I Why don't we bury Schrodinger's Cat?

[PeterDonis]

how can the first description be right if decoherence doesn't produce a definite outcome but 'merely' a classical probability distribution without interference between the two outcomes?

The answer to this is interpretation dependent. The point is that on any interpretation, there is no interference between the possible outcomes, so opening the box doesn't change anything about what's inside the box; all it does is enable interaction between what's inside and what's outside so the information about what's in the box can spread more widely. That is what the first description describes. But different QM interpretations will tell different stories about "what is actually happening" behind the scenes in order to make that description correct.

 

[PeterDonis]

How do you know "there was no change" if somehow the classical probability distribution "collapses" to one outcome?

Even on interpretations that make this claim, the collapse occurs when the cat decoheres inside the box, not when the box is opened. Opening the box still doesn't change anything about the cat.

 

[Lord Jestocost]

If I say: "The cat is either alive or dead, and I don't know which, and I'll find when I open the box. No change to the cat happens when I open the box. The only change is that I (myself) now know in what condition the cat is, and had been for some time."

That's fine so far, if you would leave out the half-sentence "and had been for some time".

 

[gmax137]

My understanding of QM is very dated, based on a handful of undergrad courses back in the 1970s. I don't recall the terms "coherence" and "decoherence," I'm not really sure what they mean. Is there a modern intro textbook that would help me catch up? I'd like to understand threads like this one better.

 

[PeterDonis]

My understanding of QM is very dated, based on a handful of undergrad courses back in the 1970s. I don't recall the terms "coherence" and "decoherence," I'm not really sure what they mean. Is there a modern intro textbook that would help me catch up? I'd like to understand threads like this one better.

I don't know if any textbooks have been updated to cover decoherence theory, but there is a good 2019 review article by Schlosshauer referenced here (as well as other PF threads):

https://www.physicsforums.com/threads/new-review-on-decoherence-by-schlosshauer.979635/

 

[haushofer]

Even on interpretations that make this claim, the collapse occurs when the cat decoheres inside the box, not when the box is opened. Opening the box still doesn't change anything about the cat.

Yes, that was my confusion due to old fashioned Copenhagen. Thanks, I get it!

 

[gmax137]

I don't know if any textbooks have been updated to cover decoherence theory, but there is a good 2019 review article by Schlosshauer referenced here (as well as other PF threads):

https://www.physicsforums.com/threads/new-review-on-decoherence-by-schlosshauer.979635/

Thanks @PeterDonis !

 

[vanhees71]

Another good book on decoherence is

E. Joos, H.D. Zeh, C. Kiefer, D. Giulini , J. Kupsch I.-0. Stamatescu
Decoherence and the Appearance of a Classical World in Quantum Theory, 2nd edition, Springer 2003

 

[bhobba]

https://www.physicsforums.com/threads/new-review-on-decoherence-by-schlosshauer.979635/

Great book. I highly recommend it. It has a prominent place in my library, along with Ballentine, which I also highly recommend as a modern treatment of QM.

Thanks
Bill

 

[bhobba]

Yes, that was my confusion due to old fashioned Copenhagen. Thanks, I get it!

The interesting thing about Copenhagen, and what annoyed both Einstein (when he was alive) and me, was the unnecessary claim of Bohr's version, where it was said to be a complete theory. For example, at the 1927 Solvay conference, Niels Bohr and Heisenberg said the Copenhagen view was a "complete" picture of quantum physics, even though they could not, or would not, visualise or otherwise explain precisely what is going on in the microscopic world. I have zero problems with not knowing precisely what is happening in the microscopic realm, but saying it was complete - the formalism does not imply that. What some call the 'modern' version of Copenhagen, Decoherent Histories makes no such claim. Nor do I think all Copenhagen adherents subscribed to that claim.

Thanks
Bill

 

[Ken G]

Personally I think the main issue behind Schroedinger's cat goes well beyond quantum mechanics, it is the deeper question of what is a physics theory for, i.e., what is it ultimately trying to describe. So the main problem is, we tend to imagine that the purpose of a physics theory is to describe what is actually happening in a place we think of as "objective reality." But such a place is never part of any physics theory, since theories just manipulate abstract notions to create predictions that can be tested. They are simply never used for any other purpose in the scientific method, which only requires objective tests, and the objective tests then become the reality (a thing that never had to be in the theories at all). It is a mistake about what a physics theory is to imagine otherwise. We just got away with it many times, before quantum mechanics, but the problem was always there, the disconnect between what we demonstrate and what we choose to believe.

So what this means is, quantum mechanics is the place where we find we cannot get away with the naive and unnecessary notion that a physics theory was ever about objective reality. There is never any problem if the components "stay in their lanes": observations establish objective reality, theory is a prescription for making testable predictions that show up in objective reality when we test them. All we have to do is take greater pains to track what science actually does, and the "Schroedinger cat problem" never appears at all.

