Insights Why the Quantum | A Response to Wheeler's 1986 Paper - Comments

[Mentz114]

I sketched this in another post a while back.

[..]

So the Born rule in my understanding requires a distinction between macroscopic coarse-grained descriptions (where the rule applies) and microscopic descriptions (where it does not).

Thanks ! I think Sewell and some of the refs therein have something about this. I will reply ( if this thread is dead I'll start a new one).

[Edit]
@stevendaryl
This a huge subject but some re-readings suggest that if the coarse graining results in a big system that has the same eigenstates as the small grained system, then the Born rule applies to both.

(I found this fun paper which is not relevant but short and interesting)
Coarse graining: lessons from simple examples
https://arxiv.org/pdf/physics/0101077.pdf

 

[Fra]

The "wave function" is a probability amplitude by definition (within the standard minimal interpretation). Thus it's probabilistic from the very beginning, without any necessity to introduce classical concepts.

A probability distribution itself is classical statistical concept, involving no uncertainties. At this level quantum mechanics is just a deterministic theory as is Newtons mechanics.

The laws of quantum theory deductively infers distribution of events, given a preparation. So the heart of the predictions is at the level of distributions.

Its just the link to single outcomes that is probabilistic. But this link, depends on a definite distribution; which IMO is anchored in the observer part of the system. And the reason this is considered to be in the realm of classical mechanics is that intercommunication within the measurment device is considered trivial in comparasion. One effectively assumes that (if we forget about relativity for a second) that all classical observers are equivalent, and thus we attain objectivity. But this objectivity (observer equivalent) only is manifested in the classical realm.

Ie. without a classical context for the measuremnt device, you can not defined a definite distribution, and not even a certain probability. Then even the probability gets "undertain", in an uncontrollable way.

/Fredrik

 

[Fra]

There is no distinction between measurements and other interactions.

In the way i am sure you mean it i fully agree.

But the distinction is in its description; and the description (and the expectations) are encoded in the observer part. The "questions asked" about an subatomic system, are in a deep way "formulated" and encoded physically in the observing system. The computational inference machinery required, for constructing questions (ie. observations) live in the observer part of the cut in my view.

If we relax this (which takes us beyond the standard theory) things become very complicated. Its to avoid this we need the "classical reference". Of course my opinon is that at some point we need to face these problems, but that is exactly the questions we need to ask to go beyond QM as it stands, to understand QG and unification imo.

/Fredrik

 

[atyy]

Why aren't there situations where quantum states are naturally falling into eigenstates of some operator - without measurement - for instance on a star?
This could happen trillions of times and thus a probability distribution. Or is any time a quantum states projects onto an eigen state of an operator a measurement by definition?

This is not standard quantum mechanics. This is what is proposed in attempts to solve the measurement problem such as the physical collapse theories like GRW. Although vanhees71 is an expert on quantum field theory, in these fundamental and basic points, he is in contradiction to almost all standard textbooks of quantum physics.

 

[vanhees71]

I am not! If you read the physics content of all standard textbooks, all there is predicted are probabilities for the outcome of measurements, and these predictions are in excellent agreement with all experiments done so far. That's the core of quantum theory, and that's the physics described by it. It's called the minimal statistical interpretation, and it's within the Copenhagen class of interpretation, taken away the unnecessary problematic parts, i.e., the collapse (in contradiction with relativistic space-time structure and causality) and a quantum-classical cut, which nobody has ever been able to demonstrate experimentally. To the contrary, the more advanced (quantum) engineering gets, the larger systems can be prepared in states that behave "quantum like" not "classical like", although the common "classical-like states" of everyday matter around us is of course also a quantum state. Classical physics is a limit for classical behavior of macroscopic properties which are coarse-grained quantities that averaged over many microscopic degrees of freedom. The rest is the math of the central-limit theorem of standard probability theory.

