Groups — Axioms, Examples, and Basic Theorems

⚡ Key Takeaways

A group is a set GG equipped with a single associative binary operation that has an identity element and in which every element has an inverse. Groups formalize the abstract notion of symmetry and invertible transformation. They appear throughout mathematics (number theory, geometry, topology) and physics (conservation laws, quantum mechanics). The modern axiomatic definition crystallized in the late 19th and early 20th centuries from work of Galois, Cayley, and Weber.

💡 The Core Insight

A group abstracts the single property all systems of invertible operations share: you can undo any operation, and the order in which you compose operations may matter but never the grouping. Strip away everything except associativity, identity, and inverses, and what remains still has astonishing rigidity — almost everything else (uniqueness, cancellation, solving equations) follows automatically.

Formal Statement

DefinitionGroup

A group is an ordered pair (G,⋅)(G,\cdot) consisting of a set GG and a binary operation ⋅:G×G→G\cdot:G\times G\to G, written (a,b)↦a⋅b(a,b)\mapsto a\cdot b (often abbreviated abab), satisfying the following axioms:

(G0) Closure. For all a,b∈Ga,b\in G, the element a⋅ba\cdot b lies in GG.

(G1) Associativity.

1
(a⋅b)⋅c=a⋅(b⋅c)∀ a,b,c∈G.\quad (a\cdot b)\cdot c = a\cdot(b\cdot c)\qquad \forall\, a,b,c\in G.

Derivation hint: this is an axiom, not derived; it asserts that bracketing is irrelevant, so finite products a1a2⋯ana_1a_2\cdots a_n are unambiguous.

(G2) Identity. There exists an element e∈Ge\in G such that

2
e⋅a=a⋅e=a∀ a∈G.\quad e\cdot a = a\cdot e = a\qquad \forall\, a\in G.

Derivation hint: ee is required to be a two-sided identity; uniqueness is proved below, not assumed.

(G3) Inverses. For each a∈Ga\in G there exists a−1∈Ga^{-1}\in G such that

3
a⋅a−1=a−1⋅a=e.\quad a\cdot a^{-1} = a^{-1}\cdot a = e.

Derivation hint: each a−1a^{-1} is a two-sided inverse relative to the identity ee of (G2). Here GG is the underlying set, ⋅\cdot the group operation, ee (also written 11 or 00) the identity element, a−1a^{-1} the inverse of aa, and ∣G∣|G| the order of the group. If additionally

4
a⋅b=b⋅a∀ a,b∈G,\quad a\cdot b = b\cdot a\qquad \forall\, a,b\in G,

Derivation hint: commutativity is an optional extra axiom; groups satisfying it are special. the group is called abelian (or commutative), after Niels Henrik Abel. A group is finite if ∣G∣<∞|G|<\infty, otherwise infinite.

DefinitionSubgroup

A subset H⊆GH\subseteq G is a subgroup, written H≤GH\le G, if HH is itself a group under the restriction of ⋅\cdot. Equivalently (the subgroup criterion), H≤GH\le G iff H≠∅H\neq\varnothing and

5
a,b∈H  ⟹  a⋅b−1∈H.\quad a,b\in H \implies a\cdot b^{-1}\in H.

Derivation hint: nonemptiness gives some a∈Ha\in H, then aa−1=e∈Haa^{-1}=e\in H, then eb−1=b−1∈Heb^{-1}=b^{-1}\in H, then a(b−1)−1=ab∈Ha(b^{-1})^{-1}=ab\in H; associativity is inherited.

TheoremElementary Consequences of the Axioms

Let (G,⋅)(G,\cdot) be a group. Then: 1. The identity ee is unique. 2. Each a∈Ga\in G has a unique inverse a−1a^{-1}. 3. Cancellation: ab=ac  ⟹  b=cab=ac\implies b=c and ba=ca  ⟹  b=cba=ca\implies b=c. 4. Socks–shoes: (ab)−1=b−1a−1(ab)^{-1}=b^{-1}a^{-1}, and (a−1)−1=a(a^{-1})^{-1}=a. 5. For all a,b∈Ga,b\in G the equations ax=bax=b and ya=bya=b have unique solutions x=a−1bx=a^{-1}b, y=ba−1y=ba^{-1}.

