A retrocausal amendment to de Broglie-Bohm pilot wave theory

[DrChinese]

Just when I thought there couldn't be any more quantum interpretations (I think @Demystifier listed 9 in his recent thread)...

Lee Smolin and several others (Cohen, Cortês, Elitzur) have published a pair of related papers discussing dBB/Bohmian Mechanics and its ability to explain the hypothetical guidance wave. Smolin is quite the believer in the value of interpretations of QM, having published at least 5 himself in recent years.

Realism and causality I: Pilot wave and retrocausal models as possible facilitators
Of all basic principles of classical physics, realism should arguably be the last to be given up when seeking a better interpretation of quantum mechanics. We examine the de Broglie-Bohm pilot wave theory as a well developed example of a realistic theory. We present three challenges to a naive reading of pilot-wave theory, each based on systems of several entangled particles. With the help of a coarse graining of pilot wave theory into a discrete system, we show how these challenges are answered. However this comes with a cost. In the description of individual systems, particles appear to scatter off empty branches of the wave function, and conversely travel through particles as if they were waves. More generally, the "particles" of pilot wave theory are led by the guidance equation to move in ways no classical particle would, involving apparent violations of the principles of inertia and momentum conservation. Once this cost is paid, the de Broglie-Bohm pilot wave theory does consistently describe the thought experiments that appeared originally to challenge it. We then suggest that a retrocausal version of the pilot wave theory, in which the particle is guided by a combination of advanced and retarded waves, might account for quantum physics with less damage to intuition. This is the first of two papers. In the second we show that, in the context of an explicit model, retrocausality, with respect to an effective, emergent spacetime metric, can co-exist with a strict irreversibility of causal processes.

The second paper, for completeness, is: Realism and Causality II: Retrocausality in Energetic Causal Sets

I am curious to hear the opinions of others, especially Bohmians, on some of the topics the authors touch on. These include topics discussed in several current/ongoing threads about non-classical Bohmian trajectories, comparing interpretations, whether interpretations are theories or not, etc. If nothing else, the citations to the paper read like a who's who of related interpretations and key ideas around this subject (what might loosely be called "designer interpretations").

Thanks,

-DrC

 

[haushofer]

Motl confuses "a clear ontology" with "classical" in that link. But I don't think this topic is meant for a general discussion on BM :P

 

[stevendaryl]

As you asked for opinions related to the papers and topics discussed in recent threads - it really can't be emphasized enough that 'theories/interpretations' like dBB are considered "ludicrous" and "nonsense" outside of the fringes of physics.

I have learned to skip over anything written by Lubos Motl once it starts sounding like an opinion.

 

[maline]

We then suggest that a retrocausal version of the pilot wave theory, in which the particle is guided by a combination of advanced and retarded waves, might account for quantum physics with less damage to intuition.

The suggestion that invoking retrocausality might lead to "less damage to intuition" is pretty hilarious...

I don't think there is anything problematic about the dynamics of BM, except for the lack of Lorentz invariance. The "problems" these authors are worried about seem to be based on a very basic misunderstanding of what BM is.
BM consists of the Schrodinger equation, the assumption that one particular position configuration is "the true one", the Guiding equation for how that configuration evolves, and the fact that at some point it time, the true configuration matched the Born Rule statistics. This reproduces the predictions of QM, in the sense that the Born probabilities are dynamically maintained. Working out the details in specific thought experiments is an interesting exercise, but little more. Yes, of course the Bohmian "particles" do not have meaningful energies or momenta- the resemblance to classical particles is weak, and more-or-less coincidental. Of course they are influenced by "empty branches" of the wavefunction - that's kind of the whole point!
I should mention, however, that I do think BM should be ruled out for a different reason- it is "MWI in denial", as Deutch and co. have pointed out.

 

[stevendaryl]

The suggestion that invoking retrocausality might lead to "less damage to intuition" is pretty hilarious...

