Skip to content

abstract algebra

1. Introduction

1.1 What is Abstract Algebra?

Abstract algebra is the study of algebraic structures — sets equipped with operations satisfying certain axioms. Rather than studying specific objects (numbers, matrices, polynomials) in isolation, abstract algebra identifies the common structural patterns they share and studies these patterns in their full generality.

The central objects of study are groups, rings, and fields. Each adds successive layers of algebraic structure:

  • Groups: A set with one binary operation satisfying closure, associativity, identity, and inverses.
  • Rings: A set with two binary operations (addition and multiplication) where addition is abelian, multiplication is associative, and the distributive laws hold.
  • Fields: A ring in which every nonzero element has a multiplicative inverse.

1.2 Motivation

Abstract algebra arises by definition from several directions:

  • Number theory: Fermat”s little theorem and Euler’s theorem are most by definition understood through the lens of group theory. The structure of Z/nZ\mathbb{Z}/n\mathbb{Z} and its units underpins modular arithmetic.
  • Equation solving: The question “which polynomial equations can be solved by radicals?” motivated Galois theory, one of the crowning achievements of 19th-century mathematics.
  • Geometry and physics: Symmetry groups describe the fundamental symmetries of physical systems, crystals, and geometric objects. Lie groups and Lie algebras bridge algebra with differential geometry and quantum mechanics.
  • Cryptography: RSA, Diffie–Hellman key exchange, and elliptic curve cryptography all rely on group-theoretic hardness assumptions.

1.3 Historical Remarks

Key figures include Galois (groups and solvability, 1830s), Cauchy and Lagrange (early group theory, late 1700s–early 1800s), Dedekind and Kronecker (rings and ideals, 1870s–80s), Noether (abstract ring theory and the isomorphism theorems, 1920s), and Artin (Galois theory reformulation, 1940s).

2. Groups

2.1 Definition

Definition. A group is a pair (G,)(G, \cdot) where GG is a set and :G×GG\cdot : G \times G \to G is a binary operation satisfying:

  1. Closure: For all a,bGa, b \in G, abGa \cdot b \in G.
  2. Associativity: For all a,b,cGa, b, c \in G, (ab)c=a(bc)(a \cdot b) \cdot c = a \cdot (b \cdot c).
  3. Identity: There exists eGe \in G such that for all aGa \in G, ea=ae=ae \cdot a = a \cdot e = a.
  4. Inverses: For each aGa \in G, there exists a1Ga^{-1} \in G such that aa1=a1a=ea \cdot a^{-1} = a^{-1} \cdot a = e.

A group is abelian (or commutative) if ab=baa \cdot b = b \cdot a for all a,bGa, b \in G.

The order of a group GG, denoted G|G|, is the number of elements in GG. The order of an element gGg \in G is the smallest positive integer nn such that gn=eg^n = e; if no such nn exists, gg has infinite order.

2.2 Examples of Groups

Example 2.1 (Integers under addition). (Z,+)(\mathbb{Z}, +) is an infinite abelian group. Identity: 00. Inverse of nn: n-n.

Example 2.2 (Symmetric group). The symmetric group SnS_n is the group of all permutations of {1,2,,n}\{1, 2, \ldots, n\} under composition. Sn=n!|S_n| = n!. For n3n \geq 3, SnS_n is non-abelian.

Example 2.3 (Dihedral group). The dihedral group DnD_n is the symmetry group of a regular nn-gon, consisting of nn rotations and nn reflections. Dn=2n|D_n| = 2n. For n3n \geq 3, DnD_n is non-abelian.

Example 2.4 (Cyclic group). A cyclic group of order nn is Z/nZ={0,1,2,,n1}\mathbb{Z}/n\mathbb{Z} = \{0, 1, 2, \ldots, n{-}1\} under addition modulo nn. Every cyclic group is abelian. Up to isomorphism, there is exactly one cyclic group of each finite order nn and one infinite cyclic group Z\mathbb{Z}.

Example 2.5 (General linear group). The general linear group GL(n,R)\mathrm{GL}(n, \mathbb{R}) is the group of all n×nn \times n invertible real matrices under matrix multiplication. It is non-abelian for n2n \geq 2.

Example 2.6 (Unit group). For n2n \geq 2, the unit group (Z/nZ)×(\mathbb{Z}/n\mathbb{Z})^\times consists of the elements of Z/nZ\mathbb{Z}/n\mathbb{Z} coprime to nn, under multiplication modulo nn. Its order is ϕ(n)\phi(n) (Euler’s totient function).

2.3 Subgroups

Definition. A subgroup of GG is a subset HGH \subseteq G that is itself a group under the operation restricted to HH.

Proposition. HGH \subseteq G is a subgroup if and only if:

  1. eHe \in H.
  2. For all a,bHa, b \in H, ab1Ha \cdot b^{-1} \in H.

Example 2.7. The alternating group AnA_n (even permutations) is a subgroup of SnS_n with An=n!/2|A_n| = n!/2.

2.4 Lagrange’s Theorem

Theorem 2.1 (Lagrange). If HH is a subgroup of a finite group GG, then H|H| divides G|G|.

Proof sketch. The left cosets aH={ah:hH}aH = \{ah : h \in H\} partition GG into disjoint subsets of equal size H|H|. Therefore G=[G:H]H|G| = [G:H] \cdot |H|, where [G:H]=G/H[G:H] = |G|/|H| is the index of HH in GG.

