Against "interpretation" - Comments

[DarMM]

How can there be a contradiction, since that's how our theories are applied by experimentelists and also the other way around, how models and maybe finally theories are developed from analysis of experiments.

I think Peres's textbook is the most (if not the only) sensible textbook on interpretational issues I know.

See Chapter 9 of Peres's book, particularly the section called "The consistency problem", p. 376 if I remember right. I suspect you already have what Peres says built into your view so there's no issue with what you call the "Minimal view", but technically it's an issue with textbook QM.

If you like his style I'd look at some of his papers from the 1980s like "The classical paradoxes of quantum theory".

 

[vanhees71]

There's very little research on the pedagogy of QM particularly for higher areas like QFT and interpretations. I'd also have opinions on the best way to teach QFT. There's no studies about it, but I do think it for reasons that come from lecturing and talking to other lecturers. I've had a bit about interpretations at the end of my QM lectures. No there is no study, but there aren't really many such studies for any higher area of physics. I have one anecdotal experience as a lecturer, you have another as a moderator here. Yes we can't say more than that since there are no studies, but I don't think that makes my view ridiculous.

I think, pedagogical studies can have there merits but I'm highly suspicious, though I like one type of empirical studies: You teach some topic in different ways to different groups and then look at the outcome of problem-solving sets/exams, including also asking conceptual questions. It's amazing what sometimes comes out from such studies. E.g., there's a recent textbook on school students' (of various grades) as well as university students' understanding of standard physics concepts. In quantum theory even at the university level there are amazing outcomes of such studies: E.g., in one study they found out that after a QM lecture students (5-11 semester at the university) just had the "particle idea of light/photons" but have not associated any wave aspects of light anymore. This is amazing given the fact that in teaching QM usually one starts with empasizing the wave-particle-duality aspect pretty much, but it seems that the "photons as miniature bullets" idea sticks, and for me this confirms that one should from the very beginning emphasize that this old picture a la Einstein is simply wrong and that for a full understanding of photons they have to wait for the QFT lecture.

I think there are some merits in discussing foundational aspects for the reason that some students are interested in them and one can use them to teach the applications of the formalism. E.g., I find an experiment like the quantum eraser with entangled photons or the entanglement-swapping experiment with two entangled photon pairs atmost fascinating from purely physics scientific point of view, and you can discuss them with quite uncomplicated basic parts of the formalism used in QED (in its quantum-optics flavor of course). I think this can be easily implemented for 1-2 90minute lectures within the QFT 1 lecture.

 

[Demystifier]

In my personal life I have strong philosophical positions to which I am quite firmly attached, e.g. I am a devoutly religious person. I don't argue about those positions at all.

I also have strong opinions about some religious questions (for instance I think I can prove that omnipotent God is a logical contradiction) and like you, I also don't argue those positions here. But ...

... the P in PF is physics, not philosophy.

... unlike you, I don't think that something is not philosophy just because it is physics. I think many topics, including quantum foundations, are both. In fact, the very first post in this thread is precisely an argument that it doesn't make sense to make a clear cut between interpretations (that is, philosophy) and theories (that is, physics).

 

[Demystifier]

I disagree about the pedagogical value of interpretations arguments. In my personal (anectdotal) case it is definitely an obstacle to learning, not a benefit.

My personal anecdotal experience is precisely the opposite. With some intuitive interpretation in my mind, I find it easier to deal with some practical questions too. For me, interpretations are not definite but mutually exclusive truths, but practical mutually complementary tools. I find it hard to think about practical details if I don't have a big intuitive picture of all this in my mind. My understanding of Bohmian mechanics or many worlds helps me to be a better "Copenhagenian" when it's appropriate. But I guess that's very individual.

 

[atyy]

I disagree about the pedagogical value of interpretations arguments.

I hope you at least agree that quantum mechanics in the minimal interpretation requires - different people say this differently - a "classical-quantum" cut or grant the observer or measurement apparatus a special status. After all, this is right at the start of Landau & Lifshitz, one of the most famous textbooks.

 

[dextercioby]

Sure. I completely agree with you here. I ask myself many non-scientific questions. I just (a) recognize that they are non-scientific (b) don’t ask them here (c) don’t bother to argue about them at all (d) keep an open mind and (e) switch positions as often as I find convenient.

For me, the biggest problem here on PF is (c). You all argue incessantly and interminably about this stuff. It is all opinion! It may be the opinion that you hold in your heart, but it is just an opinion. Why do you all argue so much about opinions?

Interpretations (beyond the minimal interpretation) have little scientific value, and only this “in your heart” emotional attachment. So please (all of you) tone down the arguments substantially.

