From: Tobias Nipkow <nipkow@in.tum.de>
Subject: [isabelle] New in the AFP: The Abel–Ruffini theorem and the unsolvability,of the quintic
The Abel–Ruffini theorem and the unsolvability of the quintic
Lawrence C. Paulson
This entry proves the Abel–Ruffini theorem — a polynomial equation solvable by
radicals has a solvable Galois group — and applies it to a concrete quintic to
conclude that the general quintic is not solvable by radicals. We follow the
corresponding mathlib formalisation [2].
The forward direction — solvable by radicals implies solvable Galois group — is
assembled on the session New_Algebra: cyclotomic and Kummer extensions have
solvable (indeed abelian) Galois groups, solvability is preserved by the tower
correspondence, and the normal closure of a radical tower is again a radical
tower obtained by splicing the Galois conjugates of the original tower.
For the concrete witness we study the rational quintic Φ=x^5 − 4x + 2 and show
that the Galois group of its splitting field over Q is not solvable. We prove
that Φ is irreducible over Q (Eisenstein’s criterion at 2) and, using the AFP
entry Sturm_Sequences [1], show that it has exactly three real roots and one
conjugate pair of non-real roots. Complex conjugation therefore acts on the five
roots as a transposition, while irreducibility makes the Galois action
transitive; a transitive subgroup of S5 containing a transposition is the whole
of S5, which is not solvable. Combining the two halves yields the headline
result that Φis not solvable by radicals.
https://isa-afp.org/entries/Unsolvable_Quintic.html
Enjoy!
Last updated: Nov 19 2026 at 01:46 UTC