What this also means is that we are untrue to the scientific method when we hold that any interpretation of quantum mechanics is a statement about objective reality, because if the theory was never demonstrably that, then so much less is an interpretation of a theory. Instead, an interpretation of a theory is a kind of lens through which to look at a theory, a way to understand or make sense of the theory, not objective reality itself (as the latter is only what we observe, and the predictions we test thereby). This is not a philosophy, it is simply a factual statement about what science actually does, and demonstrably so throughout its history. So there never was, or should be, any possibility that two valid interpretations of any physics theory could ever contradict each other, or argue with each other in any way. And there is no Schroedinger cat problem, insofar as we never have any difficulty using our theories to correctly predict the probabilistic tendencies of our observations. And that's all they were ever tasked with doing, and all they ever showed success in doing, ever since the scientific process was first invented.

So should we "do away with" the Schroedinger cat? No, we should do away with the idea that the cat paradox is a problem about the nature of physical reality. It is instead a problem about the nature of physics theories, and for that demonstration, it retains supreme importance. We can use a deeper physics theory to explain when and why a simpler theory works, but we can never use a theory to understand why that same theory works, that's just not what theories do.

 

[gentzen]

So there never was, or should be, any possibility that two valid interpretations of any physics theory could ever contradict each other, or argue with each other in any way.

That is a misunderstanding of the relation between a mathematical theory and its interpretations. No interpretation can contradict the theory. But the theory itself stays silent on certain questions, and different interpretations can give different answers to such questions on whicht the theory itself stays silent. And giving a different answer to the same question means that they can contradict each other.

Apart from that, it is quite obvious that different interpretation of quantum mechanics do contradict each other, and that the proponents of different interpretations never were shy to fight with each other over such contradictions.

 

[vanhees71]

The interesting thing about Copenhagen, and what annoyed both Einstein (when he was alive) and me, was the unnecessary claim of Bohr's version, where it was said to be a complete theory. For example, at the 1927 Solvay conference, Niels Bohr and Heisenberg said the Copenhagen view was a "complete" picture of quantum physics, even though they could not, or would not, visualise or otherwise explain precisely what is going on in the microscopic world. I have zero problems with not knowing precisely what is happening in the microscopic realm, but saying it was complete - the formalism does not imply that. What some call the 'modern' version of Copenhagen, Decoherent Histories makes no such claim. Nor do I think all Copenhagen adherents subscribed to that claim.

Thanks
Bill

Well, it's only "incomplete", if you don't accept the possibility that Nature is indeed inherently random. There's no hint that this might not be the case. It's anyway pretty unimportant, whether a physical theory is "complete" or not. I think QT is incomplete a priori not because it's probabilistic since it may well be that Nature is "really" random in the sense of QT, but because we don't have a satisfactory quantum description of the gravitational interaction and/or of spacetime. Rather the specific form of the quantum dynamics is dictated in a large part by the symmetries of the underlying classical (!) spacetime model. In which sense this classical spacetime description is to be understood as an emergent phenomenon from a more fundamental (quantum?) theory seems to be likely to be part of the puzzle, how to describe the gravitational interaction within QT.

 

[vanhees71]

The problem with that in turn is that there's not the slightest hint at something that's missing in the description of matter by QT. If there were, maybe we'd be able to find another theory, which maybe is more satisfactory for our intuition, based on our experience with macroscopic objects around us, but I doubt it, given that classical physics has so many more problems to understand matter (starting from the obvious stability of it) than QT.

 

[Ken G]

That is a misunderstanding of the relation between a mathematical theory and its interpretations. No interpretation can contradict the theory. But the theory itself stays silent on certain questions, and different interpretations can give different answers to such questions on whicht the theory itself stays silent. And giving a different answer to the same question means that they can contradict each other.

Can you give an example of something falsifiable that any of the interpretations disagree on?

Apart from that, it is quite obvious that different interpretation of quantum mechanics do contradict each other, and that the proponents of different interpretations never were shy to fight with each other over such contradictions.

You speak of contradictions, but I see only disagreements in interpretation, no contradictions at all. Why do people think QT is any different from classical mechanics on this score? One interpretation of CM says there are forces and particles, another says there are extremum principles and symmetries. They make all the same predictions, in a different language. No contradictions there, and no one thinks there are. Do classical particles move due to forces, or don't they? It's pure philosophy, nothing falsifiable, but no one seems to mind that in CM, as well they should not.

 

[Ken G]

Well, it's only "incomplete", if you don't accept the possibility that Nature is indeed inherently random.

But what does it even mean to talk about nature being "inherently" anything? This is physics, nothing is "inherent." That's my whole point, the problem is a tendency to mix physics and philosophy, as if they were the same. But science is one thing, and philosophy is another. Both have their place, and are important, but they are clearly distinct in terms of objective falsifiability. How would one falsify that nature is "inherently" something, rather than just behaving in some way?

It's anyway pretty unimportant, whether a physical theory is "complete" or not.