Corollary. The order of any element gGg \in G divides G|G|.

2.5 Cosets and Normal Subgroups

Definition. For a subgroup HGH \leq G and aGa \in G, the left coset of HH by aa is aH={ah:hH}aH = \{ah : h \in H\}. The right coset is Ha={ha:hH}Ha = \{ha : h \in H\}.

Definition. A subgroup HGH \leq G is normal (written HGH \triangleleft G) if aH=HaaH = Ha for all aGa \in G. Equivalently, aha1Haha^{-1} \in H for all aGa \in G, hHh \in H.

When HGH \triangleleft G, the set of cosets G/HG/H forms a group under the operation (aH)(bH)=(ab)H(aH)(bH) = (ab)H. This is the quotient group.

Example 2.8. AnSnA_n \triangleleft S_n for all n2n \geq 2, and Sn/AnZ/2ZS_n / A_n \cong \mathbb{Z}/2\mathbb{Z}.

Example 2.10. The centre Z(G)={zG:zg=gz for all gG}Z(G) = \{z \in G : zg = gz \text{ for all } g \in G\} is a normal subgroup of GG.

2.6 Conjugacy Classes and the Class Equation

Definition. Two elements a,bGa, b \in G are conjugate if there exists gGg \in G such that b=gag1b = gag^{-1}. Conjugacy is an equivalence relation; its equivalence classes are the conjugacy classes.

Proposition. In SnS_n, conjugacy classes are determined by cycle type. Two permutations are conjugate if and only if they have the same cycle structure.

Theorem 2.2 (Class equation). For a finite group GG:

G=Z(G)+i=1r[G:CG(gi)]|G| = |Z(G)| + \sum_{i=1}^{r} [G : C_G(g_i)]

where g1,,grg_1, \ldots, g_r are representatives of the non-central conjugacy classes, and CG(g)={xG:xg=gx}C_G(g) = \{x \in G : xg = gx\} is the centraliser of gg.

Corollary. Every finite pp-group has a non-trivial centre. (Proof: the terms [G:CG(gi)][G : C_G(g_i)] are powers of pp greater than 1, so they are divisible by pp; since G|G| is a power of pp, Z(G)|Z(G)| must also be divisible by pp.)

2.7 Group Actions

Definition. A group action of GG on a set XX is a map G×XXG \times X \to X, written (g,x)gx(g, x) \mapsto g \cdot x, satisfying ex=xe \cdot x = x and (gh)x=g(hx)(gh) \cdot x = g \cdot (h \cdot x).

Theorem 2.3 (Orbit-Stabiliser). For xXx \in X:

G=Orb(x)Stab(x)|G| = |\operatorname{Orb}(x)| \cdot |\operatorname{Stab}(x)|

where Orb(x)={gx:gG}\operatorname{Orb}(x) = \{g \cdot x : g \in G\} is the orbit and Stab(x)={gG:gx=x}\operatorname{Stab}(x) = \{g \in G : g \cdot x = x\} is the stabiliser.

This generalises both Lagrange’s theorem (acting on cosets) and the class equation (acting on itself by conjugation).

3. Group Homomorphisms

3.1 Definition and Basic Properties

Definition. A group homomorphism is a map ϕ:GH\phi : G \to H between groups such that ϕ(ab)=ϕ(a)ϕ(b)\phi(ab) = \phi(a)\phi(b) for all a,bGa, b \in G.

The kernel of ϕ\phi is ker(ϕ)={gG:ϕ(g)=eH}\ker(\phi) = \{g \in G : \phi(g) = e_H\}.

The image of ϕ\phi is im(ϕ)={ϕ(g):gG}\operatorname{im}(\phi) = \{\phi(g) : g \in G\}.

Proposition. ker(ϕ)\ker(\phi) is a normal subgroup of GG, and im(ϕ)\operatorname{im}(\phi) is a subgroup of HH.

A homomorphism is:

  • Injective (monomorphism) if and only if ker(ϕ)={eG}\ker(\phi) = \{e_G\}.
  • Surjective (epimorphism) if and only if im(ϕ)=H\operatorname{im}(\phi) = H.
  • Bijective (isomorphism) if both.

3.2 The First Isomorphism Theorem

Theorem 3.1 (First Isomorphism Theorem). If ϕ:GH\phi : G \to H is a group homomorphism, then

G/ker(ϕ)im(ϕ).G/\ker(\phi) \cong \operatorname{im}(\phi).

This is the fundamental link between homomorphisms and quotient groups. Every normal subgroup NGN \triangleleft G arises as the kernel of the natural projection π:GG/N\pi : G \to G/N.

Example 3.1. The determinant det:GL(n,R)R×\det : \mathrm{GL}(n, \mathbb{R}) \to \mathbb{R}^\times is a homomorphism with ker(det)=SL(n,R)\ker(\det) = \mathrm{SL}(n, \mathbb{R}) (matrices with determinant 1). Hence GL(n,R)/SL(n,R)R×\mathrm{GL}(n,\mathbb{R})/\mathrm{SL}(n,\mathbb{R}) \cong \mathbb{R}^\times.

Example 3.2. The sign map sgn:Sn{±1}\operatorname{sgn} : S_n \to \{\pm 1\} is a homomorphism with ker(sgn)=An\ker(\operatorname{sgn}) = A_n. Hence Sn/AnZ/2ZS_n/A_n \cong \mathbb{Z}/2\mathbb{Z}.

