Overview

  • Which cycle type follows from factorization modulo a prime?
  • When does a polynomial have an irreducible reduction?
  • How do modular factorizations prove that a quintic has group S₅?

The Dedekind–Frobenius theorem

Theorem · Dedekind–Frobenius

Let f∈Z[x]f\in\mathbb Z[x] be nonconstant, monic, and separable, and let GG be the Galois group of its splitting field over Q\mathbb Q, acting on its nn roots. If p∤disc⁡(f)p\nmid\operatorname{disc}(f) and

fˉ=f1⋯frin Fp[x]\bar f=f_1\cdots f_r\quad\text{in }\mathbb F_p[x]

is its factorization into distinct monic irreducibles, then GG contains a permutation whose cycle lengths are deg⁡f1,…,deg⁡fr\deg f_1,\ldots,\deg f_r.

The cycle statement does not require irreducibility of ff. Irreducibility is the additional condition that makes GG transitive. For nonmonic integral polynomials, also exclude primes dividing the leading coefficient.

References 2
  • ConradModP · Theorem 1 and its proofThe note treats irreducible monic polynomials; the same reduction argument gives the separable form stated here.Source
  • Milne2022 · Chapter 4, “Computing Galois groups over Q”Author’s book page

Frobenius orbits

Choose a prime p\mathfrak p of the splitting field above pp. The discriminant condition ensures that the roots remain distinct after reduction. Frobenius acts on the residue field by z↦zpz\mapsto z^p. A root of an irreducible degree-dd factor has an orbit of length exactly dd under this map. Lifting that action to the roots gives the required permutation.

A different choice of p\mathfrak p conjugates Frobenius. Thus factorization determines the cycle type. It does not identify the root labels, so permutations obtained from different primes have no specified common labeling.

References 1

Example

Example: S5S_5

Take f=x5−x−1f=x^5-x-1. Modulo 33, it has no root. The only monic irreducible quadratics over F3\mathbb F_3 are

x2+1,x2+x+2,x2+2x+2.x^2+1,\quad x^2+x+2,\quad x^2+2x+2.

The remainders on division of ff by these are −1,x−1,x−1-1,x-1,x-1, respectively. Every reducible quintic has a factor of degree at most two, so fˉ\bar f is irreducible over F3\mathbb F_3. Hence ff is irreducible over Q\mathbb Q, and its group contains a 5-cycle.

Modulo 22,

f=(x2+x+1)(x3+x2+1),f=(x^2+x+1)(x^3+x^2+1),

with distinct irreducible factors. This gives an element of type (2,3)(2,3), whose cube is a transposition.

A transitive action of prime degree is primitive. A primitive group containing a transposition is the full symmetric group, as proved in the permutation-group entry. Therefore Gal⁡(f/Q)=S5\operatorname{Gal}(f/\mathbb Q)=S_5.

Modular factorizations

Modulo 2

(x2+x+1)(x3+x2+1)(x^{2} + x + 1)(x^{3} + x^{2} + 1)
Factor degrees
2 + 3
Squarefree reduction
Yes
Possible proper factor degrees
2, 3

The factor degrees are the cycle lengths of a Frobenius element. Every proper rational factor has one of the listed degrees.

References 2
  • ConradSnAn · Theorem 2.1Recognition criterion used after the explicit finite-field computations.Source
  • Selmer1956Classical irreducibility of the broader trinomial family; the present degree-five proof is self-contained.Source

Connections and evidence

Classical

How do two modular factorizations determine S₅?

One prime proves transitivity; another supplies a transposition. Prime degree supplies primitivity, and the two branches together determine the full group.

Scope. Let f=X5−X−1f=X^5-X-1 and G=Gal⁡(f/Q)G=\operatorname{Gal}(f/\mathbb Q), acting on its five roots.