I agree, but would go even farther. No physics theory is even trying to be "complete," unless we are kidding ourselves, because that has never been the purpose of physics, nor the process of physics. The sketchy history of "completeness" of physics theories should be clear enough on that point, and we must steadfastly avoid the error of every era of physicists thinking that they have seen the final truth in a process that demonstrably simply progresses.

I think QT is incomplete a priori not because it's probabilistic since it may well be that Nature is "really" random in the sense of QT, but because we don't have a satisfactory quantum description of the gravitational interaction and/or of spacetime.

It would never make sense to claim that nature is really random, or really determinate. Randomness and determinism are both ways of thinking about nature, which have their various advantages. Nature could never actually be one or the other, that would be a category error. We must not confuse nature with how we think about nature, as the latter is tied up in the motivations we have for thinking that way. (The former is probably not a coherent notion anyway, since all we have is how we think, and how would a way we think be called "inherent" or "actually random" anyway?)

Rather the specific form of the quantum dynamics is dictated in a large part by the symmetries of the underlying classical (!) spacetime model. In which sense this classical spacetime description is to be understood as an emergent phenomenon from a more fundamental (quantum?) theory seems to be likely to be part of the puzzle, how to describe the gravitational interaction within QT.

The puzzle is indeed the key, and these are deep insights into the theories and the puzzles they represent, without equating the philosophical puzzles with the physics they inspire. Don't get me wrong, philosophy is great for motivating our thinking, it's just something different from the physics. Hence "inherentness" is an attribute of a philosophy, that an individual might take value in, but it is not an attribute of a physics theory because it is merely waiting to be falsified, even when it takes a thousand years to do so.

 

[gentzen]

Can you give an example of something falsifiable that any of the interpretations disagree on?

Most mathematical theories can be interpreted both "literal" and "in their original context". For QT, MWI would be an attempt at a "literal" interpretation, and Copenhagen the practice of interpretation close to "the original context" of QT. My reaction to any "literal" interpretation is fear of the dangers of stupid mistakes. I have no illusions about my chances to convince proponents of "literal" interpretations of the existence of such dangers. Logical arguments, like for example those given in Decoherence is Dephasement, Not Disjointness, often have very little chance to reach them. So let me instead go with the existence of scalable quantum computers that are exponentially faster than classical computers for certain problems (like factoring) as a falsifiable predictions of MWI (because of David Deutsch). Notice that quantum computers are outside of "the original context" of QT. Additionally, for Copenhagen, quantum computers are related to limits like zero temperature or perfect isolation, and it remains unclear whether those limits stand in the way of scalability or not.

MWI and Copenhagen are the extremes, but there are also interpretations in between those extremes, like Bohmian mechanics or the minimal statistical interpretation. In fact, I believe Bohmian mechanics provides the clearest mathematical examples to see how it contradicts both MWI and Copenhagen. But let me first come back to my objection: "That is a misunderstanding of the relation between a mathematical theory and its interpretations." Euclidean geometry is (proto-)typical example of a mathematical theory. As long as you cannot accept that there really was an axiom missing in it, I have serious doubts that you will be able to appreciate the difference between a mathematical theory and its interpretations:

Did you ever heard of Pasch's axiom? It was THE axiom missing from Euclid's axioms. You may think, why is it missing, isn't it SO OBVIOUS that we don't even need to write down that axiom? Well, if you just look at the theory defined by Euclid's axioms, then that theory would also allow other models for which many constructions from Euclid would not work, and many theorems from Euclid would not be true. Of course, we all know that those models were not intended by Euclid, and that is precisely why we can say that Pasch's axiom was missing.

Where Euclidean geometry is quite untypical is that after the addition of THE missing axiom, it was indeed a complete theory from a purely mathematical point of view. Still, its applicability to the world around us and its range of validity remains open to interpretation. This again is quite typical for mathematical theories, as discussed again and again by Werner Heisenberg in his book Physics and Philosophy:

After this short survey we may come back to the more general question, what one should consider as the characteristic features of such a closed system of axioms and definitions. Perhaps the most important feature is the possibility of finding a consistent mathematical representation for it. This representation must guarantee that the system does not contain contradictions. Then the system must be suited to describe a wide field of experience. The great variety of phenomena in the field should correspond to the great number of solutions of the equations in the mathematical representation. The limitations of the field can generally not be derived from the concepts. The concepts are not sharply defined in their relation to nature, in spite of the sharp definition of their possible connections. The limitations will therefore be found from experience, from the fact that the concepts do not allow a complete description of the observed phenomena.

The preface by Paul Davies and the section in the book the which he refers show that the above is not a random quote taken out of context, but really represents Heisenbergs views, both on "closed system of axioms and definitions" in general, and on Copenhagen as the practice of interpretation close to "the original context" of QT.