Example 3.3. The quotient map ZZ/nZ\mathbb{Z} \to \mathbb{Z}/n\mathbb{Z} sending aa(modn)a \mapsto a \pmod{n} is a surjective homomorphism with kernel (n)=nZ(n) = n\mathbb{Z}.

3.3 Second and Third Isomorphism Theorems

Theorem 3.2 (Second Isomorphism Theorem). If NGN \triangleleft G and KGK \leq G, then K/(NK)NK/NK/(N \cap K) \cong NK/N.

Theorem 3.3 (Third Isomorphism Theorem). If NGN \triangleleft G and MGM \triangleleft G with NMN \subseteq M, then (G/N)/(M/N)G/M(G/N)/(M/N) \cong G/M.

3.4 Cayley’s Theorem

Theorem 3.4 (Cayley). Every group GG is isomorphic to a subgroup of some symmetric group. Specifically, GG embeds into SGS_{|G|} via the left regular representation gρgg \mapsto \rho_g where ρg(x)=gx\rho_g(x) = gx.

This shows that the symmetric groups are, in a precise sense, universal — every group is a permutation group.

3.5 Direct Products and Semidirect Products

Definition. The direct product G×HG \times H of groups GG and HH is the set of ordered pairs (g,h)(g, h) with componentwise operation (g1,h1)(g2,h2)=(g1g2,h1h2)(g_1, h_1)(g_2, h_2) = (g_1 g_2, h_1 h_2).

If G,HG, H are finite, then G×H=GH|G \times H| = |G| \cdot |H|. Both GG and HH embed as normal subgroups of G×HG \times H.

Definition. A semidirect product NϕHN \rtimes_\phi H generalises the direct product. Given a homomorphism ϕ:HAut(N)\phi : H \to \operatorname{Aut}(N), the semidirect product has elements (n,h)(n, h) with multiplication (n1,h1)(n2,h2)=(n1ϕ(h1)(n2),h1h2)(n_1, h_1)(n_2, h_2) = (n_1 \phi(h_1)(n_2), h_1 h_2).

NN is normal in NHN \rtimes H, but HH need not be. Every extension 1NGH11 \to N \to G \to H \to 1 with a complement is a semidirect product.

Example 3.4. DnZ/nZZ/2ZD_n \cong \mathbb{Z}/n\mathbb{Z} \rtimes \mathbb{Z}/2\mathbb{Z}. The action of Z/2Z\mathbb{Z}/2\mathbb{Z} on Z/nZ\mathbb{Z}/n\mathbb{Z} is inversion: kkk \mapsto -k.

Example 3.5. Q8Q_8 is not a semidirect product of smaller groups — it is a non-split extension of Z/2Z\mathbb{Z}/2\mathbb{Z} by Z/2Z×Z/2Z\mathbb{Z}/2\mathbb{Z} \times \mathbb{Z}/2\mathbb{Z}.

4. Sylow Theorems

4.1 Statement

Definition. A pp-group is a group whose order is a power of a prime pp. A Sylow pp-subgroup of GG is a subgroup of order pkp^k where pkp^k divides G|G| but pk+1p^{k+1} does not.

Theorem 4.1 (First Sylow Theorem). If pkGp^k \mid |G|, then GG has a subgroup of order pkp^k. In particular, Sylow pp-subgroups exist.

Theorem 4.2 (Second Sylow Theorem). All Sylow pp-subgroups of GG are conjugate. If PP and QQ are Sylow pp-subgroups, there exists gGg \in G such that P=gQg1P = gQg^{-1}.

Theorem 4.3 (Third Sylow Theorem). The number npn_p of Sylow pp-subgroups satisfies:

  1. np1(modp)n_p \equiv 1 \pmod{p}.
  2. npn_p divides G/pk|G|/p^k.

4.2 Applications: Classifying Groups of Small Order

Example 4.1 (Groups of order pqpq). Let G=pq|G| = pq where p<qp < q are primes. The number of Sylow qq-subgroups satisfies nq1(modq)n_q \equiv 1 \pmod{q} and nqpn_q \mid p. Since p<qp < q, we must have nq=1n_q = 1, so the Sylow qq-subgroup QQ is normal.

The number of Sylow pp-subgroups satisfies np1(modp)n_p \equiv 1 \pmod{p} and npqn_p \mid q. If p(q1)p \nmid (q - 1), then np=1n_p = 1 and both Sylow subgroups are normal, so GZ/pqZG \cong \mathbb{Z}/pq\mathbb{Z}. If p(q1)p \mid (q - 1), there are two groups: Z/pqZ\mathbb{Z}/pq\mathbb{Z} and Z/qZZ/pZ\mathbb{Z}/q\mathbb{Z} \rtimes \mathbb{Z}/p\mathbb{Z}.

Example 4.2 (Groups of order p2p^2). Every group of order p2p^2 is abelian. Proof: if Z(G)=p|Z(G)| = p, then G/Z(G)G/Z(G) has order pp and is cyclic, which forces GG to be abelian — contradiction. So Z(G)=p2|Z(G)| = p^2 and G=Z(G)G = Z(G). Hence GG is isomorphic to either Z/p2Z\mathbb{Z}/p^2\mathbb{Z} or Z/pZ×Z/pZ\mathbb{Z}/p\mathbb{Z} \times \mathbb{Z}/p\mathbb{Z}.

