Erdős Problem #97
Editorial depth: Reviewed explanatory article — evidence qualified.
Snapshot classification: falsifiable (not automatically the live result).
Subject areas: geometry, distances, convex.
Read this status correctly
The status above is reproduced from GCL's immutable external-catalogue snapshot. It is not a new claim of current mathematical resolution, a source-semantic review, an independent proof check, or GCL certification. A published proof and a GCL replay are separate events.
The question in plain language
At a vertex \(x\) of a convex polygon, several other vertices may all lie on a circle centered at \(x\). Let
Erdős #97 asks whether every convex polygon has at least one vertex \(x\) with \(m(x)\le3\). Put plainly: is it impossible for every vertex to have four other vertices at some common distance? The distance \(r\) may depend on \(x\).
The problem is marked FALSIFIABLE, not disproved: one finite convex polygon with \(m(x)\ge4\) for every vertex would settle it negatively.
A small example
In a regular pentagon, every vertex sees two neighbours at the side length and two other vertices at the diagonal length. Thus \(m(x)=2\) for all five vertices. This shows the conjectured property can hold, but does not prove it for every convex polygon.
The original Erdős conjecture asked for \(m(x)\le2\). Danzer constructed a nine-vertex convex polygon with \(m(x)\ge3\) at every vertex. Fishburn and Reeds subsequently constructed a 20-vertex convex polygon in which each vertex has three others at one fixed common distance. These invalidate the earlier 3-equal-neighbours exclusion, but neither supplies four equal-distance neighbours everywhere.
What is known
OPEN, finitely falsifiable. The maintained Erdős #97 carefully distinguishes these constructions from the unresolved four-neighbour threshold. In one 1975 paper, Erdős attributed to Danzer a much stronger all-threshold counterexample, but the maintained source notes that the claim was not repeated in later papers and is likely a historical error. It must not be cited as a proven disproof.
The source also records the conditional reduction: if a bound \(m(x)\le k\) holds at some vertex of every convex polygon, then deleting that vertex and inducting yields an upper bound \(kn\) on the number of pairs at a fixed distance. Consequently an affirmative answer to #97 would imply the linear version of #96 with coefficient at most three.
The original source leads are Erdős's 1946, 1975 and 1987 papers and Fishburn–Reeds, Unit distances between vertices of a convex polygon, Computational Geometry (1992), 81–91. Their exact coordinates and full proof have not been independently retrieved for this audit.
GCL's current state
The pinned catalogue classifies #97 as FALSIFIABLE and the maintained page agrees. The original assignment is #1366. We preserve the distinction between the superseded three-neighbour claim and the still-open four-neighbour claim. The strong historical Danzer attribution is explicitly unresolved.
The original issue contains an authenticated RESULT/1 source report. This is unadjudicated source evidence only, not independent mathematical certification or editorial admission.
What remains
A bounded successor is to implement an exact checker on points with rational coordinates: test strict convex position, and for each vertex group squared distances to the other vertices. Accept a refutation only when every vertex has at least four others at a single common squared distance. Otherwise report a finite fixture, not confirmation of the universal statement. The exact Danzer nine-point and Fishburn–Reeds twenty-point coordinates remain a separate primary-source replication task.
Sources and claim boundaries
- Erdős problem #97, original references and explicit warning about the 1975 Danzer claim.
- Erdős #97 bibliographic transcription, including [Er46b], [Er75f], [Er87b] and [FiRe92].
- Original source issue #1366.
No counterexample for the four-neighbour question, mathematical certification or editorial admission is claimed.
Catalogue and evidence provenance
- Pinned catalogue source: teorth/erdosproblems at
536408716724. - GCL issue: #1366 — protected issue/assignment identity; its labels and results are live, not reproduced automatically here.
- GCL triage: OUTSIDE_STRICT_OPEN_TRIAGE.
- Formal-statement coverage: A provider statement was included in the pinned snapshot. Presence does not mean the statement or proof has been checked.
- External mathematical status: consult current problem page and its dated bibliography.