Time symmetry and the Einstein paradox. - II

Abstract

The characteristic difference between the paleoquantal calculation (addition of partial probabilities) and the neoquantal one (addition of partial amplitudes) for the correlation of photon polarizations in cascade transitions is derived in terms of elementary trigonometry. This deliberate use of simple formulae aims at a transparent rendering of the change in paradigm required by the so-called EPR paradox (which is truly the 1927 Einstein paradox), namely that 1) the two photons donot possess polarizations of their own when leaving the source C, butborrow onelater, when interacting with the analysersL andN; 2) the die is thus not cast atC, but later, atL andN; 3) the correlation between the measurements atL andN is tied throughC, in their common past. The tight connection between this « Einstein nonseparability » and the nonlocality in Feynman’s « theory of positrons » is demonstrated through an analysis of the e+e+ annihilation into two photons. Thus the Einstein paradoxcorresponds, in the « new wavelike probability calculus », to the Loschmid and Zermelo sort of paradox in the old probability calculus. That is, it contrasts theintrinsic time symmetry existing at the elementary level to thefactlike macroscopic time asymmetry. Our discussion deliberately by-passes the hidden-variable problem, our model in this being Einstein’s by-passing of the mechanical aether when proposing special relativity. We believe that here today, like there in 1905, the problem istayloring the wording after the (operationally good)mathematics. Moreover, that the change in paradigm, which is needed, comes through a victory of formalism over modelism.