We do, of course, observe that the cumulative distribution of impacts on the back screen is in line with the square of the wavefunction from the emitter. If emission occurs whenever, while the distribution of available absorbers on the screen is constrained by external factors, then we won't recover the expected distribution. — SophistiCat

That this doesn't hold true is precisely my point. While an individual transmission may depend on the precise microstate of the screen, the screen explores these microstates continuously. A statistical number of transmission events will take place over a period of time, during which one will have a statistical spread across the precise microstates explored during that time, a spread which looks like the probability of finding a given electron at a given position depends principally on the wavefunction.

But other than terminology, do you see any issues with his proposal? — SophistiCat

The cancellation depends on both waves being advanced waves, so it's not purely terminological. (Advanced waves cannot cancel retarded waves in Cramer's formulation.)

Instead of that, they should interact with the ordinary dense matter, at least 380,000 years old, as shown in Figure 2 from my recent paper on the problem of the direction of the electromagnetic arrow of time: http://philsci-archive.pitt.edu/13505 — Darko B

I agree with the gist of your idea, that advanced waves could exist and will look to us like retarded waves (photons being their own antiparticles).

I don't see the classical spherical wavefront as a particular problem to get around though. Quantum mechanically, the photon, after collapse, is never consistent with a spherical wave. For instance, the emitting object will undergo recoil in a direction inconsistent with the symmetry of spherical wave emission.

Spherical wavefronts are useful because we do not know the full boundary conditions of the transmission and, further, because they recover the correct interference effects for every possible future absorption event (sum over histories). According to the OP, we can consider the retarded and advanced parts of absorption and emission as spherical separately, but the full transmission is only the sums over all paths between the emitter and the absorber... in *either* direction of time.

That this doesn't hold true is precisely my point. While an individual transmission may depend on the precise microstate of the screen, the screen explores these microstates continuously. A statistical number of transmission events will take place over a period of time, during which one will have a statistical spread across the precise microstates explored during that time, a spread which looks like the probability of finding a given electron at a given position depends principally on the wavefunction. — Kenosha Kid

Let's take an extreme example:

1. The emitter (of electrons, photons, ...) is under experimental control, so that for instance we can ensure that a particle is emitted every millisecond.

2. The availability of receptive absorbers is so constrained by present and future boundary conditions that at any point of time at most one site is available.

Right away, if at the time when we want to make emission happen there are no available absorbers, then we have a problem: some assumption has to give. But even if an absorber is available, the cumulative distribution of impacts will be defined only by the distribution of the available absorbers on the screen over time. And at the same time, in order for the Born rule to hold, that distribution has to match the impacting wavefunction - whatever it happens to be. If we can contrive to emit a particle that hits the screen at times (t₁, t₂, ...), the screen had better supply us absorbers at such locations r_i that (r₁, r₂, ...) form the distribution that we expect to see.

So how can this happen (or can it happen)?

A. (1) and (2) hold, which means that the screen and the universe in its future lightcone have to contrive to match the impacting wavefunction. While no individual absorber is constrained to be in a fixed position at a fixed time, there is a constraint over time on all such absorbers, which is a function of the impacting wavefunction, whatever it happens to be.

B. (2) doesn't quite hold: instead of just one absorber at a time, we have "not many" absorbers. This relaxes the constraint on the screen, but does not completely eliminate it. Unless there is such a constraint, the actual distribution of impacts will inevitably be distorted.

C. (1) does not quite hold: we cannot make emissions happen at will (can we?) This distributes the constraint of producing the right cumulative distribution between the emitter and the screen or shifts it entirely to the emitter, so that now it is the emitter's responsibility to be aware of the state of the screen (and the rest of the universe in the future lightcone) and fire particles under the constraint of producing the right distribution.

The cancellation depends on both waves being advanced waves, so it's not purely terminological. (Advanced waves cannot cancel retarded waves in Cramer's formulation.) — Kenosha Kid

Why not? If I understood it correctly, the reflected wave is out of phase with the advanced wave, so it must cancel it. That the time direction is reversed means that the cancellation occurs everywhere at once, so that to an observer it is as if neither wave ever existed. Only the advanced wave back towards the BB is cancelled; the retarded wave from the emitter is out of phase with the advanced wave, which means that it is in phase with the reflected wave.