How do two modular factorizations determine S₅?One prime proves transitivity; another supplies a transposition. Prime degree supplies primitivity, and the two branches together determine the full group. Connection 1: Irreducible reduction modulo 3 leads to f irreducible over ℚ. The monic degree-five polynomial retains its degree modulo 3, so irreducible reduction proves irreducibility over the rationals. Connection 2: f irreducible over ℚ leads to G transitive on five roots. Irreducibility is equivalent to transitivity of the Galois action on the five roots. Connection 3: G transitive on five roots leads to G primitive. In a transitive action block sizes divide the degree; prime degree has no nontrivial block sizes. Connection 4: Modulo 2: degrees 2 and 3 leads to σ has cycle type (2,3). The reduction modulo 2 is squarefree, and its irreducible factors have degrees 2 and 3; the Galois group therefore contains that cycle type. Connection 5: σ has cycle type (2,3) leads to σ³ is a transposition. The two cycles are disjoint; cubing kills the 3-cycle and leaves the 2-cycle. Connection 6: G primitive is an input to AND. Primitivity is one of the two hypotheses required by the transposition criterion. Connection 7: σ³ is a transposition is an input to AND. The existence of a transposition is the second hypothesis required by the criterion. Connection 8: AND leads to G = S₅. A primitive permutation group containing a transposition is the full symmetric group.Irreducible reduction modulo 3 — related articleIrreducible reductionmodulo 3f irreducible over ℚ — related articlef irreducible over ℚG transitive on five roots — related articleG transitive on five rootsG primitive — related articleG primitiveModulo 2: degrees 2 and 3 — related articleModulo 2: degrees 2 and3σ has cycle type (2,3) — related articleσ has cycle type (2,3)σ³ is a transposition — related articleσ³ is a transpositionANDANDG = S₅ — related articleG = S₅Connection 1: The monic degree-five polynomial retains its degree modulo 3, so irreducible reduction proves irreducibility over the rationals.Reduction criterion [1]Connection 2: Irreducibility is equivalent to transitivity of the Galois action on the five roots.One irreducible factor[2]Connection 3: In a transitive action block sizes divide the degree; prime degree has no nontrivial block sizes.Prime degree 5 [3]Connection 4: The reduction modulo 2 is squarefree, and its irreducible factors have degrees 2 and 3; the Galois group therefore contains that cycle type.Dedekind–Frobenius [4]Connection 5: The two cycles are disjoint; cubing kills the 3-cycle and leaves the 2-cycle.Cube the permutation [5]Connection 6: Primitivity is one of the two hypotheses required by the transposition criterion.Primitivity [6]Connection 7: The existence of a transposition is the second hypothesis required by the criterion.A transposition [7]Connection 8: A primitive permutation group containing a transposition is the full symmetric group.Primitive + transposition[8]
Linked nodes open articles; numbered connections open the evidence below.Open full-size diagram

deduction or information flowAND joint hypotheses

The complete arithmetic certificate is

f mod 2=(X2+X+1)(X3+X2+1),f mod 3 is irreducible,disc⁡(f)=2869=19⋅151.\begin{aligned}f\bmod2&=(X^2+X+1)(X^3+X^2+1),\\ f\bmod3&\text{ is irreducible},\\ \operatorname{disc}(f)&=2869=19\cdot151.\end{aligned}

The two factors modulo 22 are irreducible: each has degree two or three and neither has a root in F2\mathbb F_2. Their distinctness makes the reduction squarefree. For an exact check modulo 33, polynomial arithmetic gives X35≡X(modf)X^{3^5}\equiv X\pmod f and gcd⁡(f,X3−X)=1\gcd(f,X^3-X)=1. The finite-field irreducibility criterion for prime degree five then applies. The discriminant calculation independently confirms that both primes are good.

The left branch excludes nontrivial block systems. The right produces a permutation exchanging exactly two roots: if σ=(ab)(cde)\sigma=(ab)(cde), then σ3=(ab)\sigma^3=(ab). Both properties are needed for the final group criterion. Transitivity and a transposition alone do not suffice in composite degree; the order-eight dihedral action on four points is a counterexample.

The same group certificate has a geometric consequence: Zarhin’s theorem applies to y2=X5−X−1y^2=X^5-X-1, whose genus-two Jacobian has geometric endomorphism ring Z\mathbb Z.

Inspect all 8 connections and exact references
Evidence for each connection
ConnectionStatement and source
1Irreducible reduction modulo 3 → f irreducible over ℚDeduction

The monic degree-five polynomial retains its degree modulo 3, so irreducible reduction proves irreducibility over the rationals.