Hi Dale, I understand your position, but not only the community here thinks the topic "interpretations of Quantum Physics" is actual real physics research, simply because there is no agreement in the literature on a common way to interpret/translate what the theory is saying.
For me, these discussions are worth reading for the sake of (educated) argumentation and I press the "like" button -like on other social media- each time I agree with what is written (content and even more so flow of ideas).

You have the option to not moderate these borderline physics/philosophy discussions which you acknowledge bring little value to you. We have Peter Donis who has lately migrated from moderating the relativity section into the QP one.

What you just did, metaphorically speaking, is just tell people "I hate the music you listen to", when you have the option to not simply listen to it...

From my point of view, I consider the whole "string/LQG theory" simply a useful exercise in mathematics (perhaps I am not aware of their true predictive power wrt experiments), but I do not go them to tell them "Hey, your work on string theory is partly/mostly/totally useless. By what argument do you expect me to believe its Quantum Gravity explanation?"

So you think some Netflix series is junk, wasted money. Then simply turn to HBO or Amazon Prime...

 

[Dale]

I find it easier to deal with some practical questions too. For me, interpretations are not definite but mutually exclusive truths, but practical mutually complementary tools. I find it hard to think about practical details if I don't have a big intuitive picture of all this in my mind.

This approach I like. If an interpretations discussion took this approach then I could believe it would be pedagogically beneficial. Something along the lines of introducing a particular type of problem, introducing a specific interpretation that is well suited to that type of problem, and allowing that interpretation to guide the solution of the problem.

Do such threads exist here?

 

[Dale]

What you just did, metaphorically speaking, is just tell people "I hate the music you listen to", when you have the option to not simply listen to it...

Here is the personal frustration: I would like to know more QM than I do, but I find the interpretations discussions here to be a severe obstacle to doing so. I hate the music you listen to partly because you are playing it so loud and so continuously that it drowns out the music that I want to listen to.

I am hoping that the new Interpretations forum will alleviate this issue in the future. But in the past this has been a problem for me personally.

So you think some Netflix series is junk, wasted money. Then simply turn to HBO or Amazon Prime.

I would but the junk Netflix series was being continuously broadcast on top of Amazon Prime. Unless they fixed it recently.

You have the option to not moderate these borderline physics/philosophy discussions

I don’t moderate them, I am not qualified. I am not here as a moderator, just a participant, and only as a participant because the title was “Against Interpretation” so I thought it would be a thread specifically for voicing my kind of opinion.

 

[Demystifier]

This approach I like. If an interpretations discussion took this approach then I could believe it would be pedagogically beneficial. Something along the lines of introducing a particular type of problem, introducing a specific interpretation that is well suited to that type of problem, and allowing that interpretation to guide the solution of the problem.

Do such threads exist here?

I'm not sure about whole threads, but posts of that form exist. See e.g. my
https://www.physicsforums.com/threads/wigners-friend-and-incompatibility.976587/page-3#post-6226655https://www.physicsforums.com/threads/wigners-friend-and-incompatibility.976587/page-2#post-6225640

 

[ftr]

I am hoping that the new Interpretations forum will alleviate this issue in the future. But in the past this has been a problem for me personally.

Physics has several main problems that many physicists (almost all bloggers) debate. How else would people solve these problems. People here with a spectrum of knowledge strength give their "opinion" , that is part of the process (whether PF or any other place).

 

[Minnesota Joe]

My personal anecdotal experience is precisely the opposite. With some intuitive interpretation in my mind, I find it easier to deal with some practical questions too. For me, interpretations are not definite but mutually exclusive truths, but practical mutually complementary tools. I find it hard to think about practical details if I don't have a big intuitive picture of all this in my mind. My understanding of Bohmian mechanics or many worlds helps me to be a better "Copenhagenian" when it's appropriate. But I guess that's very individual.

Anecdotal, yes, but it describes very well the problems I had with my first quantum mechanics class and I've heard similar things from peers. A balanced discussion of the various ideas about what it could mean would have been a very useful guide to thinking. A close second would have been "This is the recipe and we don't know why" as opposed to strong metaphysical positions on indeterminacy, non-realism, etc. I hope that has changed at least.

Anyway, what you wrote really resonates with me at least.

 

[vanhees71]

I hope you at least agree that quantum mechanics in the minimal interpretation requires - different people say this differently - a "classical-quantum" cut or grant the observer or measurement apparatus a special status. After all, this is right at the start of Landau & Lifshitz, one of the most famous textbooks.

