Erdős Problem #828
Editorial depth: Reviewed explanatory article — evidence qualified.
Snapshot classification: open (not automatically the live result).
Subject areas: number theory.
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
Does EVERY fixed integer a admit infinitely many positive n with phi(n) dividing n+a? The a=0 subcase has infinitely many trivial examples but does not solve all shifts. An auxiliary catalogue sentence incorrectly omits restrictions on a,b exponents when characterizing phi(n)|n.
A small example
phi(1)=1 divides1; phi(2)=1 divides2; phi(6)=2 divides6; phi(9)=6 does not divide9. For a=0 the infinite family 2^a (a>=1) works, but this proves nothing about every shift a.
The example can be checked exactly. A finite witness does not replace the universal, asymptotic, or infinitely-often quantifiers in the original question.
What is known
Maintained https://www.erdosproblems.com/latex/828 OPEN; Graham's conjecture. Lehmer's n-1 conjecture separately says phi(n)|n-1 only for primes, with n>1; open. Catalogue shorthand 'n=2^a3^b' is false if a=0,b>=1, e.g. n=3 gives phi(3)=2 not dividing 3. Exact positive n characterization: n=1 OR n=2^a 3^b with a>=1,b>=0; formal-conjectures Lean accompanying theorem states n<=1 or a>0. The formal conjecture's Lean sorry is not GCL proof replay.
The catalogue's auxiliary shorthand needs a boundary correction. For positive \(n\),
A positive power of 3 alone fails: \(\varphi(3)=2\) does not divide 3. The result follows by inspecting the factors \(p-1\) in \(\varphi(n)=\prod_{p^e\Vert n}p^{e-1}(p-1)\); every odd prime divisor supplies a factor two, constraining its possible values. Even a rigorous proof of the \(a=0\) shift case does not settle all other shifts \(a\in\mathbb Z\).
GCL's current state
The authenticated original issue-bound source return records STATUS_CONFLICT. Mathematical results attributed to published researchers are not re-labeled as new GCL proofs. This is an editorially reviewed explanatory article under protected batch EC-BULK-005-R5; its original STATUS_CONFLICT source remains preserved. Its historical source conflict is preserved; source checks and rendering do not certify a new theorem.
What remains
Provide elementary proof of exact phi(n)|n classification by factoring phi(n)=prod p^(e-1)(p-1) and ruling out primes p>=5 since p-1 cannot divide n without an impossible decreasing chain of prime factors. Validate finite n<=200, then source lock Graham / Lehmer statuses; never claim all shifts proved.
Independent critical review must compare the historical primary theorem to the written quantifiers, inspect the exact finite fixture, determine whether a claimed status refers to the whole question or a subcase, and validate the actual rendered mathematical output. No V2 editorial admission or formal theorem certification is claimed.
Sources and claim boundaries
- Maintained Erdős problem #828 and its LaTeX/bibliography, read 14 October 2026.
- Authenticated original GitHub source-return record. All original historical bibliography, quantified status limitations and exact source observations are recorded in that return.
- Pinned teorth/erdosproblems revision 536408716724a2402bdfa7c1d764bf83d64a8c89, data/problems.yaml. No protected source-lock alteration is claimed.
- Source theorem proof: PARTIAL and identified where unread. Editorial review: R5 protected PR #2668. Source publication: MERGED. Deployed MathJax: VERIFIED. External Lean certification by GCL: NONE.
Catalogue and evidence provenance
- Pinned catalogue source: teorth/erdosproblems at
536408716724. - GCL issue: #2099 — protected issue/assignment identity; its labels and results are live, not reproduced automatically here.
- GCL triage: PROTECTED_READY_BACKLOG.
- 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.