Quotient group

In abstract algebra, a quotient group $ G/N $ of a group $ G $ by a normal subgroup $ N $ is the set of all left cosets of $ N $ in $ G $, equipped with the group operation defined by $ (g_1 N)(g_2 N) = (g_1 g_2) N $ for $ g_1, g_2 \in G $.^[1] This construction requires $ N $ to be normal in $ G $, meaning that $ g N g^{-1} = N $ for all $ g \in G $, or equivalently, that left and right cosets of $ N $ coincide, ensuring the operation is well-defined and the resulting structure forms a group.^[2] The order of the quotient group $ G/N $ equals the index $ [G : N] $, the number of distinct cosets, which generalizes Lagrange's theorem by relating subgroup sizes to the structure of $ G $.^[2] Quotient groups play a central role in group theory, facilitating the analysis of group homomorphisms through the first isomorphism theorem, which states that for any homomorphism $ f: G \to H $, $ G / \ker f \cong \operatorname{im} f $, where $ \ker f $ is the kernel and a normal subgroup of $ G $.^[2] Additional isomorphism theorems, such as the second (for subgroups $ K $ and normal $ N $, $ K / (K \cap N) \cong KN / N $) and third (for normal chains $ N \trianglelefteq K \trianglelefteq G $, $ (G/N) / (K/N) \cong G/K $), further illustrate how quotients decompose and classify group structures.^[2] As a foundational tool in modern algebra, quotient groups enable the study of symmetries and invariants by "factoring out" subgroups, with applications extending to fields like representation theory and classification of finite simple groups.^[2] Every subgroup of an abelian group is normal, making quotient constructions particularly straightforward in commutative settings, such as the integers modulo $ n $, which yield cyclic groups essential in number theory.^[2]

Fundamentals

Definition

A quotient group, also known as a factor group, is constructed from a group $G$ and a normal subgroup $N$ of $G$.^[3] Specifically, $G/N$ denotes the set of all left cosets of $N$ in $G$, given by $\{gN \mid g \in G\}$, where each coset $gN = \{gn \mid n \in N\}$ partitions the elements of $G$.^[4] A prerequisite for this construction is that $N$ must be a normal subgroup of $G$, meaning that for all $g \in G$ and $n \in N$, the conjugate $g^{-1}ng \in N$.^[4] Normality ensures that the set of cosets forms a group under the induced operation defined by $(gN)(hN) = (gh)N$ for $g, h \in G$.^[3] This operation is well-defined, independent of the choice of representatives $g$ and $h$, because $N$ is normal.^[4] The quotient $G/N$ satisfies the group axioms: the identity element is the coset $N$ itself (corresponding to the identity in $G$); the inverse of a coset $gN$ is $g^{-1}N$; and associativity follows from that in $G$.^[3] If $G$ and $N$ are finite, the order of the quotient group is given by $|G/N| = |G| / |N|$, as established by Lagrange's theorem.^[3]

