Chapter 06
Hard

Group Homomorphisms & Isomorphism Theorems

00 · Symbol Glossary

$\varphi: G \to H$Homomorphism from G to H

A function between two groups that respects the group operation: φ(ab)=φ(a)φ(b)\varphi(ab)=\varphi(a)\varphi(b). The single defining requirement of this chapter — everything else follows from it.

$\ker\varphi$Kernel of a homomorphism

kerφ={gG:φ(g)=eH}\ker\varphi = \{g\in G : \varphi(g) = e_H\} — everything that collapses to the identity. Measures exactly how far φ\varphi is from being injective.

$\mathrm{im}\,\varphi$Image of a homomorphism

imφ={φ(g):gG}H\mathrm{im}\,\varphi = \{\varphi(g) : g\in G\} \subseteq H — everything actually hit by φ\varphi. Measures exactly how far φ\varphi is from being surjective (onto HH).

$\cong$Isomorphic

GHG \cong H means there is a bijective homomorphism between them — the two groups are structurally identical, differing only in the names of their elements. The strongest possible relationship between two groups.

$e_G,\ e_H$Identity of G, identity of H

Subscripts distinguish which group's identity is meant, since a homomorphism deals with two groups at once and their identities are, a priori, different elements of different sets.

$HN$Product of two subsets

HN={hn:hH,nN}HN = \{hn : h\in H, n\in N\}. Not generally a subgroup unless at least one of H,NH,N is normal — a fact made precise in Section 07.


01 · Homomorphisms

Chapter 05 built one structure-preserving relationship between groups: the quotient G/NG/N. Homomorphisms generalize this idea to any structure-preserving map between any two groups — quotients will turn out to be a special case.

Definition — Group Homomorphism

Let G,HG,H be groups. A function φ:GH\varphi: G\to H is a homomorphism if:

φ(ab)=φ(a)φ(b)for all a,bG\varphi(ab) = \varphi(a)\varphi(b) \qquad \text{for all } a,b\in G

Note: the multiplication on the left is GG's operation; on the right, HH's.

Example — The sign homomorphism

Define sgn:Sn{1,1}\text{sgn}: S_n \to \{1,-1\} (with {1,1}\{1,-1\} a group under ordinary multiplication) by sgn(σ)=1\text{sgn}(\sigma)=1 if σ\sigma is even, 1-1 if odd. Since parity adds when permutations compose (even∘even=even, even∘odd=odd, odd∘odd=even — exactly matching 1×1=11\times1=1, 1×(1)=11\times(-1)=-1, (1)×(1)=1(-1)\times(-1)=1), sgn(στ)=sgn(σ)sgn(τ)\text{sgn}(\sigma\tau)=\text{sgn}(\sigma)\text{sgn}(\tau) — a genuine homomorphism.

Example — Reduction mod n

Define π:ZZn\pi:\mathbb{Z}\to\mathbb{Z}_n by π(a)=[a]\pi(a) = [a] (the congruence class of aa). Then π(a+b)=[a+b]=[a]+[b]=π(a)+π(b)\pi(a+b) = [a+b] = [a]+[b] = \pi(a)+\pi(b) — exactly Theorem 1.5's well-definedness statement, now recognized as the homomorphism property.

Common mistake — Confusing a homomorphism with an arbitrary function

The function f:ZZf:\mathbb{Z}\to\mathbb{Z}, f(n)=n+1f(n)=n+1, is not a homomorphism of (Z,+)(\mathbb{Z},+): f(a+b)=a+b+1f(a+b) = a+b+1, but f(a)+f(b)=(a+1)+(b+1)=a+b+2f(a)+f(b) = (a+1)+(b+1) = a+b+2 — different, unless coincidentally 1=21=2. A homomorphism is a very specific kind of function, not just "any" map between two groups' underlying sets.


02 · Basic Consequences

Theorem 6.1 — Homomorphisms Preserve Identity and Inverses

For any homomorphism φ:GH\varphi: G\to H: (a) φ(eG)=eH\varphi(e_G) = e_H; (b) φ(a1)=φ(a)1\varphi(a^{-1}) = \varphi(a)^{-1} for every aGa\in G.

Proof

(a) φ(eG)=φ(eGeG)=φ(eG)φ(eG)\varphi(e_G) = \varphi(e_G e_G) = \varphi(e_G)\varphi(e_G). Multiply both sides on the left by φ(eG)1\varphi(e_G)^{-1} (which exists in HH): eH=φ(eG)e_H = \varphi(e_G).