This is really the bottom line of the argument, for quantum mechanics is, at its core, a mathematical scheme that relates the results of observations in a statistical fashion. And that is all. Any talk of what is 'really' going on is just an attempt to infuse the quantum world with a spurious concreteness for ease of imagination. In this connection Heisenberg examines the work of Descartes and Kant in the light of modern physics and concludes that words and their associated concepts do not have absolute and sharply defined meanings. They arise through our experiences of the world, and we do not know in advance the limits of their applicability. We cannot expect to uncover any fundamental truths about the world merely from the abstract manipulation of words and concepts. For Heisenberg the fact that certain cherished words and concepts simply cannot be transported into the relativity or quantum domains is not especially philosophically objectionable.

Any concepts or words which have been formed in the past through the interplay between the world and ourselves are not really sharply defined with respect to their meaning; that is to say, we do not know exactly how far they will help us in finding our way in the world. We often know that they can be applied to a wide range of inner or outer experience, but we practically never know precisely the limits of their applicability. This is true even of the simplest and most general concepts like `existence' and `space and time.' Therefore, it will never be possible by pure reason to arrive at some absolute truth.

The concepts may, however, be sharply defined with regard to their connections. This is actually the fact when the concepts become a part of a system of axioms and definitions which can be expressed consistently by a mathematical scheme. Such a group of connected concepts may be applicable to a wide field of experience and will help us to find our way in this field. But the limits of the applicability will in general not be known, at least not completely.

Even if we realize that the meaning of a concept is never defined with absolute precision, some concepts form an integral part of scientific methods, since they represent for the time being the final result of the development of human thought in the past, even in a very remote past; they may even be inherited and are in any case the indispensable tools for doing scientific work in our time. In this sense they can be practically a priori. But further limitations of their applicability may be found in the future.

Sadly, now this comment is already too long to also discuss Bohmian mechanics (and hope that anybody would read it). The examples would have been Bell type experiments where the decision by the experimenter can be modeled by a time dependent Hamiltonian in Bohmian mechanics, but not in MWI. And Quantum State Teleportation where the measurement result dependent unitary transformations applied to particle 3 work fine for Copenhagen, but not for Bohmian mechanics (this is the paragraph just above section 6 Conclusion in the link above):

Simply by noting the actual position ($x_0$) of the measuring device, the observer, near particles 1 and 2, immediately knows which wavepacket $x_0$ has entered, and therefore which state is active for particle 3. The observer then sends this classical information to the observer at 3 who will then apply the appropriate unitary transformation $U_1\dots U_4$ so that the initial spin state of particle 1 can be recovered at particle 3.

To see that this is incompatible with Bohmian mechanics, note that what is described here would be a backaction of the trajectories on the wavefunction, which is not possible (or at least not included) in Bohmian mechanics.

 

[Ken G]

Most mathematical theories can be interpreted both "literal" and "in their original context". For QT, MWI would be an attempt at a "literal" interpretation, and Copenhagen the practice of interpretation close to "the original context" of QT.

I'm still not seeing any observable contradictions. You are explaining what an interpretation is. I'm explaining what a physics theory isn't.

My reaction to any "literal" interpretation is fear of the dangers of stupid mistakes. I have no illusions about my chances to convince proponents of "literal" interpretations of the existence of such dangers.

And why should you try? What keeps you from making "stupid mistakes" is personal, like all philosophy. It would be valid science to test a thousand people of equal experience and aptitude who are trying to use QT to make predictions within the various interpretations, and score the number of stupid mistakes they make using each interpretation, but you can see how that is not the way debates over interpretations usually go.

So let me instead go with the existence of scalable quantum computers that are exponentially faster than classical computers for certain problems (like factoring) as a falsifiable predictions of MWI (because of David Deutsch). Notice that quantum computers are outside of "the original context" of QT.

The issue is not the "context" of QT, but the predictions of it. If what you are saying is outside the predictions of QT, then it's simply not QT. That's the point, a physics theory makes predictions, and if the predictions are different, it's a different theory, not a different interpretation. I realize that different interpretations of one theory can suggest different ways to alter that theory to create a new one, just as different interpretations of Newtonian physics led to general relativity. But that's because general relativity is a different theory, not because it's a different interpretation of Newton's theory. It is perfectly valid to debate how different interpretations of QT could lead to new different theories, indeed that's precisely what interpretations are for. But you see the pointlessness of arguing the interpretations themselves.

Additionally, for Copenhagen, quantum computers are related to limits like zero temperature or perfect isolation, and it remains unclear whether those limits stand in the way of scalability or not.

It sounds like you are saying the Copenhagen interpretation predicts something about quantum computers that other interpretations do not predict. That would be impossible, because then the CT would not be an interpretation of a theory, it would be different theory that makes different predictions using different equations. It could not use all the same equations, because how can you use all the same equations and make a different prediction? I think you are not talking about interpretations of QT, you are guessing at whatever the next theory might be. To be clear, I don't say that interpretations have no value, they have value in understanding a given theory, and a value in inspiring new theories that are different from the given theory. That's exactly what interpretations are supposed to do, but the interpretations don't make different predictions because only a theory can make a prediction.