Why It Holds / Derivation

Proof

We prove each elementary consequence using only (G0)–(G3).

(1) Uniqueness of identity. Suppose ee and e′e' both satisfy (G2). Then

6
e=e⋅e′=e′,\quad e = e\cdot e' = e',

Derivation hint: read e⋅e′e\cdot e' first using e′e' as a right identity (e⋅e′=ee\cdot e'=e is wrong order — instead use e′e' as identity to get e=e⋅e′e=e\cdot e', then ee as identity to get e⋅e′=e′e\cdot e'=e'). where the first equality uses that e′e' is an identity and the second that ee is an identity. Hence e=e′e=e'.

(2) Uniqueness of inverse. Suppose bb and cc both satisfy ab=ba=eab=ba=e and ac=ca=eac=ca=e. Then

7
b=b⋅e=b⋅(a⋅c)=(b⋅a)⋅c=e⋅c=c.\quad b = b\cdot e = b\cdot(a\cdot c) = (b\cdot a)\cdot c = e\cdot c = c.

Derivation hint: insert e=ace=ac, regroup with associativity (G1), then use ba=eba=e. So the inverse is unique.

(3) Cancellation. Assume ab=acab=ac. Multiply on the left by a−1a^{-1}:

8
a−1(ab)=a−1(ac)  ⇒  (a−1a)b=(a−1a)c  ⇒  eb=ec  ⇒  b=c.\quad a^{-1}(ab)=a^{-1}(ac)\;\Rightarrow\;(a^{-1}a)b=(a^{-1}a)c\;\Rightarrow\;eb=ec\;\Rightarrow\;b=c.

Derivation hint: apply (G1) to regroup, then (G3) to collapse a−1a=ea^{-1}a=e, then (G2). Right cancellation is symmetric, multiplying on the right by a−1a^{-1}.

(4) Socks–shoes. Compute, using associativity to reposition the inner inverse–element pair:

9
(ab)(b−1a−1)=a(bb−1)a−1=a e a−1=aa−1=e,\quad (ab)(b^{-1}a^{-1}) = a(b b^{-1})a^{-1} = a\,e\,a^{-1} = a a^{-1} = e,

Derivation hint: regroup with (G1) to bring bb−1bb^{-1} together, collapse to ee, then collapse aa−1aa^{-1}. and symmetrically (b−1a−1)(ab)=e(b^{-1}a^{-1})(ab)=e. By uniqueness (2), (ab)−1=b−1a−1(ab)^{-1}=b^{-1}a^{-1}. Since a−1a=aa−1=ea^{-1}a=aa^{-1}=e, the element aa satisfies the defining property of (a−1)−1(a^{-1})^{-1}, so (a−1)−1=a(a^{-1})^{-1}=a.

(5) Solving equations. Set x=a−1bx=a^{-1}b. Then ax=a(a−1b)=(aa−1)b=eb=bax=a(a^{-1}b)=(aa^{-1})b=eb=b, so a solution exists. If ax=ax′=bax=ax'=b, then x=x′x=x' by cancellation (3); hence it is unique. The case ya=bya=b is symmetric with y=ba−1y=ba^{-1}. ■\qquad\blacksquare

∎
⚠️ Common mistake

Axioms (G2) and (G3) are stated with two-sided identity and inverse. One can weaken them to a left identity and left inverses and still recover a group — but mixing a left identity with right inverses (or vice versa) does not in general give a group. The careful undergraduate proves the equivalence rather than assuming it.

mathematician · Évariste Galois | 1811–1832

Though the abstract axioms came later, Galois (who died in a duel at twenty) was the first to wield groups as a tool, attaching to each polynomial equation a finite group of permutations of its roots — the Galois group — and showing that the solvability of the equation by radicals is governed entirely by the structure of this group. His insight that a structure, not a formula, holds the key to a classical problem inaugurated abstract algebra. The word "group" (groupe) is his.

Building Intuition

🧠 Intuition

Think of a group as the collection of all the ways you can reversibly transform something while staying inside the same world. Pick up a square: the symmetries that map it onto itself (rotations by 0°,90°,180°,270°0°,90°,180°,270° and four reflections) form a group of order 8. Composition is "do one, then the other"; the identity is "do nothing"; the inverse undoes a move. Associativity is automatic because composing functions is associative. The single most important mental model is: group element = invertible action, group operation = composition of actions. Cayley's theorem makes this exact — every group is, up to relabeling, a group of permutations.