Example 4.3 (Groups of order 8). Up to isomorphism, the five groups of order 8 are: Z/8Z\mathbb{Z}/8\mathbb{Z}, Z/4Z×Z/2Z\mathbb{Z}/4\mathbb{Z} \times \mathbb{Z}/2\mathbb{Z}, (Z/2Z)3(\mathbb{Z}/2\mathbb{Z})^3, D4D_4, and the quaternion group Q8Q_8.

The first three are abelian. D4D_4 and Q8Q_8 are both non-abelian with n2=1n_2 = 1, and can be distinguished by the number of elements of order 2: D4D_4 has five, Q8Q_8 has one.

Example 4.4 (Simple group of order 60). A5A_5 is the smallest non-abelian simple group. Using Sylow theory on A5=60=2235|A_5| = 60 = 2^2 \cdot 3 \cdot 5: the third Sylow theorem gives n51(mod5)n_5 \equiv 1 \pmod{5} with n512n_5 \mid 12, so n5=6n_5 = 6. The action of A5A_5 by conjugation on the six Sylow 5-subgroups gives a homomorphism A5S6A_5 \to S_6. Since A5A_5 is simple, this homomorphism is injective, and the analysis shows no proper normal subgroups exist.

5. Rings and Fields

5.1 Definitions

Definition. A ring is a triple (R,+,)(R, +, \cdot) where RR is a set and ++, \cdot are binary operations satisfying:

  1. (R,+)(R, +) is an abelian group.
  2. Multiplication is associative.
  3. The distributive laws hold: a(b+c)=ab+aca(b + c) = ab + ac and (a+b)c=ac+bc(a + b)c = ac + bc for all a,b,cRa, b, c \in R.

A ring is commutative if ab=baab = ba for all a,ba, b. A ring is a ring with unity if it has a multiplicative identity 1R0R1_R \neq 0_R.

Definition. An integral domain is a commutative ring with unity in which ab=0ab = 0 implies a=0a = 0 or b=0b = 0 (no zero divisors).

Definition. A field is a commutative ring with unity in which every nonzero element has a multiplicative inverse. Equivalently, a field is an integral domain in which every nonzero element is a unit.

5.2 Examples

Example 5.1. Z\mathbb{Z} is an integral domain but not a field. Q\mathbb{Q}, R\mathbb{R}, and C\mathbb{C} are fields.

Example 5.2. Z/nZ\mathbb{Z}/n\mathbb{Z} is a field if and only if nn is prime. When n=pn = p is prime, we write Fp\mathbb{F}_p.

Example 5.3. The ring Mn(R)M_n(\mathbb{R}) of n×nn \times n real matrices is non-commutative and has zero divisors for n2n \geq 2.

5.3 Polynomial Rings

Definition. If RR is a ring, the polynomial ring R[x]R[x] consists of formal sums f(x)=k=0nakxkf(x) = \sum_{k=0}^{n} a_k x^k with akRa_k \in R, under the usual addition and multiplication of polynomials.

Theorem 5.1. If RR is an integral domain, then R[x]R[x] is an integral domain. The units of R[x]R[x] are precisely the units of RR.

Theorem 5.2 (Division algorithm). If FF is a field and f,gF[x]f, g \in F[x] with g0g \neq 0, there exist unique q,rF[x]q, r \in F[x] with deg(r)<deg(g)\deg(r) < \deg(g) such that f=qg+rf = qg + r.

5.4 Ideals and Quotient Rings

Definition. An ideal of a ring RR is a subset IRI \subseteq R such that:

  1. (I,+)(I, +) is a subgroup of (R,+)(R, +).
  2. For all rRr \in R and aIa \in I: raIra \in I and arIar \in I.

Definition. An ideal II is prime if abIab \in I implies aIa \in I or bIb \in I. An ideal IRI \neq R is maximal if there is no ideal JJ with IJRI \subsetneq J \subsetneq R.

Proposition. Every maximal ideal is prime. In a commutative ring with unity, R/IR/I is a field if and only if II is maximal, and R/IR/I is an integral domain if and only if II is prime.

When II is an ideal, the quotient R/IR/I is a ring under (a+I)+(b+I)=(a+b)+I(a + I) + (b + I) = (a + b) + I and (a+I)(b+I)=ab+I(a + I)(b + I) = ab + I.

Example 5.4. Z/nZ\mathbb{Z}/n\mathbb{Z} is the quotient ring Z/(n)\mathbb{Z}/(n) where (n)=nZ(n) = n\mathbb{Z} is the ideal generated by nn. The ideal (p)(p) is maximal (hence prime) in Z\mathbb{Z} when pp is prime.

Example 5.5. For a field FF and irreducible f(x)F[x]f(x) \in F[x], the ideal (f)(f) is maximal, so F[x]/(f)F[x]/(f) is a field.

5.5 Unique Factorisation Domains and Principal Ideal Domains

Definition. A principal ideal domain (PID) is an integral domain in which every ideal is principal (generated by a single element).

Definition. A unique factorisation domain (UFD) is an integral domain in which every nonzero non-unit element factors uniquely (up to order and units) into irreducibles.

Example 5.6. Z\mathbb{Z} is a PID (every ideal is (n)(n) for some n0n \geq 0) and hence a UFD. Q[x]\mathbb{Q}[x] is a PID; more generally, F[x]F[x] is a PID for any field FF.