Cosets and Normal Subgroups

In group theory, given a group $G$ and a subgroup $N \leq G$, the left coset of $N$ generated by an element $g \in G$ is the set $gN = \{gn \mid n \in N\}$.^[5] Similarly, the right coset is $Ng = \{ng \mid n \in N\}$.^[5] These cosets represent translates of the subgroup $N$ within $G$, and in general, left and right cosets may differ when $G$ is non-abelian. The collection of all left cosets of $N$ in $G$ forms a partition of $G$, meaning the cosets are disjoint and their union is $G$.^[6] The same holds for right cosets.^[6] Moreover, every left coset $gN$ and every right coset $Ng$ has the same cardinality as $N$, so $|gN| = |N|$ and $|Ng| = |N|$ for all $g \in G$.^[6] This equality follows from the bijection $h \mapsto gh$ between $N$ and $gN$, which preserves the group structure. To endow the set of cosets with a group operation, define multiplication of left cosets by $(gN)(hN) = (gh)N$. For this operation to be well-defined—independent of the choice of representatives $g$ and $h$—$N$ must be a normal subgroup of $G$.^[7] Specifically, if $g' N = g N$ and $h' N = h N$, then $g' = g n_1$ and $h' = h n_2$ for some $n_1, n_2 \in N$, so $g' h' = g n_1 h n_2$. The product $(g' h') N = g (n_1 h) n_2 N$ equals $g h N$ if and only if $n_1 h \in h N$, or equivalently, $n_1 h = h n_1'$ for some $n_1' \in N$. This holds for all such elements precisely when left and right cosets coincide, i.e., when $N$ is normal.^[7] A subgroup $N \leq G$ is normal if and only if every left coset equals the corresponding right coset, so $gN = Ng$ for all $g \in G$.^[8] Equivalently, $N$ is normal if it is invariant under conjugation by elements of $G$, meaning $g N g^{-1} = N$ for all $g \in G$.^[8] These criteria ensure the coset multiplication is associative and forms a group. For an illustration of a non-normal subgroup, consider the symmetric group $S_3$ on three letters and the subgroup $H = \{\mathrm{id}, (1\,3)\}$. To check normality, compute the conjugate $(1\,2) H (1\,2)^{-1}$. Since $(1\,2) \cdot \mathrm{id} \cdot (1\,2) = \mathrm{id}$ and $(1\,2) (1\,3) (1\,2) = (2\,3)$, the conjugate is $\{\mathrm{id}, (2\,3)\}$, which is not contained in $H$. Thus, $H$ is not normal in $S_3$.^[9] This failure implies that coset multiplication would not be well-defined for $H$.

Motivation and Construction

Origin of the Term "Quotient"

The term "quotient group" was introduced by William Burnside in his seminal 1897 textbook Theory of Groups of Finite Order, marking the first systematic use of the phrase in the context of abstract group theory.^[10] Earlier, in 1893, Arthur Cayley had referred to the structure G/H as a "quotient" without fully specifying the term for the group itself. This naming convention emerged in the late 19th century, paralleling Richard Dedekind's earlier introduction of quotient rings in 1871, where he developed the concept to handle factorization in rings of algebraic integers via ideals.^[11] Dedekind's work on quotient structures provided a foundational analogy that influenced group theorists, as both constructions involve dividing an algebraic object by a substructure to form a new entity. The choice of "quotient" reflects a direct analogy to division in arithmetic, particularly integer division. Just as dividing the integers ℤ by the subgroup nℤ yields the cyclic group ℤ/nℤ of order n, the quotient group G/N of a group G by a normal subgroup N has order |G|/|N| when finite, effectively "dividing out" the size of N to obtain a smaller group that captures the structure of G modulo N. This mirrors how remainders in division classify integers into equivalence classes, providing an intuitive bridge from elementary number theory to abstract algebra. Conceptually, forming the quotient group "factors out" the subgroup N by collapsing its elements to the identity, similar to how group presentations mod out by relations to define new groups. This process simplifies the original group by ignoring internal symmetries imposed by N, allowing focus on the coarser structure. In set-theoretic terms, the quotient arises from identifying elements that differ by elements of N, partitioning G into equivalence classes known as cosets, much like quotient sets in general equivalence relations.^[12] This identification preserves the group operation on the cosets, yielding a group that encodes G's behavior up to translation by N.

