Analyzing the correlations for spin-1/2 particles and singlet pairs

Abstract

We revisit the computation of correlations of spin projections onto unit vectors for spin-1/2 particles in Quantum Mechanics. We then choose one of the Boole inequalities that, as we recall, must be obeyed by collections of sequences of normalized spin-1/2 projections onto unit vectors that belong to, say, the plane that is orthogonal to the classical trajectory or to the three dimensional space. Next we define the concept of being a-p for i-indexed sequences of spin-1/2 particles: it turns out that a sequence of spin-1/2 particles is a-p if and only if the particles behave as if they had indeed been prepared (in the usual sense) to have definite spins +1 or -1 along the axis a, the sequence of signs being pre-determined. The a-p property is formulated in a way abstract enough to let one generalize it and we define the concept of being a-p and b-p. However, using our chosen Boole inequality we prove that no particle sequence can be both a-p and b-p when a and b are not co-linear. We next consider two modifications of Quantum Mechanics: - The first one by assuming both Weak Realism, the weakest form of Realism needed to develop any main theorem in Bell's Theory, and Locality. - The second one by assuming Weak Realism and an hypothesis that turns out to be none else than (relativistic) Causality extended to cover as well the observable values that make sense only if one assumes Weak Realism. In both cases we arrive at the conclusion that in the EPRB setting where successive pairs of spin-1/2 particles are prepared in the so-called singlet state, one of these particles sequences is both a-p and b-p and a and b are the opposite of the vectors respectively used to project the spin of the other particle and to get the Realism-based alternate value for a projection. Thus the usual discussions of Bell's inequalities under such hypotheses do not even make physical sense.