Entanglement swapping and Bohmian mechanics

[martinbn]

If you have a |V> photon stream (say from a laser) that you can only measure on the $\{|+\rangle, |-\rangle\}$ basis, yes, that is meaningful.

No, I meant what I said. You have a qubit (the physical realisation is not important, but if you like you can think of it as polarization of photons), and you can measure in both basis.

I would not call them the same state though. Naturally, I can distinguish them experimentally using a suitably oriented PBS.

Similarly, I can distinguish the difference between Ma's (1) and (2) experimentally. As they do. Megidish as well. Disagree?

Yes, it is exactly the same as in Ma's paper, just a simpler case, no swapping involved, so that the trees are not in the way to see the woods. And, yes, I disagree. On this you are wrong! It is very basic, your confusion is not even in the basic QM. It is much simpler it is writing the same vector in two different basis. But all this has been said many times. It is your job to clear this for yourself. Untill then you will not understand Ma's paper.

 

[martinbn]

In the actual experiments, it is not done as you describe/imagine at all.

I know. My point was to consider what I described instead of the actual experiment.

So... no. Of course the selection process does not create entanglement. It is the physical overlap, and nothing else, that varies. No more than a post-selection filter causes cancer (or the appearance thereof) in medical studies.

But your use of the word entanglement is prisicely that. You said it yourself in the other thread, entanglement = non-classical correlations. Why is there no entanglement in this case (the one that I described, not the actual experiment)?

 

[Morbert]

Sure.

Great. So do you accept Huggett when he says (1.3), decomposed using the Bell basis, is (1.4)?

 

[Demystifier]

John Cramer's transactional interpretation is NOT an interpretation. It is not a theory either. It is a story or a narrative, intended as a complement to Copenhagen.

I'm glad that someone pointed this out explicitly.

 

[DrChinese]

Suppose that in the swap set up, nothing is done to photons 2 and 3. But the usual measurments are done on 1 and 4. This goes for all the runs. Then in the data of the results of the measurements on 1 and 4 you can find a subset of runs which show non-classical correlations. According to you that would mean that post-selection can create entanglement just as the swap can. Right?

In your particular example, since you asked again in #512:

No, there are certainly no identifiable fourfold coincidence subsets featuring non-classical correlations. Unless of course, you plan to cherry pick by hand.

No ticket, no laundry. There must be physical overlap for entanglement and non classical correlations.

 

[martinbn]

In your particular example, since you asked again in #512:

No, there are certainly no identifiable fourfold coincidence subsets featuring non-classical correlations. Unless of course, you plan to cherry pick by hand.

Yes, I search through the data and find such a subset of runs. I don't even have to do it, we know there is suh subset. The point is that it will show non classical corelations, and for you that means entanglembent.

 

[DrChinese]

Yes, I search through the data and find such a subset of runs. I don't even have to do it, we know there is such subset. The point is that it will show non classical correlations…

Exactly how do propose to define criteria for this hypothetical subset? Other than selecting by hand…

 

[martinbn]

Exactly how do propose to define criteria for this hypothetical subset? Other than selecting by hand…

What difference does it make? I don't even need to find it, we know it exists. The point is that according to you entanglement means non-classical corelations. Hence there is entanglement here as well, at least if you are consistent.

 

[DrChinese]

What difference does it make? I don't even need to find it, we know it exists. The point is that according to you entanglement means non-classical correlations.

Huh? Please, quote me from an experiment. “We” don’t know it exists.

Product State experiments do not yield Entangled State statistics. You’ve made a mistake in your understanding somewhere.

 

[Demystifier]

@DrChinese I would really appreciate if you could answer my question in https://www.physicsforums.com/threa...ng-and-bohmian-mechanics.1086070/post-7324074
It would help me to better understand your point of view.

 

[martinbn]

Huh? Please, quote me from an experiment. “We” don’t know it exists.

Product State experiments do not yield Entangled State statistics. You’ve made a mistake in your understanding somewhere.