In fact, I believe Bohmian mechanics provides the clearest mathematical examples to see how it contradicts both MWI and Copenhagen.

If there is such a contradiction, it must not come from "mathematical examples," it must come from different predictions. Experiment A must come out X for one and not X for the other, where X is a quantitative prediction. Theories produce such things, interpretations do not.

Euclidean geometry is (proto-)typical example of a mathematical theory.

You are now talking about mathematical theories. I would have thought it would be very clear in that context that interpretations of a mathematical theory are not a different theory, or they would be called a different theory with different axioms and postulates, not a different interpretation of the same axioms and postulates.

Where Euclidean geometry is quite untypical is that after the addition of THE missing axiom, it was indeed a complete theory from a purely mathematical point of view.

Any mathematical theory is defined by its axioms, so it is logically impossible for an axiom to be missing from one. I believe what you mean is that a better version of the theory that Euclid was trying to create can be made by adding an additional axiom that he did not think to add. That would make it a different mathematical theory. Whether a mathematical theory is "better" by using a different set of axioms is a question completely outside mathematics, but it does relate to the philosophical reasons we do mathematics. That is also like interpretations that are outside of science, but relate to the personal and philosophical reasons we do science.

Still, its applicability to the world around us and its range of validity remains open to interpretation.

A new meaning for "interpretation." Now you are not talking about interpretations of a mathematical theory, you are talking about interpretations of real world applicability. That sounds like philosophy for sure, and certainly not mathematics.

This again is quite typical for mathematical theories, as discussed again and again by Werner Heisenberg in his book Physics and Philosophy:

Thank you for quoting the insights of Heisenberg, but they just sound to me like a careful distinction between physics and philosophy, very much along the same lines as what I am saying. Heisenberg is saying that the physics is the theory that makes predictions, but how well those predictions correspond to the real world is something outside the physics theory. It is in the realm of testing the theory, and deciding if it meets whatever needs were set out for it. The interpretations of a theory are among the latter of those two, and can be very personal, like all philosophy. Paul Davies quote seems to agree with this as well, it's just what I'm saying. All I'm adding is that I find it curious people think QT is somehow different from all other physics theories, when in fact they all possess these exact same properties. You could take the quotes by Heisenberg and Davies and replace all references to QT with references to Newtonian mechanics and it's all just the same.

The examples would have been Bell type experiments where the decision by the experimenter can be modeled by a time dependent Hamiltonian in Bohmian mechanics, but not in MWI.

But this is exactly the point, at issue is not "how it can be modeled", but rather, the testable predictions it makes. From the perspective of a scientific theory, everything else is counting angels on a pin. From the perspective of personal philosophy, and inspirations for the next theory, that's where we have interpretations. But just as Heisenberg implied, the interpretations don't tell you anything, you still need to interview reality to obtain knowledge.

To see that this is incompatible with Bohmian mechanics, note that what is described here would be a backaction of the trajectories on the wavefunction, which is not possible (or at least not included) in Bohmian mechanics.

Again that all sounds, at best, like confusing an interpretation with a theory, and at worst, a category error about what an interpretation of a theory is. It is not the purpose of a theory to say what is really happening, the purpose is to make testable predictions, period. Some hold that the job of an interpretation is to say what is really happening, but that just sounds like kidding oneself to me. If one stays within the realm of science, the job of an interpretation is to create a personal sense of understanding of a theory, not to make sense of reality (as the latter can be done by no other means than experimentation). It is pure personal philosophy to use interpretations for the latter purpose, which is fine as a personal philosophy, but its only projection onto the scientific process appears in how it can be used to inspire new theories. Whether or not it is ever "what is actually happening" is entirely unscientific, because it is not testable beyond simply testing the theory itself, which never requires any interpretation or philosophy.

 

[PeterDonis]

Thread closed for moderation.

 

[PeterDonis]

After some cleanup, thread is reopened.

 

[gentzen]

I'm still not seeing any observable contradictions. You are explaining what an interpretation is. I'm explaining what a physics theory isn't.

Well, before trying to react to your other reactions, let's discuss your points about "what a physics theory isn't" and my points about "what an interpretation is".

So the main problem is, we tend to imagine that the purpose of a physics theory is to describe what is actually happening in a place we think of as "objective reality." But such a place is never part of any physics theory, since theories just manipulate abstract notions to create predictions that can be tested.

I have no problem with that. And as you already noticed, Heisenberg defended a very similar position.

What this also means is that we are untrue to the scientific method when we hold that any interpretation of quantum mechanics is a statement about objective reality, because if the theory was never demonstrably that, then so much less is an interpretation of a theory.

Except that again and again, interpretations did just that. But you are right that Laplace was "untrue to the scientific method" when he did that. So you seem to be talking about what "interpretations in an ideal world" should be, not about how proponents of different "interpretation" behave in our world.

Instead, an interpretation of a theory is a kind of lens through which to look at a theory, a way to understand or make sense of the theory, not objective reality itself (as the latter is only what we observe, and the predictions we test thereby).

