EPR Revisited: Context-Indexed Elements of Reality and Operational Completeness

Mikołaj Sienicki, Krzysztof Sienicki · 2025

Read EPR Revisited: Context-Indexed Elements of Reality and Operational Completeness by Mikołaj Sienicki, Krzysztof Sienicki (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2511-01930.txt.

Introduction

Mikołaj Sienicki and Krzysztof Sienicki revisit the Einstein-Podolsky-Rosen argument, the 1935 challenge to quantum mechanics' completeness, with modern operational tools. The original argument leaned on elements of reality, fixed values a system supposedly carries either way. The authors replace those with context-indexed conditional states, the measurement assemblages of current quantum information theory.

Their updated reality criterion says: when Alice measures a chosen observable and gets an outcome, Bob's system must take a conditional state that ensures the matching outcome for that context. They assume operational completeness, which quantum mechanics satisfies for reachable assemblages. Then they prove rigorously a sharp result. In one-sided no-signaling theories, perfect predictions justify these context-indexed inferences. They never justify fixed values across all contexts. Examples include the qubit singlet with Pauli settings, a continuous-variable case via the Reid criteria, and a PR-box counterexample marking quantum theory's boundary.

Catalog record

CREATORMikołaj Sienicki, Krzysztof Sienicki
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE21 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2511-01930

