This action might not be possible to undo. Are you sure you want to continue?

Welcome to Scribd! Start your free trial and access books, documents and more.Find out more

J.S. Milne

Taiaroa Publishing

Erehwon

Version 4.00

February 19, 2005

Abstract

These notes give a concise exposition of the theory of ﬁelds, including the Galois theory

of ﬁnite and inﬁnite extensions and the theory of transcendental extensions. They are an

expansion of those handed out during a course taught to ﬁrst-year graduate students.

v2.01 (August 21, 1996). First version on the web.

v2.02 (May 27, 1998). Fixed about 40 minor errors; 57 pages.

v3.00 (April 3, 2002). Revised notes; minor additions to text; added 82 exercises with

solutions, an examination, and an index; 100 pages.

v3.01 (August 31, 2003). Fixed many minor errors; no change to numbering; 99 pages.

v4.00 (February 19, 2005). Minor corrections and improvements; added proofs to the

section on inﬁnite Galois theory; added material to the section on transcendental

extensions; 107 pages.

Please send comments and corrections to me at math@jmilne.org

Available at http://www.jmilne.org/math/

Copyright c ( 1996, 1998, 2002, 2003, 2005. J.S. Milne.

This work is licensed under a

Creative Commons Licence (Attribution-NonCommercial-NoDerivs 2.0)

http://creativecommons.org/licenses/by-nc-nd/2.0/

Contents

Notations. 3; References. 3

1 Basic deﬁnitions and results 1

Rings 1; Fields 1; The characteristic of a ﬁeld 2; Review of polynomial rings 3; Factoring

polynomials 4; Extension ﬁelds 6; Construction of some extension ﬁelds 7; Stem ﬁelds 9;

The subring generated by a subset 9; The subﬁeld generated by a subset 10; Algebraic and

transcendental elements 10; Transcendental numbers 12; Constructions with straight-edge

and compass. 13; Algebraically closed ﬁelds 16; Exercises 18

2 Splitting ﬁelds; multiple roots 19

Maps from simple extensions. 19; Splitting ﬁelds 20; Multiple roots 22; Exercises 24

3 The fundamental theorem of Galois theory 25

Groups of automorphisms of ﬁelds 25; Separable, normal, and Galois extensions 27; The

fundamental theorem of Galois theory 29; Examples 31; Constructible numbers revisited

33; The Galois group of a polynomial 34; Solvability of equations 34; Exercises 34

4 Computing Galois groups. 36

When is G

f

⊂ A

n

? 36; When is G

f

transitive? 37; Polynomials of degree at most three

37; Quartic polynomials 38; Examples of polynomials with S

p

as Galois group over Q

40; Finite ﬁelds 41; Computing Galois groups over Q 43; Exercises 45

5 Applications of Galois theory 46

Primitive element theorem. 46; Fundamental Theorem of Algebra 48; Cyclotomic exten-

sions 49; Dedekind’s theorem on the independence of characters 51; The normal basis

theorem 52; Hilbert’s Theorem 90. 53; Cyclic extensions. 56; Proof of Galois’s solvabil-

ity theorem 57; The general polynomial of degree n 58; Norms and traces 61; Exercises

65

6 Algebraic closures 66

Zorn’s lemma 66; First proof of the existence of algebraic closures 67; Second proof of

the existence of algebraic closures 67; Third proof of the existence of algebraic closures

67; (Non)uniqueness of algebraic closures 68; Separable closures 69

7 Inﬁnite Galois extensions 70

Topological groups 70; The Krull topology on the Galois group 71; The fundamental the-

orem of inﬁnite Galois theory 73; Galois groups as inverse limits 76; Nonopen subgroups

of ﬁnite index 77

8 Transcendental extensions 79

Algebraic independence 79; Transcendence bases 80; L¨ uroth’s theorem 83; Separating

transcendence bases 83; Transcendental Galois theory 84

A Review exercises 86

B Two-hour Examination 91

C Solutions to the Exercises 92

Notations.

We use the standard (Bourbaki) notations:

N = ¦0, 1, 2, . . .¦,

Z = ring of integers,

R = ﬁeld of real numbers,

C = ﬁeld of complex numbers,

F

p

= Z/pZ = ﬁeld with p elements, p a prime number.

Given an equivalence relation, [∗] denotes the equivalence class containing ∗.

Throughout the notes, p is a prime number: p = 2, 3, 5, 7, 11, . . ..

Let I and A be sets. A family of elements of A indexed by I, denoted (a

i

)

i∈I

, is a

function i →a

i

: I →A.

X ⊂ Y X is a subset of Y (not necessarily proper).

X

df

= Y X is deﬁned to be Y , or equals Y by deﬁnition.

X ≈ Y X is isomorphic to Y .

X · Y X and Y are canonically isomorphic (or there is a given or unique isomorphism).

References.

Dummit, D., and Foote, R.M., 1991, Abstract Algebra, Prentice Hall.

Jacobson, N., 1964, Lectures in Abstract Algebra, Volume III — Theory of Fields and

Galois Theory, van Nostrand.

Also, the following of my notes (available at www.jmilne.org/math/).

GT Group Theory, v2.11, 2003.

AG Algebraic Geometry, v5.00, 2005.

ANT Algebraic Number Theory, v2.10, 1998.

Prerequisites

Group theory (for example, GT), basic linear algebra, and some elementary theory of rings.

Acknowledgements

I thank the following for providing corrections and comments for earlier versions of the

notes: Mike Albert, Tommaso Centeleghe, Demetres Christoﬁdes, Antoine Chambert-Loir,

Dustin Clausen, Hardy Falk, Jens Hansen, Albrecht Hess, Philip Horowitz, Trevor Jarvis,

Henry Kim, Martin Klazar, Jasper Loy Jiabao, Dmitry Lyubshin, John McKay, Shuichi

Otsuka, David G. Radcliffe, Martin Ward (and class), and others.

1

1 Basic deﬁnitions and results

Rings

A ring is a set R with two composition laws + and such that

(a) (R, +) is a commutative group;

(b) is associative, and there exists

1

an element 1

R

such that a 1

R

= a = 1

R

a for all

a ∈ R;

(c) the distributative law holds: for all a, b, c ∈ R,

(a +b) c = a c +b c

a (b +c) = a b +a c.

We usually omit “” and write 1 for 1

R

when this causes no confusion. If 1

R

= 0, then

R = ¦0¦.

A subring S of a ring R is a subset that contains 1

R

and is closed under addition,

passage to the negative, and multiplication. It inherits the structure of a ring from that on

R.

A homomorphism of rings α: R →R

t

is a map with the properties

α(a +b) = α(a) +α(b), α(ab) = α(a)α(b), α(1

R

) = 1

R

, all a, b ∈ R.

A ring R is said to be commutative if multiplication is commutative:

ab = ba for all a, b ∈ R.

A commutative ring is said to be an integral domain if 1

R

= 0 and the cancellation law

holds for multiplication:

ab = ac, a = 0, implies b = c.

An ideal I in a commutative ring R is a subgroup of (R, +) that is closed under multipli-

cation by elements of R:

r ∈ R, a ∈ I, implies ra ∈ I.

We assume that the reader has some familiarity with the elementary theory of rings. For ex-

ample, in Z (more generally, any Euclidean domain) an ideal I is generated by any “small-

est” nonzero element of I.

Fields

DEFINITION 1.1. A ﬁeld is a set F with two composition laws + and such that

(a) (F, +) is a commutative group;

(b) (F

, ), where F

**= F ¦0¦, is a commutative group;
**

(c) the distributive law holds.

Thus, a ﬁeld is a nonzero commutative ring such that every nonzero element has an inverse.

In particular, it is an integral domain. A ﬁeld contains at least two distinct elements, 0 and

1. The smallest, and one of the most important, ﬁelds is F

2

= Z/2Z = ¦0, 1¦.

A subﬁeld S of a ﬁeld F is a subring that is closed under passage to the inverse. It

inherits the structure of a ﬁeld from that on F.

1

We follow Bourbaki in requiring that rings have a 1, which entails that we require homomorphisms to

preserve it.

2 1 BASIC DEFINITIONS AND RESULTS

LEMMA 1.2. A nonzero commutative ring R is a ﬁeld if and only if it has no ideals other

than (0) and R.

PROOF. Suppose R is a ﬁeld, and let I be a nonzero ideal in R. If a is a nonzero element

of I, then 1 = a

−1

a ∈ I, and so I = R. Conversely, suppose R is a commutative ring with

no nontrivial ideals. If a = 0, then (a) = R, and so there is a b in F such that ab = 1.

2

EXAMPLE 1.3. The following are ﬁelds: Q, R, C, F

p

= Z/pZ (p prime).

A homomorphism of ﬁelds α: F → F

t

is simply a homomorphism of rings. Such a

homomorphism is always injective, because its kernel is a proper ideal (it doesn’t contain

1), which must therefore be zero.

The characteristic of a ﬁeld

One checks easily that the map

Z →F, n →1

F

+ 1

F

+ + 1

F

(n copies),

is a homomorphism of rings, and so its kernel is an ideal in Z.

CASE 1: The kernel of the map is (0), so that

n 1

F

= 0 =⇒ n = 0 (in Z).

Nonzero integers map to invertible elements of F under n → n 1

F

: Z → F, and so this

map extends to a homomorphism

m

n

→(m 1

F

)(n 1

F

)

−1

: Q →F.

Thus, in this case, F contains a copy of Q, and we say that it has characteristic zero.

CASE 2: The kernel of the map is = (0), so that n 1

F

= 0 for some n = 0. The

smallest positive such n will be a prime p (otherwise there will be two nonzero elements in

F whose product is zero), and p generates the kernel. Thus, the map n → n 1

F

: Z → F

deﬁnes an isomorphism from Z/pZ onto the subring

¦m 1

F

[ m ∈ Z¦

of F. In this case, F contains a copy of F

p

, and we say that it has characteristic p.

The ﬁelds F

2

, F

3

, F

5

, . . . , Q are called the prime ﬁelds. Every ﬁeld contains a copy of

exactly one of them.

REMARK 1.4. The binomial theorem

(a +b)

m

= a

m

+

m

1

a

m−1

b +

m

2

a

m−2

b

2

+ +b

m

holds in any commutative ring. If p is prime, then p divides

p

n

r

**for all r with 1 ≤ r ≤
**

p

n

−1. Therefore, when F has characteristic p,

(a +b)

p

n

= a

p

n

+b

p

n

all n ≥ 1.

Hence a → a

p

is a homomorphism F → F, called the Frobenius endomorphism of F.

When F is ﬁnite, it is an isomorphism, called the Frobenius automorphism.

Review of polynomial rings 3

Review of polynomial rings

For more on the following, see Dummit and Foote 1991, Chapter 9. Let F be a ﬁeld.

1.5. The ring F[X] of polynomials in the symbol (or “indeterminate” or “variable”) X

with coefﬁcients in F is an F-vector space with basis 1, X, . . . , X

n

, . . . , and with the

multiplication deﬁned by

¸

i

a

i

X

i

¸

j

b

j

X

j

=

¸

k

¸

i+j=k

a

i

b

j

X

k

.

For any ring R containing F as a subring and element r of R, there is a unique homomor-

phism α: F[X] →R such that α(X) = r and α(a) = a for all a ∈ F.

1.6. Division algorithm: given f(X) and g(X) ∈ F[X] with g = 0, there exist q(X),

r(X) ∈ F[X] with deg(r) < deg(g) such that

f = gq +r;

moreover, q(X) and r(X) are uniquely determined. Thus F[X] is a Euclidean domain with

deg as norm, and so is a unique factorization domain.

1.7. From the division algorithm, it follows that an element a of F is a root of f (that is,

f(a) = 0) if and only if X −a divides f. From unique factorization, it now follows that f

has at most deg(f) roots (see also Exercise 1-3).

1.8. Euclid’s algorithm: Let f and g ∈ F[X] have gcd d(X). Euclid’s algorithm con-

structs polynomials a(X) and b(X) such that

a(X) f(X) +b(X) g(X) = d(X), deg(a) < deg(g), deg(b) < deg(f).

Recall how it goes. We may assume deg(f) ≥ deg(g) since the argument is the same in

the opposite case. Using the division algorithm, we construct a sequence of quotients and

remainders

f = q

0

g +r

0

g = q

1

r

0

+r

1

r

0

= q

2

r

1

+r

2

r

n−2

= q

n

r

n−1

+r

n

r

n−1

= q

n+1

r

n

with r

n

the last nonzero remainder. Then, r

n

divides r

n−1

, hence r

n−2

,. . . , hence g, and

hence f. Moreover,

r

n

= r

n−2

−q

n

r

n−1

= r

n−2

−q

n

(r

n−3

−q

n−1

r

n−2

) = = af +bg

and so any common divisor of f and g divides r

n

: we have shown r

n

= gcd(f, g).

Let af +bg = d. If deg(a) ≥ deg(g), write a = gq +r with deg(r) < deg(g); then

rf + (b +qf)g = d,

and b +qf automatically has degree < deg(f).

Maple knows Euclid’s algorithm — to learn its syntax, type “?gcdex;”.

4 1 BASIC DEFINITIONS AND RESULTS

1.9. Let I be a nonzero ideal in F[X], and let f be a nonzero polynomial of least degree

in I; then I = (f) (because F[X] is a Euclidean domain). When we choose f to be monic,

i.e., to have leading coefﬁcient one, it is uniquely determined by I. Thus, there is a one-

to-one correspondence between the nonzero ideals of F[X] and the monic polynomials in

F[X]. The prime ideals correspond to the irreducible monic polynomials.

1.10. Since F[X] is an integral domain, we can form its ﬁeld of fractions F(X). Its

elements are quotients f/g, f and g polynomials, g = 0.

Factoring polynomials

The following results help in deciding whether a polynomial is reducible, and in ﬁnding its

factors.

PROPOSITION 1.11. Suppose r ∈ Q is a root of a polynomial

a

m

X

m

+a

m−1

X

m−1

+ +a

0

, a

i

∈ Z,

and let r = c/d, c, d ∈ Z, gcd(c, d) = 1. Then c[a

0

and d[a

m

.

PROOF. It is clear from the equation

a

m

c

m

+a

m−1

c

m−1

d + +a

0

d

m

= 0

that d[a

m

c

m

, and therefore, d[a

m

. Similarly, c[a

0

.

2

EXAMPLE 1.12. The polynomial f(X) = X

3

− 3X − 1 is irreducible in Q[X] because

its only possible roots are ±1, and f(1) = 0 = f(−1).

PROPOSITION 1.13 (GAUSS’S LEMMA). Let f(X) ∈ Z[X]. If f(X) factors nontrivially

in Q[X], then it factors nontrivially in Z[X].

PROOF. Let f = gh in Q[X]. For suitable integers m and n, g

1

=

df

mg and h

1

=

df

nh

have coefﬁcients in Z, and so we have a factorization

mnf = g

1

h

1

in Z[X].

If a prime p divides mn, then, looking modulo p, we obtain an equation

0 = g

1

h

1

in F

p

[X].

Since F

p

[X] is an integral domain, this implies that p divides all the coefﬁcients of at least

one of the polynomials g

1

, h

1

, say g

1

, so that g

1

= pg

2

for some g

2

∈ Z[X]. Thus, we have

a factorization

(mn/p)f = g

2

h

1

in Z[X].

Continuing in this fashion, we can remove all the prime factors of mn, and so obtain a

factorization of f in Z[X].

2

PROPOSITION 1.14. If f ∈ Z[X] is monic, then any monic factor of f in Q[X] lies in

Z[X].

Factoring polynomials 5

PROOF. Let g be a monic factor of f in Q[X], so that f = gh with h ∈ Q[X] also monic.

Let m, n be the positive integers with the fewest prime factors such that mg, nf ∈ Z[X]. As

in the proof of Gauss’s Lemma, if a prime p divides mn, then it divides all the coefﬁcients

of at least one of the polynomials mg, nh, say mg, in which case it divides m because g is

monic. Now

m

p

g ∈ Z[X], which contradicts the deﬁnition of m.

2

REMARK 1.15. We sketch an alternative proof of Proposition 1.14. A complex number

α is said to be an algebraic integer if it is a root of a monic polynomial in Z[X]. The

algebraic integers form a subring of C — for an elementary proof of this, using nothing but

the symmetric polynomials theorem (5.30), see AG, Theorem 1.14. Now let α

1

, . . . , α

m

be

the roots of f in C. By deﬁnition, they are algebraic integers. The coefﬁcients of any monic

factor of f are polynomials in (certain of) the α

i

, and therefore are algebraic integers. If

they lie in Q, then they lie in Z, because Proposition 1.11 shows that every algebraic integer

in Q is in Z.

PROPOSITION 1.16 (EISENSTEIN’S CRITERION). Let

f = a

m

X

m

+a

m−1

X

m−1

+ +a

0

, a

i

∈ Z;

suppose that there is a prime p such that:

— p does not divide a

m

,

— p divides a

m−1

, ..., a

0

,

— p

2

does not divide a

0

.

Then f is irreducible in Q[X].

PROOF. If f(X) factors in Q[X], it factors in Z[X]:

a

m

X

m

+a

m−1

X

m−1

+ +a

0

= (b

r

X

r

+ +b

0

)(c

s

X

s

+ +c

0

)

b

i

, c

i

∈ Z, r, s < m. Since p, but not p

2

, divides a

0

= b

0

c

0

, p must divide exactly one of

b

0

, c

0

, say, b

0

. Now from the equation

a

1

= b

0

c

1

+b

1

c

0

,

we see that p[b

1

, and from the equation

a

2

= b

0

c

2

+b

1

c

1

+b

2

c

0

,

that p[b

2

. By continuing in this way, we ﬁnd that p divides b

0

, b

1

, . . . , b

r

, which contradicts

the condition that p does not divide a

m

.

2

The last three propositions hold with Z replaced by any unique factorization domain.

REMARK 1.17. There is an algorithm for factoring a polynomial in Q[X]. To see this,

consider f ∈ Q[X]. Multiply f(X) by a rational number so that it is monic, and then

replace it by D

deg(f)

f(

X

D

), with D equal to a common denominator for the coefﬁcients

of f, to obtain a monic polynomial with integer coefﬁcients. Thus we need consider only

polynomials

f(X) = X

m

+a

1

X

m−1

+ +a

m

, a

i

∈ Z.

6 1 BASIC DEFINITIONS AND RESULTS

From the fundamental theorem of algebra (see 5.6), we know that f splits completely

in C[X]:

f(X) =

m

¸

i=1

(X −α

i

), α

i

∈ C.

From the equation

0 = f(α

i

) = α

m

i

+a

1

α

m−1

i

+ +a

m

,

it follows that [α

i

[ is less than some bound depending only on the degree and coefﬁcients

of f; in fact,

[α

i

[ ≤ max¦1, mB¦, B = max [a

i

[.

Now if g(X) is a monic factor of f(X), then its roots in C are certain of the α

i

, and its

coefﬁcients are symmetric polynomials in its roots. Therefore, the absolute values of the

coefﬁcients of g(X) are bounded in terms of the degree and coefﬁcients of f. Since they are

also integers (by 1.14), we see that there are only ﬁnitely many possibilities for g(X). Thus,

to ﬁnd the factors of f(X) we (better Maple) have to do only a ﬁnite amount of checking.

Thus, we need not concern ourselves with the problem of factoring polynomials in

Q[X] or F

p

[X], since Maple knows how to do it. For example

>factor(6

*

Xˆ2+18

*

X-24); will ﬁnd the factors of 6X

2

+ 18X −24, and

>Factor(Xˆ2+3

*

X+3) mod 7; will ﬁnd the factors of X

2

+ 3X + 3 modulo 7,

i.e., in F

7

[X].

REMARK 1.18. One other observation is useful. Let f ∈ Z[X]. If the leading coefﬁcient

of f is not divisible by a prime p, then a nontrivial factorization f = gh in Z[X] will give

a nontrivial factorization f = gh in F

p

[X]. Thus, if f(X) is irreducible in F

p

[X] for some

prime p not dividing its leading coefﬁcient, then it is irreducible in Z[X]. This test is very

useful, but it is not always effective: for example, X

4

− 10X

2

+ 1 is irreducible in Z[X]

but it is reducible

2

modulo every prime p.

Extension ﬁelds

A ﬁeld E containing a ﬁeld F is called an extension ﬁeld of F (or simply an extension

of F). Such an E can be regarded in an obvious fashion as an F-vector space. We write

[E : F] for the dimension, possibly inﬁnite, of E as an F-vector space, and call [E : F] the

degree of E over F. We say that E is ﬁnite over F when it has ﬁnite degree over F.

2

In an earlier version of these notes, I said that I didn’t know an elementary proof of this, but several

correspondents sent me such proofs, the simplest of which is the following. It uses only that the product of two

nonsquares in F

×

p

is a square, which follows from the fact that F

×

p

is cyclic (see Exercise 1-3). If 2 is a square

in Fp, then

X

4

− 10X

2

+ 1 = (X

2

− 2

√

2X − 1)(X

2

+ 2

√

2X − 1).

If 3 is a square in Fp, then

X

4

− 10X

2

+ 1 = (X

2

− 2

√

3X + 1)(X

2

+ 2

√

3X + 1).

If neither 2 nor 3 are squares, 6 will be a square in Fp, and

X

4

− 10X

2

+ 1 = (X

2

− (5 + 2

√

6))(X

2

− (5 − 2

√

6)).

The general study of such polynomials requires nonelementary methods. See, for example, the paper

Brandl, R., Amer. Math. Monthly, 93 (1986), pp286–288, which proves that every nonprime integer n ≥ 1

occurs as the degree of a polynomial in Z[X] that is irreducible over Z but reducible modulo all primes.

Construction of some extension ﬁelds 7

EXAMPLE 1.19. (a) The ﬁeld of complex numbers C has degree 2 over R (basis ¦1, i¦).

(b) The ﬁeld of real numbers R has inﬁnite degree over Q — because Q is countable,

every ﬁnite-dimensional Q-vector space is also countable, but a famous argument of Can-

tor shows that R is not countable. More explicitly, there are speciﬁc real numbers α, for

example, π, whose powers 1, α, α

2

, . . . are linearly independent over Q (see the subsection

on transcendental numbers p12).

(c) The ﬁeld of Gaussian numbers

Q(i)

df

= ¦a +bi ∈ C [ a, b ∈ Q¦

has degree 2 over Q (basis ¦1, i¦).

(d) The ﬁeld F(X) has inﬁnite degree over F; in fact, even its subspace F[X] has

inﬁnite dimension over F (basis 1, X, X

2

, . . .).

PROPOSITION 1.20 (MULTIPLICATIVITY OF DEGREES). Let L ⊃ E ⊃ F (all ﬁelds and

subﬁelds). Then L/F is of ﬁnite degree if and only if L/E and E/F are both of ﬁnite

degree, in which case

[L : F] = [L : E][E : F].

PROOF. If Lis of ﬁnite degree over F, then it is certainly of ﬁnite degree over E. Moreover,

E, being a subspace of a ﬁnite dimensional F-vector space, is also ﬁnite dimensional.

Thus, assume that L/E and E/F are of ﬁnite degree, and let (e

i

)

1≤i≤m

be a basis for E

as an F-vector space and let (l

j

)

1≤j≤n

be a basis for L as an E-vector space. To complete

the proof, it sufﬁces to show that (e

i

l

j

)

1≤i≤m,1≤j≤n

is a basis for L over F, because then

L will be ﬁnite over F of the predicted degree.

First, (e

i

l

j

)

i,j

spans L. Let γ ∈ L. Then, because (l

j

)

j

spans L as an E-vector space,

γ =

¸

j

α

j

l

j

, some α

j

∈ E,

and because (e

i

)

i

spans E as an F-vector space,

α

j

=

¸

i

a

ij

e

i

, some a

ij

∈ F.

On putting these together, we ﬁnd that

γ =

¸

i,j

a

ij

e

i

l

j

.

Second, (e

i

l

j

)

i,j

is linearly independent. A linear relation

¸

a

ij

e

i

l

j

= 0, a

ij

∈ F,

can be rewritten

¸

j

(

¸

i

a

ij

e

i

)l

j

= 0. The linear independence of the l

j

’s now shows that

¸

i

a

ij

e

i

= 0 for each j, and the linear independence of the e

i

’s shows that each a

ij

= 0.

2

Construction of some extension ﬁelds

Let f(X) ∈ F[X] be a monic polynomial of degree m, and let (f) be the ideal gener-

ated by f. Consider the quotient ring F[X]/(f(X)), and write x for the image of X in

F[X]/(f(X)), i.e., x is the coset X + (f(X)). Then:

(a) The map

P(X) →P(x): F[X] →F[x]

is a surjective homomorphism in which f(X) maps to 0. Therefore, f(x) = 0.

8 1 BASIC DEFINITIONS AND RESULTS

(b) From the division algorithm, we know that each element g of F[X]/(f) is rep-

resented by a unique polynomial r of degree < m. Hence each element of F[x] can be

expressed uniquely as a sum

a

0

+a

1

x + +a

m−1

x

m−1

, a

i

∈ F. (*)

(c) To add two elements, expressed in the form (*), simply add the corresponding coef-

ﬁcients.

(d) To multiply two elements expressed in the form (*), multiply in the usual way, and

use the relation f(x) = 0 to express the monomials of degree ≥ m in x in terms of lower

degree monomials.

(e) Now assume f(X) is irreducible. To ﬁnd the inverse of an element α ∈ F[x], use

(b) to write α = g(x) with g(X) is a polynomial of degree ≤ m − 1, and use Euclid’s

algorithm in F[X] to obtain polynomials a(X) and b(X) such that

a(X)f(X) +b(X)g(X) = d(X)

with d(X) the gcd of f and g. In our case, d(X) is 1 because f(X) is irreducible and

deg g(X) < deg f(X). When we replace X with x, the equality becomes

b(x)g(x) = 1.

Hence b(x) is the inverse of g(x).

From these observations, we can conclude:

1.21. For a monic irreducible polynomial f(X) of degree m in F[X],

F[x] = F[X]/(f(X))

is a ﬁeld of degree m over F. Moreover, computations in F[x] reduce to computations in

F.

EXAMPLE 1.22. Let f(X) = X

2

+ 1 ∈ R[X]. Then R[x] has:

elements: a +bx, a, b ∈ R;

addition: (a +bx) + (a

t

+b

t

x) = (a +a

t

) + (b +b

t

)x;

multiplication: (a +bx)(a

t

+b

t

x) = (aa

t

−bb

t

) + (ab

t

+a

t

b)x.

We usually write i for x and C for R[x].

EXAMPLE 1.23. Let f(X) = X

3

− 3X − 1 ∈ Q[X]. We observed in (1.12) that this is

irreducible over Q, and so Q[x] is a ﬁeld. It has basis ¦1, x, x

2

¦ as a Q-vector space. Let

β = x

4

+ 2x

3

+ 3 ∈ Q[x].

Then using that x

3

−3x −1 = 0, we ﬁnd that β = 3x

2

+ 7x + 5. Because X

3

−3X −1

is irreducible,

gcd(X

3

−3X −1, 3X

2

+ 7X + 5) = 1.

In fact, Euclid’s algorithm (courtesy of Maple) gives

(X

3

−3X −1)(

−7

37

X +

29

111

) + (3X

2

+ 7X + 5)(

7

111

X

2

−

26

111

X +

28

111

) = 1.

Hence

(3x

2

+ 7x + 5)(

7

111

x

2

−

26

111

x +

28

111

) = 1,

and we have found the inverse of β.

Stem ﬁelds 9

Stem ﬁelds

Let f be a monic irreducible polynomial in F[X]. We say that F[α] is a stem ﬁeld

3

for f if

f(α) = 0. Then

α ↔x: F[α] · F[x], F[x] = F[X]/f(X).

Therefore, stem ﬁelds always exist, and each element of a stem ﬁeld F[α] for f has a unique

expression

a

0

+a

1

α + +a

m−1

α

m−1

, a

i

∈ F, m = deg(f),

i.e., 1, α, . . . , α

m−1

is a basis for F[α] over F. Arithmetic in F[α] can be performed using

the same rules as in F[x]. If F[α

t

] is a second stem ﬁeld for f, then there is a unique

F-isomorphism F[α] →F[α

t

] sending α to α

t

.

The subring generated by a subset

An intersection of subrings of a ring is again a ring. Let F be a subﬁeld of a ﬁeld E, and let

S be a subset of E. The intersection of all the subrings of E containing F and S is evidently

the smallest subring of E containing F and S. We call it the subring of E generated by F

and S (or generated over F by S), and we denote it F[S]. When S = ¦α

1

, ..., α

n

¦, we

write F[α

1

, ..., α

n

] for F[S]. For example, C = R[

√

−1].

LEMMA 1.24. The ring F[S] consists of the elements of E that can be expressed as ﬁnite

sums of the form

¸

a

i

1

in

α

i

1

1

α

in

n

, a

i

1

in

∈ F, α

i

∈ S. (*)

PROOF. Let R be the set of all such elements. Evidently, R is a subring containing F and

S and contained in any other such subring. Therefore R equals F[S].

2

EXAMPLE 1.25. The ring Q[π], π = 3.14159..., consists of the complex numbers that can

be expressed as a ﬁnite sum

a

0

+a

1

π +a

2

π

2

+ a

n

π

n

, a

i

∈ Q.

The ring Q[i] consists of the complex numbers of the form a +bi, a, b ∈ Q.

Note that the expression of an element in the form (*) will not be unique in general.

This is so already in R[i].

LEMMA 1.26. Let R be an integral domain containing a subﬁeld F (as a subring). If R is

ﬁnite dimensional when regarded as an F-vector space, then it is a ﬁeld.

PROOF. Let α be a nonzero element of R — we have to show that α has an inverse in R.

The map x →αx: R →Ris an injective linear map of ﬁnite dimensional F-vector spaces,

and is therefore surjective. In particular, there is an element β ∈ R such that αβ = 1.

2

Note that the lemma applies to subrings (containing F) of an extension ﬁeld E of F of

ﬁnite degree.

3

Following A. Albert, Modern Higher Algebra, 1937, who calls the splitting ﬁeld of a polynomial its root

ﬁeld. More formally, a stem ﬁeld for f is a pair (E, α) consisting of a ﬁeld E containing F and a generator α

such that f(α) = 0.

10 1 BASIC DEFINITIONS AND RESULTS

The subﬁeld generated by a subset

An intersection of subﬁelds of a ﬁeld is again a ﬁeld. Let F be a subﬁeld of a ﬁeld E, and let

S be a subset of E. The intersection of all the subﬁelds of E containing F and S is evidently

the smallest subﬁeld of E containing F and S. We call it the subﬁeld of E generated by F

and S (or generated over F by S), and we denote it F(S). It is the ﬁeld of fractions of F[S]

in E, since this is a subﬁeld of E containing F and S and contained in any other such ﬁeld.

When S = ¦α

1

, ..., α

n

¦, we write F(α

1

, ..., α

n

) for F(S). Thus, F[α

1

, . . . , α

n

] consists

of all elements of E that can be expressed as polynomials in the α

i

with coefﬁcients in F,

and F(α

1

, . . . , α

n

) consists of all elements of E that can be expressed as the quotient of

two such polynomials.

Lemma 1.26 shows that F[S] is already a ﬁeld if it is ﬁnite dimensional over F, in

which case F(S) = F[S].

EXAMPLE 1.27. The ﬁeld Q(π), π = 3.14 . . . consists of the complex numbers that can

be expressed as a quotient

g(π)/h(π), g(X), h(X) ∈ Q[X], h(X) = 0.

The ring Q[i] is already a ﬁeld.

An extension E of F is said to be simple if E = F(α) some α ∈ E. For example,

Q(π) and Q[i] are simple extensions of Q.

Let F and F

t

be subﬁelds of a ﬁeld E. The intersection of the subﬁelds of E containing

F and F

t

is evidently the smallest subﬁeld of E containing both F and F

t

. We call it the

composite of F and F

t

in E, and we denote it F F

t

. It can also be described as the subﬁeld

of E generated over F by F

t

, or the subﬁeld generated over F

t

by F:

F(F

t

) = F F

t

= F

t

(F).

Algebraic and transcendental elements

For a ﬁeld F and an element α of an extension ﬁeld E, we have a homomorphism

f(X) →f(α): F[X] →E.

There are two possibilities.

CASE 1: The kernel of the map is (0), so that, for f ∈ F[X],

f(α) = 0 =⇒ f = 0 (in F[X]).

In this case, we say that α transcendental over F. The homomorphism F[X] → F[α] is

an isomorphism, and it extends to an isomorphism F(X) →F(α).

CASE 2: The kernel is = (0), so that g(α) = 0 for some nonzero g ∈ F[X]. In this case,

we say that α is algebraic over F. The polynomials g such that g(α) = 0 form a nonzero

ideal in F[X], which is generated by the monic polynomial f of least degree such f(α) = 0.

We call f the minimum polynomial of α over F. It is irreducible, because otherwise there

would be two nonzero elements of E whose product is zero. The minimum polynomial is

characterized as an element of F[X] by each of the following sets of conditions:

f is monic; f(α) = 0 and divides every other polynomial g in F[X] with g(α) = 0.

f is the monic polynomial of least degree such f(α) = 0;

Algebraic and transcendental elements 11

f is monic, irreducible, and f(α) = 0.

Note that g(X) → g(α) deﬁnes an isomorphism F[X]/(f) → F[α]. Since the ﬁrst is a

ﬁeld, so also is the second:

F(α) = F[α].

Thus, F[α] is a stem ﬁeld for f.

EXAMPLE 1.28. Let α ∈ C be such that α

3

−3α −1 = 0. Then X

3

−3X −1 is monic,

irreducible, and has α as a root, and so it is the minimum polynomial of α over Q. The set

¦1, α, α

2

¦ is a basis for Q[α] over Q. The calculations in Example 1.23 show that if β is

the element α

4

+ 2α

3

+ 3 of Q[α], then β = 3α

2

+ 7α + 5, and

β

−1

=

7

111

α

2

−

26

111

α +

28

111

.

REMARK 1.29. Maple knows how to compute in Q[α]. For example,

factor(Xˆ4+4); returns the factorization

(X

2

−2X + 2)(X

2

+ 2X + 2).

Now type: alias(c=RootOf(Xˆ2+2

*

X+2));. Then

factor(Xˆ4+4,c); returns the factorization

(X +c)(X −2 −c)(X + 2 +c)(X −c),

i.e., Maple has factored X

4

+ 4 in Q[c] where c has minimum polynomial X

2

+ 2X + 2.

A ﬁeld extension E/F is said to be algebraic, or E is said to be algebraic over F, if all

elements of E are algebraic over F; otherwise it is said to be transcendental (or E is said

to be transcendental over F). Thus, E/F is transcendental if at least one element of E is

transcendental over F.

PROPOSITION 1.30. A ﬁeld extension E/F is ﬁnite if and only if E is algebraic and ﬁ-

nitely generated (as a ﬁeld) over F.

PROOF. =⇒: To say that α is transcendental over F amounts to saying that its powers

1, α, α

2

, . . . are linearly independent over F. Therefore, if E is ﬁnite over F, then it is

algebraic over F. It remains to show that E is ﬁnitely generated over F. If E = F, then it

is generated by the empty set. Otherwise, there exists an α

1

∈ E F. If E = F[α

1

], there

exists an α

2

∈ E F[α

1

], and so on. Since

[F[α

1

] : F] < [F[α

1

, α

2

] : F] < < [E: F]

this process terminates.

⇐=: Let E = F(α

1

, ..., α

n

) with α

1

, α

2

, . . . α

n

algebraic over F. The extension

F(α

1

)/F is ﬁnite because α

1

is algebraic over F, and the extension F(α

1

, α

2

)/F(α

1

)

is ﬁnite because α

2

is algebraic over F and hence over F(α

1

). Thus, by ( 1.20), F(α

1

, α

2

)

is ﬁnite over F. Now repeat the argument.

2

COROLLARY 1.31. (a) If E is algebraic over F, then any subring R of E containing F is

a ﬁeld.

(b) If in L ⊃ E ⊃ F, L is algebraic over E and E is algebraic over F, then L is

algebraic over F.

12 1 BASIC DEFINITIONS AND RESULTS

PROOF. (a) We observed above (p11), that if α is algebraic over F, then F[α] is a ﬁeld. If

α ∈ R, then F[α] ⊂ R, and so α has an inverse in R.

(b) Any α ∈ L is a root of some monic polynomial f = X

m

+a

m−1

X

m−1

+ +a

0

∈

E[X]. Now each of the extensions F[a

0

, . . . , a

m−1

, α] ⊃ F[a

0

, . . . , a

m−1

] ⊃ F is ﬁnite

(1.20), and so F[a

0

, . . . , a

m−1

, α] is ﬁnite (hence algebraic) over F.

2

Transcendental numbers

A complex number is said to be algebraic or transcendental according as it is algebraic or

transcendental over Q. First some history:

1844: Liouville showed that certain numbers, now called Liouville numbers, are tran-

scendental.

1873: Hermite showed that e is transcendental.

1874: Cantor showed that the set of algebraic numbers is countable, but that R is not

countable. Thus most numbers are transcendental (but it is usually very difﬁcult to prove

that any particular number is transcendental).

4

1882: Lindemann showed that π is transcendental.

1934: Gel’fond and Schneider independently showed that α

β

is transcendental if α and

β are algebraic, α = 0, 1, and β / ∈ Q. (This was the seventh of Hilbert’s famous problems.)

2004: Euler’s constant

γ = lim

n→∞

n

¸

k=1

1/k −log n

**has not yet been proven to be transcendental or even irrational.
**

2004: The numbers e + π and e −π are surely transcendental, but again they have not

even been proved to be irrational!

PROPOSITION 1.32. The set of algebraic numbers is countable.

PROOF. Deﬁne the height h(r) of a rational number to be max([m[, [n[), where r = m/n

is the expression of r in its lowest terms. There are only ﬁnitely many rational numbers

with height less than a ﬁxed number N. Let A(N) be the set of algebraic numbers whose

minimum equation over Qhas degree ≤ N and has coefﬁcients of height < N. Then A(N)

is ﬁnite for each N. Count the elements of A(10); then count the elements of A(100); then

count the elements of A(1000), and so on.

5

2

A typical Liouville number is

¸

∞

n=0

1

10

n!

— in its decimal expansion there are increas-

ingly long strings of zeros. We prove that the analogue of this number in base 2 is transcen-

dental.

THEOREM 1.33. The number α =

¸

1

2

n!

is transcendental.

4

In 1873 Cantor proved the rational numbers countable. . . . He also showed that the algebraic numbers. . .

were countable. However his attempts to decide whether the real numbers were countable proved harder. He

had proved that the real numbers were not countable by December 1873 and published this in a paper in 1874

(http://www-gap.dcs.st-and.ac.uk/

∼

history/Mathematicians/Cantor.html).

5

More precisely, choose a bijection from some segment [0, n(1)] of N onto A(10); extend it to a bijection

from a segment [0, n(2)] onto A(100), and so on.

Constructions with straight-edge and compass. 13

PROOF.

6

Suppose not, and let

f(X) = X

d

+a

1

X

d−1

+ +a

d

, a

i

∈ Q,

be the minimum polynomial of α over Q. Thus [Q[α] : Q] = d. Choose a nonzero integer

D such that D f(X) ∈ Z[X].

Let Σ

N

=

¸

N

n=0

1

2

n!

, so that Σ

N

→ α as N → ∞, and let x

N

= f(Σ

N

). If α

is rational,

7

f(X) = X − α; otherwise, f(X), being irreducible of degree > 1, has no

rational root. Since Σ

N

= α, it can’t be a root of f(X), and so x

N

= 0. Evidently,

x

N

∈ Q; in fact (2

N!

)

d

Dx

N

∈ Z, and so

[(2

N!

)

d

Dx

N

[ ≥ 1. (*)

From the fundamental theorem of algebra (see 5.6 below), we know that f splits in

C[X], say,

f(X) =

d

¸

i=1

(X −α

i

), α

i

∈ C, α

1

= α,

and so

[x

N

[ =

d

¸

i=1

[Σ

N

−α

i

[ ≤ [Σ

N

−α

1

[(Σ

N

+M)

d−1

, where M = max

i,=1

¦1, [α

i

[¦.

But

[Σ

N

−α

1

[ =

∞

¸

n=N+1

1

2

n!

≤

1

2

(N+1)!

∞

¸

n=0

1

2

n

=

2

2

(N+1)!

.

Hence

[x

N

[ ≤

2

2

(N+1)!

(Σ

N

+M)

d−1

and

[(2

N!

)

d

Dx

N

[ ≤ 2

2

dN!

D

2

(N+1)!

(Σ

N

+M)

d−1

which tends to 0 as N →∞because

2

d·N!

2

(N+1)!

=

2

d

2

N+1

N!

→0. This contradicts (*).

2

Constructions with straight-edge and compass.

The Greeks understood integers and the rational numbers. They were surprised to ﬁnd

that the length of the diagonal of a square of side 1, namely,

√

2, is not rational. They

thus realized that they needed to extend their number system. They then hoped that the

“constructible” numbers would sufﬁce. Suppose we are given a length, which we call 1, a

straight-edge, and a compass (device for drawing circles). A real number (better a length)

is constructible if it can be constructed by forming successive intersections of

— lines drawn through two points already constructed, and

— circles with centre a point already constructed and radius a constructed length.

6

This proof, which I learnt from David Masser, also works for

¸

1

a

n!

for any integer a ≥ 2.

7

In fact α is not rational because its expansion to base 2 is not periodic.

14 1 BASIC DEFINITIONS AND RESULTS

This led them to three famous questions that they were unable to answer: is it possible

to duplicate the cube, trisect an angle, or square the circle by straight-edge and compass

constructions? We’ll see that the answer to all three is negative.

Let F be a subﬁeld of R. For a positive a ∈ F,

√

a denotes the positive square root of

a in R. The F-plane is F F ⊂ R R. We make the following deﬁnitions:

An F-line is a line in R R through two points in the F-plane. These are the

lines given by equations

ax +by +c = 0, a, b, c ∈ F.

An F-circle is a circle in R R with centre an F-point and radius an element

of F. These are the circles given by equations

(x −a)

2

+ (y −b)

2

= c

2

, a, b, c ∈ F.

LEMMA 1.34. Let L = L

t

be F-lines, and let C = C

t

be F-circles.

(a) L ∩ L

t

= ∅ or consists of a single F-point.

(b) L ∩ C = ∅ or consists of one or two points in the F[

√

e]-plane, some e ∈ F, e > 0.

(c) C ∩ C

t

= ∅ or consists of one or two points in the F[

√

e]-plane, some e ∈ F, e > 0.

PROOF. The points in the intersection are found by solving the simultaneous equations,

and hence by solving (at worst) a quadratic equation with coefﬁcients in F.

2

LEMMA 1.35. (a) If c and d are constructible, then so also are c+d, −c, cd, and

c

d

(d = 0).

(b) If c > 0 is constructible, then so also is

√

c.

PROOF (SKETCH). First show that it is possible to construct a line perpendicular to a given

line through a given point, and then a line parallel to a given line through a given point.

Hence it is possible to construct a triangle similar to a given one on a side with given length.

By an astute choice of the triangles, one constructs cd and c

−1

. For (b), draw a circle of

radius

c+1

2

and centre (

c+1

2

, 0), and draw a vertical line through the point A = (1, 0) to

meet the circle at P. The length AP is

√

c. (For more details, see Artin, M., 1991, Algebra,

Prentice Hall, Chapter 13, Section 4.)

2

THEOREM 1.36. (a) The set of constructible numbers is a ﬁeld.

(b) A number α is constructible if and only if it is contained in a subﬁeld of R of the

form

Q[

√

a

1

, . . . ,

√

a

r

], a

i

∈ Q[

√

a

1

, . . . ,

√

a

i−1

], a

i

> 0.

PROOF. (a) Immediate from (a) of Lemma 1.35.

(b) It follows from Lemma 1.34 that every constructible number is contained in such

a ﬁeld Q[

√

a

1

, . . . ,

√

a

r

]. Conversely, if all the elements of Q[

√

a

1

, . . . ,

√

a

i−1

] are con-

structible, then

√

a

i

is constructible (by 1.35b), and so all the elements of Q[

√

a

1

, . . . ,

√

a

i

]

are constructible (by (a)). Applying this for i = 0, 1, . . ., we ﬁnd that all the elements of

Q[

√

a

1

, . . . ,

√

a

r

] are constructible.

2

COROLLARY 1.37. If α is constructible, then α is algebraic over Q, and [Q[α] : Q] is a

power of 2.

Constructions with straight-edge and compass. 15

PROOF. According to Proposition 1.20, [Q[α] : Q] divides [Q[

√

a

1

] [

√

a

r

] : Q] and

[Q[

√

a

1

, . . . ,

√

a

r

] : Q] is a power of 2.

2

COROLLARY 1.38. It is impossible to duplicate the cube by straight-edge and compass

constructions.

PROOF. The problem is to construct a cube with volume 2. This requires constructing the

real root of the polynomial X

3

− 2. But this polynomial is irreducible (by Eisenstein’s

criterion 1.16 for example), and so [Q[

3

√

2] : Q] = 3.

2

COROLLARY 1.39. In general, it is impossible to trisect an angle by straight-edge and

compass constructions.

PROOF. Knowing an angle is equivalent to knowing the cosine of the angle. Therefore, to

trisect 3α, we have to construct a solution to

cos 3α = 4 cos

3

α −3 cos α.

For example, take 3α = 60 degrees. As cos 60

◦

=

1

2

, to construct α, we have to solve

8x

3

−6x −1 = 0, which is irreducible (apply 1.11).

2

COROLLARY 1.40. It is impossible to square the circle by straight-edge and compass con-

structions.

PROOF. A square with the same area as a circle of radius r has side

√

πr. Since π is

transcendental

8

, so also is

√

π.

2

We now consider another famous old problem, that of constructing a regular polygon.

Note that X

m

−1 is not irreducible; in fact

X

m

−1 = (X −1)(X

m−1

+X

m−2

+ + 1).

LEMMA 1.41. If p is prime then X

p−1

+ +1 is irreducible; hence Q[e

2πi/p

] has degree

p −1 over Q.

PROOF. Let f(X) = (X

p

−1)/(X −1) = X

p−1

+ + 1; then

f(X + 1) =

(X + 1)

p

−1

X

= X

p−1

+ +a

2

X

2

+a

1

X +p,

with a

i

=

p

i+1

. Nowp[a

i

for i = 1, ..., p−2, and so f(X+1) is irreducible by Eisenstein’s

criterion 1.16. This implies that f(X) is irreducible.

2

In order to construct a regular p-gon, p an odd prime, we need to construct

cos

2π

p

= (e

2πi

p

+ (e

2πi

p

)

−1

)/2.

But

Q[e

2πi

p

] ⊃ Q[cos

2π

p

] ⊃ Q,

8

Proofs of this can be found in many books on number theory, for example, in 11.14 of

Hardy, G. H., and Wright, E. M., An Introduction to the Theory of Numbers, Fourth Edition, Oxford, 1960.

16 1 BASIC DEFINITIONS AND RESULTS

and the degree of Q[e

2πi

p

] over Q[cos

2π

p

] is 2 — the equation

α

2

−2 cos

2π

p

α + 1 = 0, α = e

2πi

p

,

shows that it is ≤ 2, and it is not 1 because Q[e

2πi

p

] is not contained in R. Hence

[Q[cos

2π

p

] : Q] =

p −1

2

.

Thus, if the regular p-gon is constructible, then (p−1)/2 = 2

k

for some k (later (5.12),

we shall see a converse), which implies p = 2

k+1

+ 1. But 2

r

+ 1 can be a prime only if r

is a power of 2, because otherwise r has an odd factor t and for t odd,

Y

t

+ 1 = (Y + 1)(Y

t−1

−Y

t−2

+ + 1);

whence

2

st

+ 1 = (2

s

+ 1)((2

s

)

t−1

−(2

s

)

t−2

+ + 1).

Thus if the regular p-gon is constructible, then p = 2

2

k

+1 for some k. Fermat conjectured

that all numbers of the form 2

2

k

+1 are prime, and claimed to show that this is true for k ≤ 5

—for this reason primes of this formare called Fermat primes. For 0 ≤ k ≤ 4, the numbers

p = 3, 5, 17, 257, 65537, are prime but Euler showed that 2

32

+ 1 = (641)(6700417), and

we don’t know of any more Fermat primes.

Gauss showed that

9

cos

2π

17

= −

1

16

+

1

16

√

17+

1

16

34 −2

√

17+

1

8

17 + 3

√

17 −

34 −2

√

17 −2

34 + 2

√

17

when he was 18 years old. This success encouraged him to become a mathematician.

Algebraically closed ﬁelds

We say that a polynomial splits in F[X] if it is a product of polynomials of degree 1 in

F[X].

PROPOSITION 1.42. For a ﬁeld Ω, the following statements are equivalent:

(a) Every nonconstant polynomial in Ω[X] splits in Ω[X].

(b) Every nonconstant polynomial in Ω[X] has at least one root in Ω.

(c) The irreducible polynomials in Ω[X] are those of degree 1.

(d) Every ﬁeld of ﬁnite degree over Ω equals Ω.

PROOF. The implications (a) =⇒(b) =⇒(c) =⇒(a) are obvious.

(c) =⇒ (d). Let E be a ﬁnite extension of Ω. The minimum polynomial of any element α

of E has degree 1, and so α ∈ Ω.

(d) =⇒ (c). Let f be an irreducible polynomial in Ω[X]. Then Ω[X]/(f) is an extension

ﬁeld of Ω of degree deg(f) (see 1.30), and so deg(f) = 1.

2

9

Or perhaps that

cos

2π

17

= −

1

16

+

1

16

√

17 +

1

16

34 − 2

√

17 +

1

8

17 + 3

√

17 − 2

34 − 2

√

17 −

170 − 26

√

17

— both expressions are correct.

Algebraically closed ﬁelds 17

DEFINITION 1.43. (a) A ﬁeld Ω is said to be algebraically closed when it satisﬁes the

equivalent statements of Proposition 1.42.

(b) A ﬁeld Ω is said to be an algebraic closure of a subﬁeld F when it is algebraically

closed and algebraic over F.

For example, the fundamental theorem of algebra (see 5.6 below) says that C is alge-

braically closed. It is an algebraic closure of R.

PROPOSITION 1.44. If Ω is algebraic over F and every polynomial f ∈ F[X] splits in

Ω[X], then Ω is algebraically closed (hence an algebraic closure of F).

PROOF. Let f ∈ Ω[X]. We have to show that f has a root in Ω. We know (see 1.21) that f

has a root α in some ﬁnite extension Ω

t

of Ω. Set

f = a

n

X

n

+ +a

0

, a

i

∈ Ω,

and consider the ﬁelds

F ⊂ F[a

0

, . . . , a

n

] ⊂ F[a

0

, . . . , a

n

, α].

Each extension is algebraic and ﬁnitely generated, and hence ﬁnite (by 1.30). Therefore α

lies in a ﬁnite extension of F, and so is algebraic over F — it is a root of a polynomial g

with coefﬁcients in F. By assumption, g splits in Ω[X], and so the roots of g in Ω

t

all lie in

Ω. In particular, α ∈ Ω.

2

PROPOSITION 1.45. Let Ω ⊃ F; then

¦α ∈ Ω [ α algebraic over F¦

is a ﬁeld.

PROOF. If α and β are algebraic over F, then F[α, β] is a ﬁeld (by 1.31) of ﬁnite degree

over F (by 1.30). Thus, every element of F[α, β] is algebraic over F, including α ± β,

α/β, αβ.

2

The ﬁeld constructed in the lemma is called the algebraic closure of F in Ω.

COROLLARY 1.46. Let Ω be an algebraically closed ﬁeld. For any subﬁeld F of Ω, the

algebraic closure of F in Ω is an algebraic closure of F.

PROOF. From its deﬁnition, we see that it is algebraic over F and every polynomial in

F[X] splits in it. Now Proposition 1.44 shows that it is an algebraic closure of F.

2

Thus, when we admit the fundamental theorem of algebra (5.6), every subﬁeld of C has

an algebraic closure (in fact, a canonical algebraic closure). Later ('6) we shall show that

the axiom of choice implies that every ﬁeld has an algebraic closure.

18 1 BASIC DEFINITIONS AND RESULTS

Exercises

Exercises marked with an asterisk were required to be handed in.

1-1 (*). Let E = Q[α], where α

3

−α

2

+α + 2 = 0. Express (α

2

+α + 1)(α

2

−α) and

(α −1)

−1

in the form aα

2

+bα +c with a, b, c ∈ Q.

1-2 (*). Determine [Q(

√

2,

√

3) : Q].

1-3 (*). Let F be a ﬁeld, and let f(X) ∈ F[X].

(a) For any a ∈ F, show that there is a polynomial q(X) ∈ F[X] such that

f(X) = q(X)(X −a) +f(a).

(b) Deduce that f(a) = 0 if and only if (X −a)[f(X).

(c) Deduce that f(X) can have at most deg f roots.

(d) Let G be a ﬁnite abelian group. If G has at most m elements of order dividing m for

each divisor m of (G : 1), show that G is cyclic.

(e) Deduce that a ﬁnite subgroup of F

, F a ﬁeld, is cyclic.

1-4 (*). Show that with straight-edge, compass, and angle-trisector, it is possible to con-

struct a regular 7-gon.

19

2 Splitting ﬁelds; multiple roots

Maps from simple extensions.

Let E and E

t

be ﬁelds containing F. An F-homomorphism is a homomorphism

ϕ: E →E

t

such that ϕ(a) = a for all a ∈ F. Thus an F-homorphism ϕ maps a polynomial

¸

a

i

1

im

α

i

1

1

α

im

m

, a

i

1

im

∈ F,

to

¸

a

i

1

im

ϕ(α

1

)

i

1

ϕ(α

m

)

im

.

An F-isomorphism is a bijective F-homomorphism. Note that if E and E

t

have the same

ﬁnite degree over F, then every F-homomorphism is an F-isomorphism.

PROPOSITION 2.1. Let F(α) be a simple ﬁeld extension of a ﬁeld F, and let Ω be a second

ﬁeld containing F.

(a) Let α be transcendental over F. For every F-homomorphismϕ: F(α) →Ω, ϕ(α) is

transcendental over F, and the map ϕ →ϕ(α) deﬁnes a one-to-one correspondence

¦F-homomorphisms ϕ: F(α) →Ω¦ ↔¦elements of Ω transcendental over F¦.

(b) Let α be algebraic over F with minimumpolynomial f(X). For every F-homomorphism

ϕ: F[α] → Ω, ϕ(α) is a root of f(X) in Ω, and the map ϕ → ϕ(α) deﬁnes a one-

to-one correspondence

¦F-homomorphisms ϕ: F[α] →Ω¦ ↔¦roots of f in Ω¦.

In particular, the number of such maps is the number of distinct roots of f in Ω.

PROOF. (a) To say that α is transcendental over F means that F[α] is isomorphic to the

polynomial ring in the symbol α with coefﬁcients in F. For any γ ∈ Ω, there is a unique

F-homomorphism ϕ: F[α] → Ω sending α to γ (see 1.5). This extends to the ﬁeld of

fractions F(α) of F[α] if and only if all nonzero elements of F[α] are sent to nonzero

elements of Ω, which is so if and only if γ is transcendental.

(b) Let f(X) =

¸

a

i

X

i

, and consider an F-homomorphism ϕ: F[α] → Ω. On ap-

plying ϕ to the equation

¸

a

i

α

i

= 0, we obtain the equation

¸

a

i

ϕ(α)

i

= 0, which

shows that ϕ(α) is a root of f(X) in Ω. Conversely, if γ ∈ Ω is a root of f(X), then the

map F[X] → Ω, g(X) → g(γ), factors through F[X]/(f(X)). When composed with

the inverse of the isomorphism X + f(X) → α: F[X]/(f(X)) → F[α], this becomes a

homomorphism F[α] →Ω sending α to γ.

2

We shall need a slight generalization of this result.

PROPOSITION 2.2. Let F(α) be a simple ﬁeld extension of a ﬁeld F, and let ϕ

0

: F →Ω

be a homomorphism of F into a second ﬁeld Ω.

(a) If α is transcendental over F, then the map ϕ → ϕ(α) deﬁnes a one-to-one corre-

spondence

¦extensions ϕ: F(α) →Ω of ϕ

0

¦ ↔¦elements of Ω transcendental over ϕ

0

(F)¦.

20 2 SPLITTING FIELDS; MULTIPLE ROOTS

(b) If α is algebraic over F, with minimum polynomial f(X), then the map ϕ → ϕ(α)

deﬁnes a one-to-one correspondence

¦extensions ϕ: F[α] →Ω of ϕ

0

¦ ↔¦roots of ϕ

0

f in Ω¦.

In particular, the number of such maps is the number of distinct roots of ϕ

0

f in Ω.

By ϕ

0

f we mean the polynomial obtained by applying ϕ

0

to the coefﬁcients of f:

if f =

¸

a

i

X

i

then ϕ

0

f =

¸

ϕ(a

i

)X

i

. By an extension of ϕ

0

to F(α) we mean a

homomorphism ϕ: F(α) →Ω such that ϕ[F = ϕ

0

.

The proof of the proposition is essentially the same as that of the preceding proposition.

Splitting ﬁelds

Let f be a polynomial with coefﬁcients in F. A ﬁeld E containing F is said to split f if f

splits in E[X]: f(X) =

¸

m

i=1

(X −α

i

) with α

i

∈ E. If, in addition, E is generated by the

roots of f,

E = F[α

1

, . . . , α

m

],

then it is called a splitting or root ﬁeld for f. Note that

¸

f

i

(X)

m

i

(m

i

≥ 1) and

¸

f

i

(X)

have the same splitting ﬁelds. Also, that if f has deg(f) − 1 roots in E, then it splits in

E[X].

EXAMPLE 2.3. (a) Let f(X) = aX

2

+ bX + c ∈ Q[X], and let α =

√

b

2

−4ac. The

subﬁeld Q[α] of C is a splitting ﬁeld for f.

(b) Let f(X) = X

3

+ aX

2

+ bX + c ∈ Q[X] be irreducible, and let α

1

, α

2

, α

3

be

its roots in C. Then Q[α

1

, α

2

, α

3

] = Q[α

1

, α

2

] is a splitting ﬁeld for f(X). Note that

[Q[α

1

] : Q] = 3 and that [Q[α

1

, α

2

] : Q[α

1

]] = 1 or 2, and so [Q[α

1

, α

2

] : Q] = 3 or 6.

We’ll see later (4.2) that the degree is 3 if and only if the discriminant of f(X) is a square

in Q. For example, the discriminant of X

3

+ bX + c is −4b

3

− 27c

2

, and so the splitting

ﬁeld of X

3

+ 10X + 1 has degree 6 over Q.

PROPOSITION 2.4. Every polynomial f ∈ F[X] has a splitting ﬁeld E

f

, and

[E

f

: F] ≤ (deg f)!.

PROOF. Let F

1

= F[α

1

] be a stem ﬁeld for some monic irreducible factor of f in F[X].

Then f(α

1

) = 0, and we let F

2

= F

1

[α

2

] be a stem ﬁeld for some monic irreducible factor

of f(X)/(X −α

1

) in F

1

[X]. Continuing in this fashion, we arrive at a splitting ﬁeld E

f

.

Let n = deg f. Then [F

1

: F] = deg g

1

≤ n, [F

2

: F

1

] ≤ n −1, ..., and so [E

f

: E] ≤

n!.

2

REMARK 2.5. For a given integer n, there may or may not exist polynomials of degree n

in F[X] whose splitting ﬁeld has degree n! — this depends on F. For example, there do

not for n > 1 if F = C (see 5.6), nor for n > 2 if F = F

p

(see 4.20) or F = R. However,

later (4.30) we shall see how to write down large numbers of polynomials (in fact inﬁnitely

many) of degree n in Q[X] whose splitting ﬁelds have degree n!.

EXAMPLE 2.6. (a) Let f(X) = (X

p

− 1)/(X − 1) ∈ Q[X], p prime. If ζ is one root of

f, then the remaining roots are ζ

2

, ζ

3

, . . . , ζ

p−1

, and so the splitting ﬁeld of f is Q[ζ].

Splitting ﬁelds 21

(b) Suppose F is of characteristic p, and let f = X

p

−X −a ∈ F[X]. If α is one root

of f, then the remaining roots are α + 1, ..., α + p − 1, and so any ﬁeld generated over F

by α is a splitting ﬁeld for f (and F[α] · F[X]/(f) if f is irreducible).

(c) If α is one root of X

n

− a, then the remaining roots are all of the form ζα, where

ζ

n

= 1. Therefore, if F contains all the n

th

roots of 1 (by which we mean that X

n

−1 splits

in F[X]), then F[α] is a splitting ﬁeld for X

n

−a. Note that if p is the characteristic of F,

then X

p

−1 = (X −1)

p

, and so F automatically contains all the p

th

roots of 1.

PROPOSITION 2.7. Let f ∈ F[X]. Assume that E ⊃ F is generated by roots of f, and let

Ω ⊃ F be a ﬁeld in which f splits.

(a) There exists at least one F-homomorphism ϕ: E →Ω.

(b) The number of F-homomorphisms E →Ω is ≤ [E : F], and equals [E : F] if f has

deg(f) distinct roots in Ω.

(c) If E and Ω are both splitting ﬁelds for f, then each F-homomorphism E → Ω is an

isomorphism. In particular, any two splitting ﬁelds for f are F-isomorphic.

PROOF. By f having deg(f) distinct roots in Ω, we mean that

f(X) =

¸

deg(f)

i=1

(X −α

i

), α

i

∈ Ω, α

i

= α

j

if i = j.

If f has this property, then so also does any factor of f in Ω[X].

By assumption, E = F[α

1

, ..., α

m

] with the α

i

roots of f(X). The minimum polyno-

mial of α

1

is an irreducible polynomial f

1

dividing f. As f (hence f

1

) splits in Ω, Propo-

sition 2.1 shows that there exists an F-homomorphism ϕ

1

: F[α

1

] →Ω, and the number of

ϕ

1

’s is ≤ deg(f

1

) = [F[α

1

] : F], with equality holding when f

1

has distinct roots in Ω.

The minimum polynomial of α

2

over F[α

1

] is an irreducible factor f

2

of f in F[α

1

][X].

According to Proposition 2.2, each ϕ

1

extends to a homomorphism ϕ

2

: F[α

1

, α

2

] → Ω,

and the number of extensions is ≤ deg(f

2

) = [F[α

1

, α

2

] : F[α

1

]], with equality holding

when f

2

has deg(f

2

) distinct roots in Ω.

On combining these statements we conclude that there exists an F-homomorphism

ϕ: F[α

1

, α

2

] → Ω, and that the number of such homomorphisms is ≤ [F[α

1

, α

2

] : F],

with equality holding if f has deg(f) distinct roots in Ω.

After repeating the argument several times, we obtain (a) and (b).

Any F-homomorphism E → Ω is injective, and so, if there exists such a homomor-

phism, [E : F] ≤ [Ω : F]. If E and Ω are both splitting ﬁelds for f then (a) shows that

[E: F] = [Ω: F], and so any F-homomorphism E →Ω is an isomorphism.

2

COROLLARY 2.8. Let E and L be extension ﬁelds of F, with E ﬁnite over F.

(a) The number of F-homomorphisms E →L is at most [E: F].

(b) There exists a ﬁnite extension Ω/L and an F-homomorphism E →Ω.

PROOF. Write E = F[α

1

, . . . , α

m

], and f be the product of the minimum polynomials of

the α

i

. Let Ω be a splitting ﬁeld for f regarded as an element of L[X]. The proposition

shows that there is an F-homomorphism E →Ω, and the number of such homomorphisms

is ≤ [E : F]. This proves (b), and since an F-homomorphism E → L can be regarded as

an F-homomorphism E →Ω, it also proves (a).

2

REMARK 2.9. Let E

1

, E

2

, . . . , E

m

be ﬁnite extensions of F, and let L be an extension of

F. The corollary implies that there exists a ﬁnite extension Ω/L containing an isomorphic

copy of every E

i

.

22 2 SPLITTING FIELDS; MULTIPLE ROOTS

Warning! If E and E

t

are both splitting ﬁelds of f ∈ F[X], then we know there is

an F-isomorphism E → E

t

, but there will in general be no preferred such isomorphism.

Error and confusion can result if you simply identify the ﬁelds.

Multiple roots

Let f, g ∈ F[X]. Even when f and g have no common factor in F[X], one might expect

that they could acquire a common factor in Ω[X] for some Ω ⊃ F. In fact, this doesn’t

happen — greatest common divisors don’t change when the ﬁeld is extended.

PROPOSITION 2.10. Let f and g be polynomials in F[X], and let Ω ⊃ F. If r(X) is

the gcd of f and g computed in F[X], then it is also the gcd of f and g in Ω[X]. In

particular, distinct monic irreducible polynomials in F[X] do not acquire a common root in

any extension ﬁeld of F.

PROOF. Let r

F

(X) and r

Ω

(X) be the greatest common divisors of f and g in F[X] and

Ω[X] respectively. Certainly r

F

(X)[r

Ω

(X) in Ω[X], but Euclid’s algorithm (1.8) shows

that there are polynomials a and b in F[X] such that

a(X)f(X) +b(X)g(X) = r

F

(X),

and so r

Ω

(X) divides r

F

(X) in Ω[X].

For the second statement, note that the hypotheses imply that gcd(f, g) = 1 (in F[X]),

and so f and g can’t acquire a common factor in any extension ﬁeld.

2

The proposition allows us to speak of the greatest common divisor of f and g without

reference to a ﬁeld.

Let f ∈ F[X], and let

f(X) = a

r

¸

i=1

(X −α

i

)

m

i

, α

i

distinct, m

i

≥ 1,

r

¸

i=1

m

i

= deg(f), (*)

be a splitting of f in some extension ﬁeld Ωof F. We say that α

i

is a root of f of multiplicity

m

i

. If m

i

> 1, α

i

is said to be a multiple root of f, and otherwise it is a simple root.

The unordered sequence of integers m

1

, . . . , m

r

in (*) is independent of the extension

ﬁeld Ω in which f splits. Certainly, it is unchanged when Ω is replaced with its subﬁeld

F[α

1

, . . . , α

m

], but F[α

1

, . . . , α

m

] is a splitting ﬁeld for f, and any two splitting ﬁelds are

isomorphic (2.7c).

We say that f has a multiple root when at least of the m

i

> 1, and we say that f has

only simple roots when all m

i

= 1.

We wish to determine when a polynomial has a multiple root. If f has a multiple factor

in F[X], say f =

¸

f

i

(X)

m

i

with some m

i

> 1, then obviously it will have a multiple

root. If f =

¸

f

i

with the f

i

distinct monic irreducible polynomials, then Proposition 2.10

shows that f has a multiple root if and only if at least one of the f

i

has a multiple root.

Thus, it sufﬁces to determine when an irreducible polynomial has a multiple root.

EXAMPLE 2.11. Let F be of characteristic p = 0, and assume that F has contains an

element a that is not a p

th

-power, for example, a = T in the ﬁeld F

p

(T). Then X

p

− a

is irreducible in F[X], but X

p

− a

1.4

= (X − α)

p

in its splitting ﬁeld. Thus an irreducible

polynomial can have multiple roots.

Multiple roots 23

Deﬁne the derivative f

t

(X) of a polynomial f(X) =

¸

a

i

X

i

to be

¸

ia

i

X

i−1

. When

f has coefﬁcients in R, this agrees with the deﬁnition in calculus. The usual rules for

differentiating sums and products still hold, but note that in characteristic p the derivative

of X

p

is zero.

PROPOSITION 2.12. For a nonconstant irreducible polynomial f in F[X], the following

statements are equivalent:

(a) f has a multiple root;

(b) gcd(f, f

t

) = 1;

(c) F has characteristic p = 0 and f is a polynomial in X

p

;

(d) all the roots of f are multiple.

PROOF. (a) =⇒ (b). Let α be a multiple root of f, and write f = (X−α)

m

g(X), m > 1,

in some splitting ﬁeld. Then

f

t

(X) = m(X −α)

m−1

g(X) + (X −α)

m

g

t

(X).

Hence f

t

(α) = 0, and so gcd(f, f

t

) = 1.

(b) =⇒ (c). Since f is irreducible and deg(f

t

) < deg(f),

gcd(f, f

t

) = 1 =⇒ f

t

= 0.

But, because f is nonconstant, f

t

can be zero only if the characteristic is p = 0 and f is a

polynomial in p.

(c) =⇒ (d). Suppose f(X) = g(X

p

), and let g(X) =

¸

i

(X−a

i

)

m

i

in some splitting

ﬁeld for f. Then

f(X) = g(X

p

) =

¸

i

(X

p

−a

i

)

m

i

=

¸

i

(X −α

i

)

pm

i

where α

p

i

= a

i

. Hence every root of f(X) has multiplicity at least p.

(d) =⇒ (a). Obvious.

2

DEFINITION 2.13. A polynomial f ∈ F[X] is said to be separable over Fif none of its

irreducible factors has a multiple root (in a splitting ﬁeld).

10

The preceding discussion shows that f ∈ F[X] will be separable unless

(a) the characteristic of F is p = 0, and

(b) at least one of the irreducible factors of f is a polynomial in X

p

.

Note that, if f ∈ F[X] is separable, then it remains separable over every ﬁeld Ω containing

F (condition (b) of 2.12 continues to hold — see 2.10).

DEFINITION 2.14. A ﬁeld F is said to be perfect if all polynomials in F[X] are separable

(equivalently, all irreducible polynomials in F[X] are separable).

PROPOSITION 2.15. A ﬁeld of characteristic zero is always perfect, and a ﬁeld F of char-

acteristic p = 0 is perfect if and only if every element of F is a p

th

power.

10

This is the standard deﬁnition, although some authors, for example, Dummit and Foote 1991, 13.5, give a

different deﬁnition.

24 2 SPLITTING FIELDS; MULTIPLE ROOTS

PROOF. A ﬁeld of characteristic zero is obviously perfect, and so we may suppose F to be

of characteristic p = 0. If F contains an element a that is not a p

th

power, then X

p

− a ∈

F[X] is not separable (see 2.11). Conversely, if every element of F is a p

th

power, then

every polynomial in X

p

with coefﬁcients in F is a p

th

power in F[X],

¸

a

i

X

p

= (

¸

b

i

X)

p

if a

i

= b

p

i

,

and so is not irreducible.

2

EXAMPLE 2.16. (a) A ﬁnite ﬁeld F is perfect, because the Frobenius endomorphism

a →a

p

: F →F is injective and therefore surjective (by counting).

(b) A ﬁeld that can be written as a union of perfect ﬁelds is perfect. Therefore, every

ﬁeld algebraic over F

p

is perfect.

(c) Every algebraically closed ﬁeld is perfect.

(d) If F

0

has characteristic p = 0, then F = F

0

(X) is not perfect, because X is not a p

th

power.

Exercises

2-1 (*). Let F be a ﬁeld of characteristic = 2.

(a) Let E be quadratic extension of F (i.e., [E : F] = 2); show that

S(E) = ¦a ∈ F

[ a is a square in E¦

is a subgroup of F

containing F

2

.

(b) Let E and E

t

be quadratic extensions of F; show that there is an F-isomorphism

ϕ: E →E

t

if and only if S(E) = S(E

t

).

(c) Show that there is an inﬁnite sequence of ﬁelds E

1

, E

2

, . . . with E

i

a quadratic ex-

tension of Q such that E

i

is not isomorphic to E

j

for i = j.

(d) Let p be an odd prime. Show that, up to isomorphism, there is exactly one ﬁeld with

p

2

elements.

2-2 (*). (a) Let F be a ﬁeld of characteristic p. Show that if X

p

− X − a is reducible in

F[X], then it splits into distinct factors in F[X].

(b) For any prime p, show that X

p

−X −1 is irreducible in Q[X].

2-3 (*). Construct a splitting ﬁeld for X

5

−2 over Q. What is its degree over Q?

2-4 (*). Find a splitting ﬁeld of X

p

m

−1 ∈ F

p

[X]. What is its degree over F

p

?

2-5. Let f ∈ F[X], where F is a ﬁeld of characteristic 0. Let d(X) = gcd(f, f

t

). Show

that g(X) = f(X)d(X)

−1

has the same roots as f(X), and these are all simple roots of

g(X).

2-6 (*). Let f(X) be an irreducible polynomial in F[X], where F has characteristic p.

Show that f(X) can be written f(X) = g(X

p

e

) where g(X) is irreducible and separable.

Deduce that every root of f(X) has the same multiplicity p

e

in any splitting ﬁeld.

25

3 The fundamental theorem of Galois theory

In this section, we prove the fundamental theorem of Galois theory, which gives a one-to-

one correspondence between the subﬁelds of the splitting ﬁeld of a separable polynomial

and the subgroups of the Galois group of f.

Groups of automorphisms of ﬁelds

Consider ﬁelds E ⊃ F. An F-isomorphism E → E is called an F-automorphism of E.

The F-automorphisms of E form a group, which we denote Aut(E/F).

EXAMPLE 3.1. (a) There are two obvious automorphisms of C, namely, the identity map

and complex conjugation. We’ll see later (8.18) that by using the Axiom of Choice one can

construct uncountably many more.

(b) Let E = C(X). Then Aut(E/C) consists of the maps

11

X →

aX+b

cX+d

, ad −bc = 0

(Jacobson 1964, IV, Theorem 7, p158), and so

Aut(E/C) = PGL

2

(C),

the group of invertible 22 matrices with complex coefﬁcients modulo its centre. Analysts

will note that this is the same as the automorphism group of the Riemann sphere. This

is not a coincidence: the ﬁeld of meromorphic functions on the Riemann sphere P

1

C

is

C(z) · C(X), and so there is certainly a map Aut(P

1

C

) → Aut(C(z)/C), which one can

show to be an isomorphism.

