Entanglement Swapping and Action at a Distance

Abstract

A 2015 experiment by Hanson and Delft colleagues provided further confirmation that the quantum world violates the Bell inequalities, being the first Bell test to close two known experimental loopholes simultaneously. The experiment was also taken to provide new evidence of ‘spooky action at a distance’. Here we argue for caution about the latter claim. The Delft experiment relies on entanglement swapping, and our main claim is that this geometry introduces an additional loophole in the argument from violation of the Bell inequalities to action at a distance: the apparent action at a distance may be an artifact of ‘collider bias’. In the absence of retrocausality, the sensitivity of such experiments to this ‘Collider Loophole’ (CL) depends on the temporal relation between the entanglement swapping measurement C and the two measurements A and B between which we seek to infer a causal connection. CL looms large if the C is in the future of A and B, but not if C is in the past. The Delft experiment itself is the intermediate case, in which the separation is spacelike. We argue that this leaves it vulnerable to CL, unable to establish conclusively that it avoids it.

Entanglement and the Path Integral

“…Another argument against the conventional QM perspective is that somehow the existence of "entanglement" between Alice's and Bob's photons seems to depend on one's reference frame, as this would determine the precise timeordering of the three measurements in question. [8] Nevertheless, even with these arguments, we take it to still be conventional wisdom that there really is entanglement between the wings of the experiment in Figure 4, and there really is not entanglement between the wings of the experiment in Figure 5. [7] If this viewpoint is correct, it must mean the path integral account is essentially mistaken, or at least missing a key piece of physics hiding in instantaneous states.…”

“…But conventional QM takes ES and DCES to be quite different. ES seems to require that Alice's and Bob's photons are truly entangled, while DCES seems not to have any true entanglement between Alice and Bob's photon (instead explaining those correlations as a product of post-selection [8]). If "entanglement" is thought to be an objective property of a system, then these two experiments are inherently different, and the conventional QM perspective cannot be reconciled with the path integral perspective.…”

“…And yet the conventional QM account of DCES does not consider these two experiments to be essentially identical. In fact, they are widely thought to be quite different [7,8]. While QM takes it to be incontrovertible that Alice and Bob's photons are entangled in Figure 4, it would not conclude that Alice and Bob's photons were entangled in Figure 5.…”

“…The appearance of entanglement-like correlations between Alice and Bob's measurement is widely accepted to not be an indication of true entanglement, but rather a post-selection artifact, apparent only after the final measurement is made and then certain runs of the experiment are chosen for analysis. [8] (Recall, if all the data is analyzed, in either version of the experiment, there are no correlations between A and B. )…”

Bell state measurement locally explained

“…Many physicists are convinced that this process is non-local. This conviction is ultimately based on the assumption of Bell's theorem validity [2]. It states that quantum mechanics cannot be local because it cannot be described by local realistic models with hidden variables.…”