17.

Irreducible polynomials
Definition 17.1. Let F be a field. We say that a non-constant poly-
nomial f (x) is reducible over F or a reducible element of F [x],
if we can factor f (x) as the product of g(x) and h(x) ∈ F [x], where the
degree of g(x) and the degree of h(x) are both less than the degree of
f (x),
f (x) = g(x)h(x) and d(g(x)) < d(f (x)), d(h(x)) < d(f (x)).
We say that a non-constant polynomial f (x) is irreducible if it is
not reducible.
Example 17.2. Consider the polynomial x2 − 2.
2
√ that x − 2 has no zeroes over Q. This is the same as saying
Note
that 2 is irrational, a result that goes all the way back to the time of
Euclid.
If x2 − 2 is reducible then we may write
x2 − 2 = g(x)h(x),
where the degree of g(x) and h(x) is less than two. As the degree of
the LHS is two, the only possibility is that both g(x) and h(x) have
degree one. In this case x2 − 2 has a zero in Q, a contradiction.
Thus x2 − 2 is irreducible
√ over Q.
On the other hand, 2 ∈ R so that x2 − 2 is reducible over R,
√ √
x2 − 2 = (x − 2)(x + 2).
Example 17.3. Consider f (x) = x3 + 3x + 2 over the field Z5 .
Suppose that this is reducible. Then we can write
f (x) = g(x)h(x),
where both g(x) and h(x) have degree at most two. Possibly reordering
we may assume that the degree of g(x) is at most the degree of h(x).
It follows that g(x) has degree one and h(x) has degree two, since the
sum of the degrees is three. Thus f (x) has a zero, corresponding to
the linear factor g(x).
We check this by simply plugging in the elements of Z5 .
φ0 (x3 + 3x + 2) = 03 + 3 · 0 + 2 = 2
φ1 (x3 + 3x + 2) = 13 + 3 · 1 + 2 = 1
φ2 (x3 + 3x + 2) = 23 + 3 · 2 + 2 = 1
φ3 (x3 + 3x + 2) = 33 + 3 · 3 + 2 = 3
φ4 (x3 + 3x + 2) = 43 + 3 · 4 + 2 = 3.
1
 Since we never get zero f (x) must be irreducible.
Theorem 17.4. Let f (x) ∈ F [x] be a polynomial over a field F of
degree two or three.
Then f (x) is irreducible if and only if it has no zeroes.
Proof. If f (x) has zero α then we have already seen it can be factored
as (x − α)h(x). If f (x) has degree two then g(x) has degree one and
if f (x) has degree three then g(x) has degree two. Therefore f (x) is
reducible.
Now suppose that f (x) is reducible. Then
f (x) = g(x)h(x),
where the degrees of g(x) and h(x) are less than the degree of f (x).
Possibly reordering we may assume that g(x) has degree no more than
the degree of h(x).
It follows that g(x) has degree one. If g(x) = ax + b then a 6= 0. In
this case
b
α=− .
a
is a zero of g(x) and so it is a zero of f (x). 
The most beautiful results in this area relate to irreducibility over
the rationals. The first is due to Gauss:
Theorem 17.5. If f (x) ∈ Z[x] then we can factor f (x) into two poly-
nomials of degrees r and s in Z[x] if and only if we can factor f (x)
into two polynomials of the same degrees r and s in Q[x].
The point is that it is much easier to show that we cannot factor
over Z[x].
Corollary 17.6. Let f (x) = xn + an−1 xn−1 + · · · + a0 ∈ Z[x], where
a0 6= 0.
If f (x) has a zero in Q then it has a zero m ∈ Z and m divides a0 .
Proof. If α is a zero of f (x) then (x − α) is a linear factor of Q[x].
By Gauss f (x) must have a linear factor in Z,
f (x) = (ax + b)g(x).
Looking at the leading coefficients, we must have that a divides 1. So
a = ±1. Possibly replacing g(x) by −g(x) we may assume that a = 1.
If m = −b then
f (x) = (x − m)g(x).
m ∈ Z is a zero of f (x). Considering the constant coefficients m must
divide a0 . 
2
Example 17.7. Consider x2 − 2 ∈ Q[x].
Let’s show that this is irreducible over Q. If not then since x2 − 2
is a quadratic polynomial then it would have a zero in Z and this zero