(c) The group Aut(C(X

1

, X

2

)/C) is quite complicated — there is a map

PGL

3

(C) = Aut(P

2

C

) →Aut(C(X

1

, X

2

)/C),

but this is very far from being surjective. When there are more X’s, the group is unknown.

(The group Aut(C(X

1

, . . . , X

n

)/C) is the group of birational automorphisms of P

n

C

. It is

called the Cremona group. Its study is part of algebraic geometry.)

In this section, we shall be concerned with the groups Aut(E/F) when E is a ﬁnite

extension of F.

PROPOSITION 3.2. If E is a splitting ﬁeld of a monic separable polynomial f ∈ F[X],

then Aut(E/F) has order [E : F].

PROOF. Let f =

¸

f

m

i

i

, with the f

i

monic irreducible and distinct. The splitting ﬁeld

of f is the same as the splitting ﬁeld of

¸

f

i

. Hence we may assume f is a product of

distinct monic separable irreducible polynomials, and so has deg f distinct roots in E. Now

Proposition 2.7 shows that there are [E : F] distinct F-homomorphisms E →E. Because

E has ﬁnite degree over F, they are automatically isomorphisms.

2

EXAMPLE 3.3. (a) Consider a simple extension E = F[α], and let f be a polynomial with

coefﬁcients in F having α as a root. If f has no root in E other than α, then Aut(E/F) = 1.

For example, if

3

√

2 denotes the real cube root of 2, then Aut(Q[

3

√

2]/Q) = 1. Thus, in the

proposition, it is essential that E be a splitting ﬁeld.

(b) Let F be a ﬁeld of characteristic p = 0, and let a be an element of F that is not a p

th

power. Then f = X

p

− a has only one root in a splitting ﬁeld E, and so Aut(E/F) = 1.

Thus, in the proposition, it is essential that E be a splitting ﬁeld of a separable polynomial.

11

By this I mean the map that sends a rational function f(X) to f(

aX+b

cX+d

).

26 3 THE FUNDAMENTAL THEOREM OF GALOIS THEORY

When G is a group of automorphisms of a ﬁeld E, we set

E

G

= Inv(G) = ¦α ∈ E [ σα = α, all σ ∈ G¦.

It is a subﬁeld of E, called the subﬁeld of G-invariants of E or the ﬁxed ﬁeld of G.

In this section, we shall show that, when E is the splitting ﬁeld of a separable polyno-

mial in F[X] and G = Aut(E/F), then the maps

M →Aut(E/M), H →Inv(H)

give a one-to-one correspondence between the set of intermediate ﬁelds M, F ⊂ M ⊂ E,

and the set of subgroups H of G.

THEOREM 3.4 (E. ARTIN). Let G be a ﬁnite group of automorphisms of a ﬁeld E, and let

F = E

G

; then [E : F] ≤ (G : 1).

PROOF. Let G = ¦σ

1

= 1, . . . , σ

m

¦, and let α

1

, . . . , α

n

be n > m elements of E. We

shall show that the α

i

are linearly dependent over F. In the system of linear equations

σ

1

(α

1

)X

1

+ +σ

1

(α

n

)X

n

= 0

σ

m

(α

1

)X

1

+ +σ

m

(α

n

)X

n

= 0

there are m equations and n > m unknowns, and hence there are nontrivial solutions in E

—choose one (c

1

, . . . , c

n

) having the fewest possible nonzero elements. After renumbering

the α

i

’s, we may suppose that c

1

= 0, and then (after multiplying by a scalar) that c

1

∈ F.

With these normalizations, we’ll show that all c

i

∈ F. Then the ﬁrst equation

α

1

c

1

+ +α

n

c

n

= 0

(recall that σ

1

= 1) will be a linear relation on the α

i

.

If not all c

i

are in F, then σ

k

(c

i

) = c

i

for some k and i, k = 1 = i. On applying σ

k

to

the equations

σ

1

(α

1

)c

1

+ +σ

1

(α

n

)c

n

= 0

σ

m

(α

1

)c

1

+ +σ

m

(α

n

)c

n

= 0

and using that ¦σ

k

σ

1

, . . . , σ

k

σ

m

¦ is a permutation of ¦σ

1

, . . . , σ

m

¦, we ﬁnd that

(c

1

, σ

k

(c

2

), . . . , σ

k

(c

i

), . . .)

is also a solution to the system of equations (*). On subtracting it from the ﬁrst, we obtain

a solution (0, . . . , c

i

− σ

k

(c

i

), . . .), which is nonzero (look at the i

th

coordinate), but has

more zeros than the ﬁrst solution (look at the ﬁrst coordinate) — contradiction.

2

COROLLARY 3.5. For any ﬁnite group Gof automorphisms of a ﬁeld E, G = Aut(E/E

G

).

PROOF. We know that:

[E : E

G

] ≤ (G : 1) (by 3.4),

G ⊂ Aut(E/E

G

) (obvious),

(Aut(E/E

G

): 1) ≤ [E: E

G

] (by 2.8a).

The inequalities

[E : E

G

] ≤ (G : 1) ≤ (Aut(E/E

G

) : 1) ≤ [E : E

G

]

must be equalities, and so G = Aut(E/E

G

).

2

Separable, normal, and Galois extensions 27

Separable, normal, and Galois extensions

DEFINITION 3.6. An algebraic extension E/F is said to be separable if the minimum

polynomial of every element of E is separable; otherwise, it is inseparable.

Thus, an algebraic extension E/F is separable if every irreducible polynomial in F[X]

having a root in E is separable, and it is inseparable if

— F is nonperfect, and in particular has characteristic p = 0, and

— there is an element α of E whose minimal polynomial is of the form g(X

p

), g ∈

F[X].

For example, E = F

p

(T) is an inseparable extension of F

p

(T

p

).

DEFINITION 3.7. An algebraic extension E/F is normal if the minimum polynomial of

every element of E splits in E[X].

In other words, an algebraic extension E/F is normal if every irreducible polynomial

f ∈ F[X] having a root in E splits in E.

Let f be an irreducible polynomial of degree m in F[X]. If f has a root in E, then

E/F separable =⇒ roots of f distinct

E/F normal =⇒ f splits in E

¸

=⇒ f has m distinct roots in E.

Therefore, E/F is normal and separable if and only if, for each α ∈ E, the minimum

polynomial of α has [F[α] : F] distinct roots in E.

EXAMPLE 3.8. (a) The ﬁeld Q[

3

√

2], where

3

√

2 is the real cube root of 2, is separable but

not normal over Q (X

3

−2 doesn’t split in Q[α]).

(b) The ﬁeld F

p

(T) is normal but not separable over F

p

(T

p

) — the minimum polyno-

mial of T is the inseparable polynomial X

p

−T

p

.

DEFINITION 3.9. Let F be a ﬁeld. A ﬁnite extension E of F is said to Galois if F is the

ﬁxed ﬁeld of the group of F-automorphisms of E. This group is then called the Galois

group of E over F, and it is denoted Gal(E/F).

THEOREM 3.10. For an extension E/F, the following statements are equivalent:

(a) E is the splitting ﬁeld of a separable polynomial f ∈ F[X].

(b) F = E

G

for some ﬁnite group G of automorphisms of E.

(c) E is normal and separable, and of ﬁnite degree, over F.

(d) E is Galois over F.

PROOF. (a) =⇒ (d). Let G = Aut(E/F), and let F

t

= E

G

⊃ F. Then E is also the

splitting ﬁeld of f regarded as a polynomial with coefﬁcients in F

t

, and f is still separable

when it is regarded in this way. Hence Proposition 3.2 shows that

[E : F

t

] = #Aut(E/F

t

)

[E : F] = #Aut(E/F).

Since Aut(E/F

t

) = Aut(E/F) = G, we conclude that F = F

t

, and so F = E

G

.

(d) =⇒ (b). According to (2.8a) , Gal(E/F) is ﬁnite, and so this is obvious.

(b) =⇒ (c). By Proposition 3.4, we know that [E : F] ≤ (G : 1); in particular, it is

ﬁnite. Let α ∈ E and let f be the minimum polynomial of α; we have to prove that f splits

28 3 THE FUNDAMENTAL THEOREM OF GALOIS THEORY

into distinct factors in E[X]. Let ¦α

1

= α, ..., α

m

¦ be the orbit of α under the action of G

on E, and let

g(X) =

¸

(X −α

i

) = X

m

+a

1

X

m−1

+ +a

m

.

Any σ ∈ G merely permutes the α

i

. Since the a

i

are symmetric polynomials in the α

i

,

we ﬁnd that σa

i

= a

i

for all i, and so g(X) ∈ F[X]. It is monic, and g(α) = 0, and

so f(X)[g(X) (see the deﬁnition of the minimum polynomial p10). But also g(X)[f(X),

because each α

i

is a root of f(X) (if α

i

= σα, then applying σ to the equation f(α) = 0

gives f(α

i

) = 0). We conclude that f(X) = g(X), and so f(X) splits into distinct factors

in E.

(c) =⇒ (a). Because E has ﬁnite degree over F, it is generated over F by a ﬁnite

number of elements, say, E = F[α

1

, ..., α

m

], α

i

∈ E, α

i

algebraic over F. Let f

i

be the

minimum polynomial of α

i

over F. Because E is normal over F, each f

i

splits in E, and

so E is the splitting ﬁeld of f =

¸

f

i

. Because E is separable over F, f is separable.

2

REMARK 3.11. (a) Let E be Galois over F with Galois group G, and let α ∈ E. The

elements α

1

= α, α

2

, ..., α

m

of the orbit of α are called the conjugates of α. In the course

of the proof of (b) =⇒ (c) of the above theorem we showed that the minimum polynomial

of α is

¸

(X −α

i

).

(b) Note that if F = E

G

for some ﬁnite group G, then, because E is the splitting ﬁeld

of a separable polynomial, Proposition 2.1 shows that Gal(E/F) has [E: F] elements.

Combined with Artin’s theorem (3.4), this shows that G = Gal(E/F) and (G: 1) =

[E: F].

COROLLARY 3.12. Every ﬁnite separable extension E of F is contained in a ﬁnite Galois

extension.

PROOF. Let E = F[α

1

, ..., α

m

]. Let f

i

be the minimum polynomial of α

i

over F, and take

E

t

to be the splitting ﬁeld of

¸

f

i

over F.

2

COROLLARY 3.13. Let E ⊃ M ⊃ F; if E is Galois over F, then it is Galois over M.

PROOF. We knowE is the splitting ﬁeld of some separable f ∈ F[X]; it is also the splitting

ﬁeld of f regarded as an element of M[X].

2

REMARK 3.14. When we drop the assumption that E is separable over F, we can still say

something. An element α of an algebraic extension of F is said to be separable over F if its

minimum polynomial over F is separable. The proof of Corollary 3.12 shows that any ﬁnite

extension generated by separable elements is separable. Therefore, the elements of a ﬁnite

extension E of F that are separable over F form a subﬁeld E

sep

of E that is separable over

F; write [E : F]

sep

= [E

sep

: F] (separable degree of E over F). If Ω is an algebraically

closed ﬁeld containing F, then every F-homomorphism E

sep

→ Ω extends uniquely to E,

and so the number of F-homomorphisms E →Ω is [E : F]

sep

. When E ⊃ M ⊃ F (ﬁnite

extensions),

[E : F]

sep

= [E : M]

sep

[M : F]

sep

.

In particular,

E is separable over F ⇐⇒ E is separable over M and M is separable over F.

See Jacobson 1964, I 10, for more details.

DEFINITION 3.15. A ﬁnite extension E ⊃ F is called a cyclic, abelian, ..., solvable

extension if it is Galois with cyclic, abelian, ..., solvable Galois group.

The fundamental theorem of Galois theory 29

The fundamental theorem of Galois theory

THEOREM 3.16 (FUNDAMENTAL THEOREM OF GALOIS THEORY). Let E be a Galois ex-

tension of F, and let G = Gal(E/F). The maps H → E

H

and M → Gal(E/M) are in-

verse bijections between the set of subgroups of Gand the set of intermediate ﬁelds between

E and F:

¦subgroups of G¦ ↔¦intermediate ﬁelds F ⊂ M ⊂ E¦.

Moreover,

(a) the correspondence is inclusion-reversing: H

1

⊃ H

2

⇐⇒ E

H

1

⊂ E

H

2

;

(b) indexes equal degrees: (H

1

: H

2

) = [E

H

2

: E

H

1

];

(c) σHσ

−1

↔σM, i.e., E

σHσ

−1

= σ(E

H

); Gal(E/σM) = σ Gal(E/M)σ

−1

.

(d) H is normal in G ⇐⇒ E

H

is normal (hence Galois) over F, in which case

Gal(E

H

/F) = G/H.

PROOF. For the ﬁrst statement, we have to show that H →E

H

and M →Gal(E/M) are

inverse maps.

Let H be a subgroup of G. Then, as we observed in (3.11b), Gal(E/E

H

) = H.

Let M be an intermediate ﬁeld. Then E is Galois over M by (3.13), which means that

E

Gal(E/M)

= M .

(a) We have the obvious implications:

H

1

⊃ H

2

=⇒ E

H

1

⊂ E

H

2

=⇒ Gal(E/E

H

1

) ⊃ Gal(E/E

H

2

).

But Gal(E/E

H

i

) = H

i

.

(b) As we observed in (3.11b), for any subgroup H of G, [E : E

H

] = Gal(E/E

H

).

This proves (b) in the case H

2

= 1, and the general case follows, using that

(H

1

: 1) = (H

1

: H

2

)(H

2

: 1) and [E : E

H

1

] = [E : E

H

2

][E

H

2

: E

H

1

].

(c) For τ ∈ Gand α ∈ E, τα = α ⇐⇒ στσ

−1

(σα) = σα. Therefore, Gal(E/σM) =

σ Gal(E/M)σ

−1

, and so σ Gal(E/M)σ

−1

↔σM.

(d) Let H be a normal subgroup of G. Because σHσ

−1

= H for all σ ∈ G, we must

have σE

H

= E

H

for all σ ∈ G, i.e., the action of G on E stabilizes E

H

. We therefore

have a homomorphism

σ →σ[E

H

: G →Aut(E

H

/F)

whose kernel is H. As (E

H

)

G/H

= F, we see that E

H

is Galois over F (by Theorem 3.10)

and that G/H · Gal(E

H

/F) (by 3.11b).

Conversely, assume that M is normal over F, and write M = F[α

1

, ..., α

m

]. For

σ ∈ G, σα

i

is a root of the minimum polynomial of α

i

over F, and so lies in M. Hence

σM = M, and this implies that σHσ

−1

= H (by (c)).

2

REMARK 3.17. The theorem shows that there is an order reversing bijection between the

intermediate ﬁelds of E/F and the subgroups of G. Using this we can read off more results.

(a) Let M

1

, M

2

, . . . , M

r

be intermediate ﬁelds, and let H

i

be the subgroup correspond-

ing to M

i

(i.e., H

i

= Gal(E/M

i

)). Then (by deﬁnition) M

1

M

2

M

r

is the smallest ﬁeld

containing all M

i

; hence it must correspond to the largest subgroup contained in all H

i

,

which is

¸

H

i

. Therefore

Gal(E/M

1

M

r

) = H

1

∩ ... ∩ H

r

.

30 3 THE FUNDAMENTAL THEOREM OF GALOIS THEORY

(b) Let H be a subgroup of Gand let M = E

H

. The largest normal subgroup contained

in H is N =

¸

σ∈G

σHσ

−1

(see GT 4.10), and so E

N

, which is the composite of the ﬁelds

σM, is the smallest normal extension of F containing M. It is called the normal, or Galois,

closure of M in E.

PROPOSITION 3.18. Let E and L be ﬁeld extensions of F contained in some common

ﬁeld. If E/F is Galois, then EL/L and E/E ∩ L are Galois, and the map

σ →σ[E: Gal(EL/L) →Gal(E/E ∩ L)

is an isomorphism.

PROOF. Because E is Galois over F, it is the splitting ﬁeld of a separable polynomial

f ∈ F[X]. Then EL is the splitting ﬁeld of f over L, and E is the split-

ting ﬁeld of f over E ∩ L. Hence EL/L and E/E ∩ L are Galois. Any

automorphism σ of EL ﬁxing the elements of L maps roots of f to roots

of f, and so σE = E. There is therefore a homomorphism

σ →σ[E: Gal(EL/L) →Gal(E/F).

If σ ∈ Gal(EL/L) ﬁxes the elements of E, then it ﬁxes the elements of

EL, and hence is 1. Thus, σ → σ[E is injective. If α ∈ E is ﬁxed by all

σ ∈ Gal(EL/L), then α ∈ L ∩ E. By the fundamental theorem,

EL

E L

=

E ∩ L

=

F

this implies that the image of σ →σ[E is Gal(E/E ∩ L).

2

COROLLARY 3.19. Suppose, in the proposition, that L is ﬁnite over F. Then

[EL: F] =

[E : F][L: F]

[E ∩ L: F]

.

PROOF. According to Proposition 1.20,

[EL: F] = [EL: L][L: F],

but

[EL: L]

3.18

= [E: E ∩ L]

1.20

=

[E : F]

[E ∩ L: F]

.

2

PROPOSITION 3.20. Let E

1

and E

2

be ﬁeld extensions of F contained in some common

ﬁeld. If E

1

and E

2

are Galois over F, then E

1

E

2

and E

1

∩ E

2

are Galois over F, and

σ →(σ[E

1

, σ[E

2

) : Gal(E

1

E

2

/F) →Gal(E

1

/F) Gal(E

2

/F)

is an isomorphism of Gal(E

1

E

2

/F) onto the subgroup

H = ¦(σ

1

, σ

2

) [ σ

1

[E

1

∩ E

2

= σ

2

[E

1

∩ E

2

¦

of Gal(E

1

/F) Gal(E

2

/F).

Examples 31

PROOF: Let a ∈ E

1

∩ E

2

, and let f be its minimum polynomial over F. Then f has

deg f distinct roots in E

1

and deg f distinct roots in E

2

. Since f

can have at most deg f roots in E

1

E

2

, it follows that it has deg f

distinct roots in E

1

∩ E

2

. This shows that E

1

∩ E

2

is normal

and separable over F, and hence Galois (3.10). As E

1

and E

2

are

Galois over F, they are splitting ﬁelds of separable polynomials

f

1

, f

2

∈ F[X]. Now E

1

E

2

is a splitting ﬁeld for f

1

f

2

, and hence

it also is Galois over F. The map σ → (σ[E

1

, σ[E

2

) is clearly

an injective homomorphism, and its image is contained in H. We

prove that the image is the whole of H by counting.

E

1

E

2

E

1

E

2

E

1

∩ E

2

F

From the fundamental theorem,

Gal(E

2

/F)/ Gal(E

2

/E

1

∩ E

2

) · Gal(E

1

∩ E

2

/F),

and so, for each σ

1

∈ Gal(E

1

/F), σ

1

[E

1

∩ E

2

has exactly [E

2

: E

1

∩ E

2

] extensions to an

element of Gal(E

2

/F). Therefore,

(H: 1) = [E

1

: F][E

2

: E

1

∩ E

2

] =

[E

1

: F] [E

2

: F]

[E

1

∩ E

2

: F]

,

which equals [E

1

E

2

: F] by (3.19).

Examples

EXAMPLE 3.21. We analyse the extension Q[ζ]/Q, where ζ is a primitive 7

th

root of 1,

say ζ = e

2πi/7

.

Note that Q[ζ] is the splitting ﬁeld of the polynomial

X

7

−1, and that ζ has minimum polynomial

X

6

+X

5

+X

4

+X

3

+X

2

+X + 1

(see 1.41). Therefore, Q[ζ] is Galois of degree 6 over

Q. For any σ ∈ G, σζ = ζ

i

, some i, 1 ≤ i ≤ 6, and

the map σ → i deﬁnes an isomorphism Gal(Q[ζ]/Q) →

(Z/7Z)

**. Let σ be the element of Gal(Q[ζ]/Q) such that
**

σζ = ζ

3

. Then σ generates Gal(Q[ζ]/Q) because the

class of 3 in (Z/7Z)

**generates it (the powers of 3 mod
**

7 are 3, 2, 6, 4, 5, 1). We investigate the subﬁelds of Q[ζ]

corresponding to the subgroups 'σ

3

` and 'σ

2

`.

Q[ζ]

Q[ζ +ζ]

/σ

3

)

Q[

√

−7]

/σ

2

)

Q

/σ)//σ

2

) /σ)//σ

3

)

Note that σ

3

ζ = ζ

6

= ζ (complex conjugate of ζ). The subﬁeld of Q[ζ] corresponding

to 'σ

3

` is Q[ζ + ζ], and ζ + ζ = 2 cos

2π

7

. Since 'σ

3

` is a normal subgroup of 'σ`,

Q[ζ +ζ] is Galois over Q, with Galois group 'σ`/'σ

3

`. The conjugates of α

1

=

df

ζ +ζ are

32 3 THE FUNDAMENTAL THEOREM OF GALOIS THEORY

α

3

= ζ

3

+ζ

−3

, α

2

= ζ

2

+ζ

−2

. Direct calculation shows that

α

1

+α

2

+α

3

=

¸

6

i=1

ζ

i

= −1,

α

1

α

2

+α

1

α

3

+α

2

α

3

= −2,

α

1

α

2

α

3

= (ζ +ζ

6

)(ζ

2

+ζ

5

)(ζ

3

+ζ

4

)

= (ζ +ζ

3

+ζ

4

+ζ

6

)(ζ

3

+ζ

4

)

= (ζ

4

+ζ

6

+ 1 +ζ

2

+ζ

5

+ 1 +ζ +ζ

3

)

= 1.

Hence the minimum polynomial

12

of ζ +ζ is

g(X) = X

3

+X

2

−2X −1.

The minimum polynomial of cos

2π

7

=

α

1

2

is therefore

g(2X)

8

= X

3

+X

2

/2 −X/2 −1/8.

The subﬁeld of Q[ζ] corresponding to 'σ

2

` is generated by β = ζ + ζ

2

+ ζ

4

. Let

β

t

= σβ. Then (β −β

t

)

2

= −7. Hence the ﬁeld ﬁxed by 'σ

2

` is Q[

√

−7].

EXAMPLE 3.22. We compute the Galois group of a splitting ﬁeld E of X

5

− 2 ∈ Q[X].

Recall from Exercise 2-3 that E = Q[ζ, α] where ζ is a primitive

5

th

root of 1, and α is a root of X

5

− 2. For example, we could

take E to be the splitting ﬁeld of X

5

−2 in C, with ζ = e

2πi/5

and

α equal to the real 5

th

root of 2. We have the picture at right, and

[Q[ζ] : Q] = 4, [Q[α] : Q] = 5.

Because 4 and 5 are relatively prime,

[Q[ζ, α] : Q] = 20.

Q[ζ, α]

Q[ζ]

N

Q[α]

H

Q

G/N

Hence G = Gal(Q[ζ, α]/Q) has order 20, and the subgroups N and H corresponding to

Q[ζ] and Q[α] have orders 5 and 4 respectively. Because Q[ζ] is normal over Q (it is the

splitting ﬁeld of X

5

− 1), N is normal in G. Because Q[ζ] Q[α] = Q[ζ, α], we have

H ∩ N = 1, and so G = N

θ

H. Moreover, H · G/N · (Z/5Z)

, which is cyclic,

being generated by the class of 2. Let τ be the generator of H corresponding to 2 under this

isomorphism, and let σ be a generator of N. Thus σ(α) is another root of X

5

− 2, which

we can take to be ζα (after possibly replacing σ by a power). Hence:

τζ = ζ

2

τα = α

σζ = ζ

σα = ζα.

Note that τστ

−1

(α) = τσα = τ(ζα) = ζ

2

α and it ﬁxes ζ; therefore τστ

−1

= σ

2

. Thus

G has generators σ and τ and deﬁning relations

σ

5

= 1, τ

4

= 1, τστ

−1

= σ

2

.

12

More directly, on setting X = ζ +ζ in

(X

3

− 3X) + (X

2

− 2) +X + 1

one obtains 1 +ζ +ζ

2

+ · · · +ζ

6

= 0.

Constructible numbers revisited 33

The subgroup H has ﬁve conjugates, which correspond to the ﬁve ﬁelds Q[ζ

i

α],

σ

i

Hσ

−i

↔σ

i

Q[α] = Q[ζ

i

α], 1 ≤ i ≤ 5.

Constructible numbers revisited

Earlier, we showed (1.36) that a real number α is constructible if and only if it is con-

tained in a subﬁeld of R of the form Q[

√

a

1

, . . . ,

√

a

r

] with each a

i

a positive element of

Q[

√

a

1

, . . . ,

√

a

i−1

]. In particular

α constructible =⇒ [Q[α] : Q] = 2

s

some s. (∗)

Now we can prove a partial converse to this last statement.

THEOREM 3.23. If α is contained in a subﬁeld of R that is Galois of degree 2

r

over Q,

then it is constructible.

PROOF. Suppose α ∈ E ⊂ R where E is Galois of degree 2

r

over Q, and let G =

Gal(E/Q). From a theorem on the structure of p-groups (GT 6.7), we know there will be

a sequence of groups

¦1¦ = G

0

⊂ G

1

⊂ G

2

⊂ ⊂ G

r

= G

with G

i

/G

i−1

of order 2. Correspondingly, there will be a sequence of ﬁelds,

E = E

0

⊃ E

1

⊃ E

2

⊃ ⊃ E

r

= Q

with E

i−1

of degree 2 over E

i

. The next lemma shows that E

i

= E

i−1

[

√

a

i

] for some

a

i

∈ E

i−1

, and a

i

> 0 because otherwise E

i

would not be real. This proves the theorem.

2

LEMMA 3.24. Let E/F be a quadratic extension of ﬁelds of characteristic = 2. Then

E = F[

√

d] for some d ∈ F.

PROOF. Let α ∈ E, α / ∈ F, and let X

2

+bX +c be the minimum polynomial of α. Then

α =

−b±

√

b

2

−4c

2

, and so E = F[

√

b

2

−4c].

2

COROLLARY 3.25. If p is a prime of the form 2

k

+ 1, then cos

2π

p

is constructible.

PROOF. The ﬁeld Q[e

2πi/p

] is Galois over Q with Galois group G · (Z/pZ)

, which has

order p −1 = 2

k

.

2

Thus a regular p-gon, p prime, is constructible if and only if p is a Fermat prime, i.e.,

of the form 2

2

r

+ 1. For example, we have proved that the regular 65537-polygon is con-

structible, without (happily) having to exhibit an explicit formula for cos

2π

65537

.

REMARK 3.26. The converse to (∗) is false. We’ll show below (4.9) that that the Galois

group of the splitting ﬁeld E over Q of the polynomial f(X) = X

4

−4X + 2 is S

4

. If the

four roots of f(X) were constructible, then all the elements of E would be constructible

(1.36a). Let H be a Sylow subgroup of S

4

. Then E

H

has odd degree over Q, and so the

elements of E

H

Q can’t be constructible.

34 3 THE FUNDAMENTAL THEOREM OF GALOIS THEORY

The Galois group of a polynomial

If the polynomial f ∈ F[X] is separable, then its splitting ﬁeld F

f

is Galois over F, and

we call Gal(F

f

/F) the Galois group G

f

of f.

Let f =

¸

n

i=1

(X−α

i

) in a splitting ﬁeld F

f

. We know that the elements of Gal(F

f

/F)

map roots of f to roots of f, i.e., they map the set ¦α

1

, α

2

, . . . , α

n

¦ into itself. Being

automorphisms, they deﬁne permutations of ¦α

1

, α

2

, . . . , α

n

¦, and as the α

i

generated F

f

,

an element of Gal(F

f

/F) is uniquely determined by the permutation it deﬁnes. Thus G

f

can be identiﬁed with a subset of Sym(¦α

1

, α

2

, . . . , α

n

¦) ≈ S

n

. In fact, G

f

consists

exactly of the permutations σ of ¦α

1

, α

2

, . . . , α

n

¦ such that, for P ∈ F[X

1

, . . . , X

n

],

P(α

1

, . . . , α

n

) = 0 =⇒ P(σα

1

, . . . , σα

n

) = 0.

This gives a description of G

f

without mentioning ﬁelds or abstract groups (neither of

which were available to Galois).

Note that this shows again that (G

f

: 1), hence [F

f

: F], divides deg(f)!.

Solvability of equations

For a polynomial f ∈ F[X], we say that f(X) = 0 is solvable in radicals if its solu-

tions can be obtained by the algebraic operations of addition, subtraction, multiplication,

division, and the extraction of m

th

roots, or, more precisely, if there exists a tower of ﬁelds

F = F

0

⊂ F

1

⊂ F

2

⊂ ⊂ F

m

such that

(a) F

i

= F

i−1

[α

i

], α

m

i

i

∈ F

i−1

;

(b) F

m

contains a splitting ﬁeld for f.

THEOREM 3.27 (GALOIS, 1832). Let F be a ﬁeld of characteristic zero. The equation

f = 0 is solvable in radicals if and only if the Galois group of f is solvable.

We shall prove this later (5.29). Also we shall exhibit polynomials f(X) ∈ Q[X] with

Galois group S

n

, which are therefore not solvable when n ≥ 5 by GT 4.29.

REMARK 3.28. If F has characteristic p, then the theorem fails for two reasons:

(a) f may not be separable, and so not have a Galois group;

(b) X

p

−X −a = 0 is not solvable by radicals even though it is separable with abelian

Galois group (cf. Exercise 2-2).

If the deﬁnition of solvable is changed to allow extensions of the type in (b) in the chain,

and f is required to be separable, then the theorem becomes true in characteristic p.

Exercises

3-1 (*). Let F be a ﬁeld of characteristic 0. Show that F(X

2

) ∩ F(X

2

−X) = F (inter-

section inside F(X)). [Hint: Find automorphisms σ and τ of F(X), each of order 2, ﬁxing

F(X

2

) and F(X

2

−X) respectively, and show that στ has inﬁnite order.]

3-2 (*).

13

Let p be an odd prime, and let ζ be a primitive p

th

root of 1 in C. Let E = Q[ζ],

and let G = Gal(E/Q); thus G = (Z/(p))

**. Let H be the subgroup of index 2 in G. Put
**

α =

¸

i∈H

ζ

i

and β =

¸

i∈G\H

ζ

i

. Show:

13

This problem shows that every quadratic extension of Q is contained in a cyclotomic extension of Q. The

Kronecker-Weber theorem says that every abelian extension of Q is contained in a cyclotomic extension.

Exercises 35

(a) α and β are ﬁxed by H;

(b) if σ ∈ G` H, then σα = β, σβ = α.

Thus α and β are roots of the polynomial X

2

+ X + αβ ∈ Q[X]. Compute αβ and show

that the ﬁxed ﬁeld of H is Q[

√

p] when p ≡ 1 mod 4 and Q[

√

−p] when p ≡ 3 mod 4.

3-3 (*). Let M = Q[

√

2,

√

3] and E = M[

(

√

2 + 2)(

√

3 + 3)] (subﬁelds of R).

(a) Show that M is Galois over Q with Galois group the 4-group C

2

C

2

.

(b) Show that E is Galois over Q with Galois group the quaternion group.

36 4 COMPUTING GALOIS GROUPS.

4 Computing Galois groups.

In this section, we investigate general methods for computing Galois groups.

When is G

f

⊂ A

n

?

Consider a polynomial

f(X) = X

n

+a

1

X

n−1

+ +a

n

and let f(X) =

¸

n

i=1

(X −α

i

) in some splitting ﬁeld. Set

∆(f) =

¸

1≤i<j≤n

(α

i

−α

j

), D(f) = ∆(f)

2

=

¸

1≤i<j≤n

(α

i

−α

j

)

2

.

The discriminant of f is deﬁned to be D(f). Note that D(f) is nonzero if and only if

f has only simple roots, i.e., if f is separable with no multiple factors. Let G

f

be the

Galois group of f, and identify it with a subgroup of Sym(¦α

1

, . . . , α

n

¦) (as on p34). The

choice of a numbering for the roots determines an isomorphism Sym(¦α

1

, . . . , α

n

¦) · S

n

,

and the subgroup of Sym(¦α

1

, . . . , α

n

¦) corresponding to any normal subgroup of S

n

is

independent of the choice for n = 6 (because all automorphisms of S

n

are inner when

n = 6, cf. GT 3.9).

PROPOSITION 4.1. Assume f is separable, and let σ ∈ G

f

.

(a) σ∆(f) = sign(σ)∆(f), where sign(σ) is the signature of σ.

(b) σD(f) = D(f).

PROOF. The ﬁrst equation follows immediately from the deﬁnition of the signature of σ

(see GT '4), and the second equation is obtained by squaring the ﬁrst.

2

COROLLARY 4.2. Let f(X) ∈ F[X] be of degree n and have only simple roots. Let F

f

be a splitting ﬁeld for f, so that G

f

= Gal(F

f

/F).

(a) The discriminant D(f) ∈ F.

(b) The subﬁeld of F

f

corresponding to A

n

∩ G

f

is F[∆(f)]. Hence

G

f

⊂ A

n

⇐⇒ ∆(f) ∈ F ⇐⇒ D(f) is a square in F.

PROOF. (a) The discriminant of f is an element of F

f

ﬁxed by G

f

=

df

Gal(F

f

/F), and

hence lies in F (by the fundamental theorem of Galois theory).

(b) Because f has simple roots, ∆(f) = 0, and so the formula σ∆(f) = sign(σ)∆(f)

shows that an element of G

f

ﬁxes ∆(f) if and only if it lies in A

n

. Thus, under the Galois

correspondence,

G

f

∩ A

n

↔F[∆(f)].

Hence,

G

f

∩ A

n

= G

f

⇐⇒ F[∆(f)] = F.

2

The discriminant of f can be expressed as a universal polynomial in the coefﬁcients of

f. For example:

D(aX

2

+bX +c) = (b

2

−4ac)/a

2

D(X

3

+bX +c) = −4b

3

−27c

2

.

When is G

f

transitive? 37

By completing the cube, one can put any cubic polynomial in this form (in characteristic

= 3).

The formulas for the discriminant rapidly become very complicated, for example, that

for X

5

+aX

4

+bX

3

+cX

2

+dX +e has 59 terms. Fortunately, Maple knows them: the

syntax is “discrim(f,X);” where f is a polynomial in the symbol X.

REMARK 4.3. Suppose F ⊂ R. Then D(f) will not be a square if it is negative. It is

known that the sign of D(f) is (−1)

s

where 2s is the number of nonreal roots of f in C

(see ANT 2.39). Thus if s is odd, then G

f

is not contained in A

n

. This can be proved more

directly by noting that complex conjugation acts on the roots as the product of s disjoint

transpositions.

Of course the converse is not true: when s is even, G

f

is not necessarily contained in

A

n

.

When is G

f

transitive?

PROPOSITION 4.4. Let f(X) ∈ F[X] have only simple roots. Then f(X) is irreducible if

and only if G

f

permutes the roots of f transitively.

PROOF. =⇒: If α and β are two roots of f(X) in a splitting ﬁeld F

f

for f, then they both

have f(X) as their minimum polynomial, and so F[α] and F[β] are both stem ﬁelds for f.

Hence, there is an F-isomorphism

F[α] · F[β], α ↔β.

Write F

f

= F[α

1

, α

2

, ...] with α

1

= α and α

2

, α

3

, . . . the other roots of f(X). Then the

F-homomorphism α → β: F[α] → F

f

extends (step by step) to an F-homomorphism

F

f

→F

f

(use 2.2b), which is an F-isomorphism sending α to β.

⇐= : Let g(X) ∈ F[X] be an irreducible factor of f, and let α be one of its roots. If

β is a second root of f, then (by assumption) β = σα for some σ ∈ G

f

. Now, because g

has coefﬁcients in F,

g(σα) = σg(α) = 0,

and so β is also a root of g. Therefore, every root of f is also a root of g, and so f(X) =

g(X).

2

Note that when f(X) is irreducible of degree n, n[(G

f

: 1) because [F[α] : F] = n

and [F[α] : F] divides [F

f

: F] = (G

f

: 1). Thus G

f

is a transitive subgroup of S

n

whose

order is divisible by n.

Polynomials of degree at most three

EXAMPLE 4.5. Let f(X) ∈ F[X] be a polynomial of degree 2. Then f is inseparable

⇐⇒ F has characteristic 2 and f(X) = X

2

−a for some a ∈ F F

2

. If f is separable,

then G

f

= 1(= A

2

) or S

2

according as D(f) is a square in F or not.

EXAMPLE 4.6. Let f(X) ∈ F[X] be a polynomial of degree 3. We can assume f to be

irreducible, for otherwise we are essentially back in the previous case. Then f is inseparable

if and only if F has characteristic 3 and f(X) = X

3

− a for some a ∈ F ` F

3

. If f is

separable, then G

f

is a transitive subgroup of S

3

whose order is divisible by 3. There are

38 4 COMPUTING GALOIS GROUPS.

only two possibilities: G

f

= A

3

or S

3

according as D(f) is a square in F or not. Note that

A

3

is generated by the cycle (123).

For example, X

3

− 3X + 1 ∈ Q[X] is irreducible (see 1.12), its discriminant is

−4(−3)

3

−27 = 81 = 9

2

, and so its Galois group is A

3

.

On the other hand, X

3

+ 3X + 1 ∈ Q[X] is also irreducible (apply 1.11), but its

discriminant is −135 which is not a square in Q, and so its Galois group is S

3

.

Quartic polynomials

Let f(X) be a quartic polynomial without multiple roots. In order to determine G

f

we shall

exploit the fact that S

4

has

V = ¦1, (12)(34), (13)(24), (14)(23)¦

as a normal subgroup — it is normal because it contains all elements of type 2 + 2 (GT

4.28). Let E be a splitting ﬁeld of f, and let f(X) =

¸

(X − α

i

) in E. We identify

the Galois group G

f

of f with a subgroup of the symmetric group Sym(¦α

1

, α

2

, α

3

, α

4

¦).

Consider the partially symmetric elements

α = α

1

α

2

+α

3

α

4

β = α

1

α

3

+α

2

α

4

γ = α

1

α

4

+α

2

α

3

.

They are distinct because the α

i

are distinct; for example,

α −β = α

1

(α

2

−α

3

) +α

4

(α

3

−α

2

) = (α

1

−α

4

)(α

2

−α

3

).

The group Sym(¦α

1

, α

2

, α

3

, α

4

¦) permutes ¦α, β, γ¦ transitively. The stabilizer of each of

α, β, γ must therefore be a subgroup of index 3 in S

4

, and hence has order 8. For example,

the stabilizer of β is '(1234), (13)`. Groups of order 8 in S

4

are Sylow 2-subgroups. There

are three of them, all isomorphic to D

4

. By the Sylow theorems, V is contained in a Sylow

2-subgroup; in fact, because the Sylow 2-subgroups are conjugate and V is normal, it is

contained in all three. It follows that V is the intersection of the three Sylow 2-subgroups.

Each Sylow 2-subgroup ﬁxes exactly one of α, β, or γ, and therefore their intersection V is

the subgroup of Sym(¦α

1

, α

2

, α

3

, α

4

¦) ﬁxing α, β, and γ.

LEMMA 4.7. The ﬁxed ﬁeld of G

f

∩ V is F[α, β, γ]. Hence

F[α, β, γ] is Galois over F with Galois group G

f

/G

f

∩ V .

PROOF. The above discussion shows that the subgroup of G

f

of

elements ﬁxing F[α, β, γ] is G

f

∩V , and so E

G

f

∩V

= F[α, β, γ]

by the fundamental theorem of Galois theory. The remaining state-

ments follow from the fundamental theorem using that V is nor-

mal.

2

E

F[α, β, γ]

G

f

∩V

F

G

f

/G

f

∩V

Let M = F[α, β, γ], and let g(X) = (X −α)(X −β)(X −γ) ∈ M[X] — it is called

the resolvent cubic of f. Any permutation of the α

i

(a fortiori, any element of G

f

) merely

permutes α, β, γ, and so ﬁxes g(X). Therefore (by the fundamental theorem) g(X) has

coefﬁcients in F. More explicitly, we have:

Quartic polynomials 39

LEMMA 4.8. The resolvent cubic of f = X

4

+bX

3

+cX

2

+dX +e is

g = X

3

−cX

2

+ (bd −4e)X −b

2

e + 4ce −d

2

.

The discriminants of f and g are equal.

PROOF (SKETCH). Expand f = (X−α

1

)(X−α

2

)(X−α

3

)(X−α

4

) to express b, c, d, e

in terms of α

1

, α

2

, α

3

, α

4

. Expand g = (X−α)(X−β)(X−γ) to express the coefﬁcients

of g in terms of α

1

, α

2

, α

3

, α

4

, and substitute to express them in terms of b, c, d, e.

2

Now let f be an irreducible separable quartic. Then G = G

f

is a transitive subgroup of

S

4

whose order is divisible by 4. There are the following possibilities for G:

G (G∩ V : 1) (G : V ∩ G)

S

4

4 6

A

4

4 3

V 4 1

D

4

4 2

C

4

2 2

(G∩ V : 1) = [E : M]

(G : V ∩ G) = [M : F]

The groups of type D

4

are the Sylow 2-subgroups discussed above, and the groups of type

C

4

are those generated by cycles of length 4.

We can compute (G : V ∩ G) from the resolvent cubic g, because G/V ∩ G =

Gal(M/F) and M is the splitting ﬁeld of g. Once we know (G : V ∩ G), we can de-

duce G except in the case that it is 2. If [M : F] = 2, then G∩V = V or C

2

. Only the ﬁrst

group acts transitively on the roots of f, and so (from 4.4) we see that in this case G = D

4

or C

4

according as f is irreducible or not in M[X].

EXAMPLE 4.9. Consider f(X) = X

4

−4X + 2 ∈ Q[X]. It is irreducible by Eisenstein’s

criterion (1.16), and its resolvent cubic is g(X) = X

3

− 8X + 16, which is irreducible

because it has no roots in F

5

. The discriminant of g(X) is −4864, which is not a square,

and so the Galois group of g(X) is S

3

. From the table, we see that the Galois group of

f(X) is S

4

.

EXAMPLE 4.10. Consider f(X) = X

4

+4X

2

+2 ∈ Q[X]. It is irreducible by Eisenstein’s

criterion (1.16), and its resolvent cubic is (X − 4)(X

2

− 8); thus M = Q[

√

2]. From the

table we see that G

f

is of type D

4

or C

4

, but f factors over M (even as a polynomial in

X

2

), and hence G

f

is of type C

4

.

EXAMPLE 4.11. Consider f(X) = X

4

− 10X

2

+ 4 ∈ Q[X]. It is irreducible in Q[X]

because (by inspection) it is irreducible in Z[X]. Its resolvent cubic is (X + 10)(X +

4)(X −4), and so G

f

is of type V .

EXAMPLE 4.12. Consider f(X) = X

4

− 2 ∈ Q[X]. It is irreducible by Eisenstein’s

criterion (1.16), and its resolvent cubic is g(X) = X

3

+ 8X. Hence M = Q[i

√

2]. One

can check that f is irreducible over M, and G

f

is of type D

4

.

Alternatively, analyse the equation as in (3.22).

40 4 COMPUTING GALOIS GROUPS.

As we explained in (1.29), Maple knows how to factor polynomials with coefﬁcients in

Q[α].

EXAMPLE 4.13. (From the web, sci.math.research, search for “ﬁnal analysis”.) Consider

f(X) = X

4

− 2cX

3

− dX

2

+ 2cdX − dc

2

∈ Z[X] with a > 0, b > 0, c > 0, a > b

and d = a

2

− b

2

. Let r = d/c

2

and let w be the unique positive real number such that

r = w

3

/(w

2

+4). Let mbe the number of roots of f(X) in Z (counted with multiplicities).

The Galois group of f is as follows:

— If m = 0 and w not rational, then G is S

4

.

— If m = 1 and w not rational then G is S

3

.

— If w is rational and w

2

+ 4 is not a square then G = D4.

— If w is rational and w

2

+ 4 is a square then G = V = C

2

C

2

.

This covers all possible cases. The hard part was to establish that m = 2 could never

happen.

Examples of polynomials with S

p

as Galois group over Q

The next lemma gives a criterion for a subgroup of S

p

to be the whole of S

p

.

LEMMA 4.14. For p prime, the symmetric group S

p

is generated by any transposition and

any p-cycle.

PROOF. After renumbering, we may assume that the transposition is τ = (12), and we may

write the p-cycle σ so that 1 occurs in the ﬁrst position, σ = (1 i

2

i

p

). Now some power

of σ will map 1 to 2 and will still be a p-cycle (here is where we use that p is prime). After

replacing σ with the power, we have σ = (1 2 j

3

. . . j

p

), and after renumbering again, we

have σ = (1 2 3 . . . p). Now

(i i + 1) = σ

i

(12)σ

−i

(see GT 4.25) and so lies in the subgroup generated by σ and τ. These transpositions

generate S

p

.

2

PROPOSITION 4.15. Let f be an irreducible polynomial of prime degree p in Q[X]. If f

splits in C and has exactly two nonreal roots, then G

f

= S

p

.

PROOF. Let E be the splitting ﬁeld of f in C, and let α ∈ E be a root of f. Because f is

irreducible, [Q[α] : Q] = deg f = p, and so p[[E : Q] = (G

f

: 1). Therefore G

f

contains

an element of order p (Cauchy’s theorem, GT 4.13), but the only elements of order p in S

p

are p-cycles (here we use that p is prime again).

Let σ be complex conjugation on C. Then σ transposes the two nonreal roots of f(X)

and ﬁxes the rest. Therefore G

f

⊂ S

p

and contains a transposition and a p-cycle, and so is

the whole of S

p

.

2

It remains to construct polynomials satisfying the conditions of the Proposition.

EXAMPLE 4.16. Let p≥ 5 be a prime number. Choose a positive even integer m and even

integers

n

1

< n

2

< < n

p−2

,

and let

g(X) = (X

2

+m)(X −n

1

)...(X −n

p−2

).

Finite ﬁelds 41

The graph of g crosses the x-axis exactly at the points n

1

, . . . , n

p−2

, and it doesn’t have a

local maximum or minimum at any of those points (because the n

i

are simple roots). Thus

e = min

g

(x)=0

[g(x)[ > 0, and we can choose an odd positive integer n such that

2

n

< e.

Consider

f(X) = g(X) −

2

n

.

As

2

n

< e, the graph of f also crosses the x-axis at exactly p−2 points, and so f has exactly

two nonreal roots. On the other hand, when we write

nf(X) = nX

p

+a

1

X

p−1

+ +a

p

,

the a

i

are all even and a

p

is not divisible by 2

2

, and so Eisenstein’s criterion implies that f

is irreducible. Over R, f has p−2 linear factors and one quadratic factor, and so it certainly

splits over C (high school algebra). Therefore, the proposition applies to f.

14

Finite ﬁelds

Let F

p

= Z/pZ, the ﬁeld of p elements. As we noted in '1, any other ﬁeld E of character-

istic p contains a copy of F

p

, namely, ¦m1

E

[ m ∈ Z¦. No harm results if we identify F

p

with this subﬁeld of E.

Let E be a ﬁeld of degree n over F

p

. Then E has q = p

n

elements, and so E

is a

group of order q −1. Hence the nonzero elements of E are roots X

q−1

−1, and all elements

of E (including 0) are roots of X

q

− X. Hence E is a splitting ﬁeld for X

q

− X, and so

any two ﬁelds with q elements are isomorphic.

PROPOSITION 4.17. Every extension of ﬁnite ﬁelds is simple.

PROOF. Consider E ⊃ F. Then E

**is a ﬁnite subgroup of the multiplicative group of a
**

ﬁeld, and hence is cyclic (see Exercise 1-3). If ζ generates E

as a multiplicative group,

then certainly E = F[ζ].

2

Now let E be the splitting ﬁeld of f(X) = X

q

− X, q = p

n

. The derivative f

t

(X) =

−1, which is relatively prime to f(X) (in fact, to every polynomial), and so f(X) has

q distinct roots in E. Let S be the set of its roots. Then S is obviously closed under

multiplication and the formation of inverses, but it is also closed under subtraction: if a

q

=

a and b

q

= b, then

(a −b)

q

= a

q

−b

q

= a −b.

Hence S is a ﬁeld, and so S = E. In particular, E has p

n

elements.

PROPOSITION 4.18. For each power q = p

n

there is a ﬁeld F

q

with q elements. It is

the splitting ﬁeld of X

q

− X, and hence any two such ﬁelds are isomorphic. Moreover,

F

q

is Galois over F

p

with cyclic Galois group generated by the Frobenius automorphism

σ(a) = a

p

.

14

If mis taken sufﬁciently large, then g(X) −2 will have exactly two nonreal roots, i.e., we can take n = 1,

but the proof is longer (see Jacobson 1964, p107, who credits the example to Brauer). The shorter argument in

the text was suggested to me by Martin Ward.

42 4 COMPUTING GALOIS GROUPS.

PROOF. Only the ﬁnal statement remains to be proved. The ﬁeld F

q

is Galois over F

p

because it is the splitting ﬁeld of a separable polynomial. We noted in (1.4) that x

σ

→ x

p

is an automorphism of F

q

. An element a of F

q

is ﬁxed by σ if and only if a

p

= a, but F

p

consists exactly of such elements, and so the ﬁxed ﬁeld of 'σ` is F

p

. This proves that F

q

is

Galois over F

p

and that 'σ` = Gal(F

q

/F

p

) (see 3.11b).

2

COROLLARY 4.19. Let E be a ﬁeld with p

n

elements. For each divisor m of n, m ≥ 0, E

contains exactly one ﬁeld with p

m

elements.

PROOF. We know that E is Galois over F

p

and that Gal(E/F

p

) is the cyclic group of order

n generated by σ. The group 'σ` has one subgroup of order n/m for each m dividing n,

namely, 'σ

m

`, and so E has exactly one subﬁeld of degree m over F

p

for each m dividing

n, namely, E

/σ

m

)

. Because it has degree m over F

p

, E

/σ

m

)

has p

m

elements.

2

COROLLARY 4.20. Each monic irreducible polynomial f of degree d[n in F

p

[X] occurs

exactly once as a factor of X

p

n

−X; hence, the degree of the splitting ﬁeld of f is ≤ d.

PROOF. First, the factors of X

p

n

− X are distinct because it has no common factor with

its derivative. If f(X) is irreducible of degree d, then f(X) has a root in a ﬁeld of degree

d over F

p

. But the splitting ﬁeld of X

p

n

− X contains a copy of every ﬁeld of degree

d over F

p

with d[n. Hence some root of X

p

n

− X is also a root of f(X), and therefore

f(X)[X

p

n

− X. In particular, f divides X

p

d

− X, and therefore it splits in its splitting

ﬁeld, which has degree d over F

p

.

2

PROPOSITION 4.21. Let F be an algebraic closure of F

p

. Then F contains exactly one ﬁeld

F

p

n for each integer n ≥ 1, and F

p

n consists of the roots of X

p

n

−X. Moreover,

F

p

m ⊂ F

p

n ⇐⇒ m[n.

The partially ordered set of ﬁnite subﬁelds of F is isomorphic to the set of integers n ≥ 1

partially ordered by divisibility.

PROOF. Obvious from what we have proved.

2

PROPOSITION 4.22. The ﬁeld F

p

has an algebraic closure F.

PROOF. Choose a sequence of integers 1 = n

1

< n

2

< n

3

< such that n

i

[n

i+1

for

all i, and every integer n divides some n

i

. For example, let n

i

= i!. Deﬁne the ﬁelds F

p

n

i

inductively as follows: F

p

n

1 = F

p

; F

p

n

i is the splitting ﬁeld of X

p

n

i

− X over F

p

n

i−1 .

Then, F

p

n

1 ⊂ F

p

n

2 ⊂ F

p

n

3 ⊂ , and we deﬁne F =

¸

F

p

n

i . As a union of a chain of

ﬁelds algebraic over F

p

, it is again a ﬁeld algebraic over F

p

. Moreover, every polynomial

in F

p

[X] splits in F, and so it is an algebraic closure of F (by 1.44).

2

REMARK 4.23. Since the F

p

n’s are not subsets of a ﬁxed set, forming the union requires

explanation. Deﬁne S to be the disjoint union of the F

p

n. For a, b ∈ S, set a ∼ b if a = b

in one of the F

p

n. Then ∼ is an equivalence relation, and we let F = S/ ∼.

Maple factors polynomials modulo p very quickly. The syntax is “Factor(f(X))

mod p;”. Thus, for example, to obtain a list of all monic polynomials of degree 1, 2, or 4

over F

5

, ask Maple to factor X

625

−X.

Computing Galois groups over Q 43

Finite ﬁelds were often called

15

Galois ﬁelds, and F

q

was denoted GF(q) (it still is in

Maple). Maple contains a “Galois ﬁeld package” to do computations in ﬁnite ﬁelds. For

example, it can ﬁnd a primitive element for F

q

(i.e., a generator for F

q

). To start it, type:

readlib(GF):.

Computing Galois groups over Q

In the remainder of this section, I sketch a practical method for computing Galois groups

over Q and similar ﬁelds. Recall that for a monic separable polynomial f ∈ F[X], F

f

denotes a splitting ﬁeld for F, and G

f

= Gal(F

f

/F) denotes the Galois group of F.

Moreover, G

f

permutes the roots α

1

, α

2

, . . . of f in F

f

:

G ⊂ Sym¦α

1

, α

2

, . . .¦.

The ﬁrst result generalizes Proposition 4.4.

PROPOSITION 4.24. Let f(X) be a monic polynomial in F[X] with only simple roots,

and suppose that the orbits of G

f

acting on the roots of f have m

1

, . . . , m

r

elements re-

spectively. Then f factors as f = f

1

f

r

with f

i

irreducible of degree m

i

.

PROOF. Let α

1

, . . . , α

m

, m = deg f, be the roots of f(X) in F

f

. The monic factors of

f(X) in F

f

[X] correspond to subsets S of ¦α

1

, . . . , α

m

¦,

S ↔f

S

=

¸

α∈S

(X −α),

and f

S

is ﬁxed under the action of G

f

(and hence has coefﬁcients in F) if and only if S

is stable under G

f

. Therefore the irreducible factors of f in F[X] are the polynomials fS

corresponding to minimal subsets S of ¦α

1

, . . . , α

m

¦ stable under G

f

, but these subsets S

are precisely the orbits of G

f

in ¦α

1

, . . . , α

m

¦.

2

REMARK 4.25. Note that the proof shows the following: let ¦α

1

, . . . , α

m

¦ =

¸

O

i

be the

decomposition of ¦α

1

, . . . , α

m

¦ into a disjoint union of orbits for the group G

f

; then

f =

¸

f

i

, f

i

=

¸

α

i

∈O

i

(X −α

i

)

is the decomposition of f into a product of irreducible polynomials in F[X].

Now suppose F is ﬁnite, with p

n

elements say. Then G

f

is a cyclic group generated by

the Frobenius automorphism σ: x → x

p

. When we regard σ as a permutation of the roots

of f, then distinct orbits of σ correspond to the factors in its cycle decomposition (GT 4.22).

Hence, if the degrees of the distinct irreducible factors of f are m

1

, m

2

, . . . , m

r

, then σ has

a cycle decomposition of type

m

1

+ +m

r

= deg f.

15

From a letter to the Notices of the AMS, February 2003 (P´ alfy): “full credit should be given to [Galois] for

constructing ﬁnite ﬁelds in general. In one of the few papers published during his short lifetime, entitled “Sur

la th´ eorie des nombres”, which appeared in the Bulletin des Sciences Math´ ematiques in June 1830, Galois —

at that time not even nineteen years old — deﬁned ﬁnite ﬁelds of arbitrary prime power order and established

their basic properties, e.g. the existence of a primitive element. So it is fully justiﬁed when ﬁnite ﬁelds are

called Galois ﬁelds and customarily denoted by GF(q).” True, but I prefer to use descriptive names wherever

possible.

44 4 COMPUTING GALOIS GROUPS.

LEMMA 4.26. Let R be a unique factorization domain with ﬁeld of fractions F, and let f

be a monic polynomial in R[X]. Let P be a prime ideal in R, and let f be the image of f in

(R/P)[X]. Assume that neither f nor f has a multiple root. Then the roots α

1

, . . . , α

m

of

f lie in some ﬁnite extension R

t

of R, and their reductions α

i

modulo PR

t

are the roots of

f. Moreover G

f

⊂ G

f

when both are identiﬁed with subgroups of Sym¦α

1

, . . . , α

m

¦ =

Sym¦α

1

, . . . , α

m

¦.

PROOF. Omitted — see van der Waerden, Modern Algebra, I, '61 (second edition) or ANT

3.43.

2

On combining these results, we obtain the following theorem.

THEOREM 4.27 (DEDEKIND). Let f(X) ∈ Z[X] be a monic polynomial of degree m, and

let p be a prime such that f mod p has simple roots (equivalently, D(f) is not divisible by

p). Suppose that f =

¸

f

i

with f

i

irreducible of degree m

i

in F

p

[X]. Then G

f

contains an

element whose cycle decomposition is of type

m = m

1

+ +m

r

.

EXAMPLE 4.28. Consider X

5

− X − 1. Modulo 2, this factors as (X

2

+ X + 1)(X

3

+

X

2

+ 1), and modulo 3 it is irreducible. Hence G

f

contains (ik)(lmn) and (12345), and

so also ((ik)(lmn))

3

= (ik). Therefore G

f

= S

5

by (4.14).

LEMMA 4.29. A transitive subgroup of H ⊂ S

n

containing a transposition and an (n−1)-

cycle is equal to S

n

.

PROOF. After possibly renumbering, we may suppose the (n − 1)-cycle is (123 . . . n −

1). Because of the transitivity, the transposition can be transformed into (in), some 1 ≤

i ≤ n − 1. Conjugating (in) by (123 . . . n − 1) and its powers will transform it into

(1n), (2n), . . . , (n −1 n), and these elements obviously generate S

n

.

2

EXAMPLE 4.30. Select monic polynomials of degree n, f

1

, f

2

, f

3

with coefﬁcients in Z

such that:

(a) f

1

is irreducible modulo 2;

(b) f

2

= (degree 1)(irreducible of degree n −1) mod 3;

(c) f

3

= (irreducible of degree 2)(product of 1 or 2 irreducible polys of odd degree) mod

5.

We also choose f

1

, f

2

, f

3

to have only simple roots. Take

f = −15f

1

+ 10f

2

+ 6f

3

.

Then

(i) G

f

is transitive (it contains an n-cycle because f ≡ f

1

mod 2);

(ii) G

f

contains a cycle of length n −1 (because f ≡ f

2

mod 3);

(iii) G

f

contains a transposition (because f ≡ f

3

mod 5, and so it contains the product of a

transposition with a commuting element of odd order; on raising this to an appropriate

odd power, we are left with the transposition). Hence G

f

is S

n

.

Exercises 45

The above results give the following strategy for computing the Galois group of an

irreducible polynomial f ∈ Q[X]. Factor f modulo a sequence of primes p not dividing

D(f) to determine the cycle types of the elements in G

f

— a difﬁcult theorem in number

theory, the effective Chebotarev density theorem, says that if a cycle type occurs in G

f

,

then this will be seen by looking modulo a set of prime numbers of positive density, and

will occur for a prime less than some bound. Now look up a table of transitive subgroups

of S

n

with order divisible by n and their cycle types. If this doesn’t sufﬁce to determine the

group, then look at its action on the set of subsets of r roots for some r.

See, Butler and McKay, The transitive groups of degree up to eleven, Comm. Alge-

bra 11 (1983), 863–911. This lists all transitive subgroups of S

n

, n ≤ 11, and gives the

cycle types of their elements and the orbit lengths of the subgroup acting on the r-sets of

roots. With few exceptions, these invariants are sufﬁcient to determine the subgroup up to

isomorphism.

Maple V can compute Galois groups for polynomials of degree ≤ 7 over Q. To learn

the syntax, type “?galois;”.

See also, Soicher and McKay, Computing Galois groups over the rationals, J. Number

Theory, 20 (1985) 273–281.

Exercises

4-1 (*). Find the splitting ﬁeld of X

m

−1 ∈ F

p

[X].

4-2 (*). Find the Galois group of X

4

−2X

3

−8X −3 over Q.

4-3 (*). Find the degree of the splitting ﬁeld of X

8

−2 over Q.

4-4 (*). Give an example of a ﬁeld extension E/F of degree 4 such that there does not

exist a ﬁeld M with F ⊂ M ⊂ E, [M : F] = 2.

4-5. List all irreducible polynomials of degree 3 over F

7

in 10 seconds or less (there are

112).

4-6. “It is a thought-provoking question that few graduate students would know how to

approach the question of determining the Galois group of, say,

X

6

+ 2X

5

+ 3X

4

+ 4X

3

+ 5X

2

+ 6X + 7.”

[over Q].

(a) Can you ﬁnd it?

(b) Can you ﬁnd it without using the “galois” command in Maple?

4-7 (*). Let f(X) = X

5

+aX +b, a, b ∈ Q. Show that G

f

≈ D

5

(dihedral group) if and

only if

(a) f(X) is irreducible in Q[X], and

(b) the discriminant D(f) = 4

4

a

5

+ 5

5

b

4

of f(X) is a square, and

(c) the equation f(X) = 0 is solvable by radicals.

4-8. Show that a polynomial f of degree n =

¸

k

i=1

p

r

i

i

is irreducible over F

q

if and only

if gcd(f(x), x

q

n/p

i

−x) = 1 for all i.

4-9. Let f(X) be an irreducible polynomial in Q[X] with both real and nonreal roots.

Showthat its Galois group is nonabelian. Can the condition that f is irreducible be dropped?

46 5 APPLICATIONS OF GALOIS THEORY

5 Applications of Galois theory

In this section, we apply the fundamental theorem of Galois theory to obtain other results

about polynomials and extensions of ﬁelds.

Primitive element theorem.

Recall that a ﬁnite extension of ﬁelds E/F is simple if E = F[α] for some element α of

E. Such an α is called a primitive element of E. We shall show that (at least) all separable

extensions have primitive elements.

Consider for example Q[

√

2,

√

3]/Q. We know (see Exercise 3-3) that its Galois group

over Q is a 4-group 'σ, τ`, where

σ

√

2 = −

√

2

σ

√

3 =

√

3

,

τ

√

2 =

√

2

τ

√

3 = −

√

3

.

Note that

σ(

√

2 +

√

3) = −

√

2 +

√

3,

τ(

√

2 +

√

3) =

√

2 −

√

3,

(στ)(

√

2 +

√

3) = −

√

2 −

√

3.

These all differ from

√

2 +

√

3, and so only the identity element of Gal(Q[

√

2,

√

3]/Q)

ﬁxes the elements of Q[

√

2 +

√

3]. According to the fundamental theorem, this implies that

√

2 +

√

3 is a primitive element:

Q[

√

2,

√

3] = Q[

√

2 +

√

3].

It is clear that this argument should work much more generally.

Recall that an element α algebraic over a ﬁeld F is separable over F if its minimum

polynomial over F has no multiple roots.

THEOREM 5.1. Let E = F[α

1

, ..., α

r

] be a ﬁnite extension of F, and assume that α

2

, ..., α

r

are separable over F (but not necessarily α

1

). Then there is an element γ ∈ E such that

E = F[γ].

PROOF. For ﬁnite ﬁelds, we proved this in (4.17). Hence we may assume F to be inﬁnite.

It sufﬁces to prove the statement for r = 2, for then

F[α

1

, α

2

, . . . , α

r

] = F[α

t

1

, α

3

, . . . , α

r

] = F[α

tt

1

, α

4

, . . . , α

r

] = .

Thus let E = F[α, β] with β separable over F. Let f and g be the minimum polynomials

of α and β over F. Let α

1

= α, . . . , α

s

be the roots of f in some big ﬁeld containing E,

and let β

1

= β, β

2

, . . . , β

t

be the roots of g. For j = 1, β

j

= β

1

, and so the the equation

α

i

+Xβ

j

= α

1

+Xβ

1

,

has exactly one solution, namely, X =

α

i

−α

1

β

1

−β

j

. If we choose a c ∈ F different from any of

these solutions (using that F is inﬁnite), then

α

i

+cβ

j

= α +cβ unless i = 1 = j.

Primitive element theorem. 47

Let γ = α + cβ. Then the polynomials g(X) and f(γ − cX) have coefﬁcients in

F[γ][X], and have β as a root:

g(β) = 0, f(γ −cβ) = f(α) = 0.

In fact, β is their only common root, because we chose c so that γ − cβ

j

= α

i

unless

i = 1 = j. Therefore

gcd(g(X), f(γ −cX)) = X −β.

Here we have computed the gcd in some ﬁeld splitting fg, but we have seen (Proposition

2.10) that the gcd of two polynomials has coefﬁcients in the same ﬁeld as the coefﬁcients

of the polynomials. Hence β ∈ F[γ], and this implies that α = γ − cβ also lies in F[γ].

We have shown that F[α, β] = F[γ].

2

REMARK 5.2. When F is inﬁnite, the proof shows that γ can be chosen to be of the form

γ = α

1

+c

2

α

2

+ +c

r

α

r

, c

i

∈ F.

If E is Galois over F, then an element of this form will be a primitive element provided it is

moved by every element of Gal(E/F) except 1. These remarks make it very easy to write

down primitive elements.

Our hypotheses are minimal: if two of the α’s are not separable, then the extension need

not be simple. Before giving an example to illustrate this, we need another result.

PROPOSITION 5.3. Let E = F[γ] be a simple algebraic extension of F. Then there are

only ﬁnitely many intermediate ﬁelds M,

F ⊂ M ⊂ E.

PROOF. Let M be such a ﬁeld, and let g(X) be the minimum polynomial of γ over M. Let

M

t

be the subﬁeld of E generated over F by the coefﬁcients of g(X). Clearly M

t

⊂ M,

but (equally clearly) g(X) is the minimum polynomial of γ over M

t

. Hence

[E : M

t

] = deg g = [E : M],

and so M = M

t

— M is generated by the coefﬁcients of g(X).

Let f(X) be the minimum polynomial of γ over F. Then g(X) divides f(X) in M[X],

and hence also in E[X]. Therefore, there are only ﬁnitely many possible g’s, and conse-

quently only ﬁnitely many possible M’s.

2

REMARK 5.4. (a) Note that the proof in fact gives a description of all the intermediate

ﬁelds: each is generated over F by the coefﬁcients of a factor g(X) of f(X) in E[X]. The

coefﬁcients of such a g(X) are partially symmetric polynomials in the roots of f(X) (that

is, ﬁxed by some, but not necessarily all, of the permutations of the roots).

(b) The proposition has a converse: if E is a ﬁnite extension of F and there are only

ﬁnitely many intermediate ﬁelds M, F ⊂ M ⊂ E, then E is a simple extension of F

(see Dummit and Foote 1991, p508). This gives another proof of Theorem 5.1 in the case

that E is separable over F, because Galois theory shows that there are only ﬁnitely many

intermediate ﬁelds in this case (the Galois closure of E over F has only ﬁnitely many

intermediate ﬁelds).

48 5 APPLICATIONS OF GALOIS THEORY

EXAMPLE 5.5. The simplest nonsimple algebraic extension is k(X, Y ) ⊃ k(X

p

, Y

p

),

where k is an algebraically closed ﬁeld of characteristic p. Let F = k(X

p

, Y

p

). For

any c ∈ k, we have

k(X, Y ) = F[X, Y ] ⊃ F[X +cY ] ⊃ F

with the degree of each extension equal to p. If

F[X +cY ] = F[X +c

t

Y ], c = c

t

,

then F[X + cY ] would contain both X and Y , which is impossible because [k(X, Y ) :

F] = p

2

. Hence there are inﬁnitely many distinct intermediate ﬁelds.

16

Fundamental Theorem of Algebra

We ﬁnally prove the misnamed

17

fundamental theorem of algebra.

THEOREM 5.6. The ﬁeld C of complex numbers is algebraically closed.

PROOF. Deﬁne C to be the splitting ﬁeld of X

2

+ 1 ∈ R[X], and let i be a root of X

2

+ 1

in C; thus C = R[i]. We have to show (see 1.44) that every f(X) ∈ R[X] has a root in C.

The two facts we need to assume about R are:

— Positive real numbers have square roots.

— Every polynomial of odd degree with real coefﬁcients has a real root.

Both are immediate consequences of the Intermediate Value Theorem, which says that a

continuous function on a closed interval takes every value between its maximum and mini-

mum values (inclusive). (Intuitively, this says that, unlike the rationals, the real line has no

“holes”.)

We ﬁrst show that every element of Chas a square root. Write α = a+bi, with a, b ∈ R,

and choose c, d to be real numbers such that

c

2

=

(a +

√

a

2

+b

2

)

2

, d

2

=

(−a +

√

a

2

+b

2

)

2

.

Then c

2

− d

2

= a and (2cd)

2

= b

2

. If we choose the signs of c and d so that cd has the

same sign as b, then (c +di)

2

= α and so c +di is a square root of α.

Let f(X) ∈ R[X], and let E be a splitting ﬁeld for f(X)(X

2

+ 1) — we have to show

that E = C. Since Rhas characteristic zero, the polynomial is separable, and so E is Galois

over R. Let G be its Galois group, and let H be a Sylow 2-subgroup of G.

Let M = E

H

. Then M is of odd degree over R, and M = R[α] some α (Theorem 5.1).

The minimum polynomial of α over R has odd degree and so has a root in R. It therefore

has degree 1, and so M = R and G = H.

16

Zariski showed that there is even an intermediate ﬁeld M that is not isomorphic to F(X, Y ), and Piotr

Blass showed in his thesis (University of Michigan 1977), using the methods of algebraic geometry, that there

is an inﬁnite sequence of intermediate ﬁelds, no two of which are isomorphic.

17

Because it is not strictly a theorem in algebra: it is a statement about R whose construction is part of

analysis (or maybe topology). In fact, I prefer the proof based on Liouville’s theorem in complex analysis to

the more algebraic proof given in the text: if f(z) is a polynomial without a root in C, then f(z)

−1

will be

bounded and holomorphic on the whole complex plane, and hence (by Liouville) constant. The Fundamental

Theorem was quite a difﬁcult theorem to prove. Gauss gave a proof in his doctoral dissertation in 1798 in

which he used some geometric arguments which he didn’t justify. He gave the ﬁrst rigorous proof in 1816. The

elegant argument given here is a simpliﬁcation by Emil Artin of earlier proofs.

Cyclotomic extensions 49

We now know that Gal(E/C) is a 2-group. If it is = 1, then it has a subgroup N of

index 2 (GT 4.15). The ﬁeld E

N

has degree 2 over C, and can therefore be obtained by

extracting the square root of an element of C (see 3.24), but we have seen that all such

elements already lie in C. Hence E

N

= C, which is a contradiction. Thus E = C.

2

COROLLARY 5.7. (a) The ﬁeld C is the algebraic closure of R.

(b) The set of all algebraic numbers is an algebraic closure of Q.

PROOF. Part (a) is obvious from the deﬁnition of “algebraic closure” (1.43), and (b) follows

from Corollary 1.46.

2

Cyclotomic extensions

A primitive n

th

root of 1 in F is an element of order n in F

**. Such an element can exist
**

only if F has characteristic 0 or characteristic p not dividing n.

PROPOSITION 5.8. Let F be a ﬁeld of characteristic 0 or characteristic p not dividing n.

Let E be the splitting ﬁeld of X

n

−1.

(a) There exists a primitive n

th

root of 1 in E.

(b) If ζ is a primitive n

th

root of 1 in E, then E = F[ζ].

(c) The ﬁeld E is Galois over F; for each σ ∈ Gal(E/F), there is an i ∈ (Z/nZ)

such

that σζ = ζ

i

for all ζ with ζ

n

= 1; the map σ →[i] is an injective homomorphism

Gal(E/F) →(Z/nZ)

.

PROOF. (a) The roots of X

n

− 1 are distinct, because its derivative nX

n−1

has only zero

as a root (here we use the condition on the characteristic), and so E contains n distinct n

th

roots of 1. The n

th

roots of 1 form a ﬁnite subgroup of E

**, and so (see Exercise 3) they
**

form a cyclic group. Any generator will have order n, and hence will be a primitive n

th

root

of 1.

(b) The roots of X

n

−1 are the powers of ζ, and F[ζ] contains them all.

(c) If ζ

0

is one primitive n

th

root of 1, then the remaining primitive n

th

roots of 1 are

the elements ζ

i

0

with i relatively prime to n. Since, for any automorphism σ of E, σζ

0

is

again a primitive n

th

root of 1, it equals ζ

i

0

for some i relatively prime to n, and the map

σ →i mod n is injective because ζ

0

generates E over F. It obviously is a homomorphism.

Moreover, for any other n

th

root of 1, ζ = ζ

m

0

,

σζ = (σζ

0

)

m

= ζ

im

0

= ζ

i

.

2

The map σ → [i] : Gal(F[ζ]/F) → (Z/nZ)

**need not be surjective. For example, if
**

F = C, then its image is ¦1¦, and if F = R, it is either ¦[1]¦ or ¦[−1], [1]¦. On the other

hand, when n = p is prime, we saw in (1.41) that [Q[ζ] : Q] = p − 1, and so the map is

surjective. We now prove that the map is surjective for all n when F = Q.

The polynomial X

n

− 1 has some obvious factors in Q[X], namely, the polynomials

X

d

−1 for any d[n. The quotient of X

n

−1 by all these factors for d < n is called the n

th

cyclotomic polynomial Φ

n

. Thus

Φ

n

=

¸

(X −ζ) (product over the primitive n

th

roots of 1).

50 5 APPLICATIONS OF GALOIS THEORY

It has degree ϕ(n), the order of (Z/nZ)

. Since every n

th

root of 1 is a primitive d

th

root

of 1 for exactly one d dividing n, we see that

X

n

−1 =

¸

d[n

Φ

d

(X).

