Introduction to Abstract Algebra: Groups and Symmetry

A group is a set equipped with an associative binary operation that has an identity element and inverses; groups are the mathematics of symmetry. The central result is that every symmetry of an object forms a group, and conversely every (finite) group can be realized as a group of permutations (Cayley's theorem). Groups appear wherever structure-preserving transformations occur: geometry, crystallography, particle physics, cryptography, and Rubik's cube. Foundational work by Galois (1830s), Cayley (1854), and Cayley/Klein/Lie in the late 19th century.

A group $(G, *)$ is a set $G$ together with a binary operation $*$ satisfying four axioms: closure, associativity, existence of an identity, and existence of inverses. Groups matter because they abstract the common skeleton shared by integer arithmetic, modular clocks, geometric rotations, and shuffles of a deck — capturing "symmetry" in its most precise form. The surprising fact is that despite the minimal axioms, finite groups have an extraordinarily rich and rigid structure: they can be completely classified into "simple" building blocks (the Classification of Finite Simple Groups), one of the longest proofs ever assembled by humanity. Symmetry groups encode the invariances of physical laws, and through Noether's Theorem every continuous symmetry yields a conservation law.

The Core Insight

A symmetry is anything you can do to an object that leaves it looking the same — and the crucial fact is that symmetries compose. Do one symmetry, then another, and the result is again a symmetry; every symmetry can be undone; and "doing nothing" is always a symmetry. These three facts — composition, inverses, identity — together with associativity are the group axioms. The abstract group is therefore the distilled essence of symmetry, stripped of any reference to the particular object being symmetric.

Formal Statement

Group

A group is an ordered pair $(G, *)$ where $G$ is a set and $* : G \times G \to G$ is a binary operation, satisfying:

\forall a, b \in G : a * b \in G

⟳This is the closure axiom — often folded into "

*

is a binary operation

G \times G \to G

"; it guarantees the output stays in

G

\forall a, b, c \in G : (a * b) * c = a * (b * c)

⟳Associativity; allows unambiguous writing of

a * b * c

without parentheses.

\exists e \in G \forall a \in G : e * a = a * e = a

⟳Identity element. The two-sided condition is required; later we prove

e

is unique.

\forall a \in G \exists a^{- 1} \in G : a * a^{- 1} = a^{- 1} * a = e

⟳Inverses. Each element has a two-sided inverse relative to the identity

e

from [3].

Symbols: $G$ is the underlying set; $*$ is the group operation (often written multiplicatively as $ab$ or additively as $a + b$ ); $e$ (sometimes $1$ or $0$ ) is the identity; $a^{- 1}$ (or $- a$ ) is the inverse of $a$ . The number of elements $∣ G ∣$ is the order of the group. If additionally $a * b = b * a$ for all $a, b$ , the group is abelian (commutative).

Subgroup

A subset $H \subseteq G$ is a subgroup, written $H \leq G$ , if $H$ is itself a group under the restricted operation. Equivalently (the subgroup test):

H \neq = \emptyset \land (\forall a, b \in H : a * b^{- 1} \in H) ⟹ H \leq G

⟳Nonemptiness gives some

a \in H

, so

a * a^{- 1} = e \in H

; then

e * b^{- 1} = b^{- 1} \in H

; then

a * (b^{- 1})^{- 1} = a * b \in H

gives closure; associativity is inherited.

Lagrange's Theorem

If $G$ is a finite group and $H \leq G$ , then the order of $H$ divides the order of $G$ :

∣ G ∣ = [G : H] \cdot ∣ H ∣

⟳

G

is partitioned into disjoint left cosets

g H

, each of the same cardinality

∣ H ∣

; the number of cosets is the index

[G : H]

Corollary. For any

a \in G

, the order of

a

(smallest

n > 0

with

a^{n} = e

) divides

∣ G ∣

, and hence

a^{∣ G ∣} = e

Cayley's Theorem

Every group $G$ is isomorphic to a subgroup of the symmetric group $Sym (G)$ of permutations of its underlying set. In particular, every finite group of order $n$ embeds in $S_{n}$ .

