Lorenzo Catani, Matthew Leifer, David Schmid, Robert W. Spekkens (2021), Quantum 7, 1119 (2023) — ARXIV-2111.13727
Key takeaways
- A classical statistical theory of discrete field modes with an epistemic restriction reproduces the standard Mach-Zehnder interference phenomenology (and bomb tester, delayed choice, quantum eraser), so interference by itself does not force wave-particle duality, observer-dependent reality or retrocausation.
Seeded from the abstract and a skim, before the reading in NOTE-018. What follows is what the work says about itself.
Interference is often presented, following Feynman, as the central mystery of quantum mechanics and as forcing radical conclusions: photons toggling between particle and wave, observer-dependent reality, or nonlocal/retrocausal influences. The authors construct a “toy field theory”, a classical statistical theory of a discrete field, that reproduces the relevant interference phenomenology without those conclusions. Each field mode always has both a discrete occupation number (particle-like) and a discrete phase (wave-like), but an epistemic restriction forbids knowing both. Apparent nonlocal or backwards-in-time influences become inferences about distant or past systems, and only the observer’s knowledge, not reality, is observer-dependent.
Standing in the record
Filed by the reading-time triage of 2026-09-25: 1,515 seconds of active reading over 5 sessions in the papers-feed tracker. It was filed Deferred, unread. NOTE-018 is the close reading of 2026-09-25, and it placed the work: Active — It is the worked, experiment-by-experiment defence of placing interference in the weakly nonclassical column of LIT-007’s Table II, and it has the clearest statement of the Spekkens-school methodology (define classicality, formalize it, prove a no-go). Read §3, §4.3, §5.1 and App. A. Its noncontextuality claim for the toy field theory is asserted, not proved.
When seeded, before the reading in NOTE-018, the entry gave a priority for a deeper reading: medium — the owner spent substantial time (t=1515 s), but the paper is colloquium-level and the authors’ own “essence” sections (§1, 2.3, 3, 5) were covered by this skim; remaining depth is in the appendices.
What the seeding asked a deeper reading to check (NOTE-018 is that reading):
- Strongest claim: TRAP interference is classically explainable. Verify that the toy theory’s predictions match quantum ones for exactly the listed experiments and nowhere claims more (App. A-C).
- It is an application of item 14’s programme and leans on item 71-style classicality; reading it after those avoids redundancy.
- ML link: none. Its methodological point (state a criterion, formalize, prove a no-go, rather than infer “no classical explanation” from failing to find one) is a general one, and the ML connection is only by analogy.
Access when seeded: arXiv abs page and full PDF (61 pp.) retrieved; read outline, the author note (experts may read §1, 2.3, 3, 5), §3.1 description of the toy field theory, and §5.1 take-away. DOI and venue confirmed via Crossref.