The cancellation depends on both waves being advanced waves, so it's not purely terminological. (Advanced waves cannot cancel retarded waves in Cramer's formulation.)
— Kenosha Kid

Why not? — SophistiCat

Difficult to say. He doesn't say more than 'We will not discuss Type II transactions further' or words to that effect in the original paper. (Recall that retarded and advanced waves cancelling are Type II transactions.)

But even if an absorber is available, the cumulative distribution of impacts will be defined only by the distribution of the available absorbers on the screen over time. — SophistiCat

Not *only*: the wavefunction of the emitted electron still natters; my point was rather that it can't be the *only* thing that matters.

In TQM itself, the probability of a transaction causing absorption at (r, t) is the amplitude of the retarded wavefunction arriving at (r, t) times the amplitude of the advanced wave travelling backward from (r, t). So it depends on the probability amplitude of *both* waves.

In your example of a screen that has only one absorption site at any one time, only this site can backwards-emit a hole wave. In the language of TQM, only this wave can handshake with the retarded wave, since the amplitude coming from all other sites is everywhere zero.

However, that single hole will move around the screen and, on average, should be smeared out such that the probability distribution we see forming is given only by the retarded wave.

And at the same time, in order for the Born rule to hold, that distribution has to match the impacting wavefunction - whatever it happens to be. If we can contrive to emit a particle that hits the screen at times (t1, t2, ...), the screen had better supply us absorbers at such locations ri that (r1, r2, ...) form the distribution that we expect to see. — SophistiCat

It's not that it's obliged to because of the experimental setup: it will do whether we fire electrons at it or not. In essence, this is what entropy is at the quantum mechanical level: the effectively random exploration of energetically equivalent microstates. If the screen, without us firing electrons at it, stayed in the exact same microstate, with the same single acceptor site, it would effectively be a highly ordered system. Another way to look at it is the fact that the hole is its own quasiparticle, with its own wavefunction obeying a wave equation. Just as we wouldn't expect an electron to stay put in the absence of a driving field, likewise we wouldn't expect the hole to stay put. It'll move around the screen just like an electron would.

does not quite hold: we cannot make emissions happen at will (can we?) — SophistiCat

In the case of no acceptor sites, this is what happens. It's an interesting property of current-carrying systems (which the double slit experiment basically is) as described in quantum transport theory (my particular field) that the notion of electrons being driven by external electric fields is redundant. In actual fact, all current-carrying systems behave probabilistically and thermodynamically: an electron leaves the cathode if and only if the anode has a hole available with the same energy. (Compare to a box partitioned and arranged such that only one half contains more particles than the other, both boxes being in their ground state, with particles filling up energy levels up to the box's characteristic energy. We open a hole in the partition, and higher energy particles from the higher energy box will move to fill holes in the lower energy box, but not vice versa due to Pauli exclusion. If those molecules were charged, you'd have a battery.)



Here's an illustration that contains the main point. Current (in natural units) here flows left to right not because the system exhibits an electric field across itself but purely because of the *chemical potential difference* between the source and sink. The electron source on the left has electrons filling energy levels higher than on the right, thus electrons move to the right, thus a current. If the source and sink levels were equalised, no current would flow (or, as is described by quantum transport theory, no *net* current will flow). If the sink level was higher than the source, electrons would move from right to left.

This is another example of electrons behaving as if they know where they're going before they leave, or rather not leaving because there's no place to go.

Here's an illustration that contains the main point. Current (in natural units) here flows left to right not because the system exhibits an electric field across itself but purely because of the *chemical potential difference* between the source and sink. The electron source on the left has electrons filling energy levels higher than on the right, thus electrons move to the right, thus a current. If the source and sink levels were equalised, no current would flow (or, as is described by quantum transport theory, no *net* current will flow). If the sink level was higher than the source, electrons would move from right to left. — Kenosha Kid

This is the point I made way back at the beginning of the thread, about radiant energy. The hot will only radiate to the cold, because of that disequilibrium, and reversal is not possible. But this is just a feature of how we understand that activity which is radiant energy. It is an epistemological feature which makes the idea of radiating energy into empty space somewhat incoherent, and it makes what is referred to as spontaneous emission, unintelligible, as random. But it is folly to say that the energy knows where it is going, because what is really the case is that we only understand radiation through its absorption; so the idea of radiation which appears like it does not know where it is going is something we haven't learned how to grasp yet, and is therefore unintelligible to us.