Why It Holds / Derivation

Proof of Lagrange's Theorem

Step 1 — Define cosets. For

g \in G

, the left coset is

g H = {g * h : h \in H}

. Step 2 — Cosets partition $G$ . Define

a \sim b ⟺ a^{- 1} b \in H

. Reflexivity:

a^{- 1} a = e \in H

. Symmetry: if

a^{- 1} b \in H

then

(a^{- 1} b)^{- 1} = b^{- 1} a \in H

. Transitivity: if

a^{- 1} b, b^{- 1} c \in H

then

(a^{- 1} b) (b^{- 1} c) = a^{- 1} c \in H

. Thus

\sim

is an equivalence relation, and its classes are exactly the cosets

g H

. Equivalence classes partition

G

. Step 3 — All cosets have size $∣ H ∣$ . The map

φ : H \to g H

φ (h) = g * h

, is a bijection: it is surjective by definition of

g H

, and injective because

g h_{1} = g h_{2} \Rightarrow h_{1} = h_{2}

by cancellation (left-multiply by

g^{- 1}

). Hence

∣ g H ∣ = ∣ H ∣

. Step 4 — Count. Since the

[G : H]

disjoint cosets each have

∣ H ∣

elements and together exhaust

G

∣ G ∣ = cosets \sum ∣ g H ∣ = [G : H] \cdot ∣ H ∣.

⟳Sum the equal-sized blocks of a partition; the number of blocks is by definition the index.

Therefore $∣ H ∣$ divides $∣ G ∣$ . $■$

Proof of Cayley's Theorem

Step 1 — Left-multiplication maps. For each

g \in G

define

L_{g} : G \to G

L_{g} (x) = g * x

. Each

L_{g}

is a bijection with inverse

L_{g^{- 1}}

, hence

L_{g} \in Sym (G)

. Step 2 — Homomorphism. Define

Φ : G \to Sym (G)

Φ (g) = L_{g}

. Then

L_{g} \circ L_{h} (x) = g * (h * x) = (g * h) * x = L_{g * h} (x),

⟳Apply associativity [2]; this shows

Φ (g) Φ (h) = Φ (g h)

, so

Φ

respects the operation.

so $Φ (g * h) = Φ (g) \circ Φ (h)$ .

Step 3 — Injectivity. If

Φ (g) = id

, then

g * x = x

for all

x

; taking

x = e

gives

g = e

. So

ker Φ = {e}

and

Φ

is injective. Step 4 — Conclude.

Φ

is an injective homomorphism, so

G ≅ Φ (G) \leq Sym (G)

■

Évariste Galois | 1811–1832

A French mathematician who, before dying in a duel at age 20, founded what we now call Galois theory — the use of groups (he coined the term groupe) to determine when a polynomial equation is solvable by radicals. His night-before-the-duel manuscript introduced normal subgroups, the solvability of groups, and the idea that the symmetries of the roots govern the algebra of equations. Galois's work, ignored at first and only published in 1846 by Liouville, is the historical seed from which all of abstract group theory grew. The non-solvability of $S_{5}$ explains why there is no quintic formula.

Building Intuition

Think of a square sitting on a table. Pick it up, rotate or flip it, and set it back down so it covers the same outline — you have performed a symmetry. There are exactly eight such moves: four rotations (by $0°, 90°, 180°, 270°$ ) and four reflections (across two diagonals and two midlines). These eight form the dihedral group $D_{4}$ .

The "operation" is do one move, then another. Rotate $90°$ then reflect — the net effect is some other single symmetry from the list of eight (closure). Doing nothing is the identity. Every move can be undone (inverses). And it doesn't matter how you group a sequence of three moves (associativity) — though notice it does matter what order you do them in: rotate-then-flip $\neq =$ flip-then-rotate. That failure of commutativity is why $D_{4}$ is non-abelian, and it is the source of all the interesting complexity. A group is precisely this bookkeeping of "moves that preserve structure," abstracted away from squares to anything at all.