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

EPR Revisited: Context-Indexed Elements of Reality and Operational Completeness Mikołaj Sienicki, Krzysztof Sienicki Abstract We offer a reworked, operational take on the classic Einstein–Podolsky–Rosen (EPR) argument—one that sidesteps the problematic leap from perfect predictability to the assumption of simultaneous, context-independent values for incompatible observables. The key revision lies in swapping EPR’s “elements of reality” with what we call context-indexed conditional states , collectively forming a measurement assemblage, denoted { σ a | x } \{\sigma_{a|x}\} . According to our Revised Reality Criterion (RRC), perfect predictability “without disturbance” should be understood more precisely: if Alice selects setting x x and obtains outcome a a , then Bob’s system ends up in a conditional state ω B | a , x \omega_{B|a,x} that guarantees the corresponding measurement outcome— but only within that specific context . Building on this, we define operational completeness as the requirement that a theory account for all physically accessible assemblages and their measurement statistics. Quantum theory meets this bar for the subclass of assemblages that are quantum-reachable, relying on the Born rule and standard state-update prescriptions. We show that any operational theory satisfying one-sided no-signalling (from Alice to Bob) will support the inference of context-indexed elements from EPR-style perfect predictions—but crucially, it won’t justify a universal, context-free value assignment across incompatible observables. What seems paradoxical in EPR’s original framing turns out to be a misreading of the boundary between local-hidden-state (LHS) assemblages and those that are steerable. To ground the framework, we include complete proofs, a qubit singlet case study using 2–3 Pauli settings alongside CJWR steering inequalities and noise thresholds, a continuous-variable example based on finite-squeezing Reid criteria, and a clean counterexample involving a PR-box–style assemblage that demonstrates where quantum theory falls short in terms of reachability. I Introduction The famous 1935 paper by Einstein, Podolsky, and Rosen (EPR) brought to light a fundamental tension involving three key concepts: completeness , locality or separability , and the predictions made by quantum mechanics (QM). At the heart of the paradox is a leap—from the ability to predict distant outcomes with certainty, to the assertion of simultaneous, context-free values for incompatible observables. Our goal is to offer a refined account that (i) steers clear of the original EPR phrasing, (ii) is fully grounded in operational terms and lends itself to empirical testing, and (iii) remains faithful to the formalism and spirit of standard QM. Contributions. (1) We introduce a Revised Reality Criterion (RRC), reframing “elements of reality” as context-indexed conditional states . (2) We define a notion of operational completeness and show that quantum theory satisfies it across all quantum-reachable assemblages. (3) We establish a No-Paradox Theorem , demonstrating that while EPR-type perfect predictability implies context-sensitive elements, it does not imply global, context-free value assignments. (4) To bring the ideas closer to experiment, we supply diagnostic tools via steering inequalities , including a qubit case study with two and three Pauli settings and realistic noise thresholds, plus a continuous-variable (CV) illustration using finite squeezing. (5) Finally, we clarify and cite the formal equivalence between incompatibility and steering used throughout the argument. II Related work The EPR thought experiment and Bohr’s swift response mark the conceptual battleground of the early quantum debate Einstein et al. 1935 ; Bohr 1935 . This tension gained empirical teeth through Bell’s theorem and its associated inequalities Bell 1964 . Schrödinger, meanwhile, coined the term steering to describe the remote preparation of states across entangled systems Schrödinger 1935 . In its modern, operational form, steering is captured through assemblages and local-hidden-state (LHS) models Wiseman et al. 2007 , and has since been deeply linked to measurement incompatibility Uola et al. 2014 ; Quintino et al. 2014 . For broader perspectives, see comprehensive reviews such as Cavalcanti and Skrzypczyk 2017 ; Uola et al. 2020 , and for a focused discussion on measurement incompatibility as a quantum resource, see Heinosaari et al. 2016 . III Operational framework We adopt a general, GPT-like operational framework in which quantum mechanics appears as a special instance. In this setting, physical systems are characterized by state spaces, effect spaces, and instruments—where an instrument refers to a measurement together with a rule for post-measurement state updates. We consider two agents, Alice (A) and Bob (B), who share a bipartite preparation. Notation. Let \rho_{AB} denote a joint state shared by Alice and Bob. Alice’s measurement choices are labeled by x , with outcomes a , and her effects are written M_{a|x} (these are POVM elements in QM). Given an outcome (a,x) , Bob’s resulting normalized conditional state is denoted \omega_{B|a,x} . The associated unnormalized version is given by: \sigma_{a|x}:=p(a|x)\,\omega_{B|a,x},\qquad p(a|x)=\tr[\sigma_{a|x}], (1) and the full assemblage \{\sigma_{a|x}\} consists of all such \sigma_{a|x} across Alice’s available settings x . Bob’s measurement effects are denoted \{e_{b|y}\} , and in the quantum setting, these correspond to operators E_{b|y} , with the evaluation rule e(\omega)=\tr[E\,\omega] . For POVMs with multiple outcomes, we say there is certainty for a given measurement setting y when exactly one outcome occurs with probability 1 , and all the others for that same y occur with probability 0 . Because the probabilities across outcomes must sum to one (see Definition 5 ), this notion of exclusivity is inherently per-POVM . That said, it doesn’t prevent simultaneous certainty across effects from different POVMs, provided the system’s state lies in the intersection of the relevant 1 -eigenspaces. Remark 1 (Unit-probability condition; infinite-dimensional subtlety) . Given any POVM element 0\leq E\leq I , the supremum of \tr[E\rho] over all states \rho is simply the operator norm \lVert E\rVert_{\infty}\leq 1 . If there exists a density operator \rho such that \tr[E\rho]=1 , then it must be that \lVert E\rVert_{\infty}=1 , and \rho is supported entirely within the eigenspace of E associated with eigenvalue 1 . However, in infinite-dimensional settings, it’s possible for \lVert E\rVert_{\infty}=1 even when there’s no actual eigenvector with eigenvalue 1 —this happens when the essential supremum hits 1. In such cases, no normal state achieves unit probability. Definition 1 (Assemblage) . Suppose Alice and Bob share a bipartite preparation. When Alice performs a measurement labeled by x and obtains outcome a , Bob’s unnormalized conditional state is given by \sigma_{a|x}:=p(a|x)\,\omega_{B|a,x} . The condition of operational no-signalling from Alice to Bob then requires that the marginal \sum_{a}\sigma_{a|x} does not depend on Alice’s setting x . Definition 2 (Local-Hidden-State (LHS) model) . An assemblage \{\sigma_{a|x}\} admits an LHS model if there exists a hidden variable \lambda distributed according to p(\lambda) , response functions p(a|x,\lambda) with \sum_{a}p(a|x,\lambda)=1 , and a set of local normalized states \omega_{\lambda} for Bob, such that \sigma_{a|x}=\sum_{\lambda}p(\lambda)\,p(a|x,\lambda)\,\omega_{\lambda}\quad\forall(a,x). (2) Definition 3 (Revised Reality Criterion (RRC)) . Whenever a measurement outcome on Alice’s side allows perfect, unit-probability prediction for some test on Bob’s side, we say that Bob’s system holds a context-indexed element of reality —that is, the conditional state \omega_{B|a,x} tied to the specific context (x,a) . Definition 4 (Operational completeness) . Let \mathcal{A} be an operational arena —a specified set of joint preparations and instruments on systems A and B . A theory is said to be operationally complete (relative to \mathcal{A} ) if, for any assemblage \{\sigma_{a|x}\} generated by \mathcal{A} that satisfies one-sided no-signalling, the theory provides both a representation of

Read the complete text