Example 5.7. Z[x]\mathbb{Z}[x] is a UFD but not a PID. The ideal (2,x)Z[x](2, x) \subset \mathbb{Z}[x] is not principal.

Example 5.8. Z[5]\mathbb{Z}[\sqrt{-5}] is not a UFD: 6=23=(1+5)(15)6 = 2 \cdot 3 = (1 + \sqrt{-5})(1 - \sqrt{-5}) gives two genuinely different factorisations into irreducibles.

Theorem 5.3. Every PID is a UFD. The converse fails (Z[x]\mathbb{Z}[x] is a UFD but not a PID).

Theorem 5.4 (Euclidean algorithm). Every Euclidean domain (an integral domain with a division algorithm) is a PID. Examples include Z\mathbb{Z} and F[x]F[x] for any field FF.

5.5 Field Extensions

Definition. A field extension K/FK/F is an inclusion of fields FKF \subseteq K. The degree [K:F][K : F] is the dimension of KK as a vector space over FF.

Theorem 5.3 (Tower law). If FEKF \subseteq E \subseteq K are field extensions, then [K:F]=[K:E][E:F][K : F] = [K : E][E : F].

Definition. A field FF is algebraically closed if every non-constant polynomial in F[x]F[x] has a root in FF. The algebraic closure F\overline{F} of FF is the smallest algebraically closed field containing FF.

Example 5.6. C\mathbb{C} is the algebraic closure of R\mathbb{R}, and [C:R]=2[\mathbb{C} : \mathbb{R}] = 2.

Example 5.7. The splitting field of x32x^3 - 2 over Q\mathbb{Q} is Q(23,ω)\mathbb{Q}(\sqrt[3]{2}, \omega) where ω=e2πi/3\omega = e^{2\pi i/3}, with degree [Q(23,ω):Q]=6[\mathbb{Q}(\sqrt[3]{2}, \omega) : \mathbb{Q}] = 6.

5.6 Finite Fields

Theorem 5.4. For every prime pp and integer n1n \geq 1, there exists a unique (up to isomorphism) finite field of order pnp^n, denoted Fpn\mathbb{F}_{p^n} or GF(pn)\mathrm{GF}(p^n).

The finite field Fpn\mathbb{F}_{p^n} is the splitting field of xpnxx^{p^n} - x over Fp\mathbb{F}_p. Its multiplicative group Fpn×\mathbb{F}_{p^n}^\times is cyclic of order pn1p^n - 1.

Theorem 5.5. Fpm\mathbb{F}_{p^m} is a subfield of Fpn\mathbb{F}_{p^n} if and only if mnm \mid n. In that case, [Fpn:Fpm]=n/m[\mathbb{F}_{p^n} : \mathbb{F}_{p^m}] = n/m.

Example 5.8. F4=F2[x]/(x2+x+1)={0,1,α,α+1}\mathbb{F}_4 = \mathbb{F}_2[x]/(x^2 + x + 1) = \{0, 1, \alpha, \alpha{+}1\} where α2=α+1\alpha^2 = \alpha + 1. The multiplicative group F4×={1,α,α+1}\mathbb{F}_4^\times = \{1, \alpha, \alpha{+}1\} is cyclic of order 3, generated by α\alpha (since α2=α+1\alpha^2 = \alpha + 1 and α3=1\alpha^3 = 1).

Example 5.9. Fpn×\mathbb{F}_{p^n}^\times is cyclic for all p,np, n. This generator is called a primitive element. Its existence follows from the classification of finite abelian groups: every finite subgroup of the multiplicative group of a field is cyclic.

6. Galois Theory

6.1 Field Automorphisms

Definition. A field automorphism of KK is an isomorphism σ:KK\sigma : K \to K of fields. The set Aut(K)\operatorname{Aut}(K) of all automorphisms of KK forms a group under composition.

For a field extension K/FK/F, define

Aut(K/F)={σAut(K):σ(a)=a for all aF}.\operatorname{Aut}(K/F) = \{\sigma \in \operatorname{Aut}(K) : \sigma(a) = a \text{ for all } a \in F\}.

6.2 Galois Extensions

Definition. A finite field extension K/FK/F is Galois if Aut(K/F)=[K:F]|\operatorname{Aut}(K/F)| = [K : F].

Equivalently, K/FK/F is Galois if KK is the splitting field of a separable polynomial over FF.

Definition. For a Galois extension K/FK/F, the Galois group is Gal(K/F)=Aut(K/F)\operatorname{Gal}(K/F) = \operatorname{Aut}(K/F).

Example 6.1. The extension Q(2)/Q\mathbb{Q}(\sqrt{2})/\mathbb{Q} is Galois with Gal(Q(2)/Q)Z/2Z\operatorname{Gal}(\mathbb{Q}(\sqrt{2})/\mathbb{Q}) \cong \mathbb{Z}/2\mathbb{Z}, generated by σ:a+b2ab2\sigma : a + b\sqrt{2} \mapsto a - b\sqrt{2}.

Example 6.2. The extension Q(2,3)/Q\mathbb{Q}(\sqrt{2}, \sqrt{3})/\mathbb{Q} has Galois group GalZ/2Z×Z/2Z\operatorname{Gal} \cong \mathbb{Z}/2\mathbb{Z} \times \mathbb{Z}/2\mathbb{Z}, with four automorphisms determined by the signs on 2\sqrt{2} and 3\sqrt{3}.

