Delayed Choice Quantum Erasers

[Qwertz]

I've now read several descriptions of the delayed 'choice' quantum eraser, the modified double-slit experiment which (allegedly) demonstrates reverse causation, and I'm still having difficulty understanding the setup and procedure. Is anyone familiar enough with these buggers to help explain the experiment to me? The interpretation (at the second link above) that observing one particle alters the past observation of the other is just silly, for many reasons, and I'd like to at least know why so many people insist on interpreting the experiment in this way.

-Q

[Eric Dennis]

Your reaction is of course correct. This kind of experiment does not demonstrate any backward-time causation. In short, what's actually happening is that the measurement of photon 1 at D0 is influencing which detector (D1, D2, D3, or D4) photon 2 is more likely to trigger.

One has to be careful when reading papers that touch on the foundations of quantum mechanics, even papers about experiments actually performed, because there is much confusion within the physics community itself on the matter. This delayed choice eraser business is not the only place such issues pop up (see e.g. my comment on a similarly confused Zeilinger paper). Also this isn't really a delayed *choice* experiment either: there is no choice made by the experimenter, as in Bell Inequality experiments.

The backward causation idea is not even justifiable in the usual interpretation of qm, which is itself highly problematic for other reasons. It's just some extra ideological baggage thrown on top of a legitimate experiment to try to make it sound sexy. (Unfortunately, among some guys in this field, "sexy" means bizarre, counter-intuitive, even self-contradictory. Even more unfortunately, "some guys" includes Bohr and Heisenberg and others whom they won over.)

Here's a little more detail on the experiment. First notice that if you sum up the calculated interference patterns corresponding to double detections in D1 and D2 you get back the flat, no-interference pattern. Also if you add in both of the remaining (double) detected intensities associated with D3 and D4, you are still left with a flat pattern. And *this* summed intensity is what is actually detected at D0. Decomposing the measured intensity into two cancelling interference patterns is something of a theoretical afterthought.

Also keep in mind that the intensity pattern (whether flat or interfering) is not something that is actually measured for *each* photon pair. The only thing that's measured for each photon 1 is a single position -- the position at which it registered at D0. It's only after accumulating tons and tons of entirely separate runs, with a new photon 1 and 2 each time, that one measures the full intensity pattern.

Thus the decomposition amounts to a *post-selection* of the measured intensity in such a way as to artificially produce an interference pattern. Again this is all pretty much analogous to the above Zeilinger experiment, which is perhaps simpler to understand (it's essentially a discretized version of the current experiment in which the polarization of another pair of correlated photons assumes the role of detected position here). I should mention that if you do the pure eraser expeiment (no delayed "choice", so that D0 occurs only after D1,2,3,4), then there is real interference and my criticism doesnt apply, but then there's no insinuation of backward causation either.

Notice that no one who makes these snide comments about backward causation ever really spells out a coherent (no pun intended) sequence of events, say as captured by the evolution of the relevant quantum state, that's supposed to exhibit it. It's all wishy washy talk to sex up what is eminently comprehensible in a forward time analysis.

[EC]

Here's a link to a site that explains what is really going on in these experiments in relatively non-technical terms. I also recommend going right to the source and reading Dr. Little's paper too. It's also pretty basic to follow if you have a decent understanding of QM.

http://www.physics.prodos.org/stephenspeic...ains/index.html

[Eric Dennis]

Here's a link to a site that explains what is really going on in these experiments in relatively non-technical terms. I also recommend going right to the source and reading Dr. Little's paper too. It's also pretty basic to follow if you have a decent understanding of QM.

http://www.physics.prodos.org/stephenspeic...ains/index.html

I'm afraid that document won't be of any value in finding out what's really going on in these or any other experiments because Little's theory ("TEW") is just wrong, and demonstrably so. See my synopsis of the whole sorted history at

http://www.objectivescience.com/articles/e..._dissidents.htm

Moreover, that document you link doesn't say anything about quantum eraser experiments.

[EC]

Little's theory is not "just wrong" but I will check out your site, just not with an "open mind".

You are correct, that it doesn't directly speak of "quantum eraser" experiments but using the principles of the theory and the analysis of the dual-delayed double slit experiments that it does speak of on can easily deduce what transpires in the "quantum eraser" experiments

Edited April 24, 2006 by EC

[Qwertz]

My problem isn't quite yet one of understanding the results of these experiments, but of understanding the methodology - none of the diagrams and descriptions are really making it clear even how the experiments are set up and run.

EC: I've read Little's paper on TEW before, but without a greater understanding of how quantum erasers work, I can't apply it to this situation yet. Also, as I understand it, TEW violates Bell's Theorem, which has been experimentally confirmed. Though as I'm saying, I'm not too clear about the experimental side of this thing, so...

