False statementConstruction: Literature-sourcedVerification: AI-adaptedPipeline-generatedprecheck passaudited 2026-09-30

How statement and proof provenance work

The first chip identifies the source of the statement or construction; the second identifies the source of its local proof or verification.

Literature-sourced: the exact statement appears in a cited source; only wording and notation differ.
AI-adapted: a semantically identical restatement of literature-sourced material, modulo indexing, notation, and boundary cases adopted by the library.
AI-generated: a genuinely novel statement formulated by AI, with no source for the claim itself.

These labels describe origin, not correctness: citations and verification chips remain separate evidence.

Nullity of the cut locus follows merely because it has empty interior

Statement

Assume the Axiom of Countable Choice ${A C}_{ω}$ , inherited from the cut-locus interface. The following universal claim is false: every subset of $R$ with empty interior has measure zero, that is, nullity is deducible from the emptiness of the interior alone. The claim fails already for closed sets: the Smith-Volterra-Cantor set is compact, perfect and nowhere dense, hence has empty interior, and it is not null. Consequently the nullity of the cut locus $Cut (p)$ cannot be obtained from the emptiness of its interior: the measure-theoretic argument of Cut locus of a point has riemannian volume zero is required to prove ${vol}_{g} (Cut (p)) = 0$ .

Facts & Assumptions

Given: The Axiom of Countable Choice; the Smith-Volterra-Cantor set $S \subseteq R$ ; a complete, connected, boundaryless, finite-dimensional Riemannian manifold $(M, g)$ , a point $p \in M$ , the unit tangent sphere $S_{p} M$ , the cut time $c_{p}$ and the cut locus $Cut (p)$ .

[A1]

The Axiom of Countable Choice ${A C}_{ω}$ is the standing assumption (The Axiom of Countable Choice ( ${A C}_{ω}$ )).

[F1]

$S$ is closed and bounded, hence compact, perfect, and nowhere dense (The Smith-Volterra-Cantor set is compact, perfect and nowhere dense, and does not have measure zero).

[F2]

In particular $S$ does not have measure zero (in the vocabulary of Measure zero (a countable cover by intervals of total length below every $ε$ ) and content zero (a finite such cover)): no cover of $S$ by intervals has total length below $2^{- 1}$ , let alone below every positive $ε$ (The Smith-Volterra-Cantor set is compact, perfect and nowhere dense, and does not have measure zero).

[F3]

$A \subseteq R$ is nowhere dense when the interior of its closure is empty, and a closed set is nowhere dense exactly when its interior is empty (Nowhere dense, meager (first category), residual, and second category subsets of $R$ ).

[F4]

$A \subseteq R$ has measure zero, equivalently is null, when for every real $ε > 0$ there are sequences $(a_{k})$ and $(b_{k})$ of reals with $a_{k} \leq b_{k}$ and $A \subseteq ⋃_{k} [a_{k}, b_{k}]$ whose lengths sum to at most $ε$ ; content zero is the same demand with a finite cover (Measure zero (a countable cover by intervals of total length below every $ε$ ) and content zero (a finite such cover)).

[F5]

The cut locus of $p$ is $Cut (p) = {\exp_{p} (c_{p} (v) v) : v \in S_{p} M, c_{p} (v) < \infty}$ , the finite cut-time endpoint set (Cut point and cut locus of a point).

[F6]

Assume ${A C}_{ω}$ ; for a complete, connected, boundaryless, finite-dimensional Riemannian manifold $(M, g)$ and $p \in M$ the cut locus has zero Riemannian volume (Cut locus of a point has riemannian volume zero).

[F7]

$S$ is the Smith-Volterra-Cantor set: the intersection $S = ⋂_{n \in N} S_{n}$ of the nested unions $S_{n} = ⋃_{j < N_{n}} [e_{j}^{(n)}, e_{j}^{(n)} + λ_{n}]$ of its construction (The Smith-Volterra-Cantor set: the same construction removing, at stage $n \geq 1$ , an open middle interval of length $4^{- n}$ from each of the $2^{n - 1}$ remaining intervals).