For example, Φ

1

(X) = X −1, Φ

2

(X) = X + 1, Φ

3

(X) = X

2

+X + 1, and

Φ

6

(X) =

X

6

−1

(X −1)(X + 1)(X

2

+X + 1)

= X

2

−X + 1.

This gives an easy inductive method of computing the cyclotomic polynomials. Alterna-

tively ask Maple by typing:

with(numtheory);

cyclotomic(n,X);.

Because X

n

− 1 has coefﬁcients in Z and is monic, every monic factor of it in Q[X]

has coefﬁcients in Z (see 1.14). In particular, the cyclotomic polynomials lie in Z[X].

LEMMA 5.9. Let F be a ﬁeld of characteristic 0 or p not dividing n, and let ζ be a primitive

n

th

root of 1 in some extension ﬁeld. The following are equivalent:

(a) the n

th

cyclotomic polynomial Φ

n

is irreducible;

(b) the degree [F[ζ] : F] = ϕ(n);

(c) the homomorphism

Gal(F[ζ]/F) →(Z/nZ)

is an isomorphism.

PROOF. Because ζ is a root of Φ

n

, the minimum polynomial of ζ divides Φ

n

. It is equal to

it if and only if [F[ζ] : F] = ϕ(n), which is true if and only if the injection Gal(F[ζ]/F) →

(Z/nZ)

is onto.

2

THEOREM 5.10. The n

th

cyclotomic polynomial Φ

n

is irreducible in Q[X].

PROOF. Let f(X) be a monic irreducible factor of Φ

n

in Q[X]. Its roots will be primitive

n

th

roots of 1, and we have to show they include all primitive n

th

roots of 1. For this it

sufﬁces to show that

ζ a root of f(X) =⇒ ζ

i

a root of f(X) for all i such that gcd(i, n) = 1.

Such an i is a product of primes not dividing n, and so it sufﬁces to show that

ζ a root of f(X) =⇒ ζ

p

a root of f(X) for all primes p n.

Write

Φ

n

(X) = f(X)g(X).

Proposition 1.14 shows that f(X) and g(X) lie in Z[X]. Suppose ζ is a root of f but

that, for some prime p not dividing n, ζ

p

is not a root of f. Then ζ

p

is a root of g(X),

g(ζ

p

) = 0, and so ζ is a root of g(X

p

). As f(X) and g(X

p

) have a common root, they

have a nontrivial common factor in Q[X] (2.10), which automatically lies in Z[X] (1.14).

Write h(X) →h(X) for the map Z[X] →F

p

[X], and note that

gcd

Z[X]

(f(X), g(X

p

)) = 1 =⇒ gcd

Fp[X]

(f(X), g(X

p

)) = 1.

Dedekind’s theorem on the independence of characters 51

But g(X

p

) = g(X)

p

(use the mod p binomial theorem and that a

p

= a for all a ∈ F

p

), and

so f(X) and g(X) have a common factor. Hence X

n

− 1, when regarded as an element

of F

p

[X], has multiple roots, but we saw in the proof of Proposition 5.8 that it doesn’t.

Contradiction.

2

REMARK 5.11. This proof is very old — in essence it goes back to Dedekind in 1857 —

but its general scheme has recently become popular: take a statement in characteristic zero,

reduce modulo p (where the statement may no longer be true), and exploit the existence

of the Frobenius automorphism a → a

p

to obtain a proof of the original statement. For

example, commutative algebraists use this method to prove results about commutative rings,

and there are theorems about complex manifolds that were ﬁrst proved by reducing things

to characteristic p.

There are some beautiful and mysterious relations between what happens in character-

istic 0 and in characteristic p. For example, let f(X

1

, ..., X

n

) ∈ Z[X

1

, ..., X

n

]. We can

(a) look at the solutions of f = 0 in C, and so get a topological space;

(b) reduce mod p, and look at the solutions of f = 0 in F

p

n.

The Weil conjectures (Weil 1949; proved in part by Grothendieck in the 1960’s and com-

pletely by Deligne in 1973) assert that the Betti numbers of the space in (a) control the

cardinalities of the sets in (b).

THEOREM 5.12. The regular n-gon is constructible if and only if n = 2

k

p

1

p

s

where

the p

i

are distinct Fermat primes.

PROOF. The regular n-gon is constructible if and only if cos

2π

n

(or ζ = e

2πi/n

) is con-

structible. We know that Q[ζ] is Galois over Q, and so (according to 1.37 and 3.23) ζ is

constructible if and only if [Q[ζ] : Q] is a power of 2. But (see GT 3.10)

ϕ(n) =

¸

p[n

(p −1)p

n(p)−1

, n =

¸

p

n(p)

,

and this is a power of 2 if and only if n has the required form.

2

REMARK 5.13. The ﬁnal section of Gauss’s, Disquisitiones Arithmeticae (1801) is titled

“Equations deﬁning sections of a Circle”. In it Gauss proves that the n

th

roots of 1 form a

cyclic group, that X

n

−1 is solvable (this was before the theory of abelian groups had been

developed, and before Galois), and that the regular n-gon is constructible when n is as in

the Theorem. He also claimed to have proved the converse statement

18

. This leads some

people to credit him with the above proof of the irreducibility of Φ

n

, but in the absence of

further evidence, I’m sticking with Dedekind.

Dedekind’s theorem on the independence of characters

THEOREM 5.14 (DEDEKIND’S). Let F be a ﬁeld, and let G be a group (monoid will do).

Then any ﬁnite set ¦χ

1

, . . . , χ

m

¦ of homomorphisms G →F

**is linearly independent over
**

F, i.e.,

¸

a

i

χ

i

= 0 (as a function G →F) =⇒ a

1

= 0, . . . , a

m

= 0.

18

“Whenever n−1 involves prime factors other than 2, we are always led to equations of higher degree....WE

CAN SHOW WITH ALL RIGOR THAT THESE HIGHER-DEGREE EQUATIONS CANNOT BE AVOIDED

IN ANY WAY NOR CAN THEY BE REDUCED TO LOWER-DEGREE EQUATIONS. The limits of the

present work exclude this demonstration here, but we issue this warning lest anyone attempt to achieve geo-

metric constructions for sections other than the ones suggested by our theory (e.g. sections into 7, 9, 11, 13,

19, etc. parts) and so spend his time uselessly.” Ibid. §365.

52 5 APPLICATIONS OF GALOIS THEORY

PROOF. Induction on m. For m = 1, it’s obvious. Assume it for m−1, and suppose that,

for some set ¦χ

1

, . . . , χ

m

¦ of homomorphisms G →F

and a

i

∈ F,

a

1

χ

1

(x) +a

2

χ

2

(x) + +a

m

χ

m

(x) = 0 for all x ∈ G.

We have to show that the a

i

are zero. As χ

1

and χ

2

are distinct, they will take distinct

values on some g ∈ G. On replacing x with gx in the equation, we ﬁnd that

a

1

χ

1

(g)χ

1

(x) +a

2

χ

2

(g)χ

2

(x) + +a

m

χ

m

(g)χ

m

(x) = 0 for all x ∈ G.

On multiplying the ﬁrst equation by χ

1

(g) and subtracting it from the second, we obtain

the equation

a

t

2

χ

2

+ +a

t

m

χ

m

= 0, a

t

i

= a

i

(χ

i

(g) −χ

1

(g)).

The induction hypothesis now shows that a

t

i

= 0 for all i ≥ 2. Since χ

2

(g) − χ

1

(g) = 0,

we must have a

2

= 0, and the induction hypothesis shows that all the remaining a

i

’s are

also zero.

2

COROLLARY 5.15. Let F

1

and F

2

be ﬁelds, and let σ

1

, ..., σ

m

be distinct homomorphisms

F

1

→F

2

. Then σ

1

, ..., σ

m

are linearly independent over F

2

.

PROOF. Apply the theorem to χ

i

= σ

i

[F

1

.

2

COROLLARY 5.16. Let E be a ﬁnite separable extension of F of degree m. Let α

1

, . . . , α

m

be a basis for E over F, and let σ

1

, . . . , σ

m

be distinct F-homomorphisms from E into a

ﬁeld Ω. Then the matrix whose (i, j)

th

-entry is σ

i

α

j

is invertible.

PROOF. If not, there exist c

i

∈ Ωsuch that

¸

m

i=1

c

i

σ

i

(α

j

) = 0 for all j. But

¸

m

i=1

c

i

σ

i

: E →

Ω is F-linear, and so this implies that

¸

m

i=1

c

i

σ

i

(α) = 0 for all α ∈ E, which contradicts

Corollary 5.15.

2

The normal basis theorem

DEFINITION 5.17. Let E be a ﬁnite Galois extension of F with Galois group G. A normal

basis for E is an F-basis of the form ¦σα [ σ ∈ G¦, i.e., an F-basis consisting of the

conjugates of an element α of E.

THEOREM 5.18 (NORMAL BASIS THEOREM). Every Galois extension has a normal basis.

PROOF. Let E/F be a Galois extension with Galois group G. We give two proofs, the

ﬁrst of which assumes that F is inﬁnite and the second that G is cyclic. Since every Galois

extension of a ﬁnite ﬁeld is cyclic (4.18), this covers all cases.

Assume that F is inﬁnite. This has the consequence that, if f ∈ F[X

1

, . . . , X

m

] has

the property that f(a

1

, . . . , a

m

) = 0 for all a

1

, . . . , a

m

∈ F, then f(X

1

, . . . , X

m

) = 0. We

prove this by induction on m. For m = 1 it follows from the fact that a nonzero polynomial

in one symbol has only ﬁnitely many roots. For m > 1, write

f =

¸

c

i

(X

1

, . . . , X

m−1

)X

i

m

.

For any m−1-tuple, a

1

, . . . , a

m−1

,

f(a

1

, . . . , a

m−1

, X

m

)

Hilbert’s Theorem 90. 53

is a polynomial in X

m

having every element of F as a root. Therefore, each of its coefﬁ-

cients is zero: c

i

(a

1

, . . . , a

m−1

) = 0 for all i. Since this holds for all (a

1

, . . . , a

m−1

), the

induction hypothesis shows that c

i

(X

1

, . . . , X

m−1

) is zero.

Now number the elements of G as σ

1

, . . . , σ

m

(with σ

1

= 1).

Let f(X

1

, . . . , X

m

) ∈ F[X

1

, . . . , X

m

] have the property that

f(σ

1

α, . . . , σ

m

α) = 0

for all α ∈ E. For a basis α

1

, . . . , α

m

of E over F, let

g(Y

1

, . . . , Y

m

) = f(

¸

m

i=1

Y

i

σ

1

α

i

,

¸

m

i=1

Y

i

σ

2

α

i

, . . .).

The hypothesis on f implies that g(a

1

, . . . , a

m

) = 0 for all a

i

∈ F, and so g = 0. But the

matrix (σ

i

α

j

) is invertible (5.16). Since g is obtained from f by an invertible linear change

of variables, f can be obtained from g by the inverse linear change of variables. Therefore

it also is zero.

Write X

i

= X(σ

i

), and let A = (X(σ

i

σ

j

)), i.e., A is the m m matrix having

X

k

in the (i, j)

th

place if σ

i

σ

j

= σ

k

. Then det(A) is a polynomial in X

1

, . . . , X

m

, say,

det(A) = h(X

1

, . . . , X

m

). Clearly, h(1, 0, . . . , 0) is the determinant of a matrix having

exactly one 1 in each row and each column and its remaining entries 0. Hence the rows

of the matrix are a permutation of the rows of the identity matrix, and so its determinant

is ±1. In particular, h is not identically zero, and so there exists an α ∈ E

such that

h(σ

1

α, . . . , σ

m

α) (= det(σ

i

σ

j

α)) is nonzero. We shall show that ¦σ

i

α¦ is a normal basis.

For this, it sufﬁces to show that σ

i

α are linearly independent over F. Suppose

¸

m

j=1

a

j

σ

j

α = 0

for some a

j

∈ F. On applying σ

1

, . . . , σ

m

successively, we obtain a system of m-equations

¸

a

j

σ

i

σ

j

α = 0

in the m“unknowns” a

j

. Because this system of equations is nonsingular, the a

j

’s are zero.

This completes the proof of the lemma in the case that F is inﬁnite.

Now assume that G is cyclic generated, say, by an element σ

0

of order n. Then [E :

F] = n. The minimum polynomial of σ

0

regarded as an endomorphism of the F-vector

space E is the monic polynomial in F[X] of least degree such that P(σ

0

) = 0 (as an

endomorphism of E). It has the property that it divides every polynomial Q(X) ∈ F[X]

such that Q(σ

0

) = 0. Since σ

n

0

= 1, P(X) divides X

n

−1. On the other hand, Dedekind’s

theorem on the independence of characters (5.14) implies that 1, σ

0

, . . . , σ

n−1

0

are linearly

independent over F, and so deg P(X) > n − 1. We conclude that P(X) = X

n

− 1.

Therefore, as an F[X]-module with X acting as σ

0

, E is isomorphic to F[X]/(X

n

− 1).

For any generator α of E as a F[X]-module, α, σ

0

α, . . . , σ

0

α

n−1

is a F-basis for E.

2

Hilbert’s Theorem 90.

Let G be a ﬁnite group. A G-module is an abelian group M together with an action of G,

i.e., a map GM →M such that

(a) σ(m+m

t

) = σm+σm

t

for all σ ∈ G, m, m

t

∈ M;

(b) (στ)(m) = σ(τm) for all σ, τ ∈ G, m ∈ M;

(c) 1m = m for all m ∈ M.

54 5 APPLICATIONS OF GALOIS THEORY

Thus, to give an action of G on M is the same as to give a homomorphism G → Aut(M)

(automorphisms of M as an abelian group).

EXAMPLE 5.19. Let E be a Galois extension of F with Galois group G. Then (E, +) and

(E

, ) are G-modules.

Let M be a G-module. A crossed homomorphism is a map f : G →M such that

f(στ) = f(σ) +σf(τ) for all σ, τ ∈ G.

Note that the condition implies that f(1) = f(1 1) = f(1) +f(1), and so f(1) = 0.

EXAMPLE 5.20. (a) Let f : G →M be a crossed homomorphism. For any σ ∈ G,

f(σ

2

) = f(σ) +σf(σ),

f(σ

3

) = f(σ σ

2

) = f(σ) +σf(σ) +σ

2

f(σ)

f(σ

n

) = f(σ) +σf(σ) + +σ

n−1

f(σ).

Thus, if G is a cyclic group of order n generated by σ, then a crossed homomorphism

f : G →M is determined by its value, x say, on σ, and x satisﬁes the equation

x +σx + +σ

n−1

x = 0, (*)

Conversely, if x ∈ M satisﬁes (*), then the formulas f(σ

i

) = x + σx + + σ

i−1

x

deﬁne a crossed homomorphism f : G → M. Thus, for a ﬁnite group G = 'σ`, there is a

one-to-one correspondence

¦crossed homs f : G →M¦

f↔f(σ)

←→ ¦x ∈ M satisfying (*)¦.

(b) For any x ∈ M, we obtain a crossed homomorphism by putting

f(σ) = σx −x, all σ ∈ G.

A crossed homomorphism of this form is called a principal crossed homomorphism.

(c) If G acts trivially on M, i.e., σm = m for all σ ∈ G and m ∈ M, then a crossed

homomorphism is simply a homomorphism, and there are no nonzero principal crossed

homomorphisms.

The sum and difference of two crossed homomorphisms is again a crossed homomor-

phism, and the sum and difference of two principal crossed homomorphisms is again prin-

cipal. Thus we can deﬁne

H

1

(G, M) =

¦crossed homomorphisms¦

¦principal crossed homomorphisms¦

(quotient abelian group). The cohomology groups H

n

(G, M) have been deﬁned for all

n ∈ N, but since this was not done until the twentieth century, it will not be discussed in

this course.

Hilbert’s Theorem 90. 55

EXAMPLE 5.21. Let π:

˜

X →X be the universal covering space of a topological space X,

and let Γ be the group of covering transformations. Under some fairly general hypotheses,

a Γ-module M will deﬁne a sheaf ´on X, and H

1

(X, ´) · H

1

(Γ, M). For example,

when M = Z with the trivial action of Γ, this becomes the isomorphism H

1

(X, Z) ·

H

1

(Γ, Z) = Hom(Γ, Z).

THEOREM 5.22. Let E be a Galois extension of F with group G; then H

1

(G, E

) = 0,

i.e., every crossed homomorphism G →E

is principal.

PROOF. Let f be a crossed homomorphism G → E

**. In multiplicative notation, this
**

means,

f(στ) = f(σ) σ(f(τ)), σ, τ ∈ G,

and we have to ﬁnd a γ ∈ E

such that f(σ) =

σγ

γ

for all σ ∈ G. Because the f(τ) are

nonzero, Corollary 5.15 implies that

¸

τ∈G

f(τ)τ : E →E

is not the zero map, i.e., there exists an α ∈ E such that

β

df

=

¸

τ∈G

f(τ)τα = 0.

But then, for σ ∈ G,

σβ =

¸

τ∈G

σ(f(τ)) στ(α)

=

¸

τ∈G

f(σ)

−1

f(στ) στ(α)

= f(σ)

−1

¸

τ∈G

f(στ)στ(α),

which equals f(σ)

−1

β because, as τ runs over G, so also does στ. Therefore, f(σ) =

β

σ(β)

and we can take β = γ

−1

.

2

Let E be a Galois extension of F with Galois group G. We deﬁne the norm of an

element α ∈ E to be

Nmα =

¸

σ∈G

σα.

For τ ∈ G,

τ(Nmα) =

¸

σ∈G

τσα = Nmα,

and so Nmα ∈ F. The map

α →Nmα: E

→F

**is a obviously a homomorphism.
**

EXAMPLE 5.23. The norm map C

→R

is α →[α[

2

and the norm map Q[

√

d]

→Q

is a +b

√

d →a

2

−db

2

.

We are interested in determining the kernel of the norm map. Clearly an element of the

form

β

τβ

has norm 1, and our next result show that, for cyclic extensions, all elements with

norm 1 are of this form.

56 5 APPLICATIONS OF GALOIS THEORY

COROLLARY 5.24 (HILBERT’S THEOREM 90).

19

Let E be a ﬁnite cyclic extension of F

with Galois group 'σ`; if Nm

E/F

α = 1, then α = β/σβ for some β ∈ E.

PROOF. Let m = [E : F]. The condition on α is that α σα σ

m−1

α = 1, and so (see

5.20a) there is a crossed homomorphismf : 'σ` →E

**with f(σ) = α. Theorem 5.22 now
**

shows that f is principal, which means that there is a β with f(σ) = β/σβ.

2

Cyclic extensions.

We are now able to classify the cyclic extensions of degree n of a ﬁeld F in the case that F

contains n n

th

roots of 1.

THEOREM 5.25. Let F be a ﬁeld containing a primitive n

th

root of 1.

(a) The Galois group of X

n

−a is cyclic of order dividing n.

(b) Conversely, if E is cyclic of degree n over F, then there is an element β ∈ E such

that E = F[β] and β

n

∈ F; hence E is the splitting ﬁeld of X

n

−β

n

.

PROOF. (a) If α is one root of X

n

−a, then the other roots are the elements of the form ζα

with ζ an n

th

root of 1. Hence the splitting ﬁeld of X

n

−a is F[α]. The map σ →

σα

α

is an

injective homomorphism from Gal(F[α]/F) into the cyclic group 'ζ`.

(b) Let ζ be a primitive n

th

root of 1 in F, and let σ generate Gal(E/F). Then Nmζ =

ζ

n

= 1, and so, according to Hilbert’s Theorem 90, there is an element β ∈ E such that

σβ = ζβ. Then σ

i

β = ζ

i

β, and so only the identity element of Gal(E/F) ﬁxes β — we

conclude by the fundamental theorem of Galois theory that E = F[β]. On the other hand

σβ

n

= ζ

n

β

n

= β

n

, and so β

n

∈ F.

2

REMARK 5.26. (a) Assume F contains a primitive n

th

root of 1. Then, two cyclic exten-

sion F[a

1

n

] and F[b

1

n

] of F are isomorphic if and only if a and b generate the same subgroup

of F

/F

n

.

(b) The polynomial X

n

−a, n ≥ 2, is irreducible in F[X] under the following condition:

a is not a p

th

power for any p dividing n, and, if 4[n, then a / ∈ −4F

4

. See Lang, Algebra,

Addison-Wesley, 1965, VIII, '9, Theorem 16.

(c) If F has characteristic p (hence has no p

th

roots of 1 other than 1), then X

p

−X−a

is irreducible in F[X] unless a = b

p

− b for some b ∈ F, and when it is irreducible, its

Galois group is cyclic of order p (generated by α → α + 1 where α is a root). Moreover,

every extension of F which is cyclic of degree p is the splitting ﬁeld of such a polynomial.

REMARK 5.27 (KUMMER THEORY). Theorem 5.25 and Remark 5.26a classify the cyclic

extensions of F order n in the case that F contains a primitive n

th

root of 1. Under the same

assumption on F, it is possible to extend this to a classiﬁcation of the Galois extensions of

F with abelian Galois group of exponent n (i.e., with Galois group a quotient of (Z/nZ)

r

for some r).

Let E be such an extension of F, and let

S(E) = ¦a ∈ F

[ a becomes an n

th

power in E¦.

19

This is Satz 90 in Hilbert’s book, Theorie der Algebraischen Zahlk¨ orper, 1897. The theorem was discov-

ered by Kummer in the special case of Q[ζp]/Q, and generalized to Theorem 5.22 by E. Noether. Theorem

5.22, as well as various vast generalizations of it, are also referred to as Hilbert’s Theorem 90.

For an illuminating discussion of Hilbert’s book, see the introduction to the English translation (Springer

1998) written by F. Lemmermeyer and N. Schappacher.

Proof of Galois’s solvability theorem 57

Then S(E) is a subgroup of F

containing F

n

, and the map E → S(E) deﬁnes a one-

to-one correspondence between the abelian extensions of E of exponent n and the groups

S(E),

F

⊃ S(E) ⊃ F

n

,

such that (S(E) : F

n

) < ∞. The ﬁeld E is recovered from S(E) as the splitting ﬁeld

of

¸

(X

n

−a) (product over a set of representatives for S(E)/F

n

). Moreover, there is a

perfect pairing

(a, σ) →

σa

a

:

S(E)

F

n

Gal(E/F) →µ

n

(group of n

th

roots of 1).

In particular, [E : F] = (S(E) : F

n

). (Cf. Exercise 5 for the case n = 2.)

Proof of Galois’s solvability theorem

LEMMA 5.28. Let f ∈ F[X] be separable, and let F

t

be an extension ﬁeld of F. Then the

Galois group of f as an element of F

t

[X] is a subgroup of that of f as an element of F[X].

PROOF. Let E

t

be a splitting ﬁeld for f over F

t

, and let α

1

, . . . , α

m

be the roots of f(X)

in E

t

. Then E = F[α

1

, ..., α

m

] is a splitting ﬁeld of f over F. Any element of Gal(E

t

/F

t

)

permutes the α

i

and so maps E into itself. The map σ →σ[E is an injection Gal(E

t

/F

t

) →

Gal(E/F).

2

THEOREM 5.29. Let F be a ﬁeld of characteristic 0. A polynomial in F[X] is solvable if

and only if its Galois group is solvable.

PROOF. ⇐=: Let f ∈ F[X] have solvable Galois group G

f

. Let F

t

= F[ζ] where ζ is a

primitive n

th

root of 1 for some large n — for example, n = (deg f)! will do. The lemma

shows that the Galois group G of f as an element of F

t

[X] is a subgroup of G

f

, and hence

is also solvable (GT 6.6a). This means that there is a sequence of subgroups

G = G

0

⊃ G

1

⊃ ⊃ G

m−1

⊃ G

m

= ¦1¦

such that each G

i

is normal in G

i−1

and G

i−1

/G

i

is cyclic. Let E be a splitting ﬁeld of

f(X) over F

t

, and let F

i

= E

G

i

. We have a sequence of ﬁelds

F ⊂ F[ζ] = F

t

= F

0

⊂ F

1

⊂ F

2

⊂ ⊂ F

m

= E

with F

i

cyclic over F

i−1

. Theorem 5.25b shows that F

i

= F

i−1

[α

i

] with α

[F

i

:F

i−1

]

i

∈ F

i−1

,

each i, and this shows that f is solvable.

=⇒: It sufﬁces to show that G

f

is a quotient of a solvable group (GT 6.6a). Hence it

sufﬁces to ﬁnd a solvable extension

˜

E of F such that f(X) splits in

˜

E[X].

We are given that there exists a tower of ﬁelds

F = F

0

⊂ F

1

⊂ F

2

⊂ ⊂ F

m

such that

(a) F

i

= F

i−1

[α

i

], α

r

i

i

∈ F

i−1

;

(b) F

m

contains a splitting ﬁeld for f.

58 5 APPLICATIONS OF GALOIS THEORY

Let n = r

1

r

m

, and let Ω be a ﬁeld Galois over F and containing (a copy of) F

m

and a primitive n

th

root ζ of 1. For example, choose a primitive element γ for F

m

over F

(see 5.1), and take Ω to be a splitting ﬁeld of g(X)(X

n

− 1) where g(X) is the minimum

polynomial of γ over F.

Let G be the Galois group of Ω/F, and let

˜

E be the Galois closure of F

m

[ζ] in Ω.

According to (3.17a),

˜

E is the composite of the ﬁelds σF

m

[ζ], σ ∈ G, and so it is generated

over F by the elements

ζ, α

1

, α

2

, . . . , α

m

, σα

1

, . . . , σα

m

, σ

t

α

1

, . . . .

We adjoin these elements to F one by one to get a sequence of ﬁelds

F ⊂ F[ζ] ⊂ F[ζ, α

1

] ⊂ ⊂ F

t

⊂ F

tt

⊂ ⊂

˜

E

in which each ﬁeld F

tt

is obtained from its predecessor F

t

by adjoining an r

th

root of

an element of F

t

(r = r

1

, . . . , r

m

, or n). According to (5.8) and (5.25a), each of these

extensions is cyclic, and so

˜

E/F is a solvable extension.

2

The general polynomial of degree n

When we say that the roots of

aX

2

+bX +c

are

−b ±

√

b

2

−4ac

2a

we are thinking of a, b, c as variables: for any particular values of a, b, c, the formula gives

the roots of the particular equation. We shall prove in this section that there is no similar

formula for the roots of the “general polynomial” of degree ≥ 5.

We deﬁne the general polynomial of degree n to be

f(X) = X

n

−t

1

X

n−1

+ + (−1)

n

t

n

∈ F[t

1

, ..., t

n

][X]

where the t

i

are variables. We shall show that, when we regard f as a polynomial in X with

coefﬁcients in the ﬁeld F(t

1

, . . . , t

n

), its Galois group is S

n

. Then Theorem 5.29 proves

the above remark (at least in characteristic zero).

Symmetric polynomials

Let R be a commutative ring (with 1). A polynomial P(X

1

, ..., X

n

) ∈ R[X

1

, . . . , X

n

] is

said to be symmetric if it is unchanged when its variables are permuted, i.e., if

P(X

σ(1)

, . . . , X

σ(n)

) = P(X

1

, . . . , X

n

), all σ ∈ S

n

.

For example

p

1

=

¸

i

X

i

= X

1

+X

2

+ +X

n

,

p

2

=

¸

i<j

X

i

X

j

= X

1

X

2

+X

1

X

3

+ +X

1

X

n

+X

2

X

3

+ +X

n−1

X

n

,

p

3

=

¸

i<j<k

X

i

X

j

X

k

, = X

1

X

2

X

3

+

p

r

=

¸

i

1

<<ir

X

i

1

...X

ir

p

n

= X

1

X

2

X

n

The general polynomial of degree n 59

are all symmetric because p

r

is the sum of all monomials of degree r made up out of distinct

X

i

’s. These particular polynomials are called the elementary symmetric polynomials.

THEOREM 5.30 (SYMMETRIC POLYNOMIALS THEOREM). Every symmetric polynomial

P(X

1

, ..., X

n

) in R[X

1

, ..., X

n

] is equal to a polynomial in the elementary symmetric poly-

nomials with coefﬁcients in R, i.e., P ∈ R[p

1

, ..., p

n

].

PROOF. We deﬁne an ordering on the monomials in the X

i

by requiring that

X

i

1

1

X

i

2

2

X

in

n

> X

j

1

1

X

j

2

2

X

jn

n

if either

i

1

+i

2

+ +i

n

> j

1

+j

2

+ +j

n

or equality holds and, for some s,

i

1

= j

1

, . . . , i

s

= j

s

, but i

s+1

> j

s+1

.

For example,

X

1

X

3

2

X

3

> X

1

X

2

2

X

3

> X

1

X

2

X

2

3

.

Let X

k

1

1

X

kn

n

be the highest monomial occurring in P with a coefﬁcient c = 0. Because

P is symmetric, it contains all monomials obtained from X

k

1

1

X

kn

n

by permuting the

X’s. Hence k

1

≥ k

2

≥ ≥ k

n

.

The highest monomial in p

i

is X

1

X

i

, and it follows that the highest monomial in

p

d

1

1

p

dn

n

is

X

d

1

+d

2

++dn

1

X

d

2

++dn

2

X

dn

n

. (∗)

Therefore the highest monomial of P(X

1

, . . . , X

n

) −cp

k

1

−k

2

1

p

k

2

−k

3

2

p

kn

n

is strictly less

than the highest monomial in P(X

1

, . . . , X

n

). We can repeat this argument with the poly-

nomial on the left, and after a ﬁnite number of steps, we will arrive at a representation of P

as a polynomial in p

1

, . . . , p

n

.

2

Let f(X) = X

n

+ a

1

X

n−1

+ + a

n

∈ R[X], and suppose that f splits over some

ring S containing R:

f(X) =

¸

n

i=1

(X −α

i

), α

i

∈ S.

Then

a

1

= −p

1

(α

1

, . . . , α

n

), a

2

= p

2

(α

1

, . . . , α

n

), . . . , a

n

= ±p

n

(α

1

, . . . , α

n

).

Thus the elementary symmetric polynomials in the roots of f(X) lie in R, and so the theo-

rem implies that every symmetric polynomial in the roots of f(X) lies in R. For example,

the discriminant

D(f) =

¸

i<j

(α

i

−α

j

)

2

of f lies in R.

60 5 APPLICATIONS OF GALOIS THEORY

Symmetric functions

THEOREM 5.31 (SYMMETRIC FUNCTIONS THEOREM). When S

n

acts on E = F(X

1

, ..., X

n

)

by permuting the X

i

’s, the ﬁeld of invariants is F(p

1

, ..., p

n

).

FIRST PROOF. Let f ∈ F(X

1

, . . . , X

n

) be symmetric (i.e., ﬁxed by S

n

). Set f = g/h,

g, h ∈ F[X

1

, . . . , X

n

]. The polynomials H =

¸

σ∈Sn

σh and Hf are symmetric, and

therefore lie in F[p

1

, . . . , p

n

] (5.30). Hence their quotient f = Hf/H lies in F(p

1

, . . . , p

n

).

2

SECOND PROOF. Clearly

F(p

1

, . . . , p

n

) ⊂ E

Sn

⊂ E.

On the one hand, [E: F(p

1

, . . . , p

n

)] ≤ n! because E is the splitting ﬁeld of (T−X

1

) (T−

X

n

) over F(p

1

, . . . , p

n

); on the other, [E: E

Sn

] ≥ n! by (2.8).

2

COROLLARY 5.32. The ﬁeld F(X

1

, ..., X

n

) is Galois over F(p

1

, ..., p

n

) with Galois group

S

n

(acting by permuting the X

i

).

PROOF. We have shown that F(p

1

, . . . , p

n

) = F(X

1

, . . . , X

n

)

Sn

, and so this follows from

(3.10).

2

As we noted in the second proof, F(X

1

, . . . , X

n

) is the splitting ﬁeld over F(p

1

, . . . , p

n

)

of

g(T) = (T −X

1

) (T −X

n

) = X

n

−p

1

X

n−1

+ + (−1)

n

p

n

.

Therefore, the Galois group of g(T) ∈ F(p

1

, . . . , p

n

)[T] is S

n

.

The general polynomial of degree n

THEOREM 5.33. The Galois group of the general polynomial of degree n is S

n

.

PROOF. Let f(X) be the general polynomial of degree n,

f(X) = X

n

−t

1

X

n−1

+ + (−1)

n

t

n

∈ F[t

1

, ..., t

n

][X].

If we can show that the map

t

i

→p

i

: F[t

1

, . . . , t

n

] →F[p

1

, . . . , p

n

]

is injective (i.e., the p

i

are algebraically independent over F, see p79), then it will extend to

an isomorphism

F(t

1

, . . . , t

n

) →F(p

1

, . . . , p

n

)

sending f(X) to

g(X) = X

n

−p

1

X

n−1

+ + (−1)

n

p

n

∈ F(p

1

, . . . , p

n

)[X].

Therefore the statement will follow from Corollary 5.32.

We now prove that the p

i

are algebraically independent

20

. Suppose on the contrary

that there exists a P(t

1

, . . . , t

n

) such that P(p

1

, . . . , p

n

) = 0. Equation (∗), p59, shows

that if m

1

(t

1

, . . . , t

n

) and m

2

(t

1

, . . . , t

n

) are distinct monomials, then m

1

(p

1

, . . . , p

n

) and

m

2

(p

1

, . . . , p

n

) have distinct highest monomials. Therefore, cancellation can’t occur, and

so P(t

1

, . . . , t

n

) must be the zero polynomial.

2

20

This can also be proved by noting that, because F(X1, . . . , Xn) is algebraic over F(p1, . . . , pn), the latter

must have transcendence degree n (see §8).

Norms and traces 61

REMARK 5.34. Since S

n

occurs as a Galois group over Q, and every ﬁnite group occurs

as a subgroup of some S

n

, it follows that every ﬁnite group occurs as a Galois group over

some ﬁnite extension of Q, but does every ﬁnite Galois group occur as a Galois group over

Q itself?

The Hilbert-Noether program for proving this was the following. Hilbert proved that

if G occurs as the Galois group of an extension E ⊃ Q(t

1

, ..., t

n

) (the t

i

are symbols),

then it occurs inﬁnitely often as a Galois group over Q. For the proof, realize E as the

splitting ﬁeld of a polynomial f(X) ∈ k[t

1

, . . . , t

n

][X] and prove that for inﬁnitely many

values of the t

i

, the polynomial you obtain in Q[X] has Galois group G. (This is quite a

difﬁcult theorem — see Serre, J.-P., Lectures on the Mordell-Weil Theorem, 1989, Chapter

9.) Noether conjectured the following: Let G ⊂ S

n

act on F(X

1

, ..., X

n

) by permuting

the X

i

; then F(X

1

, . . . , X

n

)

G

≈ F(t

1

, ..., t

n

) (for variables t

i

). However, Swan proved in

1969 that the conjecture is false for G the cyclic group of order 47. Hence this approach

can not lead to a proof that all ﬁnite groups occur as Galois groups over Q, but it doesn’t

exclude other approaches. For more information on the problem, see Serre, ibid., Chapter

10, and Serre, J.-P., Topics in Galois Theory, 1992.

REMARK 5.35. Take F = C, and consider the subset of C

n+1

deﬁned by the equation

X

n

−T

1

X

n−1

+ + (−1)

n

T

n

= 0.

It is a beautiful complex manifold S of dimension n. Consider the projection

π: S →C

n

, (x, t

1

, . . . , t

n

) →(t

1

, . . . , t

n

).

Its ﬁbre over a point (a

1

, . . . , a

n

) is the set of roots of the polynomial

X

n

−a

1

X

n−1

+ + (−1)

n

a

n

.

The discriminant D(f) of f(X) = X

n

− T

1

X

n−1

+ + (−1)

n

T

n

is a polynomial in

C[T

1

, . . . , T

n

]. Let ∆ be the zero set of D(f) in C

n

. Then over each point of C

n

∆, there

are exactly n points of S, and S π

−1

(∆) is a covering space over C

n

∆.

A brief history

As far back as 1500 BC, the Babylonians (at least) knew a general formula for the roots of

a quadratic polynomial. Cardan (about 1515 AD) found a general formula for the roots of a

cubic polynomial. Ferrari (about 1545 AD) found a general formula for the roots of quartic

polynomial (he introduced the resolvent cubic, and used Cardan’s result). Over the next

275 years there were many fruitless attempts to obtain similar formulas for higher degree

polynomials, until, in about 1820, Rufﬁni and Abel proved that there are none.

Norms and traces

Recall that, for an n n matrix A = (a

ij

)

Tr(A) =

¸

i

a

ii

(trace of A)

det(A) =

¸

σ∈Sn

sign(σ)a

1σ(1)

a

nσ(n)

, (determinant of A)

c

A

(X) = det(XI

n

−A) (characteristic polynomial of A).

62 5 APPLICATIONS OF GALOIS THEORY

Moreover,

c

A

(X) = X

n

−Tr(A)X

n−1

+ + (−1)

n

det(A).

None of these is changed when A is replaced by its conjugate UAU

−1

by an invertible

matrix U. Therefore, for any endomorphism α of a ﬁnite dimensional vector space V , we

can deﬁne

21

Tr(α) = Tr(A), det(α) = det(A), c

α

(X) = c

A

(X)

where A is the matrix of α with respect to any basis of V . If β is a second endomorphism

of V ,

Tr(α +β) = Tr(α) + Tr(β);

det(αβ) = det(α) det(β).

Now let E be a ﬁnite ﬁeld extension of F of degree n. An element α of E deﬁnes an

F-linear map

α

L

: E →E, x →αx,

and we deﬁne

Tr

E/F

(α) = Tr(α

L

) (trace of α)

Nm

E/F

(α) = det(α

L

) (norm of α)

c

α,E/F

(X) = c

α

L

(X) (characteristic polynomial of α).

Thus, Tr

E/F

is a homomorphism (E, +) → (F, +), and Nm

E/F

is a homomorphism

(E

, ) →(F

, ).

EXAMPLE 5.36. (a) Consider the ﬁeld extension C ⊃ R. For α = a+bi, the matrix of α

L

with respect to the basis ¦1, i¦ is

a −b

b a

, and so

Tr

C/R

(α) = 2'(α), Nm

C/R

(α) = [α[

2

.

(b) For a ∈ F, a

L

is multiplication by the scalar a. Therefore

Tr

E/F

(a) = na , Nm

E/F

(a) = a

n

, c

a,E/F

(X) = (X −a)

n

where n = [E : F].

Let E = Q[α, i] be the splitting ﬁeld of X

8

−2. To compute the trace and norm of α in

E, the deﬁnition requires us to compute the trace and norm of a 16 16 matrix. The next

proposition gives us a quicker method.

PROPOSITION 5.37. Let E/F be a ﬁnite extension of ﬁelds, and let f(X) be the minimum

polynomial of α ∈ E. Then

c

α,E/F

(X) = f(X)

[E:F[α]]

.

21

The coefﬁcients of the characteristic polynomial

cα(X) = X

n

+c1X

n−1

+ · · · +cn,

of α have the following description

ci = (−1)

i

Tr(α|

i

V )

— see Bourbaki, N., Algebra, Chapter 3, 8.11.

Norms and traces 63

PROOF. Suppose ﬁrst that E = F[α]. In this case, we have to show that c

α

(X) = f(X).

Note that α → α

L

is an injective homomorphism from E into the ring of endomorphisms

of E as a vector space over F. The Cayley-Hamilton theorem shows that c

α

(α

L

) = 0, and

therefore c

α

(α) = 0. Hence f[c

α

, but they are monic of the same degree, and so they are

equal.

For the general case, let β

1

, ..., β

n

be a basis for F[α] over F, and let γ

1

, ..., γ

m

be

a basis for E over F[α]. As we saw in the proof of (1.20), ¦β

i

γ

k

¦ is a basis for E over

F. Write αβ

i

=

¸

a

ji

β

j

. Then, according to the ﬁrst case proved, A =

df

(a

ij

) has

characteristic polynomial f(X). But αβ

i

γ

k

=

¸

a

ji

β

j

γ

k

, and so the matrix of α

L

with

respect to ¦β

i

γ

k

¦ breaks up into nn blocks with A’s down the diagonal and zero matrices

elsewhere, from which it follows that c

α

L

(X) = c

A

(X)

m

= f(X)

m

.

2

COROLLARY 5.38. Suppose that the roots of the minimumpolynomial of α are α

1

, . . . , α

n

(in some splitting ﬁeld containing E), and that [E : F[α]] = m. Then

Tr(α) = m

¸

n

i=1

α

i

, Nm

E/F

α = (

¸

n

i=1

α

i

)

m

.

PROOF. Write the minimum polynomial of α as

f(X) = X

n

+a

1

X

n−1

+ +a

n

=

¸

(X −α

i

),

so that

a

1

= −

¸

α

i

, and

a

n

= (−1)

n

¸

α

i

.

Then

c

α

(X) = (f(X))

m

= X

mn

+ma

1

X

mn−1

+ +a

m

n

,

so that

Tr

E/F

(α) = −ma

1

= m

¸

α

i

, and

Nm

E/F

(α) = (−1)

mn

a

m

n

= (

¸

α

i

)

m

.

2

EXAMPLE 5.39. (a) Consider the extension C ⊃ R. If α ∈ C ` R, then

c

α

(X) = f(X) = X

2

−2'(α)X +[α[

2

.

If α ∈ R, then c

α

(X) = (X −a)

2

.

(b) Let E be the splitting ﬁeld of X

8

−2. Then E has degree 16 over Qand is generated

by α =

8

√

2 and i =

√

−1 (see Exercise 16). The minimum polynomial of α is X

8

−2, and

so

c

α,Q[α]/Q

(X) = X

8

−2, c

α,E/Q

(X) = (X

8

−2)

2

Tr

Q[α]/Q

α = 0, Tr

E/Q

α = 0

Nm

Q[α]/Q

α = −2, Nm

E/Q

α = 4

REMARK 5.40. Let E be a separable extension of F, and let Σ be the set of F-homomorphisms

of E into an algebraic closure Ω of F. Then

Tr

E/F

α =

¸

σ∈Σ

σα

Nm

E/F

α =

¸

σ∈Σ

σα.

64 5 APPLICATIONS OF GALOIS THEORY

When E = F[α], this follows from 5.38 and the observation (cf. 2.1b) that the σα are

the roots of the minimum polynomial f(X) of α over F. In the general case, the σα are

still roots of f(X) in Ω, but now each root of f(X) occurs [E : F[α]] times (because each

F-homomorphism F[α] →Ω has [E : F[α]] extensions to E). For example, if E is Galois

over F with Galois group G, then

Tr

E/F

α =

¸

σ∈G

σα

Nm

E/F

α =

¸

σ∈G

σα.

PROPOSITION 5.41. For ﬁnite extensions E ⊃ M ⊃ F, we have

Tr

E/M

◦ Tr

M/F

= Tr

E/F

,

Nm

E/M

◦ Nm

M/F

= Nm

E/F

.

PROOF. If E is separable over F, then this can be proved fairly easily using the descriptions

in the above remark. We omit the proof in the general case.

2

PROPOSITION 5.42. Let f(X) ∈ F[X] factor as f(X) =

¸

m

i=1

(X−α

i

) in some splitting

ﬁeld, and let α = α

1

. Then, with f

t

=

df

dX

(formal derivative), we have

disc f(X) = (−1)

m(m−1)/2

Nm

F[α]/F

f

t

(α).

PROOF. Compute that

disc f(X)

df

=

¸

i<j

(α

i

−α

j

)

2

= (−1)

m(m−1)/2

¸

i

(

¸

j,=i

(α

i

−α

j

))

= (−1)

m(m−1)/2

¸

i

f

t

(α

i

)

= (−1)

m(m−1)/2

Nm

F[α]/F

(f

t

(α)) (by 5.40).

2

EXAMPLE 5.43. We compute the discriminant of

f(X) = X

n

+aX +b, a, b ∈ F,

assumed to be irreducible and separable, by computing the norm of

γ

df

= f

t

(α) = nα

n−1

+a, f(α) = 0.

On multiplying the equation

α

n

+aα +b = 0

by nα

−1

and rearranging, we obtain the equation

nα

n−1

= −na −nbα

−1

.

Hence

γ = nα

n−1

+a = −(n −1)a −nbα

−1

.

Exercises 65

Solving for α gives

α =

−nb

γ + (n −1)a

.

From the last two equations, it is clear that F[α] = F[γ], and so the minimum polynomial

of γ over F has degree n also. If we write

f

−nb

X + (n −1)a

=

P(X)

Q(X)

P(X) = (X + (n −1)a)

n

−na(X + (n −1)a)

n−1

+ (−1)

n

n

n

b

n−1

Q(X) = (X + (n −1)a)

n

/b,

then

P(γ) = f(α) Q(γ) = 0.

As

Q(γ) =

(γ + (n −1)a)

n

b

=

(−nb)

n

α

n

b

= 0

and P(X) is monic of degree n, it must be the minimum polynomial of γ. Therefore Nmγ

is (−1)

n

times the constant term of P(X), namely,

Nmγ = n

n

b

n−1

+ (−1)

n−1

(n −1)

n−1

a

n

.

Therefore,

disc(X

n

+aX +b) = (−1)

n(n−1)/2

(n

n

b

n−1

+ (−1)

n−1

(n −1)

n−1

a

n

),

which is something Maple V doesn’t know (because it doesn’t understand symbols as ex-

ponents). For example,

disc(X

5

+aX +b) = 5

5

b

4

+ 4

4

a

5

.

Exercises

5-1 (*). For a ∈ Q, let G

a

be the Galois group of X

4

+X

3

+X

2

+X +a. Find integers

a

1

, a

2

, a

3

, a

4

such that i = j =⇒ G

a

i

is not isomorphic go G

a

j

.

5-2 (*). Prove that the rational solutions a, b ∈ Q of Pythagoras’s equation a

2

+ b

2

= 1

are of the form

a =

s

2

−t

2

s

2

+t

2

, b =

2st

s

2

+t

2

, s, t ∈ Q,

and deduce that any right triangle with integer sides has sides of length

d(m

2

−n

2

, 2mn, m

2

+n

2

)

for some integers d, m, and n (Hint: Apply Hilbert’s Theorem 90 to the extension Q[i]/Q.)

5-3 (*). Prove that a ﬁnite extension of Q can contain only ﬁnitely many roots of 1.

66 6 ALGEBRAIC CLOSURES

6 Algebraic closures

In this section, we prove that Zorn’s lemma implies that every ﬁeld F has an algebraic

closure Ω. Recall that if F is a subﬁeld C, then the algebraic closure of F in C is an

algebraic closure of F (1.46). If F is countable, then the existence of Ω can be proved as

in the ﬁnite ﬁeld case (4.22), namely, the set of monic irreducible polynomials in F[X] is

countable, and so we can list them f

1

, f

2

, . . .; deﬁne E

i

inductively by, E

0

= F, E

i

= a

splitting ﬁeld of f

i

over E

i−1

; then Ω =

¸

E

i

is an algebraic closure of F.

The difﬁculty in showing the existence of an algebraic closure of an arbitrary ﬁeld F is

in the set theory. Roughly speaking, we would like to take a union of a family of splitting

ﬁelds indexed by the monic irreducible polynomials in F[X], but we need to ﬁnd a way

of doing this that is allowed by the axioms of set theory. After reviewing the statement of

Zorn’s lemma, we sketch three solutions

22

to the problem.

Zorn’s lemma

DEFINITION 6.1. (a) A relation ≤ on a set S is a partial ordering if it reﬂexive, transitive,

and anti-symmetric (a ≤ b and b ≤ a =⇒ a = b).

(b) A partial ordering is a total ordering if, for all s, t ∈ T, either s ≤ t or t ≤ s.

(c) An upper bound for a subset T of a partially ordered set (S, ≤) is an element s ∈ S

such that t ≤ s for all t ∈ T.

(d) A maximal element of a partially ordered set S is an element s such that s ≤ s

t

=⇒

s = s

t

.

A partially ordered set need not have any maximal elements, for example, the set of

ﬁnite subsets of an inﬁnite set is partially ordered by inclusion, but it has no maximal

elements.

LEMMA 6.2 (ZORN’S). Let (S, ≤) be a nonempty partially ordered set for which every

totally ordered subset has an upper bound in S. Then S has a maximal element.

Zorn’s lemma

23

is equivalent to the Axiom of Choice, and hence independent of the

axioms of set theory.

REMARK 6.3. The set S of ﬁnite subsets of an inﬁnite set doesn’t contradict Zorn’s lemma,

because it contains totally ordered subsets with no upper bound in S.

The following proposition is a typical application of Zorn’s lemma — we shall use a *

to signal results that depend on Zorn’s lemma (equivalently, the Axiom of Choice).

PROPOSITION 6.4 (*). Every nonzero commutative ring A has a maximal ideal (meaning,

maximal among proper ideals).

22

There do exist naturally occurring ﬁelds, not contained in C, that are uncountable. For example, for any

ﬁeld F there is a ring F[[T]] of formal power series

¸

i≥0

aiT

i

, ai ∈ F, and its ﬁeld of fractions is uncountable

even if F is ﬁnite.

23

The following is quoted from A.J. Berrick and M.E. Keating, An Introduction to Rings and Modules,

2000: The name of the statement, although widely used (allegedly ﬁrst by Lefschetz), has attracted the attention

of historians (Campbell 1978). As a ‘maximum principle’, it was ﬁrst brought to prominence, and used for

algebraic purposes in Zorn 1935, apparently in ignorance of its previous usage in topology, most notably in

Kuratowski 1922. Zorn attributed to Artin the realization that the ‘lemma’ is in fact equivalent to the Axiom of

Choice (see Jech 1973). Zorn’s contribution was to observe that it is more suited to algebraic applications like

ours.

First proof of the existence of algebraic closures 67

PROOF. Let S be the set of all proper ideals in A, partially ordered by inclusion. If T is a

totally ordered set of ideals, then J =

¸

I∈T

I is again an ideal, and it is proper because if

1 ∈ J then 1 ∈ I for some I in T, and I would not be proper. Thus J is an upper bound for

T. Now Zorn’s lemma implies that S has a maximal element, which is a maximal ideal in

A.

2

First proof of the existence of algebraic closures

(Bourbaki, 1959, Chap. 5 '4.)

24

An F-algebra is a ring containing F as a subring. Let

(A

i

)

i∈I

be a family of commutative F-algebras, and deﬁne

¸

F

A

i

to be the quotient of the

F-vector space with basis

¸

i∈I

A

i

by the subspace generated by elements of the form:

(x

i

) + (y

i

) −(z

i

) with x

j

+y

j

= z

j

for one j ∈ I and x

i

= y

i

= z

i

for all i = j;

(x

i

) −a(y

i

) with x

j

= ay

j

for one j ∈ I and x

i

= y

i

for all i = j,

(Bourbaki, 1989, Chap. II, 3.9)

25

. It can be made into a commutative F-algebra in an

obvious fashion, and there are canonical homomorphisms A

i

→

¸

F

A

i

of F-algebras.

For each polynomial f ∈ F[X], choose a splitting ﬁeld E

f

, and let Ω = (

¸

F

E

f

)/M

where M is a maximal ideal in

¸

F

E

f

(whose existence is ensured by Zorn’s lemma).

Note that F ⊂

¸

F

E

f

and M ∩ F = 0. As Ω has no ideals other than (0) and Ω, and it

is a ﬁeld (see 1.2). The composite of the F-homomorphisms E

f

→

¸

F

E

f

→ Ω, being

a homomorphism of ﬁelds, is injective. Since f splits in E

f

, it must also split in the larger

ﬁeld Ω. The algebraic closure of F in Ω is therefore an algebraic closure of F (by 1.44).

Second proof of the existence of algebraic closures

(Jacobson 1964, p144.). After (4.22) we may assume F to be inﬁnite. This implies that

the cardinality of any ﬁeld algebraic over F is the same as that of F (ibid. p143). Choose

an uncountable set Ξ of cardinality greater than that of F, and identify F with a subset of

Ξ. Let S be the set triples (E, +, ) with E ⊂ Ξ and (+, ) a ﬁeld structure on E such that

(E, +, ) contains F as a subﬁeld and is algebraic over it. Write (E, +, ) ≤ (E

t

, +

t

,

t

)

if the ﬁrst is a subﬁeld of the second. Apply Zorn’s lemma to show that S has maximal

elements, and then show that a maximal element is algebraically closed. (See ibid. p144

for the details.)

Third proof of the existence of algebraic closures

(E. Artin, see Dummit and Foote 1991, 13.4). Consider the polynomial ring F[. . . , x

f

, . . .]

in a family of symbols x

f

indexed by the nonconstant monic polynomials f ∈ F[X]. If 1

lies in the ideal I in F[. . . , x

f

, . . .] generated by the polynomials f(x

f

), then

g

1

f

1

(x

f

1

) + +g

n

f

n

(x

fn

) = 1 (in F[. . . , x

f

, . . .])

for some g

i

∈ F[. . . , x

f

, . . .] and some nonconstant monic f

i

∈ F[X]. Let F

t

be an exten-

sion of F such that each f

i

, i = 1, . . . , n, has a root α

i

in F

t

. Under the F-homomorphism

24

Bourbaki, N.,

´

El´ ements de math´ ematique. I: Les structures fondamentales de l’analyse. Fascicule XI.

Livre II: Alg` ebre. Chapitre 4: Polynomes et fractions rationnelles. Chapitre 5: Corps commutatifs. Deuxi` eme

´ edition. Actualit´ es Scientiﬁques et Industrielles, No. 1102 Hermann, Paris 1959 iv+222 pp. (2 inserts). MR 30

#4751

25

Bourbaki, Nicolas. Algebra. I. Chapters 1–3. Translated from the French. Reprint of the 1974 edition.

Elements of Mathematics. Springer-Verlag, Berlin, 1989. xxiv+709 pp.

68 6 ALGEBRAIC CLOSURES

F[. . . , x

f

, . . .] →F

t

sending

x

f

i

→α

i

x

f

→0, f / ∈ ¦f

1

, . . . , f

n

¦

the above relation becomes 0 = 1. From this contradiction, we deduce that 1 does not

lie in I, and so Proposition 6.4 applied to F[. . . , x

f

, . . .]/I shows that I is contained in

a maximal ideal M of F[. . . , x

f

, . . .]. Let E

1

= F[. . . , x

f

, . . .]/M. Then E

1

is a ﬁeld

containing (a copy of) F in which every nonconstant polynomial in F[X] has at least one

root. Repeat the process starting with E

1

instead of F to obtain a ﬁeld E

2

. Continue in this

way to obtain a sequence of ﬁelds

F = E

0

⊂ E

1

⊂ E

2

⊂ ,

and let E =

¸

E

i

. Then E is algebraically closed, because the coefﬁcients of any noncon-

stant polynomial g ∈ E[X] lie in E

i

for some i, and then g has a root in E

i+1

. Therefore,

the algebraic closure of F in E is an algebraic closure of F (by 1.46).

26

(Non)uniqueness of algebraic closures

THEOREM 6.5 (*). Let Ω be an algebraic closure of F, and let E be an algebraic extension

of F. There exists an F-homomorphism E → Ω, and, if E is also an algebraic closure of

F, then every such homomorphism is an isomorphism.

PROOF. Suppose ﬁrst that E is countably generated over F, i.e., E = F[α

1

, ..., α

n

, . . .].

Then we can extend the inclusion map F →Ω to F[α

1

] (map α

1

to any root of its minimal

polynomial in Ω), then to F[α

1

, α

2

], and so on (see 2.2).

In the uncountable case, we use Zorn’s lemma. Let S be the set of pairs (M, ϕ

M

) with

M a ﬁeld F ⊂ M ⊂ E and ϕ

M

an F-homomorphism M → Ω. Write (M, ϕ

M

) ≤

(N, ϕ

N

) if M ⊂ N and ϕ

N

[M = ϕ

M

. This makes S into a partially ordered set. Let T be

a totally ordered subset of S. Then M

t

=

¸

M∈T

M is a subﬁeld of E, and we can deﬁne

a homomorphism ϕ

t

: M

t

→ Ω by requiring that ϕ

t

(x) = ϕ

M

(x) if x ∈ M. The pair

(M

t

, ϕ

t

) is an upper bound for T in S. Hence Zorn’s lemma gives us a maximal element

(M, ϕ) in S. Suppose that M = E. Then there exists an element α ∈ E, α / ∈ M. Since α

is algebraic over M, we can apply (2.2) to extend ϕ to M[α], contradicting the maximality

of M. Hence M = E, and the proof of the ﬁrst statement is complete.

If E is algebraically closed, then every polynomial f ∈ F[X] splits in E[X] and hence

in ϕ(E)[X]. Let α ∈ Ω, and let f(X) be the minimum polynomial of α. Then X − α is

a factor of f(X) in Ω[X], but, as we just observed, f(X) splits in ϕ(E)[X]. Because of

unique factorization, this implies that α ∈ ϕ(E).

2

The above proof is a typical application of Zorn’s lemma: once we know how to do

something in a ﬁnite (or countable) situation, Zorn’s lemma allows us to do it in general.

REMARK 6.6. Even for a ﬁnite ﬁeld F, there will exist uncountably many isomorphisms

from one algebraic closure to a second, none of which is to be preferred over any other.

Thus it is (uncountably) sloppy to say that the algebraic closure of F is unique. All one can

say is that, given two algebraic closures Ω, Ω

t

of F, then, thanks to Zorn’s lemma, there

exists an F-isomorphism Ω →Ω

t

.

26

In fact, E is algebraic over F. To see this, note that E1 is generated by algebraic elements over F, and

so is algebraic over F (apply 1.45). Similarly, E2 is algebraic over E1 and therefore also over F (see 1.31b).

Continuing in this way, we ﬁnd that every element of every Ei is algebraic over F.

Separable closures 69

Separable closures

Let Ω be a ﬁeld containing F, and let c be a set of intermediate ﬁelds F ⊂ E ⊂ Ω with the

following property:

(*) for any E

1

, E

2

∈ c, there exists an E ∈ c such that E

1

, E

2

⊂ E.

Then E(c) =

¸

E∈c

E is a subﬁeld of Ω (and we call

¸

E∈c

E a directed union), because

(*) implies that any ﬁnite set of elements of E(c) is contained in a common E ∈ c, and

therefore their product, sum, etc., also lie in E(c).

We apply this remark to the set of subﬁelds E of Ω that are ﬁnite and separable over F.

As the composite of any two such subﬁelds is again ﬁnite and separable over F (cf. 3.14),

we see that the union L of all such E is a subﬁeld of Ω. We call L the separable closure of

F in Ω — clearly, it is separable over F and every element of Ω separable over F lies in L.

Moreover, because a separable extension of a separable extension is separable, Ω is purely

inseparable over L.

DEFINITION 6.7. (a) Aﬁeld Ωis said to be separably closed if every nonconstant separable

polynomial in Ω[X] splits in Ω.

(b) A ﬁeld Ω is said to be a separable closure of a subﬁeld F if it is separable and

algebraic over F and it is separably closed.

THEOREM 6.8 (*). (a) Every ﬁeld has a separable closure.

(b) Let E be a separable algebraic extension of F, and let Ω be a separable algebraic

closure of F. There exists an F-homomorphism E → Ω, and, if E is also a separable

closure of F, then every such homomorphism is an isomorphism.

PROOF. Replace “polynomial” with “separable polynomial” in the proofs of the corre-

sponding theorems for algebraic closures. Alternatively, deﬁne Ωto be the separable closure

of F in an algebraic closure, and apply the preceding theorems.

2

70 7 INFINITE GALOIS EXTENSIONS

7 Inﬁnite Galois extensions

In this section, we make free use of Zorn’s lemma.

Topological groups

DEFINITION 7.1. A set G together with a group structure and a topology is a topological

group if the maps

(g, h) →gh: GG →G,

g →g

−1

: G →G

are both continuous.

Let a be an element of a topological group G. Then a

L

: G

g→ag

−−−→ G is continuous

because it is the composite of

G

g→(a,g)

−−−−−→ GG

(g,h)→gh

−−−−−−→ G.

In fact, it is a homeomorphism with inverse (a

−1

)

L

. Similarly a

R

: g → ga and g → g

−1

are both homeomorphisms. In particular, for any subgroup H of G, the coset aH of H is

open or closed if H is open or closed. As the complement of H in G is a union of such

cosets, this shows that H is closed if it is open, and it is open if it is closed and of ﬁnite

index.

Recall that a neighbourhood base for a point x of a topological space X is a set of

neighbourhoods ^ such that every open subset U of X containing x contains an N from

^.

PROPOSITION 7.2. Let G be a topological group, and let ^ be a neighbourhood base for

the identity element e of G. Then

27

(a) for all N

1

, N

2

∈ ^, there exists an N

t

∈ ^ such that e ∈ N

t

⊂ N

1

∩ N

2

;

(b) for all N ∈ ^, there exists an N

t

∈ ^ such that N

t

N

t

⊂ N;

(c) for all N ∈ ^, there exists an N

t

∈ ^ such that N

t

⊂ N

−1

;

(d) for all N ∈ ^ and all g ∈ G, there exists an N

t

∈ ^ such that N

t

⊂ gNg

−1

;

(e) for all g ∈ G, ¦gN [ N ∈ ^¦ is a neighbourhood base for g.

Conversely, if G is a group and ^ is a nonempty set of subsets of G satisfying (a,b,c,d),

then there is a (unique) topology on G for which (e) holds.

PROOF. If ^ is a neighbourhood base at e in a topological group G, then (b), (c), and (d)

are consequences of the continuity of (g, h) → gh, g → g

−1

, and h → ghg

−1

respec-

tively. Moreover, (a) is a consequence of the deﬁnitions and (e) of the fact that g

L

is a

homeomorphism.

Conversely, let ^ be a nonempty collection of subsets of a group G satisfying the

conditions (a)–(d). Note that (a) implies that e lies in all the N in ^. Deﬁne | to be

the collection of subsets U of G such that, for every g ∈ U, there exists an N ∈ ^

with gN ⊂ U. Clearly, the empty set and G are in |, and unions of sets in | are in

|. Let U

1

, U

2

∈ |, and let g ∈ U

1

∩ U

2

; by deﬁnition there exist N

1

, N

2

∈ ^ with

gN

1

, gN

2

⊂ U; on applying (a) we obtain an N

t

∈ ^ such that gN

t

⊂ U

1

∩ U

2

, which

27

For subsets S and S

of G, we set SS

= {ss

| s ∈ S, s

∈ S

}, and S

−1

= {s

−1

| s ∈ S}.

The Krull topology on the Galois group 71

shows that U

1

∩ U

2

∈ |. It follows that the elements of | are the open sets of a topology

on G (and, in fact, the unique topology for which (e) holds).

We next used (b) and (d) to show that (g, g

t

) → gg

t

is continuous. Note that the

sets g

1

N

1

g

2

N

2

form a neighbourhood base for (g

1

, g

2

) in G G. Therefore, given

an open U ⊂ G and a pair (g

1

, g

2

) such that g

1

g

2

∈ U, we have to ﬁnd N

1

, N

2

∈ ^

such that g

1

N

1

g

2

N

2

⊂ U. As U is open, there exists an N ∈ ^ such that g

1

g

2

N ⊂ U.

Apply (b) to obtain an N

t

such that N

t

N

t

⊂ N; then g

1

g

2

N

t

N

t

⊂ U. But g

1

g

2

N

t

N

t

=

g

1

(g

2

N

t

g

−1

2

)g

2

N

t

, and it remains to apply (d) to obtain an N

1

∈ ^ such that N

1

⊂

g

2

N

t

g

−1

2

.

Finally, we use (c) and (d) to show that g →g

−1

is continuous. Given an open U ⊂ G

and a g ∈ G such that g

−1

∈ U, we have to ﬁnd an N ∈ ^ such that gN ⊂ U

−1

. By

deﬁnition, there exists an N ∈ ^ such that g

−1

N ⊂ U. Now N

−1

g ⊂ U

−1

, and we use

(c) to obtain an N

t

∈ ^ such that N

t

g ⊂ U

−1

, and (d) to obtain an N

tt

∈ ^ such that

gN

tt

⊂ g(g

−1

N

t

g) ⊂ U

−1

.

2

The Krull topology on the Galois group

Recall (3.9) that a ﬁnite extension Ω of F is Galois over F if it is normal and separable, i.e.,

if every irreducible polynomial f ∈ F[X] having a root in Ω has deg f distinct roots in Ω.

Similarly, we deﬁne an algebraic extension Ω of F to be Galois over F if it is normal and

separable. Clearly, Ω is Galois over F if and only if it is a union of ﬁnite Galois extensions.

PROPOSITION 7.3. If Ω is Galois over F, then it is Galois over any intermediate ﬁeld M.

PROOF. Let f(X) be an irreducible polynomial in M[X] having a root a in Ω. The min-

imum polynomial g(X) of a over F splits into distinct degree-one factors in Ω[X]. As f

divides g (in M[X]), it also must split into distinct degree-one factors in Ω[X].

2

PROPOSITION 7.4. Let Ω be a Galois extension of F and let E be a subﬁeld of Ω contain-

ing F. Then every F-homomorphism E →Ω extends to an F-isomorphism Ω →Ω.

PROOF. The same Zorn’s lemma argument as in the proof of Theorem 6.5 shows that every

F-homomorphism E →Ω extends to an F-homomorphism α: Ω →Ω. Let a ∈ Ω, and let

f be its minimum polynomial over F. Then Ω contains exactly deg(f) roots of f, and so

therefore does α(Ω). Hence a ∈ α(Ω), which shows that α is surjective.

2

Let Ω be a Galois extension of F, and let G = Aut(Ω/F). For any ﬁnite subset S of

Ω, let

G(S) = ¦σ ∈ G [ σs = s for all s ∈ S¦.

PROPOSITION 7.5. There is a unique structure of a topological group on G for which the

sets G(S) form an open neighbourhood base of 1. For this topology, the sets G(S) with S

G-stable form a neighbourhood base of 1 consisting of open normal subgroups.

PROOF. We show that the collection of sets G(S) satisﬁes (a,b,c,d) of (7.2). It satisﬁes (a)

because G(S

1

) ∩ G(S

2

) = G(S

1

∪ S

2

). It satisﬁes (b) and (c) because each set G(S) is a

group. Let S be a ﬁnite subset of Ω. Then F(S) is a ﬁnite extension of F, and so there are

only ﬁnitely many F-homomorphisms F(S) → Ω. Since σS = τS if σ[F(S) = τ[F(S),

this shows that S =

¸

σ∈G

σS is ﬁnite. Now σS = S for all σ ∈ G, and it follows that

G(S) is normal in G. Therefore, σG(S)σ

−1

= G(S) ⊂ G(S), which proves (d). It also

proves the second statement.

2

72 7 INFINITE GALOIS EXTENSIONS

The topology on Aut(Ω/F) deﬁned in the proposition is called the Krull topology. We

write Gal(Ω/F) for Aut(Ω/F) endowed with the Krull topology, and call it the Galois

group of Ω/F.

PROPOSITION 7.6. Let Ω be Galois over F. For any intermediate ﬁeld E ﬁnite and Galois

over F, the map

σ →σ[E: Gal(Ω/F) →Gal(E/F)

is a continuous surjection (discrete topology on Gal(E/F)).

PROOF. Let σ ∈ Gal(E/F), and regard it as an F-homomorphism E → Ω. Then σ

extends to an F-isomorphism Ω → Ω (see 7.4), which shows that the map is surjective.

For any ﬁnite set S of generators of E over F, Gal(E/F) = G(S), which shows that the

inverse image of 1

Gal(E/F)

is open in G. By homogeneity, the same is true for any element

of Gal(E/F).

2

PROPOSITION 7.7. All Galois groups are compact and totally disconnected.

28

PROOF. Let G = Gal(Ω/F). We ﬁrst show that G is Hausdorff. If σ = τ, then σ

−1

τ =

1

G

, and so it moves some element of Ω, i.e., there exists an a ∈ Ω such that σ(a) =

τ(a). For any S containing a, σG(S) and τG(S) are disjoint because their elements act

differently on a. Hence they are disjoint open subsets of G containing σ and τ respectively.

We next show that G is compact. As we noted above, if S is a ﬁnite set stable under G,

then G(S) is a normal subgroup of G, and it has ﬁnite index because it is the kernel of

G →Sym(S).

Since every ﬁnite set is contained in a stable ﬁnite set, the argument in the last paragraph

shows that the map

G →

¸

S ﬁnite stable under G

G/G(S)

is injective. When we endow

¸

G/G(S) with the product topology, the induced topology

on G is that for which the G(S) form an open neighbourhood base of e, i.e., it is the Krull

topology. According to the Tychonoff theorem,

¸

G/G(S) is compact, and so it remains to

show that G is closed in the product. For each S

1

⊂ S

2

, let E(S

1

, S

2

) be the set of families

(σ

S

G(S))

S

in

¸

G/G(S) such that

σ

S

1

= σ

S

2

[S

1

.

This is closed in

¸

G/G(S) because it is the set on which two continuous maps

¸

G/G(S) →

G/G(S

1

) agree. Therefore,

¸

S

1

⊂S

2

E(S

1

, S

2

) is closed, but it also equals the image of G.

Finally, for each ﬁnite set S stable under G, G(S) is a subgroup that is open and hence

closed. Since

¸

G(S) = ¦1

G

¦, this shows that the connected component of G containing

1

G

is just ¦1

G

¦. By homogeneity, a similar statement is true for every element of G.

2

PROPOSITION 7.8. For any Galois extension Ω/F, Ω

Gal(Ω/F)

= F.

PROOF. Every element of Ω F lies in a ﬁnite Galois extension of F, and so this follows

from the surjectivity in Proposition 7.6.

2

28

Following Bourbaki, we require compact spaces to be Hausdorff. A topological space is totally discon-

nected if its connected components are the one-point sets.

The fundamental theorem of inﬁnite Galois theory 73

ASIDE 7.9. There is a converse to the proposition: every compact totally disconnected

group arises as the Galois group of some Galois extension of ﬁelds of characteristic zero

(Douady, A., Cohomologie des groupes compact totalement discontinus (d’apr` es J. Tate),

S´ eminaire Bourbaki 1959/60, no. 189).

The fundamental theorem of inﬁnite Galois theory

PROPOSITION 7.10. Let Ω be Galois over F, with Galois group G.

(a) The ﬁeld Ω is Galois over every subﬁeld M containing F; moreover, Gal(Ω/M) is

closed in G, and Ω

Gal(Ω/M)

= M.

(b) For every subgroup H of G, Gal(Ω/Ω

H

) is the closure of H.

PROOF. (a) The ﬁrst assertion was proved in (7.3). For each ﬁnite subset S ⊂ M, G(S) is

an open subgroup of G, and hence it is closed. But Gal(Ω/M) =

¸

S⊂M

G(S), and so it

also is closed. The ﬁnal statement follows from (7.8).

(b) Since Gal(Ω/Ω

H

) contains H and is closed, it certainly contains the closure H of

H. On the other hand, let σ ∈ G H. Then σG(S) ∩ H = ∅ for some ﬁnite subset S of

Ω which we may assume to be stable under G. Now σG(S) ∩H = ∅ implies σ / ∈ HG(S),

and so there exists an α ∈ F(S) that is ﬁxed by H but moved by σ. This shows that

σ / ∈ Gal(Ω/Ω

H

), as required.

2

THEOREM 7.11. Let Ω be Galois over F with Galois group G. The maps

H →Ω

H

, M →Gal(Ω/M)

are inverse bijections between the set of closed subgroups of G and the set of intermediate

ﬁelds between Ω and F:

¦closed subgroups of G¦ ↔¦intermediate ﬁelds F ⊂ M ⊂ Ω¦.

Moreover,

(a) the correspondence is inclusion-reversing: H

1

⊃ H

2

⇐⇒ Ω

H

1

⊃ Ω

H

2

;

(b) a closed subgroup H of Gis open if and only if Ω

H

has ﬁnite degree over F, in which

case (G: H) = [Ω

H

: F];

(c) σHσ

−1

↔σM, i.e., Ω

σHσ

−1

= σ(E

H

); Gal(E/σM) = σ Gal(E/M)σ

−1

;

(d) a closed subgroup H of G is normal if and only if Ω

H

is Galois over F, in which

case Gal(Ω/Ω

H

) · G/H.

PROOF. For the ﬁrst statement, we have to show that H →Ω

H

and M →Gal(Ω/M) are

inverse maps.

Let H be a closed subgroup of G. Then Ω is Galois over Ω

H

and Gal(Ω/Ω

H

) = H

(see 7.10).

Let M be an intermediate ﬁeld. Then Gal(Ω/M) is a closed subgroup of G and

Ω

Gal(Ω/M)

= M (see 7.10).

(a) We have the obvious implications:

H

1

⊃ H

2

=⇒ Ω

H

1

⊂ Ω

H

2

=⇒ Gal(Ω/Ω

H

1

) ⊃ Gal(Ω/Ω

H

2

).

But Gal(Ω/Ω

H

i

) = H

i

(see 7.10).

74 7 INFINITE GALOIS EXTENSIONS

(b) As we noted earlier, a closed subgroup of ﬁnite index in a topological group is

always open. Because G is compact, conversely an open subgroup of G is always of ﬁnite

index. Let H be such a subgroup. The map σ →σ[Ω

H

deﬁnes a bijection

G/H →Hom

F

(Ω

H

, Ω)

(apply 7.4) from which the statement follows.

(c) For τ ∈ Gand α ∈ Ω, τα = α ⇐⇒ στσ

−1

(σα) = σα. Therefore, Gal(Ω/σM) =

σ Gal(Ω/M)σ

−1

, and so σ Gal(Ω/M)σ

−1

↔σM.