Not *only*: the wavefunction of the emitted electron still natters; my point was rather that it can't be the *only* thing that matters.

In TQM itself, the probability of a transaction causing absorption at (r, t) is the amplitude of the retarded wavefunction arriving at (r, t) times the amplitude of the advanced wave travelling backward from (r, t). So it depends on the probability amplitude of *both* waves. — Kenosha Kid

Well, the confirmation wave is just an echo of the offer wave: its amplitude is proportional to the amplitude of the offer wave at the would-be absorber. So the information carried back by confirmation waves is redundant, as far as the system as a whole is concerned - it's just a mechanism for establishing a transaction in accordance with the Born rule.

In your example of a screen that has only one absorption site at any one time, only this site can backwards-emit a hole wave. In the language of TQM, only this wave can handshake with the retarded wave, since the amplitude coming from all other sites is everywhere zero.

However, that single hole will move around the screen and, on average, should be smeared out such that the probability distribution we see forming is given only by the retarded wave. — Kenosha Kid

If the hole moves around independently of the impacting wave, while emissions are a Markov process, i.e. a transaction is established whenever a hole is available (as you explain below), with no "knowledge" of what comes before or after, then the resulting distribution of impacts will be independent of the impacting wave. It will only depend on the entropic movement of the hole - most likely just a uniform distribution.

Some dependence on the wavefunction will manifest when multiple holes are available at the same time, but unless there are a whole lot of them (rather than "not many"), the distribution will be blurred.

If the hole moves around independently of the impacting wave, while emissions are a Markov process, i.e. a transaction is established whenever a hole is available (as you explain below), with no "knowledge" of what comes before or after, then the resulting distribution of impacts will be independent of the impacting wave. It will only depend on the entropic movement of the hole - most likely just a uniform distribution. — SophistiCat

I'm not sure why you think so. The electron doesn't have to be transmitted at all. In fact, wherever the hole is located, we expect no electron to be transmitted most of the time. Any time the hole is at a site where the probability of finding the electron (as given by its wavefunction) is zero, then no transmission event will occur at all, for instance (i.e. you cannot slow the rate down to 0.001 Hz and get an event every 1 ms if the only available hole is sometimes inaccessible).

This is the analogy with the wire and the source and sink reservoirs. No electron ever leaves the source that cannot fit into the sink. (A nice sanity check that the idea that electrons know where they're going before they leave isn't too exotic.)

Similarly if the probability of finding the electron at a given site is 0.2, you would expect an electron to transmit there when there is a hole there at most 20% of the time.

In reality, the screen is more complex, and electrons will usually be able to squeeze in somewhere. But there should, as per Pauli, always be places it cannot squeeze, and that is neglected in ideal treatments.

I'm not sure why you think so. The electron doesn't have to be transmitted at all. In fact, wherever the hole is located, we expect no electron to be transmitted most of the time. Any time the hole is at a site where the probability of finding the electron (as given by its wavefunction) is zero, then no transmission event will occur at all, for instance (i.e. you cannot slow the rate down to 0.001 Hz and get an event every 1 ms if the only available hole is sometimes inaccessible). — Kenosha Kid

In our example of a diffraction through slits the wavefunction is non-zero almost everywhere on the back screen, so that is not an issue. If an electron is ready to fire, and there is (in the edge case) just one hole that it can fill, then it will go there almost always, because where else would it go? Which means that the distribution over time will just be the distribution of holes popping up on the screen. We could put one slit, or two, or ten - doesn't matter, the distribution will be the same.

(Unless holes and/or emission events conspire to construct the distribution that we expect to see.)

Similarly if the probability of finding the electron at a given site is 0.2, you would expect an electron to transmit there when there is a hole there at most 20% of the time. — Kenosha Kid

Not if that's the only place where it can go, or one of the few places where it can go. Perhaps the emitter is picky and won't always transact with a hole just because it's available? So in your example if the only available hole is at a 20% probability spot, then 80% of the time the electron will wait out until another hole opens up (and then make another probabilistic decision). That would work, but where is the mechanism for this process?