 

[stevendaryl]

I am not! If you read the physics content of all standard textbooks, all there is predicted are probabilities for the outcome of measurements, and these predictions are in excellent agreement with all experiments done so far. That's the core of quantum theory, and that's the physics described by it. It's called the minimal statistical interpretation, and it's within the Copenhagen class of interpretation, taken away the unnecessary problematic parts, i.e., the collapse (in contradiction with relativistic space-time structure and causality) and a quantum-classical cut, which nobody has ever been able to demonstrate experimentally. To the contrary, the more advanced (quantum) engineering gets, the larger systems can be prepared in states that behave "quantum like" not "classical like", although the common "classical-like states" of everyday matter around us is of course also a quantum state. Classical physics is a limit for classical behavior of macroscopic properties which are coarse-grained quantities that averaged over many microscopic degrees of freedom. The rest is the math of the central-limit theorem of standard probability theory.

I'm not disagreeing with the claim that quantum mechanics makes good predictions, I'm just saying that it is patently wrong to say that it makes those predictions without distinguishing measurements from non-measurements.

Bringing up the central limit theorem is just not relevant to this question. It's a non-sequitur. It's possible (in principle, if not in practice) to treat a measurement interaction quantum-mechanically, but when you do so, the probabilities disappear. To recover probabilities, you need yet another system that is not treated quantum-mechanically that will measure the measuring device. There are no probabilities associated with a pure quantum-mechanical system. At least not in the minimal interpretation. That's why I say that bringing up the central limit theorem is a non-sequitur. The central limit theorem is concerned with probabilities, and the issue is whether there are any probabilities at all involved in a quantum system where you treat everything (including observers and measurement devices) quantum-mechanically. Invoking the central limit theorem is assuming your conclusion.

I don't see why this is even controversial. The basic assumptions of the "minimalist interpretation" only say what happens when a measurement is performed. That's very different from the assumptions of Newtonian mechanics, which say what happens when massive particles interact through forces. Whether or not anything is measured it doesn't make any difference.

 

[stevendaryl]

This seems pretty straightforward: If there is no distinction between measurement-like interactions and non-measurement interactions, then it should be possible to formulate the minimalist interpretation in which the word "measurement" is replaced by its definition---something like "an interaction between two systems such that a property of one system causes a macroscopic change in the other system". If you try to do that, you will see that the minimalist interpretation inherently involves a microscopic/macroscopic distinction.

 

[vanhees71]

Again you simply make bold claims without explanation. To make my still unanswered question very simple: What's the (principle) difference between the interaction of a photon hitting a CCD screen (measurement device) and just some other plane like my desk? I don't see, where there should be a difference. It's all the good old electromagnetic interaction, isn't it? Of course, if you think photons to be too special (and they are special), just take any massive particle you like to explain clearly in physical terms the difference between interacting of the particle with a measurement device and just matter that isn't used as a measurement device.

To be honest, I think it's ridiculous to think that there are different laws for this interaction simply because once the material is used as a measurement apparatus and the other time it's not. The very design of any physical measurement device (starting from something as simple as a yardstick up to the most complicated high-accuracy devices used for high-precision measurements in (sub-)atomic physics) are based on the fundamental laws of physics, which are believed to hold true universally and do not have exception only because something is used as a measurement device. There's even no different physical law for things living or non-living. There's no "vis viva" but just the fundamental interactions of physics at work also in living organisms. This is just another example for claims in the past that physical laws might not be universal. It's one of the great achievements of science to find universal laws. Although being far from trivial to exist, all quantitative and qualitative experience shows this universality.

 

[stevendaryl]

Again you simply make bold claims without explanation.

I'm not making a claim---I'm pointing out that what you are claiming is just not true. The minimalist interpretation makes a distinction between a measurement and other kinds of interactions. It's right there in the definition of how the wave function is interpreted. I'm not making a claim about quantum mechanics; it's certainly possible that there could be an interpretation that doesn't make such a distinction (maybe Many-Worlds, or maybe Bohmian). But that isn't the minimalist interpretation.

To make my still unanswered question very simple: What's the (principle) difference between the interaction of a photon hitting a CCD screen (measurement device) and just some other plane like my desk? I don't see, where there should be a difference

I agree. There shouldn't be a difference. But the minimalist interpretation requires a difference. So the minimalist interpretation is unsatisfactory for that reason. It's fine as a rule of thumb, but it can't be literally true.

