Skip to content

Erdős Problem #1108

Editorial depth: Authored explanatory DRAFT — NOT editorially accepted.
Snapshot classification: open (not automatically the live result).
Subject areas: number theory, factorials.

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.

UNACCEPTED DRAFT — mathematical editorial review pending

This material is a provisional, source-qualified manuscript published for line-by-line scrutiny and real HTML/LaTeX testing. It has NOT received separate GCL editorial acceptance or independent mathematical certification. Its presence on this website is not evidence of a solution, reviewed coverage, or final publication.

Review the underlying draft on GitHub.

The question in plain language

Make numbers by selecting distinct factorials and adding them. For example,

\[ 1!+3!+4!=1+6+24=31. \]

Let \(A\) contain every finite such subset sum. Erdős Problem #1108 asks whether \(A\) has only finitely many perfect \(k\) th powers for each fixed \(k\ge2\), and whether it contains only finitely many powerful numbers (integers whose prime divisors all occur squared).

These are different problems. A factorial sum can be powerful without being a perfect power.

A small example

Take factorials

\[ 1!,2!,\ldots,7!=1,2,6,24,120,720,5040. \]

They have \(2^7=128\) distinct subset sums, including the empty sum \(0\).

Exact integer-square tests find the following positive squares among those 128 values:

\[ 1,9,25,121,144,729,841,5041,5184. \]

Exact prime factorization finds 14 positive powerful values, including these squares and nonsquare examples such as \(8\) and \(27\).

These findings show that the finite factorial-sum set has repeated-power structure. They do not decide whether there are finitely or infinitely many such values when arbitrarily large factorial indices are allowed.

What is known

The original 1988 question is still open, including the special problem of whether infinitely many squares have the form \(1+n!\).

The catalogue reports an important partial result of Brindza and Erdős (1991): if the number of factorial summands is a fixed number \(r\), then a powerful sum of \(r\) distinct factorials has a bounded largest factorial index, where the bound can depend on \(r\).

That does not settle the full problem, because the set \(A\) permits an unbounded number of summands as the largest factorial index grows.

Erdős also discussed a related problem of Mahler involving sums of distinct powers \(k^j\) for a fixed base \(k\). The example

\[ 1+7+7^2+7^3=400=20^2 \]

belongs to that different setting and must not be confused with factorial sums.

GCL's current state

Authenticated source audit #2379 records the fixed-summand versus unrestricted-summand distinction and reproduces the exact finite-square and powerful-number tables.

The 1991 published proof and precise treatment of \(0!\), \(1!\), and the empty sum have not been independently audited. In the bounded check, the factorial indices are \(1,\ldots,7\), with \(0\) allowed only as an empty subset sum and excluded when counting positive powerful numbers.

What remains

Read Brindza and Erdős's 1991 theorem in full, confirm its indexing conventions and how the bound depends on the number of summands, and assess the still-open unrestricted problem. Larger finite searches may test hypotheses but cannot settle finiteness of the infinite intersection.

Sources and claim boundaries

Erdős Problem #1108, Oberwolfach 1988; Brindza–Erdős (1991) reported fixed-summand theorem; GCL return #2379. The exact finite subset-sum table is independently checked, not an asymptotic result. Open; unaccepted draft.

Catalogue and evidence provenance

  • Pinned catalogue source: teorth/erdosproblems at 536408716724.
  • GCL issue: #2379 — 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.

Back to catalogue · Compendium introduction