Contextuality as an External Bookkeeping Cost under Fixed Shared-State Semantics

Song-Ju Kim

arxiv: 2601.20167 · v2 · submitted 2026-01-28 · 🪐 quant-ph · cs.AI· cs.IT· math.IT

Contextuality as an External Bookkeeping Cost under Fixed Shared-State Semantics

Song-Ju Kim This is my paper

Pith reviewed 2026-05-16 11:08 UTC · model grok-4.3

classification 🪐 quant-ph cs.AIcs.ITmath.IT

keywords contextualityclassical simulationmutual informationobstruction costfixed shared statelinear witnessquantum probability

0 comments

The pith

Contextuality requires a positive minimum external information cost in classical simulations that keep the shared internal state fixed.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper establishes that in a classical simulation model where the shared internal description stays fixed, reproducing contextual quantum statistics demands a minimum amount of mutual information between the measurement context and an auxiliary label. This quantity, called the obstruction cost, is bounded from below by any linear witness that separates the observed statistics from the set of behaviors achievable with zero such information. A sympathetic reader would care because the result supplies a quantitative, information-theoretic reading of contextuality as irreducible external bookkeeping rather than a purely qualitative feature of quantum theory. The analysis is deliberately narrow: it applies only to this minimal external-label architecture and makes no claim that the bound is tight or that the model exhausts all classical simulation possibilities.

Core claim

Under fixed shared-state semantics, any linear witness that separates the observed statistics from the zero-obstruction set yields a positive lower bound on the obstruction cost, defined as the minimum mutual information between the context and the auxiliary label that is required to reproduce those statistics in a minimal external-label simulation model.

What carries the argument

The obstruction cost, defined as the minimum mutual information between context and auxiliary label needed to reproduce observed statistics while the shared internal description remains fixed.

Load-bearing premise

The simulation is restricted to a minimal external-label model in which the shared internal description remains completely fixed and all context dependence is carried solely by an auxiliary label.

What would settle it

A concrete contextual distribution for which the minimum mutual information between context and auxiliary label required to match the statistics is exactly zero would falsify the lower bound.

read the original abstract

Contextuality is a central feature distinguishing quantum from classical probability theories, but its operational meaning is often stated only qualitatively. In this Letter, we study a simple information-theoretic question: how much additional contextual information must a classical simulation introduce when it tries to keep a shared internal description fixed across contexts? To make this question precise, we analyze a minimal external-label simulation model in which the remaining context dependence is carried only by an auxiliary label. For this model, we define an obstruction cost as the minimum mutual information between the context and the auxiliary label required to reproduce the observed statistics. We then prove a conservative quantitative lower bound: any linear witness that separates the observed statistics from the zero-obstruction set yields a positive lower bound on this cost. We do not claim that this bound is tight, and we do not claim that the simulation model covers every possible classical architecture. Its role is narrower and more explicit: under fixed shared-state semantics, contextuality can be read as a certificate of irreducible external bookkeeping cost in a simple and well-defined simulation model.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The paper turns contextuality into a mutual-information cost inside one fixed-state simulation model and shows linear witnesses give a lower bound on it.

read the letter

The main takeaway is that contextuality here means the smallest amount of mutual information an auxiliary label must carry when a classical simulator is forced to keep its shared internal description fixed across contexts. The paper defines this obstruction cost directly as that minimum mutual information and proves that any linear witness separating the data from the zero-cost set produces a positive lower bound on the cost. The argument follows from the witness being non-negative on the non-contextual set and negative on the observed statistics, which forces the label to carry some dependence to match the marginals. This framing is new in the literature they cite, and the paper handles it cleanly by staying inside one explicit model. It does not claim the bound is tight or that the model covers every classical simulation, which keeps the claims proportionate. The proof looks straightforward and internally consistent from the description. The obvious limitation is the model itself. Everything depends on the shared state staying completely fixed, with context dependence off-loaded entirely to the auxiliary label. That choice makes the cost well-defined, but it leaves open whether other architectures could reduce the cost by allowing the internal description to vary. The bound also inherits the strength of whatever witness is used, so it does not improve detection power. Readers working on resource theories of contextuality or on classical simulation overhead will find the mutual-information angle useful for organizing numbers. It is coherent on its own terms and supplies a concrete quantity without overreach, so it deserves a serious referee to check the derivation and compare it to existing measures like contextual fraction. I would send it to peer review.