(d) Let H ↔ M. It follows from (c) that H is normal if and only if M is stable under

the action of G. But M is stable under the action of G if and only it is a union of ﬁnite

extensions of F stable under G, i.e., of ﬁnite Galois extensions of G. We have already

observed that an extension is Galois if and only if it is a union of ﬁnite Galois extensions.

2

REMARK 7.12. As in the ﬁnite case (3.17), we can deduce the following statements.

(a) Let (M

i

)

i∈I

be a (possibly inﬁnite) family of intermediate ﬁelds, and let H

i

↔M

i

.

Let

¸

M

i

be the smallest ﬁeld containing all the M

i

; then because

¸

i∈I

H

i

is the largest

(closed) subgroup contained in all the H

i

,

Gal(E/

¸

M

i

) =

¸

i∈I

H

i

.

(b) Let M ↔ H. The largest (closed) normal subgroup contained in H is N =

¸

σ

σHσ

−1

(cf. GT 4.10), and so Ω

N

, which is the composite of the ﬁelds σM, is the

smallest normal extension of F containing M.

PROPOSITION 7.13. Let E and L be ﬁeld extensions of F con-

tained in some common ﬁeld. If E/F is Galois, then EL/L and

E/E ∩ L are Galois, and the map

σ →σ[E: Gal(EL/L) →Gal(E/E ∩ L)

is an isomorphism of topological groups.

EL

E L

=

E ∩ L

=

F

PROOF. The proof that the map is an isomorphism of groups (neglecting the topology) is

the same as in the ﬁnite case (3.18).

We next prove it is continuous. Let G

1

= Gal(EL/L) and let G

2

= Gal(E/E ∩ L).

For any ﬁnite set S of elements of E, the inverse image of G

2

(S) in G

1

is G

1

(S).

Finally, we prove that it is open. An open subgroup of Gal(EL/L) is closed (hence

compact) of ﬁnite index; therefore its image in Gal(E/E ∩ L) is compact (hence closed)

of ﬁnite index, and hence open.

2

COROLLARY 7.14. Let Ω be an algebraically closed ﬁeld containing F, and let E and L

be as in the proposition. If ρ: E → Ω and σ: L → Ω are F-homomorphisms such that

ρ[E∩L = σ[E∩L, then there exists an F-homomorphismτ : EL →Ω such that τ[E = ρ

and τ[L = σ.

The fundamental theorem of inﬁnite Galois theory 75

PROOF. According to (7.4), σ extends to an F-homomorphisms: EL →Ω. As s[E∩L =

ρ[E ∩ L, we can write s[E = ρ ◦ ε for some ε ∈ Gal(E/E ∩ L). According to the

proposition, there exists a unique e ∈ Gal(EL/L) such that e[E = ε. Deﬁne τ = s◦e

−1

.

2

EXAMPLE 7.15. Let Ω be an algebraic closure of a ﬁnite ﬁeld F

p

. Then G = Gal(Ω/F

p

)

contains a canonical Frobenius element, σ = (a → a

p

), and it is generated by it as a

topological group, i.e., G is the closure of 'σ`. Endow Z with the topology for which the

groups nZ, n ≥ 1, form a fundamental system of neighbourhoods of 0. Thus two integers

are close if their difference is divisible by a large integer.

As for any topological group, we can complete Z for this topology. A Cauchy sequence

in Z is a sequence (a

i

)

i≥1

, a

i

∈ Z, satisfying the following condition: for all n ≥ 1, there

exists an N such that a

i

≡ a

j

mod n for i, j > N. Call a Cauchy sequence in Z trivial

if a

i

→ 0 as i → ∞, i.e., if for all n ≥ 1, there exists an N such that a

i

≡ 0 mod n

for all i > N. The Cauchy sequences form a commutative group, and the trivial Cauchy

sequences form a subgroup. We deﬁne

´

Z to be the quotient of the ﬁrst group by the second.

It has a ring structure, and the map sending m ∈ Z to the constant sequence m, m, m, . . .

identiﬁes Z with a subgroup of

´

Z.

Let α ∈

´

Z be represented by the Cauchy sequence (a

i

). The restriction of σ to F

p

n has

order n. Therefore (σ[F

p

n)

a

i

is independent of i provided it is sufﬁciently large, and we can

deﬁne σ

α

∈ Gal(Ω/F

p

) to be such that, for each n, σ

α

[F

p

n = (σ[F

p

n)

a

i

for all i sufﬁciently

large (depending on n). The map α →σ

α

:

´

Z →Gal(Ω/F

p

) is an isomorphism.

The group

´

Z is uncountable. To most analysts, it is a little weird—its connected com-

ponents are one-point sets. To number theorists it will seem quite natural — the Chinese

remainder theorem implies that it is isomorphic to

¸

p prime

Z

p

where Z

p

is the ring of p-adic

integers.

EXAMPLE 7.16. Let Ω be the algebraic closure of Q in C; then Gal(Ω/Q) is one of the

most basic, and intractable, objects in mathematics. It is expected that every ﬁnite group

occurs as a quotient of it, and it certainly has S

n

as a quotient group for every n (and every

sporadic simple group, and every...) — cf. (5.34). We do understand Gal(F

ab

/F) where

F ⊂ C is a ﬁnite extension of Q and F

ab

is the union of all ﬁnite abelian extensions of F

contained in C. For example, Gal(Q

ab

/Q) ≈

´

Z

**. (This is abelian class ﬁeld theory — see
**

my notes Class Field Theory.)

ASIDE 7.17. A simple Galois correspondence is a system consisting of two partially or-

dered sets P and Q and order reversing maps f : P → Q and g : Q → P such that

gf(p) ≥ p for all p ∈ P and fg(q) ≥ q for all q ∈ Q. Then fgf = f, because

fg(fp) ≥ fp and gf(p) ≥ p implies f(gfp) ≤ f(p) for all p ∈ P. Similarly, gfg = g,

and it follows that f and g deﬁne a one-to-one correspondence between the sets g(Q) and

f(P).

From a Galois extension Ω of F we get a simple Galois correspondence by taking

P to be the set of subgroups of Gal(Ω/F) and Q to be the set of subsets of Ω, and by

setting f(H) = Ω

H

and g(S) = G(S). Thus, to prove the one-to-one correspondence

in the fundamental theorem, it sufﬁces to identify the closed subgroups as exactly those

in the image of g and the intermediate ﬁelds as exactly those in the image of f. This is

accomplished by (7.10).

76 7 INFINITE GALOIS EXTENSIONS

Galois groups as inverse limits

DEFINITION 7.18. A partial ordering ≤on a set I is said to be directed, and the pair (I, ≤)

is called a directed set, if for all i, j ∈ I there exists a k ∈ I such that i, j ≤ k.

DEFINITION 7.19. Let (I, ≤) be a directed set, and let C be a category (for example, the

category of groups and homomorphisms, or the category of topological groups and contin-

uous homomorphisms).

(a) An inverse system in C indexed by (I, ≤) is a family (A

i

)

i∈I

of objects of C together

with a family (p

j

i

: A

j

→A

i

)

i≤j

of morphisms such that p

i

i

= id

A

i

and p

j

i

◦ p

k

j

= p

k

i

all i ≤ j ≤ k.

(b) An object A of C together with a family (p

j

: A → A

j

)

j∈I

of morphisms satisfying

p

j

i

◦ p

j

= p

i

all i ≤ j is said to be an inverse limit of the system in (a) if it has the

following universal property: for any other object B and family (q

j

: B → A

j

) of

morphisms such p

j

i

◦ q

j

= q

i

all i ≤ j, there exists a unique morphism r : B → A

such that p

j

◦ r = q

j

for j,

B

..........

r

E

A

r

r

r

q

j

r

r

rj

e

e

e

e

e

e

e

e

q

i

d

d

d

p

j

A

j

©

p

j

i

A

i

p

i

c

Clearly, the inverse limit (if it exists), is uniquely determined by this condition up to a

unique isomorphism. We denote it lim

←−

(A

i

, p

j

i

), or just lim

←−

A

i

.

EXAMPLE 7.20. Let (G

i

, p

j

i

: G

j

→G

i

) be an inverse system of groups. Let

G = ¦(g

i

) ∈

¸

G

i

[ p

j

i

(g

j

) = g

i

all i ≤ j¦,

and let p

i

: G →G

i

be the projection map. Then p

j

i

◦ p

j

= p

i

is just the equation p

j

i

(g

j

) =

g

i

. Let (H, q

i

) be a second family such that p

j

i

◦ q

j

= q

i

. The image of the homomorphism

h →(q

i

(h)): H →

¸

G

i

is contained in G, and this is the unique homomorphism H → G carrying q

i

to p

i

. Hence

(G, p

i

) = lim

←−

(G

i

, p

j

i

).

EXAMPLE 7.21. Let (G

i

, p

j

i

: G

j

→ G

i

) be an inverse system of topological groups and

continuous homomorphisms. When endowed with the product topology,

¸

G

i

becomes a

topological group

G = ¦(g

i

) ∈

¸

G

i

[ p

j

i

(g

j

) = g

i

all i ≤ j¦,

and G becomes a topological subgroup with the subspace topology. The projection maps

p

i

are continuous. Let H be (H, q

i

) be a second family such that p

j

i

◦ q

j

= q

i

. The

homomorphism

h →(q

i

(h)): H →

¸

G

i

Nonopen subgroups of ﬁnite index 77

is continuous because its composites with projection maps are continuous (universal prop-

erty of the product). Therefore H → G is continuous, and this shows that (G, p

i

) =

lim

←−

(G

i

, p

j

i

).

EXAMPLE 7.22. Let (G

i

, p

j

i

: G

j

→G

i

) be an inverse system of ﬁnite groups, and regard

it as an inverse system of topological groups by giving each G

i

the discrete topology. A

topological group G arising as an inverse limit of such a system is said to be proﬁnite.

If (x

i

) / ∈ G, say p

j

0

i

0

(x

j

0

) = x

i

0

, then

G∩ ¦(g

j

) [ g

j

0

= x

j

0

, g

i

0

= x

i

0

¦ = ∅.

As the second set is an open neighbourhood of (x

i

), this shows that Gis closed in

¸

G

i

. By

Tychonoff’s theorem,

¸

G

i

is compact, and so G is also compact. The map p

i

: G →G

i

is

continuous, and its kernel U

i

is an open subgroup of ﬁnite index in G (hence also closed).

As

¸

U

i

= ¦e¦, the connected component of G containing e is just ¦e¦. By homogeneity,

the same is true for every point of G: the connected components of G are the one-point sets

— G is totally disconnected.

We have shown that a proﬁnite group is compact and totally disconnected, and it is an

exercise to prove the converse.

29

EXAMPLE 7.23. Let Ω be a Galois extension of F. The composite of two ﬁnite Galois

extensions of in Ω is again a ﬁnite Galois extension, and so the ﬁnite Galois subextensions

of Ω form a directed set I. For each E in I we have a ﬁnite group Gal(E/F), and for each

E ⊂ E

t

we have a restriction homomorphism p

E

E

: Gal(E

t

/F) → Gal(E/F). In this

way, we get an inverse system of ﬁnite groups (Gal(E/F), p

E

E

) indexed by I.

For each E, there is a restriction homomorphism p

E

: Gal(Ω/F) → Gal(E/F) and,

because of the universal property of inverse limits, these maps deﬁne a homomorphism

Gal(Ω/F) →lim

←−

Gal(E/F).

This map is an isomorphism of topological groups. This is a restatement of what we showed

in the proof of (7.7).

Nonopen subgroups of ﬁnite index

We apply Zorn’s lemma to construct a nonopen subgroup of ﬁnite index in Gal(Q

al

/Q).

30

LEMMA 7.24. Let V be an inﬁnite dimensional vector space. For all n ≥ 1, there exists a

subspace V

n

of V such that V/V

n

has dimension n.

PROOF. Zorn’s lemma shows that V contains maximal linearly independent subsets, and

then the usual argument shows that such a subset spans V , i.e., is a basis. Choose a basis,

and take V

n

to be the subspace spanned by the set obtained by omitting n elements from

the basis.

2

PROPOSITION 7.25. The group Gal(Q

al

/Q) has nonopen normal subgroups of index 2

n

for all n > 1.

29

More precisely, it is Exercise 3 of §7 of Chapter 3 of Bourbaki’s General Topology.

30

Contrast: “. . . it is not known, even when G = Gal(Q/Q), whether every subgroup of ﬁnite index in

G is open; this is one of a number of related unsolved problems, all of which appear to be very difﬁcult.”

Swinnerton-Dyer, H. P. F., A brief guide to algebraic number theory. Cambridge, 2001, p133.

78 7 INFINITE GALOIS EXTENSIONS

PROOF. Let E be the subﬁeld Q[

√

−1,

√

2, . . . ,

√

p, . . .] of C. For each prime p,

Gal(Q[

√

−1,

√

2, . . . ,

√

p]/Q)

is a product of copies of Z/2Z indexed by the set ¦primes ≤ p¦ ∪ ¦∞¦ (see 5.27 and

Exercise 2-1), and so

Gal(E/Q) = lim

←−

Gal(Q[

√

−1,

√

2, . . . ,

√

p]/Q)

is a direct product of copies of Z/2Z indexed by the primes l of Q (including l = ∞)

endowed with the product topology. Let G = Gal(E/Q), and let

H = ¦(a

l

) ∈ G [ a

l

= 0 for all but ﬁnitely many l¦.

This is a subgroup of G (in fact, it is a direct sum of copies of Z/2Z indexed by the primes

of Q), and it is dense in G because

31

clearly every open subset of G contains an element of

H. We can regard G/H as vector space over F

2

and apply the lemma to obtain subgroups

G

n

of index 2

n

in G containing H. If G

n

is open in G, then it is closed, which contradicts

the fact that H is dense. Therefore, G

n

is not open, and its inverse image in Gal(Q

al

/Q) is

the desired subgroup.

2

31

Better, let (a

l

) ∈ G; then the sequence

(a∞, 0, 0, 0, . . .), (a∞, a2, 0, 0, . . .), (a∞, a2, a3, 0, . . .), . . .

converges to (a

l

).

79

8 Transcendental extensions

In this section we consider ﬁelds Ω ⊃ F with Ω much bigger than F. For example, we

could have C ⊃ Q.

Algebraic independence

Elements α

1

, ..., α

n

of Ω give rise to an F-homomorphism

f →f(α

1

, ..., α

n

): F[X

1

, . . . , X

n

] →Ω.

If the kernel of this homomorphism is zero, then the α

i

are said to be algebraically inde-

pendent over F, and otherwise, they are algebraically dependent over F. Thus, the α

i

are algebraically dependent over F if there exists a nonzero polynomial f(X

1

, ..., X

n

) ∈

F[X

1

, ..., X

n

] such that f(α

1

, ..., α

n

) = 0, and they are algebraically independent if

a

i

1

,...,in

∈ F,

¸

a

i

1

,...,in

α

i

1

1

...α

in

n

= 0 =⇒ a

i

1

,...,in

= 0 all i

1

, ..., i

n

.

Note the similarity with linear independence. In fact, if f is required to be homogeneous of

degree 1, then the deﬁnition becomes that of linear independence.

EXAMPLE 8.1. (a) A single element α is algebraically independent over F if and only if it

is transcendental over F.

(b) The complex numbers π and e are almost certainly algebraically independent over

Q, but this has not been proved.

An inﬁnite set A is algebraically independent over F if every ﬁnite subset of A is

algebraically independent; otherwise, it is algebraically dependent over F.

REMARK 8.2. If α

1

, ..., α

n

are algebraically independent over F, then

f(X

1

, ..., X

n

) →f(α

1

, ..., α

n

): F[X

1

, ..., X

n

] →F[α

1

, ..., α

n

]

is an injection, and hence an isomorphism. This isomorphism then extends to the ﬁelds of

fractions,

X

i

→α

i

: F(X

1

, ..., X

n

) →F(α

1

, ..., α

n

)

In this case, F(α

1

, ..., α

n

) is called a pure transcendental extension of F. The polynomial

f(X) = X

n

−α

1

X

n−1

+ + (−1)

n

α

n

has Galois group S

n

over F(α

1

, ..., α

n

) (5.33).

LEMMA 8.3. Let γ ∈ Ω and let A ⊂ Ω. The following conditions are equivalent:

(a) γ is algebraic over F(A);

(b) there exist β

1

, . . . , β

n

∈ F(A) such that γ

n

+β

1

γ

n−1

+ +β

n

= 0;

(c) there exist β

0

, β

1

, . . . , β

n

∈ F[A], not all 0, such that β

0

γ

n

+β

1

γ

n−1

+ +β

n

= 0;

(d) there exists an f(X

1

, . . . , X

m

, Y ) ∈ F[X

1

. . . , X

m

, Y ] and α

1

, . . . , α

m

∈ A such

that f(α

1

, . . . , α

m

, Y ) = 0 but f(α

1

, . . . , α

m

, γ) = 0.

80 8 TRANSCENDENTAL EXTENSIONS

PROOF. (a) =⇒(b) =⇒(c) =⇒(a) are obvious.

(d) =⇒(c). Write f(X

1

, . . . , X

m

, Y ) as a polynomial in Y with coefﬁcients in the ring

F[X

1

, . . . , X

m

],

f(X

1

, . . . , X

m

, Y ) =

¸

f

i

(X

1

, . . . , X

m

)Y

n−i

.

Then (c) holds with β

i

= f

i

(α

1

, . . . , α

m

).

(c) =⇒(d). The β

i

in (c) can be expressed as polynomials in a ﬁnite number of elements

α

1

, . . . , α

m

of A, say, β

i

= f

i

(α

1

, . . . , α

m

) with f

i

∈ F[X

1

, . . . , X

m

]. Then (d) holds with

f =

¸

f

i

(X

1

, . . . , X

m

)Y

n−i

.

2

DEFINITION 8.4. When γ satisﬁes the equivalent conditions of Lemma 8.3, it is said to be

algebraically dependent on A (over F). A set B is algebraically dependent on A if each

element of B is algebraically dependent on A.

The theory in the remainder of this section is logically very similar to a part of linear

algebra. It is useful to keep the following correspondences in mind:

Linear algebra Transcendence

linearly independent algebraically independent

A ⊂ span(B) A algebraically dependent on B

basis transcendence basis

dimension transcendence degree

Transcendence bases

THEOREM 8.5 (FUNDAMENTAL RESULT). Let A = ¦α

1

, ..., α

m

¦ and B = ¦β

1

, ..., β

n

¦

be two subsets of Ω. Assume

(a) A is algebraically independent (over F);

(b) A is algebraically dependent on B (over F).

Then m ≤ n.

We ﬁrst prove two lemmas.

LEMMA 8.6 (THE EXCHANGE PROPERTY). Let ¦α

1

, ..., α

m

¦ be a subset of Ω; if β is al-

gebraically dependent on ¦α

1

, ..., α

m

¦ but not on ¦α

1

, ..., α

m−1

¦, then α

m

is algebraically

dependent on ¦α

1

, ..., α

m−1

, β¦.

PROOF. Because β is algebraically dependent on ¦α

1

, . . . , α

m

¦, there exists a polynomial

f(X

1

, ..., X

m

, Y ) with coefﬁcients in F such that

f(α

1

, ..., α

m

, Y ) = 0, f(α

1

, ..., α

m

, β) = 0.

Write f as a polynomial in X

m

,

f(X

1

, ..., X

m

, Y ) =

¸

i

a

i

(X

1

, ..., X

m−1

, Y )X

n−i

m

,

and observe that, because f(α

1

, . . . , α

m

, Y ) = 0, at least one of the polynomials

a

i

(α

1

, ..., α

m−1

, Y ),

Transcendence bases 81

say a

i

0

, is not the zero polynomial. Because β is not algebraically dependent on

¦α

1

, ..., α

m−1

¦,

a

i

0

(α

1

, ..., α

m−1

, β) = 0. Therefore, f(α

1

, ..., α

m−1

, X

m

, β) = 0. Since f(α

1

, ..., α

m

, β) =

0, this shows that α

m

is algebraically dependent on ¦α

1

, ..., α

m−1

, β¦.

2

LEMMA 8.7 (TRANSITIVITY OF ALGEBRAIC DEPENDENCE). If C is algebraically depen-

dent on B, and B is algebraically dependent on A, then C is algebraically dependent on

A.

PROOF. The argument in the proof of Proposition 1.44 shows that if γ is algebraic over a

ﬁeld E which is algebraic over a ﬁeld F, then γ is algebraic over F (if a

1

, . . . , a

n

are the

coefﬁcients of the minimum polynomial of γ over E, then the ﬁeld F[a

1

, . . . , a

n

, γ] has

ﬁnite degree over F). Apply this with E = F(A∪ B) and F = F(A).

2

PROOF (OF THEOREM 8.5). Let k be the number of elements that A and B have in com-

mon. If k = m, then A ⊂ B, and certainly m ≤ n. Suppose that k < m, and write B =

¦α

1

, ..., α

k

, β

k+1

, ..., β

n

¦. Since α

k+1

is algebraically dependent on ¦α

1

, ..., α

k

, β

k+1

, ..., β

n

¦

but not on ¦α

1

, ..., α

k

¦, there will be a β

j

, k + 1 ≤ j ≤ n, such that α

k+1

is algebraically

dependent on ¦α

1

, ..., α

k

, β

k+1

, ..., β

j

¦ but not

¦α

1

, ..., α

k

, β

k+1

, ..., β

j−1

¦.

The exchange lemma then shows that β

j

is algebraically dependent on

B

1

df

= B ∪ ¦α

k+1

¦ ¦β

j

¦.

Therefore B is algebraically dependent on B

1

, and so A is algebraically dependent on B

1

(by 8.7). If k + 1 < m, repeat the argument with A and B

1

. Eventually we’ll achieve

k = m, and m ≤ n.

2

DEFINITION 8.8. A transcendence basis for Ω over F is an algebraically independent set

A such that Ω is algebraic over F(A).

LEMMA 8.9. If Ω is algebraic over F(A), and A is minimal among subsets of Ω with this

property, then it is a transcendence basis for Ω over F.

PROOF. If A is not algebraically independent, then there is an α ∈ S that is algebraically

dependent on S ¦α¦. It follows from Lemma 8.7 that Ω is algebraic over F(A¦α¦).

2

THEOREM 8.10. If there is a ﬁnite subset A ⊂ Ω such that Ω is algebraic over F(A), then

Ω has a ﬁnite transcendence basis over F. Moreover, every transcendence basis is ﬁnite,

and they all have the same number of elements.

PROOF. In fact, any minimal subset A

t

of A such that Ω is algebraic over F(A

t

) will be a

transcendence basis. The second statement follows from Theorem 8.5.

2

LEMMA 8.11. Suppose that A is algebraically independent, but that A ∪ ¦β¦ is alge-

braically dependent. Then β is algebraic over F(A).

82 8 TRANSCENDENTAL EXTENSIONS

PROOF. The hypothesis is that there exists a nonzero polynomial

f(X

1

, ..., X

n

, Y ) ∈ F[X

1

, ..., X

n

, Y ]

such that f(α

1

, ..., α

n

, β) = 0, some distinct α

1

, ..., α

n

∈ A. Because A is algebraically

independent, Y does occur in f. Therefore

f = g

0

Y

m

+g

1

Y

m−1

+ +g

m

, g

i

∈ F[X

1

, ..., X

n

], g

0

= 0, m ≥ 1.

As g

0

= 0 and the α

i

are algebraically independent, g

0

(α

1

, ..., α

n

) = 0. Because β is a

root of

f = g

0

(α

1

, ..., α

n

)X

m

+g

1

(α

1

, ..., α

n

)X

m−1

+ +g

m

(α

1

, ..., α

n

),

it is algebraic over F(α

1

, ..., α

n

) ⊂ F(A).

2

PROPOSITION 8.12. Every maximal algebraically independent subset of Ω is a transcen-

dence basis for Ω over F.

PROOF. We have to prove that Ω is algebraic over F(A) if A is maximal among alge-

braically independent subsets. But the maximality implies that, for every β ∈ Ω A,

A∪¦β¦ is algebraically dependent, and so the lemma shows that β is algebraic over F(A).

2

Recall that (except in '7), we use an asterisk to signal a result depending on Zorn’s

lemma.

THEOREM 8.13 (*). Every algebraically independent subset of Ω is contained in a tran-

scendence basis for Ω over F; in particular, transcendence bases exist.

PROOF. Let S be the set of algebraically independent subsets of Ω containing the given

set. We can partially order it by inclusion. Let T be a totally ordered subset of S, and let

B =

¸

¦A [ A ∈ T¦. I claim that B ∈ S, i.e., that B is algebraically independent. If not,

there exists a ﬁnite subset B

t

of B that is not algebraically independent. But such a subset

will be contained in one of the sets in T, which is a contradiction. Now Zorn’s lemma shows

that there exists a maximal algebraically independent containing S, which Proposition 8.12

shows to be a transcendence basis for Ω over F.

2

It is possible to show that any two (possibly inﬁnite) transcendence bases for Ω over F

have the same cardinality. The cardinality of a transcendence basis for Ω over F is called

the transcendence degree of Ω over F. For example, the pure transcendental extension

F(X

1

, . . . , X

n

) has transcendence degree n over F.

EXAMPLE 8.14. Let p

1

, . . . , p

n

be the elementary symmetric polynomials in X

1

, . . . , X

n

.

The ﬁeld F(X

1

, . . . , X

n

) is algebraic over F(p

1

, . . . , p

n

), and so ¦p

1

, p

2

, . . . , p

n

¦ contains

a transcendence basis for F(X

1

, . . . , X

n

). Because F(X

1

, . . . , X

n

) has transcendence de-

gree n, the p

i

’s must themselves be a transcendence basis.

EXAMPLE 8.15. Let Ω be the ﬁeld of meromorphic functions on a compact complex man-

ifold M.

(a) The only meromorphic functions on the Riemann sphere are the rational functions

in z. Hence, in this case, Ω is a pure transcendental extension of C of transcendence degree

1.

L¨ uroth’s theorem 83

(b) If M is a Riemann surface, then the transcendence degree of Ω over C is 1, and Ω

is a pure transcendental extension of C ⇐⇒ M is isomorphic to the Riemann sphere

(c) If M has complex dimension n, then the transcendence degree is ≤ n, with equality

holding if M is embeddable in some projective space.

PROPOSITION 8.16. Any two algebraically closed ﬁelds with the same transcendence de-

gree over F are F-isomorphic.

PROOF. Choose transcendence bases A and A

t

for the two ﬁelds. By assumption, there

exists a bijection A → A

t

, which extends uniquely to an F-isomorphism F[A] → F[A

t

],

and hence to an F-isomorphism of the ﬁelds of fractions F(A) →F(A

t

). Use this isomor-

phism to identify F(A) with F(A

t

). Then the two ﬁelds in question are algebraic closures

of the same ﬁeld, and hence are isomorphic (Theorem 6.5).

2

REMARK 8.17. Any two algebraically closed ﬁelds with the same uncountable cardinality

and the same characteristic are isomorphic. The idea of the proof is as follows. Let F and

F

t

be the prime subﬁelds of Ω and Ω

t

; we can identify F with F

t

. Then show that when Ω

is uncountable, the cardinality of Ω is the same as the cardinality of a transcendence basis

over F. Finally, apply the proposition.

REMARK 8.18. What are the automorphisms of C? There are only two continuous auto-

morphisms (cf. Exercise A-8 and solution). If we assume Zorn’s lemma, then it is easy to

construct many: choose any transcendence basis A for C over Q, and choose any permu-

tation α of A; then α deﬁnes an isomorphism Q(A) → Q(A) that can be extended to an

automorphism of C. Without Zorn’s lemma, there are only two, because the noncontinuous

automorphisms are nonmeasurable,

32

and it is known that the Zorn’s lemma is required to

construct nonmeasurable functions.

33

L¨ uroth’s theorem

THEOREM 8.19 (L¨ UROTH’S THEOREM). Any subﬁeld E of F(X) containing F but not

equal to F is a pure transcendental extension of F.

PROOF. Jacobson 1964, IV 4, p157.

2

REMARK 8.20. This fails when there is more than one variable — see Zariski’s example

(footnote to Remark 5.5) and Swan’s example (Remark 5.34). The best true statement is

the following: if [F(X, Y ) : E] < ∞and F is algebraically closed of characteristic zero,

then E is a pure transcendental extension of F (Theorem of Zariski, 1958).

Separating transcendence bases

Let E ⊃ F be ﬁelds with E ﬁnitely generated over F. A subset ¦x

1

, . . . , x

d

¦ of E is a

separating transcendence basis for E/F if it is algebraically independent over F and E is

a ﬁnite separable extension of F(x

1

, . . . , x

d

).

32

A fairly elementary theorem of G. Mackey says that measurable homomorphisms of Lie groups are con-

tinuous (see David Witte Morris, Introduction to Arithmetic Groups, http://people.uleth.ca/˜dave.morris/, Ap-

pendix I.C).

33

“We show that the existence of a non-Lebesgue measurable set cannot be proved in Zermelo-Frankel set

theory (ZF) if use of the axiom of choice is disallowed...” R. Solovay, Ann. of Math., 92 (1970), 1–56.

84 8 TRANSCENDENTAL EXTENSIONS

THEOREM 8.21. If F is perfect, then every ﬁnitely generated extension E of F admits a

separating transcendence basis F.

PROOF. If F has characteristic zero, then every transcendence basis is separating, and so

the statement becomes that of (8.10). Thus, we may assume F has characteristic p = 0.

Because F is perfect, every polynomial in X

p

1

, . . . , X

p

n

with coefﬁcients in F is a p

th

power

in F[X

1

, . . . , X

n

]:

¸

a

i

1

in

X

i

1

p

1

. . . X

inp

n

=

¸

a

1

p

i

1

in

X

i

1

1

. . . X

in

n

p

.

Let E = F(x

1

, . . . , x

n

), and assume n > d + 1 where d is the transcendence de-

gree of E over F. After renumbering, we may suppose that x

1

, . . . , x

d

are algebraically

independent (8.9). Then f(x

1

, . . . , x

d+1

) = 0 for some nonzero irreducible polynomial

f(X

1

, . . . , X

d+1

) with coefﬁcients in F. Not all ∂f/∂X

i

are zero, for otherwise f would

be a polynomial in X

p

1

, . . . , X

p

d+1

, which implies that it is a p

th

power. After renumber-

ing, we may suppose that ∂f/∂X

d+1

= 0. Then F(x

1

, . . . , x

d+1

, x

d+2

) is algebraic over

F(x

1

, . . . , x

d

) and x

d+1

is separable over F(x

1

, . . . , x

d

), and so, by the primitive ele-

ment theorem (5.1), there is an element y such that F(x

1

, . . . , x

d+2

) = F(x

1

, . . . , x

d

, y).

Thus E is generated by n − 1 elements (as a ﬁeld containing F). After repeating the

process, possibly several times, we will have E = F(z

1

, . . . , z

d+1

) with z

d+1

separable

over F(z

1

, . . . , z

d

).

2

ASIDE 8.22. In fact, we showed that E admits a separating transcendence basis with d+1

elements where d is the transcendence degree. This has the following geometric interpreta-

tion: every irreducible algebraic variety of dimension d over a perfect ﬁeld F is birationally

equivalent with a hypersurface H in A

d+1

for which the projection (a

1

, . . . , a

d+1

) →

(a

1

, . . . , a

d

) realizes F(H) as a ﬁnite separable extension of F(A

d

) (see my notes on Al-

gebraic Geometry).

Transcendental Galois theory

THEOREM 8.23 (*). Let Ω be a separably closed ﬁeld and let F be a perfect subﬁeld of Ω.

If α ∈ Ω is ﬁxed by all F-automorphisms of Ω, then α ∈ F, i.e., Ω

G

= F.

PROOF. Let α ∈ Ω F. If α is transcendental over F, then it is part of a transcendence

basis A for Ω over F (see 8.13). Choose an automorphism σ of A such that σ(α) = α.

Then σ extends to an F-automorphism of F(A), which extends to an F-automorphism of

the separable closure Ω of F(A) (see 6.8).

If α is algebraic over F, then by inﬁnite Galois theory (7.8) there exists an F-automorphism

σ of the separable closure of F in Ω such that σ(α) = α. As before, σ can be extended to

an F-automorphism of Ω.

2

Let Ω ⊃ F be ﬁelds and let G = Aut(Ω/F). For any ﬁnite subset S of Ω, let

G(S) = ¦σ ∈ G [ σs = s for all s ∈ S¦.

Then, as in '7, the subgroups G(S) of G form a neighbourhood base for a unique topology

on G, which we again call the Krull topology. The same argument as in '7 shows that this

topology is Hausdorff (but it is not necessarily compact).

Transcendental Galois theory 85

THEOREM 8.24. Let Ω ⊃ F be ﬁelds such that Ω

G

= F, G = Aut(Ω/F).

(a) For every ﬁnite extension E of F in Ω, Ω

Aut(Ω/E)

= E.

(b) The maps

H →Ω

H

, M →Aut(Ω/M) (∗)

are inverse bijections between the set of compact subgroups of Gand the set of intermediate

ﬁelds over which Ω is Galois (possibly inﬁnite):

¦compact subgroups of G¦ ↔¦ﬁelds M such that F ⊂ M

Galois

⊂ Ω¦.

(c) If there exists an M ﬁnitely generated over F such that Ω is Galois over M, then G

is locally compact, and under (∗):

¦open compact subgroups of G¦

1:1

↔¦ﬁelds M such that F

ﬁntely generated

⊂ M

Galois

⊂ Ω¦.

(d) Let H be a subgroup of G, and let M = Ω

H

. Then the algebraic closure M

1

of M

is Galois over M. If moreover H = Aut(Ω/M), then Aut(Ω/M

1

) is a normal subgroup

of H, and σ → σ[M

1

maps H/ Aut(Ω/M

1

) isomorphically onto a dense subgroup of

Aut(M

1

/M).

PROOF. See 6.3 of Shimura, Goro., Introduction to the arithmetic theory of automorphic

functions. Princeton, 1971.

2

86 A REVIEW EXERCISES

A Review exercises

A-1. Let p be a prime number, and let m and n be positive integers.

(a) Give necessary and sufﬁcient conditions on m and n for F

p

n to have a subﬁeld iso-

morphic with F

p

m. Prove your answer.

(b) If there is such a subﬁeld, how many subﬁelds isomorphic with F

p

m are there, and

why?

A-2. Show that the Galois group of the splitting ﬁeld F of X

3

− 7 over Q is isomorphic

to S

3

, and exhibit the ﬁelds between Q and F. Which of the ﬁelds between Q and F are

normal over Q?

A-3. Prove that the two ﬁelds Q[

√

7] and Q[

√

11] are not isomorphic.

A-4. (a) Prove that the multiplicative group of all nonzero elements in a ﬁnite ﬁeld is

cyclic.

(b) Construct explicitly a ﬁeld of order 9, and exhibit a generator for its multiplicative

group.

A-5. Let X be transcendental over a ﬁeld F, and let E be a subﬁeld of F(X) properly

containing F. Prove that X is algebraic over E.

A-6. Prove as directly as you can that if ζ is a primitive p

th

root of 1, p prime, then the

Galois group of Q[ζ] over Q is cyclic of order p −1.

A-7. Let G be the Galois group of the polynomial X

5

−2 over Q.

(a) Determine the order of G.

(b) Determine whether G is abelian.

(c) Determine whether G is solvable.

A-8. (a) Show that every ﬁeld homomorphism from R to R is bijective.

(b) Prove that C is isomorphic to inﬁnitely many different subﬁelds of itself.

A-9. Let F be a ﬁeld with 16 elements. How many roots in F does each of the following

polynomials have? X

3

−1; X

4

−1; X

15

−1; X

17

−1.

A-10. Find the degree of a splitting ﬁeld of the polynomial (X

3

−5)(X

3

−7) over Q.

A-11. Find the Galois group of the polynomial X

6

−5 over each of the ﬁelds Q and R.

A-12. The coefﬁcients of a polynomial f(X) are algebraic over a ﬁeld F. Show that f(X)

divides some nonzero polynomial g(X) with coefﬁcients in F.

A-13. Let f(X) be a polynomial in F[X] of degree n, and let E be a splitting ﬁeld of f.

Show that [E : F] divides n!.

A-14. Find a primitive element for the ﬁeld Q[

√

3,

√

7] over Q, i.e., an element such that

Q[

√

3,

√

7] = Q[α].

A-15. Let G be the Galois group of (X

4

−2)(X

3

−5) over Q.

87

(a) Give a set of generators for G, as well as a set of deﬁning relations.

(b) What is the structure of G as an abstract group (is it cyclic, dihedral, alternating,

symmetric, etc.)?

A-16. Let F be a ﬁnite ﬁeld of characteristic = 2. Prove that X

2

= −1 has a solution in

F if and only if #F ≡ 1 mod 4.

A-17. Let E be the splitting ﬁeld over Q of (X

2

−2)(X

2

−5)(X

2

−7). Find an element

α in E such that E = Q[α]. (You must prove that E = Q[α].)

A-18. Let E be a Galois extension of F with Galois group S

n

, n > 1 not prime. Let H

1

be the subgroup of S

n

of elements ﬁxing 1, and let H

2

be the subgroup generated by the

cycle (123 . . . n). Let E

i

= E

H

i

, i = 1, 2. Find the degrees of E

1

, E

2

, E

1

∩E

2

, and E

1

E

2

over F. Show that there exists a ﬁeld M such that F ⊂ M ⊂ E

2

, M = F, M = E

2

, but

that no such ﬁeld exists for E

1

.

A-19. Let ζ be a primitive 12

th

root of 1 over Q. Howmany ﬁelds are there strictly between

Q[ζ

3

] and Q[ζ].

A-20. For the polynomial X

3

− 3, ﬁnd explicitly its splitting ﬁeld over Q and elements

that generate its Galois group.

A-21. Let E = Q[ζ], ζ

5

= 1, ζ = 1. Show that i / ∈ E, and that if L = E[i], then −1 is a

norm from L to E. Here i =

√

−1.

A-22. Let E be an extension ﬁeld of F, and let Ω be an algebraic closure of E. Let

σ

1

, . . . , σ

n

be distinct F-isomorphisms E →Ω.

(a) Show that σ

1

, . . . , σ

n

are linearly dependent over Ω.

(b) Show that [E : F] ≥ m.

(c) Let F have characteristic p > 0, and let L be a subﬁeld of Ω containing E and

such that a

p

∈ E for all a ∈ L. Show that each σ

i

has a unique extension to a

homomorphism σ

t

i

: L →Ω.

A-23. Identify the Galois group of the splitting ﬁeld K of X

4

−3 over Q. Determine the

number of quadratic subﬁelds.

A-24. Let F be a subﬁeld of a ﬁnite ﬁeld E. Prove that the trace map T = Tr

E/F

and the

norm map N = Nm

E/F

of E over F both map E onto F. (You may quote basic properties

of ﬁnite ﬁelds and the trace and norm.)

A-25. Prove or disprove by counterexample.

(a) If L/K is an extension of ﬁelds of degree 2, then there is an automorphism σ of L

such that K is the ﬁxed ﬁeld of σ.

(b) The same as (a) except that L is also given to be ﬁnite.

A-26. A ﬁnite Galois extension L of a ﬁeld K has degree 8100. Show that there is a ﬁeld

F with K ⊂ F ⊂ L such that [F : K] = 100.

A-27. An algebraic extension L of a ﬁeld K of characteristic 0 is generated by an element

θ that is a root of both of the polynomials X

3

− 1 and X

4

+ X

2

+ 1. Given that L = K,

ﬁnd the minimum polynomial of θ.

88 A REVIEW EXERCISES

A-28. Let F/Q be a Galois extension of degree 3

n

, n ≥ 1. Prove that there is a chain of

ﬁelds

Q = F

0

⊂ F

1

⊂ F

n

= F

such that for every i, 0 ≤ i ≤ n −1, [F

i+1

: F

i

] = 3.

A-29. Let L be the splitting ﬁeld over Q of an equation of degree 5 with distinct roots.

Suppose that L has an automorphism that ﬁxes three of these roots while interchanging the

other two and also an automorphism α = 1 of order 5.

(a) Prove that the group of automorphisms of L is the symmetric group on 5 elements.

(b) How many proper subﬁelds of L are normal extensions of Q? For each such ﬁeld F,

what is [F : Q]?

A-30. If L/K is a separable algebraic ﬁeld extension of ﬁnite degree d, show that the

number of ﬁelds between K and L is at most 2

d!

.

A-31. Let K be the splitting ﬁeld over Qof X

5

−1. Describe the Galois group Gal(K/Q)

of K over Q, and show that K has exactly one subﬁeld of degree 2 over Q, namely, Q[ζ +

ζ

4

], ζ = 1 a root of X

5

− 1. Find the minimum polynomial of ζ + ζ

4

over Q. Find

Gal(L/Q) when L is the splitting ﬁeld over Q of

(a) (X

2

−5)(X

5

−1);

(b) (X

2

+ 3)(X

5

−1).

A-32. Let Ω

1

and Ω

2

be algebraically closed ﬁelds of transcendence degree 5 over Q, and

let α : Ω

1

→Ω

2

be a homomorphism (in particular, α(1) = 1). Show that α is a bijection.

(State carefully any theorems you use.)

A-33. Find the group of Q-automorphisms of the ﬁeld k = Q[

√

−3,

√

−2].

A-34. Prove that the polynomial f(X) = X

3

− 5 is irreducible over the ﬁeld Q[

√

7]. If

L is the splitting ﬁeld of f(X) over Q[

√

7], prove that the Galois group of L/Q[

√

7] is

isomorphic to S

3

. Prove that there must exist a subﬁeld K of L such that the Galois group

of L/K is cyclic of order 3.

A-35. Identify the Galois group Gof the polynomial f(X) = X

5

−6X

4

+3 over F, when

(a) F = Q and when (b) F = F

2

. In each case, if E is the splitting ﬁeld of f(X) over F,

determine how many ﬁelds K there are such that E ⊃ K ⊃ F with [K : F] = 2.

A-36. Let K be a ﬁeld of characteristic p, say with p

n

elements, and let θ be the au-

tomorphism of K that maps every element to its p

th

power. Show that there exists an

automorphism α of K such that θα

2

= 1 if and only if n is odd.

A-37. Describe the splitting ﬁeld and Galois group, over Q, of the polynomial X

5

−9.

A-38. Suppose that E is a Galois ﬁeld extension of a ﬁeld F such that [E : F] = 5

3

(43)

2

.

Prove that there exist ﬁelds K

1

and K

2

lying strictly between F and E with the following

properties: (i) each K

i

is a Galois extension of F; (ii) K

1

∩ K

2

= F; and (iii) K

1

K

2

= E.

A-39. Let F = F

p

for some prime p. Let m be a positive integer not divisible by p, and

let K be the splitting ﬁeld of X

m

−1. Find [K : F] and prove that your answer is correct.

89

A-40. Let F be a ﬁeld of 81 elements. For each of the following polynomials g(X),

determine the number of roots of g(X) that lie in F: X

80

−1, X

81

−1, X

88

−1.

A-41. Describe the Galois group of the polynomial X

6

−7 over Q.

A-42. Let K be a ﬁeld of characteristic p > 0 and let F = K(u, v) be a ﬁeld extension of

degree p

2

such that u

p

∈ K and v

p

∈ K. Prove that K is not ﬁnite, that F is not a simple

extension of K, and that there exist inﬁnitely many intermediate ﬁelds F ⊃ L ⊃ K.

A-43. Find the splitting ﬁeld and Galois group of the polynomial X

3

− 5 over the ﬁeld

Q[

√

2].

A-44. For any prime p, ﬁnd the Galois group over Q of the polynomial X

5

−5p

4

X +p.

A-45. Factorize X

4

+1 over each of the ﬁnite ﬁelds (a) F

5

; (b) F

25

; and (c) F

125

. Find its

splitting ﬁeld in each case.

A-46. Let Q[α] be a ﬁeld of ﬁnite degree over Q. Assume that there is a q ∈ Q, q = 0,

such that [ρ(α)[ = q for all homomorphisms ρ: Q[α] → C. Show that the set of roots

of the minimum polynomial of α is the same as that of q

2

/α. Deduce that there exists an

automorphism σ of Q[α] such that

(a) σ

2

= 1 and

(b) ρ(σγ) = ρ(γ) for all γ ∈ Q[α] and ρ: Q[α] →C.

A-47. Let F be a ﬁeld of characteristic zero, and let p be a prime number. Suppose that

F has the property that all irreducible polynomials f(X) ∈ F[X] have degree a power of

p (1 = p

0

is allowed). Show that every equation g(X) = 0, g ∈ F[X], is solvable by

extracting radicals.

A-48. Let K = Q[

√

5,

√

−7] and let L be the splitting ﬁeld over Q of f(X) = X

3

−10.

(a) Determine the Galois groups of K and L over Q.

(b) Decide whether K contains a root of f.

(c) Determine the degree of the ﬁeld K ∩ L over Q.

[Assume all ﬁelds are subﬁelds of C.]

A-49. Find the splitting ﬁeld (over F

p

) of X

p

r

− X ∈ F

p

[X], and deduce that X

p

r

− X

has an irreducible factor f ∈ F

p

[X] of degree r. Let g(X) ∈ Z[X] be a monic polynomial

that becomes equal to f(X) when its coefﬁcients are read modulo p. Show that g(X) is

irreducible in Q[X].

A-50. Let E be the splitting ﬁeld of X

3

−51 over Q. List all the subﬁelds of E, and ﬁnd

an element γ of E such that E = Q[γ].

A-51. Let k = F

1024

be the ﬁeld with 1024 elements, and let K be an extension of k of

degree 2. Prove that there is a unique automorphism σ of K of order 2 which leaves k

elementwise ﬁxed and determine the number of elements of K

such that σ(x) = x

−1

.

A-52. Let F and E be ﬁnite ﬁelds of the same characteristic. Prove or disprove these

statements:

(a) There is a ring homomorphism of F into E if and only if #E is a power of #F.

90 A REVIEW EXERCISES

(b) There is an injective group homomorphism of the multiplicative group of F into the

multiplicative group of E if and only if #E is a power of #F.

A-53. Let L/K be an algebraic extension of ﬁelds. Prove that L is algebraically closed if

every polynomial over K factors completely over L.

A-54. Let K be a ﬁeld, and let M = K(X), X an indeterminate. Let L be an intermediate

ﬁeld different from K. Prove that M is ﬁnite-dimensional over L.

A-55. Let θ

1

, θ

2

, θ

3

be the roots of the polynomial f(X) = X

3

+X

2

−9X + 1.

(a) Show that the θ

i

are real, nonrational, and distinct.

(b) Explain why the Galois group of f(X) over Q must be either A

3

or S

3

. Without

carrying it out, give a brief description of a method for deciding which it is.

(c) Show that the rows of the matrix

¸

¸

¸

¸

3 9 9 9

3 θ

1

θ

2

θ

3

3 θ

2

θ

3

θ

1

3 θ

3

θ

1

θ

2

¸

**are pairwise orthogonal; compute their lengths, and compute the determinant of the
**

matrix.

A-56. Let E/K be a Galois extension of degree p

2

q where p and q are primes, q < p and

q not dividing p

2

−1. Prove that:

(a) there exist intermediate ﬁelds L and M such that [L : K] = p

2

and [M : K] = q;

(b) such ﬁelds L and M must be Galois over K; and

(c) the Galois group of E/K must be abelian.

A-57. Let ζ be a primitive 7

th

root of 1 (in C).

(a) Prove that 1+X+X

2

+X

3

+X

4

+X

5

+X

6

is the minimum polynomial of ζ over

Q.

(b) Find the minimum polynomial of ζ +

1

ζ

over Q.

A-58. Find the degree over Q of the Galois closure K of Q[2

1

4

] and determine the isomor-

phism class of Gal(K/Q).

A-59. Let p, q be distinct positive prime numbers, and consider the extension K = Q[

√

p,

√

q] ⊃

Q.

(a) Prove that the Galois group is isomorphic to C

2

C

2

.

(b) Prove that every subﬁeld of K of degree 2 over Q is of the form Q[

√

m] where

m ∈ ¦p, q, pq¦.

(c) Show that there is an element γ ∈ K such that K = Q[γ].

91

B Two-hour Examination

1. (a) Let σ be an automorphism of a ﬁeld E. If σ

4

= 1 and

σ(α) +σ

3

(α) = α +σ

2

(α) all α ∈ E,

show that σ

2

= 1.

(b) Let p be a prime number and let a, b be rational numbers such that a

2

+pb

2

= 1. Show

that there exist rational numbers c, d such that a =

c

2

+pd

2

c

2

−pd

2

and b =

2cd

c

2

−pd

2

.

2. Let f(X) be an irreducible polynomial of degree 4 in Q[X], and let g(X) be the resolvent

cubic of f. What is the relation between the Galois group of f and that of g? Find the Galois

group of f if

(a) g(X) = X

3

−3X + 1;

(b) g(X) = X

3

+ 3X + 1.

3. (a) How many monic irreducible factors does X

255

−1 ∈ F

2

[X] have, and what are their

degrees.

(b) How many monic irreducible factors does X

255

− 1 ∈ Q[X] have, and what are their

degrees?

4. Let E be the splitting ﬁeld of (X

5

−3)(X

5

−7) ∈ Q[X]. What is the degree of E over

Q? How many proper subﬁelds of E are there that are not contained in the splitting ﬁelds

of both X

5

−3 and X

5

−7?

[You may assume that 7 is not a 5th power in the splitting ﬁeld of X

5

−3.]

5. Consider an extension Ω ⊃ F of ﬁelds. Deﬁne a ∈ Ω to be F-constructible if it is

contained in a ﬁeld of the form

F[

√

a

1

, . . . ,

√

a

n

], a

i

∈ F[

√

a

1

, . . . ,

√

a

i−1

].

Assume Ω is a ﬁnite Galois extension of F and construct a ﬁeld E, F ⊂ E ⊂ Ω, such that

every a ∈ Ω is E-constructible and E is minimal with this property.

6. Let Ω be an extension ﬁeld of a ﬁeld F. Show that every F-homomorphism Ω → Ω is

an isomorphism provided:

(a) Ω is algebraically closed, and

(b) Ω has ﬁnite transcendence degree over F.

Can either of the conditions (i) or (ii) be dropped? (Either prove, or give a counterex-

ample.)

You should prove all answers. You may use results proved in class or in the notes, but you

should indicate clearly what you are using.

Possibly useful facts: The discriminant of X

3

+aX+b is −4a

3

−27b

2

and 2

8

−1 = 255 =

3 5 17.

92 C SOLUTIONS TO THE EXERCISES

C Solutions to the Exercises

These solutions fall somewhere between hints and complete solutions. Students were ex-

pected to write out complete solutions.

1-1. Similar to Example 1.28.

1-2. Verify that 3 is not a square in Q[

√

2], and so [Q[

√

2,

√

3] : Q] = 4.

1-3. (a) Apply the division algorithm, to get f(X) = q(X)(X − a) + r(X) with r(X)

constant, and put X = a to ﬁnd r = f(a).

(c) Use that factorization in F[X] is unique (or use induction on the degree of f).

(d) If G had two cyclic factors C and C

t

whose orders were divisible by a prime p, then G

would have (at least) p

2

elements of order dividing p. This doesn’t happen, and it follows

that G is cyclic.

(e) The elements of order m in F

**are the roots of the polynomial X
**

m

− 1, and so there

are at most m of them. Hence any ﬁnite subgroup G of F

**satisﬁes the condition in (d).
**

1-4. Note that it sufﬁces to construct α = cos

2π

7

, and that [Q[α] : Q] =

7−1

2

= 3, and

so its minimum polynomial has degree 3. There is a standard method (once taught in high

schools) for solving cubics using the equation

cos 3θ = 4 cos

3

θ −3 cos θ.

By “completing the cube”, reduce the cubic to the form X

3

− pX − q. Then construct a

so that a

2

=

4p

3

. Choose 3θ such that cos 3θ =

4q

a

3

. If β = cos θ is a solution of the above

equation, then α = aβ will be a root of X

3

−pX −q.

2-1. (a) is obvious, as is the “only if” in (b). For the “if” note that for any a ∈ S(E),

a / ∈ F

2

, E ≈ F[X]/(X

2

−a).

(c) Take E

i

= Q[

√

p

i

] with p

i

the i

th

prime. Check that p

i

is the only prime that

becomes a square in E

i

. For this use that (a +b

√

p)

2

∈ Q =⇒ 2ab = 0.

(d) Any ﬁeld of characteristic p contains (an isomorphic copy of) F

p

, and so we are

looking at the quadratic extensions of F

p

. The homomorphism a → a

2

: F

p

→ F

p

has kernel ¦±1¦, and so its image has index 2 in F

p

. Thus the only possibility for S(E)

is F

p

, and so there is at most one E (up to F

p

-isomorphism). To get one, take E =

F[X]/(X

2

−a), a / ∈ F

2

p

.

2-2. (a) If α is a root of f(X) = X

p

−X −a (in some splitting ﬁeld), then the remaining

roots are α+1, . . . , α+p−1, which obviously lie in whichever ﬁeld contains α. Moreover,

they are distinct. Suppose that, in F[X],

f(X) = (X

r

+a

1

X

r−1

+ +a

r

)(X

p−r

+ ), 0 < r < p.

Then −a

1

is a sum of r of the roots of f, −a

1

= rα + d some d ∈ Z 1

F

, and it follows

that α ∈ F.

(b) As 0 and 1 are not roots of X

p

−X−1 in F

p

it can’t have p distinct roots in F

p

, and

so (a) implies that X

p

− X − 1 is irreducible in F

p

[X] and hence also in Z[X] and Q[X]

(see 1.18, 1.13).

2-3. Let α be the real 5

th

root of 2. Eisenstein’s criterion shows that X

5

− 2 is irre-

ducible in Q[X], and so Q[

5

√

2] has degree 5 over Q. The remaining roots of X

5

− 2 are

ζα, ζ

2

α, ζ

3

α, ζ

4

α, where ζ is a primitive 5

th

root of 1. It follows that the subﬁeld of C

93

generated by the roots of X

5

− 2 is Q[ζ, α]. The degree of Q[ζ, α] is 20, since it must be

divisible by [Q[ζ] : Q] = 4 and [Q[α] : Q] = 5.

2-4. It’s F

p

because X

p

m

− 1 = (X − 1)

p

m

. (Perhaps I meant X

p

m

− X — that would

have been more interesting.)

2-5. If f(X) =

¸

(X −α

i

)

m

i

, α

i

= α

j

, then

f

t

(X) =

¸

m

i

f(X)

X −α

i

and so d(X) =

¸

m

i

>1

(X −α

i

)

m

i

−1

. Therefore g(X) =

¸

(X −α

i

).

2-6. From (2.12) we know that either f is separable or f(X) = f

1

(X

p

) for some polyno-

mial f

1

. Clearly f

1

is also irreducible. If f

1

is not separable, it can be written f

1

(X) =

f

2

(X

p

). Continue in the way until you arrive at a separable polynomial. For the ﬁnal state-

ment, note that g(X) =

¸

(X−a

i

), a

i

= a

j

, and so f(X) = g(X

p

e

) =

¸

(X−α

i

)

p

e

with

α

p

e

i

= a

i

.

3-1. Let σ and τ be automorphisms of F(X) given by σ(X) = −X and τ(X) = 1 − X.

Then σ and τ ﬁx X

2

and X

2

−X respectively, and so στ ﬁxes E =

df

F(X)∩F(X

2

−X).

But ατX = 1 + X, and so (στ)

m

(X) = m + X. Thus Aut(F(X)/E) is inﬁnite, which

implies that [F(X) : E] is inﬁnite (otherwise F(X) = E[α

1

, . . . , α

n

]; an E-automorphism

of F(X) is determined by its values on the α

i

, and its value on α

i

is a root of the minimum

polynomial of α

i

). If E contains a polynomial f(X) of degree m > 0, then [F(X) : E] ≤

[F(X) : F(f(X))] = m — contradiction.

3-2. Since 1 + ζ + + ζ

p−1

= 0, we have α + β = −1. If i ∈ H, then iH = H and

i(GH) = GH, and so α and β are ﬁxed by H. If j ∈ GH, then jH = GH and

j(GH) = H, and so jα = β and jβ = α. Hence αβ ∈ Q, and α and β are the roots of

X

2

+X +αβ. Note that

αβ =

¸

i,j

ζ

i+j

, i ∈ H, j ∈ GH.

How many times do we have i+j = 0? If i+j = 0, then −1 = i

−1

j, which is a nonsquare;

conversely, if −1 is a nonsquare, take i = 1 and j = −1 to get i +j = 0. Hence

i +j = 0 some i ∈ H, j ∈ GH ⇐⇒ −1 is a square mod p ⇐⇒ p ≡ −1 mod 4.

If we do have a solution to i + j = 0, we get all solutions by multiplying it through by the

p−1

2

squares. So in the sum for αβ we see 1 a total of

p−1

2

times when p ≡ 3 mod 4 and not

at all if p ≡ 1 mod 4. In either case, the remaining terms add to a rational number, which

implies that each power of ζ occurs the same number of times. Thus for p ≡ 1 mod 4,

αβ = −(

p−1

2

)

2

/(p − 1) =

p−1

4

; the polynomial satisﬁed by α and β is X

2

+ X −

p−1

4

,

whose roots are (−1 ±

√

1 +p −1)/2; the ﬁxed ﬁeld of H is Q[

√

p]. For p ≡ −1 mod 4,

αβ =

p−1

2

+ (−1)

(

p−1

2

)

2

−

p−1

2

/(p − 1) =

p−1

2

−

p−3

4

=

p+1

4

; the polynomial is

X

2

+X +

p−1

4

, with roots (−1 ±

√

1 −p −1)/2; the ﬁxed ﬁeld of H is Q[

√

−p].

3-3. (a) It is easy to see that M is Galois over Q with Galois group 'σ, τ`:

σ

√

2 = −

√

2

σ

√

3 =

√

3

τ

√

2 =

√

2

τ

√

3 = −

√

3

.

94 C SOLUTIONS TO THE EXERCISES

(b) We have

σα

2

α

2

=

2 −

√

2

2 +

√

2

=

(2 −

√

2)

2

4 −2

=

2 −

√

2

√

2

2

= (

√

2 −1)

2

,

i.e., σα

2

= ((

√

2 −1)α)

2

. Thus, if α ∈ M, then σα = ±(

√

2 −1)α, and

σ

2

α = (−

√

2 −1)(

√

2 −1)α = −α;

as σ

2

α = α = 0, this is impossible. Hence α / ∈ M, and so [E : Q] = 8.

Extend σ to an automorphism (also denoted σ) of E. Again σα = ±(

√

2 − 1)α and

σ

2

α = −α, and so σ

2

= 1. Now σ

4

α = α, σ

4

[M = 1, and so we can conclude that σ has

order 4. After possibly replacing σ with its inverse, we may suppose that σα = (

√

2 −1)α.

Repeat the above argument with τ:

τα

2

α

2

=

3−

√

3

3+

√

3

=

3−

√

3

√

6

2

, and so we can extend τ to

an automorphism of L (also denoted τ) with τα =

3−

√

3

√

6

α. The order of τ is 4.

Finally compute that

στα =

3 −

√

3

−

√

6

(

√

2 −1)α; τσα = (

√

2 −1)

3 −

√

3

√

6

α.

Hence στ = τσ, and Gal(E/Q) has two noncommuting elements of order 4. Since it has

order 8, it must be the quaternion group.

4-1. The splitting ﬁeld is the smallest ﬁeld containing all m

th

roots of 1. Hence it is F

p

n

where n is the smallest positive integer such that m

0

[p

n

−1, m = m

0

p

r

.

4-2. We have X

4

− 2X

3

− 8X − 3 = (X

3

+ X

2

+ 3X + 1)(X − 3), and g(X) =

X

3

+X

2

+3X+1 is irreducible over Q(use 1.4??), and so its Galois group is either A

3

or

S

3

. Either check that its discriminant is not a square or, more simply, show by examining its

graph that g(X) has only one real root, and hence its Galois group contains a transposition

(cf. the proof of 4.13??).

4-3. Eisenstein’s criterion shows that X

8

− 2 is irreducible over Q, and so [Q[α] : Q] = 8

where α is a positive 8

th

root of 2. As usual for polynomials of this type, the splitting ﬁeld

is Q[α, ζ] where ζ is any primitive 8

th

root of 1. For example, ζ can be taken to be

1+i

√

2

,

which lies in Q[α, i]. It follows that the splitting ﬁeld is Q[α, i]. Clearly Q[α, i] = Q[α],

because Q[α], unlike i, is contained in R, and so [Q[α, i] : Q] = 2. Therefore the degree is

2 8 = 16.

4-4. Find an extension L/F with Galois group S

4

, and let E be the ﬁxed ﬁeld of S

3

⊂ S

4

.

There is no subgroup strictly between S

n

and S

n−1

, because such a subgroup would be

transitive and contain an (n − 1)-cycle and a transposition, and so would equal S

n

(see

4.23). We can take E = L

S

3

.

4-5. Type: “Factor(X

343

−X) mod 7;” and discard the 7 factors of degree 1.

4-6. Type “galois(X

6

+2X

5

+3X

4

+4X

3

+5X

2

+6X +7);”. It is the group PGL

2

(F

5

)

(group of invertible 2 2 matrices over F

5

modulo scalar matrices) which has order 120.

Alternatively, note that there are the following factorizations: mod 3, irreducible; mod 5

(deg 3)(deg 3); mod 13 (deg 1)(deg 5); mod 19, (deg 1)

2

(deg 4); mod 61 (deg 1)

2

(deg 2)

2

;

mod 79, (deg 2)

3

. Thus the Galois group has elements of type:

6, 3 + 3, 1 + 5, 1 + 1 + 4, 1 + 1 + 2 + 2, 2 + 2 + 2.

95

No element of type 2, 3, 3 + 2, or 4 + 2 turns up by factoring modulo any of the ﬁrst 400

primes (or, so I have been told). This suggests it is the group T14 in the tables in Butler and

McKay, which is indeed PGL

2

(F

5

).

4-7. ⇐= : Condition (a) implies that G

f

contains a 5-cycle, condition (b) implies that

G

f

⊂ A

5

, and condition (c) excludes A

5

. That leaves D

5

and C

5

as the only possibilities

(see, for example, Jacobson, Basic Algebra I, p305, Ex 6). The derivative of f is 5X

4

+a,

which has at most 2 real zeros, and so (from its graph) we see that f can have at most 3 real

zeros. Thus complex conjugation acts as an element of order 2 on the splitting ﬁeld of f,

and this shows that we must have G

f

= D

5

.

=⇒ : Regard D

5

as a subgroup of S

5

by letting it act on the vertices of a regular

pentagon—all subgroups of S

5

isomorphic to D

5

look like this one. If G

f

= D

5

, then

(a) holds because D

5

is transitive, (b) holds because D

5

⊂ A

5

, and (c) holds because D

5

is

solvable.

4-8. Omitted.

4-9. Let a

1

, a

2

be conjugate nonreal roots, and let a

3

be a real root. Complex conjugation

deﬁnes an element σ of the Galois group of f switching a

1

and a

2

and ﬁxing a

3

. On the

other hand, because f is irreducible, its Galois group acts transitively on its roots, and so

there is a τ such that τ(a

3

) = a

1

. Now

a

3

τ

→a

1

σ

→a

2

a

3

σ

→a

3

τ

→a

1.

This statement is false for reducible polynomials — consider for example f(X) = (X

2

+

1)(X −1).

5-1. For a = 1, this is the polynomial Φ

5

(X), whose Galois group is cyclic of order 4.

For a = 0, it is X(X

3

+X

2

+X +1) = X(X +1)(X

2

+1), whose Galois group is cyclic

of order 2.

For a = −4, it is (X −1)(X

3

+ 2X

2

+ 3X + 4). The cubic does not have ±1, ±2, or ±4

as roots, and so it is irreducible in Q[X]. Hence its Galois group is S

3

or A

3

. But looking

modulo 2, we see it contains a 2-cycle, so it must be S

3

.

For any a, the resolvent cubic is

g(X) = X

3

−X

2

+ (1 −4a)X + 3a −1.

Take a = −1. Then f = X

4

+ X

3

+ X

2

+ X − 1 is irreducible modulo 2, and so it is

irreducible in Q[X]. We have g = X

3

− X

2

+ 5X − 4, which is irreducible. Moreover

g

t

= 3X

2

− 2X + 5 = 3(X −

1

3

)

2

+ 4

2

3

> 0 always, and so g has exactly one real root.

Hence the Galois group of g is S

3

, and therefore the Galois group of f is S

4

. [In fact, 4

is the maximum number of integers giving distinct Galois groups: checking mod 2, we see

there is a 2-cycle or a 4-cycle, and so 1, A

3

, A

4

, V

4

are not possible. For D

8

, a can’t be an

integer.]

5-2. We have Nm(a + ib) = a

2

+ b

2

. Hence a

2

+ b

2

= 1 if and only a + ib =

s+it

s−it

for

some s, t ∈ Q (Hilbert’s Theorem 90). The rest is easy.

5-3. The degree [Q[ζ

n

] : Q] = ϕ(n), ζ

n

a primitive n

th

root of 1, and ϕ(n) → ∞ as

n →∞.

96 C SOLUTIONS TO THE EXERCISES

A-1. (a) Need that m[n, because

n = [F

p

n : F

p

] = [F

p

n : F

p

m] [F

p

m : F

p

] = [F

p

n : F

p

m] m.

Use Galois theory to show there exists one, for example. (b) Only one; it consists of all the

solutions of X

p

m

−X = 0.

A-2. The polynomial is irreducible by Eisenstein’s criterion. The polynomial has only

one real root, and therefore complex conjugation is a transposition in G

f

. This proves that

G

f

≈ S

3

. The discriminant is −1323 = −3

3

7

2

. Only the subﬁeld Q[

√

−3] is normal over

Q. The subﬁelds Q[

3

√

7], Q[ζ

3

√

7] Q[ζ

2

3

√

7] are not normal over Q. [The discriminant of

X

3

−a is −27a

2

= −3(3a)

2

.]

A-3. The prime 7 becomes a square in the ﬁrst ﬁeld, but 11 does not: (a + b

√

7)

2

=

a

2

+7b

2

+2ab

√

7, which lies in Qonly if ab = 0. Hence the rational numbers that become

squares in Q[

√

7] are those that are already squares or lie in 7Q

2

.

A-4.(a) See Exercise 3.

(b) Let F = F

3

[X]/(X

2

+ 1). Modulo 3

X

8

−1 = (X −1)(X + 1)(X

2

+ 1)(X

2

+X + 2)(X

2

+ 2X + 2).

Take α to be a root of X

2

+X + 2.

A-5. Since E = F, E contains an element

f

g

with the degree of f or g > 0. Now

f(T) −

f(X)

g(X)

g(T)

is a nonzero polynomial having X as a root.

A-6. Use Eisenstein to show that X

p−1

+ + 1 is irreducible, etc. Done in class.

A-7. The splitting ﬁeld is Q[ζ, α] where ζ

5

= 1 and α

5

= 2. It is generated by σ = (12345)

and τ = (2354), where σα = ζα and τζ = ζ

2

. The group has order 20. It is not abelian

(because Q[α] is not Galois over Q), but it is solvable (its order is < 60).

A-8. (a) A homomorphism α: R → R acts as the identity map on Z, hence on Q, and

it maps positive real numbers to positive real numbers, and therefore preserves the order.

Hence, for each real number a,

¦r ∈ Q [ a < r¦ = ¦r ∈ Q [ α(a) < r¦,

which implies that α(a) = a.

(b) Choose a transcendence basis A for C over Q. Because it is inﬁnite, there is a

bijection α: A → A

t

from A onto a proper subset. Extend α to an isomorphism Q(A) →

Q(A

t

), and then extend it to an isomorphism C → C

t

where C

t

is the algebraic closure of

Q(A

t

) in C.

A-9. The group F

**is cyclic of order 15. It has 3 elements of order dividing 3, 1 element
**

of order dividing 4, 15 elements of order dividing 15, and 1 element of order dividing 17.

A-10. If E

1

and E

2

are Galois extensions of F, then E

1

E

2

and E

1

∩E

2

are Galois over F,

and there is an exact sequence

1 −−→ Gal(E

1

E

2

/F) −−→ Gal(E

1

/F) Gal(E

2

/F) −−→ Gal(E

1

∩ E

2

/F) −−→ 1.

97

In this case, E

1

∩ E

2

= Q[ζ] where ζ is a primitive cube root of 1. The degree is 18.

A-11. Over Q, the splitting ﬁeld is Q[α, ζ] where α

6

= 5 and ζ

3

= 1 (because −ζ is then a

primitive 6

th

root of 1). The degree is 12, and the Galois group is D

6

(generators (26)(35)

and (123456)).

Over R, the Galois group is C

2

.

A-12. Let the coefﬁcients of f be a

1

, . . . , a

n

— they lie in the algebraic closure Ω of F.

Let g(X) be the product of the minimum polynomials over F of the roots of f in Ω.

Alternatively, the coefﬁcients will lie in some ﬁnite extension E of F, and we can take

the norm of f(X) from E[X] to F[X].

A-13. If f is separable, [E: F] = (G

f

: 1), which is a subgroup of S

n

. Etc..

A-14.

√

3 +

√

7 will do.

A-15. The splitting ﬁeld of X

4

− 2 is E

1

= Q[i, α] where α

4

= 2; it has degree 8, and

Galois group D

4

. The splitting ﬁeld of X

3

−5 is E

2

= Q[ζ, β]; it has degree 6, and Galois

group D

3

. The Galois group is the product (they could only intersect in Q[

√

3], but

√

3

does not become a square in E

1

).

A-16. The multiplicative group of F is cyclic of order q −1. Hence it contains an element

of order 4 if and only if 4[q −1.

A-17. Take α =

√

2 +

√

5 +

√

7.

A-18. We have E

1

= E

H

1

, which has degree n over F, and E

2

= E

<1n>

, which has

degree (n − 1)! over F, etc.. This is really a problem in group theory posing as a problem

in ﬁeld theory.

A-19. We have Q[ζ] = Q[i, ζ

t

] where ζ

t

is a primitive cube root of 1 and ±i = ζ

3

etc..

A-20. The splitting ﬁeld is Q[ζ,

3

√

3], and the Galois group is S

3

.

A-21. Use that

(ζ +ζ

4

)(1 +ζ

2

) = ζ +ζ

4

+ζ

3

+ζ

A-22. (a) is Dedekind’s theorem. (b) is Artin’s theorem 3.4. (c) is O.K. because X

p

− a

p

has a unique root in Ω.

A-23. The splitting ﬁeld is Q[i, α] where α

4

= 3, and the Galois group is D

4

with genera-

tors (1234) and (13) etc..

A-24. From Hilbert’s theorem 90, we know that the kernel of the map N: E

→ F

**consists of elements of the form
**

σα

α

. The map E

→ E

, α →

σα

α

, has kernel F

.

Therefore the kernel of N has order

q

m

−1

q−1

, and hence its image has order q −1. There is a

similar proof for the trace — I don’t know how the examiners expected you to prove it.

A-25. (a) is false—could be inseparable. (b) is true—couldn’t be inseparable.

A-26. Apply the Sylow theorem to see that the Galois group has a subgroup of order 81.

Now the Fundamental Theorem of Galois theory shows that F exists.

A-27. The greatest common divisor of the two polynomials over Q is X

2

+ X + 1, which

must therefore be the minimum polynomial for θ.

A-28. Theorem on p-groups plus the Fundamental Theorem of Galois Theory.

98 C SOLUTIONS TO THE EXERCISES

A-29. It was proved in class that S

p

is generated by an element of order p and a transposition

(4.14). There is only one F, and it is quadratic over Q.

A-30. Let L = K[α]. The splitting ﬁeld of the minimum polynomial of α has degree at

most d!, and a set with d! elements has at most 2

d!

subsets. [Of course, this bound is much

too high: the subgroups are very special subsets. For example, they all contain 1 and they

are invariant under a →a

−1

.]

A-31. The Galois group is (Z/5Z)

**, which cyclic of order 4, generated by 2.
**

(ζ +ζ

4

) + (ζ

2

+ζ

3

) = −1, (ζ +ζ

4

)(ζ

2

+ζ

3

) = −1.

(a) Omit.

(b) Certainly, the Galois group is a product C

2

C

4

.

A-32. Let a

1

, . . . , a

5

be a transcendence basis for Ω

1

/Q. Their images are algebraically

independent, therefore they are a maximal algebraically independent subset of Ω

2

, and

therefore they form a transcendence basis, etc..

A-33. C

2

C

2

.

A-34. If f(X) were reducible over Q[

√

7], it would have a root in it, but it is irreducible

over Q by Eisenstein’s criterion. The discriminant is −675, which is not a square in any R,

much less Q[

√

7].

A-35. (a) Should be X

5

− 6X

4

+ 3. The Galois group is S

5

, with generators (12) and

(12345) — it is irreducible (Eisenstein) and (presumably) has exactly 2 nonreal roots. (b)

It factors as (X +1)(X

4

+X

3

+X

2

+X +1). Hence the splitting ﬁeld has degree 4 over

F

2

, and the Galois group is cyclic.

A-36. This is really a theorem in group theory, since the Galois group is a cyclic group of

order n generated by θ. If n is odd, say n = 2m+ 1, then α = θ

m

does.

A-37. It has order 20, generators (12345) and (2354).

A-38. Take K

1

and K

2

to be the ﬁelds corresponding to the Sylow 5 and Sylow 43 sub-

groups. Note that of the possible numbers 1, 6, 11, 16, 21, ... of Sylow 5-subgroups, only 1

divides 43. There are 1, 44, 87, ... subgroups of ....

A-39. See Exercise 14.

A-40. The group F

**is cyclic of order 80; hence 80, 1, 8.
**

A-41. It’s D

6

, with generators (26)(35) and (123456). The polynomial is irreducible by

Eisenstein’s criterion, and its splitting ﬁeld is Q[α, ζ] where ζ = 1 is a cube root of 1.

