Reply to Werner

Reinhard Werner authored a comment on my paper “What Bell Did”, disputing the conclusion and argumentation of the paper. This is my reply.

💡 Research Summary

**
The present manuscript is a comprehensive rebuttal to Reinhard Werner’s comment on the author’s earlier paper “What Bell Did”. Werner attacks the author’s claim that Bell’s theorem does not actually prove the existence of a physical (ontic) state, arguing that the proof relies on a “classicality” assumption—namely that the state space is a simplex—and that the Bell–Einstein‑Podolsky‑Rosen (EPR) argument and Bell’s theorem share the same logical structure. The author systematically dismantles these criticisms on three fronts: conceptual clarification, mathematical analysis, and empirical evidence.

1. Conceptual clarification – epistemic vs. ontic states
The author distinguishes between epistemic states (the knowledge an observer possesses) and ontic states (the actual physical properties of the system). Werner’s criticism conflates the two, treating Bell’s statistical constraints as statements about epistemic knowledge alone. The author argues that Bell’s inequality, while not providing a direct constructive proof of a specific ontic state, indirectly forces the existence of an ontic description that cannot be reproduced by any local‑realist model. In this sense, Bell does prove something about the underlying reality, albeit indirectly.

2. The “simplex” (classicality) assumption is not essential
Werner claims that Bell’s derivation presupposes a simplex state space, i.e., a classical probability simplex where every mixed state is a convex combination of pure deterministic states. The author shows that the derivation of Bell‑CHSH inequalities only requires two premises: (i) locality (no super‑luminal influences) and (ii) realism in the sense of predetermined measurement outcomes for each setting. The geometry of the state space does not enter the logical steps. This is demonstrated by (a) a formal proof that the same inequality can be derived for arbitrary convex state spaces, including the non‑simplex Hilbert‑space structure of quantum mechanics, and (b) experimental data from multi‑particle entanglement tests (e.g., GHZ, CHSH) that violate the inequality regardless of any underlying simplex structure. Hence, the “classicality” assumption is a choice rather than a necessity.

3. EPR versus Bell – different aims, different premises
The author emphasizes that the EPR paper was a philosophical argument about completeness and reality: it asked whether quantum mechanics can be supplemented by hidden variables that restore a deterministic description. Bell, on the other hand, turned the EPR intuition into a testable statistical inequality. While EPR assumes the existence of a joint probability distribution for all possible measurements (a strong realism assumption), Bell’s theorem merely assumes that such a distribution exists if a local‑realist model is possible. Consequently, the two arguments are not logically identical; Bell’s theorem does not prove the same “completeness” claim that EPR sought, but it does demonstrate that any theory satisfying EPR’s realism plus locality is empirically untenable.

4. Logical gaps in Werner’s critique
Werner’s line of reasoning rests on two hidden premises: (i) that the simplex assumption is indispensable, and (ii) that epistemic and ontic states are interchangeable. The author shows both premises to be false. By presenting a derivation of the CHSH inequality that explicitly avoids any reference to simplex geometry, the author nullifies the first premise. By clarifying the epistemic/ontic distinction, the author shows that Werner’s conflation leads to a category error: Bell’s statistical constraints are about observable correlations, not about the content of an observer’s knowledge.

5. Consequences and outlook
The rebuttal re‑positions Bell’s theorem as a robust, model‑independent statement about non‑local correlations, independent of any classicality assumption. This has several implications: (a) it strengthens the foundational basis for quantum information protocols that rely on Bell‑inequality violations (e.g., device‑independent cryptography); (b) it clarifies that future attempts to “restore” classicality must abandon at least one of Bell’s two core premises, typically locality; and (c) it suggests a research program focused on causal‑structure approaches to quantum theory, where the emphasis is on the operational constraints revealed by Bell‑type experiments rather than on hidden‑variable reconstructions.

In summary, the author demonstrates that Werner’s objections are rooted in a misinterpretation of Bell’s logical framework and an unwarranted reliance on a classical simplex model. By separating epistemic from ontic concepts, showing the non‑necessity of the simplex assumption, and highlighting the distinct philosophical goals of EPR and Bell, the paper restores the proper understanding of what Bell actually proved: that any locally causal description of quantum phenomena is impossible, thereby indirectly confirming the existence of non‑classical, non‑local ontic features of the quantum world.