(b) φ(a)φ(a1)=φ(aa1)=φ(eG)=eH\varphi(a)\varphi(a^{-1}) = \varphi(aa^{-1}) = \varphi(e_G) = e_H by part (a). So φ(a1)\varphi(a^{-1}) satisfies the defining property of φ(a)1\varphi(a)^{-1}; by uniqueness of inverses (Theorem 1.2), φ(a1)=φ(a)1\varphi(a^{-1})=\varphi(a)^{-1}.

Theorem 6.2 — Homomorphic Images of Subgroups Are Subgroups

If φ:GH\varphi: G\to H is a homomorphism and KGK\le G, then φ(K)={φ(k):kK}H\varphi(K) = \{\varphi(k):k\in K\} \le H.

Proof

Nonempty: eH=φ(eG)φ(K)e_H = \varphi(e_G) \in \varphi(K) by Theorem 6.1(a), since eGKe_G\in K.

Closure: for φ(k1),φ(k2)φ(K)\varphi(k_1),\varphi(k_2)\in\varphi(K), φ(k1)φ(k2)=φ(k1k2)\varphi(k_1)\varphi(k_2) = \varphi(k_1k_2), and k1k2Kk_1k_2\in K (closure in KK), so this is in φ(K)\varphi(K).

Inverses: φ(k)1=φ(k1)\varphi(k)^{-1} = \varphi(k^{-1}) by Theorem 6.1(b), and k1Kk^{-1}\in K, so this is in φ(K)\varphi(K).

By Theorem 3.1, φ(K)H\varphi(K)\le H.


03 · Kernel and Image

Definition — Kernel and Image

For φ:GH\varphi: G\to H: kerφ={gG:φ(g)=eH}\ker\varphi = \{g\in G : \varphi(g)=e_H\}, and imφ=φ(G)={φ(g):gG}\mathrm{im}\,\varphi = \varphi(G) = \{\varphi(g):g\in G\}.

Theorem 6.3 — The Kernel Is a Normal Subgroup of G

kerφG\ker\varphi \trianglelefteq G.

Proof

Subgroup: nonempty, since φ(eG)=eH\varphi(e_G)=e_H (Theorem 6.1a) puts eGkerφe_G\in\ker\varphi. Closure: if φ(a)=eH\varphi(a)=e_H and φ(b)=eH\varphi(b)=e_H, then φ(ab)=φ(a)φ(b)=eHeH=eH\varphi(ab)=\varphi(a)\varphi(b) = e_He_H=e_H, so abkerφab\in\ker\varphi. Inverses: if φ(a)=eH\varphi(a)=e_H, then φ(a1)=φ(a)1=eH1=eH\varphi(a^{-1}) = \varphi(a)^{-1} = e_H^{-1}=e_H (Theorem 6.1b), so a1kerφa^{-1}\in\ker\varphi. By Theorem 3.1, kerφG\ker\varphi\le G.

Normal: take any gGg\in G and kkerφk\in\ker\varphi. Then:

φ(gkg1)=φ(g)φ(k)φ(g1)=φ(g)eHφ(g)1=φ(g)φ(g)1=eH\varphi(gkg^{-1}) = \varphi(g)\varphi(k)\varphi(g^{-1}) = \varphi(g)\,e_H\,\varphi(g)^{-1} = \varphi(g)\varphi(g)^{-1} = e_H

using φ(k)=eH\varphi(k)=e_H and φ(g1)=φ(g)1\varphi(g^{-1})=\varphi(g)^{-1} (Theorem 6.1b). So gkg1kerφgkg^{-1}\in\ker\varphi for every g,kg,k, meaning g(kerφ)g1kerφg(\ker\varphi)g^{-1}\subseteq\ker\varphi for every gg; applying this with g1g^{-1} gives the reverse inclusion. Hence kerφG\ker\varphi\trianglelefteq G.

By Theorem 6.2, imφ=φ(G)H\mathrm{im}\,\varphi = \varphi(G) \le H is also automatic (applying the theorem to K=GK=G).

Example — Kernel of the sign homomorphism

ker(sgn)={σSn:sgn(σ)=1}=An\ker(\text{sgn}) = \{\sigma \in S_n : \text{sgn}(\sigma)=1\} = A_n — the even permutations. This matches Chapter 05's finding that AnSnA_n \trianglelefteq S_n, now explained: AnA_n is normal precisely because it's the kernel of a homomorphism, not merely because it happens to have index 2.