In reality, the screen is more complex, and electrons will usually be able to squeeze in somewhere. But there should, as per Pauli, always be places it cannot squeeze, and that is neglected in ideal treatments. — Kenosha Kid

Yes, but in order to explain experiments where we see nice diffraction patterns, we must conclude that the number of holes available at any given time is not too few, or else we would be seeing something different (or we need to modify the theory).

If an electron is ready to fire, and there is (in the edge case) just one hole that it can fill, then it will go there almost always, because where else would it go? — SophistiCat

That's my point, it doesn't have to go anywhere. For an electron to transmit, there will typically be a chemical potential difference between the source (here a cathode) and the destination (here a screen). If the only available site lies where the electron wavefunction is zero, no electron will transmit, exactly as if there are no available sites at all, which is the case when the source and sink are at equal chemical potential (zero voltage).

(Unless holes and/or emission events conspire to construct the distribution that we expect to see.) — SophistiCat

Effectively, yes. TQM is a conspiracy of sorts.

Perhaps the emitter is picky and won't always transact with a hole just because it's available? — SophistiCat

Exactly this. Quantum systems are whimsical :D

Yes, but in order to explain experiments where we see nice diffraction patterns, we must conclude that the number of holes available at any given time is not too few, or else we would be seeing something different (or we need to modify the theory). — SophistiCat

Well, as I said, so long as, over the lifetime of the experiment, the holes on average occupy a uniform distribution, we will obtain the characteristic banded pattern.

(Unless holes and/or emission events conspire to construct the distribution that we expect to see.) — SophistiCat

Sure they would appear to "conspire", because the fields associated with the emitting device would overlap and interact with the fields associated with the absorbing device. Fundamentally, the emitter cannot be separated from, to be considered as independent from, the absorber.

Effectively, yes. TQM is a conspiracy of sorts. — Kenosha Kid

Well, as I said, so long as, over the lifetime of the experiment, the holes on average occupy a uniform distribution, we will obtain the characteristic banded pattern. — Kenosha Kid

So if the holes (at the screen and elsewhere downstream) don't participate in the conspiracy (indeed, such an extended conspiracy would seem problematic), and there aren't many holes available at any one time, then the emitter has to time its transactions so as to build up the right pattern over time. Indeed, you need to posit that the process of picking the site of the transaction unfolds not instantaneously, as Cramer posits, but in all 3+1 dimensions. That's the only plausible solution that I can think of.

Is it plausible? I suppose the bare-bones theory (not including a specific mechanism for the Born rule) does not rule it out, and neither does its empirical basis, which consists of just such accumulation over time of apparently probabilistic events.

The appearance of conspiracy has a very limited representation in relation to what is real. As I said above, it represents something epistemic. What it represents is our inability to actually understand the true nature of emission and absorption. The one, absorption, is dependent on the other, emission, in its real existence, yet emission is not understood, and is represented as a mirror image of absorption, which has a dependence on emission, which is not understood. Therefore it is simply a vicious circle of misunderstanding, manifesting as the appearance of conspiracy.

What it represents is our inability to actually understand the true nature of emission and absorption. — Metaphysician Undercover

Therefore it is simply a vicious circle of misunderstanding, manifesting as the appearance of conspiracy — Metaphysician Undercover

From what little I know of the subject, these are reasonable comments.

(Oddly, I am working on a problem in elementary complex dynamical systems that has me caught up in a circular argument that must be broken: To show a certain sequence converges one needs to show it is bounded, but to prove it is bounded reflects back to its convergence behavior. It looks like a step-by-step argument alternating between the two will be the ticket.)

...To show a certain sequence converges one needs to show it is bounded, but the prove it is bounded reflects back to its convergence behavior. It looks like a step-by-step argument alternating between the two will be the ticket.) — jgill

That's exactly what experiment-based theorizing in science accomplishes. To see how it figures into my account of relative superposition amongst wavelengths and wavicles, give my most recent post in the "Anatomy of a Wave and Quantum Physics" thread a look. (Trying to get some more commentary on those ideas, which might associate in many ways with your guys' thread, though I haven't read much of this).