Figure
The dihedral group D4D_4: symmetries of the square. Rotations (gold) and reflection axes (gray) generate a group of order 8.
1 2 3 4 r = 90° rotation D₄ = { e, r, r², r³, s, sr, sr², sr³ }, |D₄| = 8 relations: r⁴ = e, s² = e, srs = r⁻¹

Worked Examples

Example

The integers modulo nn under addition, (Z/6Z,+)(\mathbb{Z}/6\mathbb{Z},+). Given: The set {0,1,2,3,4,5}\{0,1,2,3,4,5\} with operation a+b  6a+b \bmod 6. Step 1 — Closure & associativity. Modular addition stays inside {0,…,5}\{0,\dots,5\} and inherits associativity from Z\mathbb{Z}. Step 2 — Identity. 00 satisfies 0+a=a+0=a0+a=a+0=a, so e=0e=0. Step 3 — Inverses. The inverse of aa is 6−a6-a (and 00 is its own inverse). Check a=4a=4:

10
4+(6−4)=4+2=6≡0(mod6).\quad 4 + (6-4) = 4 + 2 = 6 \equiv 0 \pmod 6.

Derivation hint: the inverse of aa in Z/nZ\mathbb{Z}/n\mathbb{Z} is n−an-a for a≠0a\neq 0; verify by reducing a+(n−a)=n≡0a+(n-a)=n\equiv 0. *Step 4 — Solve an equation.* Find xx with 4+x≡1(mod6)4+x\equiv 1\pmod 6. Add the inverse 22 of 44:

11
x=(−4)+1=2+1=3(mod6).\quad x = (-4)+1 = 2+1 = 3 \pmod 6.

Derivation hint: apply x=a−1bx=a^{-1}b from consequence (5) with a−1=2a^{-1}=2, b=1b=1.

(Z/6Z,+) is an abelian group of order 6;x=3.\boxed{(\mathbb{Z}/6\mathbb{Z},+)\text{ is an abelian group of order }6;\quad x=3.}
Example

The general linear group GL2(R)GL_2(\mathbb{R}) — invertible 2×22\times2 real matrices under multiplication. Given: A=(2111)A=\begin{pmatrix}2&1\\1&1\end{pmatrix}, B=(1031)B=\begin{pmatrix}1&0\\3&1\end{pmatrix}. Step 1 — Membership. det⁡A=2⋅1−1⋅1=1≠0\det A = 2\cdot1-1\cdot1 = 1\neq0 and det⁡B=1≠0\det B = 1\neq 0, so both lie in GL2(R)GL_2(\mathbb{R}). Step 2 — Operation (non-commutative). Compute both products:

12
AB=(2⋅1+1⋅32⋅0+1⋅11⋅1+1⋅31⋅0+1⋅1)=(5141),BA=(2174).\quad AB=\begin{pmatrix}2\cdot1+1\cdot3 & 2\cdot0+1\cdot1\\ 1\cdot1+1\cdot3 & 1\cdot0+1\cdot1\end{pmatrix}=\begin{pmatrix}5&1\\4&1\end{pmatrix},\quad BA=\begin{pmatrix}2&1\\7&4\end{pmatrix}.

Derivation hint: ordinary row-times-column matrix multiplication; AB≠BAAB\neq BA shows the group is non-abelian. *Step 3 — Identity and inverse.* The identity is I=(1001)I=\begin{pmatrix}1&0\\0&1\end{pmatrix}. Since det⁡A=1\det A=1,

13
A−1=1det⁡A(1−1−12)=(1−1−12).\quad A^{-1}=\frac{1}{\det A}\begin{pmatrix}1&-1\\-1&2\end{pmatrix}=\begin{pmatrix}1&-1\\-1&2\end{pmatrix}.

Derivation hint: the 2×22\times2 inverse formula 1ad−bc(d−b−ca)\frac{1}{ad-bc}\begin{pmatrix}d&-b\\-c&a\end{pmatrix}; verify AA−1=IA A^{-1}=I.