The dihedral group $D_4$: the eight symmetries of a square, organized into rotations and reflections

Worked Examples

Integers mod 6 under addition

Given. Let

G = Z_{6} = {0, 1, 2, 3, 4, 5}

with operation

a * b = (a + b) mod 6

. Verify it is a group and find the order of the element

4

. Steps.

Closure: any sum mod 6 lands in ${0, \dots, 5}$ . ✓
Associativity: inherited from integer addition. ✓
Identity: $0$ , since $0 + a \equiv a$ . ✓
Inverses: $- a mod 6$ , e.g. $4^{- 1} = 2$ since $4 + 2 = 6 \equiv 0$ . ✓
Order of 4: compute multiples $4, 4 + 4 = 8 \equiv 2, 2 + 4 = 6 \equiv 0$ . So $3 \cdot 4 \equiv 0$ , smallest such positive multiple is $3$ .

Check with Lagrange's corollary: order of element $= 3$ divides $∣ G ∣ = 6$ . ✓

Answer.

(Z_{6}, +) is an abelian group of order 6; ord (4) = 3

Composing symmetries in $D_3$ (equilateral triangle)

Given. The symmetry group of an equilateral triangle,

D_{3}

, has order 6: rotations

{e, r, r^{2}}

0°, 120°, 240°

and three reflections

{s, r s, r^{2} s}

, with relations

r^{3} = e

s^{2} = e

sr = r^{- 1} s = r^{2} s

. Compute the product

(r s) (r^{2} s)

and identify it. Steps.

Write out: $(r s) (r^{2} s) = r (s r^{2}) s$ .
Push $s$ past $r^{2}$ using $sr = r^{- 1} s$ , so $s r^{2} = (sr) r = (r^{- 1} s) r = r^{- 1} (sr) = r^{- 1} r^{- 1} s = r^{- 2} s = r s$ (since $r^{- 2} = r$ ).
Substitute: $r (s r^{2}) s = r (r s) s = r^{2} s^{2} = r^{2} \cdot e = r^{2}$ .
Identify: $r^{2}$ is the rotation by $240°$ .

Answer.

(r s) (r^{2} s) = r^{2} (rotation by 24 0^{\circ})

Note this is a rotation: reflection

\times

reflection

=

rotation, mirroring the determinant rule

(- 1) \cdot (- 1) = + 1

Common Misconceptions

"All groups are commutative." False — only abelian groups are. The smallest non-abelian group is

S_{3} = D_{3}

of order 6. In general

ab \neq = ba

, and you may not rearrange factors. A related error: writing

(ab)^{- 1} = a^{- 1} b^{- 1}

. The correct socks-and-shoes rule reverses order:

(a * b)^{- 1} = b^{- 1} * a^{- 1} .

⟳Check

(ab) (b^{- 1} a^{- 1}) = a (b b^{- 1}) a^{- 1} = a e a^{- 1} = e

using associativity and inverses.

Also beware: closure is not automatic. The odd integers under addition are not a group — $3 + 5 = 8$ is even, leaving the set.

A subset closed under the operation need not be a subgroup. Consider

G = Z

under addition and the subset

H = {0, 1, 2, 3, \dots}

, the non-negative integers.