Example 6.3. The splitting field of x32x^3 - 2 over Q\mathbb{Q} is Q(23,ω)\mathbb{Q}(\sqrt[3]{2}, \omega) where ω=e2πi/3\omega = e^{2\pi i/3}, with degree [Q(23,ω):Q]=6[\mathbb{Q}(\sqrt[3]{2}, \omega) : \mathbb{Q}] = 6.

Gal(Q(23,ω)/Q)S3\operatorname{Gal}(\mathbb{Q}(\sqrt[3]{2}, \omega)/\mathbb{Q}) \cong S_3

The six automorphisms permute the three roots 23,ω23,ω223\sqrt[3]{2}, \omega\sqrt[3]{2}, \omega^2\sqrt[3]{2}. The intermediate fields are:

QQ(ω)Q(23,ω),QQ(23)Q(23,ω),QQ(ω23)Q(23,ω)\mathbb{Q} \subset \mathbb{Q}(\omega) \subset \mathbb{Q}(\sqrt[3]{2}, \omega), \qquad \mathbb{Q} \subset \mathbb{Q}(\sqrt[3]{2}) \subset \mathbb{Q}(\sqrt[3]{2}, \omega), \qquad \mathbb{Q} \subset \mathbb{Q}(\omega\sqrt[3]{2}) \subset \mathbb{Q}(\sqrt[3]{2}, \omega)

Only Q(ω)/Q\mathbb{Q}(\omega)/\mathbb{Q} is Galois among the proper intermediate extensions (corresponding to the unique normal subgroup A3S3A_3 \triangleleft S_3).

6.3 The Fundamental Theorem of Galois Theory

Theorem 6.1 (Fundamental Theorem of Galois Theory). Let K/FK/F be a Galois extension with Galois group G=Gal(K/F)G = \operatorname{Gal}(K/F). Then there is a one-to-one, order-reversing correspondence:

{Intermediate fields FEK}{Subgroups HG}\{\text{Intermediate fields } F \subseteq E \subseteq K\} \longleftrightarrow \{\text{Subgroups } H \subseteq G\}

given by EGal(K/E)E \mapsto \operatorname{Gal}(K/E) and HKHH \mapsto K^H (the fixed field of HH), satisfying:

  1. [K:E]=H[K : E] = |H| and [E:F]=[G:H][E : F] = [G : H].
  2. E/FE/F is Galois if and only if HGH \triangleleft G, in which case Gal(E/F)G/H\operatorname{Gal}(E/F) \cong G/H.

6.4 Solvability by Radicals

Definition. A polynomial fF[x]f \in F[x] is solvable by radicals if its roots can be expressed using only the operations ++, -, ×\times, ÷\div, and nn-th roots.

A group GG is solvable if there exists a chain of subgroups

G=G0G1Gn={e}G = G_0 \trianglerighteq G_1 \trianglerighteq \cdots \trianglerighteq G_n = \{e\}

where each Gi+1G_{i+1} is normal in GiG_i and each quotient Gi/Gi+1G_i/G_{i+1} is cyclic of prime order.

Theorem 6.2. A polynomial fQ[x]f \in \mathbb{Q}[x] is solvable by radicals if and only if Gal(f)\operatorname{Gal}(f) is a solvable group.

Theorem 6.3 (Abel–Ruffini). The general polynomial of degree n5n \geq 5 is not solvable by radicals. Specifically, the Galois group of x54x+2x^5 - 4x + 2 over Q\mathbb{Q} is S5S_5, and since A5A_5 is simple and non-abelian, S5S_5 is not solvable.

6.5 Cyclotomic Extensions

The cyclotomic field Q(ζn)\mathbb{Q}(\zeta_n), where ζn=e2πi/n\zeta_n = e^{2\pi i/n} is a primitive nn-th root of unity, is the splitting field of xn1x^n - 1 over Q\mathbb{Q}.

Theorem 6.4. The Galois group Gal(Q(ζn)/Q)\operatorname{Gal}(\mathbb{Q}(\zeta_n)/\mathbb{Q}) is isomorphic to (Z/nZ)×(\mathbb{Z}/n\mathbb{Z})^\times — the unit group of Z/nZ\mathbb{Z}/n\mathbb{Z}.

Example 6.4. For n=5n = 5: Gal(Q(ζ5)/Q)(Z/5Z)×Z/4Z\operatorname{Gal}(\mathbb{Q}(\zeta_5)/\mathbb{Q}) \cong (\mathbb{Z}/5\mathbb{Z})^\times \cong \mathbb{Z}/4\mathbb{Z}. The degree is ϕ(5)=4\phi(5) = 4.

Example 6.5. For n=8n = 8: Gal(Q(ζ8)/Q)(Z/8Z)×Z/2Z×Z/2Z\operatorname{Gal}(\mathbb{Q}(\zeta_8)/\mathbb{Q}) \cong (\mathbb{Z}/8\mathbb{Z})^\times \cong \mathbb{Z}/2\mathbb{Z} \times \mathbb{Z}/2\mathbb{Z}. The degree is ϕ(8)=4\phi(8) = 4.

6.6 Constructibility with Straightedge and Compass

A real number α\alpha is constructible if it can be obtained from rational numbers using only straightedge and compass operations (addition, subtraction, multiplication, division, and square roots).

