The paper argues that a causal explanation of the correlated outcomes of EPR-type experiments is desirable and possible. It shows how Bohmian mechanics and the GRW mass density theory offer such an explanation in terms of a non-local common cause.Keywords: Bell's theorem, Bohmian mechanics, EPR experiment, GRW mass density theory, local causality, non-local causes
Bell's theorem and the failure of local causalityBell's theorem (1964( ) (reprinted in Bell 2004 proves that any theory that complies with the experimentally confirmed predictions of quantum mechanics has to violate a principle of local causality. The idea behind this principle is that, in Bell's words, "the direct causes (and effects) of events are near by, and even the indirect causes (and effects) are no further away than permitted by the velocity of light" (Bell 2004, p. 239). This is one way of formulating the principle of local action that is implemented in classical field theories and that overcomes Newtonian action at a distance. Consider the EPR experiment: two elementary quantum systems are prepared in an entangled state at the source of the experiment (such as two systems of spin 1/2 in the singlet state). Later, when they are far apart in space so that there is no interaction any more between them, Alice chooses the parameter to measure in her wing of the experiment and obtains an outcome, and Bob does the same in his wing of the experiment. Alice's setting of her apparatus is separated by a spacelike interval from Bob's setting of his apparatus. The following figure illustrates this situation: Seevinck (2010, appendix) with permission of the author.
Figure 1: The situation that Bell considers in the proof of his theorem. Figure taken fromNon-local common cause explanations for EPR 2 In this figure, a stands for Alice's measurement setting, A for Alice's outcome, b stands for Bob's measurement setting, B for Bob's outcome, and λ ranges over whatever in the past may influence the behaviour of the measured quantum systems according to the theory under consideration (which may be standard quantum mechanics, or a theory that admits additional, so-called hidden variables) (see Norsen 2009 and Seevinck and Uffink 2011 for precisions).Bell's principle of local causality can then be formulated in the following manner:P a,b (A⏐B, λ) = P a (A⏐λ)P a,b (B⏐A, λ) = P b (B⏐λ) That is to say: the probabilities for Alice's outcome depend only on her measurement setting and λ. Adding Bob's setting and outcome does not change the probabilities for Alice's outcome. The same goes for Bob. Bell's theorem then proves that quantum mechanics violates (1). Furthermore, any theory that reproduces the well-confirmed experimental predictions of quantum mechanics has to violate (1). This conclusion applies not only to quantum mechanics, but also to quantum field theory (see Bell 2004, ch. 24, and Hofer-Szabó and Vescernyés 2013 as well as Lazarovici forthcoming for the current discussion). One can therefore say that Bell's theorem puts a constraint on an...