So if the holes (at the screen and elsewhere downstream) don't participate in the conspiracy (indeed, such an extended conspiracy would seem problematic), and there aren't many holes available at any one time, then the emitter has to time its transactions so as to build up the right pattern over time. — SophistiCat

Well, I'm being careful to distinguish between transmission and emission. Emission can be described as the spread of a single electron wavefunction from the tip of the cathode. Transmission is emission + absorption. In standard QM, transmission has occurred when we detect an electron on the screen. Emission by itself cannot, as MU keeps saying, be observed directly and independently (well, it can, but not without destroying the interference pattern).

So the electron wavefunction may well continue to evolve but simply not collapse. In TQM, the same holds: the retarded wavefunction can evolve indefinitely; it is only when the transaction with the advanced wave occurs that transmission occurs. As per the OP, the emission occurs precisely because the transmission occurs, i.e. it is simply one of the boundary conditions of a process that is agnostic about any arrow of time.

Is it plausible? I suppose the bare-bones theory (not including a specific mechanism for the Born rule) does not rule it out, and neither does its empirical basis, which consists of just such accumulation over time of apparently probabilistic events. — SophistiCat

True, hence my interest in Type II transactions, which, if they existed, should be empirically observable and presumably would differentiate TQM-like interpretations from others empirically.

But just to stress, the aim is not really to argue for a particular interpretation of QM. I don't really have any beliefs about it precisely because it is not something evidence can shine a light on, although the OP best describes a tentative ordering narrative going on in my little headbox. Rather, the point is about rejecting premature non-deterministic conclusions from naive quantum mechanical treatments of systems we cannot solve the many-body Schrödinger equation for. QM != non-determinism.

et al

Pardon my interjection, but could you guys briefly outline the properties of wave motion? How does the velocity or oscillation of an electron in an atom vary from one traveling in a beam or current, and how does this compare to electromagnetic radiation in various contexts? Could kinds of waves exist that travel faster or slower than what we have thus far measured? Why is the speed of light considered a top velocity in popular physics? Seems to me that wave mechanics are at the core of quantum foundations, so I want to get a summary handle on what waves do.

Transmission is emission + absorption. — Kenosha Kid

Don't you think that this is a sort of odd way of looking at things? We think of emission as occurring at a point in space and time, at the cathode, and absorption occurs at a point in space and time at the screen. However, there is a spatial-temporal separation between these points, and the concept of transmission accounts for that spatial-temporal extension. This would imply two distinct acts, emission and absorption, with a spatial-temporal separation between them.

To make emission/absorption into one single activity you need to dissolve the spatial-temporal separation between them. The cathode and screen must directly interact. Traditionally, the radio tower emits the information, the radio receives it, and the spatial-temporal separation between them is covered as transmission in the form of waves. Now, you could say that the wave field is a property of the transmitter, and this wave field interacts with the wave field of the receiver, and then you would have the premise for a direct interaction between transmitter and receiver, making your statement (above) true. The radio tower interacts directly with the radio, through the means of their wave properties, making emission and absorption one and the same act This would be like saying that the sun warms the earth by touching the earth, the electromagnetic field properly being a part of the sun which is in contact with the earth. This would negate the premise of space between the object and the eye; when seeing an object it actually touches the eye through field interaction. Also the idea that two distinct objects cannot occupy the same space would be faulty by this premise. This would allow an object to be understood in its entirety rather than just as it appears to our senses. We already know that objects really overlap each other, by the effects of gravity which is a property of objects.

Pardon my interjection, but could you guys briefly outline the properties of wave motion? How does the velocity or oscillation of an electron in an atom vary from one traveling in a beam or current, and how does this compare to electromagnetic radiation in various contexts? — Enrique

An electron in an atom is *bound*: unless it is supplied with enough energy (ionisation energy), it cannot move away from the atom. In this sense, it has no velocity, but it still has momentum, e.g. the angular momentum that, along with energy, identifies its state. Bound states have discretised energy levels: only certain energies are allowed.

