Quantum effects from an imperfect absorber

[michael879]

I've had this idea stuck in my head ever since I read up on the Feynman-Wheeler formulation of classical E&M (absorber theory), and I was wondering if anyone here had any comments on it. For those who aren't familiar with it, it is a slight modification of the well-known E&M theory that makes identical experimental predictions, but includes some very interesting features. The usual approach to E&M is to ignore all advanced solutions to Maxwell's equations under the premise that they are unphysical. This embeds a time-asymmetry into the fundamental theory that does not appear in the equations describing the theory. The radiation reaction is explained by self-interactions which cause unphysical divergences of point particles (fixed by renormalization).

The original absorber theory, on the other hand, makes three basic assumptions:
1) Self-interactions can not occur
2) Not only Maxwell's equations, but also their solutions, are time-symmetric
3) The universe is a perfect absorber
This makes use of advanced waves to explain the radiation reaction usually attributed to self-interaction, and the observed time-asymmetry of our universe is treated as a consequence of the boundary conditions (perfect absorption of retarded waves) rather than a fundamental postulate of the theory. As long as the universe is a perfect absorber, no acausal effects will ever be observed and the two theories are equivalent!

Now, the idea I had was what if the universe was just a nearly perfect absorber? At certain scales retrocausal effects would "bleed" through due to imperfect absorption of the retarded radiation. It is well known that Bell's inequality leads one to conclude that the universe is either: nondeterministic, nonlocal, or acausal (the acausal interpretation is not quite as well-known). Is it possible that the retrocausal effects caused by a nearly perfectly absorbing universe could mimic the retrocausal interpretation of QM? I find it fascinating that it might be possible to obtain quantum phenomenon from a classical theory, and so far I've only been able to come up with 2 problems with this idea:

1) Lamb-shift: the phenomenon that led Feynman to abandon absorber theory, after he concluded it was impossible to get this shift without self-interactions. However, even if he was right this doesn't quite kill the theory. The original motivation for it may have been to remove self-interactions, but that in no way means that self-interactions are forbidden by the theory! Adding self-interactions back in leaves the theory intact, it just removes one of the nicer features of it. Also, I may not be looking at this the right way but I can't understand how two theories that are identical classically can diverge so drastically after quantization? Absorber theory can reproduce all the effects of self-interaction classically, so why would quantization change this?

2) Planck's constant: the effective "strength" of quantum effects. If quantum mechanics is simply a consequence of a non-perfectly absorbing universe, the "strength" of the retrocausal effects would vary with position (certain locations would be better suited to absorb radiation). This means that Planck's constant would actually be a dynamic function over space-time, and the "constant" we observe would merely be some local/averaged value. The fact that distant electromagnetic radiation is quantized into units of the same constant we observe on Earth seems to be at odds with this idea. However, the absorption capability of the universe as far as retrocausal effects is concerned is independent of distance or time. If the retarded radiation of a system is ever absorbed fully, the advanced signal is destroyed! So it could just be that any "distant" radiation we observe is still too close to effect the absorption coefficient of the universe significantly?

 

[michael879]

^ bump

 

[rkastner]

The WF theory and TI

...Now, the idea I had was what if the universe was just a nearly perfect absorber? At certain scales retrocausal effects would "bleed" through due to imperfect absorption of the retarded radiation. It is well known that Bell's inequality leads one to conclude that the universe is either: nondeterministic, nonlocal, or acausal (the acausal interpretation is not quite as well-known). Is it possible that the retrocausal effects caused by a nearly perfectly absorbing universe could mimic the retrocausal interpretation of QM? I find it fascinating that it might be possible to obtain quantum phenomenon from a classical theory, and so far I've only been able to come up with 2 problems with this idea:

Actually, there are a number of what could be called 'retrocausal' interps of QM -- Cramer's Transactional Interpretation, which I have further developed into the relativistic regime, is one of them. None of these predict any difference from standard QM theory however, since they are just interpretations of the theory. So the answer to this would basically be 'no', although there may be a slight departure from standard QED in TI at the relativistic level; I'm currently exploring this possibility. The other 'retrocausal' intepretations involve time symmetric hidden variables, and basically require a 'block world' view -- in contrast, my proposal allows for a dynamically emergent spacetime. But the basic point is that incomplete absorption is not necessary in a W-F type theory (i.e. TI) to account for the phenomena predicted by QM such as violations of Bell's inequality. Cramer himself likes to account for Bell inequality violations in terms of 'zig-zags' that go both forward and backward in time.