A-42. Example 5.5.

A-43. Omit.

A-44. It’s irreducible by Eisenstein. Its derivative is 5X

4

− 5p

4

, which has the roots

X = ±p. These are the max and mins, X = p gives negative; X = −p gives positive.

Hence the graph crosses the x-axis 3 times and so there are 2 imaginary roots. Hence the

Galois group is S

5

.

A-45. Its roots are primitive 8

th

roots of 1. It splits completely in F

25

. (a) (X

2

+2)(X

2

+3).

99

A-46. ρ(α)ρ(α) = q

2

, and ρ(α)ρ(

q

2

α

) = q

2

. Hence ρ(

q

2

α

) is the complex conjugate of ρ(α).

Hence the automorphism induced by complex conjugation is independent of the embedding

of Q[α] into C.

A-47. The argument that proves the Fundamental Theoremof Algebra, shows that its Galois

group is a p-group. Let E be the splitting ﬁeld of g(X), and let H be the Sylowp-subgroup.

Then E

H

= F, and so the Galois group is a p-group.

A-48. (a) C

2

C

2

and S

3

. (b) No. (c). 1

A-49. Omit.

A-50. Omit.

A-51. 1024 = 2

10

. Want σx x = 1, i.e., Nx = 1. They are the elements of the form

σx

x

;

have

1 −−−−→ k

−−−−→ K

x→

σx

x

−−−−→ K

.

Hence the number is 2

11

/2

10

= 2.

A-52. Pretty standard. False; true.

A-53. Omit.

A-54. Similar to a previous problem.

A-55. Omit.

A-56. This is really a group theory problem disguised as a ﬁeld theory problem.

A-57. (a) Prove it’s irreducible by apply Eisenstein to f(X + 1). (b) See example worked

out in class.

A-58. It’s D

4

, with generators (1234) and (12).

A-59. Omit.

Solutions for the exam.

1. (a) Let σ be an automorphism of a ﬁeld E. If σ

4

= 1 and

σ(α) +σ

3

(α) = α +σ

2

(α) all α ∈ E,

show that σ

2

= 1.

If σ

2

= 1, then 1, σ, σ

2

, σ

3

are distinct automorphisms of E, and hence are linearly

independent (Dedekind 5.14) — contradiction. [If σ

2

= 1, then the condition becomes

2σ = 2, so either σ = 1 or the characteristic is 2 (or both).]

(b) Let p be a prime number and let a, b be rational numbers such that a

2

+pb

2

= 1. Show

that there exist rational numbers c, d such that a =

c

2

+pd

2

c

2

−pd

2

and b =

2cd

c

2

−pd

2

.

Apply Hilbert’s Theorem 90 to Q[

√

p] (or Q[

√

−p], depending how you wish to correct

the sign).

2. Let f(X) be an irreducible polynomial of degree 4 in Q[X], and let g(X) be the resolvent

cubic of f. What is the relation between the Galois group of f and that of g? Find the Galois

group of f if

100 C SOLUTIONS TO THE EXERCISES

(a) g(X) = X

3

−3X + 1;

(b) g(X) = X

3

+ 3X + 1.

We have G

g

= G

f

/G

f

∩V , where V = ¦1, (12)(34), . . .¦. The two cubic polynomials

are irreducible, because their only possible roots are ±1. From their discriminants, one

ﬁnds that the ﬁrst has Galois group A

3

and the second S

3

. Because f(X) is irreducible,

4[(G

f

: 1) and it follows that G

f

= A

4

and S

4

in the two cases.

3. (a) How many monic irreducible factors does X

255

−1 ∈ F

2

[X] have, and what are their

degrees?

Its roots are the nonzero elements of F

2

8, which has subﬁelds F

2

4⊃ F

2

2⊃ F

2

. There

are 256−16 elements not in F

16

, and their minimum polynomials all have degree 8. Hence

there are 30 factors of degree 8, 3 of degree 4, and 1 each of degrees 2 and 1.

(b) How many monic irreducible factors does X

255

− 1 ∈ Q[X] have, and what are their

degrees?

Obviously, X

255

−1 =

¸

d[255

Φ

d

= Φ

1

Φ

3

Φ

5

Φ

15

Φ

255

, and we showed in class that

the Φ

d

are irreducible. They have degrees 1, 2, 4, 8, 16, 32, 64, 128.

4. Let E be the splitting ﬁeld of (X

5

−3)(X

5

−7) ∈ Q[X]. What is the degree of E over

Q? How many proper subﬁelds of E are there that are not contained in the splitting ﬁelds

of both X

5

−3 and X

5

−7?

The splitting ﬁeld of X

5

− 3 is Q[ζ, α], which has degree 5 over Q[ζ] and 20 over Q.

The Galois group of X

5

− 7 over Q[ζ, α] is (by ...) a subgroup of a cyclic group of order

5, and hence has order 1 or 5. Since 7 is not a 5th power in Q[ζ, α], it must be 5. Thus

[E : Q] = 100, and

G = Gal(E/Q) = (C

5

C

5

) C

4

.

We want the nontrivial subgroups of G not containing C

5

C

5

. The subgroups of order 5

of C

5

C

5

are lines in (F

5

)

2

, and hence C

5

C

5

has 6 + 1 = 7 proper subgroups. All

are normal in G. Each subgroup of C

5

C

5

is of the form H ∩ (C

5

C

5

) for exactly

3 subgroups H of G corresponding to the three possible images in G/(C

5

C

5

) = C

4

.

Hence we have 21 subgroups of G not containing C

5

C

5

, and 20 nontrivial ones. Typical

ﬁelds: Q[α], Q[α, cos

2π

5

], Q[α, ζ].

[You may assume that 7 is not a 5th power in the splitting ﬁeld of X

5

−3.]

5. Consider an extension Ω ⊃ F of ﬁelds. Deﬁne α ∈ Ω to be F-constructible if it is

contained in a ﬁeld of the form

F[

√

a

1

, . . . ,

√

a

n

], a

i

∈ F[

√

a

1

, . . . ,

√

a

i−1

].

Assume Ω is a ﬁnite Galois extension of F and construct a ﬁeld E, F ⊂ E ⊂ Ω, such that

every a ∈ Ω is E-constructible and E is minimal with this property.

Suppose E has the required property. From the primitive element theorem, we know

Ω = E[a] for some a. Now a E-constructible =⇒ [Ω : E] is a power of 2. Take E = Ω

H

,

where H is the Sylow 2-subgroup of Gal(Ω/F).

6. Let Ω be an extension ﬁeld of a ﬁeld F. Show that every F-homomorphism Ω → Ω is

an isomorphism provided:

(a) Ω is algebraically closed, and

(b) Ω has ﬁnite transcendence degree over F.

Can either of the conditions (i) or (ii) be dropped? (Either prove, or give a counterex-

ample.)

101

Let A be a transcendence basis for Ω/F. Because σ: Ω → Ω is injective, σ(A) is

algebraically independent over F, and hence (because it has the right number of elements)

is a transcendence basis for Ω/F. Now F[σA] ⊂ σΩ ⊂ Ω. Because Ω is algebraic over

F[σA] and σΩ is algebraically closed, the two are equal. Neither condition can be dropped.

E.g., C(X)→C(X), X →X

2

. E.g., Ω = the algebraic closure of C(X

1

, X

2

, X

3

, . . .), and

consider an extension of the map X

1

→X

2

, X

2

→X

3

, . . ..

Index

algebraic, 10–12

algebraic closure, 17

in a larger ﬁeld, 17

algebraic integer, 5

algebraically closed, 17

algebraically dependent, 79

algebraically independent, 79

algorithm

division, 3

Euclid’s, 3

factoring a polynomial, 5

automorphism, 25

birational, 25

base

neighbourhood, 70

basis

separating transcendence, 83

transcendence, 81

bound

upper, 66

characteristic

p, 2

zero, 2

closure

separable, 69

cohomology group, 54

commutative, 1

composite of ﬁelds, 10

conjugates, 28

constructible, 13, 33

cubic

resolvent, 38

cyclotomic polynomial, 49

degree, 6

separable, 28

directed, 76

discriminant, 36

Eisenstein’s criterion, 5

element

maximal, 66

extension

abelian, 28

cyclic, 28

ﬁnite, 6

Galois, 27

inseparable, 27

normal, 27

separable, 27

simple, 10

solvable, 28

extension ﬁeld, 6

ﬁeld, 1

prime, 2

stem, 9

ﬁxed ﬁeld, 26

Frobenius

automorphism, 2

endomorphism, 2, 24

fundamental theorem

of algebra, 6, 13, 17, 48

of Galois theory, 29

Galois, 71

Galois closure, 30

Galois correspondence, 75

Galois ﬁeld, 43

Galois group, 27

inﬁnite, 72

of a polynomial, 34

Gaussian numbers, 7

general polynomial, 58

group

Cremona, 25

proﬁnite, 77

topological, 70

homomorphism

crossed, 54

of ﬁelds, 2

of rings, 1

principal crossed, 54

ideal, 1

integral domain, 1

invariants, 26

inverse limit, 76

inverse system, 76

Kummer theory, 56

Lemma

Gauss’s, 4

Maple, 3, 6, 8, 11, 37, 40, 42, 45, 50, 65

module

G-, 53

multiplicity, 22

norm, 55, 62

normal basis, 52

normal closure, 30

ordering

partial, 66

total, 66

perfect ﬁeld, 23

polynomial

minimum, 10

monic, 4

separable, 23

102

INDEX 103

prime

Fermat, 16

primitive element, 46

primitive root of 1, 49

regular n-gon, 51

ring, 1

root

multiple, 22

simple, 22

separable, 46

separable element, 28

separably closed, 69

solvable in radicals, 34

split, 20

splits, 16

splitting ﬁeld, 20

subﬁeld, 1

generated by subset, 10

subring, 1

generated by subset, 9

symmetric polynomial, 58

elementary, 59

theorem

Artin’s, 26

binomial in characteristic p, 2

constructibility of n-gons, 51

constructible numbers, 14, 33

cyclotomic polynomials, 50

Dedekind, 44

Galois 1832, 34

Galois extensions, 27

independence of characters, 51

Liouville, 12

normal basis, 52

primitive element, 46

topology

Krull, 72, 84

trace, 62

transcendence degree, 82

transcendental, 10–12

Abstract These notes give a concise exposition of the theory of ﬁelds, including the Galois theory of ﬁnite and inﬁnite extensions and the theory of transcendental extensions. They are an expansion of those handed out during a course taught to ﬁrst-year graduate students.

v2.01 (August 21, 1996). First version on the web. v2.02 (May 27, 1998). Fixed about 40 minor errors; 57 pages. v3.00 (April 3, 2002). Revised notes; minor additions to text; added 82 exercises with solutions, an examination, and an index; 100 pages. v3.01 (August 31, 2003). Fixed many minor errors; no change to numbering; 99 pages. v4.00 (February 19, 2005). Minor corrections and improvements; added proofs to the section on inﬁnite Galois theory; added material to the section on transcendental extensions; 107 pages. Please send comments and corrections to me at math@jmilne.org Available at http://www.jmilne.org/math/

Copyright c 1996, 1998, 2002, 2003, 2005. J.S. Milne. This work is licensed under a Creative Commons Licence (Attribution-NonCommercial-NoDerivs 2.0) http://creativecommons.org/licenses/by-nc-nd/2.0/

Contents

Notations. 3; References. 3 1 Basic deﬁnitions and results 1 Rings 1; Fields 1; The characteristic of a ﬁeld 2; Review of polynomial rings 3; Factoring polynomials 4; Extension ﬁelds 6; Construction of some extension ﬁelds 7; Stem ﬁelds 9; The subring generated by a subset 9; The subﬁeld generated by a subset 10; Algebraic and transcendental elements 10; Transcendental numbers 12; Constructions with straight-edge and compass. 13; Algebraically closed ﬁelds 16; Exercises 18 Splitting ﬁelds; multiple roots 19 Maps from simple extensions. 19; Splitting ﬁelds 20; Multiple roots 22; Exercises 24 The fundamental theorem of Galois theory 25 Groups of automorphisms of ﬁelds 25; Separable, normal, and Galois extensions 27; The fundamental theorem of Galois theory 29; Examples 31; Constructible numbers revisited 33; The Galois group of a polynomial 34; Solvability of equations 34; Exercises 34 Computing Galois groups. 36 When is Gf ⊂ An ? 36; When is Gf transitive? 37; Polynomials of degree at most three 37; Quartic polynomials 38; Examples of polynomials with Sp as Galois group over Q 40; Finite ﬁelds 41; Computing Galois groups over Q 43; Exercises 45 Applications of Galois theory 46 Primitive element theorem. 46; Fundamental Theorem of Algebra 48; Cyclotomic extensions 49; Dedekind’s theorem on the independence of characters 51; The normal basis theorem 52; Hilbert’s Theorem 90. 53; Cyclic extensions. 56; Proof of Galois’s solvability theorem 57; The general polynomial of degree n 58; Norms and traces 61; Exercises

65

2

3

4

5

6

Algebraic closures 66 Zorn’s lemma 66; First proof of the existence of algebraic closures 67; Second proof of the existence of algebraic closures 67; Third proof of the existence of algebraic closures 67; (Non)uniqueness of algebraic closures 68; Separable closures 69 Inﬁnite Galois extensions 70 Topological groups 70; The Krull topology on the Galois group 71; The fundamental theorem of inﬁnite Galois theory 73; Galois groups as inverse limits 76; Nonopen subgroups of ﬁnite index 77 Transcendental extensions 79 Algebraic independence 79; Transcendence bases 80; L¨ roth’s theorem 83; Separating u transcendence bases 83; Transcendental Galois theory 84 86 91 92

7

8

A Review exercises B Two-hour Examination C Solutions to the Exercises

. Prerequisites Group theory (for example.. Martin Ward (and class). and others. Henry Kim. Also. X=Y X≈Y X Y df X is deﬁned to be Y . X and Y are canonically isomorphic (or there is a given or unique isomorphism). R. . basic linear algebra. Dummit. 5. or equals Y by deﬁnition. 1. .}. Hardy Falk. Volume III — Theory of Fields and Galois Theory. 3. 2005. D. AG Algebraic Geometry. and Foote.org/math/). Shuichi Otsuka.00. Philip Horowitz. van Nostrand. David G. 2003. v5. Jens Hansen. Lectures in Abstract Algebra. A family of elements of A indexed by I. Prentice Hall.Notations. p a prime number. Radcliffe. C = ﬁeld of complex numbers.M. Throughout the notes. Jasper Loy Jiabao. and some elementary theory of rings. denoted (ai )i∈I . Martin Klazar. Albrecht Hess. X is isomorphic to Y . Acknowledgements I thank the following for providing corrections and comments for earlier versions of the notes: Mike Albert. R = ﬁeld of real numbers. Tommaso Centeleghe. Let I and A be sets. 7.jmilne. References. Dmitry Lyubshin. the following of my notes (available at www. Z = ring of integers..11. . v2. 2. Fp = Z/pZ = ﬁeld with p elements. . Abstract Algebra. Antoine Chambert-Loir.10. John McKay. ANT Algebraic Number Theory. Demetres Christoﬁdes. N. Given an equivalence relation. . X ⊂ Y X is a subset of Y (not necessarily proper).. p is a prime number: p = 2.. [∗] denotes the equivalence class containing ∗. 1991. . is a function i → ai : I → A. 11. Jacobson. v2. Dustin Clausen. GT). GT Group Theory. 1998. Trevor Jarvis. We use the standard (Bourbaki) notations: N = {0. 1964.

b ∈ R. in Z (more generally. any Euclidean domain) an ideal I is generated by any “smallest” nonzero element of I. Thus. We usually omit “·” and write 1 for 1R when this causes no confusion. and one of the most important. (b) · is associative. If 1R = 0. We follow Bourbaki in requiring that rings have a 1. b. A subring S of a ring R is a subset that contains 1R and is closed under addition. A ﬁeld is a set F with two composition laws + and · such that (a) (F. ﬁelds is F2 = Z/2Z = {0. +) that is closed under multiplication by elements of R: r ∈ R. c ∈ R.1. +) is a commutative group. all a. For example. passage to the negative. A subﬁeld S of a ﬁeld F is a subring that is closed under passage to the inverse. The smallest. ·). α(1R ) = 1R . A commutative ring is said to be an integral domain if 1R = 0 and the cancellation law holds for multiplication: ab = ac. +) is a commutative group. it is an integral domain. Fields D EFINITION 1. (c) the distributive law holds. In particular. which entails that we require homomorphisms to preserve it. where F × = F {0}. a ∈ I. A ring R is said to be commutative if multiplication is commutative: ab = ba for all a. 0 and 1.1 1 Basic deﬁnitions and results Rings A ring is a set R with two composition laws + and · such that (a) (R. (b) (F × . 1}. An ideal I in a commutative ring R is a subgroup of (R. It inherits the structure of a ﬁeld from that on F . and multiplication. implies b = c. 1 . a ﬁeld is a nonzero commutative ring such that every nonzero element has an inverse. and there exists1 an element 1R such that a · 1R = a = 1R · a for all a ∈ R. (c) the distributative law holds: for all a. b ∈ R. a = 0. (a + b) · c = a · c + b · c a · (b + c) = a · b + a · c. then R = {0}. A homomorphism of rings α : R → R is a map with the properties α(a + b) = α(a) + α(b). We assume that the reader has some familiarity with the elementary theory of rings. implies ra ∈ I. It inherits the structure of a ring from that on R. is a commutative group. α(ab) = α(a)α(b). A ﬁeld contains at least two distinct elements.

which must therefore be zero. The ﬁelds F2 . and so there is a b in F such that ab = 1. F5 . Conversely. C ASE 1: The kernel of the map is (0). then p divides pn − 1. . Q are called the prime ﬁelds. because its kernel is a proper ideal (it doesn’t contain 1). . In this case.2 1 BASIC DEFINITIONS AND RESULTS L EMMA 1. Thus. C. Every ﬁeld contains a copy of exactly one of them. . P ROOF. in this case. and let I be a nonzero ideal in R. The following are ﬁelds: Q. Therefore. then (a) = R. The characteristic of a ﬁeld One checks easily that the map Z → F. Suppose R is a ﬁeld.3. If a = 0. F contains a copy of Q. and so this map extends to a homomorphism m → (m · 1F )(n · 1F )−1 : Q → F. called the Frobenius automorphism. . and so I = R. R EMARK 1. so that n · 1F = 0 =⇒ n = 0 (in Z). called the Frobenius endomorphism of F . suppose R is a commutative ring with no nontrivial ideals. is a homomorphism of rings.2. F3 . C ASE 2: The kernel of the map is = (0). The binomial theorem (a + b)m = am + m 1 am−1 b + m 2 am−2 b2 + · · · + bm pn r holds in any commutative ring. The smallest positive such n will be a prime p (otherwise there will be two nonzero elements in F whose product is zero). n → 1F + 1F + · · · + 1F (n copies).4. so that n · 1F = 0 for some n = 0. If p is prime. . then 1 = a−1 a ∈ I. the map n → n · 1F : Z → F deﬁnes an isomorphism from Z/pZ onto the subring {m · 1F | m ∈ Z} of F . it is an isomorphism. Fp = Z/pZ (p prime). Such a homomorphism is always injective. R. F contains a copy of Fp . When F is ﬁnite. and we say that it has characteristic p. If a is a nonzero element of I. n Thus. A homomorphism of ﬁelds α : F → F is simply a homomorphism of rings. and so its kernel is an ideal in Z. Nonzero integers map to invertible elements of F under n → n · 1F : Z → F . and we say that it has characteristic zero. and p generates the kernel. 2 E XAMPLE 1. n n n for all r with 1 ≤ r ≤ Hence a → ap is a homomorphism F → F . A nonzero commutative ring R is a ﬁeld if and only if it has no ideals other than (0) and R. when F has characteristic p. (a + b)p = ap + bp all n ≥ 1.

”. X n . The ring F [X] of polynomials in the symbol (or “indeterminate” or “variable”) X with coefﬁcients in F is an F -vector space with basis 1.5. X. see Dummit and Foote 1991. Chapter 9. deg(b) < deg(f ). deg(a) < deg(g). hence rn−2 . Moreover. then rf + (b + qf )g = d. 1. Euclid’s algorithm: Let f and g ∈ F [X] have gcd d(X). . r(X) ∈ F [X] with deg(r) < deg(g) such that f = gq + r. . . g). For any ring R containing F as a subring and element r of R.. . .6. q(X) and r(X) are uniquely determined. rn divides rn−1 . Division algorithm: given f (X) and g(X) ∈ F [X] with g = 0.7. Using the division algorithm. 1. 1. If deg(a) ≥ deg(g). . Let af + bg = d. Recall how it goes. . Thus F [X] is a Euclidean domain with deg as norm. . From the division algorithm. type “?gcdex. and b + qf automatically has degree < deg(f ). From unique factorization. and with the multiplication deﬁned by i ai X i j bj X j = k i+j=k ai bj X k . rn = rn−2 − qn rn−1 = rn−2 − qn (rn−3 − qn−1 rn−2 ) = · · · = af + bg and so any common divisor of f and g divides rn : we have shown rn = gcd(f.8. We may assume deg(f ) ≥ deg(g) since the argument is the same in the opposite case.Review of polynomial rings 3 Review of polynomial rings For more on the following. Euclid’s algorithm constructs polynomials a(X) and b(X) such that a(X) · f (X) + b(X) · g(X) = d(X). . . we construct a sequence of quotients and remainders f = q 0 g + r0 g = q 1 r 0 + r1 r0 = q 2 r 1 + r2 ··· rn−2 = qn rn−1 + rn rn−1 = qn+1 rn with rn the last nonzero remainder. it now follows that f has at most deg(f ) roots (see also Exercise 1-3). . . moreover. write a = gq + r with deg(r) < deg(g). there exist q(X). Let F be a ﬁeld. 1. there is a unique homomorphism α : F [X] → R such that α(X) = r and α(a) = a for all a ∈ F . and so is a unique factorization domain. Then. hence g. Maple knows Euclid’s algorithm — to learn its syntax. f (a) = 0) if and only if X − a divides f . it follows that an element a of F is a root of f (that is. and hence f .

Similarly. then it factors nontrivially in Z[X]. g = 0. If f (X) factors nontrivially in Q[X]. and let r = c/d. we can form its ﬁeld of fractions F (X). and so we have a factorization mnf = g1 · h1 in Z[X]. and let f be a nonzero polynomial of least degree in I. Since Fp [X] is an integral domain.10.4 1 BASIC DEFINITIONS AND RESULTS 1.11. d|am . Its elements are quotients f /g. P ROPOSITION 1. we can remove all the prime factors of mn. and therefore. When we choose f to be monic. . and f (1) = 0 = f (−1). If f ∈ Z[X] is monic. 2 P ROPOSITION 1. and in ﬁnding its factors. Let f (X) ∈ Z[X]. If a prime p divides mn. Since F [X] is an integral domain. Suppose r ∈ Q is a root of a polynomial am X m + am−1 X m−1 + · · · + a0 . we have a factorization (mn/p)f = g2 · h1 in Z[X].e.12. P ROOF. this implies that p divides all the coefﬁcients of at least one of the polynomials g1 . then I = (f ) (because F [X] is a Euclidean domain). f and g polynomials. c. to have leading coefﬁcient one. Continuing in this fashion. Factoring polynomials The following results help in deciding whether a polynomial is reducible. 1.9. Let f = gh in Q[X]. d) = 1. The prime ideals correspond to the irreducible monic polynomials. Let I be a nonzero ideal in F [X]. It is clear from the equation am cm + am−1 cm−1 d + · · · + a0 dm = 0 that d|am cm .13 (G AUSS ’ S L EMMA ).. ai ∈ Z. c|a0 . say g1 . then.14. g1 =df mg and h1 =df nh have coefﬁcients in Z. it is uniquely determined by I. P ROPOSITION 1. and so obtain a factorization of f in Z[X]. For suitable integers m and n. so that g1 = pg2 for some g2 ∈ Z[X]. Thus. 2 E XAMPLE 1. Thus. there is a oneto-one correspondence between the nonzero ideals of F [X] and the monic polynomials in F [X]. gcd(c. h1 . Then c|a0 and d|am . The polynomial f (X) = X 3 − 3X − 1 is irreducible in Q[X] because its only possible roots are ±1. P ROOF. i. looking modulo p. d ∈ Z. we obtain an equation 0 = g1 · h1 in Fp [X]. then any monic factor of f in Q[X] lies in Z[X].

Then f is irreducible in Q[X]. that p|b2 . . ci ∈ Z. 2 p R EMARK 1. to obtain a monic polynomial with integer coefﬁcients. 2 The last three propositions hold with Z replaced by any unique factorization domain. By continuing in this way. Let m. Multiply f (X) by a rational number so that it is monic. suppose that there is a prime p such that: — p does not divide am . nf ∈ Z[X].. using nothing but the symmetric polynomials theorem (5. and then replace it by Ddeg(f ) f ( X ). say mg. say. which contradicts the condition that p does not divide am . then they lie in Z.15. . in which case it divides m because g is monic. b1 . P ROPOSITION 1. . and from the equation a2 = b0 c2 + b1 c1 + b2 c0 . . Let g be a monic factor of f in Q[X]. A complex number α is said to be an algebraic integer if it is a root of a monic polynomial in Z[X].14. Thus we need consider only polynomials f (X) = X m + a1 X m−1 + · · · + am . br . but not p2 . . ai ∈ Z. ai ∈ Z. they are algebraic integers.17. with D equal to a common denominator for the coefﬁcients D of f .Factoring polynomials 5 P ROOF. . Since p. If f (X) factors in Q[X]. — p divides am−1 . . There is an algorithm for factoring a polynomial in Q[X]. consider f ∈ Q[X]. because Proposition 1. if a prime p divides mn. Theorem 1. By deﬁnition. c0 . The algebraic integers form a subring of C — for an elementary proof of this. we ﬁnd that p divides b0 .. p must divide exactly one of b0 . To see this.11 shows that every algebraic integer in Q is in Z. s < m. As in the proof of Gauss’s Lemma.14. see AG. — p2 does not divide a0 . . r. which contradicts the deﬁnition of m. Now let α1 . then it divides all the coefﬁcients of at least one of the polynomials mg. divides a0 = b0 c0 . αm be the roots of f in C. a0 . The coefﬁcients of any monic factor of f are polynomials in (certain of) the αi . it factors in Z[X]: am X m + am−1 X m−1 + · · · + a0 = (br X r + · · · + b0 )(cs X s + · · · + c0 ) bi . We sketch an alternative proof of Proposition 1. Let f = am X m + am−1 X m−1 + · · · + a0 . we see that p|b1 .. If they lie in Q. Now m g ∈ Z[X]. P ROOF. . Now from the equation a1 = b0 c1 + b1 c0 . R EMARK 1.30). and therefore are algebraic integers. so that f = gh with h ∈ Q[X] also monic. b0 .16 (E ISENSTEIN ’ S CRITERION ). . nh. n be the positive integers with the fewest prime factors such that mg.

to ﬁnd the factors of f (X) we (better Maple) have to do only a ﬁnite amount of checking. but it is not always effective: for example. X 4 − 10X 2 + 1 is irreducible in Z[X] but it is reducible2 modulo every prime p. One other observation is useful. for example. in fact. which proves that every nonprime integer n ≥ 1 occurs as the degree of a polynomial in Z[X] that is irreducible over Z but reducible modulo all primes. Math. will ﬁnd the factors of 6X 2 + 18X − 24. We write [E : F ] for the dimension. then √ √ X 4 − 10X 2 + 1 = (X 2 − 2 2X − 1)(X 2 + 2 2X − 1). Therefore. Extension ﬁelds A ﬁeld E containing a ﬁeld F is called an extension ﬁeld of F (or simply an extension of F ). i. we need not concern ourselves with the problem of factoring polynomials in Q[X] or Fp [X]. αi ∈ C.e. We say that E is ﬁnite over F when it has ﬁnite degree over F. but several correspondents sent me such proofs.14). Now if g(X) is a monic factor of f (X). in F7 [X]. R EMARK 1. 93 (1986). Since they are also integers (by 1. will ﬁnd the factors of X 2 + 3X + 3 modulo 7. 6 will be a square in Fp . 2 . The general study of such polynomials requires nonelementary methods. if f (X) is irreducible in Fp [X] for some prime p not dividing its leading coefﬁcient. It uses only that the product of two nonsquares in F× is a square.6). it follows that |αi | is less than some bound depending only on the degree and coefﬁcients of f . the simplest of which is the following. we see that there are only ﬁnitely many possibilities for g(X). This test is very useful. then a nontrivial factorization f = gh in Z[X] will give a nontrivial factorization f = gh in Fp [X].18. Monthly. then √ √ X 4 − 10X 2 + 1 = (X 2 − 2 3X + 1)(X 2 + 2 3X + 1). See. which follows from the fact that F× is cyclic (see Exercise 1-3). then it is irreducible in Z[X]. R. of E as an F -vector space. B = max |ai |. For example >factor(6*Xˆ2+18*X-24). Thus. the paper Brandl.. If neither 2 nor 3 are squares. If the leading coefﬁcient of f is not divisible by a prime p. If 2 is a square p p in Fp ..6 1 BASIC DEFINITIONS AND RESULTS From the fundamental theorem of algebra (see 5. and its coefﬁcients are symmetric polynomials in its roots. Thus. Amer. possibly inﬁnite. mB}. the absolute values of the coefﬁcients of g(X) are bounded in terms of the degree and coefﬁcients of f . Let f ∈ Z[X]. since Maple knows how to do it. Thus. From the equation m−1 m 0 = f (αi ) = αi + a1 αi + · · · + am . Such an E can be regarded in an obvious fashion as an F -vector space. I said that I didn’t know an elementary proof of this. we know that f splits completely in C[X]: m f (X) = i=1 (X − αi ). then its roots in C are certain of the αi . and >Factor(Xˆ2+3*X+3) mod 7. |αi | ≤ max{1. If 3 is a square in Fp . and call [E : F ] the degree of E over F . and √ √ X 4 − 10X 2 + 1 = (X 2 − (5 + 2 6))(X 2 − (5 − 2 6)). In an earlier version of these notes. pp286–288.

then it is certainly of ﬁnite degree over E. On putting these together. To complete the proof. and the linear independence of the ei ’s shows that each aij = 0. because (lj )j spans L as an E-vector space. whose powers 1. 2 Construction of some extension ﬁelds Let f (X) ∈ F [X] be a monic polynomial of degree m. assume that L/E and E/F are of ﬁnite degree. some aij ∈ F . α2 . Thus. The linear independence of the lj ’s now shows that i aij ei = 0 for each j. . Let γ ∈ L. even its subspace F [X] has inﬁnite dimension over F (basis 1. (c) The ﬁeld of Gaussian numbers Q(i) = {a + bi ∈ C | a.1≤j≤n is a basis for L over F . P ROOF. P ROPOSITION 1. . is also ﬁnite dimensional. X.19. Second. b ∈ Q} has degree 2 over Q (basis {1. First. there are speciﬁc real numbers α. Moreover. . x is the coset X + (f (X)). α..j spans L. and write x for the image of X in F [X]/(f (X)). (ei lj )i. Let L ⊃ E ⊃ F (all ﬁelds and subﬁelds). Therefore. Then. but a famous argument of Cantor shows that R is not countable. E. (d) The ﬁeld F (X) has inﬁnite degree over F . . we ﬁnd that γ= i.Construction of some extension ﬁelds 7 E XAMPLE 1.e. being a subspace of a ﬁnite dimensional F -vector space. every ﬁnite-dimensional Q-vector space is also countable.). More explicitly. and let (ei )1≤i≤m be a basis for E as an F -vector space and let (lj )1≤j≤n be a basis for L as an E-vector space. in which case [L : F ] = [L : E][E : F ].j is linearly independent. f (x) = 0. (ei lj )i. X 2 . . (a) The ﬁeld of complex numbers C has degree 2 over R (basis {1. αj = i aij ei . Consider the quotient ring F [X]/(f (X)). in fact. π.j aij ei lj . it sufﬁces to show that (ei lj )1≤i≤m. and because (ei )i spans E as an F -vector space. because then L will be ﬁnite over F of the predicted degree. for example. are linearly independent over Q (see the subsection on transcendental numbers p12). Then L/F is of ﬁnite degree if and only if L/E and E/F are both of ﬁnite degree. If L is of ﬁnite degree over F . Then: (a) The map P (X) → P (x) : F [X] → F [x] is a surjective homomorphism in which f (X) maps to 0. γ= j αj lj . (b) The ﬁeld of real numbers R has inﬁnite degree over Q — because Q is countable.20 ( MULTIPLICATIVITY OF DEGREES ). . . df some αj ∈ E. i}). aij ∈ F . and let (f ) be the ideal generated by f . can be rewritten j ( i aij ei )lj = 0. A linear relation aij ei lj = 0. i}). i.

we ﬁnd that β = 3x2 + 7x + 5. (*) (c) To add two elements. It has basis {1. addition: (a + bx) + (a + b x) = (a + a ) + (b + b )x. . (e) Now assume f (X) is irreducible. b ∈ R. Euclid’s algorithm (courtesy of Maple) gives (X 3 − 3X − 1)( −7 X + 37 Hence 7 (3x2 + 7x + 5)( 111 x2 − 26 111 x 29 111 ) 7 + (3X 2 + 7X + 5)( 111 X 2 − 26 111 X + 28 111 ) = 1. Let f (X) = X 2 + 1 ∈ R[X]. Hence b(x) is the inverse of g(x). use (b) to write α = g(x) with g(X) is a polynomial of degree ≤ m − 1.8 1 BASIC DEFINITIONS AND RESULTS (b) From the division algorithm. we can conclude: 1. a. and we have found the inverse of β. ai ∈ F. Let f (X) = X 3 − 3X − 1 ∈ Q[X]. Because X 3 − 3X − 1 is irreducible. simply add the corresponding coefﬁcients. the equality becomes b(x)g(x) = 1. For a monic irreducible polynomial f (X) of degree m in F [X]. Then using that x3 − 3x − 1 = 0. + 28 111 ) = 1. Let β = x4 + 2x3 + 3 ∈ Q[x]. From these observations. multiplication: (a + bx)(a + b x) = (aa − bb ) + (ab + a b)x. computations in F [x] reduce to computations in F. d(X) is 1 because f (X) is irreducible and deg g(X) < deg f (X). expressed in the form (*). 3X 2 + 7X + 5) = 1. gcd(X 3 − 3X − 1. (d) To multiply two elements expressed in the form (*). E XAMPLE 1. and use Euclid’s algorithm in F [X] to obtain polynomials a(X) and b(X) such that a(X)f (X) + b(X)g(X) = d(X) with d(X) the gcd of f and g. E XAMPLE 1.21. Hence each element of F [x] can be expressed uniquely as a sum a0 + a1 x + · · · + am−1 xm−1 . We usually write i for x and C for R[x]. When we replace X with x. To ﬁnd the inverse of an element α ∈ F [x]. In fact. Then R[x] has: elements: a + bx. and so Q[x] is a ﬁeld.23.22. In our case. and use the relation f (x) = 0 to express the monomials of degree ≥ m in x in terms of lower degree monomials. Moreover. x2 } as a Q-vector space. multiply in the usual way. x. we know that each element g of F [X]/(f ) is represented by a unique polynomial r of degree < m. F [x] = F [X]/(f (X)) is a ﬁeld of degree m over F .12) that this is irreducible over Q. We observed in (1.

αi ∈ S. We call it the subring of E generated by F and S (or generated over F by S). . stem ﬁelds always exist.. This is so already in R[i]. . The subring generated by a subset An intersection of subrings of a ring is again a ring. The intersection of all the subrings of E containing F and S is evidently the smallest subring of E containing F and S. For example. Albert. The ring Q[i] consists of the complex numbers of the form a + bi. a stem ﬁeld for f is a pair (E. Let F be a subﬁeld of a ﬁeld E..e. We say that F [α] is a stem ﬁeld3 for f if f (α) = 0. 3 . 1937. Following A.25.. α) consisting of a ﬁeld E containing F and a generator α such that f (α) = 0. R is a subring containing F and S and contained in any other such subring. The ring F [S] consists of the elements of E that can be expressed as ﬁnite sums of the form i in ai1 ···in α11 · · · αn . F [x] = F [X]/f (X).. When S = {α1 . i. ai ∈ Q. Therefore R equals F [S]. 2 Note that the lemma applies to subrings (containing F ) of an extension ﬁeld E of F of ﬁnite degree. Arithmetic in F [α] can be performed using the same rules as in F [x]. 1. If R is ﬁnite dimensional when regarded as an F -vector space.. C = R[ −1]. then it is a ﬁeld. 2 E XAMPLE 1. Note that the expression of an element in the form (*) will not be unique in general. Let R be the set of all such elements. (*) P ROOF. α. In particular. Therefore. αm−1 is a basis for F [α] over F . The ring Q[π]. π = 3. More formally. there is an element β ∈ R such that αβ = 1. Let α be a nonzero element of R — we have to show that α has an inverse in R. If F [α ] is a second stem ﬁeld for f . a. L EMMA 1. ai ∈ F. Evidently. we √ write F [α1 .14159. L EMMA 1.. . . and let S be a subset of E. Then α ↔ x : F [α] F [x].26. m = deg(f ). ai1 ···in ∈ F. Modern Higher Algebra..Stem ﬁelds 9 Stem ﬁelds Let f be a monic irreducible polynomial in F [X].24. Let R be an integral domain containing a subﬁeld F (as a subring). αn }.. P ROOF. b ∈ Q. . The map x → αx : R → R is an injective linear map of ﬁnite dimensional F -vector spaces. and each element of a stem ﬁeld F [α] for f has a unique expression a0 + a1 α + · · · + am−1 αm−1 .. consists of the complex numbers that can be expressed as a ﬁnite sum a0 + a1 π + a2 π 2 + · · · an π n . then there is a unique F -isomorphism F [α] → F [α ] sending α to α . who calls the splitting ﬁeld of a polynomial its root ﬁeld. αn ] for F [S]. and is therefore surjective. . and we denote it F [S]..

There are two possibilities. αn ] consists of all elements of E that can be expressed as polynomials in the αi with coefﬁcients in F . f is the monic polynomial of least degree such f (α) = 0. consists of the complex numbers that can be expressed as a quotient g(π)/h(π). for f ∈ F [X]. It is irreducible. and we denote it F · F . In this case. We call it the subﬁeld of E generated by F and S (or generated over F by S). We call it the composite of F and F in E. The intersection of all the subﬁelds of E containing F and S is evidently the smallest subﬁeld of E containing F and S. We call f the minimum polynomial of α over F . . or the subﬁeld generated over F by F : F (F ) = F · F = F (F ). . we write F (α1 . we say that α is algebraic over F . αn ) consists of all elements of E that can be expressed as the quotient of two such polynomials. in which case F (S) = F [S]. In this case.. The intersection of the subﬁelds of E containing F and F is evidently the smallest subﬁeld of E containing both F and F . Thus. h(X) = 0. and let S be a subset of E. and F (α1 . we have a homomorphism f (X) → f (α) : F [X] → E. so that.. which is generated by the monic polynomial f of least degree such f (α) = 0.27.. h(X) ∈ Q[X].. C ASE 1: The kernel of the map is (0). Algebraic and transcendental elements For a ﬁeld F and an element α of an extension ﬁeld E. αn }. . π = 3. C ASE 2: The kernel is = (0). we say that α transcendental over F . Let F and F be subﬁelds of a ﬁeld E. Let F be a subﬁeld of a ﬁeld E. . F [α1 . . The polynomials g such that g(α) = 0 form a nonzero ideal in F [X].26 shows that F [S] is already a ﬁeld if it is ﬁnite dimensional over F .10 1 BASIC DEFINITIONS AND RESULTS The subﬁeld generated by a subset An intersection of subﬁelds of a ﬁeld is again a ﬁeld. αn ) for F (S). . Lemma 1. . . f (α) = 0 and divides every other polynomial g in F [X] with g(α) = 0. For example.. and it extends to an isomorphism F (X) → F (α). It can also be described as the subﬁeld of E generated over F by F . An extension E of F is said to be simple if E = F (α) some α ∈ E. . When S = {α1 . g(X). . It is the ﬁeld of fractions of F [S] in E. so that g(α) = 0 for some nonzero g ∈ F [X].. .14 . since this is a subﬁeld of E containing F and S and contained in any other such ﬁeld. The ﬁeld Q(π). and we denote it F (S). Q(π) and Q[i] are simple extensions of Q. . The homomorphism F [X] → F [α] is an isomorphism. because otherwise there would be two nonzero elements of E whose product is zero. The minimum polynomial is characterized as an element of F [X] by each of the following sets of conditions: f is monic. E XAMPLE 1. The ring Q[i] is already a ﬁeld. . f (α) = 0 =⇒ f = 0 (in F [X]).

α2 . Now repeat the argument. irreducible. Then X 3 − 3X − 1 is monic. Now type: alias(c=RootOf(Xˆ2+2*X+2)). returns the factorization (X 2 − 2X + 2)(X 2 + 2X + 2).. then any subring R of E containing F is a ﬁeld. P ROPOSITION 1. Since the ﬁrst is a ﬁeld. then it is generated by the empty set.. Let α ∈ C be such that α3 − 3α − 1 = 0. and has α as a root. . or E is said to be algebraic over F . ⇐=: Let E = F (α1 . if E is ﬁnite over F .30. α2 )/F (α1 ) is ﬁnite because α2 is algebraic over F and hence over F (α1 ). αn ) with α1 . then L is algebraic over F. α. Therefore. 2 C OROLLARY 1. A ﬁeld extension E/F is said to be algebraic. R EMARK 1. Thus. and so on. and f (α) = 0. . P ROOF. Since [F [α1 ] : F ] < [F [α1 . Then factor(Xˆ4+4.. F (α1 . α2 ] : F ] < · · · < [E : F ] this process terminates.e. The set {1. there exists an α2 ∈ E F [α1 ].28. (b) If in L ⊃ E ⊃ F . i. Thus. It remains to show that E is ﬁnitely generated over F . The extension F (α1 )/F is ﬁnite because α1 is algebraic over F .. are linearly independent over F . . (a) If E is algebraic over F . Thus. then β = 3α2 + 7α + 5. . Maple has factored X 4 + 4 in Q[c] where c has minimum polynomial X 2 + 2X + 2. and the extension F (α1 . . α2 . . For example.31.20). irreducible. If E = F [α1 ].c). if all elements of E are algebraic over F . F [α] is a stem ﬁeld for f . Note that g(X) → g(α) deﬁnes an isomorphism F [X]/(f ) → F [α]. E/F is transcendental if at least one element of E is transcendental over F . . α2 } is a basis for Q[α] over Q. The calculations in Example 1.23 show that if β is the element α4 + 2α3 + 3 of Q[α]. and β −1 = 7 2 111 α − 26 111 α + 28 111 . L is algebraic over E and E is algebraic over F .29. If E = F . αn algebraic over F . α2 ) is ﬁnite over F . =⇒: To say that α is transcendental over F amounts to saying that its powers 1. E XAMPLE 1. there exists an α1 ∈ E F . .. otherwise it is said to be transcendental (or E is said to be transcendental over F ). then it is algebraic over F . and so it is the minimum polynomial of α over Q. α.Algebraic and transcendental elements 11 f is monic. by ( 1. Maple knows how to compute in Q[α]. returns the factorization (X + c)(X − 2 − c)(X + 2 + c)(X − c). factor(Xˆ4+4). Otherwise. so also is the second: F (α) = F [α]. A ﬁeld extension E/F is ﬁnite if and only if E is algebraic and ﬁnitely generated (as a ﬁeld) over F .

1934: Gel’fond and Schneider independently showed that αβ is transcendental if α and β are algebraic. . . and so on. 1874: Cantor showed that the set of algebraic numbers is countable. are transcendental. that if α is algebraic over F . 2 Transcendental numbers A complex number is said to be algebraic or transcendental according as it is algebraic or transcendental over Q. .) / 2004: Euler’s constant n γ = lim n→∞ 1/k − log n k=1 has not yet been proven to be transcendental or even irrational. . and so α has an inverse in R. and so F [a0 . . However his attempts to decide whether the real numbers were countable proved harder. now called Liouville numbers. where r = m/n is the expression of r in its lowest terms.33. . 5 More precisely. n(2)] onto A(100). am−1 .5 2 1 A typical Liouville number is ∞ 10n! — in its decimal expansion there are increasn=0 ingly long strings of zeros. He also showed that the algebraic numbers. extend it to a bijection from a segment [0. α] is ﬁnite (hence algebraic) over F . n(1)] of N onto A(10). and β ∈ Q. Count the elements of A(10). He had proved that the real numbers were not countable by December 1873 and published this in a paper in 1874 (http://www-gap. If α ∈ R. First some history: 1844: Liouville showed that certain numbers.st-and. .dcs. then count the elements of A(100).ac.uk/∼history/Mathematicians/Cantor. . (a) We observed above (p11). 1873: Hermite showed that e is transcendental.12 1 BASIC DEFINITIONS AND RESULTS P ROOF.4 1882: Lindemann showed that π is transcendental. . choose a bijection from some segment [0. then F [α] is a ﬁeld. P ROOF. then count the elements of A(1000). Now each of the extensions F [a0 . am−1 ] ⊃ F is ﬁnite (1. α] ⊃ F [a0 .32. then F [α] ⊂ R. 1. am−1 . . .20). . and so on. were countable. 4 In 1873 Cantor proved the rational numbers countable. There are only ﬁnitely many rational numbers with height less than a ﬁxed number N . The set of algebraic numbers is countable. 2004: The numbers e + π and e − π are surely transcendental. Then A(N ) is ﬁnite for each N . α = 0. . . . We prove that the analogue of this number in base 2 is transcendental. but that R is not countable. . Thus most numbers are transcendental (but it is usually very difﬁcult to prove that any particular number is transcendental). (b) Any α ∈ L is a root of some monic polynomial f = X m +am−1 X m−1 +· · ·+a0 ∈ E[X].html). . The number α = 1 2n! is transcendental. (This was the seventh of Hilbert’s famous problems. Deﬁne the height h(r) of a rational number to be max(|m|. T HEOREM 1. Let A(N ) be the set of algebraic numbers whose minimum equation over Q has degree ≤ N and has coefﬁcients of height < N . |n|). but again they have not even been proved to be irrational! P ROPOSITION 1. .

and let xN = f (ΣN ). has no rational root. 13 be the minimum polynomial of α over Q. and so |(2N ! )d DxN | ≥ 1. and so d |xN | = i=1 |ΣN − αi | ≤ |ΣN − α1 |(ΣN + M )d−1 . in fact (2N ! )d DxN ∈ Z. and so xN = 0. 2. They then hoped that the “constructible” numbers would sufﬁce. also works for for any integer a ≥ 2. being irreducible of degree > 1. Let ΣN = N 21 . an! In fact α is not rational because its expansion to base 2 is not periodic. 2(N +1)! |xN | ≤ and 2 · (ΣN + M )d−1 2(N +1)! 2d·N ! D · (ΣN + M )d−1 2(N +1)! = 2d 2N +1 N! |(2N ! )d DxN | ≤ 2 · which tends to 0 as N → ∞ because 2d·N ! 2(N +1)! → 0. αi ∈ C. |αi |}. say. Suppose we are given a length. (*) From the fundamental theorem of algebra (see 5. it can’t be a root of f (X).Constructions with straight-edge and compass. Thus [Q[α] : Q] = d. α1 = α. 6 7 1 This proof. xN ∈ Q. and — circles with centre a point already constructed and radius a constructed length. we know that f splits in C[X]. Choose a nonzero integer D such that D · f (X) ∈ Z[X]. Evidently. otherwise. a straight-edge. where M = max{1. and let f (X) = X d + a1 X d−1 + · · · + ad .6 below). and a compass (device for drawing circles). They thus realized that they needed to extend their number system. If α n=0 n! is rational. which I learnt from David Masser. is not rational. This contradicts (*). namely. The Greeks understood integers and the rational numbers. They were surprised to ﬁnd √ that the length of the diagonal of a square of side 1. which we call 1. . d f (X) = i=1 (X − αi ). 2 Constructions with straight-edge and compass. f (X). P ROOF. so that ΣN → α as N → ∞. ai ∈ Q. 6 Suppose not.7 f (X) = X − α. Since ΣN = α. i=1 But |ΣN − α1 | = Hence 1 1 ≤ (N +1)! 2n! 2 n=N +1 ∞ 1 2n n=0 ∞ = 2 . A real number (better a length) is constructible if it can be constructed by forming successive intersections of — lines drawn through two points already constructed.

1991. . a denotes the positive square root of a in R. some e ∈ F . . see Artin. L EMMA 1.36. (a) Immediate from (a) of Lemma 1. . one constructs cd and c−1 . Conversely. By an astute choice of the triangles. (b) It follows from Lemma 1. M. If α is constructible. P ROOF ( SKETCH ). . and let C = C be F -circles.. Algebra. and √ (b) If c > 0 is constructible. First show that it is possible to construct a line perpendicular to a given line through a given point. . An F -circle is a circle in R × R with centre an F -point and radius an element of F . (b) A number α is constructible if and only if it is contained in a subﬁeld of R of the form √ √ √ √ Q[ a1 . 0). ar ] are constructible. . (For more details. a. . −c. .35b). b. . then so also is c. we ﬁnd that all the elements of √ √ Q[ a1 . Section 4. Hence it is possible to construct a triangle similar to a given one on a side with given length. . ai > 0. and [Q[α] : Q] is a power of 2. cd. √ (c) C ∩ C = ∅ or consists of one or two points in the F [ e]-plane. and draw a vertical line through the point A = (1. We make the following deﬁnitions: An F -line is a line in R × R through two points in the F -plane. For (b). . ar ].34 that every constructible number is contained in such √ √ √ √ a ﬁeld Q[ a1 . . . some e ∈ F . . then so also are c+d. (a) The set of constructible numbers is a ﬁeld. 2 L EMMA 1. Chapter 13.. or square the circle by straight-edge and compass constructions? We’ll see that the answer to all three is negative. then α is algebraic over Q. and hence by solving (at worst) a quadratic equation with coefﬁcients in F . For a positive a ∈ F . The points in the intersection are found by solving the simultaneous equations. . . These are the circles given by equations (x − a)2 + (y − b)2 = c2 . . The F -plane is F × F ⊂ R × R. b. . . draw a circle of radius c+1 and centre ( c+1 . e > 0. Applying this for i = 0. . (a) L ∩ L = ∅ or consists of a single F -point. Prentice Hall.35.35. √ Let F be a subﬁeld of R.14 1 BASIC DEFINITIONS AND RESULTS This led them to three famous questions that they were unable to answer: is it possible to duplicate the cube. a. P ROOF. ar ]. ai−1 ] are con√ √ √ structible. . and then a line parallel to a given line through a given point. trisect an angle. . . 0) to 2 2 √ meet the circle at P .34. . P ROOF. 2 C OROLLARY 1. c d (d = 0). (a) If c and d are constructible.37. √ (b) L ∩ C = ∅ or consists of one or two points in the F [ e]-plane. c ∈ F. . and so all the elements of Q[ a1 . The length AP is c. . c ∈ F. 1. . Let L = L be F -lines. These are the lines given by equations ax + by + c = 0. ai ∈ Q[ a1 . e > 0. ai ] are constructible (by (a)). if all the elements of Q[ a1 . then ai is constructible (by 1. .) 2 T HEOREM 1. ai−1 ].

The problem is to construct a cube with volume 2. . P ROOF.11). G. . . in fact X m − 1 = (X − 1)(X m−1 + X m−2 + · · · + 1). it is impossible to trisect an angle by straight-edge and compass constructions. we have to solve 2 C OROLLARY 1. p−2.. . L EMMA 1. and so [Q[ 3 2] : Q] = 3. and Wright. This requires constructing the real root of the polynomial X 3 − 2. so also is π. Now p|ai for i = 1. P ROOF. then f (X + 1) = (X + 1)p − 1 = X p−1 + · · · + a2 X 2 + a1 X + p. Knowing an angle is equivalent to knowing the cosine of the angle. 1960. Oxford. hence Q[e2πi/p ] has degree p − 1 over Q. ] ⊃ Q[cos 2π ] ⊃ Q. for example.16 for example).. p Proofs of this can be found in many books on number theory. Therefore.39. X p with ai = i+1 . E. Note that X m − 1 is not irreducible. that of constructing a regular polygon..41. we have to construct a solution to cos 3α = 4 cos3 α − 3 cos α. H.. 15 √ √ P ROOF.38. p an odd prime. Since π is 2 We now consider another famous old problem. to construct α. to trisect 3α. √ this polynomial is irreducible (by Eisenstein’s But criterion 1. in 11. According to Proposition 1. [Q[α] : Q] divides [Q[ a1 ] · · · [ ar ] : Q] and √ √ [Q[ a1 . ar ] : Q] is a power of 2. 2 C OROLLARY 1. For example. 2 In order to construct a regular p-gon. This implies that f (X) is irreducible. As cos 60◦ = 8x3 − 6x − 1 = 0. An Introduction to the Theory of Numbers.Constructions with straight-edge and compass.20. take 3α = 60 degrees. It is impossible to square the circle by straight-edge and compass constructions. Fourth Edition..16. A square with the same area as a circle of radius r has side √ transcendental8 . P ROOF. M. √ πr. we need to construct cos 2π = (e p But Q[e 8 2πi p 2πi p + (e 2πi p )−1 )/2. Let f (X) = (X p − 1)/(X − 1) = X p−1 + · · · + 1. 1 2.40. In general. It is impossible to duplicate the cube by straight-edge and compass constructions. and so f (X +1) is irreducible by Eisenstein’s criterion 1. P ROOF.14 of Hardy. . If p is prime then X p−1 + · · · + 1 is irreducible. 2 C OROLLARY 1. which is irreducible (apply 1. .

P ROPOSITION 1. Let f be an irreducible polynomial in Ω[X]. For 0 ≤ k ≤ 4. The implications (a) =⇒ (b) =⇒ (c) =⇒ (a) are obvious. 2 [Q[cos 2π ] : Q] = p Thus. are prime but Euler showed that 232 + 1 = (641)(6700417). then (p − 1)/2 = 2k for some k (later (5. whence 2st + 1 = (2s + 1)((2s )t−1 − (2s )t−2 + · · · + 1). 17. which implies p = 2k+1 + 1. Algebraically closed ﬁelds We say that a polynomial splits in F [X] if it is a product of polynomials of degree 1 in F [X]. 65537.30). (d) =⇒ (c). For a ﬁeld Ω. 5. (b) Every nonconstant polynomial in Ω[X] has at least one root in Ω. and claimed to show that this is true for k ≤ 5 — for this reason primes of this form are called Fermat primes. Fermat conjectured k that all numbers of the form 22 +1 are prime. √ 17 + 3 17 − 2 √ 34 − 2 17 − √ 170 − 26 17 . (c) =⇒ (d). (c) The irreducible polynomials in Ω[X] are those of degree 1. and so α ∈ Ω. Let E be a ﬁnite extension of Ω. and we don’t know of any more Fermat primes. we shall see a converse). because otherwise r has an odd factor t and for t odd. P ROOF. Hence p−1 . (d) Every ﬁeld of ﬁnite degree over Ω equals Ω. 257. Gauss showed that9 cos 2π 1 1√ 1 =− + 17+ 17 16 16 16 √ 1 34 − 2 17+ 8 √ 17 + 3 17 − √ √ 34 − 2 17 − 2 34 + 2 17 k when he was 18 years old. Y t + 1 = (Y + 1)(Y t−1 − Y t−2 + · · · + 1).12). shows that it is ≤ 2. This success encouraged him to become a mathematician. the following statements are equivalent: (a) Every nonconstant polynomial in Ω[X] splits in Ω[X].16 2πi p 1 BASIC DEFINITIONS AND RESULTS ] over Q[cos 2π ] is 2 — the equation p α2 − 2 cos 2π · α + 1 = 0. if the regular p-gon is constructible. The minimum polynomial of any element α of E has degree 1. Thus if the regular p-gon is constructible. and it is not 1 because Q[e 2πi p ] is not contained in R.42. 2 9 Or perhaps that 1 8 √ √ 1 1 1 cos 2π = − 16 + 16 17 + 16 34 − 2 17 + 17 — both expressions are correct. Then Ω[X]/(f ) is an extension ﬁeld of Ω of degree deg(f ) (see 1. p α=e 2πi p and the degree of Q[e . then p = 22 + 1 for some k. and so deg(f ) = 1. But 2r + 1 can be a prime only if r is a power of 2. the numbers p = 3.

C OROLLARY 1.42. . we see that it is algebraic over F and every polynomial in F [X] splits in it. Set f = an X n + · · · + a0 . and so the roots of g in Ω all lie in Ω. P ROPOSITION 1. Now Proposition 1. . . β] is a ﬁeld (by 1. β] is algebraic over F . . Let Ω ⊃ F . . then Ω is algebraically closed (hence an algebraic closure of F ). (a) A ﬁeld Ω is said to be algebraically closed when it satisﬁes the equivalent statements of Proposition 1. (b) A ﬁeld Ω is said to be an algebraic closure of a subﬁeld F when it is algebraically closed and algebraic over F . It is an algebraic closure of R. By assumption. We know (see 1.21) that f has a root α in some ﬁnite extension Ω of Ω.Algebraically closed ﬁelds 17 D EFINITION 1. From its deﬁnition. ai ∈ Ω. For any subﬁeld F of Ω. an .46. g splits in Ω[X].30). P ROOF. an ] ⊂ F [a0 .44 shows that it is an algebraic closure of F . If α and β are algebraic over F .44. We have to show that f has a root in Ω. when we admit the fundamental theorem of algebra (5. and hence ﬁnite (by 1.6). Let Ω be an algebraically closed ﬁeld. the algebraic closure of F in Ω is an algebraic closure of F. Later (§6) we shall show that the axiom of choice implies that every ﬁeld has an algebraic closure. P ROOF. including α ± β. If Ω is algebraic over F and every polynomial f ∈ F [X] splits in Ω[X]. 2 Thus. P ROOF.6 below) says that C is algebraically closed. every subﬁeld of C has an algebraic closure (in fact. a canonical algebraic closure). α/β. Thus. and so is algebraic over F — it is a root of a polynomial g with coefﬁcients in F . and consider the ﬁelds F ⊂ F [a0 . α]. α ∈ Ω. . every element of F [α. αβ. . . then F [α. . 2 The ﬁeld constructed in the lemma is called the algebraic closure of F in Ω. Therefore α lies in a ﬁnite extension of F .43.31) of ﬁnite degree over F (by 1. Each extension is algebraic and ﬁnitely generated.30).45. 2 P ROPOSITION 1. then {α ∈ Ω | α algebraic over F } is a ﬁeld. For example. In particular. Let f ∈ Ω[X]. the fundamental theorem of algebra (see 5.

show that G is cyclic. (c) Deduce that f (X) can have at most deg f roots. (d) Let G be a ﬁnite abelian group. 3) : Q]. If G has at most m elements of order dividing m for each divisor m of (G : 1). Let E = Q[α]. Let F be a ﬁeld. 1-3 (*). compass. Show that with straight-edge. 1-1 (*). it is possible to construct a regular 7-gon. Determine [Q( 2. show that there is a polynomial q(X) ∈ F [X] such that f (X) = q(X)(X − a) + f (a). and let f (X) ∈ F [X]. (a) For any a ∈ F . and angle-trisector. where α3 − α2 + α + 2 = 0. Express (α2 + α + 1)(α2 − α) and (α − 1)−1 in the form aα2 + bα + c with a. F a ﬁeld.18 1 BASIC DEFINITIONS AND RESULTS Exercises Exercises marked with an asterisk were required to be handed in. . (e) Deduce that a ﬁnite subgroup of F × . √ √ 1-2 (*). is cyclic. 1-4 (*). b. c ∈ Q. (b) Deduce that f (a) = 0 if and only if (X − a)|f (X).

and the map ϕ → ϕ(α) deﬁnes a one-to-one correspondence {F -homomorphisms ϕ : F (α) → Ω} ↔ {elements of Ω transcendental over F }. then the map F [X] → Ω. then the map ϕ → ϕ(α) deﬁnes a one-to-one correspondence {extensions ϕ : F (α) → Ω of ϕ0 } ↔ {elements of Ω transcendental over ϕ0 (F )}. (a) To say that α is transcendental over F means that F [α] is isomorphic to the polynomial ring in the symbol α with coefﬁcients in F . Thus an F -homorphism ϕ maps a polynomial i im ai1 ···im α11 · · · αm . An F -homomorphism is a homomorphism ϕ: E → E such that ϕ(a) = a for all a ∈ F . On applying ϕ to the equation ai αi = 0. there is a unique F -homomorphism ϕ : F [α] → Ω sending α to γ (see 1. An F -isomorphism is a bijective F -homomorphism. then every F -homomorphism is an F -isomorphism. ϕ(α) is transcendental over F .19 2 Splitting ﬁelds. When composed with the inverse of the isomorphism X + f (X) → α : F [X]/(f (X)) → F [α]. if γ ∈ Ω is a root of f (X). and let ϕ0 : F → Ω be a homomorphism of F into a second ﬁeld Ω. For every F -homomorphism ϕ : F (α) → Ω. (b) Let α be algebraic over F with minimum polynomial f (X). ϕ(α) is a root of f (X) in Ω. P ROOF. P ROPOSITION 2. we obtain the equation ai ϕ(α)i = 0. This extends to the ﬁeld of fractions F (α) of F [α] if and only if all nonzero elements of F [α] are sent to nonzero elements of Ω. and the map ϕ → ϕ(α) deﬁnes a oneto-one correspondence {F -homomorphisms ϕ : F [α] → Ω} ↔ {roots of f in Ω}. 2 We shall need a slight generalization of this result. (b) Let f (X) = ai X i . Let F (α) be a simple ﬁeld extension of a ﬁeld F . For any γ ∈ Ω. multiple roots Maps from simple extensions. this becomes a homomorphism F [α] → Ω sending α to γ. which shows that ϕ(α) is a root of f (X) in Ω.1. factors through F [X]/(f (X)). Let E and E be ﬁelds containing F . P ROPOSITION 2. Conversely. For every F -homomorphism ϕ : F [α] → Ω. and consider an F -homomorphism ϕ : F [α] → Ω. (a) If α is transcendental over F . ai1 ···im ∈ F. Note that if E and E have the same ﬁnite degree over F . Let F (α) be a simple ﬁeld extension of a ﬁeld F . to ai1 ···im ϕ(α1 )i1 · · · ϕ(αm )im . In particular.2.5). . which is so if and only if γ is transcendental. the number of such maps is the number of distinct roots of f in Ω. g(X) → g(γ). (a) Let α be transcendental over F . and let Ω be a second ﬁeld containing F .

and so [Q[α1 . α2 ] : Q[α1 ]] = 1 or 2. √ E XAMPLE 2. For example. For a given integer n. Every polynomial f ∈ F [X] has a splitting ﬁeld Ef . the discriminant of X 3 + bX + c is −4b3 − 27c2 . In particular. then the map ϕ → ϕ(α) deﬁnes a one-to-one correspondence {extensions ϕ : F [α] → Ω of ϕ0 } ↔ {roots of ϕ0 f in Ω}. Note that fi (X)mi (mi ≥ 1) and fi (X) have the same splitting ﬁelds.20 2 SPLITTING FIELDS.. P ROOF.. and let α = b2 − 4ac. there may or may not exist polynomials of degree n in F [X] whose splitting ﬁeld has degree n! — this depends on F . E is generated by the i=1 roots of f . we arrive at a splitting ﬁeld Ef . with minimum polynomial f (X).5. . then it is called a splitting or root ﬁeld for f . nor for n > 2 if F = Fp (see 4. If. . . MULTIPLE ROOTS (b) If α is algebraic over F . α3 ] = Q[α1 . E = F [α1 . Note that [Q[α1 ] : Q] = 3 and that [Q[α1 . and so the splitting ﬁeld of X 3 + 10X + 1 has degree 6 over Q. Also. Let n = deg f . P ROPOSITION 2. A ﬁeld E containing F is said to split f if f splits in E[X]: f (X) = m (X − αi ) with αi ∈ E.6. [F2 : F1 ] ≤ n − 1.4. (a) Let f (X) = aX 2 + bX + c ∈ Q[X]. and let α1 . and so the splitting ﬁeld of f is Q[ζ]. α2 ] : Q] = 3 or 6. Then [F1 : F ] = deg g1 ≤ n. Then f (α1 ) = 0. However. . . the number of such maps is the number of distinct roots of ϕ0 f in Ω. The proof of the proposition is essentially the same as that of the preceding proposition. in addition. later (4. and we let F2 = F1 [α2 ] be a stem ﬁeld for some monic irreducible factor of f (X)/(X − α1 ) in F1 [X]. α2 . αm ]. then it splits in E[X]. The subﬁeld Q[α] of C is a splitting ﬁeld for f . . . and so [Ef : E] ≤ n!. Let F1 = F [α1 ] be a stem ﬁeld for some monic irreducible factor of f in F [X]. 2 R EMARK 2. Continuing in this fashion. E XAMPLE 2. Then Q[α1 . p prime.2) that the degree is 3 if and only if the discriminant of f (X) is a square in Q. If ζ is one root of f . . For example. . there do not for n > 1 if F = C (see 5. We’ll see later (4. α2 ] is a splitting ﬁeld for f (X). α3 be its roots in C. then the remaining roots are ζ 2 . that if f has deg(f ) − 1 roots in E. .30) we shall see how to write down large numbers of polynomials (in fact inﬁnitely many) of degree n in Q[X] whose splitting ﬁelds have degree n!.6). By ϕ0 f we mean the polynomial obtained by applying ϕ0 to the coefﬁcients of f : if f = ai X i then ϕ0 f = ϕ(ai )X i .20) or F = R.. and [Ef : F ] ≤ (deg f )!.3. Splitting ﬁelds Let f be a polynomial with coefﬁcients in F . (a) Let f (X) = (X p − 1)/(X − 1) ∈ Q[X]. By an extension of ϕ0 to F (α) we mean a homomorphism ϕ : F (α) → Ω such that ϕ|F = ϕ0 . α2 . (b) Let f (X) = X 3 + aX 2 + bX + c ∈ Q[X] be irreducible. ζ p−1 . ζ 3 .

αm ]. and that the number of such homomorphisms is ≤ [F [α1 . (c) If E and Ω are both splitting ﬁelds for f . α2 ] : F ].1 shows that there exists an F -homomorphism ϕ1 : F [α1 ] → Ω.. and the number of ϕ1 ’s is ≤ deg(f1 ) = [F [α1 ] : F ]. and the number of such homomorphisms is ≤ [E : F ]. with equality holding if f has deg(f ) distinct roots in Ω. any two splitting ﬁelds for f are F -isomorphic. . The minimum polynomial of α2 over F [α1 ] is an irreducible factor f2 of f in F [α1 ][X]. . By f having deg(f ) distinct roots in Ω.. α2 ] → Ω. On combining these statements we conclude that there exists an F -homomorphism ϕ : F [α1 . Therefore. Let E and L be extension ﬁelds of F . Any F -homomorphism E → Ω is injective. then each F -homomorphism E → Ω is an isomorphism... then the remaining roots are all of the form ζα. If E and Ω are both splitting ﬁelds for f then (a) shows that [E : F ] = [Ω : F ]. Proposition 2. and so F automatically contains all the pth roots of 1. (a) The number of F -homomorphisms E → L is at most [E : F ]. . Write E = F [α1 .8.. we obtain (a) and (b). As f (hence f1 ) splits in Ω. By assumption. This proves (b). . α2 ] → Ω. it also proves (a). P ROOF. with equality holding when f2 has deg(f2 ) distinct roots in Ω. with equality holding when f1 has distinct roots in Ω. Assume that E ⊃ F is generated by roots of f . . and the number of extensions is ≤ deg(f2 ) = [F [α1 . According to Proposition 2. If f has this property. (c) If α is one root of X n − a. we mean that f (X) = deg(f ) i=1 (X − αi ). where ζ n = 1. with E ﬁnite over F .2. 2 R EMARK 2. α + p − 1. and f be the product of the minimum polynomials of the αi . . Let f ∈ F [X]. then so also does any factor of f in Ω[X]. if there exists such a homomorphism. [E : F ] ≤ [Ω : F ]. then X p − 1 = (X − 1)p . αm ] with the αi roots of f (X).9. The minimum polynomial of α1 is an irreducible polynomial f1 dividing f . and so. and since an F -homomorphism E → L can be regarded as an F -homomorphism E → Ω. Note that if p is the characteristic of F . Let E1 . αi = αj if i = j. and equals [E : F ] if f has deg(f ) distinct roots in Ω. (a) There exists at least one F -homomorphism ϕ : E → Ω. Em be ﬁnite extensions of F . if F contains all the nth roots of 1 (by which we mean that X n − 1 splits in F [X]). then the remaining roots are α + 1. . 2 C OROLLARY 2. and let Ω ⊃ F be a ﬁeld in which f splits. and so any F -homomorphism E → Ω is an isomorphism.. and so any ﬁeld generated over F by α is a splitting ﬁeld for f (and F [α] F [X]/(f ) if f is irreducible).Splitting ﬁelds 21 (b) Suppose F is of characteristic p.7. E = F [α1 . . . If α is one root of f . . then F [α] is a splitting ﬁeld for X n − a. each ϕ1 extends to a homomorphism ϕ2 : F [α1 . and let L be an extension of F . (b) There exists a ﬁnite extension Ω/L and an F -homomorphism E → Ω. In particular. . P ROOF. αi ∈ Ω. Let Ω be a splitting ﬁeld for f regarded as an element of L[X]. P ROPOSITION 2. α2 ] : F [α1 ]]. (b) The number of F -homomorphisms E → Ω is ≤ [E : F ]. The proposition shows that there is an F -homomorphism E → Ω. The corollary implies that there exists a ﬁnite extension Ω/L containing an isomorphic copy of every Ei . and let f = X p − X − a ∈ F [X]. After repeating the argument several times. E2 .

a = T in the ﬁeld Fp (T ). and let Ω ⊃ F .22 2 SPLITTING FIELDS. and otherwise it is a simple root. Certainly. Let rF (X) and rΩ (X) be the greatest common divisors of f and g in F [X] and Ω[X] respectively. P ROPOSITION 2. . . In fact. . Even when f and g have no common factor in F [X]. We say that f has a multiple root when at least of the mi > 1. If r(X) is the gcd of f and g computed in F [X]. g ∈ F [X]. .4 is irreducible in F [X]. say f = fi (X)mi with some mi > 1. Let F be of characteristic p = 0. αm ]. E XAMPLE 2. Multiple roots Let f. . Error and confusion can result if you simply identify the ﬁelds. Let f ∈ F [X].11. If f has a multiple factor in F [X]. note that the hypotheses imply that gcd(f. but there will in general be no preferred such isomorphism. and any two splitting ﬁelds are isomorphic (2. and let r r f (X) = a i=1 (X − αi ) mi . In particular. Let f and g be polynomials in F [X]. but X p − a = (X − α)p in its splitting ﬁeld. this doesn’t happen — greatest common divisors don’t change when the ﬁeld is extended. . and assume that F has contains an element a that is not a pth -power. one might expect that they could acquire a common factor in Ω[X] for some Ω ⊃ F . We wish to determine when a polynomial has a multiple root. αi distinct. mi ≥ 1. . Then X p − a 1. αm ] is a splitting ﬁeld for f . Thus an irreducible polynomial can have multiple roots. then obviously it will have a multiple root. distinct monic irreducible polynomials in F [X] do not acquire a common root in any extension ﬁeld of F. The unordered sequence of integers m1 . and so rΩ (X) divides rF (X) in Ω[X]. then it is also the gcd of f and g in Ω[X].7c). then Proposition 2. .10 shows that f has a multiple root if and only if at least one of the fi has a multiple root.8) shows that there are polynomials a and b in F [X] such that a(X)f (X) + b(X)g(X) = rF (X). for example. but F [α1 . it sufﬁces to determine when an irreducible polynomial has a multiple root. g) = 1 (in F [X]). mr in (*) is independent of the extension ﬁeld Ω in which f splits. MULTIPLE ROOTS Warning! If E and E are both splitting ﬁelds of f ∈ F [X]. We say that αi is a root of f of multiplicity mi . then we know there is an F -isomorphism E → E . . and we say that f has only simple roots when all mi = 1. . i=1 mi = deg(f ). P ROOF. (*) be a splitting of f in some extension ﬁeld Ω of F . it is unchanged when Ω is replaced with its subﬁeld F [α1 . 2 The proposition allows us to speak of the greatest common divisor of f and g without reference to a ﬁeld. αi is said to be a multiple root of f . and so f and g can’t acquire a common factor in any extension ﬁeld. If f = fi with the fi distinct monic irreducible polynomials. For the second statement. Certainly rF (X)|rΩ (X) in Ω[X]. Thus. . If mi > 1.10. but Euclid’s algorithm (1. . .

but note that in characteristic p the derivative of X p is zero.12 continues to hold — see 2. (a) =⇒ (b). f ) = 1 =⇒ f = 0. (c) F has characteristic p = 0 and f is a polynomial in X p . give a different deﬁnition. When f has coefﬁcients in R.10). (b) =⇒ (c). this agrees with the deﬁnition in calculus. Dummit and Foote 1991. (d) all the roots of f are multiple. and so gcd(f.5. m > 1.13. Since f is irreducible and deg(f ) < deg(f ). Let α be a multiple root of f . f can be zero only if the characteristic is p = 0 and f is a polynomial in p. in some splitting ﬁeld. (c) =⇒ (d).15.Multiple roots 23 Deﬁne the derivative f (X) of a polynomial f (X) = ai X i to be iai X i−1 . because f is nonconstant. then it remains separable over every ﬁeld Ω containing F (condition (b) of 2. although some authors. if f ∈ F [X] is separable. and let g(X) = i (X − ai )mi in some splitting ﬁeld for f . and (b) at least one of the irreducible factors of f is a polynomial in X p . Note that. For a nonconstant irreducible polynomial f in F [X]. 2 D EFINITION 2.14. all irreducible polynomials in F [X] are separable). Then f (X) = m(X − α)m−1 g(X) + (X − α)m g (X). 13. P ROPOSITION 2. gcd(f. Then f (X) = g(X p ) = i (X p − ai )mi = i (X − αi )pmi p where αi = ai . P ROPOSITION 2. f ) = 1. P ROOF.10 The preceding discussion shows that f ∈ F [X] will be separable unless (a) the characteristic of F is p = 0. (d) =⇒ (a). This is the standard deﬁnition. and a ﬁeld F of characteristic p = 0 is perfect if and only if every element of F is a pth power. The usual rules for differentiating sums and products still hold. for example. D EFINITION 2. Suppose f (X) = g(X p ). A ﬁeld of characteristic zero is always perfect. Obvious. But. Hence every root of f (X) has multiplicity at least p. A polynomial f ∈ F [X] is said to be separable over F if none of its irreducible factors has a multiple root (in a splitting ﬁeld).12. f ) = 1. A ﬁeld F is said to be perfect if all polynomials in F [X] are separable (equivalently. and write f = (X −α)m g(X). 10 . the following statements are equivalent: (a) f has a multiple root. Hence f (α) = 0. (b) gcd(f.