I don't think there is anything problematic about the dynamics of BM, except for the lack of Lorentz invariance. The "problems" these authors are worried about seem to be based on a very basic misunderstanding of what BM is.
BM consists of the Schrodinger equation, the assumption that one particular position configuration is "the true one", the Guiding equation for how that configuration evolves, and the fact that at some point it time, the true configuration matched the Born Rule statistics. This reproduces the predictions of QM, in the sense that the Born probabilities are dynamically maintained. Working out the details in specific thought experiments is an interesting exercise, but little more. Yes, of course the Bohmian "particles" do not have meaningful energies or momenta- the resemblance to classical particles is weak, and more-or-less coincidental. Of course they are influenced by "empty branches" of the wavefunction - that's kind of the whole point!
I should mention, however, that I do think BM should be ruled out for a different reason- it is "MWI in denial", as Deutch and co. have pointed out.

I agree about the "intuition" part, but I think the idea behind retrocausality is that you can (perhaps) have the same effect as Bohmian action-at-a-distance while preserving Lorentz invariance.

 

[RUTA]

Just when I thought there couldn't be any more quantum interpretations (I think @Demystifier listed 9 in his recent thread)...

Lee Smolin and several others (Cohen, Cortês, Elitzur) have published a pair of related papers discussing dBB/Bohmian Mechanics and its ability to explain the hypothetical guidance wave. Smolin is quite the believer in the value of interpretations of QM, having published at least 5 himself in recent years.

-DrC

Of course, I have to weigh in on this one (first paper only here)

I have known Avshalom for many years and he has long struggled with the use of meta-time. He appreciates the explanatory power of future boundary conditions in explaining QM mysteries, but he cannot let go of dynamical, time-evolved explanation in physics. That forces him to adopt an additional time parameter according to which the block universe evolves:

"Trajectories, evolutions and histories genuinely “grow” into the yet-non-existent future. The past, on the other hand, is where the current block universe model can be regarded as perfectly valid, i.e. events are fixed and unchangeable, obeying well defined causal relations." p. 24

"What are the pros and cons for this model? Let us begin with the latter, which are admittedly obvious. We are talking about evolution in spacetime as if there is another time parameter, about which we yet know nothing. But this is not necessarily a disadvantage for a foundational physicist who might be long suspecting that something about time is still missing in the “block-universe” view." p. 25

There are at least two problems with meta-time. First, it requires a preferred foliation of spacetime contrary to the relativity of simultaneity that motivates block universe to begin with. Second, since it violates the relativity of simultaneity, there can be only one set of observers for whom meta-time = proper time. Who is that?

The bottom line is that requiring there be "no preferred reference frame" leads to block universe where everyone's proper time is equally valid. Avshalom et al. believe this is unacceptable, since we certainly do not experience time in this way. This is a subset of the hard problem of consciousness and neither physics nor any other science can solve that problem. Thus, this simply means that physics cannot provide a theory of everything. I won't elaborate further, see chapters 7 and 8 of our book (his reference 15).

Finally, there are two characterizations of our 4D-constraint-based interpretation of modern physics that are misrepresented in this paper. First, we use the term "4D constraint" not "retrocausality" when talking about the use of future boundary conditions to explain QM correlations. Even Ken Wharton told me at the APS March 2018 Meeting that he has abandoned the term "retrocausal" in favor of "all-at-once view." My colleagues and I once tried to reconcile our constraint-based interpretation with retrocausality, but a referee saw through it writing, "I do not see how anything truly ‘retrocausal,’ in a dynamical sense, can occur given global time-symmetric constraints on spacetime. The authors seem to me to be too charitable here, a future boundary condition implies an adynamical block world, in which talk of dynamics or intervention is superfluous at best, and inconsistent at worst.” She was right of course It is best to abandon dynamical language such as "retrocausality" and admit we're dealing with a ‘4D crossword puzzle’ rather than a ‘game of chess being played backwards in time’.