H

is closed (

a + b \geq 0

), contains the identity

0

, and inherits associativity — yet

H

is not a subgroup, because it lacks inverses:

3 \in H

but

- 3 \in / H

. This shows closure

+

identity is insufficient; the subgroup test [5] (using

a * b^{- 1}

) correctly catches this, since

0 - 3 = - 3 \in / H

Connections

Groups sit at the crossroads of nearly all of mathematics. A group with a second compatible operation (multiplication distributing over addition) is a Ring Theory object; further structure gives Field Theory. The symmetries of the roots of a polynomial form a Galois Group, linking groups to solving equations. Continuous symmetry groups — circles, rotation groups $S O (3)$ , the matrices $G L_{n}$ — are Lie Groups, where group theory meets Differential Geometry. The way a group can act on a set is formalized in Group Actions, generalizing Cayley's theorem and powering counting tools like Burnside's Lemma. The quotient construction $G / N$ by a normal subgroup connects to the Isomorphism Theorems and to Homological Algebra.

Real-World Applications

Crystallography. The 230 distinct space groups classify all possible repeating atomic arrangements in crystals — direct application of group theory to X-ray diffraction patterns and mineral structure.
Particle physics (Standard Model). The gauge group $S U (3) \times S U (2) \times U (1)$ organizes quarks, leptons, and force carriers; Gell-Mann's "Eightfold Way" used $S U (3)$ representations to predict the $Ω^{-}$ baryon.
Public-key cryptography. RSA relies on the multiplicative group $(Z / n Z)^{\times}$ and Euler's theorem ( $a^{φ (n)} \equiv 1$ ); elliptic-curve cryptography uses the group law on elliptic curves.
Error-correcting codes. Linear and cyclic codes (Hamming, Reed–Solomon, used in CDs, QR codes, deep-space transmission) are subgroups of $(F_{q}^{n}, +)$ exploiting group/field structure.
The Rubik's Cube. Its $\approx 4.3 \times 1 0^{19}$ configurations form a group; solving algorithms are products of generators, and commutators isolate small changes.
Chemistry / spectroscopy. Molecular point groups ( $C_{2 v}$ , $T_{d}$ , etc.) predict which vibrational modes are IR- or Raman-active and label molecular orbitals.
Music theory. The 12 pitch classes under transposition form $Z_{12}$ ; transposition and inversion together give a dihedral group $D_{12}$ used in neo-Riemannian analysis.

Historical Development

1770 — Lagrange studies permutations of polynomial roots to analyze solvability, anticipating Lagrange's theorem and the role of symmetry in equations.
1799 — Ruffini gives an (incomplete) argument that the general quintic is not solvable by radicals.
1824 — Abel proves rigorously that no general formula in radicals exists for degree $\geq 5$ (Abel–Ruffini theorem).
1830–1832 — Galois introduces groups, normal subgroups, and solvability, definitively explaining the quintic; his ideas lie unpublished until 1846.
1854 — Cayley gives the first abstract definition of a (finite) group via a multiplication table, decoupling groups from permutations.
1872 — Klein's Erlangen Program recasts all of geometry as the study of invariants under transformation groups.
1870s–1890s — Sophus Lie develops continuous transformation groups (Lie groups) to study differential equations.
1897 — Burnside's Theory of Groups of Finite Order, the first comprehensive English treatise; later his lemma and $p^{a} q^{b}$ theorem.
1980–2004 — The classification of finite simple groups is completed, a collaborative proof spanning tens of thousands of journal pages.

Subtleties & Limits

Although the classification of finite simple groups is announced as complete, the proof is so vast (~ $10, 000 +$ pages by hundreds of authors) that an ongoing "second-generation" project led by Gorenstein, Lyons, Solomon, and Aschbacher aims to consolidate it into a self-contained, verifiable ~ $5, 000$ -page treatment — and gaps (e.g. the 2008 patch to the quasithin case) remind us that human-checked proofs of this scale carry genuine epistemic risk. For infinite groups the picture is wilder: the word problem (deciding whether a product of generators equals the identity) is undecidable in general (Novikov–Boone, 1955), so no algorithm can resolve equality in an arbitrary finitely-presented group. Open questions abound, such as the precise growth rates achievable by finitely generated groups and the structure of the Monster group's representations (the resolved moonshine conjectures connect it astonishingly to modular forms and string theory).