Let d(X) = gcd(f. (b) Let E and E be quadratic extensions of F . Construct a splitting ﬁeld for X 5 − 2 over Q. then X p − a ∈ F [X] is not separable (see 2. there is exactly one ﬁeld with p2 elements. Exercises 2-1 (*). . f ). 2-6 (*). i 2 E XAMPLE 2. What is its degree over Fp ? 2-5.16. (c) Show that there is an inﬁnite sequence of ﬁelds E1 . Let f ∈ F [X]. and so we may suppose F to be of characteristic p = 0. where F has characteristic p. 2-2 (*). (a) Let E be quadratic extension of F (i. A ﬁeld of characteristic zero is obviously perfect. m . E2 . . e Show that f (X) can be written f (X) = g(X p ) where g(X) is irreducible and separable. (a) A ﬁnite ﬁeld F is perfect.. bi X)p if ai = bp . Show that g(X) = f (X)d(X)−1 has the same roots as f (X). What is its degree over Q? 2-4 (*). (a) Let F be a ﬁeld of characteristic p. (d) Let p be an odd prime. (d) If F0 has characteristic p = 0. MULTIPLE ROOTS P ROOF.e. then F = F0 (X) is not perfect. then it splits into distinct factors in F [X]. Show that. then every polynomial in X p with coefﬁcients in F is a pth power in F [X]. Conversely. Let f (X) be an irreducible polynomial in F [X]. because X is not a pth power. show that S(E) = {a ∈ F × | a is a square in E} is a subgroup of F × containing F ×2 . ai X p = ( and so is not irreducible. Find a splitting ﬁeld of X p − 1 ∈ Fp [X]. (c) Every algebraically closed ﬁeld is perfect. show that there is an F -isomorphism ϕ : E → E if and only if S(E) = S(E ). because the Frobenius endomorphism a → ap : F → F is injective and therefore surjective (by counting). (b) For any prime p. Therefore. Deduce that every root of f (X) has the same multiplicity pe in any splitting ﬁeld. up to isomorphism.24 2 SPLITTING FIELDS. every ﬁeld algebraic over Fp is perfect. and these are all simple roots of g(X). if every element of F is a pth power. . If F contains an element a that is not a pth power. Let F be a ﬁeld of characteristic = 2.11). where F is a ﬁeld of characteristic 0. with Ei a quadratic extension of Q such that Ei is not isomorphic to Ej for i = j. show that X p − X − 1 is irreducible in Q[X]. 2-3 (*). (b) A ﬁeld that can be written as a union of perfect ﬁelds is perfect. [E : F ] = 2). Show that if X p − X − a is reducible in F [X].

) In this section. cX+d . (b) Let F be a ﬁeld of characteristic p = 0. 2 E XAMPLE 3. and so has deg f distinct roots in E. IV. (b) Let E = C(X).2. if 3 2 denotes the real cube root of 2. (The group Aut(C(X1 . Hence we may assume f is a product of distinct monic separable irreducible polynomials. Thus. If f has no root in E other than α. The splitting ﬁeld of f is the same as the splitting ﬁeld of fi .7 shows that there are [E : F ] distinct F -homomorphisms E → E. Its study is part of algebraic geometry. they are automatically isomorphisms. Theorem 7. E XAMPLE 3.25 3 The fundamental theorem of Galois theory In this section. Thus. Then f = X p − a has only one root in a splitting ﬁeld E. If E is a splitting ﬁeld of a monic separable polynomial f ∈ F [X]. Groups of automorphisms of ﬁelds Consider ﬁelds E ⊃ F . . Analysts will note that this is the same as the automorphism group of the Riemann sphere. ad − bc = 0 cX+d (Jacobson 1964. and so Aut(E/F ) = 1. then Aut(Q[ 3 2]/Q) = 1.18) that by using the Axiom of Choice one can construct uncountably many more. X2 )/C). Now Proposition 2. the identity map and complex conjugation. the group of invertible 2 × 2 matrices with complex coefﬁcients modulo its centre. with the fi monic irreducible and distinct. (a) Consider a simple extension E = F [α]. namely. then Aut(E/F ) = 1. (c) The group Aut(C(X1 . C but this is very far from being surjective. the group is unknown. We’ll see later (8. Let f = fimi . which we denote Aut(E/F ). and let a be an element of F that is not a pth power. and so Aut(E/C) = PGL2 (C). An F -isomorphism E → E is called an F -automorphism of E. . which gives a one-toone correspondence between the subﬁelds of the splitting ﬁeld of a separable polynomial and the subgroups of the Galois group of f . in the proposition. we prove the fundamental theorem of Galois theory. P ROOF. we shall be concerned with the groups Aut(E/F ) when E is a ﬁnite extension of F . P ROPOSITION 3. This is not a coincidence: the ﬁeld of meromorphic functions on the Riemann sphere P1 is C C(z) C(X). The F -automorphisms of E form a group. . and so there is certainly a map Aut(P1 ) → Aut(C(z)/C). . It is C called the Cremona group. (a) There are two obvious automorphisms of C. p158).1. X2 )/C) is quite complicated — there is a map PGL3 (C) = Aut(P2 ) → Aut(C(X1 . Xn )/C) is the group of birational automorphisms of Pn . √ For example. Because E has ﬁnite degree over F . 11 By this I mean the map that sends a rational function f (X) to f ( aX+b ). in the proposition.3. it is essential that E be a splitting ﬁeld of a separable polynomial. which one can C show to be an isomorphism. then Aut(E/F ) has order [E : F ]. When there are more X’s. Then Aut(E/C) consists of the maps11 X → aX+b . and let f be a polynomial with coefﬁcients in F √ having α as a root. it is essential that E be a splitting ﬁeld.

. Then the ﬁrst equation α1 c1 + · · · + αn cn = 0 (recall that σ1 = 1) will be a linear relation on the αi . P ROOF. [E : E G ] ≤ (G : 1) ≤ (Aut(E/E G ) : 1) ≤ [E : E G ] must be equalities. αn be n > m elements of E. For any ﬁnite group G of automorphisms of a ﬁeld E. With these normalizations. . . σk (ci ). . . After renumbering the αi ’s. . G = Aut(E/E G ). σm }. and let F = E G . then σk (ci ) = ci for some k and i. . T HEOREM 3. . we may suppose that c1 = 0. and the set of subgroups H of G. . . .). σk (c2 ).8a). 2 C OROLLARY 3. . when E is the splitting ﬁeld of a separable polynomial in F [X] and G = Aut(E/F ). . σm }. Let G be a ﬁnite group of automorphisms of a ﬁeld E. we set E G = Inv(G) = {α ∈ E | σα = α. and then (after multiplying by a scalar) that c1 ∈ F . cn ) having the fewest possible nonzero elements. we shall show that.) is also a solution to the system of equations (*). we’ll show that all ci ∈ F . P ROOF. σk σm } is a permutation of {σ1 . . F ⊂ M ⊂ E. but has more zeros than the ﬁrst solution (look at the ﬁrst coordinate) — contradiction. which is nonzero (look at the ith coordinate). ci − σk (ci ). . . and so G = Aut(E/E G ). . . . We know that: [E : E G ] ≤ (G : 1) G ⊂ Aut(E/E G ) (Aut(E/E G ) : 1) ≤ [E : E G ] The inequalities (by 3. . On applying σk to the equations σ1 (α1 )c1 + · · · + σ1 (αn )cn = 0 ··· ··· σm (α1 )c1 + · · · + σm (αn )cn = 0 and using that {σk σ1 . On subtracting it from the ﬁrst. In this section. . . Let G = {σ1 = 1. If not all ci are in F . all σ ∈ G}. (obvious). . . we ﬁnd that (c1 .26 3 THE FUNDAMENTAL THEOREM OF GALOIS THEORY When G is a group of automorphisms of a ﬁeld E. . 2 . In the system of linear equations σ1 (α1 )X1 + · · · + σ1 (αn )Xn = 0 ··· ··· σm (α1 )X1 + · · · + σm (αn )Xn = 0 there are m equations and n > m unknowns. k = 1 = i. . then the maps M → Aut(E/M ). . A RTIN ).5. then [E : F ] ≤ (G : 1). . H → Inv(H) give a one-to-one correspondence between the set of intermediate ﬁelds M . . . We shall show that the αi are linearly dependent over F . . It is a subﬁeld of E. . (by 2.4). we obtain a solution (0. and hence there are nontrivial solutions in E — choose one (c1 . and let α1 . .4 (E. called the subﬁeld of G-invariants of E or the ﬁxed ﬁeld of G. .

(a) The ﬁeld Q[ 3 2].Separable. we conclude that F = F . and it is denoted Gal(E/F ). the following statements are equivalent: (a) E is the splitting ﬁeld of a separable polynomial f ∈ F [X]. it is ﬁnite. (d) E is Galois over F . and f is still separable when it is regarded in this way. we have to prove that f splits . for each α ∈ E. (d) =⇒ (b). Hence Proposition 3. P ROOF. the minimum polynomial of α has [F [α] : F ] distinct roots in E. and Galois extensions D EFINITION 3. Let f be an irreducible polynomial of degree m in F [X]. and so this is obvious. and so F = E G . Let α ∈ E and let f be the minimum polynomial of α. In other words. where 3 2 is the real cube root of 2. g ∈ F [X].8a) .10.8. over F . By Proposition 3. is separable but not normal over Q (X 3 − 2 doesn’t split in Q[α]). and in particular has characteristic p = 0. (b) =⇒ (c). Therefore. (b) The ﬁeld Fp (T ) is normal but not separable over Fp (T p ) — the minimum polynomial of T is the inseparable polynomial X p − T p . and Galois extensions 27 Separable. A ﬁnite extension E of F is said to Galois if F is the ﬁxed ﬁeld of the group of F -automorphisms of E. D EFINITION 3. If f has a root in E. An algebraic extension E/F is said to be separable if the minimum polynomial of every element of E is separable.7. According to (2. and let F = E G ⊃ F . otherwise. E/F is normal and separable if and only if. This group is then called the Galois group of E over F . an algebraic extension E/F is separable if every irreducible polynomial in F [X] having a root in E is separable.6. (b) F = E G for some ﬁnite group G of automorphisms of E. normal. Since Aut(E/F ) = Aut(E/F ) = G. it is inseparable. and it is inseparable if — F is nonperfect. Then E is also the splitting ﬁeld of f regarded as a polynomial with coefﬁcients in F . E = Fp (T ) is an inseparable extension of Fp (T p ). an algebraic extension E/F is normal if every irreducible polynomial f ∈ F [X] having a root in E splits in E.2 shows that [E : F ] = # Aut(E/F ) [E : F ] = # Aut(E/F ). T HEOREM 3. For example. An algebraic extension E/F is normal if the minimum polynomial of every element of E splits in E[X]. For an extension E/F .4. Gal(E/F ) is ﬁnite. and of ﬁnite degree. we know that [E : F ] ≤ (G : 1). then E/F separable E/F normal =⇒ =⇒ roots of f distinct f splits in E =⇒ f has m distinct roots in E. Thus. (a) =⇒ (d). D EFINITION 3. Let F be a ﬁeld. (c) E is normal and separable. in particular.9. Let G = Aut(E/F ). and — there is an element α of E whose minimal polynomial is of the form g(X p ). √ √ E XAMPLE 3. normal.

Let fi be the minimum polynomial of αi over F . When E ⊃ M ⊃ F (ﬁnite extensions). . then applying σ to the equation f (α) = 0 gives f (αi ) = 0). If Ω is an algebraically closed ﬁeld containing F . and so f (X) splits into distinct factors in E. because each αi is a root of f (X) (if αi = σα. and let α ∈ E. write [E : F ]sep = [Esep : F ] (separable degree of E over F ). we can still say something. Let {α1 = α. .. αi ∈ E. It is monic. (b) Note that if F = E G for some ﬁnite group G. (a) Let E be Galois over F with Galois group G. if E is Galois over F . Proposition 2. . and g(α) = 0.. Combined with Artin’s theorem (3.12..4). αi algebraic over F . and so the number of F -homomorphisms E → Ω is [E : F ]sep . Since the ai are symmetric polynomials in the αi . αm ]. f is separable. Any σ ∈ G merely permutes the αi . P ROOF. Let fi be the minimum polynomial of αi over F . Let E = F [α1 . Every ﬁnite separable extension E of F is contained in a ﬁnite Galois extension. In particular.. But also g(X)|f (X). solvable extension if it is Galois with cyclic.. 2 C OROLLARY 3. In the course of the proof of (b) =⇒ (c) of the above theorem we showed that the minimum polynomial of α is (X − αi ). then every F -homomorphism Esep → Ω extends uniquely to E. 2 R EMARK 3. and so f (X)|g(X) (see the deﬁnition of the minimum polynomial p10). abelian. and let g(X) = (X − αi ) = X m + a1 X m−1 + · · · + am . We conclude that f (X) = g(X). 2 R EMARK 3... αm ]. for more details.... D EFINITION 3.. Because E is separable over F . because E is the splitting ﬁeld of a separable polynomial. Let E ⊃ M ⊃ F . When we drop the assumption that E is separable over F . each fi splits in E. and so g(X) ∈ F [X]. Therefore. it is also the splitting ﬁeld of f regarded as an element of M [X]. . then.14. [E : F ]sep = [E : M ]sep [M : F ]sep . and take E to be the splitting ﬁeld of fi over F .12 shows that any ﬁnite extension generated by separable elements is separable. solvable Galois group. Because E is normal over F . (c) =⇒ (a). E = F [α1 . we ﬁnd that σai = ai for all i. The elements α1 = α.. We know E is the splitting ﬁeld of some separable f ∈ F [X].1 shows that Gal(E/F ) has [E : F ] elements..11.. .28 3 THE FUNDAMENTAL THEOREM OF GALOIS THEORY into distinct factors in E[X]. αm } be the orbit of α under the action of G on E. Because E has ﬁnite degree over F . E is separable over F ⇐⇒ E is separable over M and M is separable over F. this shows that G = Gal(E/F ) and (G : 1) = [E : F ]. An element α of an algebraic extension of F is said to be separable over F if its minimum polynomial over F is separable.. then it is Galois over M. . C OROLLARY 3.. . abelian. The proof of Corollary 3.15.. the elements of a ﬁnite extension E of F that are separable over F form a subﬁeld Esep of E that is separable over F . A ﬁnite extension E ⊃ F is called a cyclic. it is generated over F by a ﬁnite number of elements. and so E is the splitting ﬁeld of f = fi . I 10. α2 .. See Jacobson 1964. say.13. αm of the orbit of α are called the conjugates of α. P ROOF.

10) and that G/H Gal(E H /F ) (by 3. (a) Let M1 . Moreover.. Gal(E/σM ) = σ Gal(E/M )σ −1 . 2 R EMARK 3. E σHσ = σ(E H ). Gal(E/σM ) = σ Gal(E/M )σ −1 . and write M = F [α1 . . (c) For τ ∈ G and α ∈ E.e. assume that M is normal over F . Conversely. and so lies in M .. which means that Gal(E/M ) = M . . The maps H → E H and M → Gal(E/M ) are inverse bijections between the set of subgroups of G and the set of intermediate ﬁelds between E and F : {subgroups of G} ↔ {intermediate ﬁelds F ⊂ M ⊂ E}. Let H be a subgroup of G.. We therefore have a homomorphism σ → σ|E H : G → Aut(E H /F ) whose kernel is H. and let Hi be the subgroup corresponding to Mi (i. i. [E : E H ] = Gal(E/E H ). Therefore. (a) the correspondence is inclusion-reversing: H1 ⊃ H2 ⇐⇒ E H1 ⊂ E H2 .e. hence it must correspond to the largest subgroup contained in all Hi . (d) H is normal in G ⇐⇒ E H is normal (hence Galois) over F .17. i. using that (H1 : 1) = (H1 : H2 )(H2 : 1) and [E : E H1 ] = [E : E H2 ][E H2 : E H1 ].11b).11b). σαi is a root of the minimum polynomial of αi over F . M2 . For the ﬁrst statement.. (b) As we observed in (3. P ROOF. we have to show that H → E H and M → Gal(E/M ) are inverse maps.. Therefore Gal(E/M1 · · · Mr ) = H1 ∩ . This proves (b) in the case H2 = 1. we see that E H is Galois over F (by Theorem 3. as we observed in (3. Then (by deﬁnition) M1 M2 · · · Mr is the smallest ﬁeld containing all Mi . As (E H )G/H = F . . and the general case follows. Hence σM = M . and so σ Gal(E/M )σ −1 ↔ σM. ∩ Hr . For σ ∈ G. and let G = Gal(E/F ). Mr be intermediate ﬁelds. in which case Gal(E H /F ) = G/H. Let M be an intermediate ﬁeld.e. which is Hi . E (a) We have the obvious implications: H1 ⊃ H2 =⇒ E H1 ⊂ E H2 =⇒ Gal(E/E H1 ) ⊃ Gal(E/E H2 ). Hi = Gal(E/Mi )). Let E be a Galois extension of F . αm ].16 (F UNDAMENTAL THEOREM OF G ALOIS THEORY ).. and this implies that σHσ −1 = H (by (c)). the action of G on E stabilizes E H . .13). (d) Let H be a normal subgroup of G.. we must have σE H = E H for all σ ∈ G. Then. . Gal(E/E H ) = H. . The theorem shows that there is an order reversing bijection between the intermediate ﬁelds of E/F and the subgroups of G. −1 (c) σHσ −1 ↔ σM . Using this we can read off more results. for any subgroup H of G. Then E is Galois over M by (3. τ α = α ⇐⇒ στ σ −1 (σα) = σα.The fundamental theorem of Galois theory 29 The fundamental theorem of Galois theory T HEOREM 3.. (b) indexes equal degrees: (H1 : H2 ) = [E H2 : E H1 ]. But Gal(E/E Hi ) = Hi .11b). Because σHσ −1 = H for all σ ∈ G.

[E ∩ L : F ] 2 P ROPOSITION 3. If E/F is Galois. If σ ∈ Gal(EL/L) ﬁxes the elements of E. or Galois. it is the splitting ﬁeld of a separable polynomial f ∈ F [X]. By the fundamental theorem. and E is the splitting ﬁeld of f over E ∩ L. which is the composite of the ﬁelds σM . According to Proposition 1. then E1 E2 and E1 ∩ E2 are Galois over F . σ|E2 ) : Gal(E1 E2 /F ) → Gal(E1 /F ) ×Gal(E2 /F ) is an isomorphism of Gal(E1 E2 /F ) onto the subgroup H = {(σ1 . The largest normal subgroup contained in H is N = σ∈G σHσ −1 (see GT 4. There is therefore a homomorphism E L = σ → σ|E : Gal(EL/L) → Gal(E/F ). If α ∈ E is ﬁxed by all σ ∈ Gal(EL/L). closure of M in E. Because E is Galois over F . then EL/L and E/E ∩ L are Galois.18. and so σE = E.19. but [EL : L] = [E : E ∩ L] = 3. and so E N . σ2 ) | σ1 |E1 ∩ E2 = σ2 |E1 ∩ E2 } of Gal(E1 /F ) ×Gal(E2 /F ). .20. Thus. and the map σ → σ|E : Gal(EL/L) → Gal(E/E ∩ L) is an isomorphism. P ROPOSITION 3.18 1. Then [EL : F ] = P ROOF. then α ∈ L ∩ E. is the smallest normal extension of F containing M . Then EL is the splitting ﬁeld of f over L. that L is ﬁnite over F . Let E1 and E2 be ﬁeld extensions of F contained in some common ﬁeld. C OROLLARY 3.20 E∩L F 2 [E : F ][L : F ] . and σ → (σ|E1 . [EL : F ] = [EL : L][L : F ].30 3 THE FUNDAMENTAL THEOREM OF GALOIS THEORY (b) Let H be a subgroup of G and let M = E H . [E ∩ L : F ] [E : F ] .20. then it ﬁxes the elements of EL. Suppose.10). σ → σ|E is injective. Let E and L be ﬁeld extensions of F contained in some common ﬁeld. Any EL automorphism σ of EL ﬁxing the elements of L maps roots of f to roots = of f . It is called the normal. P ROOF. this implies that the image of σ → σ|E is Gal(E/E ∩ L). If E1 and E2 are Galois over F . Hence EL/L and E/E ∩ L are Galois. in the proposition. and hence is 1.

We prove that the image is the whole of H by counting. The subﬁeld of Q[ζ] corresponding to σ 3 is Q[ζ + ζ]. 6. F From the fundamental theorem.Examples 31 P ROOF : Let a ∈ E1 ∩ E2 . Examples E XAMPLE 3. where ζ is a primitive 7th root of 1. Since σ 3 is a normal subgroup of σ . and the map σ → i deﬁnes an isomorphism Gal(Q[ζ]/Q) → (Z/7Z)× . Now E1 E2 is a splitting ﬁeld for f1 f2 . 7 Q[ζ + ζ] is Galois over Q. Gal(E2 /F )/ Gal(E2 /E1 ∩ E2 ) Gal(E1 ∩ E2 /F ). with Galois group σ / σ 3 . σ1 |E1 ∩ E2 has exactly [E2 : E1 ∩ E2 ] extensions to an element of Gal(E2 /F ). As E1 and E2 are E1 E2 Galois over F . and that ζ has minimum polynomial X6 + X5 + X4 + X3 + X2 + X + 1 (see 1. Then σ generates Gal(Q[ζ]/Q) because the class of 3 in (Z/7Z)× generates it (the powers of 3 mod 7 are 3. and its image is contained in H. Note that Q[ζ] is the splitting ﬁeld of the polynomial X 7 − 1. We analyse the extension Q[ζ]/Q. Q[ζ] is Galois of degree 6 over Q. and let f be its minimum polynomial over F . say ζ = e2πi/7 . 4. σ|E2 ) is clearly E1 ∩ E2 an injective homomorphism. [E1 : F ] · [E2 : F ] . and ζ + ζ = 2 cos 2π . The map σ → (σ|E1 . Then f has deg f distinct roots in E1 and deg f distinct roots in E2 . For any σ ∈ G. f2 ∈ F [X]. they are splitting ﬁelds of separable polynomials f1 . σζ = ζ i . Therefore. The conjugates of α1 =df ζ + ζ are . Therefore. for each σ1 ∈ Gal(E1 /F ). 5.41).21. 2.19). some i. and so. Let σ be the element of Gal(Q[ζ]/Q) such that σζ = ζ 3 . Since f can have at most deg f roots in E1 E2 . 1 ≤ i ≤ 6. and hence Galois (3. it follows that it has deg f E1 E2 distinct roots in E1 ∩ E2 . σ3 Q[ζ] σ2 Q[ζ + ζ] σ / σ3 √ Q[ −7] σ / σ2 Q Note that σ 3 ζ = ζ 6 = ζ (complex conjugate of ζ). [E1 ∩ E2 : F ] (H : 1) = [E1 : F ][E2 : E1 ∩ E2 ] = which equals [E1 E2 : F ] by (3. and hence it also is Galois over F . This shows that E1 ∩ E2 is normal and separable over F .10). We investigate the subﬁelds of Q[ζ] corresponding to the subgroups σ 3 and σ 2 . 1).

Hence the minimum polynomial12 of ζ + ζ is g(X) = X 3 + X 2 − 2X − 1. Direct calculation shows that α1 + α2 + α3 = 6 i=1 ζ i = −1. H G/N (Z/5Z)× . and the subgroups N and H corresponding to Q[ζ] and Q[α] have orders 5 and 4 respectively. α]. Because Q[ζ] · Q[α] = Q[ζ. We have the picture at right. on setting X = ζ + ζ in (X 3 − 3X) + (X 2 − 2) + X + 1 one obtains 1 + ζ + ζ 2 + · · · + ζ 6 = 0. α] where ζ is a primitive 5th root of 1. Hence G = Gal(Q[ζ. which we can take to be ζα (after possibly replacing σ by a power). Q [Q[ζ. Hence: τ ζ = ζ2 τα = α σζ = ζ σα = ζα. Note that τ στ −1 (α) = τ σα = τ (ζα) = ζ 2 α and it ﬁxes ζ. Then (β − β ) E XAMPLE 3. For example. τ στ −1 = σ 2 . . Let τ be the generator of H corresponding to 2 under this isomorphism. and let σ be a generator of N . The minimum polynomial of cos 2π = 7 α1 2 is therefore g(2X) = X 3 + X 2 /2 − X/2 − 1/8. α1 α2 + α1 α3 + α2 α3 = −2. β = σβ. We compute the Galois group of a splitting ﬁeld E of X 5 − 2 ∈ Q[X]. α]/Q) has order 20. Moreover. we have H ∩ N = 1. Let = 2 = −7. Q[ζ] G/N Q[α] Because 4 and 5 are relatively prime.22. α1 α2 α3 = (ζ + ζ 6 )(ζ 2 + ζ 5 )(ζ 3 + ζ 4 ) = (ζ + ζ 3 + ζ 4 + ζ 6 )(ζ 3 + ζ 4 ) = (ζ 4 + ζ 6 + 1 + ζ 2 + ζ 5 + 1 + ζ + ζ 3 ) = 1. α2 = ζ 2 + ζ −2 . with ζ = e2πi/5 and N H α equal to the real 5th root of 2. and α is a root of X 5 − 2. α] take E to be the splitting ﬁeld of X 5 − 2 in C. and [Q[ζ] : Q] = 4. Thus σ(α) is another root of X 5 − 2. and so G = N θ H. Because Q[ζ] is normal over Q (it is the splitting ﬁeld of X 5 − 1).32 3 THE FUNDAMENTAL THEOREM OF GALOIS THEORY α3 = ζ 3 + ζ −3 . being generated by the class of 2. we could Q[ζ. 12 τ 4 = 1. which is cyclic. α] : Q] = 20. More directly. Thus G has generators σ and τ and deﬁning relations σ 5 = 1. [Q[α] : Q] = 5. therefore τ στ −1 = σ 2 . 8 The subﬁeld of Q[ζ] corresponding to σ 2 is generated by β √ ζ + ζ 2 + ζ 4 . N is normal in G. Recall from Exercise 2-3 that E = Q[ζ. Hence the ﬁeld ﬁxed by σ 2 is Q[ −7].

Let E/F be a quadratic extension of ﬁelds of characteristic = 2. ai−1 ]. which correspond to the ﬁve ﬁelds Q[ζ i α]. R EMARK 3. . then all the elements of E would be constructible (1.23. P ROOF. If α is contained in a subﬁeld of R that is Galois of degree 2r over Q. p prime. without (happily) having to exhibit an explicit formula for cos 65537 . Let H be a Sylow subgroup of S4 . ar ] with each ai a positive element of √ √ Q[ a1 . √ α ∈ E.36a). p P ROOF. This proves the theorem.25. . r of the form 22 + 1. Suppose α ∈ E ⊂ R where E is Galois of degree 2r over Q. For example. and ai > 0 because otherwise Ei would not be real.9) that that the Galois group of the splitting ﬁeld E over Q of the polynomial f (X) = X 4 − 4X + 2 is S4 .e. . . P ROOF. (Z/pZ)× . and so the elements of E H Q can’t be constructible. . Correspondingly. we showed (1. α ∈ F . From a theorem on the structure of p-groups (GT 6.. and let G = Gal(E/Q). If the four roots of f (X) were constructible. i. . then it is constructible. and so E = F [ b2 − 4c]. then cos 2π is constructible. we have proved that the regular 65537-polygon is con2π structible. Now we can prove a partial converse to this last statement. We’ll show below (4. 33 Constructible numbers revisited Earlier. is constructible if and only if p is a Fermat prime. . which has 2 (∗) Thus a regular p-gon.7). E = E0 ⊃ E1 ⊃ E2 ⊃ · · · ⊃ Er = Q √ with Ei−1 of degree 2 over Ei . 2 L EMMA√ 3. and let X 2 + bX + c be the minimum polynomial of α. 1 ≤ i ≤ 5. The next lemma shows that Ei = Ei−1 [ ai ] for some ai ∈ Ei−1 .36) that a real number α is constructible if and only if it is con√ √ tained in a subﬁeld of R of the form Q[ a1 . Then E H has odd degree over Q. The converse to (∗) is false. The ﬁeld Q[e2πi/p ] is Galois over Q with Galois group G order p − 1 = 2k . Then E = F [ d] for some d ∈ F . σ i Hσ −i ↔ σ i Q[α] = Q[ζ i α]. we know there will be a sequence of groups {1} = G0 ⊂ G1 ⊂ G2 ⊂ · · · ⊂ Gr = G with Gi /Gi−1 of order 2.Constructible numbers revisited The subgroup H has ﬁve conjugates. there will be a sequence of ﬁelds. T HEOREM 3. If p is a prime of the form 2k + 1. . . Then Let / √ 2 α = −b± 2b −4c .26. In particular α constructible =⇒ [Q[α] : Q] = 2s some s. 2 C OROLLARY 3.24.

Also we shall exhibit polynomials f (X) ∈ Q[X] with Galois group Sn . αn }) ≈ Sn . Thus Gf can be identiﬁed with a subset of Sym({α1 . . Show: This problem shows that every quadratic extension of Q is contained in a cyclotomic extension of Q. Solvability of equations For a polynomial f ∈ F [X]. . for P ∈ F [X1 .. or. Being automorphisms. We know that the elements of Gal(Ff /F ) i=1 map roots of f to roots of f . 1832). . Note that this shows again that (Gf : 1). P (α1 .29).28. and as the αi generated Ff . αn }. [Hint: Find automorphisms σ and τ of F (X). they map the set {α1 . . . α2 . . ﬁxing F (X 2 ) and F (X 2 − X) respectively. . Exercises 3-1 (*). If the deﬁnition of solvable is changed to allow extensions of the type in (b) in the chain. divides deg(f )!. and show that στ has inﬁnite order. multiplication. Let F be a ﬁeld of characteristic zero. . . 13 Let p be an odd prime. . Xn ]. then the theorem fails for two reasons: (a) f may not be separable. Let F be a ﬁeld of characteristic 0.27 (G ALOIS . . αn } into itself. . α2 . . . more precisely. In fact. Show that F (X 2 ) ∩ F (X 2 − X) = F (intersection inside F (X)). If F has characteristic p. . T HEOREM 3. Put α = i∈H ζ i and β = i∈G\H ζ i . (b) Fm contains a splitting ﬁeld for f.29. .e. . Gf consists exactly of the permutations σ of {α1 . thus G = (Z/(p))× . . αn } such that. α2 . . We shall prove this later (5. R EMARK 3. and let G = Gal(E/Q). division. which are therefore not solvable when n ≥ 5 by GT 4. and let ζ be a primitive pth root of 1 in C. and so not have a Galois group.] 3-2 (*). subtraction. . 13 . . and we call Gal(Ff /F ) the Galois group Gf of f. . . αn ) = 0 =⇒ P (σα1 . Let H be the subgroup of index 2 in G. . hence [Ff : F ]. . and the extraction of mth roots. Let f = n (X −αi ) in a splitting ﬁeld Ff . if there exists a tower of ﬁelds F = F 0 ⊂ F1 ⊂ F2 ⊂ · · · ⊂ F m such that m (a) Fi = Fi−1 [αi ]. The Kronecker-Weber theorem says that every abelian extension of Q is contained in a cyclotomic extension. . we say that f (X) = 0 is solvable in radicals if its solutions can be obtained by the algebraic operations of addition. i. σαn ) = 0. αi i ∈ Fi−1 . (b) X p − X − a = 0 is not solvable by radicals even though it is separable with abelian Galois group (cf. then its splitting ﬁeld Ff is Galois over F . they deﬁne permutations of {α1 . α2 . Exercise 2-2). The equation f = 0 is solvable in radicals if and only if the Galois group of f is solvable. and f is required to be separable. each of order 2. an element of Gal(Ff /F ) is uniquely determined by the permutation it deﬁnes.34 3 THE FUNDAMENTAL THEOREM OF GALOIS THEORY The Galois group of a polynomial If the polynomial f ∈ F [X] is separable. This gives a description of Gf without mentioning ﬁelds or abstract groups (neither of which were available to Galois). then the theorem becomes true in characteristic p. . . Let E = Q[ζ].

Thus α and β are roots of the polynomial X 2 + X + αβ ∈ Q[X]. √ √ √ √ 3-3 (*). .Exercises 35 (a) α and β are ﬁxed by H. Compute αβ and show √ √ that the ﬁxed ﬁeld of H is Q[ p] when p ≡ 1 mod 4 and Q[ −p] when p ≡ 3 mod 4. σβ = α. then σα = β. (a) Show that M is Galois over Q with Galois group the 4-group C2 × C2 . (b) if σ ∈ G \ H. Let M = Q[ 2. (b) Show that E is Galois over Q with Galois group the quaternion group. 3] and E = M [ ( 2 + 2)( 3 + 3)] (subﬁelds of R).

Note that D(f ) is nonzero if and only if f has only simple roots. . ∆(f ) = 0. (a) The discriminant D(f ) ∈ F . i. When is Gf ⊂ An ? Consider a polynomial f (X) = X n + a1 X n−1 + · · · + an and let f (X) = ∆(f ) = n i=1 (X − αi ) in some splitting ﬁeld. The choice of a numbering for the roots determines an isomorphism Sym({α1 . (b) Because f has simple roots. Thus. and identify it with a subgroup of Sym({α1 . Hence Gf ⊂ An ⇐⇒ ∆(f ) ∈ F ⇐⇒ D(f ) is a square in F. P ROOF. . . In this section. Let f (X) ∈ F [X] be of degree n and have only simple roots. αn }) (as on p34). 4 Computing Galois groups. cf. (a) σ∆(f ) = sign(σ)∆(f ). Let Gf be the Galois group of f . (b) The subﬁeld of Ff corresponding to An ∩ Gf is F [∆(f )]. and so the formula σ∆(f ) = sign(σ)∆(f ) shows that an element of Gf ﬁxes ∆(f ) if and only if it lies in An . if f is separable with no multiple factors. . P ROPOSITION 4. D(f ) = ∆(f )2 = (αi − αj )2 .e. (b) σD(f ) = D(f ). so that Gf = Gal(Ff /F ). Gf ∩ An ↔ F [∆(f )]. we investigate general methods for computing Galois groups.36 4 COMPUTING GALOIS GROUPS. 2 The discriminant of f can be expressed as a universal polynomial in the coefﬁcients of f . αn }) corresponding to any normal subgroup of Sn is independent of the choice for n = 6 (because all automorphisms of Sn are inner when n = 6.2. under the Galois correspondence. For example: D(aX 2 + bX + c) = (b2 − 4ac)/a2 D(X 3 + bX + c) = −4b3 − 27c2 .1. and hence lies in F (by the fundamental theorem of Galois theory). Let Ff be a splitting ﬁeld for f . . . P ROOF. and the second equation is obtained by squaring the ﬁrst. Set (αi − αj ). . Hence. GT 3. . Gf ∩ An = Gf ⇐⇒ F [∆(f )] = F. where sign(σ) is the signature of σ.. . 2 C OROLLARY 4. . . The ﬁrst equation follows immediately from the deﬁnition of the signature of σ (see GT §4). αn }) Sn .9). (a) The discriminant of f is an element of Ff ﬁxed by Gf =df Gal(Ff /F ). and the subgroup of Sym({α1 . and let σ ∈ Gf . Assume f is separable. . . 1≤i<j≤n 1≤i<j≤n The discriminant of f is deﬁned to be D(f ).

R EMARK 4.When is Gf transitive? 37 By completing the cube. .4. and so β is also a root of g. g(σα) = σg(α) = 0. E XAMPLE 4. Therefore.. Then f is inseparable ⇐⇒ F has characteristic 2 and f (X) = X 2 − a for some a ∈ F F 2 . Let f (X) ∈ F [X] be a polynomial of degree 3. P ROOF. . Gf is not necessarily contained in An .5. for example. n|(Gf : 1) because [F [α] : F ] = n and [F [α] : F ] divides [Ff : F ] = (Gf : 1). one can put any cubic polynomial in this form (in characteristic = 3). Of course the converse is not true: when s is even. Write Ff = F [α1 . Let f (X) ∈ F [X] be a polynomial of degree 2. This can be proved more directly by noting that complex conjugation acts on the roots as the product of s disjoint transpositions. then Gf is not contained in An . If f is separable. There are . Then f (X) is irreducible if and only if Gf permutes the roots of f transitively.] with α1 = α and α2 . because g has coefﬁcients in F . Polynomials of degree at most three E XAMPLE 4. ⇐= : Let g(X) ∈ F [X] be an irreducible factor of f . then Gf = 1(= A2 ) or S2 according as D(f ) is a square in F or not.” where f is a polynomial in the symbol X. Thus Gf is a transitive subgroup of Sn whose order is divisible by n. Fortunately. α3 . Hence. that for X 5 + aX 4 + bX 3 + cX 2 + dX + e has 59 terms. .X). and so F [α] and F [β] are both stem ﬁelds for f .6. We can assume f to be irreducible. Suppose F ⊂ R. =⇒ : If α and β are two roots of f (X) in a splitting ﬁeld Ff for f . Thus if s is odd. . Now. Maple knows them: the syntax is “discrim(f. Then D(f ) will not be a square if it is negative. then they both have f (X) as their minimum polynomial. Let f (X) ∈ F [X] have only simple roots.39).3.. and so f (X) = g(X).2b). Then the F -homomorphism α → β : F [α] → Ff extends (step by step) to an F -homomorphism Ff → Ff (use 2. then (by assumption) β = σα for some σ ∈ Gf . When is Gf transitive? P ROPOSITION 4. Then f is inseparable if and only if F has characteristic 3 and f (X) = X 3 − a for some a ∈ F \ F 3 . α2 . If f is separable. If β is a second root of f . α ↔ β. then Gf is a transitive subgroup of S3 whose order is divisible by 3. there is an F -isomorphism F [α] F [β]. which is an F -isomorphism sending α to β. the other roots of f (X). The formulas for the discriminant rapidly become very complicated. every root of f is also a root of g. and let α be one of its roots. It is known that the sign of D(f ) is (−1)s where 2s is the number of nonreal roots of f in C (see ANT 2. for otherwise we are essentially back in the previous case. 2 Note that when f (X) is irreducible of degree n.

γ]. β. Quartic polynomials Let f (X) be a quartic polynomial without multiple roots. α − β = α1 (α2 − α3 ) + α4 (α3 − α2 ) = (α1 − α4 )(α2 − α3 ). γ. γ must therefore be a subgroup of index 3 in S4 . On the other hand. α2 . and so its Galois group is S3 . β. all isomorphic to D4 . Groups of order 8 in S4 are Sylow 2-subgroups. The above discussion shows that the subgroup of Gf of elements ﬁxing F [α. Hence F [α.11). but its discriminant is −135 which is not a square in Q. for example. (12)(34). They are distinct because the αi are distinct. β. any element of Gf ) merely permutes α. β. V is contained in a Sylow 2-subgroup. The group Sym({α1 . Any permutation of the αi (a fortiori. (13)(24). because the Sylow 2-subgroups are conjugate and V is normal. and let f (X) = (X − αi ) in E. α3 . and therefore their intersection V is the subgroup of Sym({α1 . P ROOF.12). Consider the partially symmetric elements α = α1 α2 + α3 α4 β = α1 α3 + α2 α4 γ = α1 α4 + α2 α3 . and γ. β. α3 . γ]. X 3 + 3X + 1 ∈ Q[X] is also irreducible (apply 1. its discriminant is −4(−3)3 − 27 = 81 = 92 . β. We identify the Galois group Gf of f with a subgroup of the symmetric group Sym({α1 . γ] Gf /Gf ∩V F Let M = F [α. Therefore (by the fundamental theorem) g(X) has coefﬁcients in F . and let g(X) = (X − α)(X − β)(X − γ) ∈ M [X] — it is called the resolvent cubic of f . γ} transitively. α3 . and so E Gf ∩V = F [α. or γ. (13) . It follows that V is the intersection of the three Sylow 2-subgroups. we have: . and hence has order 8. By the Sylow theorems. For example. For example. There are three of them. β. β. the stabilizer of β is (1234). (14)(23)} as a normal subgroup — it is normal because it contains all elements of type 2 + 2 (GT 4.7. β. X 3 − 3X + 1 ∈ Q[X] is irreducible (see 1. only two possibilities: Gf = A3 or S3 according as D(f ) is a square in F or not. γ] by the fundamental theorem of Galois theory. α2 .38 4 COMPUTING GALOIS GROUPS. Let E be a splitting ﬁeld of f . Each Sylow 2-subgroup ﬁxes exactly one of α. β. α4 }) ﬁxing α. α2 . and so ﬁxes g(X). in fact. γ] is Galois over F with Galois group Gf /Gf ∩ V . More explicitly. α4 }). γ] is Gf ∩ V . α4 }) permutes {α. L EMMA 4. The ﬁxed ﬁeld of Gf ∩ V is F [α.28). it is contained in all three. 2 Gf ∩V E F [α. The stabilizer of each of α. and so its Galois group is A3 . The remaining statements follow from the fundamental theorem using that V is normal. Note that A3 is generated by the cycle (123). In order to determine Gf we shall exploit the fact that S4 has V = {1. β.

d. and its resolvent cubic is g(X) = X 3 − 8X + 16. thus M = Q[ 2]. The discriminants of f and g are equal. α4 . which is irreducible because it has no roots in F5 . Expand g = (X − α)(X − β)(X − γ) to express the coefﬁcients of g in terms of α1 . From the table we see that Gf is of type D4 or C4 .16). and so the Galois group of g(X) is S3 .16). and substitute to express them in terms of b. E XAMPLE 4. From the table. Consider f (X) = X 4 +4X 2 +2 ∈ Q[X].9. which is not a square. α3 . and so (from 4.8. and Gf is of type D4 . E XAMPLE 4. We can compute (G : V ∩ G) from the resolvent cubic g. Consider f (X) = X 4 − 2 ∈ Q[X]. α3 . E XAMPLE 4. c. Consider f (X) = X 4 − 4X + 2 ∈ Q[X]. and the groups of type C4 are those generated by cycles of length 4. α2 . It is irreducible in Q[X] because (by inspection) it is irreducible in Z[X]. then G ∩ V = V or C2 . α2 . and hence Gf is of type C4 . because G/V ∩ G = Gal(M/F ) and M is the splitting ﬁeld of g. c.11. Consider f (X) = X 4 − 10X 2 + 4 ∈ Q[X]. Expand f = (X − α1 )(X − α2 )(X − α3 )(X − α4 ) to express b. 39 P ROOF (S KETCH ).22). we see that the Galois group of f (X) is S4 . It is irreducible by Eisenstein’s √ criterion (1. e. 2 Now let f be an irreducible separable quartic. Hence M = Q[i 2]. analyse the equation as in (3. we can deduce G except in the case that it is 2. E XAMPLE 4. It is irreducible by Eisenstein’s criterion (1.10. Its resolvent cubic is (X + 10)(X + 4)(X − 4). One can check that f is irreducible over M . α4 . There are the following possibilities for G: G S4 A4 V D4 C4 (G ∩ V : 1) 4 4 4 4 2 (G : V ∩ G) 6 3 1 2 2 (G ∩ V : 1) = [E : M ] (G : V ∩ G) = [M : F ] The groups of type D4 are the Sylow 2-subgroups discussed above. The discriminant of g(X) is −4864. e in terms of α1 .12. d. It is irreducible by Eisenstein’s √ criterion (1. Only the ﬁrst group acts transitively on the roots of f . The resolvent cubic of f = X 4 + bX 3 + cX 2 + dX + e is g = X 3 − cX 2 + (bd − 4e)X − b2 e + 4ce − d2 . and so Gf is of type V . Then G = Gf is a transitive subgroup of S4 whose order is divisible by 4. and its resolvent cubic is (X − 4)(X 2 − 8). and its resolvent cubic is g(X) = X 3 + 8X. Once we know (G : V ∩ G).16).4) we see that in this case G = D4 or C4 according as f is irreducible or not in M [X]. . Alternatively. but f factors over M (even as a polynomial in X 2 ).Quartic polynomials L EMMA 4. If [M : F ] = 2.

For p prime. Maple knows how to factor polynomials with coefﬁcients in Q[α]. the symmetric group Sp is generated by any transposition and any p-cycle. Let σ be complex conjugation on C. Therefore Gf contains an element of order p (Cauchy’s theorem.13). E XAMPLE 4..13. we may assume that the transposition is τ = (12). GT 4. Now some power of σ will map 1 to 2 and will still be a p-cycle (here is where we use that p is prime). P ROOF.40 4 COMPUTING GALOIS GROUPS. Choose a positive even integer m and even integers n1 < n2 < · · · < np−2 . then G is S4 . — If w is rational and w2 + 4 is a square then G = V = C2 × C2 . σ = (1 i2 · · · ip ). — If m = 1 and w not rational then G is S3 . a > b and d = a2 − b2 . The Galois group of f is as follows: — If m = 0 and w not rational. Let f be an irreducible polynomial of prime degree p in Q[X]. Let r = d/c2 and let w be the unique positive real number such that r = w3 /(w2 + 4). jp ).25) and so lies in the subgroup generated by σ and τ . Examples of polynomials with Sp as Galois group over Q The next lemma gives a criterion for a subgroup of Sp to be the whole of Sp . Let E be the splitting ﬁeld of f in C. After renumbering. (From the web. and we may write the p-cycle σ so that 1 occurs in the ﬁrst position. Because f is irreducible.14. The hard part was to establish that m = 2 could never happen. and so p|[E : Q] = (Gf : 1). — If w is rational and w2 + 4 is not a square then G = D4. sci. Let m be the number of roots of f (X) in Z (counted with multiplicities). P ROOF. and let g(X) = (X 2 + m)(X − n1 ). After replacing σ with the power. . L EMMA 4. If f splits in C and has exactly two nonreal roots. Let p≥ 5 be a prime number. we have σ = (1 2 3 . .15.16. Then σ transposes the two nonreal roots of f (X) and ﬁxes the rest. Now (i i + 1) = σ i (12)σ −i (see GT 4. ..math.(X − np−2 ). and let α ∈ E be a root of f . c > 0. This covers all possible cases. . These transpositions generate Sp . b > 0. and after renumbering again. [Q[α] : Q] = deg f = p. 2 P ROPOSITION 4. p).research. Therefore Gf ⊂ Sp and contains a transposition and a p-cycle. 2 It remains to construct polynomials satisfying the conditions of the Proposition. then Gf = Sp . but the only elements of order p in Sp are p-cycles (here we use that p is prime again). E XAMPLE 4. . As we explained in (1. search for “ﬁnal analysis”. we have σ = (1 2 j3 .29). and so is the whole of Sp .) Consider f (X) = X 4 − 2cX 3 − dX 2 + 2cdX − dc2 ∈ Z[X] with a > 0.

. . np−2 . any other ﬁeld E of characteristic p contains a copy of Fp .14 Finite ﬁelds Let Fp = Z/pZ. Then E × is a ﬁnite subgroup of the multiplicative group of a ﬁeld. Then S is obviously closed under multiplication and the formation of inverses. to every polynomial). No harm results if we identify Fp with this subﬁeld of E. Thus 2 e = ming (x)=0 |g(x)| > 0. and so S = E.18. Moreover. but the proof is longer (see Jacobson 1964. 2 Now let E be the splitting ﬁeld of f (X) = X q − X. and so f has exactly two nonreal roots. we can take n = 1. and it doesn’t have a local maximum or minimum at any of those points (because the ni are simple roots). and so it certainly splits over C (high school algebra). . For each power q = pn there is a ﬁeld Fq with q elements. E has pn elements. the ai are all even and ap is not divisible by 22 . p107. Then E has q = pn elements. P ROPOSITION 4. when we write nf (X) = nX p + a1 X p−1 + · · · + ap .e. {m1E | m ∈ Z}. then g(X) − 2 will have exactly two nonreal roots. Consider E ⊃ F . and so E × is a group of order q − 1. . and hence is cyclic (see Exercise 1-3). and so any two ﬁelds with q elements are isomorphic. Consider 2 f (X) = g(X) − . f has p − 2 linear factors and one quadratic factor. The shorter argument in the text was suggested to me by Martin Ward. Fq is Galois over Fp with cyclic Galois group generated by the Frobenius automorphism σ(a) = ap .Finite ﬁelds 41 The graph of g crosses the x-axis exactly at the points n1 .17. but it is also closed under subtraction: if aq = a and bq = b. q = pn . n 2 As n < e. If ζ generates E × as a multiplicative group. the proposition applies to f . P ROPOSITION 4. which is relatively prime to f (X) (in fact. Let E be a ﬁeld of degree n over Fp . As we noted in §1. Hence the nonzero elements of E are roots X q−1 − 1. The derivative f (X) = −1. It is the splitting ﬁeld of X q − X. and hence any two such ﬁelds are isomorphic. namely. and so Eisenstein’s criterion implies that f is irreducible. 14 . Hence E is a splitting ﬁeld for X q − X. then (a − b)q = aq − bq = a − b. and all elements of E (including 0) are roots of X q − X. and we can choose an odd positive integer n such that n < e. the graph of f also crosses the x-axis at exactly p−2 points. who credits the example to Brauer). Hence S is a ﬁeld. and so f (X) has q distinct roots in E. Over R. the ﬁeld of p elements. i. Let S be the set of its roots. On the other hand. In particular. then certainly E = F [ζ].. P ROOF. Therefore. If m is taken sufﬁciently large. Every extension of ﬁnite ﬁelds is simple.

for example.11b). The syntax is “Factor(f(X)) mod p. For example. then f (X) has a root in a ﬁeld of degree n d over Fp . and so E has exactly one subﬁeld of degree m over Fp for each m dividing m m n. and so the ﬁxed ﬁeld of σ is Fp . the factors of X p − X are distinct because it has no common factor with its derivative. Then. Only the ﬁnal statement remains to be proved. b ∈ S. E contains exactly one ﬁeld with pm elements. hence. E σ . The ﬁeld Fq is Galois over Fp σ because it is the splitting ﬁeld of a separable polynomial. The ﬁeld Fp has an algebraic closure F. Hence some root of X p − X is also a root of f (X). P ROOF. Obvious from what we have proved.22. Moreover. Because it has degree m over Fp . 2 C OROLLARY 4. σ m . to obtain a list of all monic polynomials of degree 1. ask Maple to factor X 625 − X. every polynomial in Fp [X] splits in F. 2. Since the Fpn ’s are not subsets of a ﬁxed set.”. P ROOF. For each divisor m of n. E σ has pm elements. Fpm ⊂ Fpn ⇐⇒ m|n. P ROOF. which has degree d over Fp . Let E be a ﬁeld with pn elements. it is again a ﬁeld algebraic over Fp . Moreover. Fpn1 ⊂ Fpn2 ⊂ Fpn3 ⊂ · · · . Maple factors polynomials modulo p very quickly. or 4 over F5 . If f (X) is irreducible of degree d. Choose a sequence of integers 1 = n1 < n2 < n3 < · · · such that ni |ni+1 for all i. As a union of a chain of ﬁelds algebraic over Fp . Thus.21.19. In particular. and we let F = S/ ∼.42 4 COMPUTING GALOIS GROUPS. An element a of Fq is ﬁxed by σ if and only if ap = a. Deﬁne S to be the disjoint union of the Fpn . But the splitting ﬁeld of X p − X contains a copy of every ﬁeld of degree n d over Fp with d|n. We know that E is Galois over Fp and that Gal(E/Fp ) is the cyclic group of order n generated by σ. set a ∼ b if a = b in one of the Fpn . the degree of the splitting ﬁeld of f is ≤ d.20. Deﬁne the ﬁelds Fpni n inductively as follows: Fpn1 = Fp . Each monic irreducible polynomial f of degree d|n in Fp [X] occurs n exactly once as a factor of X p − X. namely. The partially ordered set of ﬁnite subﬁelds of F is isomorphic to the set of integers n ≥ 1 partially ordered by divisibility. and therefore it splits in its splitting ﬁeld. and therefore n d f (X)|X p − X. namely. forming the union requires explanation. 2 n . P ROOF.44). and every integer n divides some ni . Then ∼ is an equivalence relation. P ROPOSITION 4. We noted in (1. and we deﬁne F = Fpni .23. let ni = i!. The group σ has one subgroup of order n/m for each m dividing n. 2 C OROLLARY 4. First. but Fp consists exactly of such elements. m ≥ 0. and Fpn consists of the roots of X p − X. f divides X p − X.4) that x → xp is an automorphism of Fq . 2 P ROPOSITION 4. and so it is an algebraic closure of F (by 1. Fpni is the splitting ﬁeld of X p i − X over Fpni−1 . Let F be an algebraic closure of Fp . Then F contains exactly one ﬁeld n Fpn for each integer n ≥ 1. P ROOF. This proves that Fq is Galois over Fp and that σ = Gal(Fq /Fp ) (see 3. For a. 2 R EMARK 4.

February 2003 (P´ lfy): “full credit should be given to [Galois] for a constructing ﬁnite ﬁelds in general.g. m2 . . The monic factors of f (X) in Ff [X] correspond to subsets S of {α1 .22).. Then f factors as f = f1 · · · fr with fi irreducible of degree mi . Recall that for a monic separable polynomial f ∈ F [X].24. . Maple contains a “Galois ﬁeld package” to do computations in ﬁnite ﬁelds. the existence of a primitive element. When we regard σ as a permutation of the roots of f . with pn elements say. but I prefer to use descriptive names wherever possible. . Then Gf is a cyclic group generated by the Frobenius automorphism σ : x → xp . Hence. P ROOF. m = deg f . In one of the few papers published during his short lifetime. . αm } into a disjoint union of orbits for the group Gf . α2 . . Note that the proof shows the following: let {α1 . αm } stable under Gf . . Let f (X) be a monic polynomial in F [X] with only simple roots. . . it can ﬁnd a primitive element for Fq (i. .25. and Gf = Gal(Ff /F ) denotes the Galois group of F . . and Fq was denoted GF (q) (it still is in Maple). but these subsets S are precisely the orbits of Gf in {α1 . Galois — e e at that time not even nineteen years old — deﬁned ﬁnite ﬁelds of arbitrary prime power order and established their basic properties. . For example. type: q readlib(GF):. . Computing Galois groups over Q In the remainder of this section. The ﬁrst result generalizes Proposition 4. . . P ROPOSITION 4. . . . mr elements respectively. . αm }. . then f= fi . Ff denotes a splitting ﬁeld for F . S ↔ fS = α∈S (X − α).}. be the roots of f (X) in Ff .4. e. From a letter to the Notices of the AMS. . To start it. entitled “Sur la th´ orie des nombres”. mr . Therefore the irreducible factors of f in F [X] are the polynomials fS corresponding to minimal subsets S of {α1 .e. which appeared in the Bulletin des Sciences Math´ matiques in June 1830. αm } = Oi be the decomposition of {α1 . α2 . 2 R EMARK 4. then σ has a cycle decomposition of type m1 + · · · + mr = deg f. . a generator for F× ). . . . . .” True. αm }. of f in Ff : G ⊂ Sym{α1 . . Now suppose F is ﬁnite. 15 . . . . αm . So it is fully justiﬁed when ﬁnite ﬁelds are called Galois ﬁelds and customarily denoted by GF (q). fi = αi ∈Oi (X − αi ) is the decomposition of f into a product of irreducible polynomials in F [X]. Gf permutes the roots α1 . and suppose that the orbits of Gf acting on the roots of f have m1 . if the degrees of the distinct irreducible factors of f are m1 . and fS is ﬁxed under the action of Gf (and hence has coefﬁcients in F ) if and only if S is stable under Gf . Let α1 . . .Computing Galois groups over Q 43 Finite ﬁelds were often called15 Galois ﬁelds. . . . I sketch a practical method for computing Galois groups over Q and similar ﬁelds. Moreover. then distinct orbits of σ correspond to the factors in its cycle decomposition (GT 4. . . .

We also choose f1 . 2 E XAMPLE 4.29. on raising this to an appropriate odd power. Therefore Gf = S5 by (4. .26. we may suppose the (n − 1)-cycle is (123 . . some 1 ≤ i ≤ n − 1. αm }. Let P be a prime ideal in R. (iii) Gf contains a transposition (because f ≡ f3 mod 5. . . n − 1).44 4 COMPUTING GALOIS GROUPS. D(f ) is not divisible by p). Because of the transitivity. . P ROOF. . Assume that neither f nor f has a multiple root. we obtain the following theorem. . Suppose that f = fi with fi irreducible of degree mi in Fp [X]. Take f = −15f1 + 10f2 + 6f3 . L EMMA 4. . . T HEOREM 4. Modulo 2. .27 (D EDEKIND ). L EMMA 4. and their reductions αi modulo P R are the roots of f . f3 to have only simple roots. . n − 1) and its powers will transform it into (1n). .28. f2 . . . I. f1 . (2n). and modulo 3 it is irreducible. (n − 1 n). §61 (second edition) or ANT 3. Moreover Gf ⊂ Gf when both are identiﬁed with subgroups of Sym{α1 . we are left with the transposition).30. and so it contains the product of a transposition with a commuting element of odd order. Hence Gf contains (ik)(lmn) and (12345).43. αm of f lie in some ﬁnite extension R of R. Omitted — see van der Waerden. (c) f3 = (irreducible of degree 2)(product of 1 or 2 irreducible polys of odd degree) mod 5. Let f (X) ∈ Z[X] be a monic polynomial of degree m. Hence Gf is Sn . E XAMPLE 4. After possibly renumbering. . and let p be a prime such that f mod p has simple roots (equivalently. P ROOF. f3 with coefﬁcients in Z such that: (a) f1 is irreducible modulo 2. f2 . . (b) f2 = (degree 1)(irreducible of degree n − 1) mod 3. 2 On combining these results. Then the roots α1 . and these elements obviously generate Sn . Select monic polynomials of degree n. . and let f be a monic polynomial in R[X]. A transitive subgroup of H ⊂ Sn containing a transposition and an (n − 1)cycle is equal to Sn . . and so also ((ik)(lmn))3 = (ik). Modern Algebra. . and let f be the image of f in (R/P )[X]. . Conjugating (in) by (123 . Consider X 5 − X − 1. Then (i) Gf is transitive (it contains an n-cycle because f ≡ f1 mod 2). Let R be a unique factorization domain with ﬁeld of fractions F . this factors as (X 2 + X + 1)(X 3 + X 2 + 1). αm } = Sym{α1 . (ii) Gf contains a cycle of length n − 1 (because f ≡ f2 mod 3). . the transposition can be transformed into (in). Then Gf contains an element whose cycle decomposition is of type m = m1 + · · · + mr .14).

a. This lists all transitive subgroups of Sn . Factor f modulo a sequence of primes p not dividing D(f ) to determine the cycle types of the elements in Gf — a difﬁcult theorem in number theory. 4-6. b ∈ Q. See also. (a) Can you ﬁnd it? (b) Can you ﬁnd it without using the “galois” command in Maple? 4-7 (*). With few exceptions. 4-4 (*). then look at its action on the set of subsets of r roots for some r. Butler and McKay. 4-2 (*). Comm. The transitive groups of degree up to eleven. then this will be seen by looking modulo a set of prime numbers of positive density. say. n ≤ 11. J. Find the Galois group of X 4 − 2X 3 − 8X − 3 over Q. Show that Gf ≈ D5 (dihedral group) if and only if (a) f (X) is irreducible in Q[X]. 4-5. and gives the cycle types of their elements and the orbit lengths of the subgroup acting on the r-sets of roots. Exercises 4-1 (*). [M : F ] = 2. type “?galois. Find the splitting ﬁeld of X m − 1 ∈ Fp [X]. Show that its Galois group is nonabelian. X 6 + 2X 5 + 3X 4 + 4X 3 + 5X 2 + 6X + 7. Find the degree of the splitting ﬁeld of X 8 − 2 over Q. Soicher and McKay. says that if a cycle type occurs in Gf . 4-8.”. xq i − x) = 1 for all i.Exercises 45 The above results give the following strategy for computing the Galois group of an irreducible polynomial f ∈ Q[X]. Now look up a table of transitive subgroups of Sn with order divisible by n and their cycle types. 20 (1985) 273–281. and (c) the equation f (X) = 0 is solvable by radicals. ri k i=1 pi is irreducible over Fq if and only 4-9. Maple V can compute Galois groups for polynomials of degree ≤ 7 over Q. 4-3 (*).” [over Q]. Let f (X) be an irreducible polynomial in Q[X] with both real and nonreal roots. Give an example of a ﬁeld extension E/F of degree 4 such that there does not exist a ﬁeld M with F ⊂ M ⊂ E. and will occur for a prime less than some bound. Algebra 11 (1983). To learn the syntax. these invariants are sufﬁcient to determine the subgroup up to isomorphism. List all irreducible polynomials of degree 3 over F7 in 10 seconds or less (there are 112). “It is a thought-provoking question that few graduate students would know how to approach the question of determining the Galois group of. 863–911. Computing Galois groups over the rationals. Show that a polynomial f of degree n = n/p if gcd(f (x). Let f (X) = X 5 + aX + b. See. Number Theory. Can the condition that f is irreducible be dropped? . and (b) the discriminant D(f ) = 44 a5 + 55 b4 of f (X) is a square. the effective Chebotarev density theorem. If this doesn’t sufﬁce to determine the group.

..46 5 APPLICATIONS OF GALOIS THEORY 5 Applications of Galois theory In this section. βt be the roots of g. . αr ] = F [α1 . where √ √ σ √2 = −√2 . Recall that an element α algebraic over a ﬁeld F is separable over F if its minimum polynomial over F has no multiple roots. √− √ (στ )( 2 + 3) = − 2 − 3. Thus let E = F [α. If we choose a c ∈ F different from any of β −β these solutions (using that F is inﬁnite). √ √ √ √ These all differ from √ + √ and so only the identity element of Gal(Q[ 2. √ √ Consider for example Q[ 2. For j = 1. αr ] = F [α1 . . We know (see Exercise 3-3) that its Galois group over Q is a 4-group σ. Recall that a ﬁnite extension of ﬁelds E/F is simple if E = F [α] for some element α of E. 3]/Q. then αi + cβj = α + cβ unless i = 1 = j. τ 3 = − 3 √ √ √ √ σ(√2 + √3) = − 2 + 3. τ . According to the fundamental theorem. We shall show that (at least) all separable extensions have primitive elements.. Then there is an element γ ∈ E such that E = F [γ]. and assume that α2 . α3 . T HEOREM 5. we proved this in (4. . . It is clear that this argument should work much more generally. . Primitive element theorem... . namely. . Such an α is called a primitive element of E. . P ROOF. . Let E = F [α1 . this implies that √ √ 2 + 3 is a primitive element: √ √ √ √ Q[ 2. . β] with β separable over F . 3] = Q[ 2 + 3]. σ 3 = 3 Note that √ √ τ √2 = √2 . αr ] be a ﬁnite extension of F . . . . and so the the equation αi + Xβj = α1 + Xβ1 . . βj = β1 . ﬁxes the elements of Q[ 2 + 3]. and let β1 = β. . It sufﬁces to prove the statement for r = 2. X = α1−αj . Hence we may assume F to be inﬁnite. we apply the fundamental theorem of Galois theory to obtain other results about polynomials and extensions of ﬁelds. αr are separable over F (but not necessarily α1 ). . Let α1 = α. . β2 . α4 .. . α2 .17). for then F [α1 . . 1 i has exactly one solution.1. For ﬁnite ﬁelds. √ √ τ (√2 + √3) = 2 3.. αs be the roots of f in some big ﬁeld containing E. 3]/Q) 2 3. . Let f and g be the minimum polynomials of α and β over F . αr ] = · · · . . .

Therefore. ﬁxed by some. We have shown that F [α. but we have seen (Proposition 2. and hence also in E[X]. These remarks make it very easy to write down primitive elements. because we chose c so that γ − cβj = αi unless i = 1 = j. Let M be such a ﬁeld. (b) The proposition has a converse: if E is a ﬁnite extension of F and there are only ﬁnitely many intermediate ﬁelds M . P ROPOSITION 5. (a) Note that the proof in fact gives a description of all the intermediate ﬁelds: each is generated over F by the coefﬁcients of a factor g(X) of f (X) in E[X]. Hence β ∈ F [γ].4. . F ⊂ M ⊂ E. Then there are only ﬁnitely many intermediate ﬁelds M . Here we have computed the gcd in some ﬁeld splitting f g. Then the polynomials g(X) and f (γ − cX) have coefﬁcients in F [γ][X]. but not necessarily all.Primitive element theorem. Let M be the subﬁeld of E generated over F by the coefﬁcients of g(X). Hence [E : M ] = deg g = [E : M ].10) that the gcd of two polynomials has coefﬁcients in the same ﬁeld as the coefﬁcients of the polynomials. and so M = M — M is generated by the coefﬁcients of g(X). the proof shows that γ can be chosen to be of the form γ = α1 + c2 α2 + · · · + cr αr . there are only ﬁnitely many possible g’s. Therefore gcd(g(X). and this implies that α = γ − cβ also lies in F [γ]. and let g(X) be the minimum polynomial of γ over M . P ROOF. p508). and have β as a root: g(β) = 0. Let f (X) be the minimum polynomial of γ over F . f (γ − cX)) = X − β. Then g(X) divides f (X) in M [X]. 47 Let γ = α + cβ. then E is a simple extension of F (see Dummit and Foote 1991. Clearly M ⊂ M . ci ∈ F. Before giving an example to illustrate this. β is their only common root. 2 R EMARK 5. The coefﬁcients of such a g(X) are partially symmetric polynomials in the roots of f (X) (that is.1 in the case that E is separable over F . of the permutations of the roots). This gives another proof of Theorem 5. but (equally clearly) g(X) is the minimum polynomial of γ over M . then an element of this form will be a primitive element provided it is moved by every element of Gal(E/F ) except 1. If E is Galois over F . β] = F [γ]. then the extension need not be simple. In fact. When F is inﬁnite. 2 R EMARK 5. because Galois theory shows that there are only ﬁnitely many intermediate ﬁelds in this case (the Galois closure of E over F has only ﬁnitely many intermediate ﬁelds). Let E = F [γ] be a simple algebraic extension of F . f (γ − cβ) = f (α) = 0.3. we need another result. Our hypotheses are minimal: if two of the α’s are not separable. and consequently only ﬁnitely many possible M ’s. F ⊂ M ⊂ E.2.

48

5 APPLICATIONS OF GALOIS THEORY

E XAMPLE 5.5. The simplest nonsimple algebraic extension is k(X, Y ) ⊃ k(X p , Y p ), where k is an algebraically closed ﬁeld of characteristic p. Let F = k(X p , Y p ). For any c ∈ k, we have k(X, Y ) = F [X, Y ] ⊃ F [X + cY ] ⊃ F with the degree of each extension equal to p. If F [X + cY ] = F [X + c Y ], c=c,

then F [X + cY ] would contain both X and Y , which is impossible because [k(X, Y ) : F ] = p2 . Hence there are inﬁnitely many distinct intermediate ﬁelds.16

**Fundamental Theorem of Algebra
**

We ﬁnally prove the misnamed17 fundamental theorem of algebra. T HEOREM 5.6. The ﬁeld C of complex numbers is algebraically closed. P ROOF. Deﬁne C to be the splitting ﬁeld of X 2 + 1 ∈ R[X], and let i be a root of X 2 + 1 in C; thus C = R[i]. We have to show (see 1.44) that every f (X) ∈ R[X] has a root in C. The two facts we need to assume about R are: — Positive real numbers have square roots. — Every polynomial of odd degree with real coefﬁcients has a real root. Both are immediate consequences of the Intermediate Value Theorem, which says that a continuous function on a closed interval takes every value between its maximum and minimum values (inclusive). (Intuitively, this says that, unlike the rationals, the real line has no “holes”.) We ﬁrst show that every element of C has a square root. Write α = a+bi, with a, b ∈ R, and choose c, d to be real numbers such that √ √ (a + a2 + b2 ) (−a + a2 + b2 ) 2 2 c = , d = . 2 2 Then c2 − d2 = a and (2cd)2 = b2 . If we choose the signs of c and d so that cd has the same sign as b, then (c + di)2 = α and so c + di is a square root of α. Let f (X) ∈ R[X], and let E be a splitting ﬁeld for f (X)(X 2 + 1) — we have to show that E = C. Since R has characteristic zero, the polynomial is separable, and so E is Galois over R. Let G be its Galois group, and let H be a Sylow 2-subgroup of G. Let M = E H . Then M is of odd degree over R, and M = R[α] some α (Theorem 5.1). The minimum polynomial of α over R has odd degree and so has a root in R. It therefore has degree 1, and so M = R and G = H.

Zariski showed that there is even an intermediate ﬁeld M that is not isomorphic to F (X, Y ), and Piotr Blass showed in his thesis (University of Michigan 1977), using the methods of algebraic geometry, that there is an inﬁnite sequence of intermediate ﬁelds, no two of which are isomorphic. 17 Because it is not strictly a theorem in algebra: it is a statement about R whose construction is part of analysis (or maybe topology). In fact, I prefer the proof based on Liouville’s theorem in complex analysis to the more algebraic proof given in the text: if f (z) is a polynomial without a root in C, then f (z)−1 will be bounded and holomorphic on the whole complex plane, and hence (by Liouville) constant. The Fundamental Theorem was quite a difﬁcult theorem to prove. Gauss gave a proof in his doctoral dissertation in 1798 in which he used some geometric arguments which he didn’t justify. He gave the ﬁrst rigorous proof in 1816. The elegant argument given here is a simpliﬁcation by Emil Artin of earlier proofs.

16

Cyclotomic extensions

49

We now know that Gal(E/C) is a 2-group. If it is = 1, then it has a subgroup N of index 2 (GT 4.15). The ﬁeld E N has degree 2 over C, and can therefore be obtained by extracting the square root of an element of C (see 3.24), but we have seen that all such elements already lie in C. Hence E N = C, which is a contradiction. Thus E = C. 2 C OROLLARY 5.7. (a) The ﬁeld C is the algebraic closure of R. (b) The set of all algebraic numbers is an algebraic closure of Q. P ROOF. Part (a) is obvious from the deﬁnition of “algebraic closure” (1.43), and (b) follows from Corollary 1.46. 2

Cyclotomic extensions

A primitive nth root of 1 in F is an element of order n in F × . Such an element can exist only if F has characteristic 0 or characteristic p not dividing n. P ROPOSITION 5.8. Let F be a ﬁeld of characteristic 0 or characteristic p not dividing n. Let E be the splitting ﬁeld of X n − 1. (a) There exists a primitive nth root of 1 in E. (b) If ζ is a primitive nth root of 1 in E, then E = F [ζ]. (c) The ﬁeld E is Galois over F ; for each σ ∈ Gal(E/F ), there is an i ∈ (Z/nZ)× such that σζ = ζ i for all ζ with ζ n = 1; the map σ → [i] is an injective homomorphism Gal(E/F ) → (Z/nZ)× . P ROOF. (a) The roots of X n − 1 are distinct, because its derivative nX n−1 has only zero as a root (here we use the condition on the characteristic), and so E contains n distinct nth roots of 1. The nth roots of 1 form a ﬁnite subgroup of E × , and so (see Exercise 3) they form a cyclic group. Any generator will have order n, and hence will be a primitive nth root of 1. (b) The roots of X n − 1 are the powers of ζ, and F [ζ] contains them all. (c) If ζ0 is one primitive nth root of 1, then the remaining primitive nth roots of 1 are i the elements ζ0 with i relatively prime to n. Since, for any automorphism σ of E, σζ0 is i again a primitive nth root of 1, it equals ζ0 for some i relatively prime to n, and the map σ → i mod n is injective because ζ0 generates E over F . It obviously is a homomorphism. m Moreover, for any other nth root of 1, ζ = ζ0 ,

im σζ = (σζ0 )m = ζ0 = ζ i . 2

The map σ → [i] : Gal(F [ζ]/F ) → (Z/nZ)× need not be surjective. For example, if F = C, then its image is {1}, and if F = R, it is either {[1]} or {[−1], [1]}. On the other hand, when n = p is prime, we saw in (1.41) that [Q[ζ] : Q] = p − 1, and so the map is surjective. We now prove that the map is surjective for all n when F = Q. The polynomial X n − 1 has some obvious factors in Q[X], namely, the polynomials X d − 1 for any d|n. The quotient of X n − 1 by all these factors for d < n is called the nth cyclotomic polynomial Φn . Thus Φn = (X − ζ) (product over the primitive nth roots of 1).

