Group Theory Introduction

A group is a set equipped with an associative binary operation that has an identity element and inverses for every element. Group theory studies symmetry abstractly: it captures the essential structure shared by integer addition, matrix multiplication, geometric rotations, and permutations. It applies anywhere symmetry, transformation, or reversible composition appears — from crystallography to quantum mechanics to error-correcting codes. Founded in the early 19th century by Évariste Galois and others to solve polynomial equations.

A group $(G, *)$ consists of a set $G$ and an operation $* : G \times G \to G$ satisfying four axioms: closure, associativity, existence of an identity, and existence of inverses. Group theory matters because it provides the rigorous language of symmetry, and symmetry underlies conservation laws in physics, solvability of equations in algebra, and structure in combinatorics. The surprising fact: despite the minimalism of its four axioms, group theory is rich enough to prove that there is no general algebraic formula (using radicals) for the roots of a degree-5 polynomial — a fact about equations decided entirely by the structure of an associated group.

The Core Insight

A group abstracts the notion of reversible, composable transformations. Forget what the objects are — numbers, rotations, shuffles — and keep only how they combine. Any system where you can (1) combine two operations into a third, (2) undo any operation, and (3) "do nothing," forms a group. This stripping-away reveals that wildly different-looking systems are secretly the same, and that the symmetries of an object form an algebraic structure you can compute with.

Formal Statement

Group

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

(G1) Closure. For all

a, b \in G

a * b \in G

. (G2) Associativity. For all

a, b, c \in G

(a * b) * c = a * (b * c)

⟳This is an axiom, not derived — it permits unambiguous writing of

a * b * c

without parentheses.

(G3) Identity. There exists

e \in G

such that for all

a \in G

e * a = a * e = a

⟳

e

is "do nothing"; uniqueness is proved below, not assumed.

(G4) Inverses. For each

a \in G

there exists

a^{- 1} \in G

such that:

a * a^{- 1} = a^{- 1} * a = e

⟳Every transformation can be undone; uniqueness of

a^{- 1}

is also a theorem.

Symbols: $G$ is the underlying set (its size $∣ G ∣$ is the order of the group); $*$ is the group operation; $e$ is the identity; $a^{- 1}$ is the inverse of $a$ . When $*$ is understood we write $ab$ for $a * b$ and call $G$ "the group."

Abelian Group

A group $(G, *)$ is abelian (or commutative) if additionally

a * b = b * a for all a, b \in G .

⟳Commutativity is an extra condition; most groups are NOT abelian (e.g. matrix groups, permutation groups).

Basic Consequences of the Axioms

In any group $(G, *)$ :

The identity $e$ is unique.
Each element's inverse is unique.
Cancellation holds: $ab = a c ⟹ b = c$ , and $ba = c a ⟹ b = c$ .
$(a^{- 1})^{- 1} = a$ and $(ab)^{- 1} = b^{- 1} a^{- 1}$ (note the order reversal).
For each $a, b \in G$ , the equations $a x = b$ and $y a = b$ have unique solutions $x = a^{- 1} b$ , $y = b a^{- 1}$ .

of Theorem: uniqueness, cancellation, socks-shoes

Uniqueness of identity (1). Suppose

e

and

e^{'}

both satisfy (G3). Then treating

e^{'}

as identity,

e * e^{'} = e

; treating

e

as identity,

e * e^{'} = e^{'}

. Hence

e = e * e^{'} = e^{'}

. Uniqueness of inverse (2). Suppose

b

and

c

are both inverses of

a

. Then

b = b * e = b * (a * c) = (b * a) * c = e * c = c .

⟳insert

e = a * c

, reassociate using (G2), then collapse

b * a = e

Cancellation (3). From

ab = a c

, left-multiply by

a^{- 1}

a^{- 1} (ab) = a^{- 1} (a c) \Rightarrow (a^{- 1} a) b = (a^{- 1} a) c \Rightarrow e b = ec \Rightarrow b = c .

⟳apply (G4) to create

e

, (G2) to reassociate, (G3) to drop

e

Socks-shoes (4). Compute

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

, and similarly on the other side. By uniqueness of inverse,

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

. Solvability (5).

x = a^{- 1} b

satisfies

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

. Uniqueness follows from cancellation.

■

Évariste Galois | 1811–1832

A French prodigy who, before dying in a duel at age 20, transformed the theory of equations. Galois introduced the concept that the symmetries of the roots of a polynomial form a group (now the Galois group), and that whether the polynomial is solvable by radicals depends on a structural property of that group (solvability in the group-theoretic sense). His night-before-the-duel manuscript laid foundations of modern abstract algebra. Though the word "group" (groupe) is his, the formal axioms crystallized only decades later through Cayley, Dyck, and Weber.

Building Intuition

Think of a group as the complete list of ways to transform an object while preserving its essential structure, together with the rule for doing two transformations in a row.

Take an equilateral triangle. Rotate it by $120°$ — it looks identical. Reflect it across a vertical axis — identical again. The set of all such symmetry operations (3 rotations, 3 reflections, 6 total) forms a group called $D_{3}$ . The operation is "do one, then the other." The identity is "leave it alone." Every move can be undone. Associativity is automatic because composing functions is associative.