Homomorphism Theorem Connection

The first isomorphism theorem establishes a fundamental connection between group homomorphisms and quotient groups. Specifically, if $\phi: G \to H$ is a group homomorphism, then $G / \ker(\phi) \cong \operatorname{im}(\phi)$, where $\ker(\phi)$ denotes the kernel of $\phi$.^[13] If $\phi$ is surjective, this simplifies to $G / \ker(\phi) \cong H$.^[14] A key prerequisite is that the kernel $\ker(\phi)$ must be a normal subgroup of $G$. To see this, let $K = \ker(\phi)$ and take any $g \in G$, $a \in K$. Then $\phi(gag^{-1}) = \phi(g) \phi(a) \phi(g)^{-1} = \phi(g) \cdot e \cdot \phi(g)^{-1} = e$, where $e$ is the identity in $H$, so $gag^{-1} \in K$. Thus, $gKg^{-1} \subseteq K$ for all $g \in G$, confirming normality.^[14] The proof of the theorem proceeds by constructing an induced map from the quotient to the image. Define $\psi: G / K \to \operatorname{im}(\phi)$ by $\psi(gK) = \phi(g)$. This is well-defined because if $gK = g'K$, then $g'^{-1}g \in K$, so $\phi(g'^{-1}g) = e$ implies $\phi(g') = \phi(g)$. Moreover, $\psi$ is a homomorphism since $\psi((gK)(hK)) = \psi(ghK) = \phi(gh) = \phi(g)\phi(h) = \psi(gK) \psi(hK)$. It is injective because $\psi(gK) = e$ implies $\phi(g) = e$, so $g \in K$ and $gK = K$. Surjectivity follows from the definition of the image. Hence, $\psi$ is an isomorphism.^[13] This theorem also explains the construction of quotient groups via projections. For any normal subgroup $N \trianglelefteq G$, the canonical projection $\pi: G \to G/N$ defined by $\pi(g) = gN$ is a surjective homomorphism with $\ker(\pi) = N$. Applying the first isomorphism theorem yields $G/N \cong \operatorname{im}(\pi) = G/N$, which is tautological but confirms the setup.^[15] The theorem motivates quotient groups as a universal mechanism for factoring out normal subgroups: any homomorphism vanishing on $N$ factors uniquely through the projection $G \to G/N$, providing a canonical way to "mod out" by $N$.^[13]

Basic Examples

Integers Modulo n

The additive group of integers, denoted $\mathbb{Z}$, forms an infinite abelian group under addition, with the subgroup $n\mathbb{Z}$ consisting of all integer multiples of a fixed positive integer $n$. Since $\mathbb{Z}$ is abelian, every subgroup is normal, making $n\mathbb{Z}$ a normal subgroup of $\mathbb{Z}$.^[3] The cosets of $n\mathbb{Z}$ in $\mathbb{Z}$ are the sets of the form $k + n\mathbb{Z}$ for integers $k = 0, 1, \dots, n-1$, each representing a distinct residue class modulo $n$. These cosets partition $\mathbb{Z}$ and form the quotient group $\mathbb{Z}/n\mathbb{Z}$, where the group operation is defined by $(a + n\mathbb{Z}) + (b + n\mathbb{Z}) = (a + b) + n\mathbb{Z}$. This operation is well-defined because if $a' \equiv a \pmod{n}$ and $b' \equiv b \pmod{n}$, then $a' + b' \equiv a + b \pmod{n}$. The quotient group $\mathbb{Z}/n\mathbb{Z}$ is isomorphic to the cyclic group $\mathbb{Z}_n$ of order $n$, generated by the coset $1 + n\mathbb{Z}$.^[16] For a concrete illustration, consider $n = 6$. The cosets are:

• $0 + 6\mathbb{Z} = \{\dots, -12, -6, 0, 6, 12, \dots\}$,
• $1 + 6\mathbb{Z} = \{\dots, -11, -5, 1, 7, 13, \dots\}$,
• $2 + 6\mathbb{Z} = \{\dots, -10, -4, 2, 8, 14, \dots\}$,
• $3 + 6\mathbb{Z} = \{\dots, -9, -3, 3, 9, 15, \dots\}$,
• $4 + 6\mathbb{Z} = \{\dots, -8, -2, 4, 10, 16, \dots\}$,
• $5 + 6\mathbb{Z} = \{\dots, -7, -1, 5, 11, 17, \dots\}$.

Addition in this quotient group corresponds to addition modulo 6; for example, $(2 + 6\mathbb{Z}) + (3 + 6\mathbb{Z}) = 5 + 6\mathbb{Z}$, since $2 + 3 = 5 \equiv 5 \pmod{6}$. The identity element is $0 + 6\mathbb{Z}$, and each element has order dividing 6. The group $\mathbb{Z}/n\mathbb{Z}$ has finite order $n$, as there are exactly $n$ distinct cosets.^[3]