Well, I disagree, even though LL is among the best textbooks I know. The classical description of macroscopic situations, including measurement devices, is an effective description which can be understood from quantum theory by basic statistical arguments (law of large numbers etc.), and then I can use the effective description of the macroscopic measurement device to understand why I measure the observable of the "quantum object" I want to measure. There is no cut but just the applicability of a "classical" description for many-body systems and some appropriate coarse grained macroscopic observables. FAPP of course you can argue with a cut. The observable outcome is the same.

Whether or not there is really a cut on the fundamental level, i.e., if there is some limit of the validity of QT at which the classical behavior cannot be understood from QT, or (in another twist of this issue) if at one point a macroscopic system cannot show specific "quantum behavior" anymore (e.g., intereference effects of objects consisting of many particles). Just recently the boarder of this "classical-only behavior" has been pushed to huge molecules with de Broglie wavelengths in the fm range. So far there's no single hint for such a "fundamental quantum-classical cut". Of course, there might be one somewhere, but at least today there's no hint for such a thing.

 

[A. Neumaier]

I have never seen a clear presentation of the measurement problem in the so called "minimal" view. The measurement problem is a logical incoherence in the theory, not just metaphysics. I don't think you can get out of it just by being positivist/empiricist. Or could somebody give me a clear explanation of the measurement problem in the minimal view?

Englert's article and the books by Ballentine's and by Peres have something. Englert and Ballentine sweep the problem under the carpet of very soft arguments. Peres admits towards the end of his books (see your post #303) that there is a problem when one applies quantum mechanics to a full model of the measurement problem. I concur with Peres, who (unlike Ballentine, Englert, and @vanhees71) is very precise in his arguments.

 

[A. Neumaier]

How can there be a contradiction, since that's how our theories are applied by experimentalists and also the other way around, how models and maybe finally theories are developed from analysis of experiments.

Successful use does not exclude logical deficiencies. It is similar to the foundations of set theory. It had been used very successfully for over 20 years before Russell discovered a contradiction in naive set theory, and it took another 20 years to fix it through axiomatic set theory. Today most people still don't care since they can use naive set theory without problems.

In quantum physics, the contradiction is not in the relation to experiment but in the theory, once Born's rule assumed together with the claim that experimental arrangements and hence measurements can be fully accounted for by quantum physics. Those like you who never model the latter will not run into the contradiction, so everything seems fine to them.

 

[vanhees71]

I still don't understand your argument. Of course, the issues with set theory were mathematical issues and had to be resolved to have a solid foundation of set theory and given its importance as a basic tool in all branches of mathematics also of mathematics as a whole.

Quantum mechanics, at least the non-relativistic flavor, has no mathematical problems as far as I can see. This of course includes Born's rule, and given this set of postulates/axioms you have a special kind of probability theory, if you just consider QM as a mathematical construct.

Now when it comes to physics all you need in addition is a relation between what's observed in nature and the formalism, and with the minimal statistical interpretation there are no problems either. If there were problems, QM wouldn't be as successful as it is in describing the observations. If there is a discrepancy, we'd learn something fundamentally new. That's how (theoretical) physics works.

Of course it's desirable to have mathematical sound and solid formulations of physical theories, but it's a different issue to consider a mathematical theory, which does not necessarily need to describe nature but just has to be mathematically sound and solid. E.g., Newtonian mechanics is a mathematically well-defined theory providing rules to derive differential equations and some general and beautiful tools for their solution. Particularly the entire field of Lie-group and -algebra theory has been triggered by it. Nevertheless it's not sufficient to describe all observations and thus had to be modified (including the necessity of new descriptions of space and time in relativity theory and QM).

Another type of trouble is classical electrodynamics. As long as you deal with continuum mechanical models of charged matter the issue from the point of view of mathematics is quite the same as with Newtonian mechanics. Here we have a well-defined classical relativistic field theory providing PDEs and some nice tools to solve them. Nevertheless we know it's incomplete as soon as it comes to models for matter in terms of point particles. There's no satisfactory closed dynamics for fields and point particles. The best one could come up with for more than 100 years is the Landau-Lifshitz approximation. Of course it's known that classical point particles are a lousy model of matter on a microscopic level and one has to use QT, and in this case the best one could come up with is relativistic QFT, which has the known problems as a mathematical theory, which however can be overcome in the typical pragmatic manner physicists ignore mathematical problems by dealing with hand-waving arguments. For a mathematician there's of course still a lot to be desired, and maybe one day there's a more comprehensive mathematically sound and solid microscopic theory that describes all observations as well as relativistic QFT does as an "effective theory" today.