Second, the authors characterize our interpretation as "instrumentalist" while calling their approach "realist." Make no mistake, our constraint-based explanation is realist, e.g., Relational blockworld: Providing a realist psi-epistemic account of quantum mechanics. They conflate wave function realism with realism since they are looking for dynamical, time-evolved stories (thus, they end up having to add meta-time to make sense of advanced waves). Once you accept 4D constraints as providing explanation rather than requiring explanation, the mysteries of modern physics, e.g., puzzle of the Big Bang, flatness problem, horizon problem, low entropy problem, paradoxes of closed timelike curves, measurement problem, conundrums associated with entanglement, all disappear as remnants of our dynamical bias.

 

[Demystifier]

I agree with @maline , except with the part that "BM is MWI in denial". I would rather say that MWI is incomplete and BM is the simplest completion of MWI. But that would be another topic.

 

[maline]

I agree about the "intuition" part, but I think the idea behind retrocausality is that you can (perhaps) have the same effect as Bohmian action-at-a-distance while preserving Lorentz invariance.

Indeed, it seems clear to me that the combination of [demanding some ontology + Lorentz invariance + Single World] implies that what's really going on is very different from our concept of cause & effect as a chain from past to future. But this paper seems to be claiming that BM dynamics, on their own, require some horrible sacrifice of intuition, to the extent that this alone might make retrocausality more attractive by comparison. This seems bizarre.

I agree with @maline , except with the part that "BM is MWI in denial". I would rather say that MWI is incomplete and BM is the simplest completion of MWI. But that would be another topic.

Hmm, I suppose BM can't be MWI "in denial" if our resident Bohmian-in-Chief refrains from "denying" that BM includes Many Worlds!
But as you say, this isn't the place for the discussion.

 

[zonde]

There is one thing common to retrocausality, adynamical constraint and superdeterminism. In Bell inequality experiments we prepare initial state and the we test this initial state with independent parameters. Of course we can not guarantee that "independent parameters" are indeed independent but we do the things to make this belief extremely likely. So interpretation of Bell inequality violation relies on assumption that there is no correlation between prepared initial state and independent parameters when communication loophole is closed.
The common thing about retrocausality, adynamical constraint and superdeterminism is that they propose such correlation exists. Well, in case of superdeterminism we call that correlation "conspiracy" and dismiss it as non scientific. But I see no reason why that correlation should be named differently in case of retrocausality and adynamical constraint as I consider it sort of obvious that the problem of superdeterminism is not it's hypothetical mechanism by which it makes this correlation appear but rather the effect of such correlation itself on scientific approach in general.

 

[RUTA]

There is one thing common to retrocausality, adynamical constraint and superdeterminism. In Bell inequality experiments we prepare initial state and the we test this initial state with independent parameters. Of course we can not guarantee that "independent parameters" are indeed independent but we do the things to make this belief extremely likely. So interpretation of Bell inequality violation relies on assumption that there is no correlation between prepared initial state and independent parameters when communication loophole is closed.
The common thing about retrocausality, adynamical constraint and superdeterminism is that they propose such correlation exists. Well, in case of superdeterminism we call that correlation "conspiracy" and dismiss it as non scientific. But I see no reason why that correlation should be named differently in case of retrocausality and adynamical constraint as I consider it sort of obvious that the problem of superdeterminism is not it's hypothetical mechanism by which it makes this correlation appear but rather the effect of such correlation itself on scientific approach in general.

4D constraints are only "conspiratorial" from the "ant's-eye view," i.e., for those who demand explanation be couched in causal, time-evolved stories. For those people, a conservation principle that only holds on average is an explanandum (something requiring explanation), not an explanans (an explanation). QM correlations do have a physical basis in this type of conservation, so whether or not you consider such conservation to be mysterious and in need of explanation or you rather understand that such conservation is the explanation depends on whether or not you can rise to Wilczek's challenge and ascend to the "all-at-once view" consistent with a truly 4D objective reality. Wilczek's challenge is not a technical challenge, our physics is already 4D, it is a conceptual challenge to accept that objective reality is really 4D, not (3+1)D as we "ants" experience it. Again, it's not just QM's conceptual mysteries that disappear in the "all-at-once view" (e.g., here, here, here, here, here, here), those of GR cosmology and closed timelike curves also disappear.

