Two Forms of Inconsistency in Quantum Foundations

Jeremy Steeger, Nicholas TehThe British Journal for the Philosophy of Science 2018 · DOI · arXiv

Abstract

Recently, there has been some discussion of how Dutch book arguments might be used to demonstrate the rational incoherence of certain hidden variable models of quantum theory. In this paper, we argue that the ‘form of inconsistency’ underlying this alleged irrationality is deeply and comprehensively related to the more familiar ‘inconsistency’ phenomenon of contextuality. Our main result is that the hierarchy of contextuality due to Abramsky and Brandenburger (2011) corresponds to a hierarchy of additivity/convexity-violations that yields formal Dutch books of different strengths. We then use this result to provide a partial assessment of whether these formal Dutch books can be given a convincing normative interpretation.

Introduction: Contextuality vs. Dutch Bookability

In quantum foundations, "inconsistency" typically manifests in two distinct ways. The first is contextuality, the principle that the outcome of a measurement depends on the context—the set of other measurements performed alongside it. This is often visualized as a "gluing failure," where local data cannot be merged into a single global classical model, much like the geometry of an impossible object.

A geometric representation of contextuality, illustrating the 'gluing failure' where local data cannot be merged into a single global model.

The second form is Dutch Bookability, a concept from formal epistemology. An agent’s beliefs are considered "rationally incoherent" if they would accept a set of bets that guarantees a loss regardless of the outcome. Recent work suggested that certain quantum models are Dutch Bookable, implying they are "rationally incoherent." This paper provides a rigorous "translation manual" between these two fields, showing that Dutch Bookability is essentially contextuality in a different mathematical suit.

The Hierarchy of Inconsistency

The authors utilize the sheaf-theoretic framework of contextuality to prove that the three-tiered hierarchy of quantum contextuality corresponds exactly to a hierarchy of "Dutch Bookability" (violations of additivity and convexity in probability):

  • Strong Contextuality (e.g., Kochen-Specker, PR Box) $\iff$ Maximal Subadditivity Violation (the most extreme Dutch Book).
  • Logical Contextuality (e.g., Hardy’s model) $\iff$ Subadditivity Violation.
  • Probabilistic Contextuality (e.g., Bell models) $\iff$ Additivity Violation in monotonic extensions.
Comparison of the Bell model (probabilistic), Hardy model (logical), and PR box (strong), representing the three tiers of the contextuality hierarchy.

Key Insights and Results

The paper’s primary achievement is Theorem 1, which formally establishes the correspondence between the Abramsky-Brandenburger hierarchy and the hierarchy of additivity violations. It demonstrates that the geometric/topological "gluing failure" of contextuality is the same phenomenon that creates the probabilistic "Dutch Book."

Despite the mathematical equivalence, the authors express normative skepticism. They argue that being "Dutch Bookable" does not automatically render a quantum model "irrational." The normative weight of a Dutch Book argument depends on interpreting hidden variables as actual "states of affairs" that an agent should have beliefs about—a controversial philosophical stance. If these variables are treated as mere predictive tools, the charge of irrationality vanishes.

Limitations and Future Work

The scope of this paper is limited to factorizable noncontextual hidden variable theories. The authors suggest that future research could explore a "cohomological formulation of additivity-violation," further linking the topology of measurement spaces to the probabilistic inconsistencies found in quantum mechanics.

Summary updated

Abstract

Recently, there has been some discussion of how Dutch book arguments might be used to demonstrate the rational incoherence of certain hidden variable models of quantum theory (Feintzeig [2015]; Feintzeig and Fletcher [2017]; Wroński and Godziszewski [2017]). In this paper, we argue that the ‘form of inconsistency’ underlying this alleged irrationality is deeply and comprehensively related to the more familiar ‘inconsistency’ phenomenon of contextuality. Our main result is that the hierarchy of contextuality due to Abramsky and Brandenburger ([2011]) corresponds to a hierarchy of additivity/convexity-violations that yields formal Dutch books of different strengths. We then use this result to provide a partial assessment of whether these formal Dutch books can be given a convincing normative interpretation.
BibTeX
@article{steeger2018two,
  author = {Jeremy Steeger and Nicholas Teh},
  title = {{Two Forms of Inconsistency in Quantum Foundations}},
  journal = {The British Journal for the Philosophy of Science},
  year = {2018},
  doi = {10.1093/bjps/axy071},
}

Loading related content...

Figure