1) Lamb-shift: the phenomenon that led Feynman to abandon absorber theory, after he concluded it was impossible to get this shift without self-interactions. However, even if he was right this doesn't quite kill the theory. The original motivation for it may have been to remove self-interactions, but that in no way means that self-interactions are forbidden by the theory! Adding self-interactions back in leaves the theory intact, it just removes one of the nicer features of it. Also, I may not be looking at this the right way but I can't understand how two theories that are identical classically can diverge so drastically after quantization? Absorber theory can reproduce all the effects of self-interaction classically, so why would quantization change this?

Wheeler/Feynman's theory was for classical em only. Davies (1971-2) developed a quantum version of the WF theory, but the em field is not quantized in that version, as it is a direct-action theory. Davies' theory however does allow for self-interaction and when the universe is a perfect absorber it leads to predictions identical with standard QED (except for the possible slight deviations I'm exploring, which would only appear in exotic systems). It's true that Feynman's original motivation was to eliminate the field, but that's certainly not my motivation -- I think that the direct-action approach provides a nice way to understand what's going on 'behind the scenes' in QM and also provides a physical basis for the otherwise ad hoc Born Rule.

For further details, you can check out my guest post on George Musser's Sci Am blog:

http://blogs.scientificamerican.com...by-stepping-out-of-space-and-time-guest-post/

Also my own blog is here:

http://transactionalinterpretation.org

It has a link to my new book on TI.

 

[michael879]

rkastner,

Thank you for the lengthy response, and I will be sure to check out those links. However, I think you misinterpreted the point of my thread. I was not talking about quantum absorber theories, but rather a classical absorber theory that can generate quantum-like effects. Whether this theory would reproduce any or ALL of the predictions of QM, I don't know and that's what I was asking about. Could a classical theory with retrocausality produce quantum effects?

 

[rkastner]

Thanks Michael for this clarification -- an interesting idea, but I'm skeptical. The quantum em field is a very different animal from the classical em field. A classical em field corresponds to a quantum em field with an indefinite number of photons (a 'coherent state' or Glauber state). (A quantum field with a definite # of photons gives zero classical field.)

 

[DrChinese]

Thanks Michael for this clarification -- an interesting idea, but I'm skeptical. The quantum em field is a very different animal from the classical em field. A classical em field corresponds to a quantum em field with an indefinite number of photons (a 'coherent state' or Glauber state). (A quantum field with a definite # of photons gives zero classical field.)

Ruth,

In your view, the act of photon absorption has a special place - it leads to collapse. Would you say, given your interpretation, that an electron must "see" an offer wave before a photon can be emitted? Another way to ask that is to say: can a truly free photon exist?

Thanks.

 

[rkastner]

Ruth,

In your view, the act of photon absorption has a special place - it leads to collapse. Would you say, given your interpretation, that an electron must "see" an offer wave before a photon can be emitted? Another way to ask that is to say: can a truly free photon exist?

Thanks.

Hi Dr. Chinese,

TI is best understood in a 'direct action' theory formulation (i.e., see Davies 1971-2) -- in which case the answer would be 'no'. Davies did consider a 'real' photon as corresponding to the pole in the Feynman propagator. However, it's important to keep in mind that Davies did not consider the transactional picture, in which a 'real' photon is an actualized transaction. He seemed to be identifying the pole only as an emitted (but not absorbed) photon. In TI, the pole can be seen as corresponding to an emitted and absorbed quantum of real energy. So in TI there certainly can be real photons, but you don't have a photon emitted without being absorbed. (To that, however, there is also a caveat: this could be possible with a perfectly reflecting past boundary condition, which would accomplish the same function as a perfectly absorbing future boundary condition.) To sum up: in TI a real photon is always due to an actualized transaction, either from an absorber response or from a reflecting past boundary. So that means there are no 'free' fields, i.e., excited states of a field without any source.

 

[michael879]

Thanks Michael for this clarification -- an interesting idea, but I'm skeptical. The quantum em field is a very different animal from the classical em field. A classical em field corresponds to a quantum em field with an indefinite number of photons (a 'coherent state' or Glauber state). (A quantum field with a definite # of photons gives zero classical field.)

In absorber theory there are no free photons, so any observation of a quantized E&M field could be due to details of absorption/emission rather than something fundamental to the E&M field. It's incredibly difficult, and maybe impossible, to isolate the E&M field in any kind of practical experiment since there will always be some kind of absorption and emission involved. Because of this I'm trying to limit the scope of this discussion to something much simpler: are there any effects we usually treat as purely quantum mechanical in nature that can be reproduced by a classical absorber theory with an imperfectly absorbing universe? Both theories are retrocausal, and they are both based on the same fundamental equations, so surely they share some properties in common.

To keep things really simple, let's just look at Bell's inequality!

Does absorber theory violate Bell's inequality in the same way as quantum mechanics?

 

[rkastner]

Thanks Michael -- I'm in the middle of a house move right now, but will keep this in mind and let you know if I have any further thoughts on it.

