(1) “Since the consolidation of quantum field theory and the Standard Model, fundamental physics has mostly confirmed and refined an inherited ontology rather than replacing it. Many Beyond-Standard-Model programs have become too flexible, mathematically rich but empirically underconstrained.”
(2) “The missing task is not a naive return to pre-Copenhagen realism, because EPR, Bell, Kochen–Specker, and modern contextuality results rule out large parts of the classical ontology Einstein hoped to preserve. The real task is to construct a post-Bell ontology: one that explains quantum phenomena, measurement, entanglement, and spacetime without hiding behind operational formalism or unconstrained mathematical elegance.”
This I’d argue says that the tests of the Bell inequality violations are the most important experiments especially Aspect/Dalibard/Roger (1982) and Hensen et al. (2015) and something in 2023 (need to read that one). This implies ontological progress is held up by experimental insufficiencies, and this may not be resolvable without unforeseen technological developments that sit in sci-fi space at present.
P.S. A nice explanatory comparison is Bell and Turing (instead of the ML example), as their structural flavor (no-go-theorem vs contradiction) is similar:
Turing: no machine can decide the halting behavior of all machines on all inputs.
Bell: no theory satisfying the Bell-local assumptions can reproduce all quantum correlations observed in Bell-type experiments.
(The difference being that Turing’s theorem is a pure mathematical impossibility result over a precisely defined class of machines, while Bell’s theorem is a mathematical constraint on a physically motivated class of theories, with experiments as source-of-truth instead of mathematical proof. But this supports your position since Turing rules out total computable halting deciders, while Bell only rules out Bell-local hidden-variable theories satisfying measurement independence and ordinary probability constraints.)
No comments yet.
