I Wheeler's delayed choice doesn't change the past

[RUTA]

Well, I don't believe much in "philosophy of science". If there is a paradigm change this must come from reproducible objective observations not from scholastic thinking about "angles on the head of a pin".

For someone who claims not to believe in philosophy of science, you certainly practice it liberally in this thread! I started teaching physics as an undergraduate physics major in 1981. I continued as a grad student then as an assistant prof, associate prof and finally a full prof. I’ve pretty much taught every subject in physics over those 36 years, publishing and refereeing journal articles, presenting at conferences, and recently writing a book representing 23 years of research. That research was in foundations of physics which includes publications in physics journals fitting dark energy and dark matter data, and refereeing papers written by notables such as Aharonov, Smolin, Price, and Elitzur, for example. I was fortunate to learn early in my career that, like Max Planck said, “There is no physics without some metaphysics,” and metaphysics is philosophy. The understanding that any practicing physicist is necessarily making philosophical assumptions has helped me greatly in crafting a worldview based on physics, because I was able to confront and challenge my philosophical assumptions in my practice of physics. I was also lucky to have a colleague in philosophy of physics whose PhD thesis was on the Bell inequality. Without him challenging my philosophical assumptions in physics for the past 23 years, we would never have put together the approach to quantum gravity outlined in “Beyond the Dynamical Universe.” Take it from this old timer, there is much about physics to learn if you’re willing to engage in philosophy of physics :-)

 

[bhobba]

Take it from this old timer, there is much about physics to learn if you’re willing to engage in philosophy of physics :-)

Eloquently expressed and I will not argue with it.

But surely you have noticed there are many physicists, and those like me interested in physics, that find philosophy leaves them cold. Of course it doesn't make it right - or wrong either - its just the way some people are.

What I have noticed on this forum there are those who like Bohr and more currently Wallace who has a PhD in both, are very imbued with philosophy and those like Dirac, Feynman and Weinberg that are not. I am in the second camp, but that does not make me correct - its just the way me and similar people are. You and others are different - and it's just the way they are.

I don't know what view will eventually make progress - as Feynman said in the past sometimes a philosophical view like logical positivism made progress as done by Einstein and Heisenberg, and other times just fooling around with equations like Dirac did made progress. When progress isn't being made its because all the 'tricks' that worked in the past now don't work - they would have been tried and because progress has paused obviously are now not working.

I have zero idea what will work to make progress in the future - the type of things you suggest, or what Dirac did - who knows. All I think is necessary is to understand we all are different and understand those differences.

And yes many have wryly noted that those anti philosophy like Feynman and Weinberg have in fact adopted a philosophical position. So I think the question is not so much whether you practice philosophy or not - but what one you have chosen.

Thanks
Bill

 

[stevendaryl]

Ahhh. But that's not what the Born rule says is it - so 'seems to me' is obviously wrong. Intuition and QM are sometimes not good friends.

Yes, I know that. The Born rule assumes a split between the observer and the thing being observed. But surely observers are quantum systems, as well.

The derivation of Gleason's theorem is the same way. It's not really assuming basis-independence. Given a split between observer and thing being observed, you can equally well assume any correlation between the states of the observed and the states of the observer. But the split itself is not basis-independent.

 

[DrChinese]

Eloquently expressed and I will not argue with it.

But surely you have noticed there are many physicists, and those like me interested in physics, that find philosophy leaves them cold. ...

I'd go further, and say your statement applies to the majority of physicists. And in another recent thread, the OP asks "What is the value of "delayed choice" experiments?" It wouldn't make sense if all physicists believed the exact same things, and similarly thought that all future research should focus on the same things. They'd all be going down the same road!

I remember reading articles written prior to 1965 arguing that the universe was "steady state" or "continuous creation"... essentially on aesthetic grounds of one type or another. I don't think the question of whether those - or the newfangled "big bang" theory - kept most physicists up at night back then. And certainly not any more than the philosophy of QM interpretations keep most physicists up at night today. But there have always been those whose curiosity DOES keep them up at night. And some of those have gone on to answer some of those deep questions. Again: I don't think it is reasonable to predict where the next big discovery is coming from... or who is going to make it. Or whether it will follow "causal" ideas or not.

Honestly, I scoff at the idea that we live in a causal universe in the first place. Other than the "deterministic" movement of bodies due to general relativity, a reasonable person can question the role of causality versus chance in most of science:

1. Evolution.
2. Human behavior.
3. Quantum mechanics.
4. The initial conditions of our universe.
5. The weather and forecasts for our planets future.

So I'm not sure where anyone gets the idea that the above subjects cannot be studied unless we first assume causality. All we need is to be able to see patterns and pattern exceptions. That would be enough to make a scientific advance.

 

[stevendaryl]

Yes, I know that. The Born rule assumes a split between the observer and the thing being observed. But surely observers are quantum systems, as well.

The derivation of Gleason's theorem is the same way. It's not really assuming basis-independence. Given a split between observer and thing being observed, you can equally well assume any correlation between the states of the observed and the states of the observer. But the split itself is not basis-independent.

A little elaboration on that: In an observation or measurement, you have two systems, the observer system (measuring device, or experimenter, or whatever) and the observed system, (an electron, say). To say that the set up is measuring property A of the observed system is to say that there is an interaction between the two systems such that different values of A in the observed system leads to macroscopically distinguishable states of the observer system (typically, in an irreversible way). The quantum formalism is independent of the basis for the observed system, but not for the composite system.

 

[bhobba]

Yes, I know that. The Born rule assumes a split between the observer and the thing being observed. But surely observers are quantum systems, as well.

Yes - that is the well known issue talked often about here with MW:
https://arxiv.org/abs/1210.8447

I think it's wrong because structure emerges naturally eg gravitational fields cause clumps of matter to form.