Even and Odd Integers

The even integers, denoted $2\mathbb{Z} = \{\dots, -4, -2, 0, 2, 4, \dots\}$, form a subgroup of the additive group of integers $\mathbb{Z}$.^[7] Since $\mathbb{Z}$ is abelian, every subgroup is normal, making $2\mathbb{Z}$ a normal subgroup of $\mathbb{Z}$.^[17] The cosets of $2\mathbb{Z}$ in $\mathbb{Z}$ partition the integers into two equivalence classes: the even integers themselves, represented as $2\mathbb{Z}$, and the odd integers, represented as $1 + 2\mathbb{Z} = \{\dots, -3, -1, 1, 3, \dots\}$.^[18] These cosets form the quotient group $\mathbb{Z}/2\mathbb{Z}$, where addition is defined by $(a + 2\mathbb{Z}) + (b + 2\mathbb{Z}) = (a + b) + 2\mathbb{Z}$.^[19] The group operation in $\mathbb{Z}/2\mathbb{Z}$ mirrors parity addition: even plus even yields even ($2\mathbb{Z} + 2\mathbb{Z} = 2\mathbb{Z}$), even plus odd yields odd ($2\mathbb{Z} + (1 + 2\mathbb{Z}) = 1 + 2\mathbb{Z}$), and odd plus odd yields even ($(1 + 2\mathbb{Z}) + (1 + 2\mathbb{Z}) = 2 + 2\mathbb{Z} = 2\mathbb{Z}$).^[1] This structure is isomorphic to the cyclic group of order 2, $\{0, 1\}$ under addition modulo 2, where $0$ corresponds to evens and $1$ to odds.^[8] Although $\mathbb{Z}$ is infinite, the quotient $\mathbb{Z}/2\mathbb{Z}$ is finite with two elements, demonstrating how quotienting by a normal subgroup can reduce an infinite group to a finite one.^[7] This quotient captures parity, which has applications in computing for error detection via parity bits—ensuring even or odd counts of 1s in binary data to identify transmission errors—and in number theory for analyzing properties like the distribution of primes or solving congruences based on evenness or oddness.^[20]

Roots of Unity

The group of $n$th roots of unity, denoted $\mu_n = \{ z \in \mathbb{C} \mid z^n = 1 \}$, forms a cyclic subgroup of order $n$ in the multiplicative group $\mathbb{C}^*$ of nonzero complex numbers.^[21] This subgroup is generated by a primitive $n$th root of unity, such as $e^{2\pi i / n}$, and consists of the points equally spaced on the unit circle in the complex plane.^[21] Since $\mathbb{C}^*$ is abelian, $\mu_n$ is a normal subgroup, and the quotient group $\mathbb{C}^*/\mu_n$ is well-defined under the induced multiplication of cosets.

Advanced Examples

Real Numbers Modulo Integers

The quotient group $\mathbb{R}/\mathbb{Z}$ is constructed from the additive group of real numbers $\mathbb{R}$ and its subgroup $\mathbb{Z}$ of integers, which is normal since $\mathbb{R}$ is abelian.^[22] The elements of $\mathbb{R}/\mathbb{Z}$ are the cosets $x + \mathbb{Z}$ for $x \in \mathbb{R}$, where each coset corresponds to the equivalence class of real numbers differing by an integer.^[23] These cosets can be represented uniquely by elements in the interval $[0, 1)$, identifying numbers that differ by integers.^[22] The group operation on $\mathbb{R}/\mathbb{Z}$ is induced from addition in $\mathbb{R}$: for cosets $x + \mathbb{Z}$ and $y + \mathbb{Z}$, their sum is $(x + y) + \mathbb{Z}$, which is equivalent to $(x + y \mod 1) + \mathbb{Z}$ using fractional parts.^[23] This makes $\mathbb{R}/\mathbb{Z}$ an abelian group, with the identity element $0 + \mathbb{Z}$ and inverses given by $(-x) + \mathbb{Z}$.^[22] The group $\mathbb{R}/\mathbb{Z}$ is isomorphic to the circle group $S^1 = \{ z \in \mathbb{C} \mid |z| = 1 \}$, the multiplicative group of complex numbers on the unit circle, via the map $\phi(x + \mathbb{Z}) = e^{2\pi i x}$.^[22][23] This isomorphism preserves the group structure, as $\phi((x + y) + \mathbb{Z}) = e^{2\pi i (x + y)} = e^{2\pi i x} e^{2\pi i y} = \phi(x + \mathbb{Z}) \phi(y + \mathbb{Z})$.^[23] Equipped with the quotient topology from the standard topology on $\mathbb{R}$, $\mathbb{R}/\mathbb{Z}$ is homeomorphic to the unit circle $S^1$, forming a compact topological group despite $\mathbb{Z}$ being an infinite discrete subgroup of $\mathbb{R}$.^[24] This compactness arises because the quotient identifies points separated by integers, "wrapping" the real line into a circle.^[23] In topology, $\mathbb{R}/\mathbb{Z}$ serves as a fundamental example of a quotient space and is used to study covering spaces and fundamental groups.^[24] In physics and analysis, $\mathbb{R}/\mathbb{Z}$ models periodic phenomena, such as waves or rotations, where the circle structure captures periodicity with period 1.^[25] It underlies Fourier analysis on the circle, where periodic functions on $[0, 1)$ (extended periodically) decompose into series of exponentials $e^{2\pi i n x}$ for $n \in \mathbb{Z}$, facilitating the study of signals and harmonic functions.^[25]