By contrast, plane waves are *unbound*: they can have any energy and momentum. However, by definition, plane waves already occupy all of space, and so don't move anywhere either. In between these two extremes, electrons starting from a confined volume (such as an ionised electron) will spread out from the region of confinement (a circular wave) or move in a more well-defined direction and spread out as it does so (a wave-packet).

Frequency is just energy with different decorative physical constants.

Well, I'm being careful to distinguish between transmission and emission. Emission can be described as the spread of a single electron wavefunction from the tip of the cathode. Transmission is emission + absorption. In standard QM, transmission has occurred when we detect an electron on the screen. Emission by itself cannot, as MU keeps saying, be observed directly and independently (well, it can, but not without destroying the interference pattern).

So the electron wavefunction may well continue to evolve but simply not collapse. In TQM, the same holds: the retarded wavefunction can evolve indefinitely; it is only when the transaction with the advanced wave occurs that transmission occurs. As per the OP, the emission occurs precisely because the transmission occurs, i.e. it is simply one of the boundary conditions of a process that is agnostic about any arrow of time. — Kenosha Kid

As per Cramer (and his predecessors), there can be no emission without transmission in the absorber theory, whether classical or quantum. "Absorber theory, unlike conventional quantum mechanics, predicts that in a situation where there is a deficiency of future absorption in a particular spatial direction, there will be a corresponding decrease in emission in that direction." (I don't think you disagree, since that is also the premise of your hypothesis - just pointing this out, because what you wrote might suggest otherwise.)

In Type II transactions there is still an "absorber" - well, let's call it a "partner emitter."

I wonder though whether the absorber theory actually rules out, logically or empirically, uncollapsed/unabsorbed waves?

True, hence my interest in Type II transactions, which, if they existed, should be empirically observable and presumably would differentiate TQM-like interpretations from others empirically. — Kenosha Kid

By the way, in the 1980 paper Cramer wrote: "Davies argues that the most general test of absorber theory would include the possibility of type II transactions." This refers to a 1975 paper by Paul Davies: On recent experiments to detect advanced radiation. Perhaps that would be a good place to start digging in that direction. (Google Scholar doesn't list Cramer's paper among the references.)

It would be more accurate to say that in a situation where there is a deficiency of future absorption in a particular spatial direction, there will be a corresponding decrease in emission in that direction towards the future. There is always a past absorber in the Big Bang Universe in any spatial direction. In spatial directions where the radiation is reduced towards the future, the radiation towards the past is increased.
In the experimental attempt to detect advanced radiation in the late 70s (published in 1980), Schmidt and Newman used a half-wave dipole as an instrument of observation: "A search for advanced fields in electromagnetic radiation". With this, they introduced a future absorber in the spatial direction in which they tried to detect advanced radiation. So in that spatial direction they increased the radiation towards the future and reduced the radiation towards the past (figure 3), therefore it is no wonder that they had a negative result.



I made the same mistake in the first attempt to replicate their experiment. I almost gave up after numerous failed attempts. Then, after I read the paper: "Radiated power and radiation reaction forces of coherently oscillating charged particles in classical electrodynamics" I decided to replace the half-wave dipole with a lambda / 20 receiving antenna, which is a much more inefficient absorber, and positive results began to appear almost immediately: "Measurement of advanced electromagnetic radiation".

As per Cramer (and his predecessors), there can be no emission without transmission in the absorber theory, whether classical or quantum. "Absorber theory, unlike conventional quantum mechanics, predicts that in a situation where there is a deficiency of future absorption in a particular spatial direction, there will be a corresponding decrease in emission in that direction." (I don't think you disagree, since that is also the premise of your hypothesis - just pointing this out, because what you wrote might suggest otherwise.) — SophistiCat

Yes, I am in the sense that I would insist at least on an ultimate absorber: an electron hole (positron) for an electron, less so on the short-range behaviour Cramer describes. But even in TQM, the retarded wavefunction is emitted whether it transacts or not. It could potentially go on forever, but the emitting system would be considered as in the same state (no emission). By contrast, standard QM has a probabilistic emission: the emitting system is in a superposition of having emitted and not emitted. When the former is 0.9, we can be 90% sure emission has occurred, whether or not we detect a corresponding absorption (in principle), and it's for this reason I was being careful to discuss transaction.