Theorem 6.5. α\alpha is constructible if and only if [Q(α):Q][\mathbb{Q}(\alpha) : \mathbb{Q}] is a power of 2.

Corollary (Classical impossibilities).

  • Doubling the cube requires 23\sqrt[3]{2}, with [Q(23):Q]=3[\mathbb{Q}(\sqrt[3]{2}) : \mathbb{Q}] = 3 — not a power of 2. Impossible.
  • Trisecting a general angle reduces to solving a cubic equation whose Galois group has order 3 — not a power of 2. Impossible.
  • Squaring the circle requires constructing π\pi, which is transcendental over Q\mathbb{Q} (Lindemann, 1882). Impossible.

7. Applications

7.1 RSA Cryptography

RSA relies on the difficulty of factoring large integers, but its correctness proof uses elementary group theory on (Z/nZ)×(\mathbb{Z}/n\mathbb{Z})^\times.

Key generation: Choose distinct primes p,qp, q; set n=pqn = pq. The public key is (n,e)(n, e) where gcd(e,ϕ(n))=1\gcd(e, \phi(n)) = 1. The private key is d=e1(modϕ(n))d = e^{-1} \pmod{\phi(n)}.

Encryption: c=me(modn)c = m^e \pmod{n} for message mm.

Decryption: m=cd(modn)m = c^d \pmod{n}.

Correctness. Since ed1(modϕ(n))ed \equiv 1 \pmod{\phi(n)}, we have cd=medm(modn)c^d = m^{ed} \equiv m \pmod{n} by Euler’s theorem (a consequence of Lagrange’s theorem applied to (Z/nZ)×(\mathbb{Z}/n\mathbb{Z})^\times).

7.2 Error-Correcting Codes

Finite fields provide the algebraic foundation for error-correcting codes. A linear code of length nn over Fq\mathbb{F}_q is a subspace of Fqn\mathbb{F}_q^n.

Example (Reed–Solomon codes). A Reed–Solomon code of dimension kk and length nn over Fpm\mathbb{F}_{p^m} encodes a message as the evaluations of a polynomial of degree <k< k at nn distinct field elements. It can correct up to (nk)/2(n - k)/2 errors.

Example (BCH codes). Bose–Chaudhuri–Hocquenghem codes generalise Reed–Solomon codes to arbitrary finite fields and provide flexible trade-offs between code rate and error correction capability.

7.3 Crystallography and Group Actions

The symmetry group of a crystal (a space group) classifies crystalline structures. The mathematical framework is that of group actions: a group GG acts on a set XX if there is a map G×XXG \times X \to X satisfying g1(g2x)=(g1g2)xg_1(g_2 x) = (g_1 g_2)x and ex=xex = x.

There are exactly 230 crystallographic space groups in 3D, 17 wallpaper groups in 2D, and 7 frieze groups in 1D — all classified by group-theoretic methods.

7.4 Fermat’s Little Theorem and Applications

Theorem (Fermat’s little theorem). If pp is prime and gcd(a,p)=1\gcd(a, p) = 1, then ap11(modp)a^{p-1} \equiv 1 \pmod{p}.

This is an immediate consequence of Lagrange’s theorem: the order of aa in (Z/pZ)×(\mathbb{Z}/p\mathbb{Z})^\times divides (Z/pZ)×=p1|(\mathbb{Z}/p\mathbb{Z})^\times| = p - 1.

Euler’s theorem generalises this: if gcd(a,n)=1\gcd(a, n) = 1, then aϕ(n)1(modn)a^{\phi(n)} \equiv 1 \pmod{n}.

Application (primality testing). The Fermat test checks whether an11(modn)a^{n-1} \equiv 1 \pmod{n} for several bases aa. If nn fails for any aa, it is composite. If nn passes for many aa, it is likely prime (though Carmichael numbers fool this test). More robust tests like Miller–Rabin use the structure of (Z/nZ)×(\mathbb{Z}/n\mathbb{Z})^\times more carefully.

7.5 Diffie–Hellman Key Exchange

The Diffie–Hellman protocol allows two parties to establish a shared secret over a public channel using cyclic groups.

Setup: Fix a cyclic group GG of prime order qq with generator gg (e.g., G=Fp×G = \mathbb{F}_p^\times).

Protocol:

  1. Alice picks random a{1,,q1}a \in \{1, \ldots, q-1\}, sends gag^a to Bob.
  2. Bob picks random b{1,,q1}b \in \{1, \ldots, q-1\}, sends gbg^b to Alice.
  3. Both compute (ga)b=(gb)a=gab(g^a)^b = (g^b)^a = g^{ab} as the shared secret.

Security relies on the discrete logarithm problem: given gg and gag^a, finding aa is computationally infeasible in sufficiently large groups.

8. Common Pitfalls

  1. “Every subgroup is normal.” False. In S3S_3, the subgroup H={e,(12)}H = \{e, (12)\} has (13)H(13)1={e,(23)}H(13)H(13)^{-1} = \{e, (23)\} \neq H. Normality is a special property, not automatic.

  2. “The converse of Lagrange’s theorem holds.” False. A4A_4 has order 12 but no subgroup of order 6. The converse holds for Sylow subgroups but not as a general principle.

  3. “Abelian implies cyclic.” False. Z/2Z×Z/2Z\mathbb{Z}/2\mathbb{Z} \times \mathbb{Z}/2\mathbb{Z} is abelian but not cyclic — every non-identity element has order 2.