Matrix Groups

The general linear group $ \mathrm{GL}(n, \mathbb{R}) $ consists of all invertible $ n \times n $ matrices with real entries, and its center $ Z $ is the subgroup of scalar matrices $ \lambda I $ where $ \lambda \neq 0 $ and $ I $ is the identity matrix.^[26] This center $ Z $ is normal in $ \mathrm{GL}(n, \mathbb{R}) $ because it is central, meaning every element commutes with all others in the group, a property that ensures the quotient construction is well-defined.^[26] The quotient group $ \mathrm{GL}(n, \mathbb{R}) / Z $ is isomorphic to the projective linear group $ \mathrm{PGL}(n, \mathbb{R}) $, which acts on projective space by linear transformations modulo scaling.^[27] In this quotient, the cosets correspond to equivalence classes of matrices up to nonzero scalar multiplication, so two matrices $ A $ and $ B $ represent the same element if $ A = \lambda B $ for some $ \lambda \neq 0 $.^[28] The group operation on cosets is induced by matrix multiplication: $ (A Z)(B Z) = (AB) Z $, preserving the associative structure of the original group while identifying scaled matrices.^[26] This construction highlights the non-abelian nature of $ \mathrm{PGL}(n, \mathbb{R}) $ for $ n \geq 2 $, as matrix multiplication is generally non-commutative, and the normality of the center ensures the quotient inherits this property without collapsing to an abelian group.^[28] A specific example occurs for $ n=2 $, where the projective special linear group $ \mathrm{PSL}(2, \mathbb{R}) $ is the quotient of the special linear group $ \mathrm{SL}(2, \mathbb{R}) $ (matrices with determinant 1) by its center $ {\pm I} $.^[29] This group $ \mathrm{PSL}(2, \mathbb{R}) $ acts via Möbius transformations on the upper half-plane, preserving the hyperbolic metric and playing a key role in the study of Fuchsian groups and modular forms.^[30]

Symmetric Groups

The symmetric group $ S_n $ consists of all permutations of $ n $ elements, forming a group under composition with order $ n! $.^[31] The alternating group $ A_n $ is the subgroup of even permutations in $ S_n $, which is normal and has index 2, hence order $ n!/2 $.^[31] This normality follows from the fact that conjugation preserves the parity of permutations.^[31] The quotient group $ S_n / A_n $ is isomorphic to the cyclic group $ \mathbb{Z}/2\mathbb{Z} $, established via the sign homomorphism $ \operatorname{sgn}: S_n \to \mathbb{Z}/2\mathbb{Z} $, which maps even permutations to the identity and odd permutations to the generator of order 2.^[31] The kernel of this homomorphism is precisely $ A_n $, and by the first isomorphism theorem, the image is $ \mathbb{Z}/2\mathbb{Z} $.^[31] The two cosets are $ A_n $ itself (the even permutations) and the coset $ \tau A_n $ for any odd permutation $ \tau $, such as a transposition.^[31] For $ n=3 $, $ S_3 $ has order 6, and $ A_3 $ is the cyclic subgroup of order 3 generated by the 3-cycle $ (1,2,3) $.^[32] The quotient $ S_3 / A_3 $ is thus isomorphic to $ \mathbb{Z}/2\mathbb{Z} $, with the non-trivial coset consisting of the three transpositions.^[32] This structure has key applications: the sign homomorphism detects the parity of permutations, distinguishing even and odd elements and aiding in the classification of permutation representations.^[31] Moreover, for $ n \leq 4 $, the symmetric groups $ S_n $ are solvable, as they admit composition series with abelian factors, including the quotient by $ A_n $; this contrasts with $ S_n $ for $ n \geq 5 $, which are non-solvable.^[33]

