22 June 2020: Paul Näger (Münster), “How Quantum Mechanics Solves the Causal Problem of Entanglement”

For the fourth FraMEPhys meeting of 2020, Dr. Paul Näger (University of Münster) gave a talk entitled “How Quantum Mechanics Solves the Causal Problem of Entanglement” via Zoom to the University of Birmingham FraMEPhys group and guests.

Recent works show that the statistics of typical experiments with entangled quantum objects (EPR/B experiments) contradict the usual principles of causal explanation, even if one disregards all spatio-temporal constraints (Wood & Spekkens 2015, Näger 2016). More precisely, this causal problem of entanglement consists in the fact that it is impossible that both central principles of the theory of causal Bayes nets (Glymour, Spirtes & Scheines 1993; Pearl 2000)—the causal Markov condition and the faithfulness condition—hold in such experiments. Any correct theory of the quantum realm must violate at least one of these conditions. This threatens the idea that the correlations in such experiments might be explained causally. In this talk I shall present a detailed analysis of the quantum mechanical formalism (in a GRW interpretation), revealing that quantum theory even violates both principles. Nevertheless, I shall argue for the claim that there are good reasons to regard the quantum mechanical explanation as a causal one. For the one, it is a well-known fact that the entangled quantum state does not screen off the correlations in such experiments. In other words, if quantum mechanics is complete, there is no screener-off for the correlations (van Fraassen 1982, Butterfield 1989, Cartwright 1989), implying that the theory violates the causal Markov condition (Spirtes, Glymour, Scheines 1993, Pearl 2000), which is a generalisation of Reichenbach’s principle of the common cause (Reichenbach 1956). Referring to the work of Cartwright (1988), however, I argue that in indeterministic worlds one should accept common causes that do not screen off. Further developing on Cartwright’s ideas, I present a generalisation of the Markov condition which is able to capture these new cases. This saves the central principle of causal explanation in the quantum realm, and makes explicit that underlying the quantum mechanical formalism is a causal structure that can explain the correlations. In a second step I show that the quantum mechanical formalism also violates the causal faithfulness condition. While being one of the central principles of the theory of causal Bayes nets, violating faithfulness does not seem to threaten a causal explanation per se: there are well-known counterexamples to the principle in perfectly causal situations. However, an unfaithfulness seems only acceptable in a causal explanation, when one indicates how it comes about (i.e. which type of unfaithfulness there is) given the causal connections in question; for not all types fit with all structures. Wood & Spekkens (2012) are tacit about which kind of unfaithfulness quantum mechanics involves; Näger (2015) claims that the theory involves an unfaithfulness of a supposedly new kind (unfaithfulness by internal cancelling paths), but only sketches its central features. In the present analysis I show explicitly how quantum mechanics explains the specific no-signalling independences by internal cancelling paths. I also provide an explanation for the unfaithfulness occurring between outcomes and local settings (for maximally entangled states), which reveals another so far unnoticed kind of unfaithfulness. In sum, my analysis shows that quantum mechanics solves the causal problem of entanglement in an astonishing and elegant way: though violating both central principles of causal explanation, the theory can still be considered as providing a causal explanation, if one moderately and reasonably modifies the original principles.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s