  4. “Every field extension is Galois.” False. Q(23)/Q\mathbb{Q}(\sqrt[3]{2})/\mathbb{Q} has degree 3 but only the identity automorphism (the other cube roots of 2 are complex and not in this field), so it is not Galois.

  5. SnS_n is solvable for all nn.” False. SnS_n is solvable only for n4n \leq 4. For n5n \geq 5, AnA_n is simple and non-abelian, making SnS_n non-solvable.

  6. “The kernel of a ring homomorphism is a subring, not an ideal.” The kernel is both an ideal and a subring. The key point is that kernels of ring homomorphisms are always ideals, and kernels of field homomorphisms are always trivial (fields have no nontrivial proper ideals).

  7. Fpn\mathbb{F}_{p^n} is the same as Z/pnZ\mathbb{Z}/p^n\mathbb{Z}.” False. Z/pnZ\mathbb{Z}/p^n\mathbb{Z} is not a field for n>1n > 1 (it has zero divisors). Fpn\mathbb{F}_{p^n} is the splitting field of xpnxx^{p^n} - x over Fp\mathbb{F}_p — an entirely different structure.

9. Summary

| Concept | Key Idea | | ------------------------------- | ------------------------------------------------------------------------------------------ | --- | ----------- | --------------------- | ----------------------------------------------------------------- | ---------------------- | --- | | Group | Set + associative binary operation with identity and inverses | | Abelian group | Group with commutative operation | | Subgroup | Subset closed under the group operation and inverses | | Lagrange’s theorem | H| H | divides G| G | for HGH \leq G | | Coset | aH={ah:hH}aH = \{ah : h \in H\}; partitions GG into equal-size subsets | | Normal subgroup | aha1Haha^{-1} \in H for all aG,hHa \in G, h \in H | | Quotient group | G/NG/N inherits group structure when NGN \triangleleft G | | Conjugacy class | Equivalence class under agag1a \sim gag^{-1} | | Class equation | G=Z(G)+[G:CG(gi)]| G | = | Z(G) | + \sum [G : C_G(g_i)]; proves pp-groups have non-trivial centre | | Group action | GG acts on XX with orbit-stabiliser G=Orb(x)Stab(x)| G | = | \operatorname{Orb}(x) | \cdot | \operatorname{Stab}(x) | | | Homomorphism | ϕ(ab)=ϕ(a)ϕ(b)\phi(ab) = \phi(a)\phi(b); kernel is normal, image is a subgroup | | First isomorphism theorem | G/ker(ϕ)im(ϕ)G/\ker(\phi) \cong \operatorname{im}(\phi) | | Cayley’s theorem | Every group embeds into a symmetric group | | Sylow theorems | Existence, conjugacy, and counting of pp-subgroups | | Ring | Set + addition (abelian group) + multiplication (associative, distributive) | | Integral domain | Commutative ring with unity and no zero divisors | | Field | Commutative ring with unity where every nonzero element is a unit | | Prime/maximal ideal | R/IR/I integral domain I\Leftrightarrow I prime; R/IR/I field I\Leftrightarrow I maximal | | PID | Integral domain where every ideal is principal; implies UFD | | UFD | Unique factorisation into irreducibles; Z[x]\mathbb{Z}[x] is UFD but not PID | | Polynomial ring R[x]R[x] | Formal polynomials; integral domain if RR is | | Field extension K/FK/F | KK is a vector space over FF; degree =[K:F]= [K:F] | | Tower law | [K:F]=[K:E][E:F][K:F] = [K:E][E:F] for FEKF \subseteq E \subseteq K | | Finite field Fpn\mathbb{F}_{p^n} | Unique field of order pnp^n; multiplicative group is cyclic | | Galois group | Automorphisms of KK fixing FF; bridges fields and groups | | Fundamental theorem | Lattice of intermediate fields \leftrightarrow lattice of subgroups | | Cyclotomic extension | Gal(Q(ζn)/Q)(Z/nZ)×\operatorname{Gal}(\mathbb{Q}(\zeta_n)/\mathbb{Q}) \cong (\mathbb{Z}/n\mathbb{Z})^\times | | Constructibility | α\alpha constructible \Leftrightarrow [Q(α):Q][\mathbb{Q}(\alpha):\mathbb{Q}] is a power of 2 | | Solvability | Polynomial solvable by radicals \Leftrightarrow Galois group is solvable | | Abel–Ruffini | General quintic (degree 5\geq 5) is not solvable by radicals | | Fermat’s little theorem | ap11(modp)a^{p-1} \equiv 1 \pmod{p}; consequence of Lagrange’s theorem |

Worked Examples

Example 1: Proving a Subgroup

Problem: Prove that the set of even permutations in S_n forms a subgroup (the alternating group A_n). Solution: A permutation is even if it can be expressed as an even number of transpositions. Identity: 0 transpositions (even). Closure: product of two even permutations has an even total number of transpositions. Inverse: reversing the sequence of transpositions preserves parity. Therefore A_n is a subgroup of S_n.

Example 2: Calculating the Order of an Element

Problem: Find the order of the element (1 2 3)(4 5) in S_5. Solution: The order of a permutation is the LCM of the cycle lengths. The cycles are (1 2 3) of length 3 and (4 5) of length 2. Order = lcm(3, 2) = 6. The element (1 2 3)(4 5)^6 = identity.

Cross-References

TopicLink
TopologyView