2f irreducible over ℚ → G transitive on five rootsDeduction

Irreducibility is equivalent to transitivity of the Galois action on the five roots.

  • R2 · ConradTheorem 2.9(b), p. 3; Example 4.17, p. 9The factor–orbit theorem and the modular computation for this quintic; taking the cube is also checked directly.Primary source
3G transitive on five roots → G primitiveDeduction

In a transitive action block sizes divide the degree; prime degree has no nontrivial block sizes.

4Modulo 2: degrees 2 and 3 → σ has cycle type (2,3)Deduction

The reduction modulo 2 is squarefree, and its irreducible factors have degrees 2 and 3; the Galois group therefore contains that cycle type.

5σ has cycle type (2,3) → σ³ is a transpositionDeduction

The two cycles are disjoint; cubing kills the 3-cycle and leaves the 2-cycle.

  • R2 · ConradTheorem 2.9(b), p. 3; Example 4.17, p. 9The factor–orbit theorem and the modular computation for this quintic; taking the cube is also checked directly.Primary source
6G primitive → ANDConstruction / input

Primitivity is one of the two hypotheses required by the transposition criterion.

7σ³ is a transposition → ANDConstruction / input

The existence of a transposition is the second hypothesis required by the criterion.

8AND → G = S₅Deduction

A primitive permutation group containing a transposition is the full symmetric group.

Sources and scope checked .

Chebotarev density

By Chebotarev, every conjugacy class of GG occurs at infinitely many unramified primes. The density for a class CC is ∣C∣/∣G∣|C|/|G|; a cycle type may comprise several classes of GG, in which case their densities add.

Consequently, an irreducible polynomial of degree nn has an irreducible reduction at some good prime if and only if GG contains an nn-cycle. Transitivity alone does not guarantee this. For x4+1x^4+1, the group is the Klein four-group and no good prime yields an irreducible quartic reduction.

References 2

Limits of modular tests

Each squarefree reduction proves that its cycle type occurs in the Galois group. Failure to observe a cycle type in a finite search does not prove its absence. To exclude a type, one needs a group-theoretic restriction or another exact argument.

For example, if the Galois group contains no nn-cycle, every good reduction of a degree-nn polynomial is reducible. If modular cycle types leave several possible groups, ramification, the discriminant or a resolvent may distinguish them.

References 1

Questions

SettledGalois group

Does every irreducible rational polynomial remain irreducible modulo some prime?

Scope. Separable irreducible polynomials over Q; good primes preserve degree and give squarefree reduction.

No. For a polynomial of degree nn, an irreducible reduction at a good prime exists exactly when its Galois group contains an nn-cycle. The polynomial x4+1x^4+1 is a counterexample: it is irreducible over Q\mathbb Q, and its Klein four-group contains no 4-cycle.

References 2
Discuss this question →

Related topics

References

  1. Galois groups over Q and factorizations mod p

    Conrad, Keith. Author-hosted expository note.

    Read the source
  2. Fields and Galois Theory

    Milne, James S. Kea Books, 2022.

    Author’s book page
  3. Recognizing Galois groups S_n and A_n

    Conrad, Keith. Author-hosted expository note.

    Read the source
  4. On the irreducibility of certain trinomials

    Selmer, Ernst S. Mathematica Scandinavica, 4, 287–302, 1956.

    Read the source
  5. Chebotarëv and his density theorem

    Stevenhagen, Peter and Lenstra, Jr., Hendrik W. The Mathematical Intelligencer, 18(2), 26–37, 1996.

    Read the source
  6. An irreducible that factors modulo all primes

    Conrad, Keith. Author-hosted expository note.

    Read the source
  7. Irreducibility tests in Q[T]

    Conrad, Keith. Author-hosted expository note.

    Read the source
  8. Galois groups as permutation groups

    Conrad, Keith. Author-hosted expository notes.

    Read the source
  9. The Galois groups of the polynomials x^n+ax^s+b, II

    Osada, Hiroyuki. Tohoku Mathematical Journal, 39(3), 437–445, 1987.

    Read the source