makes more intuitive sense once one considers the realistic details of the device. Let me try again.

Say we observe two quantities $N$ and $M$ with two separate devices and they can have outcomes $n$ and $m$. In classical mechanics their statistics are basically always modeled as some distribution over the space of pairs $(n,m)$. So even if you don't measure $M$ it can still be reasoned about.
Even if the world was fundamentally random but random in the sense of a classical stochastic theory this would be the case.

In QM however if $N$ and $M$ don't commute then this isn't true. The statistics of $N$ observations are not marginals for statistics of $(n,m)$ pairs, but simply a distribution over $n$ outcomes. This means when you measure $N$ you can't reason about some outcome $M$ had that you don't know. So basically $M$ events can't even be discussed. Only the quantity you measured has an outcome.

Now it seems to be what is true and QM models it perfectly but it's pretty weird and I don't think that weirdness goes away because of an accurate comprehension of devices. It seems that the device is embedded in one's description of the system in a way that isn't true in classical mechanics. Even in a fundamentally random classical theory.

It's only weird if you insist on a notion of "state" that is not in accordance with observations. As far as we know the notion of "state" is how QT describes it and not as how classical physics describes it. The weirdness goes away as soon as you accept that nature behaves as she does and doesn't care about what humans my consider weird.

 

[DarMM]

Well certainly, but I think most people will find it odd that you can't consider a microscopic system independently of the device like you can in other physical theories. It seems as if it should be possible to talk about things in and of themselves.

According to the standard reading of QM you can't. Your approach, i.e. just get used to it, is sensible but goes against the intuitions of many who feel science should give you a picture of the world. Not necessarily an intuitive picture. People are fine with GR despite the fact that it is unintuitive since it discusses things as they are when no measuring devices are present.

 

[Mentz114]

Conservation laws are not enough. For example, from conservation laws alone you cannot deduce that the Moon is there when nobody observes it. As far as conservation laws are concerned, the Moon could spontaneously turn into a gigantic pink elephant of the same energy-momentum as that of the Moon, whenever it is no longer observed.

Only the Moon has the same Hamiltonian and the same number of dof as the Moon. Your transformation would require dumping all internal energy dof. It is not true, except in a gross approximation.

 

[DarMM]

Only the Moon has the same Hamiltonian and the same number of dof as the Moon. Your transformation would require dumping all internal energy dof. It is not true, except in a gross approximation.

Note he did say the elephant is pink. A blue elephant would be forbidden by internal energy considerations.

 

[Jimster41]

Just trying to use this dialogue to get a better lay persons understanding: I am familiar with the complimentarity of position and momentum, but also there are spin states that are not compatible correct? There are also others?

Are the number of degrees of freedom that display such contextual inter-relation infinite or finite? I thought they were very finite ie. may be a dumb question.

 

[DarMM]

Just trying to use this dialogue to get a better lay persons understanding: I am familiar with the complimentarity of position and momentum, but also there are spin states that are not compatible correct? There are also others?

Are the number of degrees of freedom that display such contextual inter-relation infinite or finite? I thought they were very finite ie. may be a dumb question.

It depends on how you count it. For example position $x$ is incompatible with momentum $p$. However both are also incompatible with $xp$.
Technically there are an infinite number of incompatible observables, although you might want to only consider basic ones like $x$ and $p$.

In QFT however there is an infinite amount of even the basic ones.

 

[Jimster41]

In QFT however there is an infinite amount of even the basic ones.

Is that because in QFT the standard model particles are considered to be excited states of the field, and so the fundamental object is the field which could be excited a potentially infinite number of ways?

 

[DarMM]

Is that because in QFT the standard model particles are considered to be excited states of the field, and so the fundamental object is the field which could be excited a potentially infinite number of ways?

Without going into much detail, basically yes. There is still the issue mentioned above that in a Copenhagen reading "field" is a type of reaction in a device treated classically.

 

[microsansfil]

Perhaps, but the minimal instrumental view of QM says nothing about that.

Moreover, the concept of existence is a concept from the field of metaphysics, but not from the field of physics.

/Patrick

 

[Auto-Didact]

It's only weird if you insist on a notion of "state" that is not in accordance with observations. As far as we know the notion of "state" is how QT describes it and not as how classical physics describes it. The weirdness goes away as soon as you accept that nature behaves as she does and doesn't care about what humans my consider weird.

This often repeated meme that humans have a cognitive bias against QM due to natural selection isn't actually an answer but a copout; even worse, it is an incoherent philosophical ideology parading as science. The very existence of Bohmian mechanics even reduces this meme into absurdity.

The uncomfortableness isn't a matter of interpretative human psychology but a matter of mathematical self-consistency; the fact that in the minimal interpretation of QM things cannot be defined without making references to macroscopic devices simply means that this theoretical construction is de facto fundamentally logically inconsistent.