IMHO there is nothing to gain rehashing it again - its been done to death.

The derivation of Gleason's theorem is the same way. It's not really assuming basis-independence. Given a split between observer and thing being observed, you can equally well assume any correlation between the states of the observed and the states of the observer. But the split itself is not basis-independent.

I think you can dig up a proof and see the assumptions made. It's usually called non-contextuality - but is really saying the probabilities do not depend on what basis a vector may be part of eg basis independence.

Thanks
Bill

 

[stevendaryl]

I think you can dig up a proof and see the assumptions made.

Yes, and from the very beginning, it is assuming a split between observer and observed. He poses the problem as how to derive expectation values from states.

 

[bhobba]

Yes, and from the very beginning, it is assuming a split between observer and observed. He poses the problem as how to derive expectation values from states.

To cut this short - the issue is this. Yes you do have a cut between what is observed and what does the observing. The decoherence paradigm depends on it. The issue though is this - what if the cut is placed differently. General system cuts don't really make a difference in physics - you choose the simplest one. We do not have theorems if this breaks down in QM.

Thanks
Bill

 

[bhobba]

Honestly, I scoff at the idea that we live in a causal universe in the first place.

I think its very definition dependent personally - such things always worry me.

Thanks
Bill

 

[RUTA]

But surely you have noticed there are many physicists, and those like me interested in physics, that find philosophy leaves them cold. Of course it doesn't make it right - or wrong either - its just the way some people are.

All physicists make philosophical assumptions, whether they choose to acknowledge them or not. Most physicists can't be bothered to explore their philosophical assumptions and they don't need to be, they're still able to do their physics. Philosophy doesn't just leave me cold, it irritates the hell out of me! I want to do physics, not argue endlessly about my philosophical assumptions! I have told my philosophy colleagues on many occasions that I've explored my assumptions, made my choices and now I'm doing physics accordingly. They, on the other hand, don't ever stop such arguing b/c that's what it means to do philosophy. That's what they enjoy. A physicist needs to know where to draw the line in dealing with philosophers, so they can get on with doing physics. But, in PF threads about the ontology underlying delayed choice, it is precisely those philosophical assumptions that must be articulated. So, I provide my philosophical assumptions when responding to such questions on PF. I believe those assumptions are interesting, or I wouldn't bother sharing them, but I spend most of my time doing physics in accord with those assumptions (e.g., fitting the SCP Union2.1 SN data in accord with modified Regge calculus or fitting THINGS galactic rotation data, ROSAT/ASCA data on galactic cluster mass profiles and Planck CMB anisotropy data using GR contextuality, or fitting nested interferometer data on photon detections based on modified lattice gauge theory), rather than debating them. When someone starts a PF thread on dark energy or dark matter, I articulate their assumptions then show how different assumptions lead to different physics.

I suspect we agree that philosophy for the sake of philosophy, i.e., without the concordant physics, is of little or no interest to us :-)

 

[stevendaryl]

To cut this short - the issue is this. Yes you do have a cut between what is observed and what does the observing. The decoherence paradigm depends on it. The issue though is this - what if the cut is placed differently. General system cuts don't really make a difference in physics - you choose the simplest one. We do not have theorems if this breaks down in QM.

The cut is a red herring, in my opinion. The empirical predictions of QM can (in principle, at least) be obtained without a cut along the following lines:

1. You figure out the Hilbert space \mathcal{H} and Hamiltonian H for the whole shebang: observers, measuring devices, environment, observed system, everything.
2. You pick a coarse-graining for the macroscopic states. This can be done lots of different ways, but you can summarize it by coming up with a set of Hilbert space projection operators \Pi_j corresponding to each macroscopic state, j
3. You pick an initial state, |\psi\rangle
4. Then the probability of being in macroscopic state j a time t later would be \langle \psi| e^{iHt} \Pi_j e^{-iHt} |\psi\rangle

This way of looking at it, QM is a stochastic theory for the evolution of the macroscopic state. So the business of state preparation, observables, etc., would be seen as a shortcut, or rule of thumb.

 

[Lord Jestocost]

Honestly, I scoff at the idea that we live in a causal universe in the first place.

I think its very definition dependent personally - such things always worry me.

Thanks
Bill

Nothing to worry!

„The position is that the laws governing the microscopic elements of the physical World - individual atoms, electrons, quanta - do not make definite predictions as to what the individual will do next. I am here speaking of the laws that have been actually discovered and formulated on the old quantum theory and the new. These laws indicate several possibilities in the future and state the odds on each. In general the odds are moderately balanced and are not tempting to an aspiring prophet. But short odds on the behaviour of individuals combine into very long odds on suitably selected statistics of a number of individuals; and the wary prophet can find predictions of this kind on which to stake his credit - without serious risk. All the successful predictions hitherto attributed to causality are traceable to this. It is quite true that the quantum laws for individuals are not incompatible with causality; they merely ignore it. But if we take advantage of this indifference to reintroduce determinism at the basis of world structure it is because our philosophy predisposes us that way, not because we know of any experimental evidence in its favor.“

Sir A. S. Eddington in „THE NATURE OF THE PHYSICAL WORLD“, Cambridge, 1948

 

[Stephen Tashi]

I believe Hawking is exaggerating in telling us (page 83):
«...the Universe doesn't have just a single history, but every possible history, each with its own probability; and our observations of the current state affect its past and determine the different histories of the Universe, just as the observations of the particles in the double slit experiment affect the particles' past.»
Do you agree?

From the viewpoint of logic and pure mathematics, that's true by definition!

If we take the (metaphysically) controversial viewpoint that correct physics is organized as mathematics, then what "exists" in mathematical problems is certain given information and deductions made from that information. So if there are phenomena that (in the sense of common language) "existed" in the past but are not present or deducible from the current "given" information they don't exist in the sense of being possible subject matter in a mathematical problem. If, from the current given information, we can infer only certain probabilities for a past event, then what exists in the mathematical problem is those probabilities.