50

5 APPLICATIONS OF GALOIS THEORY

It has degree ϕ(n), the order of (Z/nZ)× . Since every nth root of 1 is a primitive dth root of 1 for exactly one d dividing n, we see that Xn − 1 =

d|n

Φd (X).

For example, Φ1 (X) = X − 1, Φ2 (X) = X + 1, Φ3 (X) = X 2 + X + 1, and Φ6 (X) = X6 − 1 = X 2 − X + 1. (X − 1)(X + 1)(X 2 + X + 1)

This gives an easy inductive method of computing the cyclotomic polynomials. Alternatively ask Maple by typing: with(numtheory); cyclotomic(n,X);. Because X n − 1 has coefﬁcients in Z and is monic, every monic factor of it in Q[X] has coefﬁcients in Z (see 1.14). In particular, the cyclotomic polynomials lie in Z[X]. L EMMA 5.9. Let F be a ﬁeld of characteristic 0 or p not dividing n, and let ζ be a primitive nth root of 1 in some extension ﬁeld. The following are equivalent: (a) the nth cyclotomic polynomial Φn is irreducible; (b) the degree [F [ζ] : F ] = ϕ(n); (c) the homomorphism Gal(F [ζ]/F ) → (Z/nZ)× is an isomorphism. P ROOF. Because ζ is a root of Φn , the minimum polynomial of ζ divides Φn . It is equal to it if and only if [F [ζ] : F ] = ϕ(n), which is true if and only if the injection Gal(F [ζ]/F ) → (Z/nZ)× is onto. 2 T HEOREM 5.10. The nth cyclotomic polynomial Φn is irreducible in Q[X]. P ROOF. Let f (X) be a monic irreducible factor of Φn in Q[X]. Its roots will be primitive nth roots of 1, and we have to show they include all primitive nth roots of 1. For this it sufﬁces to show that ζ a root of f (X) =⇒ ζ i a root of f (X) for all i such that gcd(i, n) = 1. Such an i is a product of primes not dividing n, and so it sufﬁces to show that ζ a root of f (X) =⇒ ζ p a root of f (X) for all primes p n. Write Φn (X) = f (X)g(X). Proposition 1.14 shows that f (X) and g(X) lie in Z[X]. Suppose ζ is a root of f but that, for some prime p not dividing n, ζ p is not a root of f . Then ζ p is a root of g(X), g(ζ p ) = 0, and so ζ is a root of g(X p ). As f (X) and g(X p ) have a common root, they have a nontrivial common factor in Q[X] (2.10), which automatically lies in Z[X] (1.14). Write h(X) → h(X) for the map Z[X] → Fp [X], and note that gcdZ[X] (f (X), g(X p )) = 1 =⇒ gcdFp [X] (f (X), g(X p )) = 1.

. but we issue this warning lest anyone attempt to achieve geometric constructions for sections other than the ones suggested by our theory (e. sections into 7. reduce modulo p (where the statement may no longer be true). I’m sticking with Dedekind. . i. and look at the solutions of f = 0 in Fpn .. The regular n-gon is constructible if and only if n = 2k p1 · · · ps where the pi are distinct Fermat primes. We know that Q[ζ] is Galois over Q. The Weil conjectures (Weil 1949. . Let F be a ﬁeld. and that the regular n-gon is constructible when n is as in the Theorem. For example.. . that X n − 1 is solvable (this was before the theory of abelian groups had been developed. 19. Contradiction. The ﬁnal section of Gauss’s. 2 and this is a power of 2 if and only if n has the required form. . .12... commutative algebraists use this method to prove results about commutative rings. . and there are theorems about complex manifolds that were ﬁrst proved by reducing things to characteristic p. The limits of the present work exclude this demonstration here. etc. 9. proved in part by Grothendieck in the 1960’s and completely by Deligne in 1973) assert that the Betti numbers of the space in (a) control the cardinalities of the sets in (b). The regular n-gon is constructible if and only if cos 2π (or ζ = e2πi/n ) is conn structible. and so get a topological space. (b) reduce mod p. “Whenever n−1 involves prime factors other than 2.13. Disquisitiones Arithmeticae (1801) is titled “Equations deﬁning sections of a Circle”. and let G be a group (monoid will do). Hence X n − 1. In it Gauss proves that the nth roots of 1 form a cyclic group. .e. n= pn(p) . parts) and so spend his time uselessly. . §365. but in the absence of further evidence... 18 .. and so f (X) and g(X) have a common factor. 11. when regarded as an element of Fp [X]. This proof is very old — in essence it goes back to Dedekind in 1857 — but its general scheme has recently become popular: take a statement in characteristic zero. and exploit the existence of the Frobenius automorphism a → ap to obtain a proof of the original statement.. But (see GT 3. but we saw in the proof of Proposition 5. For example. P ROOF.10) ϕ(n) = p|n (p − 1)pn(p)−1 .” Ibid. has multiple roots. He also claimed to have proved the converse statement18 .37 and 3. Dedekind’s theorem on the independence of characters T HEOREM 5.WE CAN SHOW WITH ALL RIGOR THAT THESE HIGHER-DEGREE EQUATIONS CANNOT BE AVOIDED IN ANY WAY NOR CAN THEY BE REDUCED TO LOWER-DEGREE EQUATIONS. Xn ]. Then any ﬁnite set {χ1 .g. ai χi = 0 (as a function G → F ) =⇒ a1 = 0. ..23) ζ is constructible if and only if [Q[ζ] : Q] is a power of 2. and so (according to 1. am = 0.11. 13. we are always led to equations of higher degree. 2 R EMARK 5. We can (a) look at the solutions of f = 0 in C. and before Galois). Xn ) ∈ Z[X1 .8 that it doesn’t. χm } of homomorphisms G → F × is linearly independent over F .14 (D EDEKIND ’ S ). let f (X1 .. There are some beautiful and mysterious relations between what happens in characteristic 0 and in characteristic p. R EMARK 5. T HEOREM 5. This leads some people to credit him with the above proof of the irreducibility of Φn .Dedekind’s theorem on the independence of characters 51 But g(X p ) = g(X)p (use the mod p binomial theorem and that ap = a for all a ∈ Fp ).

. If not. The induction hypothesis now shows that ai = 0 for all i ≥ 2. . 2 The normal basis theorem D EFINITION 5. Then the matrix whose (i.17. . For m = 1 it follows from the fact that a nonzero polynomial in one symbol has only ﬁnitely many roots.. j)th -entry is σi αj is invertible. we must have a2 = 0. Assume that F is inﬁnite. . and the induction hypothesis shows that all the remaining ai ’s are also zero. Since every Galois extension of a ﬁnite ﬁeld is cyclic (4. A normal basis for E is an F -basis of the form {σα | σ ∈ G}. Let E be a ﬁnite Galois extension of F with Galois group G. . there exist ci ∈ Ω such that m ci σi (αj ) = 0 for all j. . Since χ2 (g) − χ1 (g) = 0. We have to show that the ai are zero. . . 2 C OROLLARY 5. Then σ1 . which contradicts i=1 Corollary 5. For m = 1.. P ROOF. . . . . . . . am ∈ F . .18). But m ci σi : E → i=1 i=1 Ω is F -linear. i. . αm be a basis for E over F . . f (a1 . . this covers all cases. 2 C OROLLARY 5.52 5 APPLICATIONS OF GALOIS THEORY P ROOF. Xm ] has the property that f (a1 . . we obtain the equation a2 χ2 + · · · + am χm = 0. . Assume it for m − 1. am−1 . Let F1 and F2 be ﬁelds. . × P ROOF. a1 χ1 (x) + a2 χ2 (x) + · · · + am χm (x) = 0 for all x ∈ G. . . an F -basis consisting of the conjugates of an element α of E. . . σm are linearly independent over F2 . am−1 . . . the ﬁrst of which assumes that F is inﬁnite and the second that G is cyclic. am ) = 0 for all a1 . Let α1 . .16. Xm−1 )Xm .. .e. Every Galois extension has a normal basis. we ﬁnd that a1 χ1 (g)χ1 (x) + a2 χ2 (g)χ2 (x) + · · · + am χm (g)χm (x) = 0 for all x ∈ G. and let σ1 . Let E be a ﬁnite separable extension of F of degree m.18 (N ORMAL BASIS THEOREM ).15. Xm ) = 0. . This has the consequence that. . As χ1 and χ2 are distinct. . Apply the theorem to χi = σi |F1 . . . they will take distinct values on some g ∈ G. .. Let E/F be a Galois extension with Galois group G.. σm be distinct homomorphisms F1 → F2 . it’s obvious. ai = ai (χi (g) − χ1 (g)). . P ROOF. . We prove this by induction on m.. Xm ) i ci (X1 . For m > 1. . T HEOREM 5. if f ∈ F [X1 . and suppose that. . On replacing x with gx in the equation. then f (X1 . .. write f= For any m − 1-tuple. Induction on m. σm be distinct F -homomorphisms from E into a ﬁeld Ω. We give two proofs. and so this implies that m ci σi (α) = 0 for all α ∈ E. On multiplying the ﬁrst equation by χ1 (g) and subtracting it from the second. . a1 . and let σ1 .15. . χm } of homomorphisms G → F × and ai ∈ F . for some set {χ1 .

e. On the other hand. say. Then [E : F ] = n. . . A is the m × m matrix having Xk in the (i. . On applying σ1 . τ ∈ G. am−1 ) = 0 for all i. .e. . . . .Hilbert’s Theorem 90. i. Since g is obtained from f by an invertible linear change of variables. . . For any generator α of E as a F [X]-module. let g(Y1 . and so its determinant is ±1. . . .16). . Let G be a ﬁnite group. . . . and so deg P (X) > n − 1. . . A G-module is an abelian group M together with an action of G. . Ym ) = f ( m i=1 Yi σ1 αi . We conclude that P (X) = X n − 1. . . This completes the proof of the lemma in the case that F is inﬁnite. The minimum polynomial of σ0 regarded as an endomorphism of the F -vector space E is the monic polynomial in F [X] of least degree such that P (σ0 ) = 0 (as an endomorphism of E). σm α) (= det(σi σj α)) is nonzero.. It has the property that it divides every polynomial Q(X) ∈ F [X] n such that Q(σ0 ) = 0. 0. . σm (with σ1 = 1). Now assume that G is cyclic generated. am−1 ). . . . . Hence the rows of the matrix are a permutation of the rows of the identity matrix. . the induction hypothesis shows that ci (X1 . . each of its coefﬁcients is zero: ci (a1 . Xm ) ∈ F [X1 . say. Clearly. . σ0 are linearly independent over F . a map G × M → M such that (a) σ(m + m ) = σm + σm for all σ ∈ G. Let f (X1 . Write Xi = X(σi ). . m i=1 Yi σ2 αi . . . . the aj ’s are zero. Xm ] have the property that f (σ1 α. Suppose m j=1 aj σj α = 0 for some aj ∈ F . For a basis α1 . Xm . σ0 α. . m ∈ M . . i. . . Xm−1 ) is zero. But the matrix (σi αj ) is invertible (5. as an F [X]-module with X acting as σ0 . . am ) = 0 for all ai ∈ F . det(A) = h(X1 . . Since this holds for all (a1 . h(1. . . E is isomorphic to F [X]/(X n − 1).14) implies that 1. (c) 1m = m for all m ∈ M . . m ∈ M . . . . Dedekind’s n−1 theorem on the independence of characters (5.. . σm α) = 0 for all α ∈ E. (b) (στ )(m) = σ(τ m) for all σ. Xm ). . . Now number the elements of G as σ1 . . . αm of E over F . Therefore. . σm successively. P (X) divides X n − 1. σ0 αn−1 is a F -basis for E. . . . Since σ0 = 1. . . Because this system of equations is nonsingular. Therefore. by an element σ0 of order n. 0) is the determinant of a matrix having exactly one 1 in each row and each column and its remaining entries 0. and let A = (X(σi σj )). . Then det(A) is a polynomial in X1 . . . 53 is a polynomial in Xm having every element of F as a root. 2 Hilbert’s Theorem 90. . The hypothesis on f implies that g(a1 . h is not identically zero. α. . For this. . . . . and so there exists an α ∈ E × such that h(σ1 α. In particular. m. we obtain a system of m-equations aj σi σj α = 0 in the m “unknowns” aj . Therefore it also is zero. j)th place if σi σj = σk . We shall show that {σi α} is a normal basis. it sufﬁces to show that σi α are linearly independent over F . f can be obtained from g by the inverse linear change of variables. σ0 . . . . .). . . and so g = 0.

f (σ 3 ) = f (σ · σ 2 ) = f (σ) + σf (σ) + σ 2 f (σ) ··· f (σ n ) = f (σ) + σf (σ) + · · · + σ n−1 f (σ). Thus we can deﬁne H 1 (G. x say. Let M be a G-module. f (σ 2 ) = f (σ) + σf (σ). on σ. E XAMPLE 5. if G is a cyclic group of order n generated by σ.e. τ ∈ G. For any σ ∈ G. then a crossed homomorphism f : G → M is determined by its value. it will not be discussed in this course. (*) Conversely. all σ ∈ G. for a ﬁnite group G = σ . if x ∈ M satisﬁes (*). ·) are G-modules.20.19.. M ) = {crossed homomorphisms} {principal crossed homomorphisms} (quotient abelian group). but since this was not done until the twentieth century. E XAMPLE 5. Thus. The cohomology groups H n (G. . Let E be a Galois extension of F with Galois group G. and x satisﬁes the equation x + σx + · · · + σ n−1 x = 0. then a crossed homomorphism is simply a homomorphism. Then (E. and so f (1) = 0. (c) If G acts trivially on M . Note that the condition implies that f (1) = f (1 · 1) = f (1) + f (1). (b) For any x ∈ M . (a) Let f : G → M be a crossed homomorphism. f ↔f (σ) A crossed homomorphism of this form is called a principal crossed homomorphism. M ) have been deﬁned for all n ∈ N. to give an action of G on M is the same as to give a homomorphism G → Aut(M ) (automorphisms of M as an abelian group). A crossed homomorphism is a map f : G → M such that f (στ ) = f (σ) + σf (τ ) for all σ. The sum and difference of two crossed homomorphisms is again a crossed homomorphism.54 5 APPLICATIONS OF GALOIS THEORY Thus. σm = m for all σ ∈ G and m ∈ M . we obtain a crossed homomorphism by putting f (σ) = σx − x. and there are no nonzero principal crossed homomorphisms. Thus. then the formulas f (σ i ) = x + σx + · · · + σ i−1 x deﬁne a crossed homomorphism f : G → M . +) and (E × . and the sum and difference of two principal crossed homomorphisms is again principal. i. there is a one-to-one correspondence {crossed homs f : G → M } ←→ {x ∈ M satisfying (*)}.

We deﬁne the norm of an element α ∈ E to be Nm α = σα. a Γ -module M will deﬁne a sheaf M on X. for σ ∈ G. The norm map C× → R× is α → |α|2 and the norm map Q[ d]× → Q× √ is a + b d → a2 − db2 .23. √ E XAMPLE 5. Therefore. In multiplicative notation. 55 ˜ E XAMPLE 5. and H 1 (X. T HEOREM 5. every crossed homomorphism G → E × is principal.. Z). σ∈G τ σα = Nm α. Clearly an element of the form τβ has norm 1.15 implies that τ ∈G σγ γ for all σ ∈ G. The map α → Nm α : E × → F × is a obviously a homomorphism. σ(f (τ )) · στ (α) f (σ)−1 f (στ ) · στ (α) τ ∈G = f (σ) f (στ )στ (α). β σ(β) 2 which equals f (σ)−1 β because. so also does στ . f (σ) = and we can take β = γ −1 . then H 1 (G. Let E be a Galois extension of F with group G. τ ∈ G. for cyclic extensions. as τ runs over G. P ROOF. and our next result show that. For example.e. E × ) = 0. and let Γ be the group of covering transformations.21.e. Because the f (τ ) are f (τ )τ : E → E is not the zero map. Under some fairly general hypotheses. We are interested in determining the kernel of the norm map. Let π : X → X be the universal covering space of a topological space X. σ. i. Corollary 5. when M = Z with the trivial action of Γ . Let f be a crossed homomorphism G → E × . there exists an α ∈ E such that β= But then.22.Hilbert’s Theorem 90. and we have to ﬁnd a γ ∈ E × such that f (σ) = nonzero. . Let E be a Galois extension of F with Galois group G. this becomes the isomorphism H 1 (X. f (στ ) = f (σ) · σ(f (τ )). Z) H 1 (Γ. τ (Nm α) = and so Nm α ∈ F . σ∈G For τ ∈ G. all elements with β norm 1 are of this form. this means.. Z) = Hom(Γ. M ). i. M) H 1 (Γ. σβ = = τ ∈G τ ∈G −1 df τ ∈G f (τ )τ α = 0.

Lemmermeyer and N. (c) If F has characteristic p (hence has no pth roots of 1 other than 1). Let E be such an extension of F .e.24 (H ILBERT ’ S THEOREM 90). VIII. and generalized to Theorem 5. R EMARK 5. (b) Let ζ be a primitive nth root of 1 in F . 1897. there is an element β ∈ E such that σβ = ζβ. and let σ generate Gal(E/F ).25 and Remark 5. Noether. and. Moreover. §9. P ROOF. Hence the splitting ﬁeld of X n − a is F [α]. are also referred to as Hilbert’s Theorem 90. if E is cyclic of degree n over F . Then Nm ζ = ζ n = 1. then X p − X − a is irreducible in F [X] unless a = bp − b for some b ∈ F . and when it is irreducible. (b) The polynomial X n −a. 1965. and so only the identity element of Gal(E/F ) ﬁxes β — we conclude by the fundamental theorem of Galois theory that E = F [β]. Let F be a ﬁeld containing a primitive nth root of 1. Then. 19 This is Satz 90 in Hilbert’s book. see the introduction to the English translation (Springer 1998) written by F. every extension of F which is cyclic of degree p is the splitting ﬁeld of such a polynomial. two cyclic exten1 1 sion F [a n ] and F [b n ] of F are isomorphic if and only if a and b generate the same subgroup of F × /F ×n . it is possible to extend this to a classiﬁcation of the Galois extensions of F with abelian Galois group of exponent n (i. n ≥ 2. its Galois group is cyclic of order p (generated by α → α + 1 where α is a root). and so. For an illuminating discussion of Hilbert’s book.22. if NmE/F α = 1. 2 R EMARK 5.26a classify the cyclic extensions of F order n in the case that F contains a primitive nth root of 1. 2 Cyclic extensions. then a ∈ −4F 4 . Theorem 5. Algebra. Theorem 5. The map σ → σα is an α injective homomorphism from Gal(F [α]/F ) into the cyclic group ζ . T HEOREM 5. and so (see 5. with Galois group a quotient of (Z/nZ)r for some r).22 by E. (a) If α is one root of X n − a. and let S(E) = {a ∈ F × | a becomes an nth power in E}. and so β n ∈ F. (b) Conversely. / Addison-Wesley.27 (K UMMER THEORY ). then α = β/σβ for some β ∈ E. Theorie der Algebraischen Zahlk¨ rper. hence E is the splitting ﬁeld of X n − β n . Schappacher. Theorem 5. The theorem was discovo ered by Kummer in the special case of Q[ζp ]/Q. is irreducible in F [X] under the following condition: a is not a pth power for any p dividing n. according to Hilbert’s Theorem 90.56 5 APPLICATIONS OF GALOIS THEORY C OROLLARY 5. if 4|n. (a) Assume F contains a primitive nth root of 1. We are now able to classify the cyclic extensions of degree n of a ﬁeld F in the case that F contains n nth roots of 1.20a) there is a crossed homomorphism f : σ → E × with f (σ) = α. Then σ i β = ζ i β.. The condition on α is that α · σα · · · σ m−1 α = 1. Let m = [E : F ]. 19 Let E be a ﬁnite cyclic extension of F with Galois group σ . . On the other hand σβ n = ζ n β n = β n . then there is an element β ∈ E such that E = F [β] and β n ∈ F . (a) The Galois group of X n − a is cyclic of order dividing n.26. then the other roots are the elements of the form ζα with ζ an n th root of 1. P ROOF. See Lang.25. Under the same assumption on F . as well as various vast generalizations of it.22 now shows that f is principal. which means that there is a β with f (σ) = β/σβ. Theorem 16.

Let f ∈ F [X] be separable. Theorem 5. and let Fi = E Gi . : a F ×n In particular. . P ROOF. each i.28.6a). P ROOF.29. The lemma shows that the Galois group G of f as an element of F [X] is a subgroup of Gf . ⇐=: Let f ∈ F [X] have solvable Galois group Gf . (Cf.6a). Any element of Gal(E /F ) permutes the αi and so maps E into itself. and the map E → S(E) deﬁnes a oneto-one correspondence between the abelian extensions of E of exponent n and the groups S(E). We are given that there exists a tower of ﬁelds F = F 0 ⊂ F1 ⊂ F2 ⊂ · · · ⊂ Fm such that r (a) Fi = Fi−1 [αi ]. . . and this shows that f is solvable.. .. such that (S(E) : F ×n ) < ∞. σ) → σa S(E) × Gal(E/F ) → µn (group of nth roots of 1). =⇒: It sufﬁces to show that Gf is a quotient of a solvable group (GT 6. The ﬁeld E is recovered from S(E) as the splitting ﬁeld of (X n − a) (product over a set of representatives for S(E)/F ×n ).) Proof of Galois’s solvability theorem L EMMA 5. αm be the roots of f (X) in E . [F :F ] . Then the Galois group of f as an element of F [X] is a subgroup of that of f as an element of F [X]. Then E = F [α1 . αi i ∈ Fi−1 . Hence it ˜ ˜ sufﬁces to ﬁnd a solvable extension E of F such that f (X) splits in E[X].25b shows that Fi = Fi−1 [αi ] with αi i i−1 ∈ Fi−1 . Let F be a ﬁeld of characteristic 0. (b) Fm contains a splitting ﬁeld for f. This means that there is a sequence of subgroups G = G0 ⊃ G1 ⊃ · · · ⊃ Gm−1 ⊃ Gm = {1} such that each Gi is normal in Gi−1 and Gi−1 /Gi is cyclic. [E : F ] = (S(E) : F ×n ). F × ⊃ S(E) ⊃ F ×n . n = (deg f )! will do. We have a sequence of ﬁelds F ⊂ F [ζ] = F = F0 ⊂ F1 ⊂ F2 ⊂ · · · ⊂ Fm = E with Fi cyclic over Fi−1 . A polynomial in F [X] is solvable if and only if its Galois group is solvable.. and let F be an extension ﬁeld of F . Let E be a splitting ﬁeld of f (X) over F . Moreover. . αm ] is a splitting ﬁeld of f over F . there is a perfect pairing (a. Let E be a splitting ﬁeld for f over F . and let α1 . Let F = F [ζ] where ζ is a primitive nth root of 1 for some large n — for example. The map σ → σ|E is an injection Gal(E /F ) → Gal(E/F ). 2 T HEOREM 5.Proof of Galois’s solvability theorem 57 Then S(E) is a subgroup of F × containing F ×n . Exercise 5 for the case n = 2. and hence is also solvable (GT 6.

and let E be the Galois closure of Fm [ζ] in Ω. . Xn ). .e. Xn ] is said to be symmetric if it is unchanged when its variables are permuted. tn ). 2 The general polynomial of degree n When we say that the roots of aX 2 + bX + c are −b ± b2 − 4ac 2a we are thinking of a. c as variables: for any particular values of a.17a). . . . rm . each of these ˜ extensions is cyclic. . α2 . . . We deﬁne the general polynomial of degree n to be f (X) = X n − t1 X n−1 + · · · + (−1)n tn ∈ F [t1 . Xn ) ∈ R[X1 . We shall prove in this section that there is no similar formula for the roots of the “general polynomial” of degree ≥ 5. . if P (Xσ(1) . E over F by the elements ζ. We shall show that. A polynomial P (X1 . σα1 . .8) and (5.. b. and let Ω be a ﬁeld Galois over F and containing (a copy of) Fm and a primitive nth root ζ of 1. Xi Xj i<j<k Xi Xj Xk . . For example p1 p2 p3 pr pn = = = ··· = ··· = i Xi √ all σ ∈ Sn . . i. We adjoin these elements to F one by one to get a sequence of ﬁelds ˜ F ⊂ F [ζ] ⊂ F [ζ.Xir X1 X2 · · · Xn . αm . or n). = X1 X2 + X1 X3 + · · · + X1 Xn + X2 X3 + · · · + Xn−1 Xn . its Galois group is Sn .. . .1).. and so E/F is a solvable extension. According to (5. σ α1 . choose a primitive element γ for Fm over F (see 5. = X1 X2 X3 + · · · Xi1 .25a). . α1 . . α1 ] ⊂ · · · ⊂ F ⊂ F ⊂ · · · ⊂ E in which each ﬁeld F is obtained from its predecessor F by adjoining an rth root of an element of F (r = r1 . .. . . . . . . .58 5 APPLICATIONS OF GALOIS THEORY Let n = r1 · · · rm . Xσ(n) ) = P (X1 . when we regard f as a polynomial in X with coefﬁcients in the ﬁeld F (t1 . .. .. Symmetric polynomials Let R be a commutative ring (with 1). . c. the formula gives the roots of the particular equation.. tn ][X] where the ti are variables.29 proves the above remark (at least in characteristic zero). . σαm . Then Theorem 5. . .. σ ∈ G.. ˜ is the composite of the ﬁelds σFm [ζ]. For example. b. and take Ω to be a splitting ﬁeld of g(X)(X n − 1) where g(X) is the minimum polynomial of γ over F . i<j i1 <···<ir = X1 + X2 + · · · + Xn . and so it is generated According to (3. ˜ Let G be the Galois group of Ω/F . . . . .

. and it follows that the highest monomial in d1 p1 · · · pdn is n d d d X1 1 +d2 +···+dn X2 2 +···+dn · · · Xnn . . .. Every symmetric polynomial P (X1 . P ROOF. and so the theorem implies that every symmetric polynomial in the roots of f (X) lies in R. i. and after a ﬁnite number of steps. and suppose that f splits over some ring S containing R: f (X) = Then a1 = −p1 (α1 .. . is = js . . a2 = p2 (α1 . Xn ).30 (S YMMETRIC POLYNOMIALS THEOREM )... Hence k1 ≥ k2 ≥ · · · ≥ kn . . the discriminant D(f ) = i<j (αi − αj )2 of f lies in R. αn ).The general polynomial of degree n 59 are all symmetric because pr is the sum of all monomials of degree r made up out of distinct Xi ’s. .. Xn ) − cpk1 −k2 pk2 −k3 · · · pkn is strictly less n 1 2 than the highest monomial in P (X1 . αn ). but is+1 > js+1 . Xn ) in R[X1 . We deﬁne an ordering on the monomials in the Xi by requiring that j j i i j i X11 X22 · · · Xnn > X11 X22 · · · Xnn if either i1 + i2 + · · · + in > j1 + j2 + · · · + jn or equality holds and.. . . Because k k P is symmetric. . These particular polynomials are called the elementary symmetric polynomials. i1 = j1 . pn ]. . . .. P ∈ R[p1 . Thus the elementary symmetric polynomials in the roots of f (X) lie in R. We can repeat this argument with the polynomial on the left. we will arrive at a representation of P as a polynomial in p1 . an = ±pn (α1 . . . . For example. .. 2 Let f (X) = X n + a1 X n−1 + · · · + an ∈ R[X]. The highest monomial in pi is X1 · · · Xi . . . .. . n i=1 (X − αi ). k k Let X1 1 · · · Xnn be the highest monomial occurring in P with a coefﬁcient c = 0.e. .. . . .. . Xn ] is equal to a polynomial in the elementary symmetric polynomials with coefﬁcients in R. . pn . 3 2 2 X 1 X 2 X 3 > X1 X 2 X 3 > X1 X 2 X 3 . for some s. . αn ). . For example. . it contains all monomials obtained from X1 1 · · · Xnn by permuting the X’s. . .. . (∗) Therefore the highest monomial of P (X1 . T HEOREM 5. αi ∈ S. .

31 (S YMMETRIC FUNCTIONS THEOREM ). 2 As we noted in the second proof. .. . The Galois group of the general polynomial of degree n is Sn . Xn )Sn . . . . f (X) = X n − t1 X n−1 + · · · + (−1)n tn ∈ F [t1 . . . 2 This can also be proved by noting that. Xn ) is Galois over F (p1 . .. . Therefore the statement will follow from Corollary 5. . . . . pn ) with Galois group Sn (acting by permuting the Xi ). because F (X1 . . pn ] (5. .33. tn ) must be the zero polynomial. Clearly F (p1 . . The polynomials H = σ∈Sn σh and Hf are symmetric. . and so this follows from (3. Xn ) is algebraic over F (p1 . .. The general polynomial of degree n T HEOREM 5. . cancellation can’t occur. .e.. the Galois group of g(T ) ∈ F (p1 . . [E : E Sn ] ≥ n! by (2.. P ROOF. . Suppose on the contrary that there exists a P (t1 . . Xn ]. . tn ) such that P (p1 ... . . . When Sn acts on E = F (X1 . . the latter must have transcendence degree n (see §8). . . ﬁxed by Sn ). .. tn ) and m2 (t1 . . Xn ) be symmetric (i. tn ) are distinct monomials. . . pn ) = F (X1 . . We now prove that the pi are algebraically independent20 . [E : F (p1 . . The ﬁeld F (X1 . pn ) and m2 (p1 . then m1 (p1 . . .. pn ). F (X1 . . . . . and therefore lie in F [p1 . . . ... pn ) sending f (X) to g(X) = X n − p1 X n−1 + · · · + (−1)n pn ∈ F (p1 . . . .60 Symmetric functions 5 APPLICATIONS OF GALOIS THEORY T HEOREM 5. . pn )[X]. . Therefore. . . Hence their quotient f = Hf /H lies in F (p1 .30). . . . 2 C OROLLARY 5. .8). . .. .32. . . . p59. .. . . . . . . pn )] ≤ n! because E is the splitting ﬁeld of (T −X1 ) · · · (T − Xn ) over F (p1 . . . . . Let f (X) be the general polynomial of degree n. pn ) ⊂ E Sn ⊂ E. . . . h ∈ F [X1 .32. on the other. pn ) have distinct highest monomials. tn ][X]. and so P (t1 . F IRST P ROOF. . g. . P ROOF. . If we can show that the map ti → pi : F [t1 . Let f ∈ F (X1 . pn ). . . . .10). pn ). tn ] → F [p1 . .. see p79). . . . . the pi are algebraically independent over F . . . . We have shown that F (p1 . . . . . Xn ) by permuting the Xi ’s. then it will extend to an isomorphism F (t1 .e. . pn ) = 0. On the one hand. . . pn ] is injective (i. Set f = g/h. Xn ) is the splitting ﬁeld over F (p1 . . Equation (∗). the ﬁeld of invariants is F (p1 . .2 S ECOND P ROOF. Therefore. . pn ).. . . . 20 . tn ) → F (p1 . . . . . pn )[T ] is Sn .. . pn ) of g(T ) = (T − X1 ) · · · (T − Xn ) = X n − p1 X n−1 + · · · + (−1)n pn . shows that if m1 (t1 ..

see Serre. The discriminant D(f ) of f (X) = X n − T1 X n−1 + · · · + (−1)n Tn is a polynomial in C[T1 .. Lectures on the Mordell-Weil Theorem. cA (X) = det(XIn − A) (trace of A) (determinant of A) (characteristic polynomial of A). realize E as the splitting ﬁeld of a polynomial f (X) ∈ k[t1 . . J.. in about 1820. R EMARK 5. J. 1989. the Babylonians (at least) knew a general formula for the roots of a quadratic polynomial.. Over the next 275 years there were many fruitless attempts to obtain similar formulas for higher degree polynomials.34... an ) is the set of roots of the polynomial X n − a1 X n−1 + · · · + (−1)n an . For the proof. but does every ﬁnite Galois group occur as a Galois group over Q itself? The Hilbert-Noether program for proving this was the following. . until. Consider the projection π : S → Cn . and used Cardan’s result). Xn ) by permuting the Xi . . t1 . For more information on the problem. . Ferrari (about 1545 AD) found a general formula for the roots of quartic polynomial (he introduced the resolvent cubic.. . tn ) (the ti are symbols). . 1992.. Hence this approach can not lead to a proof that all ﬁnite groups occur as Galois groups over Q. . it follows that every ﬁnite group occurs as a Galois group over some ﬁnite extension of Q. Xn )G ≈ F (t1 . Tn ]. Its ﬁbre over a point (a1 . . ..-P. (x. Chapter 10. . tn ][X] and prove that for inﬁnitely many values of the ti . but it doesn’t exclude other approaches. . . However.. and S π −1 (∆) is a covering space over Cn ∆. tn ). there are exactly n points of S. .-P. and every ﬁnite group occurs as a subgroup of some Sn . . . and Serre. tn ) → (t1 . . . Then over each point of Cn ∆.. It is a beautiful complex manifold S of dimension n.Norms and traces 61 R EMARK 5. . for an n × n matrix A = (aij ) Tr(A) = i aii det(A) = σ∈Sn sign(σ)a1σ(1) · · · anσ(n) . . Norms and traces Recall that. Swan proved in 1969 that the conjecture is false for G the cyclic group of order 47. . Cardan (about 1515 AD) found a general formula for the roots of a cubic polynomial. ibid. . then it occurs inﬁnitely often as a Galois group over Q. the polynomial you obtain in Q[X] has Galois group G. then F (X1 . . . . Since Sn occurs as a Galois group over Q. . and consider the subset of Cn+1 deﬁned by the equation X n − T1 X n−1 + · · · + (−1)n Tn = 0. ..35. Let ∆ be the zero set of D(f ) in Cn . A brief history As far back as 1500 BC. . Rufﬁni and Abel proved that there are none.) Noether conjectured the following: Let G ⊂ Sn act on F (X1 . Chapter 9. Take F = C. tn ) (for variables ti ). Hilbert proved that if G occurs as the Galois group of an extension E ⊃ Q(t1 . Topics in Galois Theory. (This is quite a difﬁcult theorem — see Serre.. .

62 Moreover,

5 APPLICATIONS OF GALOIS THEORY

cA (X) = X n − Tr(A)X n−1 + · · · + (−1)n det(A). None of these is changed when A is replaced by its conjugate U AU −1 by an invertible matrix U . Therefore, for any endomorphism α of a ﬁnite dimensional vector space V , we can deﬁne21 Tr(α) = Tr(A), det(α) = det(A), cα (X) = cA (X) where A is the matrix of α with respect to any basis of V . If β is a second endomorphism of V , Tr(α + β) = Tr(α) + Tr(β); det(αβ) = det(α) det(β). Now let E be a ﬁnite ﬁeld extension of F of degree n. An element α of E deﬁnes an F -linear map αL : E → E, x → αx, and we deﬁne TrE/F (α) = Tr(αL ) NmE/F (α) = det(αL ) cα,E/F (X) = cαL (X) (trace of α) (norm of α) (characteristic polynomial of α).

Thus, TrE/F is a homomorphism (E, +) → (F, +), and NmE/F is a homomorphism (E × , ·) → (F × , ·). E XAMPLE 5.36. (a) Consider the ﬁeld extension C ⊃ R. For α = a + bi, the matrix of αL with respect to the basis {1, i} is a −b , and so b a TrC/R (α) = 2 (α), NmC/R (α) = |α|2 . (b) For a ∈ F , aL is multiplication by the scalar a. Therefore TrE/F (a) = na , NmE/F (a) = an , ca,E/F (X) = (X − a)n where n = [E : F ]. Let E = Q[α, i] be the splitting ﬁeld of X 8 − 2. To compute the trace and norm of α in E, the deﬁnition requires us to compute the trace and norm of a 16 × 16 matrix. The next proposition gives us a quicker method. P ROPOSITION 5.37. Let E/F be a ﬁnite extension of ﬁelds, and let f (X) be the minimum polynomial of α ∈ E. Then cα,E/F (X) = f (X)[E:F [α]] .

21

The coefﬁcients of the characteristic polynomial cα (X) = X n + c1 X n−1 + · · · + cn ,

**of α have the following description ci = (−1)i Tr(α| — see Bourbaki, N., Algebra, Chapter 3, 8.11.
**

i

V)

Norms and traces

63

P ROOF. Suppose ﬁrst that E = F [α]. In this case, we have to show that cα (X) = f (X). Note that α → αL is an injective homomorphism from E into the ring of endomorphisms of E as a vector space over F . The Cayley-Hamilton theorem shows that cα (αL ) = 0, and therefore cα (α) = 0. Hence f |cα , but they are monic of the same degree, and so they are equal. For the general case, let β1 , ..., βn be a basis for F [α] over F , and let γ1 , ..., γm be a basis for E over F [α]. As we saw in the proof of (1.20), {βi γk } is a basis for E over F . Write αβi = aji βj . Then, according to the ﬁrst case proved, A =df (aij ) has characteristic polynomial f (X). But αβi γk = aji βj γk , and so the matrix of αL with respect to {βi γk } breaks up into n × n blocks with A’s down the diagonal and zero matrices elsewhere, from which it follows that cαL (X) = cA (X)m = f (X)m . 2 C OROLLARY 5.38. Suppose that the roots of the minimum polynomial of α are α1 , . . . , αn (in some splitting ﬁeld containing E), and that [E : F [α]] = m. Then Tr(α) = m

n i=1 αi ,

NmE/F α = (

m n i=1 αi ) .

P ROOF. Write the minimum polynomial of α as f (X) = X n + a1 X n−1 + · · · + an = so that a1 = − αi , and an = (−1)n αi . Then cα (X) = (f (X))m = X mn + ma1 X mn−1 + · · · + am , n so that TrE/F (α) = −ma1 = m αi , and NmE/F (α) = (−1)mn am = ( αi )m . n E XAMPLE 5.39. (a) Consider the extension C ⊃ R. If α ∈ C \ R, then cα (X) = f (X) = X 2 − 2 (α)X + |α|2 . If α ∈ R, then cα (X) = (X − a)2 . (b) Let E be the splitting ﬁeld of X 8 −2. Then E has degree 16 over Q and is generated √ √ by α = 8 2 and i = −1 (see Exercise 16). The minimum polynomial of α is X 8 − 2, and so cα,Q[α]/Q (X) = X 8 − 2, cα,E/Q (X) = (X 8 − 2)2 TrQ[α]/Q α = 0, TrE/Q α = 0 NmQ[α]/Q α = −2, NmE/Q α = 4 R EMARK 5.40. Let E be a separable extension of F , and let Σ be the set of F -homomorphisms of E into an algebraic closure Ω of F . Then TrE/F α = NmE/F α =

σ∈Σ σα σ∈Σ σα. 2

(X − αi ),

64

5 APPLICATIONS OF GALOIS THEORY

When E = F [α], this follows from 5.38 and the observation (cf. 2.1b) that the σα are the roots of the minimum polynomial f (X) of α over F . In the general case, the σα are still roots of f (X) in Ω, but now each root of f (X) occurs [E : F [α]] times (because each F -homomorphism F [α] → Ω has [E : F [α]] extensions to E). For example, if E is Galois over F with Galois group G, then TrE/F α = NmE/F α =

σ∈G σα σ∈G σα.

P ROPOSITION 5.41. For ﬁnite extensions E ⊃ M ⊃ F , we have TrE/M ◦ TrM/F = TrE/F , NmE/M ◦ NmM/F = NmE/F . P ROOF. If E is separable over F , then this can be proved fairly easily using the descriptions in the above remark. We omit the proof in the general case. 2 P ROPOSITION 5.42. Let f (X) ∈ F [X] factor as f (X) = m (X − αi ) in some splitting i=1 df ﬁeld, and let α = α1 . Then, with f = dX (formal derivative), we have disc f (X) = (−1)m(m−1)/2 NmF [α]/F f (α). P ROOF. Compute that disc f (X) =

i<j df

(αi − αj )2 (

i

= (−1)m(m−1)/2 · = (−1)m(m−1)/2 ·

i

(αi − αj ))

j=i

f (αi ) (by 5.40).

2

= (−1)m(m−1)/2 NmF [α]/F (f (α)) E XAMPLE 5.43. We compute the discriminant of f (X) = X n + aX + b, a, b ∈ F,

assumed to be irreducible and separable, by computing the norm of γ = f (α) = nαn−1 + a, On multiplying the equation αn + aα + b = 0 by nα−1 and rearranging, we obtain the equation nαn−1 = −na − nbα−1 . Hence γ = nαn−1 + a = −(n − 1)a − nbα−1 .

df

f (α) = 0.

it must be the minimum polynomial of γ. and so the minimum polynomial of γ over F has degree n also. 2mn. let Ga be the Galois group of X 4 + X 3 + X 2 + X + a. For example. t ∈ Q. a3 . namely. . b ∈ Q of Pythagoras’s equation a2 + b2 = 1 are of the form s2 − t2 2st a= 2 . (−nb)n (γ + (n − 1)a)n = =0 b αn b and P (X) is monic of degree n. As Exercises 5-1 (*). b= 2 . it is clear that F [α] = F [γ]. 5-2 (*). which is something Maple V doesn’t know (because it doesn’t understand symbols as exponents). m2 + n2 ) for some integers d. disc(X 5 + aX + b) = 55 b4 + 44 a5 . s. m. For a ∈ Q. Find integers a1 . γ + (n − 1)a 65 From the last two equations. s + t2 s + t2 and deduce that any right triangle with integer sides has sides of length d(m2 − n2 . Prove that the rational solutions a. then P (γ) = f (α) · Q(γ) = 0.) 5-3 (*). Therefore Nm γ is (−1)n times the constant term of P (X). disc(X n + aX + b) = (−1)n(n−1)/2 (nn bn−1 + (−1)n−1 (n − 1)n−1 an ). If we write f −nb P (X) = X + (n − 1)a Q(X) P (X) = (X + (n − 1)a)n − na(X + (n − 1)a)n−1 + (−1)n nn bn−1 Q(X) = (X + (n − 1)a)n /b. a2 . a4 such that i = j =⇒ Gai is not isomorphic go Gaj .Exercises Solving for α gives α= −nb . and n (Hint: Apply Hilbert’s Theorem 90 to the extension Q[i]/Q. Therefore. Q(γ) = Nm γ = nn bn−1 + (−1)n−1 (n − 1)n−1 an . Prove that a ﬁnite extension of Q can contain only ﬁnitely many roots of 1.

either s ≤ t or t ≤ s. the Axiom of Choice). and anti-symmetric (a ≤ b and b ≤ a =⇒ a = b). Let (S. transitive. but we need to ﬁnd a way of doing this that is allowed by the axioms of set theory. Every nonzero commutative ring A has a maximal ideal (meaning. and used for algebraic purposes in Zorn 1935. Roughly speaking. we prove that Zorn’s lemma implies that every ﬁeld F has an algebraic closure Ω.22). then Ω = Ei is an algebraic closure of F . (d) A maximal element of a partially ordered set S is an element s such that s ≤ s =⇒ s=s. Ei = a splitting ﬁeld of fi over Ei−1 . . Keating. Then S has a maximal element. then the existence of Ω can be proved as in the ﬁnite ﬁeld case (4. we sketch three solutions22 to the problem. Zorn attributed to Artin the realization that the ‘lemma’ is in fact equivalent to the Axiom of Choice (see Jech 1973). Zorn’s lemma D EFINITION 6. .J. although widely used (allegedly ﬁrst by Lefschetz). The set S of ﬁnite subsets of an inﬁnite set doesn’t contradict Zorn’s lemma. it was ﬁrst brought to prominence. There do exist naturally occurring ﬁelds. (c) An upper bound for a subset T of a partially ordered set (S. After reviewing the statement of Zorn’s lemma. for any ﬁeld F there is a ring F [[T ]] of formal power series i≥0 ai T i . we would like to take a union of a family of splitting ﬁelds indexed by the monic irreducible polynomials in F [X].66 6 ALGEBRAIC CLOSURES 6 Algebraic closures In this section. As a ‘maximum principle’. An Introduction to Rings and Modules.2 (Z ORN ’ S ). but it has no maximal elements. because it contains totally ordered subsets with no upper bound in S. For example. 22 . most notably in Kuratowski 1922. The difﬁculty in showing the existence of an algebraic closure of an arbitrary ﬁeld F is in the set theory.E. not contained in C. ≤) be a nonempty partially ordered set for which every totally ordered subset has an upper bound in S.46). R EMARK 6. (b) A partial ordering is a total ordering if. L EMMA 6. and its ﬁeld of fractions is uncountable even if F is ﬁnite. Zorn’s lemma23 is equivalent to the Axiom of Choice. A partially ordered set need not have any maximal elements. has attracted the attention of historians (Campbell 1978). for example. 23 The following is quoted from A. Berrick and M. f2 . P ROPOSITION 6. E0 = F . The following proposition is a typical application of Zorn’s lemma — we shall use a * to signal results that depend on Zorn’s lemma (equivalently. the set of ﬁnite subsets of an inﬁnite set is partially ordered by inclusion. ..1. for all s. Zorn’s contribution was to observe that it is more suited to algebraic applications like ours. ≤) is an element s ∈ S such that t ≤ s for all t ∈ T . t ∈ T . maximal among proper ideals). the set of monic irreducible polynomials in F [X] is countable. and so we can list them f1 . If F is countable. then the algebraic closure of F in C is an algebraic closure of F (1. and hence independent of the axioms of set theory. 2000: The name of the statement.3.4 (*). (a) A relation ≤ on a set S is a partial ordering if it reﬂexive. deﬁne Ei inductively by. that are uncountable. Recall that if F is a subﬁeld C. namely. ai ∈ F . apparently in ignorance of its previous usage in topology.

Let S be the set triples (E. (xi ) − a(yi ) with xj = ayj for one j ∈ I and xi = yi for all i = j. Let (Ai )i∈I be a family of commutative F -algebras.] generated by the polynomials f (xf ). .9)25 . and let Ω = ( F Ef )/M where M is a maximal ideal in F Ef (whose existence is ensured by Zorn’s lemma). and it is proper because if 1 ∈ J then 1 ∈ I for some I in T . being a homomorphism of ﬁelds.44).]) for some gi ∈ F [. .). ·) contains F as a subﬁeld and is algebraic over it. . Let F be an extension of F such that each fi . . p144 for the details. which is a maximal ideal in A.4). p143). + . I: Les structures fondamentales de l’analyse. +. has a root αi in F . Chapters 1–3. The algebraic closure of F in Ω is therefore an algebraic closure of F (by 1. . xxiv+709 pp. . . Chapitre 4: Polynomes et fractions rationnelles. and then show that a maximal element is algebraically closed. II. · ) if the ﬁrst is a subﬁeld of the second. Nicolas. N. . . Thus J is an upper bound for T . Translated from the French. Artin. . Berlin. . 5 §4. 1959. . Now Zorn’s lemma implies that S has a maximal element. Reprint of the 1974 edition. . 1989. If T is a totally ordered set of ideals. n. El´ ments de math´ matique. . . . then J = I∈T I is again an ideal. (See ibid. . Chap. (Bourbaki. and I would not be proper. Apply Zorn’s lemma to show that S has maximal elements. For each polynomial f ∈ F [X]. . . 1102 Hermann. The composite of the F -homomorphisms Ef → F Ef → Ω. i = 1.] in a family of symbols xf indexed by the nonconstant monic polynomials f ∈ F [X]. After (4. If 1 lies in the ideal I in F [. it must also split in the larger ﬁeld Ω. Second proof of the existence of algebraic closures (Jacobson 1964. Deuxi` me e e ´ edition. MR 30 e #4751 25 Bourbaki. Let S be the set of all proper ideals in A. 1989. This implies that the cardinality of any ﬁeld algebraic over F is the same as that of F (ibid. Note that F ⊂ F Ef and M ∩ F = 0. e Livre II: Alg` bre. Under the F -homomorphism 24 ´ e Bourbaki. Write (E. . and identify F with a subset of Ξ. Elements of Mathematics. 3. 2 First proof of the existence of algebraic closures (Bourbaki.] and some nonconstant monic fi ∈ F [X]. choose a splitting ﬁeld Ef . xf . ..) Third proof of the existence of algebraic closures (E. Springer-Verlag.First proof of the existence of algebraic closures 67 P ROOF. . xf . No. is injective. Paris 1959 iv+222 pp. Since f splits in Ef . . Choose an uncountable set Ξ of cardinality greater than that of F . I. then g1 f1 (xf1 ) + · · · + gn fn (xfn ) = 1 (in F [. . xf . . ·) with E ⊂ Ξ and (+. . xf . and deﬁne F Ai to be the quotient of the F -vector space with basis i∈I Ai by the subspace generated by elements of the form: (xi ) + (yi ) − (zi ) with xj + yj = zj for one j ∈ I and xi = yi = zi for all i = j. and there are canonical homomorphisms Ai → F Ai of F -algebras. see Dummit and Foote 1991. Chapitre 5: Corps commutatifs. Algebra. p144. Fascicule XI. .)24 An F -algebra is a ring containing F as a subring. . . It can be made into a commutative F -algebra in an obvious fashion. +. Actualit´ s Scientiﬁques et Industrielles. partially ordered by inclusion. +. As Ω has no ideals other than (0) and Ω.22) we may assume F to be inﬁnite. ·) a ﬁeld structure on E such that (E. ·) ≤ (E .2). Chap. (2 inserts). Consider the polynomial ring F [. and it is a ﬁeld (see 1. 13.

. This makes S into a partially ordered set. ϕM ) with M a ﬁeld F ⊂ M ⊂ E and ϕM an F -homomorphism M → Ω.]. Suppose that M = E. E2 is algebraic over E1 and therefore also over F (see 1. given two algebraic closures Ω. xf .46). Then E1 is a ﬁeld containing (a copy of) F in which every nonconstant polynomial in F [X] has at least one root. and let f (X) be the minimum polynomial of α. . There exists an F -homomorphism E → Ω. αn . Then M = M ∈T M is a subﬁeld of E. and so is algebraic over F (apply 1. Since α / is algebraic over M . In the uncountable case. and we can deﬁne a homomorphism ϕ : M → Ω by requiring that ϕ (x) = ϕM (x) if x ∈ M . and let E = Ei . . . . there will exist uncountably many isomorphisms from one algebraic closure to a second. Suppose ﬁrst that E is countably generated over F .5 (*). . i. f (X) splits in ϕ(E)[X]. . . there exists an F -isomorphism Ω → Ω . fn } / the above relation becomes 0 = 1. and then g has a root in Ei+1 . xf .26 (Non)uniqueness of algebraic closures T HEOREM 6. ϕ) in S. 26 . as we just observed. . Continue in this way to obtain a sequence of ﬁelds F = E0 ⊂ E1 ⊂ E2 ⊂ · · · . Continuing in this way. ϕM ) ≤ (N. and. 2 The above proof is a typical application of Zorn’s lemma: once we know how to do something in a ﬁnite (or countable) situation. .. P ROOF. . Even for a ﬁnite ﬁeld F . .. Then we can extend the inclusion map F → Ω to F [α1 ] (map α1 to any root of its minimal polynomial in Ω). f ∈ {f1 . Ω of F . but. Then X − α is a factor of f (X) in Ω[X]. .]. Let T be a totally ordered subset of S. Because of unique factorization. . thanks to Zorn’s lemma. and so on (see 2. the algebraic closure of F in E is an algebraic closure of F (by 1. Write (M. because the coefﬁcients of any nonconstant polynomial g ∈ E[X] lie in Ei for some i. ϕ ) is an upper bound for T in S. .45). ..]/M . Let E1 = F [.6. Similarly. Hence M = E. we deduce that 1 does not lie in I.] → F sending 6 ALGEBRAIC CLOSURES xfi → αi xf → 0. we use Zorn’s lemma. α ∈ M . From this contradiction. this implies that α ∈ ϕ(E). then. Let S be the set of pairs (M.2) to extend ϕ to M [α]. . we can apply (2. none of which is to be preferred over any other.4 applied to F [. α2 ]. All one can say is that. The pair (M . and let E be an algebraic extension of F . note that E1 is generated by algebraic elements over F . . In fact. . Repeat the process starting with E1 instead of F to obtain a ﬁeld E2 .e. and the proof of the ﬁrst statement is complete. Let Ω be an algebraic closure of F . and so Proposition 6. then to F [α1 . Let α ∈ Ω. ..]/I shows that I is contained in a maximal ideal M of F [. .68 F [. . . . if E is also an algebraic closure of F . . If E is algebraically closed. Then E is algebraically closed. then every polynomial f ∈ F [X] splits in E[X] and hence in ϕ(E)[X].31b). R EMARK 6. . xf . Thus it is (uncountably) sloppy to say that the algebraic closure of F is unique. . we ﬁnd that every element of every Ei is algebraic over F . . . . xf . ϕN ) if M ⊂ N and ϕN |M = ϕM . To see this. . Hence Zorn’s lemma gives us a maximal element (M. Zorn’s lemma allows us to do it in general. contradicting the maximality of M . .2). Therefore. E = F [α1 . then every such homomorphism is an isomorphism. Then there exists an element α ∈ E. E is algebraic over F .

then every such homomorphism is an isomorphism. because (*) implies that any ﬁnite set of elements of E(E) is contained in a common E ∈ E. there exists an E ∈ E such that E1 . if E is also a separable closure of F . (b) A ﬁeld Ω is said to be a separable closure of a subﬁeld F if it is separable and algebraic over F and it is separably closed. and let Ω be a separable algebraic closure of F . (a) Every ﬁeld has a separable closure. E2 ∈ E. deﬁne Ω to be the separable closure of F in an algebraic closure. (b) Let E be a separable algebraic extension of F . As the composite of any two such subﬁelds is again ﬁnite and separable over F (cf. because a separable extension of a separable extension is separable. and therefore their product. We call L the separable closure of F in Ω — clearly. we see that the union L of all such E is a subﬁeld of Ω.. it is separable over F and every element of Ω separable over F lies in L. We apply this remark to the set of subﬁelds E of Ω that are ﬁnite and separable over F . Replace “polynomial” with “separable polynomial” in the proofs of the corresponding theorems for algebraic closures. and. 2 .8 (*). sum.14). (a) A ﬁeld Ω is said to be separably closed if every nonconstant separable polynomial in Ω[X] splits in Ω. E2 ⊂ E. D EFINITION 6. P ROOF. and apply the preceding theorems. There exists an F -homomorphism E → Ω.7. T HEOREM 6. Alternatively. and let E be a set of intermediate ﬁelds F ⊂ E ⊂ Ω with the following property: (*) for any E1 .Separable closures 69 Separable closures Let Ω be a ﬁeld containing F . Ω is purely inseparable over L. Moreover. etc. 3. also lie in E(E). Then E(E) = E∈E E is a subﬁeld of Ω (and we call E∈E E a directed union).

there exists an N ∈ N such that N ⊂ gN g −1 . this shows that H is closed if it is open. gN2 ⊂ U . Similarly aR : g → ga and g → g −1 are both homeomorphisms. and let g ∈ U1 ∩ U2 . there exists an N ∈ N with gN ⊂ U .2. there exists an N ∈ N such that e ∈ N ⊂ N1 ∩ N2 .1. Deﬁne U to be the collection of subsets U of G such that. on applying (a) we obtain an N ∈ N such that gN ⊂ U1 ∩ U2 . .g) (g. for any subgroup H of G. N2 ∈ N .d). and S −1 = {s−1 | s ∈ S}. by deﬁnition there exist N1 . N2 ∈ N with gN1 . which 27 g→ag g→(a. Conversely. (b) for all N ∈ N .h)→gh For subsets S and S of G. h) → gh : G × G → G. − −→ −−− In fact. (d) for all N ∈ N and all g ∈ G. for every g ∈ U . the coset aH of H is open or closed if H is open or closed. and let N be a neighbourhood base for the identity element e of G. there exists an N ∈ N such that N ⊂ N −1 . there exists an N ∈ N such that N N ⊂ N . If N is a neighbourhood base at e in a topological group G. h) → gh. Note that (a) implies that e lies in all the N in N . Recall that a neighbourhood base for a point x of a topological space X is a set of neighbourhoods N such that every open subset U of X containing x contains an N from N. it is a homeomorphism with inverse (a−1 )L . (c). if G is a group and N is a nonempty set of subsets of G satisfying (a. and h → ghg −1 respectively. and (d) are consequences of the continuity of (g. Then27 (a) for all N1 . (e) for all g ∈ G. Topological groups D EFINITION 7. Let G be a topological group. As the complement of H in G is a union of such cosets. g → g −1 . Then aL : G − − G is continuous −→ because it is the composite of G − − − G × G − − − → G. s ∈ S }. let N be a nonempty collection of subsets of a group G satisfying the conditions (a)–(d). g → g −1 : G → G are both continuous. {gN | N ∈ N } is a neighbourhood base for g.b. then (b). P ROOF. the empty set and G are in U. Clearly. In particular. Moreover. we set SS = {ss | s ∈ S.c. Let a be an element of a topological group G. (c) for all N ∈ N . Let U1 .70 7 INFINITE GALOIS EXTENSIONS 7 Inﬁnite Galois extensions In this section. we make free use of Zorn’s lemma. and unions of sets in U are in U. then there is a (unique) topology on G for which (e) holds. P ROPOSITION 7. (a) is a consequence of the deﬁnitions and (e) of the fact that gL is a homeomorphism. and it is open if it is closed and of ﬁnite index. Conversely. U2 ∈ U. A set G together with a group structure and a topology is a topological group if the maps (g.

Now N −1 g ⊂ U −1 . this shows that S = σ∈G σS is ﬁnite. the unique topology for which (e) holds). Therefore. 2 P ROPOSITION 7. given an open U ⊂ G and a pair (g1 . g2 ) in G × G. P ROOF. Since σS = τ S if σ|F (S) = τ |F (S). It satisﬁes (b) and (c) because each set G(S) is a group. Let S be a ﬁnite subset of Ω. N2 ∈ N such that g1 N1 g2 N2 ⊂ U . P ROPOSITION 7. Finally. g ) → gg is continuous. Similarly. then it is Galois over any intermediate ﬁeld M .2).5.d) of (7. g2 ) such that g1 g2 ∈ U . i.5 shows that every F -homomorphism E → Ω extends to an F -homomorphism α : Ω → Ω. Given an open U ⊂ G and a g ∈ G such that g −1 ∈ U . For this topology. if every irreducible polynomial f ∈ F [X] having a root in Ω has deg f distinct roots in Ω. The same Zorn’s lemma argument as in the proof of Theorem 6. and let f be its minimum polynomial over F . We next used (b) and (d) to show that (g. We show that the collection of sets G(S) satisﬁes (a. Then F (S) is a ﬁnite extension of F . Let Ω be a Galois extension of F and let E be a subﬁeld of Ω containing F . Hence a ∈ α(Ω). P ROOF. Let a ∈ Ω. we deﬁne an algebraic extension Ω of F to be Galois over F if it is normal and separable. It also proves the second statement. Clearly. It satisﬁes (a) because G(S1 ) ∩ G(S2 ) = G(S1 ∪ S2 ). and so therefore does α(Ω). Note that the sets g1 N1 × g2 N2 form a neighbourhood base for (g1 . 2 The Krull topology on the Galois group Recall (3. it also must split into distinct degree-one factors in Ω[X]. Ω is Galois over F if and only if it is a union of ﬁnite Galois extensions. we have to ﬁnd an N ∈ N such that gN ⊂ U −1 . Now σS = S for all σ ∈ G. For any ﬁnite subset S of Ω. which proves (d).3.9) that a ﬁnite extension Ω of F is Galois over F if it is normal and separable. Then Ω contains exactly deg(f ) roots of f . Then every F -homomorphism E → Ω extends to an F -isomorphism Ω → Ω. and it follows that G(S) is normal in G. The minimum polynomial g(X) of a over F splits into distinct degree-one factors in Ω[X]. By deﬁnition.The Krull topology on the Galois group 71 shows that U1 ∩ U2 ∈ U. there exists an N ∈ N such that g1 g2 N ⊂ U . Therefore. It follows that the elements of U are the open sets of a topology on G (and. then g1 g2 N N ⊂ U . But g1 g2 N N = −1 g1 (g2 N g2 )g2 N . There is a unique structure of a topological group on G for which the sets G(S) form an open neighbourhood base of 1. Let f (X) be an irreducible polynomial in M [X] having a root a in Ω. there exists an N ∈ N such that g −1 N ⊂ U . let G(S) = {σ ∈ G | σs = s for all s ∈ S}. As f divides g (in M [X]). and so there are only ﬁnitely many F -homomorphisms F (S) → Ω. P ROOF. in fact.e. 2 . 2 Let Ω be a Galois extension of F . If Ω is Galois over F .c. which shows that α is surjective. and we use (c) to obtain an N ∈ N such that N g ⊂ U −1 . and it remains to apply (d) to obtain an N1 ∈ N such that N1 ⊂ −1 g2 N g2 . σG(S)σ −1 = G(S) ⊂ G(S).4.. and let G = Aut(Ω/F ). the sets G(S) with S G-stable form a neighbourhood base of 1 consisting of open normal subgroups. and (d) to obtain an N ∈ N such that gN ⊂ g(g −1 N g) ⊂ U −1 .b. As U is open. Apply (b) to obtain an N such that N N ⊂ N . we use (c) and (d) to show that g → g −1 is continuous. P ROPOSITION 7. we have to ﬁnd N1 .

the argument in the last paragraph shows that the map G/G(S) G→ S ﬁnite stable under G is injective. P ROOF. ΩGal(Ω/F ) = F . As we noted above. For any intermediate ﬁeld E ﬁnite and Galois over F .4). By homogeneity. For any Galois extension Ω/F . Therefore. P ROOF. We ﬁrst show that G is Hausdorff. 28 .6.72 7 INFINITE GALOIS EXTENSIONS The topology on Aut(Ω/F ) deﬁned in the proposition is called the Krull topology.e. then G(S) is a normal subgroup of G. the same is true for any element of Gal(E/F ).7. Gal(E/F ) = G(S).8. 2 P ROPOSITION 7. For any ﬁnite set S of generators of E over F .. if S is a ﬁnite set stable under G. which shows that the map is surjective.6. S2 ) be the set of families (σS G(S))S in G/G(S) such that σS1 = σS2 |S1 .28 P ROOF. for each ﬁnite set S stable under G. Hence they are disjoint open subsets of G containing σ and τ respectively. Since every ﬁnite set is contained in a stable ﬁnite set. 2 P ROPOSITION 7. If σ = τ . By homogeneity. let E(S1 . A topological space is totally disconnected if its connected components are the one-point sets. All Galois groups are compact and totally disconnected. When we endow G/G(S) with the product topology. the map σ → σ|E : Gal(Ω/F ) → Gal(E/F ) is a continuous surjection (discrete topology on Gal(E/F )). i. This is closed in G/G(S) because it is the set on which two continuous maps G/G(S) → G/G(S1 ) agree. it is the Krull topology. P ROPOSITION 7. this shows that the connected component of G containing 1G is just {1G }. i. the induced topology on G is that for which the G(S) form an open neighbourhood base of e. G/G(S) is compact. and so it moves some element of Ω. According to the Tychonoff theorem.. then σ −1 τ = 1G . Let Ω be Galois over F . Finally. Let G = Gal(Ω/F ). S1 ⊂S2 E(S1 . and call it the Galois group of Ω/F . We write Gal(Ω/F ) for Aut(Ω/F ) endowed with the Krull topology. 2 Following Bourbaki. We next show that G is compact. Since G(S) = {1G }. there exists an a ∈ Ω such that σ(a) = τ (a). G(S) is a subgroup that is open and hence closed. and regard it as an F -homomorphism E → Ω. Every element of Ω F lies in a ﬁnite Galois extension of F . we require compact spaces to be Hausdorff. Then σ extends to an F -isomorphism Ω → Ω (see 7. S2 ) is closed. For each S1 ⊂ S2 . For any S containing a. a similar statement is true for every element of G.e. and so it remains to show that G is closed in the product. Let σ ∈ Gal(E/F ). and so this follows from the surjectivity in Proposition 7. σG(S) and τ G(S) are disjoint because their elements act differently on a. and it has ﬁnite index because it is the kernel of G → Sym(S). which shows that the inverse image of 1Gal(E/F ) is open in G. but it also equals the image of G.

(b) For every subgroup H of G. On the other hand. (a) The ﬁrst assertion was proved in (7. For the ﬁrst statement. Let H be a closed subgroup of G. it certainly contains the closure H of H. The ﬁnal statement follows from (7.The fundamental theorem of inﬁnite Galois theory 73 A SIDE 7. Gal(E/σM ) = σ Gal(E/M )σ −1 . Gal(Ω/ΩH ) is the closure of H. i.10).e. and so it also is closed.10). M → Gal(Ω/M ) are inverse bijections between the set of closed subgroups of G and the set of intermediate ﬁelds between Ω and F : {closed subgroups of G} ↔ {intermediate ﬁelds F ⊂ M ⊂ Ω}.10). Cohomologie des groupes compact totalement discontinus (d’apr` s J. Let Ω be Galois over F . with Galois group G. But Gal(Ω/M ) = S⊂M G(S). let σ ∈ G H. (b) Since Gal(Ω/ΩH ) contains H and is closed. and hence it is closed. (a) The ﬁeld Ω is Galois over every subﬁeld M containing F . But Gal(Ω/ΩHi ) = Hi (see 7. (b) a closed subgroup H of G is open if and only if ΩH has ﬁnite degree over F .3). in which case Gal(Ω/ΩH ) G/H. Then σG(S) ∩ H = ∅ for some ﬁnite subset S of Ω which we may assume to be stable under G. / and so there exists an α ∈ F (S) that is ﬁxed by H but moved by σ. / 2 T HEOREM 7. Then Ω is Galois over ΩH and Gal(Ω/ΩH ) = H (see 7. Let M be an intermediate ﬁeld. . Tate). e S´ minaire Bourbaki 1959/60. and ΩGal(Ω/M ) = M . For each ﬁnite subset S ⊂ M .. Moreover. The maps H → ΩH .9. e The fundamental theorem of inﬁnite Galois theory P ROPOSITION 7. P ROOF. This shows that σ ∈ Gal(Ω/ΩH ). There is a converse to the proposition: every compact totally disconnected group arises as the Galois group of some Galois extension of ﬁelds of characteristic zero (Douady. (a) the correspondence is inclusion-reversing: H1 ⊃ H2 ⇐⇒ ΩH1 ⊃ ΩH2 .11. (a) We have the obvious implications: H1 ⊃ H2 =⇒ ΩH1 ⊂ ΩH2 =⇒ Gal(Ω/ΩH1 ) ⊃ Gal(Ω/ΩH2 ). in which case (G : H) = [ΩH : F ]. Let Ω be Galois over F with Galois group G. ΩσHσ = σ(E H ). as required. P ROOF. no. A. Then Gal(Ω/M ) is a closed subgroup of G and ΩGal(Ω/M ) = M (see 7. Now σG(S) ∩ H = ∅ implies σ ∈ HG(S). Gal(Ω/M ) is closed in G. (d) a closed subgroup H of G is normal if and only if ΩH is Galois over F .. moreover. 189). −1 (c) σHσ −1 ↔ σM . we have to show that H → ΩH and M → Gal(Ω/M ) are inverse maps.8). G(S) is an open subgroup of G.10.

An open subgroup of Gal(EL/L) is closed (hence compact) of ﬁnite index.12. We have already observed that an extension is Galois if and only if it is a union of ﬁnite Galois extensions. Let H be such a subgroup.10). of ﬁnite Galois extensions of G. GT 4.4) from which the statement follows. τ α = α ⇐⇒ στ σ −1 (σα) = σα. (d) Let H ↔ M . conversely an open subgroup of G is always of ﬁnite index. then there exists an F -homomorphism τ : EL → Ω such that τ |E = ρ and τ |L = σ.14. Therefore. For any ﬁnite set S of elements of E. i. Because G is compact. We next prove it is continuous. which is the composite of the ﬁelds σM . then EL/L and E/E ∩ L are Galois. As in the ﬁnite case (3. If E/F is Galois. a closed subgroup of ﬁnite index in a topological group is always open. Finally. (b) Let M ↔ H. therefore its image in Gal(E/E ∩ L) is compact (hence closed) of ﬁnite index. (c) For τ ∈ G and α ∈ Ω. Let Mi be the smallest ﬁeld containing all the Mi . But M is stable under the action of G if and only it is a union of ﬁnite extensions of F stable under G. is the σ σHσ smallest normal extension of F containing M . Let Ω be an algebraically closed ﬁeld containing F .e.13.18). The largest (closed) normal subgroup contained in H is N = −1 (cf. and so ΩN .74 7 INFINITE GALOIS EXTENSIONS (b) As we noted earlier. If ρ : E → Ω and σ : L → Ω are F -homomorphisms such that ρ|E ∩ L = σ|E ∩ L. then because i∈I Hi is the largest (closed) subgroup contained in all the Hi .. The proof that the map is an isomorphism of groups (neglecting the topology) is the same as in the ﬁnite case (3. Let E and L be ﬁeld extensions of F contained in some common ﬁeld. Let G1 = Gal(EL/L) and let G2 = Gal(E/E ∩ L). Ω) (apply 7. we prove that it is open. Gal(Ω/σM ) = σ Gal(Ω/M )σ −1 .17). It follows from (c) that H is normal if and only if M is stable under the action of G. and hence open. we can deduce the following statements. The map σ → σ|ΩH deﬁnes a bijection G/H → HomF (ΩH . and let Hi ↔ Mi . and so σ Gal(Ω/M )σ −1 ↔ σM. the inverse image of G2 (S) in G1 is G1 (S). P ROPOSITION 7. and let E and L be as in the proposition. (a) Let (Mi )i∈I be a (possibly inﬁnite) family of intermediate ﬁelds. Gal(E/ Mi ) = i∈I Hi . . E = EL = L E∩L F P ROOF.2 R EMARK 7. 2 C OROLLARY 7. and the map σ → σ|E : Gal(EL/L) → Gal(E/E ∩ L) is an isomorphism of topological groups.