04 · Injectivity and the Kernel

Theorem 6.4 — Injective iff Trivial Kernel

A homomorphism φ:GH\varphi: G\to H is injective if and only if kerφ={eG}\ker\varphi = \{e_G\}.

Proof

(\Rightarrow) Suppose φ\varphi injective. If gkerφg\in\ker\varphi, then φ(g)=eH=φ(eG)\varphi(g)=e_H=\varphi(e_G); injectivity forces g=eGg=e_G. So kerφ{eG}\ker\varphi\subseteq\{e_G\}, and since eGkerφe_G\in\ker\varphi always, kerφ={eG}\ker\varphi=\{e_G\}.

(\Leftarrow) Suppose kerφ={eG}\ker\varphi=\{e_G\}. If φ(a)=φ(b)\varphi(a)=\varphi(b), then φ(a)φ(b)1=eH\varphi(a)\varphi(b)^{-1}=e_H, i.e. φ(a)φ(b1)=eH\varphi(a)\varphi(b^{-1})=e_H (Theorem 6.1b), i.e. φ(ab1)=eH\varphi(ab^{-1})=e_H. So ab1kerφ={eG}ab^{-1}\in\ker\varphi=\{e_G\}, giving ab1=eGab^{-1}=e_G, i.e. a=ba=b. So φ\varphi is injective.

A one-element check replaces an all-pairs check

Ordinarily, proving a function injective means checking "φ(a)=φ(b)    a=b\varphi(a)=\varphi(b)\implies a=b" for every pair a,ba,b. Theorem 6.4 reduces this, for homomorphisms specifically, to checking a single set — the kernel — for a single property: does it contain anything besides eGe_G? This shortcut is used constantly from here on.


05 · Isomorphisms

Definition — Isomorphism

A homomorphism φ:GH\varphi: G\to H that is also a bijection is called an isomorphism. If one exists, GG and HH are isomorphic, written GHG\cong H.

Isomorphic groups are indistinguishable as groups

Every structural property phrased purely in terms of the group operation — order, abelian-ness, element orders, number of subgroups of each order, and so on — transfers exactly across an isomorphism. GHG\cong H means "the same group wearing different labels," which is why D3S3D_3 \cong S_3 (both size 6, both nonabelian, matching element-order profiles) was asserted without further proof back in Chapter 02.

Example — Verifying $D_3 \cong S_3$

Define φ:D3S3\varphi: D_3 \to S_3 by sending each symmetry to the permutation it induces on the triangle's vertices {1,2,3}\{1,2,3\}: r(123)r\mapsto(1\,2\,3), s(23)s\mapsto(2\,3) (fixing vertex 1), and extend to all 66 elements by composing. Composing two symmetries and then reading off vertex permutations gives the same answer as reading off vertex permutations first and then composing — so φ\varphi is a homomorphism. Since both groups have order 66 and φ\varphi can be checked directly to hit all six permutations without repeats, φ\varphi is bijective. D3S3D_3\cong S_3.


06 · The First Isomorphism Theorem

This theorem is the payoff for building kernels, images, and quotients side by side: every homomorphism factors, essentially uniquely, into a quotient map followed by an isomorphism.

Theorem 6.5 — First Isomorphism Theorem

For any homomorphism φ:GH\varphi: G\to H:

G/kerφ    imφG/\ker\varphi \;\cong\; \mathrm{im}\,\varphi

Proof

Write K=kerφK=\ker\varphi (normal by Theorem 6.3, so G/KG/K is a genuine group by Theorem 5.3). Define φˉ:G/Kimφ\bar\varphi: G/K \to \mathrm{im}\,\varphi by φˉ(gK)=φ(g)\bar\varphi(gK) = \varphi(g).

Well-defined. If gK=gKgK=g'K, then by Theorem 4.1, g1gK=kerφg^{-1}g'\in K=\ker\varphi, so φ(g1g)=eH\varphi(g^{-1}g')=e_H, i.e. φ(g)1φ(g)=eH\varphi(g)^{-1}\varphi(g')=e_H (Theorem 6.1b applied inside this computation), giving φ(g)=φ(g)\varphi(g')=\varphi(g). So φˉ(gK)\bar\varphi(gK) doesn't depend on which representative gg of the coset was used.

Homomorphism. φˉ(gKgK)=φˉ(ggK)=φ(gg)=φ(g)φ(g)=φˉ(gK)φˉ(gK)\bar\varphi(gK\cdot g'K) = \bar\varphi(gg'K) = \varphi(gg') = \varphi(g)\varphi(g') = \bar\varphi(gK)\bar\varphi(g'K).