In standard QM, the electron can be emitted without a future absorption event.
In TQM, the electron wavefunction is likewise emitted regardless of any future absorption event, but will only transact if there is a corresponding advanced wave.
In the OP, there is no emitted wavefunction without n absorbed one.

I wonder though whether the absorber theory actually rules out, logically or empirically, uncollapsed/unabsorbed waves? — SophistiCat

The wavefunction spreads out irrespective of future absorbers, just like in standard QM. It's just that the probability of emission does not approach 1 in the absence of absorbers in TQM.

By the way, in the 1980 paper Cramer wrote: "Davies argues that the most general test of absorber theory would include the possibility of type II transactions." This refers to a 1975 paper by Paul Davies: On recent experiments to detect advanced radiation. — SophistiCat

Ah, great minds. And my little one. Yes, it should be experimentally detectable and discernible from other interpretations of QM.

INTERESTING UPDATE

This concerns a paper from last year that only came to my attention today. The paper is here:

https://www.nature.com/articles/s41586-019-1287-z

and a frankly poorly written layperson article from Scientific American is here:

https://www.scientificamerican.com/article/new-views-of-quantum-jumps-challenge-core-tenets-of-physics/

(I actually found it's analogies harder to follow than the original paper, but others might get something out of it.)

A Wiki article on Quantum Trajectory Theory is here:

https://en.wikipedia.org/wiki/Quantum_Trajectory_Theory

This has nothing to do with advanced waves or reversible processes, but rather concerns a deterministic element of quantum transitions (historically the sole purview of the non-deterministic wavefunction collapse) and the role of the measurement apparatus in its unpredictable aspect.

TL;DR version: The experimenters built an electronic qubit and found that, when weakly coupled with the measurement apparatus, the transition between two if it's energy levels was both deterministic and predictable. This transition constitutes a trajectory from the lower to the higher energy level through a predictable continuum of superpositions in a finite amount of time.

When the measurement apparatus is more strongly coupled, this finite interval is no longer guaranteed: the system might never reach the end of the trajectory. Furthermore, at any time, it might collapse unpredictability back to its initial state.

Relevance: Part of the OP and much of my conversation with Cat discussed the importance of the precise state of the measurement apparatus to the experimental outcome, something that generally cannot be known because of the intractably large number of degrees of freedom (entropy). In the OP, this is discussed in the context of the transactional interpretation of QM, in which the evolution of the electron wave as it moves toward the screen is 'realised' (made real) by a time-reversed advanced wave coming from the screen to the electron source. Without knowledge of the advanced wavefunction, we cannot determine where the electron will go: the best we can muster is to realise the electron wavefunction with itself (the Born rule) and get a statistical distribution of possible outcomes.

One of the discussions Cat and I had regarded making the macroscopic screen more quantum by preparing it such that there was only one possible acceptor site, in which case one would expect a deterministic flow of electrons from the source to that one site, the rate given by the voltage between the source and the acceptor, and by the amplitude of the electron wavefunction at that site. (Here, keeping the site as an acceptor site is equivalent to giving it a continuous supply of advanced electron holes.)

This is not dissimilar to the experiment in this paper, which also describes a deterministic trajectory from source (here two nearby states |G> and |B>) to the only possible destination |D>. When there is negligible coupling between |D> and its environment, transitions to that state are smooth, deterministic and predictable. When it is more strongly coupled, the determinism is apparently destroyed.

We can liken this to an acceptor site that is inconstant in its supply of holes, or, rather, is coupled to its environment such that electrons from outside the apparatus can dip their toes in the acceptor site. Since this is not something we can possibly describe, it is effectively stochastic, giving us the apparent non-determinism we are familiar with in QM.

The importance of this paper is that it significantly narrows down the source of this apparent non-determinism to precisely the thing we are ignorant of: the measurement apparatus itself! In a perfectly deterministic universe, ignorance about important causal factors will yield unpredictable effects. We do not need an additional intrinsic non-determinism to explain the apparent probabilism of these sorts (i.e. reversible) of events.