I don't know how to explain it any clearier. Here is one more try. Consider the standard entanglement swap protocol. Two pairs 1&2 and 3&4 of entangled qubits are prepared, 2 and 3 are send to Victor, 1 to Alice, and 4 to Bob. Alice and Bob do the usual measurments, Victor can do a BSM or do nothing. We run the experiment a billion times were Victor does the measument, and a billion times where he doesn't. The results of 1 and 4 are statisticaly indistinquishable in each set of billion runs. In the fist case you can partition the results of 1&4 into subsets depending on the results of Victors measurments. The results of 1 and 4 in these subsets will show the Bell state non-classical correlations. In the second case where Victor did nothing it is not easy to do the partitioning, but it exists, this is just basic QM because the density matrix of the combined system of 1 and 4 is the same no matter what Victor does. So there is definitely such subsets in the second billion runs too.

Now, I am not expresing an opinion on what is going on. I am just askining why you are incosistent. According to you entanglement = non-classical correlations. You also say that in the first case we have an entanglement of 1 and 4. Then why do you say that there is no entanglement in the second case!!!

 

[Demystifier]

Now, I am not expresing an opinion on what is going on. I am just askining why you are incosistent. According to you entanglement = non-classical correlations. You also say that in the first case we have an entanglement of 1 and 4. Then why do you say that there is no entanglement in the second case!!!

Because by correlations he means observed correlations.

 

[martinbn]

Because by correlations he means observed correlations.

What does this mean? How are they different in the two cases that i discribed?

 

[DrChinese]

Because by correlations he means observed correlations.

@martinbn "where Victor did nothing it is not easy to do the partitioning, but it exists, this is just basic QM"
We’re talking about 4-fold coincidence count, where all 4 are in the time window. Ma did that experiment already, and demonstrated there were no non classical stats for Product States. That was one of the primary objectives of the experiment. That was what was observed and reported.

We all agree that if you look at only 2-fold coincidences on photons 1 and 4, the results are purely random. In that case, the BSM is irrelevant.

 

[DrChinese]

@DrChinese I would really appreciate if you could answer my question in https://www.physicsforums.com/threa...ng-and-bohmian-mechanics.1086070/post-7324074
It would help me to better understand your point of view.

Sure.

I sorta skipped by it because I don't think my answer will be very satisfying to you. I also question its relevance to the Bohmian view of DCES. But, oh well... Working on it now.

 

[DrChinese]

OK, I'm glad that you accept that it is so according to (my version of) Bohmian Mechanics (BM). But what I try to emphasize is that such a view is not an exclusive feature of BM. Instead, that's a feature of any interpretation that accepts the following assumptions:
1. Universality: Everything, including the lab and the scientists, is ultimately described by QM.
2. Objectivity: The quantum state of the system is not merely a description of knowledge of an observer, but represents some objective feature of the physical system not depending on the observer's knowledge.
3. Unitarity: The time evolution of the quantum state (of a closed system) is unitary, i.e., at the fundamental level, there is no collapse of the state.

Of course, it is not trivial to satisfy all these requirements in a single interpretation. BM satisfies them all, and so does the many-world interpretation, but many interpretations don't. Also note that the Bohmian point of view is "more objective" in the sense that it satisfies the objectivity assumption 2., not in the sense that it is "more right" than some other interpretation.

At the end, I have a question for you: Which of the assumptions 1, 2 and 3 do you find implausible or unacceptable? (It seems to me that you accept 1. and 2., but deny 3. But I'm not sure, so I would like if you could be explicit about it.)

None of these seem to have anything to do with DCES or BM specifically, and they seem quite general. And I agree that many elements of your explanations are not exclusive to BM, and I would certainly expect that.

1. I'm not really sure what this is supposed to mean. I guess I agree the scientist+lab operates under all of the laws of physics, including QM. But the quantum state of the scientist+lab is to be ignored in experiments. Excluding experiments involving human behavior, 99.9% of all scientific experiments ignore the state of the the scientist (a human) completely.
2. I don't tend to align with psi-epistemic interpretations. I consider PBR to be a strong case against them; and yes, I am aware of the limits of PBR. I am certainly open to demonstrations in favor of psi-epistemic interpretations. And I follow your idea that BM is "objective" in the sense you intend.
3. I would say I follow orthodox views on evolution of the quantum state. I don't think that precludes some variations on collapse either way, specifically considering Swapping experiments*. Specifically: I can have 2 entangled particles that never coexist or share any common lightcone. Are they two closed systems, or one? The answer might tend to vary. Does a measurement on one affect the other in a manner that might affect Unitarity? I can't say. If it did, it could not be instantaneous, could it?

These are my thoughts, but I'm not pushing them in any way, shape or form. I am interested in your thoughts. You say BM satisfies all 3 you list, then fine. No argument from me.