The genius of abstraction is that this same group $D_{3}$ also describes the permutations of 3 objects, and a certain group of $2 \times 2$ matrices. They are isomorphic — structurally identical. Group theory lets you prove a fact once and apply it everywhere the structure appears.

Symmetries of the equilateral triangle forming the dihedral group $D_3$

<svg viewBox="0 0 460 300" xmlns="http://www.w3.org/2000/svg">
<rect width="460" height="300" fill="#0a0e17"/>
<text x="230" y="26" fill="#e2e8f0" font-size="15" text-anchor="middle">Dihedral group D₃ : 3 rotations + 3 reflections</text>

<!-- triangle -->
<polygon points="120,230 240,230 180,110" fill="none" stroke="#94a3b8" stroke-width="2"/>
<circle cx="120" cy="230" r="4" fill="#d4a853"/>
<circle cx="240" cy="230" r="4" fill="#d4a853"/>
<circle cx="180" cy="110" r="4" fill="#d4a853"/>
<text x="108" y="248" fill="#e2e8f0" font-size="13">1</text>
<text x="244" y="248" fill="#e2e8f0" font-size="13">2</text>
<text x="180" y="100" fill="#e2e8f0" font-size="13" text-anchor="middle">3</text>

<!-- rotation arrow -->
<path d="M 180 175 m 32 0 a 32 32 0 1 1 -16 -28" fill="none" stroke="#d4a853" stroke-width="2"/>
<polygon points="196,147 206,150 198,158" fill="#d4a853"/>
<text x="180" y="180" fill="#d4a853" font-size="11" text-anchor="middle">r (120°)</text>

<!-- reflection axes (dashed) -->
<line x1="180" y1="100" x2="180" y2="240" stroke="#94a3b8" stroke-width="1" stroke-dasharray="4 3"/>
<text x="186" y="262" fill="#94a3b8" font-size="11">mirror</text>

<!-- element list -->
<text x="320" y="120" fill="#e2e8f0" font-size="13">Elements:</text>
<text x="320" y="142" fill="#94a3b8" font-size="12">e,  r,  r²        (rotations)</text>
<text x="320" y="162" fill="#94a3b8" font-size="12">s,  rs,  r²s   (reflections)</text>
<text x="320" y="190" fill="#e2e8f0" font-size="12">Relations:</text>
<text x="320" y="210" fill="#d4a853" font-size="12">r³ = e,  s² = e</text>
<text x="320" y="228" fill="#d4a853" font-size="12">srs = r⁻¹  (non-abelian)</text>
</svg>

Worked Examples

The integers mod 6 under addition

Given.

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

with operation

* =

addition modulo

6

. Verify it's a group.

Closure: $a + b mod 6 \in {0, \dots, 5}$ . ✓
Associativity: inherited from integer addition. ✓
Identity: $e = 0$ , since $a + 0 = a$ . ✓
Inverses: $a^{- 1} = 6 - a$ for $a \neq = 0$ (and $0^{- 1} = 0$ ). E.g. $4^{- 1} = 2$ because $4 + 2 = 6 \equiv 0$ . ✓

Compute. Find the inverse of

5

and verify

5 + (5^{- 1}) = 0

5^{- 1} = 6 - 5 = 1, 5 + 1 = 6 \equiv 0 (mod 6) .

⟳additive inverse mod n is

n - a

Order of element 2. Repeatedly add:

2, 4, 0

— three steps to return to identity.

ord (2) = 3 (since 2 + 2 + 2 = 6 \equiv 0) .

⟳order = smallest

k > 0

with

k \cdot a \equiv 0

; here

k = 6/ g cd (2, 6) = 3

Boxed answer.

Z_{6} is an abelian group of order 6; 5^{- 1} = 1, ord (2) = 3.

A non-abelian group: $2\times 2$ invertible matrices

Given.

G = G L_{2} (R)

= real

2 \times 2

matrices with nonzero determinant, under matrix multiplication. Take

A = (1011), B = (1101) .

Step 1 — Group axioms. Identity is

I = (1001)

; inverses exist because

det \neq = 0

; associativity holds for matrix multiplication. Step 2 — Show non-commutativity. Compute both products.

A B = (1011) (1101) = (2111), B A = (1101) (1011) = (1112) .

⟳row-times-column entrywise; note

A B \neq = B A

Step 3 — Inverse of $A$ . For an upper-triangular unipotent matrix, invert the off-diagonal sign:

A^{- 1} = (10 - 1 1), A A^{- 1} = I .

⟳use

det A = 1

and the formula

A^{- 1} = \frac{1}{d e t A} (d - c - b a)

Boxed answer.

G L_{2} (R) is a non-abelian group; A B \neq = B A, A^{- 1} = (10 - 1 1) .

Common Misconceptions

Pitfalls when first learning groups:

"Groups must be commutative." False. Abelian groups are special; matrix and permutation groups generally are not. Always verify commutativity before assuming $ab = ba$ .