 

[vanhees71]

It's really hard to discuss with people making claims without explaining them sufficiently so that a simple-minded physicist can follow. Now you claim again that the minimal statistical interpretation requires a difference, and again I ask, which difference that might be! I've really no clue, and I'm curious about the answer!

 

[stevendaryl]

In the minimalist interpretation, a measurement plays two different roles:

1. It's a physical interaction between a small system and a larger system. Presumably this interaction is describable by quantum mechanics.
2. It serves to pick out a basis.

Quantum amplitudes are not probabilities until a basis is chosen. You cannot (or at least, I've never seen it done) make sense of amplitudes as probabilities without picking a basis. It's the second role of a measurement that distinguishes measurements from other interactions.

 

[stevendaryl]

It's really hard to discuss with people making claims without explaining them sufficiently so that a simple-minded physicist can follow.

That's my complaint about what you have said with regard to the minimalist interpretation. They make no sense to me. You have a theory whose assumptions explicitly mention measurement, and then you claim that there is nothing special about measurement. That seems like you're contradicting yourself.

Maybe there is a way to resolve the contradiction, but the minimal interpretation certainly doesn't.

 

[stevendaryl]

It's really hard to discuss with people making claims without explaining them sufficiently so that a simple-minded physicist can follow. Now you claim again that the minimal statistical interpretation requires a difference, and again I ask, which difference that might be! I've really no clue, and I'm curious about the answer!

Please state the assumptions of the minimalist interpretation without using the words measurement or macroscopic or observer. Until you can do that, what you're saying makes no sense to me.

 

[vanhees71]

In the minimalist interpretation, a measurement plays two different roles:

1. It's a physical interaction between a small system and a larger system. Presumably this interaction is describable by quantum mechanics.
2. It serves to pick out a basis.

Quantum amplitudes are not probabilities until a basis is chosen. You cannot (or at least, I've never seen it done) make sense of amplitudes as probabilities without picking a basis. It's the second role of a measurement that distinguishes measurements from other interactions.

The basis chosen is dicated by the measured observable in the usual way (eigenstates of the corresponding self-adjoint operator representing this observable). That's part of the basic postulates of minimally interpreted QT. At least in statement 1. we start to agree (I hope): There's no difference in interactions between measurement devices and any other piece of matter, which isn't used as a measurement device.

That the contrary is a pretty strange idea becomse also clear as follows: Suppose there's a difference on a fundamental level between a measurement apparatus and just an arbitrary piece of matter, that difference occurs as soon as the apparatus is used to measure something. So I let the measured system interact with the apparatus. At this point it's a "usual interaction" according to your previous claim, as far as I understand. Now I (or my dog or an amoeba?) decides to look at the pointer reading of the device, and all of a sudden the "usual interaction" turns to an "unusual measurement", or how else should I understand the claim? I think this view is due to some Copenhagen flavors, claiming that the mind of an observer is important part of the measurement process. In its extreme form, the Princeton interpretation, only when reading the pointer of the measurement device, the "state collapses", and this collapse is not within the laws of QT. This is precisely what's avoided in the minimal statistical interpretation, not claiming that there's a collapse or any other necessity for "extra rules for measurements".

 

[vanhees71]

Please state the assumptions of the minimalist interpretation without using the words measurement or macroscopic or observer. Until you can do that, what you're saying makes no sense to me.

Why should I do that, because I never claimed that this is the goal. Physics is about measurements and thus to some degree also observers (if you call a computer storage, that saves the outcome of measurements automatically an observer is your choice). The only thing I'm saying is that according to the minimal interpretation there's (a) no difference in the physical laws between situations where a measurement apparatus is used and where this is not the case and (b) that there's no difference between the physical laws concerning many-body systems making up measurement devices and any other quantum system, large or small. Of course, to make a measurement we need a macroscopic device to be able to make a measurement. I've never claimed the contrary. The only thing I'm saying is that the classical behavior of macroscopic observables does not contradict the fundamental laws of quantum theory but are well explained by standard (quantum!) statistical physics.

 

[vanhees71]

That's my complaint about what you have said with regard to the minimalist interpretation. They make no sense to me. You have a theory whose assumptions explicitly mention measurement, and then you claim that there is nothing special about measurement. That seems like you're contradicting yourself.

