Fabric of reality shadow photon partitioned off into parallel universes among themselves? In fabric of reality interpretation of QM chapter 2 how does David Deutsch  conclude that the "shadow photons" in the one-particle-at-a-time double slit experiments are themselves partitioned off into separate universes individually ?
If the property of the photons is that the "tangible photons" of a universe only interacts with the "shadow photon" counterpart, then in another universe where that particular "tangible photon" ("shadow photon" as seen from our universe) also interacts with other "shadow photons" as seen from that universe, so how can D.Deutsch come to the conclusion that as viewed from our universe, each "shadow photon" is partitioned off in a separate universe as opposed to the conclusion that all of the "shadow photons" come from the same universe ?
Can someone who has read and understood this chapter, help provide a way to understand how the "shadow photons" are partitioned off into parallel universes among themselves in terms of David Deutsch's interpretation of this phenomena ?
 A: I'm inclined to say that Deutsch is just wrong.
What he calls parallel universes in all of his writings are what everyone else calls paths or histories (in the context of the Feynman path integral / sum over histories). I don't know of anyone else who uses the term in the way he does. It doesn't match other concepts of parallel universes such as brane worlds or the worlds of the many-worlds picture.
What he calls the tangible photon is one of the terms in the sum over histories, and the shadow photons are the other terms. Since they are all terms on the same footing, it makes sense to say that the shadow photons are all related to each other in the same way that they are related to the tangible photon.
What doesn't make sense is singling out one of the histories as "tangible". There is no theoretical or experimental basis for that distinction. His justification for it seems to be that if you put detectors at each slit, at most one of them clicks per photon, and that shows where the "tangible photon" went. But if you put detectors at each slit, there is no multi-slit interference pattern on the screen. In the experiment without detectors at the slits, the pattern on the screen is different, so the behavior of the photons is evidently different, and it doesn't make sense to assume that what you found at the slits in the other experiment will still apply.
It's simply not true that one photon history is the one that triggers detectors, and the others merely interfere with that one. Whatever one may believe about the reality of individual histories, they're all equally real.
