My understanding from careful study of Bells theorem is that there may be a deep subtlety that is not ruled out, but is probably very minimally studied in the literature. However the literature is vast, and ruling out that it is not studied at all seems questionable verging on unlikely.

  • suppose that the hidden variable(s) control whether the particle is detected. is this ruled out by bells thm?
  • if not, is that postulate enough to lead to a consistent theory, compatible with the known QM formalism?

There is a lot of analysis of "biased vs unbiased sampling" in some of the literature, but it seems reasonable to me to argue that its anthropomorphic thinking to talk about it in those terms, and maybe one should just start from the (actually very reasonable?) assumption that the only possible explanation is that the hidden variable(s) are controlling detection of the particle (ie presence vs absence). There are also some recently declared "loophole free bell tests", but I am also seriously wondering if those are not ruling out this above scenario either.

There may even be circumstantial evidence in the form of many, many experiments that have to carefully consider/ account for detection efficiency which is not really directly considered in bells original (thought) experiments. Maybe "detection efficiency" is really optical/ technical shorthand for "hidden variables controlling detection."

The above scenario is briefly discussed in the book The infamous boundary by Wick and was supposedly considered decades in the past and labled "a conspiracy theory" but again that seems like possibly biased thinking to me.

See also https://en.wikipedia.org/wiki/Loopholes_in_Bell_test_experiments .


In Bell's original experiment, the only measurement outcomes are +1 and −1, but you can still prove a version of Bell's inequality for an experiment in which the possible outcomes are +1, −1, and "no particle detected", and you discard pairs for which either result was "no particle detected" when doing the statistical analysis.

Note that the proof depends on no assumptions about the underlying cause of these measurement outcomes. The particles can lie about whether they're present or not, or hang around and affect the results of later measurements, etc.

There probably are loopholes in the attempts to test Bell's inequality to date, but for a local hidden variable theory to reproduce all of the predictions of quantum mechanics, it would have to be the case that no experiment for which a version of Bell's inequality can be proven can ever be conducted in practice. That's a pretty high bar.

| cite | improve this answer | |
  • $\begingroup$ thx for succinct conventional wisdom. however throwing out pairs where either is not detected seems like it runs the risk of creating an illusion of nonlocality (in remaining pairs) where none actually exists because its fundamentally not a local operation. have not seen this treated very carefully in references. forgot to mention that CHSH 1969 converts bells analysis into one that takes optical detector efficiency into account. believe it may be subject to this subtle caveat. en.wikipedia.org/wiki/CHSH_inequality $\endgroup$ – vzn Sep 11 at 18:01

Not the answer you're looking for? Browse other questions tagged or ask your own question.