I've witnessed flame wars over TEW, and I don't want to start one. I'm trying to understand the experiments first, not the theories which explain them.

Also, I stumbled across this:

http://arxiv.org/ftp/quant-ph/papers/0303/0303036.pdf

today, which seems to address the results, but doesn't clearly explain the methodology, either.

So far, my understanding of the setup is this: I make a photon and throw it at the slits. After it passes through the slits, I make a copy of it and send the copy off in another direction, while the original continues on course to the detectors. The second one is sent off to another set of detectors some greater distance away, so that these detectors detect the copy some time after the original has been detected. What I have yet to grasp is what is experimentally observed?

-Q

Edited April 24, 2006 by Qwertz

[EC]

Bell's theorem is confirmed experimentally. That does not mean that the theory that explains the theory is correct. How can ANYONE on this forum argue against causality or identity? Is Little's theory complete? NO, but it MUST be correct within it's limitation's if causality and identity are to be preserved. Anyone that argues to the contrary is wrong or else Objectivism is wrong. I'm so SICK and tired of people here offering their attacks on O'ism in some half-hidden way just so they can get around violating the forums rules. I can nearly ALWAYS see right through these indirect attacks and they are REALLY beginning to piss me off. Bring 'em to me first and don't try to pull some slick BS on us.

[Eric Dennis]

Qwertz: You've got it half right (almost). You have two atoms A and B that are both excited by a laser. These atoms are going to simultaneously produce two correlated photons, moving in opposite directions. The following is important to understand: it's not that each atom produces one photon, it's that the pair of photons are produced in a unitary event (no pun!) by both atoms together.

Now photon 1 goes through the slits and is detected at D0. Photon 2 goes backward towards some beam splitters and then additional detectors (D1,2,3,4). Really, each of these "photons" is described (at least) by a wavefunction which is extended over space and can do things like go through both slits or beam splitters simultaneously and recollect on the other side.

Before photon 2 gets anywhere near any beam splitter or detector, photon 1 is already detected at D0. After that, what we have is photon 2 spreading out through all the beam splitters (partially transmitted, partially reflected through each one) and then impinging on all the detectors D1,2,3,4. One of these four detectors clicks, registering photon 2.

This is repeated a billion times and we build up two records: (i) an intensity pattern of photons at D0, and (ii) for each blip (photon 1) going into that intensity pattern, which detector (D1,2,3 or 4) the corresponding photon 2 was registered at. Then we can do things like determine the component of the intensity pattern at D0 due to photon 1s whose corresponding photon 2s all wound up at D1 (or at D2,3,4). Those are the graphs in that paper.

A crucial aspect of all this is the fact that not only is the photon wavefunction spread out over space, it's also "spread out" over the two "possiblities" of having been originally produced by atom A or atom B. But, as I write this it's rather clear to me that if you're not already comfortable with quantum mechanics to the extent that you could have at least started to construct this analysis yourself, it may not be doing much for you to just read it from me. If that's the case, you really need to start out with simpler experimental arrangements and gain a solid uderstanding of things like quantum superpositions (not necessarily just in space, but also say in photon polarization, etc.), Hilbert spaces, operators on states in those spaces, +++.

I make no guarantees about ensuing flame wars regarding TEW. My only guarantee is about which side is right. And yes, TEW, having forbidden faster-than-light causation, satisfies Bell's Inequality, which has been conclusively disproved in the lab. See my link above.

Bell's theorem is confirmed experimentally. That does not mean that the theory that explains the theory is correct. How can ANYONE on this forum argue against causality or identity? Is Little's theory complete? NO, but it MUST be correct within it's limitation's if causality and identity are to be preserved. Anyone that argues to the contrary is wrong or else Objectivism is wrong.

Apparently you believe TEW, a theory of physics (well, kinda), follows from Objectivism, a philosophy. That assertion is false and itself contrary to Objectivism on a very deep level. Consider the possibility that Objectivism is right, that TEW is wrong, and that some other theory of physics is capable of rationally explaining all the currently available data (about quantum systems).

I'm so SICK and tired of people here offering their attacks on O'ism in some half-hidden way just so they can get around violating the forums rules. I can nearly ALWAYS see right through these indirect attacks and they are REALLY beginning to piss me off. Bring 'em to me first and don't try to pull some slick BS on us.

It seems like you're a bit shell shocked. Perhaps I can relieve your fear that my comments are some stealthy attack on Objectivism. I am an Objectivist and have been for well over a decade. I know a lot about physics. My interest here is primarily to make sure physicists or people knowledgeable of physics don't get bad ideas about Objectivism because: (i) they assume that it's somehow related to TEW, which it isn't, and (ii) they know that TEW is an amateurish, failed attempt to explain quantum mechanics.

