Scientific Realism and Quantum Mechanics: Revisiting a Controversial Relation

ABSTRACT

The article examines the controversial relation of scientific realism with quantum mechanics. To this end, two distinct discussions are invoked: the discussion about ‘realism’ in the context of quantum mechanics and the discussion about ‘scientific realism’ in the context of the general philosophy of science. The aim is to distinguish them in order, first, to argue that the former—revolving around ‘local realism’ and the theorems of Bell and Kochen–Specker—unjustifiably identifies realism with features of a particular worldview, and thereby fosters the impression that the failure of ‘local realism’ in quantum experiments constitutes a failure of scientific realism too; and, second, in light of the latter discussion, to claim that scientific realism and quantum mechanics can be compatible—there is ground for dealing with quantum measurement simply as a physical mind-independent interaction. Therefore, a realist approach to the theory is possible despite its notorious measurement problem, even if the issue of its interpretation is still disputed.

Acknowledgements

I am deeply indebted to Aristidis Arageorgis, Stathis Psillos, and Vassilis Sakellariou for their constant encouragement and invaluable remarks on earlier versions of this paper. I wish to thank three anonymous referees for ISPS for providing me with thoughtful comments and criticism. I also wish to thank the editor, James McAllister, for his patience and kind assistance in finalising the paper.

Notes

1 Clauser et al. (1969) offered a version of Bell’s inequality for ‘deterministic’ hidden variables theories (the CHSH inequality) that could apply to realisable experiments.

2 I do not include in the ‘main discrepancies’ the intrinsic probabilistic character of quantum theory because it does not result from the two above-mentioned theorems.

3 Perhaps this is not the place to argue about this; suffice it to say that it is reflected in the way scientists exercise their method to acquire knowledge of the world. In broad terms, they are gathering data on a particular phenomenon and studying them in order to discover regularities allowing them to propose generalisations, testing them again and, if they are not falsified, formulating a scientific theory about the phenomenon.

4 Pessimistic meta-induction is a powerful argument that aims at undermining the commitment of scientific realism to unobservable entities, hence the explanationist defence of realism. According to the argument, the history of science teaches us that many of the unobservable entities that were posited by empirically successful theories of the past turned out to be false and were abandoned. Therefore, it is most likely that the same will happen with many of the unobservable entities of our current successful scientific theories; that is, terms alluding to them will turn out to be false and non-referential. In general terms, scientific realists respond to the pessimistic meta-induction by insisting that they do not commit themselves to all unobservable entities of modern science but only to those which have ‘confidently’ and ‘necessarily’ been posited, allowing for ontological error in science (Devitt 1997, 109).

5 Metaphysical scepticism also results from the second argument against scientific realism: the underdetermination of theories by evidence. According to this, two rival theories that are empirically equivalent—i.e. are equally supported by all observational evidence—entail the same observational consequences and are epistemically equivalent too; hence, we do not know how the world actually is since there are no good reasons to believe in one theory rather than another.

6 Travis Norsen (2007, 311) focuses on the question ‘what exactly is the “realism” in “local realism” supposed to mean?’. After surveying four different ‘realism’ concepts (naïve, scientific, perceptual, and metaphysical) he concludes that ‘none of them has provided a promising candidate for what users of the phrase “local realism” mean by “realism”’ (Norsen 2007, 335). And he continues by saying that that leads him ‘to speculate that the users of that phrase don’t, themselves, know what they mean, and that the phrase has in fact, become widespread through sheer, unthinking inertia’ (Norsen 2007, 335). I disagree concerning the users of the phrase. As detailed in this section, they know exactly what they mean when they use the phrase ‘local realism’. They give descriptions of the part of ‘realism’ and of the ‘local’ part. Also, the concept of scientific realism that Norsen (2007, 320) provides is too narrow, compared to the modern view of scientific realism described in the previous section, to draw any conclusions. Nevertheless, for different reasons, I agree with Norsen (2007, 314) that the use of the term ‘realism’ has no valid place in the discussions about Bell’s theorem and ‘that the underlying confusions are quite serious’.

7 Of course Jarrett refers to the paper of Clauser and Shimony (1978). It is worth recalling that Jarrett’s dissertation in 1983 was ‘Bell’s Theorem, Quantum Mechanics, and Local Realism’.

8 Condition that Jarrett extracted from the analysis of Bell’s ‘locality’. I am one of those who believe that Jarrett’s naming of this condition as ‘completeness’—hence ‘incompleteness’ for its violation—is unfortunate and in certain respects misleading.

9 Simon Gröblacher et al. (2007) carried out such an experiment and the results led them to the conclusion that giving up the concept of ‘locality’ is not sufficient for consistency with quantum experiments. Certain features of ‘realistic’ descriptions, such as the concept of ensembles of particles carrying definite properties, could fail and must be abandoned (Gröblacher et al. 2007, 871–875). Federico Laudisa claims that Leggett and Gröblacher et al. base their conclusions on a mistaken understanding of Bell’s theorem, because the theorem ‘has demonstrably nothing to do with the “realism” as defined by these authors’ (Laudisa 2008, 1110). According to Laudisa (2008, 1113), the Gröblacher et al. approach to ‘realism’ (called by him Realism_G&A and presupposing pre-existing properties for the physical systems) ‘is derived and not assumed’ in the EPR-Bohm argument and Bell theorem. Matthias Egg (2013) defended the foundational significance of Leggett’s work, and Laudisa (2014) replied to Egg. The last thing I want is to join the dispute, since the ‘realism’ I am concerned with is the scientific realism I described and not an assumption about pre-existing properties called ‘realist’. But, even if the Bell or the CHSH inequality ‘can well be derived from the only assumption of locality’ (Laudisa 2014, 301), we must not forget that ‘Bell’s locality assumption’ conflates two weaker conditions, named locality and separability by Howard (see section 2 above). As mentioned above, the feature of non-separability that two entangled quantum systems manifest is closely related to the feature of contextuality that Kochen-Specker’s theorem pointed out. The context dependence in value assignments to observables of a quantum system is expressed in the case of two sub-systems as non-separability. Therefore, ‘Bell’s locality assumption’ involves matters of assigning definite values (or probabilities) of possible measurement outcomes to quantum systems that share an entangled state; in other words, it involves matters concerning ‘realism’ as defined in the context of ‘local realism’. So, in my opinion, a significant aspect of Leggett’s work is that, by relaxing his postulate of locality together with the violation of his inequality in experiments, as it also happens in local cases, he redirects us to seek inconsistency with quantum experiments in other pertinent factors—‘realism’, say.

Additional information

Funding

Research for this paper has been co-financed by the European Union (European Social Fund—ESF) and Greek national funds through the Operational Program ‘Education and Lifelong Learning’ of the National Strategic Reference Framework (NSRF)—Research Funding Program: THALIS—University of Athens—Aspects and Prospects of Realism in the Philosophy of Science and Mathematics.