Injective. If φˉ(gK)=eH\bar\varphi(gK)=e_H, then φ(g)=eH\varphi(g)=e_H, so gKg\in K, meaning gK=eKgK=eK — the identity coset. By Theorem 6.4 (applied to φˉ\bar\varphi itself), trivial kernel means injective.

Surjective onto imφ\mathrm{im}\,\varphi. Every element of imφ\mathrm{im}\,\varphi has the form φ(g)\varphi(g) for some gGg\in G, and φˉ(gK)=φ(g)\bar\varphi(gK)=\varphi(g) hits it directly.

φˉ\bar\varphi is a bijective homomorphism G/KimφG/K\to\mathrm{im}\,\varphi: an isomorphism. Hence G/kerφimφG/\ker\varphi\cong\mathrm{im}\,\varphi.

Example — Recovering $S_n/A_n \cong \{1,-1\}$

ker(sgn)=An\ker(\text{sgn}) = A_n, and im(sgn)={1,1}\mathrm{im}(\text{sgn}) = \{1,-1\} (sgn hits both values, e.g. sgn(e)=1, sgn of a transposition =1=-1). By Theorem 6.5:

Sn/An    {1,1}S_n/A_n \;\cong\; \{1,-1\}

This instantly confirms Sn/An=2|S_n/A_n|=2, matching Lagrange's Theorem applied to [Sn:An]=2[S_n:A_n]=2 from Chapter 05 — but now with an explicit isomorphism identifying which group of order 2 it is, not merely its size.

Example — $\mathbb{Z}/n\mathbb{Z} \cong \mathrm{im}$ of reduction mod n

For π:ZZn\pi:\mathbb{Z}\to\mathbb{Z}_n (reduction mod nn), kerπ=nZ\ker\pi = n\mathbb{Z} (exactly the multiples of nn, which map to [0][0]) and π\pi is surjective onto Zn\mathbb{Z}_n (every class [a][a] is hit by π(a)\pi(a)). Theorem 6.5 gives Z/nZZn\mathbb{Z}/n\mathbb{Z}\cong\mathbb{Z}_n — confirming, via the general machinery, exactly what Chapter 05 observed directly by hand.


07 · The Second and Third Isomorphism Theorems

Two further theorems, both consequences of the First, describe how quotients interact with subgroups and with each other.

Theorem 6.6 — Second Isomorphism Theorem

Let HGH\le G and NGN\trianglelefteq G. Then HN={hn:hH,nN}HN=\{hn:h\in H,n\in N\} is a subgroup of GG, HNHH\cap N \trianglelefteq H, and:

HN/N    H/(HN)HN/N \;\cong\; H/(H\cap N)

Proof

HNGHN\le G: nonempty (e=eeHNe=ee\in HN). Closure: (h1n1)(h2n2)=h1h2(h21n1h2)n2(h_1n_1)(h_2n_2) = h_1h_2(h_2^{-1}n_1h_2)n_2; since NGN\trianglelefteq G, h21n1h2Nh_2^{-1}n_1h_2\in N, so (h21n1h2)n2N(h_2^{-1}n_1h_2)n_2\in N, giving (h1n1)(h2n2)=(h1h2)(something in N)HN(h_1n_1)(h_2n_2) = (h_1h_2)\cdot(\text{something in } N) \in HN. Inverses: (hn)1=n1h1=h1(hn1h1)(hn)^{-1}=n^{-1}h^{-1} = h^{-1}(hn^{-1}h^{-1}), and hn1h1Nhn^{-1}h^{-1}\in N by normality, so (hn)1HN(hn)^{-1}\in HN. By Theorem 3.1, HNGHN\le G.

Define ψ:HHN/N\psi: H \to HN/N by ψ(h)=hN\psi(h) = hN. This is a homomorphism: ψ(h1h2)=h1h2N=(h1N)(h2N)=ψ(h1)ψ(h2)\psi(h_1h_2) = h_1h_2N = (h_1N)(h_2N) = \psi(h_1)\psi(h_2), valid since NHNN\trianglelefteq HN (as NGN\trianglelefteq G, normality holds against every element of GG, in particular every element of HNHN), so HN/NHN/N is a genuine group by Theorem 5.3.

