nucleation

LIT-980: What is nonclassical about uncertainty relations?

2 min read

Properties5
StatusActive — read 2026-10-14 (NOTE-825) in full; worth reading as the clearest case where the form of an observed trade-off, not its existence, witnesses preparation contextuality, in a prepare-measure scenario with two binary measurements that are never made together, so that no sheaf or Kochen–Specker test applies. For a state with a realizable four-state reflection orbit, noncontextuality forces |⟨X⟩| + |⟨Z⟩| ≤ 1, with no perfectly distinguishable states needed. Read it knowing that the bound covers a whole uncertainty relation only in a theory where every state has the orbit, that the operational equivalence is used exactly and the claimed noise robustness is asserted, not shown, and that the three-measurement bound is verified by computer only. Filed for QUESTION-067; the account is THEORY-315.
Version2
Tagscontextuality, quantum-foundations
Cited as “paper” byNOTE-825 — Catani et al. — What is nonclassical about uncertainty relations?
Cited as “source” byTHEORY-315 — For a state whose four-state reflection orbit is realizable, generalized noncontextuality bounds two binary predictabilities by |⟨X⟩| + |⟨Z⟩| ≤ 1, so the form of a predictability trade-off, not its existence, can witness contextuality

Graph View

Lorenzo Catani, Matthew Leifer, Giovanni Scala, David Schmid and Robert W. Spekkens (2022), Physical Review Letters 129, 240401.

Key takeaways

From the reading, NOTE-825.

  • Existence is not the mark. The qubit stabilizer theory, Gaussian quantum mechanics and odd-prime stabilizer theories all have nontrivial predictability trade-offs and admit a noncontextual model (the latter two cited, the first exhibited).
  • The bound. If a state s1 has counterparts s2, s3, s4 with the same |⟨X⟩| and |⟨Z⟩| and the other sign patterns, and ½s1 + ½s3 = ½s2 + ½s4, then every preparation-noncontextual model gives |⟨X⟩| + |⟨Z⟩| ≤ 1 (Eq. 12). The proof is analytic, the bound is tight, and it needs no perfectly distinguishable states.
  • From a state to a theory. Where every state has such an orbit (A²₁ symmetry), Eq. 12 constrains the uncertainty relation itself. The qubit’s ⟨X⟩² + ⟨Z⟩² ≤ 1 violates it, by up to √2 at |⟨X⟩| = |⟨Z⟩| = 1/√2; the stabilizer octahedron saturates it; the gbit reaches 2.
  • Why classical joint predictability is no counterexample. The two-bit simplicial theory has perfectly predictable X and Z, but the states that achieve it fail the equivalence, so the bound does not apply to them.
  • Three measurements. |⟨X⟩| + |⟨Y⟩| + |⟨Z⟩| ≤ 1 under the analogous condition, verified by computational algebra, not proved.
  • Limits. Exact operational equivalence is assumed. The noise robustness is asserted with references, not shown.

Standing in the record

Filed on 2026-10-12 from the owner’s papers-feed for QUESTION-067, and read on 2026-10-14 in NOTE-825. It gives that question a worked instance of what it asks for: a set of operational equivalences among preparations and the noncontextuality inequality they imply, in a scenario where the two measurements are never made together and a sheaf or Kochen–Specker test has nothing to test. It also shows that one of the conditions QUESTION-067’s readings set, perfectly distinguishable decompositions, is not needed for every preparation inequality. The account filed from it is THEORY-315. It belongs to the program of Catani, Leifer, Schmid and Spekkens (LIT-019), and uses the definition of LIT-266.

Backlinks