Commit 2025-04-11 13:00 0ac690b5
View on Github →feat(RingTheory/Polynomial/Eisenstein/Generalized): a generalized version of the Eisenstein criterion (#23710)
A generalized form of the Eisenstein criterion
Let R
be an integral domain, P
a prime ideal of R
and
let K
be the field of fractions o f R ⧸ P
.
Let q : R[X]
be a monic polynomial which is irreducible in K[X]
.
Let f : R[X]
be a monic polynomial of strictly positive degree
whose image in K[X]
is a power of q
.
Assume moreover that f.modByMonic q
is not zero in (R ⧸ (P ^ 2))[X]
.
Then f
is irreducible.
The Eisenstein criterion is the particular case where q := X
.
The case of a polynomial q := X - a
is interesting,
then the mod P ^ 2
hypothesis can rephrased as saying
that f.derivative.eval a ∉ P ^ 2
. (TODO)
The case of cyclotomic polynomials of prime index can be proved directly using that, with a = 1
.
We give a (possibly non convincing) application to the irreducibility
of the polynomial X ^ 4 - 10 * X + 1
in ℤ[X]
.
One argues modulo 3, with q := X ^ 2 + 1
.
Remark
The result can also be generalized to the case where
the leading coefficients of f
and q
do not belong to P
.
(By localization at P
, make these coefficients invertible.)
There are two obstructions, though :
- Usually, one will only obtain irreducibility in
F[X]
, whereF
is the field of fractions ofR
. (IfR
is a UFD, this will be close to what is wanted, but not in general.) - The mod
P ^ 2
hypothesis will have to be rephrased to a condition in the second symbolic power ofP
. WhenP
is a maximal ideal, that symbolic power coincides withP ^ 2
, but not in general.