Maybe there is a way to resolve the contradiction, but the minimal interpretation certainly doesn't.

I think, we go in circles here. It is very clear that physical theories are about describing measurements, i.e., quantative observations of Nature. What else should physics be about?

The measurement devices used are just made of ordinary matter and are thus described by standard quantum physics as any other lump of matter. That's all I'm saying, and that's how experimentalists construct their measurement devices, i.e., using standard (quantum) physics.

 

[stevendaryl]

Why should I do that, because I never claimed that this is the goal.

Because if you are able to do that, that would demonstrate that the minimalist interpretation does not treat measurement different from other interactions. Nothing short of that would suffice.

Suppose I have a law of physics that states that cats always land on their feet. Does that treat cats differently than other objects? Maybe, maybe not. To prove that it doesn't treat cats specially, you should be able to restate the laws in a way that doesn't mention cats, and the specific claim about cats should be derivable from that. If you can't do that, that means that your laws are treating cats specially.

If you can't restate the minimalist interpretation in a way that doesn't mention measurement (or something equivalent) then to me, that's an indication that it treats measurements differently.

You keep saying that all theories of physics treat measurement special in the same way, but that's absolutely false. Newtonian physics describes objects and their motion and the forces acting on them. Anything you want to say about measurement follows from Newton's laws plus the assumption that the measurement device is a particular physical system obeying Newton's laws.

 

[stevendaryl]

I think, we go in circles here. It is very clear that physical theories are about describing measurements

No, they are not. Newton's laws are about objects and forces and motion.

 

[stevendaryl]

The measurement devices used are just made of ordinary matter and are thus described by standard quantum physics as any other lump of matter. That's all I'm saying

Yes, I agree that measurements can be described by quantum mechanics. But that is not sufficient for your claim that there is nothing special about measurements. Measurements in the minimal interpretation also serve as picking out a basis, which is necessary for the interpretation of amplitudes as probabilities. A non-measurement interaction does not pick out a basis.

 

[stevendaryl]

The basis chosen is dicated by the measured observable in the usual way (eigenstates of the corresponding self-adjoint operator representing this observable).

What makes something the "measured observable"?

 

[vanhees71]

Because if you are able to do that, that would demonstrate that the minimalist interpretation does not treat measurement different from other interactions. Nothing short of that would suffice.

I'm too stupid to understand your demand. Sorry for that. Again: Physics is about observations of nature in a quantitative way, i.e., about measurements. All I'm saying is that measurement devices and the interactions of measured objects with them are following the universal natural laws discovered by physics, and these rules are quantum. It doesn't make sense to talk about physics at all if you don't talk about observations and measurements, because that's the topic of physics.

Suppose I have a law of physics that states that cats always land on their feet. Does that treat cats differently than other objects? Maybe, maybe not. To prove that it doesn't treat cats specially, you should be able to restate the laws in a way that doesn't mention cats, and the specific claim about cats should be derivable from that. If you can't do that, that means that your laws are treating cats specially.

That's a statement about properties of cats, and of course you can observe it and see, whether it's right or not. That cats very often land on their feet is even an interesting biomechanical issue and well investigated by physicists. I'm only totally unaware what this has to do with the interpretational issues of quantum theory.

If you can't restate the minimalist interpretation in a way that doesn't mention measurement (or something equivalent) then to me, that's an indication that it treats measurements differently.

Why should it treat measurements differently? Measurements are defined by a measurement apparatus, and the very construction of all measurement apparati I know use the known universal laws of physics. There is not difference whatsoever concerning the applicability of the physical laws to construct a measurement apparatus than any other technical gadget like a car or a smartphone (although particularly the latter also contains a lot of measurement apparati you can even use to do interesting measurements in physics classes).

You keep saying that all theories of physics treat measurement special in the same way, but that's absolutely false. Newtonian physics describes objects and their motion and the forces acting on them. Anything you want to say about measurement follows from Newton's laws plus the assumption that the measurement device is a particular physical system obeying Newton's laws.