(e.g. If we are given only that Bob has 3 apples and Alice has twice as many apples as Bob then how many eggs Bob ate for breakfast does not exist as part of the given information.)

From a classical point of view, one might seek to refute this proof-by-definition experimentally by secretly recording some event and then arguing the contradiction that some other physicist would later treat the event as non-existent. However, the total information for that experiment includes the fact that the event was recorded by us.

From a classical point of view, one could argue from (empirical) induction that observed trees that fall make a sound and have other effects - therefore unobserved trees that fall make sounds and have similar effects even if these effects are unknown in our present state of knowledge. That gives the philosophical satisfaction of being able to say that the past includes "definite but unknown" events. From the mathematical point of view, that affects the existence of given information insofar as axioms based on such an induction add to what can be deduced.

I'll distinguish 3 definitions for the "existence" of the past - or anything else
1) The common language meaning - whatever that is!
2) Existence in the sense of specific given or deducible information in a mathematical problem
3) Existence in the mathematical sense of "given but unknown" (e.g. There exists an x such that...) This would include contexts where a mathematical theory has axioms about unobserved events having definite outcomes etc.

In the sense of 2), the quotation above says that information about the current state of the universe does not allow us to deduce a unique past history, so that unique history does not exist (as given information) in any correct physical theory. Whether the author of that quotation had this interpretation in mind, I don't know.

In the sense of 3) the quotation would assert that a mathematical theory of physics that assumes the past has a definite but unknown state is self contradictory or, at best, no better than a theory that omits such an assumption.

 

[bhobba]

I suspect we agree that philosophy for the sake of philosophy, i.e., without the concordant physics, is of little or no interest to us :-)

Yes.

My issue with philosophy isn't that I think its useless (I do - but that is just an opinion and I acknowledge as always opinions are like bums - everyone has one - but it doesn't make it right) it's I don't often actually even understand it. To me it's not discussing angels dancing on a pin - its discussing what I find often not even understandable. Take complementary - I can't even make sense of it. And that's something pretty basic. We had a philosopher come here discussing his view of MW - to me it was gibberish, and other science advisers thought the same. This makes it really really hard IMHO.

Check this out:
https://www.scienceandnonduality.com/wp-content/uploads/2014/09/einstein_tagore.pdf

When I read it I thought - is this guy for real? Why did Einstein even waste his time. To counter this down the ANU I did a course in philosophy as a postgraduate student for a year at their center for continuing studies. Practically it was good because it let me have official access to their library - but nobody really worried anyway - they more or less let anyone go in and read. I mentioned a lot of things I thought very profound like Noether's Theorem - but was consigned to the merely material. The maddening thing was my teacher, Petra was her name, had a very keen mind. So they are actually quite intelligent - but on a different wavelength.

I have come to the conclusion its simply, to paraphrase another saying, Physicists are from Mars, Philosophers from Venus.

Thanks
Bill

 

[bhobba]

This way of looking at it, QM is a stochastic theory for the evolution of the macroscopic state. So the business of state preparation, observables, etc., would be seen as a shortcut, or rule of thumb.

Docoherent histories maybe?

It was what Feynman thought in the end.

Thanks
Bill

 

[Lord Jestocost]

Check this out:
https://www.scienceandnonduality.com/wp-content/uploads/2014/09/einstein_tagore.pdf

When I read it I thought - is this guy for real? Why did Einstein even waste his time.

With all due respect, it seems that you have no idea at all who Rabindranath Tagore was. Your remark is misplaced.

 

[bhobba]

If we take the (metaphysically) controversial viewpoint that correct physics is organized as mathematics,

I don't even know what that means. Do you mean the laws are written in the language of math? If so I agree. If not can you be more specific? Then the rest you wrote likely will be clearer.

Thanks
Bill

 

[bhobba]

With all due respect, it seems that you have no idea at all who Rabindranath Tagore was. Your remark is misplaced.

Could be since I understood nothing of what he said. Of course I looked up who he was and what he did which was of great importance in India and probably elsewhere - but that doesn't change I couldn't understand it at all. Again I emphasize that means nothing - quite possibly just my own shortcomings. I did however understand what Einstein said - do you think Einstein understood him?

Thanks
Bill

 

[Stephen Tashi]

Do you mean the laws are written in the language of math?

Yes, and further (from that viewpoint) what "exists" in a physical situation is exactly what exists in the correct mathematical formulation of that situation. So if information about something doesn't exist in the mathematical formulation (e.g. which slit the particle went through) then this something does not exist physically. (I'm not saying this is my personal viewpoint, but I think it is a defensible viewpoint.)

 

[bhobba]

Yes, and further (from that viewpoint) what "exists" in a physical situation is exactly what exists in the correct mathematical formulation of that situation. So if information about something doesn't exist in the mathematical formulation (e.g. which slit the particle went through) then this something does not exist physically. (I'm not saying this is my personal viewpoint, but I think it is a defensible viewpoint.)

I liked it because I now understand what you said - or think I do anyway. But I hold a different view.

My view is the connection between applied mathematics and what its applied to is hard to pin down. That's the understanding part of a theory - it takes a while to build up. Now let's stay on topic with this thread. In the delayed choice experiment on the surface it looks like something really weird is going on - but when analysed carefully that isn't he case:
http://quantum.phys.cmu.edu/CQT/chaps/cqt20.pdf

The purpose of the math is to keep our thinking clear and on the straight and narrow. It turns out our laws, for reasons I do not know, are written in the language of math. Its very strange that God is a mathematician - but he seems to be. Its so strange I believed at one time in Penrose's view - but changed my mind when I realized what a strange view I was led to - but it is just so seductive.

