ON AN ALTERNATIVE INTERPRETATION OF THE BELL INEQUALITIES

An alternative interpretation is discussed on which the Bell inequalities are not consequences of local realism and on which Einstein-Podolsky-Rosen'-like experiments are no tests of.the inequalities, because the Bell inequalities only need to be satisfied if all observables are measured jointly. The violation of the Bell inequalities by quantum mechanics i generally interpreted asa demonstration of the impossibility to reproduce all results of quan-tum mechanics by means of a local realistic (hidden variables) theory. Moreover, the discrepancy between quantum mechanics and local realistic theories i thought o be liable to an experimental test by means of Einstein-Podolsky-Rosen-like experiments. The general agreement of these experiments with quan-tum mechanics i widely accepted as a proof of non-locality. In this note an alternative interpretation f the Bell inequalities i discussed, or which (i) the Bell inequalities are not a consequence oflocal realism but follow from the assumption that a joint measure-ment procedure xists for the four observables that are involved, and (ii) EPR-like experiments are not tests of the Bell inequalities because in such measure-ments only two observables are measured jointly. We will start our considerations from the point of view that a derivation of the Bell inequalities does not need a recourse to realism. Indeed, as was de-monstrated by Fine [ 1] by deriving the inequalities (in the generalized form deployed by Clauser and Home [2] ) from the mere existence of a joint prob-ability distribution of the four observables that are involved, the Bell inequalities have a meaning if we restrict ourselves to the phenomenalistic level of measurement results. Thus, if p(a 1, a 2, a 3, a 4) is the joint probability distribution of the measurement results of observables af 1, s ~ 2, af3 and ~f4, an