would divide 2. The only possible choices are ±1 and ±2. It is easy to
check that none of these √are zeroes of x2 − 2. Thus x2 − 2 is irreducible
over Q. In other words, 2 is irrational.
Example 17.8. Consider f (x) = x4 + 3x2 − 7x + 1 ∈ Q[x].
Let’s show that this is irreducible over Q. We first check it does not
have a linear factor. If it has a linear factor it has a zero in Q and so
by (17.6) it must have a zero α in Z and this zero must divide 1. Thus
α = ±1. But
f (1) = 1 + 3 + 1 − 7 = −2 and f (−1) = 1 + 3 + 7 + 1 = 12.
Thus f (x) has no linear factors. The only other possibility is that it
factors as two quadratic polynomials. In this case we may write
x4 + 3x2 − 7x + 1 = (x2 + ax + b)(x2 + cx + d),
and by (17.6) we may assume that a, b, c and d are integers. Note
that we may assume that both factors are monic, that is, their leading
coefficients are 1, as the LHS is monic.
If we equate coefficients then we get the following equations:
bd = 1, ad + bc = −7, b + d + ac = 3, and a + c = 0.
Note that either b = 1 and d = 1 or b = −1 and d = −1. Either way
we have b = d. The second equation then reads
(a + c)b = −7.
But the last equation says that a + c = 0, which is a contradiction.
Thus f (x) = x4 + 3x2 − 7x + 1 is irreducible over Q.
Theorem 17.9 (Eisenstein’s Criteria). Let
f (x) = an xn + an−1 xn−1 + · · · + a0
be a polynomial with integer coefficients. Suppose that there is a prime
p such that p divides ai , i ≤ n − 1, p does not divide an and p2 does
not divide a0 .
Then f (x) is irreducible in Q[x].
Proof. By Gauss’ Lemma, we only have to rule out the possibility that
f (x) factors into polynomials of lower degree with integer coefficients.
Suppose that
f (x) = g(x)h(x)
3
is a factorisation of f (x) over the integers. Suppose that
f (x) = an xn + an−1 xn−1 + · · · + a0
g(x) = bd xd + bd−1 xd−1 + · · · + b0
h(x) = ce xe + ce−1 xe−1 + · · · + c0 .
for some n, d and e > 1.
As a0 = b0 c0 is not divisible by p2 either b0 or c0 is not divisible
by p. Possibly switching g(x) and h(x) we may assume that b0 is not
divisible by p. As an = bd ce and an is not divisible by p, then neither
is bd nor ce .
Let m be the smallest integer such that cm is not divisible by p. We
have
am = b0 cm + b1 cm−1 + b2 cm−2 + b3 cm−3 + . . . .
Every term on the RHS but the first is divisible by p. The first term
is not divisible by p as neither b0 nor cm is divisible by p. Thus the
RHS is not divisible by p. So the LHS is not divisible by p. The only
coefficient of f (x) not divisible by p is an . So we must have that m = n
and so h(x) is a polynomial of degree n.
Thus f (x) is irreducible. 
Note that we can apply Eisenstein to the polynomial x2 − 2 with the
prime p = 2 to conclude that x2 − 2 is irreducible over Q. Here is a
more interesting example:
Example 17.10. Let
f (x) = 2x7 − 15x6 + 60x5 − 18x4 − 9x3 + 45x2 − 3x + 6.
Then f (x) is irreducible over Q. We apply Eisenstein with p = 3.
Then the top coefficient is not divisible by 3, the others are, and the
smallest coefficient is not divisible by 9 = 32 .
Corollary 17.11. Let p be a prime. Then
f (x) = xp−1 + xp−2 + · · · + x + 1,
is irreducible over Q.
Proof. By Gauss, it suffices to consider factorisations of f (x) over Z.
First note that
xp − 1
f (x) = ,
x−1
as can be easily checked. Consider the map
Z[x] −→ Z[x] given by f (x) −→ f (x + 1).
4
This is an example of an evaluation homomorphism; in this case we
evaluate f (x) at x + 1. Thus we get a ring homomorphism. This
map is an isomorphism, since the inverse map sends f (x) to f (x − 1)
(evaluation at x − 1).
Note that
g(x) = f (x + 1)
(x + 1)p − 1
=
x     
p−1 p p−2 p p−3 p
=x + x + x + ··· +
1 2 p−1
p−1 p−2
=x + px + · · · + p.