Believe me, I know it's not easy, those of us who espouse the "all-at-once view" occasionally find ourselves confused because we have lapsed into the "ant's-eye view." 4D-constraint-based explanation is very counterintuitive, i.e., it runs contrary to our experience as "ants." But, once you wrap your head around it, you'll realize how incredibly powerful and simple 4D explanation is and you will appreciate Wheeler's quote: “we will all say to each other, how could it have been otherwise? How could we have been so stupid for so long?”

 

[zonde]

4D constraints are only "conspiratorial" from the "ant's-eye view," i.e., for those who demand explanation be couched in causal, time-evolved stories.

Would you say that demanding event based explanation vs statistical explanation is fine as long as events are not required to be presented as time ordered causal sequences?

QM correlations do have a physical basis in this type of conservation, so whether or not you consider such conservation to be mysterious and in need of explanation or you rather understand that such conservation is the explanation depends on whether or not you can rise to Wilczek's challenge and ascend to the "all-at-once view" consistent with a truly 4D objective reality.

How do you know when someone can use "all-at-once view" consistent with a truly 4D objective reality? If your only criterion about the ability to use "all-at-once view" is whether someone agrees with your conclusions or not you are excluding any possibility that you might be wrong (that "all-at-once view" by itself does not imply your conclusions).

Believe me, I know it's not easy, those of us who espouse the "all-at-once view" occasionally find ourselves confused because we have lapsed into the "ant's-eye view." 4D-constraint-based explanation is very counterintuitive, i.e., it runs contrary to our experience as "ants."

Can you using "all-at-once view" predict something more about things that "ants" can actually observe compared to things that can be predicted using "ant's view"?

 

[RUTA]

Would you say that demanding event based explanation vs statistical explanation is fine as long as events are not required to be presented as time ordered causal sequences?

I teach intro physics, E&M, statics, acoustics, special relativity, general relativity, quantum mechanics, ... and I explain virtually everything in causal, time-evolved story form, i.e., dynamically. I ONLY use constraint-based explanation in those situations where dynamical explanation is lacking or inconsistent (which is rarely).

How do you know when someone can use "all-at-once view" consistent with a truly 4D objective reality? If your only criterion about the ability to use "all-at-once view" is whether someone agrees with your conclusions or not you are excluding any possibility that you might be wrong (that "all-at-once view" by itself does not imply your conclusions).

You can always use constraint-based explanation if you want. I'm a teacher, so I use whichever is easiest for my students to understand. Since dynamical explanation works in most cases and my students are "ants," I mostly use dynamical explanation. For example, I'm teaching GR now in my modern physics course and I stated at the outset: "Thus, solutions of Einstein's equations (EEs) are simply self-consistent sets of spatiotemporal measurement, force, momentum, and energy on the spacetime manifold M." That's the 4D-constraint view of GR which avoids inexplicable initial conditions of GR cosmology and consistency paradoxes of closed timelike curves. And, when we solved EEs for the Einstein-de Sitter metric, we solved for both zero and non-zero scaling factor at t = 0, highlighting the 4D-constraint view. But, when we solved problems with the Schwarzschild and Einstein-de Sitter metrics, I talked about orbiting astronauts and photons moving through space from large-redshift quasars to Earth, it was all very dynamical.

Can you using "all-at-once view" predict something more about things that "ants" can actually observe compared to things that can be predicted using "ant's view"?