Thanks
Bill

 

[vanhees71]

What's your view of the Wheeler-Feynman absorber theory?

I am not a fan of that or the transactional interpretation, but that scientifically means diddly squat. As usual I agree with you but I don't think the situation is quite as clear cut as you said - or maybe its is - but you can best explain what you mean because I think its a bit context dependent ie we can't send information backward in time - that would be exactly as you say.

Thanks
Bill

The Wheeler-Feynman absorber theory is not acausal, but it's classical. In one of Feynman's semi-autobiographical books (very amusing to read) Feynman tells about it that Wheeler promised the quantum theory, but this was too optimistic. So it's an interesting point of view on classical electrodynamics but a dead end of research towards a reformulation of QED.

 

[Lord Jestocost]

I did however understand what Einstein said - do you think Einstein understood him?

Tagore points out that it is impossible to think “your consciousness” away from reality: “…it is a relative world, depending for its reality upon our consciousness“. Thus, Einstein has finally to admit that his conception of „the world as a reality independent of the human factor“ is nothing but a religion, without any scientific basis.

 

[RUTA]

Keep in mind what we’re trying to explain is the correlation b/w spacelike separated events that violates Bell’s inequality, not the simple fact that the Source emission event resides in their mutual causal past lightcones. The violation of Bell’s inequality says simply that you cannot explain that correlation in strictly forward-time-evolved and timelike fashion. That is generally what people mean by “causal,” although you may have a different definition in which case the discussion is strictly semantics. If you’re willing to drop “timelike fashion” and allow “spacelike fashion,” then you can use superluminal forward-time-evolved causation to explain the correlation. The problem with that is SR tells us the order of spacelike separated events is frame dependent, so you either have to select a preferred frame or allow backward-time-evolved causation. But, if you’re willing to allow backward-time-evolved causation (aka retrocausation), then you can still restrict your time-evolved explanation to the past lightcones of the detection events (“timelike fashion”) anyway. So, you see that forward-time-evolved causation is out unless you want a preferred frame. The only way to maintain time-evolved causation is retrocausation. That’s why some have chosen another route, i.e., an adynamical block universe explanation. Accordingly, the most fundamental explanation for the correlation is a spatiotemporally global rule that, in this case, is just not amenable to forward-time-evolved and timelike storytelling (dynamism). It’s a very powerful explanatory method (as we show in our book) and quite consistent with the Lagrangian formalism of physics, i.e., all current physics is perfectly reasonable in the “God’s-eye view,” as Wilczek calls it. It’s our anthropocentric “ant’s-eye” bias that keeps us from using the more powerful adynamical (“acausal” as defined above) explanation in conjunction with our very successful physics.

 

[bhobba]

Thus, Einstein has finally to admit that his conception of „the world as a reality independent of the human factor“ is nothing but a religion, without any scientific basis.

Its true - I don't think Einstein finally had to admit that - its almost trivially obvious. I get him saying that - that's not the issue. Einstein on many occasions admitted to being quite religious - believing in the god of Spinoza - he was quite open about it. But before jumping to conclusions look up what the god of Spinoza is if you don't already know..

That said things like the below leave me cold:
'When our universe is in harmony with man, the eternal, we know it as truth, we feel it as beauty'

Being nice I would say I don't understand it - but really for me I think the above is mystical nonsense. I think Einstein may have felt the same - but of course we will never know. I get the 'feeling' he was having a bit of 'fun' with this guy - just a feeling - we will never know. Einstein nearly always, unlike some contemporaries like Bohr generally expressed himself clearly - not always correctly - but clearly. This guy did not do that which I think would have irked Einstein. But, as I said, we will never know.

Thanks
Bill

 

[bhobba]

Keep in mind what we’re trying to explain is the correlation b/w spacelike separated events that violates Bell’s inequality, not the simple fact that the Source emission event resides in their mutual causal past lightcones. The violation of Bell’s inequality says simply that you cannot explain that correlation in strictly forward-time-evolved and timelike fashion.

I think I will have to read your book (of which I will get a copy when released) to understand what you are saying.

Just a few points:
1. Right at the very foundations of ordinary QM is the Galilean transformations. So non-locality is built in from the start. You have to go to QFT to discuss locality and it is embodied in the cluster decomposition property:
https://www.physicsforums.com/threads/cluster-decomposition-in-qft.547574/

2. Note the exclusion in that property - correlations. EPR is just a correlation so IMHO is no mystery at all - the key thing is if you want to have CFD you need non-locality. Don't care about CFD - then no issue - we just have some different kind of statistical properties to those correlations than classically. That's hardly surprising because we also know QM as a generalized probability model is different to ordinary probability - ordinary probability is the simplest - QM the next simplest.

3. Regarding delayed choice QM explains it just fine:
http://quantum.phys.cmu.edu/CQT/chaps/cqt20.pdf

Thanks
Bill

 

[vanhees71]

Keep in mind what we’re trying to explain is the correlation b/w spacelike separated events that violates Bell’s inequality, not the simple fact that the Source emission event resides in their mutual causal past lightcones. The violation of Bell’s inequality says simply that you cannot explain that correlation in strictly forward-time-evolved and timelike fashion. That is generally what people mean by “causal,” although you may have a different definition in which case the discussion is strictly semantics. If you’re willing to drop “timelike fashion” and allow “spacelike fashion,” then you can use superluminal forward-time-evolved causation to explain the correlation. The problem with that is SR tells us the order of spacelike separated events is frame dependent, so you either have to select a preferred frame or allow backward-time-evolved causation. But, if you’re willing to allow backward-time-evolved causation (aka retrocausation), then you can still restrict your time-evolved explanation to the past lightcones of the detection events (“timelike fashion”) anyway. So, you see that forward-time-evolved causation is out unless you want a preferred frame. The only way to maintain time-evolved causation is retrocausation. That’s why some have chosen another route, i.e., an adynamical block universe explanation. Accordingly, the most fundamental explanation for the correlation is a spatiotemporally global rule that, in this case, is just not amenable to forward-time-evolved and timelike storytelling (dynamism). It’s a very powerful explanatory method (as we show in our book) and quite consistent with the Lagrangian formalism of physics, i.e., all current physics is perfectly reasonable in the “God’s-eye view,” as Wilczek calls it. It’s our anthropocentric “ant’s-eye” bias that keeps us from using the more powerful adynamical (“acausal” as defined above) explanation in conjunction with our very successful physics.