*Huggett says: "But whatever the outcome, particles 2 and 4 will end up in a Bell state, and hence be entangled. The measurement of 1 and 3 alone induces the entanglement of 2 and 4; no interaction, either unitary or involving their measurement, occurs between 2 and 4." This is the kind of thing I'm thinking about in my somewhat waffling answer.

 

[Demystifier]

What does this mean? How are they different in the two cases that i discribed?

In your second case, if I understood it correctly, the correlations are implicitly there, but nobody explicitly measures them. From an experimental point of view, that's the same as if there are no correlations at all.

 

[Demystifier]

1. I'm not really sure what this is supposed to mean. I guess I agree the scientist+lab operates under all of the laws of physics, including QM. But the quantum state of the scientist+lab is to be ignored in experiments. Excluding experiments involving human behavior, 99.9% of all scientific experiments ignore the state of the the scientist (a human) completely.

That points to a deep, but often too deep to be clearly recognized, difference between the experimental and the theoretical way of thinking. When you do experiment, you can indeed ignore such things. But when you attempt to explain what is going on in an experiment, then you are no longer doing the experiment, you are dealing with the theory. (Here "theory" is meant as an opposite to experiment, so different interpretations with the same measurable predictions are viewed as different "theories".) The theory often contains elements that cannot directly be measured, but must be taken into account in a theoretical explanation. One such thing, whether you like it or not, is the quantum state of the scientist+lab. I claim that many quantum experiments, including the DCES, cannot be properly understood without taking the quantum state of scientist+lab into account. I can understand that this concept does not make much sense to you from an experimental point of view, but to explain the experiment, one must step out from the experiment itself and think of it from an outside perspective that includes some theoretical ideas that an experimentalist naturally ignores.

2. I don't tend to align with psi-epistemic interpretations. I consider PBR to be a strong case against them; and yes, I am aware of the limits of PBR. I am certainly open to demonstrations in favor of psi-epistemic interpretations. And I follow your idea that BM is "objective" in the sense you intend.

Good, so we basically agree on this part and there is no need for additional comments.

3. I would say I follow orthodox views on evolution of the quantum state. I don't think that precludes some variations on collapse either way, specifically considering Swapping experiments*. Specifically: I can have 2 entangled particles that never coexist or share any common lightcone. Are they two closed systems, or one? The answer might tend to vary. Does a measurement on one affect the other in a manner that might affect Unitarity? I can't say. If it did, it could not be instantaneous, could it?

You seem to be pretty unsure what a collapse is, why we need it, and how we can avoid it. But such questions are crucial for understanding any quantum experiment, including the DCES. You can't hope to understand DCES before you develop your own strong opinion about the collapse.

The upshot of all this is what I said before. You think like an experimentalist who wants the understand DCES in its own terms, without any interpretational prejudices. While it may look like a right way to do science, I don't think it's efficient. As Thomas Kuhn (a famous philosopher of science) pointed out, a scientist cannot think without some paradigm, namely, a theoretical framework with some a priori assumptions and prejudices. You seem to be attempting to understand DCES with a minimal set of assumptions and prejudices, and that's why you are confused. You must extend your set of assumptions and prejudices, namely, accept one specific interpretation of QM as your working paradigm, and then explain DCES within that paradigm. You cannot remain agnostic about the notion of collapse, or about the notion of the state of scientist+lab, if you want to have an explanation of DCES that makes sense to you. Experiments establish facts, but explanations are provided by theories. You cannot explain anything in science without understanding and accepting some theory.

 

[Matterwave]

You think like an experimentalist who wants the understand DCES in its own terms, without any interpretational prejudices. While it may look like a right way to do science, I don't think it's efficient.

If you hardcore "shut up and calculate" enough then there is also no mystery or paradox. You just follow the mathematical formalism of QM to the end. Peres himself (for other, perhaps new, readers of the thread: Peres wrote "Delayed Choice for Entanglement Swapping" and "Quantum Theory Needs No 'Interpretation'") finds no paradox in DCES:

... if we attempt to attribute an objective meaning to the quantum state of a single system, curious paradoxes appear:...

In summary, there is nothing paradoxical in the experiments outlined above. However, one has to clearly understand quantum mechanics and to firmly believe in its correctness to see that there is no paradox.