All these examples show that there's a strong overlap between math and theoretical physics, but the goals are partially different. As a physical theory QM is amazingly successful, and there are no foundational problems. You just have to accept probabilistic arguments as a fundamental feature of nature. Quibbles with this view are metaphysical and have nothing to do with either math or the natural sciences.

 

[atyy]

I have never seen a clear presentation of the measurement problem in the so called "minimal" view. The measurement problem is a logical incoherence in the theory, not just metaphysics. I don't think you can get out of it just by being positivist/empiricist. Or could somebody give me a clear explanation of the measurement problem in the minimal view?

I think the measurement problem is most clearly stated in the minimal view, since it begins with a classical/quantum cut or gives the observer or measurement apparatus a special status. You can get out of it by just saying it's fine FAPP, and physics is just FAPP. But many also choose to say there is clearly a measurement problem. Either way, it is meta-physics, which is fine, at least as fine as meta-mathematics.

If by the minimal view we mean what Peres has in his book that's fine. If we mean textbook QM/what most people mean by "shut up and calculate" then no that does have a contradiction.

I think the minimal view is what Peres has (maybe Peres has very minor errors, but I think his book is beautifully written), and that should be the same as shut-up-and-calculate. Any minimal view that is not like what Peres has is just not properly informed by tradition.

 

[Minnesota Joe]

Against "interpretation"
Continue reading the Original Blog Post.

To what extent do you think stressing "interpretation" instead of "theory" is a holdover from the logical positivism that gripped some of the founders of quantum mechanics (and their students, and their students, and...)?

Logical positivism is a very attractive view for people who do not want to worry about what they cannot observe. It is ultimately a theory about meaning, about the content of a theory. According to the positivists, a theory says no more than its observable consequences.

Logical positivism has been killed many times over by philosophers. But no matter how many stakes are driven through its heart, it arises unbidden in the minds of scientists. For if the content of a theory goes beyond what you can observe, then you can never, in principle, be sure that any theory is right. And that means there can be interminable arguments about which theory is right that cannot be settled by observation.

Tim Maudlin

 

[Lord Jestocost]

As remarked by Paul K. Feyerabend in “Physics and Philosophy: Volume 4: Philosophical Papers”:

"The issue concerning the foundations of the quantum theory can therefore be solved only by the construction of a new theory as well as by the demonstration that this new theory is experimentally at least as valuable as the theory that is being used at the present time; it cannot be solved by alternative interpretations of the present theory.⁴⁹"

And in footnote 49 he adds:

“If I am correct in this, then all those philosophers who try to solve the quantum riddle by trying to provide an alternative interpretation of the current theory which leaves all laws of this theory unchanged are wasting their time. Those who are not satisfied with the Copenhagen point of view must realize that only a new theory will be capable of satisfying their demands.”

[Italics in original, LJ]

 

[A. Neumaier]

I still don't understand your argument. Of course, the issues with set theory were mathematical issues and had to be resolved to have a solid foundation of set theory and given its importance as a basic tool in all branches of mathematics also of mathematics as a whole.

Quantum mechanics, at least the non-relativistic flavor, has no mathematical problems as far as I can see. This of course includes Born's rule

It has no problems if one considers (as you do) Born's rule to be just an axiom without asking further what a measurement is on the level of theory. But the measurement problem appears once you want to give the notion of measurement a mathematical meaning (and only then Quantum mechanics including Born's rule is a mathematical theory).

As a physical theory QM is amazingly successful, and there are no foundational problems.

There are no foundational problems at the level where QM is actually applied - namely in the microscopic domain. There are foundational problems if one wants to extend QM to the macroscopic regime and still stay faithful to its textbook foundations.

I think the minimal view is what Peres has (maybe Peres has very minor errors, but I think his book is beautifully written), and that should be the same as shut-up-and-calculate. Any minimal view that is not like what Peres has is just not properly informed by tradition.

You gave this statement your Like - so you can probably explain why Peres wrote Chapter 12 of his book, titled ''The measurement problem'' if the latter does not exist. Peres deliberately and faithfully follows the minimal interpretation dogma that you also advocate until he reveals that he has to give it up in this chapter:

There are questions, such as “what is the numerical value of the pointer position,” which make sense in the classical language, but are meaningless in the quantum language.

This would cause no conceptual difficulty with quantum theory if the Moon, the planets, the interstellar atoms, etc., had a well defined state ρ. However, I have insisted throughout this book that ρ is not a property of an individual system, but represents the procedure for preparing an ensemble of such systems. How shall we describe situations that have no preparer?

You must have noted the difference between the present pragmatic approach and the dogmas held in the early chapters of this book. It was then asserted that any operator which can be written by a theorist can also be measured in the laboratory. This fiction was needed in order to establish a formal framework for quantum theory.