Putnam et al. have argued on this basis that QM actually falsifies standard logic and a new form of logic is needed, e.g. quantum logic; I myself have argued this point for years. The problem is that the 'necessity of such non-standard logics'-argument just seems to be flat out wrong.

 

[Jimster41]

...The problem is that the 'necessity of such non-standard logics'-argument just seems to be flat out wrong.

I didn’t quite follow that. You mean the argument for these non-standard logics is the wrong argument for the right logics?

 

[Auto-Didact]

I didn’t quite follow that. You mean the argument for these non-standard logics is the wrong argument for the right logics?

The Putnam argument - that QM falsifies the universal validity of standard logic and that there is therefore a necessity for a non-standard logic such as quantum logic - is wrong.

Contrary to Putnam et al., QM in fact does not falsify the validity of standard logic, because Bohmian mechanics can be completely described and understood using standard logic.

If anything the logical - and therefore mathematical - self-inconsistency of QM is exposed as being an inadequacy of the idealized mathematical framework underlying QM, which completely disappears once the extended mathematical framework of Bohmian mechanics is adopted.

This extended mathematical framework is essentially a proper complex analytic formulation of Hamilton-Jacobi theory; Bohmian mechanics is based on this formulation, while textbook QM instead makes do with the more limited Hamiltonian mechanics and then just pretends - through purely philosophical rhetoric - that the existence or construction of any such more extended mathematical frameworks is just impossible.

 

[Jimster41]

Is there any easy way to summarize the key difference between Bohmian Mechanics and QFT. My cartoon of QFT is that the fields are space-time non-local, which is also a big part of my cartoon of Bohm’s pilot wave. I could really use a cartoon of their disagreement.

 

[Auto-Didact]

Bohmian mechanics, just like QM and Newtonian mechanics, is a theory which respects Galilean relativity. The non-locality of BM (and QM) is due to the wavefunction existing and evolving in configuration space.

Quantum field theory on the other hand is a field theoretic extension of QM, which moreover respects special relativity. In this sense, QFT is a completely local relativistic field theory, where the quantum fields exist in flat spacetime.

 

[atyy]

A way out might be that we at first accept that this what we call 'reality' is just a state of mind.
https://www.theguardian.com/science/blog/2009/mar/17/templeton-quantum-entanglement

Reality is just a tool to predict the results of observations. Unfortunately, our minds seem to think that that our state of mind is ordered - then it asks - are there laws that govern the state of mind? - then it ends up again with the measurement problem.

 

[PeterDonis]

Reality is just a tool to predict the results of observations.

This doesn't seem right. I would say reality is whatever-it-is that is producing the actual results of our observations, and models are the tools we use to predict the results of our observations; we then compare the predicted results with the actual results to see how accurate our models are, and to improve them.

 

[atyy]

It's only weird if you insist on a notion of "state" that is not in accordance with observations. As far as we know the notion of "state" is how QT describes it and not as how classical physics describes it. The weirdness goes away as soon as you accept that nature behaves as she does and doesn't care about what humans my consider weird.

That is precisely why QM is weird - it doesn't allow one to describe how nature behaves as she does without caring about the observer. QM does not describe nature. As Bohr said, "There is no quantum world. There is only an abstract quantum physical description. It is wrong to think that the task of physics is to find out how nature is. Physics concerns what we can say about nature."

 

[Auto-Didact]

This doesn't seem right. I would say reality is whatever-it-is that is producing the actual results of our observations, and models are the tools we use to predict the results of our observations; we then compare the predicted results with the actual results to see how accurate our models are, and to improve them.

With respect to physics, reality is just another word for ontology, while our observations and models thereof are phenomenology; that the two need not immediately coincide conceptually is true for any epistemic question i.e. for any scientific question.

That is precisely why QM is weird - it doesn't allow one to describe how nature behaves as she does without caring about the observer. QM does not describe nature. As Bohr said, "There is no quantum world. There is only an abstract quantum physical description. It is wrong to think that the task of physics is to find out how nature is. Physics concerns what we can say about nature."

In other words, Bohr believed that QM was purely an epistemic theory without any ontology, just like statistics is an epistemic theory without any ontology. Physics has always been about ontology; any suggestion that this is not so is purely postmodern philosophical rhetoric which is astutely unaware of the history and philosophy of physics and mathematics and their relationship.

 

[atyy]

This doesn't seem right. I would say reality is whatever-it-is that is producing the actual results of our observations, and models are the tools we use to predict the results of our observations; we then compare the predicted results with the actual results to see how accurate our models are, and to improve them.

Of course. What I mean is that whether one takes the common sense view of reality or an operational view of reality as a tool, one ends up with the measurement problem. One can say I don't care about the problem, but one cannot say it doesn't exist (unless one believes it is already solved by MWI etc).