I guess I wouldn't use the word efficient as you used it. An interpretation can elucidate a deeper understanding. But if all you want is efficiency, what more can you ask for than to shut up and calculate?

To be clear, I am with you in terms of wanting a deeper understanding than the purely operationalist / instrumentalist view. But efficiency is perhaps not the right hook.

 

[martinbn]

In your second case, if I understood it correctly, the correlations are implicitly there, but nobody explicitly measures them. From an experimental point of view, that's the same as if there are no correlations at all.

What do you mean by measuring correltions? For the point i made all that matters is that they are there. But @DrChinese doen't read what i write or he doesn't understand it. In any case he doesn't see his inconsistansy.

By the way, I looked at Huggett's paper. It seems that the analysis he gives is the orthodox explantion plus BM explanation of collapse. Which means that swapping is no different than any other QM experiment.

 

[Demystifier]

I guess I wouldn't use the word efficient as you used it. An interpretation can elucidate a deeper understanding. But if all you want is efficiency, what more can you ask for than to shut up and calculate?

Right, the word "efficient" may be misleading, so let me rephrase my point. The shut-up-and-calculate is a theory too, and this theory makes correct predictions for the DCES experiment, so it is a legitimate explanation of the experiment, provided that one finds the shut-up-and-calculate theory satisfying. But @DrChinese obviously doesn't find it satisfying, so he is looking for a more detailed explanation, namely for some specific interpretation. And yet, it looks as if he also wants to remain agnostic about all specific interpretations that exist. That's inconsistent, if his goal is to find a convincing interpretation, he cannot achieve that (it's not "efficient") if he is also agnostic about the interpretations.

 

[javisot]

Right, the word "efficient" may be misleading, so let me rephrase my point. The shut-up-and-calculate is a theory too, and this theory makes correct predictions for the DCES experiment, so it is a legitimate explanation of the experiment, provided that one finds the shut-up-and-calculate theory satisfying. But @DrChinese obviously doesn't find it satisfying, so he is looking for a more detailed explanation, namely for some specific interpretation. And yet, it looks as if he also wants to remain agnostic about all specific interpretations that exist. That's inconsistent, if his goal is to find a convincing interpretation, he cannot achieve that (it's not "efficient") if he is also agnostic about the interpretations.

Following the analogy of languages and interpretations, what does "shut up and calculate" mean to you?

 

[Morbert]

But @DrChinese obviously doesn't find it satisfying, so he is looking for a more detailed explanation

@DrChinese's mistake is more fundamental. He does not believe QM can actually reproduce the experiment without invoking some retrocausality because he does not understand the forward-in-time calculations.

 

[Demystifier]

Following the analogy of languages and interpretations, what does "shut up and calculate" mean to you?

It this analogy, the shut-up-and-calculate would be the English language that anyone should understand.

 

[DrChinese]

What do you mean by measuring correltions? For the point i made all that matters is that they are there. But @DrChinese doen't read what i write or he doesn't understand it. In any case he doesn't see his inconsistansy.

By the way, I looked at Huggett's paper. It seems that the analysis he gives is the orthodox explantion plus BM explanation of collapse. Which means that swapping is no different than any other QM experiment.

There are no identifiable correlations in product state experiments that match entangled state correlations. Thousands of experiments come to the same conclusion. A CHSH value of S>2 is usually considered evidence of entanglement. Product states will always be less than 2.

I would mostly agree with you about Huggett’s paper. He also speaks extensively about particles having a definite position at all times.

As to swapping being no different than other types of experiments: there are plenty of complexities in DCES. Clearly, it touches the bounds of our understanding of the quantum world. Anyone is free to simply hand wave and say there’s nothing more to be understood.

 

[DrChinese]

Right, the word "efficient" may be misleading, so let me rephrase my point. The shut-up-and-calculate is a theory too, and this theory makes correct predictions for the DCES experiment, so it is a legitimate explanation of the experiment, provided that one finds the shut-up-and-calculate theory satisfying. But @DrChinese obviously doesn't find it satisfying, so he is looking for a more detailed explanation, namely for some specific interpretation. And yet, it looks as if he also wants to remain agnostic about all specific interpretations that exist. That's inconsistent, if his goal is to find a convincing interpretation, he cannot achieve that (it's not "efficient") if he is also agnostic about the interpretations.