But if you look at a sphere from the front, no lens will change the fact that you cannot see the back. And if somebody else looks at the back, he can see something unrelated and different from what you see. And just like in the parable of the blind men and an elephant, he might start to argue that your observations contradict his observations.

When I write that you misunderstand the relation between theory and interpretation, then I have something in mind like my own surprise, when I learned that a group representation is defined as

More formally, a "representation" means a homomorphism from the group to the automorphism group of an object. If the object is a vector space we have a linear representation. Some people use realization for the general notion and reserve the term representation for the special case of linear representations.

Intuitively, I would have expected that it should say monomorphism. But it says homomorphism. But the words realization and representation suggested to me that the "entire group" should be encoded, not just some arbitrary aspect of it. But then I learned that things just work better this way. And it is the same with interpretations. It is OK that Bohmian mechanics can only interpret non-relativistic QM, and has trouble interpreting QFT. It keeps its status as an interpretation, even if it cannot overcome its troubles with QFT. And other interpretations are allowed to miss aspects of QT too. For example, I would argue that MWI is currently blind when it comes to temperature.

---

I'm still not seeing any observable contradictions. You are explaining what an interpretation is. I'm explaining what a physics theory isn't.

In total, I gave four examples. The MWI example and the Euclidean geometry example were carefully selected "especially for your request", and are in principle open to experimental falsification. However, your reaction was a sort of disbelief, or at least the feeling that I somehow lost you, or didn't get your point.

I initially thought about using the Eikonal equation and the inviscid Burgers' equation as examples. Here, the math stays silent about what happens in the singularities (i.e. the points where the characteristics intersect), and there are different "interpretational attitudes", from "singularities only occur in pathological artificially constructed mathematical examples," over "obviously the entropy solution is the correct one, and the math told you that from the start", and "the caustic with multivalued functions is the correct solution, because the equation comes from geometric optics", to "we are agnostic as to what happens in the singularities, because this is outside of the domain of application for which the equations have been derived".

But I was unsure whether you would accept those equations as physical or mathematical theories. They are not theories in my sense, because nobody calls them theories, they are called equations. (For me, the actual usage of the words in practice is what is most important.) The investigation of partial differential equations on the other hand is called a theory, but "today" those interpretational ambiguities are already part of that theory. But maybe it would have been clearer for you that there are different interpretations with observable consequences, which can be right or wrong, depending on context. (They can contradict each other, but not the equation.)

It sounds like you are saying the Copenhagen interpretation predicts something about quantum computers that other interpretations do not predict.

Why are you so sure that the prediction from MWI is correct? I just say that Copenhagen is agnostic, especially with respect to scalability, as long as the context of temperature and non-isolation is missing. Maybe MWI should stay agnostic too, but some of its proponents claim that scalable quantum computers would be one of its predictions.

That would be impossible, because then the CT would not be an interpretation of a theory, it would be different theory that makes different predictions using different equations.

Maybe MWI simply tempted some of its proponents to make a stupid mistake? Maybe "stupid" was an unfortunate word from my side. What I mean is "unnecessary" or "avoidable", in the sense that making that mistake is not necessary for defending the interpretation.

It could not use all the same equations, because how can you use all the same equations and make a different prediction? I think you are not talking about interpretations of QT, you are guessing at whatever the next theory might be.

But this is exactly the point, at issue is not "how it can be modeled", but rather, the testable predictions it makes. From the perspective of a scientific theory, everything else is counting angels on a pin. From the perspective of personal philosophy, and inspirations for the next theory, that's where we have interpretations.

Guessing at the next theory is not what I do. I admit that my writing may be hard to follow, or that I didn't get your point. But this "next theory" thing was nowhere contained in my text. You have put it in there yourself. Also for Euclidean geometry, it is not that general relativity is a better theory where it is no longer "literally" valid. It might simply not be "literally" valid in the world out there (without any next theory having anything to do with that).

 

[kered rettop]

Well, Schroedinger's Cat started, as Schroedinger himself said, as a "quite ridiculous case". It was intended to highlight a serious problem with the Copenhagen Interpretation. However the essential problem has not gone away.

 

[vanhees71]

Which problem, do you think, "has not gone away"? Of course, the Copenhagen Interpretation is indeed a problem, particularly with those variants which postulate a collapse.

 

[kered rettop]

Which problem, do you think, "has not gone away"? Of course, the Copenhagen Interpretation is indeed a problem, particularly with those variants which postulate a collapse.

 

[kered rettop]

Yeah that was careless of me. I just meant that we still need to understand the superposition, not that there is an unsolved problem in "ordinary" QM. Of course, my preferred understanding is MWI, which many people would say is "quite ridiculous". But that's another problem.