Similarly. m.) — cf.17. to prove the one-to-one correspondence in the fundamental theorem. The map α → σ α : Z → Gal(Ω/Fp ) is an isomorphism. we can complete Z for this topology. From a Galois extension Ω of F we get a simple Galois correspondence by taking P to be the set of subgroups of Gal(Ω/F ) and Q to be the set of subsets of Ω. According to the proposition. σ extends to an F -homomorphism s : EL → Ω. there exists a unique e ∈ Gal(EL/L) such that e|E = ε.. Then G = Gal(Ω/Fp ) contains a canonical Frobenius element. Gal(Qab /Q) ≈ Z× . and the map sending m ∈ Z to the constant sequence m. gf g = g. G is the closure of σ .e.10). To number theorists it will seem quite natural — the Chinese remainder theorem implies that it is isomorphic to p prime Zp where Zp is the ring of p-adic integers. σ = (a → ap ). i. . because f g(f p) ≥ f p and gf (p) ≥ p implies f (gf p) ≤ f (p) for all p ∈ P . The Cauchy sequences form a commutative group. if for all n ≥ 1. we can write s|E = ρ ◦ ε for some ε ∈ Gal(E/E ∩ L). The group Z is uncountable.e.16. E XAMPLE 7. ai ∈ Z. . This is accomplished by (7. A Cauchy sequence in Z is a sequence (ai )i≥1 . and it certainly has Sn as a quotient group for every n (and every sporadic simple group. The restriction of σ to Fpn has order n. and by setting f (H) = ΩH and g(S) = G(S). (This is abelian class ﬁeld theory — see my notes Class Field Theory. Thus. Endow Z with the topology for which the groups nZ. it sufﬁces to identify the closed subgroups as exactly those in the image of g and the intermediate ﬁelds as exactly those in the image of f . there exists an N such that ai ≡ 0 mod n for all i > N . and intractable. j > N . and the trivial Cauchy sequences form a subgroup. Deﬁne τ = s ◦ e−1 . As for any topological group. it is a little weird—its connected components are one-point sets. Let Ω be an algebraic closure of a ﬁnite ﬁeld Fp . for each n. and every. m. n ≥ 1.. then Gal(Ω/Q) is one of the most basic. satisfying the following condition: for all n ≥ 1. i. It has a ring structure.. Therefore (σ|Fpn )ai is independent of i provided it is sufﬁciently large.15. form a fundamental system of neighbourhoods of 0. (5. Then f gf = f .34). and we can deﬁne σ α ∈ Gal(Ω/Fp ) to be such that. According to (7.The fundamental theorem of inﬁnite Galois theory 75 P ROOF.2 E XAMPLE 7. and it follows that f and g deﬁne a one-to-one correspondence between the sets g(Q) and f (P ). σ α |Fpn = (σ|Fpn )ai for all i sufﬁciently large (depending on n).) A SIDE 7. Call a Cauchy sequence in Z trivial if ai → 0 as i → ∞. . identiﬁes Z with a subgroup of Z. It is expected that every ﬁnite group occurs as a quotient of it. As s|E ∩L = ρ|E ∩ L. Thus two integers are close if their difference is divisible by a large integer. For example. .4). We do understand Gal(F ab /F ) where F ⊂ C is a ﬁnite extension of Q and F ab is the union of all ﬁnite abelian extensions of F contained in C. Let α ∈ Z be represented by the Cauchy sequence (ai ). objects in mathematics.. Let Ω be the algebraic closure of Q in C. A simple Galois correspondence is a system consisting of two partially ordered sets P and Q and order reversing maps f : P → Q and g : Q → P such that gf (p) ≥ p for all p ∈ P and f g(q) ≥ q for all q ∈ Q. there exists an N such that ai ≡ aj mod n for i. and it is generated by it as a topological group. We deﬁne Z to be the quotient of the ﬁrst group by the second. To most analysts.

qi ) be a second family such that pj ◦ qj = qi . and the pair (I.. Let H be (H.19. Let (H. is uniquely determined by this condition up to a unique isomorphism. We denote it lim(Ai . Let (I. pi ) = lim(Gi . or the category of topological groups and continuous homomorphisms).76 7 INFINITE GALOIS EXTENSIONS Galois groups as inverse limits D EFINITION 7.. the category of groups and homomorphisms. ≤) is a family (Ai )i∈I of objects of C together with a family (pj : Aj → Ai )i≤j of morphisms such that pi = idAi and pj ◦ pk = pk i j i i i all i ≤ j ≤ k.. A r e rr qj d pj d e rrd e j r e qi Aj pi e e e j e c © pi r Ai Clearly. The projection maps pi are continuous. Let i G = {(gi ) ∈ Gi | pj (gj ) = gi all i ≤ j}. ≤) is called a directed set.. pj ). j ∈ I there exists a k ∈ I such that i.21.. D EFINITION 7. i ← − E XAMPLE 7. E B . The i homomorphism h → (qi (h)) : H → Gi .18. Hence (G.20. or just lim Ai . the inverse limit (if it exists). Let (Gi . and let C be a category (for example. pj ). pj : Gj → Gi ) be an inverse system of groups. i ← − ← − E XAMPLE 7. if for all i. The image of the homomorphism i h → (qi (h)) : H → Gi is contained in G. Then pj ◦ pj = pi is just the equation pj (gj ) = i i gi . i and G becomes a topological subgroup with the subspace topology. Gi becomes a topological group G = {(gi ) ∈ Gi | pj (gj ) = gi all i ≤ j}. Let (Gi . ≤) be a directed set. there exists a unique morphism r : B → A i such that pj ◦ r = qj for j. i and let pi : G → Gi be the projection map. (a) An inverse system in C indexed by (I.. pj : Gj → Gi ) be an inverse system of topological groups and i continuous homomorphisms. qi ) be a second family such that pj ◦ qj = qi . When endowed with the product topology. (b) An object A of C together with a family (pj : A → Aj )j∈I of morphisms satisfying pj ◦ pj = pi all i ≤ j is said to be an inverse limit of the system in (a) if it has the i following universal property: for any other object B and family (qj : B → Aj ) of morphisms such pj ◦ qj = qi all i ≤ j. and this is the unique homomorphism H → G carrying qi to pi ... j ≤ k.. A partial ordering ≤ on a set I is said to be directed.

we get an inverse system of ﬁnite groups (Gal(E/F ). pi ) = lim(Gi . the connected component of G containing e is just {e}. H. E For each E. and its kernel Ui is an open subgroup of ﬁnite index in G (hence also closed).23. pj ). i. it is Exercise 3 of §7 of Chapter 3 of Bourbaki’s General Topology. 2 P ROPOSITION 7. By homogeneity. is a basis. . As Ui = {e}. these maps deﬁne a homomorphism Gal(Ω/F ) → lim Gal(E/F ).25.Nonopen subgroups of ﬁnite index 77 is continuous because its composites with projection maps are continuous (universal property of the product). More precisely. The composite of two ﬁnite Galois extensions of in Ω is again a ﬁnite Galois extension. 0 If (xi ) ∈ G. The group Gal(Qal /Q) has nonopen normal subgroups of index 2n for all n > 1. and so G is also compact. and so the ﬁnite Galois subextensions of Ω form a directed set I. ← − This map is an isomorphism of topological groups.. then / i G ∩ {(gj ) | gj0 = xj0 . this is one of a number of related unsolved problems. the same is true for every point of G: the connected components of G are the one-point sets — G is totally disconnected. This is a restatement of what we showed in the proof of (7. this shows that G is closed in Gi . We have shown that a proﬁnite group is compact and totally disconnected. i ← − E XAMPLE 7. it is not known. whether every subgroup of ﬁnite index in G is open. Choose a basis. Zorn’s lemma shows that V contains maximal linearly independent subsets. As the second set is an open neighbourhood of (xi ).29 E XAMPLE 7. A brief guide to algebraic number theory. because of the universal property of inverse limits. all of which appear to be very difﬁcult. gi0 = xi0 } = ∅. F. Gi is compact. P.” Swinnerton-Dyer. A topological group G arising as an inverse limit of such a system is said to be proﬁnite.. and it is an exercise to prove the converse.30 L EMMA 7.22. and this shows that (G. For each E in I we have a ﬁnite group Gal(E/F ). 2001. even when G = Gal(Q/Q). and take Vn to be the subspace spanned by the set obtained by omitting n elements from the basis. Cambridge. . and then the usual argument shows that such a subset spans V . Nonopen subgroups of ﬁnite index We apply Zorn’s lemma to construct a nonopen subgroup of ﬁnite index in Gal(Qal /Q). P ROOF. p133. there exists a subspace Vn of V such that V /Vn has dimension n. Let V be an inﬁnite dimensional vector space. there is a restriction homomorphism pE : Gal(Ω/F ) → Gal(E/F ) and. 30 29 .e. For all n ≥ 1. In this E way.7). and for each E ⊂ E we have a restriction homomorphism pE : Gal(E /F ) → Gal(E/F ). pj : Gj → Gi ) be an inverse system of ﬁnite groups. Contrast: “. Therefore H → G is continuous. and regard i it as an inverse system of topological groups by giving each Gi the discrete topology. By Tychonoff’s theorem.24. say pj0 (xj0 ) = xi0 . The map pi : G → Gi is continuous. Let Ω be a Galois extension of F . Let (Gi . pE ) indexed by I.

). . a2 . . .27 and Exercise 2-1). p]/Q) is a product of copies of Z/2Z indexed by the set {primes ≤ p} ∪ {∞} (see 5. .78 7 INFINITE GALOIS EXTENSIONS √ √ √ P ROOF.). which contradicts the fact that H is dense. . . 2. 2. 0. . p. . a3 . and its inverse image in Gal(Qal /Q) is the desired subgroup. Gn is not open. . . . p]/Q) ← − is a direct product of copies of Z/2Z indexed by the primes l of Q (including l = ∞) endowed with the product topology. 2. . . . let (al ) ∈ G. . 0. . √ √ √ Gal(Q[ −1. (a∞ . then it is closed. it is a direct sum of copies of Z/2Z indexed by the primes of Q). 0. . If Gn is open in G. . For each prime p. then the sequence (a∞ . a2 . 2 31 Better. Let E be the subﬁeld Q[ −1. . . Therefore. . .] of C. We can regard G/H as vector space over F2 and apply the lemma to obtain subgroups Gn of index 2n in G containing H. 0. converges to (al ). and it is dense in G because31 clearly every open subset of G contains an element of H.). and let H = {(al ) ∈ G | al = 0 for all but ﬁnitely many l}. (a∞ . Let G = Gal(E/Q). . This is a subgroup of G (in fact. 0. and so √ √ √ Gal(E/Q) = lim Gal(Q[ −1. 0. . . . . .

. .. βn ∈ F [A]. it is algebraically dependent over F . In fact. . they are algebraically dependent over F . .. Thus... . Y ) = 0 but f (α1 . .. .. (a) A single element α is algebraically independent over F if and only if it is transcendental over F... . (b) The complex numbers π and e are almost certainly algebraically independent over Q. . . (b) there exist β1 .33).... .. αn of Ω give rise to an F -homomorphism f → f (α1 .... αn ) : F [X1 . γ) = 0. αn ) : F [X1 .... .. For example. .in α11 . (c) there exist β0 . such that β0 γ n + β1 γ n−1 + · · · + βn = 0.1. Algebraic independence Elements α1 . and otherwise.. β1 ... The following conditions are equivalent: (a) γ is algebraic over F (A)..... F (α1 . . the αi are algebraically dependent over F if there exists a nonzero polynomial f (X1 . . . Xn ] such that f (α1 . but this has not been proved. Note the similarity with linear independence. αm . Y ) ∈ F [X1 . .3... and hence an isomorphism. . Y ] and α1 .. . Xn ] → F [α1 . . . R EMARK 8.79 8 Transcendental extensions In this section we consider ﬁelds Ω ⊃ F with Ω much bigger than F . αm ∈ A such that f (α1 . αn ] is an injection. Let γ ∈ Ω and let A ⊂ Ω. .... . This isomorphism then extends to the ﬁelds of fractions. L EMMA 8.. . and they are algebraically independent if ai1 . Xn ) → f (α1 .. The polynomial f (X) = X n − α1 X n−1 + · · · + (−1)n αn has Galois group Sn over F (α1 .. . . βn ∈ F (A) such that γ n + β1 γ n−1 + · · · + βn = 0.in ∈ F. .2.. ... . . ... .. αm . i in ai1 . (d) there exists an f (X1 . . An inﬁnite set A is algebraically independent over F if every ﬁnite subset of A is algebraically independent. αn ) = 0. ... If the kernel of this homomorphism is zero. then the deﬁnition becomes that of linear independence. then f (X1 . . . .. . αn ) In this case.... if f is required to be homogeneous of degree 1. not all 0. Xn ) ∈ F [X1 . αn ) (5... we could have C ⊃ Q. . Xn ] → Ω. Xm ... . αn are algebraically independent over F . in . otherwise. . Xi → αi : F (X1 . αn ) is called a pure transcendental extension of F .. If α1 . Xn ) → F (α1 .. .. E XAMPLE 8.αn = 0 =⇒ ai1 .in = 0 all i1 . .. Xm . .. . . then the αi are said to be algebraically independent over F .. . ..

. .. f (X1 . Then (c) holds with βi = fi (α1 .. . β}. αm . . βi = fi (α1 . αm−1 }. .. 2 D EFINITION 8. . βn } be two subsets of Ω. P ROOF. αm ) with fi ∈ F [X1 ... .... . ... if β is algebraically dependent on {α1 . . Write f (X1 . . Write f as a polynomial in Xm . Y ) = fi (X1 ... Y )Xm .. .. . . αm−1 .. (d) =⇒ (c). it is said to be algebraically dependent on A (over F ).. . . . Let {α1 . . . Xm . . αm . . The βi in (c) can be expressed as polynomials in a ﬁnite number of elements α1 .. Xm ]. Y ) = 0. .. Y ) with coefﬁcients in F such that f (α1 .. . . Because β is algebraically dependent on {α1 . . αm } and B = {β1 . . . because f (α1 . there exists a polynomial f (X1 . . Let A = {α1 . αm−1 . ... . and observe that. Xm . αm of A. .. . . Y ) as a polynomial in Y with coefﬁcients in the ring F [X1 .. Xm ]. . . L EMMA 8.. αm } but not on {α1 . Y ) = 0. . Xm−1 . It is useful to keep the following correspondences in mind: Linear algebra linearly independent A ⊂ span(B) basis dimension Transcendence algebraically independent A algebraically dependent on B transcendence basis transcendence degree Transcendence bases T HEOREM 8.. β) = 0.. . αm . . . Then m ≤ n.80 8 TRANSCENDENTAL EXTENSIONS P ROOF. Y ) = i n−i ai (X1 . Xm . Y ). . We ﬁrst prove two lemmas.. . . αm ). . .. . Then (d) holds with f = fi (X1 . . . When γ satisﬁes the equivalent conditions of Lemma 8. (a) =⇒ (b) =⇒ (c) =⇒ (a) are obvious.. then αm is algebraically dependent on {α1 . . Xm . αm } be a subset of Ω. αm }. . .. . ..5 (F UNDAMENTAL RESULT ). . (b) A is algebraically dependent on B (over F ).4. . Assume (a) A is algebraically independent (over F ). . .. .. Xm )Y n−i .3. The theory in the remainder of this section is logically very similar to a part of linear algebra. at least one of the polynomials ai (α1 . Xm )Y n−i . say. A set B is algebraically dependent on A if each element of B is algebraically dependent on A. f (X1 .6 (T HE EXCHANGE PROPERTY )... f (α1 . (c) =⇒ (d)..

P ROOF... Because β is not algebraically dependent on {α1 .8.5. an .. β) = 0. βk+1 . Suppose that k < m. β) = 0. but that A ∪ {β} is algebraically dependent. Therefore.. If C is algebraically dependent on B. and B is algebraically dependent on A. . P ROOF... every transcendence basis is ﬁnite. 81 ai0 (α1 . .. αm−1 . .7 (T RANSITIVITY OF ALGEBRAIC DEPENDENCE ). then it is a transcendence basis for Ω over F . and A is minimal among subsets of Ω with this property. Then β is algebraic over F (A).. Let k be the number of elements that A and B have in common. If there is a ﬁnite subset A ⊂ Ω such that Ω is algebraic over F (A). If A is not algebraically independent. k + 1 ≤ j ≤ n. A transcendence basis for Ω over F is an algebraically independent set A such that Ω is algebraic over F (A)... Therefore B is algebraically dependent on B1 . . .... αk . L EMMA 8.... this shows that αm is algebraically dependent on {α1 . and m ≤ n. βj } but not {α1 . . In fact.. The exchange lemma then shows that βj is algebraically dependent on B1 = B ∪ {αk+1 } df {βj }. γ] has ﬁnite degree over F ). . 2 L EMMA 8. an are the coefﬁcients of the minimum polynomial of γ over E...10.Transcendence bases say ai0 . .7 that Ω is algebraic over F (A {α}).. αm−1 }. Eventually we’ll achieve k = m. and certainly m ≤ n.. αk . If k = m. 2 D EFINITION 8. If k + 1 < m.... β}. then there is an α ∈ S that is algebraically dependent on S {α}.. there will be a βj ..11. βk+1 . is not the zero polynomial. 2 L EMMA 8. then the ﬁeld F [a1 . any minimal subset A of A such that Ω is algebraic over F (A ) will be a transcendence basis. . αm ... Moreover. repeat the argument with A and B1 . . Since αk+1 is algebraically dependent on {α1 .. P ROOF. and they all have the same number of elements. .. . It follows from Lemma 8. . such that αk+1 is algebraically dependent on {α1 . then γ is algebraic over F (if a1 . βn } but not on {α1 . .44 shows that if γ is algebraic over a ﬁeld E which is algebraic over a ﬁeld F . βk+1 .. Apply this with E = F (A ∪ B) and F = F (A). . then C is algebraically dependent on A. . βj−1 }. .9. and write B = {α1 . αm−1 ..7).. and so A is algebraically dependent on B1 (by 8.5).. βn }... .2 T HEOREM 8... αk }. αm−1 . then A ⊂ B. βk+1 . The argument in the proof of Proposition 1. .. . f (α1 . β) = 0. αk . Xm . Since f (α1 . ... . If Ω is algebraic over F (A). 2 P ROOF ( OF T HEOREM 8. αk . then Ω has a ﬁnite transcendence basis over F . . Suppose that A is algebraically independent. The second statement follows from Theorem 8..

A∪{β} is algebraically dependent. β) = 0. it is algebraic over F (α1 .. .e. Therefore f = g0 Y m + g1 Y m−1 + · · · + gm . . g0 = 0.12 shows to be a transcendence basis for Ω over F . . . Let Ω be the ﬁeld of meromorphic functions on a compact complex manifold M . which Proposition 8.. Because A is algebraically independent... 2 It is possible to show that any two (possibly inﬁnite) transcendence bases for Ω over F have the same cardinality.15. . . . Xn ) has transcendence degree n. pn ). . The ﬁeld F (X1 .. for every β ∈ Ω A. Xn ) has transcendence degree n over F . Every algebraically independent subset of Ω is contained in a transcendence basis for Ω over F . αn )X m + g1 (α1 . Xn .. and let B = {A | A ∈ T }. and so the lemma shows that β is algebraic over F (A). If not. and so {p1 . We can partially order it by inclusion. As g0 = 0 and the αi are algebraically independent.. . . gi ∈ F [X1 . T HEOREM 8..2 Recall that (except in §7). m ≥ 1. . Xn ) is algebraic over F (p1 .... Xn ).. We have to prove that Ω is algebraic over F (A) if A is maximal among algebraically independent subsets. the pi ’s must themselves be a transcendence basis. P ROOF. . .. .12. .. Hence. Every maximal algebraically independent subset of Ω is a transcendence basis for Ω over F .. . E XAMPLE 8. transcendence bases exist. .. . which is a contradiction. Ω is a pure transcendental extension of C of transcendence degree 1.13 (*). . . p2 . . . .. Let p1 . . But the maximality implies that.. .. in this case. . pn } contains a transcendence basis for F (X1 .14. αn )X m−1 + · · · + gm (α1 . For example.. .. . (a) The only meromorphic functions on the Riemann sphere are the rational functions in z. αn ) ⊂ F (A). in particular. . pn be the elementary symmetric polynomials in X1 . . . . Xn ].... Let S be the set of algebraically independent subsets of Ω containing the given set. there exists a ﬁnite subset B of B that is not algebraically independent. . P ROOF. Y does occur in f . Y ] such that f (α1 . the pure transcendental extension F (X1 .. E XAMPLE 8.. . But such a subset will be contained in one of the sets in T . Y ) ∈ F [X1 . i. αn ∈ A.. The cardinality of a transcendence basis for Ω over F is called the transcendence degree of Ω over F . Because β is a root of f = g0 (α1 .. that B is algebraically independent. Let T be a totally ordered subset of S. some distinct α1 . . g0 (α1 . Xn . .82 8 TRANSCENDENTAL EXTENSIONS P ROOF. αn . . αn ). . 2 P ROPOSITION 8. The hypothesis is that there exists a nonzero polynomial f (X1 .. . Now Zorn’s lemma shows that there exists a maximal algebraically independent containing S.. . . . Because F (X1 .. I claim that B ∈ S. Xn . we use an asterisk to signal a result depending on Zorn’s lemma. . αn ) = 0.

P ROOF.L¨ roth’s theorem u 83 (b) If M is a Riemann surface. 1958). then E is a pure transcendental extension of F (Theorem of Zariski.ca/˜dave. p157.C). then the transcendence degree of Ω over C is 1. R EMARK 8.. there exists a bijection A → A .” R. then the transcendence degree is ≤ n. Then show that when Ω is uncountable. Any subﬁeld E of F (X) containing F but not equal to F is a pure transcendental extension of F . and Ω is a pure transcendental extension of C ⇐⇒ M is isomorphic to the Riemann sphere (c) If M has complex dimension n. Any two algebraically closed ﬁelds with the same transcendence degree over F are F -isomorphic. 92 (1970).5). . Without Zorn’s lemma.17. 1–56. Use this isomorphism to identify F (A) with F (A ). .uleth. and hence to an F -isomorphism of the ﬁelds of fractions F (A) → F (A ). Then the two ﬁelds in question are algebraic closures of the same ﬁeld.32 and it is known that the Zorn’s lemma is required to construct nonmeasurable functions. Choose transcendence bases A and A for the two ﬁelds. there are only two.morris/. http://people. then α deﬁnes an isomorphism Q(A) → Q(A) that can be extended to an automorphism of C. apply the proposition. Ann. IV 4. What are the automorphisms of C? There are only two continuous automorphisms (cf. Jacobson 1964. xd ). Exercise A-8 and solution). Any two algebraically closed ﬁelds with the same uncountable cardinality and the same characteristic are isomorphic. with equality holding if M is embeddable in some projective space. . because the noncontinuous automorphisms are nonmeasurable. Separating transcendence bases Let E ⊃ F be ﬁelds with E ﬁnitely generated over F . .20.. Y ) : E] < ∞ and F is algebraically closed of characteristic zero. The idea of the proof is as follows. . xd } of E is a separating transcendence basis for E/F if it is algebraically independent over F and E is a ﬁnite separable extension of F (x1 . Finally. Introduction to Arithmetic Groups. This fails when there is more than one variable — see Zariski’s example (footnote to Remark 5. which extends uniquely to an F -isomorphism F [A] → F [A ]. Appendix I.33 ¨ Luroth’s theorem ¨ T HEOREM 8. and choose any permutation α of A. Let F and F be the prime subﬁelds of Ω and Ω . 32 A fairly elementary theorem of G. of Math. If we assume Zorn’s lemma.19 (L UROTH ’ S THEOREM ). . A subset {x1 . Solovay. P ROPOSITION 8. 33 “We show that the existence of a non-Lebesgue measurable set cannot be proved in Zermelo-Frankel set theory (ZF) if use of the axiom of choice is disallowed.. .16. The best true statement is the following: if [F (X. By assumption. . . the cardinality of Ω is the same as the cardinality of a transcendence basis over F . Mackey says that measurable homomorphisms of Lie groups are continuous (see David Witte Morris. 2 R EMARK 8. then it is easy to construct many: choose any transcendence basis A for C over Q. 2 R EMARK 8. P ROOF. we can identify F with F . and hence are isomorphic (Theorem 6.5) and Swan’s example (Remark 5.34).18.

σ can be extended to an F -automorphism of Ω. xd+2 ) = F (x1 . . we may suppose that ∂f /∂Xd+1 = 0. P ROOF.13). . . . ad+1 ) → (a1 . . If F is perfect. Let E = F (x1 . there is an element y such that F (x1 .. . . xd+1 . xd ). let G(S) = {σ ∈ G | σs = s for all s ∈ S}. After repeating the process. . and so the statement becomes that of (8. . This has the following geometric interpretation: every irreducible algebraic variety of dimension d over a perfect ﬁeld F is birationally equivalent with a hypersurface H in Ad+1 for which the projection (a1 . If α is algebraic over F . After renumbering. Then F (x1 . Xn with coefﬁcients in F is a pth power in F [X1 . then it is part of a transcendence basis A for Ω over F (see 8. Transcendental Galois theory T HEOREM 8. . Xnn p = i i aip1 ···in X11 . and assume n > d + 1 where d is the transcendence degree of E over F . . Not all ∂f /∂Xi are zero. . we showed that E admits a separating transcendence basis with d + 1 elements where d is the transcendence degree. Xd+1 .10). . Then f (x1 . .22. .1). .23 (*). ad ) realizes F (H) as a ﬁnite separable extension of F (Ad ) (see my notes on Algebraic Geometry). xn ). . . . In fact. . . If α is transcendental over F . After renumbering. . . . Let α ∈ Ω F . . which implies that it is a pth power. If F has characteristic zero. . we may suppose that x1 . For any ﬁnite subset S of Ω. as in §7. . . then every ﬁnitely generated extension E of F admits a separating transcendence basis F . . Then σ extends to an F -automorphism of F (A).e. If α ∈ Ω is ﬁxed by all F -automorphisms of Ω. . ΩG = F . possibly several times. . 2 Let Ω ⊃ F be ﬁelds and let G = Aut(Ω/F ). . . . . . . y). xd ) and xd+1 is separable over F (x1 . for otherwise f would p p be a polynomial in X1 . Thus. . . 2 A SIDE 8. Xd+1 ) with coefﬁcients in F .84 8 TRANSCENDENTAL EXTENSIONS T HEOREM 8. .8). . As before. . which we again call the Krull topology. P ROOF. . then every transcendence basis is separating. by the primitive element theorem (5. then α ∈ F . . . . . we may assume F has characteristic p = 0. . . . .8) there exists an F -automorphism σ of the separable closure of F in Ω such that σ(α) = α. xd . . . Then. . Choose an automorphism σ of A such that σ(α) = α. Let Ω be a separably closed ﬁeld and let F be a perfect subﬁeld of Ω. i. xd are algebraically independent (8. . . p p Because F is perfect.21. the subgroups G(S) of G form a neighbourhood base for a unique topology on G. we will have E = F (z1 . . The same argument as in §7 shows that this topology is Hausdorff (but it is not necessarily compact). then by inﬁnite Galois theory (7. . . Thus E is generated by n − 1 elements (as a ﬁeld containing F ). xd+1 ) = 0 for some nonzero irreducible polynomial f (X1 . and so. zd ). zd+1 ) with zd+1 separable over F (z1 . every polynomial in X1 . .9). . Xn ]: i i ai1 ···in X11 p . . . xd+2 ) is algebraic over F (x1 . . . which extends to an F -automorphism of the separable closure Ω of F (A) (see 6. . . . Xnn 1 p . .

Transcendental Galois theory T HEOREM 8. 2 . and σ → σ|M1 maps H/ Aut(Ω/M1 ) isomorphically onto a dense subgroup of Aut(M1 /M ). 1971. If moreover H = Aut(Ω/M ). Princeton.24. (a) For every ﬁnite extension E of F in Ω. P ROOF. then G is locally compact. G = Aut(Ω/F ). Galois (d) Let H be a subgroup of G. M → Aut(Ω/M ) 85 (∗) are inverse bijections between the set of compact subgroups of G and the set of intermediate ﬁelds over which Ω is Galois (possibly inﬁnite): {compact subgroups of G} ↔ {ﬁelds M such that F ⊂ M ⊂ Ω}. and under (∗): {open compact subgroups of G} ↔ {ﬁelds M such that F 1:1 ﬁntely generated Galois ⊂ M ⊂ Ω}. ΩAut(Ω/E) = E.3 of Shimura. Goro. (b) The maps H → ΩH . (c) If there exists an M ﬁnitely generated over F such that Ω is Galois over M . Then the algebraic closure M1 of M is Galois over M . and let M = ΩH .. See 6. Introduction to the arithmetic theory of automorphic functions. Let Ω ⊃ F be ﬁelds such that ΩG = F . then Aut(Ω/M1 ) is a normal subgroup of H.

p prime. X 17 − 1.86 A REVIEW EXERCISES A Review exercises A-1. A-9. Determine whether G is abelian. Prove your answer. √ a primitive element for the ﬁeld Q[ 3. (a) Prove that the multiplicative group of all nonzero elements in a ﬁnite ﬁeld is cyclic.. A-6. Let f (X) be a polynomial in F [X] of degree n. Let F be a ﬁeld with 16 elements. √ √ A-14. and let m and n be positive integers. A-5. A-12. . Find the degree of a splitting ﬁeld of the polynomial (X 3 − 5)(X 3 − 7) over Q. Show that [E : F ] divides n!. and exhibit a generator for its multiplicative group. Determine the order of G. Prove as directly as you can that if ζ is a primitive pth root of 1. Let X be transcendental over a ﬁeld F . (a) (b) (c) Let G be the Galois group of the polynomial X 5 − 2 over Q. then the Galois group of Q[ζ] over Q is cyclic of order p − 1. X 15 − 1. 7] over Q. Let p be a prime number. and let E be a splitting ﬁeld of f . (a) Show that every ﬁeld homomorphism from R to R is bijective. Show that f (X) divides some nonzero polynomial g(X) with coefﬁcients in F . A-10. (b) If there is such a subﬁeld. A-13. Show that the Galois group of the splitting ﬁeld F of X 3 − 7 over Q is isomorphic to S3 . A-11. and exhibit the ﬁelds between Q and F . Determine whether G is solvable. and let E be a subﬁeld of F (X) properly containing F .e. Find the Galois group of the polynomial X 6 − 5 over each of the ﬁelds Q and R. (a) Give necessary and sufﬁcient conditions on m and n for Fpn to have a subﬁeld isomorphic with Fpm . Prove that X is algebraic over E. How many roots in F does each of the following polynomials have? X 3 − 1. A-15. A-7. A-4. Which of the ﬁelds between Q and F are normal over Q? √ √ A-3. X 4 − 1. 7] = Q[α]. i. A-8. The coefﬁcients of a polynomial f (X) are algebraic over a ﬁeld F . Prove that the two ﬁelds Q[ 7] and Q[ 11] are not isomorphic. Let G be the Galois group of (X 4 − 2)(X 3 − 5) over Q. (b) Prove that C is isomorphic to inﬁnitely many different subﬁelds of itself. (b) Construct explicitly a ﬁeld of order 9. an element such that √ Find Q[ 3. and why? A-2. how many subﬁelds isomorphic with Fpm are there.

n > 1 not prime.) A-25. Show that there exists a ﬁeld M such that F ⊂ M ⊂ E2 . Let E be a Galois extension of F with Galois group Sn . Show that there is a ﬁeld F with K ⊂ F ⊂ L such that [F : K] = 100. Find the degrees of E1 . σn be distinct F -isomorphisms E → Ω. and that if L = E[i]. 2. and E1 E2 over F . (a) If L/K is an extension of ﬁelds of degree 2.) A-18. and let Ω be an algebraic closure of E. A-21. (b) Show that [E : F ] ≥ m. . Let Ei = E Hi . M = F . How many ﬁelds are there strictly between Q[ζ 3 ] and Q[ζ]. Prove or disprove by counterexample. Show that i ∈ E. A-23. (b) What is the structure of G as an abstract group (is it cyclic. ﬁnd the minimum polynomial of θ. Let E be an extension ﬁeld of F . n). A-24. dihedral. ζ 5 = 1. A-27. (c) Let F have characteristic p > 0. . Prove that X 2 = −1 has a solution in F if and only if #F ≡ 1 mod 4. Let F be a ﬁnite ﬁeld of characteristic = 2. E2 . Prove that the trace map T = TrE/F and the norm map N = NmE/F of E over F both map E onto F . An algebraic extension L of a ﬁeld K of characteristic 0 is generated by an element θ that is a root of both of the polynomials X 3 − 1 and X 4 + X 2 + 1. Let E be the splitting ﬁeld over Q of (X 2 − 2)(X 2 − 5)(X 2 − 7). (You may quote basic properties of ﬁnite ﬁelds and the trace and norm. . Identify the Galois group of the splitting ﬁeld K of X 4 − 3 over Q. ﬁnd explicitly its splitting ﬁeld over Q and elements that generate its Galois group. ζ = 1. then −1 is a / √ norm from L to E.)? A-16. . Show that each σi has a unique extension to a homomorphism σi : L → Ω. Let ζ be a primitive 12th root of 1 over Q. A-19. . i = 1. E1 ∩ E2 . and let L be a subﬁeld of Ω containing E and such that ap ∈ E for all a ∈ L. M = E2 . Let F be a subﬁeld of a ﬁnite ﬁeld E. . . then there is an automorphism σ of L such that K is the ﬁxed ﬁeld of σ. . alternating. (a) Show that σ1 . A-26. For the polynomial X 3 − 3. . Determine the number of quadratic subﬁelds. A-22. Let σ1 . A-17. as well as a set of deﬁning relations. A ﬁnite Galois extension L of a ﬁeld K has degree 8100. symmetric. etc. Let E = Q[ζ]. and let H2 be the subgroup generated by the cycle (123 . but that no such ﬁeld exists for E1 . (You must prove that E = Q[α]. Let H1 be the subgroup of Sn of elements ﬁxing 1. Find an element α in E such that E = Q[α]. (b) The same as (a) except that L is also given to be ﬁnite. Here i = −1. σn are linearly dependent over Ω. . . Given that L = K. A-20.87 (a) Give a set of generators for G.

Show that α is a bijection. (State carefully any theorems you use. Find the group of Q-automorphisms of the ﬁeld k = Q[ −3. 0 ≤ i ≤ n − 1. Prove that there is a chain of ﬁelds Q = F0 ⊂ F1 ⊂ · · · Fn = F such that for every i. √ A-34. −2]. . when (a) F = Q and when (b) F = F2 . prove that the Galois group of L/Q[ 7] is isomorphic to S3 . what is [F : Q]? A-30. Let Ω1 and Ω2 be algebraically closed ﬁelds of transcendence degree 5 over Q. A-31. (b) How many proper subﬁelds of L are normal extensions of Q? For each such ﬁeld F . A-32. and let θ be the automorphism of K that maps every element to its pth power. n ≥ 1. Describe the splitting ﬁeld and Galois group.88 A REVIEW EXERCISES A-28. determine how many ﬁelds K there are such that E ⊃ K ⊃ F with [K : F ] = 2. Let F = Fp for some prime p. Describe the Galois group Gal(K/Q) of K over Q. over Q. (a) Prove that the group of automorphisms of L is the symmetric group on 5 elements. Find Gal(L/Q) when L is the splitting ﬁeld over Q of (a) (X 2 − 5)(X 5 − 1). In each case. Suppose that L has an automorphism that ﬁxes three of these roots while interchanging the other two and also an automorphism α = 1 of order 5. Show that there exists an automorphism α of K such that θα2 = 1 if and only if n is odd. namely. Identify the Galois group G of the polynomial f (X) = X 5 −6X 4 +3 over F . show that the number of ﬁelds between K and L is at most 2d! . Find [K : F ] and prove that your answer is correct. and show that K has exactly one subﬁeld of degree 2 over Q. If L/K is a separable algebraic ﬁeld extension of ﬁnite degree d. (b) (X 2 + 3)(X 5 − 1). Let K be the splitting ﬁeld over Q of X 5 − 1. A-29. Let m be a positive integer not divisible by p. α(1) = 1). A-37. L is the splitting ﬁeld of f (X) over Q[ 7]. and (iii) K1 K2 = E. Prove that the polynomial f (X) √ X 3 − 5 is irreducible over the ﬁeld Q[ √ If = 7].) √ √ A-33. A-35. ζ = 1 a root of X 5 − 1. [Fi+1 : Fi ] = 3. (ii) K1 ∩ K2 = F . Let L be the splitting ﬁeld over Q of an equation of degree 5 with distinct roots. Suppose that E is a Galois ﬁeld extension of a ﬁeld F such that [E : F ] = 53 ·(43)2 . of the polynomial X 5 − 9. if E is the splitting ﬁeld of f (X) over F . A-36. say with pn elements. Prove that there exist ﬁelds K1 and K2 lying strictly between F and E with the following properties: (i) each Ki is a Galois extension of F . Q[ζ + ζ 4 ]. and let α : Ω1 → Ω2 be a homomorphism (in particular. and let K be the splitting ﬁeld of X m − 1. Prove that there must exist a subﬁeld K of L such that the Galois group of L/K is cyclic of order 3. A-38. Let K be a ﬁeld of characteristic p. Let F/Q be a Galois extension of degree 3n . Find the minimum polynomial of ζ + ζ 4 over Q. A-39.

Let F be a ﬁeld of characteristic zero. A-46. List all the subﬁelds of E. Let K be a ﬁeld of characteristic p > 0 and let F = K(u. Find the splitting ﬁeld and Galois group of the polynomial X 3 − 5 over the ﬁeld √ Q[ 2]. and let K be an extension of k of degree 2. Deduce that there exists an automorphism σ of Q[α] such that (a) σ 2 = 1 and (b) ρ(σγ) = ρ(γ) for all γ ∈ Q[α] and ρ : Q[α] → C. Assume that there is a q ∈ Q. Let g(X) ∈ Z[X] be a monic polynomial that becomes equal to f (X) when its coefﬁcients are read modulo p. Let F be a ﬁeld of 81 elements. [Assume all ﬁelds are subﬁelds of C. Find the splitting ﬁeld (over Fp ) of X p − X ∈ Fp [X]. A-51. Show that g(X) is irreducible in Q[X]. Let K = Q[ 5. (c) Determine the degree of the ﬁeld K ∩ L over Q. Prove that K is not ﬁnite. Prove that there is a unique automorphism σ of K of order 2 which leaves k elementwise ﬁxed and determine the number of elements of K × such that σ(x) = x−1 . (a) Determine the Galois groups of K and L over Q. √ √ A-48. X 81 − 1. For each of the following polynomials g(X). Prove or disprove these statements: (a) There is a ring homomorphism of F into E if and only if #E is a power of #F . A-42. that F is not a simple extension of K. determine the number of roots of g(X) that lie in F : X 80 − 1. Find its splitting ﬁeld in each case. Show that every equation g(X) = 0. such that |ρ(α)| = q for all homomorphisms ρ : Q[α] → C. q = 0. and that there exist inﬁnitely many intermediate ﬁelds F ⊃ L ⊃ K.89 A-40. A-45. −7] and let L be the splitting ﬁeld over Q of f (X) = X 3 − 10. (b) F25 . Let E be the splitting ﬁeld of X 3 − 51 over Q. X 88 − 1.] A-49. r r . and deduce that X p − X has an irreducible factor f ∈ Fp [X] of degree r. ﬁnd the Galois group over Q of the polynomial X 5 − 5p4 X + p. Let Q[α] be a ﬁeld of ﬁnite degree over Q. A-43. A-52. Let F and E be ﬁnite ﬁelds of the same characteristic. For any prime p. A-41. A-47. (b) Decide whether K contains a root of f . is solvable by extracting radicals. Factorize X 4 + 1 over each of the ﬁnite ﬁelds (a) F5 . Let k = F1024 be the ﬁeld with 1024 elements. and (c) F125 . and let p be a prime number. A-44. Suppose that F has the property that all irreducible polynomials f (X) ∈ F [X] have degree a power of p (1 = p0 is allowed). g ∈ F [X]. v) be a ﬁeld extension of degree p2 such that up ∈ K and v p ∈ K. Show that the set of roots of the minimum polynomial of α is the same as that of q 2 /α. and ﬁnd an element γ of E such that E = Q[γ]. Describe the Galois group of the polynomial X 6 − 7 over Q. A-50.

Find the degree over Q of the Galois closure K of Q[2 4 ] and determine the isomorphism class of Gal(K/Q). √ (b) Prove that every subﬁeld of K of degree 2 over Q is of the form Q[ m] where m ∈ {p. Let p. Prove that: (a) there exist intermediate ﬁelds L and M such that [L : K] = p2 and [M : K] = q. and (c) the Galois group of E/K must be abelian. √ √ A-59. A-53. A-54. and consider the extension K = Q[ p. q] ⊃ Q. Prove that M is ﬁnite-dimensional over L. and distinct. A-58. q. (a) Prove that the Galois group is isomorphic to C2 × C2 . 1 . Let E/K be a Galois extension of degree p2 q where p and q are primes.90 A REVIEW EXERCISES (b) There is an injective group homomorphism of the multiplicative group of F into the multiplicative group of E if and only if #E is a power of #F . Let L be an intermediate ﬁeld different from K. (a) Show that the θi are real. compute their lengths. Let K be a ﬁeld. Without carrying it out. nonrational. give a brief description of a method for deciding which it is. (c) Show that the rows of the matrix 3 9 9 9 3 θ θ θ 1 2 3 3 θ 2 θ 3 θ 1 3 θ3 θ1 θ2 are pairwise orthogonal. 1 (b) Find the minimum polynomial of ζ + ζ over Q. A-55. pq}. and compute the determinant of the matrix. (c) Show that there is an element γ ∈ K such that K = Q[γ]. A-56. A-57. Let θ1 . Prove that L is algebraically closed if every polynomial over K factors completely over L. (b) Explain why the Galois group of f (X) over Q must be either A3 or S3 . (a) Prove that 1 + X + X 2 + X 3 + X 4 + X 5 + X 6 is the minimum polynomial of ζ over Q. θ3 be the roots of the polynomial f (X) = X 3 + X 2 − 9X + 1. q < p and q not dividing p2 − 1. θ2 . Let ζ be a primitive 7th root of 1 (in C). X an indeterminate. (b) such ﬁelds L and M must be Galois over K. and let M = K(X). q be distinct positive prime numbers. Let L/K be an algebraic extension of ﬁelds.

(a) Let σ be an automorphism of a ﬁeld E. or give a counterexample. Assume Ω is a ﬁnite Galois extension of F and construct a ﬁeld E. . . and what are their degrees? 4. ai−1 ]. and (b) Ω has ﬁnite transcendence degree over F . If σ 4 = 1 and σ(α) + σ 3 (α) = α + σ 2 (α) all α ∈ E. . and what are their degrees. . Let Ω be an extension ﬁeld of a ﬁeld F . .) You should prove all answers. (b) g(X) = X 3 + 3X + 1. .] 5. . d such that a = c2 +pd2 and b = c22cd 2 . Show 2 2 that there exist rational numbers c. . 3. What is the relation between the Galois group of f and that of g? Find the Galois group of f if (a) g(X) = X 3 − 3X + 1. 6. but you should indicate clearly what you are using. (b) Let p be a prime number and let a. show that σ 2 = 1. F ⊂ E ⊂ Ω. b be rational numbers such that a2 + pb2 = 1. Let E be the splitting ﬁeld of (X 5 − 3)(X 5 − 7) ∈ Q[X].91 B Two-hour Examination 1. c −pd −pd 2. an ]. Show that every F -homomorphism Ω → Ω is an isomorphism provided: (a) Ω is algebraically closed. and let g(X) be the resolvent cubic of f . Consider an extension Ω ⊃ F of ﬁelds. Let f (X) be an irreducible polynomial of degree 4 in Q[X]. (a) How many monic irreducible factors does X 255 − 1 ∈ F2 [X] have. Deﬁne a ∈ Ω to be F -constructible if it is contained in a ﬁeld of the form √ √ F [ a1 . √ √ ai ∈ F [ a1 . You may use results proved in class or in the notes. Possibly useful facts: The discriminant of X 3 + aX + b is −4a3 − 27b2 and 28 − 1 = 255 = 3 × 5 × 17. Can either of the conditions (i) or (ii) be dropped? (Either prove. such that every a ∈ Ω is E-constructible and E is minimal with this property. What is the degree of E over Q? How many proper subﬁelds of E are there that are not contained in the splitting ﬁelds of both X 5 − 3 and X 5 − 7? [You may assume that 7 is not a 5th power in the splitting ﬁeld of X 5 − 3. . (b) How many monic irreducible factors does X 255 − 1 ∈ Q[X] have.

Hence any ﬁnite subgroup G of F × satisﬁes the condition in (d). 1-4. as is the “only if” in (b). Suppose that. (a) is obvious. E ≈ F [X]/(X 2 − a). reduce the cubic to the form X 3 − pX − q. a ∈ F 2 . Verify that 3 is not a square in Q[ 2]. . (d) If G had two cyclic factors C and C whose orders were divisible by a prime p. Note that it sufﬁces to construct α = cos 2π . then G would have (at least) p2 elements of order dividing p.18. 2-1. Let α be the real 5th √ of 2. . (d) Any ﬁeld of characteristic p contains (an isomorphic copy of) Fp . then α = aβ will be a root of X 3 − pX − q. Then construct a 4q so that a2 = 4p . (b) As 0 and 1 are not roots of X p − X − 1 in Fp it can’t have p distinct roots in Fp . 3] : Q] = 4.28. 1-3. 2-3. The homomorphism a → a2 : F× → F× p p has kernel {±1}. Students were expected to write out complete solutions. where ζ is a primitive 5th root of 1. and it follows that α ∈ F . Eisenstein’s criterion shows that X 5 − 2 is irreroot ducible in Q[X]. 1-1. . For this use that (a + b p)2 ∈ Q =⇒ 2ab = 0. / √ (c) Take Ei = Q[ pi ] with pi the ith prime. Moreover. f (X) = (X r + a1 X r−1 + · · · + ar )(X p−r + · · · ).13). in F [X]. and so its image has index 2 in F× . and so Q[ 5 2] has degree 5 over Q. and 7 2 so its minimum polynomial has degree 3. α + p − 1. take E = p F [X]/(X 2 − a). The remaining roots of X 5 − 2 are ζα. Check that pi is the only prime that √ becomes a square in Ei . ζ 2 α. 1. (e) The elements of order m in F × are the roots of the polynomial X m − 1. and put X = a to ﬁnd r = f (a). and so [Q[ 2. . and so (a) implies that X p − X − 1 is irreducible in Fp [X] and hence also in Z[X] and Q[X] (see 1. Similar to Example 1. (a) If α is a root of f (X) = X p − X − a (in some splitting ﬁeld). ζ 4 α. and it follows that G is cyclic. −a1 = rα + d some d ∈ Z · 1F . and so there are at most m of them. 0 < r < p. and so there is at most one E (up to Fp -isomorphism). Then −a1 is a sum of r of the roots of f . This doesn’t happen. There is a standard method (once taught in high schools) for solving cubics using the equation cos 3θ = 4 cos3 θ − 3 cos θ. to get f (X) = q(X)(X − a) + r(X) with r(X) constant. (c) Use that factorization in F [X] is unique (or use induction on the degree of f ). and so we are looking at the quadratic extensions of Fp . √ √ √ 1-2. then the remaining roots are α + 1. It follows that the subﬁeld of C . / p 2-2. ζ 3 α. Choose 3θ such that cos 3θ = a3 . Thus the only possibility for S(E) p is F× . a ∈ F2 . To get one. By “completing the cube”. and that [Q[α] : Q] = 7−1 = 3. they are distinct. For the “if” note that for any a ∈ S(E). which obviously lie in whichever ﬁeld contains α. (a) Apply the division algorithm. If β = cos θ is a solution of the above 3 equation.92 C SOLUTIONS TO THE EXERCISES C Solutions to the Exercises These solutions fall somewhere between hints and complete solutions.

) 2-5. For p ≡ −1 mod 4. it can be written f1 (X) = f2 (X p ). Therefore g(X) = 2-6. which implies that [F (X) : E] is inﬁnite (otherwise F (X) = E[α1 . αn ]. αβ = −( p−1 )2 /(p − 1) = p−1 . . and so (στ )m (X) = m + X. .12) we know that either f is separable or f (X) = f1 (X p ) for some polynomial f1 . 2-4. p−1 2 +X + with roots (−1 ± √ the polynomial is √ 1 − p − 1)/2. Note that αβ = i. Continue in the way until you arrive at a separable polynomial. we have α + β = −1. and so f (X) = g(X p ) = (X − αi )p with e p αi = ai . the ﬁxed ﬁeld of H is Q[ −p]. (Perhaps I meant X p − X — that would have been more interesting. which is a nonsquare. It’s Fp because X p − 1 = (X − 1)p . . Let σ and τ be automorphisms of F (X) given by σ(X) = −X and τ (X) = 1 − X. If f (X) = (X − αi )mi . 3-2. αβ = X2 p−1 2 + (−1) ( p−1 )2 − 2 p−1 4 . (a) It is easy to see that M is Galois over Q with Galois group σ. 3-3. Since 1 + ζ + · · · + ζ p−1 = 0. and α and β are the roots of X 2 + X + αβ. the polynomial satisﬁed by α and β is X 2 + X − p−1 . How many times do we have i+j = 0? If i+j = 0. the remaining terms add to a rational number. 2 4 4 √ √ whose roots are (−1 ± 1 + p − 1)/2. conversely.j ζ i+j . if −1 is a nonsquare. and its value on αi is a root of the minimum polynomial of αi ). σ 3= 3 τ 3=− 3 . i ∈ H. In either case. α] is 20. τ : √ √ √ √ σ √ =− 2 2 τ√ 2 = √ 2 √ . ai = aj . an E-automorphism of F (X) is determined by its values on the αi . take i = 1 and j = −1 to get i + j = 0. then f (X) = and so d(X) = mi >1 (X m m m mi f (X) X − αi (X − αi ). then −1 = i−1 j. If E contains a polynomial f (X) of degree m > 0. the ﬁxed ﬁeld of H is Q[ p]. But ατ X = 1 + X. /(p − 1) = p−1 2 − p−3 4 = p+1 4 . Thus for p ≡ 1 mod 4. So in the sum for αβ we see 1 a total of 2 times when p ≡ 3 mod 4 and not at all if p ≡ 1 mod 4.93 generated by the roots of X 5 − 2 is Q[ζ. and so jα = β and jβ = α. If we do have a solution to i + j = 0. The degree of Q[ζ. j∈G H. α]. If i ∈ H. Then σ and τ ﬁx X 2 and X 2 − X respectively. − αi )mi −1 . then jH = G H and j(G H) = H. note that g(X) = (X − ai ). Hence αβ ∈ Q. Thus Aut(F (X)/E) is inﬁnite. j∈G H ⇐⇒ −1 is a square mod p ⇐⇒ p ≡ −1 mod 4. From (2. For the ﬁnal statee e ment. since it must be divisible by [Q[ζ] : Q] = 4 and [Q[α] : Q] = 5. which implies that each power of ζ occurs the same number of times. and so α and β are ﬁxed by H. 3-1. Hence i + j = 0 some i ∈ H. αi = αj . If j ∈ G H. then [F (X) : E] ≤ [F (X) : F (f (X))] = m — contradiction. . we get all solutions by multiplying it through by the p−1 p−1 2 squares. then iH = H and i(G H) = G H. If f1 is not separable. Clearly f1 is also irreducible. and so στ ﬁxes E =df F (X) ∩ F (X 2 − X).

1 + 1 + 2 + 2. = = 2 α 4−2 2+ 2 2 √ √ i. i] = Q[α]. √ 2 3− 3 √ . the proof of 4. and g(X) = X 3 + X 2 + 3X + 1 is irreducible over Q (use 1. the splitting ﬁeld √ is Q[α. Hence α ∈ M . Eisenstein’s criterion shows that X 8 − 2 is irreducible over Q. irreducible. We can take E = LS3 . Finally compute that √ √ √ 3− 3 √ 3− 3 √ ( 2 − 1)α. 1 + 1 + 4. τ σα = ( 2 − 1) √ α. 4-1. unlike i. and so we can conclude√ σ has that order 4. 4-6.94 (b) We have C SOLUTIONS TO THE EXERCISES √ √ √ 2 √ (2 − 2)2 σα2 2− 2 2− 2 √ = √ = ( 2 − 1)2 . Again σα = ±( 2 − 1)α and σ 2 α = −α. (deg 2)3 . i] : Q] = 2. and let E be the ﬁxed ﬁeld of S3 ⊂ S4 . we may suppose that σα = ( 2 − 1)α. this is impossible. and so would equal Sn (see 4. We have X 4 − 2X 3 − 8X − 3 = (X 3 + X 2 + 3X + 1)(X − 3). if α ∈ M . (deg 1)2 (deg 4). and so [E : Q] = 8. 4-5. There is no subgroup strictly between Sn and Sn−1 . Either check that its discriminant is not a square or. and Gal(E/Q) has two noncommuting elements of order 4.e. Thus the Galois group has elements of type: 6. Therefore the degree is 2 × 8 = 16. because Q[α].13??). Find an extension L/F with Galois group S4 . m = m0 pr . Alternatively.. 4-2. 4-3. 2 + 2 + 2. Repeat the above argument with τ : τ α2 α2 = √ 3−√3 3+ 3 = an automorphism of L (also denoted τ ) with τ α = is 4. As usual for polynomials of this type. is contained in R. ζ can be taken to be 1+i . Type “galois(X 6 + 2X 5 + 3X 4 + 4X 3 + 5X 2 + 6X + 7). √ / Extend σ to an automorphism (also denoted σ) of E. mod 79. 1 + 5. σ 4 |M = 1. 3 + 3. 4-4. as σ 2 α = α = 0.” and discard the 7 factors of degree 1. and so [Q[α. mod 19. Since it has order 8. For example. Thus.”. 2 which lies in Q[α. The order of τ 6 can extend τ to . and so [Q[α] : Q] = 8 where α is a positive 8th root of 2. i]. mod 13 (deg 1)(deg 5). It follows that the splitting ﬁeld is Q[α. it must be the quaternion group.4??). more simply. The splitting ﬁeld is the smallest ﬁeld containing all mth roots of 1. and so we √ 6 3− 3 √ α. Type: “Factor(X 343 − X) mod 7. mod 5 (deg 3)(deg 3). Clearly Q[α. and so σ 2 = 1. and hence its Galois group contains a transposition (cf. because such a subgroup would be transitive and contain an (n − 1)-cycle and a transposition. It is the group PGL2 (F5 ) (group of invertible 2 × 2 matrices over F5 modulo scalar matrices) which has order 120. and √ √ σ 2 α = (− 2 − 1)( 2 − 1)α = −α. show by examining its graph that g(X) has only one real root. Hence it is Fpn where n is the smallest positive integer such that m0 |pn − 1. ζ] where ζ is any primitive 8th root of 1. then σα = ±( 2 − 1)α. After possibly replacing σ with its inverse. i]. mod 61 (deg 1)2 (deg 2)2 . and so its Galois group is either A3 or S3 . σα2 = (( 2 − 1)α)2 . στ α = − 6 6 Hence στ = τ σ. Now σ 4 α = α. note that there are the following factorizations: mod 3.23).

because f is irreducible. a2 be conjugate nonreal roots. and let a3 be a real root. which is indeed PGL2 (F5 ). whose Galois group is cyclic of order 4. The degree [Q[ζn ] : Q] = ϕ(n). 4-8. The derivative of f is 5X 4 + a. If Gf = D5 . Jacobson. The rest is easy. and so there is a τ such that τ (a3 ) = a1 . this is the polynomial Φ5 (X). 4-9. We have Nm(a + ib) = a2 + b2 . For a = 0. Basic Algebra I. a can’t be an integer. for example. For D8 .95 No element of type 2. The cubic does not have ±1. which has at most 2 real zeros. t ∈ Q (Hilbert’s Theorem 90). the resolvent cubic is g(X) = X 3 − X 2 + (1 − 4a)X + 3a − 1. and condition (c) excludes A5 . Hence a2 + b2 = 1 if and only a + ib = some s. which is irreducible. A3 . and therefore the Galois group of f is S4 . Take a = −1. or ±4 as roots. and so g has exactly one real root. and ϕ(n) → ∞ as n → ∞. Now a3 → a1 → a2 a3 → a3 → a1. ζn a primitive nth root of 1. and this shows that we must have Gf = D5 . 5-1. That leaves D5 and C5 as the only possibilities (see. This suggests it is the group T 14 in the tables in Butler and McKay. (b) holds because D5 ⊂ A5 . 3 + 2. or 4 + 2 turns up by factoring modulo any of the ﬁrst 400 primes (or. so I have been told). Ex 6). This statement is false for reducible polynomials — consider for example f (X) = (X 2 + 1)(X − 1). =⇒ : Regard D5 as a subgroup of S5 by letting it act on the vertices of a regular pentagon—all subgroups of S5 isomorphic to D5 look like this one. 4-7. and (c) holds because D5 is solvable. But looking modulo 2. Moreover 1 g = 3X 2 − 2X + 5 = 3(X − 3 )2 + 4 2 > 0 always. s+it s−it σ τ τ σ for 5-3. condition (b) implies that Gf ⊂ A5 . Hence its Galois group is S3 or A3 . so it must be S3 . We have g = X 3 − X 2 + 5X − 4. [In fact. it is (X − 1)(X 3 + 2X 2 + 3X + 4). . and so it is irreducible in Q[X]. Then f = X 4 + X 3 + X 2 + X − 1 is irreducible modulo 2. For a = −4. and so it is irreducible in Q[X]. Complex conjugation deﬁnes an element σ of the Galois group of f switching a1 and a2 and ﬁxing a3 . we see it contains a 2-cycle. then (a) holds because D5 is transitive. it is X(X 3 + X 2 + X + 1) = X(X + 1)(X 2 + 1). whose Galois group is cyclic of order 2. Thus complex conjugation acts as an element of order 2 on the splitting ﬁeld of f .] 5-2. For any a. 4 is the maximum number of integers giving distinct Galois groups: checking mod 2. On the other hand. 3 Hence the Galois group of g is S3 . 3. ⇐= : Condition (a) implies that Gf contains a 5-cycle. and so 1. p305. For a = 1. its Galois group acts transitively on its roots. and so (from its graph) we see that f can have at most 3 real zeros. A4 . ±2. we see there is a 2-cycle or a 4-cycle. V4 are not possible. Omitted. Let a1 .

which implies that α(a) = a. because C SOLUTIONS TO THE EXERCISES n = [Fpn : Fp ] = [Fpn : Fpm ] · [Fpm : Fp ] = [Fpn : Fpm ] · m. where σα = ζα and τ ζ = ζ 2 . Only the subﬁeld Q[ −3] is normal over Gf ≈ S3 . If E1 and E2 are Galois extensions of F .(a) See Exercise 3. it consists of all the m solutions of X p − X = 0. The polynomial is irreducible by Eisenstein’s criterion. which lies in Q only if ab = 0. and then extend it to an isomorphism C → C where C is the algebraic closure of Q(A ) in C. A-2. Use Eisenstein to show that X p−1 + · · · + 1 is irreducible. Hence. E contains an element f (T ) − is a nonzero polynomial having X as a root. It has 3 elements of order dividing 3. and therefore preserves the order. (b) Only one. It is not abelian (because Q[α] is not Galois over Q). The group has order 20. Hence the rational numbers that become √ squares in Q[ 7] are those that are already squares or lie in 7Q×2 . This proves that G 3 72 .96 A-1. A-10. but it is solvable (its order is < 60). then E1 E2 and E1 ∩ E2 are Galois over F . A-9. {r ∈ Q | a < r} = {r ∈ Q | α(a) < r}. but 11 does not: (a + b 7)2 = √ a2 + 7b2 + 2ab 7. The polynomial has only one real root. Take α to be a root of X 2 + X + 2. The group F × is cyclic of order 15. Use Galois theory to show there exists one. A-5. Done in class. (b) Let F = F3 [X]/(X 2 + 1). A-4. and therefore complex conjugation is a transposition in √ f . [The discriminant of X 3 − a is −27a2 = −3(3a)2 . The splitting ﬁeld is Q[ζ. (b) Choose a transcendence basis A for C over Q. there is a bijection α : A → A from A onto a proper subset. (a) Need that m|n. A-6. It is generated by σ = (12345) and τ = (2354). A-8. − − − − f g with the degree of f or g > 0. 15 elements of order dividing 15. for each real number a. and 1 element of order dividing 17. for example. Q[ζ 3 7] Q[ζ 2 3 7] are not normal over Q. Because it is inﬁnite. The subﬁelds Q[ 3 7]. The discriminant is √ −1323 = √ −3 √ Q. The prime 7 becomes a square in the ﬁrst ﬁeld. and it maps positive real numbers to positive real numbers. Now f (X) g(T ) g(X) . hence on Q. Modulo 3 X 8 − 1 = (X − 1)(X + 1)(X 2 + 1)(X 2 + X + 2)(X 2 + 2X + 2). Since E = F . (a) A homomorphism α : R → R acts as the identity map on Z. 1 element of order dividing 4. and there is an exact sequence 1 − → Gal(E1 E2 /F ) − → Gal(E1 /F ) × Gal(E2 /F ) − → Gal(E1 ∩ E2 /F ) − → 1.] √ A-3. α] where ζ 5 = 1 and α5 = 2. etc. A-7. Extend α to an isomorphism Q(A) → Q(A ).

the splitting ﬁeld is Q[α. A-27. We have Q[ζ] = Q[i. √ √ √ A-17. A-15. . because X p − ap has a unique root in Ω. (b) is true—couldn’t be inseparable.. The Galois group is the product (they could only intersect in Q[ 3]. and hence its image has order q − 1. A-21. 3 + 7 will do. The splitting ﬁeld is Q[i. E1 ∩ E2 = Q[ζ] where ζ is a primitive cube root of 1. If f is separable. and Galois group D4 . The degree is 18. .. an — they lie in the algebraic closure Ω of F . (c) is O.. Etc. and the Galois group is D4 with generators (1234) and (13) etc. A-19. A-28. which must therefore be the minimum polynomial for θ. The splitting ﬁeld is Q[ζ. A-16. The splitting ﬁeld of X 3 − 5 is E2 = Q[ζ. Use that (ζ + ζ 4 )(1 + ζ 2 ) = ζ + ζ 4 + ζ 3 + ζ A-22. Let the coefﬁcients of f be a1 . ζ ] where ζ is a primitive cube root of 1 and ±i = ζ 3 etc. Over R. and the Galois group is S3 . the coefﬁcients will lie in some ﬁnite extension E of F .97 In this case. has kernel F × . The map E × → E × . √ A-20. Now the Fundamental Theorem of Galois theory shows that F exists. A-25.K. Let g(X) be the product of the minimum polynomials over F of the roots of f in Ω. and the Galois group is D6 (generators (26)(35) and (123456)). the Galois group is C2 . The greatest common divisor of the two polynomials over Q is X 2 + X + 1.4. [E : F ] = (Gf : 1). it has degree 8. The multiplicative group of F is cyclic of order q − 1. Alternatively. A-18. A-11. 3 3]. (a) is Dedekind’s theorem. α → σα . The splitting ﬁeld of X 4 − 2 is E1 = Q[i. etc. . A-24. √ √ A-14. it has degree 6. From Hilbert’s theorem 90. and Galois √ √ group D3 . but 3 does not become a square in E1 ). ζ] where α6 = 5 and ζ 3 = 1 (because −ζ is then a primitive 6th root of 1). we know that the kernel of the map N : E × → F × consists of elements of the form σα . A-12. Over Q. There is a similar proof for the trace — I don’t know how the examiners expected you to prove it. Apply the Sylow theorem to see that the Galois group has a subgroup of order 81. which is a subgroup of Sn . We have E1 = E H1 . A-23. A-13. β]. A-26. This is really a problem in group theory posing as a problem in ﬁeld theory. Take α = 2 + 5 + 7. α] where α4 = 3. which has degree n over F . Hence it contains an element of order 4 if and only if 4|q − 1. and E2 = E <1···n> . Theorem on p-groups plus the Fundamental Theorem of Galois Theory. The degree is 12. and we can take the norm of f (X) from E[X] to F [X]. which has degree (n − 1)! over F . α m α −1 Therefore the kernel of N has order qq−1 . .. α] where α4 = 2. (a) is false—could be inseparable. (b) is Artin’s theorem 3. .

[Of course. Let a1 . and the Galois group is cyclic. . For example. therefore they are a maximal algebraically independent subset of Ω2 . It was proved in class that Sp is generated by an element of order p and a transposition (4. . (ζ + ζ 4 )(ζ 2 + ζ 3 ) = −1. A-43. A-30. generators (12345) and (2354). . Hence the Galois group is S5 . hence 80.98 C SOLUTIONS TO THE EXERCISES A-29. √ A-34. and a set with d! elements has at most 2d! subsets. (b) Certainly. since the Galois group is a cyclic group of order n generated by θ. The splitting ﬁeld of the minimum polynomial of α has degree at most d!. A-37. ζ] where ζ = 1 is a cube root of 1. and it is quadratic over Q. A-40. A-32.. they all contain 1 and they are invariant under a → a−1 . Let L = K[α]. √ much less Q[ 7]. the Galois group is a product C2 × C4 . say n = 2m + 1. which is not a square in any R. Its derivative is 5X 4 − 5p4 .. A-36. of Sylow 5-subgroups. The group F × is cyclic of order 80. A-38. Its roots are primitive 8th roots of 1. The discriminant is −675. Their images are algebraically independent. The Galois group is S5 . it would have a root in it. Hence the graph crosses the x-axis 3 times and so there are 2 imaginary roots. 11. A-45.] A-31. with generators (26)(35) and (123456). (b) It factors as (X + 1)(X 4 + X 3 + X 2 + X + 1). with generators (12) and (12345) — it is irreducible (Eisenstein) and (presumably) has exactly 2 nonreal roots. C2 × C2 . A-44. See Exercise 14..5. 44. 21. There are 1. 1. Take K1 and K2 to be the ﬁelds corresponding to the Sylow 5 and Sylow 43 subgroups. A-33. . which cyclic of order 4.. (a) (X 2 +2)(X 2 +3). The polynomial is irreducible by Eisenstein’s criterion. 6. a5 be a transcendence basis for Ω1 /Q. but it is irreducible over Q by Eisenstein’s criterion. A-39. A-41. Hence the splitting ﬁeld has degree 4 over F2 . then α = θm does. If f (X) were reducible over Q[ 7]. If n is odd. . subgroups of . (a) Omit. Omit. . and therefore they form a transcendence basis.14). generated by 2. The Galois group is (Z/5Z)× . A-35. It splits completely in F25 . this bound is much too high: the subgroups are very special subsets. (ζ + ζ 4 ) + (ζ 2 + ζ 3 ) = −1. It’s irreducible by Eisenstein. only 1 divides 43.. which has the roots X = ±p. 87. These are the max and mins... There is only one F . Example 5. X = −p gives positive.. (a) Should be X 5 − 6X 4 + 3. It has order 20. 8. X = p gives negative. . 16. etc. A-42. This is really a theorem in group theory. Note that of the possible numbers 1. It’s D6 . and its splitting ﬁeld is Q[α.

true. Show c2 +pd2 that there exist rational numbers c. A-56. What is the relation between the Galois group of f and that of g? Find the Galois group of f if . and hence are linearly independent (Dedekind 5. The argument that proves the Fundamental Theorem of Algebra. 1 A-49. and let g(X) be the resolvent cubic of f . A-48. N x = 1. (b) See example worked out in class. This is really a group theory problem disguised as a ﬁeld theory problem. A-50. 1. (a) Let σ be an automorphism of a ﬁeld E.99 A-46. 1024 = 210 .e. Solutions for the exam. with generators (1234) and (12).] (b) Let p be a prime number and let a. σ. so either σ = 1 or the characteristic is 2 (or both). If σ 2 = 1. (a) C2 × C2 and S3 . Hence ρ( q ) is the complex conjugate of ρ(α).. −− −− −− → Hence the number is 211 /210 = 2. 2. then 1. Omit. and let H be the Sylow p-subgroup. A-58. A-59. Let f (X) be an irreducible polynomial of degree 4 in Q[X]. Omit. They are the elements of the form have 1 − − → k× − − → K × − − x K × . (a) Prove it’s irreducible by apply Eisenstein to f (X + 1). A-52. ρ(α)ρ(α) = q 2 . and so the Galois group is a p-group. x→ σx σx x . σ 3 are distinct automorphisms of E. Omit. Omit. b be rational numbers such that a2 + pb2 = 1. [If σ 2 = 1. Similar to a previous problem. 2 2 show that σ 2 = 1. A-51. Let E be the splitting ﬁeld of g(X). False. Pretty standard. Omit. d such that a = c2 −pd2 and b = c22cd 2 . −pd √ √ Apply Hilbert’s Theorem 90 to Q[ p] (or Q[ −p]. then the condition becomes 2σ = 2. A-53. A-54. (c). It’s D4 . i. and ρ(α)ρ( q ) = q 2 . A-57. α α Hence the automorphism induced by complex conjugation is independent of the embedding of Q[α] into C. shows that its Galois group is a p-group. Want σx · x = 1. σ 2 . A-55. Then E H = F . If σ 4 = 1 and σ(α) + σ 3 (α) = α + σ 2 (α) all α ∈ E. depending how you wish to correct the sign). A-47.14) — contradiction. (b) No.

√ √ ai ∈ F [ a1 . 32.100 C SOLUTIONS TO THE EXERCISES (a) g(X) = X 3 − 3X + 1. The two cubic polynomials are irreducible. 2. Thus [E : Q] = 100. X 255 − 1 = d|255 Φd = Φ1 Φ3 Φ5 Φ15 · · · Φ255 . Q[α. . We want the nontrivial subgroups of G not containing C5 × C5 . 3 of degree 4.. . or give a counterexample. Each subgroup of C5 × C5 is of the form H ∩ (C5 × C5 ) for exactly 3 subgroups H of G corresponding to the three possible images in G/(C5 × C5 ) = C4 . 8. and their minimum polynomials all have degree 8. . Consider an extension Ω ⊃ F of ﬁelds. 5 [You may assume that 7 is not a 5th power in the splitting ﬁeld of X 5 − 3. Suppose E has the required property. (b) g(X) = X 3 + 3X + 1. where H is the Sylow 2-subgroup of Gal(Ω/F ). Assume Ω is a ﬁnite Galois extension of F and construct a ﬁeld E.}. From the primitive element theorem. (12)(34). because their only possible roots are ±1..] 5. cos 2π ]. What is the degree of E over Q? How many proper subﬁelds of E are there that are not contained in the splitting ﬁelds of both X 5 − 3 and X 5 − 7? The splitting ﬁeld of X 5 − 3 is Q[ζ. (a) How many monic irreducible factors does X 255 − 1 ∈ F2 [X] have. Let Ω be an extension ﬁeld of a ﬁeld F . We have Gg = Gf /Gf ∩ V . F ⊂ E ⊂ Ω. 3. . α]. an ]. which has subﬁelds F24 ⊃ F22 ⊃ F2 . and we showed in class that the Φd are irreducible. one ﬁnds that the ﬁrst has Galois group A3 and the second S3 . α] is (by . . Typical ﬁelds: Q[α]. The subgroups of order 5 of C5 × C5 are lines in (F5 )2 . 4. Since 7 is not a 5th power in Q[ζ. and hence C5 × C5 has 6 + 1 = 7 proper subgroups. Let E be the splitting ﬁeld of (X 5 − 3)(X 5 − 7) ∈ Q[X]. 6. and hence has order 1 or 5. 4. and 1 each of degrees 2 and 1. ai−1 ]. and 20 nontrivial ones. 16. They have degrees 1. Take E = ΩH . 128.) . . All are normal in G. Hence we have 21 subgroups of G not containing C5 × C5 . and (b) Ω has ﬁnite transcendence degree over F . 4|(Gf : 1) and it follows that Gf = A4 and S4 in the two cases. we know Ω = E[a] for some a. and what are their degrees? Obviously. Q[α. . Deﬁne α ∈ Ω to be F -constructible if it is contained in a ﬁeld of the form √ √ F [ a1 . . Because f (X) is irreducible. α]. Can either of the conditions (i) or (ii) be dropped? (Either prove. .) a subgroup of a cyclic group of order 5. . Show that every F -homomorphism Ω → Ω is an isomorphism provided: (a) Ω is algebraically closed. where V = {1. it must be 5. There are 256 − 16 elements not in F16 . which has degree 5 over Q[ζ] and 20 over Q. The Galois group of X 5 − 7 over Q[ζ. 64. (b) How many monic irreducible factors does X 255 − 1 ∈ Q[X] have. ζ]. From their discriminants. . such that every a ∈ Ω is E-constructible and E is minimal with this property. and what are their degrees? Its roots are the nonzero elements of F28 . Now a E-constructible =⇒ [Ω : E] is a power of 2. and G = Gal(E/Q) = (C5 × C5 ) C4 . Hence there are 30 factors of degree 8.

101 Let A be a transcendence basis for Ω/F . . X2 → X3 . . . X → X 2 . C(X)→ C(X). and hence (because it has the right number of elements) is a transcendence basis for Ω/F . Because Ω is algebraic over F [σA] and σΩ is algebraically closed. E. X2 ... the two are equal. . . Now F [σA] ⊂ σΩ ⊂ Ω.g.. and consider an extension of the map X1 → X2 . Ω = the algebraic closure of C(X1 . . Because σ : Ω → Ω is injective. Neither condition can be dropped. .). σ(A) is algebraically independent over F . X3 .g. E.

54 commutative. 17. 10 monic. 27 separable. 8. 6 Galois. 27 simple. 30 Galois correspondence. 5 element maximal. 54 ideal. 9 ﬁxed ﬁeld. 62 normal basis. 76 discriminant. 17 in a larger ﬁeld. 75 Galois ﬁeld. 6 ﬁeld. 28 cyclic. 27 inseparable. 27 normal. 3 Euclid’s. 45. 25 proﬁnite. 26 Frobenius automorphism. 49 degree. 10–12 algebraic closure. 2 zero. 71 Galois closure. 23 102 . 17 algebraically dependent. 2 stem. 25 birational. 2 closure separable. 70 basis separating transcendence. 13. 6 separable. 10 solvable. 76 Kummer theory. 28 directed. 1 composite of ﬁelds. 76 inverse system. 33 cubic resolvent. 6. 79 algorithm division. 26 inverse limit. 1 principal crossed. 5 algebraically closed. 2 of rings.Index algebraic. 58 group Cremona. 27 inﬁnite. 42. 83 transcendence. 28 extension ﬁeld. 79 algebraically independent. 22 norm. 48 of Galois theory. 2 endomorphism. 25 base neighbourhood. 10 conjugates. 24 fundamental theorem of algebra. 37. 66 total. 29 Galois. 28 constructible. 1 invariants. 66 extension abelian. 50. 81 bound upper. 66 perfect ﬁeld. 43 Galois group. 54 of ﬁelds. 1 integral domain. 11. 3 factoring a polynomial. 3. 56 Lemma Gauss’s. 30 ordering partial. 23 polynomial minimum. 1 prime. 17 algebraic integer. 36 Eisenstein’s criterion. 34 Gaussian numbers. 77 topological. 40. 7 general polynomial. 53 multiplicity. 4 separable. 72 of a polynomial. 66 characteristic p. 5 automorphism. 70 homomorphism crossed. 28 ﬁnite. 65 module G-. 55. 6. 69 cohomology group. 13. 38 cyclotomic polynomial. 4 Maple. 2. 52 normal closure.

28 separably closed. 20 subﬁeld. 2 constructibility of n-gons. 10 subring. 10–12 . 14. 9 symmetric polynomial. 1 root multiple. 51 constructible numbers. 46 primitive root of 1. 58 elementary. 49 regular n-gon. 1 generated by subset. 34 split. 51 Liouville. 59 theorem Artin’s. 27 independence of characters. 26 binomial in characteristic p. 16 primitive element. 20 splits. 22 separable. 33 cyclotomic polynomials. 82 transcendental. 51 ring. 46 separable element. 12 normal basis. 84 trace. 52 primitive element. 44 Galois 1832. 22 simple.INDEX 103 prime Fermat. 50 Dedekind. 34 Galois extensions. 69 solvable in radicals. 46 topology Krull. 1 generated by subset. 16 splitting ﬁeld. 72. 62 transcendence degree.

- Visual Abstract Algebra(2003.03.25. 243s)_Marcus Pivato
- integral transforms
- Notes on Linear Algebra-Peter J Cameron
- Notes for Sf 2735
- Andrew Rosenberg- Lecture 2.1
- Solusi soal Rings
- Matrices L
- 02
- Lecture 5
- Misc Matrix Algebra
- (ebook - pdf) - ALGEBRAIC GEOMETRY
- lecture5
- Eckert
- RC4901B54-MATHS
- Linear Algebra Ch 1 and 2 Notes
- fermat talk
- Riemann Zeta Alternative_0709.4173v25
- 1666
- Ji 2416271633
- Eigenvalues Eigenvectors
- 3revsoln
- Math 185 Topology
- Matrix Algebra[1]
- 2 Vector Algebra
- Ko or Winder
- H2 Model Reduction
- inner1
- Factorization of Permutations into n-cycles
- Partial differential equations

Are you sure?

This action might not be possible to undo. Are you sure you want to continue?

We've moved you to where you read on your other device.

Get the full title to continue

Get the full title to continue listening from where you left off, or restart the preview.

scribd