Referee Report

0 major / 3 minor

Summary. The manuscript introduces a minimal external-label simulation model for contextuality in which the shared internal description remains fixed across contexts, with all remaining context dependence carried solely by an auxiliary label. It defines an obstruction cost as the minimum mutual information I(context; auxiliary label) required to reproduce the observed statistics. The central result is a conservative quantitative lower bound: any linear witness separating the observed statistics from the zero-obstruction set yields a strictly positive lower bound on this cost. The authors explicitly restrict the claim to this model, do not assert that the bound is tight, and do not claim coverage of all possible classical architectures.

Significance. If the lower bound holds within the stated scope, the work supplies a quantitative operational reading of contextuality as an irreducible external bookkeeping cost under fixed shared-state semantics. This is a modest but useful contribution that moves beyond purely qualitative distinctions between quantum and classical theories. Credit is due for the conservative nature of the bound (derived directly from linear witnesses), the explicit scoping of the model, and the absence of overclaiming. The approach may usefully inform resource analyses of classical simulations of contextual statistics.

minor comments (3)

[§2] §2 (model definition): the notation for the auxiliary label and its dependence on context should be introduced with an explicit equation or diagram to make the minimal external-label model fully precise before the lower-bound argument.
[§3] The proof of the lower bound (likely §3) relies on the linear functional being non-negative on the zero-cost set; a short explicit verification that the witness value directly implies a positive lower bound via the definition of mutual information would strengthen readability.
A brief comparison to prior information-theoretic quantifications of contextuality (e.g., those based on non-contextual hidden-variable models or communication cost) would help situate the obstruction cost within the existing literature.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for their supportive review, accurate summary of the manuscript's scope, and recommendation for minor revision. We appreciate the recognition of the conservative nature of the lower bound and the explicit limitations we placed on the model. No specific major comments were raised requiring point-by-point rebuttal.

Circularity Check

0 steps flagged

No significant circularity in the derivation chain

full rationale

The paper defines the obstruction cost explicitly as the minimum mutual information between context and auxiliary label needed to reproduce observed statistics under a fixed shared internal state. The claimed lower bound follows from applying any linear witness that is non-negative on the zero-obstruction set (by the model's own definition) and negative on the observed contextual statistics; this separation directly implies a positive lower bound on the defined mutual-information cost without any reduction to fitted parameters, self-citations, or ansatzes internal to the paper. The manuscript states its scope is narrow and does not claim tightness or coverage of all classical models, confirming the argument remains self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 1 invented entities

The model introduces an auxiliary label as a new bookkeeping device whose only justification is the simulation setup itself; mutual-information properties are standard but the cost definition is paper-specific.

axioms (1)

standard math Mutual information is non-negative and satisfies the usual chain-rule and data-processing inequalities
Invoked implicitly when defining the obstruction cost as a minimum mutual information

invented entities (1)

auxiliary label no independent evidence
purpose: to carry all context dependence while keeping the shared internal state fixed
Postulated in the minimal external-label simulation model; no independent falsifiable prediction outside the model is given

pith-pipeline@v0.9.0 · 5478 in / 1237 out tokens · 84473 ms · 2026-05-16T11:08:21.504555+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

IndisputableMonolith/Cost/FunctionalEquation.lean washburn_uniqueness_aczel unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

O_π(P) := inf η,τ I_π(C;M) subject to Eq. (3) reproducing P(y|c)
IndisputableMonolith/Foundation/RealityFromDistinction.lean reality_from_one_distinction unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

Theorem 1: O_π(P) ≥ (2/ln 2) ΔW(P)² / L_W,π² via Pinsker

What do these tags mean?

matches: The paper's claim is directly supported by a theorem in the formal canon.
supports: The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends: The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses: The paper appears to rely on the theorem as machinery.
contradicts: The paper's claim conflicts with a theorem or certificate in the canon.
unclear: Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

Contextual Control without Memory Growth in a Context-Switching Task
cs.AI 2026-04 unverdicted novelty 7.0

Intervention on a fixed-size recurrent state enables contextual control in sequential decisions without memory growth or direct context input.
Contextual Chain: Single-State Ledger Design for Mobile/IoT Networks with Frequent Partitions
cs.DC 2026-04 unverdicted novelty 6.0

Simulation at N=20 across 500 seeds finds that adaptive synchronization, not quarantine, primarily drives final agreement and recovery-time improvement after partitions in noisy regimes.