Sure, and Newtonian physics is as well about quantitative observations and thus measurements in nature. An in principle you are right, Newtonian mechanics also is in principle the sufficient basis to construct all the measurement devices you need to measure the quantities described by Newtonian physics (i.e., times, lengths, and masses; everything else is derived). Of course, the same physical laws are needed and the corresponding theory to define what's measurable (i.e., what are the observables) and at the same time are used to test this very theory. In this sense all experimental tests of physical theories are in fact consistency tests.

Nowadays you need a lot more than just Newtonian mechanics to construct measurement apparati; at least some Faraday and Maxwell electrodynamics is usually applied. Many high-precision measurements use in fact quantum theory. The entire SI units will be redefined soon, making use of the accuracy that can only be achieved by using the properties of quantum theory. E.g., to define the second (which will stay the same as before) you use the stability of atomic transitions (or maybe in the future nuclear transitions, which are even more stable and accurate), only describable by quantum theory. The representation of the ampere will hinge on quantum effects providing accurate quantities described by fundamental constants (among the THE quantum one par excellence, $\hbar$) like the quantization of magnetic moments, Josephson junctions, etc. etc. Classical physics is way to inaccurate to be used to define the base units of the SI for use in the 21st century!

 

[vanhees71]

What makes something the "measured observable"?

That's simple: By measuring it. I think we just are unable to explain to each other what the issue is. Maybe it's better to leave it at that :-(.

 

[stevendaryl]

I'm too stupid to understand your demand.

I don't think that's true. I think that you are unable to answer because you are holding onto two incompatible beliefs.

Sorry for that. Again: Physics is about observations of nature in a quantitative way, i.e., about measurements.

I'm not asking a philosophical question. You have a tendency to turn everything into philosophy, and then state how much you dislike philosophy.

I'm asking a technical question: Is it possible to formulate the minimal interpretation of quantum mechanics in a way that does not mention measurement?

The answer seems to be no. That's very different from the case with every other theory of physics.

Newton's laws are not formulated in terms of measurements. They make predictions about the results of measurements, which is all that you want for a theory to have empirical content.

That's a statement about properties of cats, and of course you can observe it and see, whether it's right or not. That cats very often land on their feet is even an interesting biomechanical issue and well investigated by physicists. I'm only totally unaware what this has to do with the interpretational issues of quantum theory.

It shows that you are claiming two contradictory things. If my axioms mention cats, then either the axioms can be reformulated so that cats are not specifically mentioned, or else it's false to claim that they don't treat cats specially. If your axioms mention measurements, then either the axioms can be reformulated so that measurements are not explicitly mentioned, or it's false to claim that they don't treat measurements specially.

 

[Fra]

there's (a) no difference in the physical laws between situations where a measurement apparatus is used and where this is not the case and (b) that there's no difference between the physical laws concerning many-body systems making up measurement devices and any other quantum system, large or small.

But WHERE are the physical laws manifested without a classical context? In the mathematical realm?

Of course, to make a measurement we need a macroscopic device to be able to make a measurement. I've never claimed the contrary.

Yes, but I get the impression that you might think this is not a major point, but a practicality?

Without the classical realm, we would not only have problems to make a reliable measurement, we would not have been able to reliable infer the laws of particle physics in the first place from large amounts of measurements! Without this, we could not compute the expectation values because the algorithm is unknown.

I may be taking this a step further here, but i think that the whole notion of physical law becomes fluid once we remove the classical observer. And thus fluidity may be necessary to face, but there is not fluidity in current theory, thanks to relating things to a classical measurement device. Here i think Bohr is very minimalist. He does not assume anything. He just notes that we need the classical context, to construct the questions that define the P-distributions.

The only thing I'm saying is that the classical behavior of macroscopic observables does not contradict the fundamental laws of quantum theory but are well explained by standard (quantum!) statistical physics.

A catch is the the laws of standard physics are inferred in the classical realm. You can not first abduce statistical laws, then remove the basis for the statistical processing, and claim that you still have a valid inference. Its a fallacy.

It is one thing to in principle explain a macroscopic piece of metal from QM as a manybody problem, because from the point of view of the human Earth based laboratory both are "small". Both are relative to our lab, "small subsystems". But if we make cosmological observations, or scale the classical laboratory down to grain level, this logic breaks.

