Originally posted to 13.7 Cosmos & Culture on 2/21/2018

Who has no regrets about things done in the past? Wouldn't it be nice if, somehow, we could go back to tweak a couple of bad decisions?

This sounds (and as we will see, is, to a certain extent) like science fiction.

The laws of physics prohibit traveling backwards in time for many reasons. If we did travel backwards in time and changed the course of events, we would be altering the course of history. An example often cited is the grandfather's paradox: If you traveled back in time and murdered your grandfather when he was still a high school student, he wouldn't have met your grandmother and your father and you wouldn't exist.

A popular example of traveling back in time is the fascinating Canadian TV series Travelers: In a distant future, the Earth is in shambles; humans are controlled by a benevolent AI that finds a way to project the consciousness of people into unaware hosts in the 21st century. The idea is that travelers from the future take over the minds of people in the 21st century right before they are about to die. There is some obscure talk about quantum entanglement of consciousness between traveler and host, but this is secondary. The point of the show is that the travelers go back to try and change the course of history — so that the future looks better.

Putting humans or consciousness traveling back in time aside for the moment, is there anything in science even remotely similar? Surprisingly, yes. At the level of quantum particles (we are talking individual photons, elementary particles, or individual atoms), there is something called Wheeler's Delayed-Choice experiments that show that actions in the present can influence the past.

The experiments use something called the wave-particle duality of light and of matter, the fact that the physical nature of quantum objects is undetermined until it is measured. In other words, this means that a particle of light, or of matter, can behave either as a wave (spreading out in space, showing interference) or as a particle (staying together as a small bundle) depending on the measuring apparatus. Long and ongoing discussions about the nature of quantum physics are still trying to work out what this actually means. Do our minds determine the nature of physical reality? The interested reader can see what I wrote about this is more detail here.

But experiments measure — they don't ask questions of meaning. John Wheeler, the Princeton physicist who proposed such experiments in the 1970s, would have been amazed if he had seen the current results. It does seem that the present can influence the past, at least at the level of quantum objects.

The picture below explains the story. Imagine that there is a source of photons (or other small quantum particle). The photons can pass through a double slit. Behind the double slit there is a screen. If the photons hit the screen, those conducting the experiment observe an interference pattern of bright and dark fringes, typical of waves. If the screen is not there, and there are photon counters aligned with the slits, the photons will hit either one or the other, behaving like little bullets (or particles). So far, this is the typical set up for a double-slit experiment.

The "mystery which cannot go away" (as physicist Richard Feynman famously remarked) of the double-slit experiment, is that the person performing the experiment determines the physical nature of the particle — i.e., whether it is a wave or a particle. And, with Wheeler, the mystery deepens.

Imagine that the set-up — having or not having the screen — is decided after the photon goes through the slits. That's what the arrows in the diagram represent, the possibility that the screen is there — or is not. In 2007, a group in France did exactly that, letting a single photon pass through a double slit and then, after it passed through, having a random number generator choose whether the screen would be there or not to detect it. As Wheeler wrote, "Thus one decides the photon shall have come by one route or by both routes after it has already done its travel." Since then, many other groups (for example, here) have performed refined versions of the experiment, confirming Wheeler's intuition.

An important detail is that the switching over of detecting apparatus must be faster than the time the photon has to travel to the detectors. This way, there is no way the photon could "know" what to do. (If a photon knows anything, anyway.) Experiments presented in October extended the range of the photon's trip to about 2,200 miles, and still the photon seems to always choose the path consistent with the delayed choice. It is, as Mike McRae wrote in this Science Alert piece, "that even after the horse has bolted 2,200 miles out of the gate, it can still wait until the finish line to decide which race it ran." That is, which path to the finish line it took.

Of course, photons are not people, and to sustain a quantum superposition is very difficult, especially as the size of the object increases. Still, there is something quite amazing and mysterious about this behavior, where the path in space taken by an object seems to be impervious to time; it is as if the two choices (particle or wave; one slit or two) are suspended in time and are only enacted once the spatial arrangement is decided upon. No wonder Wheeler liked to called such ideas as being indicative of a "participatory universe", that is, of a universe where our minds are somehow deeply connected with the very fabric of space and time. After all, the choices of the apparatus may be made by a random number generator, but the apparatus and the interpretation of the data require our intent and design.

Food for thought for the future. Or maybe the past?

Unfortunately, these experiments say very little about how we could interfere with the past in events relevant to the human scale. Better to think carefully about decisions than to try to fix them backwards.