Everett’s many-worlds or multiverse theory is an attempt to find an alternative to the standard Copenhagen interpretation of quantum mechanics. Everett’s theory is often claimed to be local in the Bell sense. Here, we show that this is not the case and debunk the contradictions by analyzing in detail the Greenberger–Horne–Zeilinger (GHZ) nonlocality theorem. We discuss and compare different notions of locality often mixed in the Everettian literature and try to explain the nature of the confusion. We conclude with a discussion of probability and statistics in the many-worlds theory and stress that the strong symmetry existing between branches in the theory prohibits the definition of probability and that the theory cannot recover statistics. The only way out from this contradiction is to modify the theory by adding hidden variables à la Bohm and, as a consequence, the new theory is explicitly Bell-nonlocal.
The interpretation of measurements in quantum mechanics has been disputed for over a century. The scope of this study is to find an experimental test of Everett’s many-worlds interpretation that could potentially verify the existence of a multiverse. Our proposal is based on a quantum field theory formulation of the many-worlds interpretation through the path integral formalism and a careful choice of the vacuum state. By generalizing the earlier quantum suicide thought experiment to quantum field theory, we find that the many-worlds interpretation implies a measurable change in the lifetime of elementary particles.
Abstract One influential argument for the Everett interpretation of quantum mechanics is the idea that the Everettian view allows us to preserve locality. Evidently this argument assumes that preserving locality is something we should all aspire to—but what is the justification for treating locality as a theoretical desideratum? But there is no good reason for Everettians—or indeed anyone else—to aspire to “locality” as a general feature of their theory. It is true that certain specific ways in which nonlocality may be implemented have undesirable properties and should probably be avoided, but that does not mean that nonlocality itself is always undesirable—nonlocal theories should be assessed on their individual merits, rather than dismissed as a general class. Thus the “argument from locality” for the Everett interpretation is not compelling unless it is coupled with some justification for why the specific kind of locality instantiated by the Everettian picture is worth pursuing.