Not sure if I should resurrect this thread, but feel the compulsion so who am I to resist? Reading a book on quantum physics that describes retarded/advanced wave theory, and while trying to envision this thought experiment it occurred to me that the mechanism seems to be the same as a lightning strike, though the rates of transmission may differ. Does lightning demonstrate this theory observationally, and what level of insight might be lent to a "quantum handshake" model of the double-slit experiment? Since I'm no expert at this point, I'll let you guys ponder the idea from scratch instead of awkwardly trying to explain it all at the outset and see what you think.

I followed along with this very interesting dialogue on the essence of symbolic reality. I sense sometimes in my journey the philosophical strokes which I just read engaged my mind much more completely than today. I wish to save this conversation to revisit at my leisure, but I'm afraid I'll lose part, if not all of it. If I provide a mailing address, would someone volunteer sending me a copy, or perhaps better, tell me how to safely save it in its entirety on my PC.thanks, abe

it occurred to me that the mechanism seems to be the same as a lightning strike — Enrique

Yes, very apt. The possible trajectories are explored down from the sky to the ground, but only once it reaches the ground does it become lightning. Of course, lightning will still explore different endpoints, but it's still a canny comparison.

You could just bookmark the URL? There's usually a little star or something in the address bar. What internet browser do you use.

If this is being preserved for posterity, I might go back and edit my embarrassing snapping at MU :fear:

...a canny comparison. — Kenosha Kid

I'm glad you liked it! I think I can actually specify this pretty well since I've been ruminating on it awhile.

Almost immediately after the retarded "wave" leaves (whichever side it starts from, probably the absorber side paradoxically, just as a lightning bolt originates from the ground, induced by the emission charge that is comparable to a thunder cloud), the complementary advanced wave arrives, then the two interfere to produce a new retarded wave in the forward direction (towards the emitter) while the retarded wave in the backward direction cancels out the original retarded wave.

The new advanced wave dissipates into the absorber somehow and similarly interferes with the emitter's retarded wave. The stair-stepping retarded/advanced waves from the emitter and those from the absorber rapidly close the distance between them until they contact each other and the quantum handshake occurs, a surge of charge briefly connecting the emitter and absorber directly, like a lightning bolt, in this case invisible. (whew, hopefully I didn't retard that description!)

The double-slit produces a symmetry in the chamber’s charge that makes the absorber’s charge-active sites comparably symmetrical, resulting in what looks like a precise interference pattern on the florescent screen despite the haphazard, seemingly randomized nature of each individual transmission “handshake” and its equal likelihood of passing through either slit (though abiding the path integral concept).

It would be fantastic to run this in slow motion or while measuring the charge distribution in the double-slit chamber, and maybe analyzing lightning could give a clue as to the coordination of causal vectors and rates in this type of process.

Some uncertainties I'm curious about:

Is an analogy between the double-slit mechanism which produces an interference pattern from a beam of light and that generating a statistical distribution from a beam of electrons entirely fallacious?

Does a statistical distribution similar to that of the electron experiment arise from molecules with up to a thousand atoms by this "lightning bolt" mechanism also, perhaps involving sizable charge polarities and "smearing" within the particle as it travels that temporarily reconstitute its structure into an entirely different form?

What do you guys think?

The double-slit experiment holds for light too. If you fire individual photons through the slits, you still get the interference patterns. It's not just an analogy: they genuinely behave the same, which speaks to a fundamental law.

The photon shows up as a single speck on the florescent screen, with an interference pattern built up as the statistics of many individual particles? The original double-slit experiment diffracted a beam of light into a spreading field before it reached the double-slit, so it was most certainly traveling through both slits simultaneously to interfere with itself, but the modern double-slit experiment could be different.

The photon shows up as a single speck on the florescent screen, with an interference pattern built up as the statistics of many individual particles? — Enrique

Exactly right.

The original double-slit experiment diffracted a beam of light into a spreading field before it reached the double-slit, so it was most certainly traveling through both slits simultaneously to interfere with itself, but the modern double-slit experiment could be different. — Enrique

What the low-frequency experiment tells us is that it isn't one photon from the beam interfering with another photon, but each photon interfering with itself, likewise for electrons. If there was an additional many-photon interference effect, it should be evident in the interference pattern on the screen.