DrChinese said:
When Swapping is enabled via BSM on photons 2/3, each and every 1/4 pair becomes entangled in one of 4 Bell states.
This is obviously true, but we must remember that there is no quantum state we can assign to a system unless we know how it was prepared. In the case of photons 1 and 4, if we consider their initial preparation at the start of the experiment, each pair is in a non-entangled (product) state, whereas if we take the swap performed on photons 2 and 3 as the "preparation", then each pair of photons 1 and 4 is in an entangled state corresponding to the Bell state measured on photons 2 and 3. Both statements are true simultaneously. The fact that we can treat the swap performed on photons 2 and 3 as a form of "future preparation" for photons 1 and 4, thereby assigning them an entangled state, does not in any way mean that the swap physically affects photons 1 and 4. As clearly stated in Ma's paper:
"
If one viewed the quantum state as a real physical object, one could get the paradoxical situation that future actions seem to have an influence on past and already irrevocably recorded events. However, there is never a paradox if the quantum state is viewed as no more than a `catalogue of our knowledge'."
In light of this, it is obvious that what
@martinbn says is correct. As also noted in Ma's paper, the initial quantum state can be rewritten (in the Bell state basis) as:
##\ket{\psi}_{1234} = \frac{1}{2} (\ket{\psi^+}_{14} \otimes \ket{\psi^+}_{23} - \ket{\psi^-}_{14} \otimes \ket{\psi^-}_{23} - \ket{\phi^+}_{14} \otimes \ket{\phi^+}_{23} + \ket{\phi^-}_{14} \otimes \ket{\phi^-}_{23})##
which means that the outcomes obtained by Alice and Bob on photons 1 and 4, respectively, can be separated into four sets exhibiting Bell's inequalities violations. Victor's decision to perform the swap on photons 2 and 3 physically changes the state of photons 2 and 3, allowing us to identify (postselect) the subsets of photons 1 and 4 that will show violation of Bell's inequalities. The authors are very clear about this point:
"
What, however, is important is to relate the lists of Alice, Bob and Victor's measurement results. On the basis of Victor's measurement settings and results, Alice and Bob can group their earlier and locally totally random results into subsets that each have a different meaning and interpretation. This formation of subsets is independent of the temporal order of the measurements."
As
@martinbn said, if Víctor does not perform the swap, the measurement results for photons 1 and 4 will continue to show subsets that violate Bell's inequalities (as concluded from the equation I wrote above), but there will be no "preparation" that allows them to be distinguished beforehand.
Lucas.