ψ\psi is surjective onto HN/NHN/N: every coset in HN/NHN/N has the form (hn)N=hN(hn)N = hN (since nN=NnN=N) for some hHh\in H, which is ψ(h)\psi(h).

kerψ=HN\ker\psi = H\cap N: ψ(h)=N\psi(h)=N (the identity of HN/NHN/N)     hN=N    hN\iff hN=N \iff h\in N. Combined with hHh\in H (domain of ψ\psi), this is exactly hHNh\in H\cap N. By Theorem 6.3, kerψ=HNH\ker\psi=H\cap N \trianglelefteq H.

By the First Isomorphism Theorem applied to ψ\psi: H/kerψimψH/\ker\psi \cong \mathrm{im}\,\psi, i.e. H/(HN)HN/NH/(H\cap N) \cong HN/N.

Theorem 6.7 — Third Isomorphism Theorem

Let N,KGN,K\trianglelefteq G with NKN\le K. Then K/NG/NK/N \trianglelefteq G/N, and:

(G/N)/(K/N)    G/K(G/N)\big/(K/N) \;\cong\; G/K

Proof

Define Ψ:G/NG/K\Psi: G/N \to G/K by Ψ(gN)=gK\Psi(gN) = gK.

Well-defined: if gN=gNgN=g'N, then g1gNKg^{-1}g'\in N \subseteq K, so g1gKg^{-1}g'\in K, giving gK=gKgK=g'K.

Homomorphism: Ψ(gNgN)=Ψ(ggN)=ggK=(gK)(gK)=Ψ(gN)Ψ(gN)\Psi(gN\cdot g'N) = \Psi(gg'N) = gg'K = (gK)(g'K) = \Psi(gN)\Psi(g'N).

Surjective onto G/KG/K: every coset gKgK is Ψ(gN)\Psi(gN).

kerΨ=K/N\ker\Psi = K/N: Ψ(gN)=K\Psi(gN) = K (identity of G/KG/K)     gK=K    gK\iff gK=K \iff g\in K. The set of such cosets gNgN (for gKg\in K) is exactly K/N={kN:kK}K/N=\{kN : k\in K\}. By Theorem 6.3, K/N=kerΨG/NK/N=\ker\Psi \trianglelefteq G/N.

By the First Isomorphism Theorem: (G/N)/kerΨimΨ(G/N)/\ker\Psi \cong \mathrm{im}\,\Psi, i.e. (G/N)/(K/N)G/K(G/N)/(K/N) \cong G/K.

A recurring template

Every proof in this section follows the exact same three-step template: construct a natural map, identify its kernel, invoke Theorem 6.5. This is deliberate — the First Isomorphism Theorem is doing essentially all of the work every time; the Second and Third are really just clever choices of which map to feed into it.


08 · Exercises

EXERCISE 6.1

Apply Theorem 6.1 directly, using that det(I)=1\det(I)=1 is GLn(R)GL_n(\mathbb{R})'s identity's image.

det:GLn(R)R{0}\det: GL_n(\mathbb{R}) \to \mathbb{R}\setminus\{0\} satisfies det(AB)=det(A)det(B)\det(AB)=\det(A)\det(B) — a homomorphism. By Theorem 6.1(a), det(I)=1\det(I) = 1 (the identity of R{0}\mathbb{R}\setminus\{0\} under multiplication), confirming what's already known about determinants directly, but now as a forced consequence of the homomorphism property alone.

Confirm that det:GLn(R)R{0}\det: GL_n(\mathbb{R}) \to \mathbb{R}\setminus\{0\} is a homomorphism, and use Theorem 6.1(a) to state what det(I)\det(I) must equal.

EXERCISE 6.2

ker(det)\ker(\det) is exactly SLn(R)SL_n(\mathbb{R}) from Exercise 3.1. Apply Theorem 6.3 directly.

ker(det)={AGLn(R):detA=1}=SLn(R)\ker(\det) = \{A\in GL_n(\mathbb{R}) : \det A=1\} = SL_n(\mathbb{R}), exactly the subgroup from Exercise 3.1. By Theorem 6.3, the kernel of any homomorphism is normal: SLn(R)GLn(R)SL_n(\mathbb{R}) \trianglelefteq GL_n(\mathbb{R}) — a fact that would require direct conjugation computation with matrices to verify by hand, but follows immediately here.

Using det:GLn(R)R{0}\det: GL_n(\mathbb{R})\to\mathbb{R}\setminus\{0\} and Theorem 6.3, show that SLn(R)={A:detA=1}SL_n(\mathbb{R})=\{A:\det A=1\} is normal in GLn(R)GL_n(\mathbb{R}).

