Overview
- What are the hypotheses of the Schönemann–Eisenstein criterion?
- When does a translation make the criterion apply?
- How does the one-edge Dumas criterion extend Eisenstein’s criterion?
The Schönemann–Eisenstein criterion
Theorem · Schönemann–Eisenstein
Let , with . Suppose a prime satisfies , for , and . Then is irreducible over .
Proof
A factorization into positive degrees can be taken in . Reducing it modulo gives a factorization of the monomial . Both reduced factors are monomials of positive degree, because their leading coefficients remain nonzero. Thus both original constant terms are divisible by , which would make divisible by .
Schönemann's work of 1846 precedes Eisenstein's 1850 paper. Cox's historical account explains their different motivations and the connection with finite fields and the division of the lemniscate.
References 3
- schoenemann1846Original historical source.Source
- eisenstein1850Eisenstein’s original treatment.Source
- cox2011eisenstein · pp. 3–4Historical attribution and the elementary form of the criterion.Source
Changes of variable
For and , substitution is an automorphism of and preserves irreducibility. It may change the coefficient divisibility required by the criterion. Reversing coefficients, , also preserves irreducibility when .
Example
The right side is Eisenstein at , proving irreducibility of . More generally, if is prime,
is Eisenstein at : all intermediate binomial coefficients are divisible by , while the constant coefficient is exactly .
References 1
- cox2011eisenstein · p. 4The prime cyclotomic example after translation.Source
Schönemann’s residue-power criterion
The following form allows a higher-degree irreducible factor in the reduction. Let be monic with irreducible reduction modulo , let , and let satisfy and . Then
is irreducible over . Indeed, a monic integral factorization would reduce to , , where and . Write , . Comparing with , then dividing by and reducing modulo , forces , a contradiction.
For example, is irreducible: is irreducible over , and . The reduction is a power of an irreducible quadratic.
References 2
- schoenemann1846Historical source of the residue-power criterion.Source
- bonciocat2015multiprime · IntroductionModern statement of Schönemann’s criterion.Source
The Dumas criterion
Theorem · one-edge Dumas criterion
Let , with , , and . Suppose and
where . Then is irreducible over . For a rational polynomial, this also proves irreducibility over .
The Newton polygon has one segment of slope , so every root has valuation . A monic factor of degree over would have constant coefficient of valuation , an integer. Coprimality forces . Hence no proper positive factor degree is possible. Eisenstein is the case .
References 3
- dumas1906Classical source for the Newton-polygon criterion.Source
- bonciocat2015multiprime · IntroductionIntroduction states Dumas’s criterion in reversed-coefficient form; the form here follows by coefficient reversal.Source
- Hajir2005 · Theorem 2.1Roots and slopes over a discretely valued field.Source
Example
Example ·
At , the nonzero coefficient points are . The middle point lies above the segment joining to . Its slope is , and . Thus is irreducible over , hence over .
Eisenstein’s criterion at does not apply because . Dumas’s criterion instead uses the root valuation , whose denominator excludes every proper factor degree.
References 1
- bonciocat2015multiprime · IntroductionThe one-edge criterion used here; the valuation diagram and polynomial are an explicit worked computation.Source
Limitations and translation
A Newton polygon with several slopes proves reducibility over , but not necessarily over . A single edge whose horizontal and vertical increments have a common divisor may correspond to either an irreducible or a reducible polynomial over . Other primes, residual polynomials, or a change of variable can give further information. Local irreducibility alone does not determine the global Galois group.
Example
The polynomial is irreducible: a root generates the degree-four field , as shown in Field extensions. No integral translation makes Eisenstein, since
For an odd prime , divisibility of the cubic coefficient forces , while the constant coefficient is then . At , the constant coefficient is odd for even , and divisible by eight for odd . In each case an Eisenstein hypothesis fails.
The stronger one-edge criterion succeeds after translation: at the polynomial becomes . Its -adic coefficient valuations are , and the lower polygon is a single edge of slope . Dumas therefore proves irreducibility even over .
References 3
- ConradAllPrimes · Theorems 1 and 2Irreducibility and failure of every integral Eisenstein shift.Source
- dumas1906The one-edge criterion applies to the translated quartic, whose coefficients and valuations are calculated here.Source
- Hajir2005 · Theorem 2.1Distinct Newton slopes give factors over the complete local field; they do not imply reducibility over the rational field.Source
Questions
Can an integral translation always make an irreducible polynomial Eisenstein?
Scope. Translations x↦x+c with c∈Z
No. The irreducible polynomial is a counterexample. If were Eisenstein at an odd prime , its cubic coefficient would force , but its constant coefficient would be . At two, the constant coefficient is odd when is even and divisible by eight when is odd. This rules out every integral translation; other changes of variable or valuation criteria may still apply.
References 1
- ConradAllPrimes · Theorems 1 and 2Explicit counterexample, proved in the article. The Claimed label records that the cited source is unpublished exposition.Source
Related topics
Foundations: polynomials, fields, and Galois groups
Gauss’s lemma gives an integral factorization from rational reducibility.
UsesNewton polygons and Galois groups
The Dumas criterion is the one-edge case of the Newton-polygon argument.
Contrasts withReduction modulo primes
An Eisenstein polynomial has repeated reduction at the prime used.
Applies toTaylor polynomials
Factorial valuations determine Newton polygons of exponential truncations.
References
- Von denjenigen Moduln, welche Potenzen von Primzahlen sind. (Fortsetzung)
Schönemann, Theodor. Journal für die reine und angewandte Mathematik, 32, 93–105, 1846.
Read the source - Über die Irreductibilität und einige andere Eigenschaften der Gleichung, von welcher die Theilung der ganzen Lemniscate abhängt
Eisenstein, Gotthold. Journal für die reine und angewandte Mathematik, 39, 160–179, 1850.
Read the source - Why Eisenstein Proved the Eisenstein Criterion and Why Schönemann Discovered It First
Cox, David A. The American Mathematical Monthly, 118(1), 3–21, 2011.
Read the source - Schönemann–Eisenstein–Dumas-type irreducibility conditions that use arbitrarily many prime numbers
Bonciocat, Nicolae Ciprian. Communications in Algebra, 43(8), 3102–3122, 2015.
Read the source - Sur quelques cas d'irréductibilité des polynômes à coefficients rationnels
Dumas, Gustave. Journal de Mathématiques Pures et Appliquées, 6e série, 2, 191–258, 1906.
Read the source - On the Galois group of generalized Laguerre polynomials
Hajir, Farshid. Journal de théorie des nombres de Bordeaux, 17(2), 517–525, 2005.
Read the source - An irreducible that factors modulo all primes
Conrad, Keith. Author-hosted expository note.
Read the source
Comments
Any comments are welcome. Comments appear after the site owner approves them. Privacy policy.