[EC]

So what "theory" of QM do you subsribe too that also doesn't violate causality and identity?

[Qwertz]

Bell's theorem is confirmed experimentally. That does not mean that the theory that explains the theory is correct. How can ANYONE on this forum argue against causality or identity? Is Little's theory complete? NO, but it MUST be correct within it's limitation's if causality and identity are to be preserved. Anyone that argues to the contrary is wrong or else Objectivism is wrong. I'm so SICK and tired of people here offering their attacks on O'ism in some half-hidden way just so they can get around violating the forums rules. I can nearly ALWAYS see right through these indirect attacks and they are REALLY beginning to piss me off. Bring 'em to me first and don't try to pull some slick BS on us.

Calm down! I didn't say TEW was wrong - I merely said it's validity was not my current question. I am as yet unable to fully understand it. I acknowledge and agree that the 'Copehagen Interpretation' is severely flawed, and I recognize that TEW might present a more rational explanation, but I do not have an understanding yet of the underlying observations. TEW is not even part of what I'm asking about.

...you really need to start out with simpler experimental arrangements and gain a solid uderstanding of things like quantum superpositions (not necessarily just in space, but also say in photon polarization, etc.), Hilbert spaces, operators on states in those spaces, +++.

Can you make a recommendation, book or otherwise?

[EC]

Calm down!

Yeah, sorry I'm just not in the best of moods tonight.

[Qwertz]

I can thoroughly relate.

<grin>

-Q

Edited to add:

PS to Eric Dennis:

Currently it is known with reasonable certainty, that this speed is at least 10,000 times the speed of light.

Dang that's fast. I don't suppose there's any way to practically use that speed for anything, is there?

Edited April 25, 2006 by Qwertz

[Eric Dennis]

EC: Bohm's theory (usually referred to as his "interpretation of qm") is easily the best one around. You might start here.

Qwertz: Unfortunately there's no real good textbook which both introduces you to the formalism of qm and gets the foundational issues right. And if you want to understand nature, you really need both. See the above link for a start on foundational issues, and for the formalism and a real understanding of the experimental landscape you're going to have to hunker down with one of the standard textbooks, e.g. Schiff or Cohen-Tannoudji (sp?). Bohm and Hiley is an attempt at a textbook, but it's not a good choice for trying to learn qm. (Warning: Bohm's old textbook from the '50s was written before he discovered his theory of qm.)

Re: 10^4 times c: unfortunately qm (in any of its various interpretations) does not allow us to harness these non-local effects. The only way we know they exist is when we bring back together the space-like separated measurement results and compare them -- there are no Bell phones. I suspect someday (in the far future) a better theory may change this.

[EC]

Bohm's theory with the "supernatural" non-local "guide wave" is what you find more promising?

[Eric Dennis]

Bohm's theory with the "supernatural" non-local "guide wave" is what you find more promising?

Yes. I'm also a fan of Newton's theory with the "supernatural" non-local "force of gravity." You see, I'm waging a secret war against the law of identity.

[Qwertz]

The term 'non-local,' does that mean to say that there is a lack of causation, or merely to say that there is a lack of sub-luminal causation? Is spooky-action-at-a-distance only spooky because it's really, really, obscenely fast?

-Q

[EC]

Yeah it means essentially means actions or events that transpire either instantaniously or in time periods that violate limits placed on it by special relativity, i.e., superluminal action.

[Qwertz]

So it is conceiveable that there is some mechanism which, much like well-documented gravity, allows this interaction to take place super-luminally without violating causation? A 'non-local' theory of QP could still be rational? In what way then is Bohm's 'guide wave' (I'm still trudging through that link) supernatural?

PS: With Mr. Dennis' mention of gravity in this context, I can see how the 'curvature of space-time' described by general reletivity is really more of a mental exercise to describe the near-instantaneous action of gravity over great distances, rather than an actual geometric status applied to the purely relational concepts of space and time. With gravity, and perhaps with entanglement?, there is likely some (clearly massless) intermediary force particle (graviton?) which mediates the super-luminal effects?

-Q

[EC]

No. And gravity is NOT instantaneous. Einstein has shown through General Relativity that all gravitational interactions occur at exactly c. I would like to see this "well-documented proof" of superluminal gravitational interactions (or any other superluminal interaction). Note, that I am not speaking of some wave actions that do to special relativity can appear to travel superluminally.