But in standard QED there are no causally connected spacelike separated events possible, and this is by construction. I don't see any necessity to invent new theories, where the well-established relativistic spacetime and causality structure is violated. The correlations are indeed there due to the preparation of the polarization-entangled biphoton state, and there's no assumption about causal effects between the later measurements necessary. This becomes the more clear if you consider a real postselection setup, which is an extreme way of a delayed-choice experiment. Take Walborn's realization of the quantum eraser. You can do this experiment also by just making a measurement protocol of both detectors, using time marks of registration events. In the eraser setup you can then erase the which-way-information long after all photons are gone by just bringing both measurement protocols together and only select the signal photons for which also the idler has gone through the polarizer (oriented parallel to the one or the other quarter-wave plate in the slits), i.e., you can erase the WWI information even after the measurement is irreversibly fixed in the measurement protocols.

I guess, to understand what you are after with your alternative interpretation, I'd have to read your book.

 

[RUTA]

I think I will have to read your book (of which I will get a copy when released) to understand what you are saying.

Just a few points:
1. Right at the very foundations of ordinary QM is the Galilean transformations. So non-locality is built in from the start. You have to go to QFT to discuss locality and it is embodied in the cluster decomposition property:
https://www.physicsforums.com/threads/cluster-decomposition-in-qft.547574/

2. Note the exclusion in that property - correlations. EPR is just a correlation so IMHO is no mystery at all - the key thing is if you want to have CFD you need non-locality. Don't care about CFD - then no issue - we just have some different kind of statistical properties to those correlations than classically. That's hardly surprising because we also know QM as a generalized probability model is different to ordinary probability - ordinary probability is the simplest - QM the next simplest.

3. Regarding delayed choice QM explains it just fine:
http://quantum.phys.cmu.edu/CQT/chaps/cqt20.pdf

Thanks
Bill

Regarding point 1, here is an excerpt from our book:

Before continuing, we should point out that using a block universe interpretation of QM is supported by more than the Feynman path integral. (kaiser, BohrUlfbeck, anandan) all showed independently that the non-commutivity of the position and momentum operators in QM follows from the non-commutivity of the Lorentz boosts and spatial translations in SR, i.e., the relativity of simultaneity. Per Kaiser:
begin{quote}
For had we begun with Newtonian spacetime, we would have the Galilean group instead of [the restricted Poincar{\'e} group]. Since Galilean boosts commute with spatial translations (time being absolute), the brackets between the corresponding generators vanish, hence no canonical commutation relations (CCR)! In the [c $\rightarrow\infty$ limit of the Poincar{\'e} algebra], \textit{the CCR are a remnant of relativistic invariance where, due to the nonabsolute nature of simultaneity, spatial translations do not commute with pure Lorentz transformations} \cite[p. 706]{kaiser}. [Italics in original].
end{quote}

title ={Phase-space approach to relativistic quantum mechanics III: Quantization, relativity, localization and gauge freedom},
author ={Kaiser, G.},
journal ={Journal of Mathematical Physics},
volume ={22},
page ={705--714},
year ={1981}

title ={Laws, Symmetries, and Reality},
author ={Anandan, J.},
journal ={International Journal of Theoretical Physics},
volume ={42},
pages ={1943--1955},
year ={2003},
url ={https://arxiv.org/abs/quant-ph/0304109}

title ={Primary manifestation of symmetry. Origin of quantal indeterminacy},
author ={Bohr, A., and Ulfbeck, O.},
journal ={Reviews of Modern Physics},
volume ={67},
number ={},
pages ={1--35},
year ={1995},

I’m in a hurry to make a meeting, so I don’t have time to clean up the LaTeX more than that, but I know you’re interested in math, so I wanted to get that to you.

 

[vanhees71]

Regarding point 1, here is an excerpt from our book:

Before continuing, we should point out that using a block universe interpretation of QM is supported by more than the Feynman path integral. (kaiser, BohrUlfbeck, anandan) all showed independently that the non-commutivity of the position and momentum operators in QM follows from the non-commutivity of the Lorentz boosts and spatial translations in SR, i.e., the relativity of simultaneity. Per Kaiser:
begin{quote}
For had we begun with Newtonian spacetime, we would have the Galilean group instead of [the restricted Poincar{\'e} group]. Since Galilean boosts commute with spatial translations (time being absolute), the brackets between the corresponding generators vanish, hence no canonical commutation relations (CCR)! In the [c $\rightarrow\infty$ limit of the Poincar{\'e} algebra], \textit{the CCR are a remnant of relativistic invariance where, due to the nonabsolute nature of simultaneity, spatial translations do not commute with pure Lorentz transformations} \cite[p. 706]{kaiser}. [Italics in original].
end{quote}

Well, you can in full generality get position observables only for all kinds of massive particles or massless particles with spin $\leq 1/2$. Of course, you are right concerning the commutation relations of position and momentum, if position observables exist, because they are simply the Lie algebra of spatial translations, which are both part of the continuous parts of the Poincare (proper orthochronous Poincare) group and the Galilei group.

