Causal Inference and EPR

Abstract

The status of causality in the EPR experiment has always been a source of controversy. A condition of local causality is implicit in the original EPR criterion of reality: “If, without in any way disturbing the system, we can predict with certainty (i.e., with probability equal to unity) the value of a physical quantity, then there exists an element of physical reality corresponding to this physical quantity.” In the EPR set-up both systems have separated and are no longer interacting so it is assumed that “no real change can take place in the second system in consequence of anything that may be done to the first system” [1, p. 779]. The non-disturbance clause in the antecedent is hence satisfied, and we may predict with certainty the values of properties in the distant wing. In other words: although the theory does not represent causal influences, there seems prima facie to be physical determination of values across a spatial gap. This notoriously led EPR to draw the conclusion that the theory is incomplete; but in the aftermath of ► Bell's theorem it is customary to draw the alternative conclusion — that there is non-local causation in nature. Indeed Bell's theorem has been the driving force of scepticism regarding local causality in the literature. In the last two decades the scepticism has linked up to a more general worry concerning the inference of causal hypotheses from statistical correlations in quantum mechanics. For physicists these issues matter to the evaluation of the compatibility of quantum mechanics with special relativity theory, and the prospects of a unified quantum gravitational theory. For philosophers these issues are key to a thorough assessment of the philosophical implications of quantum mechanics; and in addition EPR has become one benchmark against which all methodologies of causal inference are routinely tested.