$(ab)^{- 1} = a^{- 1} b^{- 1}$ . Wrong in non-abelian groups — the order reverses: $(ab)^{- 1} = b^{- 1} a^{- 1}$ (socks-shoes principle).

"The operation must be addition or multiplication of numbers." The operation can be anything closed and associative: function composition, symmetric difference of sets, etc.

Confusing the identity with the number 1 or 0. The identity is whatever element acts as "do nothing" under the given operation; for $(Z, +)$ it is $0$ , for $(R^{\times}, \cdot)$ it is $1$ .

Forgetting closure / total definition. $Z$ under subtraction is not a group (subtraction is not associative). Always check all four axioms.

Structures that fail to be groups

$(N, +)$ (naturals $0, 1, 2, \dots$ ): no inverses — there is no natural number $n$ with $3 + n = 0$ . Not a group (only a monoid).
$(Z, \times)$ (integers under multiplication): identity is $1$ , but $2$ has no integer inverse ( $1/2 \in / Z$ ). Not a group.
$(R, \div)$ : division is not associative, e.g. $(8 \div 4) \div 2 = 1$ but $8 \div (4 \div 2) = 4$ . Not a group.
$3 \times 3$ matrices under multiplication (all of them): singular matrices ( $det = 0$ ) have no inverse. Not a group — you must restrict to $G L_{3}$ .

Connections

Group theory sits at the hub of abstract algebra and connects outward:

Subgroups and Cosets partition a group and lead to Lagrange's Theorem ( $∣ H ∣$ divides $∣ G ∣$ ), the first deep structural result.
Group Homomorphisms and Isomorphism Theorems formalize when two groups are "the same" or how one maps onto another via Normal Subgroups and Quotient Groups.
Adding a second compatible operation gives Ring Theory and Field Theory; a group with a topology gives Lie Groups.
Group Actions connect abstract groups back to concrete symmetry and to combinatorics via Burnside's Lemma.
In physics, Representation Theory realizes group elements as matrices, linking to Noether's Theorem (symmetry ⇒ conservation law).
The classification of Finite Simple Groups is the "periodic table" of group theory.

Real-World Applications

Crystallography. The 230 space groups classify all possible crystal symmetries; X-ray diffraction patterns are interpreted via group theory (used by the International Tables for Crystallography).
Quantum mechanics & particle physics. The gauge group $S U (3) \times S U (2) \times U (1)$ of the Standard Model organizes elementary particles; selection rules follow from representation theory of symmetry groups.
Public-key cryptography. RSA relies on the multiplicative group $(Z / n Z)^{\times}$ ; elliptic-curve cryptography (ECC) uses the group law on elliptic curves; Diffie–Hellman key exchange uses cyclic groups.
Error-correcting codes. Linear and cyclic codes (used in CDs, QR codes, deep-space communication) are subgroups of $F_{2}^{n}$ ; the Mathieu groups underlie the Golay code.
Rubik's Cube. The legal configurations form a group of order $\approx 4.3 \times 1 0^{19}$ ; solving algorithms exploit its subgroup structure (commutators, conjugates).
Chemistry / spectroscopy. Molecular point groups ( $C_{2 v}, T_{d}, O_{h}, \dots$ ) predict vibrational modes and which spectral transitions are allowed.
Music theory. Transposition and inversion of pitch classes form the dihedral group $D_{12}$ , analyzed in mathematical music theory ("neo-Riemannian" transformations).

Historical Development

1770 — Lagrange studies permutations of roots of polynomials, implicitly using what becomes Lagrange's theorem.
1799 — Ruffini gives an (incomplete) argument that the general quintic is unsolvable by radicals, using permutation ideas.
1824 — Abel rigorously proves the quintic is not solvable by radicals (Abel–Ruffini theorem).
1830–32 — Galois introduces "groups" of permutations and the solvability criterion; his work is published posthumously in 1846 by Liouville.
1854 — Cayley gives the first abstract definition of a (finite) group via a multiplication table, in "On the theory of groups."
1872 — Klein's Erlangen Program reframes geometry as the study of invariants under transformation groups.
1882 — von Dyck presents groups by generators and relations; Weber (1882, 1895) gives the modern axioms close to today's.
1872–1880s — Sylow proves his theorems on subgroups of prime-power order, a cornerstone of finite group theory.
2004 — Completion of the Classification of Finite Simple Groups, a collaborative effort spanning ~10,000 journal pages and decades, including the discovery of the Monster group.

Subtleties & Limits

Word problem. For groups given by generators and relations, deciding whether two words represent the same element is in general algorithmically undecidable (Novikov–Boone theorem, 1955–58). Group theory thus brushes the boundary of computability.
Classification beyond finite simple groups. While finite simple groups are classified, understanding all finite groups (built from simple ones via extensions) remains intractable; the "extension problem" is genuinely hard.
Infinite groups. Many natural questions about infinite groups (e.g. amenability, growth, the Burnside problem on whether a finitely generated torsion group can be infinite — answered yes by Golod–Shafarevich) reveal that infinite group theory is far wilder than the finite case.
Inverse Galois problem. Whether every finite group occurs as the Galois group of some extension of $Q$ remains open in general — a direct descendant of Galois's original questions.