 

[vanhees71]

What do you think is the problem with superposition? I have never understood what the point of the MWI is. For me, QT predicts probabilities for the outcome of measurements, and in everyday lab experience there's a clear outcome of each measurement (modulo if there are technical problems with the equipment, but this has of course nothing to do with QT as a theory). The outcome is random with probabilities as predicted by QT, and QT tells me that this randomness is an objective property of Nature. That leaves just the statistical (ensemble) interpretation of QT, and with this "minimal interpretation" there are never contradictions or logical problems, and until today, even under very stringent tests, no example for a failure of QT has become known.

If there is some problem with QT, it's not a physical one, as far as the realm of applicability is concerned. The only real problem is the apparent incompatibility between QT and general relativity, i.e., a QT of the gravitational interaction. The problem here seems to be, however, not a mere interpretational/philosophical problem but the lack of empirical input for figuring out, how a QT of gravitation should look like.

 

[kered rettop]

What do you think is the problem with superposition? I have never understood what the point of the MWI is. For me, QT predicts probabilities for the outcome of measurements, and in everyday lab experience there's a clear outcome of each measurement (modulo if there are technical problems with the equipment, but this has of course nothing to do with QT as a theory). The outcome is random with probabilities as predicted by QT, and QT tells me that this randomness is an objective property of Nature. That leaves just the statistical (ensemble) interpretation of QT, and with this "minimal interpretation" there are never contradictions or logical problems, and until today, even under very stringent tests, no example for a failure of QT has become known.

If there is some problem with QT, it's not a physical one, as far as the realm of applicability is concerned. The only real problem is the apparent incompatibility between QT and general relativity, i.e., a QT of the gravitational interaction. The problem here seems to be, however, not a mere interpretational/philosophical problem but the lack of empirical input for figuring out, how a QT of gravitation should look like.

I didn't say there was a problem with superposition, in fact I said there wasn't a problem in ordinary QT.
Thanks for explaining your preferred interpretation.

 

[kered rettop]

One has to start the experiment by putting the cat, poison and trigger in the box so the state is completely specified when the box is closed. What then causes the state to become entangled?

The triggering atom evolves into a superposition of |decayed> and |not decayed> The |decayed> state includes the emitted particle, which triggers the killing mechanism and kills the cat. The |not-decayed> state essentially leaves the cat unaffected. So the complete state is a superposition of |dead-cat stuff> and |living-cat stuff>
I'm not sure where the term entanglement came into the discussion, but you can define an entanglement as a superposition of correlated states. So just divide the "stuff" into the atom and the cat etc. You then have a superposition of |atom intact>|cat alive> with |atom decayed>|cat dead> which is, by definition, an entanglement between the cat etc and the atom.

 

[vanhees71]

Indeed, in the same article Schrödinger introduced his famous cat, he also coined the notion of entanglement as THE key aspect of the interpretational issues of QM in his time. For me it's incomprehensible, why this very clearly written article is so much less famous than the cloudy EPR article with the opaque response by Bohr. For the English translation, see

https://www.jstor.org/stable/986572

 

[PeterDonis]

I'm not sure where the term entanglement came into the discussion

Because there is an interaction between the radioactive atom and the cat (mediated by the poison that the decay of the atom releases) that entangles their respective degrees of freedom.

you can define an entanglement as a superposition of correlated states.

This is not the correct definition of entanglement. The correct definition is that an entangled state of a system cannot be expressed as a product state of its subsystems.

 

[kered rettop]

I'm not sure where the term entanglement came into the discussion

Because there is an interaction between the radioactive atom and the cat (mediated by the poison that the decay of the atom releases) that entangles their respective degrees of freedom.

You misunderstand me. I was not asking why entanglement come into it. I was querying why a discussion that had been using the term superposition exclusively, suddenly became a question about entanglement. To answer it was necessary to bridge between the concept of superposition and that of entanglement. Which I attempted to do.

you can define an entanglement as a superposition of correlated states.

This is not the correct definition of entanglement. The correct definition is that an entangled state of a system cannot be expressed as a product state of its subsystems.

True. I should probably have said something like "a superposition of correlated states is a sufficient condition for the state to be an entanglement according to the correct definition". But the rest of what I said follows anyway, and I really don't think anyone is likely to have been confused.

Have a Happy New Year!

 

[kered rettop]

There is never any problem if the components "stay in their lanes": observations establish objective reality, theory is a prescription for making testable predictions that show up in objective reality when we test them. All we have to do is take greater pains to track what science actually does, and the "Schroedinger cat problem" never appears at all.

That's a fairly common stance, but it only crops up in connection with QM, all other sciences expect their theories to be ontic. But since 1957 we've had ontic QM in the shape of MWI. You don't need to exclude onticity in order to get rid of the SC problem.

 

[gentzen]

That's a fairly common stance, but it only crops up in connection with QM, all other sciences expect their theories to be ontic. But since 1957 we've had ontic QM in the shape of MWI. You don't need to exclude onticity in order to get rid of the SC problem.

We don‘t need to wait for MWI, de Broglie Bohm was there much earlier, and is ontic too. Don‘t get me wrong, MWI has its place and motivations too, and it would be good for PF to have some true MWI proponent(s) again.