[Eric Dennis]

No. And gravity is NOT instantaneous. Einstein has shown through General Relativity that all gravitational interactions occur at exactly c. I would like to see this "well-documented proof" of superluminal gravitational interactions (or any other superluminal interaction). Note, that I am not speaking of some wave actions that do to special relativity can appear to travel superluminally.

You are struggling to attribute a view to me that is unrelated to anything I've said. I didn't say gravitational effects occur faster than light. I don't know why you put quotes around words that I haven't written ("well-documented proof"). One must distinguish between a particular theory of gravity, Newton's, and gravity itself, the thing. Gravity is local. Newton's theory is non-local (in fact, instantaneous). And yet Newton's theory was and still is an extremely valuable tool in understanding gravity. It captures many aspects of gravity, and still is *right* -- as a scientific theory, an abstraction, operating in a delimited context.

That is why it would have been childish to refuse to consider Newton's theory just because it was non-local. Until various experimental developments around the end of the 19th century, there was no way for even Newton to improve upon his theory. If some contemporary had concocted an alternative theory with a retarded gravitational force going like ~1/r, one does not accept such a theory on the grounds that anything is better than this damned non-locality.

I've had these discussions many times with people interested in physics. There is a common misunderstanding about different ways a scientific theory can be wrong. Some ways a theory can be wrong are quite natural epsitemologically and just imply an incompleteness that may be resolved later. This is the case for Newtonian gravity and arguably for Bohmian mechanics. There are others ways a theory can be wrong that indicate a grave epistemological malfunction. For instance: Copenhagen QM, which exhibits a true blue primacy of consciousness, or Little's TEW, which re-invents itself every couple months or years in a new attempt to square the circle.

Incidentally, if you want "well-documented proof" of superluminal effects actually observed, crack open a book on Bell Inequality experiments. Because that's exactly what they are.

[Starblade]

Bell's theorem is confirmed experimentally. That does not mean that the theory that explains the theory is correct. How can ANYONE on this forum argue against causality or identity?

Define causality and identity first. Then we'll apply logical standards to see if these definitions contradict anything already known about our universe.

Is Little's theory complete? NO, but it MUST be correct within it's limitation's if causality and identity are to be preserved. Anyone that argues to the contrary is wrong or else Objectivism is wrong.

Or you are wrong in that your interpretation of modern physics and Objectivism within eachother's contexts is wrong. If you can tell us why your premises and the conclusions that follow are correct, then you'll at least have a case.

I'm so SICK and tired of people here offering their attacks on O'ism in some half-hidden way just so they can get around violating the forums rules. I can nearly ALWAYS see right through these indirect attacks and they are REALLY beginning to piss me off. Bring 'em to me first and don't try to pull some slick BS on us.

There have been many people here who have argued something different than what Objectivism says. This is allowed. What ISN'T allowed is illogical adherence to alternate philosophies. People must bare out the logic behind their judgements.

The rules may say something that is different in its details, but from what I read, simply holding a view different from what Ayn Rand says is not against the rules. Can you read specifically where it says one must agree with Ayn Rand in all instances?

[EC]

You are struggling to attribute a view to me that is unrelated to anything I've said. blah, blah, blah

This was in response to what Qwertz said not you.... but I'll respond to the rest and the post by Starblade tomarrow when I have more energy. Edited April 27, 2006 by EC

[ttn]

A good chunk of this material (Bell's Theorem, Bohm's theory, nonlocality, TEW, etc.) was covered in some detail in a thread here about a year ago:

http://forum.ObjectivismOnline.com/index.p...topic=3957&st=0

Those interested in the current thread might benefit from reviewing the old one. I would also like to offer a 2nd order appeal-to-authority: anybody reading this should take the comments of Eric Dennis very seriously. He is extremely knowledgeable about these issues, and it's clear (to this authority, at least) that some of the other posters here could learn a lot from him.

[Starblade]

A good chunk of this material (Bell's Theorem, Bohm's theory, nonlocality, TEW, etc.) was covered in some detail in a thread here about a year ago:

http://forum.ObjectivismOnline.com/index.p...topic=3957&st=0

Those interested in the current thread might benefit from reviewing the old one. I would also like to offer a 2nd order appeal-to-authority: anybody reading this should take the comments of Eric Dennis very seriously. He is extremely knowledgeable about these issues, and it's clear (to this authority, at least) that some of the other posters here could learn a lot from him.

I don't see Eric Dennis as having posted on that link.

Also, I believe it's intellectually sloppy to hold onto intrinicist views of space and time. Same with subjectivist views of space and time. I will elaborate on what I mean later.

Edited April 27, 2006 by Starblade

[ttn]

I don't see Eric Dennis as having posted on that link.

He didn't. But I did. (My suggestion to take Eric's comments seriously was meant in reference to this current thread.)