The dogmatic part is mathematical physics and has unsolved consistency problems in which people like me or DarMM are interested. The pragmatic way out is FAPP only but satisfies those like you who don't care about mathematical consistency of theoretical physics as long as the predictions are successful.

 

[vanhees71]

Though I like Peres's book, I don't have to agree with everything he writes. If he writes that there's a measurement problem I disagree with him, but as Peres seems to me also to be a "minimal interpreter" I don't think that he consideres the measurement problem a physics problem either.

I disagree with him in the sense that I don't think one has to explain that there are clearly defined pointer readings in an experiment, because that's one of the assumptions you make as a theoretical physicist because that's established by experience. If there were no clear pointer readings the entire probabilistic interpretation of the formalism wouldn't be sufficient too. Maybe QT would have been forgotten as some curiosity and some other theory would have been discovered, but now there are in fact clear pointer readings. You don't need to derive this from theory but you just take it as an empircal fact and use it as such to formulate the theory. Of course, as any empirical fact, it's subject to further empirical tests, and if it's found to be invalid under some circumstances, it may well be that QT would have to be updated by something more comprehensive.

 

[A. Neumaier]

I don't think one has to explain that there are clearly defined pointer readings in an experiment, because that's one of the assumptions you make as a theoretical physicist because that's established by experience.

Yes. But this is where the mathematics of quantum physics has a consistency problem when it tries to model the whole experiment in terms of a microscopic description. You say you don't care because its established by experience. Others like me care because its not established by the theory.

Thus whether there is a measurement problem depends on which standards one requires for the theory. A mathematical physicist has higher standards than a theoretical physicist.

I still don't understand your argument.

Clearly there is no way we can agree. But perhaps you can at least understand why others, including me, consider the measurement problem as something real.

 

[Demystifier]

To what extent do you think stressing "interpretation" instead of "theory" is a holdover from the logical positivism that gripped some of the founders of quantum mechanics (and their students, and their students, and...)?

To a great extent. Let me also add that logical positivism is self-contradictory. That's because the main claims of logical positivism (e.g. that "only the measurable has a meaning") are themselves philosophical claims not obtained from measurements. What logical positivists really say is: "All philosophy is meaningless, except our philosophy."

 

[Minnesota Joe]

To a great extent. Let me also add that logical positivism is self-contradictory. That's because the main claims of logical positivism (e.g. that "only the measurable has a meaning") are themselves philosophical claims not obtained from measurements. What logical positivists really say is: "All philosophy is meaningless, except our philosophy."

Yes it is infamously self-defeating and you would have to special plead to hold it. As Maudlin points out, you still encounter the attitude in the sciences however.

 

[Reductionist89]

Which interpretation makes the least amount of assumptions in your opinion - Copenhagen or Many Worlds? Both take the data at face value but in radically different ways.

 

[vanhees71]

The interpretation making the least amount of assumptions is the minimal statistical interpretation, which I'd consider one flavor of the Copenhagen interpretation. I think it comes pretty close to Bohr's version, though I think Bohr is in favor of a quantum-classical cut on a fundamental basis though there's not the slightest hint for it ever observed.

I can't comment on Many Worlds since I've never understood how it relates the formalism to the observable facts.

 

[ftr]

I wonder which is the more serious "problem"(in quotation to avoid ..) the fact that in QM the particles have undefined properties(before measurement) or that measurement is classical in the end.

 

[vanhees71]

There's no problem in either. The fact that in QM a particle has precisely the probabilistic properties due to its preparation, described by the corresponding quantum state $\hat{\rho}$ is, as far as all experiments show, precisely what's observed when measuring an observable of this particle.

That measurement devices, as macroscopic objects, behave classical is also no surprise but just due to sufficient course graining of the macroscopic observables. A "pointer position" of a measurement device is such a coarse-grained macroscopic observable.

 

[ftr]

The fact that in QM a particle has precisely the probabilistic properties due to its preparation, described by the corresponding quantum state ^ρρ^\hat{\rho} is

Is it "due" to or inherently, because that is how the electron in hydrogen atom is modeled?

 

[ftr]

That measurement devices, as macroscopic objects, behave classical is also no surprise but just due to sufficient course graining of the macroscopic observables. A "pointer position" of a measurement device is such a coarse-grained macroscopic observable.

In the age of gravity wave detection experiments isn't possible to do the same for quantum. I mean couldn't they use a thin sheet of single carbon or whatever atom and have some miniature source using femto cameras and such. Money should be no problem, right?