For non-relativistic particles, the massless case doesn't exist and thus you have always a position operator. The main difference between Poincare and Galilei group is the notion of mass: In the Poincare group it's defined as a Casimir operator via $p \cdot p=m^2$ (in natural units, $\hbar=c=1$) and for the Galilei group it's a non-trivial central charge of the corresponding Lie algebra. The case $m=0$ doesn't lead to physically sensible dynamics, as shown for the first time in the famous paper by Inönü and Wigner [1]. That's why the naive commutation relations between boosts and momentum operators are not applying, but you have the mass as a central charge, i.e., you have to lift the 10D classical Galilei group to the 11D quantum Galilei group, and then the commutators read
$$[\hat{K}_i,\hat{P}_j]=\mathrm{i} \delta_{ij} \mathbb{1}.$$
For a detailed analysis, see my QM manuscript (which however is in German)

http://theory.gsi.de/~vanhees/faq/quant/node74.html

or the textbook by Ballentine, which gives the same derivation in a bit different form.

[1] E. Inönü and E. P. Wigner, Representations of the Galilei group, Il Nuovo Cimento, 9 (1952), p. 705–718.
http://dx.doi.org/10.1007/BF02782239

 

[bhobba]

Regarding point 1

Its over my head - I will wait for the book.

But Ballentine showed, chapter 3, using Galilean transformations (see page 66) the dynamics of standard QM follow eg the form of the Schrodinger equation. This is hardly surprising because from the the path integral approach the PLA immediately follows and Landau - in Mechanics - showed exactly the same thing. Take the Hamiltonian used in hydrogen atom. Move the object and the force changes instantaneously. IMHO its part of the reason standard QM fails with things like spontaneous emission - its basis is wrong - or at least just approximate. Of course for many purposes an excellent approximation like classical mechanics itself - but still fundamentally wrong.

Thanks
Bill

 

[vanhees71]

Of course. @RUTA 's statement in #77 doesn't apply, because in QT a symmetry isn't realized by proper unitary transformations of the symmetry group but by ray representations, and the Galilei group has non-trivial ray representations, which are precisely the physically relevant ones, and the proper unitary representations do not lead to a physically interpretable particle dynamics (see my previous posting). You can of course always lift any ray representation of the symmetry group to unitary representation of a central extension of the original group, and that's precisely the 11 dimensional quantum Galilei group. The latter interpretation of Galilei invariance leads to the mass superselection rule of non-relativistic QT. If I remember right, Ballentine's book has all this in great detail.

 

[bhobba]

Of course. @RUTA If I remember right, Ballentine's book has all this in great detail.

You are right - it does. And I even got the book, found the key page and double checked. Commutation relation etc - the lot - follows from the Galilean transformations. One must go to QFT for locality to be a worry.

Of course it in no way falsifies, changes, or in anyway alters Bell - just adds a slight twist in interpretation.

Thanks
Bill

 

[vanhees71]

Sure, all the Bell tests today are completely understandable using standard QFT (mostly QED since the experiments are usually using photons). I don't see, what one should seek for an acausal theory as long as we can explain everything with the standard theory!

 

[Stephen Tashi]

Now let's stay on topic with this thread.

That would be a miracle. As far as topics go, the thread seems to be in a superposition.

In the delayed choice experiment on the surface it looks like something really weird is going on - but when analysed carefully that isn't he case:
http://quantum.phys.cmu.edu/CQT/chaps/cqt20.pdf

That's one chapter of an entire book http://quantum.phys.cmu.edu/CQT, which, in previous chapters develops an approach to QM not found in all textbooks. I found the review of the book by Faris http://math.arizona.edu/~faris/consis.pdf helpful.

So (on the topic of that book) there are at least two questions. 1) Is the general method advocated in the book good, correct, nice etc ? and 2) Does the general method advocated in the book successfully resolve the paradoxes it considers in chapter 20 ?

 

[bhobba]

Is the general method advocated in the book good, correct, nice etc

Yes - it resolves all known supposed paradoxes.

But if that isn't to your taste ie its within a specific interpretation, there are tons of papers explaining what's going on. This one concentrates on the so called separation fallacy:
http://philsci-archive.pitt.edu/10216/1/SeparationFallacy-rev.pdf
We have seen the same fallacy of interpretation in two-slit experiments, which-way interferometer experiments, polarization analyzers, and Stern-Gerlach experiments. The common element in all the cases is that there is some separation apparatus that puts a particle into a certain superposition of spatially-entangled eigenstates in such a manner that when an appropriately spatially-positioned detector induces a collapse to an eigenstate, then the detector will only register one of the eigenstates. The separation fallacy is that this is misinterpreted as showing that the particle was already in that eigenstate in that position as a result of the previous "separation." In fact the superposition evolves until some distinction is made that constitutes a measurement, and only then is the state reduced to an eigenstate. In general, when a measurement shows a specific eigenstate, it should not be assumed that the quantum system was already in that eigenstate (e.g., "goes through upper slit/arm or through lower slit/arm") prior to the measurement. The quantum erasers are more elaborate versions of these simpler experiments, and a similar separation fallacy arises in that context.

Thanks
Bill

 

[RUTA]

But in standard QED there are no causally connected spacelike separated events possible, and this is by construction. I don't see any necessity to invent new theories, where the well-established relativistic spacetime and causality structure is violated. The correlations are indeed there due to the preparation of the polarization-entangled biphoton state, and there's no assumption about causal effects between the later measurements necessary. This becomes the more clear if you consider a real postselection setup, which is an extreme way of a delayed-choice experiment. Take Walborn's realization of the quantum eraser. You can do this experiment also by just making a measurement protocol of both detectors, using time marks of registration events. In the eraser setup you can then erase the which-way-information long after all photons are gone by just bringing both measurement protocols together and only select the signal photons for which also the idler has gone through the polarizer (oriented parallel to the one or the other quarter-wave plate in the slits), i.e., you can erase the WWI information even after the measurement is irreversibly fixed in the measurement protocols.