 

[DrChinese]

... So that means there are no 'free' fields, i.e., excited states of a field without any source.

I thought that might be the case, and as far as I know there is nothing about this view that is inconsistent or contradictory. However, I have puzzled over this previously. Haven't really ever seen any analysis on this, and wondered what you might think.

Let's assume that the current standard cosmological model of an accelerating expansion is correct. With that model, the Milky Way and a few other galaxies is all that will be visible to us at some date in the far future. In fact, some of the light currently emitted by our sun - in the direction of deep space - could never move to a region where it could be absorbed. The distance is just too great, and it is growing.

According to your viewpoint, such light could not be emitted as there is nothing for it to transact with. It seems to me that it would be possible to construct an experiment with pairs of entangled photons. By aiming Alice's beam to the wall across the room, Bob's beam will have a certain intensity. By instead aiming Alice's beam to deep space, where it cannot be absorbed ever (per above), it should affect Bob's beam intensity. Since Alice's photon cannot be emitted, Bob's can't either! So we could observe that. Such would be a direct confirmation of TI and other interpretations with similar elements.

On the other hand: if we observed a null result, it could mean either that our cosmological model is not correct, or TI is not correct. Thoughts, anyone?

 

[michael879]

I think it would be difficult to prove that there is any region of space where total absorption will never occur, and your "experiment" assumes their existence. I'm not trying to speculate on whether or not any absorber theory is valid, I'm just exploring what they predict in the absense of a perfect future absorber

 

[DrChinese]

I think it would be difficult to prove that there is any region of space where total absorption will never occur, and your "experiment" assumes their existence. ...

I am assuming an expanding universe, of course. Perhaps there is a big crunch.

 

[rkastner]

I thought that might be the case, and as far as I know there is nothing about this view that is inconsistent or contradictory. However, I have puzzled over this previously. Haven't really ever seen any analysis on this, and wondered what you might think.
... It seems to me that it would be possible to construct an experiment with pairs of entangled photons. By aiming Alice's beam to the wall across the room, Bob's beam will have a certain intensity. By instead aiming Alice's beam to deep space, where it cannot be absorbed ever (per above), it should affect Bob's beam intensity. Since Alice's photon cannot be emitted, Bob's can't either! So we could observe that. Such would be a direct confirmation of TI and other interpretations with similar elements.

On the other hand: if we observed a null result, it could mean either that our cosmological model is not correct, or TI is not correct. Thoughts, anyone?

By a 'null result' I assume you mean no change in Bob's intensity. But (as I think you acknowledge here) that depends heavily on the idea that there really is some area of deep space without any absorbers of any kind. Also, even if that were the case, there is an alternative boundary condition that can provide the same effect as the 'light-tight box' -- a perfectly reflecting t=0 boundary condition. So a null result really can't be taken as ruling out absorber theory unless we could somehow know for a fact that the universe is not a light-tight box or lacks such a past reflecting BC.

BTW, your motto "the map is not the territory" is the basic thesis of Chapter 2 of my book ;)

 

[michael879]

Also, even if that were the case, there is an alternative boundary condition that can provide the same effect as the 'light-tight box' -- a perfectly reflecting t=0 boundary condition.

Could you explain this further? I haven't come across any mention of this, and I don't really see why it would be true.

One thing that has been bothering me though (and Dr Chinese alluded to) is that its pretty clear we don't live in an even nearly perfect future absorber. Of course our cosmological models could be wrong, but as long as there is a big bang it would appear that in fact we live in a perfect past absorber! So why would the arrow of time point away from the big bang instead of towards it?? Your statement reminded me of this issue, and I'm hoping it can resolve it

 

[DrChinese]

... So why would the arrow of time point away from the big bang instead of towards it?? Your statement reminded me of this issue, and I'm hoping it can resolve it

Symmetry considerations might lead me to postulate that there is another universe, similar to ours, that was created at the Big Bang. The time direction we reference as the "future" would be the "past" to them. Our "present" would be receding from their "present" in the time continuum.

 

[rkastner]

The relevant reference is by Cramer, (1983). “The Arrow of Electromagnetic Time and the Generalized Absorber Theory,” Foundations of Physics 13, 887-902."

But on further thought, I don't think the proposed experiment can work. The photons (remember these are just offer waves in TI) are emitted from a single common source. The beams themselves are entangled, so neither Alice nor Bob really have access to an independent beam of photons. They couldn't measure the intensity of either of their beams, unless they themselves were somehow absorbing the photons. But this of course precludes probing open space. The best they can do is use mirrors to direct each beam, but a mirrow can't measure the intensity of either beam.