GL2(R) is a non-abelian group; AB=(5141), A−1=(1−1−12).\boxed{GL_2(\mathbb{R})\text{ is a non-abelian group;}\ AB=\begin{pmatrix}5&1\\4&1\end{pmatrix},\ A^{-1}=\begin{pmatrix}1&-1\\-1&2\end{pmatrix}.}

Common Misconceptions

⚠️ Common mistake
  • Closure is part of the definition. "⋅\cdot is a binary operation on GG" already means a⋅b∈Ga\cdot b\in G; listing (G0) separately is a teaching emphasis, not a fifth law to forget.
  • Groups need not be commutative. ab=baab=ba is false in general (see GL2GL_2). Beginners routinely "cancel" or reorder illegally.
  • The identity belongs to every subgroup. A subgroup must contain ee and be closed under inverses; the empty set is not a subgroup.
  • One operation only. A group has a single operation. Rings and fields have two; conflating them is a common error.
Counterexample

The natural numbers under addition (N,+)(\mathbb{N},+) are NOT a group. Closure, associativity, and an identity (00, if you include it) hold, but there is no inverse: there is no x∈Nx\in\mathbb{N} with 1+x=01+x=0. Likewise (Z,⋅)(\mathbb{Z},\cdot) fails — 22 has no multiplicative inverse in Z\mathbb{Z}. And (R,⋅)(\mathbb{R},\cdot) fails because 00 has no inverse; only (R∖{0},⋅)(\mathbb{R}\setminus\{0\},\cdot) is a group. These show how delicate the inverse axiom is: removing a single bad element (like 00) can rescue or break the structure.

Connections

connection

Groups sit at the root of algebra and branch outward. Quotienting a group by a Normal Subgroups and Quotient Groups yields new groups; structure-preserving maps are studied in Group Homomorphisms and Isomorphism Theorems; the action of a group on a set, formalized in Group Actions and the Orbit-Stabilizer Theorem, makes precise the slogan "groups are symmetries." Finite groups decompose via Lagrange's Theorem and Cosets and the Sylow Theorems. Adding a second operation produces Rings and Fields, and equipping a group with a compatible topology or smooth structure leads to Lie Groups. The classification of finite simple groups is the deepest descendant of the basic axioms here.

Real-World Applications

🔬 Where this appears

Historical Development

⏳ History
- 1770 — Lagrange. Studies permutations of polynomial roots, implicitly using what becomes Lagrange's theorem; first stirrings of group thinking.
- 1799 — Ruffini. Attempts (incompletely) to prove the quintic is unsolvable, manipulating permutations.
- 1824–1826 — Abel. Proves the general quintic is unsolvable by radicals; commutative groups later named "abelian" in his honor.
- 1831–1832 — Galois. Introduces the term groupe and the decisive idea that a finite group governs solvability of equations; writes his foundational ideas the night before a fatal duel.
- 1854 — Cayley. Gives the first abstract definition of a finite group via a multiplication table, divorced from permutations.
- 1872 — Klein (Erlangen Program). Reorganizes geometry around its symmetry groups, making group theory central to all geometry.
- 1882 — von Dyck. Introduces generators and relations (presentations), and clarifies the abstract axioms.
- 1893–1896 — Weber & Frobenius. Weber gives essentially the modern axiomatic definition of an abstract (including infinite) group; Frobenius develops representation theory.
- 1980–2004 — Classification of Finite Simple Groups. A collaborative effort spanning tens of thousands of journal pages completes the periodic table of finite simple groups.

Subtleties & Limits

open-problem

The basic axioms are settled, but their consequences run deep and partly unresolved. The inverse Galois problem — is every finite group the Galois group of some extension of Q\mathbb{Q}? — remains open in general. In infinite group theory, decision problems are subtle: the word problem (does a given product of generators equal the identity?) is undecidable for general finitely presented groups (Novikov–Boone, 1955). The classification of finite simple groups, though declared complete, is so vast that its verification and simplification (the "second-generation proof") is an ongoing decades-long program. Even the count of groups of a given order grows mysteriously: there are 49,487,365,42249{,}487{,}365{,}422 groups of order 1024=2101024=2^{10}, and exact growth laws for pp-groups remain only partially understood.

See Also

Permutations and Symmetric Groups · Cyclic Groups and Generators · Lagrange's Theorem and Cosets · Group Homomorphisms and Isomorphism Theorems · Lie Groups · Rings and Fields · Sylow Theorems