I guess, to understand what you are after with your alternative interpretation, I'd have to read your book.

The analysis of a standard Bell-inequality-violating experiment requires only simple Hilbert space analysis (see Dehlinger, D., and Mitchell, M.W.: Entangled photons, nonlocality, and Bell inequalities in the undergraduate laboratory. American Journal of Physics 70(9), 903--910 (2002), for example), which must follow from any QFT in the proper limit (by definition). And, there is nothing wrong with the physics that fits the data so well in these experiments; it's not an issue with the physics. So, there is nothing in the formalism of QFT that resolves the ontological mysteries of quantum nonlocality -- you have to develop an ontology to resolve ontological mysteries.

 

[RUTA]

Of course. @RUTA 's statement in #77 doesn't apply, because in QT a symmetry isn't realized by proper unitary transformations of the symmetry group but by ray representations, and the Galilei group has non-trivial ray representations, which are precisely the physically relevant ones, and the proper unitary representations do not lead to a physically interpretable particle dynamics (see my previous posting). You can of course always lift any ray representation of the symmetry group to unitary representation of a central extension of the original group, and that's precisely the 11 dimensional quantum Galilei group. The latter interpretation of Galilei invariance leads to the mass superselection rule of non-relativistic QT. If I remember right, Ballentine's book has all this in great detail.

The statement follows from the results of the cited papers (all of which I have checked). Nothing you've said here refutes that work.

 

[Stephen Tashi]

there are tons of papers explaining what's going on. This one concentrates on the so called separation fallacy:
http://philsci-archive.pitt.edu/10216/1/SeparationFallacy-rev.pdf

If paper P1 explains paradox W one way and paper P2 explains paradox W a different way then has paradox W really been explained?

The paper linked above offers a conceptually simple explanation -given the we accept the idea of a Quantum state as simple. By contrast, what is Griffiths' consistent histories approach trying to explain?

The above paper deals with the fallacy of assuming that a device which creates a superposition of states also separates particles into being in one state or the other. It doesn't deal with a paradox claiming that the evolution of superposed quantum states after the particle exits the device is different or evolves differently in time depending on the event of a detector being used at a later time. Is that what Griffths is explaining?

 

[RUTA]

Sure, all the Bell tests today are completely understandable using standard QFT (mostly QED since the experiments are usually using photons). I don't see, what one should seek for an acausal theory as long as we can explain everything with the standard theory!

Again, there is nothing in the quantum physics that needs to be changed (as we explain in our book, QM and QFT don't need to be replaced, they work very well). The entangled states of the Hilbert space formalism that can be used so effectively to fit the data in Bell-inequality-violating experiments follow from QFT in the proper limit. There is no mystery in that, the physics works. On the flip side, the physics cannot resolve the ontological mysteries associated with quantum nonlocality. I use Zeilinger's delayed choice experiment to show students the mystery of quantum entanglement precisely because it requires no formalism to understand (Anton Zeilinger, ``Why the quantum? `It' from `bit'? A participatory universe? Three far-reaching challenges from John Archibald Wheeler and their relation to experiment,'' in Science and Ultimate Reality: Quantum Theory, Cosmology and Complexity, John D. Barrow, Paul C.W. Davies and Charles L. Harper, Jr. (eds.), (Cambridge Univ Press, Cambridge, 2004), pp 201-220). In that case, it is obvious that the measurement setting the experimentalist controls is associated with the later measurement outcome. So, could the experimentalist's (delayed) choice have been otherwise? That's the mystery. No formalism dispels that mystery (on the contrary, the formalism predicts the outcomes that generate it!). The mystery has to do with ontology and ontology is always underdetermined by the formalism. Thus, more sophisticated experiments have not reduced the number of QM interpretations, indeed they have proliferated. Everyone picks their personal favorite and continues doing physics accordingly, i.e., everyone is using the same formalism and successfully fitting the same data, but offering different explanations afterwards. That's fine for the status quo, the book argues that to advance physics from the status quo (e.g., obtain quantum gravity and explain dark energy and dark matter data) we should consider moving to the "God's-eye view" per Wilczek. It's that simple.

 

[bhobba]

If paper P1 explains paradox W one way and paper P2 explains paradox W a different way then has paradox W really been explained?

Why would it not. We can use Newtons laws or the PLA to solve mechanics problems. Is that an issue as well?

Griffiths concentrates on Histories which is what Consistent Histories is all about. The other doesn't - but I think on separability. Many ways to skin a cat. Griffiths is the more complex answer because of that:
Let us summarize the results obtained by using a quantum coin and studying various consistent families related to the counterfactual statement of the delayed choice paradox. We have looked at three different frameworks, (20.22), (20.23), and (20.24), and found that they give somewhat different answers to the question of what would have happened if the beam splitter had been left in place, when what actually happened was that the photon was detected in E with the beam splitter out. (Such a multiplicity of answers is typical of quantum and—to a lesser degree—classical stochastic counterfactual questions; see Sec. 19.4.) In the end, none of the frameworks supports the original paradox, but each framework evades it for a somewhat different reason. Thus (20.22) does not have photon states localized in the arms of the interferometer, (20.23) has such states, but they cannot be used as a pivot for the counterfactual argument, and remedying this last problem by using (20.24) results in the counterfactual question being answered in terms of MQS states, which were certainly not in view in the original statement of the paradox.

Sometimes Newtons laws make mechanics problems easier to solve - sometime the PLA. Here Consistent Histories is the more complex actually giving slightly different solutions depending on the framework chosen.

Just goes to show it's probably a good idea to understand a number of different interpretations.

Thanks
Bill

 

[zonde]

The problem with that is SR tells us the order of spacelike separated events is frame dependent

You attach physical meaning to convention. You can not blame reality for not respecting particular set of conventions.

 

[zonde]

But in standard QED there are no causally connected spacelike separated events possible, and this is by construction.