Refutation

technique · direct

1.1

The witness has empty interior. [F1, F3, F7] By [F1] the set $S$ is closed and nowhere dense, so [F3] applies to the closed set $S$ and gives that $S$ has empty interior. Thus the antecedent of the refuted claim holds for $S$ .

1.2

The witness is not null. [F2, F4] By [F2] the set $S$ does not have measure zero, and by [F4] having measure zero is exactly being null; hence $S$ is not null. The consequent of the refuted claim therefore fails for $S$ .

2.1

The implication fails, with an explicit witness. [step 1.1, step 1.2] The set $S$ has empty interior (step 1.1) and is not null (step 1.2), a closed, compact and perfect set. Hence "empty interior implies measure zero" is false; no topological smallness of that kind forces nullity.

3.1

Consequence for the cut locus, and boundary audit. [A1, F5, F6, step 2.1] By [F5] the cut locus $Cut (p)$ is the radial image of the finite cut-time endpoints, and by [F6], under the inherited [A1] and on the complete connected boundaryless manifold, it has zero Riemannian volume, proved in this library through the radial-graph measure argument. Step 2.1 shows that the emptiness of the interior of a closed set cannot replace such an argument, so the nullity of $Cut (p)$ does not follow merely because its interior is empty; the measure-theoretic input of [F6] is required. Boundary cases: the empty set has empty interior and is null, so it is not a witness; the exhibited witness $S$ is not null, hence nonempty, and the refutation never relies on the empty set. Zero-length intervals add nothing to the total length of a cover in [F4], although singleton intervals can cover nonempty countable sets. They do not evade [F2]: its lower bound applies to every countable interval cover of $S$ , including covers with degenerate intervals. The witness is one-dimensional, in $R$ ; no higher-dimensional or vector-space object is involved, and no distinguished "zero" element is used. Degenerate and endpoint conventions match: [F4] uses closed intervals $[a_{k}, b_{k}]$ with $a_{k} \leq b_{k}$ , endpoints included, exactly as [F2] does, so the quantitative lower bound of [F2] applies to every cover the definition of nullity allows. No choice is made in this refutation: the witness $S$ is a single set given by [F7]; the assumption [A1] is inherited from the cut-locus interface and is spent only through [F6]. The refuted claim is a universal implication and contains no biconditional, so there is no converse direction to check; the failing instance is the pair of steps 1.1 and 1.2, antecedent true and consequent false.

$□$

Source locator

Lee, Riemannian Manifolds: An Introduction to Curvature, Chapter 10, printed pp.173-190, defines the cut locus of a point as the set of finite cut-time endpoints, and Datar, Lectures on Riemannian Geometry, Section 23.3, printed pp.170-171, records that the cut locus may be discarded in measure-theoretic integration. The mathematical witness of this refutation is the library's own published theorem The Smith-Volterra-Cantor set is compact, perfect and nowhere dense, and does not have measure zero on the Smith-Volterra-Cantor set, whose construction is in The Smith-Volterra-Cantor set: the same construction removing, at stage $n \geq 1$ , an open middle interval of length $4^{- n}$ from each of the $2^{n - 1}$ remaining intervals; no source text is quoted, and the impossibility of deducing nullity from empty interior is exhibited by that witness.

Depends on

Used by

Nothing in the library uses this result yet.

Dependency tree · two levels

104 results within two dependency steps of this one, each drawn at its shortest distance from it. An arrow runs from a result to what uses it, so the chart reads left to right and ends at this result, which carries a heavier outline. Every node is a link to that result. Click elsewhere on the chart to enlarge it.

Sources

John M. Lee, Riemannian Manifolds: An Introduction to Curvature (1997), Chapter 10 (standard reference, not scraped)
Ved Datar, Lectures on Riemannian Geometry (2025) (standard reference, not scraped)