However, consider a single photon source whose power utilization we could measure, and beam it first at a wall and second into open space. If you could see a dip in the power utilization when it's aimed into open space, AND assuming that our current ideas about cosmology are correct (i.e. probably not a light tight box), AND assuming that there is no perfectly reflecting past boundary condition, then this would provide corroboration of the absorber theory. But because there are so many unknowns here concerning the cosmological model, a null result couldn't be taken as a refutation of absorber theory.

 

[DrChinese]

... However, consider a single photon source whose power utilization we could measure, and beam it first at a wall and second into open space. If you could see a dip in the power utilization when it's aimed into open space, AND assuming that our current ideas about cosmology are correct (i.e. probably not a light tight box), AND assuming that there is no perfectly reflecting past boundary condition, then this would provide corroboration of the absorber theory. But because there are so many unknowns here concerning the cosmological model, a null result couldn't be taken as a refutation of absorber theory.

I had considered your power usage option in my first consideration of the idea. But I think there is way to use entangled photon pairs.

1. Alice and Bob each receive a beam of entangled photons. Each has a beam splitter (PBS) oriented at 45 degrees (or any same angle), and 2 detectors. Alice has path A1 and A2 with detectors at each; Bob has B1 and B2 with detectors. We would expect A1=A2 since each photon has a 50-50 probability of going to either A1 or A2. Due to perfect correlations, counts should be about: A1=A2=B1=B2.

2. Next: Alice points her A1 path to deep space where it cannot be absorbed (hypothetically). If her A1 path photons are not absorbed, then they must not be emitted from the PDC crystal (at least not as part of an entangled pair). There is no associated click at B1. Therefore Bob’s B1 is no longer equal to Alice’s A2 or Bob’s B2.

3. Hmmm, but that can’t be right. Because then Alice could send an FTL signal to Bob simply by pointing path A1 to deep space. Since Bob would instantly notice that B1<>B2. So I would again expect a null result. That forces me to conclude:

4. A Null result means at least one of below is true:
a. We are not in an eternally expanding universe (there may be a big crunch or some other unimaginable something in the far distant future).
b. Absorber theory is wrong (or there is some other compensating element).
c. No FTL signaling is possible (ie somehow we are running afoul of uncertainty principle).

5. A Non-Null result indicates all of below are true:
a. We are in an eternally expanding universe.
b. Absorber theory is right.
c. FTL signaling possible.

-DrC

 

[rkastner]

This sounds like a reasonable summary; however there are some big cosmological unknowns here, so I think it would be a mistake to try to rule out absorber theory based on this kind of analysis. Keep in mind what we get from a direct action theory of fields:

1. A clear physical basis for von Neumann's projection postulate -- so that it's no longer a 'postulate' but arises from QM theory itself. I.e., a real solution to the measurement problem in physical terms.
2. A straightforward connection between nonrelativistic and relativistic QM-- not available in any other QM interpretation.
3. An explanation for both the radiative and thermodynamic arrows of time (see Chapter 9 of my book).

And also -- to take a null result as a refutation of absorber theory would mean one would have to certain that *all* these are also true:
(a) path A1 offer wave component really did not encounter a single absorber, ever;
(b) the universe is open
(c) there is no reflecting t=0 boundary condition

 

[DrChinese]

...And also -- to take a null result as a refutation of absorber theory would mean one would have to certain that *all* these are also true:
(a) path A1 offer wave component really did not encounter a single absorber, ever;
(b) the universe is open
(c) there is no reflecting t=0 boundary condition

Well, I am a fan of the class of interpretations/theories that include a time symmetric component of some type. But it is an interesting idea for an experiment nevertheless.

 

[rkastner]

Yes, and I think it's a mark of methodological superiority that the direct-action approach is testable at least in principle, as opposed to assuming retarded solutions only, which is ad hoc.

 

[rkastner]

BTW, there is a new review of my book acknowledging that Maudlin's objection and related concerns are put to rest: http://www.tandfonline.com/doi/abs/10.1080/00107514.2013.825322#.UlGhncpNYZ8

Excerpt: "There have been many objections to the TI over the years, most famously by Tim Maudlin in his book
‘Quantum Nonlocality and Relativity’. This is where Kastner comes into her own, with her own updated version
of TI called ‘possibilist transactional interpretation’ which keeps all the key ideas of Cramer but extends the
philosophical basis of TI to eliminate the original dependence on ‘pseudotime’ and replace it with a more fundamental view of Hilbert space describing pre-spacetime possibilities, out of which spacetime arises from the actualised transactions. From this viewpoint, the author systematically refutes all the latest and most sophisticated philosophical challenges to the TI."

 

[mitchell porter]

michael879, I suggest you ask https://www.math.ucdavis.edu/~deckert/UC_Davis/Dirk_-_Andre_Deckert.html your main question, he did a thesis on classical absorber theory just a few years ago.

You may also be interested in Mark Hadley, who wants to get quantum mechanics from classical general relativity plus small time loops.