How can QED say this is impossible when QED does not speak about individual events? Or I am missing something? In QT's minimal interpretation as given by Ballentine statistical nature of predictions is clearly stated. Is it somehow different in QED? Can QED go beyond that statistical nature and say something about individual realizations of experiment (except in some special cases where predicted probabilities approach 1)?

 

[vanhees71]

The commutator of local observables vanishes at space-like separated arguments identically. That's why actions at a distance (i.e., over space-like distances in Minkowski space) have probability 0, i.e., they don't happen with certainty!

 

[vanhees71]

The mystery has to do with ontology and ontology is always underdetermined by the formalism. Thus, more sophisticated experiments have not reduced the number of QM interpretations, indeed they have proliferated. Everyone picks their personal favorite and continues doing physics accordingly, i.e., everyone is using the same formalism and successfully fitting the same data, but offering different explanations afterwards. That's fine for the status quo, the book argues that to advance physics from the status quo (e.g., obtain quantum gravity and explain dark energy and dark matter data) we should consider moving to the "God's-eye view" per Wilczek. It's that simple.

For me the only mystery is why this is considered a mystery since the formalism tells us precisely what to expect, and that's what's empirically confirmed with high significance. You cannot expect more from any natural-science theory. It's about observable objective facts not some mysterious "ontology behind the phenomena".

 

[vanhees71]

The statement follows from the results of the cited papers (all of which I have checked). Nothing you've said here refutes that work.

The statement that the generators of boosts commute with the momentum operators (the generators of spatial translations) is wrong within standard non-relativistic QM, as shown already by Inönü and Wigner. For non-relativistic particles a vanishing mass doesn't make sense, because it doesn't lead to a physically interpretable quantum dynamics. For non-0 mass the commutator is
$$[\hat{K}_j,\hat{p}_k]=\mathrm{i} \delta_{jk} m \hat{1} \neq 0.$$
The group-representation theoretical reasons for this are quite obvious (see posting #78).

 

[Lord Jestocost]

The correlations are indeed there due to the preparation of the polarization-entangled biphoton state, and there's no assumption about causal effects between the later measurements necessary.

Maybe, as an experimental physicist, I don’t get the point when reasoning about “due to the preparation“. Does that mean that each entangled photon pair emerges from the source with, in effect, a set of pre-programmed, no longer modifiable, instructions for how each photon of the pair has to respond to a measurement on one of its observables?

 

[vanhees71]

Indeed, "due to preparation" means that the photon pairs have been created (today usually using parametric downconversion) in the polarization-entangled state. With appropriate "optical elements" you can produce the needed Bell states. I don't know, what you mean by "pre-programmed". The meaning of the state is very clearly given by Born's rule, and I don't know what you mean by "no longer modifiable". Of course, when the photons interact with something later, the state is modified. E.g., when the photon hits a detector usually it's absorbed and thus you have a state with this photon gone. So the state clearly changes through interactions. There are also no "instructions for how each photon of the pair has to respond to a measurement on one of its observables". The state of the single photon in the entangled pair is given according to the standard rules of QT by tracing out the other photon, i.e.,
$$\hat{\rho}_A=\mathrm{Tr}_{B} \hat{\rho}_{AB}=\sum_{j} |ij\rangle \langle ij |\hat{\rho}_{AB}|\langle kj \rangle \langle kj|.$$
For the "singlet state" you have
$$\hat{\rho}_{AB}=|\Psi \rangle \langle \Psi|, \quad |\Psi \rangle=\frac{1}{\sqrt{2}} (|HV \rangle-|VH \rangle),$$
and
$$\hat{\rho}_A=\frac{1}{2} \hat{1}.$$
I.e., the single photons in the entangled pair each are totally unpolarized, i.e., a polarizer in any direction let's the photon through with probability 1/2 and absorbs it with probability 1/2. There's nothing more known about the single photons than that.

 

[RUTA]

For me the only mystery is why this is considered a mystery since the formalism tells us precisely what to expect, and that's what's empirically confirmed with high significance. You cannot expect more from any natural-science theory. It's about observable objective facts not some mysterious "ontology behind the phenomena".

Do you understand why those who do wonder about the nature of reality (ontology) are mystified? Or is that why you're participating in such threads?

 

[RUTA]

The statement that the generators of boosts commute with the momentum operators (the generators of spatial translations) is wrong within standard non-relativistic QM, as shown already by Inönü and Wigner. For non-relativistic particles a vanishing mass doesn't make sense, because it doesn't lead to a physically interpretable quantum dynamics. For non-0 mass the commutator is
$$[\hat{K}_j,\hat{p}_k]=\mathrm{i} \delta_{jk} m \hat{1} \neq 0.$$
The group-representation theoretical reasons for this are quite obvious (see posting #78).

You're making the point for me. Clearly "Galilean boosts commute with spatial translations (time being absolute)" so in order to get the proper commutation relationship for QM, you need boosts that don't commute with spatial translations and for that you have to go to the Poincare group. That's Kaiser's point and yours.

 

[RUTA]

You attach physical meaning to convention. You can not blame reality for not respecting particular set of conventions.

The statement you quoted here is true and doesn't entail any interpretation. What you choose to believe ontologically about it can and does differ between different interpretations of QM.

 

[vanhees71]

You're making the point for me. Clearly "Galilean boosts commute with spatial translations (time being absolute)" so in order to get the proper commutation relationship for QM, you need boosts that don't commute with spatial translations and for that you have to go to the Poincare group. That's Kaiser's point and yours.

You don't need the Poincare group to get this result. You can get it from a careful analysis of the unitary ray representations of the Galilei group. Of course, it's equivalent to use the appropriate "deformation" of the Poincare group. I guess, I simply misunderstood your statement, and I think all in all we agree in this matter :-).