Hello John.

Kudos for your impressive work describing the process matrix experiment, but there's something I think I'm missing. Maybe you can help me understand it better.

I don't see how this experiment states anything different than the Wheeler's delayed-choice experiment. It looks like a microscopic version of this experiment to me.

For instance, in the cosmic version of the delayed choice experiment (https://en.wikipedia.org/wiki/Wheeler%27s_delayed-choice_experiment#Cosmic_interferometer), the final outcome depends on the presence or absence of a beam splitter placed just before detection.

• If the beam splitter is present, the results imply that each photon from a distant quasar traveled as a wave through all possible paths around a galaxy located between the quasar and the detectors.
• If the beam splitter is absent, the results imply that each photon traveled through only one side of the galaxy.

The history we infer for each photon (the causal order of events, or where it was during its journey) seems to depend on a decision the observer can make after the fact.

Should we imagine the photon as if it is in a superposition of all possible paths, and also in any of the paths while it has not yet reached the place where the beam splitter may be?

Then we would have exactly the same image you show in figure 2 for the process matrix experiment.

There you show how the causal order (case A, case B or indefinite) depends on the state of the control qubit detected.

In the case of the delayed choice experiment, we have the same kind of inferences and outcomes:

• Case A (control qubit in state 0, implying WA causal order) would be no beam splitter present and sensor A detection (implying the photon followed just one path on one side of the galaxy).
• Case B (control qubit in state 1, implying WB causal order) would be no beam splitter present and sensor B detection (implying the photon followed one path on the other side of the galaxy).
• Indefinite order (control qubit in superposition, implying no causal order) would be the case when the beam splitter is present (implying the detected photons travelled as waves through all possible paths around the galaxy, and interfered with themselves at the beam-splitter position).

This puzzling behaviour (massless quanta choosing in the past how they must travel to meet later conditions) is usually explained with some kind of retrocausal mechanism (waves travelling backwards in time from detector to source), or a wavefunction that accounts for all the posibilities until a sudden collapse happens, or by postulating that all outcomes in the universe are already fixed since the Big Bang (superdeterminism).

But I think the real explanation is simpler than all that: we need all these tricks because we insist on analyzing the situation from a spacetime point of view.

But we should not forget that Special Relativity tells us that no time passes for photons and other massless quanta that seem to us to have traveled at the speed of light. This literally means that massless quanta don't know about our space and time dimensions. It means that photons and other massless quanta are, in fact, INSTANTANEOUS. They feel leaving emission and arriving at detection at the same time, so they don't know about one path or a thousand, because there is not such a concept for them. They don't have to choose from a superposition of possibilities or a sequence of events, because for them there's just one possibility and one event: the realization of their instantaneous existences.

It is up to us to decide how best to describe their instantaneous existences, how to translate their instant happenings to our spacetime scaffolding of distances and durations.