Résumé
In 1953 de Broglie's student, Olivier Costa de Beauregard, raised what he took to be an objection to the EPR argument. He pointed out that the EPR assumption of Locality might fail, without action-at-a-distance, so long as the influence in question is allowed to take a zigzag path, via the past lightcones of the particles concerned. (He argued that considerations of time-symmetry counted in favour of this proposal.) As later writers pointed out, the same idea provides a loophole in Bell's Theorem, allowing a hidden variable theory to account for the Bell correlations, without irreducible spacelike influence. (The trick depends on the fact that retrocausal models reject an independence assumption on which Bell's Theorem depends, thereby blocking the derivation of Bell's Inequality.) Until recently, however, it seems to have gone unnoticed that there is a simple argument that shows that the quantum world must be retrocausal, if we accept three assumptions (one of them time-symmetry) that would have all seemed independently plausible to many physicists in the years following Einstein's 1905 discovery of the quantisation of light. While it is true that later developments in quantum theory provide ways of challenging these assumptions - different ways for different views of the ontology of the quantum world - it is interesting to ask whether this new argument provides a reason to re-examine the Costa de Beauregard's "Paris interpretation".