Properties

Universal Property

The universal property of a quotient group characterizes it as the "freest" construction that factors out a normal subgroup. Specifically, let $G$ be a group and $N \trianglelefteq G$ a normal subgroup. Let $\pi: G \to G/N$ be the canonical projection homomorphism sending $g \mapsto gN$. Then, for any group $H$ and any homomorphism $\phi: G \to H$ such that $N \subseteq \ker \phi$, there exists a unique homomorphism $\overline{\phi}: G/N \to H$ such that $\phi = \overline{\phi} \circ \pi$.^[34] To see this, define $\overline{\phi}(gN) = \phi(g)$. This is well-defined because if $gN = g'N$, then $g'^{-1}g \in N \subseteq \ker \phi$, so $\phi(g') = \phi(g)$. Moreover, $\overline{\phi}$ preserves the group operation: $\overline{\phi}((gN)(g'N)) = \overline{\phi}(gg'N) = \phi(gg') = \phi(g)\phi(g') = \overline{\phi}(gN) \overline{\phi}(g'N)$. Finally, uniqueness follows since $\pi$ is surjective, so $\overline{\phi}$ is determined on all cosets by its values on generators via $\phi$.^[34] In categorical terms, $G/N$ together with $\pi$ is the initial object in the category whose objects are pairs $(K, \psi)$ where $K$ is a group and $\psi: G \to K$ is a homomorphism with $N \subseteq \ker \psi$, and whose morphisms are homomorphisms commuting with the maps from $G$. This means that for any other such pair $(K, \psi)$, there is a unique morphism $G/N \to K$ making the diagram commute.^[35] This property implies that the quotient $G/N$ is unique up to unique isomorphism: if $Q$ is another group with a surjective homomorphism $\rho: G \to Q$ such that $N = \ker \rho$, then there is a unique isomorphism $\iota: G/N \to Q$ with $\rho = \iota \circ \pi$. Thus, quotients provide a canonical way to classify homomorphic images modulo normal subgroups.^[35]

Isomorphism Theorems

The second isomorphism theorem states that if $G$ is a group, $H$ is a subgroup of $G$, and $K$ is a normal subgroup of $G$, then $HK$ is a subgroup of $G$ with $K$ normal in $HK$, $H \cap K$ is normal in $H$, and there is a natural isomorphism $HK / K \cong H / (H \cap K)$ given by $hK \mapsto h(H \cap K)$ for $h \in H$.^[36] To prove this, note that the map $\phi: H \to HK/K$ defined by $\phi(h) = hK$ is a surjective homomorphism because for any $hkK = hK$ with $h \in H$, $k \in K$, and the kernel of $\phi$ is precisely $H \cap K$, so the first isomorphism theorem yields the desired result.^[36] The third isomorphism theorem states that if $G$ is a group, $K$ is a normal subgroup of $G$, and $N$ is a normal subgroup of $G$ contained in $K$, then $K/N$ is normal in $G/N$, and there is a natural isomorphism $(G/N)/(K/N) \cong G/K$ given by $gN \cdot (K/N) \mapsto gK$ for $g \in G$.^[36] The proof proceeds by defining the map $\psi: G/N \to G/K$ via $\psi(gN) = gK$, which is a well-defined surjective homomorphism with kernel $K/N$, and thus the first isomorphism theorem applies.^[36] These theorems facilitate simplifying the structure of subgroups within quotient groups; for instance, the second isomorphism theorem identifies the projective special linear group $\mathrm{PSL}_2(\mathbb{C})$ as isomorphic to $\mathrm{SL}_2(\mathbb{C})/\{\pm I_2\}$ via the relation $\mathrm{GL}_2(\mathbb{C})/\mathbb{C}^\times I_2 \cong \mathrm{SL}_2(\mathbb{C})/\{\pm I_2\}$, where $\mathbb{C}^\times I_2$ is the center.^[36] Similarly, the third theorem shows that for integers $n$ dividing $m$, $(\mathbb{Z}/m\mathbb{Z}) / (n\mathbb{Z}/m\mathbb{Z}) \cong \mathbb{Z}/n\mathbb{Z}$, aiding computations of successive quotients.^[36]