EXERCISE 6.3

Apply the First Isomorphism Theorem to det:GLn(R)R{0}\det: GL_n(\mathbb{R})\to\mathbb{R}\setminus\{0\}: what is its image?

det\det is surjective onto R{0}\mathbb{R}\setminus\{0\} (for any nonzero real cc, the diagonal matrix with cc in one entry and 11s elsewhere has determinant cc). By Theorem 6.5:

GLn(R)/SLn(R)    R{0}GL_n(\mathbb{R})/SL_n(\mathbb{R}) \;\cong\; \mathbb{R}\setminus\{0\}

State the conclusion of the First Isomorphism Theorem applied to det:GLn(R)R{0}\det: GL_n(\mathbb{R}) \to \mathbb{R}\setminus\{0\}.

EXERCISE 6.4

Use Theorem 6.4: compute the kernel directly, then decide if it's trivial.

φ:Z6Z3\varphi:\mathbb{Z}_6\to\mathbb{Z}_3 by φ([a]6)=[a]3\varphi([a]_6) = [a]_3 (reduce further mod 3). Check well-defined and homomorphism properties hold (similar to Theorem 1.5's argument). kerφ={[a]6:a0(mod3)}={0,3}\ker\varphi = \{[a]_6 : a\equiv0\pmod3\} = \{0,3\} — not just {0}\{0\}.

By Theorem 6.4, since kerφ{e}\ker\varphi\neq\{e\}, φ\varphi is not injective. (In fact kerφ=2|\ker\varphi|=2, so exactly two elements of Z6\mathbb{Z}_6 map to 00.)

Let φ:Z6Z3\varphi: \mathbb{Z}_6 \to \mathbb{Z}_3 send [a]6[a]3[a]_6 \mapsto [a]_3. Find kerφ\ker\varphi and determine, using Theorem 6.4, whether φ\varphi is injective.

EXERCISE 6.5

Apply the First Isomorphism Theorem to the φ\varphi from Exercise 6.4, using kerφ={0,3}\ker\varphi=\{0,3\} and surjectivity onto Z3\mathbb{Z}_3.

φ\varphi from Exercise 6.4 is surjective onto Z3\mathbb{Z}_3 (every class mod 3 is hit). By Theorem 6.5:

Z6/{0,3}    Z3\mathbb{Z}_6/\{0,3\} \;\cong\; \mathbb{Z}_3

Sanity check via order: Z6/{0,3}=6/2=3=Z3|\mathbb{Z}_6/\{0,3\}| = 6/2 = 3 = |\mathbb{Z}_3|. ✓ — consistent, and now with an explicit isomorphism rather than just matching cardinalities.

Using the map from Exercise 6.4, state what the First Isomorphism Theorem gives for Z6/kerφ\mathbb{Z}_6/\ker\varphi, and check the orders match.


09 · Chapter Summary

ConceptStatement
Homomorphismφ:GH\varphi:G\to H with φ(ab)=φ(a)φ(b)\varphi(ab)=\varphi(a)\varphi(b)
Preserves identity, inversesφ(eG)=eH\varphi(e_G)=e_H; φ(a1)=φ(a)1\varphi(a^{-1})=\varphi(a)^{-1} (Thm 6.1)
Image of a subgroupφ(K)H\varphi(K)\le H when KGK\le G (Thm 6.2)
Kernelkerφ={g:φ(g)=eH}\ker\varphi = \{g:\varphi(g)=e_H\}; always normal in GG (Thm 6.3)
Injective testφ\varphi injective     kerφ={eG}\iff \ker\varphi=\{e_G\} (Thm 6.4)
IsomorphismBijective homomorphism; GHG\cong H means structurally identical
First Isomorphism TheoremG/kerφimφG/\ker\varphi \cong \mathrm{im}\,\varphi (Thm 6.5)
Second Isomorphism TheoremHN/NH/(HN)HN/N \cong H/(H\cap N) (Thm 6.6)
Third Isomorphism Theorem(G/N)/(K/N)G/K(G/N)/(K/N) \cong G/K (Thm 6.7)

Next: Chapter 07 — Direct Products & Direct Sums builds larger groups out of smaller ones, giving the tools to construct (and recognize) groups like Z2×Z2\mathbb{Z}_2\times\mathbb{Z}_2 that Exercise 5.2 hinted might differ from Z4\mathbb{Z}_4 despite sharing the same order.