Whichever version of the equations is easiest to solve, do it that way. Hamiltonian or Lagrangian, you better get the same answer. It's only when I teach GR and QM that I use the 4D-constraint-based explanation to avoid "faux" mysteries associated with the dynamical view. I explained above where that enters GR. In QM, for example, we read Mermin, N. (1981). Bringing home the atomic world: quantum mysteries for anybody. American Journal of Physics, 49(10):940–43 and fill in all the missing QM, so they can be "mystified." Then I show them the 4D-constraint-based explanation of that experiment. My entire course on QM is like that, reading a paper, filling in all the missing QM details, and writing computer programs (like the MATLAB spectral analysis in Danan, A., Farfurnik, D., Bar-Ad, S., & Vaidman, L.: Asking Photons Where They Have Been. Physical Review Letters 111, 240402 (2013)), then resolving the mystery via 4D-constraint-based explanation. But, QM is the exception obviously.

 

[RUTA]

To follow up on my last post (#13), it’s a bit like when I teach astronomy and talk about the motion of the celestial sphere about Earth. We don’t think Earth is the center of the universe, i.e., not all observations are best understood geocentrically, but astronomy textbooks present it that way, since that’s the way we experience it. Here’s a quote from Price, H., & Wharton, K.: Dispelling the Quantum Spooks – a Clue that Einstein Missed? (2013) http://arxiv.org/abs/1307.7744:

In putting future and past on an equal footing, this kind of approach is different in spirit from (and quite possibly formally incompatible with) a more familiar style of physics: one in which the past continually generates the future, like a computer running through the steps in an algorithm. However, our usual preference for the computer-like model may simply reflect an anthropocentric bias. It is a good model for creatures like us, who acquire knowledge sequentially, past to future, and hence find it useful to update their predictions in the same way. But there is no guarantee that the principles on which the universe is constructed are of the sort that happens to be useful to creatures in our particular situation.

Physics has certainly overcome such biases before – the Earth isn’t the center of the universe, our sun is just one of many, there is no preferred frame of reference. Now, perhaps there’s one further anthropocentric attitude that needs to go: the idea that the universe is as “in the dark” about the future as we are ourselves.

 

[stevendaryl]

I have an intuitive reluctance to accept constraint-based theories. It might be just over-reliance on making the common-sensical distinction between time and the other dimensions. You can have a physical system that is elegantly described by a constraint. For example, a soap bubble is said to form a surface of minimal area (it doesn't, actually, except for simple situations). Or you can describe mechanics in terms of a particle taking a path that minimizes the action. However, at a deeper level, you can understand how local, causal laws can produce something that satisfies the constraint. So my first instinct when told that a system evolves in such a way as to satisfy some constraint is to think that there is some evolution equation that leads to the constraint being satisfied.

However, I admit that this is a prejudice from thinking in terms of systems evolving over time. From a 4-dimensional point of view, there is no hard and fast distinction between something satisfying a 4-dimensional constraint and something evolving over time according to causal rules. The latter is a special case of the former.

Interestingly, the Cramer transactional interpretation of QM manages to understand a 4-dimensional constraint in terms of a kind of evolution: When a photon is emitted by one particle and absorbed by a second, you can think of the first particle as being the source of a retarded "offer wave", and the second particle is the source of an advanced "acceptance wave". That's an intuitively satisfying (to me) way to think about it. However, mathematically, it's not clear what it means in terms of evolution. The story has a temporality that is not captured in the theory: First, an offer wave is made, then a response wave is created, and finally the particle is transmitted. But this notion of first this then this is unconnected with any time parameter.

 

[RUTA]

Since the advanced wave is emitted at the sink and goes backwards in time from the reception of the retarded wave, its reception at the source coincides with the emission of the retarded wave at the source. Thus, information about the sink is already at the source when the retarded wave is emitted. Clearly, as you suspect, nothing is really “happening” in the transactional interpretation, as Cramer himself admits https://arxiv.org/pdf/1011.2287.pdf:

Cramer suggests that the ‘dynamical process’ be understood as occurring in a “pseudotime” sequence:
The account of an emitter-absorber transaction presented here employs the semantic device of describing a process extending across a lightlike or a timelike interval of space-time as if it occurred in a time sequence external to the process. The reader is reminded that this is only a pedagogical convention for the purposes of description. The process is atemporal and the only observables come from the superposition of all “steps” to form the final transaction. (1986, p. 661, fn. 14)