Exact Sequences

In group theory, quotient groups arise naturally in the context of short exact sequences, which provide a framework for understanding extensions and homomorphisms between groups. A short exact sequence of groups is a sequence $1 \to N \xrightarrow{i} G \xrightarrow{\pi} Q \to 1$ where $i$ is an injective homomorphism, $\pi$ is a surjective homomorphism, and the image of $i$ equals the kernel of $\pi$.^[37] In the specific case of a quotient group, if $N$ is a normal subgroup of $G$, the sequence takes the form $1 \to N \xrightarrow{i} G \xrightarrow{\pi} G/N \to 1$, where $i$ embeds $N$ as a subgroup of $G$ and $\pi$ is the canonical projection onto the cosets $G/N$. This exactness ensures that $N$ captures precisely the elements mapping to the identity in $G/N$, highlighting the structural relationship between the subgroup, the group, and its quotient.^[38] Such sequences are fundamental in homological algebra, where they facilitate the study of derived functors and invariants of groups.^[39] A short exact sequence splits if there exists a homomorphism $s: G/N \to G$ such that $\pi \circ s = \mathrm{id}_{G/N}$, meaning the quotient can be "retrieved" as a subgroup complementing $N$ in $G$. In this case, $G$ is isomorphic to a semidirect product $N \rtimes (G/N)$.^[37] However, not all sequences split; for example, consider the sequence $1 \to \mathbb{Z} \xrightarrow{\times n} \mathbb{Z} \to \mathbb{Z}/n\mathbb{Z} \to 1$ for $n > 1$, where the first map multiplies by $n$. This is exact, but it does not split because there is no homomorphism $s: \mathbb{Z}/n\mathbb{Z} \to \mathbb{Z}$ satisfying $\pi \circ s = \mathrm{id}$, as $\mathbb{Z}$ is torsion-free while $\mathbb{Z}/n\mathbb{Z}$ has torsion.^[37] Non-splitting illustrates how quotient constructions can encode indecomposable structures. More generally, the group $G$ in the sequence $1 \to N \to G \to G/N \to 1$ represents an extension of the quotient $G/N$ by the normal subgroup $N$. In the abelian case, where all groups are abelian, such extensions are classified up to equivalence by the Ext group $\mathrm{Ext}^1_{\mathbb{Z}}(G/N, N)$, which parametrizes the possible ways to "glue" $N$ onto $G/N$ via Baer sum.^[40] The zero element in $\mathrm{Ext}^1_{\mathbb{Z}}(G/N, N)$ corresponds to the split (direct product) extension. For non-abelian groups, non-split extensions are classified by the second cohomology group $H^2(G/N, N)$, where $N$ is viewed as a module over $G/N$ via conjugation; this cohomology captures the obstructions to splitting through 2-cocycles defining the group operation.^[41][42]

