Quelques arguments en faveur de l'interprétation de la mécanique quantique par la ramification d'Everett

Abstract

In Part 1, Bell's inequality for a pair of protons starting from a singlet state is briefly described. Since it has not been possible so far to conclusively verify the inequality experimentally, the usual hidden-variables interpretation of quantum mechanics is tentatively abandoned in favour of Everett's ramification (or many-universe) interpretation. Several details need then to be made precise, among them the following: (1) ramification must propagate itself at the velocity of light to satisfy relativistic requirements; (2) the ramification process starts from those atoms of the detection screen where impacts are quantum-mechanically possible. These quantum-mechanically possible impacts become real impacts, while space-time is progressively split into several branches resulting in several copies of the observers and their apparatus. The future-cones (i.e. the spatial spheres growing at light velocity) which are splitting space-time are called ramifiers. In the simple case where both protons are submitted to spin measurements along the same direction, the known minus-one correlation between the measurements is realized. This happens not at the impacts, but only at the intersection of both ramifiers, the common future of the impacts; then the branch of spin + from the first impact is identified with the branch of spin − from the second one, and conversely.

Part 2 contains a cursory study of the case of a wave bearing only one particle. We recall the existence, in usual formalisms (e.g., of Gordon-Klein or of Dirac), of a conservative presence quadrivector, consisting of its time component (the presence density) and the current trivector. The screen being a tridimensional hypersurface, the probability of finding the impact is the flux of the presence quadrivector through the screen. According to our point of view, every possible impact does occur, and it becomes the origin of a branch in which this impact alone is observed; the wave is then rubbed out at the velocity of light. Part 2 ends with a comment on the logic specific to ramification, i.e. on the virgin form of the screen and wave: the form they would have if the impact never occurred.

We consider the impacts of a two-proton wave in Part 3 for the case of a spatial screen and again in Part 5 for the general case. The presence quadrivector is then replaced by a presence tensor with 16 components, propagating itself in configuration space. This tensor is conservative and symmetric, taking into account the indiscernibility of protons. All considerations from Part 2 are extended to the case involving two particles. In the case of a non-spatial screen, each impact is again the origin of a ramifier, generating at light velocity a wave reduced to one proton since only one second impact is still possible. In the common future of any pair of impacts, that wave is rubbed out.

In Parts 4 and 5, we examine how a wave is constantly renormalized, not only by the observing physicist, but by nature itself. This argues in favor of an interaction between branches, which could lead in the future to an experimental verification of ramification.

A note is added about Schrödinger's cat paradox and the selection by a ramifier of some of the components in a wave, e.g. in a ramification description of the Compton collision (Part 6). The paper concludes with a discussion of several critical objections to the ramification interpretation (Part 7).