 

[vanhees71]

The trouble with de Broglie-Bohm is that there's no satisfactory version for relativistic QFT. Also, I don't get the obsession about a theory being "ontic" or "epistemic". I think all physical theories are about a quantative description of observable facts from as little assumptions as possible.

 

[PeterDonis]

I was querying why a discussion that had been using the term superposition exclusively, suddenly became a question about entanglement.

I would say, because entanglement is the correct term for this discussion, whereas superposition is not.

 

[kered rettop]

Ha ha! Yes, quite so.

 

[tomj]

Do you have a reference for superconducting qubits? AFAIK quantum computing experiments are done using polarizations of single photon states as qubits.

A cat made up of super conducting qubits would be one cool cat

 

[PeroK]

A cat made up of super conducting qubits would be one cool cat

 

[Ken G]

That's a fairly common stance, but it only crops up in connection with QM, all other sciences expect their theories to be ontic. But since 1957 we've had ontic QM in the shape of MWI. You don't need to exclude onticity in order to get rid of the SC problem.

Indeed I'm arguing something like the converse of this. It's true that if our goal is to regard ontology as the primary goal of science, then even QM can be shoehorned into that picture. However, I'm saying that QM is the place where we encounter what we should have known all along: ontology is a convenience of science not the goal of science. The goal of science is purely epistemological, as all science must be, by its very definition. QM is trying to remind us of that, but if we are not listening, we can ignore it there too.

 

[Ken G]

The trouble with de Broglie-Bohm is that there's no satisfactory version for relativistic QFT. Also, I don't get the obsession about a theory being "ontic" or "epistemic". I think all physical theories are about a quantative description of observable facts from as little assumptions as possible.

That means you are arguing for epistemics. The distinction between ontics and epistemics is the former is about what exists, and the latter is about our knowledge. A quantitative description that checks with observed facts is all clearly about knowledge. Knowledge of the quantities, knowledge of the observations, knowledge about how to compare the two in some kind of fair or useful way, all decided inside our brains. None of it has anything to do with "what actually is", it's all about "what we can know, and how we can use that knowledge to achieve our goals." That's literally exactly what epistemology is all about, or so it seems to me-- perhaps you think of something else associated with that term. (What meanings we associate with terms is also super important, so we should probably say that science combines epistemology with semantics, and uses ontology as a convenience along the way, but we should use that convenience carefully for fear that we begin to think that science is fundamentally ontological. Regarded as ontology, the history of science is a history of dismal failure. Regarded as epistemology, the history of science is a history of spectacular success.)

 

[kered rettop]

Indeed I'm arguing something like the converse of this. It's true that if our goal is to regard ontology as the primary goal of science, then even QM can be shoehorned into that picture. However, I'm saying that QM is the place where we encounter what we should have known all along: ontology is a convenience of science not the goal of science. The goal of science is purely epistemological, as all science must be, by its very definition. QM is trying to remind us of that, but if we are not listening, we can ignore it there too.

Well thanks for that, Ken, but I really have no palate for teleological or prescriptive theories of science. So "I'm out".

 

[vanhees71]

That means you are arguing for epistemics. The distinction between ontics and epistemics is the former is about what exists, and the latter is about our knowledge. A quantitative description that checks with observed facts is all clearly about knowledge. Knowledge of the quantities, knowledge of the observations, knowledge about how to compare the two in some kind of fair or useful way, all decided inside our brains. None of it has anything to do with "what actually is", it's all about "what we can know, and how we can use that knowledge to achieve our goals." That's literally exactly what epistemology is all about, or so it seems to me-- perhaps you think of something else associated with that term. (What meanings we associate with terms is also super important, so we should probably say that science combines epistemology with semantics, and uses ontology as a convenience along the way, but we should use that convenience carefully for fear that we begin to think that science is fundamentally ontological. Regarded as ontology, the history of science is a history of dismal failure. Regarded as epistemology, the history of science is a history of spectacular success.)

We can only know "what actually is" by observations. So what should ontology add to science, which is about what we can objectively observe in Nature and finding descriptions in terms of a few fundamental "natural laws" that describe a lot of different phenomena. Asking for more in the sense of "what it really is" is not a question answered by the Natural Sciences, and this restriction makes it much more successful and applicable than such metaphysical speculations. That's why, if I'm sick, I'll rather take some medication based on science rather than speculations like homeopathy ;-)).

 

[kered rettop]

We can only know "what actually is" by observations. So what should ontology add to science, which is about what we can objectively observe in Nature and finding descriptions in terms of a few fundamental "natural laws" that describe a lot of different phenomena. Asking for more in the sense of "what it really is" is not a question answered by the Natural Sciences, and this restriction makes it much more successful and applicable than such metaphysical speculations. That's why, if I'm sick, I'll rather take some medication based on science rather than speculations like homeopathy ;-)).

Disproven speculations!

 

[PeterDonis]

This thread has run its course and is now closed. Thanks to all who participated.