Generalizations

Quotients of Lie Groups

In the context of Lie groups, which are smooth manifolds equipped with a group structure compatible with the manifold topology, the construction of quotient groups extends naturally to preserve the differential structure under certain conditions. Specifically, if $G$ is a Lie group and $N$ is a closed normal subgroup of $G$, then the quotient group $G/N$ inherits a unique smooth manifold structure making the canonical projection $\pi: G \to G/N$ a smooth submersion, and $G/N$ becomes a Lie group with this structure.^[43] The Lie algebra of $G/N$ is then the quotient Lie algebra $\mathfrak{g}/\mathfrak{n}$, where $\mathfrak{g} = \mathrm{Lie}(G)$ and $\mathfrak{n} = \mathrm{Lie}(N)$, reflecting the infinitesimal symmetries modulo those of the subgroup.^[43] This smoothness ensures that the group operations on $G/N$ are diffeomorphisms, allowing the quotient to model continuous symmetries in a differentiable manner. A prominent example arises in the special orthogonal group $SO(3)$, the Lie group of 3D rotations, which is diffeomorphic to the real projective space $\mathbb{RP}^3$. This identification can be viewed through the double cover by the special unitary group $SU(2)$, where $SU(2)/\{\pm I\} \cong SO(3) \cong \mathbb{RP}^3$, with $\{\pm I\}$ the center of $SU(2)$, a closed normal subgroup. Another illustrative case is the quotient $SU(2)/U(1)$, where $U(1)$ is embedded as the subgroup of diagonal matrices with determinant 1; this yields a smooth manifold diffeomorphic to the 2-sphere $S^2$, arising from the Hopf fibration and highlighting how quotients capture spherical geometries in Lie theory.^[44] Beyond normal subgroups, the notion of homogeneous spaces generalizes quotients to $G/H$, where $G$ is a Lie group and $H$ is any closed subgroup (not necessarily normal). In this setting, $G/H$ is a smooth manifold on which $G$ acts transitively by left multiplication via the projection map, with $H$ as the isotropy subgroup at the base point $[e]$.^[45] The smoothness follows from the closedness of $H$, ensuring the quotient topology is a manifold, and the action preserves the differential structure. For instance, $SO(3)/SO(2) \cong S^2$, where $SO(2)$ is the stabilizer of the z-axis, modeling the sphere as a space of directed axes under rotations.^[45] These constructions find significant applications in differential geometry and physics, particularly in modeling spaces of constant curvature and symmetry reductions. In physics, quotients like $G/H$ describe the coset spaces arising in spontaneous symmetry breaking, where a full symmetry group $G$ is reduced to a subgroup $H$ of unbroken symmetries, with Goldstone bosons parameterizing the manifold $G/H$; for example, the breaking of $SO(3)$ to $SO(2)$ yields $S^2$ as the space of vacua in certain field theories.^[46] Such homogeneous spaces underpin the study of invariant metrics and geodesic flows, essential for understanding symmetric configurations in general relativity and particle physics.

Abelianization

The abelianization of a group $ G $, denoted $ G^{\mathrm{ab}} $ or $ G/[G,G] $, is the quotient of $ G $ by its commutator subgroup $ [G,G] $, which is the normal subgroup generated by all commutators $ [x,y] = xyx^{-1}y^{-1} $ for $ x,y \in G $. This quotient is always abelian, as the relations imposed by the commutators force all elements to commute in the quotient. The commutator subgroup $ [G,G] $ is the smallest normal subgroup $ N \trianglelefteq G $ such that $ G/N $ is abelian, making $ G^{\mathrm{ab}} $ the largest abelian quotient of $ G $.^[35][47] The abelianization satisfies a universal property in the category of groups: for any group homomorphism $ \phi: G \to A $ where $ A $ is abelian, there exists a unique homomorphism $ \tilde{\phi}: G^{\mathrm{ab}} \to A $ such that $ \tilde{\phi} \circ \pi = \phi $, where $ \pi: G \to G^{\mathrm{ab}} $ is the canonical projection. This property characterizes the abelianization as the left adjoint to the inclusion functor from the category of abelian groups into the category of groups. As a functor $ (-)^{\mathrm{ab}}: \mathbf{Grp} \to \mathbf{Ab} $, it preserves certain structures and is essential in homological algebra for computing homology groups.^[35][48] Examples illustrate the abelianization's role in simplifying non-abelian groups. For the alternating group $ A_4 $, the commutator subgroup is the Klein four-group $ V_4 $, so $ A_4^{\mathrm{ab}} \cong \mathbb{Z}/3\mathbb{Z} $. Similarly, for the dihedral group $ D_4 $ of order 8, the commutator subgroup is $ \langle r^2 \rangle $, yielding $ D_4^{\mathrm{ab}} \cong V_4 $. In the case of free groups, the abelianization of the free group on $ n $ generators is the free abelian group $ \mathbb{Z}^n $.^[47]