If we stay away from such extremes, and study only small subsystems - from the point of view of a classical boundary, then current physics works fine. I mainly care about this as i want to develop this. But to develp this its good to first understand the premises of current framework.

/Fredrik

 

[stevendaryl]

That's simple: By measuring it.

I'm asking: What does it mean to measure something? Informally, I measured some property if I performed an action so that afterward, I know its value. That way of phrasing it sounds very solipsistic. Must there be a person around in order for quantum probabilities to be meaningful?

An alternative is to say that system A measures a property of system B if through interacting, the state of system A becomes correlated with that of system B and the alternative values of the property are macroscopically distinguishable. But that way of understanding it makes a macroscopic/microscopic distinction, which you claim not to be making.

 

[vanhees71]

I'm asking a technical question: Is it possible to formulate the minimal interpretation of quantum mechanics in a way that does not mention measurement?

It is not possible to do physics without measurements, so any physical theory is about measurements. Your question doesn't make sense to begin with, and that's not philosophy but the simple definition of what physics is about.

The answer seems to be no. That's very different from the case with every other theory of physics.

It's not different as with any other thery of physics, because physics is about measurements. Without measurements there's no physics.

Newton's laws are not formulated in terms of measurements. They make predictions about the results of measurements, which is all that you want for a theory to have empirical content.

Of course are Newton's laws about measurements, because all of physics is about measurements. You start with the postulates about space and time, which implies that you talk about measurable quantities like the period of a pendulum or a planet orbiting the Sun and about distances and angles of bodies in space. Without at least these kinematical observables you cannot even start to state the postulates!

It shows that you are claiming two contradictory things. If my axioms mention cats, then either the axioms can be reformulated so that cats are not specifically mentioned, or else it's false to claim that they don't treat cats specially. If your axioms mention measurements, then either the axioms can be reformulated so that measurements are not explicitly mentioned, or it's false to claim that they don't treat measurements specially.

What you stated is a prediction about the behavior of cats. You use (implicitly) the definition of "cat" and it's a statement about the mechanics of cats, which can be checked by observation. Of course, if you make a statement about something it's a statement about this entity. However, I don't see at all what this has to do with the foundations of quantum theory and particularly what this has to do with the existence of a classical-quantum cut (which you seem to insist on as vehemently as I deny any empirical foundation for its existence) or, for me even on the edge of esoterics, that a piece of matter cannot be described by the universal physical laws of nature only because it's used as a measurement device. Are you really claiming that a piece of wood changes to obey the known physical laws, only because I put some marks of it to use it as a yardstick? For me that would be utter nonsense.

 

[Lord Jestocost]

Why should it treat measurements differently?

As Maximilian Schlosshauer remarks in “ELEGANCE AND ENIGMA, The Quantum Interviews”:

“Measurement-as-interaction, by contrast [to measurement-as-axiom], leads to an entangled quantum state for the composite system-plus-apparatus. The system has been sucked into a vortex of entanglement and no longer has its own quantum state. On top of that, the entangled state fails to indicate any particular measurement outcome.”

 

[stevendaryl]

It is not possible to do physics without measurements,

That wasn't the question. The question was whether it is possible formulate the minimal interpretation without mentioning measurements. Can you answer that question?

 

[stevendaryl]

Of course are Newton's laws about measurements, because all of physics is about measurements.

No, Newton's laws are not about measurements. They are about particles and forces and motion. You can deduce facts about measurements from those laws (under the assumption that your measurement devices themselves are physical systems made up of particles and affected by forces.)

 

[PeterDonis]

It is not possible to do physics without measurements, so any physical theory is about measurements.

No, any physical theory has to be able to model measurements. But the mathematical machinery of QM, the thing that makes predictions, does much more than that: it tells you, "when a measurement occurs, use the Born rule to calculate the probabilities of the possible outcomes". No other physical theory has a rule like that embedded in its mathematical machinery. Newton's Laws, to use the example you have been using, don't tell you "when a measurement occurs, use F = ma", for example. They just say "F = ma".