You’re correct, I don’t find it satisfying. Not the shut up and calculate side in particular. Mainly it bothers me that various common statements about various interpretations are subject to conflict with experiment. Of course, IMHO.

I will be posting later today in the other thread, asking about some specific issues. These will relate to the Huggett paper, which looks better and better the more I study it. However, it does not begin to address my questions about Bohmian views on both general entanglement or swapping.

 

[DrChinese]

@DrChinese's mistake is more fundamental. He does not believe QM can actually reproduce the experiment without invoking some retrocausality because he does not understand the forward-in-time calculations.

I understand them just fine. Mjelva’s paper opened my eyes to some of the conflicts between experiment and various interpretations. Your arguments essentially follow the same line as his. Of course, he also makes several other critical mistakes, not directly present in your mathematical development.

Delayed-choice entanglement swapping experiments: No evidence for timelike entanglement

I am not pushing retrocausality. Huw Price and Ken Wharton published a number of works attempting to summarize interpretations along that line. I’m not sure such interpretations solve all the problems either.

A New Class of Retrocausal Models
Bell's Theorem and Locally-Mediated Reformulations of Quantum Mechanics

In this thread, we are focused on BM - and I don’t want us to stray from that. I included these references for anyone interested in venturing into other territory.

 

[martinbn]

There are no identifiable correlations in product state experiments that match entangled state correlations. Thousands of experiments come to the same conclusion. A CHSH value of S>2 is usually considered evidence of entanglement. Product states will always be less than 2.

You keep ignoring the part about the subsets. The full set has no entnglement, swap or no swap. The subsets do.

I would mostly agree with you about Huggett’s paper. He also speaks extensively about particles having a definite position at all times.

Of course, he uses BM.

As to swapping being no different than other types of experiments: there are plenty of complexities in DCES. Clearly, it touches the bounds of our understanding of the quantum world. Anyone is free to simply hand wave and say there’s nothing more to be understood.

 

[DrChinese]

You keep ignoring the part about the subsets. The full set has no entanglement, swap or no swap. The subsets do.

1. Your understanding of Entanglement Swapping is incorrect. When Swapping is enabled via BSM on photons 2/3, each and every 1/4 pair becomes entangled in one of 4 Bell states. Using current technology: Only half (or so) can be identified as belonging to a specific subset, and those (and only those) demonstrate Entangled State correlations. Those subsets are identified using specific criteria chosen in advance.

2. When Swapping is NOT enabled on photons 2/3, there is no entanglement of any 1/4 pairs. Using the same criteria as for identifying the Entangled subsets, there is no correlation at all.

3. Also when Swapping is NOT enabled on photons 2/3 : There is absolutely no objective criteria under which non-classical correlations will appear. Product State experiments produce Product State statistics. I.e. there are no identifiable subsets of the type you are attempting to describe.

You keep posting as if what you say is so obvious, it needs no support. If you are correct, a suitable quote should be easy to locate. OTOH: Megidish: "The first and last photons, that did not share between them any correlations, become entangled." See their Fig. 3. "In order to fully characterize the first and last photons’ state, a quantum state tomography (QST) procedure is required."

 

[DrChinese]

This is a general comment to any reader here: I am NOT pushing any interpretation(s). I am NOT trying to bash Bohmian Mechanics. And I am NOT saying orthodox QM answers these questions more than BM does.

However: There are what I consider to be consequences to Bohmian hypotheses when applied to Delayed Choice Entanglement Swapping. Some of these - to me anyway - seem to be obvious, and seem to lead to contradictions. One glaring obvious example is how a theory featuring instantaneous action at a distance* - but not action across time - can explain the Megidish results**. I realize Bohmians are applying concepts that superficially explain this; and sure, they look fine as long as you don't question the application of various concepts. But that is precisely why I am asking various questions. It's the questions not being usually asked where I see conflicts. But as posters are helping to understand the underlying nature of BM, I hope to answer these questions.

I'm only saying this because some posters are pointing out - correctly I might add - that some of the BM concepts we are discussing are present either in orthodox QM or in other popular interpretations. Yes, I get that and accept that. But there are differences too, which makes BM unique.

"Thank you for your attention to this matter." -Author Unknown

-DrC

*A theory featuring instantaneous action at a distance has (or should have) a huge advantage in explaining Quantum Nonlocality.
**In which detection order is flipped around. No need to answer this, I am just giving an example. It's just a rhetorical point.