Observe that pi is divisible by p, for all 1 ≤ i < p, so that we can

apply Eisenstein to the polynomial g(x), using the prime p, to conclude
that g(x) is irreducible.
Suppose that f (x) = h(x)k(x) is a factorisation of f (x) over the
integers. Then
g(x) = f (x + 1) = h(x + 1)k(x + 1),
is a factorisation of g(x) over the integers. Here we use the fact that the
map f (x) −→ f (x+1) is a ring homomorphism. As we already decided
we cannot factor g(x) into polynomials of lower degree, it follows that
we cannot factor f (x) either.
Thus f (x) is irreducible. 

It seems worth pointing out a rather nice fact about factorisation of

polynomials over a field F .
Theorem 17.12. Let p(x) be an irreducible polynomial over a field F .
If p(x) divides the product f (x)g(x) of two polynomials over F then
p(x) must divide one of the factors f (x) or g(x).
Corollary 17.13. Let p(x) be an irreducible polynomial over a field F .
If p(x) divides the product f1 (x)f2 (x) . . . fk (x) of the polynomials over
the field F then p(x) must divide one of the factors fi (x), for some index
1 ≤ i ≤ k.
Proof. Follows by induction on k, using (17.12). 
Theorem 17.14. If F is a field then every nonconstant polynomial
f (x) can be factored into irreducible polynomials. Morever this factori-
sation is unique up to order and units.
5
Proof. We first show that f (x) can be factored into irreducibles.
If f (x) is irreducible then we are done. If not, then f (x) = g(x)h(x),
where both g(x) and h(x) have smaller degree than f (x). If g(x) and
h(x) are irreducible then we are done. Otherwise one of g(x) or h(x)
factors and we keep going. Eventually this process must stop, since at
every stage we have a reducible polynomial the degree of the factors
goes down.
Now suppose that we can find two factorisations,
p1 (x)p2 (x)p3 (x) . . . pm (x) = q1 (x)q2 (x)q3 (x) . . . qn (x).
of f (x) into irreducibles. As p1 (x) divides the LHS it must also divide
the RHS. But then p1 (x) must divide one of the factors. Re-ordering
the factors on the RHS we may assume that p1 (x) divides q1 (x). As
q1 (x) is irreducible it follows that
q1 (x) = u1 p1 (x)
where u1 is a constant polynomial. As u1 is non-zero, u1 is a unit.
Replacing q1 (x) with u1 q(x) and cancelling q1 (x) from both sides we
obtain an equality of the form
p2 (x)p3 (x) . . . pm (x) = u1 q2 (x)q3 (x) . . . qn (x).
Now we repeat the same argument with p2 (x). It must divide one of
the factors, which we may assume is q2 (x), so that q2 (x) = u2 p2 (x).
Cancelling p2 (x) from both sides we obtain an equality of the form
p3 (x)p4 (x) . . . pm (x) = u1 u2 q3 (x)q4 (x) . . . qn (x).
Continuing in this way we eventually arrive at
1 = u1 u2 . . . um qm+1 (x)qm+2 (x) . . . qn (x).
The only way this is possible is if m = n so that we get
1 = u1 u2 . . . um .
It follows that p1 , p2 , . . . , pm and q1 , q2 , . . . , qn are the same, up re-
ordering and unit factors. 
Example 17.15. It is not hard to see that
x4 + 3x3 + 2x + 4 = (x − 1)3 (x + 1) ∈ Z5 [x].
We may also rewrite this factorisation as
x4 + 3x3 + 2x + 4 = (x − 1)2 (2x − 2)(3x + 3) ∈ Z5 [x].
Here we didn’t change the order, but we messed around a little bit with
the units.